12183 World Development Report 1993 Investing in Health Published for the World Bank Oxford University Press Oxford University Press OXFORD NEW YORK TORONTO DELHI BOMBAY CALCUTTA MADRAS KARACHI KUALA LUMPUR SINGAPORE HONG KONG TOKYO NAIROBI DARES SALAAM CAPE TOWN MELBOURNE AUCKLAND and associated companies in BERLIN IBADAN © 1993 The International Bank for Reconstruction and Development I THE WORLD BANK 1818 H Street, N. W, Washington, D.C. 20433 U.S.A. Published by Oxford University Press, Inc. 200 Madison Avenue, New York, N.Y. 10016 Oxford is a registered trademark of Oxford University Press. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Manufactured in the United States of America First printing June 1993 The maps that accompany the text have been prepared solely for the convenience of the reader; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city, or area, or of the authorities thereof, or concerning the delimitation of its boundaries or its national affiliation. The map on the cover, which shows the eight demographic regions used in the analysis in this Report, seeks to convey an impression of the general improvement in health experienced worldwide during the past forty years. ISBN 0-19-520889-7 clothbound ISBN 0-19-620890-0 paperback ISSN 0163-5085 Text printed on recycled paper that conforms to the American National Standard for Permanence of Paper for Printed Library Materials, 239.48-1984 Foreword World Development Report 1993, the sixteenth in this Second, government spending on health should annual series, examines the interplay between hu- , be redirected to more cost-effective programs that man health, health policy, and economic develop- do more to help the poor. Government spending ment. The three most recent reports-on the envi- accounts for half of the $168 billion annual expen- ronment, on development strategies, and on diture on health in developing countries. Too poverty-have furnished an overview of the goals much of this sum goes to specialized care in ter- and means of development. This year's report on tiary facilities that provides little gain for the health, like next year's on infrastructure, examines money spent. Too little goes to low-cost, highly in depth a single sector in which the impact of effective programs such as control and treatment public finance and public policy is of particular of infectious diseases and of malnutrition. Devel- importance. oping countries as a group could reduce their bur- Countries at all levels of income have achieved den of disease by 25 percent-the equivalent of great advances in health. Although an unaccepta- averting more than 9 million infant deaths-by re- bly high proportion of children in the developing directing to public health programs and essential world-one in ten-die before reaching age 5, this clinical services about half, on average, of the gov- number is less than half that of 1960. Declines in ernment spending that now goes to services of low poverty have allowed households to increase con- cost-effectiveness. sumption of the food, clean water, and shelter nec- Third, governments need to promote greater di- essary for good health. Rising educational levels versity and competition in the financing and deliv- have meant that people are better able to apply ery of health services. Government financing of new scientific knowledge to promote their own public health and essential clinical services would and their families' health. Health systems have leave the coverage of remaining clinical services to met the demand for better health through an ex- private finance, usually mediated through insur- panded supply of services that offer increasingly ance, or to social insurance. Government regula- potent interventions. tion can strengthen private insurance markets by Yet developing countries, and especially their improving incentives for wide coverage and for poor, continue to suffer a heavy burden of disease, cost control. Even for publicly financed clinical ser- much of which can be inexpensively prevented or vices, governments can encourage competition cured. (If the child mortality rate in developing and private sector involvement in service supply countries were reduced to the level that prevails in and can help improve the efficiency of the private high-income countries, 11 million fewer children sector by generating and disseminating key infor- would die each year.) Furthermore, increasing mation. The combination of these measures will numbers of developing countries are beginning to improve health outcomes and contain costs while face the problems of rising health system costs enhancing consumer satisfaction. now experienced by high-income countries. Significant reforms in health policy are feasible, This Report advocates a three-pronged ap- as experience in several developing countries has proach to government policies for improving shown. The donor community can assist by fi- health in developing countries. First, governments nancing the transitional costs of change, especially need to foster an economic environment that en= in low-income countries. The reforms outlined in abies households to improve their own health. this Report will translate into longer, healthier, and Growth policies (including, where necessary, eco- more productive lives for people around the nomic adjustment policies) that ensure income world, and especially for the more than 1 billion gains for the poor are essential. So, too, is ex- poor. panded investment in schooling, particularly for The World Health Organization (WHO) has girls. been a full partner with the World Bank at every iii step of the preparation of the Report. I would like World Bank is grateful to them as well . Specific to record my appreciation to WHO and to its many acknowledgments are provided elsewhere in the staff members at global and regional levels who Report. facilitated this partnership . The Report has bene- Like its predecessors, World Development Report fited greatly from WHO's extensive technical ex- 1993 includes the World Development Indicators, pertise. Starting from the Report's conception, which offer selected social and economic statistics WHO participated actively by providing data on on 127 countries. The Report is a study by the various aspects of health development and sys- Bank's staff, and the judgments made herein do tematic input for many technical consultations. not necessarily reflect the views of the Board of Perhaps WHO's most significant contribution was Directors or of the governments they represent . in a jointly sponsored assessment of the global burden of disease, which is a key element of the Report. I look forward to continued collaboration between the World Bank and WHO in the discus- Lewis T. Preston sion and implementation of the messages in this President Report. The United Nations Children's Fund The World Bank (UNICEF), bilateral agencies, and other institu- tions also contributed their expertise, and the May 31, 1993 This Report has been prepared by a team led by Dean T. jamison and comprising jose-Luis Bobadilla, Robert Hecht, Kenneth Hill, Philip Musgrove, Helen Saxenian, jee-Peng Tan, and, part-time, Seth Berkley and Christopher J. L. Murray. Anthony R. Measham drafted and coordinated contributions from the Bank's Population, Health, and Nutrition Department. Valuable contributions and advice were provided by Susan Cochrane, Thomas W. Merrick, W. Henry Mosley, Alexander Preker, Lant Pritchett, and Michael Walton. Extensive input to the Report from the World Health Organization was coordinated through a Steering Committee chaired by jean-Paul jardel. An Advisory Committee chaired by Richard G. A. Feachem provided valuable guidance at all stages of the Report's prepara- tion. Members of these committees are listed in the Acknowledgments . Peter Cowley, Anna E. Maripuu, Barbara J. McKinney, Karima Saleh, and Abdo S . Yazbeck served as research associates, and interns Lecia A. Brown, Caroline) . Cook, Anna Godal, and Vito Luigi Tanzi assisted the team . The work was carried out under the general direction of Lawrence H. Summers and Nancy Birdsall. Many others inside and outside the Bank provided helpful comments and contributions (see the Bibliographical note). The Bank's International Economics Department contributed to the data appen- dix and was responsible for the World Development Indicators . The production staff of the Report included Ann Beasley, Stephanie Gerard, jane Gould, Kenneth Hale, jeffrey N . Lecksell, Nancy Levine, Hugh Nees, Kathy Rosen, and Walton Rosenquist. The support staff was headed by Rhoda Blade-Charest and included Laitan Alii and Nyambura Kimani. Trinidad S. Angeles served as admin- istrative assistant. John Browning was the principal editor, and Rupert Pennant-Rea edited two chapters . Preparation of this Report was immensely, aided by contributions of the participants in a series of consultations and seminars; the subjects and the names of participants are listed in the Acknowledg- ments. The consultations could not have occurred without financial cooperation from the following organizations, whose assistance is warmly acknowledged: the Canadian International Development Association, the Danish International Development Agency, the Edna McConnell Clark Foundation, the Norwegian Ministry of Foreign Affairs, the Rockefeller Foundation, the Swiss Development Cooperation, the U.S . Agency for International Development, the Overseas Development Adminis- tration of the United Kingdom, and the Environmental Health Division and the Special Programme for Research and Training in Tropical Diseases of the World Health Organization. The World Health Organization and the United Nations Children's Fund contributed to the preparation of the statistical appendices. Three academic institutions-the Harvard Center for Population and Development Studies, the London School of Hygiene and Tropical Medicine, and the Swiss Tropical Institute- provided important support for the preparation of the Report . iv Contents Definitions and data notes x Overview 1 Health systems and their problems 3 The roles of the government and of the market in health 5 Government policies for achieving health for all 6 Improving the economic environment for healthy households 7 Investing in public health and essential clinical services 8 Reforming health systems: promoting diversity and competition 11 An agenda for action 13 1 Health in developing countries: successes and challenges 17 Why health matters 17 The record of success 21 Measuring the burden of disease 25 Challenges for the future 29 Lessons from the past: explaining declines in mortality 34 The potential for effective action 35 2 Households and health 37 Household capacity: income and schooling 38 Policies to strengthen household capacity 44 What can be done? · 51 3 The roles of the government and the market in health 52 Health expenditures and outcomes 53 The rationales for government action 54 Value for money in health 59 Health policy and the performance of health systems 65 4 Public health 72 Population-based health services 72 Diet and nutrition 75 Fertility 82 Reducing abuse of tobacco, alcohol, and drugs 86 Environmental influences on health 90 AIDS: a threat to development 99 The essential public health package . 106 5 Clinical services 108 Public and private finance of clinical services 108 Selecting and financing the essential clinical package 112 Insurance and finance of discretionary clinical services 119 Delivery of clinical services 123 Reorienting clinical services and beyond 132 v 6 Health inputs 134 Reallocating investments in facilities and equipment 134 Addressing imbalances in human resources 139 Improving the selection, acquisition, and use of drugs 144 Generating information and strengthening research 148 7 An agenda for action 156 Health policy reform in developing countries 156 International assistance for health 165 Meeting the challenges of health policy reform 170 Acknowledgments 172 Bibliographical note 176 Appendix A. Population and health data 195 Appendix B. The global burden of disease, 1990 213 World Development Indicators 227 Boxes 1 Investing in health: key messages of this Report 6 2 The World Summit for Children 15 1.1 Controlling river blindness 19 1.2 The economic impact of AIDS 20 1.3 Measuring the burden of disease 26 1.4 The demographic and epidemiological transitions 30 2.1 Progress in child health in four countries 38 2.2 Teaching schoolchildren about health: radio instruction in Bolivia 48 2.3 Violence against women as a health issue 50 3.1 Paying for tuberculosis control in China 58 3.2 Cost information and management decisions in a Brazilian hospital 60 3.3 Cost-effectiveness of interventions against measles and tuberculosis 63 3.4 Priority health problems: high disease burdens and cost-effective interventions 64 4.1 Women's nutrition 76 4.2 The Tamil Nadu Integrated Nutrition Project: making supplementary feeding work 80 4.3 World Bank policy on tobacco 89 4.4 After smallpox: slaying the dragon worm 92 4.5 The costs and benefits of investments in water supply and sanitation 93 4.6 Environmental and household control of mosquito vectors 94 4.7 Air pollution and health in Central Europe 97 4.8 Pollution in Japan : prevention would have been better and cheaper than cure 98 4.9 Coping with AIDS in Uganda 104 4.10 HIV in Thailand: from disaster toward containment 105 5.1 Making pregnancy and delivery safe 113 5.2 Integrated management of the sick child 114 5.3 Treatment of sexually transmitted diseases 115 5.4 Short-course treatment of tuberculosis .1 16 5.5 Targeting public expenditure to the poor 119 5.6 Containing health care costs in industrial countries 122 5.7 Health care reform in the OECD 125 5.8 Traditional medical practitioners and the delivery of essential health services 129 5.9 "Managed competition" and health care reform in the United States 132 6.1 International migration and the global market for health professionals 141 6.2 Community health workers 143 vi 6.3 Buying right: how international agencies save on purchases of pharmaceuticals 146 6.4 The contribution of standardized survey programs to health information 149 6.5 Evaluating cesarean sections in Brazil 150 6.6 An unmet need: inexpensive and simple diagnostics for STDs 154 7.1 Community financing of health centers: the Bamako Initiative 159 7.2 Health sector reforms in Chile 162 7.3 Reform of the Russian health system 164 7.4 Health assistance and the effectiveness of aid 168 7.5 World Bank support for reform of the health sector 169 7.6 Donor coordination in the health sector in Zimbabwe and Bangladesh 170 Text figures 1 Demographic regions used in this Report 2 2 Burden of disease attributable to premature mortality and disability, by demographic region, 1990 3 3 Infant and adult mortality in poor and nonpoor neighborhoods of Porto Alegre, Brazil, 1980 7 1.1 Child mortality by country, 1960 and 1990 22 1.2 Trends in life expectancy by demographic region, 1950-90 23 1.3 Age-standardized female death rates in Chile and in England and Wales, selected years 24 1.4 Change in female age-specific mortality rates in Chile and in England and Wales, selected years 24 1.5 Disease burden by sex and demographic region, 1990 28 1.6 Distribution of disability-adjusted life years (DALYs) lost, by cause, for selected demographic regions, 1990 29 1.7 Trends in life expectancy and fertility in Sub-Saharan Africa and Latin America and the Caribbean, 1960-2020 30 1.8 Median age at death, by demographic region, 1950, 1990, and 2030 32 1. 9 Life expectancy and income per capita for selected countries and periods 34 2.1 Mutually reinforcing cycles: reduction of poverty and development of human resources 37 2.2 Child mortality in rich and poor neighborhoods in selected metropolitan areas, late 1980s 40 2.3 Declines in child mortality and growth of income per capita in sixty-five countries 41 2.4 Effect of parents' schooling on the risk of death by age 2 in selected countries, late 1980s 43 2.5 Schooling and risk factors for adult health, Porto Alegre, Brazil, 1987 44 2.6 Deviation from mean levels of public spending on health in countries receiving and not receiving adjustment lending, 1980-90 46 2.7 Enrollment ratios in India, by grade, about 1980 47 3.1 Life expectancies and health expenditures in selected countries: deviations from estimates based on GDP and schooling 54 3.2 Benefits and costs of forty-seven health interventions 62 4.1 Child mortality (in specific age ranges) and weight-for-age in Bangladesh, India, Papua New Guinea, and Tanzania 77 4.2 Total fertility rates by demographic region, 1950-95 82 4.3 Risk of death by age 5 for fertility-related risk factors in selected countries, late 1980s 83 4.4 Maternal mortality in Romania, 1965-91 86 4.5 Trends in mortality from lung cancer and various other cancers among U.S. males, 1930-90 88 4.6 Population without sanitation or water supply services by demographic region, 1990 91 4.7 Simulated AIDS epidemic in a Sub-Saharan African country 100 4.8 Trends in new HIV infections under alternative assumptions, 1990-2000: Sub-Saharan Africa and Asia 101 5.1 Income and health spending in seventy countries, 1990 110 5.2 Public financing of health services in low- and middle-income countries, 1990 117 6.1 The health system pyramid: where care is provided 135 6.2 Hospital capacity by demographic region, about 1990 136 6.3 Supply of health personnel by demographic region, 1990 or most recent available year 140 7.1 Disbursements of external assistance for the health sector, 1990 166 vii Tables 1 Population, economic indicators, and progress in health by demographic region, 1975-90 2 2 Estimated costs and health benefits of the minimum package of public health and essential clinical services in low- and middle-income countries, 1990 10 3 Contribution of policy change to objectives for the health sector 14 1.1 Burden of disease by sex, cause, and type of loss, 1990 25 1.2 Burden of five major diseases by age of incidence and sex, 1990 28 1.3 Evolution of the HIV-AIDS epidemic 3~ 2.1 Poverty and growth of income per capita by developing region, 1985 and 1990, and long- and medium-term trends 42 3.1 Global health expenditure, 1990 52 3.2 Actual and proposed allocation of public expenditure on health in developing countries, 1990 66 3.3 Total cost and potential health gains of a package of public health and essential clinical services, 1990 68 4.1 Burden of childhood diseases preventable by the Expanded Programme on Immunization (EPI) by demographic region, 1990 73 4.2 Costs and health benefits of the EPI Plus cluster in two developing country settings, 1990 74 4.3 Direct and indirect contributions of malnutrition to the global burden of disease, 1990 76 4.4 Cost-effectiveness of nutrition interventions 82 4.5 Estimated burden of disease from poor household environments in demographically developing countries, 1990, and potential reduction through improved household services 90 4.6 Estimated global burden of disease from selected environmental threats, 1990, and potential worldwide reduction through environmental interventions 95 4.7 Costs and health benefits of public health packages in low- and middle-income countries, 1990 106 5.1 Rationales and directions for government action in the finance and delivery of clinical services 109 5.2 Clinical health systems by income group 111 5.3 Estimated costs and health benefits of selected public health and clinical services in low- and middle- income countries, 1990 117 5.4 Social insurance in selected countries, 1990 120 5.5 Strengths and weaknesses of alternative methods of paying health providers 124 5.6 Policies to improve delivery of health care 126 6.1 Annual drug expenditures per capita, selected countries, 1990 145 6.2 Some priorities for research and product development, ranked by the top six contributors to the global burden of disease 152 7.1 The relevance of policy changes for three country groups 157 7.2 Official development assistance for health by demographic region, 1990 167 Appendix tables A.1 Population (midyear) and average annual growth 199 A.2 GNP, population, GNP per capita, and growth of GNP per capita 199 A.3 Population structure and dynamics 200 A.4 Population and deaths by age group 202 A.5 Mortality risk and life expectancy across the life cycle 203 A.6 Nutrition and health behavior 204 A.7 Mortality, by broad cause, and tuberculosis incidence 206 A.8 Health infrastructure and services 208 A.9 Health expenditure and total flows from external assistance 210 A.10 Economies and populations by demographic region, mid-1990 212 B.1 Burden of disease by age and sex, 1990 215 B.2 Burden of disease in females by cause, 1990 216 B.3 Burden of disease in males by cause, 1990 218 viii B.4 Burden of disease by age and the three main groups of causes, 1990 220 B.S Burden of disease by consequence, sex, and age, 1990 221 B.6 Distribution of the disease burden in children in demographically developing economies, 3howing the ten main causes, 1990 222 B.7 Distribution of the disease burden in the adult and elderly populations in demographically developing economies, showing the ten main causes, 1990 223 B.8 Deaths by cause and demographic group, 1990 224 ix Definitions and data notes Selected terms related to health, as used in this calculated as the present value of the future years Report of disability-free life that are lost as the result of the premature deaths or cases of disability occurring in Child mortality. The probability of dying between a particular year. (See Box 1.3 and Appendix B for birth and age 5, expressed per 1,000 live births. further details.) The term under-five mortality is also used . Population-based health services. Services, such as Median age at death. The age below which half of immunization, that are directed toward all mem- all deaths in a year occur. This measure is deter- bers of a specific population subgroup . mined both by the age distribution of the popula- Tertiary care facility. A hospital or other facility tion and by the age pattern of mortality risks . It that offers a specialized, highly technical level of does not represent the average age at which any health care for the population of a large region. group of individuals will die, and it is not directly Characteristics include specialized intensive care related to life expectancy. units, advanced diagnostic support services, and Total fertility rate. The number of children that highly specialized personnel. would be born to a woman if she were to live to the end of her childbearing years and bear children at Country groups each age in accordance with prevailing age-specific fertility rates. For operational and analytical purposes the World Externality. A spillover of benefits or losses from Bank's main criterion for classifying economies is one individual to another. gross national product (GNP) per capita . Every Intervention (in health care). A specific activity economy is classified as low-income, middle-in- meant to reduce disease risks, treat illness, or palli- come (subdivided into lower-middle and upper- ate the consequences of disease and disability. middle), or high-income. Other analytical groups, A/locative efficiency. The extent of optimality in based on regions, exports, and levels of external distribution of resources among a number of com- debt, are also used. peting uses. Because of changes in GNP per capita, the coun- Technical efficiency. The extent to which choice try composition of each income group may change and utilization of input resources produce a spe- from one edition to the next. Once the classifica- cific health output, intervention, or service at low- tion is fixed for any edition, all the historical data est cost. presented are based on the same country group- Cost-effectiveness (in health care). The net gain in ing. The income-based country groupings used in health or reduction in disease burden from a this year's Report are defined as follows. health intervention in relation to the cost. Mea- • Low-income economies are those with a GNP per sured in dollars per disability-adjusted life year capita of $635 or less in 1991. (see next two entries) . • Middle-income economies are those with a GNP Global burden of disease (GBD) . An indicator de- per capita of more than $635 but less than $7,911 in veloped for this Report in collaboration with the 1991. A further division, at GNP per capita of World Health Organization that quantifies the loss · $2,555 in 1991, is made between lower-middle- of healthy life from disease; measured in disabil- income and upper-middle-income economies. ity-adjusted life years. • High-income economies are those with a GNP Disability-adjusted life year (DALY) . A unit used per capita of $7,911 or more in 1991. for measuring both the global burden of disease • World comprises all economies, including and the effectiveness of health interventions, as economies with sparse data and those with less indicated by reductions in the disease burden. It is than 1 million population; these are not shown X separately in the main tables but are presented in The regional grouping of economies in the WDI Table 1a in the technical notes to the World Devel- differs from that used in the main text of this Re- opment Indicators (WDI). port. Part 1 of the table "Classification of econ- omies" at the end of the WDI lists countries by the Demographic regions WDI's income and regional classifications. For purposes of demographic and epidemiological Low-income and middle-income economies are analysis, this year's Report (including its health sometimes referred to as developing economies. data appendices but not the WDI) groups econ- The use of the term is convenient; it is not in- omies into eight demographic regions, defined as tended to imply that all economies in the group are follows: experiencing similar development or that other • Sub-Saharan Africa comprises all countries economies have reached a preferred or final stage south of the Sahara including Madagascar and of development. Classification by income does not South Africa but excluding Mauritius, Reunion, necessarily reflect development status. (In the and Seychelles, which are in the Other Asia and WDI, high-income economies classified as devel- islands group. oping by the United Nations or regarded as devel- • India oping by their authorities are identified by the • China symbol t.) The use of the term "countries" to refer • Other Asia and islands includes the low- and to economies implies no judgment by the Bank middle-income economies of Asia (excluding India about the legal or other status of a territory. and China) and the islands of the Indian and Pa- Analytical groups cific oceans except Madagascar. • Latin America and the Caribbean comprises all For some analytical purposes, other overlapping American and Caribbean economies south of the classifications that are based predominantly on ex- United States, including Cuba. ports or external debt are used, in addition to in- • Middle Eastern crescent consists of the group of come or geographic groups. Listed below are the economies extending across North Africa through economies in these groups that have populations the Middle East to the Asian republics of the for- of more than 1 million. Countries with sparse data mer Soviet Union and including Israel, Malta, and those with less than 1 million population, al- Pakistan, and Turkey. though not shown separately, are included in • Formerly socialist economies of Europe (FSE) in- group aggregates. cludes the European republics of the former Soviet • Fuel exporters are countries for which exports Union and the formerly socialist economies of of petroleum and gas accounted for at least 50 per- Eastern and Central Europe. cent of exports in the period 1987-89. They are • Established market economies (EME) includes all Algeria, Angola, Brunei, Congo, Gabon, Islamic the countries of the Organization for Economic Co- Republic of Iran, Iraq, Libya, Nigeria, Oman, operation and Development (OECD) except Tur- Qatar, Saudi Arabia, Trinidad and Tobago, Turk- key, as well as a number of small high-income menistan, United Arab Emirates, and Venezuela. economies in Europe. • Severely indebted middle-income economies (ab- These eight regions fall into two broad demo- breviated to "Severely indebted" in the WDI) are graphic groups. The first consists of the FSE and twenty-one countries that are deemed to have en- EME, where relatively uniform age distributions countered severe debt-servicing difficulties. These are leading to older populations. The other six re- are defined as countries in which, averaged over gions are referred to as demographically developing, 1989-91, either of two key ratios is above critical in the sense that their age distributions are youn- levels: present value of debt to GNP (80 percent) ger but aging. The demographically developing or present value of debt to exports of goods and all economies correspond approximately to the low- services (200 percent). The twenty-one countries and middle-income economies. Figure 1 of t_he are Albania, Algeria, Angola, Argentina, Bolivia, Overview depicts these regional groups. Table Brazil, Bulgaria, Congo, Cote d'Ivoire, Cuba, Ec- A.10 of Appendix A lists all economies by demo- uador, Iraq, Jamaica, Jordan, Mexico, Mongolia, graphic region and indicates their mid-1990 popu- Morocco, Panama, Peru, Poland, and Syrian Arab lation. Appendix tables A.3 through A.9 provide Republic. demographic and health data by economy within • In the WDI, DECO members, a subgroup of these regions for economies with populations high-income economies, comprises the members greater than 3 million. of the OECD except for Greece, Portugal, and Tur- xi key, which are included among the middle-income Acronyms and initials economies. In the main text of the Report, the term "OECD countries" includes all OECD mem- AIDS Acquired immune deficiency syn- bers unless otherwise stated. drome ARI Acute respiratory infection Data notes BCG Bacillus of Calmette and Guerin vac- cine (to prevent tuberculosis) • Billion is 1,000 million. DALY Disability-adjusted life year • Trillion is 1,000 billion. DPT Diphtheria, pertussis, and tetanus vac- • Tons are metric tons, equal to 1,000 kilograms, cine or 2,204.6 pounds. EPI Expanded Programme on Immuniza- • Dollars are current U.S. dollars unless other- tion (immunization against diphtheria, wise specified. pertussis, tetanus, poliomyelitis, mea- • Growth rates are based on constant price data sles, and tuberculosis) and, unless otherwise noted, have been computed EPI Plus EPI with additional components: im- with the use of the least-squares method. See the munization against hepatitis B and yel- technical notes to the WDI for details of this low fever and, where appropriate, vi- method. tamin A and iodine supplementation • The symbol I in dates, as in "1988/89," means GBD Global burden of disease that the period of time may be less than two years GDP Gross domestic product but straddles two calendar years and refers to a GNP Gross national product crop year, a survey year, or a fiscal year. HIV Human immunodeficiency virus • The symbol .. in tables means not available. HMO Health maintenance organization • The symbol - in tables means not applicable. NGO Nongovernmental organization (In the WDI, a blank is used to mean not OECD Organization for Economic Coopera- applicable.) tion and Development (Australia, Aus- • The number 0 or 0.0 in tables and figures tria, Belgium, Canada, Denmark, Fin- means zero or a quantity less than half the unit land, France, Germany, Greece, shown and not known more precisely. Iceland, Ireland, Italy, Japan, Lux- The cutoff date for all data in the WDI is April embourg, Netherlands, New Zealand, 30, 1993. Norway, Portugal, Spain, Sweden, Historical data in this Report may differ from Switzerland, Turkey, United Kingdom, those in previous editions because of continuous and United States) updating as better data become available, because STD Sexually transmitted disease of a change to a new base year for constant price UNDP United Nations Development Pro- data, or because of changes in country composi- gramme tion of income and analytical groups. UNICEF United Nations Children's Fund Economic and demographic terms are defined in UNPF United Nations Population Fund the technical notes to the WDI. WHO World Health Organization xii Overview Over the past forty years life expectancy has im- ratios are, on average, thirty times as high in de- proved more than during the entire previous span veloping countries as in high-income countries. of human history. In 1950 life expectancy in devel- Although health has improved even in the poor- oping countries was forty years; by 1990 it had est countries, the pace of progress has been un- increased to sixty-three years. In 1950 twenty-eight even. In 1960 in Ghana and Indonesia about one of every 100 children died before their fifth birth- child in five died before reaching age 5-a child day; by 1990 the number had fallen to ten. Small- mortality rate typical of many developing coun- pox, which killed more than 5 million annually in tries. By 1990 Indonesia's rate had dropped to the early 1950s, has been eradicated entirely. Vac- about one-half the 1960 level, but Ghana's had cines have drastically reduced the occurrence of fallen only slightly. Table 1 provides a summary of measles and polio. Not only do these improve- regional progress in mortality reduction between ments translate into direct and significant gains in 1975 and 1990. (Figure 1 illustrates the demo- well-being, but they also reduce the economic bur- graphic regions used in Table 1 and frequently den imposed by unhealthy workers and sick or throughout this Report.) absent schoolchildren. These successes have come In addition to premature mortality, a substantial about in part because of growing incomes and in- portion of the burden of disease consists of disabil- creasing education around the globe and in part ity, ranging from polio-related paralysis to blind- because of governments' efforts to expand health ness to the suffering brought about by severe psy- services, which, moreover, have been enriched by chosis . To measure the burden of disease, this technological progress. Report uses the disability-adjusted life year Despite these remarkable improvements, enor- (DALY), a measure that combines healthy life mous health problems remain. Absolute levels of years lost because of premature mortality with mortality in developing countries remain unac- those lost as a result of disability. ceptably high: child mortality rates are about ten There is huge variation in per person loss of times higher than those in the established market DALYs across regions, mainly because of differ- economies. If death rates among children in poor ences in premature mortality; regional differences countries were reduced to those prevailing in the in loss of DALYs as a result of disability are much rich countries, 11 million fewer children would die smaller (Figure 2) . The total loss of DALYs is re- each year. Almost half of these preventable deaths ferred to as the global burden of disease. are a result of diarrheal and respiratory illne'Ss, The world is facing serious new health chal- exacerbated by malnutrition . In addition, every lenges. By 2000 the growing toll from acquired im- year 7 million adults die of conditions that could be mune deficiency syndrome (AIDS) in developing inexpensively prevented or cured; tuberculosis countries could easily rise to more than 1.8 million alone causes 2 million of these deaths. About deaths annually, erasing decades of hard-won re- 400,000 women die from the direct complications ductions in mortality. The malaria parasite's in- of pregnancy and childbirth. Maternal mortality creased resistance to available drugs could lead to 1 The first six regions named in the key are at intermediate stages of the demographic transition. Figure 1 Demographic regions used in this Report ., ·: · \:· ;·. '" Sub-Saharan Africa Latin America and the Caribbean India Middle Eastern crescent China Formerly socialist economies of Europe Other Asia and islands Established market economies Table 1 Population, economic indicators, and progress in health by demographic region, 1975-90 Income per capita Growth rate, Population, Deaths, 1975-90 Life expectancy at 1990 1990 Dollars, (percent per Child mortality birth (years) Region (millions) (millions) 1990 year) 1975 1990 1975 1990 Sub-Saharan Africa 510 7.9 510 -1.0 212 175 48 52 India 850 9.3 360 2.5 195 127 53 58 China 1,134 8.9 370 7.4 85 43 56 69 Other Asia and islands 683 5.5 1.320 4.6 135 97 56 62 Latin America and the Caribbean 444 3.0 2,190 -0.1 104 60 62 70 Middle Eastern crescent 503 4.4 1,720 -1.3 174 111 52 61 Formerly socialist economies of Europe (FSE) 346 3.8 2,850 0.5 36 22 70 72 Established market economies (EME) 798 7.1 19,900 2.2 21 11 73 76 Demographically developing group• 4,123 39.1 900 3.0 152 106 56 63 FSE and EME 1.144 10.9 14,690 1.7 25 15 72 75 World 5,267 50.0 4,000 1.2 135 96 60 65 Note: Child mortality is the probability of dying between birth and age 5, expressed per 1,000 live births; life expectancy at birth is the average number of years that a person would expect to live at the prevailing age-specific mortality rates . a. The countries of the demographic regions Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent. Source: For income per capita, World Bank data; for other items, Appendix A. 2 a doubling of malaria deaths, to nearly 2 million a role in the developing world is not in doubt. Public year within a decade . Rapid progress in reducing health measures brought about the eradication of child mortality and fertility rates will create new smallpox and have been central to the reduction in demands on health care systems as the aging of deaths caused by vaccine-preventable childhood populations brings to the fore costly noncommuni- diseases. Expanded and improved clinical care has cable diseases of adults and the elderly. Tobacco- saved millions of lives from infectious diseases and related deaths from heart disease and cancers injuries. But there are also major problems with alone are likely to double by the first decade of the health systems that, if not resolved, will hamper next century, to 2 million a year, and, if present progress in reducing the burden of premature smoking patterns continue, they will grow to more mortality and disability and frustrate efforts to re- than 12 million a year in developing countries in spond to new health challenges and emerging dis- the second quarter of the next century. ease threats. • Misallocation. Public money is spent on health Health systems and their problems interventions of low cost-effectiveness, such as surgery for most cancers, at the same time that Although health services are only one factor in ex- critical and highly cost-effective interventions, plaining past successes, the importance of their such as treatment of tuberculosis and sexually The disease burden is highest in poor countries, but disability remains a problem in all regions. Figure 2 Burden of disease attributable to premature mortality and disability, by demographic region, 1990 []] Prem,1turc mortality I2Ll Disability DALYs lost per 1,000 population Sub-Saharan India Middle Otner Asia Latin China Formerly Established Africa Eastern and islands America socialist market crescent and the economies economies Caribbean of Europe Source: Appendix B. 3 transmitted diseases (STDs), remain under- lose out in health because public spending in the funded. In some countries a single teaching hospi- sector is heavily skewed toward high-cost hospital tal can absorb 20 percent or more of the budget of services that disproportionately benefit better-off the ministry of health, even though almost all cost- urban groups. In Indonesia, despite concerted effective interventions are best delivered at lower- government efforts in the 1980s to improve health level facilities. services for the poor, government subsidies to • Inequity. The poor lack access to basic health health for the richest 10 percent of households in services and receive low-quality care. Government 1990 were still almost three times the subsidies spending for health goes disproportionately to the going to the poorest 10 percent of Indonesians. affluent in the form of free or below-cost care in In middle-income countries governments fre- sophisticated public tertiary care hospitals and quently subsidize insurance that protects only the subsidies to private and public insurance. relatively wealthy-a small, affluent minority in the case of private insurance in South Africa and • Inefficiency. Much of the money spent on Zimbabwe and, in Latin America, the larger indus- health is wasted: brand-name pharmaceuticals are trial labor force covered by compulsory public in- purchased instead of generic drugs, health surance (so-called social insurance). The bulk of workers are badly deployed and supervised, and the population, especially the poor, relies heavily hospital beds are underutilized. on out-of-pocket payments and on government • Exploding costs. In some middle-income devel- services that may be largely inaccessible to them. oping countries health care expenditures are grow- In Peru, for example, more than 60 percent of the ing much faster than income. Increasing numbers poor have to travel for more than an hour to obtain of general physicians and specialists, the availabil- primary health care, as compared with less than 3 ity of new medical technologies, and expanding percent of the better-off. The quality of care is also health insurance linked to fee-for-service pay- low: drugs and equipment are in short supply; ments together generate a rapidly growing de- patient waiting times are long and medical consul- mand for costly tests, procedures, and treatments. tations are short; and misdiagnoses and inap- World health spending-and thus also the po- propriate treatment are common. tential for misallocation, waste, and inequitable In the formerly socialist economies, where govern- distribution of resources-is huge. For the world ments have historically been responsible for both as a whole in 1990, public and private expenditure the financing and the delivery of health care, on health services was about $1,700 billion, or 8 health care is free in principle, and wide coverage percent of total world product. High-income coun- of the population has been achieved. This has led tries spent almost 90 percent of this amount, for an to greater apparent equity. But in reality, better-off average of $1,500 per person. The United States consumers make informal out-of-pocket payments alone consumed 41 percent of the global total- to get better care: about 25 percent of health costs more than 12 percent of its gross national product in Romania and 20 percent in Hungary, for exam- (GNP). Developing countries spent about $170 bil- ple, are under-the-table payments for phar- lion, or 4 percent of their GNP, for an average of maceuticals and gratuities to health care providers. $41 per person-less than one-thirtieth the amount Inefficiency is also widespread because the gov- spent by rich countries. ernment-run health system is highly centralized, In the low-income countries government hospitals bureaucratic, and unresponsive to citizens. Gov- and clinics, which account for the greatest part of ernments have been slow to regulate workplace the modern medical care provided, are often ineffi- safety and environmental pollution and have cient, suffering from highly centralized decision- failed to mount effective campaigns against un- making, wide fluctuations in budgetary alloca- healthy personal behaviors-especially alcohol tions, and poor motivation of facility managers consumption and cigarette smoking. In recent and health care workers. Private providers-. years real government spending for health has mainly religious nongovernmental organizations fallen dramatically in the course of the transition to (NGOs) in Africa and private doctors and un- more market-oriented economies. The public sec- licensed practitioners in South Asia-are often tor has suffered from serious shortages of drugs more technically efficient than the public sector and equipment and a lack of skills to manage and offer a service that is perceived to be of higher changing health institutions. The consequences quality, but they are not supported by government have been declining staff morale and falling qual- policies. In low-income countries the poor often ity of care. 4 The roles of the government and of the market ill. A second has to do with "moral hazard": in- in health surance reduces the incentives for individuals to avoid risk and expense by prudent behavior and Three rationales for a major government role in the can create both incentives and opportunities for health sector should guide the reform of health doctors and hospitals to give patients more care systems. than they need. A third has to do with the asym- • Many health-related services such as informa- metry in information between provider and pa- tion and control of contagious disease are public tient concerning the outcomes of intervention; goods. One person's use of health information does providers advise patients on choice of treatment, not leave less available for others to consume; one and when the providers' income is linked to this person cannot benefit from control of malaria- advice, excessive treatment can result. As a conse- carrying mosquitoes while another person in the quence of these last two considerations, in unregu- same area is excluded. Because private markets lated private markets costs escalate without appre- alone provide too little of the public goods crucial ciable health gains to the patient. Governments for health, government involvement is necessary have an important role to play in regulating pri- to increase the supply of these goods. Other health vately provided health insurance, or in mandating services have large externalities: consumption by alternatives such as social insurance, in order to one individual affects others. Immunizing a child ensure widespread coverage and hold down costs. slows transmission of measles and other diseases, If governments do intervene, they must do so conferring a positive externality. Polluters and intelligently, or they risk exacerbating the very drunk drivers create negative health externalities. problems they are trying to solve. When govern- Governments need to encourage behaviors that ments become directly involved in the health sec- carry positive externalities and to discourage those tor-by providing public health programs or fi- with negative externalities. nancing essential clinical services for the poor- • Provision of cost-effective health services to policymakers face difficult decisions concerning the poor is an effective and socially acceptable ap- the allocation of public resources. For any given proach to poverty reduction. Most countries view amount of total spending, taxpayers and, in some access to basic health care as a human right. This countries, donors want to see maximum health perspective is embodied in the goal, "Health for gain for the money spent. An important source of All by the Year 2000," of the conference held by guidance for achieving value for money in health the World Health Organization (WHO) and the spending is a measure of the cost-effectiveness of United Nations Children's Fund (UNICEF) at different health interventions and medical pro- Alma-Ata in 1978, which launched today's pri- cedures-that is, the ratio of costs to health bene- mary health care movement. Private markets will fits (DALYs gained). not give the poor adequate access to essential clini- Until recently, little has been done to apply cost- cal services or the insurance often needed to pay effectiveness analysis to health. This is, in part, for such services. Public finance of essential clini- because it is difficult. Cost and effectiveness data cal care is thus justified to alleviate poverty. Such on health interventions are often weak. Costs vary public funding can take several forms: subsidies to between countries and can rise or fall sharply as a private providers and NGOs that serve the poor; service is expanded. Some groups of interventions vouchers that the poor can take to a provider of are provided jointly, and their costs are shared. their choice; and free or below-cost delivery of Nonetheless, cost-effectiveness analysis is already public services to the poor. demonstrating its usefulness as a tool for choosing • Government action may be needed to com- among possible health interventions in individual pensate for problems generated by uncertainty and countries and for addressing specific health prob- insurance market failure. The great uncertainties sur- lems such as the spread of AIDS. rounding the probability of illness and the effica_cy Just because a particular intervention is cost- of care give rise both to strong demand for insur- effective does not mean that public funds should ance and to shortcomings in the operation of pri- be spent on it. Households can buy health care vate markets. One reason why markets may work with their own money and, when well informed, poorly is that variations in health risk create incen- may do this better than governments can do it for tives for insurance companies to refuse to insure them. But households also seek value for money, the very people who most need health insurance- and governments, by making information about those who are already sick or are likely to become cost-effectiveness available, can often help im- 5 Box 1 Investing in health: key messages of this Report This Report proposes a three-pronged approach to provements and a productive asset-better health - to government policies for improving health. the poor. • Improve management of government health ser- Foster a11 cnvir011111ellf that e11ables llousellolds vices through such measures ,,s decentralization of ad- lo improve Ilea/Ill ministrati\'l' and budgl'lary authority and contracting Household decisions shape health, but these decisions out of services. are constrained by the income and education of house- Prcmrofe diversity alilt competition hold members. In addition to promoting overall L 'Co- nomic growth, governments can help to improvL' those Government finance of public health and of a nation- decisions if they : ally defined package of essential clinical services would • Pursue economic growth policies that will benefit leave the remaining cl inical services to be financed pri- the poor (including, where necessary, adjustment poli- vately or by social insurance within the context of a cies that preserve cost-effective health expenditures) policy framework establishL•d by the government. Gov- • Expand investment in schooling, particularly for ernments Ciln promote diversity and compl'lition in girls provision of health services and insurance by adopting • Promote the rights and status of women through policies that: political and economic empowerment and legal protec- • Encourage social or private insurance (with regula- tion against abuse. tory incentives for equitable access and cost contain- ment) for clinical services outside the ess~ntial Improve govem11renf spe11di11g 011 health p<1ckage. • Encourage suppliers (both public and private) to The challenge for most governments is to concentrate compL•te both to deliver clinical services and to provide resources on compensating for market failures and effi- inputs, such as drugs , to publicly and privatL·Iy fi" ciently financing services that will particularly benefit nanced health services. Domestic suppliers should not the poor. Several directions for policy respond to this be protected from international competition. challenge: • Generate and disseminate information on pro- • Reduce government expenditures on tertiary facil- vider performance, on essential equipment and drugs, ities, specialist training, and interventions that provide on the costs and effectiveness of interventions, and on little health gain for the money spent. the accredit,1tion status of institutions and providers. • Finance and implement a package of public health interventions to deal with the substantial externalities Increased scientific knowledge has accounted for much surrounding infectious disease control, prevention of of the dramatic improvement in health that hils oc- AIDS, environmental pollution , and behaviors (such as curred in this century-by providing information that drunk driving) that put others at risk. forms the basis of household and government action • Finance and ensure delivery of a package of essen- and by underpinning the development of preventive, tial clinical services. The comprehensivenc·ss and com- curative, and diagnostic technologies. lnvestml•nt in position of such a package can only be defined by each continued scientific advance will amplify the effective- country, taking into account epidemiological condi- ness of e<1ch element of the three-pronged <1pproach tions, local preferences, and income. In most countries proposed in this Report. Because thL' fruits of science public finance, or publicly mandated finance, of the benefit all countries, internationaily collabor<1tive ef- essential clinical package would provide a politically forts, of which there are several excellent exampll's, acceptable mechanism for distributing both welfare im- will often be the right way to proceed . prove the decisions of private consumers, pro- portance of continued investment in scientific viders, and insurers. advance. • Since overall economic growth-particularly Government policies for achieving health for all poverty-reducing growth-and education are cen- tral to good health, governments need to pursue This Report focuses primarily on the relation be- sound macroeconomic policies that emphasize re- tween policy choices, both inside and outside the duction of poverty. They also need to expand basic health sector, and health outcomes, especially for schooling, especially for girls, because the way in the poor. Box 1 summarizes the Report's three key which households, particularly mothers, use in- messages for government policy and notes the im- formation and financial resources to shape their 6 dietary, fertility, health care, and other life-style The poor suffer far higher levels of mortality at all choices has a powerful influence on the health of ages than do the rich. household members. • Governments in developing countries should spend far less-on average, about 50 percent less- Figure 3 Infant and adult mortality in poor than they now do on less cost-effective interven- and nonpoor neighborhoods of tions and instead double or triple spending on ba- Porto Alegre, Brazil, 1980 sic public health programs such as immunizations and AIDS prevention and on essential clinical ser- Infant mortality Adult mortality vices . A minimum package of essential clinical ser- (ages 45-64) vices would include sick-child care, family plan- ning, prenatal and delivery care, and treatment for Deaths per 1,000 Deaths per year per 100,000 tuberculosis and STDs. Low-income countries live births persons in age group would have to redirect current public spending for 50 2,000 health and increase expenditures (by government, donors, and patients) to meet needs for public 40 health and the minimum package of essential clini- 1,500 cal services for their populations; less reallocation would be needed in middle-income countries. Ter- 30 tiary care and less cost-effective services will con- 1,000 tinue, but public subsidies to them, if they mainly benefit the wealthy, should be phased out during a 20 transitional period. • Because competition can improve quality and 500 drive down costs, governments should foster com- 10 petition and diversity in the supply of health ser- vices and inputs, particularly drugs, supplies, and 0 0 equipment. This could include, where feasible, Infants Males Females private supply of health care services paid for by governments or social insurance. There is also con- • Poor • Nonpoor siderable scope for improving the quality and effi- ciency of government health services through a Note: Poor neighborhoods were defined according to combination of decentralization, performance- specific criteria. They are, broadly, squntter settlements based incentives for managers and clinicians, and with substandard housing and infrustructure. Source: Barcellos und others 1986. related training and development of management systems . Exposing the public sector to competition with private suppliers can help to spur such im- countries in which average incomes rose by more provements. Strong government regulation is also than 1 percent a year. Economic policies conducive crucial, including regulation of privately delivered to sustained growth are thus among the most im- health services to ensure safety and quality and of portant measures governments can take to im- private insurance to encourage universal access to prove their citizens' health. coverage and to discourage practices-such as fee- Of these economic policies, increasing the in- for-service payment to providers reimbursed by a come of those in poverty is the most efficacious for "third-party" insurer-that lead to overuse of ser- improving health. The reason is that the poor are vices and escalation of costs. most likely to spend additional income in ways that enhance their health: improving their diet, ob- Improving the economic environment taining safe water, and upgrading sanitation and for healthy households housing. And the poor have the greatest remain- Advances in income and education have allowed ing health needs, as Figure 3 illustrates for Porto households almost everywhere to improve their Alegre, Brazil. Government policies that promote health. In the 1980s, even in countries in which equity and growth together will therefore be better average incomes fell, death rates of children under for health than those that promote growth alone. age 5 declined by almost 30 percent. But the child In the 1980s many countries undertook macro- mortality rate fell more than twice as much in economic stabilization and adjustment programs 7 designed to deal with severe economic imbalances tions are currently saving an estimated 3 million and move the countries onto sustainable growth lives a year. Social marketing of condoms to pre- paths. Such adjustment is clearly needed for long- vent transmission of human immunodeficiency vi- run health gains. But during the transitional pe- rus (HIV) has proved highly successful in Uganda, riod, and especially in the earliest adjustment pro- Zaire, and elsewhere. Information on the risks of grams, recession and cuts in public spending smoking, and taxes on both tobacco and alcohol, slowed improvements in health. This effect was are changing behavior in some countries-al- less than originally feared, however-in part be- though mostly, so far, in the richer countries. cause earlier expenditures for improving health Governments need to expand these efforts and and education had enduring effects. As a result of to move forward with other promising public this experience, most countries' adjustment pro- health initiatives. Several activities stand out be- grams today try to rationalize overall government cause they are highly cost-effective: the cost of spending while maintaining cost-effective expen- gaining one DALY can be remarkably low-some- ditures in health and education. Despite these im- times less than $25 and often between $50 and provements, much is still to be learned about more $150. Activities in this category include: efficient ways of carrying out stabilization and ad- • Immunizations justment programs while protecting the poor. • School-based health services Policies to expand schooling are also crucial for • Information and selected services for family promoting health. People who have had more planning and nutrition schooling seek and utilize health information more • Programs to reduce tobacco and alcohol effectively than those with little or no schooling. consumption This means that rapid expansion of educational • Regulatory action, information, and limited opportunities-in part by setting a high minimum public investments to improve the household standard of schooling (say, six full years) for all-is environment a cost-effective way of improving health. Educa- • AIDS prevention. tion of girls and women is particularly beneficial to Intensified government support is required to household health because it is largely women who extend the Expanded Programme on Immuniza- buy and prepare food, maintain a clean home, care tion (EPI), which currently protects about 80 per- for children and the elderly, and initiate contacts cent of the children in the developing world with the health system. Beyond education, gov- against six major diseases at a cost of about $1.4 ernment policies that support the rights and eco- billion a year. Expanding EPI coverage to 95 per- nomic opportunities of women also contribute to cent of all children would have a significant impact overall household well-being and better health. on children in poor households, who make up a disproportionately large share of those not yet Investing in public health reached by the EPI. Other vaccines, particularly and essential clinical services those for hepatitis B and yellow fever, could be added to the six currently included in the EPI, as The health gain per dollar spent varies enormously could vitamin A and iodine supplements. In most across the range of interventions currently fi- developing countries such an "EPI Plus" cluster of nanced by governments. Redirecting resources interventions in the first year of life would have from interventions that have high costs per DALY the highest cost-effectiveness of any health mea- gained to those that cost little could dramatically sure available in the world today. reduce the burden of disease without increasing A second high priority for governments should expenditures. A limited package of public health be to provide inexpensive and highly efficacious measures and essential clinical interventions is a medications to treat school-age children afflicted top priority for government finance; some govern- with schistosomiasis, intestinal worm infections, ments may wish, after covering that minimum for and micronutrient deficiencies. Treatment of these everyone, to define their national essential pack- conditions through distribution of medications age more broadly. and micronutrient supplements in schools would greatly improve the health, school attendance, and Public health learning achievement of hundreds of millions of children, at a cost of $1 to $2 per child per year. In Government action in many areas of public health addition to treatment, schoolchildren can be has already had an important payoff. lmmuniza- taught by their teachers or by radio about the hu- 8 man body and about avoiding risks to health-for gun to spread through human populations, it has example, from smoking or unsafe sex. so far caused 2 million deaths and infected about Governments need to encourage healthier be- 13 million individuals. Some parts of the develop- haviors on the part of individuals and households ing world are already heavily infected: in Sub- by providing information on the benefits of breast- Saharan Africa an average of one in forty adults feeding and on how to improve children's diets. has the virus, and in certain cities the rate is one in Programs in Colombia, Indonesia, and elsewhere three. In Thailand one adult in fifty is infected. show the potential for success. Information on the More than 90 percent of the infected individuals benefits of family planning and on the availability are in their economically most productive years, of family planning services is also critical. Govern- ages 15-40. They will be developing AIDS and ment dissemination of this information can take a dying over the next decade. Projections of the fu- number of creative forms, as the effective use of ture course of the epidemic are gloomy: conserva- radio drama and folk theater in Kenya and Zim- tive estimates from WHO are that by 2000, 26 mil- babwe demonstrates. lion individuals will be HIV-infected and 1.8 Measures to control the use of tobacco, alcohol, million a year will die of AIDS. By destroying indi- and other addictive substances-through informa- viduals' immune systems, HIV will also vastly tion campaigns, taxes, bans on advertising, and, worsen the spread of other diseases, especially tu- in certain cases, import controls-can help sub- berculosis. In highly affected areas demand for stantially to reduce chronic lung disease, heart dis- AIDS treatment will overwhelm capacity for clini- ease, cancer, and injuries. Unless smoking behav- cal treatment and cause a deterioration of care for ior changes, three decades from now premature other illnesses. deaths caused by tobacco in the developing world What governments need to do is clear: intervene will exceed the expected deaths from AIDS, tuber- early, before a major epidemic gets under way. culosis, and complications of childbirth combined. Countries as diverse as Bangladesh, Ghana, and Governments must do more to promote a Indonesia share the preconditions for rapid trans- healthier environment, especially for the poor, mission of HIV-substantial numbers of pros- who face greatly increased health risks from poor titutes and high rates of prevalence of other STDs, sanitation, insufficient and unsafe water supplies, such as syphilis, gonorrhea, and chancroid, which poor personal and food hygiene, inadequate gar- facilitate the spread of the AIDS virus. Strong pub- bage disposal, indoor air pollution, and crowded lic action is required to reduce HIV transmission. and inferior housing. Collectively, these risks are Particularly important are efforts targeted to high- associated with nearly 30 percent of the global bur- risk groups: information to promote change in den of disease. To help the poor improve their sexual behavior; distribution of condoms; and household environments, governments can pro- treatment for other STDs. Early reduction in HIV vide a regulatory and administrative framework transmission by high-risk individuals is very cost- within which efficient and accountable providers effective, but later in an AIDS epidemic the cost- (often in the private sector) have an incentive to effectiveness of interventions declines substan- offer households the services they want and are tially. Current expenditures on AIDS prevention willing to pay for, including water supply, sanita- in developing countries-totaling less than $200 tion, garbage collection, clean-burning stoves, and million a year-are woefully inadequate. Five to housing. The government has a vital role in dis- ten times this level of spending is needed to deal seminating information about hygienic practices. with the emerging epidemic. It can also improve the use of public resources by eliminating widespread subsidies for water and Essential clinical services sanitation that benefit the middle class. Govern- ment legislation and regulations to increase secu- The components of a package of essential clinical rity of land tenure for the poor would encourage services of high cost-effectiveness will vary from low-income families to invest more in safer, country to country, depending on local health healthier housing. needs and the level of income. At a minimum, the A special challenge for concerted public health package should include five groups of interven- action is to reduce the spread of AIDS. The AIDS tions each of which addresses very large disease epidemic has already become a dominant public burdens. The five groups are: health concern in many countries. Although HIV, • Services to ensure pregnancy-related (prena- the virus that causes AIDS, has only recently be- tal, childbirth, and postpartum) care; strength- 9 ened efforts could prevent most of the almost half- broader range of interventions than this mtm- million maternal deaths that occur each year in mum. At modest increases in spending, relatively developing countries. cost-effective measures for the treatment of some • Family planning services; improved access to common noncommunicable conditions could be these services could save as many as 850,000 chil- included. Examples are low-cost protocols for dren from dying every year and eliminate as many treatment of heart disease using aspirin and anti- as 100,000 of the maternal deaths that occur hypertensive drugs; treatment for cervical cancer; annually. drug treatment of some psychoses; and removal of • Tuberculosis control, mainly through drug cataracts. therapy, to combat a disease that kills more than 2 Many health services have such low cost-effec- million people annually, making it the leading tiveness that governments will need to consider cause of death among adults. excluding them from the essential clinical package. • Control of STDs, which account for more than In low-income countries these might include heart 250 million new cases of debilitating and some- surgery; treatment (other than pain relief) of times fatal illness each year. highly fatal cancers of the lung, liver, and stom- • Care for the common serious illnesses of ach; expensive drug therapies for HIV infection; young children-diarrheal disease, acute respira- and intensive care for severely premature babies. tory infection, measles, malaria, and acute malnu- It is hard to justify using government funds for trition-which account for nearly 7 million child these medical treatments at the same time that deaths annually. much more cost-effective services which benefit These clinical interventions are all highly cost- mainly the poor are not adequately financed. effective-often costing substantially less than $50 Widespread adoption of an essential clinical per DALY gained. package would have a tremendous positive impact A minimal package of essential clinical services on the health of people in developing countries. If would also include some treatment for minor in- 80 percent of the population were reached, 24 per- fection and trauma and, for health problems that cent of the current burden of disease in low- cannot be fully resolved with existing resources, income countries and 11 percent of that in middle- advice and alleviation of pain. The provision of income countries could be averted (Table 2). The hospital-based emergency care other than the in- estimated impact of implementing the minimum terventions mentioned above would depend on clinical services is more than twice that for the day-to-day capacity and availability of resources. public health package outlined above; when com- This emergency care includes, for example, treat- bined with the public health package, the share of ment of most fractures, as well as appendec- current illness that could be eliminated rises to tomies. Depending on resource availability and so- perhaps 32 percent for low-income countries and cial values, some countries may define their 15 percent for middle-income countries. This re- essential clinical package to include a much duction in disease is equivalent, in terms of DALYs Table 2 Estimated costs and health benefits of the minimum package of public health and essential clinical services in low- and middle-income countries, 1990 Cost Cost Approximate (dollars per capita as a percentage reduction in burden Group per year) of income per capita of disease (percent) Low-income countries (Income per capita = $350) Public health 4.2 1.2 8 Essential clinical services a 7.8 2.2 24 Total 12.0 3.4 32 Middle-income countries (Income per capita = $2,500) Public health 6.8 0.3 4 Essential clinical services a 14.7 0.6 11 Total 21.5 0.9 15 a. The estimated costs and benefits are for a minimum essential package of clinical services, as defined in the text. Many countries may wish, if they have the resources, to define their essential clinical package more broadly. Source: World Bank calculations. 10 gained, to saving the lives of more than 9 million groups can lead to erosion of political support for infants each year. the essential package and to decreased funding and lower quality of care. Furthermore, problems Paying for the package of cost escalation and access to insurance on the part of high-risk groups can complicate private fi- The most sophisticated facility required to deliver nance. For these reasons, in most member coun- the minimum elements of the essential clinical tries of the Organization for Economic Coopera- package is a district hospital. Providing services in tion and Development (OECD), governments lower-level facilities allows costs to be contained at finance (or mandate the financing of) comprehen- modest levels for minimal versions of the essential sively defined essential packages for virtually all clinical package. The cost is about $8 per person their citizens. each year in low-income countries and $15 in In low-income countries, where current public middle-income countries. The cost differences are spending for health is less than the cost of an es- the result of distinct demographic structures, epi- sential package, some degree of targeting is inevi- demiological conditions, and labor costs in the two table. If the wealthy are already opting out of gov- settings. When the cost of the public health inter- ernment-financed services because of the higher ventions described above is added, total costs rise quality and convenience of privately financed ser- to $12 per capita in low-income countries and $22 vices, targeting is fairly easy. Community-financ- per capita in middle-income countries. ing schemes, whereby patients at local health cen- Adoption of the package in all developing coun- ters and pharmacies pay modest fees, are another tries would require a quadrupling of expenditures option that can help both to improve the quality of on public health, from $5 billion at present to $20 care and, when fees are retained and managed lo- billion a year, and an increase from about $20 bil- cally, to sustain services. A large number of coun- lion to $40 billion in spending on essential clinical tries in Africa have had some early success with services. In the poorest countries governments community financing as part of the Bamako Initia- typically spend about $6 per person for health and tive led by UNICEF and WHO. Nonetheless, expe- total health expenditures are about $14 per person. rience to date suggests that introduction of user There, paying for an essential package will require fees at levels that do not discourage the poor is a combination of increased expenditures by gov- likely to be more useful for improving technical ernments, donor agencies, and patients and some efficiency (for example, by facilitating drug sup- reorientation of current public spending for ply) than for raising substantial revenues on ana- health. In middle-income countries, where public tionwide basis. spending for health averages $62 per person, the $22 cost of the package is financially feasible if the Reforming health systems: promoting diversity political commitment exists for shifting existing re- and competition sources away from discretionary services with lower cost-effectiveness toward public health pro- Ensuring basic public health services and essential grams and essential clinical care. These major clinical care while the rest of the health system changes cannot be made overnight, but it is impor- becomes self-financed will require substantial tant to start and complete them as swiftly as possi- health system reforms and reallocations of public ble, before interest groups and bureaucratic inertia spending. Only by reducing or eliminating spend- undermine reform. ing on discretionary clinical services can govern- A critical question in designing an essential clini- ments concentrate on ensuring cost-effective clini- cal package is the extent of government financing. cal care for the poor. One way to do so is by Should governments pay for everyone, or only for charging fees to affluent patients who use govern- the poor? The main problem with universal gov- ment hospitals and services. In Chile, Kenya, ernment financing is that it subsidizes the wealthy, Lesotho, and other countries governments are who could afford to pay for their own services, increasing user fees for the wealthy and for those and thus leaves fewer government resources for covered by insurance and are strengthening the the poor. A policy requiring those who can pay all legal and administrative systems for billing pa- or part of their own costs to do so may make sense tients and collecting revenues. on equity grounds, but it also has disadvantages. Promoting self-financed insurance, thus elim- Often, the administrative costs of targeting are inating large and inequitable subsidies to the more high, and exclusion of wealthy and middle-income affluent groups who are covered by insurance, 11 would also help to free government funds for pub- keep patients longer than necessary and are lic health programs and essential clinical care. Sub- poorly organized and managed. Countries pay too sidies in the form of tax relief for contributions to much for drugs of low efficacy, and drugs and sup- private insurance are equal to nearly a fifth of total plies are stolen or go to waste in government ware- government spending for health in South Africa. houses and hospitals. In Latin America subsidies to the social insurance In the short term, reforms in pharmaceutical us- systems are widespread and include tax relief, di- age offer the greatest gains in efficiency. Govern- rect transfers to cover the operating deficits of so- ments that have introduced competition in the cial security health funds, and matching govern- procurement of drugs have typically achieved sav- ment funds for employee payroll contributions. ings of 40 to 60 percent. Governments can also Where these subsidies benefit only the better-off in develop national essential drug lists, consisting of society, they need to be scaled back. a limited number of inexpensive drugs that ad- Reforms entail shifting new government spend- dress the important health problems of the popu- ing for health away from specialized personnel, lation. Many countries have such lists, but not all equipment, and facilities at the apex of health sys- use them to guide the selection and procurement tems and "down the pyramid" toward the broad of drugs for the public sector. New treatment pro- base of widely accessible care in community facili- tocols and alternative uses of facilities can also ties and health centers. Very few cost-effective in- raise efficiency. Outpatient surgery can replace terventions depend on sophisticated hospitals and some procedures customarily performed on an in- specialized physicians-all the services contained patient basis, at considerable savings. in the minimum essential clinical package pro- In the long run, decentralization can help to in- posed in this Report can be provided by health crease efficiency when there is adequate capacity centers and district hospitals. Yet specialized facili- and accountability at lower levels of the national ties everywhere absorb a large amount of public health system. Some countries, such as Botswana resources, a problem that has frequently been ex- and Ghana, have delegated a wide range of man- acerbated by donor investments in tertiary care fa- agement responsibilities to regional and district- cilities. In the 1980s Papua New Guinea, to correct level offices of the ministry of health; others, overconcentration of resources on higher-level fa- including Chile and Poland, have devolved au- cilities, limited public spending on hospitals to 40 thority and resources to local government agen- percent of the recurrent budget of the Ministry of cies. Their experience provides evidence that Health-well below the level in most developing success is possible-but also that hasty and countries. unplanned decentralization, sometimes purely in Governments need to use more effective policies response to political pressures, can create new for financing training (including use of national problems. service mechanisms) to help meet the need for pri- Greater reliance on the private sector to deliver mary care providers, particularly nurses and mid- clinical services, both those that are included by a wives, and for public health, health policy, and country in its essential package and those that are management personnel. At the same time, gov- discretionary, can help raise efficiency. The private ernments should limit or eliminate subsidies for sector already serves a large and diverse clientele specialist training. Increased government support in developing countries and often delivers services for health information systems and operations re- of higher quality without the long lines and inade- search would help to guide public policies for quate supplies frequently found in government fa- health. Estimates of the national burden of disease cilities. In many countries private doctors and along the lines of the global burden of disease pharmacies face unnecessary legal and administra- methodology used in this Report, and local infor- tive barriers, and these need to be removed. But mation on the cost-effectiveness of different inter- the tendency for profit-making providers to over- ventions, would enable governments to establish prescribe drugs, procedures, and diagnostics health priorities. needs to be countered; encouraging the for-profit In every developing country decisive steps are sector to move away from fee-for-service to pre- needed to correct the pervasive inefficiency of clin- paid coverage (through, for example, encouraging ical health programs and facilities and especially of health maintenance organizations) is one feasible government services. Clinics and outreach pro- approach. grams operate poorly because of shortages of Governments could also subsidize private drugs, transport, and maintenance. Hospitals health care providers who deliver essential clinical 12 services to the poor. This is already beginning to private health maintenance organizations and in happen and needs to go further. In many African the British National Health Service. Another is for countries, including Malawi, Uganda, and Zam- insurers jointly to negotiate uniform fees with doc- bia, governments subsidize the operating expendi- tors and hospitals, as is done in Japan's social in- tures of church hospitals and clinics in rural areas surance system and Zimbabwe's private medical and the training of their health personnel. In Ban- aid insurance system; or insurers themselves can gladesh, Kenya, Thailand, and other countries . set fixed payments for specified medical diag- governments, with assistance from donors, are noses, as in Brazil. Yet a third approach, which has supporting the work of traditional birth attendants been tested on a limited scale in the United States, in safe pregnancy and delivery care and of tradi- is "managed competition." This scheme pursues tional healers in controlling infectious diseases the three objectives of cost-effective health spend- such as malaria, diarrhea, and AIDS. ing, universal insurance coverage, and cost con- Regulation is an essential element of govern- tainment simultaneously through tightly regulated ment efforts to encourage private health care sup- competition among companies that provide a spe- pliers. In most countries, governments have an cified package of health care for a fixed annual fee. important role to play in ensuring the quality of Each of these approaches has proved workable, private sector health care-through accreditation but each also has its limits and disadvantages. of hospitals and laboratories, licensing of medical There are no simple answers for health schools and physicians, regulation of drugs, and policy makers. reviews of medical practices. Some countries in which the government's ability to regulate is par- An agenda for action ticularly weak could explore self-regulation for health care providers, while building up govern- Adoption of the main policy recommendations of ment capacity. In Brazil experiments with self- this Report by developing country governments regulation for local hospital associations and medi- would enormously improve the health status of cal ethics boards are now under way. their people, especially poor households, and Government regulation of insurance is equally would also help to control health care spending important. In some countries part of the popula- (Table 3). Millions of lives and billions of dollars tion is denied insurance because of selection bias could be saved. Implementation of the public under private voluntary insurance. In the United health and essential clinical care packages, pursuit States millions of people with high health risks- of economic growth strategies that reduce poverty, and thus high need for health insurance-are un- and increased imrestment in schooling for girls able to obtain affordable coverage. Some types of would have the largest payoffs in averting deaths insurance schemes also seem to contribute to and reducing disability. Scaling back public spend- pushing up health care costs; this is particularly ing for tertiary care facilities, specialist training, true of third-party systems and of systems that and clinical care with lower cost-effectiveness reimburse hospitals and physicians item by item would help to increase the effectiveness of health for any and all services performed. In both the spending. So would encouragement of competi- Republic of Korea, which relies on universal social tion in delivery of health services and regulation of insurance, and the United States, which uses insurance and of provider payment systems. mostly private insurance, health care already ab- These recommendations will facilitate progress sorbs an unusually high share of GNP-and costs toward the goal contained in the declaration from are still rising. During the 1980s, for example, the historic 1978 Alma-Ata conference: "The at- health expenditures in Korea increased from 3.7 to tainment of all peoples of the world by the year almost 7 percent of GNP, in large part because of 2000 of a level of health that will permit them to expansion of third-party insurance coverage com- lead a socially and economically productive life." bined with fee-for-service provider compensation. Continued momentum toward this goal was pro- To eliminate selection bias and expand insur~ vided by the 1990 World Summit for Children. Al- ance coverage, governments can require insurers most 150 countries have now signed commitments to pool risks across large numbers of people. To to specific goals for their countries to improve the control costs, governments have a number of op- health of children and women (Box 2). These goals tions for limiting payments to health providers. include reduction of child mortality rates by one- One approach is to encourage prepayment of a third (or to 70 per 1,000 births, whichever would fixed amount for each person, as is now done in be less) over the course of the decade of the 1990s, 13 Table 3 Contribution of policy change to objectives for the health sector Contribution to goals Government objectives and policies Foster au enabling environme11t for llousellolds to improve healt/1 Pursue economic growth policies that benefit the poor Exp11nd investment in educ11tion, particularly for females Promote the rights 11nd status of women through political and economic empowerment and legal protection against abuse Improve govemment investments i11 l1ealt11 Reduce government expenditures for tertiary care facilities, specialist training, and discretionary services Finance and ensure delivery of a public health pack11ge, including AIDS prevention Finance and ensure delivery of essential clinical services, at least to the poor Improve the man<1gement of public health services Facilitate involveme11t by tl1 e private sector Encourage private finance and provision of insurance (with incentives to contain costs) for all discretionary clinical services Encourage private sector delivery of clinical services (including those that arc publicly financed) Provide information on performance and cost • Very favorable II F11vorable D Somewhat favorable D No impact expected reduction of maternal mortality rates by half, erad- nancing of essential clinical services should be at ication of polio, and major reductions in morbidity the top of the policy agenda . In most middle- and mortality from several other diseases . Com- income countries these policies are still germane, mitments to specific improvements in education, but reducing public subsidies for insurance and nutrition, water supply, and sanitation were also discretionary care would also yield large benefits made. These commitments underscore the politi- and should therefore be a key element of policy cal potential of action agendas for improving change . In the formerly socialist economies there health . are two particularly crucial policy areas-improv- The relevance of the main recommendations of ing the management of government health ser- this Report varies from one setting to another. In vices and developing sustainable health-financing low-income countries renewed emphasis on basic systems that maintain universal coverage while schooling for girls, strengthening of public health encouraging competition among cost-conscious programs, and support for expanded public fi- suppliers. 14 Box 2 The World Summit for Children The declaration and plan of action adoptt>d at the gets and external aid if priorities .for human devdop- World Summit for C hildrL' n, held in New York in 1990, ment are to be met. The health goals of the summit's incorporate a politically salient ,,genda for health. The plan of action include: summit focused, in p.uticular, on the needs of children • The L'radication of polio by 2000 and women but w.1s sl't in the bro.1der context of hu- • The elimination of neonatal tetanus by 1995 man and community goals. The seventy-one heads of • A 90 percent reduction in measles cases and a 95 state who attended and the seventy-seven more who percent reduction in measles deaths subsequently signed the declaration committed their • Achievement (by 2000) and maintenance of at least countries to developing national progr.1ms of action 90 percent immunization cover,1ge of one-year-old chil- (NPAs) for achieving these goals. To date, ,,bout dren, as well as universal tetanus immuniza tion for eighty-five countries h.we drawn up NPAs, and ,,n- women of childbearing ,,ge other sixty are in thL' process of prep.1ring them . • A halving of child deaths caused by diarrhea and. a NPAs typically cover, ,,mung other concerns, pri- one-quarter reduction in the incidence of diarrheal mary health care, family planning, safe water, environ- disease mental sanitation, nutrition, and basic education. Be- • A reduction by one-third in child deaths caused by cause oftheir concentration on the welfare of children, acute respiratory infections NPAs are able to transcend political differences. They • Virtual elimination of vitamin A deficiency and io- offer a means of mobiliz ing the whoiL' of civil society- dine deficiency disorders neighborhood and civic associations, religious groups • A reduction in the incidence of low · birth weight and professional bodies, businesses, voluntary agen- (2.5 kilograms or less) to no more than 10 percent cies, organized labor, and universities-in the cause of • A one-third reduction fn1m 1990 levels in iron d efi- investment for health. ciency anL'mia among women NPAs are being integrated into national develop- • Access for all women to prenatal care, trained at- ment planning . They set forth measurable, attainable tendants during childbirth, and referral for high-risk goals- to be met by 2000 or earlier- that are adapted to pregnancies and obstetric emergencies . the realities of the country. By quantifying the re- The agenda for action of the children's health summit sources required to achieve these goals, NPAs help to is broadly consistent with the messages of this Report. identify the changes that are needed in national bud- At first glance, it might appear that adoption of central hospitals and has concentrated on improv- this Report's major recommendations will be easy. ing health centers and other district-level infra- To reach most people living in the developing structure. Tunisia has converted eleven large gov- world with the minimum package of cost-effective ernment hospitals to semiautonomous institutions public health and essential clinical services, about with strong incentives for improved performance. half of current government expenditures on other, During the 1980s Chile delegated responsibility for more discretionary care would have to be redi- its entire primary clinical care system to local gov- rected. But in reality, change will be difficult, since ernments and fostered more public and private an array of interest groups may stand to lose- competition in health service delivery and in in- from suppliers of medical services to rich benefici- surance. Costa Rica and Korea achieved universal aries of public subsidies to protected drug com- health coverage through social insurance. panies. Many of the changes will take years to im- The international community can do more to plement because they mean a major redirection of support health policy reforms. In 1990 donors dis- public resources and require the development of bursed about $4.8 billion of assistance for health, new institutional capabilities. or about 2.5 percent of all health spending in de- A number of developing countries have already veloping countries. The share of total develop- shown in recent years that broad reforms in the ment aid for health declined slightly in the 1980s, health sector are possible when there is sufficient from 7 to 6 percent, despite widespread calls for political will and when changes to the health sys- increased investment in human resource develop- tem are designed and implemented by capable ment, including health . As an immediate first planners and managers . Zimbabwe has imposed a step, donors need to restore this share to its former decade-long moratorium on new investments in level. A more substantial increase can be easily 15 justified, given the importance of health in reduc- essential clinical care-especially for tuberculosis ing poverty and the large gap between current and control, the EPI Plus program, AIDS prevention, needed spending for public health programs and and reduction of tobacco consumption-would be minimum clinical services. An additional $2 billion a significant contribution to policy reform. So a year from donors would meet about one-quarter would support for capacity-building. Countries of the costs of stabilizing the AIDS epidemic ($500 that are willing to undertake major changes in million) and one-sixth of the extra resources health policy should be strong candidates for in- needed to provide the public health and clinical creased aid, including donor financing of recurrent care package for low-income countries ($1.5 billion costs. An increasing number of donors, among of the $10 billion required). them the World Bank, are now supporting this Increased external assistance for health research kind of broad sectoral reform. Stronger donor co- that focuses on the major health problems of de- ordination, especially at the level of individual de- veloping countries-such as the search for new an- veloping country clients, would improve the posi- timalarial drugs and new or improved vaccines- tive impact of aid on health, as shown by the could have a very high payoff and would build on experience of Bangladesh, Senegal, and the comparative advantage of donor countries in Zimbabwe. conducting scientific research. That most health The benefits to the developing world from research benefits many countries further justifies adopting sound policies for health are enormous. donor support, particularly through such effective There is great potential for change during the internationally collaborative mechanisms as the closing years of this decade as more countries en- Special Programme for Research and Training in courage broad political participation and public ac- Tropical Diseases. countability, as levels of education and knowledge Donors and developing country governments improve, and as understanding of human biology, can also do much to improve the effectiveness of public health, and health care systems increases. If aid for health. This is especially important in low- the right policy choices are made, the payoff will income Africa, where aid already accounts for an be high. The momentum of past reductions in the average 20 percent of health spending-and for burden of infectious disease in developing coun- over half in Burundi, Chad, Guinea-Bissau, tries can be maintained and accelerated. The AIDS Mozambique, and Tanzania. Even in other devel- epidemic can be slowed or reversed. The emerging oping regions, where aid amounts to 2 percent or problems of noncommunicable disease in aging less of health expenditures, better targeting and populations can be managed without rapid in- management of this assistance can catalyze policy creases in health expenditures. In the end, this will change. translate into longer, healthier, and more produc- Redirecting donor money from hospitals and tive lives for people around the world, especially specialist training to public health programs and the more than 1 billion now living in poverty. 16 Health in developing countries: successes and challenges On October 22, 1977, Ali Maow Maalin, a twenty- Few investments of any kind generate human three-year-old cook living in the town of Merca, and financial benefits on that scale . Yet in many Somalia, developed a fever and rash that was sub- ways the Intensified Smallpox Eradication Pro- sequently diagnosed as smallpox. Vaccination gramme exemplifies the potential of today' s medi- teams immediately descended on Merca and cine. Around the world, the past half century has within three weeks had vaccinated more than seen startling improvements in health . Progress in 50,000 people. They also began an intensive search drugs, vaccines, epidemiological knowledge, and for other cases in Merca and along the road and organizational experience continually expands the footpaths leading to it. By December 29 the World range of options for tomorrow. Tools and methods Health Organization (WHO) had removed Merca for combating and eliminating much of the re- from its list of potential outbreaks of smallpox and maining burden of disease are now affordable, had initiated a two-year surveillance for the dis- even by the poorest countries. Good policy, how- ease throughout the Horn of Africa. It turned out, ever, is essential for achieving good health . Some however, that Mr. Maalin had experienced the countries have made full use of the potential of world's last case of smallpox. He survived, and medicine; others have barely tapped it, despite WHO's twelve-year-long Intensified Smallpox heavy spending. This Report draws from this var- Eradication Programme was brought to a trium- ied experience lessons that will assist policymakers phant end . in realizing the enormous potential returns from In 1967, the year when the program began, their countries' investments in health . somewhere between 1.5 million and 2 million peo- ple died from smallpox. Perhaps half a million Why health matters more were blinded, and more than 10 million were seriously and permanently disfigured . In the early Good health, as people know from their own ex- 1950s the toll from smallpox had been three or four perience, is a crucial part of well-being, but spend- times greater. Then more and more countries un- ing on health can also be justified on purely eco- dertook vaccination programs, and by the time the nomic grounds. Improved health contributes to global program began, the disease had been vir- economic growth in four ways: it reduces produc- tually eradicated in 125 countries. Even so, the cos_t tion losses caused by worker illness; it permits the of smallpox vaccination, quarantine programs, use of natural resources that had been totally or and treatment totaled more than $300 million in nearly inaccessible because of disease; it increases 1968 alone. The eradication program, by contrast, the enrollment of children in school and makes cost $300 million over the whole of its twelve-year them better able to learn; and it frees for alterna- life and has therefore saved hundreds of millions tive uses resources that would otherwise have to of dollars a year in direct, measurable costs, as well be spent on treating illness . The economic gains as unquantifiable amounts of human suffering. are relatively greater for poor people, who are typ- 17 ically most handicapped by ill health and who ability in their income. In Paraguay, for example, stand to gain the most from the development of farmers in malarious areas choose to grow crops underutilized natural resources. that are of lower value but that can be worked outside the malaria season. Gains in worker productivity Improved utilization of natural resources The most obvious sources of gain are fewer work days lost to illness, increased productivity, greater Some health investments raise the productivity of opportunities to obtain better-paying jobs, and land. In Sri Lanka the near-eradication of malaria longer working lives. To take a classic example, during 1947-77 is estimated to have raised national leprosy is a disease that affects people in the prime income by 9 percent in 1977. The cumulative cost of life, with peak incidence rates among young was $52 million, compared with a cumulative gain adults. As many as 30 percent of those affected in national income over the thirty-one years of $7.6 may be seriously deformed, and their working billion, implying a spectacular benefit-cost ratio of lives will be shortened as well. A study of lepers in more than 140. Areas previously blighted by mos- urban Tamil Nadu, India, estimates that the elim- quitoes became attractive for settlement; migrants ination of deformity would more than triple the moved in, and output increased. In Uganda mas- expected annual earnings of those with jobs. The sive migration to fertile but underexploited land prevention of deformity in all of India's 645,000 followed the partial control of river blindness (on- lepers would have added an estimated $130 mil- chocerciasis) in the 1950s. The Onchocerciasis lion to the country's 1985 GNP. This amount is the Control Programme, conducted in eleven coun- equivalent of almost 10 percent of all the official tries of the Sahel, is a more recent example of the development assistance received by India in 1985. same benefits (see Box 1.1). Yet leprosy accounted for only a small proportion of the country's disease burden, less than 1 per- Benefits in the next generation through education cent in 1990. Healthier workers earn more because (as re- There is no question that schooling pays off in search in Bangladesh has demonstrated) they are higher incomes. Four years of primary education more productive and can get better-paying jobs. In boosts farmers' annual productivity by 9 percent Cote d'Ivoire daily wage rates are estimated to be on average, and workers who do better at school 19 percent lower, on average, among men who are earn more. Studies in Ghana, Kenya, Pakistan, likely to lose a day of work per month because of and Tanzania indicate that workers who scored 10 illness than among healthier men. percent above the sample mean on various cogni- When illness strikes, an individual's lost output tive tests have a wage advantage ranging from 13 and earnings often go undetected in economic sta- to 22 percent; in Nepal farmers with better mathe- tistics because they are borne by the household. In matical skills are more likely to adopt profitable many developing countries unemployment (or new crops. disability) insurance is rare, and healthier mem- Poor health and nutrition reduce the gains of bers of the household work harder or longer to schooling in three areas: enrollment, ability to make up for the loss in income. In a sample of 250 learn, and participation by girls. Children who en- Sudanese households, each of which lost, on aver- joy better health and nutrition during early child- age, forty working hours per year because of ma- hood are more ready for school and more likely to laria alone, this extra work made up for 68 percent enroll. A study in Nepal has found that the proba- of the lost agricultural labor. Similar findings have bility of attending school is only 5 percent for nu- come from research in Paraguay and Colombia. tritionally stunted children, compared with 27 per- In the long run, the benefits of improved health cent for those at the norm. are also likely to influence the way work is orga- Health and nutrition problems affect a child's nized and carried out. With a healthy work force, ability to learn. Nutritional deficiencies in early employers can reduce the costs of building slack childhood can lead to lasting problems: iron defi- into their production schedules, invest more in ciency anemia reduces cognitive function, iodine staff training, and exploit the benefits of specializa- deficiency causes irreversible mental retardation, tion. Similar gains are likely among farmers, who and vitamin A deficiency is the primary cause of often hedge against sickness by being risk-averse; blindness among children. Older children are sub- they forgo higher output in return for less vari- ject to other kinds of disease. In a recent study in 18 Box 1.1 Controlling river blindness Onchocerciasis, or river blindness as it is more com- a team of entomologists, epidemiologists, field staff, monly known, is caused by a parasitic worm which and pilots; 97 percent of the staff are nationals of the produces millions of larvae that move through the participating countries. The World Bank organizes the body, causing intense itching, debilitation, and eventu- finances and manages them through a trust fund. It ally blindness. The disease is spread by a small, fiercely also supports socioeconomic development in the areas biting blackfly that transmits the larvae from infected to affected by the disease . uninfected people. The program is widely regarded as a great success. It The goals of the Onchocerciasis Control Programme protects from river blindness about 30 million people, (OCP), set up in 1974 and covering eleven Sahelian including more than 9 million children born since the countries, are to control the blackfly by destroying its ocr began, at an annual cost of less than $1 per per- larvae with insecticides sprayed from the air. The envi- son. More than 1.5 million people who were once seri- ronmental impact of the insecticides is continuously ously infected have completely recovered. It is esti- monitored by an independent ecological committee, in mated that the program will have prevented at least cooperation with the national governments. The com- 500,000 cases of blindness by the time it is wound up mittee has full authority to screen insecticides and to around the end of the century. And it is already freeing approve or reject their use. The program has also col- approximately 25 million hectares of previously laborated with the pharmaceutical industry to develop blighted land for resettlement and cultivation, boosting for human use a drug, ivermectin,.that safely and effec- agricultural production. tively kills the larvae in the body. Ivermectin, however, The estimated cost of the OCP during the whole of has little impact on the adult worm and so must be its existence, from 1974 to 2000, is about $570 million. supplemented with vector .control by aerial spraying. Its estimated internal rate of return is in the range of 16 The producer of ivermectin, Merck & Co., has commit- to 28 percent (depending on the pace at which the ted itself to provide the drug free of charge as long as it newly available land is settled, the incremental output is needed to combat river blindness. added by the new land, the income level of the ocr The OCP's four sponsoring agencies-the Food and area, and the productivity growth rate that is pro- Agriculture Organization, the United Nations Devel- jected). These estimated benefits do not include the opment Programme (UNDP), the World Bank, and program's favorable effects on income distribution; its WHO-through a steering committee chaired by th e main beneficiaries are subsistence farmers whose in- World Bank, make broad policy decisions and oversee comes are well below average. operations. WHO has executive responsibility through Jamaica children with moderate whipworm infec- Girls are particularly liable to suffer from iodine tion scored 15 percent lower before treatment than or iron deficiency-reasons why fewer of them uninfected children in the same school. When re- complete primary school. Other health-related tested after treatment, those same children did al- reasons include dropping out as a result of preg- most as well as the uninfected children. nancy and parental concern about sexual violence. In a sample of children in a poverty-stricken In societies where girls' education is given lower area of northeast Brazil, inadequately nourished priority than boys', girls miss school because they children lagged 20 percent behind the average gain have to stay home to look after sick relatives. in achievement score over a two-year period. The same study also shows the harm done by a simple Reduced costs of medical care and easily remedied handicap: children with bad eyesight lagged 27 percent behind the average Spending that reduces the incidence of disease can gain over the two years. Both groups had below- produce big savings in treatment costs. For some average promotion rates and above-average drop- diseases the expenditure pays for itself even when out rates. In China a child at the twentieth percen-- all the indirect benefits-such as higher labor pro- tile in height-for-age (a sign of poor health) aver- ductivity and reduced pain and suffering-are ig- ages about one-third of a year behind the grade nored. Polio is one example. Calculations for the normally reached by children of that age. In Thai- Americas made prior to the eradication of polio in land children whose height-for-age is 10 percent the region showed that investing $220 million over below average are 14 percent lower in grade fifteen years to eliminate the disease would pre- attainment. vent 220,000 cases and save between $320 million 19 Box 1.2 The economic impact of AIDS The AIDS epidemic, through its effects on savings and That AIDS kills so many skilled adults adds to its productivity, poses a threat to economic growth in economic impact. At a large hospital in Kinshasa, for many countries that are already in distress . World Bank example, more than 1 percent per year of the health simulations indicate a slowing of growth of income per personnel, including highly trained staff, become in- capita by an average 0.6 percentage point a year in the fected (through sexual rather than occupational con- ten worst-affected countries in Sub-Saharan Africa . In tact) . Among the (largely male) employees at a Kin- Tanzania, where income per capita has already fallen shasa textile mill, managers had a higher infection rate 0.2 percent a year in recent years, the estimated slow- than foremen, who in turn had a higher rate than down ranges between 0.1 and 0.8 percentage point, workers. The cost of replacing skilled workers will be depending on the assumptions used. In Malawi, which substantial. A study of Thailand estimates that through has had a recent growth rate of 0.9 percent a year, the 2000 the cost of replacing long-haul truckers lost to simulated reduction ranges from 0.3 to 0.5 percentage AIDS will be $8 million, and another study, of Tan- point. These calculations include the effect of the epi- zania, projects the cost of replacing teachers at $40 mil- demic on population growth, which will slow slightly lion through 2010. in severely affected countries. The death of an adult can tip vulnerable households The heavy macroeconomic impact of AIDS comes into poverty. Even in Tanzania, where the government partly from the high costs of treatment, which divert pays a large share of health costs, a World Bank study resources from productive investments . Tanzanian cli- shows that affected rural households in 1991 spent nicians estimate that, on average, an HIV-infected $60-roughly the equivalent of annual rural income per adult suffers 17 episodes of HIV-related illnesses prior capita-on treatment and funerals . The study also to death and a child suffers 6.5 episodes. Depending showed that the effects of losing an adult persist into on how much medical care a patient gets, in the typical the next generation as children are withdrawn from developing country the total cost per adult death school to help at home. School attendance of young ranges from 8 to 400 percent of annual income per cap- people ages 15-20 is reduced by half if the household ita; the average is about 150 percent of annual income has lost an adult female member in the previous year. per capita. and $1.3 billion (depending on the number of peo- previously uninfected person every four years . At ple treated) in annual treatment costs. The pro- this rate, there will be six HIV-positive persons in gram's net return, after discounting at even as 2000 for every one today. If the transmission rate much as 12 percent a year, was calculated to be could be slowed to one every five years, that num- between $18 million and $480 million. ber could be reduced to only four infected persons AIDS is another example. Although it remains in 2000 for every one today. The corresponding much less common in the developing world than reduction in medical costs, after discounting at 3 diseases such as malaria, its economic impact per percent a year, amounts to $750 by 2000 for each case is greater for two reasons: it mainly affects currently HIV-positive person in India, for a total adults in their most productive years, and the in- saving of $750 million . Similar calculations for fections resulting from it lead to heavy demand for Thailand suggest savings of $1,250 per currently expensive health care (Box 1.2). For example, be- HIV-positive person, for a potential total of $560 cause individuals with AIDS are typically more million. prone to pneumonia, diarrhea, and tuberculosis, Health investments and poverty the cost of medical care is high even though there is no effective treatment as yet for the disease it- The goal of reducing poverty provides a different self. Research in nine developing and seven high- but equally powerful case for health investments. income countries suggests that preventing a case The adverse effects of ill health are greatest for of AIDS saves, on average, about twice GNP per poor people, mainly because they are ill more of- capita in discounted lifetime costs of medical care; ten, but partly because their income depends ex- in some urban areas the saving may be as much as clusively on physical labor and they have no sav- five times GNP per capita. Calculations for India ings to cushion the blow . They may therefore find show that, given prevailing transmission patterns, it impossible to recover from an illness with their each currently HIV-positive person infects one human and financial capital intact . 20 The health consequences of poverty are severe: of growth in income per capita between 1960 and the poor die younger and suffer more from disabil- 1990 in about seventy countries to the initial level ity. In Porto Alegre, Brazil, adult mortality rates in of national income, the initial educational level, poor areas in the late 1980s were 75 percent higher and an indicator of initial health status (the child than in rich areas, and in Sao Paulo rates were two mortality rate, used in this Report to mean the risk to three times higher for nonprofessionals than for of dying by age 5 per 1,000 live births). The health professionals. In the late 1970s among Kenyan status indicator is found to be a highly significant families in which the mother had no schooling, the predictor of economic performance. For the aver- probability of dying by age 2 averaged 184 per age country in the sample, the annual growth rate 1,000 in regions where half of the families lived of income per capita is 1.40 percent and the child below the poverty line but 100 per 1,000 in regions mortality rate is 116 per 1,000. An otherwise aver- where only one-fifth of the families lived in pov- age country with a child mortality rate of 106 erty. The poor are exposed to greater risks from would have a growth rate of income per capita of unhealthy and dangerous conditions, both at 1.55 percent, whereas one with a child mortality home and at work. Malnourishment and the leg- rate of 126 would have a growth rate of 1.26 acy of past illness mean that they are more likely to percent. fall ill and slower to recover, especially as they Not surprisingly, the health status variable is have little access to health care. strongly correlated with educational stock, but the When a family's breadwinner becomes ill, other significant association between income growth members of the household may at first cope by and health remains strong and of similar magni- working harder themselves and by reducing con- tude across time periods and for a range of model sumption, perhaps even of food. Both adjust- formulations. Although it is possible that unob- ments can harm the .health of the whole family. If served factors such as government capacity to im- free health care is not available, the costs of treat- plement effective policies could explain the appar- ment may drive a household deeper into debt. Al- ent association, the data do suggest that better though ill health is only one of many factors that health means more rapid growth. can cause financial distress, its potential for disas- ter means that it should be explicitly recognized in The record of success formulating policies. Investments to reduce health risks among the poor and provision of insurance Mortality started to decline in Europe, North against catastrophic health care costs are impor- America, and Australasia about two centuries ago, tant elements in a strategy for reducing poverty. but slowly at first. A century ago life expectancy in Spending on health is a productive investment: the United States, then the world's richest coun- it can raise incomes, particularly among the poor, try, was only forty-nine years, and child mortality and it reduces the toll of human suffering from ill was about 180 per 1,000. The rate of improvement health. Good health, however, is a fundamental accelerated in the first half of this century; by 1950 goal of development as well as a means of acceler- life expectancy in the United States had increased ating it. Targeting health as part of development to sixty-six years, and child mortality had fallen to efforts is an effective way to improve welfare in 34 per 1,000. Progress was also being made in de- low-income countries. Evidence gathered over the veloping countries: in Chile, for example, life ex- past thirty years indicates that in health, unlike pectancy increased from thirty-seven years in 1930 income, the gap between poor and rich countries to forty-nine in 1950, and child mortality fell from has been narrowing. 350 to 209 per 1,000. Putting the effects together Mortality transitions since 1950 The detrimental effects of poor health on individ- Health conditions around the world have im- uals and households and on the use of resources proved more in the past forty years than in all suggest that better health should lead to better previous human history. Life expectancy at birth economic performance at the national level. A in developing countries increased from forty to number of analyses have found a positive relation- sixty-three years, and child mortality fell from 280 ship between growth of income per capita and the to 106 per 1,000. In a high-income country life ex- initial national educational stock. A similar analy- pectancy is more than seventy-five years; in a low- sis carried out for this Report examines the relation mortality developing country it is seventy years or 21 Child mortality has fallen sharply in the past thirty years, with particularly rapid declines in parts of Asia and Latin America. Figure 1.1 Child mortality by country, 1960 and 1990 ,, Under-five 111ortality rate • 175 or more ·• 125- 174 D 75-124 D 50-74 D 25-49 • Less than 25 ,, Source: Appendix A. 22 more; and in Sub-Saharan Africa, the region Life expectancy has increased substantially where least progress has been made, it is about everywhere over the past fo'rty years. fifty years. Much of what is known about the decline in mortality in the developing world since 1950 is Figure 1.2 Trends in life expectancy limited to the mortality of children and has come by demographic region, 195Q-90 from a series of standardized, internationally funded demographic surveys . Enormous reduc- Life expectancy at birth (years) tions in child mortality occurred almost every- 80 where around the world between 1960 and 1990 (Figure 1.1). For example, child mortality in Chile dropped from 155 to 20 per 1,000, in Tunisia from 245 to 45, and in Sri Lanka from 140 to 22. 70 The statistics for adult mortality in the develop- ing world are much less satisfactory than those for child mortality. Approximate estimates for all de- 60 veloping countries suggest that the adult mortality rate (defined as the probability of dying between ages 15 and 60 per 1,000 persons reaching age 15) fell from about 450 in 1950 to about 230 in 1990. In 50 Chile, a country with excellent statistics, the rate dropped from 466 in 1930 to 152 in 1990. The decline in mortality has accelerated over the 40 past thirty years. In the 1960s child mortality fell 1950 1960 1970 1980 1990 by approximately 2 percent a year in about seventy developing countries for which estimates are avail- able. The annual decline increased to more than 3 Sub-Saharan Africa percent in the 1970s and to more than 5 percent in India the 1980s. This result could be skewed by changes China in the mix of countries with reliable data; there Other Asia and islands were, however, twenty-one countries with a con- Latin America and the Caribbean tinuous series of acceptable estimates of child mor- Middle Eastern crescent tality from the early 1%0s to the late 1980s, and for Formerly socialist economies of Europe this group as a whole the fall in child mortality Established market economies averaged 3 percent a year in the 1960s but 6 per- cent a year in the 1980s. In seventeen of the Source: Appendix A. twenty-one the pace of decline increased over the period. Regional patterns There are strong parallels between the pattern of The extent of success has varied significantly be- mortality decline in the high-income countries and tween regions. Between 1950 and 1990 all eight the accelerated progress of developing countries demographic regions used for this Report enjoyed over the past forty years. In both groups the con- increases in life expectancy at birth, but China and trol of communicable diseases, particularly those the Middle Eastern crescent did particularly well of childhood, accounts for most of the gains. (The (see Figure 1.2). Sub-Saharan Africa showed the term ''communicable diseases,'' in the analyses slowest improvement, with life expectancy in- for this Report, includes deaths from maternal and creasing only from thirty-nine to fifty-two years- perinatal causes.) Progress against noncommuni- although even this compares well with European cable diseases-primarily those of the circulatory experience in the nineteenth century. (It took En- and respiratory systems, which principally affect gland and Wales more than half a century to raise adults-has been much slower. In both Chile (from life expectancy by a similar amount.) The formerly 1930 to 1987) and England and Wales (over the socialist economies of Europe showed a rapid im- longer period 1891 to 1990) mortality from commu- provement in the 1950s and 1960s, but the rise was nicable disease fell to less than 5 percent of its much slower in the 1970s and 1980s. initial level, whereas mortality from noncommuni- 23 Mortality from communicable diseases has fallen Similar patterns in Chile and in England and Wales much faster than that from noncommunicable show how mortality rates have declined much more diseases or injuries. sharply for the young than for the old. Figure 1.3 Age-standardized female death Figure 1.4 Change in female age-specific rates in Chile and in England and Wales, mortality rates in Chile and in England selected years and Wales, selected years • Earlier period" 0 Later period b Ratio of mortality rates at two time periods Chile 10 Communicable 0.8 I diseases< Noncommunicable diseasesd 0.6 ~ I Chile " ./ 0.4 ,.- v ' ,r-- Injuries I 0.2 b l--" ~ II' England and England and Wales o I" ~ - vyales~ I 0 5 15 25 35 45 55 65 75 85 Communicable Age diseases< a. Ratio of 1990 rates to 1930 rates. b. Ratio of 1981 rates to 1891 rates. Noncommunicable Source: For Chile 1930 and England and Wales 1891 ; diseasesd Preston, Keyfitz, and Schoen 1972; for Chile 1990, World Health Organization datil; for England and Wales 1990, United Nations, Dc111ogmpl1ic Ycarlook, 1991. Injuries 0 2 4 6 8 10 12 14 adults. In Chile, for example, mortality risks up to Age-standardized death rate age 30 fell by more than 90 percent between 1930 per 1,000 population and 1990; the decline was at least 60 percent at ages 30-70, but above age 70 the gains were much .a. For Chile, 1930; for England and Wales, 1891 . smaller (Figure 1.4). The age pattern of mortality b. For Chile, 1'1H7; for England and Wales, 1990. decline in Chile over sixty years is strikingly simi- c. Includes ;naternal and neonatal mortality . d . For earlier period, includes "other and unknown" lar to the pattern in England and Wales during the category. ninety years from 1891 to 1981. Source: For C hile 1'130 and England and Wales 1891 , Preston, Keyfitz, and Schoen 1'172; for 1987 and 1990! The only exception to this broad similarity be- WHO, World /lcai/IJ Stat istics Aunua/, 1989 and 1991. tween industrial and developing countries has been in the formerly socialist economies. In these countries child mortality has continued to decline, as has the mortality of women, albeit more slowly. cable disease fell much less rapidly (Figure 1.3). The mortality of adult men, on the other hand, has One result of this change is that mortality risks stopped declining in the past two decades and has have fallen much faster for children than for actually started to increase. This excess male mor- 24 tality is largely the result of extremely high death Table 1.1 Burden of disease by sex, cause, rates from cardiovascular disease, associated with and type of loss, 1990 (millions of DALYs) heavy smoking and drinking. Disease category Measuring the burden of disease Sex and outcome Communicable" Noncommunicable Injuries Male The health improvements of the past few decades Premature death 259 152 70 Disability 47 146 39 have done much to enhance human welfare, both directly and indirectly. But much more remains to Female Premature death 244 135 33 be done. Communicable (and largely preventable) Disability 74 142 20 diseases are still common. Health systems also Note: DALY, disability-adjusted life year. have to cope with the aging of populations, which a. Includes maternal and perinatal causes. leads to an increased burden of the more expen- Source: Appendix B. sive noncommunicable diseases. New illnesses, such as AIDS, have emerged. One simple statistic gives a sense of the remaining burden of disease: arrhea, childhood diseases such as measles, respi- about 12.4 million children under age 5 died in ratory infections, worm infections, and malaria ac- 1990 in the developing world. Had those children count for one-quarter of the GBD. The burden of faced the mortality risks of children in the estab- these largely preventable or inexpensively curable lished market economies, the number of deaths diseases of children is far larger in Sub-Saharan would have been cut by more than 90 percent, to Africa (43 percent of all DALYs lost) than any- 1.1 million. where else, although it is still substantial in India Any discussion of health policy must start with (28 percent), Other Asia and islands (29 percent), a sense of the scale of health problems. These and the Middle Eastern crescent (29 percent). For problems are often assessed in terms of mortality, adults too, communicable diseases are far from but that indicator fails to account for the losses that trivial: sexually transmitted diseases (STDs) and occur this side of death because of handicap, pain, tuberculosis together contribute 7 percent of the or other disability. A background study for this GBD. Report, undertaken jointly with the World Health Even as broad a measure as the GBD does not Organization, measures the global burden of dis- capture all the consequences of disease or injury. It ease (GBD) by combining (a) losses from prema- excludes the social costs of disfigurement, such as ture death, which is defined as the difference be- that arising from river blindness or leprosy, and of tween actual age at death and life expectancy at dysfunction-for example, marital breakups re- that age in a low-mortality population, and (b) loss sulting from obstetric fistula (permanent damage of healthy life resulting from disability. The GBD is to the reproductive tract incurred during delivery). measured in units of disability-adjusted life years And some health-related factors are likely to be (DALYs). Worldwide, 1.36 billion DALYs were lost underreported. A clear example is violence against in 1990, the equivalent of 42 million deaths of new- women, much of which goes undetected-but not born children or of 80 million deaths at age 50. unsuffered. Premature deaths were responsible for 66 percent Comparisons of absolute numbers of DALYs lost of all DALYs lost and disabilities for 34 percent. In may be misleading because the sizes and age struc- the developing world 67 percent of all DALY loss tures of the populations at risk are not the same. was a result of premature death; in the established The effects of population size can be allowed for by market economies and the formerly socialist econ- expressing the 1990 burden per 1,000 population. omies of Europe the figure was only 55 percent. Figure 1.5 shows the resulting rates by sex and Table 1.1 shows the GBD broken down by sex, regional group. This index is 259 for the world as a category of disease, and type of loss (premature whole, but it varies widely among regions. Sub- death or disability). The three categories of disease Saharan Africa loses 574 DALYs for every 1,000 used are the group of communicable diseases, population, more than twice the global average. noncommunicable diseases, and injuries. India, the Middle Eastern crescent, and Other Asia The derivation and interpretation of the GBD and islands all have values between 250 and 350. are explained in Box 1.3. The results of research on For China, the formerly socialist economies of Eu- the GBD challenge the belief that the war against rope, and Latin America and the Caribbean, the infectious and parasitic diseases has been won. Di- figures are between 150 and 250. The burden per 25 Box 1.3 Measuring the burden of diseas·e Most assessments of the relative importance of differ- healthy life lost was then obtained by multiplying the ent diseases are based on how many deaths they cause. expected duration of the condition (to remission or to This convention has certain merits: death is an unam- death) by a severity weight that measured the severity biguous event, and the statistical systems of many of the disability in comparison with loss of life. Dis- countries routinely produce the data· required . There eases were grouped into six classes of severity of dis- are, however, many diseases or conditions that are not ability; for example, class 2, ..yhich includes most cases fatal but that are responsible for great loss of healthy of leprosy and half· the cases of pelvic inflammatory life: examples are chronic depression and paralysis disease, was g~ven a severity weight of 0.22, and class caused by polio. These conditions are common, can 4, which includes 30 percent of cases of dementia and last a long ti!Tie, and frequently lead to significant de- 50 percent-·of those of blindness, was assigned a sever- mands on health systems . ity weight o£.0.6. The death and disability losses were To quantify the full loss of healthy life, the World then combined, and allowance was made for a dis- Bank and the World Health Organization undertook a coun't rate of 3 percent (so that future years of healthy joint exercise for this Report. Diseases were classified life were valued at progressively lower levels) and for into 109 categories on the basis of the li1lernational C/as- age weights (so that years of life lost at different ages sificatio'' of Diseases (ninth revision). These categories were given different relative values). The value for cover all possible causes of death and about 95 percent each year of life lost, shown in the left-hand panel of of the possible causes of disability. Using the recorded Box figure 1.3, rises steeply from zero at birth to a peak cause of death where available, and expert judgment at age 25 and then declines gradually with increasing when records were not available, the study assigned all age. These age ·weights reflect a consensus judgment, deaths in 1990 to these categories by age, sex, and .de- but other patterns could be used-for example, uni- mographic region . For each death, the number of years form age weights, with each year of life having the of life lost was defined as the difference between the same value, which would increase the relative impor- actual age at death and the expectation of life at that, tance of childhood diseases. age in a low-mortality population . For disability, the The combination of discounting and age weights incidence of cases by age, sex, and demographic region· produces the pattern of DALYs (disability-adjusted life was estimated on the basis of community surveys or, years) lost by a death at -each age . As the right-hand failing that, expert opinion; the number of years of panel of Box figure 1.3 shows, the death of a newborn Box figure 1.3 Age patterns of age weights and DALY losses Value of a year of life DALYs lost by death at given year (females) Relative value of a year of life at age x Disability-adjusted life years (DALYs) 1.6 40 1.2 / 1---- ~ 30 v r--..1'-. If ' 0.8 1 I " "" !'...r---.... 20 "' ~ ' "" "" ~ 10 ~ ........ 0 I 0 .......... 0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 90 Agex Age at death in years 1-------- Sou rce: World Bank data . 26 Box table 1.3 Distribution of DALY loss by cause and demographic region, 1990 (percent) Fo rmerlv Sub' Latin Middle sodalisi Established Saharan Oth er Asia America and Eusft•n1 ccmowics of t.• market Cause World Africa India C!zi>m and islands the Caribbean cresCe11t Europe economies Population (millions) 5,267 510 850 1,134 683 444 503 346 798 Communicable diseases 45.8 71.3 50.5 25 .3 48.5 42.2 51.0 8.6 9.7 Tuberculosis 3.4 4.7 3.7 2.9 5.1 2.5 2.8 0.6 0.2 STDsand HIV 3. 8 8.8 2.7 1.7 1.5 6.6 0.7 1.2 3.4 Di.1rrhe~ 7.3 10.4 9.6 2.1 8.3 5.7 10.7 0.4 0.3 Vaccine-preventable childhood infections 5.0 9.6 6.7 0.9 4.5 1.6 6.0 0.1 0.1 Malaria 2.6 10.8 0..3 1.4 0.4 0.2 Worm infections 1.8 1.8 0.9 3.4 3.4 2.5 0.4 Respiratory infections 9.0 10.8 10.9 6.4 11 .1 6.2 11.5 2.6 2.6 Maternal causes 2.2 2.7 2.7 1.2 2.5 1.7 2.9 0.8 0.6 Perin.1tal causes 7.3 7.1 9.1 5.2 7.4 9.1 10.9 2.4 2.2 Other 3.5 4.6 4.0 1.4 3.3 5.8 4.9 0.6 0.5 Noncommunicable diseases 42.2 19.4 40.4 58.0 40.1 42.8 36.0 74.8 78.4 Cancer 5.8 1.5 4.1 9.2 4.4 5.2 3.4 14.8 19.1 Nutritional deficiencies 3.9 2.8 6.2 3.3 4.6 4.6 3.7 1.4 1.7 Neuropsychiatric disease 6.8 3.3 6.1 8.0 7.0 8.0 5.6 11 .1 15.0 Cerebrovascular disease 3.2 1.5 2.1 6.3 2.1 2.6 2.4 8.9 5.3 Ischemic heart disease 3.1 0.4 2.8 2.1 3.5 2.7 1.8 13.7 10.0 Pulmonary obstruction 1.3 0.2 0.6 5.5 0.5 0.7 0.5 1.6 1.7 Other 18.0 9.7 18.5 23.6 17.9 19.1 18. 7 23.4 25.6 Injuries 11.9 9.3 9.1 16.7 113 15.0 13.0 16.6 11.9 Motor vehicle 2.3 1.3 1.1 2.3 2.3 5.7 3.3 3.7 3.5 Intentional 3.7 4.2 1.2 5.1 3.2 4.3 5.2 4.8 4.0 Other 5.9 3.9 6.8 9.3 5.8 5.0 4.6 8.1 4.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Millions of DALYs 1,362 293 292 201 177 103 144 58 94 Equivalent infant deaths (millions) 42 .0 9.0 9.0 6.2 5.5 3.2 4.4 1.8 2.9 DALYs per 1,000 population 259 575 344 178 260 233 286 168 11 7 • Less than 0.05 percent. Note: DALY, disability-adjusted life year; STD, sexually transmitted disease; HIV. human immunodeficiency virus. Sou rce: World Bank data. baby girl represents a loss of 32.5 DALYs; a female one .year per person. A baby who died in 1990 contrib- death at age 30 means the loss ~f 29 DALY~; and a uted about thirty-two years (the discount ed value of female death at age 60 represents 12 lost DALYs. about eighty years of expected life) to the burden but (Values are slightly lower for males .) The sum across all counted as one in the population . To take an extreme ages, conditions, and regions is referred to as the case, if the entfre population of the world were to be global burden of disease (GBD) . More details on the killed in one year, the burden per 1,000 population in GBD are presented in Appendix B. that year wottld exceed 20,000 DALYs. There is there- The global burden measures the present value of the ·fore no absolute scale w ith which the GBD per 1,000 future stream of disability-free life lost as a result of population can be compared; the only co mparisons death, disease, or injury in 1990. It is thus based on t\1at make sense are those between categories-of re- events that occurred in 1990 but includes the loss of gions, risk factors, disease groups, or sex. Box table 1.3 disability-free life in future yea.rs. This Report ex- shows the GBD by cause and demographic region. presses the burden in three distinct ways: as the num- The approach used to compute the GBD can also be ber of DALYs, as a percen.tage of some larger aggregate used to track improvements in a nation ' s health over (such as the percentage of total loss attributable to a time by following changes in the national burden of specific disease), and in relation to population size in disease. Preliminary plans for initial national assess- 1990. This last measure tails for careful interpretation ments have been developed for Costa Rica, South Af- because all future loss is expressed in relation to the rica, and Andhra Pradesh State in India . .current population, and the measure can easily exceed 27 Disease burdens per 1,000 population vary substantially by region. Figure 1.5 Disease burden by sex and demographic region, 1990 Disease burden per 1,000 population 0 100 200 300 400 500 600 650 Sub-Saharan Africa India China Other Asia and islands Latin America and the Caribbean Middle Eastern crescent Formerly socialist economies of Europe Established market economies World liZ! Females [] Males Sottrce: Appendix B. 1,000 population for the established market econ- bers of "missing" (and presumed dead) women in omies is easily the lowest, at 117. It turns out that relation to the expected population balance be- these broad rankings are not significantly affected tween the sexes. In China illegal female infanticide by differences in age distributions between (and, in the recent past, illegal sex-selective abor- regions. tion) is thought to be the main reason. Females have abcut a 10 percent lower disease burden per 1,000 population than males for the world as a whole. They lose fewer DALYs from Table 1.2 Burden of five major diseases by age of incidence and sex, 1990 premature mortality, but their DALY loss from dis- (millions of DALYs) ability is about the same as for males. Within the Age (years) disability category, however, the female disease Disease and sex 0-4 5- 14 15-44 45-59 60+ Total burden from the group of communicable causes is Diarrhea considerably higher than that for males, partly be- Male 42.1 4.6 2.8 0.4 0.2 50.2 cause of a substantial toll from maternal causes but Female 40.7 4.8 2.8 0.4 0.3 48.9 also because of the much greater female burden Wonn infection associated with STDs. Effective interventions exist Male 0.2 10.6 1.6 0.5 0.1 13.1 for much of this excess female burden. For all Female 0.1 9.2 0.9 0.5 0.1 10.9 causes together, the female advantage ranges from Tuberculosis more than 30 percent in the formerly socialist · Male 1.2 3.1 13.4 6.2 2.6 26.5 economies, where adult mortality is much higher Female 1.3 3.8 10.9 2.8 1.2 20.0 for men than for women, to negative in India, Ischemic heart disease where females suffer a disadvantage of 8 percent. Male 0.1 0.1 3.6 8.1 13.1 25.0 Both India and China show a female disadvantage Female 1.2 3.2 13.0 17.5 •• Less than 0.05 m illion. in disease burden per 1,000 population, and, not Note: DALY, disability-adjusted life year. coincidentally, both countries also have large num- Source: World Bank data. 28 The figures on disease burden by age (Table 1.2) The share of communicable diseases in the disease suggest how health officials should target their burden declines as mortality rates fall. programs. More than 80 percent of the DALY loss from diarrhea is a result of infections in children under age 5. Worm infections are concentrated Figure 1.6 Distribution of disability- among children ages 5-14. More than half the bur- adusted life years (DALYs) lost, by cause, den of tuberculosis is borne by the 15-44 age for selected demographic regions, 1990 group. More than 60 percent of the burden of isch- (percentage of total DALYs lost) emic heart disease falls on the population over age 60. Sub-Saharan Latin America The higher the disease burden, the higher the Africa and the Caribbean proportion attributable to the communicable group of causes (Figure 1.6). Sub-Saharan Africa has the highest disease burden per 1,000 popula- tion, and 71 percent of this is from the communica- ble disease group, whereas in Latin America (a medium-burden region) the figure is 42 percent and in the established market economies it is only 10 percent. Noncommunicable diseases show the opposite pattern, accounting for 19 percent of the total burden in Sub-Saharan Africa, 43 percent in Latin America, and 78 percent in the established Established market market economies. Despite these marked differ- economies ences in relative burden, however, the absolute rates of loss for both groups are highest in Sub- Saharan Africa and lowest in the established mar- ket economies. The pattern is plain: as health im- proves, the burden from all types of disease declines, but the distribution of the burden shifts dramatically from a preponderance of communica- ble disease to a preponderance of noncommunica- ble disease. Despite the sharp improvements in health around the world, the GBD calculations show that D Communicable diseases" a large burden of premature mortality and disabil- D Noncommunicable diseases ity still remains, particularly in the world's poorer • Injuries regions. There are inexpensive and effective ways to eliminate the share caused by communicable a. Includes maternal and perinatal causes. Source: Appendix B. diseases (other than maternal and perinatal condi- tions), which is roughly 35 percent of the world burden and more than 60 percent in Sub-Saharan Africa. The remaining 65 percent of the world bur- den is less responsive to such measures, and re- ber of drug-resistant disease strains; and the con- ducing it will require changes in the behavior and tinued use of health-damaging substances such as life-styles of adults. tobacco. Although nobody can forecast the impact of these challenges with any precision, reasonable Challenges for the future projections are possible. For example, outside the established market economies the number of New health challenges will emerge over the next deaths attributable to smoking is expected to in- few decades. Some are certain: these involve the crease from 1.7 million in 1990 (40 percent of which significant increase in noncommunicable diseases were in the formerly socialist economies of Eu- arising from the continuing demographic transi- rope) to more than 3 million by 2005 and to about tion. Others are less certain: the spread of HIV and 4.5 million by 2015. Other challenges are poten- the increase in AIDS deaths; the increasing num- tially important but not forecastable: possible ex- 29 Worldwide, fertility and mortality declines go hand in hand. Box .1.4 The demographic and epidemiological transitions Figure 1.7 Trends in life expectancy and Changes in the pattern of diSl'.!Sl' pnKL'l'd in two steps. fertility in Sub-Saharan Africa and Latin The first is the demographic transition, wlwn mort.1lity America and the Caribbean, 1960-2020 from infectious diseaSL' declines ,md, p<~rtly as" n•stiit, fertility decreases as well. Thl' second, ,1 consequl·nCl' of declining fertility and diffen•nti,tl r·'•-•• ... .. "' : .;.;~ 5-9 ·;;, 0-4 10 8 6 4 2 0 10 20 30 40 50 10 8 6 4 2 0 10 20 30 40 50 Percentage of Percentage of Percentage of Percentage of total population total deaths' total population total deaths +- Median age af deathb a . Projected. ·b. The age below which half of all deaths in a year occur. Source: For England and Wales 1891, Preston, Keyfitz, and Schoen ·1972; for England and Wales 1966, United Nations, Dcllwgmplric Yearbook, 1978; for Latin America and the Caribbean,. 1955 and 2030, World Bank data. 31 in tandem beginning in the late nineteenth cen- eases will increase sharply, both absolutely and tury, and population growth rates rarely exceeded proportionately. At the same time, the challenge of 2 percent a year.) communicable diseases of the young will persist. These demographic changes are having, and Despite declines in fertility, the number of births will continue to have, dramatic effects on age dis- each year in developing countries will rise some- tributions in developing countries. As fertility de- what, from 127 million in 1990 to 145 million in clines, age structures in these countries are evolv- 2020, before decreasing to 142 million in 2030. The ing toward the existing patterns in the established number of children under age 5 will increase more market economies and the formerly socialist econ- rapidly, from 552 million in 1990 to 682 million in omies (see Box 1.4) . The proportion of the popula- 2030. These changes, which took a century to com- tion age 65 and over is expected to increase from 4 plete in today's high-income countries, are occur- percent in 1990 to 9 percent by 2030 (in absolute ring within fifty years or less in parts of the devel- numbers, from 184 million to 678 million). As a oping world. result, the burden from noncommunicable dis- In judging the importance of the health prob- lems of the young in comparison with those of the elderly, one useful guide is the median age at death. For all six regions of the developing world, All regions will experience the health transition, but the median was below 20 in 1950, indicating the the timing will differ. dominance of the health problems of children . By 1990 the median had risen close to age 60 in China Figure 1.8 Median age at death, by and in Latin America but was scarcely above 5 in demographic region, 1950, 1990, and 2030 Sub-Saharan Africa and was still below 25 in the Middle Eastern crescent (see Figure 1.8) . By 2030, assuming that current trends continue, the median Median age at death (years)" age at death will have risen above 60 in all regions 80 except Sub-Saharan Africa, where it will still be close to 40 . The message from these population projections is that health services must plan for a modest in- crease in child-related demands over the next forty years. At the same time, the numbers of the el- derly, with very different health needs, will be ris- ing sharply. The pace of demographic change has been, and is expected to continue to be, faster in the developing world than it was in the high- income countries, and the problems of adaptation are therefore greater. Because treatments for non- communicable diseases are often expensive, there is a danger that these diseases will absorb re- sources needed to combat communicable diseases 1950 1990 2030 b (which will still be widespread). This kind of di- lemma has already been noted in World Bank studies in Brazil and China . • Sub-Saharan Africa • Middle Eastern crescent HIVandAIDS D India D Other Asia and islands More than 80 percent of the estimated 8.8 million • Latin America and the Caribbean people infected with HIV in 1990 lived in develop- • China ing countries. There the disease is primarily one of heterosexual adults, with substantial perinatal in- a. The age below w hich half of all deaths in a year occur. fection of young children . Of the eight demo- b. Projected . Source: World Bank data. graphic regions used in this Report, only the for- merly socialist economies, the Middle Eastern crescent, and China have little recorded spread of 32 Table 1.3 Evolution of the HIV-AIDS epidemic HIV incidence HlV prevalence AIDS-related (millions) (millimzs) deaths (millions) Region 1990 2000·' 1990 2000• 1990 2000·' Demographically developing groupb 1.6 2.5 7.4 25 0.3 1.7 Sub-Saharan Africa 1.1 1.0 5.8 12 0.3 0.9 Asia' 0.3 1.3 0.4 9 0.6 EME and FSEd 0.1 1.4 0.1 0.1 Total 1.7 2.5 8.8 26 0.4 1.8 ** Less than 0.05 million. Note: Incidence refers to new infections in a given year; prevalence refers to the total number of persons infected. a. Conservative estimates. b. The countries of the demographic regions Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent. c. India, China, and the demographic region Other Asia and islands. d. EME, established market economies; FSE, formerly socialist economies of Europe. Source: World Health Organization data. the virus. Spread of the virus may be about to mission, incubation periods, survival times, and occur even in these three regions. It takes six to ten the role of such factors as STDs-are not well years, on average, for an HIV-infected adult to de- quantified, and accurate projections are therefore velop AIDS. Thus, regardless of future changes in impossible. In the African communities that are transmission of the virus, there will certainly be an most severely affected, early assessments pre- increasing number of AIDS cases over the next few dicted absolute declines in population. Later views years. suggest that population growth will continue, al- It is difficult to predict the future course of the beit at a reduced rate. Trial projections for Sub- epidemic because so little is known about the dy- Saharan Africa, based on a high assumption of namics of HIV transmission. WHO has projected HIV prevalence of 60 million infections worldwide that in 2000, 2.5 million people will be newly infec- in 2000, suggest a reduction in life expectancy by ted with HIV, HIV prevalence will have reached 26 2010 of about six years, in comparison with a low- million, and AIDS deaths will total 1.8 million (see HIV model, and a 25 percent increase in adult mor- Table 1.3). These estimates are conservative, since tality. The effect on population growth would still they assume that the rate of new infections in Af- be modest: a reduction of about 0.25 percentage rica will slow somewhat and that new transmis- point a year, to an annual rate of 2.7 rather than sion will be concentrated in India and in the Other 2.95 percent in 2005-10. In areas such as Thailand Asia and islands region. If no effective interven- where fertility and mortality rates are much lower tions to slow transmission are introduced, the total than in Sub-Saharan Africa, AIDS may well con- number of deaths may be twice as large, in which tribute to actual population declines over a period case AIDS would be responsible for 8 percent of of thirty years or more. the global burden of disease by 2000 instead of the 3.5 percent implied by the estimates in Table 1.3. Drug-resistant diseases If, however, sexual behavior changes dramatically over the next decade, even the conservative pro- Microbes evolve as a result of natural mutation, jections given here may prove too pessimistic. which throws up new threats, and of drug thera- Relatively modest reductions in numbers of casual py-induced selection, which fosters drug resis- sexual partners, or in the prevalence of STDs-or, tance. Two major new threats have arisen in this alternatively, substantial increases in condom century: the influenza virus responsible for the use-could reduce transmission significan!ly. 1918-19 worldwide epidemic, and HIV. Early (and still tentative) findings from Thailand The evolution of drug resistance, partly driven are encouraging; perhaps behavior really will by incomplete or inadequate treatment, is more change. gradual and less dramatic but no less serious. The Opinions differ concerning the effects of AIDS everyday bacteria responsible for pneumonias and on population growth. The variables needed to diarrheas have become resistant to the older anti- model the epidemic-including baseline rates of biotics and will gradually do the same with the infections, behavioral risk factors, efficacy of trans- newer antibiotics developed over the past few dec- 33 Life expectancy is related to income, but the emerging. In the developing world the diseases for relationship has shifted upward during the twentieth which drug resistance is already a major issue (re- century. spiratory infections, tuberculosis, STDs, and ma- laria) accounted for almost one-fifth of the GBD in 1990. Figure 1.9 Life expectancy and income per This steady evolution of drug-resistant microbes capita for selected countries and periods poses challenges for research and for health care. Better understanding of infectious agents is needed as a basis for the development of new ther- Life expectancy (years) apies. Health providers must consider the effects 80 of drug use on the evolution of resistant microbial strains. Basic scientific advances can contribute to tracking resistance, as recently shown by develop- 70 ments in identifying drug-resistant strains of tu- berculosis. There is little reason to hope for perma- nent success in humanity's struggle against 60 infection; investments in scientific research and vigilance on the part of public health authorities will remain indispensable. 50 These problems arising from microbial evolution are most severe in Sub-Saharan Africa. If efforts to control the spread of HIV fail, by 2000 an addi- 40 • tional 1 million people in the region will be dying from AIDS each year. Most of them will be young adults who would otherwise have gone on to live 30 ~----L-----~----~----~----~ healthy lives. If malaria develops resistance to all 0 5,000 10,000 15,000 20,000 25,000 available drugs, the number of people it kills every Income per capita year could increase sharply, from the expected 1.5 (1991 international dollars) million deaths in 2000 to 2.3 million. Sub-Saharan Africa might also suffer from a tuberculosis epi- demic, driven partly by drug resistance and partly Note: International dollars are derived from national currencies not by use of exchange rates but by assessment by the spread of the disease by people with HIV of purchasing power. The effect is to raise the relative Extrapolation of current trends indicates an annual incomes of poorer countries, often substantially. For total of 8.5 million premature deaths in Sub- illustrative country comparisons and a more detailed explanation, see Table 30 in the World Development Saharan Africa by the end of the century. But it is Indicators. all too easy to project a figure as high as 11.5 mil- Source: Preston, Keyfitz, and Schoen 1972; World Bank data. lion, accompanied l:>y a sharp reduction in life expectancy. Lessons from the past: explaining declines in mortality ades. Tuberculosis resistant to the standard mix of antibiotics is becoming more common in the in- Three factors have been important in the dramatic dustrial world, and it kills many of those who con- and unprecedented mortality declines of the past tract it. Chloroquine-resistant malaria has now hundred years and in the still more dramatic de- spread to practically all endemic areas, and al- clines in developing countries since World War II. though new drug therapies are available, wide- These factors are income growth, improvements in spread resistance even to several of these is being medical technology, and public health programs reported. Malaria has thus reemerged as a signifi- combined with the spread of knowledge about cant health risk in urban areas that had been free health. of it for several decades. Resurgence of the disease has been abetted by the reduced effectiveness of Income growth vector control, which is partly attributable to the increasing resistance of mosquitoes to standard in- Increased income allows people, particularly the secticides. Resistant strains of many STDs are also poor, to buy more food, better housing, and more 34 health care. Throughout the twentieth century life health measures were responsible for only a small expectancy has been strongly associated at the na- part of the progress made. In the late nineteenth tional level with income per capita, as seen in Fig- century Robert Koch showed that the bacterium ure 1. 9. Life expectancy rises rapidly with income M. tuberculosis causes tuberculosis, and people be- at low levels of income, particularly when income gan to understand about germs. They took simple per capita is less than $3,000 (1991 purchasing precautions-preparing food and disposing of power dollars). The figure shows, however, that waste hygienically, eliminating flies, and quaran- the relationship has shifted upward over each tining sick family members-that had far-reaching thirty-year period, so that more health is realized benefits. Recent research has shown that child for a given income. For example, in 1900 life expec- mortality differed little by education or even by tancy in the United States was about forty-nine income in the United States in the last decade of years and income per capita in 1991 dollars was the nineteenth century but that differences wid- about $4,800. In 1990 that income per capita would ened sharply as child mortality fell in the early be associated with a life expectancy of about sev- twentieth century. The implication is that afflu- enty-one years. This upward shift shows that ence and education made little difference until sci- health depends on more than income alone. entific knowledge showed households how to reduce the dangers to their health. Since better- Improvements in medical technology educated individuals acquire and use new infor- mation more quickly, this emphasis on knowledge Before the 1930s medical technology had little to helps to explain the large differences in child mor- offer humanity, with the exception of smallpox in- tality by mother's education observed in develop- oculation, the use of which was widespread in Eu- ing countries today. rope from the late eighteenth century onward, and diphtheria antitoxin, discovered in 1894. Starting The potential for effective action in the 1930s, with the introduction of antibacterial drugs and new vaccines, a wide range of effective The recent declines in mortality in the developing interventions has become available to counter world have been sharper than the earlier declines most communicable diseases. in the high-income countries and more influenced The effect of these technological improvements by technical advances. To take one example, Sri on health has depended on other factors, such as Lanka achieved a remarkable decrease in mortality income gains for the poor, increased schooling, after World War II; the crude death rate fell from and public policies that affect health systems. As a 21.5 per 1,000 in 1945 to 12.4 in 1950. Some 23 result, outcomes have varied widely by country, percent of that drop has been attributed to the ma- even within the same region. For example, in the laria eradication program, which mainly involved early 1980s child mortality was three times higher spraying of insecticide from the air. The same ap- in Mali than in Botswana, six times higher in Boli- proach also did much to control yellow fever, on- via than in Chile, and five times higher in Ban- chocerciasis, and many other diseases. Wide- gladesh than in Sri Lanka. Between the early 1960s spread use of newly available antibiotics against and the early 1980s child mortality fell 20 percent conditions such as yaws in Africa helped to reduce in Bangladesh but 65 percent in Sri Lanka, 10 per- STDs and (probably) acute respiratory infections. cent in Uganda but 50 percent in Kenya, and 10 Improvements in water and sanitation curbed the percent in Haiti but nearly 80 percent in Costa spread of disease, particularly in towns and cities. Rica. Some countries have clearly made better use Whereas at the beginning of this century child of the available technology than others. mortality rates in today's high-income countries were much higher in urban than in rural areas, the Public health and the spread of knowledge opposite has been true of the developing world since 1950. The introduction of public health measures-par- Vaccination, too, has produced dramatic results, ticularly clean water, sanitation, and food regula- including the eradication of smallpox and the elim- tion-certainly contributed to the decline in child ination of paralytic polio in the Western Hemi- mortality in the late nineteenth century and to the sphere. About 80 percent of the world's children accelerated decline in the early twentieth century. are now vaccinated against the main infectious dis- The geographic distribution of mortality declines eases of childhood, thanks largely to the Expanded suggests, however, that until people began to un- Programme on Immunization (EPI) sponsored by derstand the sources of poor health, such public WHO and UNICEF. It is estimated that the EPI 35 prevented the deaths of 2.6 million children in more than 20 percent of the GBD in 1990. Epide- 1990 alone. Substantial benefits have also come miological advances are giving governments and from simple curative measures such as oral re- households warning of the enormous health toll hydration to avert death from diarrhea and a short from smoking. But if the full benefits of scientific course of drugs for curing tuberculosis. But there advances are to be realized, parallel developments is much more still to be done: in 1990 childhood are needed to empower households so that they deaths from diarrhea and immunizable diseases can put the advances into practice. The key devel- alone accounted for 12 percent of the GBD. opments are schooling, particularly of girls; in- The march of science has increased both the come growth, particularly of the poor; and a flex- range of inexpensive clinical treatments and prac- ible, responsive health system able to provide the tices and the potential performance of health sys- necessary preventive and curative care. The poli- tems. It is now possible to treat at low cost tuber- cies needed to achieve these developments are the culosis, STDs, many respiratory infections, and subject of the remainder of this Report. risky deliveries, which together accounted for 36 Households and health What people do with their lives and those of their their effects on the conditions facing households children affects their health far more than any- and individuals, can be important to people ' s thing that governments do. But what they can do health. Especially in the poorest countries, policies is determined, to a great extent, by their income that accelerate income growth and reduce poverty and knowledge-factors that are not completely make it possible for people to afford better diets, within their control . In every society, moreover, healthier living conditions, and better health care . the capabilities, income, and status of women ex- Policies to expand educational opportunities, par- ert a powerful influence on health. Because of ticularly for girls, help households achieve health- these interrelations, government actions, through ier lives by increasing their access to information Economic growth and investments in human resources interact to improve well-being. Figure 2.1 Mutually reinforcing cycles: reduction of poverty and development of human resources Health /;mpm"'m''"~ \ Expansion of Empowerment ..___.. schooling of women opportunities Economic growth ~ policies that ~ benefit the poor 37 and their ability to make good use of it. The same Household capacity: income and schooling goes for policies that work to ensure effective and accessible health services for all. When all these Within the household, health improves rapidly as policies are combined, they create a virtuous cycle people escape from poverty and low education in which reduction of poverty and improvements (Box 2.1). Beyond the household, every society's in health reinforce each other (Figure 2.1). health services are affected by its national income, Box 2.1 Progress in child health in four countries In the 1960s a child born in the developing world had a only 60 percent as fast . Thus, improvements in school- 77 percent chance of surviving the first five years of ing were most significant in Cote d'lvoire, whereas in life . About thirty years later, the chances of survival Egypt growth in income per capita accounted for fully have improved to 89 percent. How much did income half of the gain in child health. growth and expansion of schooling contribute to this Costa Rica and japan followed the same pattern as gain? What was the role of other factors, such as prog- Egypt: growth of income per capita contributed sub- ress in science and medicine' Some answers to these stantially more to child health gains than did educa- questions emerge from data on child survival from sev- tional improvements. Technical progress (estimated enty-five industrial and developing countries for the using the passage of time as a proxy), however, was period between 1960 and 1987 (see note to Appendix important in.Japan, whereas in Costa Rica and Egypt it table A.3). This box reviews the results for four coun- mattered less than improvements in education. Except tries with different income levels-Costa Rica, Cote in japan, where people were already quite well edu- d ' lvoire, Egypt, and japan (see Box table 2.1 and Box cated in 1960, the analysis probably underestimated figure 2.1). the contribution of schooling because it dealt with the In all four countries, part of the gain in child health schooling of all adults rather than of women alone. depends on the initial levels in 1960 of schooling in the Child health is particularly affected by maternal educa- population and of income per capita. Because school- tion, and the number of years of schooling received by ing and income per capita produce health benefits that younger women is likely to have risen much faster be- often persist through time, health in a population may tween 1960 and 1987 than was the case for the adult be improved simply by maintaining initial levels of population as a whole. schooling and income. In Costa Rica, where in 1960 income per capita was relatively high and schooling Box table 2.1 Child health, income per capita, was already widespread, initial conditions accounted and schooling in Costa Rica, Cote d'lvoire, for 58 percent of the gain in child· health between 1960 Egypt, and Japan, 1960-87 and 1987. In Cote d'lvoire and Egypt, where the levels Cote of schooling and income per capita were modest in Indicator Costa Rica d'lvoire·' Egvrt Japan 1960, initial conditions contributed only about one- 1960 fifth to one-quarter of the gains. In japan, too, these Child survivaJb 0.89 0.72 0.74 0.96 initial conditions contributed a fifth of the gains in child Income per capita health, but this is not surprising in a ·country where a (1987 international baby's chance of survival was already very good in dollars)< 2,160 1,021 557 2,701 1960. Average schooling In reality, of course, income and schooling have im- of adults (years) 4.0 .0.2 3.0 10.7 proved in all these countries, and these improvements Average a111111al percmtage clra11ge, 1960-87 contributed to further gains in child survival. In Cote Child survivalb 0.4 0.8 0.6 0.1 d'lvoire ·educational improvements did the most for Income per capita 2.3 3.2 5.2 5.3 child health, accounting for 66 percent of the gains be- Schooling of adult tween 1960 and 1980. For .Egypt, by contrast, the .figure population 2.0 11.8 2.4 0.2 was only 21 percent. A comparison between Cote Elasticity of child sun>ival witlr respect to: d'lvoire and Egypt is illuminating. The probability of Income per capita" 0.04 0.06 0.06 0.02 surviving the first five years of life started at similaF Schooling of adult levels in both countries and improved at comparable populationd 0.03 0.04 0.04 0.02 rates. In both, too, the responsiveness of child survival a . Data refer to 1960-80. b. Child survival refers to the probability of surviving from birth to income per capita and to the schooling of adults was through age 5. comparable. In Cote d'Ivoire, however, adult schooling c. Income is adjusted for differences in purchasing power parity. started from much lower levels than in Egypt but in- d . Elasticities denote the percentage change in the probability of surviving from birth through age 5 corresponding to a 1 percent creased five times faster. Income per capita in Cote change in the indicated v.uiable. d'lvoire was nearly twice Egypt's in 1960 but then grew Soura: Lau and others, b.Kkground paper. 38 and its ability to acquire and apply new scientific the more likely its people are to live long and knowledge depends on the level of schooling in healthy lives. Of course, this effect tapers off as the population. income rises: a doubling of income per capita (ad- justed for purchasing power parity) from, say, The influence of income on health $1,000 in 1990 corresponds to a gain of eleven The higher a country's average income per capita, years in life expectancy, whereas a doubling from Box figure 2.1 Gains in child health, 1960-87, and share contributed by various factors Costa Rica Cote d'Ivoire 5 Egypt 3 Japan 9 3 • Initial levels of schooling and income per capita 50 1!1 Increase in schooling • Increase in income per capita D Technicill progress Note: The <1rea of the circle is proportional to the absolute increase, over the period 1960-87, in the probability of surviving to age 5. For Cote d'Ivoire ch11 nges for the period are extrapolated from the observed change during 1960-80. Source: Lau and others, background paper. 39 Within the same city, health status is worse in poorer people in poverty matter as well. In industrial areas. countries life expectancy depends much more on income distribution than on income per capita, and it has been rising faster in countries with im- Figure 2.2 Child mortality in rich and poor proving income distribution. Japan and the United neighborhoods in selected metropolitan Kingdom had similar income distributions and life areas, late 1980s expectancies in 1970, but they have diverged since then. Japan now has the highest life expectancy in 1!!11 Poor neighborhoods the world and a highly egalitarian income distribu- 0 Rich neighborhoods tion. In the United Kingdom, where income dis- parity has widened since the mid-1980s, life expec- tancy is now more than three years shorter than in Percentage deviation from national mean Japan. 0.50 In developing countries the number of people in poverty is an especially important reason for dif- ferences in health. One study looked at twenty- 0.25 two developing countries with comparable data on poverty (defined as the share of the population consuming less than $1 a day at 1985 purchasing power parity prices) and found that variation in 0 the prevalence of poverty and in per capita public spending on health is important in explaining cross-country variation in life expectancy. Differ- -0.25 ences in income per capita became unimportant once those two factors were taken into account. This does not mean that income growth is irrele- vant to increased life expectancy; rather, its main -0.50 effect lies in how much it reduces poverty and Cairo, Egypt Colombo, Lima, Peru supports public health services. In the twenty-two Sri Lanka countries, roughly one-third of the effect of eco- nomic growth on life expectancy came through Note: Child mortality indexes for each neighborhood are poverty reduction and the remaining two-thirds calculated by dividing the observed number of deaths through increased public spending on health. In among children of women in the sampled households of a neighborhood by the expected number (given the Sri Lanka an increase in per capita public spending distrioution of women by the length of time they have on health was twenty-two times more effective in been bearing children and the national average child mortality levels at each duration of childoearing); reducing infant mortality than was the same in- Percentage deviations from the national average are crease in average income. obtained by subtracting 1 from a neighborhood's index and multiplying the result by 100. Neighborhoods in Within countries, too, health correlates strongly each city were ranked according to the proportion of with poverty. In India, Indonesia, and Kenya child houses with concrete floors. Poor neighborhoods were , the lowest 25 percent in this ranking; rich.neighborhoods' ' mortality is higher in states or provinces with were the top 23 percent. ,, larger proportions of poor people. Within cities, Source: Calculated from data from national Demographic there are large differences in child survival be- and Health Surveys. tween rich and poor neighborhoods (Figure 2.2). And children in poor families are less healthy. In Madurai, the second largest city in India's Tamil $4,000 is matched by a gain of only four years (see Nadu State, children ages 2-9 in the poorest Figure 1.9 in Chapter 1). Income growth has more. households were more than twice as likely to suf- impact in poor populations because additional re- fer from serious physical or mental disabilities as sources buy basic necessities, particularly food children from slightly better-off families. and shelter, that yield especially large health Poor people are vulnerable to disease not only benefits. because of poor living conditions but often also for Because poverty has a powerful influence on work-related reasons. In Adana, Turkey, the risk health, it is not just income per capita that is rele- of malaria is significantly greater among migrant vant; the distribution of income and the number of workers than for the local population; the average 40 number of anopheline mosquito bites per person Child mortality falls faster in countries where income was five times greater in the tents of these workers per capita is growing rapidly. than in the houses of village residents. In Sri Lanka one of the commonest causes of pesticide poisoning is leaky knapsack sprayers; surveys Figure 2.3 Declines in child mortality show that although farmers are aware of the risks and growth of income per capita involved, they continue to use broken equipment in sixty-five countries because they cannot afford to replace or repair it. The distribution of income within households Annual rate of change in child also affects health. Increasing women's access to mortality, 1970-88 (percent) income can be especially beneficial for the health 1 of children. In Brazil income in the hands of the 0 r• • ~-+----------------------------__, mother has a bigger effect on family health than income controlled by the father. In Jamaica house- • • -2 • holds headed by women eat more nutritious food •• • •• • than those headed by men; they also spend more of their income on child-centered goods and signif- • • -4 • • icantly less on alcohol. In Cote d'Ivoire a doubling of household income under women's control re- •• -6 duces the share of alcohol in the family budget by • • 26 percent and the share of cigarettes by 14 per- • ••• • • cent. In Guatemala it takes fifteen times more -8 • • spending to achieve a given improvement in child • nutrition when income is earned by the father than -10 • -1 0 2 3 4 5 6 7 when it is earned by the mother. Although a work- Annual income per capita growth ing mother may breastfeed less and have less time rate, 1960-88 (percent) for child care-both of which could be detrimental to her children's health-evidence from numerous developing countries suggests that this harm can Note: Child mortality refers to the probability of dying be offset by the health benefits that her earnings between birth and age 5; the period over which the rate of- mortality decline is averaged differs from the period bring. used for income per capita growth to take account of lags Because fewer people live in poverty as average in the relation between the two rates. Source: World Bank data. incomes rise, there is generally a strong link be- tween incomes and health status. Across coun- tries, more than 75 percent of the difference in health is associated with income differences. In- deed, this relation is not merely associative but countries and by as much as 4 to 8 percent in Cote causal and structural: income growth leads di- d'Ivoire and Ghana. rectly to better health. In a sample of fifty-eight These findings highlight the costs to health of developing countries, a 10 percent increase in in- slow economic growth. Child health has been im- come per capita, all else being equal, reduced in- proving everywhere, but gains are much less rapid fant and child mortality rates by between 2.0 and in countries with slow income growth (Figure 2.3). 3.5 percent and increased life expectancy by a During the 1980s the economic performance of de- month. This estimate reflects the total impact of veloping countries was mixed, with income per income on health; it includes effects working di- capita constant or falling, and in some countries rectly through income (such as food consump- the incidence of poverty rose (Table 2.1). Had eco- tion), as well as indirectly through factors that an; nomic growth been as fast in the 1980s as in the themselves mainly determined by income (access period between 1960 and 1980, in 1990 alone an to safe water and sanitation, availability of physi- estimated 350,000 infant deaths, or 6 percent of cians, and so on). Studies based on individual total infant deaths, would have been averted in households corroborate the cross-country results. developing countries (excluding India and China). A 10 percent advantage in income reduces infant In Africa and Latin America, where average mortality by between 1 and 2 percent in Nigeria, growth was 2.5 percentage points slower during Sri Lanka, Thailand, and several Latin American the 1980s, the saving in babies' lives in 1990 would 41 Table 2.1 Poverty and growth of income per Bangladesh, for example, found that over a period capita by developing region, 1985 and 1990, of two years following the death of a mother, mor- and long- and medium-term trends tality rates, in comparison with those of children Annual percentage with living mothers, were twice as high for boys Head-count index change in income ofpovertya per capita and three times as high for girls. Region 1985 1990 1970-92 1982-92 Education greatly strengthens women's ability All developing to perform their vital role in creating healthy countries 30.5 29.7 1.7 0.8 households. It increases their ability to benefit Sub-Saharan from health information and to make good use of Africa 47.6 47.8 -0.2 c-1.1 health services; it increases their access to income East Asia 13.2 11.3 5.3 6.3 and enables them to live healthier lives. It is not South Asia 51.8 49.0 2.0 3.0 surprising, therefore, that a child's health is af- Eastern Europe 7.1 7.1 1.2 1.7 fected much more by the mother's schooling than Middle East and North Africa 30.6 33.1 0.1 -1.6 by the father's schooling. Data for thirteen African Latin America countries between 1975 and 1985 show that a 10 and the percent increase in female literacy rates reduced Caribbean 22.4 25.2 1.1 -0.2 child mortality by 10 percent, whereas changes in Note: Regional data on annual change in income per capita refer to male literacy had little influence. Demographic unweighted country averages. The regions used in this table are as defined in the World Development Indicators, except for Eastern and Health Surveys in twenty-five developing Europe, which includes Albania, Bulgaria, Hungary, Poland, Ro- countries show that, all else being equal, even one mania, the former Czechoslovakia, and the former Socialist Federal Republic of Yugoslavia. Disaggregated data for the last two are not to three years of maternal schooling reduces child yet available. mortality by about 15 percent, whereas a similar a. Estimated share of the population consuming less than $32 per person per month at 1985 purchasing power parity prices. level of paternal schooling achieves a 6 percent Source: For poverty index, World Bank 1993c; for change in income reduction. The effects increase when mothers have per capita, World Bank data. had more education; in Peru, for example, seven or more years of maternal schooling reduces the have been as much as 7 and 12 percent, respec- mortality risks nearly 75 percent, or about 28 per- tively. Latin America's recession in 1983 is esti- cent more than the reduction for the same level of mated to have caused 12,000 additional deaths of paternal schooling (Figure 2.4). Countries that in babies, or 2 percent of all infant deaths in that year. 1965 had achieved near-universal enrollment for And because slow economic growth hampers pov- boys but much less for girls had about twice the erty reduction and constrains spending on health, infant mortality in 1985 of countries with a smaller schooling, and other services, it is highly likely boy-girl gap. that the health of the poor suffered dispropor- The advantages that a mother's schooling con- tionately in the 1980s. fers on her children's health are felt even before birth. In developing countries better-educated The influence of schooling on health women marry and start their families later, dimin- ishing the risk to child health associated with early Households with more education enjoy better pregnancies. Educated women also tend to make health, both for adults and for children. This result greater use of prenatal care and delivery assis- is strikingly consistent in a great number of tance. In a study in Lima that controlled for service studies, despite differences in resec:rch methods, availability and socioeconomic status, 82 percent time periods, and population samples. of women with six or more years of education sought prenatal care, compared with only 62 per- MATERNAL SCHOOLING AND CHILD HEALTH. In cent for women with no education. most households women have the main responsi- Following birth, the children of educated bility for a broad range of activities that affest mothers continue to enjoy other health-enhancing health. They manage household chores, keep the advantages: better domestic hygiene, which re- house clean, process foods and prepare meals, duces the risk of infection; better food and more feed and care for young children, and look after immunization, both of which reduce susceptibility the sick. Women's own health and their efficiency to infection; and wiser use of medical services. A in using available resources have an important study of women in Bangladesh documented how bearing on the health of others in the family, par- educated women kept their homes and children ticularly children. A study of children under 10 in tidier and cleaner than uneducated women and 42 Child health depends more on the mother's than on the father's schooling. Figure 2.4 Effect of parents' schooling on the risk of death by age 2 in selected countries, late 1980s 4-6 years of schooling 7 or more years of schooling Percentage reduction in child mortality Percentage reduction in child mortality (in relation to parents with no schooling) (in relation to parents with no schooling) 80 60 40 20 0 0 20 40 60 80 Indonesia Kenya Morocco Peru • Mother Ell Father Source: Hobcraft 1993. expressed a preference for water from tanks or example, 24 percent of the children of mothers tubewells at home rather than from canals or with no education were stunted, compared with rivers. In Brazil, India, and Nigeria better-edu- only 11 percent of children of mothers with some cated households are willing to pay 6 to 50 percent elementary schooling. Educated women are an im- more than other households for improved water portant part of the reason for the impressive supplies. health achievements of China, Costa Rica, India's Educated mothers are also better at getting in- Kerala State, and Sri Lanka, despite relatively low formation on health and acting on it. In Brazil the incomes. child health benefits of a father's education work mostly through his income, whereas almost all the SCHOOLING AND ADULT HEALTH. Personal habits effect of maternal education comes from learning and life-style choices affect adult health enor- about health through newspapers, television, and mously. Because educated people tend to make radio. In Thailand mothers with primary educa- choices that are better for their health, there is of- tion were 30 percent more likely than mothers ten a strong relation between schooling and with no education to treat childhood diarrhea with health. A study of U.S. life expectancy at age 25 oral rehydration therapy or a homemade solution found that between the highest and the lowest of salt and sugar; this figure rose to 90 percent for levels of education, the difference was about six mothers with secondary or higher education. Sim- years for white men and about five years for white ilar results have been reported in countries as di- women. These differences-which may partly re- verse as Burundi, Colombia, Ghana, Morocco, and flect differences in income associated with educa- Nigeria. And well-educated mothers often manage tion-have persisted since the 1960s. to reduce the damage that poverty does to health. The same pattern occurs in developing coun- Among poor rural households in Cote d'Ivoire, for tries. Surveys in Cote d'Ivoire, Ghana, Pakistan, 43 Schooling reduces the risk of adult ill health. ample, when the AIDS epidemic began, infection was initially concentrated among well-educated elites, but these same groups were the first to Figure 2.5 Schooling and risk factors for change their life-styles as information became adult health in Porto Alegre, Brazil, 1987 available about the disease and its prevention. In Brazil between 1982 and 1985, 79 percent of those infected had completed postsecondary education; Prevalence of risk (percent) by the late 1980s this group's share of cases had 100 fallen to 33 percent. Even more striking is the way that well-educated people have changed their be- havior on smoking. In the United Kingdom the proportion of smokers among adults declined be- tween 1958 and 1975 by 50 percent among the most educated but hardly changed among the least educated. In the United States between 1974 and 1987 the smoking habit declined nine times faster in the highest education group than in the lowest. The corresponding difference was twofold in Canada and threefold in Norway. Policies to strengthen household capacity Because people's ability to improve their own D No schooling health depends so much on income and educa- • Postsecondary schooling tion, the policy conclusions are clear: governments should work to boost economic growth, reduce poverty, expand schooling (particularly for girls), Source: Achutti and others 1988. and help strengthen women's ability to care for their families. This section deals with each of these points in turn. and Peru show that respondents whose parents Promoting growth and reducing poverty were educated were more likely to have living par- ents than those with uneducated parents. In Peru During the 1980s the pattern of economic growth 72 percent of the educated fathers of respondents in developing countries was very uneven. Income ages 25-29 were still alive at the time of the survey, per capita grew at more than 6 percent a year in compared with only 55 percent of the uneducated East Asia but remained constant or fell in many fathers. In Jamaica education had a bigger influ- other countries. The disappointing record re- ence on adult health than did income, particularly flected the impact of adverse external shocks as before age 50. Death rates for specific diseases also well as poor domestic policies. Nonetheless, some show educational differentials. In Russia death economies grew rapidly despite the external from coronary heart disease was two to three shocks, showing that a great deal can be done by times more common for the poorly educated than developing countries themselves. for those with higher education. In Brazil those Because it is difficult to reduce poverty without who were illiterate or who had only primary economic growth, establishing sound economic schooling were about five times more likely to policies for growth is one of the most valuable have high blood pressure than those with post- things a government can do. Development strate- secondary schooling. The first group was also sub- gies also need to emphasize broadly based growth stantially more inclined to obesity, alcohol and to- to give the poor better income-earning oppor- bacco consumption, and lack of exercise (Figure tunities and better access to a range of social ser- 2.5). vices. To protect the most vulnerable members of The advantages of education continue to show society, it is appropriate for governments to make up when new types of health risk appear. For ex- transfers and other special arrangements. 44 POLICY REFORM AND ADJUSTMENT LENDING. As a equal, middle-income countries in the "intensive" consequence of the economic crisis of the early group boosted their growth rates during 1986-90 1980s, many developing countries changed their by an estimated average of about 4 percentage economic policies. They adopted macroeconomic points a year over what would probably otherwise reforms intended to achieve price stability and sus- have occurred. The low-income countries, espe- tainable internal and external monetary balance cially in Sub-Saharan Africa, did less well; for and made microeconomic and institutional re- them, the benefit was 2 percentage points. forms to promote the efficient use of resources and Since health is helped by economic recovery and faster economic growth. These changes typically faster long-term growth, adjustment lending, by involved cuts in public spending, the opening of facilitating economic progress, benefits health in the economy to competition, liberalization of the long run. When a government has to adjust- prices, measures to improve the efficiency of pub- in response to economic shocks or to rectify mis- lic expenditure, and the development of a sound taken past policies-the whole society, poor and financial system and other institutions needed in a nonpoor, may suffer short-run reductions in em- well-functioning market economy. ployment and wages. But the resulting fall in in- To support these reforms, the World Bank and come is caused not so much by policies associated the International Monetary Fund have extended with adjustment lending as by the necessity for the adjustment lending. The purpose of this lending is country to curb its consumption; without adjust- to cushion an economy during the transitional ment loans, even greater decreases in consump- phase to its new growth path. Adjustment lending tion would probably have been necessary. None- is therefore essentially an investment in a more theless, adjustment lending can take five or more productive future. It has been central to the re- years to bear fruit, and the transition can be pain- forms in Latin America and Sub-Saharan Africa ful because incomes may fall in the short run. Evi- and important in other regions as well. Its role will dence from Sub-Saharan Africa and Latin America continue in the 1990s: it is already a major channel suggests that economic downturns are associated of assistance for the formerly socialist economies; _ with less favorable child mortality outcomes than it is being used for the first time in India; and it has would be predicted from long-term trends. In both old and new clients in other parts of the countries where child mortality rates are declining world. over time, for example, adjustment lending would Nonetheless, adjustment lending remains con- be associated, in the short run, with a slower rate troversial. Does it really raise long-term growth? of decline. To minimize such adverse effects, some Do the poor suffer as a consequence of such ad- countries have begun to use resources, including justment policies as cuts in public spending and adjustment loans, to support nutrition programs liberalization of food and other prices? How is for vulnerable children, as well as basic health and health affected? The answers to these questions other social services targeted to the poor. are complicated because adjustment lending is nei- ther necessary nor sufficient for policy reform. ADJUSTMENT LENDING AND PUBLIC EXPENDITURE Some of the most dramatic "adjustment" reforms ON HEALTH. Because cuts in government spending took place without adjustment lending (as in Chile are usually central to an adjustment program, and Viet Nam), and some countries that received health spending is likely to be reduced. In many adjustment loans did little or nothing to pursue countries early cuts were indiscriminate and failed reforms (for example, Tanzania and Zambia). In to preserve those elements of the health system addition, because a country's economic perfor- with the strongest long-term benefits for health. mance is affected by many factors, it is hard to Drugs were often cut more heavily than personnel isolate the part played by adjustment lending. because it is difficult to lay off public employees. Despite these difficulties, World Bank studies on Cote d'Ivoire's experience illustrates the mistakes the impact of adjustment lending are reveali!lg. that occurred in some early programs of economic The research looked at countries in the "intensive adjustment. With real income per capita falling 19 adjustment lending" group (which includes coun- percent between 1980 and 1984, the government tries that received at least two structural adjust- cut public spending, among other measures. ment loans or three sectoral adjustment loans by Health expenditure dropped in real terms by 12 1990, with the first loan started by mid-1986) and percent between 1981 and 1984. But personnel found that in general they did achieve faster costs were not cut; instead public expenditures on growth than in other countries. All else being medicines and materials absorbed the reduction, 45 Public spending on health recovered faster in countries that received adjustment lending. Figure 2.6 Deviation from mean levels of public spending on health in countries receiving and not receiving adjustment lending, 1980-90 Deviation in share of country income spent Deviation in per capita spending (percentage points) (1991 dollars) 0.35 8 0.30 6 0.20 I\ I . 1. NonreCipients I 4 \ Recipients _.... ~' - / .\ [.)\ ~ Nonrecipients I "" I 0.10 2 \ 1\ I 0 Recipients 0 \ ......... -I - -0.10 -0.20 1980 81 82 " ~ I 83 84 ' 85 1 / 86 87 """" -v88 89 v 1-- 90 -2 -4 -6 " "" \/ / / 1980 81 82 83 84 85 86 87 88 89 90 Nole: Recipients of adjustment lending include countries that received two structural adjustment loans or three or more adjustment operations, all effecti ve by June 1990, with the first operation effective in or before June 1986. Data were available for the following countries in this group: Bolivia, Brazil, Chile, Costa Rica, Kenya, Republic of Korea, Mauritius, Mexico, Pakistan, Philippines, Turkey, and Uruguay. Nonrecipients are countries that had not received adjustment lending by June 1990. Data were availabll' for Burkina Faso, Dominic,m Republic, Egypt, El Salvador, Guatemala, Liberia, Malaysia, and Papua New Guinea . Source: Yazbeck, Ta n, <1lld Tanzi, background paper. shrinking in real terms by more than one-third ticularly in rural areas, have been improving as a during the first half of the 1980s. In rural clinics, result . already precarious supplies of basic consumables Various studies have assessed the effect of ad- became even scarcer. justment programs on public spending on health. The implications for child health looked grim. Most have found that central government expen- Cross-sectional data show that the nutritional sta- diture on health in countries with adjustment tus of Ivorian children is strongly related to the lending programs did not suffer more than else- availability of drugs in the community. All else be- where; this result, however, is not definitive be- ing the same, the difference in height-for-age (a cause state and local governments are often re- measure of long-run nutritional status) of children sponsible for a substantial share of public in communities lacking basic medicines and those spending on health . More comprehensive data in well-supplied communities was equivalent to available for twenty countries during 1980-90 more than one-third of the difference between the show that in both countries with and without ad- average child in Cote d'lvoire and in the United justment loans, public spending on health as a States. The health of children from poor families percentage of total country income declined in the suffers even more when drugs are unavailable. early 1980s in relation to the average for the dec- Since 1990 the government has begun putting ade. In 1985-90, however, health spending recov- more resources into nonwage health inputs: their ered much faster in countries with adjustment pro- share of the health budget rose from 20 percent in grams . Similarly, per capita public spending on 1991 to 24 percent in 1993. Health services, par- health also recovered faster in such countries (Fig- 46 ure 2.6). Unfortunately, the data are not good they complete even the first few years of basic edu- enough to allow any judgment on whether adjust- cation. Fewer than 60 percent of first-graders in ment programs directly helped to ensure that pub- the lowest-income countries and about 70 percent lic spending on health was efficient. (And, as this of those in the lower-middle-income countries Report will show, not all health spending deserves reach the last year of primary school. to be protected; some of it is inefficient and Enrollments are particularly low in isolated rural regressive.) areas, for lower socioeconomic groups, and for girls. In developing countries as a group, about 10 Expanding and improving schooling percent of boys ages 6-11 do not enroll; for girls in the same age group the figure is 40 percent. Espe- In general, developing countries have made much cially in poor countries, the gaps can be substan- progress in expanding schooling since the 1960s, tial, as Figure 2.7 illustrates for India. But Sri but the trends conceal some shortcomings. In the Lanka's experience shows that this gender gap is poorest countries, especially in Africa, many chil- not an inevitable consequence of poverty. dren never go to school at all. In Mali, for example, Leaving aside the gaps in enrollment, education fully 77 percent of all school-age children never go in many countries is inadequate. Even children to school-a figure that has remained largely un- who complete primary school fail to acquire basic changed since 1980. Of those who do go to school, literacy and numeracy skills and scientific under- many often enroll late-thus missing the benefits standing. These weaknesses in the education sys- of early learning opportunities-and leave before tem reduce the potential impact of schooling on Substantial male-female gaps in schooling persist in some low-income countries. ' Figure 2.7 Enrollment ratios in India, by grade, about 1980 Enrollment ratio (percent) 100 1 2 3 4 5 6 7 8 9 10 11 12 Primary grades Middle grades Secondary grades fill Boys • Girls Note: The enrollment ratio is the share of children in the age group corresponding to a given grade who are enrolled in that grade. Source: Tan 1md Mingat 1992. 47 Box 2.2 Teaching schoolchildren about health: radio instruction in Bolivia Many personal habits and life-style choices that have basic health concepts and practices at a young age are important consequences for health are formed early in more likely to maintain them as parents. life. Health education in schools can help young people It is still too early to assess the long-term health im- make informed choices and so reinforce the effect of pact of teaching health lessons through radio. None- schooling on health. theless, pupils already show significant gains in several Bolivia has had success with health education areas, including ability to recognize symptoms of dehy- through radio lessons. Radio instruction was first intro- dration and knowledge of the proper mixture and ap- duced in 1987 for teaching mathematics and proved to plication of homemade oral rehydration solution. be both inexpensive (with costs per pupil averaging There is also evidence of increased hand-washing, and less than $1 a year) and effective. In 1989 the health and more households are using simple water filters . education ministries began to ·try out the use of radio The radio program is now being expanded. In 1993 for teaching disease prevention to schoolchildren, more than 1,000 third- and fourth-grade classrooms starting with a module on diarrhea prevention and oral will receive broadcasts of a new curriculum that in- rehydration. Children ages 8-13 were targeted because cludes lessons on cholera, personal and dental hy- they often take care of younger siblings and perform giene, acute respiratory infections, immunizations, in- household chores involving food preparation and sani- fectious diseases, and accident prevention. Nutrition, tation . They thus have a strong influence on their own environmental health, and self-esteem are to be added health and that of younger siblings. The radio health in 1994. In response to parents' requests, a comple- program emphasizes actions that a child can do for ·mentary community-based radio program is also being himself or herself or can do for or teach to a younger developed and tested. sibling. It rests on the belief that children who learn health. More important, they also reduce parents' - than a third of girls reach this level, and in China willingness to enroll their children, thus perpetu- and Latin America only 60 percent do . Achieving ating a vicious cycle of poor schooling and poor 5,000 hours of schooling for all children will thus health. In India, for example, more than 40 percent require significant and sustained policy effort in of parents in a nationwide survey cited either "not large parts of the world. To reinforce the effects of interested in education/further study" or "fail- school expansion on health, it may be useful to ure'' as the main reason for not sending their chil- include health topics in school curricula (see Box dren to school. 2.2). Much more needs to be done to extend educa- Incomplete enrollments reflect the combined in- tion in developing countries. Government support fluence of weak demand for education (which is for schooling at the lower levels and for girls is partly caused by low achievement) and inadequate especially justified: the benefits for society are schooling opportunities. To overcome these obsta- large, and poor families in low-income countries cles requires a combination of policies . Govern- typically undervalue the benefits of sending chil- ments can do more to ensure that lower levels of dren to school or are unaware of them. In addi- schooling receive priority in the allocation of pub- tion, for such families the opportunity costs of lic spending. In some countries, current levels of sending children to school are often high. A policy resources for primary schooling are insufficient to priority is to ensure that every child receives a support even minimal conditions for instruction. minimum quantity of schooling-say, 5,000 hours, In India public spending per primary pupil in rela- or roughly six full years of schooling. This would tion to income per capita averages only one-third be consistent with the aims of the 1990 World Con- that in Korea, basically because much more of In- ference on Education for All, sponsored by the dia's public expenditure on education goes to United Nations Educational, Scientific, and Cul: higher education. In Burkina Faso, Mali, Mauri- tural Organization (UNESCO), UNICEF, the tania, and Niger one-quarter of the education bud- UNDP, and the World Bank. Most pupils in devel- get is for higher education, and between 60 and 80 oping countries currently receive much less than percent of that quarter is devoted to scholarships 5,000 hours of schooling in the primary grades (be- and other forms of student aid. This, it can be cause of pupil and teacher absences caused by argued, is inefficient, and it is also extremely re- sickness, among other factors). In India no more gressive because most of the benefits of higher ed- 48 ucation are captured privately in the form of in- and that of their families. Removing discrimina- creased earnings and because students tend to tion-in the labor market, in access to credit, in come from higher-income families. property law, and so on-can boost women's On its own, spending more resources for pri- earnings and financial security, which (as an ear- mary schools is not enough. Whatever is spent lier section has shown) can promote family health. must also be used efficiently. And women need to be healthy themselves to ful- Although the health and nutrition of school-age fil their roles as mothers and household managers. children are not normally thought of as education They have specific health needs, including protec- issues, in fact they do affect a child's school atten- tion against violence. More than one-third of the dance and performance. Allocating resources to global burden of disease for women ages 15-44, address health problems in this population can of- and over one-fifth of that for women ages 45-59, is ten be an efficient way to improve schooling. (Spe- caused by conditions that afflict women exclu- cific interventions are discussed in Chapter 4.) sively (maternal mortality and morbidity and cer- Teachers and pedagogical materials are the main vical cancer) or predominantly (anemia, sexually school inputs at lower levels of schooling (with transmitted diseases, osteoarthritis, and breast teacher salaries absorbing the bulk of spending). cancer). Most of these problems can be addressed Efficient use of these inputs is thus crucial, par- cost-effectively, but health services in many devel- ticularly in countries where rapid population oping countries have typically focused narrowly growth threatens to reverse progress in expanding on women as mothers. enrollments. In Mauritania, for example, if spend- What is lacking is a strategy for engaging ing per pupil and the share of primary schooling in women in health care from adolescence onward. total government expenditure remain constant, Often this failing occurs because health services the enrollment ratio in primary schools is pro- are insensitive to the cultural needs of women: in jected to drop from 51 percent in 1988 to 45 percent many Middle Eastern countries, for example, most by 2000. To forestall such regress, maximizing the physicians are men, but there is a strong belief that learning gain per unit cost and making the correct women should not be seen after puberty by men tradeoffs between unit costs and coverage are of who are not part of their family. Inconvenience is particular importance. A recent World Bank-spon- another deterrent; in many countries individual sored review found expenditures on nonsalary in- health services (for example, prenatal care and im- puts such as textbooks and interactive radio in- munizations) are offered on different days, mean- struction to be most cost-effective in improving ing that women have to return repeatedly with learning outcomes. their children. The solution is often as feasible as it In some circumstances it may also be right to is clear: to provide child health services, prenatal spend more to lower the barriers to schooling for care, treatment of sexually transmitted diseases, girls and other disadvantaged groups. This can be and family planning services jointly at convenient done in many ways: through scholarships (used in times. The Bangladesh Women's Health Coalition Bangladesh to encourage girls to go to secondary and the Chilean Institute of Reproductive Medi- school); by offering free textbooks or fee-exemp- cine, for example, offer integrated family planning tions; or by siting schools close to pupils' homes so services at the same time as child health services, that parents are less worried about their daugh- and Thailand is experimenting with mobile health ters' safety. In Pakistan, for example, girls are as clinics to reach women in their homes. likely to enroll as boys when there is a school in The design of health services must also be sensi- the village but are 10 percent less likely to do so tive to the stigmas surrounding certain diseases, when the school is nearby but not in the village. In especially any that are sexually transmitted or several African countries distance education- physically disfiguring. Women are more likely to whereby radio and correspondence materials re- seek treatment for sexually transmitted diseases if place classroom teachers as the principal medium health centers offer multiple services, with privacy of instruction-has sometimes helped to overcome in consultations, so that it is not obvious why a the physical barriers to schooling for girls. person is visiting the center. Diseases that damage the skin (such as leprosy, onchocerciasis, and Empowering women leishmaniasis) have severe psychological implica- tions for girls and women, reduce their marriage In addition to education, other policies can en- prospects, and may lead to marital separation. In hance women's capacity to improve their health Colombia and India women tend to seek treatment 49 Box 2.3 Violence against women as a health issue Data from many industrial and developing countries woman's age ()r unhealthy habits (such ·as smoking). In reveal that anywhere between one-fifth and more than addition to physical injury and emotional trauma, rape half of women surveyed say they have been beaten by Yictims run the risk of becoming pregnant or contract- their partners. Often, this abuse is systematic and dev- ing sexually transmitted diseases, including AIDS. A astating. In Papua New Guinea, for example, 18 per- rape crisis center in Bangkok reports that 10 percent of cent· of all urban wives surveyed had sought hospital its clients contract STDs as a result of rape and 15 to 18 treatment for injJJries inflicted by their husbands . In percent become pregnant, a figure consistent with data the United States domestic violence is the leading from Korea and Mexico. In countries where abortion is cause of injury among women of reproductive age; be- restricted or illegal, rape victims often resort to unsafe tween 22 and 35 percent of women who visit emer- abortions, greatly increasing the danger of infertility or gency rooms are there for that reason. even death. Research has shown that battered women run twice Another form of violence against women and girls is the risk of miscarriage and four times the risk of having fe!Jla)e genital mutilation, popularly known as female a baby that is below average weight. In some places circumcision. An estimated 85 million to 114 million violence also accounts for a sizable portion of maternal women in the world today have experienced genital deaths. In Matlab Thana, Bangladesh, for example, in- mutilation. The practice. is reported in twenty-six Afri- tentional injury during pregnancy-'-motivated by can countries, among minorities in India, Malaysia, dowry disputes or shame over a rape or a pregnancy and Yemen, and among some immigrant populations outside wedlock- caused 6 percent of all ~aternal in Western countries. If current trends continue, more deaths between 1976 and 1986. Research from the than 2 million girls will be at risk of genital mutilation United States indicates that battered women ;~re four to every y.e ar, five times as likely to require psychiatric treatment as Clitoridectomies account for 80 to 85 percent of nonbattered women and are five times as likely to at- cases worldwide. Infibulation, which involves removal tempt suicide. They are also more prone to alcohol of more tissue, is more common in eastern Africa. abuse, drug dependence, chronic pain, and These initiation rituals pose a health risk to girls and depression . women and are a threat to their psychological, sexual, Rape and sexual abuse also damage women' s health and reproductive well-being . The consequences of and are widespread in all regions, classes, and cul- both procedures can include hemorrhage, tetanus, in- tures. In Seoul 17 percent of women report being vic- fection, urine retention, and shock. Infibulation carries tims of attempted or actual rape. In one study of U.S: the added risk of long-term complications because of women a history of rape or assault was a stronger pre- the repeated cutting and stitching at marriage and with dictor of how many times women sought medical help each childbirth, and it can limit a woman's choice of and of the severity of their health problems than was a contraceptive method . for leprosy later than men do, when patches have cent of the total disease burden among women already reached the face and hands; they are reluc- ages 15-44 in developing countries, where the bur- tant to ask for help when the first patches appear, den from maternal and communicable causes still on the buttocks. Again, sensitivity is needed to overwhelms that from other conditions . In indus- encourage women to come forward. trial countries, where the total disease burden is The same is true of another category of danger much smaller, this share rises to 19 percent. By to women's health: domestic violence and rape. damaging a woman's physical, mental and emo- Violence against women is widespread in all coun- tional capacity to care for her family, domestic vio- tries in which it has been studied (see Box 2.3). lence and rape also hurt the health of other family Although this has only recently been viewed as a members, particularly young children. public health issue, it is a significant cause of fe- This is an issue with complex economic, cul- male morbidity and mortality, leading to psycho- tural, and legal roots, and it is therefore not easily logical trauma and depression, injuries, sexually dealt with by public policies. Prevention will re- transmitted diseases, suicide, and murder. Rape quire a coordinated response on many fronts . In and domestic violence cause a substantial and the short to medium term, the right measures in- roughly comparable level of disease burden per clude training health workers to recognize abuse, capita to women in developing and industrial expanding treatment and counseling services, and countries . These problems account for about 5 per- enacting and enforcing laws against battering and 50 rape. In the long term, much depends on changing achievements of the past point to the requirements cultural beliefs and attitudes toward violence of the future-above all, to economic growth and against women. In Africa women's groups have the expansion of schooling and health services. worked to break the practice of female circumci- According to World Bank projections, income per sion, partly by informing people of its severe con- capita in Sub-Saharan African countries will grow sequences for health. In the United States the by only 0.8 percent a year over the next ten years. American Medical Association launched a major Even this modest increase will bring about a de- campaign in 1991 to educate the public and physi- cline in the infant mortality rate of between 2 and 4 cians about family violence. Research shows that percent. In South Asia, where faster growth-3.3 even health professionals often fail to identify percent a year-is projected, infant mortality de- cases of battering. Recently, the U.S . Joint Com- clines of 15 percent can be expected. mission on Hospital Accreditation issued new These benefits can be powerfully reinforced by standards requiring all hospitals to develop proto- better education and health services. In Africa in- cols and train their staffs to respond to different creasing female literacy rates by 10 percent is likely forms of abuse. In Colombia the Ministry of to lower the infant mortality rate by an estimated Health has begun to document the scale of the 10 percent. In India and Kenya two maternal problem in its most recent Demographic and deaths and about forty-five infant deaths would be Health Survey. These efforts come on the heels of averted for every 1,000 girls provided with one almost two decades of organizing efforts by extra year of primary schooling . Even in poor women around the world; in Latin America alone countries governments can enhance people's abil- there are now nearly 400 separate organizations ity to improve their own health by expanding working to reduce violence against women. schooling opportunities for all children-with spe- cial efforts to encourage parents to enroll their What can be done? daughters-and by widening access to health ser- vices, particularly for women and children. Such Around the world, much has already been done to investments pay off in better health and provide a enable people to live longer, healthier lives. The foundation for future economic growth. 51 The roles of the government and the market in health World spending on health totaled about $1,700 bil- tion, and other sectors important for health, as lion in 1990, or 8 percent of global income. Of this, well as through regulation of health systems, governments spent more than $1,000 billion, or health providers, and insurers. Governments fur- nearly 60 percent. Of the $170 billion spent on ther affect health by their impact on household health in the developing countries of Africa, Asia, income and educational levels (as discussed in and Latin America, governments spent half the Chapter 2), by financing public health services, total amount-2 percent of those regions' GNP. In and by providing care directly. What governments the established market economies, where total . do varies enormously from country to country, but health spending was almost $1,500 billion, govern- every government plays an important role. ments spent just over $900 billion-more than 5 Three economic rationales justify and guide a percent of GNP (Table 3.1) . The sheer size of these government role in health . They are discussed in expenditures on health makes it critical to under- greater detail in ''The rationales for government stand the impact of government policies on peo- action,'' below. ple's health. But governments profoundly influ- • The poor cannot always afford health care ence health in less direct ways, through their that would improve their productivity and well- policies toward education, water supply, sanita- being. Publicly financed investment in the health Table 3.1 Global health expenditure, 1990 Public sector health Total health Health expenditure Percentage Per capita Ratio of Percentage expenditure expenditure as percentage of GNP health per capita of world (billions of as percentage of regional spent on expenditure spending Demographic region population dollars) of world total total health (dollars) (SSA ~ 1) Established market economies 15 1,483 87 60 9.2 1,860 78.9 Formerly socialist economies of Europe 7 49 3 71 3.6 142 6.0 Latin America 8 47 3 60 4.0 105 4.5 Middle Eastern crescent 10 39 2 58 4.1 77 3.3 Other Asia and islands 13 42 2 39 4.5 61 2.6 India 16 18 22 6.0 21 0.9 China 22 13 59 3.5 11 0.5 Sub-Saharan Africa 10 12 55 4.5 24 1.0 Demographically developing countries 78 170 10 50 4.7 41 1.7 World 100 1,702 100 60 8.0 329 13 7 Note: SSA, Sub-Saharan Africa. Source: Appendix table A.9. 52 of the poor can reduce poverty or alleviate its education bring. The second factor is the amount consequences. and effectiveness of expenditure in the health sys- • Some actions that promote health are pure tem. The third factor is the range of diseases pre- public goods or create large positive externalities. sent, which is determined largely by climate and Private markets would not produce them at all or geography. Effective health policy takes account of would produce too little. different disease prevalences but is not simply de- • Market failures in health care and health in- termined by them. surance mean that government intervention can Differences in health spending are an obvious raise welfare by improving how those markets starting point in the search for an explanation of function. differences in health. In 1990 total annual health Any potential benefits from greater public sector spending ranged from less than $10 per person in involvement in health must be weighed against several African and Asian countries to more than the risk that governments will in fact make matters $2,700 in the United States. There was also consid- worse. For example, to satisfy special interest erable variation within regions. In Africa, Tanzania groups, governments may adopt policies that re- spent only $4 per capita for health in 1990, while duce the general welfare. Even when they choose Zimbabwe spent $42 per person. In Asia, Ban- correct policies, they may fail to implement them gladesh spent $7 per person each year, as against properly. $377 in Korea. Since the share of GNP devoted to Governments have a responsibility to spend health tends to rise with income, rich countries well, to get "value for money," whenever they differ from poor ones even more in health expen- devote public resources to health. This means al- diture than in income. locating resources so as to obtain the most im- But health spending alone cannot explain all the provement in health per public dollar, taking into variation in health among countries. Nor can in- account the private market's response to public come and education, or even spending, income, sector spending. Because private health care mar- and schooling taken together. Figure 3.1 illustrates kets can also fail to achieve value for money, gov- the discrepancies. The vertical axis shows how far ernment policy has a role in providing information life expectancy in a country differs from the value and incentives to improve the allocation of re- predicted on the basis of that country's income sources by the private sector. In most of the world and average schooling. France, Haiti, Singapore, a great deal of additional health could be obtained and Syria have almost exactly the life expectancy from a relatively small number of cost-effective in- predicted. China, Costa Rica, Honduras, and Sri terventions that could be delivered at modest cost Lanka, in the top half of the figure, all achieve five and with little need for high-level facilities or years or more of life beyond what would be ex- medical specialties. pected. Egypt, Ghana, Malawi, Uganda, the United States, and Zambia, in the bottom half of Health expenditures and outcomes the figure, all have a life expectancy about five years lower than expected, given their levels of Chapter 1 showed how greatly health status dif- income and education. fers among populations. Life expectancy ranges The horizontal axis of Figure 3.1 shows how far from forty years or less in some countries of Sub- total health spending differs from the value pre- Saharan Africa to seventy-five or more in the es- dicted by income and education. Egypt, Morocco, tablished market economies. In Sub-Saharan Af- Paraguay, Singapore, and Syria, in the left half of rica half of all deaths occur under age 5; in the the figure, spend relatively little. France, Haiti, In- established market economies half occur after age dia, Mozambique, and the United States, in the 74. Child mortality rates exceed 200 per 1,000 in right half, spend more than expected. several African countries but are below 20 in the At any level of income and education, higher richest countries. The burden of disease is five health spending should yield better health, all else times higher, per capita, in the worst-off than in being equal. But there is no evidence of such a the healthiest regions. relation. Countries are scattered in all quadrants of Three factors help to explain these huge differ- the figure. The countries that appear in the upper- ences. The first is human behavior. Chapter 2 left quadrant obtain better health for less money. showed that both health and the capacity to im- China, for instance, spends a full percentage point prove health are related to income and education less of its GNP on health than other countries at and to the changes in behavior that wealth and the same stage of development but obtains nearly 53 Health expenditure, income, and schooling only partly explain variation in life expectancy. Figure 3.1 Life expectancies and health expenditures in selected countries: deviations from estimates based on GOP and schooling Deviation from predicted life expectancy (years) f- 0 10 Better outcome, China Better outcome, lower expenditure higher expenditure Costa Rica Sri Lanka 0 0 0 Honduras 5 I- Greece 0 • • • • ... . Paraguay 0 India o o• • • • .. • 0 0 Singapore O Morocco syria • • • • • -- • - • • . ..., • ... • •• .._~ • • ''· O France Haiti - ..., 0 Mozambique \ 0 • United States 0 -5 . I- Egypt Malawi Ghana 0 • 0 0 O uganda Worse outcome, Zambia Worse outcome, lower expenditure higher expenditure ,, I I I I I. i it _ -10 -5 -4 -3 -2 I -1 0 1 I 2 3 ' 4 5 6 Deviation from predicted percentage of GDP spent on health So11rce: World Bank data. ten years of additional life expectancy. Singapore than predicted to achieve several years less of life spends about 4 percent less of its income on health expectancy than would be typical for its high in- than others at the same level of development but come and high educational level. achieves the same life expectancy. Other coun- Analyses using other measures of health status, tries, of which Costa Rica and India are examples, such as child mortality, yield similar results. This obtain relatively good health results but also spend raises obvious and important questions. What ac- relatively more. (In the case of India health spend- counts for these large deviations? How much is ing is low and health status is poor, but even lower attributable to the characteristics of health sys- spending and worse status would be expected for tems? How can public policy help to provide better a country with such low levels of income and health outcomes for a given national effort? schooling.) Egypt and Zambia, by contrast, get poor health for a lower-than-predicted level of The rationales for government action spending. Finally, it is possible both to spend more than predicted on health care and still achieve un- Public policy in health is successful if it leads to expectedly poor results. The United States is an increased welfare through better health outcomes, extreme case, spending 5 percent more of GNP greater equity, more consumer satisfaction, or lower 54 total cost than would occur in the absence of public finance care: spending more can translate into action. Of course, the pursuit of one or more of more services for the poorest or the same services these objectives does not by itself justify govern- for more people, including the less poor. In prac- ment intervention. There must be a basis for be- tice, very poor countries must target if they are to lieving that the government can achieve a better offer the poor any meaningful health care. outcome than private markets can. There are three Public goods and externalities are forms of market broad reasons why that belief may be true: one failure that may justify government intervention. centers on poverty and the equitable distribution The key characteristic of public goodswhich may of health care and the other two involve market be products or servicesis that one individual can failures. use them or benefit from them without limiting Reduction or alleviation of poverty provides a others' consumption or benefit. As long as some- straightforward rationale for public intervention in body pays, everybody benefitswhich makes it health. Success in reducing poverty requires two difficult or impossible to find anybody altruistic equally important strategies: promoting the use of enough to pay. Many public health interventions, the most important asset of the poortheir labor such as wide-area control of disease vectors and and increasing their human capital through access radio-based health information campaigns, are to basic health care, education, and nutrition. As nearly pure public goods for which only the gov- Chapter 1 showed, investment in the health of the ernment can ensure provision. Another public poor raises their educability and productivity. It good, new scientific information, has contributed gives them both the assets they need to lift them- enormously to the rapid improvements in health selves from poverty and the immediate welfare during this century. Its continued creation will de- gain of relief from physical suffering. Further- pend at least in part on governments. The right more, in most societies providing health and edu- choice of interventions and the proper level of pro- cation for the poor commands a degree of political vision of any public good require careful analysis assent that is altogether lacking for transfers of in- of the health benefits in relation to the costs. Prices come or of assets such as land. Investing in the provide no indication of what benefits are worth health of the poor is an economically efficient because private markets do not supply public and politically acceptable strategy for reducing goods. Nonprofit nongovernmental organizations poverty and alleviating its consequences, as World (NGOs) may supply such goods but cannot fully Development Report 1990 emphasized. substitute for government action. If "the poor" are all those living on less than $1 Externalities, or spillovers of benefits or losses (in real purchasing power) per day, they can typ- from one individual to another, characterize cases ically neither afford much health care nor borrow in which a private market might function but to pay for it. Simply transferring small amounts of would produce too much or too little. For example, income to poor people would create relatively little curing an individual of tuberculosis also prevents additional demand for health care. But because the transmission of the disease. But an individual's poor are more sensitive to the price of medical care demand to be cured of tuberculosis (or of mild or and also suffer a greater burden of disease than the asymptomatic sexually transmitted disease) is nonpoor, access to free or low-cost care can pro- probably not affected by consideration of the risk duce large increases in their consumption of health to others. If the externality is not taken into ac- care. count, treatment will be priced too high in private To ensure that subsidized health services actu- markets, and too little treatment will be given. ally reach the poor, however, may require restric- Subsidies for treatment are therefore justified. An tions, particularly on the kind of care that is paid example of negative externalities is a person's use for by the public sector. Offering free care of all of antibiotics, which may, by increasing microbial kinds to everybody typically leads to rationing of resistance to a drug, reduce the drug's value to servicesgeographically or according to quality. others and increase their risks. Such universal programs may not reach the poor Failures in markets for health care and health insur- or improve their health. They may, however, com- ance provide a third rationale for government ac- mand more political support than targeting, and tion to improve efficiency and, in the case of fail- they more easily address the problems of insur- ures in the market for health insurance, to improve ance markets that are discussed below. Who equity. One source of market failure, "adverse se- should receive free care depends on the preva- lection," arises because individuals face different lence of poverty and on the country's capacity to risks. Customers who know themselves to be at 55 high risk are motivated to buy more insurance and There is some moral hazard in the markets for are more likely to use it. So it is in the insurer's house and vehicle insurance. The extreme form is interest to find out who the high-risk customers when somebody burns down a house to collect the are and either to exclude them or to compensate insurance or abandons a car and reports it as for their greater risk by charging them higher pre- stolen. But unlike consumption of too much health miums. (Higher prices for all customers would re- care, these actions are crimes, with penalties that duce demand by low-risk people and therefore may greatly exceed the value of the asset. In any push prices still higher.) Defensive efforts to ob- case, the insurer's potential liability is limited to tain valuable information about risks add to the the (easily determined) market value of the asset. cost of insured health care without improving All the limitations on moral hazard and adverse health outcomes. selection are weaker in health insurance. It is Adverse selection presents a serious problem for harder to identify individual risks, and still harder risks existing at the time insurance is taken out, to attribute them to behavioral choices. There is no but an even more complex problem arises from the market value for the human body and no possi- fact that an initially low-risk person may become bility of abandoning one that is worn out and ac- high-risk later in life. In principle, there should be quiring a new one. The lack of a natural limit on insurance available specifically against this likeli- costs (since the asset being insured, the body, has hood of increased risk, or else insurance should no price with which costs can be compared) distin- cover a person's entire lifetime, with sharing of guishes health from other insurable risks. risks that may arise in the distant future, as well as The difficulties in insurance markets carry over of current ones. Neither solution is easy to imple- directly into markets for health care. If people have ment because of the extreme uncertainty; insur- "too much" health insurance, they will have an ance can cover known risks but not uncertainty incentive to use "too much" health care at too about risks. high prices. Unfortunately the difficulty of judging Another problem is the tendency of consumers health care risks and the impossibility of placing a to use more of a service when its marginal cost to value on a living body make it impossible to deter- them decreases. Insurance reduces or eliminates mine how much is "too much" in health care and the marginal cost of health care to consumers. So, health insurance. Nor is making a consumer pay providing insurance does not simply shift the way more for health care a sure way of reducing only a given amount of health care is paid for but in- "unnecessary" demand. creases the amount of care demanded. Failures of information make matters even Because the financial cost of disease is reduced, worse. A patient who knew the likely outcome people may take less care of their health, leading and the cost to him or her of every possible treat- to more illness and more subsequent demand for ment might yet be able to choose rationally be- care. Or they may protect their health more by tween gains and costs. But patients do not have way of health care, paid for by insurance, and less such knowledge, and the medical professional through their own behavior. Passing costs on to generally knows far more than the customer. This others such as insurers because one does not bear asymmetry of information means that the provider the full consequences of one's actions is called not only provides services but also decides what "moral hazard." It arises because of uncertainty services should be provided. The result is a poten- and because insurers cannot fully monitor con- tial conflict of interest between what the provider sumers' behavior and make them responsible for stands to gain from selling more services and his their decisions. Moral hazard also results when or her duty to do what is best for the patient. The providers induce demand for services that neither patient is at even more of a disadvantage when they nor consumers will pay for. sick and unable to make decisions or when deci- Both adverse selection and moral hazard have sions must be made quickly because of threats to more pernicious effects in markets for health in- life. surance than in markets for insurance on houses The same potential for consumption of unneces- or cars. Risks to houses are higher in areas prone sary services can arise any time a supplier is better to earthquakes or hurricanes, but they are easy to informed than a customer. It is a notorious prob- determine, and insurers respond by charging lem in car repair and home improvement services. higher premiums in those areas. Similarly, car in- But in these sectors the insurer has more oppor- surance premiums are higher for young drivers tunity to supervise the service provider, and the and other identifiable groups at greater risk of ve- insurer may decide simply to replace the item hicular accidents. rather than to repair it. Health insurers have no 56 replacement option. They may try to review pro- Since poor people typically cannot buy such care fessionals' recommendations before agreeing to for themselves, there is a straightforward case for pay for services, but health professionals often dis- public finance. Public health measures and essen- agree about expected medical outcomes, and wait- tial clinical care together constitute a package of ing for a second opinion may cause pain, suffer- health care that might justifiably be financed by ing, and increased risk for the patient. general revenues, with perhaps some contribution These problems constitute the market failure pe- from user fees. This strategy is also compatible culiar to health: expenditure on medical care can with the argument that basic health care is a fun- be extremely high, yet not all justified care is pro- damental right. Although most of the population vided and much care of doubtful value is paid for. may be able to pay for such care, the government Some people are denied insurance, while others has a responsibility to ensure that the poor, too, may be overprotected. Those who do not pay the can exercise their rightat least to the extent that full costs of treatment may take poorer care of their society can afford. own health than they could. Many of the extra Third, the rationale that the government should costs are paid by society as a whole. intervene in health care markets because of signifi- The market for health care goods and services cant market failures applies particularly to the reg- can also fail through imperfect competition among ulation of health care and health insurance. The providers, which allows excess profits, inefficient government cannot finance all medical care for use of resources, poor quality, and too little pro- which insurance might be desirable without wors- duction. Sometimes governments themselves arti- ening the tendency toward higher costs and risk- ficially stifle competition. For example, govern- ing de facto rationing of health care, which par- ments may prohibit or interfere unduly with the ticularly hurts the poor. Beyond a well-defined operation of private health care providers, particu- package of essential services, therefore, the role of larly NGOs. Governments often also protect do- the government in clinical services should be lim- mestic producers of drugs and vaccines. In Ban- ited to improving the capacity of insurance and gladesh tetanus vaccine produced domestically at health care markets to provide discretionary care government insistence had such low potency that whether through private or through social insur- its use in 1989-92 risked thousands of lives before ance (earmarked taxes such as social security or it was replaced with imported vaccine. other mandated arrangements). Of course, the Economies of scale in productionwhich occur range of services included in the nationally de- when a single large producer is much more effi- fined essential package will vary substantially cient than many small onesalso lead to noncom- from country to country. To provide equitable ac- petitive situations. In many parts of the world hos- cess for the poor, to address problems of adverse pitals and specialists face little or no competition selection, and to contain costs, the governments of because of economies resulting from large-scale almost all OECD countries have made available a operation. Such situations may call for regulation comprehensive essential package with public (or of the private market. publicly mandated) financing. Poorer countries The three rationales for government interven- must, of necessity, define their essential packages tion in the health sectorprovision of public more narrowly. goods, reduction of poverty, and market failure Governments can further improve how markets correspond roughly to three different kinds of ser- function by providing information about the cost, vices. First, the services classified as public goods, quality, and outcome of health care. Simply by de- and some of those characterized by large exter- fining an essential clinical package, the public sec- nalities, constitute what is known as "public tor provides valuable guidance on what is and health." Public health includes those services pro- what is not cost-effective. This distinction may vided to the population at large or to the environ- then influence the design of private or social insur- ment, such as spraying to control malaria. It also ance packages and the behavior of individual pro- typically includes some services such as immun- viders or patients. Information on the relative cost- zations that are not public goods but that carry effectiveness of different discretionary procedures substantial externalities. is similarly valuable and might be used by insurers Second, the inclusion of health care as part of a and providers to reduce costs and attract clients. strategy for combating poverty justifies public fi- Neither theory nor experience points to a gen- nancing of "essential" clinical or individual ser- eral rule on the extent to which the public sector vices. These are highly cost-effective services that should provide health care directly, as distinct would greatly improve the health of the poor. from financing it. Governments might have to 57 Box 3.1 Paying for tuberculosis control in China Tuberculosis kills or debilitates more adults than any Charging tuberculosis patients had perverse effects. other single infectious agent. Without appropriate When doctors and institutions expected to be reim- treatment, 60 percent of those with the full-blown dis- bursed by insurance, they provided excessive diagnos- ease will die. In China, it is estimated, more than tic tests and examinations during treatment and dis- 360,000 peoplemost of them poor peasantsdie of pensed higher-cost antibiotics that should have been tuberculosis every year. Tuberculosis is best prevented reserved for the most resistant cases. Daunted by the by curing infectious persons early in the course of dis- costs, many low-income victims failed to enter treat- ease, thus interrupting transmission to others. Well- ment or dropped out early. There were no incentives to run programs can cure 80 to 90 percent of patients; ensure that patients completed treatment or were poorly administered programs cure 30 percent or less, cured. Because health system records showed very leading to larger numbers of sustained cases of infec- high rates of cure for those who completed treatment, tion and related deaths and to new infections. the government remained largely unaware of the dete- China made substantial progress against tuber- riorating situation. The direct cost to the health system culosis during the 1960s and 1970s, using standard of a poorly functioning program was nil, but the indi- long-term (twelve to eighteen months) antibiotic ther- rect costs to the economyand the personal costs to apy that was essentially free of charge. Since the early patients and their familieswere enormous. 1980s, however, infection rates in about half the coun- An estimated I million to 1.5 million additional tu- try's provinces have stagnated or increased, despite berculosis cases remained infectious during the 1980s the adoption of an improved short-course (six to eight because treatment was no longer free. Tens of millions months) therapy. Much of the trend is attributable to of new infections were produced, and many of those changed health-financing policies and, in particular, to infected will fall victim to the disease later in life. the government's decision that health facilities should The development of drug-resistant strains was also ac- be encouraged to charge patients for virtually all ser- celerated. Given appropriate policies, many of the vices. Starting in 1981, health institutions had to earn more than 3 million persons who died of tuberculosis much of their operating costs from sales of drugs and in China during the decade could have been saved, services. Although base salaries were still funded from and the risk of infection for society could have been the public budget, workers' bonuses, housing, and re- halved. tirement benefits depended in part on institutional in- China, having recognized the problems caused by come from service provision. Managers' investment charging for tuberculosis therapy, has begun a major budgets were also linked to revenues from fees. A few national tuberculosis-control effort that provides sub- public health services such as immunizations remained sidies for treatment and appropriate incentives for pro- partially subsidized, but tuberculosis diagnosis and viders of care. Early results of this policy show dra- treatment were not, despite drug costs of $30 to $80 per matic increases in the number of cases cured. treatment. -I supply a package of essential health services di- In some circumstances market failure may im- rectly where private care would not be feasible pose only slight welfare losses, and the benefit of without high subsidiesfor example, in lightly correcting it may be outweighed by the costs of populated, very poor areas. (In many parts of the government action. In other cases the losses from developing world an alternative method of provid- failure to take account of positive externalities and ing such services is to subsidize an NGO.) In most supplier-induced demand can be enormous. Pol- circumstances, however, the primary objective of icy toward tuberculosis control in China provides public policy should be to promote competition an example: elimination of some free health care among providersincluding between the public and introduction of profit incentives in the provi- and private sectors (when there are public pro- sion of health services dramatically reduced treat- viders), as well as among private providers, ment rates, reversing progress against the disease whether nonprofit or for-profit. CompetitionS and causing much needless suffering (Box 3.1). should increase consumer choice and satisfaction Failures of government intervention can arise, how- and drive down costs by increasing efficiency. ever, even when government action might be Government supply in a competitive setting may sound policy. improve quality or control costs, but noncompeti- Governments may misjudge how an interven- tive public provision of health services is likely to tion will work in practice. Governments have only be inefficient or of poor quality. partial control over how private actors respond, 58 and those responses can undermine the intended ments are most appropriate for affecting the objective. Since 1971 physicians' fees in all pro- behavior of insurers, providers, and patients. This vinces of Canada have been set by negotiation raises the question of how far the government with provincial governments, and fees are no should itself act as an insurer, through social insur- longer rising faster than the general price level. To ance, and how far it should regulate private in- protect their incomes, particularly during the infla- surers. Each of these decisions involves tradeoffs tionary period 1971-75, physicians carried out a among the objectives of health policy: better greater number of procedures. This reaction was health outcomes, lower costs, more equity, and strongest where real fees fell the most. The saving greater consumer satisfaction with the health sys- in government expenditure was therefore much tem as a whole and with individual care. less than had been anticipated. Governments may not have the capacity to Value for money in health administer or implement policies well. Indeed, they may suffer from corruption and from sheer No matter how health services are organized and incompetence. The examples of two donor- paid for, what they actually provide are health in- financed public hospitals, each with 500 to 600 terventions: specific activities meant to reduce dis- beds, in two Latin American countries illustrate ease risks, treat illness, or palliate the conse- the problem. One was simply too large to adminis- quences of disease and disability. Debates about ter and operate and therefore could not be used at whether health services should concentrate on more than 60 percent of capacity. The other was so "vulnerable groups" such as children, pregnant badly designed that it could not accommodate women, or the elderly, or about the relative roles more than one-third the planned number of of hospitals versus health centers, or about pre- patients. ventive versus curative activities, are at bottom de- Governments are vulnerable to special inter- bates concerning the proper mixture of interven- ests both within and outside the health system. By tions. In health, as in every other sector, customers financing the training of unneeded physicians, by want value for the money spent on such interven- paying for low-value discretionary services for bet- tionswhether they pay directly or indirectly, in ter-off patients, and by protecting domestic indus- their roles as taxpayers or as buyers of health tries, governments help create the interests that insurance. later impede good policy, especially when quick Knowing the cost-effectiveness of a health inter- responses are needed to meet changing circum- ventionthe net gain in health (compared with stances or new opportunities. Even when society doing nothing) divided by the costcan be ex- as a whole would gain, public action may fail be- tremely useful for both public and private deci- cause it does not overcome the resistance of those sions. Governments can generate such informa- who would lose as a result. tion, and they can use it in two ways. First, they Perhaps the most fundamental problem facing can use it in determining whether a particular pub- governments is simply how to make choices about lic sector intervention is cost-effective: this means health care. Too often, government policy has con- judging the improvement in health compared with centrated on providing as much health care as pos- what would have happened through private deci- sible to as many people as possible, with too little sions in the absence of public action. (Chapter 4 attention to other issues. If governments are to addresses these issues for public health measures finance a package of public health measures and and Chapter 5 for the public finance of essential clinical services, there must be a way to choose clinical services.) Second, they can supply infor- which services belong in the package and which mation about the outcomes and costs of different will be left out. (The next section describes a mea- health interventions to consumers, providers, and sure of cost-effectiveness for health interventions insurers, and this knowledge can increase the that helps with this choice.) If financing is public value per health dollar spent in the private sector, but provision is private, governments must decide including what is spent on discretionary services. how to subsidize private care. The question of in- Private providers have no more incentive than centives to providers raised by that issue also ap- public providers to measure health outcomes, but plies to paying for publicly provided careto the they do face greater incentives to know their costs. "internal market" in the public sector. And if gov- Cost information alone can promote allocative effi- ernments are to influence the market for discre- ciency, as the experience of a Brazilian nonprofit tionary services, they must decide what instru- maternal and child hospital demonstrates (Box 59 Box 3.2 Cost information and management decisions in a Brazilian hospital The Instituto Materno-Infantil de Pernambuco (IMIP) is lations, it was evident that closing the intensive care a private, nonprofit hospital founded in 1960 to serve unit (except for newborns) and strengthening other the metropolitan area of Recife. In 1992 it received the services would save a greater number of children's first UNICEF award to a "child-friendly" hospital in lives. In particular, since the children who died in the Brazil in recognition of its work, particularly in the pro- hospital typically arrived very sick and often severely I motion of breastfeeding. IMIP depends for 95 percent malnourished, it appeared more cost-effective to try to of its revenue on contracts from the Instituto Nacional find high-risk children and treat them earlier. The strat- de Assistência Médica e Previdência Social (INAMPS) egy used was to expand the network of small commu- of Brazil's Ministry of Health. Annual spending runs nity health posts in the slum neighborhoods of Recife. about $6 million. IMIP opened the first such posts in 1983; by 1986 infant Starting in 1989, IMIP organized an accounting sys- mortality in those neighborhoods had fallen from 147 tem that divides services according to eleven cost cen- to 101 per 1,000 births. ters for final output. Administrative, laundry, food, ra- The experience of IMIP illustrates three lessons diology, laboratory, transport, and other nonfinal about cost-effective delivery of essential care. One is services were assigned to these final outputs in propor- that allocative efficiency can be improved without com- I tion to their measured or estimated use. plete information: medical professionals know much IMIP must match its average costs to average reve- about outcomes and often need only to know more nues determined by the price schedule of INAMPS, about costs. A second lesson is that autonomy facili- which is organized by treatment groups rather than by tates such changes: since private facilities generally individual services. Losses in any cost center must be have much more autonomy than public ones, this is an I offset by surpluses elsewhere. Gravely ill children are argument for more public finance of private provision referred to the hospital from all over northeast Brazil, or for decentralization of public systems. The third les- and there are three infant deaths per day among them. son is that even prices that are not based Ofl cost-effec- reduce mortality, IMIP created a pediatric intensive tiveness criteria can guide decisions about what care to care unit. The treatments provided, however, cost provide. It is more useful for government to set those much more than INAMPS would pay. And mortality prices correctly than to try to make all the allocative did not decline. Even without cost-effectiveness calcu- choices. Lb 3.2). By estimating costs for "cost centers" and value to human life, as would be necessary if costs relating them to outputs, the hospital discovered and gains were to be put in the same units. that its pediatric intensive care unit would drain Only in the past decade have costs and effective- resources from other departments, given the ness been systematically estimated for a wide prices the government paid for various services. range of health care interventionsalthough the The decision was made to limit the intensive care first such calculations had been made many years unit to newborns; community-level health posts earlier. Only a small share of the thousands of appeared more cost-effective for other cases. known medical procedures has been analyzed, but the approximately fifty studied would be able to Measuring the cost-effectiveness of health interventions deal with more than half the world's disease bur- den. Just implementing the twenty most cost- Given a common currency for measuring cost and effective interventions could eliminate more than a unit for measuring health effects, different inter- 40 percent of the total burden and fully three-quar- ventions can be compared by what it costs to ters of the health loss among children. achieve one additional year of healthy life. Out- The cost and effectiveness estimates used in this comes are measured in the same unit of disability- Report are based, as far as possible, on actual con- adjusted life years (DALYs) used to estimate the ditions in developing countries. Some fixed costs burden of disease. Nonhealth burdens, such as in- of operating a health system that cannot be attrib- come lost because of disease, are not included in uted to particular interventions are not consid- the measure. The ratio of cost and effect, or the ered, but the costs of intervention-specific capacity unit cost of a DALY, is called the cost-effectiveness are taken into account. Costs are assessed at mar- of the intervention; the lower that number, the ket prices. For inputs that cannot be traded inter- greater the value for money offered by the inter- nationally (such as semiskilled labor), costs will be vention. This approach avoids assigning a dollar lower in developing countries. For drugs, most 60 equipment, and high-level manpower, costs are either the public or the private sector costs lives. likely to be equal across countries, leaving aside An expenditure of $100,000 on chemotherapy for the effects of tariffs or other barriers. Indirect tuberculosis could directly save about 500 patients. costs, such as patients' costs of travel to treatment It would also prevent them from infecting others, or the income they forgo, can be substantial for for a total gain of about 35,000 DALYs. The same some interventions and perhaps particularly for expenditure on management of diabetes would women. Because these costs are difficult to deter- also benefit 500 patients but would save only 400 mine, they were largely ignored; more study is DALYs; each patient would gain less than one needed of how these barriers affect the utilization healthy year from a year of treatment, and there of health services. would be no benefit from reducing incidence. In- For some common health service packages such sisting on value for money is not only fully con- as immunizations, costs are computed on a joint sistent with compassion for the victims of disease, basis rather than separately for each intervention it is the only way to avert needless suffering. in the package. The estimates do not represent an The results of cost-effectiveness analysis confirm unattainable ideal; they assume that medically cor- the value of the primary health care interventions rect procedures are followed and that reasonable included in programs to reduce childhood malnu- care is taken as to quality, but they also allow for trition and mortality, chiefly from infectious dis- incomplete coverage or compliance. Whenever eases. Several hitherto neglected interventions are possible, actual experience is used to guide esti- also very cost-effective: chemotherapy against tu- mates of such things as how many patients will fail berculosis, integrated prenatal and delivery care, to complete a course of treatment. Future gains mass programs to deworm children, provision of from current interventions are discounted at 3 per- condoms along with information and education to cent per year, which has little effect on the ranking combat AIDS, and measures against smoking, of interventions the effects of which are felt such as education, consumer taxes on tobacco quickly, although it does reduce measured gains products (an effective deterrent for adolescents from interventions when the health effects are felt who are not yet addicted), and prohibition of only in the long run. smoking in public places. Many of the most cost- This Report found huge differences in both the effective health interventions are preventive in cost and the effectiveness of various health inter- character. But not all preventive measures are cost- ventions. Figure 3.2 presents both dollar costs and effective: spraying to control the mosquitoes that gains in DALYs for each of forty-seven different carry dengue is an example of relatively poor value interventions. Higher points represent interven- for money. At the same time, a small number of tions that are more effective in improving health; neglected but cost-effective clinical (mostly cura- points farther to the right represent lower-cost in- tive) interventions could eliminate a substantial terventions. Some interventions cost more than fraction of the burden of disease in many $10,000 per person benefited, while others cost countries. less than $1. Some interventions add more than In general, most cost-effective interventions can ten years of healthy life; for others the gain is be performed outside hospitals. By treating a small equivalent to only a few hours or days of full number of severe cases of disease, however, hos- health. Both axes are scaled in logarithms so that pitals can sometimes improve health at a lower the diagonal lines show equal cost-effectiveness cost per DALY than lower-level facilitiespro- ratios in dollars per DALY. These ratios vary vided that clinics or health posts treat most cases widely, from as little as $1 to as much as $10,000. and refer to hospitals only those requiring more Higher lines represent more cost-effective inter- sophisticated care. ventions. Four specific interventions illustrate ex- treme combinations of cost and health gain: vita- Complications in the use of cost-effectiveness min A supplementation in areas where the risk of blindness from vitamin deficiency is high (very Both the cost and the effectiveness of an interven- low cost, high gain), chemotherapy for tuber- tion can be affected by the incidence and preva- culosis (high cost, very high gain), environmental lence of the disease and the probability of dying control of dengue (low cost, low gain), and treat- from it. Preventive interventions are less cost- ment of childhood leukemia (very high cost, mod- effective for relatively rare diseases because more erately high gain). people have to be reached to prevent one case. The Because interventions can differ so much in cost- fatality rate matters because preventing or control- effectiveness, making allocative decisions badly in ling a disease saves more lives if there is a high 61 The costs and effectiveness of health interventions vary greatly. Figure 3.2 Benefits and costs of forty-seven health interventions Increase in DALYs (log scale) 100 Chemotherapy for tuberculosis . 10 Vitamin A supplementation . . 0.1 Treatment of $1/DALY leukemia S 0.01 $10/DALY Greater effectiveness Environmental 0.001 Lower control of dengue $100/DALY cost $10,000/DALY $1,000/DALY 0.0001 10,000 1,000 100 10 1 0.10 Cost per intervention or per intervention-year (dollars, log scale) Target: Children under age 15 Adults age 15 or older a Note: DALY, disablility-adjusted life year. Interventions are specific activities intended to reduce disease risks, treat illness, or palliate the consequences of disease and disability; an intervention-year is an intervention repeated throughout the year rather than provided only once. a. Jncludes some interventions that benefit all age groups. Source: Jamison and others forthcoming; Worlc Bank data. probability of dying. Immunization in an environ- information is needed to judge which interven- ment in which children are undernourished and tions should have priority. National or regional as- many die from preventable diseases is more cost- sessments are also important for estimating the effective than if children are otherwise healthy and expenditure levels required and the probable im- face little risk of dying. (Box 3.3, on measles and pact on the national burden of disease. tuberculosis, illustrates these issues.) Fortunately, If providing an intervention did not impose any differences in cost-effectiveness between one in- fixed costs in infrastructure and program adminis- tervention and another are often much larger than tration, then a low cost per DALY saved would either the variation from one locale to another or suffice to justify the intervention. In practice, there the uncertainty in the estimates. Where this is not may be substantial fixed costs to share over a num- the case, as exemplified by malaria, detailed local ber of interventions, and administrative capacity 62 Box 3.3 Cost-effectiveness of interventions against measles and tuberculosis The costs and effects of measles vaccination were esti- average cost of the lower-coverage strategy, but the mated for a model urban area based on data from higher-coverage approach continues to be extremely Lagos and Kinshasa. Data from Matlab, Bangladesh, cost-effective. Similar calculations were made for che- were used to model measles in rural areas. In each motherapy for tuberculosis, for both a standard course area, 36,400 cases of measles were assumed to occur in of treatment (twelve to eighteen months) and a short the absence of vaccination, with 1,452 urban deaths course (six to eight months), each with and without and 806 rural deaths. All the health damage from mea- hospitalization. Using data from Malawi, Mozam- sles comes from deaths, each of which costs thirty bique, and Tanzania, the average incremental cost DALYs. The simulations considered three different (marginal cost plus the average cost attributable to the strategies: no vaccination, immunization at nine fixed costs of the tuberculosis control program but ex- months (the earliest age at which the standard vaccine cluding other fixed costs of the health system) was esti- is effective) with 60 percent coverage, and nine-month mated at about $80 to $110 per cure for ambulatory vaccination plus efforts to increase coverage to 80 per- treatment, and $160 to $300 when hospitalization was cent, which raises supervision costs by 10 percent. For required. Cost per death directly averted was in the the last strategy, incremental as well as average costs range of $75 to $275, but cost per total death averted, were calculated to highlight the effect of raising cover- taking account of the interruption of transmission, can age. Costs were related to each of three effects: cases be as low as $20 and never exceeds $100. These very averted, deaths averted, and DALYs gained. Box table low costs translate into costs per DALY saved of about 3.3 shows the results. $1 to $3, making chemotherapy for tuberculosis one of Measles strikes later in childhood in rural areas, so the most cost-effective of all interventions. These costs cases are easier to prevent. But because earlier cases (in do not vary with the annual rate of infection. The cost- urban areas) cause more deaths, the cost per death effectiveness of BCG vaccine, by contrast, is extremely averted or per DALY is higher. In both urban and rural sensitive to infection rates; the vaccine is cost-effective areas, the marginal cost of raising coverage exceeds the only when the risk of infection is high. Box table 3.3 Health costs and gains from measles immunization Urban, by percentage vaccinated Rural, hr percentage vaccinated Item 60 80 60 80 Cases prevented (thousands) 10.7 14.2 16.2 22.6 Total cost per case prevented (dollars) 17 18 11 12 Incremental cost (dollars) - 22 - 15 Deaths averted (thousands) 0.4 0.6 0.4 0.5 Total cost per death averted (dollars) 432 462 525 561 Incremental cost (dollars) 552 670 DALYs gained (thousands) 12.3 16.4 10.2 13.5 Total Cost per DALY (dollars) 15 16 18 19 Incremental cost (dollars) - 19 - 23 Source: Foster, McFarland, and John forthcoming. may be limited. Spending on interventions that Since only relative ranking is possible, the at- are very cost-effective but resolve very small dis- tractiveness of an intervention also varies accord- ease burdens could waste resources by making it ing to what other health problems and treatments difficult to deal with diseases that impose much are locally prevalent. Oral rehydration therapy is larger burdens. Priority should go to those health an example: in environments in which child mor- problems that cause a large disease burden and for tality is low, it is much less cost-effective than im- which cost-effective interventions are available munization because it may have to be given re- (Box 3.4). If a particular health problem causes the peatedly during a child's first few years, but as loss of many healthy life years but there are no mortality rises it becomes more cost-effective. In meansor only very costly meansfor dealing general, the cost-effectiveness ratio varies not only with it, then it should be a priority not for health with local conditions but also with the degree to care but for research on development of cost-effec- which an intervention penetrates or covers a tive interventions. population. Cost-effectiveness is also influenced 63 r Box 3.4 Priority health problems: high disease burdens and cost-effective interventions It is easy to determine which health problems among ease burden in women ages 45-59 and for smaller children under age 5 deserve priority. As Appendix shares in other age and sex groups. Among communi- table B.6 shows, nine diseases each account for more cable diseases tuberculosis, AIDS, and respiratory in- than 1 percent of the total disease burden in this age fections deserve priority, but they cause less than 10 group. These diseases range from acute respiratory in- percent of all ill health after age 45 and only 20.1 per- fections (more than 17 percent in both boys and girls) cent in men ages 15-44. to iodine deficiency (1.2 to 1.3 percent). Of these prob- Large disease burdens and cost-effective interven- lems, which cause fully 80 percent of young children's tions coincide for only one group of adults, women ill health, eight can be addressed by interventions cost- ages 15-44. Six of the ten main sources of ill health can ing less than $100 per DALY saved. The only exception be prevented or treated for less than $100 per DALY. is congenital problems, which are responsible for more These range from maternal health problems (18.0 per- than 6 percent of the disease burden but for which no cent of the burden) to respiratory infections and ane- cost-effective interventions are known. mia (2.5 percent of the burden each) and account in The situation is much more complicated for adults total for 44 percent of ill health among women of repro- (Appendix table B.7). For example, cerebrovascular ductive age. Two other problemsdepression and self- disease is the leading cause of healthy life years lost in inflicted injuryeach cause at least 3 percent of the both sexes after age 60 and in women ages 45-59, but disease burden, but dealing with them is much more interventions to deal with it cost $1,000 or more per problematic. DALY saved. Ischemic heart disease is the second or These calculations illustrate the chief problem a third leading cause of ill health in both sexes after age health care system faces as the population ages: the 45, but the cost per DALY of dealing with it is $250 to marginal cost of a year of healthy life gained rises $1,000. Among the ten principal noncommunicable sharply, leading to difficult choices between increased causes of ill health in this age group, interventions spending and lower health gains. However, many costing less than $100 per DALY saved exist only for health problems of the elderly that cannot be fully re- I cataracts, anemia, and cancers of the respiratory sys- solved may be palliated at low cost. And much can be tem (through reduction of smoking) and the cervix. done at earlier ages to improve the health of future These problems account for only 7.9 percent of the dis- generations of old people. by the presence of other interventions that might An important limitation on the use of cost-effec- affect costs (through sharing of joint costs) or out- tiveness analysis of resource allocation in health is comes. Sometimes combining two interventions, that a number of interventions with important one preventive and one curative, is the appropri- health consequences also affect income or welfare ate way to deal with a particular disease, as is the in other ways. Chemotherapy for tuberculosis has case for tuberculosis and malaria. Some treatment no value beyond the DALY gain associated with for malaria is necessary because preventive mea- curing tuberculosis, but investing in girls' school- sures do not protect everyone; even if treatment is ing has both important consequences for health more costly, both interventions should be applied. (as documented in Chapter 2) and for income and In exceptional circumstances it may be worth status later in life. Similarly, family planning, in paying high marginal costs to extend coverage of addition to its health benefits, permits family an intervention to the entire population because choice about the number and spacing of children; the disease can be eradicated permanently (as has improved water supply and sanitation create ame- occurred with smallpox and may now be possible nity and time-saving benefits; increased food con- with polio). The gains in such cases include not sumption allows higher levels of physical activity; only the DALYs saved at the margin from the last and improved road safety reduces property dam- people immunized but all the healthy years that age and saves lives. For some of these interven- would otherwise be lost to the disease in the tions (for example, family planning and girls' future. A similar argument holds if a low-cost schooling) the cost per DALY is sufficiently low to intervention that has to be applied continually make them attractive on health grounds alone; is replaced by one with large initial costs other benefits only strengthen the case. For other but permanent effects, such as sanitation to reduce interventions the cost per DALY gained may be the need for treating many fecally borne diseases. too high to justify investment on health grounds 64 alone, but consumer willingness to pay for non- has the most easily measured output. Coverage in health benefits means that costs to the health sys- many regions remains incomplete. Immunization tem can be low. Many water and sanitation invest- against measles and against diphtheria, pertussis, ments are in this category. and tetanus has reached 90 percent or more in Using cost-effectiveness to select health inter- Chile, China, Cuba, Korea, and Saudi Arabia, but ventions for public financing does not necessarily it is still below 50 percent in some Asian and many mean spending the most resources where the bur- Sub-Saharan African countries. Many of the other den of disease is greatest. Instead, it means con- components of an adequate public health package centrating on the interventions that offer the great- scarcely exist. est possible gain in health per public dollar spent. As far as clinical services are concerned, the The relevant comparison is usually not with a situ- principal government failing in most countries is ation in which nothing is done but with the situa- the attempt to provide everything to everybody, tion created by privately financed health interven- with no distinction between more and less essen- tions. The most justified public measures will tial care and more or less needy patients. For some therefore combine a strong rationale for public ac- health services provided by the public sector, the tion with a cost-effective health intervention. Be- system of provision is so grossly inefficient that it cause individuals differ in how they value the is unlikely to be cost-effective no matter what in- present in relation to the future and in how they terventions the system tries to provide. Such inef- judge the seriousness of different health condi- ficiencies have been criticized so clearly and for so tions, a uniform ceiling on what the government long that it is evident they will only be overcome pays to gain one DALY may leave some people by radical changes in the organization of health with more publicly financed care than they would caresuch as a shift in the government's role from choose and some with less. But of all possible uni- providing care to financing care and stimulating form criteria by which to judge what interventions competition among providers. These changes will to pay for, cost-effectiveness appears to yield the in turn require a clear distinction between essential most efficient distribution of health resources. and discretionary spending and a new determina- tion by governments to achieve value for money in Data limitations health services. There is no other equitable way to control government spending. Cost-effectiveness analysis requires data on expen- Most governments also perform poorly in regu- ditures for particular interventions and on out- lating markets for private services, including in- comesinformation that health facilities and sys- surance. As recent research in Brazil has shown, tems, particularly those in the public sector, the quality of medical care could be substantially typically do not collect. Such information could improved at low cost if government discharged promote substantial gains in efficiency, but it will this role better. The rapid growth and almost total take considerable time and effort for most public lack of regulation of private insurance in such systems in developing countries to learn how to countries as Brazil and Korea present another chal- gather and use it. Budgets often disaggregate only lenge for which governments are ill-prepared. by inputs, not by programs, and costs per consul- tation or per bed-day mix many different interven- Allocation of spending to cost-effective services tions. Outcomes are seldom quantified. For inter- ventions the cost-effectiveness of which vary There are no calculations of how many years of greatly with local conditions, there is no substitute healthy life are currently saved by health systems. for information on both costs and results. Nonetheless, it is clear that many of them perform much worse than they might. Many governments Health policy and the performance of health spend too much on sophisticated hospital services systems of low cost-effectiveness and too little on essential public health and clinical services. The share of All governments subscribe to the view that the public expenditures for health absorbed by tertiary state must ensure the provision of certain basic and secondary care hospitals, for example, is as public health services. But few achieve this goal high as 70 to 75 percent in Jordan and Venezuela. even for immunization, which is probably the Tertiary care hospitals alone may consume 30 to 50 health intervention that has received the greatest percent of the health budget. Only a quarter of government attention and donor support and that government spending, and often less, is devoted 65 Table 3.2 Actual and proposed allocation of public expenditure on health in developing countries, 1990 (dollars per capita) Estimated Spending under the proposed package actual Low- Middle- All spending, all income income developing developing Package component count ries count ri cs' countries countries Contents Public health 4 7 5 1 EPI Plus; school health programs; tobacco and alcohol control; health, nutrition, and family planning information; vector control; STD prevention; monitoring and surveillance Essential clinical services 8 15 10 4-6 Tuberculosis treatment; (minimum package) management of the sick child; prenatal and delivery care; family planning; STD treatment; treatment of infection and minor trauma; assessment, advice, and pain alleviation Total, public health and 12 22 15 5-7 minimum essential clinical services Discretionary clinical services' 6 40 6 13-15 All other health services, including low-cost- effectiveness treatment of cancer, cardiovascular disease, other chronic conditions, major trauma, and neurological and psychiatric disorders Total 6 62 21 21 Note: Current spending on essential clinical services is estimated to be 20-30 percent of total public expenditure on health on the basis of estimates in World Bank health sector reports. The numbers reported should be regarded as approximations. Estimated for an income level of $350 per capita. Estimated for an income level of $2,500 per capita. Estimated residually. The negative number for low-income countries reflects total spending below the cost of the package. Source: World Bank calculations; World Bank sector reports on Ghana (1989), India (1992), Indonesia (1991), Jamaica (1989), Jordan (1989), Mexico (1989), Nigeria (1991), Pakistan (1992), Turkey (1990), and Venezuela (1992). to cost-effective public health measures and essen- propriate package of cost-effective care would ad- tial clinical care, delivered mainly in health centers dress. Even the best-designed care package could and communities. In many countries the share of not prevent all the health damage from these dis- public spending devoted to these basic services eases because of the low cost-effectiveness of some has been falling in recent years. In Brazil 64 per- interventions and the increasing marginal costs of cent of public spending on health in 1965 was for even the best ones. Still, because of the size of the preventive and public health activities, but by the burden and the low cost per DALY of the interven- mid-1980s the share had dropped to 15 percent, tions, it is reasonable to conclude that public ex- and hospitals absorbed fully 70 percent of expendi- penditure on health should initially be concen- ture. The resulting weakness of the primary care trated on those conditions. What this implies for network leads patients to seek care in hospitals; Up the distribution of spending by type of input or to 80 percent of the cases crowding hospital emer- level of facility is less clear, but it probably means gency rooms could be treated as effectively, but that facilities above the district hospital level more cheaply, at the primary level. should account for only a small share of the total, In the world as a whole, almost half the existing primarily for dealing with referrals. disease burden is from communicable diseases, Table 3.2 illustrates the degree of misallocation nutritional disorders, and maternal and perinatal of health spending by comparing estimated actual causes. It is primarily these problems that an ap- expenditure with what would be spent for a pack- 66 age of health services designed to address most cluding private spending, is about $14, about the effectively the burden of disease in the developing same as the proposed package. This means that world. This package consists of public health ser- either substantial private resources will have to be vices that would cost just over $4 per capita in poor used or additional government resources will be countries (with average income per capita of $350) needed; even if all public expenditure on discre- and a minimum package of essential clinical ser- tionary services were eliminated, current govern- vices that would cost about $8 per capita more. In ment spending on health would not meet the costs middle-income countries (with average income per of the package. capita of $2,500), the same package of public Since the minimum package would cost only health measures and essential clinical services about $60 billion for all developing countries to- would cost about 80 percent more, or $22 per per- gether, the task is to reallocate resources in mid- son. This difference partly reflects different epi- dle-income countries and to find additional re- demiological conditions, but input costs, particu- sources of about $10 billion in low-income larly salaries, would also be higher in middle- countries. The $8 to $10 per capita needed in extra income countries. Many countries will define the spending on public health measures and essential essential clinical package much more broadly than clinical services is less than the $13 to $15 per cap- the minimum discussed here. Even in relatively ita now spent, on average, on discretionary or poor countries, targeting public finance of essen- nonessential clinical services. In fact, spending on tial services to the poor would allow creation of a these less cost-effective services is now roughly broader and more generous package. Such a pack- double the amount that countries spend on the age can be built up by adding interventions in or- recommended package of public health measures der of decreasing cost-effectiveness until the addi- and essential clinical care. The right combination tional health gain is judged not to be worth the of reallocation and additional expenditure would cost, given the country's resources. To ensure po- allow governments to achieve a large improve- litical support and to deal with problems of market ment in overall health. failure and equity, countries may choose to finance Table 3.3 indicates how large this gain could be. the essential package universally from public, or Properly allocated, an expenditure of only $12 per publicly mandated, sources. person in low-income countries (excluding China) Governments in developing countries spend an would be enough to reduce the disease burden by estimated $21 per capita on health, for a total of almost one-third. This is 226 million DALYs, about $84 billion. It is estimated that only a little equivalent to 7.0 million infant deaths per year. In more than $1 per person, or a total of $5 billion, middle-income countries the proposed package goes for cost-effective public health measures. To could deal with only 15 percent of the disease bur- buy the package described here, countries would den, despite the higher expenditure per person. need almost to quintuple what they spend on pub- The total reduction in ill health in middle-income lic health. About $4 to $6 per capita, or $17 billion countries would be about 45 million DALYs, the to $25 billion total, goes for clinical services deliv- equivalent of 1.4 million infant deaths per year. ered through lower-level facilities or classified as The smaller gain in these countries reflects the fact primary health care. These services commonly in- that they have already eliminated much of the bur- clude many of those in the essential clinical care den from easily controlled communicable diseases. package, but they are usually not delivered to the A large part of the remaining burden is caused by entire population. And this expenditure includes chronic disease and disability. some less cost-effective services that should be re- It is assumed that the disease burden would de- garded as discretionary. Paying for a minimum cline by the same share in China as in a middle- package of essential clinical care would require ex- income country because China has already sub- penditure of an additional $16 billion to $24 billion, stantially reduced the burden from the diseases doubling the current expenditure level. If total addressed by the package. About 30 million spending did not change, this would imply a re- DALYs could be gained, the equivalent of 930,000 duction by about half in what is now spent on infant deaths prevented. The cost per capita would discretionary services. be the same as in low-income countries. For some countries, paying for the proposed Full coverage with the minimum package would package of services poses a severe challenge. In cost an estimated $22 billion in low-income coun- fact, in the poorest countries total current public tries, $14 billion in China, and $26 billion in spending of $6 per person is about $6 short of the middle-income countries. The total cost would be cost of the package. Total per capita spending, in- about $62 billion, or $15 per person in the develop- 67 Table 3.3 Total cost and potential health gains of a package of public health and essential clinical services, 1990 Cost per ca pit a Total cost (billions Reduction in disease burden Country group and package component (dollars) of dollars) Percent Millions of DALYs Low-income countrjes 12 22 32 226 Public health 4 8 8 57 Essential clinical services 8 14 24 170 Chinab 12 14 15 30 Public health 4 5 4 8 Essential clinical services 8 9 11 22 Middle-income countrjes 22 26 15 45 Public health 7 8 4 12 Essential clinical services 15 18 11 33 All developing countries 15 62 25 301 Public health 5 21 6 77 Essential clinical services 10 41 19 225 Estimated from data for Bangladesh, Egypt, India, Indonesia, Pakistan, and Sub-Saharan Africa. China is shown separately because its cost per capita is assumed to be that of a low-income country but its percentage reduction in disease burden is assumed to be that of a middle-income country. Estimated from data for Latin America and the Caribbean, Other Asia and islands except for Indonesia and Bangladesh, and the Middle Eastern crescent except for Egypt and Pakistan. Source: World Bank calculations. ing world as a whole. This figure includes what covered by a combination of greater public spend- countries are already spending on the services in ing, increased donor contributions, and more pri- the package, estimated at $20 billion to $30 billion. vate expenditure by those able to pay. Shifting The incremental cost would therefore be only $30 some part of the cost to higher-income con- billion to $40 billion a year. sumersfor example, through private or social in- The gain in health would be about 300 million surancewould allow for an expansion of the DALYs, which is equivalent to 9.3 million infant package or a reduced burden of public deaths. Universal application of the package expenditure. would therefore yield about the same health gain There are several reasons why developing coun- as eliminating nearly all the infant deaths in the tries fail to allocate sufficient resources to cost- world today. These gains could be achieved for an effective health interventions. Health providers average cost of about $50 per DALY for the public often lack incentives to provide cost-effective ser- health measures and about $100 per DALY for the vices. Doctors' pay, promotion, and professional minimum package of essential clinical services. recognition are enhanced by specialization and by The cost per DALY of the interventions in the the use of expensive new medical technologynot package ranges from less than $5 to more than by serving as public health doctors or district $200; average costs also reflect those public health medical officers in poor rural areas. Badly de- measures that do not improve health directly but signed government salary schedules and price sys- that are essential to the functioning of a health tems may exacerbate this trend. In China hospitals system. currently have a strong incentive to use new diag- In both low-income and middle-income coun- nostic and therapeutic technologies, for which full tries the marginal cost per DALY would be less costs can be charged, instead of older and less ex- than the average cost because all the fixed costs of pensive technologies for which government-set infrastructure are included in these estimates and prices are far below actual costs. there would be spare capacity for producing small Consumer demand for cost-effective services is additional amounts of most services. Because of often weak. This may reflect lack of information. joint costs, it is difficult to separate the cost per In rural Africa, for example, goiter and impaired DALY for every intervention in the package. mental abilities from iodine deficiency have in In middle-income countries the package could many places become accepted as the normal state be entirely financed by reallocating current public of affairs. Low demand may also reflect deficien- spending. In poor countries there would be a cies in supply. Most cost-effective interventions shortfall of about $10 billion a year, which could be can be delivered at primary care sites, but in poor 68 countries such as Burkina Faso and Mali, more health care center or doctor's office. In Indonesia than half of the population lives more than 10 kilo- in 1991, for example, rural households in the top meters from the nearest primary care center. income decile were three times more likely to live At a more fundamental level, the distribution of in a village with a health center than those in the political power explains much of the misallocation bottom decile. of government resources for health. The urban Partly because of difficulties in access, the poor population is better organized than rural groups in developing countries generally consume fewer and more vocal in demanding health facilities and health services. Household surveys from Sub- services. Similarly, middle-class workers in wage Saharan Africa and Latin America demonstrate employment, who frequently belong to powerful clearly that among people who report themselves trade unions, are more effective than self-em- to be sick, those in urban areas seek and obtain ployed farmers and workers in the informal sector medical care more often than those in rural areas, in lobbying for government-subsidized health and the wealthy contact a care provider more often benefits. Health professionals are also often better than do the poor. The differences can be large: in organized than the population they serve, and in Côte d'Ivoire in the mid-1980s, for example, an promoting their own interests they may make the urban household was nearly twice as likely to seek health system less efficient. Despite these prob- care as a rural household (60 versus 36 percent), lems, many countries have succeeded in dramati- and within the rural population a family in the top cally improving the health of their people. This income quintile was almost twice as likely to seek success can be accelerated, as newly available in- care as a family in the bottom quintile (44 versus 23 formation makes it clear how costly misallocation percent). is and how much health can be gained for rela- A study of Peru showed similar inequalities tively modest levels of spending. among geographic regions and educational groups. There was little variation in self-declared Equity in health status, utilization, and finance illness, but the likelihood of obtaining medical care when sick was nearly three times higher in some Data on health status, physical access to health parts of the country than in others. Regional dif- services, consumption of health care, the distribu- ferences in immunization rates were highest for tion of the financial burden of health care spend- uneducated mothers, whose children were only ing, and public expenditures for health all tell the one-third as likely to be fully immunized as the same story of severe inequities in developing children of women with secondary schooling. The countries. In Bangladesh, for example, the infant 36 percent of all the self-declared sick who lived in mortality rate for the urban poor (13.4 percent) is the capital city accounted for 53 percent of all Min- nearly twice the urban average (6.8 percent) and istry of Health ambulatory consultations, 41 per- about 50 percent higher than the average rate for cent of hospital admissions, and 47 percent of all the entire country (10 percent). In China, despite public expenditure attributable to care for individ- remarkable overall progress in health (infant mor- ual patients. At the other extreme was Piura, a tality fell from 20 percent in 1950 to 4.6 percent in poor mountainous region with 10 percent of the 1982), there is considerable geographic variation, sick but only 4 percent of public spending and of which is strongly related to income. Poor regions consultations. such as Yunnan, Xinjiang, and Tibet have infant Inequity in public spending for health both ac- mortality rates of more than 7 percent, compared counts for and reflects marked inequalities in ac- with less than 2 percent in more affluent Beijing, cess to and utilization of care. In Indonesia, for Guangdong, Shanghai, and Tianjin. To take an- example, despite significant investments in lower- other example, in Kenya the probability of a level health facilities in the 1980s, only 12 percent child's dying before age 2 varied among ethnic of public spending for health in 1990 went for ser- groups from 7.4 to 19.7 percent, and in Cameroon vices consumed by the bottom 20 percent of these probabilities ranged from 11.6 to 20.5 households, while the top 20 percent obtained 29 percent. percent of the government subsidy. This bias in The poor also have considerably worse access to favor of the wealthy was mainly a result of the health care. A number of surveys show that low- distribution of government spending for hospital income households, especially in rural areas, have inpatient and ambulatory care, services that were to travel considerably farther or longer to reach the used more frequently by the rich. Much more un- first level of referral services, usually a primary equal situations can be found in many countries 69 that concentrate government spending even more Household surveys systematically show that on high-level facilities. people choose whether to seek care and which The few countries in which public spending on provider to consult on the basis of many factors health is biased toward the poor show that govern- hours of service, travel time or cost, waiting time, ment policies can help reduce inequities in access availability of doctors or of drugs, and how pa- and health status. In Malaysia the government has tients are personally treated. The time required to followed a pro-poor policy since the 1970s, with get care can be valued according to local wages the lowest-income groups receiving a larger share and treated as a cost of service together with of public subsidies for health than the middle class money payments. On this basis, free public medi- and the wealthy. Similarly, in Costa Rica govern- cal care often is more costly than unsubsidized pri- ment spending for health has continued to favor vate care for which patients do not have to travel the poor, despite economic shocks and a major so far or wait so long. It is not surprising that, adjustment program in the 1980s that entailed cuts under these circumstances, even poor people ex- in public expenditure. In 1988 about 30 percent of press their dissatisfaction with public services by government spending for health went to the poor- paying for a great deal of private outpatient care. est 20 percent of households and just over 10 per- In both El Salvador and the Dominican Republic cent of spending to the richest 20 percent. The residents of the poorest quintile of the capital city poverty-oriented pattern of public spending for obtain more than half their ambulatory care from health in Costa Rica can be explained largely by private physicians. Although the price of private the high degree of coverage by the social security care to the poorest quintile is, on average, half that health system (the entire population is covered in for patients in the richest quintile, it is still fifteen principle, even though only 63 percent of the times higher than Ministry of Health fees. Differ- working population contributes) and the relatively ences in waiting timeone hour for private pa- equal access to and quality of care enjoyed by all tients as against two and a half hours at Ministry Costa Ricans. It also helps that the wealthy get of Health facilitiesaccount for much of this effect. most of their outpatient care from the private Sensitivity to price and travel time is also found sector. in rural Peru and Côte d'Ivoire. But private hospi- Consumer satisfaction with health care tal care is still much too expensive for the poor; even those who use private doctors go to public How satisfied people are with their own health hospitals. The excessive use of hospital care in re- and their health care can be only partly explained lation to ambulatory services often seen in public by objective criteria; subjective expectations mat- health systems partly reflects dissatisfaction with ter. People can also be pleased with their own the high cost in time and the perceived poor qual- health care and dissatisfied with their country's ity of ambulatory care. In the absence of incentives health system as a whole. A comparison of ten to improve lower-level facilities and service, this OECD countries with different health systems overuse reinforces the tendency to concentrate re- found that in eight countries public satisfaction sources on hospitals, urban areas, and less cost- was related to the level of spending. Canada, with effective interventions. the second-highest expenditure, had the highest The importance of public satisfaction with a satisfaction rating, and people were generally bet- health care system raises two issues for the pack- ter satisfied with the costlier health systems of age of publicly financed services proposed here. France, Germany, and the Netherlands than with First, it suggests that quality can be maintained the lower-spending systems in Australia, Italy, Ja- only if coverage is broad enough. Services de- pan, and the United Kingdom. Both very high ex- signed only for the poor will almost inevitably be penditure and great dissatisfaction were found in low in quality and will not receive the political sup- the United States. The study also showed that port necessary for adequate provision. This is a having a unified national health system did not difficult political issue because it may be hard to guarantee a high level of satisfaction. In most maintain equity and control costs if coverage is countries 30 to 50 percent of those polled sup- universal. The proper balance between more care ported "fundamental changes" in the health sys- for fewer people and the same amount of care for tem. In Italy and the United States many people more people depends on ensuring that the poor thought that such changes would not suffice and have access to the same quality of care as everyone that the health system should be "completely else and on limiting public finance to cost-effective rebuilt." services for which there is a sound rationale. Sec- 70 ond, reform of public provision alone, important more cost-effective care, but they also cannot as it may be, may have much less effect on health provide equitably what care they do offer be- outcomes, costs, and satisfaction than reforms that cause facilities will inevitably be geographically also try to stimulate competition and improve peo- concentrated. ple's access to a variety of providers. One of the principal responsibilities of govern- ment is to match the available instruments of poi- Matching means and ends icythe levers the public sector actually controls to the objectives. Much of governments' failure to The objectives of a health system are to improve achieve better health outcomes derives not from outcomes, control costs, increase equity, and sat- the wrong choice of objectives but from the wrong isfy users. Policy instruments, however, do not choice of instrumentsin particular, from too correspond to individual objectives. What govern- much reliance on direct provision of care and cen- ments actually do is build facilities, buy equipment tral control of health facilities and too little use of and supplies, hire and train people, set fees or the financial, informational, and regulatory instru- other service conditions, regulate providers and ments at the disposal of the government. These insurers, disseminate information, determine instruments are particularly important for improv- overall policy, and maintain surveillance of disease ing performance in the private market. When gov- conditions or other variables. Misallocation and in- ernments pay for health care in addition to regulat- equity are caused by mistakes in deciding what ing it, they have a further responsibility to provide facilities to build, where to locate them, how to value for money by ensuring that public resources staff them, and what services to provide. If gov- go first to cost-effective public health and essential ernments spend too much on tertiary care, for ex- clinical services so as to buy the largest health gain ample, not only can they not adequately finance possible. 71 Public health Health services interact with households in two ated, in collaboration with UNICEF, a global effort fundamentally different ways. Public health pro- to prevent a range of childhood diseases by immu- grams strike against health problems of entire nization. The EPI now reaches about 80 percent of populations or population subgroups. Their objec- children in developing countries and averts an es- tive is to prevent disease or injury and to provide timated 3.2 million deaths a year at a cost of $1.4 information on self-cure and on the importance of billion a year. seeking care. Clinical services respond to demand Population-based health services such as the EPI from individuals. They generally seek to cure or to rely on personnel with limited training to provide ease the pain of those already sick. This chapter drugs, vaccines, or specific health services directly discusses public health; Chapter 5 turns to clinical to specific populationsin schools, at worksites, services. or in households. Government finance for such Public health programs work in three ways: they programs is justified because the objective is usu- deliver specific health services to populations (for ex- ally to provide services to all in a community, be- ample, immunizations), they promote healthy be- cause the services create externalities or indirect havior, and they promote healthy environments. benefits, and because the diseases they typically Governments play a leading role, and provision of combat are particular problems for the poor. Three information through public education is a central types of interventions are immunization, mass feature of most programs, especially those de- treatment for worms and other infections, and signed to change behavior. But difficult choices screening and referral. Information, education, have to be made about the best use of public and communication are critical to many such pro- money. The Expanded Programme on Immuniza- grams, both to attract participation and to achieve tion (EPI), described below, is highly cost-effec- durable change in behavior. tive, at about $25 per DALY gained, but not all programs offer such good value for money. This Immunization chapter examines six particularly cost-effective public health services in the realms of population- Vaccines to prevent tuberculosis, measles, diph- based services (including immunization), nutri- theria, pertussis, tetanus, and polio have revolu- tion, fertility, tobacco and other drugs, the house- tionized preventive medicine over the past two hold and external environment (including control decades. Costs are less than $10 per DALY gained of insect vectors of disease), and AIDS. Public for measles immunization and less than $25 for a health packages in developing countries should in- combination of polio plus DPT (diphtheria, per- clude components in most or all of these six areas. tussis, and tetanus). These vaccines, together with BCG immunization against tuberculosis and lep- Population-based health services rosy and immunization of pregnant women against tetanus, form the EPI. In 1979 the World Health Organization declared As a result of the EPI, the proportion of children that smallpox had been eradicated. It then initi- immunized rose from less than 5 percent in 1977 to 72 20 to 30 percent by 1983. By 1990 coverage with paigns continue to be justified. In areas with better polio, DPT, and measles vaccines had reached ap- infrastructure, routine services make more sense. proximately 80 percent of all children, and about An ambitious current goal, established in 1988 35 percent of pregnant women were receiving tet- by WHO's governing body, the World Health As- anus toxoid. The lowest vaccine coverage is re- sembly, is to eradicate polio by 2000. Current ported in Sub-Saharan Africa. trends suggest that even if eradication is not Had vaccination coverage remained at the low achieved on that schedule, it will be soon there- levels of the 1970s, as many as 120 million DALYs a after. And substantial success has already been year (the equivalent of 23 percent of the global achieved: there has been no naturally occurring burden of disease among children under age 5 in case of polio in the Western Hemisphere since Au- 1990) would be lost to diseases preventable by the gust 1991. EPI. At current levels of vaccination coverage, Two extensions to the EPI appear to be justified. these diseases cause a loss of 55 million DALYs, or First, coverage should be extended, probably to 95 10 percent of the disease burden among children percent of all children born. The costs of expand- under age 5 (Table 4.1). ing coverage are relatively high, but so are the The cost of fully immunizing a child in low- gains. Those not covered at present often lack any income countries is about $15, with a range of $6 to health services and are disproportionately vulner- more than $20, depending on the prices of labor able to the diseases. Second, it makes sense to and other local inputs. Reducing the number of include additional items in the package: hepatitis B contacts needed to immunize each child fully and yellow fever vaccines for selected countries could cut costs dramatically, by as much as 70 per- and vitamin A and iodine supplements in regions cent if only one contact instead of the current five where deficiency of these micronutrients is highly were needed. This prospect depends on the suc- prevalent. If micronutrients are not delivered cess of ongoing research efforts. Technical im- through the EPI, some other vehicle must be provements in the cold chain (by which vaccines found for reaching very young children. Adding are kept refrigerated until use), good administra- these two vaccines and two micronutrients to the tion, widespread deployment of delivery teams, EPI (EPI Plus) would improve health substantially, and effective social mobilization efforts can also particularly in the poorest households, for a mod- contribute to dramatic cost reductions. In the est increase of about 15 percent in the cost of Gambia the cost of immunizations fell from $19 in reaching each child with complete services (vac- 1982 to $6 in 1988. Costs also depend on the immu- cine and micronutrients). Table 4.2 summarizes nization strategy: campaigns achieve high initial the estimated costs and health benefits of the EPI coverage, but routine services are more cost-effec- Plus cluster in two different settings. Total annual tive. In Ecuador campaigns cost $66 per DALY costs range between $2.2 billion and $2.4 billion for gained compared with $30 for routine services. Be- EPI Plus, or less than 2 percent of the public health cause many countries lack the infrastructure to de- expenditure of developing countries. Expanding liver vaccines routinely in remote rural areas, cam- coverage from 80 to 95 percent would probably Table 4.1 Burden of childhood diseases preventable by the Expanded Programme on Immunization (EPI) by demographic region, 1990 Share of the Burden per 1,000 Burden (millions total burden in children children under age 5 Region of DALYS per year) under age 5 (percent) (DALYs) Sub-Saharan Africa 23 15 242 India 16 12 137 China I 3 8 Other Asia and islands 7 10 81 Latin America and the Caribbean .1 3 18 Middle Eastern crescent 7 10 86 * Formerly socialist economies of Europe I * Established market economies I World 55 10 87 * Less than 1. Note: The EPI includes immunizations for pertussis, polio, diphtheria, measles, tetanus, and tuberculosis. These estimates exclude the burden from tuberculosis because most of it falls on adults. Source: Calculated from Murray and Lopez, background paper. 73 Table 4.2 Costs and health benefits of the EPI Plus cluster in two developing country settings, 1990 Middle-income countries Low-income countries (low mortality and Costs and benefits (high mortality and fertility) medium fertility) Cost per capita (dollars) 0.5 0.8 Cost per fully immunized child (dollars) 14.6 28.6 Cost per DALY gained (dollars) 12-17 25-30 Cost per DALY gained as a percentage of income per capita 0.14 0.03 Potential health gains as a percentage of the global burden of disease 6.0 1.0 Note: Figures are based on 95 percent coverage. a. Income per capita in 1990 was assumed to be $350 for low-income countries and $2,500 for middle-income Countries. Source: World Bank data and authors' calculations. increase annual costs by between $500 million and tion can be reduced by environmental improve- $750 million. In low-income countries the increase ments, especially sanitation, but where this is im- in coverage would reduce by about 6 percent the practical or unaffordable, it is cost-effective to global disease burden. repeat the therapy at regular intervals. The benefits of individual treatment can be sig- Mass treatment for parasitic worm infection nificantly enhanced by community-wide treatment which, by lowering the overall levels of contam- The most common intestinal wormsround- ination of the environment with infective stages of worms, hookworms, and whipwormseach infect the worms, slows the rate of reinfection. Treat- between 170 million and 400 million school-age ment programs targeted at the most heavily infec- children annually. Schistosomiasis infection, also ted group (school-age children) reduce infection caused by parasitic worms, affects almost 100 mil- immediately both among those treated and in the lion school-age children annually. The immediate rest of the population. Treatment through schools effects of infectionincluding failure to thrive, also allows delivery at relatively low cost: a pro- anemia, and impaired cognitioncan now be rap- gram in Montserrat was estimated to cost less than idly reversed by low-cost, single-dose oral ther- $1.50 per person for a cycle of eight treatments. A apy. Studies of single-course treatment of school- program managed by a nongovernmental organi- children with hookworm or schistosomiasis in zation in Jakarta initially cost $0.74 per capita per Kenya, with worm-induced disease in India, and year, but after expansion to almost 1,000 schools with trichuriasis in the West Indies showed re- the costs fell to $0.26. Such programs are ex- markable spurts in growth and development in all tremely cost-effective, at $15 to $30 per DALY the populations studied, including the large per- gained. In light of this cost-effectiveness and the centages of children with asymptomatic infections. burden of disease addressed, the Rockefeller And treatment also appears to have improved cog- Foundation and the UNDP are initiating a major nitive development. program to document and explore the potential of Curing worm infections is simple with inexpen- school-based health interventions that focus on sive modern drugs such as albendazole and prazi- deworming and provision of micronutrient quantel because it is not necessary to determine supplements. which species are present. Furthermore, the high level of safety of these drugs has led WHO to de- Mass screening and referral velop protocols for their use on a mass basis (where a high prevalence of infection exists) and Mass screening for disease control involves the ex- by providers who are not medically traineda amination of asymptomatic individuals to identify combination that makes for high cost-effective- and treat those affected by disease. Although this ness. Treatment usually cures the current infec- method has been used to control some infectious tion, but in endemic areas children will inevitably diseases, such as tuberculosis, it is mostly used become reinfected. A return to pretreatment levels for noncommunicable diseases. Mass screening of infection typically takes about twelve months makes sense for highly prevalent diseases that can for roundworm and whipworm and twenty-four be cured by early treatment, especially when la- months or more for hookworm. Rates of reinfec- tency periods span many years. An example is cer- 74 vical cancer, which is the leading cause of death of adding energy or protein to an initially poor from cancer among women in developing coun- diet, the effect of deficiencies in either or both tries, accounting for 150,000 deaths each year. components is combined under the term "protein- Screening with Papanicolaou (Pap) smears is com- energy malnutrition." Foods rich in protein, such mon in industrial countries, but attempts to repli- as soybeans and animal products, tend to be rela- cate those efforts in developing countries have tively costly per unit of energy, and low-cost rarely been successful. Such programs could, sources of energy such as cassava tend to be ex- however, be made cost-effective by the use of a pensive per unit of protein. Because food takes up simplified design that targets women over 35, much of a poor household's budget, choosing the screens only every five to ten years, and uses inex- protein-energy balance that is right for health can pensive outpatient treatment (such as freezing ab- be difficult. normal cells) for severe precancerous conditions. Protein-energy malnutrition raises the risk of When backed up with good follow-up services, death and may reduce physical and mental capac- such interventions are cost-effective, at $150 to ity. Worldwide, about 780 million people are esti- $200 per DALY gained. mated to be energy deficient according to WHO standards. It is not known how many of them are Diet and nutrition also protein deficient, or how many people who get enough energy may still suffer from shortage Eating well is necessary for good health. Either of protein. Exploratory studies of the determinants directly or in association with infectious diseases, of human growth suggest that at the margin, the inadequate diets account for a large share of the importance of additional protein may be greater world's disease burden, including as much as a than is recognized. Malnutrition is not syn- quarter of that among children. Much of this suf- onymous with hunger because people who have fering stems from poverty-related underconsump- become accustomed to a deficient diet may not tion of protein and energy, but equally important consider themselves hungry. If malnutrition is are deficiencies of key micronutrientsiodine, widespread in the community, underweight and vitamin A, and ironfrom which children and lethargic children look normal to parents who do women suffer disproportionately. Increasing the not know how healthy children behave. incomes of the poor is the most effective means of Iron deficiency is the most common micro- reducing protein-energy malnutrition, but govern- nutrient disorder. It reduces physical productivity ments can play an effective direct role through nu- and children's capacity to learn in school. By re- trition education, measures to increase consump- ducing appetite, it may diminish children's intake tion of micronutrients, and reduction in diarrheal and growth. Women suffer especially because and parasitic infections among children. Public ac- menstruation and childbearing raise their need for tion is also essential for preventing crop failures iron, and anemia, a shortage of iron in the blood, from leading to famines. increases the risk of death from hemorrhage in childbirth. The problem is worst in India, where 88 Malnutrition and ill health percent of pregnant women are anemic. Almost 60 percent of women are anemic in other parts of Low height for a given age, or stunting, is the most Asia, but the proportion does not exceed 40 per- prevalent symptom of protein-energy malnutri- cent in China, Africa, or Latin America. Anemia tion; approximately 40 percent of all two-year-olds affects 15 percent of pregnant women in the estab- in developing countries are short for their age (see lished market economies. Appendix table A.6). The prevalence of stunting Iodine deficiency causes mental retardation, de- may be as high as 65 percent in India; it is more layed motor development, and stunting, as well as than 50 percent in Asia other than India and China neuromuscular, speech, and hearing disorders. It and about 40 percent in China and Sub-Sahran is the leading preventable cause of intellectual im- Africa. Stunted children are often also under- pairment in the world. Cretinism from iodine defi- weight or have low weight for their age. Wasting ciency affects about 5.7 million people, and lack of (low weight for a given height) is less prevalent- iodine causes another 20 million to be mentally 11 percent or less worldwide except in India, retarded. where it reaches 27 percent. Vitamin A deficiency causes varying degrees of Diets must contain both energy and protein. Be- vision loss and is the primary cause of acquired cause little is known about the relative importance blindness in children. It also increases the severity 75 Table 4.3 Direct and indirect contributions of malnutrition to the global burden of disease, 1990 (millions of DALYs, except as specified) Latin America Middle Formerly Established Sub-Saha ran Other Asia and the Eastern socialist market Type of malnutrition Africa India China and islands Caribbean crescent economies economies World Direct effects Protein-energy malnutrition 2.2 5.6 1.7 0.9 1.0 1.0 0.2 0.2 12.7 Vitamin A deficiency 2.2 4.1 1.0 2.5 1.4 0.5 0.0 0.0 11.8 Iodine deficiency 1.7 1.4 1.0 1.3 0.5 1.4 0.0 0.0 7.2 Anemia 1.0 4.5 2.7 2.3 1.0 1.5 0.4 0.6 14.0 Total direct 7.0 15.5 6.3 7.0 3.9 4.5 0.6 0.9 45.7 Total DALYs per 1,000 population 13.8 18.3 5.6 10.3 8.9 8.9 1.7 1.1 8.7 Indirect effects (minimum estimate) Mortality from other diseases attributed to mild or moderate underweight' 23.6 14.9 3.3 8.0 2.4 8.0 0.0 0.0 60.4 Mortality from other diseases attributed to vitamin A deficiencyb 13.4 14.0 1.0 7.0 1.8 2.0 0.0 0.0 39.1 Based on the global burden of disease (GBD) attributable to deaths from tuberculosis, measles, pertussis, malaria, and diarrheal and respira- tory diseases in children under age 5; in developing countries 25 percent of those deaths are attributed to mild or moderate underweight. Based on estimated deaths attributable to vitamin A deficiency in the age groups 6-Il months and 1-4 years. These account for, respectively, 10 and 30 percent of all such deaths in high-risk countries and for 3 and 10 percent of all such deaths in other countries. Thirty lost DALY5 are attributed to each child death; losses are redistributed to the regional classification used in this Report. Source: For GBD calculations, Appendix B; for estimate of mortality from underweight, Pelletier 1991; for estimate of mortality from vitamin A deficiency, Humphrey, West, and Sommer 1992. Box 4.1 Women's nutrition Women suffer more than men from iron deficiency anemia, from stunting caused by protein-energy mal- normalities. A significant proportion of pregnancies end in poor maternal or infant health as a direct conse- 1 nutrition, and from iodine deficiency. The largest gap quence of maternal malnutrition. is for iron deficiency anemia, which affects 458 million Iodine and vitamin A deficiencies tend to be lo- adult women but 238 million men. About 450 million calized rather than widely distributed and could be vir- women are stunted because of protein-energy malnu- tually eliminated through targeted, sporadic interven- trition, compared with 400 million men. Iodine defi- tions, given a reasonable health infrastructure and a ciency also affects substantial numbers of women, high level of political will. Anemia and protein-energy probably more than for men. Corneal lesions and malnutrition, by contrast, affect much larger numbers blindness caused by vitamin A deficiency afflict both of women and require more continuous intervention. sexes equally, but deficiency as such is twice as com- Distribution of a regular supply of ferrous sulfate tab- mon for girls as for boys. Women's nutritional prob- lets can prevent or cure anemia among pregnant and lems are worst in South Asia, where prevalences of lactating women. Such efforts should include all anemia, protein-energy malnutrition, and vitamin A women of reproductive age, certainly where the preva- deficiency are the highest in the world and where, as a lence of anemia among women in general exceeds 50 result of widespread discrimination, girls and women percent. To reduce protein-energy malnutrition, much suffer disproportionately. must be done outside the health sector toward making Small pelvic size among stunted women increases more food available to households, increasing employ- the risk of maternal and infant mortality, as does ma- ment opportunities for women, decreasing the time ternal anemia. Iodine-deficient mothers give birth to and energy costs of women's home production, and more infants with cretinism and other congenital ab- reducing discrimination against women and girls. 76 of and mortality from a variety of infections, espe- Childhood mortality drops sharply as ,iutritional cially measles and diarrhea. WHO calculates that status improves. 13.8 million children have some degree of eye damage because of vitamin A deficiency; of these, 250,000 to 500,000 go blind every year, and two- Figure 4.1 Child mortality (in specific age thirds of the blinded children die. Both vitamin A ranges) and weight-for-age in Bangladesh, and iodine deficiency are particularly common in India, Papua New Guinea, and Tanzania Asia and Sub-Saharan Africa. These four diseases of malnutrition caused a di- rect loss of almost 46 million DALYs in 1990, or 3.4 percent of the global burden of disease (Table 4.3). (The estimates do not include the health damage from deficiencies of other micronutrients. Calcium deficiency may be the most important of these; it causes bone deformities and slows skeletal growth 500 4, in children, and it may contribute to osteoporosis in the elderly.) The estimated burden is slightly 269 larger for females than males because anemia af- fects mostly women ages 15-44 (Box 4.1); anemia accounts for 1.3 percent of the total female disease ,J 119 burden but for 24 percent among women in the reproductive ages. Children under 5 are the princi- 47 pal victims of vitamin A deficiency, iodine defi- ciency, and protein-energy malnutrition. The nu- o 18 tritional disease burden for young children is 32 0 million DALYs, or 6 percent of their total burden of illness. 6.7 The total impact of malnutrition on health is much larger, however, because mild or moderate 2.5 protein-energy malnutrition and micronutrient de- 55 65 75 85 ficiencies (as well as overcorisumption of energy, Weight-for-age fat, salt, and sugar) are risk factors for illness and (percentage of NCHS reference median) death. Studies in Asia and Africa consistently show that mild to moderate stunting or under- - Tanzania, 6-30 months weight in children raises the risk of death (Figure - Papua New Guinea, 6-30 months 4.1), contributing to 25 to 50 percent of childhood Matlab, Bangladesh, 12-59 months mortality. The greatest risk occurs for children in Punjab, India, 12-36 months their second year, after they are weaned. Malnour- ished children die principally from measles, diar- Note: NCHS, (U.S.) National Center for Health Statistics. rheal and respiratory disease, tuberculosis, per- The vertical axis shows the child mortality rate, CMR, in tussis, and malaria. Child deaths from these log-odds ratio form, log[CMR/(1000 - CMR)l. Thus, for a child in Punjab weighing 60 percent of the reference diseases cost 231 million DALYs, making the total median, the probability of dying between 12 and 36 burden attributable to malnutrition at least one- months is 18 per 1,000. fourth that amount, or 60 million DALYs. Source: Pelletier 1991. Vitamin A deficiency, too, raises the risk of death from other causes. Of the 8 million deaths of children with vitamin A deficiency that occur ech year, between 1.3 million and 2.5 million might be prevented by eliminating the vitamin deficiency, the total direct and indirect damage from malnutri- for a gain of 39 million to 74 million DALYs. Dam- tion is at least 20 to 25 percent of the disease bur- age from being underweight and from vitamin A den in children. deficiency cannot be added together because many Links between nutrition and growth in child- children suffer from both problems. Nonetheless, hood persist into adulthood. Both height and 77 weight affect the risk of adult mortality. For men part of the year when malnutrition is most preva- and women at all ages, greater height is associated lent. Small variations in diet can be fatal to chil- with greater survival. Stunted adults are at par- dren already at risk. ticularly high risk of cardiovascular disease and The extreme form of this risk is widespread fam- obstructive lung disease. If the relative risk of ine as a result of a breakdown in food production, death associated with stunting is the same in low- food distribution, or the flow of income with income countries as for richer populations, ap- which people buy food. Famines occurred in proximately 300,000 adult female deaths between China in 1959-61, in Bangladesh in 1974, in Ethio- the ages of 15 and 59 can be attributed to stunting. pia and the Sahel in 1973-74, in Ethiopia and Somalia in the 1980s, and in Somalia and Sudan in Sources of malnutrition the early 1990s. As many as 30 million people are believed to have died in the Chinese famine and Both food consumption and communicable dis- hundreds of thousands in the recent famines in ease affect nutritional status by way of a "malnu- Sub-Saharan Africa. A relatively small number of trition-infection complex." Food consumption de- people die from outright starvation; many die of pends both on people's capacity to acquire food infectious diseases, to which people weakened by and on their knowledge of how to choose a nutri- hunger are especially susceptible. tious diet. For infants the chief determinant of nu- Public action is critical in preventing a food crisis tritional status is whether they are exclusively from becoming a famine. A combination of actions breastfed for at least the first four to six months of is required to ensure that food is available in fam- life. In southern Brazil infants who were not ine areas (through both market and nonmarket breastfed were eighteen times as likely to die from mechanisms) and to sustain the incomes of vulner- diarrhea and three times as likely to die from respi- able households (through public employment or ratory illness as breastfed babies, both because other transfers). This is particularly difficult when they got less to eat and because of increased risk of there is a breakdown in order: the major African infection. After six months children need solid famines of the past decade were mostly associated food even if they are still breastfed. The composi- with war. tion and hygiene of this food are crucial to contin- Hunger and crowding into refugee camps facili- ued good health. tate the spread of infectious disease and raise the risk of death from it, particularly when such INCOME AND FOOD SECURITY. Chronic malnutri- camps are first established. Control of communica- tion is mostly a consequence of poverty. Higher ble disease is as crucial as the provision of food or income allows people to buy a more balanced diet, of money to buy food. Even when refugee popula- as well as better hygiene and medical care. In In- tions are protected from starvation, they are often donesia during 1984-87 rising incomes translated exposed to micronutrient deficiencies because they into reduced malnutrition in nearly all fifty-two are dependent on just a few foodstuffs. In recent regions of the country. The fraction of families eat- years there have been outbreaks of scurvy (vitamin ing less than 2,200 calories per person per dayan C deficiency) in Ethiopia and Somalia, pellagra energy intake adequate for only light physical ac- (niacin deficiency) among Mozambican refugees in tivitydeclined only 2 percentage points, but the Malawi, and ben-ben (thiamine deficiency) among decline was 9 percentage points at 1,800 calories Cambodian refugees in Thailand. In nonfamine and 26 percentage points at 1,400 calories. Con- conditions these diseases make no contribution to versely, increases in food prices in Côte d'Ivoire the world's burden of disease. during the 1980s reduced the weight of both chil- Beyond ensuring food distribution and control- dren and adults. Nutrition is also affected by who ling the diseases that can easily become epidemics in the household controls the money; women's in conditions of social and sanitary breakdown, income is more likely than men's to be spent on governments have two overriding responsibilities better nutrition. in famines. The first is to recognize the early signs Chronic food insecurity for poor people is often of trouble and act before large numbers of people made worse by seasonal fluctuations in availability have become destitute. The second is to allow free and prices. In India and the Philippines temporal flow of information about conditions during the variation in children's food intake is greatest famine so that relief agencies and others can react. among poor households, and severely malnour- Hiding the extent of a disaster only makes it ished children are more likely to die during that worse. 78 DIET AND DISEASE. Some children receive so lit- nicable diseases but can also increase diseases of tle to eat that getting more food is by far the most dietary excess. As diets change, usually to include important means of improving their growth. But a smaller proportion of complex carbohydrates for those with a barely adequate diet, controlling and more sugar and animal fat, people become infectious disease can be as important as getting more susceptible to cardiovascular disease and to more food. (This is the most common situation in cancers of the colon, prostate, and breast. Obesity poor countries.) In the absence of diarrhea, studies becomes more prevalent and increases the risk of have found little difference in growth in children premature death, particularly from cardiovascular up to thirty-six months of age despite significant diseases and diabetes. Increased sugar consump- differences in energy intake. Children with both tion contributes to dental caries and may raise the low energy intake and diarrhea grow less. Diar- risk of diabetes. High salt intake increases hyper- rhea has little effect on the growth of adequately tension, raising the risks of stroke and cardio- fed children. However, a diet adequate for healthy vascular disease. For people eating a modern children may be inadequate under the additional "Western" diet, a 60 percent reduction in lifetime demand imposed by infection. salt intake would reduce the risk of death from Control of parasitic worms can also help im- coronary disease at age 55 by 16 percent and the prove nutrition for older children. Hookworm and risk of death from stroke by 23 percent. The share other intestinal worm infections cause anemia, of the disease burden attributable to these dietary and roundwormsthe most prevalent of all changes is unknown but may be quite large. wormscompete with the host for food. All these The diseases of overnutrition, which strike peo- infections may suppress appetite and reduce food ple later in life, may seem unimportant compared intake. More than 1 billion people are infected with those of undernutrition. The burden of car- with one or more of these parasites, and about 100 diovascular disease and of some cancers, however, million of them suffer from stunting or wasting. is already important in Brazil, China, and other Recent intervention studies show rapid spurts in developing countries; the demographic and epi- height after children are treated for worm infec- demiological transitions documented in Chapter 1 tions, suggesting that these infections may be sig- will accelerate this trend. Treatment of chronic dis- nificant contributors to malnutrition and that it ease is often expensive or ineffective, so modifying may be easier to reverse stunting in older children diet and other risks is the best way of avoiding than was previously believed. growth in the burden from these diseases and in Malaria is the other major infection leading to unnecessary health care expenditure. malnutrition, particularly anemia. It is an impor- Other interventions for better nutrition tant risk factor in much of Africa. Control of ma- laria has led to substantial decreases in the preva- Six other interventions, in addition to control of lence of anemia. infectious diseases, may help to reduce malnutri- tion: nutrition education, control of intestinal LACK OF NUTRITIONAL KNOWLEDGE. People may parasites, micronutrient fortification of food, mi- eat more poorly than their incomes allow because cronutrient supplementation, food supplementa- of ignorance. This is most true for vitamin A; defi- tion, and food price subsidies. Control of parasites ciencies persist although almost everywhere in the was discussed above; the other five interventions world foods rich in vitamin A can be grown at low are reviewed here. cost in family gardens or commercially The effect of ignorance cannot be quantified, but it is striking NUTRITION EDUCATION. Inducing behavioral that it often increases the gender or age bias of changethus enabling families to improve their malnutrition. In a number of Asian and African diets even without additional incomeis often the countries, children and women, especially preg- most cost-effective way to improve nutritional sta- nant women, are discouraged from eating eggs tus. In Indonesia a large-scale program to teach and fruit. Ignorance also interacts with economic mothers about child feeding has reduced malnutri- factors. When the price of leafy vegetables rich in tion among children at very low cost. In a Colom- vitamin A rises in the Philippines, people switch to bian program using food supplements and "ma- vegetables containing much less of the vitamin. ternal tutoring," the effects of education on children's height and weight were as large as the THE NUTRITION TRANSITION AND CHRONIC effects of extra food. Education about feeding chil- DISEASE. Better nutritional status reduces commu- dren adequately during illness is particularly im- 79 portant in dealing with the interaction of malnutri- and may require efforts to compensate for the cost tion and infection. It may also ensure that in time that breastfeeding imposes on women. additional food is actually consumed by the family Breastfeeding may be incompatible with some oc- members who need it most. cupations, but in most cases modification of work- Probably the most valuable form of nutrition ed- place practices can facilitate nursing by working ucation is promotion of breastfeeding. The princi- mothers. pal gain is improved child health, but the nursing mother also benefits from conservation of iron MICRONUTRIENT FORTIFICATION. Fortifying the stores (because menstruation is suppressed), bet- foods people already eat raises micronutrient in- ter spacing of births, decreased risk of breast or takes even without a change in eating habits. Bra- ovarian cancer, and possibly less postpartum zil's national salt iodization program, which began bleeding. The economic benefits to families and in 1978, greatly reduced endemic goiter in areas of health facilities can be substantial. Breast-milk sub- iodine deficiency. Many experimental programs stitutes would cost an estimated $15 billion a year have also shown the value of fortification. In Chile for the 120 million infants now relying on mother's the addition of iron to powdered milk and soy- milk. Promotion of breastfeeding in a large Philip- based infant formula decreased anemia in nine- pine hospital saved 8 percent of the budget by re- month-old babies from 32 to 12 percent and in fif- ducing the cost of substitute foods and the time teen-month-olds from 30 to 6 percent. Iron is also spent in feeding by nursery staff. Such programs often added to flour. Vitamin A can be added to a require education of both mothers and health pro- variety of foods, including sugar, milk, cereals, fessionals (who often discourage breastfeeding) and monosodium glutamate (MSG). Fortification Box 4.2 The Tamil Nadu Integrated Nutrition Project: making supplementary feeding work In the late 1970s the government of the state of Tamil feeding was continued for up to 180 days. Intensive Nadu in south India was operating twenty-five differ- nutrition education was directed at mothers of at-risk ent supplementary feeding programs. Evaluation children. Food supplementation was also offered to showed these programs to be ineffective and identified women whose children were being fed, to those who several reasons. The programs were not directed to- had numerous children, and to those who were nurs- ward malnourished children; they provided food that ing while pregnant. was often not suitable for small children and was eaten The project cut severe malnutrition in half and pre- by other family members; they replaced rather than vented many at-risk children from becoming malnour- supplemented home consumption of food; they did ished. Of those receiving food supplementation, 67 not educate mothers; and they failed to provide percent gained enough weight to graduate in ninety needed nutrition-related health care. The Tamil Nadu days; all except the severely malnourished graduated Integrated Nutrition Project, the first phase of which within 150 days. Because participants were fed only ran from 1980 to 1989, was accordingly designed to when required, food was only 13 percent of the proj- target services more effectively, to improve family nu- ect's total cost, much less than is typical in supplemen- trition and health practices, and to improve maternal tation programs. (The initial share dropped during the and child health services. course of the project as the number of children who Children ages 6-36 months were weighed each needed feeding declined.) When the program began, month. Of every 100 children selected for feeding, 44 in 1980, 45 to 50 percent of the children required feed- were normal in weight but faltering in growth, 34 were ing; by 1988 the project had brought the share down to moderately malnourished and faltering, and 22 were 24 percent. Selective, limited-duration supplementary severely malnourished. Supplementary feeding was feeding worked in Tamil Nadu because the community I provided immediately to those who were severely mal- nutrition workers were well trained and highly moti- nourished, and feeding for children with faltering vated and because mothers came to understand the growth was provided after one month (for children importance of feeding for healthy growth and were ages 6-12 months) or three months (for children ages pleased when their children grew well. The experience I 12-35 months). The children selected were fed for at of Tamil Nadu suggests that appropriate supplemen- least ninety days. If they failed to gain at least 500 tary feeding is both an inexpensive and an effective grams in weight, they were referred to health care, and form of nutrition education. I 80 of MSG in Indonesia cut child mortality by 30 per- nomic stagnation. There was rio improvement in cent. Whenever a food consumed by the target districts not participating in the program. A large- population can be fortified at reasonable cost, for- scale program in Chile substantially reduced child- tification can provide the same benefits as promot- hood malnutrition while increasing the use of the ing changes in diet and may be quicker and easier. public health system. In many countries free meals for schoolchildren may have little effect on their MICRONUTRIENT SUPPLEMENTATION. Supplying nutritional status but improve school attendance micronutrients separately from food requires regu- and performance. In general, food supplementa- lar, sometimes frequent, contact with the target tion works best when it is used to motivate and population. This may make it more difficult to sus- educate mothers to care for their children's health, tain high coverage. It may also make supplements when it can be concentrated within a crucial inter- more costly than fortification of foodsalthough val (during pregnancy, for example), or when it micronutrient supplementation can be added at provides additional, rionnutritional benefits. very low marginal cost to immunization programs or school-based dewormirig programs. Vitamin A FOOD PRICE SUBSIDIES. Letting people buy basic can be given in capsules at intervals of one week to foodstuffs more cheaply can, in theory, increase six months, reducing the risk of blindness substan- intake of particular foods, but there are often prac- tially. Vitamin A supplementation can reduce mor- tical problems in targeting subsidies to needy tality from measles and diarrheal disease by about households. Targeting by locale or by commodities 30 percent but has little effect on deaths from res- eaten primarily by poor people is more efficient piratory disease. than wasteful general subsidies but less precise Iodine can also be provided as a supplement to than targeting according to specific needs. Ineffi- diets. Oral doses of iodized oil protect for two to ciencies in administration can eat up much of the four years, and injectable oil protects for three to potential benefit. One large urban subsidy pro- five years. Side effects are usually not serious and gram in Brazil has often sold food for nearly the occur mostly in older adults. Supplements for same price as private markets, despite a nominal women of reproductive age prevent mental retar- 20 percent price reduction. When such waste is dation in their children and reduce the risks of avoided, targeted subsidies can effectively transfer infant mortality. Iron deficiency is the most diffi- income to poor households. As with direct trans- cult micronutrient shortage to combat by supple- fers of income or of food, subsidies are more likely mentation: tablets must be taken every day, and to improve nutrition and health when they are they often cause side effects. Because these prob- combined with nutrition education and related lems limit compliance, supplements are usually health interventions. Unless that is done, sub- given only to pregnant women, who suffer most sidies are not cost-effective. from anemia. When the principal cause of iron de- There is a strong case for government interven- ficiency is infection with hookworm and other tion to improve health by improving nutrition, but parasites, however, iron supplements are also not for interfering generally in food markets, ex- given to all those treated for a limited period after cept in extraordinary conditions such as famine. deworming. Government action in nutrition has often been wasteful because it has duplicated what private FOOD SUPPLEMENTATION. Programs that provide markets do and has paid too little attention to the food instead of micronutrient supplements are causes of poverty and to cost-effective measures harder to implement effectively. Inadequate tar- that improve families' knowledge and capacity to geting, replacement of food from the normal diet, feed themselves adequately. Reductions in mortal- or lack of attention to other causes of malnutrition ity, blindness, mental impairment, and anemia can often mean that the food is wasted. With proper make fortification and supplementation extremely targeting and attention to changing behavior, cost-effective, comparable to the best control mea- however, some supplementation programs have sures for other diseases (Table 4.4). A year of been made to work. A program in Tamil Nadu, healthy life can be bought for less than $10 with India, achieved remarkable gains by distributing some micronutrient interventions and for less than food only when children's growth faltered, while $100 with programs that provide food supple- providing information to mothers continuously ments sparingly and combine them with behav- through highly motivated community nutrition ioral change and health care. Improved adult workers (Box 4.2). This success came despite eco- health, more productive schooling, higher in- 81 Table 4.4 Cost-effectiveness of nutrition interventions Approximate cost (dollars) Intervention Target group Per death averted Per DALY saved Iron supplementation Pregnant women 800 13 Iron fortification Entire population 2,000 4 Iodine supplementation Women of reproductive age 1,250 19 Iodine supplementation Entire population 4,650 37 lodization of salt or water Entire population 1,000 8 Vitamin A supplementationa Children under age 5 50 I Vitamin A fortification Entire population 154 4 Food supplementation Children under age 5 1,942 63 Food supplementation Pregnant women" 733 24 Semiannual mass dose. Deaths averted and DALYs saved are for fetal deaths. Source: Pinstrup-Andersen and others forthcoming; Levin and others forthcoming. Fertility has been declining worldwide, but at comes, and other benefits that come with good different paces. childhood nutritional status strengthen the case for appropriate nutrition interventions. Fertility Figure 4.2 Total fertility rates by demographic region, 1950-95 All pregnancies and births carry some health risks to the mother and the child. But the risks are higher when women have health problems (such Total fertility rate as high blood pressure, heart disease, malaria, or 7 diabetes) that could be aggravated by pregnancy, when pregnancies come too early or too late in a woman's reproductive life, when they are too closely spaced or are unwanted, and when they occur to high-parity women (for example, those who have already had four or more babies). The use of family planning services by couples is an effective means of avoiding many of these fer- tility-related health risks, and it enables families to achieve their fertility goals. In many parts of the world, fertility has been falling over time as the use of family planning spreads (Figure 4.2). Gov- ernments can do much to help couples by promot- ing family planning as a socially acceptable prac- tice, by providing information on the health effects 1950 60 70 80 90 95 of fertility regulation, by teaching couples about effective methods of contraception, and by remov- ing restrictions on the marketing of contraceptives. - Sub-Saharan Africa India Subsidies may be justified in low-income popula- tions, in rural areas, and for programs targeted to - China young people. Nongovernmental organizations Other Asia and islands and the private sector will often have a large role in - Latin America and the Caribbean service provision. Ensuring access to safe abortion - Middle Eastern crescent can complement family planning services in im- - Formerly socialist economies of Europe proving health. Established market economies Fertility patterns and health Note: Dotted lines represent projected values. Source: World Bank data. Births to very young women elevate the health risks to both mother and child. Births that are too 82 Child mortality risks are higher for babies born shortlij after a zious hi ung mothers. Figure 4.3 Risk of death by age 5 for fertility-related risk factors in selected countries, late 1980s Age of mother Length of previous birth interval Percentage increase in child mortality Percentage increase in child mortality (in relation to children (in relation to children born 24 to 48 born to women ages 20-34) months after the previous birth) 220 200 160 120 80 40 0 0 40 80 120 160 200220 Egypt Guatemala Indonesia Kenya Under 18 0 o Less than 18 months 18-19 0 18-24 months Source: Hobcraft 1991. closely spaced increase the risk of child mortality; to avoid such births. If the closely spaced births births at older ages and higher parities are riskier were delayed until mothers wanted them, overall to mothers, as are unwanted pregnancies that lead child mortality in those countries might be re- to unsafe abortion or to neglect of prenatal care. duced by more than 20 percent. The reduction could be as large as 30 percent in Brazil, Colombia, THE TIMING OF BIRTHS AND CHILD HEALTH. Short Ecuador, Peru, and Tunisia, where between 40 and birth intervals pose substantial risks to child health 50 percent of births are spaced too closely. throughout the first five years of life (Figure 4.3). In Kenya, for example, infants born within eigh- CHILDBEARING AND MATERNAL HEALTH. Each teen months of the birth of a previous child are year about 430,000 women in developing countries more than twice as likely to die as those born after die from complications associated with childbear- a longer interval. In Egypt the risks are more than ing. In the absence of obstetric care, women who triple. Babies born to teenage women are also at have a birth before age 18 are three times as likely greater risk of dying. In Indonesia, for example, to die in childbirth as those who have a birth be- babies born to mothers age 18 and under are 50 tween ages 20 and 29; for women over 34, the risk percent more likely to die than those born to of maternal mortality is five times as high. First women ages 20-24. births are often riskier than second or third births, Surveys taken in twenty-five developing coun- but as parity rises thereafter, the risk of maternal tries in the 1980s show that, on average, nearly 35 mortality also climbs. In Jamaica the risk increases percent of births occur within twenty-four months by 65 percent after the third birth. In Kenya, the of the previous birth and that many women wish Philippines, and Zimbabwe 30 to 60 percent of 83 pregnant women are estimated to be at elevated tion of contraceptives, and abandonment of un- risk of death on account of either age or parity. necessary prescription requirements. Experience Maternal deaths also arise from unsafe abortion. in the formerly socialist economies of Europe has Almost 30 percent of pregnancies end in abortion, shown that all such constraints reduce contracep- for a total of about 55 million induced abortions in tive use and often damage maternal and child the world each year; 25 million of these are per- health. Demand for many contraceptive supplies formed under unsafe conditions. The damage to and services can be met by private doctors and maternal health arises mainly from infection (the commercial outlets, especially in towns and, for long-run consequences of which include ectopic some methods (such as condoms), in rural areas as pregnancy, chronic pain, and infertility), hemor- well. rhage, damage to the cervix or uterus, and reaction Use of contraceptives is the best way to avoid to anesthesia and the drugs used to induce abor- unwanted pregnancies, but it is not foolproof. For tion. About 60,000 women a year are estimated to women who wish to terminate their pregnancies, die from unsafe abortions (see Appendix table access to safe abortion as a complement to contra- B.8); other estimates range as high as 200,000. ceptive services is also important to women's Treatment of abortion-related complications can health. consume significant resources. In Brazil in 1988 about 2 percent of all hospital admissions in the REACHING LOW-INCOME AND RURAL POPULATIONS. publicly financed, privately provided health sys- The health infrastructure in poor countries is often tem were for abortion-related complications, and limited in its ability to reach highly dispersed the costs amounted to about 6 percent of all spend- populations in rural areas. In rural Uganda, for ing on obstetrics and 1 percent of all hospital example, travel time to the nearest family planning spending in that system. facility averages one hour, whereas it is only fif- Better health through family planning services teen minutes in rural Thailand. Long waits at the facility are another problem; a study of clinics in Family planning services can help women reduce several Latin American countries found that wait- the health risks from mistimed and unwanted ing times for initial visits averaged one hour and pregnancies. In low-income populations and in twenty minutes. In many countries rural women rural areas there is a strong case on equity grounds have no access to family planning fieldworkers for the government to subsidize and organize the who can provide information and simple services. provision of family planning services, using public In Guatemala, for example, 86 percent of rural as well as nongovernmental and private channels women live in communities without a family plan- as appropriate. In these settings subsidized family ning fieldworker; in Egypt the figure is only planning services are often the most effective way 33 percent. Community-based strategies have of transmitting family planning information to the been used with success in some countries to reach poor. They can also be an efficient means of im- low-income women. In Colombia, Zaire, and proving the welfare of poor families, especially Zimbabwe community-based-distribution (CBD) when private medical care is unavailable. For both workers serve the dual purpose of spreading infor- reasons, family planning services are part of the mation about family planning and providing the minimum essential clinical package discussed in most isolated populations with family planning the next chapter. Special efforts are also appropri- methodsprimarily barrier methods, such as con- ate for addressing the needs of adolescents, both doms and foaming tablets, but also oral because they tend to be particularly uninformed contraceptives. about reproductive health risks and because Family planning services provided through they often misjudge the consequences of early community-based distribution are a highly cost- childbearing. effective means of improving maternal and child Beyond providing subsidized services to specific health. In countries where both mortality and fer- populations, the government also has a role in en- tility are still relatively high, the cost per child suring access to family planning services for those death averted is extremely low. In Mali, for exam- able and willing to pay. Encouraging better ser- ple, it averages about $130, which corresponds to a vices and availability of more contraceptive mere $4 to $5 per DALY gained. In other countries, methods requires various changes, including re- such as Colombia, Mexico, and Thailandwhere moval of price controls and bans on contraceptive mortality and fertility are substantially lower- advertising, easing of restrictions on the importa- CBD family planning services cost no more than 84 $25 per DALY gained and thus remain highly cost- women to discontinue contraceptive use. In Thai- effective. land and Colombia, where the programs offer good services, about 6 percent of users quit each REACHING YOUNG PEOPLE. In developing coun- year for these reasons; in Paraguay as many as 18 tries childbearing among teenage women (ages 15- percent of users discontinue annually. 19) is common. Surveys in the 1980s in Liberia, There is considerable scope for broadening the Mali, and Uganda show that more than one in five range and quality of contraceptive methods. India, teenage women had had at least one child or was for example, has for a long time heavily empha- pregnant at the time of the interview. In Latin sized sterilization and offered attractive financial American and Caribbean countries 16 percent of incentives to both clients and providers. A nation- all births in 1992 were to teenage mothers. Adoles- wide survey in 1986-87 showed that among non- cent pregnancies are often unintended and tend to sterilized couples seeking a temporary method of be more prevalent among low-income women. In contraception, nearly 75 percent of those who both Ghana and Kenya, for example, about 40 per- wanted intrauterine devices (IUDs) reported fail- cent of married teenagers who have had children ure to get them, 67 percent reported failure to get said their first pregnancies were unintended; contraceptive pills, and 40 percent reported not among unmarried teenagers the proportion of un- being able to get condoms. In China, where steel- intended births rose to 58 percent in Ghana and 77 ring IUDs have been in widespread use, the gov- percent in Kenya. A 1986 study of Brazilian ernment, because of concern about the risk to women showed that 65 percent of those who be- women's health, recently decided to switch to the came mothers before age 20 came from poor fami- safer Copper-T IUDs. In some countries the range lies (that is, those with household income below of available methods is constrained because public the national median), in contrast to 48 percent for sector providers are required to use products on an women who delayed childbearing. essential drug list and the list mistakenly excludes Family life education in schools and other some contraceptive methods. Other constraints on venues can help teenagers make informed choices method availability include excessively restrictive about sexual behavior and the prevention of sexu- medical screening requirements, unnecessary or ally transmitted disease (STD). Family planning duplicative approval procedures, packaging and services are needed to help sexually active adoles- labeling requirements that perform no useful func- cents prevent pregnancies. And programs to help tion but increase costs, and import restrictions or teenagers cope with unintended pregnancies, es- tariffs. A study in Indonesia that surveyed a group pecially premarital ones, can be especially valu- of women eighteen months after they started able. In Jamaica the Women's Center Program has using contraceptives found that, all else being the had some success in helping young mothers to same, women who failed to get their contraceptive complete their schooling after childbirth and to method of choice were more than three times as avoid another mistimed pregnancy. likely to have discontinued use as women who did receive their preferred method. IMPROVING SERVICES AND ENCOURAGING GREATER Providing an appropriate mix of contraceptive VARIETY IN METHODS. The quality of family plan- methods can also help to reduce the spread of ning services in developing countries has been im- STDs and human immunodeficiency virus (HIV). proving, but more can be done. Providing good Linking the provision of family planning services counseling to clients is important, in part because with screening programs for STDs requires a clini- women's contraceptive needs change over the re- cal setting in which positive diagnoses may be fol- productive life cycle. Temporary methods are lowed up with treatment. The discussions of AIDS more appropriate earlier in the cycle, while perma- in this chapter, below, and in Chapter 5 address nent methods are more appropriate toward the this point in more detail. end. And certain methods are more or less suitable depending on the duration of protection desired ENSURING ACCESS TO SAFE ABORTION. In 1990 and on whether the woman is breastfeeding. Com- about 40 percent of the world's population lived in petent advice offered with sensitivity can help cli- countries where induced abortion was permitted ents choose the right method at each stage and use on request, 25 percent lived where it was allowed it effectively while also addressing their concerns only if the woman's life was in danger, and the about possible side effects. Dissatisfaction with remaining 35 percent lived in places where abor- services and contraceptive failure often cause tion laws varied in strictness between these ex- 85 In Roman ía, maternal deaths shot up when abortion maternal mortality rate had risen by nearly 40 per- was banned and fell sharply when it was legalized. cent above the level in 1965 (Figure 4.4). Before 1966 Romania's maternal mortality rate was simi- lar to the rates in other Eastern European coun- Figure 4.4 Maternal mortality in Romania, tries. By 1989 it was at least ten times the rate of 1965-91 almost any other European country. In 1990 Ro- mania's new government legalized abortion, and Maternal deaths per 100,000 live births the decline in maternal mortality was immediate 180 and even sharper than its rise following the ban: only one year after abortion was legalized, mater- nal mortality had fallen to just 40 percent of the A 1989 level. The percentage of all maternal deaths All causes caused by abortion dropped from nearly 90 per- cent before the ban on abortion was lifted to just IA over 60 percent in 1990. COSTS AND POTENTIAL GAINS IN HEALTH. Family planning services, particularly when delivered through community-based distribution, are among the most cost-effective means of improving maternal and child health. There is much scope for improving services in developing countries, where more than one women in five who wants to avoid pregnancy is not using contraception. In Bolivia, Ghana, Kenya, Liberia, and Togo at least one in three women ages 15-49 falls into this category. Lack of access to family planning services is one reason for not using themalthough it is certainly Abortion legalized not the only one. The cost of supplying family planning services to women without access (num- Abortion made illegal bering an estimated 120 million in the developing world) is estimated at about $2 billion annually for developing countries as a whole. Selective alloca- N tion of public resources to address the needs of these women, particularly those in poor families, Source: Adapted from Stephenson and others 1992, which would be a cost-effective means of promoting their used Romanian Ministry of Health data. well-being, as well as that of their children. Satis- fying the expressed wish of women to space or limit future births might each year avert as many as 100,000 maternal deaths and 850,000 deaths tremes. In countries where abortion is illegal, among children under 5. women resort to clandestine, and often unsafe, abortions at high risk to their health. Legalizing Reducing abuse of tobacco, alcohol, and drugs abortion is inadequate for protecting maternal health when problems with access continue. In In- Decisions about the use of tobacco, alcohol, and dia, for example, abortion is legal but not readily other drugs are among the most important health- available, and many women continue to rely on related choices that individuals can make. Because unsafe abortion, with detrimental effects to their individual options are limited by the strongly ad- health. dictive character of these substances, and because Romania's experience is the most striking exam- addiction is often established in adolescence, deci- ple of the impact of abortion laws on maternal sions about the control of tobacco and other addic- health. In 1966 the government banned abortion tive substances are among the most important and contraception and took steps to enforce the health-related choices that societies can make col- law. The consequences were dramatic: by 1970 the lectively. In many populations prolonged cigarette 86 smoking is already the greatest single cause of pre- crease in consumption from 500 billion cigarettes mature death. Alcohol and other drugs also con- in 1978 to 1,700 billion in 1992 has produced smok- tribute to disease and disability. The damage from ing patterns that, if they persist, will eventually substance abuse is not limited to the individuals result in about 2 million deaths a year from to- involved; others also suffer indirectly because of bacco. Similar consumption patterns exist in sev- drunk driving, fires, passive smoking, and drug- eral other countries. If, as now, about one-third of related crime and violence. the world's young adults become regular cigarette Several sorts of government policy can be used smokers and, as in industrial countries, more than to discourage consumption of tobacco, alcohol, one-third of them die prematurely because of the and other drugs. Educating the public about the habit, then, of the 120 million who reach adult life harmful effects of these substances is essential. each year, more than 10 percentmore than 12 Appropriate action will often involve special em- million a yearwill die prematurely because of to- phasis not only on reaching school-age children bacco. On current smoking patterns, the chief un- but also on helping adults to escape from addic- certainty is not whether mortality from tobacco tion. Tax policies on tobacco and alcohol have also will reach 12 million a year in the second quarter of reduced consumption, especially by discouraging the next century, but exactly when it will do so. use by young adults before they become addicted. Largely because of the long delay between cause Governments can ban all direct or indirect adver- and full effect, people tend to misjudge the haz- tising or promotion of tobacco goods or trade- ards of tobacco. When a generation of young marks and could do the same for alcohol. adults begins to smoke, they do not witness the Tobacco high mortality associated with their behavior until they reach middle age. The best-documented ex- Tobacco is in legal use everywhere in the world, ample of this delay is that of men in the United yet it causes far more deaths than all other psycho- States, among whom the main increase in smok- active substances combined. About 3 million pre- ing took place before 1945. In 1945 smoking was mature deaths a year (6 percent of the world total) common but lung cancer rare, as in developing are already attributable to tobacco smoking. If cur- countries today. Over the next forty years (1945- rent trends continue, deaths from tobacco world- 85) the smoking habit did not change greatly wide are projected to reach 10 million a year, or among young men in the United States, but lung more than 10 percent of total deaths, by the second cancer in this population rose sharply (Figure 4.5). quarter of the next century. Tobacco is already re- Among U.S. nonsmokers lung cancer remained sponsible for 30 percent of all cancer deaths in de- approximately constant at a low level during 1965- veloped countries, including deaths from cancers 85, but among smokers the rates increased twen- of the lung, oral cavity, larynx, esophagus, blad- tyfold. In 1985 tobacco caused the large majority der, pancreas, and kidney. Even more people die (110,000) of all lung cancer deaths, among both from tobacco-related diseases other than cancer, males and females, in the United States, as well as including stroke, myocardial infarction, aortic an- an even larger number (290,000) of deaths from eurysm, and peptic ulcer. In countries where other diseases, for about 20 percent of 2 million smoking has long been widespread, tobacco use is U.S. deaths. About half of those killed by tobacco now responsible for about 30 percent of all male were still in middle age (35-69) and thereby lost deaths in middle age. Smoking also harms the almost twenty-five years of nonsmoker life health of others. Among nonsmokers, exposure to expectancy. environmental tobacco smoke increases the risk of Effective discouragement of addiction to tobacco lung cancer. And the babies of mothers who involves slow social changes that take place over smoke weigh, on average, 200 grams less at birth many years. Public education is central to this pro- than those of nonsmokers. cess. In China, the United Kingdom, and the Per capita consumption of tobacco is decreasing United States, autonomous national action groups slowly in industrial countries and has remained such as Action on Smoking and Health have relatively unchanged in the formerly socialist helped sustain serious efforts to alert people to the economies. By contrast, per capita tobacco con- hazards of tobacco consumption and, through le- sumption is rising in many developing countries gal action, to protect the public from the harmful among both men and women and is expected to health effects of the habit. Governments can con- increase by about 12 percent between 1990 and tribute to the efforts of citizen groups by, for exam- 2000 (see Appendix table A.6). In China the in- ple, requiring prominent health warnings on ciga- 87 The spread of cigarette smoking among U.S. males free zones. Tobacco consumption per adult ap- was followed by mounting lung cancer ratesbut pears to have fallen between 1975 and 1990. China, only after a decades-long delay. with 300 million smokers, is following a similar path: in 1992 it banned most tobacco advertising, mandated health education for youths, prohibited Figure 4.5 Trends in mortality from lung smoking in many public places, and required pro- cancer and various other cancers among gressive reduction of tar levels. In support of U.S. males, 1930-90 countries' efforts to discourage tobacco consump- tion, the World Bank in 1992 set forth a new policy Deaths per 100,000 males a on tobacco (Box 4.3). 80 Tobacco has traditionally been taxed, although probably because it is a good source of revenue rather than for the health gains. Taxation reduces consumption, especially among the young. In in- dustrial countries a 10 percent price increase re- duces consumption by about 4 percent in the gen- eral population and about 13 percent among adolescents. Besides having few resources, most adolescent smokers probably have not been smok- ing long enough to be fully addicted and so tend to be more price sensitive than other smokers. In In- dia cigarette sales declined by 15 percent after the excise tax on most of the popular cigarette brands more than doubled in 1986. In Papua New Guinea a 10 percent increase in the tobacco tax reduced consumption by 7 percent. 1930 40 50 60 70 80 85 Alcohol and illegal drugs Alcohol-related diseases affect 5 to 10 percent of Lung cancer the world's population each year and accounted -. Various other cancers for about 3 percent of the global burden of disease in 1990. Of the 2 million alcohol-related deaths Note: Other cancers shown include leukemia and cancers that occur worldwide each year, about 50 percent of the bladder, esophagus, pancreas, liver, prostate, stem from cirrhosis of the liver, about 35 per- stomach, and colon and rectum. cent from cancer of the liver or esophagus, 10 a. Adjusted to the age structure of the U.S. population in 1970. percent from alcohol dependence syndrome, and Source: Boring, Squires, and Tong 1993. 5 percent from injuries caused by motor vehicles. The problems caused by alcohol abuse consume scarce medical resources and extend beyond the damage that drinkers do to themselves. In many Latin American countries in the 1980s, 20 percent rette packages and advertisements, as well as by of all hospital and emergency room admissions targeting of clear messages not only to school-age were alcohol-related. In Papua New Guinea more children but also to adults. Reaching adults is im- than 85 percent of fatal road accidents in the 1980s portant because over the next few decades it is involved either drunk drivers or drunk pedes- those who are already smoking who will account trians. Within households, drinking often leads to for nearly all of the tens of millions of deaths per assault and injury, although the scale of the prob- decade caused by tobacco. Some countries go fur- lem is hard to quantify. ther by banning commercial promotion of tobacco Alcohol consumption is stable in the industrial goods and tobacco trademarks and by placing re- world but is on the rise in many developing coun- strictions on public smoking. Singapore has been tries. Between 1960 and 1981 annual beer con- in the forefront of public activism in Asia: it has sumption per capita rose from 12 liters to 135 liters prohibited advertising (since 1971), issued strong in Gabon and from 3 liters to 20 liters in Côte warnings on health effects, and created smoke- d'Ivoire. Total world production of beer nearly 88 Box 4.3 World Bank policy on tobacco In 1992, in recognition of the adverse effects of tobacco requirements. The World Bank seeks to help these consumption on health, the World Bank articulated a countries diversify away from tobacco. formal policy on tobacco. The policy contains five main To the extent practicable, the World Bank does not points. lend indirectly for tobacco production activities, al- The World Bank's activities in the health sector though some indirect support of the tobacco economy including sector work, policy dialogue, and lending may occur as an inseparable part of a project that has a discourage the use of tobacco products. broader set of objectives and outcomes (for example, The World Bank does not lend directly for, invest rural roads). in, or guarantee investments or loans for tobacco pro- Unmanufactured and manufactured tobacco, to- duction, processing, or marketing. However, in the bacco-processing machinery and equipment, and re- few countries that are heavily dependent on tobacco as lated services are included on the negative list of im- a source of income and of foreign exchange earnings ports in loan agreements and so cannot be included (for example, those where tobacco accounts for more among imports financed under loans. than 10 percent of exports) and especially as a source of Tobacco and tobacco-related producer or con- income for poor farmers and farmworkers, the World sumer imports may be exempt from borrowers' agree- Bank treats the subject within the context of respond- ments with the Bank to liberalize trade and reduce tar- ing most effectively to these countries' development iff levels. doubled between 1970 and 1989, far surpassing (that is, about one to two months' wages) are esti- population growth, with much of the increase oc- mated to reduce traffic fatalities by about 5 per- curring in developing regions. cent. Mandatory jail sentences for drunk driving As with alcohol, abuse of illegal drugs causes have also been weakly effective. The effect of in- serious health and social problems. Individuals formation campaigns concerning alcohol con- run the risk of death from infectious, circulatory, sumption has not been quantified, but there is evi- respiratory, and digestive diseases, as well as from dence that in countries where alcohol is legal but violence, overdose, and AIDS. Users of cocaine, commercial promotion is not, per capita alcohol especially in the form of "crack," often suffer consumption is 30 percent lower than elsewhere acute cardiovascular problems that require emer- and deaths from motor vehicle accidents are 10 gency room services, and the babies of pregnant percent fewer. As a successful alcohol rehabilita- users of cocaine are often born with severe health tion program in south India demonstrates, com- problems. munity efforts are generally more effective than Reliable data on trends and patterns in illegal medical interventions in helping individuals to drug use are scarce. Users typically fall in the age overcome alcohol dependence, in part because of group 15-44, although most are in their mid-twen- the importance of sustained encouragement, ties. In the past decade the production and con- which is more easily offered by the community sumption of illicit drugs, especially cocaine, ap- than by health service institutions. pear to have increased considerably worldwide. In Prohibition is a common approach to drug some developing countries the use of psychoactive abuse. In the United States prohibition as an ap- drugs such as inhalants is also a serious problem. proach to control of alcohol failed early in this cen- Taxes and judicial penalties have been used to tury. It appears to be having, at best, limited suc- discourage alcohol abuse. A 1982 U.S. study indi- cess in controlling use of other drugs now. In other cated that an increase in the liquor tax of about settings, including Malaysia and Singapore, prohi- $3.50 (at 1991 prices) per gallon equivalent of pure bition coupled with a mandatory death penalty for alcohol would lower demand enough to reduce drug trafficking appears to have been more effec- the incidence of liver cirrhosis by 5 percent in the tive. The successes achieved in controlling the use short run and perhaps twice as much in the long of alcohol and tobaccothrough restrictions on run. In industrial countries mandatory license promotion and access, high taxation, rehabilitation sanctions on drunk drivers are estimated to de- of addicts, and public educationmay also be rele- crease traffic fatalities by about 10 percent; the im- vant for efforts against other drugs. For alcohol position of a minimum legal drinking age and the and tobacco, past successes with these measures assessment of relatively large mandatory fines should spur efforts toward full implementation. 89 Environmental influences on health compounded by poor hygiene), inadequate gar- bage disposal and drainage, heavy indoor air pol- The environment in which people live has a huge lution, and crowding. The diseases associated influence on their health. For poor people and with poor household environments occur mainly poor regions, it is the household environment that in developing countries, where they account for carries the greatest risks to health. By providing nearly 30 percent of the total burden of disease information, reducing poverty, and facilitating and (Table 4.5). Modest improvements in household stimulating private sector action, governments can environments would avert almost a quarter of this deploy potent mechanisms to improve this envi- burden, mostly as a result of reductions in diar- ronment. Potential health gains from these efforts rhea and respiratory infections. total nearly 80 million DALYs a year in developing countries. Other government actions, designed to WATER AND SANITATION. About 1.3 billion people ameliorate or remedy unsafe conditions in the in the developing world lack access to clean and workplace and pollution of the ambient environ- plentiful water, and nearly 2 billion people lack an ment, could save 36 million and 8 million DALYs a adequate system for disposing of their feces (Fig- year, respectively. Finally, feasible reductions in ure 4.6). Feces deposited near homes, contami- the toll taken by road traffic injuries could avert nated drinking water (sometimes caused by poorly the loss of 6 million DALYs a year. designed or maintained sewerage systems), fish The household environment from polluted rivers and coastal waters, and agri- cultural produce fertilized with human waste are Poor households generally live in a domestic envi- all health hazards. Water quantity is as important ronment with high health risks caused by poor as water quality. Washing hands after defecation sanitation and inadequate water supply (often and before preparing food is of particular impor- Table 4.5 Estimated burden of disease from poor household environments in demographically developing countries, 1990, and potential reduction through improved household services Burden from these diseases in Reduction Burden averted Burden averted developing achievable by feasible per 1,000 Principal diseases related countries through feasible interventions population to poor household (millions of interventions (millions of (DALYs per environmentsa Relevant environmental problem DALYs per year) (percent)b DALYs per year) year) Tuberculosis Crowding 46 10 5 1.2 Diarrhea' Sanitation, water supply, hygiene 99 40 40 9.7 Trachoma Water supply, hygiene 3 30 1 0.3 Tropical clusterd Sanitation, garbage disposal, vector breeding around the home 8 30 2 0.5 Intestinal worms Sanitation, water supply, hygiene 18 40 7 1.7 Respiratory infections Indoor air pollution, crowding 119 15 18 4.4 Chronic respiratory Indoor air pollution diseases 41 15 6 1.5 Respiratory tract Indoor air pollution cancers 4 10 * 0.1 All the above 338 - 79 19.4 * Less than one. Note: The demographically developing group consists of the demographic regions Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent. The diseases listed are those for which there is substantial evidente of a relationship with the household environment and which are listed in Appendix B. Examples of excluded conditions are violence related to crowding (because of lack of evidence) and guinea worm infection related to poor water supply (not listed in Appendix B). Estimates derived from the product of the efficacy of the interventions and the proportion of the burden of disease that occurs among the exposed. The efficacy estimates assume the implementation of improvements in sanitation, water supply, hygiene, drainage, garbage disposal, indoor air pollution, and crowding of the kind being made in poor communities in developing countries. Includes diarrhea, dysentery, cholera, and typhoid. Diseases within the tropical cluster most affected by the domestic environment are schistosomiasis, South American trypanosomiasis, and Bancroftian filariasis. Based on very inadequate data on efficacy. Source: Appendix tables B.2 and B.3 and authors' calculations. 90 Many people worldwide still lack safe water and adequate sanitation. Figure 4.6 Population without sanitation or water supply services by demographic region, 1990 Percentage of population without services Millions of people without services 100 80 60 40 20 0 0 500 1,000 1,500 2,000 2,500 Sub-Saharan Africa IIIIIIIIIIIUIL I ndia Ilijilh!!! China Other Asia and islands 11111 Latin America and the Caribbean II !MII Middle Eastern crescent D Formerly socialist economies of Europe I Established market 11 economies World Sanitation W Water Note: Coverage is defined in accordance with local standards. Source: World Health Organization data. tance in reducing disease transmission, but with- suspended particulates, smoky houses in Nepal out abundant water in or near the home, hygiene and Papua New Guinea have peak levels of 10,000 becomes difficult or impossible. The lack of water or more. Rural people in developing countries may supply and sanitation is the primary reason why receive as much as two-thirds of the global expo- diseases transmitted via feces are so common in sure to particulates. Women and young children developing countries. The most important of these suffer the greatest exposure. diseases, diarrhea and intestinal worm infections, Indoor air pollution contributes to acute respira- account for an annual burden of 117 million tory infections in young children, chronic lung dis- DALYs, or 10 percent of the total burden of disease ease and cancer in adults, and adverse pregnancy in developing countries. In addition, an inade- outcomes (such as stillbirths) for women exposed quate water supply increases the risk of schis- during pregnancy. Acute respiratory infections, tosomiasis, skin and eye infections, and guinea principally pneumonia, are the chief killers of worm disease (Box 4.4). young children, causing a loss of 119 million DALYs a year, or 10 percent of the total burden of INDOOR AIR POLLUTION. Indoor air pollution, disease in developing countries. Data from the which World Development Report 1992 identified as Gambia, Nepal, South Africa, the United States, one of the four most critical global environmental and Zimbabwe suggest that reducing indoor air problems, probably exposes more people world- pollution from very high to low levels could poten- wide to important air pollutants than does pollu- tially halve the incidence of childhood pneumonia. tion in outdoor air. Whereas air in such cities as Adults can suffer chronic damage to the respira- Delhi, India, and Xian, China, contains a daily av- tory system from indoor pollution. Studies in erage of 500 micrograms per cubic meter of total China, India, Nepal, and Papua New Guinea have 91 Box 4.4 After smallpox: slaying the dragon worm Guinea worm disease, or dracunculiasis ("infection out of sources of drinking water when they have with a little dragon"), was endemic from ancient times guinea worm blisters on their legs and to filter their in a belt stretching from West Africa through the Mid- water with a cloth if they do not have a safe water dle East to India and Central Asia. It has been suc- source. cessfully eliminated from the Central Asian republics Eradication of dracunculiasis by the end of 1995 has and from Iran, where the last case was seen in the been adopted as an international goal. Pakistan may 1970s, and it has spontaneously disappeared from have achieved eradication in 1992, and Cameroon, In- most of the Middle East and from several African coun- dia, and Senegal may do so in 1993. Between 1987 and tries, such as the Gambia and Guinea. There are now 1992 cases reported per year fell from 653,000 to probably fewer than a million cases worldwide. 201,000 in Nigeria, from 180,000 to 33,000 in Ghana, The disease does not kill people, but it causes pain and from 17,000 to 900 in India. In general, these ad- and disability to its victims for several weeks in the vances have been achieved through "vertical" pro- year as the 60-centimeter-long female worm emerges gramsthat is, programs specific to dracunculiasis. from a blister, usually on the leg. In some cases the The eradication of the disease from the poor, sparsely disability is permanent. The worms usually emerge in populated endemic countries in West Africa will, how- the early rainy season, the time when the incapacitated ever, require integrated programs in which the re- victims would otherwise be planting and weeding their sources available for guinea worm control are shared crops. Children whose parents are stricken by guinea with other activities, such as immunization. A by-prod- worm are more likely to be malnourished in the follow- uct of guinea worm eradication will be community- ing year. based surveillance systems, which can be used by com- Because dracunculiasis can only be caught by drink- munities to monitor and improve their own health and ing infected water, improving the water supply is an by public health workers to combat other diseases, important preventive measure. Health education is such as polio. also essential. Villagers need to be persuaded to stay shown that up to half of adult women (few of ford the household improvements, including bet- whom smoke) suffer from chronic lung and heart ter water and sanitation services, that they desire. diseases. Nonsmoking Chinese women exposed to As people acquire more education, their hygiene indoor coal smoke (which is especially harmful) improves, and their responsiveness to public in- have a risk of lung cancer similar to that of men formation programs increases. To support house- who smoke lightly. Comprehensive improvement holds' efforts, governments have an important in indoor air quality in the developing countries role in setting and enforcing appropriate environ- might avert a loss of 24 million DALYs each year mental standards and disseminating information by reducing the burden of acute respiratory infec- on, for example, the health benefits of good hy- tions and chronic respiratory diseases by 15 per- giene and the effects of exposure (especially of cent and of respiratory tract cancers by 10 percent babies) to smoke. Governments should also con- (Table 4.5). centrate on strengthening security of tenure HousiNG. In many cities 30 to 60 percent of the (which is essential for encouraging households to population live in overcrowded and deteriorating invest in their housing) and on establishing a legal, shanties, tenements, and boardinghouses. Crowd- regulatory, and administrative framework that fa- ing is associated with increased airborne infection cilitates responsive, accountable, and efficient pro- and personal violence. Poor structures lead to vision, often by private suppliers, of services that greater exposure to heat, cold, noise, dust, rain, people want and are willing to pay for. And they insects, and rodents. And housing locations ate should refrain from supplying services directly often unhealthy because of, for example, poor and from granting indiscriminate, widespread drainage. subsidies. Such subsidies are often captured by wealthier consumers, go for improvements that POLICIES FOR IMPROVING THE HOUSEHOLD ENVI- households would make anyway, or encourage RONMENT. The most powerful forces for reducing consumption patterns that are detrimental to domestic risks to health are rising incomes and health. (For example, subsidies for coal used in increased education for household members. cooking lead to more indoor air pollution than Higher incomes make it possible for people to af- would be the case with cleaner liquid or gas fuels.) 92 Past experience in water and sanitation illus- poorer class that receives little or no service; and a trates the limitations of direct government provi- ripe environment for political patronage. sion of household services. Despite technical prog- The poor usually miss out on both services and ress in developing affordable engineering solu- subsidies. They suffer the substantial health con- tions to the problems of water, sanitation, drain- sequences described in Table 4.5 and pay high age, and housing, the delivery and maintenance of prices for inadequate services. In Lima poor peo- these services, especially by governments, have ple may pay $3 for a cubic meter of contaminated been disappointing. At the end of the Interna- water collected by bucket from a private vendor, tional Drinking Water Supply and Sanitation De- while the middle class pays 30 cents per cubic me- cade (the 1980s), most people in the poorer regions ter for treated water provided on tap in their of the world still lacked sanitation, and the num- houses by the publicly subsidized water company. ber of urban residents without water had not been Broadly based subsidies are not necessary for reduced. ensuring access to safe water and sanitation. In Supply-side failures are largely caused by ineffi- most urban communities households are willing to cient and unresponsive public sector monopolies pay the full costs of water service and often the full which, in the water sector, typically provide subsi- cost of sanitation services. Willingness to pay for dized services at between one-third and two-thirds water may be high in rural areas as well, but what of the full economic cost. Massive public invest- people can afford is commonly not enough to ments, often supported by the donor community cover the high costs of supply. Subsidy may be and the World Bank, have been made in public or justified in such situations. But the rationale quasi-public agencies responsible for the delivery should be primarily one of redistribution: a society and maintenance of household services. The net re- may choose to provide cheap water or other ser- sult has often been bloated public agencies with low vices to the poor as one of many alternative means accountability to their customers and few incen- of improving their welfare. Health benefits alone tives for improving efficiency; a middle class that is do not generally provide a rationale for public sub- increasingly well served with subsidized services; a - sidy of water and sanitation (see Box 4.5). Box 4.5 The costs and benefits of investments in water supply and sanitation People want safe water and good sanitation and are discharge. The cost of water and sanitation services can willing to pay for these services, especially for plentiful range from $15 per person per year for simple rural water in or very near the home. Improvements in wa- systems to $200 for full-fledged urban systems. Poor ter supply raise productivity through savings in the households cannot afford the design standards of in- fuel used to boil polluted water and, even more impor- dustrial countries, but such standards are not neces- tant, through the time and energy savings for women sary on health grounds. Completely eliminating fecal who have to collect water from distant sources. Provi- bacteria requires expensive chlorination, but low con- sion of public handpumps in Imo State, Nigeria, re- centrations present little health hazard and should be duced the median time that each household spent on tolerated. water collection in the dry season from six hours a day If households pay the total cost of water and sanita- to forty-five minutes. In Lesotho, not an especially dry tion services because of the productivity and amenity country, the benefits in time saved alone are sufficient benefits, substantial health gains are an added bonus to justify investments in rural water supply. Sanitation achieved at no cost per DALY gained. When willing- improvements have high amenity value, making possi- ness to pay is much less than costs, it is usually a mis- ble a cleaner and more pleasant environment. take to justify subsidies on the basis of health benefits The costs of water supply and sanitation services alone. First, such subsidies compromise the demand- vary by technology, population density, the hydrologic driven approach to service provision (that is, provision and geologic environment, and design standards..De- of services that people want and are willing to pay for); sign standards for water supply can range from one lack of accountability and inefficiency are the inevitable handpump per 250 people, supplying 20 liters per per- consequences. And second, if publicly financed invest- son per day, to multiple-tap in-house connections that ments in these services are being considered for health supply several hundred liters of fully treated water per reasons, it should be noted that such investments gen- person per day. Design standards for sanitation can erally cost more per DALY gained than other health vary from a pit latrine to flush toilets connected to a interventions recommended in this Report. sewerage system, with downstream treatment prior to 93 Box 4.6 Environmental and household control of mosquito vectors Diseases transmitted by insect vectors account for Polystyrene beads losses of 44 million DALYs worldwide each year (35 million in Sub-Saharan Africa), or 3 percent of the The application of polystyrene beads to pit latrines has world burden (12 percent in Sub-Saharan Africa). Al- proved successful in reducing the breeding of Culex though widespread application of insecticides is help- mosquitoes and the transmission of filariasis. The ing to control river blindness in West Africa (see Box beads form a floating layer that discourages egg laying 1.1) and Chagas' disease in South America, it is no and suffocates any mosquito larvae that do hatch. In longer the mainstay of vector control against other dis- the town of Makunduchi (population 12,000) in eases. Emphasis has shifted to a range of targeted bio- Zanzibar, Tanzania, a combination of polystyrene-bead logical, physical, and behavioral approaches supported application and mass drug treatment of the population by insecticides when necessary. Two examples are between January 1988 and June 1989 virtually elimi- given here. nated biting by infective mosquitoes, and the propor- tion of people infected by filariasis fell from 50 to 10 Impregnated bednets percent. By January 1993 the proportion of people in- fected had fallen to 3 percent. The polystyrene bead Bednets impregnated with a pyrethroid insecticide of layers remained intact and effective for several years low mammalian toxicity form lethal traps for mosqui- and were disrupted only by exceptional flooding. toes attracted by the carbon dioxide and body odor In Zanzibar Town researchers are studying whether emitted by the occupants. In Sichuan Province, China, it is better to make beads freely available so that house- up to 2.25 million netsalready owned by nearly all holders can apply them to their own pits or to have householdershave been treated each year since 1987. trained teams identify and treat all pits requiring treat- If nothing else, the cost is much lower than spraying ment. In Dar es Salaam polystyrene beads are being the same houses with DDT. In Emei County, Sichuan, used to control Culex nuisance biting and thereby in- the number of malaria cases had been steady at about crease public acceptance of house spraying against the 4,000 between 1980 and 1986. After bednet treatment Anopheles vectors of malaria. The effectiveness of the began, the number declined steadily, to 352 in 1991. In Culex control measures is evidenced by declining sales the Gambia a combination of net treatment and chemo- of mosquito coils in local shops. In Madras, India, poly- prophylaxis, carried out by primary health care person- styrene beads are being applied to water tanks to con- nel, reduced overall child mortality by 63 percent. trol the local vectors of malaria and dengue; the quality of the water is not affected. An important policy issue, on which there is an sanitation servicesas in urban areasthe de- apparent tension between health objectives and mand for improved sanitation has invariably risen the demand-driven approach advocated here and automatically as the demand for water services is in World Development Report 1992, concerns the se- satisfied. Second, where the demand-driven ap- quencing and packaging of investments in water proach has not been followed, service provision and sanitation. Given the patterns of household has almost always been characterized by ineffi- choice, a demand-driven approach will usually ciency and lack of accountability. For the provision mean that provision of water supply services pre- of water supply and wastewater collection ser- cedes that of sanitation services. It is frequently vices, therefore, the demand-driven approach argued that this sequence would produce few should be compromised only in rare health benefits because rapid increases in water circumstances. use can overwhelm existing waste disposal capac- Households are less willing to pay for the cost of ity and because health benefits are maximized only trunk sewers and treatment of excreta and waste- when households utilize both better water and water. Because these investments benefit the better sanitation services. These arguments are whole community and are important for environ- plausible, but experiences in many countries sug- mental quality and health, there is potentially a gest that close adherence to the demand-driven case for using public funds to finance them. A few approach remains appropriate in most places, in- other situations may also justify direct government cluding low-income settings. First, where rapid in- action or subsidies. Householders tend to under- crease in water use is likely to cause environmental value such investments as areawide pollution and health problems in the absence of household abatement, vector control involving actions within 94 households (see Box 4.6), and research and devel- The occupational environment opment. There may thus be grounds for public subsidy or other interventions in these areas. It Many women work in the home and thus suffer will often be difficult to disentangle environmental disproportionately from the health risks in the and health benefits, and judgments will be neces- household environment just described. Both men sary concerning the use of public funds. and women may also encounter health risks in Large institutional and cultural shifts are needed workplaces outside the home. A burden of 36 mil- to create an efficient system for allocating scarce lion DALYs, or 3 percent of the global burden of public and private resources to improve the house- disease, is caused each year by preventable inju- hold environment. Many developing countries ries and deaths in high-risk occupations and by have inheritedand then elaborated onthe for- chronic illness stemming from exposure to toxic mer colonial powers' worst traditions of public chemicals, noise, stress, and physically debilitat- sector inertia and professional inflexibility. En- ing work patterns (Table 4.6). couragingly, however, private sector involvement The International Labour Office has estimated is increasing rapidly in both industrial and devel- that the cost of occupational injuries and associ- oping countries. SODECI, the privately run utility ated production losses in a sample of industrial in Abidjan, is considered one of the best-run water countries is between 1 and 4 percent of GNP. In companies in Africa. EMOS, the utility that serves developing countries this proportion is likely to be Santiago, has used private sector contracts for greater because accident rates tend to be higher. such functions as meter reading, pipe mainte- Rates of fatal occupational injuries among con- nance, billing, and vehicle leasing and is one of the struction workers, for example, are more than ten most efficient utilities in Latin America. The role of times higher in Kenya and Thailand than in Fin- community organizations and NGOs may also be land. Agriculture, which employs more than half significant, particularly in drainage and sanitation of all adults in most developing countries, is improvements. In cities such as Karachi and São among the world's most dangerous occupations. Paulo, community groups have significantly accel- Not only do agricultural workers suffer injuries, erated the provision of low-cost water supply and but they are also exposed to disease-carrying ani- sanitation services to poor households, as well as mals and to poisonous agrochemicals. Health risks helping to maintain and manage local services. are high in other sectors as well. Miners, construc- Table 4.6 Estimated global burden of disease from selected environmental threats, 1990, and potential worldwide reductions through environmental interventions Reduction achievable Burden averted by Burden from these through feasible feasible interventions Burden averted per Type of environment and diseases (millions of interventionsb (millions of DALYs 1,000 population principal related diseases' DALYs per year) (percent) per year) (DALYs per year) Occupational 318 - 36 7.1 Cancers 79 5 4 0.8 Neuropsychiatric 93 5 5 0.9 Chronic respiratory 47 5 2 0.5 Musculoskeletal 18 50 9 1.8 Unintentional injury 81r 20 16 3.1 Urban air 170 - 8 1.7 Respiratory infections 123 5 6 1.2 Chronic respiratory 47 5 2 0.5 Road transport (motor vehicle injuries) 32 20 6 1.2 Alltheabove 473' - 50 10.0 The diseases shown are those for which there is substantial evidence of a relationship with the particular environment and which are listed in Appendix B. Estimates derived from the product of the efficacy of the interventions and the proportion of the global burden of disease that occurs among the exposed. All estimates of efficacy are speculative and assume the implementation of known, feasible, and affordable interventions in the circumstances encountered in developing countries. Computed by subtracting motor vehicle injuries (32 million DALY5) from all unintentional injuries (113 million DALY5). Adjusted for double counting. Source: Appendix tables B.2 and B.3 and authors' calculations. 95 tion workers, migrant workers, and child laborers able. Under the assumption that achievable reduc- all suffer increased risk of disease because of their tions in urban air pollution can prevent 5 percent occupations. Small workplaces may have espe- of all infectious and chronic respiratory disease, cially low standards of safety, yet such risks are these reductions could avert a burden of 8 million often overlooked by government agencies and DALYs each year, or 0.6 percent of the global bur- trade unions alike. A survey of companies in den of disease (Table 4.6). Local impacts and the Samud Prakhan, Thailand, found that smaller effects on especially vulnerable groups can be plants, with fewer than fifty workers, had substan- much greater (Box 4.7). tially lower levels of sanitation, health services, Lead poisons many systems in the body and is safety provisions, and environmental control mea- particularly dangerous to children's developing sures than larger enterprises. Workers suffered brains and nervous systems. Airborne lead con- more than twice as much noise and a third more centrations are high in polluted urban environ- lead fumes and vapors. And they experienced sig- ments, where lead comes mainly from the exhaust nificant work-related health problems: 22 percent of vehicles burning leaded gasoline. Elevated lead had lead poisoning or absorption, 27 percent had levels in children have been associated with im- upper respiratory symptoms, and 6 percent had paired neuropsychologic development as mea- chronic obstructive pulmonary disease, even sured by loss of IQ, poor school performance, and though most workers were below age 30. behavioral difficulties. Alleviating occupational risk depends on safety education for workers and managers, use of ap- WATER POLLUTION. Newly industrialized coun- propriate equipment and technology, and sound tries, as well as many industrial countries, have management practices. Governments can encour- polluted or are polluting their rivers, lakes, and age these initiatives through legislation and regu- coastal waters with a variety of chemical and bio- lation, financial incentives, investment in educa- logical wastes of both industrial and domestic ori- tion, and research and development. Where gin. The practice of letting raw wastewater from worker organizations are strong, they have played industry and residential areas flow into rivers or a major role in identifying and reducing occupa- the sea is common but unwise. Investment in pre- tional risks. Tripartite agreements between venting it may be justified because of the possibly workers, employers, and governments can lead to severe local health consequences (as illustrated in speedy progress. Box 4.8) and because generalized water pollution, The ambient environment by reducing the number of water sources available for domestic supply, can foreclose cost-effective Radiation and pollution of air and water are addi- options for responding to demand for domestic tional health hazards. Since there is no market for services. clean air and water, government action is fre- quently justified. RADIATION. Individuals are exposed to natural background ionizing radiation and to radiation AIR POLLUTION. Many cities suffer from air p01- used for medical and dental diagnosis. Only a tiny lution caused by industry, power plants, road amount of additional radiation comes from safely transport, and domestic use of coal. About 1.3 bil- operated nuclear power stations or other installa- lion urban residents worldwide are exposed to air tions (roughly one-thousandth of the background pollution levels above recommended limits. Air dose for those living within 50 kilometers of a nu- quality in the established market economies has clear power station). Current evidence suggests generally improved in the past two decades. But in that the health effects of this radiation on the gen- many developing countries and in the formerly so- eral population are extremely small or nonexistent. cialist economies, air quality has deteriorated be- Accidents and occupational risks to workers in nu- cause of rising industrial activity, increasing power clear industries and to miners of radioactive ores, generation, and the congestion of streets with however, are different matters. The consequences poorly maintained motor vehicles that use leaded of the nuclear power plant accident at Chernobyl, fuel. Ukraine, in 1986 have yet to be fully documented Air pollution damages the human respiratory but are undoubtedly large. (The risk of such acci- and cardiorespiratory systems in various ways. dents is particularly high in the formerly socialist The elderly, children, smokers, and those with economies because of their large number of poorly chronic respiratory difficulties are most vulner- designed nuclear facilities.) Standards and safe- 96 Box 4.7 Air pollution and health in Central Europe Contrary to expectations, public ownership and cen- mortality of 1 percent for every 10 micrograms per cu- trally planned economies have neither controlled poi- bic meter of particulates and sulfur dioxide. These esti- lution nor brought health benefits to the populations of mates suggest air pollution causes up to 3 percent of the formerly socialist economies. The countries of this total mortality in the Czech Republic and is responsible region face a variety of serious environmental health for roughly 9 percent in the gap in mortality rates be- threats, of which the greatest are particulates and gases tween the Czech Republic and Western Europe. Simi- in air, lead in air and soil, and nitrates and metals in lar estimates have been obtained for Silesia in Poland. water. A substantial gap in health status between these The effect of air pollution on mortality is greater for countries and those of Western Europe has opened up certain causes of death in specific age groups. A recent since the early 1960s: life expectancy is roughly five study of postneonatal respiratory mortality showed a years shorter in the formerly socialist economies, and rate 2.4 times higher in the most polluted districts of mortality rates in middle-aged males are roughly the Czech Republic than in the least polluted, after double. There has been considerable speculation adjusting for a battery of socioeconomic factors. An among scientists and the public in Central Europe increase in particulates of 25 micrograms per cubic me- about how much of this health gap is attributable to ter was associated with an increase in postneonatal res- environmental pollution. piratory mortality of 58 percent. Air pollution is the environmental factor that has had The contribution of air pollution to morbidity in the the greatest negative effect on health in Central Eu- Czech Republic is likely to be considerably greater than rope. Of the many air pollution 'hotspots" through- the effect on mortality and to have larger economic out the region, the worst-affected area is the 'Black consequences through health expenditures, lost Triangle," which covers northern Bohemia and Mo- schooling, and lost productivity. Children in heavily ravia, Silesia, and Saxony and has a population of polluted areas may suffer twice the rates of respiratory roughly 6.5 million. In August 1991 the three govern- morbidity of those in clean areas. Overall, air pollution ments involvedthe Czech Republic, Germany, and may be responsible for up to one-quarter of all respira- Polandand the Commission of the European Com- tory morbidity in Czech children. munities formed a Working Group for Neighbourly "Hotspots" of lead exposure exist throughout the Cooperation on Environmental Issues to deal with the formerly socialist economies. Average blood levels of extremely high levels of air pollutants in the area. more than 25 micrograms per deciliter in children have The overall effect of air pollution on mortality in the been reported in, for example, Pribram, Czech Repub- I Czech Republic has been estimated using data on the lic, and Katowice, Poland. In comparison with normal distribution of the population, the ambient levels of levels, these higher levels could double the proportion particulates and sulfur dioxide, and the relationship of children requiring special education and halve the between excess mortality and pollution. This relation- proportion in the exceptionally gifted group (IQ greater ship is derived from studies in Canada, the United than 130). Kingdom, and the United States that indicate excess guards against accidents and occupational hazards appropriate coping strategies) and healthier. The have been greatly improved, but risks may re- best preparation at the national level for these un- main, and continued research and vigilance are certain future events is therefore to pursue sound required. Putative links of certain cancers with ex- economic and health policies in the medium term. posure to radon in houses and with electromagne- tic fields created by high-voltage cables are being IMPROVING THE AMBIENT ENVIRONMENT. Improv- investigated in several industrial countries. ing health is only one of several reasons why soci- eties may choose to invest in a cleaner environ- GLOBAL THREATS. Depletion of the atmospheric ment. The policies and actions needed to clean up ozone layer and global warming pose potential the air in a given city or area will depend on the threats of unknown magnitude to health. Internã- origins of the pollution at that site. In most cities in tional agreements are limiting or will limit the re- developing countries motor vehicles are a signifi- lease of chlorine compounds that can harm the cant source of air pollution and need to be specifi- ozone layer and of the greenhouse gases that con- cally targeted. A few cities in the developing tribute to global warming. The societies that will world, among them Bangkok and Mexico City, are suffer least from these global changes are those pursuing systematic policies to reduce motor vehi- that are wealthier (and therefore able to invest in cle emissions, and their experience will be valuable 97 Box 4.8 Pollution in Japan: prevention would have been better and cheaper than cure In the 1950s and 1960s Japan experienced a period of in 1956 patients with a severe neurological affliction, rapid industrialization and economic growth, hut little later to be called Minamata disease, were observed. In attention was paid to the environmental consequences. 1968, following extensive research, the disease was The result was high levels of pollutants in the air, wa- linked to the ingestion of seafood containing high con- ter, and soil in some areas and several infamous out- centrations of methyl mercury, a compound dis- breaks of disease. Strong corrective action was taken in charged into Minamata Bay by the Chisso Corporation the 1970s and 1980s to redress the severest problems. as a by-product of the manufacture of acetaldehyde. Three conclusions emerge from the examples given be- The discharge of methyl mercury peaked in 1959; it low: allowing the release of toxic substances into the ended in 1968 when the company ceased production of environment can lead to serious health consequences acetaldehyde, but by then the floor of the bay and its and economic losses; prevention, as Japan is now do- aquatic life had become heavily contaminated. Starting ing, is less costly than cleaning up; and taking correc- in 1974, 1.5 million cubic meters of polluted sediment tive action now is less costly than allowing problems to were dredged and removed. persist. By 1991, 2,248 people (1,004 of whom had died) had been certified as suffering from Minamata disease and Case 1: sulfur dioxide in the air were eligible for compensation. An additional 2,000 people are pursuing claims for compensation. Had dis- Between 1956 and 1973 one of Japan's largest petro- charge of mercury continued, the estimated annual chemical complexes was constructed at Yokkaichi City. costs of the damage, including patient treatment and By 1960 air pollution was causing local concern, and by compensation, sediment dredging, and losses to fish- 1963 one-hour average sulfur dioxide levels exceeded eries, would have been $97 million a year. If 2,800 micrograms per cubic meter, far above WHO's acetaldehyde production had continued, pollution suggested maximum of 350 micrograms per cubic me- abatement through in-plant waste recycling would ter. In 1967 local residents successfully sued six com- have cost only $1 million a year. panies, claiming medical costs and compensation for lost income. Seven percent of the total population of Case 3: cadmium in the soil the district were certified to have been medically af- fected by ambient air pollution. Increasingly stringent In the late 1940s a disease characterized by extreme pollution measures were introduced starting in 1970, generalized pain, kidney damage, and loss of bone and by 1976 sulfur dioxide levels were in compliance strength appeared in the Jinzu River Basin. The dis- with local standards. ease, which primarily afflicted women, was called itai- Air pollution control costs since 1971including itai (''It hurts, it hurts!'') after the cries of the sufferers. technical installations and their operation, monitoring, Two decades of research led, in 1968, to the conclusion and creation of environmental buffer zoneshave been that the cause was chronic cadmium poisoning, which $114 million a year. Without this investment, however, was traced to the effluent from the Mitsui Mining and medical expenses and compensation would have been Smelting Company located in the upper reaches of the more than $160 million a year. basin. The route for the cadmium poisoning was from river water to irrigation water to soil to rice. By 1991, Case 2: mercury in the water 129 people had been certified as itai-itai sufferers, and 116 of them had died. At the turn of the century Minamata was a scenic A major program of soil restoration was initiated in coastal town of 12,000 people who made their living 1979. By 1992, 36 percent of the contaminated area of from wood products, oranges, and fish. In 1908 a fertil- 1,500 hectares had been treated. Had the further re- izer plant was established that eventually became the lease of cadmium not been prevented, the annual costs Chisso Corporation, one of Japan's largest manufac- from medical compensation, agricultural losses, and turers of chemicals. By the 1920s compensation for soil restoration would have been $19 million a year. The damage to fisheries had already become an issue, and costs of prevention were $5 million a year. in designing the next generation of programs. Suc- lead levels in gasoline and are using price differen- cessful policies include incentives and regulations tials to encourage consumers to switch to lead-free to improve fuel quality, enhance engine perfor- products. Lead concentrations in the air have mance and maintenance, and reduce traffic vol- fallen by 50 percent or more in response to these ume. Most industrial countries and an increasing measures, and average blood lead levels in urban number of developing countries have set limits on areas have also declined substantially. 98 Clean technologies and practices can reduce lo- the eradication of smallpox, and one great medical cal industrial pollution levels even as output ex- tragedy, AIDS. Unknown prior to 1981, AIDS now pands. To encourage adoption of such technolo- dominates public health programs and health ser- gies, governments need to pursue policies that vices in several countries and may come to domi- improve the efficiency with which energy is used. nate in many more. The human immunodeficiency Such policies include the elimination of subsidies virus (HIV) that causes AIDS is transmitted for power generation and, in many countries, for through sexual intercourse. Like other STDs, it can vehicle fuels and coal. Efficient reforms help re- also be transmitted by contact with contaminated duce pollution while raising a country's economic blood (notably from transfusions) and from output. Policy options are described in full in World mother to child during the perinatal period. Ca- Development Report 1992. sual transmission from person to person does not occur. In developing countries more than 85 per- The road transport environment cent of infections occur through heterosexual in- tercourse. There is no cure, and discovery of a Motor vehicle crashes are responsible for an in- vaccine is unlikely before 2000. Action is needed creasing burden of injury and death in developing now to combat the spread of the disease. countries. Each year throughout the world road traffic injuries cause a loss of 32 million DALYs, or Why AIDS is a special case more than 2 percent of the global burden of dis- ease (Table 4.6). Men suffer roughly twice the bur- AIDS deserves special attention because failure to den from road traffic injuries as women. The control the epidemic at an early stage will result in young and the old are particularly vulnerable, as far more damaging and costly consequences in the are drivers of nonmotorized vehicles and pedes- future. trians. The number of road fatalities and injuries in developing countries is rising rapidly with urban- The HIV epidemic is bad and is getting worse. An ization and growth in the volume of traffic. Road estimated 9 million people worldwide carried the fatalities in Africa increased fourfold between 1968 HIV virus in 1990; as many as 26 million could be and 1988, whereas in Europe they declined by infected by 2000, according to WHO estimates (see more than 20 percent during the same period. Table 1.3 in Chapter 1). AIDS will then contribute A multipronged approach to road safety can re- about 3.3 percent to the global burden of disease, duce crashes at reasonable cost. Public investment and 1.8 million people will die of AIDS each year. in improved road infrastructure and highway op- Given the short time it takes infection rates to eration systems, remedial action at known "black- double in many developing countries and the spots" with high accident rates, and expanded rapid spread of the disease to countries that previ- public transport systems all make a difference. ously had low numbers of infections, total figures Legislation, financial incentives, and programs of in 2000 may be two or three times higher than the road safety education can improve driver behav- above projections. ior, reduce traffic speeds, promote use of seat More than 80 percent of those infected lived in belts, improve vehicle safety, and reduce drunk developing countries in 1990; by 2000 this will in- driving. The insurance and legal liability systems crease to an estimated 95 percent. In Thailand one may also offer powerful incentives for road safety. in fifty adults is infected. In Sub-Saharan African A carefully designed package of measures such as one in forty adults is already infected, and in cer- those mentioned above can, over time, reduce tain cities of Africa the prevalence of infection is as road fatalities and injuries by at least one-fifth, high as one in three. In some of these high-preva- thus preventing the loss of at least 6 million lence communities AIDS is already starting to re- DALYs a year worldwide (Table 4.6). Several coun- verse long-term declines in child mortality. tries, including Kenya and Malaysia, have set The cost-effectiveness of interventions drops more ambitious targets for reductions in deaths sharply when infections cross from high-risk groups to and injuries over the next decade. the general population. Since there is no vaccine or cure for AIDS, primary prevention is the only way AIDS: a threat to development to fight the disease. In the absence of adequate preventive action, AIDS spreads rapidly in the Historians will look back on the latter half of this "core" groups (such as sex workers and their cli- century as having had one great medical triumph, ents), followed by a slower and then accelerating 99 Early intervention against AIDS prevents spread of mand for health care for AIDS patients will crowd the disease to the general population. out the needs of other patients. Furthermore, the number of tuberculosis cases is increasing dramati- cally as a direct result of HIV, and the presence of Figure 4.7 Simulated AIDS epidemic HIV worsens problems with other sexually trans- in a Sub-Saharan African country mitted diseases. (STDs both facilitate HIV trans- mission and are harder to treat in HIV-infected individuals.) AIDS cases per 1,000 sexually active population Prevention of AIDS involves sensitive and politi- 100 cally charged issues. Preventing HIV infection often necessitates working with socially marginalized groups (including, in many cultures, homosex- 80 uals), and with people who pursue illegal activities such as drug use or prostitution. In addition, an effective preventive program must reach out to in- 60 A form young people frankly about sexual practices and risks. These activities offer little political bene- 40 +-.- Spread into general population fit and may be highly controversial. Strong gov- ernment will and commitment are therefore essen- tial to effective programs, the more so because the seven-to-ten-year lag between HIV infection and 20 the development of AIDS makes it tempting for countries and individuals to put off dealing with Spread through AIDS issues until it is too late to avert a wide- core groups spread epidemic. 0 0 20 40 60 80 100 Prevention: an absolute necessity Time from start of epidemic (years) A combination of strategies, backed up with ade- Source: Adapted from Potts, Anderson, and Boily 1991. quate resources, is required for stemming the spread of AIDS. Crucial elements in these strate- gies are providing information on how to avoid infection, promoting condom use, treating other spread in the general population (Figure 4.7). sexually transmitted diseases, and reducing blood- Early and effective targeting of HIV interventions borne transmission. These measures are especially is critical because these interventions diminish in cost-effective when targeted at the relatively few cost-effectiveness as the infection moves out of the people in the core groups. Unless effective preven- high-risk, high-transmission core groups. The tive action is taken, the number of new HIV infec- large number of new sexual contacts in the core tions can be expected to grow, especially in parts groups means that each HIV case avoided in this of Asia. But a comprehensive AIDS prevention group can avert more than ten times as many addi- program could check the growth of the disease tional infections as can a case avoided in the gen- (Figure 4.8). eral population. Current annual worldwide expenditure on AIDS has catastrophically costly consequences. AIDS prevention is about $1.5 billion a year. Per- AIDS, affecting as it does mainly people in the haps less than $200 million of this is spent in devel- economically productive adult years, has powerful oping countries, where 85 percent of all infections negative economic effects on households, produc- occur. Among developing countries Thailand tive enterprises, and countries (see Box 1.2 in spends the most for AIDS prevention, with 1992 Chapter 1). Because so many of its victims are spending of $45 million, more than 75 percent of heads of households or parents, AIDS devastates which was from government funds. Total AIDS families. Heavily infected countries have found spending on prevention in all Sub-Saharan Africa their health systems burdened with costly cases of was only twice this amount, with a mere 10 per- AIDS-related opportunistic infections. If the AIDS cent from government funds. A recent study for epidemic continues unchecked, the accelerated de- WHO's Global Program on AIDS suggested that 100 comprehensive AIDS and STD prevention services Effective prevention can markedly slow the rate of for all developing countries would cost $1.5 billion new infection with HIV to $2.9 billion a year. This is ten to fifteen times current spending, but it would yield enormous benefits. The estimated number of new adult HIV Figure 4.8 Trends in new HIV infections infections averted by such spending between 1993 under alternative assumptions, 1990-2000: and 2000 would be about 9.5 million-4.2 million Sub-Saharan Africa and Asia in Africa, 4.2 million in Asia, and 1.1 million in Latin America. Sub-Saharan Africa Groups to be targeted New infections per year (millions) 3 Preventive efforts must reach populations with di- verse needs: people at particularly high risk of ac- quiring and transmitting HIV infection (core groups), young people, and women. Preventive programs for the population at large are less cost- case effective than targeted programs but are needed to increase awareness and understanding of AIDS and STDs, reduce discrimination against infected Optimistic case persons, and prepare the way for subsequent in- 0 terventions when levels of infection rise. Monog- 1990 1995 2001) amy might be encouraged as part of public infor- mation efforts to curb the spread of HIV, but it cannot be the only strategy; even where it is the Asia societal norm, not all individuals adhere. High-risk groups may include sex workers, mi- New infections per year (millions) grants, members of the military, truck drivers, and 4 drug users who share needles. For these groups, prevention of sexual transmission essentially Worstd means education on safer sex, promotion of con- dom use, and prevention and treatment of STDs. 2 It is important not simply to provide information on condoms but also to ensure their availability and to empower members of the core group, espe- I cially female sex workers, to use them. Brothel managers and clients must also be persuaded of the need to change their behavior; experience from 1990 1995 2000 Zaire and other countries shows that promotion of condoms to male clients substantially improves the success of programs targeted at sex workers. Areas of high STD prevalence warrant aggressive Note: Asia includes China, India, and Other Asia and attempts to control STDs through condom promo- islands. Source: World Health Organization and World Bank data. tion, case management and counseling, notifica- tion of partners, and surveillance. These can be provided through a wide spectrum of health insti- tutions such as family planning clinics and pri- mary health centers. particularly important because men so often domi- Young people, both in and out of school, need nate the sexual relationship. The curriculum comprehensive education on reproduction and re- should be sensitive to local cultural conditions but productive health issues. To be most effective, ed- should provide explicit, honest explanations of ucation must begin before the onset of sexual ac- sexuality, gender issues, safe sexual practices, tivity (ages 12-14 in many countries) and must be STDs and HIV, safe motherhood, and family plan- targeted at boys as well as girls. Reaching boys is ning. All potential behavioral choices, including 101 abstinence and condom use, should be presented. PROVIDING INFORMATION. Informing people There is no evidence to support the objection that about the steps they can take to protect themselves providing sex education encourages promiscuity. against HIV infection is central to any strategy for In societies in which it is unacceptable for teachers combating AIDS. Individuals need to know that to provide sex education, the task can be delegated the risk of infection can be minimized by reducing to qualified voluntary groups. the number of new sexual partners they have, by Women are biologically more susceptible to ac- choosing partners of lowest risk, by avoiding con- quiring HIV infection through heterosexual inter- tact with contaminated blood, by using condoms course than are men, and social factors often add and refraining from risky sexual practices such as to the risks. In Uganda, for example, more than 60 anal sex, and by avoiding or seeking treatment for percent of infected persons are women. Preventive cofactors such as STDs. Intravenous drug users efforts addressed to women, especially those of can lower their risks by using clean needles. In childbearing age, can protect both maternal and communities where the HIV virus is present, peo- child health. In many African countries AIDS and ple should be aware that unprotected sex is safe HIV-related illnesses are already among the top only with a person known unequivocally to be un- ten causes of childhood mortality. At present there infectedfor example, someone who has not yet is no way to prevent HIV transmission from an been sexually active and has no other risk factors infected woman to her fetus; about 30 percent of (such as intravenous drug use or transfusion) or the babies of infected women are born with the who has recently undergone HIV testing and has infection. Most such babies survive their first year been found to be uninfected. but succumb to opportunistic infections during their second or third year. The uninfected children ENCOURAGING CONDOM USE. Condom use is ef- of infected mothers are also at increased risk of fective in slowing the spread of both HIV and dying because they are likely to lose one or both of STDs and needs to be encouraged in all risky sex- their parents. The only strategy for fighting child- ual encounters. Programs to promote condom use hood AIDS is to target preventive efforts to in highly vulnerable groups such as clients of sex women of childbearing age. workers are cost-effective. One such program tar- Babies can contract HIV through breast milk, geted to low-income sex workers in Nairobi re- creating difficult tradeoffs between the risk of in- duced the mean annual incidence of gonorrhea fection and the benefits of breastfeeding for child from 2.8 episodes per woman in 1986 to 0.7 epi- health. Recent studies suggest that the risk is sub- sode in 1989. The program averted an estimated stantial (about 30 percent) for babies breastfed by 6,000 to 10,000 new HIV infections a year at an mothers who develop an HIV infection after child- approximate cost of $0.50 per DALY gained. Com- birth; the risk for babies of women who are infec- munity-wide interventions are also being tried ted prenatally is smaller, although still significant. with success. In Zimbabwe a community interven- Randomized controlled studies are under way in tion estimated to cost $85,000 successfully reached Haiti, Kenya, and Rwanda to determine the risks more than 1 million persons, distributed more more accurately. In the meantime, in areas where than 5.7 million condoms, and reduced STDs in the primary causes of infant deaths are infectious the general population by 6 to 50 percent in differ- diseases and malnutrition, breastfeeding should ent areas. The intervention also changed behavior probably continue to be recommended. In areas among sex workers: the proportion reporting con- where a safe alternative to breastfeeding exists, dom use with their last client rose to 72 percent, testing of pregnant women would provide an op- from only 18 percent before the intervention. portunity to advise those infected about the health Social marketing is another strategy for promot- risk of breastfeeding for their babies. ing condom use. In Zaire distribution outlets from pharmacies to traditional healers and from Specific preventive interventions nightclubs to street vendorswere saturated with condoms. Condom sales rose from 20,000 in 1987 Widespread experience with national AIDS con- to 18.3 million in 1991. Consumer research indi- trol programs in industrial and developing coun- cates that 90 percent of the condoms were bought tries is already on hand. It suggests some areas in by men and that about 60 percent were intended which action needs to be taken and provides im- for casual sex. Estimates suggest that the program portant lessons for programs to control AIDS. averted about 25,000 HIV infections in 1991 alone. 102 REDUCING BLOOD-BORNE TRANSMISSION. Blood ning clinics has revealed infection rates as high as transfusions account for less than 5 percent of HIV 20 percent. transmission worldwide, but transfusion with in- fected blood almost always leads to infection. Mea- TESTING AND SCREENING. Voluntary HIV testing sures for preventing transmission of HIV through provides individuals with useful information blood transfusions include reducing the need for about themselves and their partners. Studies sug- transfusions, eliminating payments for donated gest that counseling and testing can help individ- blood (because paid donors tend to have a higher uals and couples adopt safer sexual behavior. The risk of HIV), and screening donors. Effective early once-prohibitive cost of testing has been declining treatment of health problems, combined with edu- sharply; a couple can now be screened for less cation for health care providers about the proper than $2 (excluding the cost of counseling). The indications for transfusion, can cut the need for test, however, is not always reliable because there transfusions by more than 50 percent. Public is a short period during which HIV may not be health programs such as helminth control in detected in a newly infected person; furthermore, schools, iron supplementation, prenatal care, and a negative test result is no guarantee of continuing malaria control can reduce the severity of existing risk-free behavior. As a result, testing is currently anemia and thus diminish the need for transfu- most useful for couples within or planning a long- sions. When transfusion is unavoidable, blood do- term relationship. Governments need to ensure nors can be screened to ensure a supply of unin- that testing remains voluntary and anonymous, fected blood. The cost-effectiveness of blood meets quality standards, and is accompanied by screening varies dramatically depending on the appropriate counseling or information. prevalence of HIV. To maintain the population's overall confidence in the medical community, SURVEILLANCE. Public health surveillance for however, blood screening has been advocated HIV is critical in areas where extensive spread of even where HIV prevalence is low. When blood the virus has not yet occurred. Countries that es- banks exist, screening adds only about 5 percent to tablish a timely and reliable system of surveillance the total cost of each unit of transfused blood. are able to give policymakers early warning of an impending spread of the virus. Useful activities INTEGRATING AIDS PREVENTION AND STD SER- include regular surveillance for HIV and syphilis VICES. Wide availability of STD services is crucial in a few prenatal clinics and in centers serving cli- for fighting AIDS because HIV transmission is fa- ents at high risk of infection, such as sex workers cilitated by the genital lesions and inflammation and patients with STDs. associated with STDs. Treatment of STDs is often Several of the preventive interventions dis- highly cost-effective in its own right. It becomes cussed above have been incorporated in the Indian even more cost-effective when the benefits of re- government's National AIDS Control Project, duced HIV transmission are added. Curing each launched in 1992 with the assistance of the World case of gonorrhea in a core group saves 120 Bank and WHO. The Indian program emphasizes DALYs, at a cost well below $1 per DALY gained if promoting public awareness about AIDS, promot- the benefits of fewer secondary cases and reduced ing health in core groups through NGOs, control- risk of HIV transmission are included. ling STDs, improving the safety of blood supplies, Because many STDs are asymptomatic (espe- strengthening surveillance and institutional capa- cially in women), infected individuals may have bilities for control of HIV-AIDS, and encouraging little motivation to be tested and treated. Even for the humane treatment of people with AIDS or HIV those with symptoms, charges for clinical services infections. may reduce access to treatment and therefore in- crease the spread of STDs. In Nairobi, for exam- Care of AIDS patients ple, the introduction of fees at the main STD clinic reduced attendance by 60 percent among men and In 1992 developing countries spent about $340 mil- 35 percent among women. Subsidizing STD ser- lion to care for AIDS patients. Although this is vices therefore makes sense as part of an AIDS only a small fraction of the $4.7 billion spent by control program. Another good strategy is to com- industrial countries to care for their AIDS patients, bine STD and family planning services; screening it is still nearly twice the amount spent on AIDS for asymptomatic STD infections in family plan- prevention in the developing world. If spending 103 per patient remains constant, the amount spent on tial sum. Palliative home care, by contrast, costs the care of AIDS patients in developing countries between $30 and $75 per DALY gained, but it may will more than triple, to $1.1 billion in 2000. To be a substantial burden to the family. Uganda's date, many AIDS control programs have not de- innovative activities have made it possible to pro- veloped guidelines for the cost-effective provision vide caring responses, at modest cost, to those af- of care for AIDS patients. Antiviral drugs such as flicted (Box 4.9). azidothymidine (AZT) are enormously expensive, have severe side effects, and may, at best, delay The need for national and international action the onset of AIDS and prolong life to some extent. One year of AZT costs more than $3,000, a prohib- At present, most national AIDS programs are in- itively high figure. Treatment options in many adequate, despite international attention and the low-income nations are therefore limited to allevia- significant effort by WHO to help design and im- tion of pain and management of the opportunistic plement plans for controlling AIDS. Most pro- infectionsmost commonly, tuberculosis, diar- grams use only the resources available to minis- rhea, and candidiasisthat afflict HIV-infected tries of health, are too standardized, and neglect persons. Strategic planning can greatly reduce the control of STDs. AIDS has to be approached as costs through the use of a small number of less- a national development issue. National leadership expensive drugs and outpatient or community is crucial; the most effective programs, such as treatment where possible. Basic care, including Thailand's, pursue strategies that involve many outpatient treatment of opportunistic infections, agencies, in and outside government, in an atmo- can cost $200 to $400 per DALY gained, a substan- sphere of openness and frankness (Box 4.10). Box 4.9 Coping with AIDS in Uganda By June 1992 Uganda had reported 33,971 AIDS cases; TASO has grown to include ninety-seven counselors, the true number may be between 100,000 and 300,000, three supervisors, and six trainers in eight locations. and it is estimated that I million to 1.5 million Ugan- Services, which reach more than 30,000 people a year, dans are infected with HIV. In Kampala more than 30 include counseling, condom education and distribu- percent of all pregnant women are infected, and in tion, home care, income-generating activities, feeding many parts of the country AIDS is the most common programs, and payment of orphans' school fees. cause of admission and death among hospitalized In 1990, to address the demand for personal testing, adults. With this immense burden, care of infected in- Uganda's first anonymous HIV testing and counseling dividuals and management of the social consequences center was established. The enormous demand has of infection are perceived to be as important as preven- made individual pretest counseling impossible, but tion of further cases of HIV. group counseling has become popular. Individual post- In response, a variety of innovative activities have test counseling continues to be offered, and HIV-posi- been undertaken. In 1987 the first AIDS clinic was tive patients are referred to TASO for further support. opened, with a small staff, a few drugs, and little out- AIDS awareness in Uganda is so high that many peo- side support. The clinic recently enrolled its 8,000th ple assume they are infected. Couples who are tested patient. Patients regard the care they get there as much and found to be negative report they are more moti- higher in quality than that available elsewhere. The vated to be monogamous, and a small follow-up study founder of the clinic, Dr. Ely Katabira, and another found that such clients have fewer casual sex partners physician at the national teaching hospital have pro- and use condoms regularly. Additional centers have duced a 104-page manual on AIDS care that recom- been established in other areas, as well as an executive mends simple diagnostic and treatment strategies for testing center for businessmen and parliamentarians AIDS; for example, nine relatively inexpensive drugs uncomfortable about being served in the busy public used in combination with tuberculosis therapy can clinic. High demand indicates that Ugandans want to achieve a high degree of relief for patients with AIDS. know whether they are infected, particularly before Also in 1987 sixteen Ugandans who were personally embarking on important life events such as marriage. affected by AIDS (because of their own infection or that Uganda's experience demonstrates that an AIDS-test- of a family member) set up a new voluntary organiza- ing program in a country with a high prevalence of tion, The AIDS Support Organization (TASO), to pro- heterosexual transmission can have a more positive in- vide emotional support for AIDS sufferers. Twelve of fluence on behavior than results from the industrial the founding members have since died of AIDS, but world would indicate. 104 Box 4.10 HIV in Thailand: from disaster toward containment As late as 1988 Thailand and the test of Asia were con- dination Bureau in the office of the permanent secre- sidered to be relatively free of HIV infection, leading tary of the prime minister. The multisectoral bureau senior Thai health experts to conjecture that Asians coordinates the planning and budgeting of AIDS activ- might be less susceptible to the disease. That year, ities among fourteen ministries, international funding however, an explosive HIV epidemic started its march agencies, and local sources of support. The bureau also through Thailand, affecting all levels of society. Today facilitates the planning of joint activities with private it is estimated that 2 percent of sexually active adults, businesses and NGOs. or 400,000 to 600,000 people, are infected. Without ef- Thailand's strategy has led to a broad consensus fective prevention, by 2000 the number infected may be within the country on the importance of taking action. as high as 2 million to 4 million. Spending for AIDS prevention was $28 million in 1991 Faced with the HIV epidemic, Thai officials have and $45 million in 1992. To monitor the epidemic, Thai- moved quickly from complacency to action. Thailand, land has established the world's most comprehensive they realized, could not sustain its 10 percent annual national HIV surveillance system, which reports twice growth of GNP in the presence of a huge AIDS epi- a year on HIV prevalence in all risk groups in all pro- demic. Indeed, in 1991 researchers projected that the vinces of the country. Acknowledging that commercial aggregate direct and indirect cost of AIDS could be as sex is ingrained in Thai society and will remain so in high as $8 billion over the next decade and that AIDS the short run, the government has decided to mandate could have negative effects on tourism, foreign invest- and enforce a policy of 100 percent condom use at the ment, and labor remittances from abroad. They ar- brothels. This ensures that brothels cannot compete for gued, however, that a major preventive effort, with the customers seeking condom-free sex. Preliminary evi- goals of reducing numbers of sexual partners by at least dence shows very high rates of condom use, with de- one-half, doubling condom use, and treating STDs, mand increasing from 10 million a year to about 120 could mean 3.5 million fewer infections and more than million a year and reductions in the incidence of STDs. $5 billion in savings by 2000. The prime minister's office is launching national AIDS prevention is now being accorded the highest campaigns through the mass media in 1993-94 to pro- priority in Thailand, and a national AIDS prevention mote changes in the sexual culture and the sexual and control committee chaired by the prime minister norms of the population. Only time will determine has been formed. In 1992 the cabinet approved the es- whether intervention has been prompt and effective tablishment of the AIDS Policy and Planning Coor- enough to halt the further spread of HIV No single strategy in the fight against AIDS will Areas with an HIV epidemic but as yet little meet the needs of every country. Three main crite- disease (for example, Thailand, and urban areas of ria can guide the choice of priorities from the range India) need to develop AIDS prevention programs of HIV-AIDS interventions listed above. These cri- for the entire population while continuing to target teria are current HIV prevalence, risk of future high-risk groups. Voluntary HIV testing and coun- spread, and existing AIDS burden. Strategies for seling and preparation for the care of AIDS pa- different countries and regions within countries tients should also begin. fall into four main groups. Finally, areas with a major epidemic and a Areas with little HIV and few STDs (for ex- high disease burden (for example, Uganda and ample, rural areas of northern China and North Zambia) have to combine a broadly based preven- Africa) should emphasize comprehensive repro- tive strategy with attention to care for AIDS pa- ductive health education for youth, with some at- tients (see Box 4.9). tention to AIDS prevention among high-risk Nongovernmental organizations can play a vital groups, and should establish a sensitive HIV sur- role in prevention, care, and community support veillance system. programs, using their credibility and access to Areas at high risk of an epidemic from early reach those at highest risk, such as intravenous spread of HIV or having a high rate of STDs (for drug users and sex workers. Such groups have example, Yunnan Province in China and Sura- been highly effective in using social marketing to baya, Indonesia) should undertake massive, tar- reach individuals at the grass-roots level, particu- geted preventive activities for high-risk groups, in- larly by initiating peer education and media pro- cluding sex workers, supplemented by general grams that reinforce behavior change and work to education and by testing of the blood supply. modify the perceived social norms. A recent 105 Table 4.7 Costs and health benefits of public health packages in low- and middle-income countries, 1990 Annual cost (dollars) Disease burden Country groupand component of package averted Per participant Per capita Per DALY (percent) Low-income (income per capita = $350) EPI Plus 14.6 0.5 12-17 6.0 School health program 3.6 0.3 20-25 0.1 Other public health programs (including family planning, health, and nutrition information)b Tobacco and alcohol control program 2.4 0.3 1.4 - - 0.3 35-50 0.1' AIDS prevention program° 112.2 1.7 3-5 2.0 Total - 4.2 - 8.2 (1.2) Middle-income (income per capita = $2,500) EPI Plus 28.6 0.8 25-30 1.0 School health program 6.5 0.6 38-43 0.4 Other public health programs (including family planning, health, and nutrition information)" 5.2 3.1 Tobacco and alcohol control program 0.3 0.3 45-55 Q3d AIDS prevention programe 132.3 2.0 13-18 2.3 Total 6.8 (0.3) - 4.0 Note: Numbers in parentheses refer to per capita cost as a percentage of income per capita. Although costs are estimated for 100 percent coverage, the health benefits are based on 95 percent coverage for EN Plus and 80 percent coverage for the school health, AIDS prevention, and tobacco and alcohol programs. Includes information, communication, and education on selected risk factors and health behaviors, plus vector control and disease surveil- lance and monitoring. The health benefits from information and communication and from disease surveillance are counted in the other public and clinical services in the health package. The health benefits from vector control are unknown. Calculation of the potential disease burden averted through this program assumes no change in the prevalence of smoking and alcohol consumption; if such prevalence were to rise, the potential benefits would be larger. Exiudes treatment of STDs, which are in the clinical services package; see Table 5.3. Source: Authors' calculations. needs-assessment study conducted in a number of gains at modest cost. Local conditions vary, but an developing countries showed that the full poten- essential public health package is likely to include: tial of NGOs was not being realized for lack of The Expanded Programme on Immunization, financial, managerial, and technical support. Plan- including micronutrient supplementation ning is under way for a program to provide inno- School health programs to treat worm infec- vative mechanisms for simple and flexible assis- tions and micronutrient deficiencies and to pro- tance to nongovernmental groups working on vide health education AIDS. Programs to increase public knowledge about The world must do more to deal with the global family planning and nutrition, about self-cure or challenge of AIDS. No country is immune from a indications for seeking care, and about vector con- future HIV epidemic, and the costs of delay are trol and disease surveillance activities high. A global coalition is needed that will encour- Programs to reduce consumption of tobacco, age and assist governments to take bold action be- alcohol, and other drugs fore it is too late. Without a substantial increase in AIDS prevention programs with a strong STD political commitment and leadership-as well as component. additional resources to support the $1.5 billion to $2.9 billion needed annually for effective preven- This public health package would yield large bene- tion of AIDS-the HIV epidemic could cause a fits at low cost (Table 4.7). In low-income countries health disaster and an enormous setback for it would avert more than 8 percent of the burden development. of disease at a cost of just $4 per capita (1.2 percent The essential public health package of income per capita), while in middle-income countries it might avert 4 percent of the burden of Public health programs that address the problems disease at a cost of $7 per capita (0.3 percent of described above can produce substantial health income per capita). Because it is difficult to quan- 106 tify the health gains from the activities under services. These are included in the essential pack- "Other public health programs," the correspond- age of clinical measures discussed in the next ing cost per DALY is not estimated. chapter. Health will also be served if governments Provision of information is needed in every as- do less in a number of areasif they avoid inter- pect of the program. Information should cover the vening in food markets, cut indiscriminate sub- benefits of healthy eating, contraceptive use, and sidies for water and sanitation, remove most re- hygienic practices in the household; the health ef- strictions on contraceptive services, and abolish fects of smoking and of alcohol and drug abuse; subsidies on fuels. Appropriate government regu- and prevention of HIV infections. Some public latory action on the ambient environment, occupa- health measures will involve providing services in tional conditions, and road safety can also safe- clinics, including family planning and STD-related guard people's health. 107 Clinical services This chapter analyzes the roles of the public and Although both the public and the private sectors private sectors in paying for and delivering clinical have important roles in the delivery of clinical ser- services. It examines in depth an important con- vices, government-run health systems in many de- clusion of Chapter 3: that governments have a fun- veloping countries are overextended and need to damental responsibility for ensuring universal ac- be scaled back. This can be done through legal and cess to an essential package of clinical services, administrative changes designed to facilitate pri- with special attention to reaching the poor (Table vate (NGO and for-profit) involvement in provi- 5.1). The choice of services to be included in such a sion of health services, by public subsidies to package for each country will be strongly influ- NGOs for supplying the essential package, and by enced by information on the distribution of dis- curtailment of new investments in public tertiary ease and the relative cost-effectiveness of clinical hospitals. At the same time, the efficiency of pub- interventions. A minimum package of clinical ser- lic sector health services can be greatly enhanced vices could reduce the present burden of disease through decentralization and improved manage- by about one-quarter in low-income countries and ment of government hospitals and programs. by about one-tenth in middle-income countries. This package is affordablebut only if govern- Public and private finance of clinical services ments carry out significant health-financing re- forms that will affect the allocation of public funds Around the world, clinical services are financed and the roles of insurance and of user charges. through four main channels. Twoout-of-pocket Only by reducing or eliminating spending on payments and voluntary insuranceare private. clinical services that are outside the nationally de- The other two are public: compulsory insurance fined essential package can governments concen- (sometimes known as social insurance) that is trate on ensuring essential clinical care for the either publicly managed or heavily regulated by poor. Two key ways to reallocate government governments, and funding from general govern- spending are to increase cost recovery, especially ment revenues. by charging the wealthy for services in govern- In the poorest countries total health expenditure ment hospitals, and to promote unsubsidized in- may be as low as $2 per person a year, and more surance for middle- and upper-income groups. than half of this comes from private sources, Governments can avoid the explosive increases in mainly in the form of out-of-pocket payments. In- health expenditures that many countries are now surance mechanisms in those countries are weak, confronting by encouraging competition among and the amount of government revenues devoted providers and prepayment for care, generating to health is low. As incomes increase, so do both and disseminating information on providers' costs the percentage of income spent on health (as and insurers' products, and, in some cases, setting shown in the upper panel of Figure 5.1) and the limits on compensation of physicians and share of health spending that comes from public hospitals. sources (illustrated in the lower panel). In the for- 108 Table 5.1 Rationales and directions for government action in the finance and delivery of clinical services Conditions that may call for government action: market failure Area and poverty Directions for government action Essential clinical services Failure to treat, for example, Finance essential clinical services by reallocating current tuberculosis and STDs government spending. In low-income countries this may creates risks for the general mean increasing public expenditures for health. population. Public Require through legislation that social insurance or financing can help offset mandated private insurance cover an essential package. the additional external Encourage more private and NGO provision of essential costs to society. Poor services, through appropriate legislation and targeted people have limited ability public subsidies. to save or borrow to meet unexpected and uninsured health expenses. Families, including children, can fall into poverty because of ill health. Clinical services outside the In insurance markets Reduce or eliminate subsidization of clinical services outside essential package selection bias leads to lack the essential package. Subsidies for public provision of of coverage for high-risk services at less than cost and tax relief for employer and groups. "Moral hazard," employee health insurance payments often cover services by insulating patient and with low cost-effectiveness and primarily benefit the provider from the cost wealthy. implications of their Legislate compulsory social insurance or mandated private decisions, results in insurance, or define the national essential package overuse of services. The comprehensively. asymmetry of information Limit government involvement in delivery of nonessential between patient and services and encourage competition in service delivery by provider can cause government, NGOs, and the private sector. suppliers to induce excess demand. Regulate private insurance by, for example, requiring community risk-rating and forbidding the rejection of high-risk consumers. Define the exact content of prepaid packages of care to serve as the products bought and sold in the insurance market. Encourage the use of prepayment or salary-based approaches to provider compensation. Foster improvements in the quality of private provision by encouraging self-regulation of hospitals, medical schools, and physicians and by disseminating performance indicators. merly socialist economies and the established mar- surance pays for much care and, if so, what is the ket economies (excluding the United States) public dominant type of insurance (Table 5.2). spending accounts for a full three-quarters of total In low-income countries private out-of-pocket pay- health expenditure. ments account for more than half of the mere $2 to In addition to the four sources of health financ- $40 per person spent each year for health care. ing, there are three ways of organizing clinical Most of this sum goes for doctors' fees, payments health services: public, private nonprofit, and pri- to traditional healers, and drugs. NGOs, particu- vate for-profit. All national health systems use at larly those related to religious institutions, make least two of the twelve possible combinations of important contributions to the provision of health financing method and health service organization, services in many low-income countries. In Tan- and sometimes the different combinations serve zania and Haiti NGOs operate nearly half of the sharply differentiated populations. Even so, it is hospitals, and in Cameroon and Uganda they possible to group countries according to income manage 40 percent of health facilities. In Ghana level and the predominant system of providing and Nigeria about a third of all hospital beds are health care. A principal distinction is whether in- located in mission hospitals. Government spend- 109 As countries get richer, the i spend more of their income on health, and the public share grows larger. Figure 5.1 Income and health spending in seventy countries, 1990 Share of GDP spent on health Percent 13 United States 11 France . . Canada 9 - Lesotho 0 0 . 0 7 0 - India 0 0 0 Korea, Rep. of 00 . o 0 Spain S 0 Norway0 0 0 0 Japan 0 0 0 Jamaica 0 United Kingdom 5 0 enya S 00 - 000 000 0Brazil o Cyprus 0 0 3 0 0 - 0 Egypt I 0 5,000 10,000 15,000 20,000 25,000 GDP per capita (1991 international dollars) Public share of total health spending Percent Norway. 0 90 0 0 Hungary 5 0 United Kingdom. 0 0 0 0 5 Japan 0 0 Spain Greece 5Can da 70 - Brazil 0 0 0 o Jamaica S 0 France 0 Cyprus 00 -' 0 00 0 50 0 0 S Israel 0 0 0 Unit d States Kenya S S 0 'Egypt . Korea, Rep. of Lesotho 0 0 30 0 'Philippines 0 0 Hong Kong 'India 0 S sUganda 10 0 5,000 10,000 15,000 20,000 25,000 GDP per capita (1991 international dollars) Source: Murray, Govindaraj, and Chellaraj, background paper. 110 Table 5.2 Clinical health systems by income group Health expenditure, 1990 As share Dollars Country group and 1990 per of GNP per Main capita income (dollars) (percent) capita characteristics Examples Low-income (100-600) 2-7 2-40 High private spending for traditional Bangladesh, India, medicine and for drugs Pakistan, most Public services financed from general Sub-Saharan revenues African Little insurance countries Middle-income (600-7,900) Private insurance 2-7 20-350 Government services for middle- and low- South Africa, income groups financed from general Zimbabwe revenues Private insurance and private provision for affluent (less than 10 percent of population) Social insurance 3-7 20-400 Public health and clinical care for low-income Costa Rica, Republic groups financed from general revenues of Korea, Turkey Social insurance for wage labor force, with mixed provision Formerly socialist 3-6 30-200 Public services (which are low in quality or Czech Republic, economies of Europe collapsing) financed from general revenues Poland, Slovak (650-6,000) Large underground market in privately Republic, republics provided services of former U.S.S.R. Established market 6-10 400-2,500 Universal or near-universal coverage through France, Germany, economies, excluding general revenue financing or compulsory Japan (social United States social insurance insurance); Norway, (5,000-34,000) Use of capped third-party payments and Sweden, United global budgets Kingdom (general tax revenues) United States (22,000) 12 2,800 Combination of private voluntary insurance United States and use of general revenue from taxes Unregulated and open-ended fee-for-service compensation High administrative costs associated with health provision and insurance a. Although China is a low-income country, its health system is closer to that of a middle-income country with social insurance. Source: For expenditure, Appendix table A.9. ing from general tax revenues generally amounts weakened considerably. The rural population in- to less than half of the 2 to 7 percent of GNP allo- creasingly relies on a system of government- cated to health services. There is little or no provided health care financed in part out of gen- insurance. eral revenues, but with substantial cost recovery Until recently, China was an important excep- through user charges, not unlike systems prevail- tion among low-income countries. There, between ing in other low-income countries. 1960 and 1980, state enterprises provided health In the middle-income countries there are two major care directly to their workers or contracted with types of health systems, distinguished by whether government hospitals to do so. Rural communes the government or the private sector provides were required to earmark a portion of their finan- health insurance. Health spending, at $20 to $400 cial resources for health services for all their mem- per capita, is higher than in low-income countries, bers. By the late 1970s insurance covered virtually and both public and private managerial capacity is all the urban population and 85 percent of the rural stronger. In countries with private insurance, such populationa unique achievement for a low- as South Africa and Zimbabwe, the government income developing country. Since the elimination uses general revenues to pay for health care for of communal agriculture and the liberalization of middle- and low-income groups, while upper- industry in the early 1980s, however, these forms income households (less than 20 percent of the of health insurance and service delivery have population) use private insurance to pay for pri- 111 vate physicians and hospitals or for private rooms of essential clinical services. But what should be in government hospitals. the content of this package? Although political In countries with social insurance, mandatory considerations will inevitably affect the decision, contributions from employees and employers, and the most important factors in selecting the essen- sometimes government funds, finance insurance tial package should be the relative cost-effective- for part of the population, including most middle- ness of interventions, the size and distribution of class workers. Health care for the poor is financed the health problems affecting the population, and from general revenues. This is the system that pre- the resources available. vails in Korea, Turkey, and most of Latin America. In the formerly socialist economies of Eastern Eu- Defining the essential package rope and the Soviet Union, general revenue financing with government provision of health A patient's health needs often require several in- services was until recently the only officially recog- terrelated interventions. A child with fever and di- nized form of health care. Public spending on arrhea may require treatment for both acute respi- health now accounts for 3 to 6 percent of GNP in ratory and gastrointestinal infections. A pregnant these countries, or $30 to $200 per capita. Prior to woman needs to receive both prenatal and deliv- the period of political and economic liberalization ery care. For this reason, it makes sense to group in the late 1980s, private payments were fre- certain interventions when analyzing their costs quently made for "public" health services (for ex- and benefits. Five groups, or clusters, of clinical ample, gratuities were given physicians in govern- interventions are likely to be important in every ment hospitals), and drugs often leaked from the country's essential clinical package: prenatal and public sector into private markets. Since the politi- delivery care; family planning services; manage- cal and economic reforms that swept across these ment of the sick child; treatment of tuberculosis; countries in the late 1980s, the health systems and case management of sexually transmitted dis- there have been in crisis. Dwindling public fund- eases (STD5). The first two groups are often dis- ing and deteriorating government services have cussed under the umbrella of "safe motherhood" created strong pressures for new forms of public activities, but for the purposes of cost and benefit and private insurance. estimates they are presented separately here. All The established market economies, with the excep- five sets of interventions are highly cost-effective; tion of the United States, rely on one of the two each costs $50 or less per DALY in low-income types of public financing for more than three-quar- settings and $150 or less per DALY in middle- ters of their health expenditures, which range income settings. Moreover, they deal with wide- from $400 to $2,500 per person per year. Norway, spread health problems that affect the poor. Sweden, and the United Kingdom use general tax Details on the health problems that these five revenues to pay for health services that are pro- groups of clinical interventions help to resolve, on vided directly by the government. In France, Ger- ways of delivering the interventions efficiently to many, and Japan, among others, social insurance patients, and on the cost-effectiveness of the inter- is the dominant mode of financing. The United ventions are given in Boxes 5.1 through 5.4. The States, with annual health spending of about problems addressed are among the largest afflict- $2,800 per capita, has a bewildering combination ing developing countries. Four preventable or eas- of systems, including voluntary private employ- ily treatable infectious diseases of children account ment-based insurance, compulsory insurance for for nearly 7 million child deaths annually. Unsafe federal workers with each employee having a childbirth is responsible for half a million maternal choice of alternative insurers and packages, and deaths each year. Tuberculosis kills more than 2 full public finance and provision for veterans. A million people annually, making it the leading single-payer approach financed from general reve- cause of death among adults. More than 250 mil- nue is used at the federal level for health care for lion new cases of debilitating and potentially fatal the elderly (Medicare) and at the state level for thd STDs occur each year. poor (Medicaid). In addition to these five groups of clinical inter- Selecting and financing the essential clinical ventions, in any realistic setting an essential pack- package age would have to include treatment of minor infection and trauma, as well as advice and allevia- A basic responsibility accepted by governments al- tion of pain for health problems that cannot be most everywhere is to ensure access to a package fully resolved with existing resources and technol- 112 Box 5.1 Making pregnancy and delivery safe Under optimal conditions, about 990 of every 1,000 others to danger signs that may occur during preg- pregnancies that reach the seventh month of gestation nancy and childbirth, and mobilize communities for conclude with a healthy newborn and a healthy transport of women with complications to district mother. For most women in the developing world, hospitals however, childbirth is unsafe. About one in 50 women Community-based obstetrics with trained nurse-mid- in developing countries dies as a consequence of com- wife staff to provide prenatal care, including tetanus plications of pregnancy and childbirth, compared with toxoid immunization, treatment for syphilis, provision only one in 2,700 in the established market economies. of micronutrients (iron, folate, and iodine), and detec- Maternal mortality has profound consequences within tion of complications of pregnancy and delivery; nor- the household; the chances of dying for children under mal delivery, including prophylactic application of an- 5 increase by up to 50 percent when the mother dies. tibiotics against gonorrheal ophthalmia; obstetric first In [987 the international health community, includ- aid, including sedatives for early eclampsia (preg- ing the World Bank, WHO, the United Nations Popula- nancy-related seizures) and manual removal of the pla- tion Fund (UNPF), and agencies in forty-five countries, centa; effective early referral of severe complications; launched the Safe Motherhood Initiative. The prime and safe abortion. goal is to reduce by half the number of maternal deaths District hospital facilities to provide essential obstet- by 2000. The health programs recommended under the ric services (cesarean section, anesthesia, blood re- initiative include family planning and pregnancy- placement, manual procedures, and monitoring of la- related care, prenatal care, and delivery care. (Family bor) and neonatal resuscitation (aspiration of secretions planning and abortion services are discussed more and assisted respiration with oxygen). fully in Chapter 4.) The marginal cost-effectiveness of The emphasis given the different components will pregnancy-related care varies with circumstances, but depend on local conditions. At one extreme are dis- the World Bank has estimated that the average cost per tricts where resources are limited and women are DALY is between $30 and $110, the equivalent of less highly isolated. Here, high priorities would be prenatal than $2,000 per death averted. care aimed mainly at correcting micronutrient deficien- The extension of prenatal, delivery, and postpartum cies and infections such as STDs and malaria. At the care to 80 percent of the world's population would re- other extreme are urban and periurban areas where duce by 40 percent the burden of disease associated referral centers are overwhelmed with normal deliv- with unsafe childbirth, at a cost of between $90 and eries and the quality of care is typically low; here, $255 per birth attended, or $4 to $9 per capita. A rea- health centers should be improved so that they can sonable program of pregnancy-related care would in- deal with normal births, and the quality of hospital care clude three components: should be enhanced to provide better treatment of ob- Information, education, and communications designed stetric complications. to create demand for clinical services, alert women and ogies. Hospital capacity would be sufficient to trained staffto deliver essential services and that handle some emergency care, including most frac- inputs for services outside the nationally defined tures and infrequently needed procedures such as package are not supplied. A district hospital with appendectomies. Local discretion in the provision about one bed per 1,000 population served is of these services would depend on the availability needed to provide inpatient and specialized out- of inputs and on day-to-day capacity. This "lim- patient care, but the hospital would have to per- ited care" and the five groups of interventions to- form only basic surgery. No higher-level hospital gether constitute a minimum package of essential clin- is required for delivery of the minimum package. ical services. Efficient delivery of these essential Although doctors are needed for supervising es- services requires a well-functioning district health sential clinical care and handling more compli- system consisting of health posts and health cn- cated cases, most of the services in the minimum ters as the first point of patient contact and district package can be delivered by nurses and midwives. hospitals as referral facilities, with the two levels A ratio of fully qualified nurses to physicians of linked by emergency transport. between 2 and 4 to 1 (estimates vary to accommo- Governments must ensure that publicly pro- date the availability of physicians and nurses in vided facilities have the necessary inputsdrugs, different regions) and 0.1 to 0.2 physician per supplies, facilities, equipment, and properly 1,000 population would be adequate. Although 113 Box 5.2 Integrated management of the sick child Four groups of infectious diseasediarrhea! diseases, experience with these two disease clusters can be ex- acute respiratory infections (ARIs), measles, and ma- panded to include children with malaria, measles, and lanaaccount for more than half of the 12.7 million malnutrition. Evidence that malaria and pneumonia deaths every year of children under age 5. In the devel- overlap in their clinical presentation and can be treated oping world measles alone causes 860,000 deaths in with the same antibiotic strengthens the case for treat- children under age 5 and accounts for 6 percent of ing several diseases together. DALYs lost in that age group. Malaria causes 4 percent Under the integrated management approach, the of the disease burden in the under-five group. Sick sick child is initially assessed by means of a limited children taken by their mothers to health centers for range of questions and observation of easily recognized diarrhea! disease and for ARTs such as pneumonia of- symptoms. The child's nutritional and immunization ten receive inappropriate diagnosis and treatment, status is measured, and immunization is given if leading to unnecessary complications and deaths. needed. The child's condition is classified according to Whereas preventing diarrhea! diseases and ARIs has disease grouping and severity guidelines, which are proved difficult and is probably not cost-effective, case used as a basis for treatment and possible referrals. The management in community-based programs is feasible final step is to give the mother advice Ofl follow-up and extremely effective. WHO and UNICEF have re- care. cently begun to support national programs on the Inte- The core of the package is to train primary health grated Management of the Sick Child. This initiative care providers to diagnose diseases and prescribe the builds on more than fifteen years of experience with appropriate treatment at the health center level or refer case management of diarrheal diseases, mainly by oral immediately to a district hospital those cases with com- rehydration therapy (ORT), and about seven years of plications. An adequate supply of antibiotics, anti- research on and program implementation of case man- malarial drugs, and other drugs is critical for success. agement of ARIs. The integrated cluster of treatments, including hospital In Nepal a controlled intervention trial that relied services, would cost between $30 and $100 per DALY exclusively on indigenous community health workers saved. Since the walk-in component accounts for ap- (CHWs) to detect and treat pneumonia without hospi- proximately 60 to 70 percent of the reduction in the talization led to a 28 percent reduction in the risk of disease burden, district hospitals are not indispensable death from all causes by the third year of service. Addi- for starting the program, but their presence and proper tional benefits were obtained from the reduction in functioning add substantial health benefits. If high deaths caused by diarrhea and measles. Other research rates of health service use can be achieved, child deaths on similar community-based strategies for children un- in high-mortality communities, according to WHO esti- der age 5 indicates decreases of approximately 50 per- mates, could be reduced by between 50 and 70 percent. cent in infant mortality from ARTs. In Egypt the use of This fact and the relatively low technology involved ORT has in some areas led to a reduction of 50 percent make the management of the sick child a high priority in mortality from diarrhea and 40 percent in overall in countries with child mortality rates of more than mortality among children ages I month to 5 years. The thirty deaths per 1,000 children under age 5. many developing countries can already deliver the drugs) to reduce the risk of cardiovascular disease minimum package, some low-income nations in high-risk individuals, and inexpensive manage- would require additional investments in person- ment of angina and heart attacks. Other treat- nel, equipment, and facilities. ments that might fit into an expanded package in- In developing countries with the financial re- clude hernia repair, meningitis treatment for sources and political will to go beyond the mini- children, management of gastrointestinal ulcers, mum clinical package, a more comprehensive set cataract removal, and treatment of moderately se- of services could cover other interventions with vere injuries and of complications of diabetes. slightly lower cost-effectiveness than those in the Given the large contribution of disability to the minimum package. This set might include a num- burden of disease, inclusion of low-cost rehabilita- ber of interventions for chronic disease, such as tive measures will often be a priority for interven- use of oral hypoglycemics or insulin to control dia- tions beyond the minimum essential package. betes, medical treatment for schizophrenia and These interventionsmost of which respond to manic-depressive illness, screening and treatment conditions that will become increasingly common for breast and cervical cancer, measures (for exam- with the epidemiological transition documented in ple, use of aspirin and of simple antihypertensive Chapter 1can cost as little as $200 to $300 per 114 DALY. Including them and other interventions of essential package of services in their allocation of similar cost-effectiveness would reduce the current public spending on health. In Botswana and Zim- disease burden by 5 to 10 percent. babwe the rapid decline in infant mortality and Many health procedures have such low cost- rise in life expectancy during the 1980s were effectiveness that governments should exclude strongly influenced by government action to ex- them from the essential clinical package. In low- pand the health infrastructure and by the use of income countries these might include heart sur- general tax revenues to finance an array of public gery; treatment (other than pain relief) of highly health and clinical services. Some key services that fatal cancers of the lung, liver, and stomach; ex- were initially left out of the package but are now pensive drug therapies for HIV infection; and in- being incorporated include vitamin A supplemen- tensive care for severely premature babies. It is tation and improved control of STDs. Similarly, difficult to justify using government funds for dramatic health gains in Costa Rica in the 1970s these medical treatments when much more cost- were largely brought about by new basic public effective services that benefit mainly the poor are health and clinical services, financed almost en- not receiving adequate financing. tirely by the Ministry of Health and the national Several developing countries that have been social security agency. The remarkable improve- highly successful at improving the health status of ments in health status in China, Kerala State in their populations have emphasized access to an India, and Sri Lanka are attributable in part to gov- Box 5.3 Treatment of sexually transmitted diseases Sexually transmitted diseases (STDs) are extremely person affected is a member of a "core" group and common infections: according to a 1990 estimate by hence likely to transmit syphilis to others, and Ofl the WHO, there are more than 250 million new cases each case-detection strategy used, curing a case of syphilis year worldwide. These diseases have severe and often can cost as little as $0.10 or as much as $40 per DALY. In irreversible consequences that disproportionately af- Zambia a syphilis treatment demonstration project for fect women, who bear 80 percent of the total DALYs pregnant women achieved a two-thirds reduction in lost to STDs (excluding HIV). Women are more likely stillbirths, low birth weights, and neonatal deaths as- than men to acquire STDs because of the greater effi- sociated with syphilis, even though attendance, ciency of male-to-female transmission for most STD screening, and treatment were not optimal. The cost pathogens, the lack of female-controlled preventive was $12 per adverse outcome prevented, or less than methods, and, in many settings, gender power dy- $1 per DALY. In a similar program in Kenya prevention namics that limit women's ability to determine the con- of one neonatal death cost $50. Unfortunately, in most ditions under which sexual intercourse occurs. Women countries a comprehensive program of perinatal syph- are less likely than men to obtain care for STDs because ilis screening has not been implemented. the majority of infected women are asymptomatic and Effective interventions exist for other STDs, but lack those with symptoms may be deterred by fear of social of simple, appropriate, rapid, and inexpensive diag- stigma. Since STDs also inflict a heavy burden of ill- nostics for use in the field have made these measures ness on men, and since men are responsible for much much more difficult and expensive to implement. Con- STD transmission, disease control measures, to be ef- sequently, when resources are limited, such interven- fective, must be targeted to both sexes. tions must be targeted specifically to core groups to be Because STDs increase the efficacy of transmission of cost-effective. In these groups, case management of HIV, controlling these infections is one of the most im- chlamydia (a bacterial infection of the reproductive portant interventions for containing the spread of tract) and gonorrhea can be highly cost-effective ($10 to AIDS. But even in the absence of AIDS, STDs cause $40 per DALY), as can treatment of chancroid in areas substantial morbidity and mortality. They usually af- where it is common. Much of the cost of intervention fect people in the 15-44 age group, the most economi- for these diseases is related to diagnosis. Calculations cally productive ages. One of the main causes of .neo- suggest that if prevalence of an STD is more than 10 natal morbidity and mortality in some countries is percent, treating everyone in the risk group may be congenital syphilis. Because treatment is simple (one to more cost-effective than screening. This, however, three injections of penicillin) and inexpensive, screen- does not take into account the problems of widespread ing and treatment for syphilis during prenatal care has use of the antibiotics used to treat STDs, including ad- been recommended for the minimum package. De- verse reactions, development of microbial resistance, pending on the prevalence of disease, on whether the and change in microbial flora. 115 Box 5.4 Short-course treatment of tuberculosis Tuberculosis (TB) kills or debilitates more adults ages teen months. Drugs for the short course cost about $50 15-59 than any other disease and is responsible for to $80 per patient. Those for the standard course cost about 2 to 4 percent of the burden of disease. It is the only $10 to $15, but the cost per death averted is higher single leading cause of death in developing countries, because only 30 percent of patients complete treatment accounting for about 2 million deaths a year, or ap- and are cured, as against 60 percent for the short proximately 5 percent of all deaths and 25 percent of course. Other benefits of the short course include a preventable adult deaths in those countries. More smaller number of resistant organisms and less need women of childbearing age die from TB than from for expensive retreatment. (This discussion applies to causes associated with pregnancy and childbirth. More the treatment of sputum-smear-positive TB. Once than half of the world's population is infected with the other forms of TB have been identified, treatment costs TB bacillus. People who are malnourished or have an- should be similar except for serious forms of smear- other severe illness are at particular risk for TB, as are negative TB.) those infected with HIV. The relationship between TB Walk-in treatment is less expensive than hospitaliza- and HIV is highly significant, as each person infected tion, but if this care cannot be closely monitored (as in with HIV and TB could infect twelve other persons many rural areas), hospitalization may be more cost- with TB per year. effective. The program described is modeled on pas- Annual incidence rates of all forms of clinical TB vary sive case investigation, assuming that a person with TB from 50 to 260 per 100,000 in the developing world; will have symptoms such as cough and weight loss and more than half of these cases are infectious (sputum- will seek care and that infected persons discovered by smear-positive). For most forms of TB, 50 to 60 percent active searches will be less likely to continue treatment of those infected will die if untreated. All ages are at than those who seek care. Although the BCG vaccina- risk, but the peak is in young adulthood. In Sub- tion is important in TB control for children, its effec- Saharan Africa the annual risks of infection remain tiveness in adults is still under investigation. high, partly because of poverty and overcrowding The cost of treatment is less than $10 per DALY in all (which are risk factors for TB) and partly because any chemotherapy scenarios. It is estimated that tuber- decrease in the annual risk of infection is offset by the culosis treatment of infectious (smear-positive) individ- HIV epidemic. In India and Sub-Saharan Africa TB is uals prevents one to four new cases by stopping trans- the leading cause of death and the biggest contributor mission. The positive externalities of short-course to the disease burden; it is responsible for about 8 to 11 chemotherapy explain in part the extremely favorable percent of the DALYs lost in the 15-59 age group. cost-effectiveness and justify government intervention. There are two effective approaches to treating TB: Because the cost of drugs, at $50 to $80 per patient, is short-course chemotherapy, which uses three to five probably too high for the poor, public subsidy is espe- drugs over six to eight months, and the "standard" cially warranted for low-income households. course of two to three drugs taken over twelve to eigh- ernment policies that emphasized the financing of tries. This reduction in the burden of disease cost-effective clinical services directed especially at would be equivalent to saving the lives of more the rural poor. than 9 million infants each year. Benefits, costs, and financing of the essential clinical Delivery of the minimum clinical package would package cost an average of about $8 per person each year in low-income countries and about $15 in middle- Widespread adoption of the minimum clinical income countries. Approximately half of these package would have a tremendous positive effect amounts would be for prenatal and delivery care on the health of people in developing countries. If alone. When the cost of selected public health in- 80 percent of the population were reached, 24 per- terventions is added, total costs rise to $12 per cap- cent of the current burden of disease in low- ita in low-income countries and $22 in middle- income countries and 11 percent in middle-income income countries. The differences are the result of countries could be averted (Table 5.3). When the different demographic structures, epidemiological minimum clinical package is combined with the conditions, and labor costs in the two settings. public health package outlined in Chapter 4, the In low-income countries, where governments share of current illness that could be eliminated typically spend about $5.50 per person for health rises to approximately 32 percent in low-income and where total health expenditures are about countries and 15 percent in middle-income coun- $14 per person (Figure 5.2), the affordability of the 116 Table 5.3 Estimated costs and health benefits of selected public health and clinical services in low- and middle-income countries, 1990 Annual cost (dollars)' Per capita cost as share of Disease burden Per case or Per income per averted Country group and package per participant Per capita DALY capita (percent) (percent)" Low-income (per capita income = $350) Public health packager 4.2 - 1.2 8 Minimum essential package of clinical services - 7.8 - 2.2 24 Short-course chemotherapy for tuberculosis 500 0.6 3-5 1 Management of the sick child 9 1.6 30-50 14 Prenatal and delivery care 90 3.8 30-50 4 Family planning 12 0.9 20-30 3 Treatment of STDs' 11 0.2 1-3 1 Limited caree 6 0.7 200-350 1 Total, public health and clinical services 12.0 - 3.4 32 Middle-income (per capita income = $2,500) Public health packager - 6.8 0.3 4 Minimum essential package of clinical services - 14.7 - 0.6 11 Short-course chemotherapy for tuberculosis 275 0.2 5-7 1 Management of the sick child 8 1.1 50-100 4 Prenatal and delivery care 255 8.8 60-110 3 Family planning 20 2.2 100-150 1 Treatment of STDsd 18 0.3 10-15 1 Limited care' 13 2.1 400-600 1 Total, public health and clinical services - 21.5 - 0.9 15 Note: Figures assume coverage of 80 percent of the population. Average costs. Marginal benefits. Includes EPI Plus; school health including deworming, micronutrient supplementation, and health education; information on health, nutrition, and family planning; tobacco and alcohol control programs; monitoring and surveillance; vector control; and programs for prevention of AIDS. Benefits were calculated assuming an AIDS epidemic comparable to that in Sub-Saharan Africa today. Limited care includes assessment, advice, alleviation of pain, treatment of infection and minor trauma, and treatment of more complicated conditions as resources permit. Source: World Bank calculations. Because the poorest countries spend far less on health than do middle-income countries and the public share is smaller, thei have fewer resources available for reallocation. Figure 5.2 Public financing of health services in low and middleincome countries, 1990 El l'ublic and private spending Public spending Health spending per capita (1990 dollars) Percentage of GNP 120 100 80 60 40 20 0 0 1 2 3 4 Lowincome countries Middleincome countries Source: Appendix table A.9. 117 $12 minimum package is problematic. Paying for sential package, some degree of targeting is almost the package would require an increase in public inevitable. In countries where the wealthy do not spending for health (part of which could be fi- use government-financed services because of the nanced by donors in the short term), as well as a greater quality and convenience of privately f i- reorientation of current government expenditure nanced services, targeting may be fairly easy. In for health from discretionary to essential care. In other cases, however, user charges will have to be middle-income countries, where public spending applied selectively, relying on means testing and for health averages about $62 per person, the other targeting devices (see Box 5.5). Special ame- $22 minimum package is financially feasible if nities in teaching hospitals, for example, can be there is adequate political commitment to shift ex- priced at or above cost, since they will be con- isting resources in the direction of public health sumed exclusively by the wealthy. User charges and essential clinical care. Indeed, upper-middle- can help generate additional revenues for the es- income countries can afford public finance of an sential package, amounting to perhaps 10 to 20 essential package that goes well beyond this percent of total government spending for health. minimum. Finally, directing donor funding to public health Who should pay for the essential clinical pack- and essential clinical care can significantly expand age? There are strong efficiency arguments for di- the total resource basket available for the mini- rect financing by developing country governments mum package. of the selected public health interventions dis- User charges for public health services in devel- cussed in Chapter 4. A number of the essential oping countries have sparked much debate since clinical services, including treatment of tuber- the World Bank endorsed the concept in 1987 in a culosis and STDs, have large positive externalities. policy study on health financing. Critics argue that What is more important is that the poor are dispro- fees restrict access to care, especially for the poor. portionately affected by the disease burden of the Yet many developing countries, particularly in conditions listed in Table 5.3 and that, because of Sub-Saharan Africa, have been forced to rely in- larger family sizes, they would benefit dispropor- creasingly on fees to supplement strained budgets. tionately from prenatal, delivery, and childhood Studies on the effect of user fees are inconclusive services. Public finance of a basic package of ser- and contradictory. One reason is that some re- vices is an effective mechanism for reaching the searchers have failed to calculate the true cost to poor. patients of treatment at government clinics. People The main problem with universal government often pay dearly for supposedly "free" health financing of an essential package is that it leads to care. Recent household surveys in India, Indo- public subsidies to the wealthy, who could afford nesia, and Viet Nam indicate that each visit to a to pay for their own services, with the result that government health center actually costs patients fewer government resources go to serve the poor. two to three times the amount of the low official A policy of concentrating public resources on ser- fees. Bribes aside, the indirect costs such as trans- vices for the poor and requiring others to pay all or port and the opportunity cost of time spent seek- part of their own costs makes sense on equity ing care are substantial. grounds but also has some disadvantages. Often, Since patients are already paying for supposedly the administrative costs of targeting are high, and free or low-cost health care, new user fees, when the exclusion of wealthy and middle-income accompanied by a reduction in indirect costs and groups can erode political support for the essential improvement of services, may increase utilization. package, causing a decline in funding and quality Recent studies in four African countriesBenin, of care. Furthermore, problems of cost escalation Cameroon, Guinea, and Sierra Leoneindicate and access to insurance on the part of high-risk that this is the case and, encouragingly, that the groupsdiscussed in Chapter 3can complicate poor benefited most from these changes. Since fa- reliance on private finance. Perhaps for these rea- cilities used fees to fund services not previously sons, most member governments of the OECD fi- available locally, poor patients avoided costly nance (or mandate finance of) comprehensively travel, and the actual cost of care declined. Studies defined essential packages for virtually all their indicate that user fees amounting to less than 1 citizens. percent of annual household income have little im- In low-income countries, where current public pact on the utilization of health services, even by spending for health is less than the cost of an es- the poor. Because higher fees do decrease utiliza- 118 Box 5.5 Targeting public expenditure to the poor When public spending on health is not targeted to the tively well-off. Targeting public finance to those not pooras often happens, according to numerous participating in social insurance will reach the poor, studiesno other source of funds is likely to compen- and administration will be relatively simple. In coun- sate. Which targeting mechanisms work best in prac- tries with established social insurance mechanisms, tice will depend on their impact on demand, their ad- this targeting mechanism will often prove best. ministrative costs, their technical and managerial Let individuals self-target. The essential services requirements, and the level of political support. In are available free of charge to all, but the program is countries where incomes are too low for a minimum designed in such a way as to deter the better-off from essential package to be provided universally, there are using them. Time costs, stigma, and fewer amenities four main mechanisms for targeting the essential pack- associated with services are the usual mechanisms for age of services: encouraging self-targeting. Unfortunately, these same Assess individuals seeking services on the basis of characteristics may discourage the poor as well as the income, nutritional status, or other criteria and, de- better-off. Low-income working mothers, for example, pending on the assessment, provide services from the may find that the time, for themselves and their chil- essential clinical package free of charge or according to dren, involved in using subsidized services is an insur- a sliding scale of fees. In evaluating income, direct mountable obstacle. measurements or proxies (such as housing characteris- Target by type of service. Offer free of charge, or tics) can be used, but this tends to be more adminis- subsidize heavily, services that are needed dispropor- tratively costly than other mechanisms. tionately by the poor. This sort of targeting mechanism Subsidize essential clinical services for easily iden- is inherent in much of the essential package of clinical tified subgroups of the population (for example, all services. Prenatal and delivery services, management those living in certain low-income regions or neighbor- of the sick child, and STD and tuberculosis treatment hoods or all children in public schools). Where social are all services that, if universally available, would es- insurance mechanisms (usually financed through pay- pecially benefit the poor. roll taxes) exist, they generally tend to cover the rela- tion, reducing charges or exempting the poor from ary clinical health careservices outside the essen- the fees may be warranted. tial packageis far less compelling. In fact, gov- In middle-income countries, where a significant ernments can promote both efficiency and equity part of the population may be covered by private by reducingor, when possible, eliminating or social insurance, governments can target public public funding for these services. Doing so re- monies to essential clinical services for the poor by quires recovering the cost of discretionary services legally defining and mandating that the national provided in government health facilities and cut- essential package be covered in all insurance poli- ting subsidies to private and public insurance cies, thereby freeing government resources to tar- schemes that finance discretionary care. By reduc- get the poor. Surprisingly, emerging managed care ing spending on these services, governments can institutions in developing countries often fail to concentrate public expenditure where it will do the cover benefits that would be in such an essential most goodin public health and cost effective clin- package. In the 1980s in Brazil, for example, many ical services. health maintenance organizations failed to include Out-of-pocket payments are the main source of immunizations and family planning in the basic financing for discretionary care in low-income benefits package for their clienteles. countries. They remain substantial in middle- income countries, but insurance becomes increas- Insurance and finance of discretionary clinical ingly important as incomes rise. Because, except services for the very rich, out-of-pocket financing cannot cover expensive care or deal with catastrophic ill- Public financing of a national essential clinical ness, widespread financing of discretionary care is package can be justified because the package cre- possible only through insurance. Countries have ates positive externalities and reduces poverty. two main options for meeting a growing demand The case for government financing of discretion- and need for insurance. One is to move toward the 119 current U.S. system, which relies substantially on the Philippines, a combination of limited private private voluntary insurance. The other is to follow insurance and the ability of upper-income groups the examples of Canada, Japan, and most Eu- to pay makes it feasible for governments to charge ropean countries, where general government reve- for discretionary care delivered in public hospitals. nues or social insurance cover the cost of relatively In Kenya the government is currently attempting comprehensive essential packages, leaving only a to recover the cost of caring for the insured at the small discretionary residual for private insurance. national referral hospital in Nairobi. In Lesotho Government policy can improve the functioning charges in the private ward of the central hospital of insurance markets in three ways. It can strive to in Maseru were increased in 1990 to recover costs eliminate unfair subsidies to insurance. It can from wealthier patients. work to maximize the population covered by in- In middle-income countries insurance becomes surance by preventing selection biasthe ten- more important as a mechanism for financing dis- dency of insurers to discriminate against bad cretionary services. In South Africa private insur- health risks. And it can help to eliminate another ance covers about 15 percent of the population and potential problem with insurance: the explosive accounts for more than a third of total health increases in health care costs that are closely asso- spending. In Brazil, even though everyone is eligi- ciated with fee-for-service payment of health pro- ble for publicly financed health services, about a viders by third-party insurers. fifth of the population is also privately insured. Social insurance, in which payroll deductions are Redirecting public funding from discretionary care earmarked for health care, is widespread in mid- dle-income countries, especially in Latin America. There is substantial scope in the developing world Such payroll taxes account for a quarter or more of for redirecting current public spending away from national health spending in Costa Rica, Korea, and discretionary services. Cost recovery in govern- Panama (Table 5.4). In countries with broader in- ment hospitals, especially from the wealthy and surance coverage there is even greater potential for insured, is one important mechanism. Even in cost recovery than in poorer countries. Public hos- low-income countries, where insurance may ac- pitals in Chile are now being encouraged to count for less than 5 percent of total health spend- charge, particularly for patients who have private ing, as in Ethiopia, Kenya, Lesotho, Pakistan, and insurance. Table 5.4 Social insurance in selected countries, 1990 (percent) Social insurance as share of Share of population covered public sector health Social insurance as share of Health expenditure as Group and country by social insurance expenditure total health expenditure share of GNP Low-income India 5 9 2 6.0 Kenya 10 7 4 4.3 Indonesia 13 17 6 2.0 Middle-income Dominican Rep. 6 9 6 3.7 Ecuador 9 11 7 4.1 Colombia 15 18 8 4.0 Paraguay 18 24 13 2.8 Philippines 38 12 6 2.8 Panama 50 55 43 7.1 Turkey 58 26 14 4.0 Costa Rica 82 85 62 6.5 Korea, Rep. of 90 50 25 6.6 High-income Germany 75 76 63 8.0 Japan 100 64 56 6.5 France 100 95 71 8.9 Netherlands 100 94 73 7.9 Source: World Bank data; Mesa-Lago 1991; de Geyndt 1991; Vogel 1990; Brotowasisto and others 1988; Ikegami 1992; Hurst 1992; and Solon and others 1992. 120 The other way to redirect government spending experience with such arrangements in developing away from discretionary care is to phase out public countries. subsidies to insurance. These subsidies, which are large and widespread, take the form of both direct Eliminating unequal access to clinical care budgetary transfers to insurance institutions and under insurance tax concessions for employers' and employees' in- surance contributions. They benefit the better-off A serious problem with relying on insurance to and are therefore regressive. pay for discretionary care in developing countries In Latin America governments in Guatemala, is that individuals and groups often have unequal Honduras, Mexico, Nicaragua, and Venezuela access to insurance coverage. The problem is espe- contribute a percentage of individual workers' cially acute with voluntary private insurance be- wages to social security sickness and maternity cause of selection bias. When insurers rate individ- funds. In Chile and Uruguay the government cov- ual risks, they often either refuse to insure the sick ers the operating deficits of the funds, and in Co- and elderly or make insurance prohibitively ex- lombia and El Salvador it pays directly for a part of pensive for these clients because their expected the cost of social security health services and ad- costs are so high. They also commonly exclude ministration. Private insurance in South Africa many health conditions that should ideally be cov- and Zimbabwe receives large public subsidies in ered. In Brazil, for example, where government the form of tax deductions for employer and em- regulation of insurance is weak, private insurers ployee contributions. Fees to insured patients frequently refuse to enroll persons who are poor using government hospitals in Zimbabwe are set health risks, and they fail to cover costly risks such at perhaps a third to a quarter of actual costs, pro- as HIV-AIDS. In South Africa and Zimbabwe pri- viding another subsidy to the better-off. vate health insurance schemes known as medical Once subsidies to private and public insurance aid societies have historically covered entire indus- are established, they are extremely difficult to tries and occupational groups, but the recent mar- eliminate. Recipientsgenerally the better-off- keting of individual insurance policies has intro- view them as an important benefit. The current duced discrimination based on risk. Selection bias debate in the United States over reducing the em- is an important reason for the incomplete insur- ployer tax deduction for contributions to medical ance coverage in the United States, where 37 mil- insurance illustrates this strong political resis- lion people, or about 15 percent of the population, tance. To date, very few developing countries have are uninsured. successfully cut subsidies. An exception is Zim- Very few countries, developing or industrial, babwe, where major political shiftsnational inde- have managed to eliminate selection bias under pendence and the advent of a democratically elec- private insurance, even though it could theo- ted government in 1980led to a significant retically be done by prohibiting insurers from rat- scaling back of tax breaks for medical aid pre- ing individuals' health risks and requiring them to miums. The Chilean government is actively con- rate only large groups, or "communities," in sidering eliminating the existing tax concession for which high risks are spread over a large number of employers' contributions to social insurance. people to minimize the cost effect of the risks. Subsidies to insurance systems that cover only Eliminating selection bias is easier with social in- part of the population are invariably regressive, surance because contributions are compulsory and but subsidies can be progressive when insurance are normally a fixed percentage of wages for all has become universal. At that point, public sub- employees regardless of their number of depen- sidies end up benefiting mainly those outside the dents or their individual health risks. The snag is formal labor force, notably the elderly and the that since compulsory insurance generally uses poor. Examples of progressive subsidies are the 30 employment-based contributions, it is able to percent of national social insurance funding that achieve broad coverage only when most of the eco- the Korean government contributes on behalf of nomically active population is in the formal labor low-income households and the 20 percent of Ja- force. pan's social insurance spending that derives from Where only part of the population is covered by government budget transfers for retirees and the insurance, as in most developing countries, access poor. Governments could provide equally pro- to clinical services for the uninsured poor can be gressive, targeted subsidies to buy regulated pri- much more difficult than for insured, better-off vate health insurance for the poor, but there is no people. The quality of care, including essential 121 clinical services, may also be vastly inferior for the successfully upgraded separate clinical services poor. In low-income Africa the wealthy consult targeted to the poor. private doctors for their clinical care while the poor are often forced to use understaffed government Containing the costs of clinical care health centers that lack the most basic drugs and Escalating health spending is perceived as a crisis equipment. In much of Latin America middle-class when it begins to crowd out other sectors of the families receive better care in hospitals and clinics economy or to raise the cost of labor, threatening a belonging to the social security agency than the country's international competitiveness. This is poor are given in the badly deteriorated facilities currently happening in the United States and to a run by the ministry of health. varying extent in other high-income countries (Box One way to eliminate these disparities would be 5.6), and it is about to happen in several middle- to put all health facilities under a single adminis- income developing countries, including Chile and tration and open them to all. Few countries have Korea. taken on this politically and administratively diff i- The sources of excess health costs and growth of cult task. Costa Rica, in which the social security costs are complex and much debated. Health ser- agency manages all government hospitals, is a rare vices have a high labor content, and their produc- exception. Another solution is for the government tivity has grown slowly in comparison with that of to focus spending on the poor by investing heavily other areas of the economy. In the United States in the infrastructure (facilities, equipment, and relatively high levels of underlying morbidity and transport) needed to improve essential clinical care greater amenities in hospitals are part of the rea- for the poor and spending substantially more on son, but inefficiencies are also important. Two the associated personnel and drugs. By focusing types of inefficiency stand out: high administrative investments on peripheral health units (health costs and unnecessary use of an ever-expanding centers, subcenters, and health posts) and on staff array of sophisticated and ever more costly tech- for these facilities, Malaysia and Zimbabwe have nologies for diagnostic tests and surgical pro- Box 5.6 Containing health care costs in industrial countries In 1990 the United States devoted 12.7 percent of its mark, Germany, and Sweden) set overall limits on pay- GNP to health, as against 9.1 percent in Canada, 8.9 ments to both doctors and hospitals. In the latter percent in France, 8.0 percent in Germany, 6.5 percent group, the method of limiting payments to doctors in Japan, and 6.1 percent in the United Kingdom. The varies widely: capitation in Britain, fee-for-service in 2.7 percent annual increase in the health-to-GNP ratio Germany, and salaries in Sweden. In Germany, as a for the United States during the 1980s was the highest means of controlling expenditure, fees are reduced if among the OECD countries. U.S. health expenditures the volume of services exceeds the anticipated level. In of $2,800 per person in 1990 were nearly $1,000 above general, the OECD countries that have contained costs the average for the OECD countries. The price of better have greater government control of health health care services in relation to other goods and ser- spending and a larger public sector share of total health vices also rose much more rapidly in the United States expenditures. This is also the case in eleven developing (2.2 percent per year) than in the other OECD countries with income per capita of more than $6,000. countries. In the poorer countries there is no apparent link be- An important factor in explaining the rapid growth tween the public and private shares of health expendi- in health care costs in the United States is that doctors ture and the proportion of income devoted to health. and hospitals are paid predominantly on a fee-for-ser- In the U.S. health insurance system, with its large vice basis. Countries experiencing moderate spending numbers of insurers reimbursing providers at different growth (Canada and Japan) also use fee-for-service fot rates, administrative costs absorb about 15 percent of outpatient physician care but have devised other ways health expenditures, compared with 5 percent or less of controlling expenditures: a uniform fee structure in the other OECD countries. If the United States re- and aggressive peer review of doctors' spending pat- duced its administrative outlays to the level in these terns in Japan and fixed overall budgets for hospitals in other countries, a total of $80 billion annually could be Canada. Countries with low levels of spending (the saved, equivalent to about a third of total health spend- United Kingdom) or low recent growth of costs (Den- ing for all developing countries combined. 122 cedures. These two kinds of inefficiency appear to outstripping Korea's robust overall economic be closely linked to basic features of the U.S. growth. The share of GNP devoted to health rose health system. Open-ended fee-for-service com- from 3.7 percent in 1980 to 6.6 percent in 1990. By pensation for health providers encourages the de- 1990 Korea was spending $377 per capita for velopment of new equipment, drugs, and pro- health, putting it 50 percent above the expected cedures and leads to exploding costs because level for its income. neither providers nor patients have strong incen- To control health care costs, countries need to tives to hold down utilization or spending. A com- limit payments to health care providers (Table 5.5). plex system of multiple insurance institutions and One approach is to pay a fixed amount for each other payers, each with its own procedures, raises person, as is now done by health maintenance or- administrative overheads substantially. ganizations in Brazil and the United States and by The findings concerning health cost escalation in the British National Health Service. Another the United States and other industrial countries method, used in several OECD countries, is to are especially relevant for middle-income develop- give each hospital or network of physicians a fixed ing countries. Those countries are under pressure total budget (Box 5.7). from medical professionals, manufacturers, and Insurers may jointly negotiate uniform fees for consumers to use new medical technologies, and physicians, as is done by Japan's social insurance they face difficult policy choices related to insur- system and by Zimbabwe's private insurance sys- ance institutions and compensation of providers. tem. They can also set fixed payments for specified Korea's problems with escalating health expendi- medical procedures or standard per diem pay- tures parallel those of the United States and may ments for hospital stays. Brazil's social insurance hold important lessons for other developing system, for example, is paying standard fees for countries. hospital care according to a modified version of the The Korean social insurance system is a nation- system of diagnostic-related groups that was de- wide network of "sickness funds" covering occu- veloped in the United States for Medicare, the pational and regional groups. Worker payroll con- government-funded system for the elderly. Chile's tributions are compulsory. All Koreans are public health system is introducing comparable in- covered, and the government subsidy to extend ternal prices for its hospitals. coverage to the elderly and indigent is highly As worldwide experience amply demonstrates, progressive. As in the United States, Korea's there is no perfect insurance system: every coun- health providers are predominantly private-72 try's insurance institutions have their problems. percent of physicians and 80 percent of hospital But there are two important generalizations: third- beds are in this categoryand they are paid on a party insurance leads to cost escalation, and, in fee-for-service basis. There is little control of the general, social insurance and regulated private in- acquisition of medical technologies. Korea already surance with community rating avoid selection has more sophisticated new medical equipment, bias far better than voluntary private insurance. It such as imaging machines and lithotripters (used is difficult to achieve wide coverage of the popula- to treat kidney stones), per capita than either Can- tion in most developing countries, and without ada or Germany. During 1989-91 spending for wide coverage, public subsidies to insurance will medical devices and diagnostic products grew by inevitably be regressive. Countries, whether they more than 20 percent a year. As in Japan, physi- use social or private insurance, are finding it ex- cians often sell the drugs they prescribe, and in- tremely hard to eliminate these deeply entrenched surance reimburses the cost of nonprescription subsidies. Once widespread insurance coverage is drugs, which creates strong incentives to over- achieved, costs can easily spin out of control un- prescribe and overuse drugs. Prescription phar- less provider compensation is tightly regulated or maceuticals now account for 36 percent of health determined in ways that give incentives for cost spending, one of the highest shares anywhere. Fi- containment. Developing country governments nally, the administrative costs associated with the must be prepared to deal forthrightly with this dif- more than 300 independent insurance funds are 10 ficult set of interlocking issues. to 20 percent of operating expenses, which is simi- lar to the administrative burden of private insur- Delivery of clinical services ance in the United States. Policies related to the delivery of health services in All these causes contributed to a dramatic in- developing countries should have two main objec- crease in health spending in the past decade, far tives. The first is to improve access to essential 123 Table 5.5 Strengths and weaknesses of alternative methods of paying health providers Payment method Strengths Weaknesses Fee for service Provider's reward closely linked to Tends to cause cost inflation level of effort and output Allows for easy analysis of Creates incentives for excessive and provider's practice unnecessary treatment Per case Provider's reward fairly well tied to Technical difficulty of forcing all cases (for example, using output into standard list can lead to mismatch diagnostic-related between output and reward groups) Gives provider incentive to minimize Providers may misrepresent diagnosis resource use per individual treated in order to receive higher payment Capifation Administratively simple; no need to Gives provider incentive to select (per patient under break down physician's work into patients based on risk and to reject continuous care) procedures or cases high-cost patients Facilitates prospective budgeting May create incentives for provider to underservice accepted patients Gives provider incentive to minimize Difficult to analyze provider's practice cost of treatment Allows for consumer clout if patient can select own provider Salary Administratively simplest Loss of patient influence over provider (straight payment per behavior unless patient choice links period of work) provider salary to patient satisfaction Facilitates prospective budgeting Can easily create incentives for provider to underservice patient and to reduce productivity Source: Adapted from Reinhardt 1989. clinical services, especially for the poor. The sec- low efficacy, and drugs and supplies are stolen or ond is to increase the efficiency with which ser- go to waste in government warehouses and hospi- vices are delivered. In the public sector inefficiency tals. Although the problems are many and deep, is widespread. Clinics and outreach programs op- appropriate government policies can do much to erate poorly because of shortages of drugs, trans- reduce or eliminate these kinds of inefficiency. port, and maintenance. Hospitals are poorly orga- Most countries have mixed systems of both pub- nized and managed and keep patients longer than lic and private delivery of clinical services and both necessary. Countries pay too much for drugs of public and private financing. The type of financing 124 a country chooses does not dictate the kind of cient public sector health centers and district hos- health delivery system it should have, or the other pitals (for example, in Chile, China, Sri Lanka, and way around. Some countries, such as Botswana, Zimbabwe), and there are circumstances in which Malaysia, Sweden, and the United Kingdom, have it is impossible or too costly to persuade the pri- mainly public financing and public delivery; some, vate sector to deliver careparticularly in geo- including Brazil and Korea, have public financing graphically remote or extremely poor areas. and private delivery; and still others, such as the The existence of alternative suppliers, both pub- Philippines, the United States, and Zaire, rely pri- lic and private, creates pressure for improved per- marily on private financing and private delivery. formance. In developing countries where the pub- The private health sector typically serves a di- lic system has a near-monopoly on health care verse clientele, and it typically delivers services delivery, a mixed system that exposes the public that are perceived to be high in quality and more services to competition is likely to be more efficient responsive to consumer demand than the govern- and to improve quality of care. Furthermore, in ment's. But there are also examples of highly effi- countries where government health services are -S Box 5.7 Health care reform in the OECD The OECD countries face persistent difficulties in the ing the productivity of rationed resources. There has financing and delivery of health services. Problems in- been some convergence of systems toward contracting clude inequitable access to services, gaps in insurance between public insurers and private providers. Bel- coverage, unacceptably rapid increases in health ex- gium and France have introduced tighter contracts into penditure, inefficiency, and poor quality. These diffi- their reimbursement systems, and the United King- culties are partly the result of circumstances outside the dom has moved away from its integrated National control of governmentsfor example, demographic Health Service and toward more autonomous and and technological change. To some extent, however, competitive physicians and hospitals. Other countries they arise from flaws in the design of the financing, reformed the contractual model itself by emphasizing payment, and regulation systems for health care. Dur- consumer choice, active informed purchasing rather ing the 1980s most OECD countries initiated moderate than passive funding by third parties, and managed or major reforms of their health care systems to correct competition among providers. These reforms are most these flaws. marked in Germany, but elements of such changes Three principal types of reforms can be distin- have also been implemented in Belgium and the guished. Several countries, including Ireland, the Netherlands in the form of mixed payment systems Netherlands, and Spain, have taken steps to extend eli- that combine budgetary caps with work-related pay- gibility for public medical care, bringing the last remain- ment of providers. ing groups of their populations into the public system Finally, the Netherlands and the United Kingdom of coverage for basic medical care. Despite widespread have embarked on differing experiments to introduce calls for privatization of finance, no country has re- competition within their public systems. In the Nether- duced its commitment to public coverage. lands it is envisaged that consumers will be able to Many governments have taken important initiatives choose among quasi-public sickness funds and private to contain costs, through increased cost sharing or insurers, with a central health care fund taking income- through supply-side reforms. Belgium, France, Ger- related premiums and paying out risk-related pre- many, and the Netherlands have set fixed budgets for miums to the competing insurers. This amounts to a hospital expenditure. Virtually all the OECD countries sophisticated health voucher scheme. In the United greatly reduced the rate of growth of health expendi- Kingdom part of the hospital budget is given to large tures during the 1980s. The biggest exception was the (competing) general practices, which will enable gen- United States. Those countries that still relied to some eral practitioners to purchase certain hospital services extent on the reimbursement of patients for medical on behalf of their patients. Since the reforms were in- bills, with no connection between insurers and pro- troduced only recently in the United Kingdom and are viders, were less successful in containing costs han being carried out gradually in the Netherlands, it is not those in which insurers had direct contracts or in which yet possible to evaluate them fully. There are, however, a public agency was both public insurer and health already signs that general practitioner "Iundholders" provider. in the United Kingdom are using their new purchasing Probably the most important reforms of the 1980s power to negotiate a higher quality of hospital services involved the introduction of improved incentives and reg- for their patients. ulations for providers and insurers, with the aim of rais- 125 Table 5.6 Policies to improve delivery of health care Potential impact on / / o Provider and policy Public sector Protect nonsalary recurrent spending Complete the district health delivery infrastructure Retain fees at point of collection Decentralize financial resources and operational authority Subcontract ancillary services to private sector Improve drug selection, procurement, and use Nongovernmental organizations Legalize and simplify registration Provide government subsidies (per case, per diem, or block grants) for essential clinical services Subsidize training for district health workers Private (for-profit) sector Remove legal barriers to practice Promote health maintenance organizations Establish managed competition among suppliers Regulate private hospitals and physicians Provide public subsidies for essential clinical services and selected public health interventions Significant U Moderate 0 Modest 0 No significant impact both overextended and excessively concentrated care facilities to the private sector. At the same on discretionary care at the expense of essential time, governments have to improve the equity and services for the poor, the public system needs to be efficiency of their remaining health programs and scaled back. This means reducing public invest- facilities, through selective and progressive user ment in tertiary care facilities and specialist train- charges, decentralization, managerial incentives, ing and, in some cases, transferring discretionary and better information systems. The key policies 126 for improving delivery of clinical services by gov- operating expenditures for its thirteen predomi- ernment, NGO, and private for-profit providers nantly rural health regions. Countries can also are shown in Table 5.6. protect nonsalary spending by introducing more flexibility into their hiring arrangements. In India Delivering the essential package the Ministry of Health is planning to hire 8,000 In a competitive health system, people seeking workers for a leprosy control project on a per diem health services can choose from a diversity of pro- basis rather than engage them as civil servants viderspublic, private nonprofit, and private for- with virtual lifetime guarantees of employment. profit. As developing countries move toward such Another policy for ensuring adequate spending a competitive system, they face a wide range of on essential clinical services for the poor is to allow policy options that can improve the delivery of the primary care facilities to retain user charges and essential clinical package. spend these revenues on drug supplies and incen- tive bonuses for health workers. In Cameroon THE PUBLIC SECTOR. For many countries an im- when fees were introduced in a group of rural portant step in improving access to government- health centers and the revenues collected were provided essential clinical services is to complete used to replenish drug supplies, utilization of the the basic district health infrastructure by building centersespecially by low-income familiesin- health centers and health posts and training more creased substantially. The percentage of sick peo- nurses, midwives, and other providers of primary ple living near those centers who sought care rose care. Provision of good-quality housing for rural by more than 25 percent compared with the areas health workers as part of the district health infra- where fees were not charged, demonstrating that structure can improve efficiency by encouraging user fees can actually increase services for the staff to move to rural areas. Zimbabwe doubled poor. the number of its rural health centers, from 500 to more than 1,000, during the period 1980-90 with NONGOVERNMENTAL ORGANIZATIONS. Private the goal of making essential clinical services avail- nonprofit institutions, both local and foreign, pro- able within 8 kilometers of home for the whole vide a significant share of health services in devel- population. When housing for rural physicians oping countries, delivering essential clinical ser- and nurses was added to the district health pro- vices to low-income households in the poorest gram in the mid-1980s, the effect on staff recruit- countries. NGOs provide a third or more of all ment and retention was enormous: vacancy rates clinical carein Cameroon, Ghana, Malawi, for physicians and nurses in the areas with new Uganda, and Zambia. They own a quarter of the housing fell to zero, as against 20 to 30 percent health facilities in Bolivia's three largest cities, and elsewhere. they supply more than 10 percent of clinical ser- Government recurrent spending for primary vices in India and Indonesia. Although it is diffi- care inputs other than salaries is particularly vul- cult to compare the performances of NGO and nerable to budget cuts. When revenue shortfalls government health facilities, recent data from Af- occuror tertiary care hospitals overspendser- rica suggest that NGOs are often more efficient vices in peripheral facilities and communities often than the public sector. In Uganda physicians in suffer. In countries such as Chad, Haiti, Mozam- church mission hospitals treated an average of five bique, and Nepal there are widespread reports of times as many patients as physicians in govern- health centers that have staff but few or no drugs ment facilities, and NGO nurses attended more and of mobile health teams with little or no gas- than twice as many patients as their government oline for their vehicles. Under these circumstances counterparts. Governments that have excluded technical efficiency falls to near zerogovern- NGOs or heavily restricted their operations have ments continue to pay staff salaries, but virtually seen essential services deteriorate. When Mozam- no meaningful health services can be delivered. bique decided after independence in 1975 to ban Protecting the nonsalary part of the budget is NGO health activities in favor of government-run extremely difficult but critical. In Senegal the gov- facilities, a wide range of health services in rural ernment is currently committed to increasing its areas suddenly disappeared. Where such bans or budgetary allocation for drugs and medical sup- barriers to NGO activity exist, they should be plies by 10 percent a year in real terms during removed. 1992-96. Mauritania has set targets for annual gov- Beyond this, there are important opportunities ernment spending on drugs and other nonsalary for governments to form constructive partnerships 127 with NGOs to deliver essential clinical services. small-town families are prepared to pay for good One approach being followed in Sub-Saharan Af- care. The average outpatient visit costs $2 and a rica and in some states of India is to incorporate normal delivery $8. NGO health centers into the network of public f a- In Africa and Asia traditional medicine remains cilities by nominating appropriately located NGO an important part of the health care system, some- hospitals as district (first level of referral) hospi- times accounting for more than 10 percent of total tals. The NGOs are expected to provide a range of spending. The number of traditional healersin- public health and clinical services and to perform cluding herbalists, bonesetters, faith healers, and specific districtwide functions such as health plan- traditional birth attendantsis typically many ning, supervision of lower-level clinics and com- times larger than the number of medical physi- munity activities, and maintenance of emergency cians. The ratio of traditional to modern health transport. In return, the government pays some of practitioners has been estimated at nine to one in the NGOs' costs. Sri Lanka, seventeen to one in Indonesia, twenty- This kind of government-NGO collaboration five to one in Ghana, and twenty-eight to one in takes a variety of forms. In Lesotho nine of the Nigeria. Under these circumstances there may be country's eighteen health service areas (districts) opportunities for governments to improve the de- are headed by a church mission hospital that car- livery of essential health services by using tradi- ries out comprehensive health planning and man- tional practitioners. Successful examples include agement for its entire area. In Zimbabwe govern- the use of traditional healers to screen for malaria ment funds for rural health improvement are and to distribute antimalarial drugs in Thailand, to being used to expand mission ("designated dis- promote modern contraceptives in Kenya, and to trict") hospitals and to purchase ambulances for distribute condoms to reduce HIV and STD trans- the NGOs. Ministries of health pay the salaries of mission in Uganda and Zimbabwe. Traditional nursing staff in mission hospitals in Zaire and birth attendants have also been enlisted to im- most of the recurrent costs of NGO facilities in prove pregnancy outcomes in many countries, in- Botswana. Government donation of free vaccines cluding Bangladesh (Box 5.8). and contraceptives to NGO health providers has Making delivery of clinical services more efficient also become a common way to target public sub- sidies to specific health intervention programs. There are many routes by which developing coun- tries can improve the efficiency of clinical services. THE PRIVATE SECTOR. In recognition of the grow- Policies to increase the efficiency of government ing importance of the modern private sector in health services through decentralization and better developing countries' health systems, some gov- hospital management could have an especially ernments have begun to encourage private practi- large positive effect, as could policies for strength- tioners to deliver essential services. In India sev- ening government regulation of a more competi- eral states and nonprofit groups are working with tive private sector. the private sectorincluding the country's 1 mil- lion semiqualified urban and rural medical practi- DECENTRALIZATION. A policy that can improve tionersto improve the quality and effectiveness both efficiency and responsiveness to local needs of basic care. Some innovative approaches are be- is decentralization of the planning and manage- ing tried: private distribution outlets are being ment of government health services. In Africa stocked with condoms and oral rehydration solu- some central ministries of health have given pro- tion to make both more widely available, and the vincial and district offices responsibility for plan- Indian Rural Medical Association is trying to im- ning, day-to-day management of funds, person- prove the skills of rural private practitioners nel, training, maintenance, and other functions. through education on such subjects as family plan- Many problems have arisen: local governments or ning, immunization, and oral rehydration. In local offices of the central health ministry have not Mali, one of the world's poorest countries, the lo had the capacity to plan and manage health activ- cal medical school has started a program to train ities; devolution of responsibility has not always graduates to set up private practices in small been accompanied by allocation of the needed towns of 15,000 to 50,000 inhabitants. Although funds; and local officials have not necessarily been experience is recent, a number of physicians have accountable to their constituents. There have been already established successful private practices in some successes, however, that offer lessons for these towns, and they have demonstrated that other countries. 128 Box 5.8 Traditional medical practitioners and the delivery of essential health services Many simple health activities do not require extensive outreach workers, the volunteers improve case detec- professional training or major facilities and equipment. tion and save the government considerable expense. Health workers based in clinics or in their own commu- A WHO study found that traditional healer volun- nities play an important role in delivering these ser- teers in Thailand were more active in pursuing and vices. The millions of community-based traditional identifying malaria cases than other volunteers and health practitioners have enormous potential as public that they tended to remain in the program longer be- health workers and providers of essential clinical ser- cause their service enhanced their standing in the Com- vices if governments can give them the appropriate munity. Villagers indicated that they felt more confi- training, information, and incentives. dent about having someone they already knew as the Thus far, the experience with modern-traditional col- village traditional healer draw their blood and adminis- laboration has been mixed. A number of projects have ter treatment. failed because of poorly designed training and inade- In western Kenya the African Medical and Re- quate supervision, and many governments need to do search Foundation (AMREF) has trained male and fe- more to curb unnecessary and dangerous practices by male traditional health practitioners who live in remote traditional healers. But there have also been instructive villages to dispense drugs and some types of contra- successes. ceptives. Since the project began, the share of women An evaluation of workers participating in a volun- of reproductive age using modern contraception in six teer program for detection of malaria in northern Thai- pilot sites has risen from less than 10 percent to more land found that the performance of volunteers who than 25 percent. The Kenyan government has asked were traditional healers was superior to that of other AMREF to expand the project. volunteers. The program, which began in 1961, had by In Bangladesh a program to train and support mid- 1988 more than 40,000 malaria-control volunteers dis- wives to work with traditional birth attendants helped tributed across 34,000 Thai villages. Volunteers trained to lower maternal mortality rates by 60 percent over a by the Ministry of Health are expected to examine vil- ten-year period. The results of the program indicate lagers, take blood samples, prepare smears to be sent that, given adequate support systems, community- to the district malaria clinic for analysis, and treat fever. based services could bring about a substantial decline They also provide malaria-related education to the vil- in maternal mortality. lagers. In comparison with alternatives such as paid In Ghana, until recently, decisions on health sponsibility for primary care to local district coun- spending were highly centralized, with inflexible cils. The process began on a pilot basis in 1973 and expenditure levels set by the Ministry of Health for was gradually expanded to cover the entire coun- specific "vertical" programs such as immuniza- try. To support decentralization, the Ministry of tion, control of tuberculosis and leprosy, and fam- Health funded the creation of district health teams ily planning. In the late 1980s the ministry agreed consisting of a medical officer, a public health to delegate financial authority to health teams in nurse, and a health inspector. The central govern- each of the country's 110 districts. Unfortunately, ment continues to finance, through annual block most district health officials did not know the pro- grants, most of the recurrent primary care expen- cedures for obtaining and accounting for funds. To ditures of the district councils, and the councils' remedy this problem, members of the district proposals for capital spending are included in the health management teams were given training ministry's investment budget. But the day-to-day that enabled them to make more timely budgeting management of primary care centers, including and spending decisions, and expenditure levels in- purchases of supplies and hiring of personnel, is creased as much as fivefold. In districts where ear- in the hands of the councils. marked funds from the center were pooled and Those countries that have gone furthest in de- reallocated according to local priorities, technicil centralization have devolved responsibility for efficiency improved because of joint planning of health servicesincluding implementation of gov- work schedules and sharing of transport for out- ernment health programs and management of reach services and supervision. government health facilitiesto subriational levels In Botswana, although secondary and tertiary of government. Such devolution has been going health care is the responsibility of the central Min- on for many years in some large developing coun- istry of Health, the government has devolved re- tries with federal systems of government (for ex- 129 ample, Brazil, India, Mexico, and Nigeria), but it is ficient systems that conform to the expected pat- also becoming increasingly common in a number tern. But in others, such as Indonesia, Jamaica, of other Latin American countries and in the for- and Lesotho, there are no significant differences merly socialist economies. between the different types of hospitals. In Papua The gradual devolution of Chile's health system New Guinea the ideal pattern is reversed: the low over the past decade suggests that success can be turnover rate for district hospitals implies that the achieved through a measured process accom- hospitals either have too many beds or offer such panied by training and institutional development. poor services that patients go directly to higher Chile began the progressive decentralization of its levels. The shorter average stay at the tertiary hos- publicly provided health services in 1979, when pital suggests that many patients who do not re- twenty-six health service areas (HSAs) were estab- quire tertiary care are treated there. lished to cover the country's thirteen administra Disparities in average length of stay across hos- tive regions. Each HSA was given responsibility pitals with a similar mix of cases are another indi- (and, to go with it, additional personnel and a cation of substantial inefficiencies. Data from Latin share of the health budget) for managing all the America for 1980-85 show that average stays in government health facilities in its area. The second hospitals run by ministries of health varied from step in decentralization was taken in 1987, when five days in Colombia to thirteen days in Uruguay, management of the primary care network (consist- and stays in social insurance hospitals varied from ing of more than 2,500 urban and rural clinics, five days in Mexico to twelve days in Peru. In pub- rural health posts, and rural medical stations as lic hospitals in Argentina the average length of well as about 14,000 health personnel) was trans- stay ranged from eight to twenty-seven days. In ferred to the local government or municipalities. Malawi average stays in six government hospitals The HSAs were responsible for monitoring the with a similar case mix varied from five to thirteen municipalities' actions. In a third phase of the de- days. centralization process, scheduled to begin in late To remedy these inefficiencies, major steps to 1993, the central Ministry of Health will withdraw improve the organization and management of completely from service provision, leaving this public hospitals are required. One approach task entirely to the HSAs, which will enter into would be to delegate responsibility for health ser- formal performance contracts with the ministry. vice delivery to individual public institutions, but But not all decentralization has been a success. doing so will require changes in accounting and Countries such as Colombia that have devolved management practices. Government health bud- responsibility in a short period of time, without gets are now often highly aggregated, covering all the requisite financial resources and institutional facilities and programs in a given district, region, capacity at lower levels of government, have or even country. This prevents any detailed analy- found that decentralization can be counterproduc- sis of spending and services. Tracking costs should tive, aggravating existing inefficiencies and ineq- become part of the responsibility of the facility uities in the health system. In Brazil the munici- manager and the district management team. Cost pality of Rio de Janeiro refused to accept analysis would make it easy for these managers to responsibility for all the "decentralized" federal monitor areas of over- or underfunding and for health facilities within its borders on the grounds higher-level officials to compare cost profiles and that federal budget transfers to the municipality unit costs. were inadequate. Some countries have converted public hospitals into semiautonomous foundations or public enter- IMPROVEMENT OF HOSPITAL MANAGEMENT. Low prises in order to improve performance by grant- rates of hospital utilization in developing countries ing greater budgetary and management auton- point to significant inefficiencies in the use of omy. These foundations or parastatals are under buildings and equipment (to which scarce capital fewer restrictions than public facilities in manag- has been devoted). Most common medical condi- ing their budgets and sometimes in hiring and fir- tions can be treated in relatively simple facilities. ing. They can recover costs and collect charitable In a well-run hospital system district hospitals donations. The Tunisian government, for exam- would have the highest turnover rates and the ple, has converted eleven large public hospitals shortest stays, while tertiary hospitals would have into semiautonomous entities over the past two the lowest turnover rates and the longest stays. years and will convert another ten in the next few Some countries, including China and Fiji, have ef- years. Under the new arrangement, each hospital 130 manages its own operational budget and is free to when they are already sick or have a pressing reassign funds across budget categories as needed. health problemand because, in the case of third- To ensure accountability, the hospital is required to party reimbursement, they do not bear the full operate within its annual budget and to provide to cost of the services they consume. the Ministry of Health detailed reports about ser- Governments can encourage efficiency-pro- vices provided and unit costs. Autonomy in per- moting competition among suppliers of health ser- sonnel matters is more limited. All staff except the vices by requiring them to offer a standard pack- general manager are still governed by civil service age of services at a price fixed in advance. regulations. The hospital management cannot fire Consumers can then pick the supplier that offers employees, but it can ask the ministry to reabsorb the most attractive combination of price, service, staff and can use contracted personnel instead. and quality, with competition spurring suppliers The effect of these reforms will be fully assessed to improve quality and reduce costs. This is the over the next few years, but some gains in effi- basic approach taken in "managed competition" ciency are already apparent. One of the first hospi- proposals for health care reform in the United tals converted has fully contracted out all food, States (Box 5.9). Although managed competition cleaning, and security functions and now obtains requires a high degree of government administra- services of much higher quality, at similar or even tive capacity to set the rules and to monitor pro- lower unit cost. vider performance, it may be relevant in some Elsewhere too, an increasing body of evidence middle-income developing countries. suggests that the technical efficiency of govern- As developing countries take steps to encourage ment health facilities can be improved by contract- a diversified system of health service delivery, in- ing out ancillary services. In Venezuela, for exam- cluding use of NGOs and private providers, they ple, Health Ministry hospitals contract out will also need to strengthen their governments' maintenance of large equipment, and social secu- capacity to regulate the private sector. Regulations rity hospitals frequently contract for laundry, gar- are required to ensure that quality standards are dening, food services, and security, as well as for met, that financial fraud and other abuses do not maintenance. This arrangement has several ad- take place, that those entitled to care are not de- vantages. It can be less costly than publicly pro- nied services, and that confidentiality of medical vided services; services can be of better quality; information is respected. Regulation can be carried and they can be more reliable (because less subject out in a number of ways: by inspecting private to strikes and other industrial action). health facilities; by accrediting medical schools; by licensing physicians, nurses, and other health pro- COMPETITION AND REGULATION. Competition fessionals; and by prohibiting certain insurance among health providers in developing countries practices such as exclusion of prior medical condi- can improve the quality of services as perceived by tions. When the government allocates funds to patients and thus increase consumer satisfaction. NGOs, reimburses private providers under public This applies to the poor as well as the rich: compe- insurance schemes, or subcontracts with the pri- tition among private physicians in the slums of vate sector for ancillary services such as catering Bombay, for example, is intense, with private prac- and laundry, it can require an independent audit titioners offering convenient evening hours, short of these private contractors. waiting times, and readily available drugs to win In practice, few developing countries have es- patients from other private practitioners and from tablished such regulatory mechanisms, but the sit- public clinics. uation may be changing. In Brazil, where social It is much less clear, however, whether competi- insurance finances the bulk of health care, and pri- tion among suppliers of health services always vate hospitals and physicians provide 80 percent leads to greater efficiency. In fact, the contrary of hospital services and half of all outpatient care, sometimes happens, especially when competition important regulatory changes are now under dis- among private providers is combined with third- cussion. These include comprehensive accredita- party reimbursement of fees paid for services. Ex- tion of facilities by state governments, standard cessive tests, procedures, and drugs are supplied, licensing examinations for medical school gradu- and costs increase. This supplier-induced ineffi- ates, and the inclusion of representatives of citi- ciency takes place because most patients are un- zens' groups and consumer advocacy organiza- able to judge the value of specific services or to tions in medical ethics boards, which are currently compare prices among suppliersall the more so composed exclusively of physicians. The private 131 Box 5.9 "Managed competition" and health care reform in the United States 'Managed competition," which has attracted wide- zation would enable consumers to choose among com- spread interest in the United States, refers to a health peting packages in an informed and more price- services purchasing strategy designed to promote com- conscious way. One example of "managed competi- petition and to reward those health care providers with tion" is the California Public Employees Retirement the best performance in terms of cost, quality, and pa- System, which operates like the proposed HIPC, ar- tient satisfaction. The strategy is designed to address ranging health coverage and managing competition on the fundamental problems of the current health care behalf of almost I million state employees and retirees financing and delivery system in the United States. and their families. There, health care coverage is mainly employment- Analysts expect that over time competition would based and is far from universal: approximately 37 mil- force third_party insurers to drop out or move in the lion people under age 65 lacked insurance coverage in direction of managed care networks because these net- 1990. Costs are increasing rapidly; if current trends works can use financial incentives and management continue, spending is expected to grow from 12 to 18 tools to achieve efficient care. Indeed, insurers are al- percent of GNP by 2000. Under third-party insurance, ready shifting in this direction: nearly half of all health provider reimbursement methods often create financial insurers now offer some sort of plan involving man- incentives to provide more care. Insurers seek profits aged care. Competition would also give health care by excluding higher-risk individuals rather than by ag- providers clear incentives to become more efficient. gressively pursuing greater efficiency in providing One question about the managed competition model is health services. A major indirect cost of the system is whether it would work in rural areas and areas of low reduced labor mobility as a result of the risks of exclu- population density, and, if not, what alternatives sion from insurance at a new place of employment. would be best. Another concern is the cost of extend- Under managed competition, a health insurance ing universal coverage. purchasing cooperative (HIPC) would be formed to or- Although this proposal was developed to respond to ganize purchasers of health care within a region. The the particular problems of the U.S. system, it has rele- HIPC would establish standards for the region's health vance elsewhere. The Netherlands, which is introduc- plan by, for example, defining a basic benefit package ing choice of insurer under a universal social health of comprehensive health services, and would contract insurance scheme, faces the same challenges regarding with eligible providers for this basic package. During control of risk-selection behavior on the part of in- the annual open-enrollment period the HIPC would surers. Chile has, since 1981, encouraged the growth of provide information about the price of the basic pack- private prepaid health insurance schemes, known as age from different providers and about the quality of ISAPREs. (Box 7.2 provides details on these plans.) care offered. Equity would be improved by requiring Problems have come up with Chile's reforms, how- providers to open their rolls to all consumers, regard- ever: the lack of a standard package has limited effec- less of risk. Universal coverage would be achieved tive competition, and weak regulation has allowed pri- through public subsidies to those not otherwise cov- vate insurers to deny coverage to high-risk individuals. ered so that they could purchase packages. Standardi- Brazilian Association of Hospitals is debating the Reorienting clinical services and beyond establishment of its own accreditation system, and the medical association in São Paulo has begun a It is a fundamental responsibility of governments voluntary pilot effort to certify hospitals in that everywhere to ensure access to a package of essen- state on the basis of adherence to norms, assess- tial clinical health care, with special attention to ment of patient data, and patient satisfaction sur- the poor. Utilization of a minimum package de- veys. Qualifying examinations for licensing physi- fined by its high cost-effectiveness would reduce cians are being tested in another state where, at the total burden of illness in developing countries present, all medical school graduates are automaf- dramatically, by an average of 25 percent. Such a ically licensed for life, without requirements for package is affordable in low-income countries if continuing education or recertification. And the governments reallocate current health expendi- Federal Council of Medicine has proposed legisla- tures and increase public spending and if they im- tion that would expand its authority to monitor the plement policies that encourage selective pay- quality of health care and to discipline poorly per- ments directly from better-off patients and from forming doctors and hospitals. existing insurance schemes. Middle-income coun- 132 tries could easily pay for the minimum package, providers. Key government measures include using the resources currently devoted to health, strengthening legal and financial support to and might wish to enrich the package by adding NGOs that provide health services and creating a services. positive environment for the private sector com- Government efforts to improve health insurance bined with important regulatory safeguards should be aimed at increasing the portion of the against abuse. Gains in technical efficiency can be population covered, reducing subsidies for insur- achieved through a combination of careful decen- ance that benefit primarily the wealthy and the tralization of government health services and im- middle class, and controlling health care spending proved management of public hospitals. financed from insurance. This will require stronger These efforts to reorient government recurrent regulation of private insurance and policies to ex- spending toward essential clinical care and to pro- pand compulsory social insurance based on pay- mqte diversity and competition in the supply of roll taxes. It also means shifting provider payment health services must be accompanied by changes away from open-ended fee-for-service methods to in longer-term investments in health inputsfacil- prepayment through capitation and preset bud- ities and equipment, health personnel, phar- gets for hospitals. maceutical management systems, health informa- Greater diversity and competition in the supply tion, and health research infrastructure. Policies to of health services can do much to improve the de- bring about this reorientation are taken up in livery of an essential clinical package and raise the Chapter 6. technical efficiency of doctors, hospitals, and other 133 Health inputs In recent decades developing countries have in- spending subsidizes high-end facilities, equip- vested heavily in health. Often with help from do- ment, and human resources for private markets. nors, they have constructed hospitals and build- The challenge for public policy is to redress the ings and purchased equipment to fill them. They balance and so permit the efficient delivery of pub- have educated doctors, nurses, and other health lic health and essential clinical services. Where care professionals. And they have set up new sys- cost containment of health spending is a concern, tems to supply drugs, research, and information. public policy can play a useful role in limiting the Worldwide, the number of hospital beds rose be- growth of both public and private investments in tween 1960 and 1980 from 5 million to almost 17 specialist training, equipment, and tertiary facili- million, which more than doubled the per capita ties. For some inputs, such as buildings and hu- supply. The number of physicians increased more man resources, changes will necessarily be slow. than fivefold between 1955 and 1990, from 1.2 mil- For others, such as pharmaceuticals, a new policy lion to 6.2 million. Such investments have created can alter inputs rapidly. This chapter suggests how new opportunities, but they have also led to to set about these tasks. It also assesses how public problems. support for information and research can help Once built, hospitals are extremely difficult to improve health sector performance today and close. Once trained, physicians create pressure to create new health systems and technologies for be employed. In virtually every developing coun- tomorrow. try, facilities, equipment, human resources, and drugs are skewed toward the top of the health Reallocating investments in facilities system pyramid (Figure 6.1). Yet the cost-effective and equipment public health and clinical interventions discussed in preceding chapters of this Report are best deliv- Investments to support delivery of essential clini- ered at the level of the district hospital or below. cal services are best directed at health centers and That they are often delivered through tertiary hos- district hospitals and at improving access in under- pitals simply increases costs without improving served areas. Some public investments in tertiary quality. This problem is found in poor countries in facilities are needed to support research and train- which the principal tertiary teaching hospital in ing, but at levels well below current levels of pub- the capital city consumes a large proportion of the lic financing in most countries. Investments in spe- total resources available for health. It is also found cialized facilities can be left largely to the private in cities such as London, where numerous spe- sector, and public subsidies, where they exist, can cialized teaching hospitals absorb large amounts often be greatly reduced. Redirecting public of resources while failing to address the most spending toward lower-level facilities is difficult common and pressing health problems of city politically, but some countries are moving in this residents. direction. In Papua New Guinea, for example, In many countries public investments are con- public spending on hospitals has for the past dec- centrated unduly on tertiary services, and public ade been limited to 40 percent of the Ministry of 134 Most health care should.take place toward the bottom whole cityan example is Zambia's University of this pyramid. Teaching Hospital of Lusaka, with 1,835 beds. Hospitals absorb the bulk-40 to 80 percentof public spending on health in developing coun- Figure 6.1 The health system pyramid: tries. Industrial countries have much higher health where care is provided expenditures and more chronic disease problems, but the share allocated to hospitals is slightly smaller, 35 to 70 percent. Figure 6.2 shows the marked variations in hospital supply across the hospitals eight demographic regions used in this Report, ASpecialized from about eleven beds per 1,000 population in Central and Eastern Europe to less than one bed District A District per 1,000 population in India. In most developing health countries more than 60 percent of all hospital beds hospitals system are public. The data used in Figure 6.2 unfor- Health centers tunately fail to distinguish between tertiary and and clinics district-level hospitals. The minimum package of essential clinical services described in Chapter 5 requires about one district hospital bed per 1,000 Households population. Given that some of India's and Sub- Saharan Africa's hospital beds are devoted to care outside the essential package, there is likely to be a shortage of district-level hospital beds in parts of those regions. In some countries the underfunding of lower- Health's recurrent budget, which is well below the level facilities has been exacerbated by the creation average in most developing countries. of multiple levels of outpatient facilities (health posts, dispensaries, and rural health centers), Facilities none of which functions well. At the same time, tertiary care hospitals are crowded with patients Tertiary hospitals provide the most specialized and who could be treated in less costly and more acces- sophisticated services and are where most clinical sible district hospitals or health centers. A study in research, education, and training take place. They Chad, for example, revealed that 71 percent of all are usually located in large urban areas. One step central hospital consultations were for problems down the health hierarchy are district hospitals, that could have been treated at lower-level facili- which are typically located in towns or smaller ties. An obvious way to reduce spending without cities serving rural areas but are valuable in large sacrificing any health gains is to make full use of cities too. District hospitals generally have 100 to existing lower-level facilities. Measures for achiev- 400 beds, serve 50,000 to 200,000 inhabitants, and ing this include charging a higher fee to patients include departments of medicine, surgery, pedi- who go straight to tertiary facilities without refer- atrics, obstetrics and gynecology, and dentistry. rals, except in emergencies, and making a referral They also provide basic anesthesia, radiology, and from the primary care provider a mandatory con- clinical laboratory services. The district hospital is dition for specialized services. At the same time, the first level of referral from health centers and however, the quality and responsiveness of ser- provides complementary services such as basic vices at lower-level facilities need to be improved. surgery. It mainly offers inpatient care but also typically provides some outpatient care, day sur- INCENTIVES AND INVESTMENT DECISIONS. Public gery (in which the patient is operated on and dis- sector budgetary procedures often obscure the real charged on the same day), and emergency services costs of investments in health facilities and bias not available at health centers. In many cities that them toward high-profile investments in large have grown rapidly over the past twenty years, hospitals. Major investments, including donor- periurban areas do not have enough health centers financed projects, may be approved by a govern- and district hospitals. In some African capitals one ment body that does not have to face the recurrent extremely large tertiary public hospital serves the costs of operating the facility. Regions can argue 135 Installed hospital ca pacify is lowest in India, Sub-Saha ran Africa, and Asia and highest in the formerly socialist economies of Europe. Figure 6.2 Hospital capacity by demographic region, about 1990 Beds per 1,000 population 0 2 4 6 8 10 12 Sub-Saharan Africa India China Other Asia and islands Latin America and the Caribbean Middle Eastern crescent Formerly socialist economies of Europe Established market economies Source: Organization for Economic Cooperation and Development data; World Health Organization data. for new facilities without having to weigh the bud- zations (HMOs) in the United States have incen- getary consequences if federal-state resource tives for providing care efficiently. As a result, transfers, instead of following predictable and they operate with much less hospital infrastruc- transparent funding formulas (such as population- ture than the health system overall. Large HMOs based schemes), are heavily politicized. (with hospitals) have about 1.5 hospital beds per International assistance has frequently exacer- 1,000 memberswell below the average of 8 beds bated the problem of unsustainable health invest- per 1,000 for established market economies and 5 ments. Donor assistance, particularly for tertiary for the United States overall and slightly below facilities and teaching hospitals, has sometimes regional averages for China and Latin America. been provided even if the incremental recurrent Evidence indicates that quality of care in HMOs is costs from these investments are too high. Too maintained even with significantly lower levels of many donor-financed hospitals have opened only hospitalization and hospital infrastructure. Pri- partially, not at all, or at the expense of existing vately financed health insurance with uncon- facilities. In Rwanda, for example, a 200-bed hos- strained fee-for-service payment, by contrast, pro- pital was completed in 1991 but has not yet opened vides no incentives to rationalize physical capacity. because of the difficulty of financing its high recur- rent costs, which are estimated at about 15 percent DUPLICATION OF HEALTH FACILITIES IN THE PUBLIC of the Ministry of Health's already tightly con- SECTOR. Excess public facilities in urban areas are a strained budget. In Chad, where external assis- problem in many countries. The historical growth tance amounts to about 30 percent of national in- of hospitals, especially in capital cities, has led to a come, a national development plan proved useful proliferation of specialist tertiary services that are for screening out inappropriate donor financing. often linked to medical education and research. Two new hospital construction proposals that Another cause of duplication is the public provi- were found to conflict with the plan are being sion of health services to different subgroups of reconsidered. the population, each with its own hospitals and In the private sector, financial incentives drive health centers. In Poland, for example, parallel investment decisions. Health maintenance organi- health systems exist for workers and their families 136 in the railway, mining, police, and military sectors PRODUCTIVITY. The potential for improving the and for prisoners; another set of facilities serves productivity of installed hospital capacity is large. the general population. This can result in excess In addition to the financing and management re- capacity, with no facilities achieving economies of forms discussed in Chapter 5, efficiency gains can scale. be achieved by taking the following measures, There are two solutions for duplication: creating which will need to be supported by investments in internal markets and instituting central or regional training and infrastructure: planning linked to health budgets. If effective in- Convert some acute care hospital capacity to ternal markets are created within the public sector, less costly extended or chronic care facilities for money will follow patientsand patients, together patients who require less-intensive care for long- with their general practitioners, will have a choice term recovery and for rehabilitation of chronic as to which hospital to use. The availability of conditions. Extended care facilities operate at a good information about quality and price will help lower cost per bed-day than acute care hospitals. efficient providers of specialist services to prosper, In the absence of such lower-level facilities, pa- while less-efficient hospitals will close. The alter- tients occupy high-cost acute care beds. native is rationalization of services by central or Perform outpatient diagnostic tests before ad- regional planning. In the largely publicly financed mitting the patient to the hospital. health systems of the Nordic countries, health re- Support home care as an alternative to long- sources are allocated by region. Each region of term hospitalization for some ailments. about 350,000 inhabitants elects representatives Modify treatment protocolsfor example, re- who make decisions about health care spending. duce unnecessary surgeries, perform low-risk de- These representatives have incentives to avoid du- liveries at maternity centers, and treat tuberculosis plication of services and to capture economies of patients and many surgical cases on an outpatient scale in service delivery. If left to individual hospi- basis. In Cali, Colombia, costs per procedure for tals, decisionmaking for large investments will day surgery are less than 30 percent of the cost tend to reflect the interests of that hospital, not the of traditional treatment in hospital. Outpatient region. Multiple hospitals will want to provide surgery has grown rapidly in many industrial specialized, "prestige" services, leading to countries but is used much less in developing overinvestment. countries. EQuITY CONSIDERATIONS. For the rural poor, lack MAKING THE TRANSITION. The 1985 earthquake of physical infrastructure is the largest obstacle to in Mexico City destroyed about 20 percent of pub- use of health services. Distance to health facilities lic hospital capacity. The Ministry of Health chose limits people's willingness and ability to seek care, to concentrate reconstruction and new construc- particularly when transport is limited. There is a tion in low-income periurban areas that had hith- heavy urban bias in the distribution of health facili- erto been poorly served, and six new 144-bed dis- ties. Large cities are much better served by both trict hospitals were built in these areas. But such public and private health infrastructure than possibilities for rapidly reconfiguring capacity to- would be expected from their roles of serving ur- ward lower-level facilities and underserved areas ban populations and providing referral services for are seldom available. The alternative is to reduce the surrounding population. (Referral hospitals or refrain from public investment in tertiary hospi- are needed for only a small proportionno more tals while simultaneously increasing investment than 10 percentof total hospitalizations.) Wealth- and operating budgets for health centers and dis- ier regions also have better access to infrastruc- trict hospitals. Over time, the tertiary hospitals can ture. In India the richer states of Maharashtra and be operated on a self-financing basis, or they can Gujarat have 1.5 and 1.1 beds, respectively, per be closed, converted to chronic care facilities or 1,000 population; the poorer states of Bihar and district hospitals if these are needed, or even sold Madhya Pradesh have only 0.3 and 0.4 bed, re- to the private sector. But in most countries this spectively, per 1,000 population. Public invest- process will necessarily be slow. ments need to address inequities in the present distribution of health infrastructure. Donors have Equipment an important role in this regard, especially where a significant proportion of investment is donor Developing countries account for about $5 billion, financed. or 7 percent, of the $71 billion spent each year on 137 medical equipment worldwide. This global esti- proving maintenance to increase operating life and mate includes medical and dental supplies, surgi- reduce downtime of equipment is more efficient cal instruments, electromedical and X-ray equip- than buying new equipment. ment, diagnostic tools, and implanted products. Because of the many products on the market The ability of the medical equipment industry to and the speed of change, carrying out technology develop new health care technologies has vastly assessments can be extremely costly. The interna- exceeded the capacity of purchasers to evaluate tional community could help by developing and the clinical value and the cost-effectiveness of such disseminating information on the availability, ef- innovations. At present, approximately 6,000 dis- fectiveness, and prices of equipment and on user tinct types of medical devices (equipment, sup- guidelines. Essential equipment lists could be de- plies, and reagents) and more than 750,000 veloped along the lines of the essential drug lists brands, models, and sizes, produced by perhaps already used by many countries. 12,000 manufacturers worldwide, are on the Equipment procurement would also benefit market. from greater use of competitive buying. Purchas- Efficiency losses from poor selection and main- ing is commonly restricted to local distributor- tenance of medical equipment can be very large. ships, and some countries also heavily protect lo- WHO estimates that less than half of all medical cal industry. These policies reduce competition equipment in developing countries is usable. In and can easily double the purchase price of equip- Brazil an estimated 20 to 40 percent of the $2 billion ment. Developing countries can cut costs by to $3 billion worth of public sector medical equip- adopting competitive purchasing methods or by ment is not working. A study of twelve Kenyan purchasing equipment from international agen- hospitals in 1984 found that sterilizers operated for ciessuch as UNICEF, Equipment for Charity an average of two years instead of the six expected Hospitals Overseas (ECHO), and the International and that incubators lasted only two years rather Dispensary Associationthat offer procurement than eight. Equipment failed prematurely because services for some medical equipment at competi- maintenance budgets were only about 1 percent of tive prices. the value of the capital stock (10 percent might be There are several reasons for government in- considered optimal). In Viet Nam 39 percent of volvement in the development of health infra- urban health centers and 29 percent of urban poly- structure. The government itself, as a provider of clinics surveyed in 1991 lacked a working ster- health services, may finance and use infrastruc- ilizera critical piece of equipment for developing ture. It may also intervene to compensate for mar- countries that have to reuse such supplies as ket failures that can lead to greater investment, syringes. particularly in specialized health inputs, than is Investments in medical equipment can be ratio- socially optimal. Finally, the government has a role nalized by controlling the purchase of expensive, in undertaking technology assessment of medical sophisticated equipment and rejecting most do- equipment, which is a costly public good. nated medical equipment, new or used. To contain To reduce both capital and recurrent costs with- costs, Belgium, France, and Portugal directly con- out sacrificing quality of care, governments can: trol the acquisition of state-of-the-art medical tech- Reallocate public spending toward the facili- nologies by both the public and the private sectors. ties and equipment required for providing public In Canada major capital acquisitions require prior health programs and essential clinical services. approval by the provincial or territorial ministry of Improve the efficiency of installed capacity by health on the basis of a needs assessment and considering alternative uses of facilities, as well as other factors. Alternatively, governments can en- new diagnostic and treatment protocols. (Exam- courage public hospitals to make tough choices by ples are the conversion of some costly acute care limiting their budgets. Even assuming that do- capacity to less costly extended care beds and nated equipment meets local equipment require-. treatment of some surgeries on an outpatient ments, very little of it ever becomes operational, basis.) Such reconfiguration may require modest for a variety of reasons, including missing or dam- new investment. aged parts, lack of disposable inputs and of user When cost containment is a concern, consider and service manuals, and problems with power controls on the purchase of expensive, specialized supply. Standardization of equipment could sim- technologies, whether by public or by private plify management and maintenance and reduce providers. inventory costs. Purchasing decisions could be an- Support and disseminate technology assess- alyzed on a life-cycle cost basis. In many cases im- ments to purchasers. 138 Reduce or eliminate subsidies to private in- too, has long relied on graduates of three-year (in- vestors in facilities and equipment. stead of five-year) medical schools to meet the needs of rural areas. Addressing imbalances in human In some countries tasks traditionally performed resources by physicians have been successfully delegated to lower-level primary care providers as a way of im- Nearly all countries face the same fundamental proving the efficiency of health services. By spe- problems with human resources in the health sec- cializing in certain common procedures (as mid- tor. There are not enough primary care providers wives specialize in deliveries, for example), such and too many specialists. Health workers are con- providers may become better at their tasks than a centrated in urban areas. Training in public health, generalist physician. Surgical technicians in health policy, and health management has been Mozambique perform hysterectomies and ce- relatively neglected. Medical training is subsidized sarean sections and remove ectopic pregnancies. even though physicians may earn high incomes Some nongovernmental organizations (NGOs) in and many work in the private sector. Bangladesh use graduate nurses to do steriliza- There are several ways in which governments tions, and in Thailand public sector nurse-mid- can do something about these problems. Public wives perform this procedure. In these cases, eval- sector pay and employment policies can be im- uations indicate no differences in outcomes proved to be more competitive with the private compared with procedures done by physicians. sector and to relate pay to performance. Career Ophthalmic clinical officers, who are not physi- development paths and in-service training are cians, have performed cataract surgery in Kenya needed to retain staff, especially in managerial po- on a pilot basis, and evaluations indicate accept- sitions. Policies on accreditation and licensing can able results. Africa has only one ophthalmologist be used to limit enrollments in training programs, per 1 million people; without the use of nonphysi- to shape curricula (all physicians might spend time cian services, many patients would not be able to in rural practice during their medical training or be get cataract surgery. required to pass examinations in public health), The distribution of nurses and physicians by re- and to set minimum standards for providers. Edu- gion is shown in Figure 6.3. Appropriate staffing cation finance policies can be used to curtail educa- ratios depend heavily on the organization and f i- tion opportunities for physicians and specialists nancing of care and the specific tasks health per- and to expand them for workers in primary care, sonnel carry out. Health maintenance organiza- public health, health policy, and management. But tions in the United States, for example, operate where oversupply is greatest, as for specialist phy- with about 1.2 physicians per 1,000 enrollees, com- sicians, the only effective solution may be to set pared with about 4.5 in the fee-for-service sector. quotas for training, or at the very least for publicly Evaluations of health outcomes and user satisfac- subsidized training. tion indicate that these savings in resources do not come at the expense of quality. Sub-Saharan Africa Improving the balance between primary care providers has the fewest physicians and nurses of any re- and specialists gion, which is an obstacle to the delivery of the public health interventions and essential clinical A central role in delivery of most cost-effective services described in Chapters 4 and 5 because health interventions belongs to primary care pro- some of the existing personnel are providing other viders, a category that can include physicians, services. The public health and minimum essential nurses, nurse practitioners, or midwives, depend- clinical interventions require about 0.1 physician ing upon how the jobs are defined. Nonphysician per 1,000 population and between 2 and 4 gradu- primary care providers have many advantages. ate nurses per physician. Given resource con- They cost less to train (data from Myanmar, straints, however, the relatively high ratio of Pakistan, and Sri Lanka indicate that between 2.5 nurses to physicians in Sub-Saharan Africa is a and 3 nurses can be trained for the cost of training good sign. There is no optimal level of physicians one physician), and they receive lower salaries. per capita or optimal nurse-to-physician ratio, but They are easier to attract to rural areas and usually a rule of thumb is that nurses should exceed physi- communicate more effectively with their patients. cians by at least two to one. (The ratio is five to one In Sub-Saharan Africa, where the few local physi- in Africa but well under two to one in China, cians are concentrated in urban hospitals, nurses India, Latin America, and the Middle Eastern often function as primary care providers. China, crescent.) 139 The availability and mix of health personnel vary widely across regions. Figure 6.3 Supply of health personnel by demographic region, 1990 or most recent available year Ratio of nurses and midwives Physicians per 1,000 population to physicians 5 6 D Sub-Saharan Africa fl Other Asia and islands D India O Middle Eastern crescent O Latin America and the Caribbean o China I 0 Established market economies Formerly socialist economies Source: See Appendix table A.8. PHYSICIAN OVERSUPPLY. During the 1960s and nonmedical jobs, and 11 percent were in low- 1970s many governments encouraged, primarily income medical jobs or were seeing very few pa- through subsidies to education, rapid expansion in tientswhich is a concern because the physicians physician training to meet the need for primary may not see enough patients to maintain their care providers. In many countries the excess of competence. The quality of medical education also physicians in relation to nurses and of specialists declined with the rapid growth in medical schools. in relation to other physicians has created prob- Furthermore, the expansion of medical training lems. By the early 1980s the established market did nothing to solve the problem of attracting phy- economies, Latin America, and parts of Asia were sicians to rural areas. In 1983 an interinstitutional having trouble absorbing growing numbers of body was created, with representatives from the physicians. These policies have been costly, and it ministries of health and education, health care in- will take many years to correct the imbalances. stitutions, and universities. This group has, by Mexico illustrates the problems. Medical enroll- agreement, reduced enrollments and contained ments in 1970 stood at about 29,000 in twenty- the number of medical schools. More recently, the seven schools. Within ten years there were 93,000 government has begun publishing average exam- in fifty-six schools. Many of the schools offered ination scores of medical school graduates by highly subsidized or free tuition, and some of the school to provide information on educational qual- largest had open enrollment policies. At the same ity for prospective students and employers. time, health services were growing only modestly. Other countries responded to physician over- In 1960 there were 20,600 physicians in Mexico; by supply by restricting medical immigration (Canada 1990 there were 166,000. A survey of physicians in and the United Kingdom), by reducing working major cities in 1986 revealed that 7 percent were hours (Denmark), and by indirectly promoting unemployed, another 11 percent were working in outmigration of medical personnel. The last two 140 Box 6.1 International migration and the global market for health professionals Over the past several decades, large numbers of physi- trained human resources can cause shortages of health cians and nurses have migrated across national bor- workers. In Jamaica vacancy rates of more than 50 per- ders. WHO estimates that 14,000 nurses did so in the cent in nursing positions, in large part because of mas- early 1970s and that in 1972 more than 140,000 physi- sive migration of nurses, have forced the Ministry of cians (or 6 percent of the total) resided outside the Health to close whole wards and to reduce the services countries in which they were born or had been trained. offered in many facilities. In addition, emigrating Over the past half century the main flow of physicians health workers deprive their own countries of the and nurses has been from developing to industrial benefits of (often state-financed) investments in their countries. Developing countries donate a full 56 per- education. For example, the 111 registered nurses who cent of all migrating physicians and receive less than 11 resigned from government service in Jamaica in 1990 percent. The principal donating countries for physi- took with them nearly $1.7 million in government in- cians are India and the Philippines. More than 90 per- vestment in training and education. cent of the nurses who migrate go to North America, Europe, and the high-income countries of the western Policy responses Pacific, while only about 7 percent go to developing countries. Migrating nurses come overwhelmingly In an attempt to alter the patterns of migration, many from the Philippines, which exports each year 2,000 to countries have changed their immigration and licens- 3,000 nurses, many of whom go to North America. In ing laws and regulations. During the 1980s, for exam- 1970 more Filipino nurses were registered in Canada ple, the United States, to address its own nursing and the United States than in the Philippines, and the shortage, changed its policy on immigration of nurses, trend has continued to the present. Other major coun- making it relatively easy for nurses wishing to come to tries of origin for migrating nurses are Australia, Can- the United States to obtain a visa. This had a profound ada, the United Kingdom, and certain West Indian effect on a number of neighboring countries. In the countries. Philippines during the same period it became increas- ingly difficult for registered nurses to obtain travel doc- Consequences uments because of the enormous outflow of nurses from that country. The migration of health professionals has both positive Short-term immigration restrictions, however, may and negative effects. It can help alleviate shortages in have only a limited effect. Other possibilities for en- the receiving countries, and large remittances or tax couraging health professionals to remain in their home revenues from overseas workers can improve the stan- countries include reforming education finance to re- dard of living in the countries of origin. (The Philip- quire that individuals repay some or all of the costs of pines received an estimated $680 million from expatri- state-financed training, through student loans or en- ate workers in all fields in 1986, and an estimated $8 forced service bond requirements. And publicly fi- billion in remittances went to developing countries as a nanced opportunities for overseas training could be re- group in 1975.) On the negative side, the net outflow of stricted because of its tendency to lead to outmigration. L solutions waste valuable resources (see Box 6.1). In cedures. This, in turn, pushes up health care costs some countries the government can limit enroll- and reduces the quality of care. The United States ments in medical schools directly. Egypt has re- has the highest number of cardiologists and car- duced medical enrollments by half since 1982; in diac operating suites per capita in the world and France the Ministry of Health used quotas to cut correspondingly higher rates of surgerya signifi- new enrollments from about 11,000 in 1975 to less cant proportion of which is inappropriate. than 5,000 in 1989. In other countries, such as Ger- While practice guidelines and incentive struc- many and Mexico, universities have autonomy in tures can be important policy tools for curbing determining enrollments, and cuts must be made overuse of procedures, training fewer cardiologists through consensus or through education finance would also help. A classic U.S. study showed that policies. a 10 percent increase in surgeons would bring about a 3 to 4 percent increase in surgical opera- CURTAILING SPECIALIST TRAINING. Most govern- tionsthe phenomenon of "supplier-induced ments will need to limit not only total enrollments demand." but also the training of medical specialists. High Even when specialists function as generalists, numbers of specialists tend to increase the fre- they have more costly styles of practice, ordering quency of unnecessary and often risky pro- more tests and procedures. Analysis of geographic 141 variation in expenditures in the United States indi- slow. The fifty-five member institutions of the in- cates that expenditures on physician services are ternational Network of Community-Oriented Edu- unrelated to the total number of physicians per cational Institutions for Health Sciences have capita but are related to the ratio of primary care adopted curriculum reforms that emphasize com- physicians to specialists. Many OECD countries munity-based and problem-based learning. The limit the number of specialist training oppor- goal is to produce graduates whose competencies tunities. This policy instrument is increasingly and experience correspond closely to community relevant for middle-income countries interested in health needs. Significantly larger proportions of cost containment. Subsidized medical education graduates from these schools have followed ca- has already led to overproduction in some middle- reers in primary care, Similar reform efforts have income developing countries such as Chile, where taken place in nursing education. In Nigeria, Sene- 75 percent of all physicians are specialists, and gal, and Uganda (for basic nurse training) and in Venezuela, where about 55 percent of all physi- Thailand (for public health nurse practitioners) the cians employed in the public sector are specialists nursing curriculum has been oriented more to- In contrast, only 25 to 50 percent of physicians in ward community settings and preventive services. Belgium, France, Germany, and the Scandinavian countries are specialists, and regulatory bodies Attracting primary care providers to underserved areas and committees determine the number to be trained. Health providers are concentrated in urban areas. Few, if any, specialists are needed to deliver the Professional isolation, lack of additional work op- cost-effective clinical interventions discussed in portunities, substandard housing, and other dis- Chapter 5, even with a modest expansion in con- amenities often make staffing rural health facilities tent beyond the minimum essential package. difficult. If public sector wages cannot be in- Some specialists are required for services outside creased, other methods must be found to increase the essential package. The overall proportion of the attractiveness of rural posts. Many countries physician generalists to specialists is an important require a period of rural service following publicly indicator for governments to monitor, but this in- financed medical training. Canadian provinces formation is not at present readily available in have used many incentives, including differential many countries. A reasonable benchmark for the pay scales, settlement allowances, payment of ex- maximum proportion of specialists to physicians penses for continuing education, and provision of in developing countries might be 25 percent, scholarships for later study in return for a certain which is about the lowest proportion found in the number of years of service. established market economies. In many develop- In some settings lack of female health providers ing countries the proportion could be much lower, is an obstacle to utilization of health services. In given the epidemiological characteristics of the Egypt, for example, most physicians are male, but population and the smaller share of the population cultural beliefs constrain women from being seen using clinical services beyond the essential pack- after puberty by men who are not family mem- age. Public regulation and rationing of specialist bers. Even when trained, female primary care pro- training, in addition to the elimination of training viders are hard to attract to underserved areas be- subsidies, may be needed to achieve this. cause of security concerns and the importance of living with their families. The Aga Khan Develop- CONTENT OF TRAINING. Primary care training ment Network in Pakistan, recognizing this prob- should include, at a minimum, the skills necessary lem, has trained women to work in their own com- to provide the essential clinical services discussed munities as lady health visitors. in Chapter 5. In fact, however, basic curricula in Community health workers can complement the medical schools often fail adequately to cover work of primary care providers in rural areas. some of these services, such as family planning Burkina Faso, the Gambia, Ghana, and other services and the proper diagnosis and treatment of countries have trained large numbers of commu- sexually transmitted diseases (STDs). For more nity health workers as part of the national strategy than two decades there have been calls to stop for primary health care; in many other settings training health professionals in high-technology much smaller programs have been set up by NGO tertiary institutions and to expose them thor- groups. Evaluations sometimes show disappoint- oughly to health problems and practice at the ing results: community health workers have often grass-roots level. But progress has been extremely had little impact on health service utilization and 142 Box 6.2 Community health workers Over the past twenty years many countries have exper- to the health system, but their availability to the com- imented with the use of community health workers munity diminished. The program has since been (CHWs) to provide primary health care. Several Afri- greatly reduced. can countries introduced CHW programs in the 1970s Other efforts have been more successful. Perhaps as a way of extending primary health care services at the largest scale NGO-run community health worker low cost nationwide. Health workers' responsibilities program is the Pastoral da Crianca, operated by the typically include providing education on sanitation, Catholic Church in Brazil. This program, initiated in nutrition, family planning, child health, and immuni- 1983, receives strong support from the Ministry of zations, in addition to carrying out some basic health Health and some technical and financial support from interventions. They can also be valuable as a referral UNICEF and from the Bernard Van Leer Foundation point between health centers and the community. Re- and other NGOs. It now has 47,000 CHWs throughout grettably, CHW programs have had mixed results. Brazil. An estimated 1.5 million children were enrolled Studies have shown that in the Gambia and Indonesia in the program in 1992. CHWs provide health educa- traditional birth attendants who were not backed up by tion to low-income mothers regarding the importance skilled services were unable to decrease the risk of ma- of prenatal care, good diet during pregnancy, breast- ternal mortality. feeding, proper weaning, immunizations, and man- A Jamaican program, launched in 1977, that used agement of diarrhea, and they monitor the growth of CHWs in primary health care efforts is an example of a infants and young children. The training process for well-intentioned effort gone awry. Problems emerged CHWs follows a central guideline but is adapted to fit from the beginning, with the selection of personnel. the characteristics of different regions. Special care is CHWs generally demonstrate greater dedication when given to the training programs for illiterate volunteers, they serve the communities in which they live. Unfor- and supervision of CHWs is closely integrated with tunately, too few CHWs were recruited from the target continuing education and motivational support. An communities, and workers who lived elsewhere had to evaluation carried out in 1990 found that health and be enlisted. Inability to recruit male volunteers limited nutritional indicators for young children enrolled in the the success of family planning and STD-prevention program were significantly better than indicators from programs. The CHWsa large groupsought and ob- similar communities in which the Pastoral da Crianca tained civil service benefits, including a set salary struc- had no activities. ture and promotional opportunities. In 1985 salaries for Community health workers are also central to the briefly trained CHWs were to be equivalent to two- successful Aga Khan Health Service primary health thirds those of registered nurses with three years' care programs in remote mountainous areas of rural training. Health center buildings were altered to serve Pakistan. The CHWsvolunteers selected by the vil- as bases for CHW operations. Shortages of higher-level lagerscollect epidemiological information, provide staff prompted many health centers to substitute health education, identify problems, and provide sim- CHWs for nurses, even though the workers lacked the ple treatment and referrals. They are backed up by mo- necessary training. CHWs became increasingly linked bile teams of physicians and nurses. health indicators (Box 6.2). These same evalua- These skills are in short supply in most developing tions point to four necessary (but difficult) condi- countries. Public health often receives little atten- tions for success: community health workers must tion in basic medical curricula, specialty training is be well trained, well supervised, well provided often inadequate, and courses in public health with logistical support, and linked to well-func- schools may be too academic and not relevant to tioning district health systems for referral when local problems and needs. In Sub-Saharan Africa, needed. where public health capacity is weakest, fewer than 100 people receive specialty training in public Increasing training in public health, management, health annually. Some countries are exploring and policy, and planning implementing multidisciplinary training programs that include management and communication Improvements in health systems performance can techniques as well as the traditional public health be facilitated by training adequate numbers of poli- sciences. An innovative example of public health cymaking and management personnel, including training designed to produce future leaders is the public health specialists, policy analysts, hospital Union School of Public Health in Beijing, estab- managers, and drug management specialists. lished in 1989 to stimulate public health training in 143 the entire country. The school offers a master's in a priority sector (such as primary care or public degree in public health and draws students and health) or in an underserved location. Not only teachers both from health disciplines and from would professionals be better distributed and economics, management, and the social and envi- used, but there would be substantial savings of ronmental sciences. Training is based on problem public resources. solving, and more than half of the educational ex- Almost every country today is grappling with perience is in the form of community service. In problems in the mix and quality of its health pro- Zimbabwe, under a new public health training fessionals. Government financial policies can play program, students spend 75 percent of their time a constructive and central role in correcting market in the field. failures that lead to distortions in access to training Health policy and planning and good manage- and in the supply of professionals in different ment are fundamental (albeit insufficient) condi- fields. (For example, if credit is not widely access- tions for better performance of health services. ible, only the better-off may be able to go to medi- Over the past thirty years the role of managers, cal school; if the private rate of return for a certain economists, and planners in health services has specialty greatly exceeds the social rate of return, expanded in the industrial countries. For example, more professionals may choose that field than in many of these countries professional (nonphysi- would be socially optimal.) Government policy cian) hospital managers commonly run hospitals, can: in contrast to developing countries, where hospi- Help meet the need for training primary care tals tend to be run by physicians. As developing providers and other health professionals by im- countries seek to boost efficiency and as they move proving capital marketsusing student loan pro- toward decentralized management of health facili- grams, where feasibleand through national ser- ties, the need for trained managers increases. In vice mechanisms. most developing countries, however, training pro- Increase spending on training of, and im- grams in these areas are poorly developed. prove public sector wages and benefits for, health Distance education can facilitate training in pub- professionals in areas in which social benefits cur- lic health, health economics, and management by rently exceed private returns. These include, in allowing rapid implantation of what are often new particular, nonphysician primary care providers, curricula without the time-consuming task of health care managers, and staff in rural areas. training a new generation of teachers. Distance Limit or eliminate subsidies and financial in- learning has been used, for example, to build centives for specialist training. health research capacity in China. The University of Newcastle in Australia, in collaboration with Improving the selection, acquisition, and use Chinese universities, has set up a postgraduate of drugs distance-learning program in clinical epidemiol- ogy. The printed materials and academic stan- Drugs and vaccines embody much of the power of dards of the distance-learning program are equiva- modern medicine. Governments can enhance lent to those in the Australian program. Chinese their own utilization of drugs and assist the private professors help the students with applied labora- sector in increasing its efficiency through policies tory and research work. that improve selection, rationalize acquisition and production, and encourage better use. Through Reforming the finance of health training drug regulation and the development of a national list of essential drugs of established cost-effective- Many of the problems with human resources in ness, governments can help providers and con- the health sector derive from the fundamental flaw sumers make better choices among the approxi- of public subsidization of medical training. If phy- mately 100,000 different drugscomposed of more sicians paid the full costs of their training, it would than 5,000 different active substancesnow avail- be of no concern if they were later employed in able worldwide. Governments can encourage nonmedical work. Public subsidies could be specif- health systems to buy drugs of assured quality ically targeted to encourage those training and ca- from the lowest-cost supplier, whether domestic reer choices that are in the public interest. or international. They can eliminate the incentives Student loans could replace most of the current that in many countries induce physicians to over- public subsidies for training. Repayment of loans prescribe drugs because of the profits they earn might then be forgone if the trainee agreed to work from directly dispensing them. In China, Japan, 144 and Korea such incentives helped to drive drug therapeutic products. Evidence from the United spending up to 35 to 50 percent of total health Kingdom and other countries shows that the spending. adoption of formularies can contribute to consider- able savings in drug costs if physicians are in- Selecting essential drugs volved in their development and are educated about the results. The Model List of Essential Drugs developed by Governments are also responsible for carrying WHO suggests a basic list of drugs that WHO con- out regulatory functions to ensure that all drugs siders important and effective for dealing with on the market are of acceptable quality, safety, and health problems in developing countries. First efficacy. Building up a national regulatory author- drawn up in 1977 by art expert panel, the original ity requires the creation of a core group of trained list has been revised and updated seven times and staff, enactment of supporting legislation for ad- now includes about 270 products. It is designed to ministrative drug review, and the establishment of serve as a template from which countries can de- quality assurance laboratories. These are impor- velop their own still more specific lists of essential tant areas for donor assistance and perhaps for drugs. internationally shared efforts. Drugs on the national essential list are intended to be available at all times and in the appropriate Acquiring and producing drugs dosage forms in publicly provided health services. At the health center level about thirty to forty In 1990 the public and private sectors in develop- drugs can treat almost all complaints. District hos- ing countries spent an estimated $44 billion, or $10 pitals require no more than 120 drugs. If properly per capita, on pharmaceuticals. Global expendi- purchased, these drugs tend to be relatively inex- tures on pharmaceuticals amounted to about $220 pensive; almost all have multiple suppliers on in- billion, or $40 per capita. Total expenditures on ternational markets. Drugs are listed by interna- human vaccines, excluding those made in devel- tional, nonproprietary (generic) names. Although oping countries, were between $1.6 billion and many countries have created these essential drug $2.0 billion in 1992. Drug expenditures vary lists, only a few have used them to guide purchas- widely, from a low of $2 per capita in parts of Sub- ing and management of public sector (or publicly Saharan Africa and in Bangladesh to a high of $412 financed) drug supplies. And occasionally national in Japan (Table 6.1). drug lists have omitted important products, par- ticularly contraceptives. Table 6.1 Annual drug expenditures per Bangladesh and Sudan use limited lists not only capita, selected countries, 1990 to select drugs for public financing but also to Expenditure guide the national drug registration process, Country (dollars) thereby affecting the mix of drugs available in the Japan 412 private sector as well. Norway has limited the Germany 222 number of drugs registered by incorporating cost- United States 191 effectiveness, among other factors, into the review Canada 124 process. Since 1991 Zimbabwe has used its na- United Kingdom 97 Norway 89 tional list to determine which drugs can be im- Costa Rica 37 ported by the private sector without a permit. Chile 30 The applicability of the essential drug concept is Mexico 28 not limited to developing countries; drug formul- Turkey 21 aries, which are detailed lists of essential drugs, Morocco 17 Brazil 16 are widely used by institutional health providers Philippines 11 (public or private) and insurance companies in in- Ghana 10 dustrial countries. The formulary contains the China 7 names of drugs that are approved or recom- Pakistan 7 mended for health providers and supply systems. Indonesia 5 Kenya 4 It also provides useful information for individual India 3 prescribers. In creating formularies, drugs are as- Bangladesh 2 sessed on the basis of their safety, effectiveness, Mozambique 2 and cost-effectiveness in comparison with other Source; Ballance, Pogany, and Forstner 1992. 145 In most established market economies phar- cedures, and selection of generic drugs on the maceuticals and vaccines account for between 5 basis of its national essential drug list. This, of and 20 percent of health care spending, and, ex- course, is facilitated by Costa Rica's political stabil- cept in Canada and the United States, more than ity. In 1986 several Caribbean islands joined to- half of all drug expenditures are publicly financed. gether to carry out international tenders through In developing countries, households' out-of- the Caribbean Development Bank. In the first year pocket expenditures make up a much larger pro- they saved 44 percent over previous prices. portion of total drug spending. In Côte d'Ivoire The first step toward efficient procurement is and Pakistan, more than 90 percent of household careful quantification of drug and vaccine supply health expenditure is devoted to drugs. In the needs over a given period, using essential drug public sector drugs generally account for between lists or formularies where possible. Large stocks of 10 and 30 percent of total recurrent costs, making low-priority drugs have high opportunity costs: them the second largest category after salaries. they tie up resources and may expire before they Given this high volume of expenditure, achieving can be used. Shortages of high-priority drugs are the substantial improvements in efficiency of also costly; emergency purchases from local sup- procurement that are possible becomes a high pliers are always expensive. Good forecasting per- priority. mits economical purchasing. Some governments and many donors purchase PURCHASING DRUGS AND VACCINES EFFICIENTLY. drugs through international agencies (see Box 6.3). Some countries have achieved savings of 40 to 60 These agencies use international tendering and, percent in pharmaceutical expenditure by improv- because of the scale of their purchases and their ing selection and by competitive purchasing. For low operating margins, pass on very low prices. example, for several years the Costa Rican social (The total amount of drugs procured in this way is, security agency has been able to purchase drugs at however, small in relation to total drug expendi- approximately half the price of its counterpart in- tures in developing countries.) UNICEF purchased stitutions in other Central American countries, about $160 million worth of pharmaceuticals, vac- partly because of its use of centralized purchasing, cines, and related supplies for developing coun- more open and transparent purchasing pro- tries in 1992. Ethiopia, Sudan, Tanzania, and Box 6.3 Buying right: how international agencies save on purchases of pharmaceuticals UNICEF and several nonprofit organizations offer pur- plies on behalf of governments and nonprofit organiza- chasing services that enable countries to obtain favor- tions in more than eighty developing countries. Its cur- able prices for drugs, vaccines, and some medical rent annual turnover amounts to $80 million. IDA also equipment. UNICEF, the biggest in the field, has sup- carries out quality assurance, checking that manufac- plied basic drugs and vaccines since the 1960s. In 1983 turers produce in accordance with internationally ac- it issued its first international invitation to tender for cepted standards. When the drugs are received, IDA the bulk purchase of pharmaceuticals for Tanzania. The tests samples for quality and verifies labels and certifi- prices quoted against the invitation to tender were up cates of analysis. to 50 percent lower than previous price quotations. As Price lists from UNICEF and IDA provide valuable a result of this experience, UNICEF has continued to market information for countries' own procurement. use international tendering for the bulk purchase of Competitive tendering in Mali reduced prices by 40 pharmaceuticals and to pass on these favorable prices percent. In Kenya bulk purchasing of carefully selected to developing countries. UNICEF contracts with the essential drugs was estimated in 1985 to save nearly 40 Danish National Board of Health to provide advice on percent (or $700,000) of the annual drug bill for church quality assurance for pharmaceutical products. In 1992 health institutions. In 1992 the Chinese government UNICEF's purchases of drugs ($61.2 million), vaccines carried out international competitive bidding for drugs ($63.6 million), and refrigeration equipment, syringes, for tuberculosis treatment and_perhaps because of the needles, and sterilizers ($33.4 million) were delivered very large scale of procurement involved, the low-cost to more than 120 countries. packaging requirements, and the desire of manufac- The International Dispensary Association (IDA), es- turers to enter the Chinese marketachieved savings tablished in 1972, is a nonprofit supplier of drugs to of about 70 percent of UNICEF's published prices. developing countries. IDA procures drugs and sup- 146 Zambia have all relied heavily on nonprofit inter- Brazil, Indonesia, and Turkeyhave primary man- national drug suppliers. ufacturing capabilities or the ability to produce But many other developing countries fail to take both therapeutic ingredients and finished prod- advantage of international competition or interna- ucts. But most developing countries either have tional agencies. Purchasing methods, as well as only the capacity to produce finished products import restrictions, tend to restrict competition from imported ingredients or have no manufactur- and thereby raise prices. In addition, price compe- ing capability whatsoever. (Countries in the latter tition is restricted by the industry's extensive drug group are typically very small.) promotion practices and, in the case of patented Except in the largest countries that have primary products, by monopoly power. Some countries, manufacturing capabilities, local pharmaceutical such as Venezuela and Zimbabwe, protect local production in developing countries is likely to pharmaceutical industries from international com- make sense only for intravenous fluids, which petition (imported drugs will not be approved for have relatively high transport costs; for local pack- import and sale if there is a local producer), and aging of bulk imports in finishing plants; and for Belize and other countries impose import tariffs packaging of oral rehydration salts. Even in these even if there is no local production. This results in activities local production may be inefficient and great variation in prices for pharmaceuticals and waste scarce resources. State-run drug and vaccine supplies in developing countries. Cross-country companies, from which the public sector pur- data on the retail price of condoms show remark- chases preferentially, are common in many coun- able variation: condoms cost only $2 to $3 per 100 tries, including Bangladesh, Brazil, India, and in China, Egypt, and Tunisia, $15 to $30 per 100 in Laos. In some countries the local pharmaceutical Costa Rica, Ecuador, and Mexico, and more than industry (public or private) produces drugs that $70 per 100 in Brazil, Burundi, Myanmar, and Ven- could be purchased less expensively elsewhere. ezuela. This price variation is attributable to a com- Such industries survive only because of the protec- bination of factors, including import tariffs, import tion accorded through the prohibition of compet- restrictions, and wholesale and retail marketing ing imports, through import tariffs, or through structures. guaranteed agreements for public purchase re- Some countries purchase directly from a few lo- gardless of price. cal suppliers because of liquidity constraints. Inter- The combination of protection and poor regula- national agencies do not extend credit, and they tion can be particularly damaging. A 1990 study of require payment in hard currency. Local suppliers more than 6,000 infants in Bangladesh revealed often extend credit in exchange for significantly that the mothers' tetanus toxoid vaccinations did higher prices. Changing this practice to take ad- not reduce the risk of tetanus. Subsequent testing vantage of benefits from competitive procurement in reference laboratories of Bangladesh-produced would require the ministries of both health and vaccine indicated no potency in several consecu- finance to make budgetary funds and foreign ex- tive batches, raising questions about the efficacy of change available when needed for large-scale drug the more than 40 million doses already adminis- purchases. Governments can also improve drug tered. Since Bangladesh has no independent na- procurement by passing legislation to facilitate ge- tional control authority for certifying vaccine neric drug prescribing. This can increase the af- safety, all testing had been done by the production fordability of drugs purchased from private facility itself. Evidence suggests that few public outlets. sector pharmaceutical and vaccine producers have been able to operate competitively, in terms of PHARMACEUTICAL PRODUCTION. The cost of de- price and quality, in the highly competitive and veloping a sophisticated pharmaceutical industry rapidly changing pharmaceuticals market. Im- with a significant research base is huge. During proved selection and purchasing practicesrather 1961-90, 90 percent of the approximately 2,000 than protection of drug manufacturingwill usu- "new chemical entities" (new drugs) brought on ally be the best ways to counter the market power the market were discovered in only ten OECD of international suppliers of drugs. countries. Five countries in the developing worldArgentina, China, India, Korea, and Mex- IMPROVING STORAGE AND DISTRIBUTION. Theft, icodiscovered, developed, and marketed at least spoilage, and shortages are major problems facing one new chemical entity between 1961 and 1990. public distribution in many countries. Systems for Several other developing countriesamong them, inventory control, port clearing, storage, and de- 147 livery can address many of these problems. In providers cannot possibly review all the informa- Zimbabwe a standard nationwide system of stock tion available on the quality, safety, and efficacy of control was fundamental to recent reforms in the drugs and vaccines, governmental involvement in drug supply system. Surveys show a gradual im- regulation and in provision of information is nec- provement in drug availability: in 1991 the facili- essary. In addition, the government must manage ties surveyed had 78 percent of the representative drug selection, procurement, and distribution for essential drugs in stock, up from 38 percent in publicly provided health services. To support the 1987. In hospitals, management information sys- rational use of drugs, governments can: tems help to track periods of drug validity and to Develop a national list of essential drugs and analyze rotation rates and drug consumption. direct public finance to those drugs that support the essential package of clinical services and public Influencing prescription and self-medication health interventions. patterns Purchase drugs competitively and reduce or eliminate protection of local pharmaceutical pro- Significant efficiencies can be achieved by improv- duction of vaccines and drugs. These policies work ing prescription and self-medication practices. to consumers' benefit. Efficient local industry is Widespread overprescription and inappropriate best created under competitive conditions. prescription have been documented in most coun- Provide information to public and private pro- tries. For example, recent surveys found that the viders and consumers on drug use and cost-effec- average number of drugs prescribed per single tiveness and establish regulations that discourage consultation in public health centers ranged from overuse or overprescription. 1.3 in Zimbabwe and Ecuador to as high as 3.3 in Indonesia and 3.8 in Nigeria. These surveys also Generating information and strengthening documented that unnecessarily high proportions research of drugs were being administered in the form of injections (which carry the risk of abscesses, nerve In health, as elsewhere, good information facili- injuries, and transmission of infectious disease) tates sound decisionmaking. Although some basic and that extensive overuse of antibiotics was oc- health information is generated by the private sec- curring. A survey of seventy-five pharmacies in tor without government involvement, the govern- three Asian countries found that only sixteen gave ment has a central role in requiring, standardizing, appropriate advice regarding oral rehydration for and financing the collection, analysis, and dissem- treatment of diarrhea in infants. ination of health information, as well as in financ- Public policies for improving prescription and ing health systems research. Governments are al- medication practices include: ready heavily involved in data collection. Distribution to health care providers and Unfortunately, the data are often irrelevant to p01- pharmacists of regularly updated essential drug icy and program design. And too often, the private lists or formularies that include descriptions of sector is ignored when statistics are being gath- use, dose, adverse reactions, and costs; examples ered. Revamping health information systems is an include the British National Formulary and the attractive investment, both because it is relatively Uganda Drug Formulary inexpensive and because poor decisions based on Strengthening of medical and nursing train- inadequate information can be very costly. But the ing regarding pharmacology, appropriate prescrib- impact of information systems depends crucially ing practices, and problems caused by over- on the decisionmaking environment. Even the prescription and unnecessary use of injections best systems may be seen as irrelevant if managers Public education on appropriate drug use, the have no incentive or scope for using information to disadvantages of injections when oral doses are improve efficiency. Information helps guide available, and the importance of compliance with choices among the existing options, and invest- the full course of therapy ments in research and development create new Removal of financial incentives that encour- options, both for households and for providers of age physicians to overprescribe. care. It can be argued that investments in research Unlike facilities, equipment, and human re- have been the source of the enormous improve- sources, pharmaceuticals and vaccines are an area ments in health in this century. This section dis- in which government policies can alter input use cusses ways of ensuring continued benefits from relatively quickly. And good policies could make a research, as well as the role of the international significant contribution. Because consumers and community in this task. 148 Understanding health status and health risks Some countries have established surveillance systems that rely on sentinel districts selected to be An essential step toward improving health is to understand the distribution of disease, death, and roughly representative of the country. To improve disability. This requires the systematic collection, the speed and accuracy of reporting, data collec- analysis, and dissemination of timely and accurate tion systems are upgraded in these districts to a information on mortality, morbidity, and risk fac- greater degree than could be done for the country tors. Such data are a cornerstone of public health as a whole. Cause-specific death rates, vaccine efforts in any country, and the government's role coverage, the effectiveness of vaccines, and the is central in creating them because the private sec- impact of specific health interventions are then tor has little interest in producing such public monitored intensively within the district. National goods. Epidemiological data are used to estimate household surveys can also generate a wealth of the magnitude of health problems, study risk fac- information on health status, risk factors, and the utilization of health services according to age, sex, tors, evaluate health programs and the effective- ness of interventions, detect epidemics, facilitate region, and racial and ethnic group (see Box 6.4). planning, and monitor changes in health practices. Unlike government health service statistics, popu- These data could be used to estimate a national lation-based surveys cover nonusers as well as users of public services. burden of disease similar to the global burden of disease estimated for this Report. The national Monitoring health spending and equity burden of disease would quantify the loss of healthy life from the diseases that are important in Previous chapters have recommended redirecting the specific country. It could be used to monitor public spending to nationally defined essential and track over time improvements in both mortal- clinical services, targeted largely to the poor, and ity and morbidity. to public health interventions, leaving to private Box 6.4 The contribution of standardized survey programs to health information Three internationally supported standardized house- immunizations, health care behavior, and other aspects hold survey programs have contributed immensely to of child health. DHS survey information has been used knowledge of health conditions, particularly those of for purposes as diverse as examining the effects of eco- children, in the developing world over the past three nomic reversals on demographic outcomes and study- decades. The World Fertility Survey (WFS) sponsored ing small area variations in child mortality risks in ur- forty-three surveys between 1974 and 1982, with fund- ban areas. ing from the U.S. Agency for International Develop- Neither survey program has collected detailed eco- ment (USAID) and the United Nations Population nomic information on households and communities. Fund (UNPF) and some country contributions toward The World Bank's Living Standards Measurement Sur- the costs of survey fieldwork. The Demographic and vey (LSMS) was designed to fill this need by studying Health Surveys (DHS) program, started in 1984, has so the determinants and interactions of poverty, health, far implemented thirty-nine surveys in thirty coun- education, nutrition, and labor activities. The survey tries; the third phase of the survey program, with a collects a wealth of information about incomes, pro- planned twenty-five surveys, is about to begin. The duction, and prices. Some LSMS surveys are funded DHS program has received funding from USAID, with through World Bank-financed projects, but many have contributions from countries and other donors. received grant support from a variety of bilateral do- Both the WFS and the DHS program have used a nors, the UNDP, and other agencies. common core questionnaire around which special The experience with these standardized surveys in- topics could be explored. The core WFS questionnaire dicates the great value of using comparable survey pro- was primarily concerned with fertility and fertility- cedures and instruments across countries and the im- related behavior; for each eligible woman it included a portance of rigorous supervision at all stages of the birth history, recording the date of each birth and, if survey operation, from sampling to data processing. the child had died, the age at death. This information The LSMS and DHS programs have been particularly base has provided much of what is known about child successful with respect to turnaround time; prelimi- mortality trends and the relationships between child nary findings from a survey are available within six mortality and birth spacing, maternal education, and weeks of the conclusion of fieldwork, and a final report household characteristics. The DHS questionnaire, in typically becomes available within one year. addition to a birth history, includes questions about .- 149 finance health services outside the essential pack- National research priorities age. Private expenditures are always difficult to estimate, but even in the public sector, spending is Governments have a role in supporting the re- poorly disaggregated by use. By revamping infor- search necessary for understanding specific local mation systems, estimates can be made of spend- health problems and for guiding public policymak- ing on public health interventions and on catego- ing and program design. This "essential national ries of inputs (essential drugs, nonessential drugs, health research," which is also undertaken by the primary care physicians, other primary care pro- private sector, examines health strategy in more viders, specialists, health centers, district hospi- depth than is done with day-to-day budgetary and tals, and tertiary hospitals). Although still imper- management information. The international com- fect, such estimates better capture the nature of munity can help both in gathering data for interna- government spending. In addition, public expen- tional comparisons and in assisting local institu- ditures need to be regularly consolidated across tions to build up capacity in epidemiology, health federal, state, and local levels for analysis. In Bra- economics, health policy, and management. Re- zil, where state and local governments account for search priorities in this area include cost-effective- about half of all public spending on health, expen- ness analysis of health interventions, evaluations diture estimates are available only for federal of medical practice and of variations in practice spending (except for 1984). Much less information (see Box 6.5), and studies of drug utilization, eq- is compiled from state and municipal levels, de- uity, consumer satisfaction, and women's health. spite their importance. Household surveys can Where the national burden of disease is high collect appropriate information for monitoring and cost-effective interventions already exist, re- who benefits from public health spending. In part search can guide program implementation. One because such data are lacking, analyses of equity such example is the problem of intestinal parasitic in health care have been carried out in only a worms. How can local programs be best designed handful of developing countries, among them Co- to reach children? How can involvement of school lombia, Costa Rica, Côte d'Ivoire, Indonesia, Ma- officials be fostered? Another area is tuberculosis, laysia, and Peru. where treatment compliance is a chronic problem; Box 6.5 Evaluating cesarean sections in Brazil Operations research can examine variations in medical systematic variations by region, type of hospital, socio- practice with a view to identifying areas in which economic status of the woman, and reimbursement changes in practice are needed, as well as possible in- patterns. Rates in 1981 were higher in the more pros- struments for modifying provider practice. In the early perous Southeast (38 percent) and lowest in the poor 1980s Brazil was estimated to have the highest overall Northeast (20 percent). In every region the incidence of cesarean section rate in the world-31 percent of all cesarean section increased with family income. A 1986 hospital births in 1981. Although cesarean sections are survey showed that rates were highest for women with a life-saving procedure in certain circumstances, their a university education (61 percent) and for births in unnecessary use raises costs and poses medical risks private hospitals (57 percent). Other studies showed for the mother and the newborn. The financial cost of that rates were lowest among women with no insur- unnecessary publicly financed cesareans in Brazil was ance. Women covered under the social security system estimated at about $60 million annually in the late had higher rates of cesarean section, and women with 1980s. Medical risks stem from incorrect estimation of private insurance had the highest. the length of gestation (leading to premature deliv- The country's social security institute changed its re- eries), infection from surgery, and the use of general imbursement policies in the early 1980s to remove anesthesia. Among the many factors responsible for some of the financial incentives for cesarean sections, the rising rate of cesareans in Brazil are the financial and education campaigns for physicians were initiated. and administrative incentives for hospitals and doctors But it is clear that even stronger policies are needed to to perform cesarean deliveries, the desire to use a Ce- reverse these trends, as cesarean section rates have sarean delivery as a vehicle for obtaining a sterilization, continued at high levels. A large sample of births in the and the widespread view that cesarean section is the state of São Paulo in 1991, for example, indicated a preferred, "modern" way to deliver. cesarean section rate of 47 percent. Brazilian studies of cesarean section rates illustrate 150 patients often stop taking medication once they laundry, food preparation, and laboratory testing. feel better, but before the problem has been effec- Systems that gather information on vaccine utiliza- tively treated. What program approaches work tion, equipment and vehicle inventories, preven- best in different settings to ensure compliance tive maintenance for buildings and equipment, with directions? In nutrition, how can policies and personnel management, and the like are also programs promote dietary change most effec- fundamental. tively? Solutions to these problems are not univer- Ministries of health frequently pay little atten- sal. Research must be local, and often public sup- tion to the activities of private providers, instead port is needed. focusing all data collection efforts on public pro- In its 1990 report the Commission on Health Re- viders. To remedy this, governments can collect search for Development recommended the forma- basic information about private providers and the tion of international partnerships or networks to population covered under private insurance plans. focus on ensuring that national health resources They can require standardized reporting from are used to maximum effect. The International both public and private hospitals through uniform Network for the Rational Use of Drugs (INRUD), hospital discharge data. The information can then established in 1989, is one such network. Another be synthesized to provide consumers, health re- is the International Clinical Epidemiology Net- searchers, and communities with information work (INCLEN), which was started in the early about the quality of care given by providers, both 1980s by the Rockefeller Foundation to build up a public and private, and about variations in medical critical mass of researchers in clinical epidemiol- practice. These systems can generate sophisticated ogy, including epidemiologists, health econo- information; consumers in California, for example, mists, social scientists, and biostatisticians. IN- can obtain risk-adjusted mortality rates by hospital CLEN enrolls midcareer academic physicians who for common procedures. But relatively simple hold positions of influence in the medical systems measures can also be useful; an example is ce- of developing countries. It provides overseas sarean section rates, by hospital, which can help study opportunities, support for research, and the identify overuse of this procedure (see Box 6.5). opportunity to participate in annual scientific Such standardized information about hospital per- meetings. The network concept has permitted formance can help consumers make better choices units to share experiences and teaching materials about health care and can help central authorities and to carry out collaborative research between identify problems to be corrected. clinical epidemiology units, training centers, and If there are incentives for using information in the international health community. Capacity decisionmaking, improvements in data gathering building is a lengthy process, but INCLEN has al- can often be inspired simply by giving those who ready influenced health policy. Research on the need the information more training in how to col- effectiveness and efficiency of hepatitis B immu- lect it and more responsibility for doing so. District nization in the Philippines brought about the addi- medical officers, hospital superintendents, and tion of hepatitis B vaccine to the national EPI pro- health care managers are usually not trained to gram. Studies on the cost-effectiveness of short- make the best use of data. Whenever possible, tab- course chemotherapy for tuberculosis have led to a ulation of data should be decentralized so that lo- change in national treatment policies in Brazil, the cal decisionmakers can immediately use the infor- Philippines, and Thailand. mation instead of relying on feedback from central levels. In Papua New Guinea, for example, when Improving information at the district and facility levels local-level staff began to see the relevance of man- Health organizations also benefit from improve- agement information for their work, they sought ment of the information needed to make everyday to verify data and to eliminate reporting that was management decisions. In publicly provided dis- irrelevant. trict health facilities, simple management informa- To summarize, governments have a twofold role tion systems for measuring costs, inputs, and pro- in health information systems and operational re- duction could be helpful for monitoring program search: generating the information necessary to efforts over time and for making decisions about guide health policies and public spending and pro- how to combine inputs efficiently. Yet many public viding certain types of information about provider facilities operate without such information. With- performance that would be too costly for con- out basic data on costs, it is difficult to decide, for sumers to collect. To this end, governments can: example, whether to contract out services such as Gather and synthesize epidemiological and 151 Table 6.2 Some priorities for research and product development, ranked by the top six contributors to the global burden of disease Associated DALY loss (millions) Demographically developing Disease or injury countries FSE and EME Priority areas Perinatal and 125 4 Methods of lowering costs of intervention and maternal causes improving delivery in rural areas. Respiratory infections 119 4 Impact of indoor air pollution on pneumonia (to guide interventions designed to reduce pneumonia by use of improved stoves); inexpensive or simplified antibiotic regimens; inexpensive, simple, reliable diagnostics; pneumococcal vaccine Diarrheal diseases 99 - Rotavirus and enterotoxigenic E. coli vaccines; improved cholera vaccine; ways of improving hygiene; better case management of persistent diarrhea; prevention of diarrhea by the promotion of breastfeeding and improved weaning practices Ischemic heart and 58 27 Low-cost prevention, diagnosis, and management cerebrovascular disease methods Childhood cluster: 67 Development of new and improved vaccines to reduce diphtheria, polio, patient contacts, permit immunization at younger pertussis, measles, and ages, and improve heat stability of some vaccines tetanus Tuberculosis 46 1 Methods of ensuring compliance; monitoring tools for drug resistance; simpler diagnostics; new and cheaper drugs All conditions" 1,210 152 Note: The demographically developing countries are those in the Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent regions. FSE, formerly socialist economies of Europe. EME, established market economies. DALYs, disability-adjusted life years; see Box 1.3. Less than 0.5 million. Total for all conditions presented in Appendix B. Source: Appendix B. other information necessary to monitor health sta- malaria; oral rehydration therapy; antibiotics and tus, detect disease outbreaks, and guide public other antimicrobials; and synthetic hormonal policy and program design contraceptives. Support research, where needed, to generate Basic research and product development are local solutions to local problems public goods that require support through govern- Facilitate standardization of information ment subsidy or intervention (for example, grants about health production and health outcomes by of patents). In addition, because the poor in devel- district health systems and other major health pro- oping countries lack market power, the system of viders; where necessary, synthesize and publicize patent protection fails to provide incentives to the this information to aid consumers in making in- commercial sector for developments related to dis- formed choices about health care. eases of the poor. Thus, there is a clear argument for government and international assistance to cat- Expanding the range of choice alyze technological development. In the develop- ing world many serious health problems do not A revolution in health care technology has taken present sufficiently attractive commercial markets place in the course of this century. Significant bio- to induce the development by private companies medical breakthroughs that have generated inter- of better methods of prevention, diagnosis, and national benefitsfor developing countries as well treatment. Developing countries account for al- as for the established market economiesinclude most 90 percent of the global burden of disease, the development of measles, pertussis, polio, and and much of that burden is from conditions such tetanus vaccines; chioroquine for the treatment of as malaria or tuberculosis that primarily occur in 152 those countries. Only about 5 percent of the $30 eases: measles, tetanus, pertussis, diphtheria, billion global investment in health research in 1986 polio, and tuberculosis. It requires at least seven went to health problems unique to developing patient contacts (two for the pregnant mother and countries. five for the infant). Possible improvements in vac- cine technology would reduce multidose vaccines Setting priorities to a single dose, improve the heat stability of vac- Where is extra research really likely to pay off? cines, simplify administrative requirements (to Table 6.2 suggests priorities for research on pre- permit greater use of oral vaccines as compared vention, diagnosis, and case management for the with injections, for example), create new combina- six conditions that make the largest contributions tions of vaccines to reduce patient contacts, inte- to the global burden of disease. These conditions grate new vaccines into the immunization sched- account for about 40 percent of the DALYs lost in ule, permit vaccination earlier in life to reduce demographically developing countries and for infant deaths caused by vaccine-preventable dis- about 25 percent of the losses in industrial coun- eases, and add to the menu of interventions new tries (where cardiovascular disease accounts for vaccinesfor example, against diarrhea and pneu- much of the burden). If the global burden caused monia. These innovations would reduce some of by a disease is large, if no cost-effective interven- the costs and improve the effectiveness of vaccina- tions exist, and if experts believe that such inter- tion programs. An important source of support for ventions might be developed, there is a case for this research is the Children's Vaccine Initiative greater investment in research and product devel- (CVI), which is identifying measures for catalyzing opment. One example that meets these criteria is technological development in these areas. The inexpensive, simple, and reliable diagnostics for CVI, which has its secretariat at WHO, is an inter- respiratory infections. For problems that create a national effort to harness new technologies that large burden of disease and for which cost-effec- can advance the immunization of children. tive interventions already exist, there is a need to direct efforts more toward program development TROPICAL DISEASES. It is primarily the rural poor and operational research to guide implementation. who suffer from tropical diseases such as malaria, For example, little is known about low-cost schistosomiasis, lymphatic filariasis, onchocer- methods of managing ischemic heart disease in ciasis (river blindness), trypanosomiasis, and lep- developing country settings. One low-cost ap- rosy. These diseases create a high burden, and ex- proach that is being adopted in many industrial isting interventions are inadequate against many countries is the use of low daily doses of aspirin to of them. The UNDP-World Bank-WHO Special reduce the risk of obstructive blood clots inside the Programme for Research and Training in Tropical arteries. This approach, developed on the basis of Diseases (TDR) is developing partnerships with the results of large-scale assessments of the effi- commercial entities, national governments, scien- cacy of the intervention, illustrates the potential tists, and NGOs to support research and drug de- benefits of research on low-cost case management. velopment for these diseases. One strategy the International agencies and governments can program has adopted is to look for new applica- stimulate research on health and product develop- tions of drugs already in use in human or veteri- ment in several ways. They can provide informa- nary medicine. An example is the use of ivermec- tion on potential markets for new products, in- tin, a drug originally marketed by Merck & Co. for cluding epidemiological data about the disease, treating worms in animals, in the fight against on- the target population, and the technical require- chocerciasis in human populations (see Box 1.1). ments of desirable innovations. They can subsi- The TDR program facilitated the field testing of dize a portion of the development costs. They can this product on a large scale for human use. The facilitate or finance field evaluations in a variety of results showed that the drug was very safe, that it settings and support introduction of the technol- could be distributed by primary health care ogy in the field. Finally, they can provide procure- workers, and that one oral dose per year could ment guarantees for new or improved products at prevent or arrest blindness. As a veterinary prod- an agreed-on price. A few examples illustrate the uct, ivermectin has estimated annual sales of $500 potential. million; Merck & Co. agreed to supply the drug without charge to governments for treatment of Nuw AND IMPROVED CHILDHOOD VACCINES. The human onchocerciasis. The TDR's network of in- EPI currently includes vaccines against six dis- ternationally funded research centers in develop- 153 Box 6.6 An unmet need: inexpensive and simple diagnostics for STDs This Report recommends that concerted efforts be stick, developed by the Program for Appropriate Tech- made to develop or strengthen effective programs for nology for Health with support from Canada's Interna- control of STDs. Such efforts will be hampered by the tional Development Research Centre (IDRC) and from challenges of diagnosing STDs, particularly in women, private funds, uses synthetic peptides and a color for whom the vast majority of infections are asympto- change to provide an easily performed test. The test matic. Current methods are often unreliable and ex- takes twenty minutes, requires only three simple steps, pensive, and their use requires refrigeration, electric- is stable for six months at tropical temperatures, has a ity, and sophisticated equipment and training. In pattern of sensitivity and specificity similar to commer- addition, certain tests require patients to return in one cially available tests, and costs less than $0.20. Thus, or two days, which is not feasible when, as is often the the per patient cost for testing can be brought down to case, the patient must travel a long distance to receive less than $1.00, including a confirmatory second test. health care. Even if patients return, the period of infec- This test is now being commercially produced in India tivity is prolonged by this delay in therapy. Syndromic- and Thailand. The Canadian International Develop- based approaches to treating STDs are currently being ment Agency (CIDA) is funding the establishment of a used to bridge this gap and are effective for men. For production facility in Cameroon, and there is interest women, however, these approaches are less accurate. in production in Brazil, Indonesia, and Zimbabwe. New diagnostics that are inexpensive, simple, and The STD Diagnostics Initiative, which is funded by convenient to use and provide rapid, stable, and accu- multiple donors, was established to facilitate develop- rate results would overcome these problems. An exam- ment of appropriate diagnostics for resource-limited ple of such a tool is a new HIV test. The availability of settings. The initiative, being carried out in collabora- HJV testing has been limited by high cost, complexity, tion with industry, clarifies and validates performance and requirements for reagents that need refrigeration criteria for STD diagnostics, organizes and supports and have a short shelf life. Even when labor costs are field trials, provides seed money for the development excluded, testing and confirmation can cost $25 to $50 of new diagnostics, and brokers bulk purchases to cre- (although this cost is declining rapidly). The HIV dip- ate markets of adequate size. ing countries allowed it to respond quickly and cancer, rapid plasma finger-stick diagnostic tools flexibly to the opportunity to test ivermectin. for syphilis, and new diagnostic tests for malaria Women spend up to half of their reproductive for use at the local level. (Box 6.6 provides another lives pregnant or lactating. Many protocols for example.) Rapid diagnostic tests avoid reliance on treating tropical diseases exclude these women other levels of the health system because the and sometimes even large numbers of women health center, if supplied with the necessary who might be pregnant (such as adolescent girls). drugs, can treat the problem on the spot. Innova- Blanket exclusion of pregnant or lactating women tions in medical equipment to reduce the cost or has been the result not of clear evidence of prob- improve the effectiveness of preventing and treat- lems but of reluctance to carry out appropriate ing problems at the health center level are high drug trials on pregnant women. There is an urgent priorities for research and development. need to evaluate drug treatments for such women so that health services can offer them better treat- International aspects ment. This is part of a much broader problem of the common omission of women from medical Some types of research and product development studies and clinical trials in both developing and are costly; it can cost more than $100 million to industrial countries. bring a new drug to market. But several break- throughs in medical technology have been inex- MEDICAL EQUIPMENT. Another priority area for pensive. (One of them, oral rehydration therapy, research and development is the development of is now widely recognized as an effective way of low-cost and efficient diagnostic technologies for treating acute watery diarrhea, which, untreated, use in health centers in developing countries can weaken or kill young children.) The need for where sophisticated laboratories are unavailable. public support for certain types of research is Examples of potentially important new technolo- widely understood. The international community gies are visual methods of screening for cervical has played an important role in supporting health 154 research, and most governments support some re- dertaken by the private sector at socially optimal search as well. Over the short to medium term, levels. (Even research that is internationally fi- developing countries can use their scarce public nanced will take place principally in developing resources best if: countries and will be done increasingly by scien- Governments reduce or eliminate finance of tists from developing countries.) The total invest- basic biomedical research that generates interna- ment in health technology research relevant to the tional benefits (which is best supported by the in- needs of the developing countries is woefully in- ternational community) and redirect it toward re- adequate in relation to its potential benefits. And search efforts that generate primarily national the level of international coordination and cooper- benefits ation falls well short of what is required. An inter- The international community directs research national mechanism with stable funding over the support toward new and improved technologies medium to long term could more effectively build where the expected social returns are highest and research capacity in developing countries. Donors would benefit many countries. and governments also need to give more support International financing is needed for important to activities for testing new technologies and incor- biomedical research when the benefits transcend porating them into health systems. national borders and the research will not be un- 155 An agenda for action The policy conclusions of this Report can be tai- quire donor backing for major reforms in the al- lored to the widely varying circumstances of de- location of public spending for health and in veloping countries. This chapter highlights the pri- health policy more generally. ority policy issues and actions that are likely to be The effectiveness of donor spending can be im- most relevant for three groups of countries: low- proved through increased investment in basic income countries in Africa and South Asia, mid- public health measures and essential clinical care, dle-income countries in Latin America and East steps to strengthen the policy and regulatory Asia, and the formerly socialist countries of Eu- framework for insurance and for delivery of ser- rope and Central Asia. It describes the reforms vices, and backing for research to expand the needed in the health sector and assesses their fea- range of cost-effective treatments available to the sibility, examines the principal obstacles to reform, poor in developing countries. Aid for lower-prior- and outlines possible strategies for overcoming ity items, including tertiary care hospitals and these obstacles. Although policy reform must deal training of medical specialists, needs to be corre- with difficult underlying problems, the experience spondingly reduced or eliminated. of a number of developing countries with imple- Finally, improved coordination among donors menting significant policy changes shows that suc- could raise the effectiveness of aid. Despite the cess is possible. many serious obstacles, the recent experience of a This chapter also examines the role of the inter- number of African and Asian countries shows that national community in supporting improvements such coordination can be achieved. in health policies and programs in developing countries. Despite widespread calls for more do- Health policy reform in developing countries nor investment in human resources and in poverty reduction programs, aid flows to the health sector The policies that this Report suggests should be at declined from 7 percent of total development assis- the top of the agenda for developing countries and tance in the early 1980s to 6 percent in the latter the donor community are summarized in Table half of the 1980s. Donors need to match their ver- 7.1. This section describes those policies and pro- bal commitments with actions: the share of aid for vides examples of successful policy reforms in var- health should be restored to its previous level im- ious developing countries. mediately and should be increased substantially over the next five years. An additional $2 billion in Low-income countries aid would help to finance the transitional cost of health policy reforms, as well as priority pro- Previous chapters have outlined the main charac- grams, including AIDS prevention. At the same teristics of health systems in low-income coun- time, donors and developing countries need to tries. In general, there is little public or private focus on measures to improve the effectiveness of insurance. Out-of-pocket spending for drugs, tra- external assistance for health. Doing so will re- ditional medicine, and user fees usually accounts 156 Table 7.1 The relevance of policy changes for three country groups Government objectives and policies Foster an enabling environment for households to improve health Pursue economic growth policies that benefit the poor Expand investment in education, particularly for females Promote the rights and status of women through political and economic empowerment and legal protection against abuse Improve government investments in health Reduce government expenditures for tertiary care facilities, specialist training, and discretionary services Finance and ensure delivery of a public health package, including AIDS prevention Finance and ensure delivery of essential clinical services, at least to the poor Improve the management of public health services Facilitate involvement by the private sector Encourage private finance and provision of insurance (with incentives to contain costs) for all discretionary clinical services Encourage private sector delivery of clinical services, including those that are publicly financed Provide information on performance and cost Very relevant 0 Relevant 0 Somewhat relevant Not relevant for more than half of total spending for health. unlicensed practitioners in South Asia) account for Government financing from general tax reve- the remainder of the health facilities and deliver nuesand sometimes substantial donor contribu- most outpatient care. They offer a service that is tionsaccount for the remainder. Government perceived to be of higher quality than that pro- hospitals and clinics provide the bulk of modern vided by the public sector. Large segments of the medical care, but they suffer from highly central- population, especially the rural poor, do not have ized decisionmaking, wide fluctuations in annual access to modern health services. Female literacy budget allocations, and poor motivation of both and enrollment of girls in primary and secondary facility managers and health care workers. Minis- school are low. tries of health and other government agencies of- Five policies for better health are crucial in this ten have only limited capacity to formulate health environment: providing solid primary schooling policy, implement health plans, and regulate the for all children, especially girls; investing more re- private sector. Private providers (mainly religious sources in highly cost-effective public health activ- organizations in Africa and private physicians and ities that can substantially improve the health of 157 the poor; shifting health spending for clinical ser- sary radio and telephone networks; and building vices from tertiary care facilities to district health the capacity to plan and manage health services at infrastructure capable of delivering essential clini- the district level and in individual facilities. In cal care; reducing waste and inefficiency in gov- many low-income countries, focusing on district ernment health programs; and encouraging in- health infrastructure will mean limiting new in- creased community control and financing of vestment in central hospitals and reorienting those essential health care. facilities toward research and teaching activities that are more relevant to key national health prob- INCREASED SCHOOLING. Despite the often formi- lems. At the same time, there is considerable scope dable obstaclesboth in providing access to for improving the efficiency of large government schools and in eliminating cultural barriers that hospitals, especially through performance-linked keep girls outa number of low-income countries incentives for managers and staff and expanded have proved that dramatic change is possible in a cost recovery from the wealthy and insured. short period of time. Between 1970 and 1990 Indo- To deliver essential clinical services, a greater nesia and Kenya, for example, achieved rapid and share of government health budgets needs to be sustained growth of primary school enrollments devoted to the operations of lower-level facilities and raised the proportion of girls to nearly half of and especially to nonsalary recurrent items. Initial all pupils. These gains were brought about by a emphasis needs to be placed on building capacity combination of high-level political commitment to to deliver the services included in the minimum universal primary schooling, information pro- essential package described in Chapter 5. This is grams that created stronger demand on the part of now happening in a number of countries. Senegal parents, and support from the international has set annual targets for increasing its spending community. for drugs, transport, and maintenance. Ghana is INVESTMENT IN PUBLIC HEALTH ACTIVITIES. The trying to reduce the number of civil servants work- public health activities with the largest payoff will ing for the Ministry of Health. In India, where vary from country to country: vitamin A and io- state governments account for more than three- dine supplementation in India and Indonesia, quarters of total public spending for health, the antismoking campaigns in China, and policies to central government is attempting to act as a cata- reduce traffic injuries in urban areas of Sub- lyst for more cost-effective resource allocation by Saharan Africa. Completion of immunization cov- earmarking its funds for immunization, treatment erage should be a high priority in all low-income of leprosy and tuberculosis, and AIDS control. countries, especially in India and in much of Sub- Some low-income countries will need to in- Saharan Africa, where coverage remains low. Sim- crease government outlays for health if they are to ilarly, a greatly intensified effort to reduce trans- finance a package of public health measures and mission of HIV and other sexually transmitted dis- essential clinical services for the poor. In 1990 gov- eases is warranted. In the parts of Africa in which ernment spending for health in low-income coun- the AIDS epidemic is already widespread, behav- tries averaged only $6 per capita-1.5 percent of ioral change through education and condom distri- GNP if foreign assistance is excluded and 1.6 per- bution should be high on the list of public health cent including aid. The analysis in this Report in- actions. And where, as in Bangladesh and Indo- dicates that provision of a minimum package will nesia, the preconditions (widespread commercial cost about $12 per capita in low-income countries, sex and high prevalence of other STD5) exist for or nearly 3 percent of GNP. Effective targeting of rapid spread of HIV, governments urgently need publicly subsidized clinical services to the poor, to take steps to halt the spread of AIDS from high- and corresponding efforts to encourage cost recov- risk groups into the population at large. ery from more affluent groups, would help stretch limited government budgets. Modest fees col- BETrER ALLOCATION OF SPENDING ON CLINICAL lected at health centers could also be retained and SERVICES. Governments should invest in distfict reinvested locally to improve the quality and re- health infrastructure by (as described in Chapter 6) liability of basic services. expanding training programs for primary care pro- But even with these efforts, many governments viders, particularly nurses and midwives; target- in low-income countries will have to increase the ing construction funds to improve health posts, share of the budget allocated to health. (In Sub- health centers, and district hospitals; financing Saharan Africa health spending declined during ambulances and other vehicles needed for effec- the 1980s to an average of less than 4 percent of tive emergency transport, together with the neces- public expenditure and less than 2 percent of 158 Box 7.1 Community financing of health centers: the Bamako Initiative The principal aim of the Bamako Initiative, launched in The initiative is only five years old, but its achieve- 1988, is to "revitalize the public sector health care de- ments are impressive. Eighteen African countries were livery system [byl strengthening district management participating as of late 1991, and nearly 1,800 health [andi capturing some of the resources the people them- centers located in 221 districts were part of the pro- selves are spending on health" (UNICEF 1992). gram. In Benin the first forty-four health centers tar- Both revolving funds for drug purchases and com- geted by the initiative are covering 42 to 46 percent of munity-managed health centers have existed for many their operating costs with user charges, and in the first years in developing countries, but the Bamako Initia- seventeen centers in Guinea's program, user fees cover tive is attempting to implement these schemes on a 38 to 49 percent of expenditures. Utilization of health much larger scale in Africa and other low-income coun- centers has increased. In Benin average monthly visits tries. The initiative is based on two premises: that to pilot health centers rose from 100 in 1987 to 250 in where public institutions are weak, as they are in many 1989. low-income countries, bottom-up action by commu- Despite the initiative's promising accomplishments, nities is badly needed to complement top-down health it is not yet certain that the reforms can be sustained Ofl policy reforms, and that even poor households are will- a large scale. A number of health centers covered by ing to pay for higher-quality and more reliable health the initiative have received both financial and technical services. assistance from UNICEF, WHO, and other donors Under the initiative, members of local communities more than $36 million has come from UNICEF alone. who use a health center or pharmacy agree to pay mod- Problems may emerge when this external assistance est charges for outpatient care, including drugs. The ends, particularly in converting local revenues gener- revenues generated from fees are retained by the ated through user charges into the foreign exchange health centers and managed by local elected commit- needed to purchase imported drugs. In addition, ef- tees. The committees reinvest in additional drugs forts to encourage local private financing of health care (through a revolving fund), in incentive payments for by poor urban and rural households may allow govern- health workers, and in other improvements. The gov- ments to avoid tackling basic reforms of their health ernment and donors assist health centers in purchasing systems, especially the reallocation of public revenues inexpensive generic drugs, thus increasing the cost- from tertiary care hospitals to more basic services. effectiveness of services at the health center. GNP.) Some countries are already moving in this in part through the use of essential national drug direction. Mozambique, for example, is increasing lists, and by purchasing drugs competitively. Nu- government outlays for health in 1992 and 1993 as merous successes have already been recorded. part of a broader program of economic reform, and Bulk procurement of drugs enabled a group of Mauritania is committed to substantial rises in church-run African health associations to save 40 government health spending during 1992-96. percent of their annual drug bill. Similar efforts by Since shifts in domestic budget resources between several Caribbean states led to an average reduc- sectors take several years to implement, donor tion of 44 percent in the price paid for the twenty- funds could play a significant role in increasing five most frequently used drugs. An essential- government health expenditures, including recur- drugs revolving fund for several Central American rent spending, in the early years. The budget re- nations yielded savings of 65 percent of the costs forms in both Mozambique and Mauritania are be- of pharmaceuticals. ing supported by transitional financing from the donor community. COMMUNITY CONTROL AND FINANCING. Com- munity financing, in the form of user charges and REDUCTION OF WASTE AND INEFFICIENCY. There is prepaid insurance schemes, has become a practical substantial scope for reduction of waste and ineffi- necessity in a number of low-income countries. ciency in government health programs, especially But community financing is also a virtuous neces- in drug management. Pharmaceuticals, which ac- sity: it can help to improve the quality and re- count for 10 to 30 percent of public spending for liability of services, in part by making health health in most countries, are the most promising workers more accountable to their clienteles. area for efficiency gains in the short run. Very This is the approach being taken in the Bamako large savings can be achieved by improving the Initiative, sponsored by WHO and UNICEF (Box selection and quantification of drug requirements, 7.1). Recent experience from a number of African 159 countries shows that rural households are pre- of the population receives services financed pared to pay modest charges for drugs in govern- through general tax revenues. Other countries use ment health centers, provided that the quality of social insurance, with part of the population cov- services improves, that fees are retained and uti- ered by mandatory employment-based contribu- lized at the point of service, and that the local tions, usually pooled in a single fund run by a population has a strong voice in the operation of parastatal agency. The share of the population pro- the facility. In Guinea, for example, about half of tected by social insurance varies widely, from less the country's 350 health centers were practicing than 10 percent in the Dominican Republic, Ecua- community financing in 1991. Of these, all the ur- dor, and El Salvador to more than 80 percent in ban-based facilities and a third of the rural clinics Brazil, Costa Rica, and Cuba. Brazil and Chile em- were able to cover their operating expenses with ploy hybrids of private and public insurance. In income from fees. Governments should act cau- Brazil every citizen is legally entitled to services tiously, however. Experience suggests that fees financed from a combination of general revenues substantial enough to cover the full cost of clinical and social security contributions, and social insur- services can discourage utilization by the poor. ance is deducted from the wages of every salaried Under these circumstances, the poor should be worker. Yet more than one-fifth of the country's charged reduced fees or should be exempted from population currently opts for some form of private payment. insurance coverage. Middle-income countries need to focus on at PROBLEMS AND PROSPECTS. Health policy re- least four key areas of policy reform: phasing out forms face formidable obstacles in low-income de- public subsidies to better-off groups; extending in- veloping countries. The health ministry often surance coverage more widely; giving consumers makes only a weak case for a larger share of the a choice of insurer; and encouraging payment (sometimes shrinking) budget. Politicians, doc- methods that control costs. tors, and the urban population exert strong pres- sures for higher spending on tertiary care facilities REDUCTION OF SUBSIDIES TO BETFER-OFF GROUPS. in the major metropolitan areas at the expense of Governments should reduce and eventually elimi- the district health infrastructure. Professional as- nate public subsidies to relatively affluent groups. sociations and trade unions representing doctors This can be done by charging full-cost fees to in- and nurses strongly resist both staff cuts designed sured persons who use government hospitals and to increase nonsalary spending and efforts to re- clinics for services not included in the national es- deploy health workers to rural areas. Despite these sential clinical package and by cutting tax deduc- obstacles, some low-income countries are cur- tions for insurance contributions. In South Africa rently carrying out major health policy reforms. and Zimbabwe privately insured individuals have Malawi, for example, is implementing sweeping been charged less than the full cost of the services changes as part of a World Bank project. It is in- they receive in government health facilities. In ad- creasing the share of the government budget allo- dition, they have been allowed to deduct from tax- cated to health from 7.1 percent in 1991 to 9.1 per- able income part or all of their out-of-pocket pay- cent by 1995, raising the fraction of health ments for health care, as well as their health spending for district health services from 15 to 23 insurance premiums. Employers can also deduct percent, and reducing the share devoted to the their insurance contributions. These measures re- country's three central hospitals from 35 to 25 per- duce the amounts available for financing essential cent. To strengthen the district health system, the services. In South Africa individual tax deductions government is also engaging more than 3,500 new were estimated to be equivalent to 18 percent of lower-level health workers to serve in rural clinics total public sector health expenditures in 1990. In a and communities. Donor funds are being used to recent effort to reverse a similar situation, Zim- help pay for these workers. babwe has sharply limited tax deductions for health care and insurance, raised fees, and inten- Middle-income countries sified efforts to collect fees from privately insured In middle-income developing countries out-of- patients. Government hospitals have learned that pocket payments for health usually account for they can often identify insured patients by offering less than a third of total spending. Some middle- them extra nonmedical amenities, such as private income countries, such as South Africa and Zim- hospital rooms, and can then target them for ag- babwe, have private insurance, even though most gressive cost recovery if they accept. 160 In countries where social insurance covers only government, as in Germany, tend to be more ac- a fraction of the population, governments can in- countable to their members. In a number of Latin crease the extent to which health services are self- American countries monolithic social security "in- financing by eliminating public subsidies to social stitutes" are already heavily discredited because of insurance. These subsidies, which are widespread their past inefficiencies and corruption. Greater in Latin America, mostly benefit the middle classes competition and accountability are two of the main and are therefore regressive. Elimination of the objectives of current proposals for reforming social subsidies would free resources for health services insurance in Argentina. for the poor. Eliminating subsidies also imposes more financial discipline on the social insurance COST CONTAINMENT. Copayment by insured in- agencies, which are often allowed to run deficits dividuals for some services can help to restrain that are later covered by transfers from other social their use of the services but is unlikely to be a very security programs or from the general government powerful cost-containment method. Copayments budget. In Venezuela, for example, the govern- amounting to an average 40 percent of expendi- ment subsidizes contributions to the medical assis- tures in Korea have done little to slow the rate of tance fund within the parastatal social security increase in health spending, which grew from 3.7 agency. Despite this subsidy, in 1990 the fund ran to 6.6 percent of GNP during the 1980s. Similarly, a deficit equivalent to 37 percent of its health the practice, introduced by private U.S. insurers, expenditures. of retrospective reviews of utilization of medical care appears to lead to a modest one-time savings EXTENSION OF INSURANCE. Where the bulk of the in health spending but does not have long-lasting labor force is already employed, government poli- effects on the rate of growth of expenditures. cies that extend insurance coverage to the rest of By contrast, prepayment of health care pro- the populationincluding the self-employed, the viders is a promising approach to containing elderly, and the poorremove the inequities in- health expenditures. Governments could help to herent in multitiered systems of health financing promote such schemes by removing legal barriers and expand the content of the universally available that in many countries prevent the same institu- package of care. When insurance coverage be- tion from acting as both insurer and provider. In comes universal, as in Costa Rica and Korea, sub- South Africa the government recently decided to sidies actually end up targeting the poor and are allow the creation of health maintenance Organiza- thus progressive. But only a few middle-income tions (HMO5), mainly as a way of containing countries that have adequate financial resources, health costs. More than twenty such organizations political resolve, and administrative capacity will have been established in just a few years. They be able to achieve such universal insurance cover- have introduced capitation and negotiated fees, age. Korea's bold initiative to create a national which limit costs more effectively than did the health insurance system from scratch between open-ended fee-for-service payment arrangements 1978 and 1989 and Costa Rica's efforts in the 1980s historically used in South Africa. to universalize a system that had previously cov- Governments can do much to improve the in- ered only the industrial labor force show that this centives created by social insurance. Where the in- is a difficult but achievable goal. Attaining univer- sured use private providers, fee-for-service pay- sal coverage would be more feasible if govern- ment schemes need to be replaced with an ments limited the essential package of insured ser- alternativecapitation or annually negotiated uni- vices to those with high cost-effectiveness. form fees for doctors and hospitals (based on diag- nostic-related groups of procedures, for example) CONSUMER CHOICE. Competition among sup- or preset overall budgets for hospitals. Where so- pliers of a clearly specified prepaid package of cial insurance covers services by government hos- health services would improve quality and encour- pitals, competition with the private sector can im- age efficiency. And even where there is little or no prove performance. Other promising approaches direct competition among insurance funds, as in are to allow government hospitals to compete with Japan and Korea, multiple semi-independent in- one another as semiautonomous enterprises, as in surance institutions may still have advantages the United Kingdom in recent years, and to give over a single large parastatal agency. Local insur- hospital managers financial and career incentives ance funds managed by boards composed of rep- to meet performance targets, as in Chile. resentatives of workers, employers, and local The example of Chile (Box 7.2) illustrates the 161 benefits and perils of health sector reform in a mid- the formerly socialist countries of Eastern Europe dle-income country. Chile has been able to im- and the Soviet Union. Health expenditures were prove efficiency, quality of care, and consumer financed from general revenues. In principle, they choice, but the reforms have also created new were provided free of cost to the population at problems regarding administration, financing, and government clinics and hospitals and at facilities equity. run by state enterprises, but in practice, "infor- Formerly socialist countries mal" payments oiled the wheels of bureaucracy. Today the health systems in these countries are in Historically, the government was responsible for severe crisis. Many doctors and pharmacists are both the finance and the delivery of health care in leaving the government health services to practice Box 7.2 Health sector reforms in Chile Over the past fifteen years Chile has undertaken dra- had few incentives to help supervise municipal matic reforms of its health sector, Its experience shows facilities. that reform is a permanent process, not a one-time ef- Because municipalities were reimbursed for each fort, and that countries undertaking reform must have unit of service delivered, they tended to provide too both the capacity and the political will to review and much high-cost curative care and too few preventive revise health policies continuously. services, which caused costs to explode. The govern- Starting in the late 1970s, Chile (then under a mili- ment then moved to cap allocations to local authorities, tary government) decentralized the government-run using as a basis historical budget shares that favored health system and created private health insurance in- the wealthier municipalities. stitutions. Responsibility for operating primary care The ISAPREs, by targeting the richest segments of services was devolved to the country's 325 munici- Chilean society, impoverished the rest of the social in- palities. The Ministry of Health transferred its primary surance system. Each salaried beneficiary who chose to care budget and about half of its personnel to the mu- shift to an ISAPRE cost the public system 2.5 times the nicipalities, which could also draw for financing on lo- contribution of an average salaried worker. Because the cal tax revenues and on resources from the central gov- ISAPREs are permitted to rate individual health risks, ernment's Municipal Common Fund. More important, they have "skimmed" the population for good risks, the government encouraged the establishment of pri- leaving the public sector to care for the sick and the vately owned and operated health insurance funds, elderly. known as ISAPREs. The roughly 70 percent of the The democratically elected government that came to population covered by social security schemes had the power in 1989 has chosen to maintain the broad thrust option of using their payroll deduction to buy a pre- of the health reforms while seeking ways to overcome paid private health plan. The competing plans were their adverse effects. Municipal elections have been regulated by a new oversight unit (superintendencia) in held to ensure that popularly chosen and accountable the Ministry of Health. By 1990 about 2.5 million peo- officials look after primary health services. Training ple, or 18 percent of the population, were covered by programs have been organized for municipal health thirty-five ISAPREs. officers. Responsibility for hospitals is being decentral- Both decentralization and the creation of the private ized to twenty-seven health service areas that will enter insurers brought about some improvements in the into management contracts with the Ministry of health system. The municipalities expanded primary Health. Finally, under a new proposal, central funds care services. The ISAPREs introduced more competi- would be allocated to the municipalities on a capitation tion and consumer choice into the financing and deliv- basis, with a further adjustment to favor the poorest ery of services and spurred growth in the numbers of localities. private doctors and hospitals. The government is also beginning to look at ways to But the reforms also created new problems. In the reduce inequities in the ISAPRE health financing sys- early years of the reforms, when local officials were tem. The superintendencia that regulates ISAPRE is be- appointed by the military regime, municipal health ser- ing strengthened. It is considering requiring the pri- vices were not responsive to the local population. vate plans to use community risk-rating and to accept Transfers of Ministry of Health staff to the munici- all applicants able to pay the community-rated pre- palities created job insecurity and caused a decline in miums; making it mandatory for all ISAPREs to offer a staff morale. Many municipalities lacked the capacity similar basic medical plan in order to promote direct to plan and manage primary health services. The mu- competition among suppliers (as in the managed care nicipalities tended to overrefer patients to hospitals, systems being developed in the United States); and which were still funded by the ministry. The ministry eliminating the deduction for employer contributions. 162 fee-for-service medicine in the private sector. Since covering two to four provinces, but the provinces real government spending for health has fallen are reluctant to finance such regions. Moreover, dramatically during the recent transition toward a there are political pressures for further decentral- market economy, the government health system is ization to the level of the district governments, also experiencing serious shortages of drugs and where there is now very little capacity for manag- equipment. ing health systems. Largely because they know all too well the prob- At the same time that they decentralize, govern- lems of repressive central government control, ments will have to reduce the size of publicly policymakers, medical professionals, and con- owned health services, which have far too many sumers in the formerly socialist countries are look- hospitals, hospital beds, and physicians. In this ing to systems of public and private insurance in way, governments can free resources for vital pub- industrial countries as possible models for reform. lic health services, including immunization, work- Some countriesfor example, the Czech Republic, place and food safety, environmental regulation, Hungary, and Polandhave much in common measures such as education and higher taxes to with upper-middle-income countries such as Ar- discourage consumption of alcohol and tobacco, gentina, Costa Rica, and Korea. They may be also and quality control of privately delivered clinical able to adapt some features of the systems of the care. The clinical and managerial skills of the re- Nordic countries and the United Kingdom, which maining government health personnel need to be are financed from general revenues, or of the uni- substantially upgraded and reoriented from the versal social insurance approaches of Germany previous system of centralized bureaucratic con- and Japan. Others in this groupincluding the trol toward the emerging system of semi- relatively poor Central Asian republicsface many autonomous health facilities. of the same issues currently confronting lower- middle-income and even low-income countries, NEw MODES OF FINANCING. The examples of such as Pakistan and Yemen. other countries could help the formerly socialist Despite this diversity, the governments of all the countries establish insurance systems that pre- formerly socialist countries need to consider health serve the main virtue of their old systemwide- sector reforms in at least three main areas: improv- spread coverage of the population. It could also ing the efficiency of government health facilities help them to recognize the circumstances under and services, partly by reducing the size of the which general government revenues can play a public system; finding new ways to finance health positive role, as the dominant source of funding care; and encouraging private supply of health (the pattern in the United Kingdom) or as a com- services while strengthening public regulatory plement to insurance (as in Japan). Experience capacity. elsewhere offers important lessons on how to cre- ate financing systems that are sustainable and that EFFICIENCY OF GOVERNMENT SERVICES. De- contain costs by, for example, discouraging fee- centralization of government health services is po- for-service compensation. The formerly socialist tentially the most important force for improving countries will also want to avoid the large and in- efficiency and responding to local health condi- equitable government subsidies commonly pro- tions and demands. It will be successful only when vided to private insurance for the wealthy or to local government health agencies and hospitals social insurance for the middle class. have a sound financial base, solid administrative Most formerly socialist countries are already on capacity, and incentives for improving efficiency the road to reform. The Czech and Slovak repub- and when they are accountable to patients and lo- lics and Hungary are experimenting with forms of cal citizens. Extreme and hasty decentralization social insurance. Because the Czech system in- can create inefficiencies. In Poland, for example, cluded a very comprehensive package of health the government has decentralized health careto benefits and paid private doctors on a fee-for-ser- the level of the country's forty-nine provinces. The vice basis, it encountered serious financial diff i- average provincial population of less than a mil- culties after just a few months of operation. Under lion is proving too small to make efficient use of the recently revised Hungarian health-financing the tertiary care hospitals being built in each pro- system, public sector doctors will be salaried em- vince, and the available medical personnel are be- ployees of the central and local governments, and ing spread too thin. For these reasons, the govern- private general practitioners will be paid on a cap- ment is now experimenting with health regions itation basis. Russia and Ukraine are also prepar- 163 Box 7.3 Reform of the Russian health system Before the political upheavals of 1990-91 that led to the They will sign contracts for care with public and private breakup of the Soviet Union, the 3 to 4 percent of GNP providers. Individuals can then voluntarily purchase that the Russian republic spent on health care for its supplementary private insurance to cover additional nearly 150 million inhabitants was financed from gen- health services. eral government revenues and delivered through a The health insurance legislation has been in effect vast network of public facilities, programs, and em- since late 1991, but progress in implementing it has ployees. This highly centralized and bureaucratic sys- been slow. Some important issues in the design of the tem led to excessive numbers of doctors and hospitals. system still need to be resolved. These include the role It gave few incentives for efficiency or for providing and extent of competition among public and private quality care, and it neglected the preventive measures insurers; whether risks are to be rated on an individual needed to combat the country's most serious environ- basis or across larger pools of individuals; and how the mental and behavioral problems: industrial pollution, insurance funds will pay providerson a fee-for-ser- alcohol and tobacco dependency, and poor nutrition. vice basis, through capitation, or by some other Consequently, the health status of Russians stagnated method or combination of methods. during the 1970s and 1980s. In 1990 life expectancy for The practical obstacles to the implementation of the Russian men was just sixty-four years, a full ten years new system are formidable, partly because of the un- less than in Western Europe, and the infant mortality settled administrative and economic environment. The rate, at twenty-two per 1,000 live births, was twice the regional governments lack the capacity to manage and Western European average. regulate the health system they are inheriting. The The new Russian government has pursued several economy and the government budget are under severe fundamental reforms of the old Soviet health system. strain. Real wages have fallen dramatically in the past Health financing and management are being decentral- few years. The costs of drugs and equipment have in- ized to eighty-eight regions. Much medical practice is creased faster than inflation, leading to serious short- being privatized, and a recent health insurance law ages. Payroll taxes to cover employee benefits already provides for the introduction and regulation of new absorb 38 percent of wages, making it difficult to fi- forms of insurance. Under the law and its proposed nance an affordable package of health services through amendments, each region is to have a social insurance the social insurance system. To help overcome these fund, and a national fund will equalize resources problems, a number of international agencies, includ- across regions. These insurance funds will receive a ing the World Bank, are working closely with Russian combination of compulsory payroll deductions and health officials on designing and carrying out health budget transfers from general government revenues. policy reforms. ing to implement mixed systems of social insur- sion to issue lifetime licenses to doctors without ance and general revenue financing. Box 7.3 establishing strict standards of practice or recer- describes the current efforts in Russia. tification requirements. Since government regula- tory capacity is likely to be weak in the next few COMPETITIVE PROVISION AND PUBLIC REGULATION. years, health system reforms should be designed Although private medical practice is now permit- in ways that minimize the need for direct govern- ted in most of the formerly socialist countries, the ment regulation. Encouraging self-regulation legal and regulatory environment for private doc- through associations of private medical schools, tors, hospitals, and insurance institutions is often doctors, and hospitals would be one such ap- either nonexistent or hostile. With large numbers proach. In the long run, better regulation will re- of private doctors establishing practices and pri- quire both training of government inspectors and vate hospitals and clinics being created, regulation other regulatory personnel and development of of providers will be critical for reducing the inci- government institutions such as medical licensing dence of medical malpractice and financial fraud: boards and national and local medical ethics It is also essential that regulation encourage the committees. development of efficient institutions, such as health maintenance organizations, for financing Directions and prospects for reform and providing clinical care for the bulk of the population. Already there are signs of poorly con- The world's diversity of health care systems is ceived regulations, such as Romania's recent deci- matched by the diversity of reform movements. 164 But several common themes are beginning to Nigeria, and South Africa and in Eastern Europe. emerge. First, governments are increasingly recog- A free press is important, as are consumer advo- nizing the centrality of their own role in public cacy groups, for conveying a diversity of views on healthfor example, in achieving the enormous health reform and for stimulating debate. global gains in immunization coverage. Second, In many countries, maintaining the support of governments are exploring ways to introduce the middle class and of urban groups for health more competition and foster a diversity of public policy reformsincluding the reallocation of pub- and private institutions in the delivery of clinical lic spending from tertiary care to basic public services. Third, governments are examining new health and clinical care for the poorwill require a approaches to finance and insurance, including se- gradual shift in resources rather than wholesale lective user fees in the public sector, systems that changes in just one or two years. For this reason, discourage third-party reimbursement, systems universal government financing (or government- that mix finance from compulsory social insurance mandated financing) for a nationally defined es- and from general tax revenues, and systems that sential package of services will often be more suc- set fixed budgets for each patient or each case. cessful than a highly targeted approach that may Everywhere, health sector reform is a contin- undermine the political base for reform. Similarly, uous and complex struggle. Neither governments continued government ownership of some hospi- nor free markets can by themselves allocate re- tals that offer high-quality tertiary care, with a sources for health efficiently. As policymakers try phased reduction in public subsidies to the to reach compromises, they must deal with power- wealthy for this care, may be more feasible politi- ful interest groups (private doctors, drug com- cally than rapid divestiture to the private sector. panies, medical equipment manufacturers, and External financial assistance can help countries insurers) and strong political constituencies, in- handle these politically difficult tradeoffs and can cluding urban dwellers and industrial workers. ease the process of policy change. Strategies for overcoming these obstacles to health sector reform will vary from country to International assistance for health country, but some common approaches are dis- cernible. Political leadership, beginning with the After growing rapidly in the 1970s, aid for health head of state, is an indispensable element in re- stagnated during the 1980s. As a share of official form programs almost everywhere. The 1990 development assistance, aid for health declined World Summit for Children proved an effective from an average of 7 percent for the period 198 1-85 means for engaging the attention and commitment to 6 percent during 1986-90. Total aid flows to the of heads of state (see Box 2 in the Overview). Se- health sector in 1990 were $4.8 billionalmost $4 nior officials of ministries of health can be strongly billion in official development assistance and $0.8 influenced by the prevailing views of the interna- billion from NGOs and foundations (Figure 7.1). tional health community, particularly those of This amounts to about one dollar per person in WHO and other major donors, and by participa- developing countries. (The figure for official devel- tion in international meetings and seminars on opment assistance is based on reports from donor health policy and management. governments. Only $3.3 billion of the $4 billion can Professional associations may be able to bring be accounted for as receipts by individual coun- about some reorientation of health workers, espe- tries; this is the amount that appears in the total cially physicians. Appeals to the sense of social health expenditure estimated in Chapter 3 and in responsibility of these associations have helped Appendix table A.9.) Bilateral agencies accounted advance agendas for preventive health in the for the largest share (40 percent), followed by United States and elsewhere. Such groups, how- United Nations agencies (33 percent), NGOs (17 ever, are often the sources of the strongest resis- percent), development banks (8 percent), and tance to change. Reshaping the training curricula foundations (2 percent). of medical and nursing schools to include a greater The trend is for donors to provide aid for health emphasis on public health and general practice is through multilateral channels. The share of multi- likely to be a more effective way to enlist the sup- lateral assistance has grown from 25 percent in port of physicians and nurses. 1980 to 40 percent in 1990 and is likely to exceed 50 Public opinion can be a powerful force for health percent by 1995. As a result of the quadrupling of reform, not only in industrial countries but also in World Bank lending for health over the past six developing countries such as Brazil, Chile, years, disbursements of Bank funds are expected 165 External assistance to developing countries for health comes from many sources, public and private. Figure 7.1 Disbursements of external assistance for the health sector, 1990 (millions of dollars) N 1,913 I 3,925 / Bilateral agencies 4,794 / Public agencies 382 242 I 671 / 4,794 Development > banks" N Donor 869 1,601 382 / Developing countries COO 0 tries United 58(1 Private Nations agencies agencies 830 Nongovernmental organizations (1 N 7 68 -J K Foundations a. Includes $84 million in nonconcessional loans. Source: Michaud and Murray, background paper. to grow from about $350 million in 1992 to about $1 Programme (UNDP), in its annual Human Develop- billion in 1995, making the World Bank the largest ment Report, has argued for more donor spending single source of external funding for health. Since on health, and in its recent World Development Re- the portion of aid going to middle-income coun- ports the World Bank has made similar recommen- tries from the World Bank and other development dations. World Development Report 1990 proposed a banks is nonconcessional lending, some of the 3 percent annual increase in aid during the 1990s, projected increase in lending for health will in- to be targeted at poverty-reducing activities, in- volve a hardening of terms. It would be desirable cluding basic health care. The donor community for bilateral grant-funding agencies and conces- needs to review these goals and targets in light of sional arms of the development banks (such as the the actual trends in aid flows for health. World Bank's International Development Associa- The share of aid going to health should be re- tion) to increase their assistance to health as well. stored immediately to its earlier level of 7 percent The amount of health aid has stagnated, and its of total official development assistance and should share in total development assistance has de- rise substantially over the next five years. Such an clined, even as donors continue to express concern increase would have a significant impact on the about health. Over the past ten years the United health status of the poor, particularly if it is di- Nations and other international agencies have rected toward the transitional costs of reallocating called for increased investments in the develop- government spending to public health measures ment of human resources, including health, both and essential clinical care and to seriously under- by developing countries themselves and by the do- funded disease control efforts such as those for nor community. The United Nations Development tuberculosis and AIDS. A rise in donor assistance 166 of $2 billion, for example, could finance a quarter large share of health expenditure. In Africa aid of the estimated additional costs of a basic package makes up an average 10 percent of national health in low-income countries and of strengthened ef- spending (Table 7.2), or 20 percent if South Africa forts to prevent AIDS. Such an increase, which is excluded. Aid covers more than half of all health would boost from 6 to 9 percent the share of total expenditures in countries such as Burkina Faso, official aid going to health, would be feasible if Chad, Guinea-Bissau, Mozambique, and Tan- other donors matched the rise in World Bank dis- zania. In these countries donors finance an impor- bursements for health that is expected to occur in tant share of recurrent costs, as well as investment coming years. It would also be consistent with the items. In Mozambique, for example, aid accounted proposal in the UNDP's Human Development Report for more than half of recurrent spending in 1991 1993 (also endorsed by UNICEF) that 20 percent of and for 90 percent of capital expenditures for aid be spent on health, education, water and sani- health. Even when aid amounts to 2 percent or less tation, and environmental protection for the of total health spending, as in the other develop- world's poor. ing regions, improvements in its use would still be There are a number of ways, in addition to the an important catalyst for reform. traditional annual and multiyear programming of General lessons on improving aid effectiveness aid by individual donors, for the international apply equally to the health sector (Box 7.4). Do- community to mobilize more financial resources nors need to set their priorities carefully and allo- for health. Coordinated sectorwide pledging at cate their resources in accordance with these prior- consultative group meetings and donor round- ities. The productivity of aid would increase tables has been used successfully in countries such substantially if donors were to direct more of their as Tanzania and Zambia. Another approach is pro- assistance to public health measures and essential gram-specific pledging, as illustrated by the clinical services, especially in low-income coun- dozens of national AIDS-control donor meetings tries. They might also usefully focus on capacity chaired by WHO in recent years. The role of debt- building, research, and reform of health policy. for-development swaps as a means of generating Countries that show a willingness to improve ac- extra resources for both government and NGO- cess to health services for the poor and to under- provided health services should be assessed in this take reforms of the health system should be strong context. Ecuador, Sudan, and Zimbabwe have al- candidates for aid. ready carried out swaps, and Nigeria is exploring a The World Bank increasingly stresses policy re- major swap of its debt currently held by donors in form in its lending for health, which has grown return for increased public spending for essential nearly fourfold in recent years (Box 7.5). For some health services. donors, adjustment of priorities would mean spending less on hospitals, sophisticated medical Improving the effectiveness of aid for health equipment, and training for medical specialists It is crucial that the donor community and devel- During 1988-90 Japan spent more than 33 percent oping countries focus on ways to improve the ef- of its bilateral assistance for health on construction fectiveness of existing and future assistance to the of hospitals, France spent 25 percent, and Ger- health sector, particularly in the low-income coun- many and Italy spent nearly 15 percent each. tries where donor assistance already accounts for a Within the domain of public health and essential clinical care, several areas of intervention deserve greater attention from donors, including tuber- Table 7.2 Official development assistance culosis control, the EPI Plus program, micro- for health by demographic region, 1990 nutrient supplementation, AIDS prevention and Health aid control, and programs to reduce tobacco consump- Health aid as a tion. These problems impose a large burden of ill- received Health aid percentage (millions per capita of health ness, in some cases because rapid growth of the Region of dollars) (dollars) expend?ture threat has gone unrecognized. Their control offers Sub-Saharan Africa 1,251 2.45 10.4 large externalities or economies of scale. Often, so- Other Asia and islands 594 0.87 1.4 lutions will require a global effort. Latin America and the The efficiency of aid for health can be greatly Caribbean 591 1.33 1.3 Middle Eastern crescent 453 1.31 1.3 enhanced through better coordination of donor India 286 0.34 1.6 projects and policies. Fragmentation of external China 77 0.07 0.6 support in the health sector is a long-standing Source: Michaud and Murray, background paper. problem in many countries and imposes a heavy 167 r Box 7.4 Health assistance and the effectiveness of aid Recent evaluations of the effectiveness of aid, includ- ronment. When it comes to coordination, both sides ing a classic 1986 study commissioned by the world have been at fault. Donors have pursued their OWfl donor community, point toward the same conclusion: objectives without attempting to ensure that their aid most aid has been successful, but a considerable share, complements that of others. And all too often, aid re- perhaps a third or more, has been much less so, and a cipients have played one donor off against another, small percentage has failed completely or has even while ministers and ministries have focused on their been harmful. These broad-brush averages hide signifi- own concerns rather than looking to the national good. cant regional differences: in Asia and Latin America Aid for health has generally had a good technical performance has been better; in Sub-Saharan Africa it record. It has fit in well with development priorities, has been worse. Aid has been least effective in the especially in recent years, as the concentration on hos- poorest countries, where success is most needed. pitals and high-technology curative medicine has been The reasons for inferior performance lie with both replaced by an emphasis on primary and preventive donors and recipients. Poor countries and those experi- care. There have also been major successes_mainly encing political conflict and instability constitute a diffi- highly focused initiatives such as the program for the cult environment for aid, as they have little administra- eradication of smallpox, the drive against child mortal- tive capacity or infrastructure. But these difficulties ity, and the effort to control river blindness in Africa. have in many cases been compounded by unfortunate What is still lacking is the ability of the aid system to policies. Aid projects have been poorly designed, both help set in place and sustain locally appropriate public technically and because of inadequate understanding health programs and essential clinical services. of the human, social, institutional, and political envi- burden on already overextended government offi- strengthening the public institutions that finance cials. In the extreme, fragmentation can lead to and deliver health services, both through broad conflicting policies being put into effect. Recently civil service reform and through changes within in one West African country, for example, three the health sector. Donors can play an important different cost recovery policies, each sponsored by role in these areas by supporting decentralization a different donor agency, were being applied in and other organizational reforms and by assisting separate regions of the country. The dangers of the groups that formulate national health policies. fragmentation are especially great in poor coun- Additional support is required for initiatives such tries where different donors choose to focus their as the foundation-backed International Health Pol- health sector activities on different provinces or icy Program and for bilateral projects to train districts and either lose sight of or undermine the health planners and managers, economists, and formulation of national policies. sociologists. Much can be done to improve donor coordina- International programs for research and development in tion, globally and regionally, but especially at the health country level. Donors can agree with countries on overall national health and assistance strategies. Investments in health research and development This is especially effective when the government have yielded high returns in better health. For ex- takes the lead in planning and in coordinating the ample, the programs for tropical disease research donors, as has happened recently in Zimbabwe. and human reproduction funded by donors and Another approach is for donors to form large con- executed by WHO have produced a number of sortia to fund national programs, as in the case of new or improved drugs and diagnostic tests and maternal and child health and family planning in have strengthened research capacity in developing Bangladesh. (The experiences of these two coun- countries. Yet according to the 1990 report of the tries are reviewed in Box 7.6.) At a minimum, do- Commission on Health Research for Develop- nors should create informal local groups that meet ment, only 5 percent of global expenditures on periodically to review progress and problems in health research are directed at the health problems the health sector, as in Mozambique and Senegal. unique to developing countries, and less than 10 The efficiency with which aid for health is spent percent of donor assistance for health is devoted to depends critically on building local capacity to research, both biomedical and in the social plan and manage health systems. This requires sciences. 168 The commission identified several serious defi- compounded by donors' limited capacity to stay ciencies in the international health research and abreast of the latest research proposals and to development system. The expertise of the global assess the relative priorities for funding this pharmaceutical industry is not being adequately research. applied to the development of drugs and vaccines To help stabilize funding, to improve the setting that could reduce the toll of early childhood dis- of priorities, and to boost efficiency, developing eases. Technology assessment is weak, as is the countries, donors, and scientists should consider health policy research needed to determine more the development of a global mechanism for better equitable and efficient ways to finance and deliver coordination of international health research. A health services. Most important, the commission number of institutional arrangements are possible, noted, local research capacity in developing coun- including well-defined networks of research cen- tries is woefully inadequate. A number of promis- ters, informal consultative bodies, and large global ing research efforts, including the Children's Vac- funds that pool donor assistance. Examples of cine Initiative and programs to deal with acute these institutional arrangements in other sectors, respiratory infections, tuberculosis, micronutrient such as the Consultative Group for International deficiencies and worm infections, suffer from Agricultural Research and the Global Environment weak and uncertain donor funding. In general, the Facility, may provide models for improving the co- problems of constrained funding for research are ordination of international health research. Box 7.5 World Bank support for reform of the health sector World Bank support for the health sector has grown hensive reforms, including granting greater manage- dramatically over the past six years. The number of ment autonomy to health facilities and decentralizing new World Bank-financed health, population, and nu- resources to the regional level. Doctors, nurses, and trition projects approved each year increased from an other health personnel are being encouraged to work average of eight during fiscal 1987-89 to twenty-one in better-equipped health centers and other basic facili- during fiscal 1990-92, and the value of credits and ties. And health-financing mechanisms are being re- loans committed each year rose from $317 million to vised, with updated fee schedules, new exemption $1,151 million over the same period. As of June 1992, procedures for the poor, and changes to the health eighty-one Bank-financed health projects were being benefits covered by existing insurance schemes. The implemented. As a share of new World Bank lending, Hospital Restructuring Project, supported by the projects for health, population, and nutrition grew World Bank, is assisting the improvement of manage- from less than 1 percent in 1987 to nearly 7 percent in ment systems and the quality of health services in the 1991. largest government hospitals, which were recently Whereas most of these projects continue the Bank's granted autonomous legal status. The project dovetails traditional support for basic health servicesincluding with the concurrent World Bank-financed Population district health infrastructure and personnel, maternal and Family Health Project, designed to improve the and child health, and control of infectious diseases quality and efficiency of public health services and es- World Bank lending for health is increasingly focusing sential clinical care, especially for mothers and chil- on broad policy reforms in the health sector. For exam- dren. It is expected that better basic services at the ple, in connection with a recent Bank project, the gov- health center level will reduce the demand for hospital ernment of Maurifania has developed a financing plan care, thus slowing the expansion of the country's to improve the availability of basic health services for hospitals. its widely dispersed population. The share of the gen- The Rumania Health Rehabilitation Project supports eral recurrent budget going to the Ministry of Public government efforts to diversify sources of health fi- Health will increase from 5.5 percent in 1992 to 7.5 nancing and thus to reduce dependence on the public percent in 1996. The project is introducing commurtity- budget, which is under pressure because of weak and based cost recovery in three of the country's thirteen unstable macroeconomic conditions and rising health regions as a way of improving the efficiency and qual- care costs. The government is pilot testing decentral- ity of services. Revenues are being raised mainly ization of health sector policymaking, planning, man- through the sale of drugs, organized and managed by agement, and evaluation in three subregions. It is also local health communities. discussing how to create a legal and regulatory envi- In Tunisia the government is carrying out compre- ronment to support reform of health financing. 169 Box 7.6 Donor coordination in the health sector in Zimbabwe and Bangladesh The Zimbabwe Second Family Health Project (1992-96) review and the administrative burden on the is the culmination of a long period of interaction be- government. tween the government of Zimbabwe, the World Bank, In Bangladesh the World Bank and ten bilateral agen- and other multilateral and bilateral donors. The $120 cies together are contributing $440 million to the Fourth million project, which supports the government's five- Population and Health Project, and the government is year investment program for population, health, and providing $165 million, for a total of $605 million over nutrition, is designed to benefit directly low-income the five-year project period. The United Nations Popu- households, especially women and children. Zim- lation Fund, WHO, and UNICEF are supplying project babwe has entered a period of economic adjustment management, procurement, and technical assistance. that will necessitate spending Cuts. The project will All the partners in the project belong to the Bangladesh help protect poor and vulnerable households from Population and Health Consortium, which has some adverse effects of adjustment by mobilizing addi- emerged as an important collective force in the health tional resources for human resource development and sector. The Asian Development Bank has joined the by improving the equity and efficiency of spending. consortium with a view to ensuring consistency be- Zimbabwean participation was emphasized from the tween the project and its own $60 million investment in project's beginning. The project preparation committee population and health in Bangladesh. included representatives from various central govern- During project formulation the government and the ment departments, provincial governments, and the donors held several workshops in Dhaka and a special Zimbabwe National Family Planning Council. The conference in Geneva. These workshops were instru- committee set planning guidelines, including a prelimi- mental in forging a consensus on population and nary outline of project components, costs, and financ- health strategies. The consortium approach enables the ing. Using these guidelines, proposals were prepared government and the donor community to agree on an locally and were then reviewed by the committee and overall strategy and to work out a consistent financing by interested donors. plan for the sector. The consortium operates on the Virtually all the major donors to the health sector in basis of strict equality of all the partners, independent Zimbabwe helped with planning the project by assign- of the size of their financial contributions. ing agency officials and technical specialists to the do- In addition to strengthening Bangladesh's popula- nor team that advised the government on design is- tion program and its delivery system for family plan- sues. In the end, the project received financing from ning services, the consortium is attempting to reorient Denmark, the European Community, Norway, the health care system toward public health, including Sweden, the United Kingdom, and the World Bank. maternal health. It is also trying to make basic services Donors monitor project implementation jointly rather more easily accessible to the rural and urban poor. than separately, reducing the time needed for donor Meeting the challenges of health policy reform Policymakers in developing countries and off i- cials of the international donor community face a If policymakers are to accelerate the substantial number of difficult challenges in pursuing this health gains of recent decades, especially for the agenda. The changing demographic profile of the poor in developing countries, the agenda for re- developing world, including the aging of the form is clear. It includes increasing overall rates of population, is creating new patterns of disease. economic growth and expanding basic schooling, Emerging microbial threats, such as AIDS and particularly for girls; reallocating government drug-resistant strains of tuberculosis and malaria, spending for health from tertiary care and special- call for changes in personal behavior, new drugs, ist training to public health measures and essential and new ways of delivering services effectively. clinical services; encouraging more diversity and In virtually every country interest groups will competition in the provision of clinical care and resist health policy reforms of the kind suggested the development of cost-containing approaches to in this Report. Health workers will object to insurance; increasing the efficiency of government changes that threaten their job security, income health services; and fostering greater involvement levels, and degree of professional autonomy. Drug of communities and households in promoting companies, medical equipment manufacturers, healthier behavior on their own part and in man- and other suppliers will try to block policies that aging their local health services. they see as having an adverse effect on their mar- 170 kets, revenues, and profits. Political and economic about reallocating resources, improving access, elites and organized labor groups will seek to pre- and increasing efficiency. To do this, higher and serve existing public subsidies for insurance and sustained rates of macroeconomic growth are re- health services from which they benefit and to quired. In many cases countries will also need to maintain their privileged access to clinical care. enact fundamental political reforms designed to Beyond this, policymakers will have to wrestle increase participation and to improve the account- with the reality that in the area of health there is no ability of governments for their health spending, simple paradigm for policy choice. Free markets service delivery, and regulatory performance. for public health activities and clinical care often The donor community has a major responsibility fail, and when governments intervene in financing to back up with concrete actions its verbal commit- and delivery, as they frequently do, they can fail ment to poverty reduction and to investment in just as badly. Effective government regulation of health and human resources. In particular, donors private suppliers of health services and inputs, should do more to support the formulation of im- combined with public financing of cost-effective proved health policies and more effective health packages of public health and essential clinical ser- sector reform programs in developing countries. vices, is needed to deal with these failures. But this As suggested in this Report, they can do this by in turn requires strong private and public institu- financing some of the transitional costs of real- tionsand institutional capacity is seriously lack- locating government budgets to public health mea- ing in many developing countries. sures and essential clinical care, by building local Despite these obstacles, there have been a num- planning, management, and research capacity, ber of successes in specific intervention programs and by providing sound assessments of the world- such as polio eradication and river blindness con- wide experience with the cost-effectiveness of in- trol and a smaller but still important number of terventions and with reform of systems. successes in broader health sector reform in such If developing country governments and donors countries as Chile, Tunisia, and Zimbabwe, as well accept the challenges and embrace the key health as in many OECD countries. These successes now policy reforms outlined above, improvements in need to be multiplied, especially in the area of sec- human welfare in the coming years will be enor- tor reform, if countries are to address the acute mous. A large share of the current burden of dis- weaknesses in existing institutional structures and easeperhaps as much as one-quarterwill be to lay the foundation for major improvements in prevented. And people around the world, espe- future living standards. cially the more than 1 billion people now living in Developing country governments need to do poverty, will live longer, healthier, and more pro- more to translate into practice today's rhetoric ductive lives. 171 Acknowledgments This Report benefited greatly from ideas, technical Adolfo MartInez-Palomo, David N. Nabarro, Gen inputs, and critical review from a broad range of indi- Ohi, Richard Peto, Kenneth I. Shine, Pravin Visaria, viduals and organizations. Contributions to specific and Richard Zeckhauser chapters are acknowledged in the Bibliographical note. In addition, valuable input was provided Consultations through four other mechanisms: a World Health Or- Ministerial Review of Health Transition Issues ganization Steering Committee, an Advisory Com- June 22-26, 1992, Bellagio Study and Conference mittee, a series of consultations on specific subjects, Center, Villa Serbelloni, Bellagio, Italy. Partial finan- and a series of seminars, mostly held at the World cial support for the consultation was provided by the Bank. Those from outside the World Bank who con- Rockefeller Foundation. tributed to the Report through these mechanisms are listed below. Participants: Alfredo R. A. Bengzon, Demissie Habte (provided written comments), Richard C. A. Fea- World Health Organization Steering Committee chem, Julio Frenk, Mamdouh Cabr, Scott Halstead, Jean-Paul Jardel, Jorge Jiménez de la Jara, Jeffrey R. This committee provided the focal point for the major Koplan, Marthini K. Budi Salyo, Adolfo MartInez- contributions from WHO to the World Development Re- Palomo, Piotr Mierzewski, Rajiv L. Misra, W. Henry port, for WHO's participation in consultations, and Mosley, Samuel Ofosu-Amaah, Raphael Owor, for its critical review of various drafts. Olikoye Ransome-Kuti, and Leonardo Santos Simão Chair: Jean-Paul Jardel Interventions for Nervous System Disorders Members: Andrew L. Creese, Michel Jancloes, Yuji July 6-7, 1992, Pan American Health Organization, Kawaguchi, R. Srinivasan, and Muthu Subramanian Washington, D.C. Chair: Benedetto Saraceno Advisory Committee Participants: Antonio Campino, Vijay Chandra, Brian The Advisory Committee met on October 7-9, 1992, Cooper, Bulent Coskun, Marcelo E. Cruz, Mary Jane at the World Bank in Washington, D.C.; in addition, England, John T. Farrar, Jefferson Fernandes, R. Juan committee members later provided critical review of Ramon de la Fuente, Walter Gulbinat, Itzhak Levav, an early draft of the Report. Thomas McGuire, Kay Redfield Jamison, Norman Sartorius, Carole Siegel, Donald H. Silverberg, T. Chair: Richard C. A. Feachem Takayanagi, and Richard Jed Wyatt Members: Jane C. Baltazar, José Barzellato, Mayra International Aid Flows to the Health Sector Buvini, Lincoln C. Chen, Antoine Degrémont, Nicholas Eberstadt, John Evans, Mahbub ul Haq, Pe- August 27, 1992, Harvard Center for Population and ter Heller, Abraham Horwitz, Jean-Paul Jardel, Rich- Development Studies, Cambridge, Massachusetts ard Jolly, Somkid Kaewsonthi, Pangu Kasa-Asila, Chair: David Bell 172 Participants: Lincoln C. Chen, Nick Drager, Ramesh Participants: Antoine Degremont, Maria Elena Ducci, Govindaraj, Eva Jespersen, Catherine Michaud, Lilia Durán Gonzales, Paul Garner, Greg Goldstein, Christopher J. L. Murray, and David Parker Emile Jeannée, Matthias Kerker, Peter Kilima, Nicolaus Lorenz, Ngudup Paijor, Voahangy Investing in Health Research Ramahatafandry, Allessandro Rossi-Espagnet, John September 16, 1992, World Health Organization, Ge- Seager, Gustavo A. Torres, and Charles Yesudian neva, Switzerland. Financial support was provided Human Resources for Health by the WHO/UNDP/World Bank Special Programme for Research and Training in Tropical Diseases. November 18-19, 1992, McMaster University, Ham- ilton, Ontario, Canada. Financial support was pro- Chair: Carlos Morel vided by the Canadian International Development Participants: Eusebe Alihouno, Barry R. Bloom, David Agency. J. Bradley, Barbro Carlsson, Gelia T. Castillo, Jac- Co-chairs: Julio Frenk and Victor Neufeld queline Cattani, Anthony Cerami, Joseph A. Cook, Henry Danielsson, Ronald W. Davis, Peter de Raadt, Participants: Orvill Adams, Barbara Carpio, Gilles David Evans, Esmat Ezzat, Tore Godal, Melba Dussauld, John Evans, Alfonso MejIa, Hiroshi Na- Gomes, E. Robert Greenberg, Win E. Gutteridge, katani, Kenneth Ojo, Una Reid, Charas Suwanwela, Ralph H. Henderson, H. Robert Horvitz, Nathan K. and Peter Tugwell Kere, Veronique Lawson, David N. Nabarro, Richard Health of the Elderly Peto, Vulimiri Ramalingaswami, Peter Reeve, Hans Remme, Lateef Akinola Salako, Ebrahim M. Samba, November 23-24, 1992, Voksenasen, Norway. The Vladimir Petrovich Sergiev, Carol Viassoff, Gabisiu consultation was organized by the London School of A. Williams, and Richard Wilson Hygiene and Tropical Medicine. Financial support was provided by the Norwegian Ministry of Foreign Resource Flows to the Health Sector Affairs. September 21, 1992, Harvard Center for Population Chair: Alex Kalache and Development Studies, Cambridge, Massachusetts Participants: Jordi Alonso, Nana Apt, Chris Beer, Felix Chair: William Hsiao Bermejo, Ruth Bonita, Carol Brayne, Paul Chen, Participants: Sudhir Anand, Peter Berman, Mirnal Yolande Coombes, Xianglin Du, Denise Eldemire, J. Dutta Choudray, Gnanaraj Chellaraj, Lincoln C. Grimley Evans, Richard G. A. Feachem, Joe Hamp- Chen, Michel Cichon, Andrew Creese, Ramesh son, Hana Hermanova, Benedicte Ingstad, Zhang Govindaraj, Catherine Michaud, Christopher J. L. Kaiti, Roberto Kaplan, Luis Ramos, Melba Sanchez- Murray, Sudhakar Rao, and George Scheiber Ayéndez, Bela Shah, Alberto Spagnoli, Knight Steel, and Renato Veras AIDS Policy District Health Services November 5-6, 1992, Rockefeller Foundation, New York. Financial support was provided by the Rock- November 24-27, 1992, Institute of Health and Devel- efeller Foundation and the Danish International De- opment, University of Dakar, Senegal. Financial sup- velopment Agency. port was provided by Swiss Development Coopera- tion and the Swiss Tropical Institute. Chair: Robert S. Lawrence Co-chairs: Antoine Degremont and Ibrahima Wone Participants: Roy Anderson, José Barzellato, Seth Berkley, Robert Black, Kevin De Cock, Richard G. A. Participants: Abdel Wahed Abassi, Waya Amoula, Feachem, Penelope Hitchcock, King Holmes, Robert Anarfi Asamoa-Baah, Hubert Balique, Wolfgang E. Howells, Jane Hughes, Peter Lamptey, Jonathan Bichmann, Malang Coly, Christian Darras, Pierre Mann, Michael Merson, Daan Mulder, Peter Piot, Daveloose, Annemarie Demazy, Issakha Diallo, Isseu Peer Sieben, Werasit Sittirai, Judith Wasserheit, Fer- Diop-Touré, Gina Etheridge, Georges Fournier, Lucy nando Zacharias, and Richard Zeckhauser Gilson, Kathia Janovsky, Emile Jeannée, Pangu Kasa- Asila, Matthias Kerker, Vincent Litt, Mandiaye Urban Health Loum, Javier Martinez, Sigrun Mogedal, Maty Cissé Samb Ndao, Sène Touré Ngone, Cornelius Oepen, November 9-11, 1992, Basel, Switzerland. Financial Bakary Sambou, Lamine Cissé Sarr, Malick Sarr, support was provided by Swiss Development Coop- Peter Schubarth, Michael Singleton, Thierno Maine eration and the Swiss Tropical Institute. Aby Sy, Al Hadji Ali Tahirou, James Tumwine, Jean- Co-chairs: Marcel Tanner and Trudy Harpham Pierre Unger, Adamou Yada, and Alfredo Zurita 173 Child Health Romer, Elizabeth Sherwin, Peter Smith, Jan Stjern- November 30-December 2, 1992, Baltimore, Mary- swàrd, Rand Stoneburner, Muthu Subramanian, land. The consultation was organized by the Johns Carole Torel, and Godfrey Walker Hopkins School of Hygiene and Public Health. Finan- Health Finance cial support was provided by the U.S. Agency for International Development. Dkember 14-16, 1992, Montebello, Quebec, Canada. Financial support was provided by the Canadian In- Chair: W. Henry Mosley ternational Development Agency. Participants: Fernando Barros, Al Bartlett, Mark Bel- Chair: Stephen Simon sey, Seth Berkley, Robert E. Black, David Boyd, Donald A. P. Bundy, Carlos C. Campbell, Dennis Participants: Nicholas Barr, David Bell, Ricardo Bitran, Carroll, Robert Clay, Felicity Cutts, Steve Esrey, Ake Blomqvist, Joseph Brunet-Jailly, Claude Cas- Ronald Gray, Jerry Gibson, Bill Hausdorf, Jim Heiby, tonguay, Andrew L. Creese, Robert G. Evans, Donald A. Henderson, Terrel Hill, Sandra L. Huff- Claude Forget, William Hsiao, Naoki Ikegami, Daniel man, Jessica Jitta, Pamela Johnson, Charlotte Neu- M. Le Touzé, Mario Taguiwalo, Abdelmajid Tibouti, man, Alok Perti, Phyllis Piotrow, Kenneth F. Schulz, Bokar Touré, and Katarzyna Tymowska Jim Shepperd, William A. Smith, Sally Stansfield, Review of WDR Findings Hope Sukin, Nebiat Tafari, Taha el Tahir Taha, Carl E. Taylor, James L. Tulloch, Roxann Van Dusen, Caby C. January 26, 1993, Institute of Medicine, Washington, Verzosa, Kenneth S. Warren, and Vivian Wong D.C. Women and Health Chair: William H. Foege December 7-9, 1992, Cumberland Lodge, Windsor, Participants: Abdelmonem A. Afifi (provided written England. The consultation was organized by the Lon- comments), Carolyn Asbury, David E. Bell, Richard don School of Hygiene and Tropical Medicine. Finan- Bissell, Barry R. Bloom, Margaret Catley-Carlson, J. cial support was provided by the U.K. Overseas De- Jarrett Clinton, Joseph A. Cook, Richard G. A. Fea- velopment Administration. chem, Harvey V. Fineberg, Julio Frenk, Susan Gibb, Polly F. Harrison, Donald A. Henderson, Jeffrey R. Organizers: Oona Campbell, Wendy Graham, and Veronique Filippi Koplan, Adetokunbo 0. Lucas, Christopher J. L. Murray, June E. Osborn, Adeline Wynante Patterson, Participants: Uche Amazigo, Carmen Barroso, Loretta David P. Rall, Frederick C. Robbins, Timothy Ro- Brabin, Mayra Buvinic, Mirai Chatterjee, Ann Coles, thermel, Kenneth I. Shine, Alfred Sommer, Roxann Richard G. A. Feachem, Zuzana Feachem, Aleya El Van Dusen, Noel S. Weiss, Barbara L. Wolfe, and Bindari Hammad, Sioban Harlow Gillian Holmes, James Wyngaarden Susan Joekes, Marjorie Koblinsky, Joanne Leslie, Environment and Health Claudia Garcia Moreno, Jacky Mundy, Cynthia Myntti, David N. Nabarro, Phoebe Roome, Kasturi February 4-5, 1993, World Bank, Washington, D.C. Sen, Jacqueline Sherris, Godfrey Walker, and Judith Financial support was provided by the Environmen- Wasserheit tal Health Division, World Health Organization. Global Burden of Disease Chair: Wilfried Kreisel December 10-11, 1992, World Health Organization, Participants: Hendrik De Koning, Devra Lee Davis, Geneva, Switzerland. The consultation was orga- Richard G. A. Feachem, Jacobo Finkelman, Gregory nized by the World Health Organization. Financial Goldstein, Tord Kjellstrom, Anthony J. McMichael, support was provided by the Edna McConnell Clark Horst Otterstetter, David P. Rall (provided written Foundation. comments), and Kirk Smith Chair: Jean-Paul Jardel Improving the Effectiveness of International Assistance to Health Participants: Carla Abou-Zahr, David Barmes, Monika Blössner, Luis Lopez Bravo, Anthony Burton, Yan- February 9-10, 1993, World Bank European Office, kum Dadzie, Richard G. A. Feachem, Jacques Ferlay, Paris, France. Partial financial support was provided Tore Godal, Ann Goerdt, Sandra Gove, Walter by the Danish International Development Agency. Gulbinat, Habib Rachmat Hapsara, Joachim Hempel, Mark Kane, Hilary King, Jeffrey R. Koplan, Jacob Chair: Anthony R. Measham Kumaresan, Marie-Hélène Leclerq, Alan Lopez, In- Participants: Marja Antilla, Lynn Bailey, Jose Bar- grid Martin, Alvaro Moncayo, Christopher J. L. Mur- zellato, Alfredo R. A. Bengzon, Luciano Carrino, Ge- ray, Jenny Pronczuk, Jean-Marie Robine, Claude nevieve Chedville-Murray, Zafrullah Chowdhury, 174 Immita Cornaz, Göran Dahigren, Francois Decaillet, Chair: Jeffrey R. Koplan Nicolas de Riviere, Tore Godal, Klaus Gordel, Arm- Participants: Ruth L. Berkelman, Ruth A. Etzel, Fran- elle George-Guiton, Jacques Hallak, Kyo Hanada, çoise F. Hamers, Jeffrey R. Harris, Nancy C. Lee, Anne Kristin Hermansen, Gillian Holmes, Jean-Paul Alan Lopez, Christopher J. L. Murray, Mark L. Jardel, Eva Jesperseri, Jorge Jimenez de la Jara, Mat- Rosenberg, Richard B. Rothenberg, Frank M. Vinicor, thias Kerker, Robert Kestell, Irene Klinger, Roif and Ray Yip Korte, Louise Lassonde, Jean-Marie Laure, Robert S. Lawrence, Rune Andreas Lea, Dominique Maroger, Seminars Catherine Michaud, Rajiv L. Misra, Bernard Mon- taville, W Muchenje, David N. Nabarro, Francois An important source of ideas for this Report was a Orivel, Tom Ortiz, Aagje Papineau Saim, Liu series of seminars. Most were held at the World Bank Peilong, Ines Penn, Martin Pinero, Peter Poore, Vu!- and were cosponsored by the World Bank's Popula- imiri Ramalingaswami, Olikoye Ransome-Kuti, Brett tion, Health, and Nutrition Department. The Har- Ridgeway, Jon Rohde, Yolanda Richardson, Kenneth vard Center for Population and Development Studies Ross, Timothy Rothermel, Philippa Saunders, Chris- held a series of five seminars to assist in developing topher Shaw, Leonardo Santos Simão, Stephen Si- World Development Report themes; these were orga- mon, Margareta Sköld, Guillermo Soberon, Birgit nized by Lincoln C. Chen and Julio Frenk. The Storgaard, Muthu Subramanian, Carl Wahren, Ronald Wilson, Robert Wrin, Carlos Yanez-Barneuvo, George Washington University Center for Interna- tional Health held a seminar, organized by RosalIa and Pat Youri Rodrigues-GarcIa, to critically review the findings of Violence against Women the World Development Report. The World Bank series included presentations by February 12, 1993, Washington, D.C. This follow-up Henry Aaron, John Akin, Kenneth Arrow, Amie Bat- consultation to that on Interventions for Nervous son, Jere Behrman, David Bloom, Michael Cichon, Systems Disorders focused on the health outcomes of Andrew Creese, Anil Deolalikar, Avi Dor and Janet violence against women for the global burden of dis- Hunt-McCool, Alain Enthoven, Michelle Fryer, Paul ease exercise. Gertler, Eric Hanushek, Estelle James, Lawrence J. Chair: Helen Saxenian Lau, Beryl Levinger, Joseph Newhouse, Abdel Omran, Joel Nobel, Francois Orivel, Charles Phelps, Participants: Jacqueline Campbell, Walter Gulbinat, Lori Heise, Dean Kilpatnick, and Christopher 1. L. Samuel Preston, Barry Popkin, Uwe Reinhardt, George Scheiber, T. Paul Schultz, Donald Shepard, Murray John Strauss, Duncan Thomas, Carol Vlassoff, and Review of the Global Burden of Disease Beverly Winikoff. March 15, 1993, Centers for Disease Control and Pre- vention, Atlanta, Georgia 175 Bibliographical note The Report has drawn on a wide range of World Bank Those outside the World Bank who contributed reports and advice and on numerous outside sources. substantially with comments and material include Special thanks go to the World Health Organization Aloysio Achutti, Universidade Federal do Rio Grande (WHO) for providing extensive expert advice, techni- do Sul, who assisted with the preparation of Figure 3; cal materials, and helpful comments. A. A. Afifi, University of California, Los Angeles; The principal sources are noted below and are also Jere Behrman, University of Pennsylvania; Marit listed alphabetically by author or organization in two Berggrav, Einar Heldal, Rune Andreas Lea, Johanne groups: background papers commissioned for this Sundby, and Ann-Karin Valle, Norwegian Agency Report and a selected bibliography. for International Development (NORAD); Barry In addition to the sources listed, many people both Bloom, Albert Einstein College of Medicine; Robert inside and outside the World Bank helped with the H. Cassen, International Development Centre, Ox- Report. In particular, helpful comments were re- ford University; Immita Cornaz, Swiss Development ceived from World Bank staff and consultants, includ- Cooperation; Göran Dahlgren, Swedish International ing Alexandre Abrantes, Masood Ahmed, Michael Development Authority (SIDA); Joe H. Davis, Cen- Azefor, Howard Barnum, Alan Berg, Eduard Bos, Pa- ters for Disease Control and Prevention; Antoine De- tricia Daly, Willy De Geyndt, Janet de Merode, Jean- grémont, Swiss Tropical Institute; David Fraser, Aga Jacques de St. Antoine, Dennis de Tray, Alfred Duda, Khan Institute; Lucy Gilson, London School of Hy- Graham Dukes, Oscar Echeverri, A. Edward Elmen- giene and Tropical Medicine; Ted Greiner, Uppsala dorf, James Green, Charles Griffin, Ann Hamilton, University; Davidson Gwatkin, International Health Jeffrey Hammer, Barbara Herz, Janet Hohnen, Ishrat Policy Program; David J. Halliday, UNICEF; Gillian Z. Husain, Estelle James, Emmanuel Jimenez, Eliz- Holmes and David Nabarro, U.K. Overseas Develop- abeth King, Timothy King, Mubina Kirmani, Kathie ment Administration (ODA); William C. Hsiao, Har- Krumm, Joseph Kutzin, Jean-Louis Lamboray, Kye vard School of Public Health; Valerie Hull, Australian Woo Lee, Danny M. Leipziger, Maureen Lewis, Sam- International Development Assistance Bureau uel Lieberman, Bernhart Liese, James Listorti, Mar- (AIDAB); Pamela Johnson, Richard Seifman, and lame Lockheed, Jack Maas, Jo Martins, Judith Robert Wrin, U.S. Agency for International Develop- McGuire, Mohan Munasinghe, Rieko Niimi, Mead ment (USAID); Joanne Leslie, UCLA School of Public Over, Lisa Pachter, Ok Pannenborg, David Peters, Health; Adetokunbo 0. Lucas, Harvard University; Ian Porter, Juan Prawda, George Psacharopoulos, A. J. McMichael, University of Adelaide; Peter Poore, Sandra Rosenhouse, Anna Sant'Anna, Miguel John Seaman, and David Woodward, Save the Chil- Schloss, Julian Schweitzer, lona Sebastian, Paul dren (U.K.); Barry Popkin, University of North Caro- Shaw, James Socknat, Lyn Squire, Andrew Steer, lina; Vulimiri Ramalingaswami, Task Force on Health Susan Stout, Vinod Thomas, Erik Thulstrup, Anne Research for Development; Patricia L. Rosenfield, Tinker, Vincent Turbat, Jagadish Upadhyay, Denise Carnegie Corporation; Timothy S. Rothermel, United Vaillancourt, Armand Van Nimmen, Herman van der Nations Development Programme (UNDP); A. Pa- Tak, Dominique van de Walle, Claudia Von Monbart, pineau Saim, Ministry of Foreign Affairs, Nether- Marie-Odile Waty, Kin Bing Wu, Guillermo Yepes, lands; Philippa Saunders, OXFAM; Alfred Sommer Mary E. Ming Young, and Shahid Yusuf. and Carl Taylor, Johns Hopkins School of Hygiene 176 and Public Health; Birgit Storgaard, Ministry of For- Bank-UNICEF exercise to be used in UNICEF forth- eign Affairs, Denmark; Noel S. Weiss, University of coming and described in Hill and Yazbeck, back- Washington; and Hans Emblad, Tore Godal, Marcus ground paper. Eduard Bos and My Vu of the World Grant, Fritz Kaferstein, Wilfried Kreisel, Alan D. Bank's Population, Health and Nutrition Department Lopez, and James C. Tulloch, WHO. provided invaluable assistance and advice with the base regional population projections underlying Chapter 1 much of Chapter 1 and Appendix A. Mortality as- sumptions were updated and revised in the light of This chapter draws on technical materials from the discussions with Larry Heligman of the United Na- World Health Organization and the World Bank and tions Population Division, Gareth Jones of UNICEF, on the scientific literature. The smallpox story is and recent data from the Demographic and Health adapted from Fenner and others 1988. The discussion Surveys program provided by Ties Boerma or extrac- of the gains in worker productivity from better health ted from recent reports. Nicholas Eberstadt contrib- draws on studies by Castro and Mokate 1988, Conly uted useful ideas on mortality differentials in adult- 1975, Max and Shepard 1989, Nur and Mahran 1988, hood, and the section further benefited from Pitt, Rosenzweig, and Hassan 1990, Sagan and Afifi Feachem and others 1992. 1979, Schultz and Tansel 1993, and Hill and others, The results in the section on the global burden of background paper, as well as on helpful materials disease are taken from a joint World Bank-World provided by John Caldwell, Gavin Jones, and John Health Organization study (Murray and Lopez back- Anarfi. Anil Deolalikar provided additional material ground paper); many collaborators are listed in Ap- on the economic impact of improved nutrition in In- pendix B. Material on measuring the burden of dis- dia, and John Akin made available unpublished notes ease for Ghana, to establish health care priorities, on the relationship between health and income. The came from Ghana Health Assessment Project Team cost-benefit calculations of malaria eradication in Sri 1981. Feachem 1988 stresses the importance of "mac- Lanka are derived from Barlow and Grobar 1985. roepidemiology" for health planning. Aehyung Kim and Bruce Benton contributed to Box The section on challenges for the future is based on 1.1. Institute of Medicine 1992, Mackay 1993, and WHO The discussion of the education benefits of im- 1992b and 1992c and on information on HIV and proved health and the related economic benefits of AIDS provided by Seth Berkley, Rand Stoneburner, improved education is based on studies by Behrman and WHO staff. D. A. Henderson provided informa- and others 1991, Boissiere, Knight, and Sabot 1985, tion on emerging microbial infections; Tore Godal, Glewwe 1991, Gomes-Neto and Hanushek 1991, Jam- Tekle Haimanot, and Hans Remme on malaria; and ison and Leslie 1990, Jamison and Moock 1984, Nokes Alan D. Lopez and Neil Collishaw on smoking. and others 1992, and Psacharopoulos 1993. Jacobs and others 1993 report on the development of The section on reduced costs of medical care draws a test for drug resistance in tuberculosis. on studies by Ainsworth and Over 1992, notes by The discussion on demographic and epidemiologic David Bloom and Ajay Mahal on the implications of transition draws on studies by Bobadilla and others reducing the rate of HIV transmission among Se- forthcoming, Frenk and others 1989, and Omran ropositive individuals (with additional personal com- 1971. Country-specific discussions of the implications munication from the authors) and Musgrove 1988. of epidemiological transition for health policy may be Martha Ainsworth and Mead Over drafted Box 1.2. found in World Bank 1984a, World Bank 1990a, and Bloom and Lyons 1993 provide analyses pointing to World Bank 1992a. The purchasing power parity per the economic gains associated with AIDS prevention capita incomes used in Figure 1.9 were provided by in a number of Asian countries. Angus Maddison. The discussion of the factors ex- The discussion of the impact of health investments plaining mortality declines is based on Ewbank and on poverty draws on the work of Henry Mosley and Preston 1990, McKeown 1976, and Preston and on World Bank 1980 and 1990a. The record of success Haines 1991. draws heavily on statistical publications of the United The structure and content of the chapter benefited Nations (Demographic Yearbook, various years) and th from presentations made by Abdel Omran and Sam- World Health Organization (Statistics Annual, various uel Preston. Valuable comments on earlier drafts years). Ingram 1992 discusses the greater conver- were made by Joseph Cook and Richard Morrow. gence of social (including health) indicators than of income across countries. Chapter 2 Child mortality estimates are derived in part from United Nations 1988 but were mostly calculated from This chapter draws on academic sources, presenta- data in United Nations 1992 as part of a joint World tions by speakers in the seminar series cosponsored 177 by the World Development Report and the World Bank and 2.5, respectively. Robert Anda, David Bradley, Population, Health and Nutrition Department, and John Briscoe, Mayra Buvinic, Brigitte Duces, Luis Es- on numerous World Bank documents. The discussion cobedo, Paul Gertler, A. K. Shiva Kumar, Joanne on household capacity (income and schooling) was Leslie, Ruth Levine, Jack Molyneaux, Damianos informed by studies that included Anand and Ray- Odeh, Nick Prescott, Luis Serven, John Strauss, and allion 1993, Behrman 1990, Benefo and Schultz 1992, Molly Tees contributed helpful data and resource ma- Fuchs 1979, Grossman 1975, Hill and Palloni 1992, terials. Valuable comments on earlier drafts were re- Jeyaratnam 1985, Lau and others, background paper, ceived from Sue Berryman, Joseph Bredie, Barbara Luft 1978, Natale and others 1992, Oganov 1992, Pal- Bruns, Ishac Diwan, Edward Henevald, Eva Jarawan, loni 1981, Pierce 1989, Pritchett and Summers, back- Himelda Martinez, Kenneth Shine, and David ground paper, Rodgers 1979, Rogot, Sorlie, and John- Woodward. son 1992, Strauss and others 1992, United Nations 1985, Wilkinson 1992, and World Bank 1990a. Chapter 3 The discussion on women's schooling and child health drew on the extensive literature, including This chapter draws on a wide range of published Bhargava and Yu 1992, Bruce and Lloyd 1992, Cald- and unpublished sources, including documentation well 1986, Cleland 1990, Elo 1992, Engle 1991, Hod- and expertise from the World Health Organization dinott and Haddad 1991, Kennedy 1992, King and and the World Bank and on the academic literature. Hill 1993, Leslie 1989b, Lindenbaum, Chakraborty, Discussion of the role of government draws on World and Elias 1985, Louat, Grosh, and van der Gaag 1992, Bank 1991. Jamison and others forthcoming summa- Over and others 1992, Sahn 1990, Summers 1992, rizes the methods and findings of the cost effective- Thomas 1990, Thomas, Strauss, and Henriques 1990, ness analysis that forms the starting point for the and World Bank Water Demand Research Team 1993. analyses used in this report. The data on health ex- The discussion on economic policy reform and ad- penditures in the first section were compiled from a justment lending drew on sources that included background paper by Murray, Govindaraj, and Behrman 1992, Berg and Hunter 1992, Edwards forth- Chellaraj, which used a wide range of government coming, Kakwani, Makonnen, and van der Gaag health budgets, World Bank reports, and other coun- 1990, Serageldin, Elmendorf, and El-Tigani forthcom- try studies of health financing. Heller and Diamond ing, Summers and Pritchett 1993, Thomas, Lavy, and 1990 also treat this issue. Data on equity in health Strauss 1992, Woodward 1992, World Bank 1990b, status, access, and expenditure were drawn from the World Bank 1992e, and World Bank 1993c. Duncan World Bank's Living Standard Measurement Study Thomas contributed materials on protecting nonsal- (LSMS) and were further analyzed by Kalpana ary spending during economic adjustment. Mehra. The analysis of costs and benefits of packages The presentation on education policies was in- of public health measures and essential clinical ser- formed by Alderman and others 1992, Jamison and vices draws on the background paper by Bobadilla Leslie 1990, Jarousse and Mingat 1992, Lockheed, and others. Verspoor, and associates 1991, Minhas 1991, Over Box 3.1 was drafted by Richard Bumgarner. Box 3.2 and Piot forthcoming, Tan and Mingat 1992, and is based on unpublished data provided by the Insti- World Bank 1988. tuto Materno-Infantil de Pernambuco and on The work on policies for empowering women drew UNICEF and IMIP 1992. Box 3.3 is from the chapters on Akin and others 1985, Birdsall and McGreevey on measles and on tuberculosis in Jamison and others 1983, and Leslie 1989a. The discussion of women and forthcoming. The discussion of cost-effective inter- violence benefited from assistance from Jacquelyn ventions also draws on Halstead, Walsh, and Warren Campbell, Rosemary Garner, Lori Heise and Dean 1985, Walsh 1988, and Walsh and Warren 1979. Basic Kilpatrick and drew on Archavanitkui and Pram- economic issues and their application are treated in aualrantan 1990, Bradley 1988, CAMVAC 1985, Coun- Over 1991. The discussion of market failures in health cil on Scientific Affairs 1992, COVAC 1990, Fauveau draws particularly on Arrow 1963. Insurance and reg- and Blanchet 1989, Handwerker 1991, Hosken 1988, ulation are discussed generally in Diamond 1992 and Koop 1989, Koss, Koss, and Woodruff 1991, Plitcha for Brazil in Piola and Vianna 1991. The section on 1992, Shim 1992, Stark 1984, and Stark and Flitcraft government failures in health policy takes examples 1991. from Evans, Barer, and Labelle 1988, Hlady and Lawrence Lau contributed to the drafting of Box others 1992, and 1DB 1988. Equity examples are 2.1. Box 2.2 is based on material provided by Michelle drawn from Black and others 1982, Meerman 1980, Fryer. Carmen Barroso, Lori Heise, and Nahid Toubia Musgrove 1986 and 1993, and President's Commis- contributed to Box 2.3. John Hobcraft and Aloysio sion 1983, as well as from the work of Prescott and Achutti assisted with the preparation of Figures 2.4 others on social spending in Indonesia. The discus- 178 sion of satisfaction with health care uses Bitran and 1990. Joanne Leslie contributed Box 4.1, and Jayshree Mclnnes 1993, Blendon and others 1990, and Gertler Balachander contributed Box 4.2. Harold Alderman, and van der Gaag 1990. Leslie 1989a discusses the George Beaton, Robert Black, Barry Bloom, Leslie time cost of health interventions, an issue that is not Elder, Paul Elliott, Abraham Horwitz, Suraiya Ismail, explicitly addressed in the cost-effectiveness calcula- Francisco Mardones, Reynaldo Martorell, John tions reported here but that deserves further work. Mason, Paul McKeigue, Daan Mulder, Philip Payne, Brook and Lohr 1986 provide evidence pointing to David Pelletier, and Peter Piot provided helpful infor- huge overuse of medical care in the United States mation or comments. beyond what is of value even at zero cost, resulting in The section on fertility drew on Cochrane and Mer- part from third-party financing. rick, background paper, AbouZahr and Royston Nicholas Barr, Peter Diamond, Robert Evans, and 1991, Amadeo, Chernichovsky, and Ojeda 1991, Fernando Figueira provided valuable ideas and Bertrand and Brown 1992, Population Information comments. Program 1992, Population Reference Bureau 1992a Chapter 4 and 1992b, Sanderson and Tan forthcoming, Ste- phenson and others 1992, United Nations forthcom- This chapter draws on documentation and expertise ing, World Bank 1984a, 1992c, and 1993a, and from the World Health Organization and the World Zinanga 1992. Birgitta Bucht, Parker Mauldin, Vin- Bank and from the academic literature, as well as on cent Miller, Richard Osborn, Warren Sanderson, Bev- expert consultations and on papers and discussions erley Winikoff, and the staff at the Rockefeller Foun- in the seminar series sponsored by the World Develop- dation provided helpful materials and advice. John ment Report and the World Bank Population, Health, Hobcraft assisted in the preparation of Figure 4.3. and Nutrition Department. The section on tobacco, alcohol, and drugs bene- The discussion on immunization and other popula- fited from materials and comments from Jerry Husch, tion-based health services draws on ideas and data Judith Mackay, Richard Peto, and Derek Yack. The discussed at the consultation on Child Health held in discussion drew on background materials from James Baltimore in 1992. Berkley and Jamison 1991 discuss Cercone and from the U.S. Surgeon-General's 1992 the cost and effectiveness of school-based programs report on smoking in the Americas, as well as on for mass treatment of worm infections and micro- Gutierrez-Fisac, Regidor, and Ronda 1992, Pierce nutrient deficiencies. Assistance was also provided 1991, Walsh and others forthcoming, Wasserman and by Amie Batson, Donald Bundy, Pamela Johnson, others 1991, and WHO 1991b and 1992e. Marjorie Koblinsky, Jim Shepperd, Jacqueline The section on the environment benefited from the Sherris, and Nebiat Tafari. Other sources were Bour- contributions of participants in a joint WHO-World don, Orivel, and Perrot 1993, Brenzel 1990, Nokes Bank consultation (see Acknowledgments) and from and others 1992, Robertson and others 1992, Shepard additional assistance provided by Carl Bartone, and others 1989, and chapters in Jamison and others David Bates, Sue Binder, Gloria Davis, Roger Detels, forthcoming on measles, polio, hepatitis B, tetanus, John Dixon, Mohamed T. El-Ashry, Gunnar Eske- and helminth infection. land, Ruth Etzel, Philip Graitcer, Peter Kolsky, Tony The section on diet and nutrition drew on Levin McMichael, David Rail, Anand Seth, and Anthony and others forthcoming, Pinstrup-Anderson and Zwi. Data on the health impact of water supply and others forthcoming, and a variety of other sources. sanitation were taken from the extensive literature Valuable summaries of particular topics are given in and from recent reviews by Cairncross 1990, Esrey ACC/SCN 1991, 1992a, and 1992b, Beaton and and others 1991, and Huttly 1990. The material on Ghassemi 1987, Beaton and others 1993, Berg 1987, water and sanitation policy drew on Briscoe 1992, Drèze and Sen 1989, Elliott 1988, Humphrey, West, World Bank 1992f, and World Bank Water Demand and Sommer 1992, Keusch and Scrimshaw 1986, Les- Research Team 1993. Box 4.4 was drafted by Sandy lie 1987, Leslie, Jamison, and Musgrove forthcoming, Cairncross. Box 4.5 relies on Blum and others 1990 McGuire and Popkin 1990, Monteiro 1988, National and on Feachem and others 1978 for time spent col- Research Council 1989, Pelletier 1991, Pollitt 1990, lecting water. Michael Garn, Letitia Obeng, and Popkin 1993, Sen 1981, Tomkins and Watson 1989, Guillermo Yepes contributed data on water and sani- and U.S. Centers for Disease Control and Prevention tation costs. Greg Watters collated the data in Figure 1992. The section also drew on studies by Adair and 4.6. The discussion of indoor air pollution rests on the others 1993, Bhargava 1992, Black 1991, Bouis 1990, reviews by Betty Kirkwood and colleagues and on the Lutter and others 1992, Mardones and Zamora 1989, work of Kirk Smith. Relevant literature included An- Martorell and others 1992, Musgrove 1990, Stamler derson 1979, Chapman and others 1989, Chen and and others 1989, Thomas, Lavy, and Strauss 1992, others 1990, Norboo and others 1991, Pandey and Waaler 1984, and Walter, Olivares, and Hertrampf others 1989, Smith forthcoming, Smith and Liu 1993, 179 and Smith and Rodgers 1992. Christopher Curtis pre- Material on the success of prevention was drawn pared Box 4.6, with assistance from Cohn Leake, from the literature and from a meeting at GPA/WHO making use of data from Alonso and others 1991, in 1992. The costs and benefits of the public health Curtis 1992, and Maxwell and others 1990. Discus- measures in the essential package are presented in sion of housing policy was informed by World Bank Bobadilla and others, background paper. 1993b. The discussion of the wider environment drew on Chapter 5 Doll 1992 and on the comprehensive accounts con- tained in WHO 1992d and World Bank 1992f. The The costs and benefits of the clinical services in the discussion on occupational health drew on Andreoni essential package are described in Bobadilla and 1986, El Batawi and Husbumrer 1987, and Wegman others, background paper. The discussion of the cost 1992. The discussion of the ambient environment of the essential package of clinical services and mech- drew on the extensive literature, including Bellinger anisms for delivering it drew on the work of the and others 1987, Bradley and others 1992, Faiz and World Bank's Africa Technical Department and others forthcoming, Lancet 1992, MRC 1989, Needle- World Bank forthcoming. The components of the es- man and others 1990, Romieu 1992, Romieu, sential package of clinical services for children drew Weitzenfeld, and Finkelman 1990, Schwartz and on analyses from UNICEF 1993 and on priorities pro- Dockery 1992, WHO 1992a, and WHO/UNEP 1992. posed in UNICEF, WHO, and UNESCO 1991. The The material for Box 4.7 is taken from Bobak 1993, analyses of sources of health financing, provider Bobak and Feachem 1992, Bobak and Leon 1992, and compensation, and alternative modes of service de- World Bank 1992d. Box 4.8 is taken from Study livery are based on the work of Arrow 1963, Barr Group for Global Environment and Economics 1991, 1992, Griffin 1992, Hsiao 1992, Hurst 1992, Reinhardt supplied by Tord Kjellstrom. José Carbajo, Paul Gui- 1991, Schneider and others 1992, van Doorslaer, Wag- tink, Zmarak Shalizi, and John Wootton assisted with staff, and Rutten 1993, and World Bank 1992a. It the section on transport risks, which also drew on draws on and is closely linked with a series of papers Barss and others forthcoming, Downing 1991, John- by the World Health Organization: WHO 1991a, on ston 1992, Smith and Barss 1991, TRRL 1991, WHO health care reform in Eastern and Central Europe; 1989a, and Zwi 1992. WHO 1991c, on the public-private mix; and WHO The section on AIDS benefited from the contribu- 1993. tions made by members of the AIDS consultation (see The analysis of user charges and community fi- Acknowledgments) and from additional assistance nancing draws on the work of Abel-Smith and Dua from Richard Hayes, Daan Mulder, Peter Piot, 1988, Gertler and van der Gaag 1990, Hecht, Over- Wendy Roseberry, Allan Rosenfield, Gary Slutkin, holt, and Holmberg 1993, Korte and others 1992, and Peter Smith. Projections of numbers of infections Lewis and Parker 1991, Litvack and Bodart 1993, were generated by Tony Burton, Rand Stoneburner, McPake, Hanson, and Mills 1992, and others. The and other staff of the Global Programme on AIDS of discussion of health insurance in developing coun- the World Health Organization (GPA/WHO). Mate- tries draws on Abel-Smith 1992b, De Geyndt 1991, rial on the core groups is drawn from Moses and Ikegami 1992, Kutzin and Barnum 1992, McGreevey others 1991 and Over and Piot forthcoming. The ac- 1990, Mesa-Lago 1992, Vogel 1989, Yang 1991, and Yu count of community intervention in Zimbabwe is and Anderson 1992. Analysis of the determinants of drawn from material supplied by David Wilson. Ma- health spending in the OECD countries draws on terial on HIV and breastfeeding is drawn from Dunn Gerdtham and others 1992. The review of options for and others 1992 and various WHO materials. Box 4.9 improving public and private delivery of clinical ser- draws on Goodgame 1990, Katabira and Goodgame vices draws on Bennett 1992, Foster 1991, and World 1989, and Muller and others 1992. Information on Bank 1992g. The discussion of managed competition voluntary testing and counseling is drawn from Fos- relies on Enthoven 1988 and Relman 1993. The dis- ter 1990, Muller and others 1992, and WHO Global cussion of decentralization of health services draws Programme on AIDS 1993b. Estimations of the cost of on Mills and others 1990 and World Bank 1992b. worldwide prevention were drawn from WHO Box 5.1 draws on material produced by Marjorie Global Programme on AIDS 1993a and WHO forih- Koblinsky and on Tinker and Koblinsky 1993 and coming, and from work by Doris Schopper. The dis- Walsh and others forthcoming. Box 5.2 was prepared cussion of the cost-effectiveness of treating sexually with information provided by James C. Tulloch and transmitted diseases is drawn from Moses and others Sandra Gove of WHO. Box 5.3 draws on material 1992, and Over and Piot forthcoming. Box 4.10 is provided by Judith Wasserheit. Box 5.4 was based on drawn from Viravaidya, Obremsky, and Myers 1991 Murray, Styblo, and Rouillon forthcoming. Box 5.5 and from materials contributed by Werasit Sittitrai. draws on Grosh 1992. Box 5.6 uses material from 180 Schieber, Poullier, and Greenwald 1992 and the re- coming, Reyes and Picazo 1990, Richards and Fulop suits of analysis by the WDR team of the relationship 1987, Schmidt and others 1991, Schroeder 1984 and between the public share of health spending and 1992, Schwab 1987, Tarlov 1986 as cited in Reinhardt health care costs. Box 5.7 is based on material from 1991, Welch and others 1993, Whitfield 1987, and un- Hurst 1992. Werner 1987 discusses approaches to published material from Ruth Roemer and WHO. In low-cost but effective rehabilitation from disability. Box 6.2 the discussion of community health workers Peter Berman and Louis Vassiliou provided useful in Jamaica draws on Cumper and Vaughan 1985. The material and Alfred Bartlett, Ricardo Bitran, Michael discussion of the Pastoral da Crianca draws on mate- Cichon, Andrew Creese, Jennie Litvack, Kasa Asila rials provided by the Coordenacao Nacional da Pas- Pangu, John Rohde, Abdelmajid Tibouti, Jacques van toral da Crianca and on Victora and Barros 1990. der Gaag, Ronald Wilson, and Zia Yusuf made valu- The drug discussion draws on Andersson 1992, able comments. Caplan 1985, Foster 1990, Hlady and others 1992, Holly and Lee 1992, Kanji and others 1992, Laing Chapter 6 1990, Management Sciences for Health 1992, Nazerali 1992, Office of Technology Assessment 1993, This chapter draws on technical materials from the Thomas, Lavy, and Strauss 1992, Tomson and Sterky World Health Organization and the World Bank and 1986, and WHO 1988a, 1988b, and 1988c. Box 6.5 from the scientific literature. Regional estimates of draws on World Bank material and on Barros and hospital beds, physicians, and nurses are from others 1986 and Fadndes and Cecatti 1993. Informa- OECD, WHO, World Bank, and national statistics. tion on the INCLEN program is from Halstead, Tug- The hospital and district health system discussion well, and Bennet 1991. The information and research draws on Barnum and Kutzin 1993 and World Bank to guide decisionmaking draws on Enthoven 1989. forthcoming. Estimates of global spending on health research are Valuable comments and materials were provided from the Commission on Health Research for Devel- by Orvil Adams, Uche Amazigo, Harvey Bale, Wil- opment 1990. The health research discussion also bert Bannenberg, Pascal Brudon-Jakobowicz, Robert draws on Free 1991, Godal 1993, and WHO 1991d. Cassen, Gilles Dussault, Anibal Faündes, Enrique Feffer, Michael Free, Julio Frenk, John Gil-Martin, Chapter 7 Wendy Graham, Richard Heller, Richard Laing, John Lloyd, Alfonso MejIa, Violaine Mitchell, Hiroshi Na- Information on health policy reform was provided by katani, Vic Neufeld, Joel Nobel, Joao Batista Oliveira, Jonathan Broomberg for South Africa, Louise Fox for Diego Palacio, David Porter, Michael Porter, Jim Romania, Salim Habayeb for India, Evangeline Javier Rankin, Una Reid, and staff of the Aga Khan Devel- for Chile, and Mary E. Ming Young for Poland. The opment Network and the Aga Khan Foundation. section on aid flows is based on the background pa- Box 6.1 was prepared by Tamara Fox and Ruth Le- per on aid by Michaud and Murray, which used a vine. Estimates of medical equipment expenditure wide range of data from Organization for Economic are from Rozynski and Gallivan 1992. Cooperation and Development, United Nations, and Estimates on the efficiency of outpatient surgery in bilateral sources. The discussion of international Colombia are from Shepard and others 1990. Hospital health research draws heavily on Commission on planning experience in the United States is based on Health Research for Development 1990. John Barton Davis and others 1990. The discussion of regional and Selcuk Ozgediz furnished material on the Con- planning is based in part on Jonsson 1989. The rela- sultative Group for International Agricultural Re- tionship between the volume of surgery and health search (CGIAR) and its relevance for health research. outcomes is based on Hughes, Hunt, and Luft 1987. Valuable comments on the draft chapter were re- Papua New Guinea's experience with containing ceived from John Evans. Box 7.1 is based on UNICEF! hospital spending is based on Newbrander 1987. The Bamako Initiative Management Unit 1990 and 1992 medical equipment discussion draws on Bloom 1989, and on the evaluation study by McPake, Hanson, and Bruley 1991, Garber and Fuchs 1991, Gelijns and Mills 1992. Box 7.2 draws on World Bank reports on Halm 1991, Halbwachs 1992, and WHO various Chile and on material provided by Thomas Bossert years. Abel-Smith 1992a and Rublee 1989 provided on evaluations of health-financing reforms and de- information on medical technology policies in indus- centralization. Box 7.3 uses material provided by Dov trial countries. Chernichovsky and George Schieber. Box 7.4 encap- The human resources for health discussion draws sulates the findings of Cassen and others 1986 and on Abel-Smith 1986, Enthoven and Vorhaus 1992, Riddell 1987 in particular. Guy Ellena and Joseph Evans 1981, Foster 1987, Frenk and others 1991, Kutzin assisted with Box 7.5. Box 7.6 draws heavily Fuchs 1978, Institute of Medicine 1988, Javitt forth- on World Bank 1993d. 181 Appendix A Deborah Symmons, B.-I. Thylefors, Ian Timeus, Carol Torel, James C. Tulloch, Ronald Waldman, Appendix A benefited from contributions from many Godfrey Walker, Jay Wenger, William Whang, Erica institutions; particularly valuable were the data re- Wheeler, Russell Wilkins, G. Yang, R. Yip, and ceived from WHO and UNICEF. Richard Bumgarner Anthony Zwi. and Godfrey Walker of WHO provided information Others who contributed include M. Adrian, Ann for many of the health indicators. Gareth Jones, Ashworth-Hill, P. Blake, Uwe Brinkman, C. Broome, UNICEF, contributed sources of data for nutritional Richard Bumgarner, Jacqueline Campbell, P. Car- indicators. Shea Rutstein from the Demographic and levaro, Mary Chamie, Lincoln C. Chen, D. P. J. Health Surveys project provided data on breastfeed- Daumerie, Hans Emblad, R. Etzel, Paul Fine, A. ing. Robert Hartford and Francis Notzon made avail- Galazka, Marito Garcia, S. Gillespie, Marcus Grant, able the database on perinatal and infant mortality R. J. Guidotti, Francoise Hamers, H. R. Hapsara, J. from the National Center of Health Statistics. Roy Harris, Peter Heller, Alan Hill, C. J. Hong, H. Jamai, Miller, USAID, commented on an earlier version of Fritz Kaferstein, Alex Kalache, N. Khaltaeve, Betty the appendix. My Vu of the World Bank's Population, Kirkwood, Arata Kochi, Jacob Kumaresan, N. Lee, Health, and Nutrition Department was responsible Anthony Mann, Ingrid Martin, G. Mayberly, Juan for preparing the statistical appendix for World Bank Menchaca, Michel Mercier, T. R. Mertens, Alvaro forthcoming, which was extensively used in the ap- Moncayo, Richard Morrow, Y. Motarjemi, Shaik Na- pendix, and processed raw data on several of the deen, William Newbrender, M. Noel, Godfrey selected health indicators. James Cercone helped pro- Oakley, D. Peterson, A. Pio, G. R. Quinke, C. P. cess data on mortality by broad causes of death. Ramachandran, M. Rosenberg, Norman Sartorius, Alan M. Schapira, Gordon Smith, Peter Smith, T. Appendix B Studwick, M. Thuriaux, Andrew Tomkins, Patrick Vaughan, S. Vidwans, F. Vinicor, and Diana Weil. The global burden of disease study was directed by The methodology used for this study drew in part Christopher Murray and Alan D. Lopez. The results on the established literature on quality-adjusted life reported here come from Murray and Lopez, back- years (see Torrance 1986). One of the first applica- ground paper; a much expanded discussion will ap- tions to developing countries was Ghana Health As- pear in Murray and Lopez forthcoming (a). Contribu- sessment Project Team 1981. tions from the United Kingdom were coordinated by Jonathan Broomberg. Substantial contributions and comments on spe- Background papers cific diseases and injuries came from the following individuals: Carla AbouZahr, Mike Adams, Paul Ar- Bobadilla, José-Luis, Peter Cowley, Helen Saxenian, and thur, Robert Ashley, Kenneth Bailey, David Barmes, Philip Musgrove. "The Essential Package of Health Ser- L. Barnes, Robert Beaglehole, Mark Belsey, Stephen vices in Developing Countries." Berman, Barry Bloom, M. Blossner, Loretta Brabin, Cochrane, Susan, and Thomas W. Merrick. "Improving Ma- Donald Bundy, A. Burton, P. D. Cattand, Jacqueline ternal and Child Health through Family Planning Cattani, Chen Chunming, Caroline J. Cook, Edward Services." Cooper, P. M. P. Desjeux, Jacques Ferlay, 1. Fomey, Hecht, Robert M., and Vito L. Tanzi. "The Role of NGOs in the Delivery of Health Services in Developing Jean-Claude Funck, Michel Garenne, Tore Godal, Countries." Anne Goerdt, Johnathan Gorstein, Sandra Gove, Hill, Kenneth, and Abdo Yazbeck. "Trends in Child Mortal- Ramesh Govindaraj, Walter Gulbinat, Ivan Gyarfas, ity, 1960-90: Estimates for 84 Developing Countries." Lori Heise, Larry Heligman, Joachim Hempel, Em- Hill, Kenneth, Dean T. Jamison, Lawrence J. Lau, Jee-Peng manuel Jimenez, Mark Kane, Patrick Kenya, Dean Tan, and Abdo Yazbeck. "The Impact of Health Status on Kilpatrick, Hilary King, Jeffrey Koplan, Marie-Helene Economic Growth." Lau, Lawrence, Abdo Yazbeck, Kenneth Hill, Dean T. Jam- Leclerq, Linda Lloyd, Julian Lobb-Levyt, Luis Lopez ison, and Jee-Peng Tan. "Sources of Child Health Gains Bravo, David Mabey, Prasanta Mahapatra, Paul since the 1960s: An International Comparison." McKeigue, Graham Medley, Edwin Michael, Cath- Michaud, Catherine, and Christopher Murray. "Aid Flows erine Michaud, Kenneth Mott, A.-D. Negrel, Mag- to the Health Sector in Developing Countries." daline Orzeszyna, Max Parkin, Richard Peto, P. Pi- Murray, Christopher, and Alan D. Lopez. "The Global Bur- sani, Jenny Pronczuk, E. Pupulin, Xinjian Qiao, Ravi den of Disease in 1990." Murray, Christopher, Ramesh Govindaraj, and G. Rannan-Eliya, Hans Remme, Jean-Marie Robine, Chellaraj. "Global Domestic Expenditures in Health." Claude J. Romer, Richard Rothenberg, Peter Sand- Murray, Christopher, Jay Kreuser, and William Whang. iford, Elizabeth Sherwin, Alan Silman, Buranaj "Cost-Effectiveness Model for Allocating Health Sector Smutharaks, Jan Stjernsward, Rand Stoneburner, Resources." 182 Pritchett, Lant, and Lawrence H. Summers. "Wealthier Is Amadeo, Jesus, Dov Chernichovsky, and Gabriel Ojeda. Healthier." 1991. "The Profamilia Family Planning Program, Colom- Yazbeck, Abdo, Jee-Peng Tan, and Vito L. Tanzi. "Public bia." Policy, Research, and External Affairs Working Pa- Spending on Health in the 1980s: The Impact of Adjust- per Series 759. World Bank, Population and Human Re- ment Lending Programs." sources Department, Washington, D.C. Anand, Sudhir, and Martin Ravallion. 1993. "Human De- Selected bibliography velopment in Poor Countries: On the Role of Private In- comes and Public Services." Journal of Economic Perspec- tives 7(1): 133-50. Aaron, Henry J. 1991. Serious and Unstable Condition: Financ- ing America's Health Care. Washington, D.C.: Brookings Anderson, H. R. 1979. "Chronic Lung Disease in the Papua Institution. New Guinea Highlands." Thorax 34: 647-53. Abel-Smith, Brian. 1986. "The World Economic Crisis, Part Andersson, Fredrik. 1992. "Norway's Pivotal Role in Regu- 2: Health Manpower out of Balance." Health Policy and latory Health Economics and Pharmaceutical Cost Con- Planning 1(4): 309-16. tainment-What Can Other Countries Learn?" Health Pol- icy 21: 17-34. 1992a. "Cost Containment and New Priorities in the European Community." Milbank Memorial Fund Quar- Andreoni, Diego. 1986. "The Cost of Occupational Acci- terly 70(3): 393-416. dents and Diseases." Occupational Safety and Health Se- ries 54. International Labour Office, Geneva. 1992b. Cost Containment and New Priorities in Health Care. Aldershot, U.K.: Ashgate Publishing. Archavanitkui, K., and A. Pramaualrantan. 1990. "Factors Affecting Women's Health in Thailand." Paper presented 1992c. "Health Insurance in Developing Coun- tries: Lessons from Experience." Health Policy and Plan- at the Workshop on Women's Health in Southeast Asia, ning 7(3): 215-26. Population Council, Jakarta, October 29-31. Arrow, Kenneth. 1963. "Uncertainty and the Welfare Eco- Abel-Smith, Brian, and Ajay Dua. 1988. "Community Fi- nomics of Medical Care." American Economic Review 53: nancing in Developing Countries: The Potential for the 941-73. Health Sector." Health Policy and Planning 3(2): 95-108. Ballance, Robert, Janos Pogany, and Helmut Forstner. 1992. AbouZahr, Carla, and Erica Royston, comps. 1991. Maternal The World's Pharmaceutical Industries: An International Per- Mortality: A Global Fact book. Geneva: World Health Organization. Published for spective on Innovation, Competition, and Policy. the United Nations Industrial Development Organization. ACC!SCN (Administrative Committee on Coordination! Aldershot, U.K.: Edward Elgar Publishing. Subcommittee on Nutrition). 1991. Controlling Iron Defi- ciency. Nutrition Policy Discussion Paper 9. Geneva: Barcellos, Tanya, Carmen Barros, Naia Oliveira, Vera Gar- World Health Organization. cia, Laureane Pastore, and Mercedes Rabelo. 1986. "Seg- 1992a. Nutrition and Population Links: Breastfeeding, regacao Urbana e Mortalidade em Porto Alegre." Fun- Family Planning and Child Health. Nutrition Policy Discus- dacao de Economia e Estatistica, Porto Alegre, Brazil. sion Paper 11. Geneva: World Health Organization. Barlow, Robin, and Lisa M. Grobar. 1985. "Costs and Bene- fits of Controlling Parasitic Diseases." World Bank, Popu- 1992b. Second Report on the World Nutrition Situa- tion. Vol. 1: Global and Regional Results. Geneva: World lation, Health, and Nutrition Department, Washington, D.C. Health Organization. Achutti, Aloysio, B. B. Duncan; Luis Ruiz, and M. I. Barnum, Howard. 1987. "Evaluating Health Days of Life Gained from Health Projects." Social Science and Medicine Schmidt. 1988. "Programas de Saüde do Adulto." Paper 24(10): 833-41. prepared for the World Bank Study on Adult Health in Brazil. World Bank, Washington, D.C. Barnum, Howard, and Joseph Kutzin. 1993. Public Hospitals in Developing Countries: Resource Use, Cost, Financing. Balti- Adair, L., Barry M. Popkin, J. VanDerslice, John S. Akin, more, Md.: Johns Hopkins University Press. David K. Guilkey, R. Black, John Briscoe, and W Flieger. 1993. "Growth Dynamics during the First Two Years of Barr, Nicholas. 1992. "Economic Theory and the Welfare State: A Survey and Interpretation." Journal of Economic Life: A Prospective Study in the Philippines." European Literature 30 (June): 741-803. Journal of Clinical Nutrition 47: 42-51. Barros, F. C., and others. 1986. "Why So Many Cesarean Ainsworth, Martha, and Mead Over. 1992. "The Economic Sections: The Need for Further Policy Change in Brazil." Impact of AIDS: Shocks, Responses, and Outcomes." Health Policy and Planning 1(19). Technical Working Paper 1. World Bank, Africa Technical Barss, P., G. S. Smith, D. Mohan, and S. P. Baker. Forth- Department, Washington, D.C. coming. Injuries in Adults in Developing Countries: Epide- Akin, John S., David K. Guilkey, Charles C. Griffin, and miology and Policy. New York: Oxford University Press. Barry M. Popkin. 1985. The Demand for Primary Health Ser- Baumol, William J. 1992. "Private Affluence, Public Squa- vices in the Third World. Totowa, N.J.: Roman and br." C. V. Starr Center for Applied Economics, New York. Allanheld. Beaton, George H., and Hossein Ghassemi. 1987. "Supple- Alderman, Harold, Jere Behrman, Shahrukh Khan, David mentary Feeding Programs for Young Children in Devel- Ross, and Richard Sabot. 1992. "Public Schooling Expen- oping Countries: A Summary of Lessons Learned." In J. ditures in Rural Pakistan: Efficiently Targeting Girls and a Price Gittinger, Joanne Leslie, and Caroline Hoisington, Lagging Region." Paper presented at a World Bank con- eds., Food Policy: Integrating Supply, Distribution, and Con- ference on Public Expenditure on the Poor, July 17-19, sumption. Baltimore, Md.: Johns Hopkins University Washington, D.C. Press. Alonso, P. L., S. W. Lindsay, J. R. M. Armstrong, M. Con- teh, P. H. David, G. Fegan, A. de Francisco, A. J. Hall, F. C. Shenton, K. Cham, and B. M. Greenwood. 1991. "The Effect of Insecticide Treated Bed-Nets on Mortality of Gambian Children." Lancet 337: 1499-1515. 183 Beaton, George H., R. Martorell, K. A. L'Abbe, B. Ed- Birdsall, Nancy, and William McGreevey. 1983. "Women, monston, G. McCabe, A. C. Ross, and B. Harvey. 1993. Poverty and Development." In Mayra Buvinic, Margaret 'Effectiveness of Vitamin A Supplementation in the Con- Lycette, and William Paul McGreevey, eds., Women and trol of Young Child Morbidity and Mortality in Develop- Poverty in the Third World. Baltimore, Md.: Johns Hopkins ing Countries." A project of the International Nutrition University Press. Program, Department of Nutritional Sciences, University Bitran, Ricardo A., and D. Keith Mclnnes. 1993. The Demand of Toronto. for Health Care in Latin America: Lessons from the Dominican Behrman, Jere R. 1990. The Action of Human Resources and Republic and El Salvador. Economic Development Institute Poverty on One Another: What We Have Yet to Learn. Living Seminar Paper 46. Washington, D.C.: World Bank. Standards Measurement Study Working Paper 74. Wash- Black, Douglas, J. N. Morris, Cyril Smith, Peter Townsend, ington, D.C.: World Bank. and Nick Davidson, eds. 1982. The Black Report. New 1992. "The Effect of Structural Adjustment on York: Penguin Books. Food Policy and Nutrition." Paper presented at the Inter- Black, Robert E. 1991. "Would Control of Childhood Infec- national Union for the Scientific Study of Population tious Diseases Reduce Malnutrition?" Acta Paediatrica (IUSSP) seminar on the Demographic Consequences of Scandinavica (Supplement) 374: 133-40. Structural Adjustment in Latin America, Belo Horizonte, Blendon, Robert J., R. Leitman, 1. Morrison, and K. Don- Brazil, September 29.-October 2. elan. 1990. "Satisfaction with Health Systems in 10 Na- Behrman, jere R., David Ross, Richard Sabot, and Matthew tions." Health Affairs 9(2): 185-92. Tropp. 1991. "Improving the Quality versus Increasing Bloom, Barry R., and Christopher J. L. Murray. 1992. "Tu- the Quantity of Schooling." University of Pennsylvania, berculosis: Commentary on a Reemergent Killer." Science Philadelphia, Pa. 257(21 August): 1055-64. Bellinger, D., A. Leviton, C. Waternaux, H. Needleman, Bloom, David E., and Joyce V. Lyons, eds. 1993. Economic and M. Rabinowitz. 1987. "Longitudinal Analyses of Pre- Implications of AIDS in Asia. UNDP Regional Progamme natal and Postnatal Lead Exposure and Early Cognitive Division (Regional Bureau for Asia and the Pacific). New Development." New England Journal of Medicine 316: Delhi: HIV/AIDS Regional Project. 1037-43. Bloom, Gerald. 1989. "The Right Equipment. . . in Working Benefo, Kofi, and T. Paul Schultz. 1992. "Fertility and Child Order." World Health Forum 10(1): 3-10. Mortality in Côte d'Ivoire and Ghana." World Bank, Afri- Blum, D., R. N. Emeh, S. R. A. Huttly, 0. Dosunmu- can Technical Department, Population, Health, and Nu- Ogunbi, N. Okeke, M. Ajala, J. I. Okoro, C. Akujobi, B. trition Division, Washington, D.C. R. Kirkwood, and R. G. Feachem. 1990. "The lmo State Bennett, Sara. 1992. "Promoting the Private Sector: A Re- Nigeria Drinking Water Supply and Sanitation Project. 1. view of Developing Country Trends." Health Policy and Description of the Project, Evaluation Methods, and Im- Planning 7(2): 97-110. pact on Intervening Variables." Transactions of the Royal Berg, Alan. 1987. Malnutrition: What Can Be Done? Lessons Society of Tropical Medicine and Hygiene 84: 309-15. from World Bank Experience. Baltimore, Md.: Johns Hopkins Bobadilla, José-Luis, Julio Frenk, Tomas Freijka, Rafael University Press. Lozano, and Claudio Stern. Forthcoming. "The Epidem- Berg, Elliot, and Graeme Hunter. 1992. "Social Costs of Ad- iological Transition and Health Priorities." In Dean T justment: The Case of Latin America and the Caribbean." Jamison, W. Henry Mosley, Anthony R. Measham, and Prepared for the U.S. Agency for International Develop- José-Luis Bobadilla, eds., Disease Control Priorities in Devel- ment. Development Alternatives, Inc., Bethesda, Md. oping Countries. New York: Oxford University Press. Berkley, Seth, and Dean jamison. 1991. "Summary Re- Bobak, M. 1993. "Air Pollution and Health in Central and port." Conference on the Health of School-Age Children, Eastern Europe: How Bad Is the Problem?" London sponsored by the United Nations Development Pro- School of Hygiene and Tropical Medicine. gramme and the Rockefeller Foundation, Bellagio, Italy, Bobak, M., and R. G. A. Feachem. 1992. "Health Status in August 12-16. the Czech and Slovak Federal Republic." Health Policy and Bertrand, Jane T., and Judith Brown. 1992. "Family Plan- Planning 7: 234-42. ning Success in Two Cities in Zaire." Policy, Research, Bobak, M., and D. A. Leon. 1992. "Air Pollution and Infant and External Affairs Working Paper Series 1042. World Mortality in the Czech Republic, 1986-88." Lancet 340: Bank, Population and Human Resources Department, 1010-14. Washington, D.C. Boissiere, M., J. B. Knight, and R. H. Sabot. 1985. "Earn- Bhargava, Alok. 1992. "Malnutrition and the Role of Indi- ing, Schooling, Ability and Cognitive Skills." American vidual Variation with Evidence from India and the Philip- Economic Review 75: 1026-30. pines." Journal of the Royal Statistical Society 155. Boring, Catherine C., Teresa S. Squires, and Tony Tong. Bhargava, Alok, and hang Yu. 1992. "A Longitudinal Analy- 1993. "Cancer Statistics 1993." CA-Cancer Journal for Cli- sis of Infant and Child Mortality Rate in Africa and Non- nicians 43 (Jan-Feb.): 7-26. African Developing Countries." Background paper pre- Bos, Eduard, My T. Vu, Ann Levin, and Rodolfo A. Bulatao. pared for the Africa Health Study. World Bank, Africa 1992. World Population Projections 1992-93 Edition: Estimates Technical Department, Washington, D.C. and Projections with Related Demographic Statistics. Balti- Birdsall, Nancy. Forthcoming. "Pragmatism, Robin Hood, more, Md.: Johns Hopkins University Press. and Other Themes: Good Government and Social Well- Bouis, Howarth E. 1990. "The Determinants of Household- Being in Developing Countries." In Lincoln C. Chen, Ar- Level Demand for Micronutrients: An Analysis for Philip- thur Kleinman, and Norma C. P. Ware, eds., Health and pine Farm Households." International Food Policy Re- Social Science in International Perspective. Harvard Series on search Institute, Washington, D.C. Population and International Health. Cambridge, Mass.: Harvard University Press. 184 Bourdon, Jean, François Orivel, and Jean Perrot. 1993. Cleland, John. 1990. "Maternal Education and Child Sur- "Peut-on Apprecier un Lien entre les Acquisitions vival: Further Evidence and Explanations." In John Cald- Scolaires et Ia Sante: Le Cas de l'Enseignement Primaire well, Sally Findley, Pat Caldwell, Gigi Santow, Wendy au Tchad." Paper presented at meetings of the Associa- Cosford, Jennifer Braid, and Daphne Broers-Freeman, tion Francais de Sciences Economiques, Dijon. May eds., What We Know about Health Transition: The Cultural, 27-28. Social, and Behavioural Determinants of Health. Health Tran- Bradley, Christine. 1988. "How Can We Help Rural Beaten sition Series, Vol. 1. Canberra: Australian National Wives? Some Suggestions from Papua New Guinea." Pa- University. per presented at International Women's Aid Conference, Coale, Ansley J., and Paul Demeny, with Barbara Vaughan. Cardiff, Wales. 1983. Regional Model Life Tables and Stable Populations. 2d Bradley, David John, Sandy Cairncross, Trudy Harpham, ed. New York: Academic Press. and Carolyn Stephens. 1992. A Reviezt' of Environmental Comanor, William S. 1986. "The Political Economy of the Health Impacts in Developing Country Cities. Urban Manage- Pharmaceutical Industry." Journal of Economic Literature 24 ment Program Paper 6. Washington, D.C.: World Bank. (September): 1178-1217. Brenzel, Logan. 1990. 'The Costs of EPI: Lessons Learned Commission on Health Research for Development. 1990. from Cost and Cost-Effectiveness Studies of Immuniza- Health Research: Essential Link to Equity in Development. tion Programs." Resources for Child Health (REACH) New York: Oxford University Press. Project. John Snow, Inc., Arlington, Va. Conly, G. N. 1975. The Impact of Malaria on Econo,nic Develop- Briscoe, John. 1992. "Poverty and Water Supply: How to ment: A Case Study. Scientific Publication 297. Washing- Move Forward." Finance and Development 29: 16-19. ton, D.C.: Pan American Health Organization. Brook, Robert H., and Kathleen N. Lohr. 1986. "Will We Council on Scientific Affairs, American Medical Association. Need to Ration Effective Health Care?" Issues in Science 1992. "Violence against Women: Relevance for Medical and Technology 3(1): 68-77. Practitioners." JAMA: Journal of the American Medical Asso- Brotowasisto, Oscar Gish, Ridwan Malik, and Paramita ciation 267(23): 3184-89. Sudharto. 1988. "Health Care Financing in Indonesia." COVAC. 1990. "Evaluación de Proyecto para Educación, Health Policy and Planning 3(2): 131-40. Capacitación, y Atención a Mujeres y Menores de Edad en Bruce, Judith, and Cynthia B. Lloyd. 1992. "Finding the Ties Materia de Violencia Sexual, Enero a Diciembre 1990." That Bind: Beyond Headship and Household." Working Asociación Mexicana Contra Ia Violencia a las Mujeres, Paper 41. Population Council, New York. Mexico City. Bruley, Mark E. 1991. "Global Trends in Healthcare Tech- Cumper, George, and Patrick Vaughan. 1985. "Community nology." Keynote address at Health-Tech 91: Health Tech- Health Aides at the Crossroads," World Health Forum 6: nology in Developing Countries: Assessment, Procure- 365-67. ment, and Maintenance, Yaoundé, Cameroon, March 11. Curtis, C. F. 1992. "Spraying Bed-Nets with Deltamethrin in ECRI, Plymouth Meeting, Pa. Sichuan, China." Tropical Diseases Bulletin 89(8): R1-R6. Cairncross, A. M. 1990. "Health Impacts in Developing Davis, Karen, Gerard F. Anderson, Diane Rowland, and Countries: New Evidence and New Prospects." Journal of Earl P. Steinberg. 1990. Health Care Cost Containment. the Institution of Water and Environmental Management 4(6): Johns Hopkins Studies in Health Care Finance and Ad- 571-77. ministration. Baltimore, Md.: Johns Hopkins University Cairncross, Sandy. 1992. "Sanitation and Water Supply: Press. Practical Lessons from the Decade." UNDP-World Bank De Geyndt, Willy. 1991. Managing Health Expenditures under Water and Sanitation Discussion Paper Series 9. World National Health Insurance: The Case of Korea. World Bank Bank, Washington, D.C. Technical Paper 156. Washington, D.C. Caldwell, J. C. 1986. "Routes to Low Mortality in Poor Diamond, Peter. 1992. "Organizing the Health Insurance Countries." Population and Development Review 12(2): Market." Econometrica 60(6): 1233-54. 171-220. Doll, Richard. 1992. "Health and the Environment in the CAMVAC (Centro de Apoyo a Mujeres Violadas). 1985. 1990s." American Journal of Public Health 82: 933-43. "Carpeta de Información Básica para Ia AtenciOn Soli- Downing, A. J. 1991. "Pedestrian Safety in Developing dana y Feminista a Mujeres Violadas." Mexico City. Countries." In Proceedings of the Vulnerable Road User: Inter- Caplan, David. 1985. "Consumption, Production, World national Conference on Traffic Safety, New Delhi, January Trade and Industry Structure." World Bank, Industry and 27-30. New Delhi: Macmillan India. Energy Department, Washington, D.C. Drèze, Jean, an