Document of The World Bank Report No: ICR0000576 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-37310 IDA-H0190) ON A DEVELOPMENT CREDIT IN THE AMOUNT OF SDR 43.5 MILLION (US$ 57.6 MILLION EQUIVALENT) AND DEVELOPMENT GRANT IN THE AMOUNT OF SDR 24.5 MILLION (US$ 32.4 MILLION EQUIVALENT) TO THE REPUBLIC OF GHANA FOR A HEALTH SECTOR PROGRAM SUPPORT PROJECT II December 21, 2007 Human Development Sector Management Unit Ghana Country Management Unit Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective December 21, 2007) Currency Unit = Cedis Cedis 1 = US$0.0001079 US$1.00 = Cedis 8375.0 FISCAL YEAR January 1 ­ December 31 ABBREVIATIONS AND ACRONYMS ADHA Additional Duty Hours Allowance ICR Implementation Completion Report AIDS Acquired Immune Deficiency IDA International Development Syndrome Association ANC Antenatal care IGF Internally generated funds BMC Budget Management Center M&E Monitoring and evaluation CAS Country Assistance Strategy MMR Maternal mortality ratio CHAG Christian Health Association of MOFEP Ministry of Finance and Economic Ghana Planning CHPS Community-based Health Planning MOH Ministry of Health Services NGO Non-governmental organization CSO Civil society organization NHIS National Health Insurance Scheme DAIA Deprived Area Incentive Allowance ORT Oral re-hydration therapy DCA Development Credit Agreement PAD Project Appraisal Document DFA Development Financing Agreement PDO Project Development Objective DHS Demographic and Health Survey PNC Prenatal care DP Development partner POW Program of Work GHS Ghana Health Service PRSC Poverty Reduction Strategy Credit GOG Government of Ghana SWAp Sector-wide approach GPRS Ghana Poverty Reduction Strategy TB Tuberculosis HIV Human Immunodeficiency Virus TTL Task Team Leader IBRD International Bank for U-5MR Under-five mortality rate Reconstruction and Development Vice President: Obiageli Katryn Ezekwesili Country Director: Ishac Diwan Sector Manager: Eva Jarawan Project Team Leader: Laura L. Rose ICR Team Leader: Laura L. Rose GHANA Health Sector Program Support Project II (HSPSP-II) CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph Executive Summary........................................................................................................ 1 1. Project Context, Development Objectives and Design............................................... 8 2. Key Factors Affecting Implementation and Outcomes ............................................ 13 3. Assessment of Outcomes.......................................................................................... 20 4. Assessment of Risk to Development Outcome......................................................... 37 5. Assessment of Bank and Borrower Performance ..................................................... 38 6. Lessons Learned ....................................................................................................... 42 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners .......... 43 Annex 1. Project Costs and Financing.......................................................................... 44 Annex 3. Economic and Financial Analysis................................................................. 55 Annex 4. Bank Lending and Implementation Support/Supervision Processes ............ 56 Annex 5. Beneficiary Survey Results........................................................................... 57 Annex 6. Stakeholder Workshop Report and Results................................................... 58 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR..................... 59 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders....................... 69 Annex 9. List of Supporting Documents ...................................................................... 70 Annex 10. Sector-wide Indicator Comparison Table .................................................. 75 Annex 11. Sector-wide Indicators 1998 ­ 2006 .......................................................... 80 MAP Project Data A. Basic Information Second Health Sector Country: Ghana Project Name: Program Support Project Project ID: P073649 L/C/TF Number(s): IDA-37310,IDA-H0190 ICR Date: 12/21/2007 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: SIL Borrower: GHANA Original Total XDR 68.0M Disbursed Amount: XDR 68.0M Commitment: Environmental Category: B Implementing Agencies: Ministry of Health, Ghana Cofinanciers and Other External Partners: B. Key Dates Process Date Process Original Date Revised / Actual Date(s) Concept Review: 10/03/2001 Effectiveness: 05/22/2003 05/22/2003 Appraisal: 11/25/2002 Restructuring(s): Approval: 02/06/2003 Mid-term Review: 12/31/2005 03/31/2006 Closing: 06/30/2007 06/30/2007 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately Unsatisfactory Risk to Development Outcome: Moderate Bank Performance: Moderately Unsatisfactory Borrower Performance: Moderately Unsatisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Moderately Unsatisfactory Government: Moderately Satisfactory Quality of Supervision: Moderately SatisfactoryImplementing Moderately Agency/Agencies: Unsatisfactory Overall Bank Moderately Overall Borrower Moderately Performance: Unsatisfactory Performance: Unsatisfactory 1 C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Performance Indicators (if any) Rating Potential Problem Project No Quality at Entry Satisfactory at any time (Yes/No): (QEA): Problem Project at any Quality of Yes None time (Yes/No): Supervision (QSA): DO rating before Satisfactory Closing/Inactive status: D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Health 100 100 Theme Code (Primary/Secondary) Child health Secondary Secondary HIV/AIDS Secondary Secondary Health system performance Primary Primary Population and reproductive health Secondary Secondary Public expenditure, financial management and Secondary Secondary procurement E. Bank Staff Positions At ICR At Approval Vice President: Obiageli Katryn Ezekwesili Callisto E. Madavo Country Director: Ishac Diwan Mats Karlsson Sector Manager: Eva Jarawan Alexandre V. Abrantes Project Team Leader: Laura L. Rose Francois Decaillet ICR Team Leader: Laura L. Rose ICR Primary Author: David H. Peters Jessica St. John 2 F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The overall objective of the 2002-2006 MOH Program of Work (POW-II) is to improve the health status of the population while reducing the geographical, socioeconomic and gender inequalities in health outcomes. This will be achieved by improving access, quality, and efficiency of health services. Specific elements of the strategy include enhancing the existing infrastructure and support services; reforming the organization and financing of the health system and increasing overall financing for the sector; strengthening management information systems, and in particular improving financial managment and monitoring of performance of the system; developing human resources for the sector; strengthening links with traditional and alternative medicine; and fostering partnerships with other stakeholders. Revised Project Development Objectives (as approved by original approving authority) (a) PDO Indicator(s) Original Target Formally Actual Value Indicator Baseline Value Values (from Revised Achieved at approval Target Completion or documents) Values Target Years Indicator 1 : Infant mortality rate (IMR) Value quantitative or 56.7/1000 50/1000 71/1000 Qualitative) Date achieved 12/31/1998 06/30/2007 12/31/2006 Comments (incl. % As explained in the main text, this was not an appropriate indicator to have been achievement) chosen to measure progress towards the PDO Indicator 2 : Under 5 mortality rate (U5MR) Value quantitative or 108 95 111 Qualitative) Date achieved 12/31/1998 06/30/2007 12/31/2003 Comments (incl. % achievement) 3 (b) Intermediate Outcome Indicator(s) Original Target Formally Actual Value Indicator Baseline Value Values (from Achieved at approval Revised Target Completion or documents) Values Target Years Indicator 1 : Supervised deliveries Value (quantitative 44% 60% 44.5 or Qualitative) Date achieved 12/31/1998 12/31/2006 12/31/2006 Comments (incl. % achievement) Indicator 2 : % of GOG recurrent budget spent on health Value (quantitative 11% 15% 18% or Qualitative) Date achieved 12/31/2003 12/31/2006 12/31/2006 Comments (incl. % achievement) Indicator 3 : Outpatient visit per capita Value (quantitative 0.49 0.6 .52 or Qualitative) Date achieved 06/30/2001 12/31/2006 12/31/2006 Comments (incl. % achievement) G. Ratings of Project Performance in ISRs Actual No. Date ISR Archived DO IP Disbursements (USD millions) 1 06/30/2003 Satisfactory Satisfactory 0.00 2 08/20/2003 Satisfactory Satisfactory 8.35 3 01/30/2004 Satisfactory Satisfactory 8.35 4 10/12/2004 Unsatisfactory Satisfactory 31.61 5 10/27/2004 Unsatisfactory Satisfactory 46.89 6 01/27/2005 Satisfactory Satisfactory 49.95 7 11/16/2005 Satisfactory Satisfactory 94.32 8 12/23/2005 Satisfactory Satisfactory 98.28 9 06/29/2006 Satisfactory Moderately Satisfactory 100.21 10 02/13/2007 Satisfactory Satisfactory 100.26 11 06/29/2007 Moderately Unsatisfactory Moderately Satisfactory 100.72 4 H. Restructuring (if any) Not Applicable I. Disbursement Profile 5 Executive Summary This Implementation Completion Report evaluates the Government of Ghana's 2002-2006 Health Sector Program of Work (POW-II) and World Bank support to this sector wide approach (SWAp) through the Second Health Sector Support Project (HSPSP-II). The POW-II outlined a vision to improve the health status of all Ghanaians while reducing the geographical, socioeconomic and gender inequalities in health outcomes, building on strategic pillars to improve access, quality, efficiency, partnerships, and financing of the health sector. Ghana's progress towards realizing the vision of the POW-II has been mixed, in contrast to the high expectations. Health status was tracked by measurements of infant and under-five mortality and childhood malnutrition. Overall reductions in infant and under-five mortality have stagnated, though most malnutrition indicators have shown substantial improvements. Disparities in mortality have been significantly reduced for the poor compared to less poor and for rural compared to urban groups. The majority of malnutrition indicators also show improvements for the poor and a reduction in relative inequality. Most of the lost ground in mortality occurs because of recent increases in neonatal deaths (under 1 month of age), particularly for wealthy and urban groups. Whereas this finding raises questions about the validity of the mortality data used to track progress and the need for better measurement, it also calls for more focused attention on maternal and neonatal health. There were important gains in achieving the strategic objectives, notably related to improvements in access to health services, equity of services, and the introduction of the National Health Insurance Scheme (NHIS). Changes in overall access to health services and equity of access during the POW-II have been largely positive, with most indicators of access showing an increase or maintenance of high levels. There is less information concerning changes in the equity of health service delivery, but the available data show that there have been reductions in Regional disparities in access to some priority health services (e.g. immunizations, skilled birth attendance and family planning), though for other indicators the results are less clear (e.g. outpatient and hospital use). The data on changes in disparities of health services access for the poor also show a largely positive picture. There were reductions in disparities for most health services that could be measured (use of oral rehydration therapy (ORT) for diarrhea, DPT3 immunization coverage, antenatal care coverage, and maternal tetanus toxoid vaccination coverage), but not for skilled birth attendance and family planning use. There were also considerable shortcomings in some areas of strategic importance, notably those related to quality of services, efficiency, and partnerships. Despite the importance of quality of health services delivery, there was a lack of focus on strategies for institutionalizing quality improvement, and little attention paid to tracking changes in quality of services. Similarly, efficiency in the health sector was not well monitored despite being a strategic pillar of the POW- II. There were large increases in public expenditures on health worker salaries that may have slowed down the exodus of health workers from Ghana, though studies of health workforce productivity demonstrated declines in productivity, and data on hospital occupancy showed substantial unused capacity. At the same time, the expansion of community services and delivery of priority interventions were hampered by a lack of funding for recurrent expenditures, while capital expenditures did not reflect stated priorities and were not well managed. In the area of partnerships, there were gains made in working with donors and developing common management arrangements, though there was limited progress in collaborations with the private health sector and communities. The re-launching of the Ghana Health Service (GHS) in 2003 also 6 led to institutional conflict and duplication of efforts with the Ministry of Health, which diverted attention and resources away from the task of implementing the POW-II. The report also highlights some important lessons for future work. One lesson is that a stronger analytic base is needed even when things appear to be going well, as was the case when the POW-II was initiated. In-depth analysis and debate concerning equity of health services and outcomes, efficiency of health services, and institutional analyses would have been particularly helpful in anticipating problems and finding ways to assure implementation of the POW-II. Donor behavior is also important, and the report notes that even shifts to general budget support by some donors can contribute to disruptions in financing, increases in off-budget project funding, and an overall increase in fragmentation of health sector support. There were design flaws that were not well addressed during the POW-II, and the Bank's conceptualization of investment lending helped to reinforce some of the design flaws. The dialogue between MOH and development partners spent considerable time dealing with unmet expectations of inappropriately defined project development objectives, even though the expected results could not have been measured during the POW-II period, much less have been attributable to the POW-II. There was also inordinate attention on results that were not under control of the health sector, using a monitoring and evaluation system that was not sufficiently aligned with strategic priorities, limiting the opportunity to move towards a performance-based health system. These factors can seriously undermine achievement of sector objectives, and require attention to technical design, institutional and political factors. 7 1. Project Context, Development Objectives and Design (this section is descriptive, taken from other documents, e.g., PAD/ISR, not evaluative) 1.1 Context at Appraisal (brief summary of country and sector background, rationale for Bank assistance) Since embarking on economic reforms in the mid-1980s, Ghana has made considerable progress in laying the foundation for sustainable growth and poverty reduction. This has resulted in sustained per capita growth and increased private sector activity and investment. Social indicators have also improved. In parallel with the economic reforms, the country completed the political transition, moving to a democratic reform of Government. Overall, progress in social and economic developments has been compromised by periodic interruptions of weak macroeconomic management associated with the electoral cycle. With improved fiscal discipline and a growing economy, improvements in the country's macroeconomic performance have been anticipated. Given this background, the Second Health Sector Program Support Project (HSPSP-II) fit clearly within the context of the Bank's Country Assistance Strategy (CAS) of 2000, which was designed to support the Government of Ghana's (GOG) strategy to reduce poverty. The need to provide sector-wide support to the health sector was highlighted as a key part of that strategy. After elections in 2000, a new Government took office in January 2001, adopted the HIPC initiative, and formulated a new Ghana Poverty Reduction Strategy (GPRS) which further reinforced the approach. The GPRS aimed to eliminate hard-core poverty, with a specific strategy to redefine the role of the state to provide public goods and services and ensure equitable distribution of those benefits. In 2003, the GPRS highlighted improving human service delivery as one of its five strategic pillars. Health-specific components included: (a) expanding access to health services and enhancing quality; (b) improving the efficiency and equity of health services; and (c) ensuring sustainable financing arrangements that protect the poor. In 2004, the Bank's strategy to support the GPRS was articulated in the CAS as focusing on: (a) sustainable growth and jobs creation; (b) service provision for human development; and (c) governance and empowerment. The Government of Ghana and the World Bank have been working together in the health sector since 1986. This includes a number of projects focused on improving health systems, including service delivery, management, and financing. During the implementation of the Second Health and Population Project, which was rated as having a positive impact, the GOG sought support from the World Bank in developing and implementing one of the first sector-wide approaches (SWAp) anywhere. This created a new way of doing business for the GOG and other development partners, involving a common policy and strategic framework, a coherent financing plan, and development of common implementation and review arrangements. The SWAp served to strengthen the Government's sector stewardship by developing institutional, management, and system capacity. In this way, by bolstering national capacity and leadership, the Government was better equipped to address national priorities and improve health outcomes. The Health Sector Program Support Project (HSPSP-I), which was also known as the Health Sector Support Program, operated from 1997 to 2001, and was the first SWAp in Ghana. Building on the perceived success Ghana was experiencing in improving overall national health indicators (e.g., under-five mortality and infant mortality rates) as well as health service indicators (e.g., percentage of family planning acceptors), the HSPSP-I was designed to improve institutional capacity and strengthen the sector to improve health outcomes, reduce inequalities, and improve service delivery and quality. HSPSP-I's stated objective was to support the GOG's Program of 8 Work (POW-I) for 1997-2001, which aimed to improve service delivery, quality, and efficiency, as well as foster intersectoral linkages to improve health outcomes, such as malnutrition, and other outcomes such as girls' education achievement. The HSPSP-I was considered successful in improving institutional capacity, particularly in improving procurement; budgeting, planning, and financial management; and improving health financing. The Second Health Sector Support Project (HSPSP-II) built on the HSPSP-I and supported the Government of Ghana's (GOG) second five-year Program of Work (POW-II) for 2002-2006. The Bank's decision to design another SWAp to support POW-II was reasonable, given the perceived successes of the first SWAp in capacity building, donor collaboration, and improving the predictability and availability of health financing through the pooled funds. 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) Project Development Objectives The Project Appraisal Document (PAD) describes the Project Development Objectives by indicating that the HSPSP-II will support the Government of Ghana's efforts towards implementing the Government's 2002-2006 Health Sector Program of Work (POW-II), described in Partnerships for Health: Bridging the Inequalities Gap. The POW-II outlined a vision statement and set of five strategic pillars (see Box 1), which was translated by the Bank in the PAD as an "overall objective...to improve the health status of the population while reducing the geographical, socioeconomic and gender inequalities in health outcomes." In addition, the POW- II's "five strategic pillars" were also identified in Annex 1 of the PAD as project development objectives. It would be inappropriate to consider a clearly defined vision statement alone as the benchmark for achievement of the project development objectives, even though this appears to be the case during implementation of the POW-II. In assessing the PDOs, this ICR attempts to be consistent with the intentions of the PAD and the POW-II, which involves assessing progress towards fulfillment of a vision, and the achievement of strategic pillars and their related objectives. Box 1. POW-II's Vision, Mission, Policy Goal, and Five Strategic Pillars Vision Improved overall health status and reduced inequities in health outcomes of people living in Ghana Mission statement "The Ministry of Health will work in collaboration with all partners in the health sector to ensure that every individual, household and community is adequately informed about health; and has equitable access to high quality health and related interventions." Policy goal Working together for equity and good health for people living in Ghana The five strategic pillars · To improve quality of health delivery · To increase access to health services · To improve the efficiency of health service delivery · To foster partnerships in improving health · To improve financing of the health sector 9 The PAD describes the five strategic pillars/objectives of the POW-II as the following: (i) to increase (geographical, financial, and sociocultural) access to health services; (ii) to improve quality of health delivery, improving health-worker-performance and responsiveness to client needs; (iii) to improve the efficiency of health service delivery, improving cost effectiveness and planning, management, and administration; (iv) to foster partnerships in improving health (with households and communities, between public and private providers, other ministries, departments, and agencies) with development partners; and (v) to improve financing of the health sector through increased financing and increasing financial access of the poor to health care by extending prepayments schemes to replace "cash and carry" systems, while developing an appropriate policy and regulatory environment for health insurance, as well as increasing public expenditure on the poor and vulnerable. Key Indicators The PAD is inconsistent in its identification of the key indicators for HSPSP-II. For instance, Annex 12 of the PAD makes reference to twenty-five sector-wide indicators that were identified for monitoring implementation of the HSPSP-II and POW-II (Table 1.5.1). However, twenty-nine sector-wide indicators are also listed in the same Annex, and additional project indicators and recorded in Annex 1 of the PAD. Although there is a great deal of overlap among these indicators, they are often worded differently and are thus inconsistently defined. The MOH and its development partners (including the World Bank) often refer to the "agreed upon twenty-five sector-wide indicators for POW-II." However, the number of the sector-wide indicators varies depending the document one refers to. Annex 10 highlights the different sector- wide indicators used in the PAD, the POW-II, the annual POW for 2006, and the Review of POW-2005. There were often more than twenty-five indicators and some of these indicators evolved over time. These indicators are largely based on the categories of the five strategic objectives (access, quality, efficiency, partnership, and financing). Because data is more comparable, this report uses the sector-wide indicators that were reported in the reviews of the annual POWs as the key indicators by which the HSPSP-II and POW-II are assessed. 10 Table 1.5.1 Sector-wide indicators for POW-II (as identified in the PAD) Health Status · Infant mortality rate · Under-five mortality rate · Maternal mortality rate · % under-five years who are malnourished Access · Number of outreach services carried out by specialist from tertiary, secondary and district hospital by region · Population to doctor and nurse ratio by region · Outpatient visit per capita · Hospital admission rate · Number of community resident nurse per district/region Quality · % of maternal audits to maternal deaths · Under-five malaria case fatality rate · % tracer drug availability Efficiency · HIV seroprevalence (among reproductive age, 15-19, 20-24) · Tuberculosis cure rate · Number of guinea worm cases · AFP non-polio rate · % family planning acceptors · % ANC coverage · % PNC coverage · % supervised deliveries (skilled attendants) · Bed occupancy rate · EPI coverage (DPT3, measles) Partnership · % recurrent budget from GOG and Health Fund used by private sector, NGOs, CSOs, and other MDAs Financing · % GOG budget spent on health · % GOG recurrent budget for health · % of earmarked/direct donor funds to total donor funds (per partner) · % of IGFs coming from pre-payment and community insurance scheme · % recurrent and capital expenditure by level, by region, and by source · Total amount spent on exemptions by exemption category Note: Sector-wide indicators are identified in Annex 12 of the Project Appraisal Document 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification The project development objectives and key indicators were not formally revised in Bank documents, though as discussed in Sections 1.2 and 2.3, the descriptions and indicators of the POW-II changed from year to year. 11 1.4 Main Beneficiaries (original and revised, briefly describe the "primary target group" identified in the PAD and as captured in the PDO, as well as any other individuals and organizations expected to benefit from the project) Based on the POW-II, the PAD identifies the main beneficiaries as: 1. All Ghanaians, who will benefit from improved management, financing, and quality of health care delivery; 2. The poor, particularly the rural poor and women, due the focus on priority health interventions, provision of basic health services, and reducing geographical, socioeconomic, and gender inequalities. 3. Ministry of Health (MOH), Ghana Health Service (GHS), and Budget and Management Center (BMC) staff, and those health workers in the non-state sector, due to capacity building activities. 1.5 Original Components (as approved) The HSPSP-II was designed to support the POW-II, which did not define project components. The definition of components was done by the Bank in order to fit its own design templates, but was done in an inconsistent way. The ICR structure also requires an assessment of the outputs from a set of components, even though this does not fit well with the design of the POW-II. The PAD and the Development Financing Agreement (DFA) highlight ten areas of focus that are consistent with the ten chapters of the POW-II that follow the POW-II definition of the strategic vision. These areas are also treated as components in the logical framework of the PAD and in the DFA. However, the PAD also identifies each year's overall Program of Work as a component when describing the financing components of the program. The description of focus areas changed substantially by the 2003 Program of Work, and continued to be modified annually. However, they were not treated as components in the POW-II, and they were never formally revised with the Bank. To be consistent with the Bank's evaluation methodology and the spirit of the program, outcomes and outputs of the program are assessed according to the five strategic pillars of the POW-II (described in Section 1.2), and the original ten focus areas of the DFA and PAD logical framework. The ten focus areas that will serve as "components" for this evaluation are: 1. Strengthening priority health interventions; 2. Developing human resources for health services; 3. Enhancing infrastructure and support services; 4. Fostering partnerships for health; 5. Improving regulation; 6. Reforming institutions arrangements; 7. Improving the health sector financing; 8. Improving financial management systems; 9. Further strengthening management; and 10. Linking with traditional medicine. 12 1.6 Revised Components The POW-II was not organized around project components. The components were an artifact of Bank documentation for the PAD of HSPSP-II (one set of components for describing the financing components, another set for the description of the program). They were not formally revised during implementation, even though the identified "components" changed considerably during implementation of the POW-II. 1.7 Other significant changes (in design, scope and scale, implementation arrangements and schedule, and funding allocations) The strategic objectives outlined in the POW-II set priorities that remained main themes throughout implementation of the project between 2002 and 2006. However, the Ministry of Health produced and implemented annual Programs of Work with its partners, which served as the main strategy document for the health sector for each respective year. The annual POWs generally maintained the strategic objectives outlined in the POW-II, but the articulation of the focus areas changed from year to year. This was particularly true of the articulation of activities and schemes planned to achieve objectives each year. The focus areas became the main thrust of the annual POWs, with the strategic objectives providing a general overview of the sector's direction. For instance, "human resources development" was treated as a part of "health sector investments" within each POW until 2006, when POW 2006 made "human resources" its own primary area of focus. An indisputably crucial change to the implementation arrangements during HSPSP-II was the re- launching of the Ghana Health Service in 2003. Under powerful and capable leadership, the GHS came into direct conflict with the MOH over the authority to make policy and management decisions related to service delivery. This conflict created gridlock at the central level and diverted attention and resources away from the task at hand--implementing the POW-II. The conflict also caused a great deal of confusion, including the failure to track capital expenditures and follow through on implementation arrangements with non-governmental providers, such as CHAG. In terms of funding allocations, the following trends have affected health sector financing: (a) establishment of the National Health Insurance Scheme in 2003 changed implementation arrangements, as the National Health Insurance Fund became another fiduciary mechanism to capture IGF and replace the exemptions policy; (b) the proportion of donors contributing to the Health Fund have increasingly moving to earmarked and off-budget project financing; and (c) some donors contributing to the Health Fund have moved to general budget support. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry (including whether lessons of earlier operations were taken into account, risks and their mitigations identified, and adequacy of participatory processes, as applicable) Soundness of background analysis The Second Health Sector Support Project (HSPSP-II) for Ghana was designed as a health SWAp, based on the perceived successes of HSPSP-I and the strong demonstrated leadership of the Ministry of Health. Early evaluations of the first health SWAp (HSPSP-I) documented marked 13 improvements in government system capacity, both at the regional and district level, particularly with regard to procurement and financial management systems. The first health SWAp achieved decentralization of financial resources with the certification of budget management centers (BMCs) and appeared to have a significant effect in improving health outcomes, particularly in priority areas. The Bank's decision to continue supporting the SWAp was based on the rationale of continuing to support a Government-led process of improving the health sector. Engaging in the SWAp allowed the Bank to continue its close collaboration with other core donors to build upon the progress of using national systems and procedures and strengthening institutional capacity to achieve health outcomes. One of the important outcomes of the POW-I was the documentation and political recognition of some important lessons for improvement. The overall objective of POW-II and its strategic pillars were based on these lessons and were devised to overcome shortcomings, namely: (a) the persistence of deep inequalities in health; (b) the chronic under-funding of the health sector; (c) the intense brain drain and unequal distribution of staff; (d) the untapped potential contribution from private providers; and (e) the potential devastating impact of HIV/AIDS on the heath sector, the economy. Notwithstanding these understandings of the limitations from the POW-I, the POW-II was largely accepted as a continuation as "business as usual." In retrospect, this should have been questioned more seriously. There was insufficient attention to the analytic basis for dealing with equity and efficiency dimensions of the program, and particularly on the institutions needed for implementation of strategies to address them. Although the potential for "turbulence" caused by the split of the MOH and GHS was recognized as a substantial risk, there was a clear need for institutional analyses and more pro-active risk reduction strategies that were not undertaken. The Bank promised to support the SWAp and pay particular attention to the areas of: i) Reducing inequalities in health--particularly to define and implement a more balanced capital development program with a clear focus on basic health services for the poorest and most deprived; and a new strategy to attract and retain health staff in the most deprived areas. ii) Improving physical and financial accessibility of health care--define and implement a more adequate and equitable allocation formula for the allocation of public resources; and develop a definition and implementation plan for insurance and prepayment schemes for both the formal and informal sectors. iii) Better exploit the public sector comparative advantage--define and implement contractual arrangements with both mission and private sectors for the provision of core health services. iv) Enhance access to, quality, and efficiency in the delivery of priority health interventions with a major focus on HIV/AIDS--support the Government to focus on financing services that tackle the major public health priorities, particularly those that disproportionately affect the poor and the vulnerable. v) Improve public service management--continue to help the MOH improve its procurement and financial management procedures; provide oversight on procurement and day-to-day implementation support; and contribute to the preparation of the annual procurement and financial audit reports. vi) Performance monitoring--assess the quality of reporting on performance indicators. 14 However, the Bank did not undertake more rigorous or systematized analysis of these areas as part of its background analysis, and appeared to rely on the information generated from the routine SWAp review processes as its primary source of evaluation. Assessment of project design The HSPSP-II was designed to support the strategies of the POW-II to improve the population's overall health status and reduce inequities in health access, quality, and efficiency. The rationale for continuing with the overall SWAp at the beginning of the POW-II was sound, particularly to support Government's strong ownership over the program, to continue the good relationships between MOH and donors, and to follow-through on the stated vision and strategies of the Government. Adequacy of the Government's commitment The Government of Ghana and, more specifically, the Ministry of Health's commitment to implementing the POW-II was exemplary. The MOH was involved early on in international discussions and conceptualizations of the first health sector SWAp. Thus, the MOH's commitment was unswerving during the implementation of POW-I, even after a new political party took office in 2001. The GOG exemplified good capacity and leadership in overseeing the donor-pooled Health Fund and made substantial progress in achieving institutional strengthening and capacity development. In addition, the GOG engaged all of its development partners in establishing priorities for the POW-II and in devising its strategies. This inclusive process was largely successful and a testament to the GOG's leadership capacity. Assessment of risks The sustainability of HSPSP-II was gauged by two parameters: (i) ownership; and (ii) availability of financial resources (including donor funds). Taking into consideration the successes of the HSPSP-I and the evident commitment of the GOG, international partners, and other stakeholders (including civil society, the mission, and the private sector), risks to achieving ownership of the SWAp were appropriately considered to be low. The availability of financial resources was deemed to be a potential risk due to macroeconomic factors, the unpredictability of GOG allocations to the health sector, and the potential volatility of donor funds. These risks were assessed to be potential but low threats to sustainability since the financial viability of the GOG looked promising and the GOG and donors' commitments to health seemed stalwart and steady. However, as discussed above, there was insufficient attention to the institutional risks posed by the split of the GHS and the MOH, and consequently to the way in which the strategies could be implemented. As a result of the conflict between the two agencies, there ensued conflict over control of resources, duplication of roles and responsibilities, a breakdown in the partnerships that were to be developed, and a undermining of the performance management basis needed to implement many of the strategies. Whereas not all of these outcomes could have been anticipated when HSPSP-II was approved, a closer analysis of the incentives framework and institutions involved would likely have helped. At the time of appraisal, the risks involving human resources were considered in terms of brain drain and staff attrition. It would have been difficult to anticipate all risks in the labor market that followed, but more attention to the institutional analysis might have pointed to the risks involving labor strife and the financing of the wage bill, both of which became major problems for the health sector during the POW-II. 15 2.2 Implementation (including any project changes/restructuring, mid-term review, Project at Risk status, and actions taken, as applicable) Implementation of the HSPSP-II in terms of the Bank's supervision of implementation went largely according to plan, with satisfactory oversight and engagement. There were no project changes, restructuring, or project at risk status changes. During the early years of HSPSP-II implementation, the Bank responded to early calls for a greater field presence and by moving the task team leader (TTL) of the project from headquarters in Washington, DC to the Ghana Country Office. Disbursement of project funds differed greatly from the disbursement plan determined at the time of appraisal (see Section A, Annex 1). Funds did not disburse until 2004 and then almost all of the remaining funds were spent in 2005 due to an unforeseen budget shortfall in the MOH. Because Bank financing was part of the Health Fund, this pattern of funding is unlikely to have dramatically affected financing the health sector, yet this was a less predictable, more volatile type of aid. Implementation of the HSPSP-II in terms of supporting implementation of POW-II encountered a number of unforeseen challenges. An evolving political climate and Parliamentary Acts to address health sector challenges created a number of moving pieces that significantly challenged the MOH's ability to carve out a real strategic plan for POW-II implementation. The years of POW-II implementation (2002-2006) were marked by institutional conflict in defining the roles and responsibilities between the MOH and Ghana Health Service (GHS). In addition, a reorganization of the public health sector in 2003 by the MOH and the passing of the wage bill in 2004 changed the landscape of the health sector, creating significant financial burdens on the MOH to cover the Additional Duty Hour Allowance (ADHA) and Deprived Area Incentive Allowance (DAIA) for health workers. Dialogue with development partners also experienced a shift during implementation. In the early years of POW-II, collaboration between development partners (DPs) and the MOH was both respectful and productive with a focus on mutual achievement; later in implementation, some DPs became more adversarial, adopting a "watchdog" attitude to expose Government inefficiencies and shortcomings. Previously, during POW-I and the early part of POW-II, such matters would have been viewed as joint challenges to be faced and addressed by DPs and Government together. This shift in development dialogue was observable, and implementation of the HSPSP-II was consequently affected by this change. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization Design Progress towards meeting the objectives of POW-I was monitored by twenty sector-wide indicators that were agreed upon by stakeholders and the MOH. The list served as the basis for annual performance reviews. At the conclusion of the POW-I, an assessment carried out in 2001 deemed many of the indicators not sufficiently focused on measuring outputs or outcomes. In the M&E design for POW-II, a new set of sector-wide indicators were identified to track progress in achieving the strategic objectives. However, the number of the sector-wide indicators varies between sector documents, highlighting inconsistencies in identifying the sector-wide indicators for POW-II. Annex 10 documents how the sector-wide indicators changed over time, outlining the indicators that were originally designed in the PAD and POW-II as well as those that were actually used for reporting for progress made in achieving the 2006 POW. Although the 16 sector-wide indicators changed over time, about twenty-five to thirty sector-wide indicators were repeatedly monitored and evaluated between 2002 and 2006. Data for these indicators are presented in Annex 11. The sector-wide indicators were based on the strategic objectives of the POW-II: (a) health outcomes (e.g., infant mortality and maternal mortality rates); (b) access to health services (e.g., immunization coverage, outpatient visits per capita); (c) quality of health services (e.g., under- five malaria case fatality rate, tracer drug availability); (d) efficiency (e.g., antenatal coverage, number of guinea worm cases); (e) partnership (e.g., % of recurrent budget from GOG and Health Fund used by private sector, etc.); and (f) financing (e.g., the proportion of GOG budget spent on health). Although the sector-wide indicators reflected the objectives of POW-II, the design of the results framework did not sufficiently enable accurate and timely monitoring of valid and reliable measures of the sector's performance. For instance, population-based mortality indicators based on relatively small surveys are not suitable to be used to monitor progress on an annual basis (e.g., under-five mortality is measured every five years and gives an estimate over the previous five or ten years), as these indicators are rather insensitive to changes in performance of health services, often depend on other determinants of health, and are less affected by changes in performance alone. Most notably, the sector-wide indicators insufficiently measure equity, especially geographic differences in health services and health outcomes. Regional differentials for health service indicators are relatively easy to monitor and evaluate, yet these were not captured in the M&E design. Furthermore, financial protection, a key indicator of equity, was not included as an indicator in the M&E framework. The sector-wide indicators did not adequately identify the links between building capacity and improving service delivery. For example, tracking vacancy rates for human resources would have been one way to measure improving capacity that has a direct impact on service delivery. More attention and effort should have been spent ensuring that appropriate measures of equity and intermediate outcomes were in place to inform the policy dialogue. Implementation The MOH and Ghana Health Service were responsible for collecting, aggregating, and reporting on the sector-wide indicators. There are still variations in data quality and management, as with implementation of POW-I, despite significant investment and improvements in information systems. As mentioned earlier, reporting on the sector-wide indicators has been inconsistent, with definitions of some indicators changing over time. Such inconsistencies compromise the opportunity to identify trends over time and create confusion among stakeholders over which indicators are valid measures of objectives. Indicators for health financing are often subject to change, oftentimes with the definition of the denominator differing over the years and subject to interpretation. Utilization Despite delays in data collection and reporting, the MOH and GHS have developed systems to collect detailed health service use and outcome data at both regional and district levels. However, 17 regional and district-level data, originally envisioned to inform regional priorities and serve as equity indicators, were hardly ever used in annual reports. All too often external reviews were required to collect, aggregate, and present health data, while this could have been done by the MOH's agencies. Despite the lack of effective use at the national level of regional and district-level data, there is evidence that some regions have made progress in this area. For instance, a few regions have developed league tables to create a performance-based system of evaluating district performance and improvement. Resources, thereby, may be allocated to reward performance and/or to address an identified need of a poorly-performing district. Such use of data monitoring should be encouraged and brought to national scale. However, on the national level, there was little evidence that data monitoring was used to improve resource allocation. Furthermore, relevant operational research that was undertaken on some of the new policies and programs were not used to inform their implementation. For example, an independent review of the information management system led to a relevant evaluation paper but there is little indication to demonstrate this paper's recommendations were used to reform the information system. These missed opportunities should be of high concern to the MOH/GHS, especially with the development of new annual POWs and the possibility of using data to inform regional health priorities and resource allocations. 2.4 Safeguard and Fiduciary Compliance (focusing on issues and their resolution, as applicable) There were two safeguards policies triggered by this project: (i) environmental; and (ii) involuntary resettlement. The environmental assessment considered the impact of the project in terms of health waste management and impregnation of bednets with insecticides. The Ghana Environmental Protection Agency (EPA) developed and disseminated guidelines on proper procedures for handling health care waste and the MOH agreed to use environmentally-friendly insecticides in bednets. The risk of involuntary resettlement was mitigated by issuing the Environmental and Resettlement/Compensation Framework to stakeholders within the MOH, GHS, academic institutions, NGOs working in environment, as well as to development partners. The project was in compliance with these safeguard measures throughout implementation of HSPSP-II. 2.5 Post-completion Operation/Next Phase (including transition arrangement to post-completion operation of investments financed by present operation, Operation & Maintenance arrangements, sustaining reforms and institutional capacity, and next phase/follow-up operation, if applicable) The Bank provided an adequate transitional arrangement when support to the health SWAp (through the HSPSP-II) ended on June 30, 2007. The last year and a half of the Bank's support to the SWAp was largely transitional, as the Bank had already moved to general budget support in 2006 after almost all of the remaining HSPSP-II funds were spent in 2005 due to an unforeseen budget shortfall in the MOH. The Bank remained involved in technical discussions and reviews during this last year, in addition to maintaining its role of reviewing procurement contracts. During this transitional year, it became obvious that the MOH faced a number of new challenges due to changes in the sources of revenue for health and new institutional arrangements, namely: (i) the National Health Insurance Scheme, established in 2003, has in principle replaced the "cash 18 and carry system" and the internally generated funds (IGF); (ii) some donors have shifted financial support from the Health Fund to budget support; and (iii) there has been a reduction in the proportion of contributions to the Health Funds, particularly in relation to earmarked funds. The MOH still needs to learn how to deal with MoFEP budgeting and disbursement procedures that are more cumbersome, and less flexible and accessible to BMCs than the Health Fund. The Government and development partners need to be concerned that the capacity gained in developing BMCs to utilize and report on financing is not lost. Furthermore, as general budget support increases, the role of BMCs needs to be carefully considered. If BMCs will continue as the district level financing modality, the sustainability of this system and its adequacy in advancing performance-based incentives should be considered. The Bank's new IDA-financed Health Insurance Project, approved on July 3, 2007, will serve as an important link to strengthening institutional capacity in order to carry out the NHIS, including its coverage of exempt categories. This project builds on the institutional capacity that was established during HSPSP-I and HSPSP-II and will hopefully provide another opportunity for the Government to implement performance-based management and financing systems, while improving equitable access to quality health services. The logic of a project-support to initiate a new institution (the NHIS) is well reasoned. The Bank is also supporting a new Nutrition and Malaria Control for Survival Project (US$25 million in IDA financing), approved on July 3, 2007. Although clearly addressing priority needs, the rationale for project-based support in these areas is more complicated. With the recognition of stagnating infant and under-five mortality rate, an unexpected increase in available IDA funds and the possibility of leveraging more funds from other donors, the Bank was opportunistic in creating a project focused on improving health outcomes through improved nutrition and malaria prevention and treatment. The new project is quite innovative in using existing financing modalities (i.e., no earmarked funding, but using the Health Fund) to create a results-based financing model. Although the Bank has gone to great lengths to design a relevant project that is harmonized with the international aid agenda, it is uncertain whether this financing will be a better alternative to the sector-wide approach or if it is a regression towards specific, project- based funding. In addition to these two new projects, the Bank has planned on continued involvement in the health sector dialogue through its support of the Poverty Reduction Strategy Credit (PRSC), which states strengthening health systems as an objective. The Bank plans to focus its support of the PRSC in the areas of health financing, human resources for health, and budget planning and management, which are consistent with Bank's areas of expertise. In this way, the Bank has tried to maintain its sector-wide dialogue in health despite the shift to project lending. The Bank has also actively participated in the discussions concerning the development of the next five year program of work (POW-III). The POW-III has proposed another set of sector-wide indicators for monitoring and evaluation that are improved, but still not optimally aligned with the POW's strategic objectives. There are now three indicators of equity to be measured at the national level, though none assesses equity at a regional, district, or BMC level, and one will be measured only once during the POW-III. The set of indicators chosen are thus unlikely to improve the ability to identify variations in equity (or other areas of performance) across management units. This is a lost opportunity for efforts to improve accountability of health services at operational levels, as well as to promote a system that would like to align financing with performance (either to reward high performers, or to provide extra help for low-performing areas). 19 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation (to current country and global priorities, and Bank assistance strategy) The PAD describes the Project Development Objectives (PDOs) in terms of "supporting the Ministry of Health in implementing its 2002-2006 Program of Work (POW-II)," and also translated what is clearly defined in the POW-II as a vision statement into the main PDO of a project, namely to: "improve the health status of the population while reducing the geographical, socioeconomic, and gender inequalities in health outcomes." This vision and the five strategic objectives defined above remain highly relevant to country and global priorities in health and development, and also fit in well with the Bank's assistance strategy, as articulated in the Ghana Poverty Reduction Strategy, the global Millennium Development Goals, and the Ghana Country Assistance Strategy. However, the vision statement is a mis-specification as a PDO, and likely contributed to the way sectoral dialogue between the MOH and development partners evolved, with the vision statement being treated more as a project objective rather than a vision statement. From a management sciences perspective, a vision statement is an aspirational proclamation about where the organization (in this case referring to the health sector) would like to be in the future. A vision statement is not defined in a way that an objective is monitored (e.g. with specific, measurable, achievable, relevant, and timely attributes), though it is important that strategies and objectives are consistent with the vision (and mission). The monitoring and evaluation systems of the POW-II were designed to assess the health status of the population over large time periods, but not in a way that can link changes in the inputs, activities, and outputs of the POW-II to achievement of the vision as an objective, as might be planned in a traditional project. The data for the sector-wide indicators are also unable to detect anything but the largest changes in population averages of selected health outcomes. The sector- wide monitoring system was designed to assess overall health outcomes of the population, but not the inequities, much less to detect reductions (changes) in inequities. No country has yet been be able to demonstrate how sectoral inputs and activities are causally linked to changes in health status on a national scale, as would be the standard of evidence if the vision statement were treated as a project objective. For these reasons, it is inappropriate to only consider the defined vision statement as the benchmark for achievement of the PDO. The POW-II strategic objectives are more appropriate as PDOs. The PAD also combines the strategic objectives at the level of a PDO in its logical framework, though it does not treat them as such in the ISRs. For purposes of evaluation, this ICR will assess the degree to which progress is being made towards fulfilling its vision, but will not treat the vision as if it were the sole PDO. In keeping with the intent of the POW-II and the PAD, this ICR will also assess the degree to which the identified strategies and related objectives of the POW-II have been accomplished. 3.2 Achievement of Project Development Objectives (including brief discussion of causal linkages between outputs and outcomes, with details on outputs in Annex 2) In considering the achievement of the PDOs, the evaluation will consider both the stated vision of the POW-II, to "improve the health status of the population while reducing the geographical, 20 socioeconomic, and gender inequalities in health outcomes," and the POW-II's five strategic pillars, both of which are identified as PDOs in the PAD logical framework. Progress Towards Realizing the Vision The assessments of health status used in the formal review of the POW-II (Table 3.2.1) provide some indication of the concern over the overall lack of improvements in mortality and HIV seroprevalence. It also highlights the lack of information needed to assess changes in these parameters as development objectives. There is insufficient data to assess changes in maternal mortality, and interpretation of HIV seroprevalence is fraught with difficulties. This is because HIV seroprevalence can increase if those living with HIV are living longer due to treatment, or if people with HIV infection become more likely to be tested, both of which would be the results of successful programs (rates can also decrease if people die sooner). Prevalence can also increase if more people are becoming infected, which is not a sign of success. This evaluation therefore focuses on further assessing what can be learned from data on changes in infant and under-five mortality and childhood malnutrition. To be consistent with an assessment of the statement of vision, we examine the longer trends over time, and assess changes in inequities. Table 3.2.1. Sector-wide indicators of health status 2002 2003 2004 2005 2006 Target Infant Mortality Rate per 1,000 live births 64 71 50 Under Five Mortality Rate per 1,000 live births 111 111 95 Maternal Mortality Ratio per 100,000 live births 560 150 Children Under Five who are Malnourished (%) 22.1 17.8 20 HIV seroprevalence (%) 3.4 3.6 3.1 2.7 2.9 2.6 Source: Health Sector Programme of Work (2002-2006) Independent Review of POW ­ 2006 (2007). Note: 2006 data for infant and under-five mortality actually refer to the average mortality experience over the previous 10 years; 2004 represents estimates of the previous five years. Malnourishment is defined as weight for age. The reference population is the WHO/CDC/NCHS reference, capturing the percentage of children scoring more than 2 standard deviations below the mean. The best estimates of the multi-agency working group on estimating global mortality rates provides a picture of mortality trends in Ghana that uses as wide a set of information as possible. Their data suggest that nationwide, overall infant and under-five mortality rates have leveled off since the late 1990s (Figure 3.2.1). Since these estimates of mortality rely on point estimates and not their error (variance), it has limitations when used to assess whether there are significant changes over time. When incorporating the level of variance in the estimates from the most recent (2003) DHS data, it is clear that only very large differences can be detected from one year to the next, such as would be found with reductions of 30 percent or more in one year (Figure 3.2.2.). The results also do not show statistically significant changes from one five year period to the next since the 1989-1993 period (Figure 3.2.3). 21 Figure 3.2.1 Best Estimates of Infant and Under-Five Mortality Rates in Ghana GHANA- INFANT MORTALITY 140 )shtrib 120 00 100 10 erp( 80 etar yti 60 altro 40 mtnafnI 20 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year Cen71 GFS79i DHS88d DHS88i GFS79d DHS93d DHS93i DHS98i DHS98d WDI DHS03d DHS03i Sources: Cen71 = Census 1971; GFS = Ghana Fertility Survey; DHS = Demographic & Health Survey; WDI = World Development Indicators. Numbers represent year of survey. "d" represents direct method of estimation; "i" represents indirect method of estimation. GHANA- UNDER-FIVE MORTALITY 250 )shtrib 200 1000 erp( 150 etar yti altro 100 m veif-rednU 50 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year Cen71 GFS79i DHS88d DHS88i GFS79d DHS93d DHS93i DHS98i DHS98d WDI DHS03d DHS03i Sources: Cen71 = Census 1971; GFS = Ghana Fertility Survey; DHS = Demographic & Health Survey; WDI = World Development Indicators. Numbers represent year of survey. "d" represents direct method of estimation; "i" represents indirect method of estimation. 22 Figure 3.2.2. Annual Trends in Ghana's Infant Mortality with Confidence Intervals 180 160 140 s rth 120 bi veil 100 000 1, 80 per 60 Deaths 40 20 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Source: Ghana DHS (2003); Hatt & Peters estimates using bootstrapped confidence intervals Note: Linear trend-line shown as dotted line. Figure 3.2.3 Trends in Childhood Mortality in Ghana (1984-2003) 180 Underfive deaths 160 Infant deaths Neonatal deaths 140 rthsib 120 veil 100 000 1,rep 80 60 eathsD 40 20 0 1984-1988 1989-1993 1994-1998 1999-2003 Period Source: DHS 1988, 1993, 1998, 2003; Hatt & Peters estimates. 23 A breakdown of infant mortality rates into its neonatal and post-natal components provides some useful observations that may explain some of these trends. Neonatal mortality appeared to increase during the last survey period, and is the driving force behind the infant and under-five mortality estimates. As shown in Figure 3.2.4, estimates of post-neonatal mortality (age 1-11 months) and child mortality (age 12-59 months) do not show an upward trend during the last survey period. There are well-known errors and biases in estimating neonatal deaths (especially under-counting and misclassification as still births). As discussed below, the increases in neonatal deaths also occurs largely among the wealthier and urban populations, suggesting a possible reporting bias that over-estimates neonatal deaths in the latter survey relative to earlier periods. Caution should be exercised when attributing causes to the trends in infant and under-five mortality. Figure 3.2.4 Trends in neonatal, postneonatal, and child mortality with 95% confidence intervals Ghana DHS (1988, 1993, 1998, and 2003) 100 90 80 70 births veil 60 50 1,000r pe 40 Deaths 30 20 10 0 1983-1988 1988-1993 1993-1998 1998-2003 Period Deaths ages 12-59 months Deaths ages 1-11 months Deaths in first month Are health outcomes becoming more equitable? We build on these analyses to assess whether there has been progress towards the long-term vision of improving equity, by defining socioeconomic differences though a standard asset index (Filmer & Pritchett 2001), and assessing trends over time. As shown in Figure 3.2.6, a steady, almost linear decline is observed for the poorest 40% of the sample, while under five-mortality rates among the middle 40% first decline and then level off. For the richest 20%, declines from 1983 through 1998 are erased by a statistically significant increase in the 1998-2003 survey period. Estimates of under-five mortality rates for all wealth groups converged in this most recent period. The situation is similar for infant mortality (figure not shown); where the poorest 40% appear to have the most consistent decline in infant mortality rates over the 20-year period. The rate of decline appeared to flatten out between the last two surveys. The middle 40% show a 24 very slight decline overall, while the pattern for the wealthiest 20% is erratic, declining precipitously in the 1998 survey period and then rising dramatically in the 2003 survey period. Figure 3.2.5 Trends in underfive mortality, by wealth group 5-year estimates with bootstrapped 95% confidence intervals Ghana DHS (1988, 1993, 1998, and 2003) 250 200 hst bir live 00 1,0 150 per Poorest 40% s Middle 40% year Richest 20% 5ts 100 fir in s Death 50 0 1983-1988 1988-1993 1993-1998 1998-2003 Period Supplementing these analysis with Weibull regression analysis for under-five mortality (Table 3.2.5), there was an average of 13% decline per survey period for the poorest 40% of the sample (p<0.01), whereas the trends for the middle and richest groups were not statistically significant. The results were similar when examining infant mortality. Incorporating socio-economic determinants into these models showed that whereas improvements were greater in rural compared to urban areas, there were otherwise no strong geographic, cultural, or gender differences in the rates of change in infant or under-five mortality (further details are to be provided in the forthcoming Country Status Report). Table 3.2.5 Weibull regression analysis: Trends in underfive mortality, by poverty group Pooled Ghana DHS surveys (1988, 1993, 1998, 2003) Average trend per survey Wealth group Hazard ratio SE Poorest 40% 0.87 0.04 ** Middle 40% 0.92 0.04 Richest 20% 0.99 0.10 *p<0.05 **p<0.01 25 Weibull regression analysis: Trends in infant mortality, by poverty group Pooled Ghana DHS surveys (1988, 1993, 1998, 2003) Average trend per survey Wealth group Hazard ratio SE Poorest 40% 0.90 0.04 * Middle 40% 0.97 0.05 Richest 20% 1.06 0.12 p<0.05 Source: DHS data, Hatt & Peters calculations Since neonatal mortality strongly influences the estimates of infant and under-five mortality, it is worthwhile to examine how it is distributed across different wealth groups. Although there are insufficient sample sizes to detect statistically significant trends, it is clear from Figure 3.2.6 that most of recent increase in neonatal mortality estimates is occurring in the richest quintile (also urban areas ­ data not shown). Although it is possible that wealthier families would have higher mortality rates, this would be highly unusual, and suggests that a reporting bias may be contributing to the apparent increases in infant mortality (as wealthier families are more likely to deliver in hospitals and report neonatal deaths). Nonetheless, the overall neonatal mortality rates are sufficiently high to demand more attention in terms of improving maternity and neonatal health services. Figure 3.2.6 Trends in neonatal mortality in Ghana according to wealth group 80 hst 70 bir Poorest 40% veli 60 Middle 40% 0 Richest 20% 1,00 50 per 40 monthtsrfi 30 in 20 hs at 10 De 0 1984-1988 1989-1993 1994-1998 1999-2003 Period Note: DHS 1988, 1993, 1998, 2003; Hatt & Peters calculations Bars indicate 95% confidence intervals. 26 Childhood Malnutrition For childhood malnutrition, the trends are very different for the different measures of malnutrition (Figure 3.2.7). Stunting (low height-for-age) declined from 1988 to 1998, but increased in the 2003 survey (p<0.05). Wasting (low weight-for-height), showed an opposite pattern, with substantial improvements between 1998 and 2003 (p<0.01), whereas underweight (low weight- for-age) showed a more consistent linear improvement. All three measures were strongly associated with poverty. Time trends in the prevalence of wasting showed that the large poverty differentials of the early survey rounds were gone by 2003, as the bottom 80% of the population enjoyed significant improvements from 1998 to 2003. The reductions were much larger than those made among the wealthiest quintile, and relative inequalities were eliminated. Trends in the prevalence of underweight showed significant reductions in all poverty groups (p< 0.05), but with slightly larger declines in the richer quintiles (so that relative inequality increased). However, time trends show that there is no change in rates of stunting across poverty groups. Figure 3.2.7 Trends in prevalence of malnutrition among children under 3, with 95% confidence intervals Ghana DHS 35% 30% 25% 3 under 20% Underweight dren Stunted chil Wasted of 15% centreP 10% 5% 0% 1988 1993 1998 2003 Survey Conclusions on Progress Towards Realizing the Vision An overall assessment of Ghana's progress towards realizing its vision would give it a barely passing grade. Overall improvements in infant and under-five mortality have slowed down, though most malnutrition indicators showed substantial improvements. Yet inequities in mortality for the poor and rural groups have been significantly reduced. The majority of malnutrition indicators show improvements for the poor and a reduction in relative inequality. Most of the lost ground in mortality is attributable to recent setbacks in neonatal survival, particularly for wealthy and urban groups, and while raising questions about the validity of the mortality data and the 27 need for better measurement of results, it also prompts a need for more focused attention to maternal and neonatal health. Achievement of Strategic Objectives Strategy 1: Increase Access to Services There are many definitions of access to health services, with most evaluators recognizing that access is related to the timely use of services according to need (Campbell et al 2000; Peters et al 2007). In the POW-II, it is clear that access to health services includes utilization of services and service coverage, and considers geographic, financial and sociocultural dimensions. An assessment of the sector wide indicators of health service coverage (Table 3.2.6) shows that there has been little change in most of the general indicators, and the (arbitrary) targets have not been met. The results for the delivery of priority health services, however, showed substantial gains and/or achievement of appropriate targets in most areas that were measured (see Annex 2 for further details). There are also improvements in structural access to care, as demonstrated by increased physician and nursing ratios, increased availability of drugs, and increases in functional Community-based Health Planning and Services (CHPS) zones (though not reaching the proposed target). Table 3.2.6 Sector wide indicators of health service coverage 2001 2002 2003 2004 2005 2006 Target General Service Coverage Indicators Outpatient visits per 0.49 0.49 0.5 0.52 0.53 0.52 0.6 capita Hospital admission rates 34.9 35.3 36 34.5 36.5 32.6 40 per 1,000 population No. of specialized 141 158 175 158 164 170 200 outreach services carried out Priority Service Coverage Indicators Penta 3/ DPT3 coverage 76.3 78.0 76.0 75.0 85.0 84.2 80 (%) Measles coverage (%) 82.4 83.7 79.0 78.0 83.0 85.1 80 Family planning acceptors 24.9 21 22.6 24.3 22.6 26.8 40 (%) Antenatal care coverage 98.4 93.7 91.2 89.2 88.7 88.4 70 (%) Supervised deliveries (%) 50.4 32.0 55.0 53.4 54.1 44.5 60* Postnatal care coverage 53.6 55.0 53.3 52.7 55.9 65 (%) 52.5 Structural Access Indicators Doctor to population ratio 1:20,036 1:18,274 1:16,759 1:17,615 1:10,380 1:10,700 1:16,500 Nurse to population ratio 1:1,728 1:1,675 1:1,649 1:1,510 1:1,508 1:1,587 1:1,500 % Tracer drug availability 70.0 85.0 85.0 87.5 84.7 73.8 80 No. of functional CHPS 19 39 55 84 186 .. 400 zones *The target proposed in the PAD was 50%, with a baseline of 44% 28 There was little explicit monitoring of equity improvements in health services during the POW-II, even though this was a key dimension of the strategy. Data on immunization coverage and outpatient visits for the years during POW-II was made available for analysis across Regions. This analysis demonstrates that equity in immunization coverage and health service delivery has improved in disadvantaged regions (defined as Northern, Upper East, and Upper West Regions). 1 Immunization coverage for all five tracked vaccines (BCG, Measles, DPT3/Penta 3, OPV3, and Yellow Fever) showed overall improvement in Ghana by the end of POW-II. Immunization rates of all five vaccines follow a very similar pattern of decreases in coverage from 2002 to 2003 and then continuing to stagnate in 2004, followed by significant increases in coverage in 2005 and 2006 (Figure 3.2.8). The increase in coverage rates increased most noticeably between 2004 and 2005, but with significant improvements maintained in 2006. When the data are aggregated by regional groups, disadvantaged regions (Northern, Upper East, and Upper West Regions) outperform the rest of the country (Greater Accra and the other regions) by statistically significant differences every year between 2002 and 2006 (Figure 3.2.9). Figure 3.2.8 National Immunization Coverage, All Regions, 2002-2006 105.00 100.00 Coverage (%) 95.00 90.00 BCG 85.00 Measles 80.00 DPT3 OPV3 75.00 Yellow Fever 70.00 65.00 60.00 2002 2003 2004 2005 2006 Year 1Central Region has also been considered a disadvantaged Region in some of the annual POWs. 29 Figure 3.2.9 Measles Coverage by Region, 2002-2006 Coverage (%) 110.00 100.00 90.00 80.00 70.00 60.00 2002 2003 2004 2005 2006 Year All other regions GAR Northern, UE, UW Regions All regions (national average) The patterns are more complicated for outpatient visits per capita. Health worker strikes in late 2002 and all of 2003 stopped of almost all reporting of outpatient services in these years, making interpretation of trends more difficult. After adjusting for periods of missing data, outpatient per capita appears to be increasing for some regions and is more variable in others (Figure 3.2.10). A more rigorous tracking of staff attendance patterns and health service outpatient use in each region would help to better assess whether there are changes in outpatient use across Regions. Figure 3.2.10 Outpatient Visits per Capita By Region, 2000-2006 ati 2.0 caprep 1.8 2000 1.5 2001 sts 1.3 2002 1.0 2003 vitnei 0.8 2004 0.5 2005 atptuO0.3 2006 0.0 Western ntral cra foNortheUp rn per Eastper West TION AL Ceeater Ac Eastern Ashanti Up NA Gr Brong Aha Region 30 Analyses of trends in disparities of health services by wealth group and Region were also conducted using the Ghana Demographic and Health Surveys (DHS), which were conducted in 1988, 1993, 1998, and 2003. The results show a number of improvements in access to services. For example, there were improvements in coverage in deprived Regions for skilled birth attendance and family planning use, reducing the disparity gaps by improving at greater rates than more advantaged Regions. Improvements for the poorest groups of Ghanaians were also achieved for: ORT use, ARI treatment, DPT3 coverage, antenatal care coverage, tetanus toxoid vaccination, and family planning use, but not for skilled birth attendance. In looking at time trends for the poor compared to wealthier groups, the gap between rich and poor for use of ORT had been eliminated by 2003. There were also significant improvements in relative rates for DPT3 coverage, antenatal care coverage, and tetanus toxoid vaccination coverage. However, wealth disparities in skilled birth attendance and family planning use actually increased. More details of these analyses will be available in the upcoming Country Status Report. Conclusions. Weighing the range of information available on health service access, the overall assessment is that a positive change has been made in overall access and equity of access during the POW-II. There has generally been an increase or maintenance of indicators for the availability of health services across Ghana during the POW-II. There is relatively little information to assess changes in equity of service delivery, though the available data suggest that there have been reductions in Regional disparities in access to some priority health services (e.g. immunizations, skilled birth attendance and family planning), though in other areas the results are less clear (e.g. outpatient and hospital use). When examining the changes in disparities for the poor, it is also a largely positive, though mixed, picture. There were improvements for most services that were measurable (ORT use, DPT3 coverage, antenatal care coverage, and tetanus toxoid vaccination coverage), but not for two others (skilled birth attendance and family planning use). Strategy 2: Improvement in Quality of Health Delivery There are several ways to assess quality of health services. The health status indicators described above can be considered measures of quality of health services, though there are many other factors that influence these outcomes, and so they are not good direct measures of the quality of care. Similarly, some of the structural measurements showing increased access to care, such as increasing the number of doctors and nurses, can also be considered as improvements in quality of services. In this section, we consider the two sector-wide indicators that are most attributable to the quality of clinical or public health service delivery (Table 3.2.7). The data indicate that there was a high level of drug availability during the POW-II, with the target being achieved in all years except 2006, and all showing an improvement over the baseline year. Tuberculosis cure rates steadily improved and achieved the stated target (a modest one by international standards) in all years of the POW-II. On this basis, the limited information that is available on quality of health services at a sectoral level leaves a largely positive picture. Table 3.2.7 Sector wide indicators of quality of service delivery 2001 2002 2003 2004 2005 2006 Target Tracer drug availability (%) 70.0 85.0 85.0 87.5 84.7 73.8 80 Tuberculosis cure rates (%) 44.9 53.8 63.9 65 67.6 .. 60 Despite the importance of quality of health services delivery, there is a lack of focus on strategies for institutionalizing quality improvement. For example, there was little attention paid to tracking 31 quality improvements at the sectoral level. Although supervision and quality management strategies did occur in the GHS, the work is not systematically documented or used as a basis for changing strategies or re-allocating resources. Given the central role of equity in the POW-II, more attention to the Regional and wealth disparities in quality of health services should have been part of the system for performance review of health services. Overall, the strategy to improving quality of health services likely made some positive gains during the POW-II, but it was largely a lost opportunity that should have led to a more concerted effort and demonstration of results. Strategy 3: Improvement in Efficiency of Health Service Delivery There is really only one of the sector wide indicators that can be used to directly examine efficiency of health services delivery, which is one measure of hospital efficiency (bed occupancy rate). On its own, it is a limited indicator of efficiency, which has even less value when not assessed by level of facility or Region. As shown in Table 3.2.8, the overall levels appear to decline in 2005 and 2006, and do not reach the 80% target in any year. Even without data on other hospital efficiency indicators, the low bed occupancy rates suggest that there are gross problems with unused bed capacity in the hospital sector. Table 3.2.8 Sector wide indicator of efficiency of health services 2001 2002 2003 2004 2005 2006 Target Bed occupancy rates (%) 61.6 65.5 64.1 62.7 58.4 50.9 80 A more rigorous assessment of health workforce productivity based on available data from 116 districts across Ghana and an index constructed from six types of inpatient, outpatient, and outreach health services demonstrated that productivity decreased slightly between 2004 and 2006 (Vujicic et al 2006). Indirect evidence concerning sectoral spending and health service outputs also suggest that there have been efficiency losses during the POW-II. Throughout the POW-II period, the overall recurrent health budget has increased substantially. This is largely attributable to significant increases in the GHS wage bill, which has increased predominantly because of increases in salaries, not from hiring more people. However, during this time, there has not been a concomitant increase in the number or quality of health service delivered ­ the increases in health services outputs have been very modest, with little evidence about changes in quality. However, without substantial increases in wages, it is also likely that "brain drain" and labor unrest would have been worse than it was during the POW-II, and that there would have been even greater losses in efficiency had there not been an increase in wages. Despite its place as a strategic pillar, there was relatively little attention placed on institutionalizing assessment and decision-making related to efficiency concerns. The expansion of community services and delivery of priority interventions was hampered by lack of funding for recurrent expenditures, including the reimbursement of exemptions. Overall, the conclusion is that there were major shortcomings in the strategy to improving efficiencies in health services delivery. 32 Strategy 4: Fostering Partnerships Progress towards fostering partnerships has been less than planned. There are no directly relevant indicators used at the sector level or in the PAD to monitor progress for the activities described in this area, which include partnerships with the private sector, empowerment of households and communities, and collaboration across sectors. The CHPS strategy was intended to empower communities, though there are little demonstrable changes in empowerment. A Patients' Charter of Rights was passed into law and disseminated throughout the country. This was accompanied by a provider "Code of Conduct", which was intended to improve staff behavior. Anecdotal reports highlight increases in malpractice lawsuits and improvements in staff behavior. Yet the dissemination and education efforts of the charter and code of conduct were not sustained. It is difficult to assess how much effect they have had on consumer empowerment or on improving quality, though it is unlikely to have much of an enduring effect without continued attention. It is useful to note that the strategic objective concerning partnerships also included partnerships with development partners, which was not part of the description of the area of focus in the POW- II (or project "component" ­ see Annex 2). In dealing with development partners, there were some clear successes in working through common management arrangements and continuing the policy and planning dialogue of the SWAp. Yet some development partners increasing moved towards project management support with earmarked funding, including off-budget financing, undermining the effectiveness of the SWAp partnership. A major strategic thrust of POW-II was to improve and establish formal commissioning arrangements with non-governmental service providers. Yet during the five-year POW-II, only one Memorandum of Understanding was signed very late and with one umbrella group for mission hospitals: the Christian Health Association of Ghana (CHAG). Very little progress was made in engaging the non-governmental sector in health services provision during POW-II. Private sector facilities are still not accounted for during health sector planning or budgeting, nor are they included in monitoring progress in the sector. Although the GOG has long recognized the value and importance of including the non-governmental sector into its dialogue, planning, and monitoring processes, few actions have matched the rhetoric. Strategy 5: Improving Financing of the Health Sector There were considerable accomplishments made in improving health financing during implementation of the POW-II, with some significant shortcomings in making health financing more equitable (see Table 3.2.9). The proportion of the GOG budget allocated to health has doubled from 2001 to 2006, reaching the Abuja target2 of 15% in 2005 and surpassing it in 2006. The proportion of GOG allocations to health of total GOG public expenditure is slightly lower-- at about 13% in 2006. The proportion of GOG recurrent funds allocated to health has also increased from 2002-2006. This increase, however, is due in large part to increases instituted by the wage bill and dedicated funding for personal emoluments. Thus, the increase in the wage budget was not met with a commensurate increase in the non-wage budget. The GOG, therefore, has continually overspent 2 In 2001, African Union countries established the Abuja target of attaining a 15% of national budgets for the health sector. 33 on its capital and wage budgets, and under spent on its non-wage recurrent budgets-- to the detriment of needed resources for delivering health services. Table 3.2.9. Sector-wide health financing indicators Indicators 2001 2002 2003 2004 2005 2006 Target % GOG budget on health 8.7 9.3 9.1 8.2 15 18 15* % GOG recurrent budget on 10.2 11.5 11.2 11.9 14.5 14 15 health % GOG recurrent health on 8.1 5.9 6.9 5.4 6.6 7 non-salary items (2+3) % spending on districts and 40.9 35.4 37.9 36 40 43 below, items 2+3 % Earmarked / total DP 62.3 32.8 39.5 26.3 40 61 40 % IGF from pre-payment 3 10 schemes % Recurrent funds from 1.2 3.1 2 GOG+HF allocated to CSOs % Recurrent funds. on 3.6 8 2.2 8 exemptions Per capita expenditure on 6.3 8.1 10.5 13.5 19 25.4 health (USD) Source: Annual sector review, 2006. * Abuja target of 15%. Although the funding of exemptions for the poor and other priority groups was a priority for the sector, this funding has been chronically insufficient throughout POW-II. Exemptions funding increased in 2005, partly as a result of the maternal exemptions policy which was implemented nationally that year. The policy was only partially implemented, however, when funds abruptly ran out in the middle of the year. Funding for exemptions has since dropped to 2.2% in 2006. The National Health Insurance Scheme (NHIS) is expected to replace the exemption policy, but full implementation of this arrangement will take time. Enrollment for those in the "exempt" category of the NHIS has surpassed expectations, yet there are delays and inconsistencies in obtaining full coverage. This transition may causes gaps in access to services for the poor and, therefore, may ultimately negatively impact their health status. There is some evidence that equity in public financing health is improving slightly. An analysis of the Ghana Living Standards Surveys from 1991-92 to 2005-06 suggests that benefit incidence of public spending at health clinics has improved to a point where it equally distributed across income groups (Coulombe & Wodon 2007). However, the distribution of public spending at hospitals has changed little, and still favors richer segments of the population. The introduction of the NHIS in 2003 was a landmark achievement for Ghana, which abolished its "cash and carry" system, though it continues to exist in practice to varying degrees across the country. This financing modality creates new opportunities to leveraging institutional changes to improve quality and productivity in health services in addition to improving financial protection to impact access and equity. 34 External financing for the health sector has recently changed, presenting several challenges that the MOH must manage. The Health Fund, which provided flexible funding to BMCs, has largely disappeared, as more donors have moved to budget support or earmarked funding. The loss of the Health Fund means that BMCs have less flexibility in how they can spend their money, which was reported to be a valuable tool for BMC managers to improve health service delivery. Similarly, donor funds have increased as a proportion of total funding, allowing national priority health programs to be financed by unpredictable financing. And finally, several donors have shifted their support to general budget support, slowing and reducing the accessibility of funds for the MOH from the MOFEP. These challenges must be managed carefully in order to maintain the progress made in health financing during the POW-II. 3.3 Efficiency (Net Present Value/Economic Rate of Return, cost effectiveness, e.g., unit rate norms, least cost, and comparisons; and Financial Rate of Return) Overall, efficiency in the health sector was not well monitored despite being a strategic pillar of the POW-II. All the direct and indirect evidence available point to a deterioration of efficiency during the POW-II (see Section 3.2, Strategy 3). Financial analysis in the PAD predicted that GOG financial contributions to the health sector would increase from 7.5% of its overall budget in 2001 to 8.0% in 2006. This target was below the Abuja target of 15% of total Government spending on health, but was considered realistic with budget constraints and health sector spending during project design. Likewise, the GOG's recurrent budget spent on health was expected to increase from 11% in 2001 to up to 13% in 2006. The Government of Ghana exceeded the financial targets set forth in the beginning of the project. In 2005 and 2006, about 15% and 18%, respectively, of the total Government budget (which includes all expenditures) was spent on health. Moreover, 14.5% in 2005 and 14% in 2006 of the recurrent budget was spent on health. The Government increased its commitment to funding the health sector in real financial terms, achieving the Abuja targets. 3.4 Justification of Overall Outcome Rating (combining relevance, achievement of PDOs, and efficiency) Rating: Moderately unsatisfactory The Bank's evaluation criteria for the extent to which the operation's major relevant objectives are achieved (or are expected to be achieved) involves a rating scale that is based on "weighing possible shortcoming in the achievements of the operation's objectives, in its efficiency, or in its relevance." (OPCS 2007) The overall outcome rating is thus defined according to the degree of shortcomings in these areas. Although the PDOs (the vision and strategic objectives of the POW-II) remain highly relevant to the country's development and the Bank's country assistance strategy and overall corporate goals, the weight of the evidence suggests that there were significant shortcomings in the operation's achievement of its objectives, as well as in its efficiency, which is consistent with the Bank's criteria for a moderately unsatisfactory rating. The POW-II did, on balance, show greater success than failure in progress towards its vision, largely because of equity gains in mortality and malnutrition, as well as the overall improvements in malnutrition. However, there was a slowing down of reductions in the measured mortality indicators. There were also substantial gains in achieving it strategic objectives, such as in increasing access to priority services, reducing 35 inequalities in service delivery, and introducing a rapidly expanding NHIS. However, there were also significant shortcomings in efficiency, institutional reforms, and partnerships with the private sector. Other than the introduction of NHIS, where an assessment of results is premature, there have also been significant shortcomings with the allocation and use of public funds for recurrent and capital expenditures. 3.5 Overarching Themes, Other Outcomes and Impacts (if any, where not previously covered or to amplify discussion above) (a) Poverty Impacts, Gender Aspects, and Social Development Poverty and equity aspects are addressed in more detail in Sections 3.2 and Annex 2 of this report, since the POW-II clearly highlights its intention to improve geographic, poverty, and gender disparities in health. The available data indicate that there have been substantial improvements in several measurements of equity in health outcomes, as well as improvements in equity of health services. However, management decision-making and the regular review of sector performance did not incorporate sufficient attention to poverty and equity considerations -- none of the sector- wide indicators of the POW-II focused on health equity, and incentives were not aligned with improving health equity. (b) Institutional Change/Strengthening (particularly with reference to impacts on longer-term capacity and institutional development) The institutional conflict between the MOH and GHS had a major impact on the implementation of the POW-II. Both agencies sought to have authority over control of resources and decisions. As a result, the performance contracts, which was intended to be the major mechanism for aligning incentives for improving service quality and productivity was undermined. The same approach for developing a more constructive and transparent relationship with non-state providers also suffered as a result of the conflict between MOH and GHS. The question of who had the authority to spend on capital works became a major point of conflict between the MOH and GHS, resulting in a loss of transparency in the selection of capital works. In addition, the agencies set up duplicated structures, which often appeared to be in competition with each other, and tended to fragment efforts to develop more coherent approaches in human resources, quality assurance, and partnerships. Fortunately, as the POW-II ended, there appeared to be diminishment of the conflict, though it's not clear that there is a long-term resolution. During POW-II, substantial achievements were still made in terms of developing mature financial management systems, procurement systems, and decentralization. The institutional capacity to support decentralized BMCs in their planning, budgeting, and implementing their own work programs remains strong, even if more could have been done if the GHS and MOH had worked more synchronously. Perhaps one of the most important institutional accomplishments of POW-II is the establishment of the National Health Insurance Scheme (NHIS). The NHIS, borne out of a law enacted in 2003, has increased significantly in terms of size and scope over the past few years. The NHIS has demonstrated significant capacity early on, as enrollment has surpassed expectations and the Scheme has tried to expand quickly to meet demand. There are challenges to the Scheme's success, mainly in achieving timely financial payments, ensuring exemption coverage, and maintaining positive public opinion. However, the institutional capacity development for the NHIS is promising and is a significant achievement during POW-II, and promises a new way to align incentives and performance in health service delivery. 36 (c) Other Unintended Outcomes and Impacts (positive or negative) One of the main purposes of the SWAp's "basket" funding (the Health Fund) was that the Government would be able to set its own priorities by applying predictable, fungible, and available funds to support the work of its budget and management centers (BMCs). However, contrary to logic, as POW-II progressed, donor funding became more unpredictable and less accessible to the health sector. This phenomenon hinged on two occurrences: (a) donors moving from the Health Fund to budget support; and (b) donors earmarking a greater proportion of their funds for the health sector. The shift to budget support, although in line with the principles of the Paris Declaration, compromised the predictability and accessibility of funds to the health sector. In the long-term, the hope is the MOH and MOFEP will be able to work out their internal processes to allow for predictable financing. However, in the meantime, the health sector has faced a huge shock as the Health Fund has been reduced drastically in size and more donors are funding off-budget, earmarked programs. Thus, one short-term impact of shifting to budget support has been a reversion back to project-like funding. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops (optional for Core ICR, required for ILI, details in annexes) Stakeholder workshops and a beneficiary survey were not carried out in preparation for this ICR. 4. Assessment of Risk to Development Outcome Rating: Moderate Diligent efforts will be required to ensure that the past gains made under the SWAp are not lost. The sector currently faces the following challenges that may jeopardize progress: (a) financing of the health sector has become more complicated, both internally and externally; (b) scale-up of the NHIS to nation-wide coverage may delay health services, particularly for the poor; (c) continued failure to properly monitor human resources for health; and (d) decentralization of budgets to district governments may cause fragmentation and confusion. Donor commitments have dropped off following the shift from pooling funds to budget support and there are signs that the Government's execution rate is retarding as processes become more complicated. The shift by some donors to budget support has also been characterized by an increase in earmarked funding. As a result, the MOH has funded its national priority health programs with earmarked funds, since this is one of the only ways to access such funding. The result is that Ghana's priority health programs are financed largely by unpredictable financing, compromising their long-term viability. BMCs were established and functioned well when using flexible funds available through the Health Fund. As the Health Fund has been replaced by funds provided through the MOFEP, there is a risk that not having flexible funds available will undermine the capacity of peripheral BMCs to manage resources and deliver services. In addition, it is not clear how the poor and the vulnerable will fair as the NHIS is rolled-out nationwide. There has already been concern that the scheme will not be financially viable enough to cover those in exempt categories until about 50% of the country is enrolled. While this prediction may be a little pessimistic, there is still little evidence that the NHIS is having a positive affect on improving access, quality, and equity in health services. It will be critical to monitor health service indicators among equity groups to determine the effect of the NHIS on equitable distribution and access. 37 The human resources for health crisis continues to plague Ghana. With so much of the Government's recurrent budget for health tied up in personal emoluments and salaries, there is very little room for creating performance incentives for medical staff or further developing its human resources strategy. The Government needs to modify its human resources for health performance indicators. Staff to population ratios are insufficient indicators for monitoring human resources as a whole. The monitoring system needs to focus on new staff training rates, retention rates, geographic distribution, and performance in terms of coverage, equity, and quality. Finally, the potential decentralization of budgets and financial management to district governments may cause fragmentation in health financing and conflict over resources and implementation responsibilities. This could increase the potential conflict and duplication between the MOH, GHS, local governments, and the NHIS. 5. Assessment of Bank and Borrower Performance (relating to design, implementation and outcome issues) 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry (i.e., performance through lending phase) Rating: Moderately unsatisfactory The Bank's performance in ensuring quality at entry is rated moderately unsatisfactory. Although the Bank ensured that the project was strategically relevant, the Bank failed to conduct the sufficient analysis and evaluation that could have contributed greatly to the design of the project. Nevertheless, the design of another sector-wide approach for this project was appropriate, given the perceived success of the first SWAp and a strong donor contingency interested in supporting further nationally-owned strategies and processes. The financial and economic aspects of project design were sound, with adequate consideration of the environmental and fiduciary aspects. Implementation arrangements were clearly stated in the PAD, with oversight of procurement retained by the Bank. In addition, a Memorandum of Understanding (MOU) that was signed by the MOH and all cooperating partners in December 2002, built on the Common Management Arrangement endorsed by all development partners in 2001. The Bank fell short in its analysis of institutions and their potential roles of contribution or hindrance to achieving the objectives of POW-II. More effort should have been spent evaluating the policy and institutional interactions for implementation of POW-II. Furthermore, and perhaps most importantly, preparation for the HSPSP-II failed to adequately analyze and evaluate the poverty differentials in terms of health status and access to health services. Although a stated priority in both the PAD and POW-II (in fact, equity is the centerpiece of POW-II's vision), the Bank did not, at the time of project design nor thereafter, successfully evaluate or analyze how differentials in health equity (geographic, financial, and sociocultural) were to be addressed. This is unfortunate, not only because improving equity was essential to achieving POW-II's vision, but also because this is a key area where the Bank is well-positioned to contribute substantively. The Bank's inability to ensure that POW-II adequately addressed equity is evidenced in its monitoring and evaluation system. Mortality indicators were specified to include regional data to indicate disparities; however, these data were largely unavailable and insensitive to annual sector- wide changes or improvements. Consequently, regional health disparities were mostly unreported and therefore not available to inform sector dialogue or policy. Similarly, a public expenditure tracking survey was not conducted until the very end of the project, even though it was cited as a 38 needed exercise and could have been instrumental in informing resource allocation to improve equity. The monitoring and evaluation system should have included sector-wide indicators that would be direct measures of equity, and particularly to compare equity performance across functional units of the sector. In addition, the sector-wide indicators focused mostly on health outcomes and service provision data, with insufficient measures of intermediate outcomes in terms of achieving institutional change. Thus, although implementation of the POW-II was very successful in improving sector capacity and institutional change, particularly at district and regional levels, monitoring and evaluation of the sector failed to capture these improvements. (b) Quality of Supervision (including of fiduciary and safeguards policies) Rating: Moderately satisfactory The Bank's quality of supervision is rated as moderately satisfactory, because, on balance, there were moderate shortcomings in the Bank's identification of opportunities and resolution of threats, the basis for assessment on the Bank's rating scale. The Bank was noted for being a cooperative and supportive partner of the MOH throughout implementation of POW-II. Despite that development partners' discussions with the MOH had become adversarial and contentious at times in the last few years of POW-II, the Bank was cited time and again as maintaining a professional dialogue with the MOH to foster donor cooperation and adherence to the principles of the SWAp. The Bank benefited greatly from a greater presence in the field when the TTL of the project was moved from Headquarters in Washington, DC to the Ghana Country Office in order to establish much more fluid and frank discussion with the MOH and development partners. The Bank also closely cooperated with the Government in assessing human resources productivity and the related fiscal constraints imposed by the Government's policy decisions concerning the wage bill. Implementation of POW-II also benefited from the Bank's technical guidance on financial management and procurement matters. This technical support was a successful and meaningful contribution to the MOH's implementation of the project and further developed institutional capacity. In addition, the Bank's candor and quality of performance reporting was high, through the identification and explanation of some difficult and important issues during implementation of the POW-II. The moderate shortcomings noted in the Bank's performance during implementation of POW-II relate to the: (a) focus on development impact; and (b) adequacy of supervision of inputs and processes. The Bank consistently called attention to concerns over health outcomes, but tended to be complacent about relying on national averages of health outcomes and service delivery rather than more rigorously investigating concerns about health equity or variation in performance. This is problematic given that health disparities and performance monitoring were highlighted at appraisal as areas where the Bank would focus its attention. The reporting on the sector-wide indicators was inadequate in terms of measuring equity (and efficiency), and the Bank was not able to correct this despite the investigations it initiated into disparities in health towards the end of the POW-II. As a result, poverty differentials in health were not accounted for and resource allocations were not changed accordingly. Despite raising issues of the capital investment plan with the Government, the Bank was unable to successfully influence Government to follow through on its early efforts to prepare and 39 implement a capital investment plan that would reflect the priorities of the GPRS and the POW-II. Control over capital investment became a major point of contention between the MOH and GHS, and donors seemed to use the conflict between agencies to create uncertainty and push their own agendas. The Bank had a crucial role to play in ensuring that the capital investment plan was adequately based on priorities. Instead, capital works spending between 2002 and 2005 was twice the forecasted amount and there is little evidence that resources were well allocated to address underlying poverty and efficiency problems. Although the Bank was a major contributor to the Health Fund, it is unclear how much leverage the Bank had over the implementation of the overall capital investment program, though more might have been done in 2003 to 2005 to independently investigate expenditures ­ a potentially risky intervention. A long-promised public expenditure tracking survey would have helped in assessing both recurrent and capital expenditures (it was conducted in 2007), though it's not clear that there was sufficient cooperation to conduct it properly in early years. Finally, the Bank also stated its intention at appraisal to focus on improving access to quality HIV/AIDS services. It reasonably pursued this largely through the development and financing of a separate HIV/AIDS project. (c) Justification of Rating for Overall Bank Performance Rating: Moderately unsatisfactory The overall rating for the Bank's performance is rated as moderately unsatisfactory, based on the guidelines used by the Bank. The Bank's ratings guidelines state that when a rating for one dimension of Bank performance is in the satisfactory range while the rating for the other dimension is in the unsatisfactory range, the overall rating depends on the outcome rating. Because the outcome rating is in the unsatisfactory range, this overall rating is also on the unsatisfactory side. 5.2 Borrower Performance (a) Government Performance Rating: Moderately satisfactory The Government's performance is rated as moderately satisfactory. In general, the Government exemplified a strong commitment to achieving the POW-II's strategic development objectives, surpassing the financing targets set out in the beginning of POW-II and instituting relevant policies to improve the health sector (e.g., the wage bill and National Health Insurance Scheme). Implementation of POW-II occurred shortly after a new Government came into power. It was notable that much of focus of POW-I remained unchanged, and the health sector was able to proceed with its five-year strategic plan. The Government also engaged with stakeholders and development partners in SWAp discussions through annual Health Summits and other relevant dialogue venues. The Government could have expanded its consultations by including more civil society and non-governmental providers into the sector dialogue. The Government created systems that facilitated implementation from fiduciary oversight to transition arrangements. The Government fell short in properly ensuring that processes were in place and transitional arrangements were made to transfer funds from MOFEP to MOH when donors, specifically the Bank, moved from the Health Fund to general budget support in 2006. However, this type of shortfall is not necessarily indicative of a long-term Government failure, as such an experience should serve as an impetus for the MOFEP and MOH to work out their internal processes to allow for predictable financing. 40 The Government's performance faced moderate shortcomings in the design, implementation, and utilization of monitoring and evaluation. Although the Government set out to improve equity in health, almost none of the indicators for the evaluation system measured equity. Moreover, the indicators failed to measure capacity indicators that could have served as milestones for improving equity, such as certain measures for improved human resources capacity. In addition, the cornerstone of the Government's vision for its health sector was not evaluated appropriately to inform policy and resource allocation. The MOH often relied on the annual sector reviews to aggregate and evaluate data, leaving the critical evaluation role of the Government out of the equation. The Government maintained good relationships and coordination with some of its donors, partners, and stakeholders, but not all. Sector-wide dialogue sometimes excluded certain partners, particularly those who were not contributing to the pooled Health Fund. Discussions with partners also became adversarial at times. Nevertheless, the Government made adequate transition arrangements to support the operation of activities after the Bank's credit and grant ended, as the SWAp continues with the POW-III and a new POW for 2007. (b) Implementing Agency or Agencies Performance Rating: Moderately unsatisfactory The rating is based on the overall assessment of significant shortcomings in implementing agencies' performance, the standard used for Bank evaluation. The performance of the implementing agencies, the Ministry of Health and Ghana Health Service, suffered due to their failure to be able to work together more coherently, which undermined policy and implementation of the POW-II. Following the re-launch of the GHS in 2003, the MOH and GHS clashed repeatedly over which organization had the authority over many policy and implementation decisions. This resulted in a collapse of the intended approach to align incentives and financing through performance agreements, and the intended approach to better exploit the potential of the private sector. Ultimately, this internal conflict diverted time and resources away from the strategies set to achieve POW-II's objective, particularly at headquarters and regional levels. In many other respects, the performance of the implementing agencies was appropriate for implementing the POW-II. The capabilities of the district level BMC's and the Regional offices was instrumental in achieving modest improvements in service coverage and equity during the POW-II, aside from periods of labor unrest. Improvement in some management systems, such as financial management and procurement of goods also improved over the POW-II. The successful introduction of the NHIS is also an important accomplishment from this period. (c) Justification of Rating for Overall Borrower Performance Rating: Moderately unsatisfactory The overall rating for Borrower performance is moderately unsatisfactory, based on the guidelines used by the Bank. The Bank's ratings guidelines state that when a rating for one dimension of Borrower performance is in the satisfactory range while the rating for the other dimension is in the unsatisfactory range, the overall rating depends on the outcome rating. Because the outcome rating is in the unsatisfactory range, this overall rating is also on the unsatisfactory side. 41 6. Lessons Learned (both project-specific and of wide general application) With the completion of HSPSP-II's support to Ghana's health SWAp, reflection on its achievements as well as weaknesses reveals several lessons. One lesson is that a stronger analytic base is needed even when things appear to be going well, as was the case when the HSPSP-II was approved. In-depth analysis and debate concerning equity of health services and outcomes, efficiency of health services, and institutional analyses would have been particularly helpful in anticipating problems and furthering the agenda set by the POW-II and the GPRS. It is also clear that donor behavior is critical to the success or failure of a SWAp. Donors need to be disciplined about their funding support to a sector and hold other donors accountable for their actions. Donors can easily revert back to earmarked, off-budget funding that compromises the Government's attention to the SWAp's implementation and success, while tempting other donors to follow suit and push for their own agendas as well. Other development partners moved towards budget support, which can also cause disruptions at the sectoral level. Safeguards should be instituted to ensure that shifts to general budget support are upheld by processes to ensure a smooth transition in financing modalities. Donor shifts to budget support are in line with the Paris Declaration, however, countries need to be ready for this transition ahead of time to ensure that the predictability of funds is not interrupted. There were also design flaws that were not well addressed during the POW-II, and the Bank's conceptualization of investment lending helped to reinforce some of the design flaws. In particular, treating the long-term vision as a project development objective helped to change expectations (that a set of investments and plans would lead to a set of measurable mortality changes on a national scale during a project period). The dialogue between MOH and development partners spent considerable time dealing with unmet expectations, even though the results could not have been measured during the POW-II period, much less have been attributable to the POW-II. The artificial structure of components imposed by Bank documentation is unlikely to have affected implementation of the program in a significant way, but it did not help to support SWAp strategies or contribute to meaningful monitoring and evaluation. In Ghana's case, there was inordinate attention on results that were not under control of the health sector, using a monitoring and evaluation system that was not sufficiently aligned with strategic priorities, and so was not able to move further towards a performance-based health system. These factors can seriously undermine achievement of sector objectives. Intermediate and long- term outcomes need to be appropriately identified to properly measure progress towards achievement of objectives and, if needed, respond and adjust to risks to those objectives. It is extremely difficult to improve and redirect policies and programs if they are not appropriately measured and monitored. In addition, the lack of valid, measurable progress underscores the need to link financing with measurable performance targets to ensure timely and appropriate monitoring and evaluation so that programs and policies may be adapted, as necessary. In Ghana, the problems were both technical and institutional. Institutional competition and overlap and the failure to resulting capture the contributions of the non-governmental sector are issues that have both technical and political dimensions, and an area where outside facilitation and input may be helpful. 42 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies The Government prepared its own implementation completion report based largely on an analysis of the independent review of the sector performance and additional interviews with key stakeholders in government and development partners (see Annex 7 for details of the report). The report describes a mixed picture in terms of achieving the intended objectives of the POW-II, as identified by the changes in the sector-wide indicators. Difficulties in managing the transition of the MOH and its agencies are noted, along with considerable successes in developing common management arrangements, and in the dialogue and support provided by the Bank. The overall conclusion is that Government merits a "pass mark" for the implementation of the POW-II, and that there remain considerable challenges and optimism for addressing inequity and effectiveness of the health sector in the future. These findings and conclusions are consistent with the analysis of this ICR. (b) Cofinanciers No additional comments. (c) Other partners and stakeholders (e.g. NGOs/private sector/civil society) No additional comments. 43 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Appraisal Estimate Actual/Latest Components Percentage of (USD millions) Estimate (USD millions) Appraisal YEAR 2002 7.50 0.00 0% YEAR 2003 15.00 0.00 0% YEAR 2004 20.00 29.97 150% YEAR 2005 20.00 59.62 298% YEAR 2006 27.50 10.62 38% YEAR 2007 0.00 0.51 -- YEAR 2008 0.00 0.31 -- Total Baseline Cost 90.0 101.03 112% Physical Contingencies 0.00 0.00 0.00 Price Contingencies 0.00 0.00 0.00 Total Project Costs 0.00 0.00 Project Preparation Fund 0.00 0.00 .00 Front-end fee IBRD 0.00 0.00 .00 Total Financing Required 0.00 0.00 (b) Financing Appraisal Actual/Latest Source of Funds Type of Estimate Estimate Percentage of Cofinancing (USD (USD Appraisal millions) millions) Borrower 638.00 627.19 98% Local Communities 75.00 70.00 93% International Development Association (IDA) 57.30 68.40 119% IDA GRANT FOR POOREST COUNTRY 32.30 38.50 119% FOREIGN SOURCES (UNIDENTIFIED) 310.40 253.26 82% 44 Annex 2. Outputs by Component The Bank's evaluation structure requires an assessment of the outputs from a set of components, even though this does not fit well with the design of the POW-II. Although the overall vision, goals, and strategic objectives of the program remained the same throughout, there are no components described in the POW-II. The PAD and the Development Financing Agreement highlight ten areas of focus that are consistent with the ten chapters of the POW-II that follow the strategic vision. These areas are treated as components in the logical framework of the PAD. However, the PAD also identifies each year's program of work as a component when describing the financing of the program. Although the ten focus areas were identified as components in the DFA description of the project, they were not formally revised, and were not treated as components in the POW-II. The description of focus areas changed substantially by the 2003 Program of Work, and continued to be modified annually. To be consistent with the Bank's evaluation methodology, outcomes and outputs of the program could be assessed according to the five strategic pillars and the original ten focus areas of the DFA. Since we have already considered the strategic pillars in the context of the project development objective (Section 3.2), this annex focuses more specifically on the outputs of the ten original focus areas that can serve as "components": 1. Strengthening priority health interventions; 2. Developing human resources for health services; 3. Enhancing infrastructure and support services; 4. Fostering partnerships for health; 5. Improving regulation; 6. Reforming institutions arrangements; 7. Improving the health sector financing; 8. Improving financial management systems; 9. Further strengthening management; and 10. Linking with traditional medicine. 1) Strengthening priority health interventions Priority health interventions were identified by the GOG as those having a direct impact on health outcomes and improving health status. Although the priority interventions change from year to year, the original priority interventions that are monitored on a sector-wide basis include: · HIV prevention, treatment, and care · TB services · Malaria prevention · Maternity services (antenatal care, delivery, and post-natal care) · Childhood immunization · Family planning · Guinea worm eradication · Polio eradication This component can be seen as being largely successful in producing targeted outputs. As shown in Table 1, targets were achieved in 2006 for five of the eight sector-wide output indicators related to priority programs. The results demonstrate that the high rates of immunization and antenatal care that were achieved during the previous POW were sustained at even higher levels. The AFP rate suggests that the surveillance system was succeeding in helping Ghana fulfill 45 commitments for polio eradication. The three indicators that did not reach their target are related to pregnancy and delivery. There was very little change in the family planning and post-natal care coverage. Table 1: Sector wide outputs for priority interventions 2001 2002 2003 2004 2005 2006 Target (Baseline) Penta 3 coverage (%)* 76.3 78.0 76.0 75.0 85.0 84.2 80 Measles coverage (%) 82.4 83.7 79.0 78.0 83.0 85.1 80 Family planning acceptors (%) 24.9 21 22.6 24.3 22.6 26.8 40 Antenatal care coverage (%) 98.4 93.7 91.2 89.2 88.7 88.4 70 Supervised deliveries (%) 50.4 32.0 55.0 53.4 54.1 44.5 60** Postnatal care coverage (%) 52.5 53.6 55.0 53.3 52.7 55.9 65 Maternal deaths audited (%) 10.0 84.0 85.0 55.9 75.6 52 50 Acute flaccid paralysis (AFP) 2.8 1.9 1.3 1.5 1.68 1.55 >1 non polio rate (%) * DPT3 was replaced by Penta 3 vaccine in 2004 ** The target proposed in the PAD was 50%, with a baseline of 44% The case of supervised deliveries merits further attention. As part of an effort to reduce maternal mortality, much emphasis was placed on increasing supervised deliveries and audit maternal deaths. A policy was introduced in 2003 to exempt users from delivery fees in order to encourage mothers (particularly the poor) to delivery at health facilities. A review of the exemption policy in 2005 found that the policy proved successful in dramatically increasing the amount of women delivering in health care facilities until the funds ran out, leading to facility indebtedness. Consequently, the 2006 exemption allocation was used to pay this debt, allowing for virtually no funds to be available for the exemption program for deliveries (Witter & Adjei 2007). The results in Table 1 suggest that higher levels were being achieved from 2003 to 2005 (surpassing the original target of 50%). Table 2 indicates the health outcome and impact indicators that are specifically related to priority programs. The targets for tuberculosis cure rates were met by the second year of the POW-II, whereas the target to eliminate guinea worm during the POW was not achieved. There seem to have been program failures in specific parts of the country during 2003 and 2004, due in part to local armed conflict and internal migration (the "Dagbon Crisis"). Although the target was not met, surveillance and control activities were mobilized to respond to these conditions, leading to reduced levels in 2005 and 2006. Interpretation of the other indicators is not as straightforward. HIV seroprevalence is not a good indicator of program performance, as the rate can increase if patients are living longer due to treatment, or if people with HIV are more likely to be tested, both indicators of successful programs (they can also decrease if people die). Prevalence can also increase if more people are becoming infected, which is not a successful trend. The under-five malaria case fatality indicator did not have a target, and requires consistent case definition and reporting. Since most deaths do not occur in hospitals or where a definitive diagnosis can be made, it is hard to tell whether changes in rates are related to reporting or to real differences. 46 Table 2: Sector-wide health outcomes and impact indicators for priority interventions 2001 2002 2003 2004 2005 2006 Target Tuberculosis cure rate (%) 44.9 53.8 63.9 65 .. 67.6 60 Guinea worm cases 4733 5545 8290 7275 3992 2968 0 HIV seroprevalence (%) 2.9 3.4 3.6 3.6 3.1 2.9 2.6 Under five malaria case 1.7 3.7 3.6 2.7 2.4 2.7 n/a fatality rate (%) 2) Developing human resources for health A number of interventions to develop human resources for health (HRH) were implemented during the POW-II. However, a coherent human resources strategy was lacking, and the efforts were not well linked to the strategic objectives to improve quality of services, responsiveness to client needs, efficiency in service provision, improved financing, or to the vision of reducing inequities. The crude indicators used to demonstrate increased success in human resources through tracking the numbers of doctors and nurses per capita were actually achieved (Table 3). The HRH indicators proposed in the PAD were not used during the POW-II. Notwithstanding the achievement of the overall human resource targets in the sector program, it is difficult to see if the increase in doctors and nurses has had positive effects in the strategic areas of the POW-II. The additional duty hour allowance (ADHA) put in place significant financial incentives to retain staff in the public sector, yet these salary increases were not tied to increased performance, whether measured as increased productivity or enhancement of service quality. They also did not appear to reduce geographic imbalances. In addition, the deprived area incentive allowance (DAIA), designed to improve equity in geographic access to health workers, was only partially implemented, and could not be sustained. By excluding some categories of staff (e.g. accounts staff), it also seemed to build resentment among those who could not benefit from the scheme. Moreover, the ADHA, although successful in improving retention of doctors and reducing health worker strikes that were particularly problematic in 2003, it has also crippled the GOG health budget. In 2005, almost 97% of GOG's health expenditures went to salaries and ADHA, leaving the Government little flexibility to use the money for other purposes. Table 3: Sector-wide indicators on human resources 2001 2002 2003 2004 2005 2006 Target Doctor to population ratio 1:20,036 1:18,274 1:16,759 1:17,615 1:10,380 1:10,700 1:16,500 Nurse to population ratio 1:1,728 1:1,675 1:1,649 1:1,510 1:1,508 1:1,587 1:1,500 3) Enhancing infrastructure and support services An early effort was made to prepare a Capital Investment Plan (2002-2006) that would reflect the priorities of the GPRS and the POW-II. In addition to listing infrastructure projects, it outlined a shift allocations in capital works so that 74.8% of the expenditures would be made at district and subdistrict levels, 8.2% to the Regional level (especially to fund Regional training centers), and 17.0% to the tertiary level. It also addressed concerns about how to deal with the recurrent cost implications of capital expenditures, and to improve the transparency in prioritization of sites and procurement. 47 Considerable infrastructure investments were made in constructing, rehabilitating and equipping clinical and administrative facilities. Other achievements included the enactment of the Procurement Act to codify procurement procedures, the adoption of a health transport policy, and application of procedures to the use of standardized designs and equipment lists for small facilities for CHPS compounds, health centers and district hospitals. However, there was relatively little attention to monitoring the capital investment plan during implementation of the POW-II. Systematic records on capital investment across the sector are not maintained. There are no relevant indicators included in the sector wide indicators, and those identified in the PAD were not used. There are multiple sources of funding, increasingly dependent on donor financing, and multiple authorizations. It appears that about $204.6 million was actually spent on capital works compared to the original forecast of about $100.8 million for 2002-2005. About 68% of spending occurred at the district level and below from 2002-2005, 6.7% at the Regional level, and 25.2% at the tertiary level. The large spending at the district level and below reflect the large number of health centers and health compounds that were built and rehabilitated. The Community-based Health Planning and Services (CHPS) strategy made significant progress in scaling-up from 19 functional CHPS zones in 2001 to about 270 in 2005. Much of the strategies achievements were related to completion of infrastructure components, even as some zones lacked health workers, resources, or the community involvement necessary to make them functional. The increase at the tertiary level represents funding of hospital and central administration of the GHS. The question of who had the authority to spend on capital works became a major point of conflict between the GHS and the MOH, resulting in a loss of transparency in the selection of capital works. In summary, although there was extensive investment in health infrastructure during the POW-II and some improvements in policies and procedures, there remains no overall planning framework and an inadequate monitoring system for capital investments. The health sector is overdeveloping facilities and equipment in some locations and under-investing in other areas. Public sector health facilities appear to have expanded beyond the limits of available operating funds and staffing. 4) Fostering partnerships for health Although fostering partnerships is one of the strategic pillars of the POW-II, progress has been less than planned. There are no directly relevant indicators used at the sector level or in the PAD to monitor progress for the activities described in this area, which include partnerships with the private sector, empowerment of households and communities, and collaboration across sectors. The CHPS strategy was intended to empower communities, though there are little demonstrable changes in empowerment. A Patients' Charter of Rights was passed into law and disseminated throughout the country, including campaigns to reduce "poor staff attitudes". This was accompanied by a provider "Code of Conduct", which was intended to improve staff behavior. Anecdotal reports highlight increases in malpractice lawsuits and improvements in staff behavior. Yet the dissemination and education efforts of the charter and code of conduct were not sustained. It is difficult to assess how much effect they have had on consumer empowerment or on improving quality, though it is unlikely to have much of an enduring effect without continued attention. 48 It is useful to note that the strategic objective concerning partnerships also included partnerships with development partners, which was not part of the description of the focus area. In dealing with development partners, there were some clear successes in working through common management arrangements and continuing the policy and planning dialogue of the SWAp. Yet some development partners increasing moved towards project management support with earmarked funding, including off-budget financing, undermining the effectiveness of the SWAp partnership (see section 3.2 for further details). A major strategic thrust of POW-II was to improve and establish formal commissioning arrangements with non-governmental service providers. Yet during the five-year POW-II, only one Memorandum of Understanding was signed with the Christian Health Association of Ghana (CHAG), an umbrella organization for mission hospitals. This was another example of an area of dispute between the GHS and the MOH, as both organizations felt that they had the authority and responsibility to oversee such compacts. Other initiatives, such as a "Strategic Initiatives Fund" intended to bring together NGOs were labor intensive and small scale operations that did not get beyond the pilot scale. As a result, very little progress was made in engaging the non- governmental sector in health services provision during POW-II. Private sector facilities are still not accounted for during health sector planning or budgeting, nor are they included in monitoring progress in the sector. Although the GOG has long recognized the value and importance of including the non-governmental sector into its dialogue, planning, and monitoring processes, few actions have matched the rhetoric. 5) Improving regulation The POW-II intended to increase consumer protection and empower and make statutory bodies more accountable, and has been partially successful. As described above, the Patient's Charter and provider Code of Conduct seemed to have initial positive effects that have not been sustained. Wider regulatory reforms that are relevant to the health sector include the establishment of a Commissioner for Human Rights and Administrative Justice, which hears complaints to protect consumers, and the passing of a Procurement Act to improve the transparency in procurement processes in the public sector. However, the MOH and the GHS were not very successful in introducing institutions to improve the quality of health care, an important part of the POW-II strategies. The National Health Insurance Council (NHIC) has since established a process for accreditation of health facilities which is needed to receive funds from the NHIF. This includes a Council on Accreditation that collaborates with the GHS and MOH. This appears to be a promising development, but one that was not planned as part of the POW-II, and has yet to demonstrate its effectiveness. 6) Reforming institutional arrangements This component was intended to complete the reorganization of the MOH and the GHS, and implement the service agreements at all levels of the health sector. This is an area that clearly failed during the POW-II. None of the proposed indicators in the PAD were used during the POW-II, but in retrospect, many of the effects of the planned institutional reforms are clear. Probably the most successful institutional reform (though still evolving) involves the introduction and expansion of the National Health Insurance Fund, which was not conceptualized as an institutional reform in the POW-II. The GHS was re-launched in 2003 with renewed efforts to fully implement its legal charter to manage the delivery of health services. Yet institutional conflict and confusion between MOH 49 and GHS consumed much of the time and energy of staff at central level till the end of the POW- II. This conflict was manifest in poor communications, and conflict over the right to control training institutions, contracts with service providers, and the procurement of civil works, commodities, and technical assistance. There were several attempts to mediate differences between the GHS and MOH, usually with the ambition of agreeing on common interests and clarifying roles, but these were not immediately successful. Only as the POW-II ended did relationships between the MOH and GHS begin to improve, but the effects of the conflict not only distracted attention away from implementing the POW-II, but had longer term effects. Donors used the confusion between agencies to bypass the accountability of SWAp over prioritization and financing, in order to push their own agendas. As a result, more earmarked and off-budget spending occurred, and many small projects with project units and special relationships were created. Service agreements were not agreed at the central level between the MOH and the GHS, and at lower levels, the service agreements were not backed up with resources or monitoring and so were not maintained. These service agreements became irrelevant to staff at lower levels, and was reported to be a de-motivating factor. As discussed above, only recently did service agreements with NGO providers come into operation, and it is not clear what effect they have had. 7) Improving the health sector financing This area comprises the fifth strategic pillar of the POW-II, and is one where there were considerable accomplishments throughout implementation of POW-II, with some significant shortcomings in making health financing more equitable. The POW-II identified four main objectives for improvement in health financing: (i) to increase GOG health expenditures; (ii) to enhance prepayment schemes and explore health insurance provisions; (iii) to rationalize and implement a clear exemption policy; and (iv) to allocate resources according to health needs, poverty, and gender needs. There were considerable accomplishments made in improving health financing during implementation of the POW-II, with some significant shortcomings in making health financing more equitable and efficient. As shown in Table 4, the proportion of the GOG budget allocated to health has doubled from 2001 to 2006, reaching the Abuja target of 15% in 2005 and 2006. However, this indicator requires careful interpretation. The total allocation of GOG expenditures on health includes donor, IGF, and statutory funds (the total allocation therefore includes private contributions [IGF] and non-discretionary expenditures [e.g., NHIS]). The proportion of GOG allocations to health out of total GOG public expenditure is lower--about 13% in 2006. 50 Table 4. Sector-wide health financing indicators Indicators 2001 2002 2003 2004 2005 2006 Target* % GOG budget on health 8.7 9.3 9.1 8.2 15 18 15 % GOG recurrent budget on 10.2 11.5 11.2 11.9 14.5 14 health % GOG recurrent health on 8.1 5.9 6.9 5.4 6.6 7 non-salary items (2+3) % spending on districts and 40.9 35.4 37.9 36 40 below, items 2+3 % Earmarked / total DP 62.3 32.8 39.5 26.3 40 61 % IGF from pre-payment 3 schemes % Recurrent funds from 1.2 3.1 GOG+HF allocated to CSOs % Recurrent funds. on 3.6 8 2.2 exemptions Per capita expenditure on 6.3 8.1 10.5 13.5 19 25.4 health (USD) Source: Annual sector review, 2006. * Abuja target of 15%. Although the proportion of GOG recurrent funds allocated to health has been increasing, this has been largely due to the wage bill and dedicated funding for personal emoluments. The proportion of GOG recurrent health spending on non-salary items has wavered around 6%, with an increase to 9% in 2006. Thus, the observed increase in the wage budget has not been met with a commensurate increase in the non-wage budget. As a consequence, the GOG has continually overspent on its capital and wage budgets, and under spent on its non-wage recurrent budgets, compromising the delivery of health services. The decentralization strategy for health financing has met some challenges. The proportion of GOG expenditures at the district level has progressively declined from 2002. This decline indicates less financing at the district level to fill funding gaps, which are often a result of project funding. In addition, the decentralization of budgets has increased confusion in timely reporting, auditing, and budget planning. There is an increasing trend of donors to move to off-budget and other earmarked (project) funding, creating funding gaps at the district level and potentially causing inefficiencies and inconsistencies among national health programs. Figure 1 depicts this increase in donor earmarked funding. However, it is important to note that the percentage of donor funds measured as a percentage of all donor funds may be falsely inflated as it does not include donor contributions to multi-donor budget support. 51 Figure 1. Ghana Health Sector Funding by Source, 2002-2006 Ghana health sector funding, 2002-6, by source 6,000,000 NHIF 5,000,000 Project funding ) 4,000,000 Financial Credits mn( HIPC sid 3,000,000 IGF Ce2,000,000 Donor earmarked 1,000,000 Donor HF GOG - 2002 2003 2004 2005 2006 Source: Annual sector review, 2006. IGF from prepayment schemes has not been easy to track in the past. In 2006, for the first time, it could have been used to distinguish those paying out of pocket from those benefiting from NHIS membership. Exemptions funding increased from 3.6% in 2001 to 8% in 2005, partly as a result of the maternal exemptions policy which was implemented nationally that year. In 2006, exemptions funding dropped to 2.2%. There is some evidence that equity in public financing health is improving slightly. An analysis of the Ghana Living Standards Surveys from 1991-92 to 2005-06 suggests that benefit incidence of public spending at health clinics has improved to a point where it equally distributed across income groups (Coulombe & Wodon 2007). However, the distribution of public spending at hospitals has changed little, and still favors richer segments of the population. The National Health Insurance Scheme (NHIS) is expected to replace the exemption policy, but full implementation of this arrangement will take time. Enrollment for those in the "exempt" category of the NHIS has surpassed expectations, yet there are delays and inconsistencies in obtaining full coverage. This transition may causes gaps in access to services for the poor and, therefore, may ultimately negatively impact their health status. The introduction of the NHIS in 2003 was a landmark achievement for Ghana, which abolished its "cash and carry" system, though in practice it continues to exist to a varying degree in the country. The NHIS enacted substantial financial protection policies, and created a system that would replace the often problematic exemption policy. It also creates new opportunities to leveraging institutional changes to improve quality and productivity in health services, which it has initiated. During this introduction phase, there have been significant delays in reimbursement of claims, leading to cash flow problems to providers, and contributing to ad hoc and undesirable patient charges, undisciplined expenditure controls, and threats to the trust required for the system to work effectively. One effect of the NHIS reimbursement is that funding is skewed toward clinical services over preventive and public health services, and particularly to specialized hospital services over peripheral primary care, which is not in line with the relative priorities of the POW-II. Another effect of the introduction of NHIS is the current confusion among service 52 providers regarding exemption policies and fee for services, such as fees for maternal deliveries, which in many places are being determined on an ad hoc basis. There are several factors that are crucial to the health financing in the sector. Government spending has increased, though it has largely been consumed by increases in salaries rather than numbers of staff, and it has not been linked to performance. The Health Fund, which provided flexible funding to BMCs, has largely disappeared, as more donors have moved to budget support or earmarked funding. The loss of the Health Fund means that BMCs have less flexibility in how they can spend their money, which was reported to be a valuable tool for BMC managers to improve health service delivery. Finally, donor funds have increased as a proportion of total funding, allowing national priority health programs to be financed by more unpredictable financing. 8) Improving financial management systems There was significant progress in improving financing management systems during POW-II in terms of planning, budgeting, and financial management. Much of this happened as a result of using the Health Fund. The Health Fund was initially established with blessing of MOFEP to help establish new financial management systems and to enhance accountability. This may have developed capacity at BMCs and accountability within the Health Fund, but there was no transfer to the regular system itself. As financing of the Health Fund has been cut back, there is a reliance on the procedures of the MOFEP. These procedures have proved more cumbersome and less flexible than the Health Fund, this arrangement poses a significant risk to timely and full disbursement of funds for the health sector. District level financial management was achieved with Budget Management Centers (BMCs) that operated service planning, non-staff budgeting, and procurement. Continuing challenges remain, however, in decentralizing human resources budgets, timely and predictable disbursements, and moving to resource-based budgeting. The further decentralization of BMCs from the district to sub-district level has not really occurred in practice. Although independent auditors assess each of the BMCs, there remains a need to assess how all funds are being spent. The proposed public expenditure tracking survey was not conducted during the POW-II (it was conducted in 2007), even though it was frequently cited as a needed exercise. 9) Further strengthening management This area of work was intended to improve information systems and performance monitoring, build capacity for contract management, operational research, and use of information technology. There were no indicators in the POW-II or the PAD that are directly related to these activities, and there is sparse documentation concerning these activities. Performance contracts implemented at the Regional and District levels were used in an effort to strengthen management capacity for several years. These contracts were largely inconsequential since the sector lacked the ability to finance the contracts. There is some evidence of performance indicator development, and some Regions began creating League tables to evaluate district performance. This evaluation tool allowed for resource allocation to respond to need and/or performance--that is, to provide necessary funds for improvement and/or reward good performance. Although operational research has not played a major role in the dialogue surrounding sector performance, there has been substantial operational research conducted during the POW-II. A 53 systematic review all the operational research conducted in the Ghana health sector during the POW-II is beyond the scope of the ICR, but there are at least three well done studies that are particularly relevant to the POW-II strategies and merit much greater attention: (1) Nyonator et al (2006) concerning the successes and limitations of scaling up CHPS; (2) Binka et al (2007) concerning the effects of a trial of placing nurse and volunteers on reducing child mortality in Navrongo; and (3) Witter & Ajei (2007) on the unintended effects of inconsistent application of exemptions for delivery care. 10) Linking with traditional medicine It is apparent that this was not a priority area for action during the POW-II. There were no substantial resources allocated for traditional medicine, no sector wide indicators to monitor this area, and the proposed indicators of the PAD were not monitored in the SWAp. There were, however, activities to assist in the development of training in traditional medicine and clinical trials of traditional medicines. Conclusions In considering an overall evaluation of the "component" outputs, it is important to realize that not all ten areas are equally important. Indeed, they were never formulated as project components in the POW-II, and were only treated as such by the Bank to fit the documentation requirements of the PAD and the ICR. Some of these areas never received sufficient attention to be given sector- wide indicators (areas 3, 4, 5, 6, 9, 10). Some of the initial focus areas were not carried through as priorities after the first year, as changes in activities focus was intended to be part of the design of the POW-II. In terms of contributing to strategic objectives and vision of the sector, there are some areas, such as the delivery of priority services (area 1), improving financing of the sector (7), and fostering partnerships (4), which are directly relevant. Developing human resources (2) and reforming organizational arrangements (6) also consume large parts of the funding and effort of the SWAp, and are important contributors to the strategic objectives and vision. In examining the outputs in these five areas, the picture is one of qualified success in health service delivery. The introduction of the NHIS is important both to sector financing and reforming organizational arrangements. However, this success is mixed with some clear failures in reforming organizational arrangements of the GHS and MOH, and although there are some successes in increasing numbers and retaining human resources, little success in demonstrating improvements in their productivity or effects on equity. 54 Annex 3. Economic and Financial Analysis (including assumptions in the analysis) There was a limited economic and financial analysis of the program completed at the time of appraisal. No NPV, economic or financial rate of return was calculated a priori for the project, and no analyses are available to assess them as economic or financial results. An analysis of the financing, efficiency, and equity effects of the program are described in Sections 3.2, 3.3, and Annex 2. 55 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Responsibility/ Specialty Lending Supervision/ICR Ferdinand Tsri Apronti Procurement Spec. AFTPC Sylvester Kofi Awanyo Sr Procurement Spec. EAPCO Evelyn Awittor Operations Officer AFTH2 Aissatou Chipkaou Senior Program Assistant AFTH2 Adriana M. Da Cunha Costa Language Program Assistant AFTH2 Gregoria Dawson-Amoah Program Assistant AFCW1 Manush A. Hristov Counsel LEGAF Karen Alexandra Hudes Sr Counsel LEGKM Bernhard H. Liese Consultant AFTH2 Johan Mathisen Consultant AFTP4 Mbuba Mbungu Sr Procurement Spec. AFTPC Jonathan Nyamukapa Sr Financial Management Specia AFTFM Alexander S. Preker Lead Economist, Health AFTH2 Marko Vujicic Economist (Health) HDNHE Frederick Yankey Sr Financial Management Specia AFTFM David Peters Sr. Public Health Specialist HDNHE ICR Jessica St. John Junior Professional Associate HDNHE ICR (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle No. of staff weeks USD Thousands (including travel and consultant costs) Lending FY01 6 41.14 FY02 31 123.79 FY03 35 176.46 FY04 0.00 FY05 0.00 FY06 0.00 FY07 0.00 Total: 72 341.39 Supervision/ICR FY01 0.00 FY02 0.00 FY03 9 41.71 FY04 39 123.51 FY05 44 142.61 FY06 26 100.97 FY07 23 60.63 Total: 141 469.43 56 Annex 5. Beneficiary Survey Results (if any) No beneficiary survey was conducted. 57 Annex 6. Stakeholder Workshop Report and Results (if any) Stakeholders were interviewed during the ICR mission and their comments were incorporated into the main text of the ICR report. No stakeholder workshop was conducted. 58 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR 1.0 Introduction The report is the Ministry of Health's contribution to the Interim Country Report (ICR) prepared the World Bank. ICR is assessed the implementation of the POW II/ HSPSP II which was implemented from 2002 ­ 2006. The first section looks at the background to the assignment and this is followed by the performance outcome of the programme. The next section looks at the challenges and areas of concern and finally the conclusion. 1.1 Background As part of the Health Sector Program Support Project (HSPSP) II, which closed on June 30, 2007, the Bank and the Government of Ghana (GOG) are required to jointly evaluate the HSPSP-II, as is the case for all completed lending operations. Since the HSPSP-II project is designed to support the Ministry of Health's 2002-2006 Program of Work II (POW-II), the evaluation is organized around the POW-II. The vision of the POW-II; "to improve the health status of the population while reducing the geographical, socio-economic and gender inequalities of health outcomes", was used in articulating the HSPSP-II project development objectives. The HSPSP-II also incorporated the sector-wide strategies and performance indicators used in the POW-II, which were reviewed and updated on annual basis during the implementation of the POW II/HSPSP-II. 1.1.1 Objectives The main objective of this evaluation assignment will be to assess performance in the implementation of the recently completed Health Sector Program Support Project II. In specific terms, this assignment seeks to find out the following: · The extent to which the major objectives of the HSPSP-II were achieved, along with other significant outcomes of the POW-II and prospects for sustainability · GOG (Borrower) performance; and · Bank performance and that of other co-financiers of the program. The evaluation report will also as much as possible, provide data and analyses to substantiate assessments made, including the views of GOG and co-financiers of the POW-II on implementation and results, highlighting lessons learnt. 1.1.2 Scope of work Areas to include in the evaluation of the HSPSP-II include: 1. The development impact of the HSPSP-II and the specific questions to be answered in this evaluation report are; · What is the extent to which the major objectives were achieved (or expected to be achieved), are they still relevant, and were they achieved efficiently? · What are the risks that the development objectives will not be achieved or maintained, and what is the potential impact if these risks materialize? 59 · What are the important factors that played a role in the achievement of the development objectives? 2. The effectiveness of GOG and the MOH in preparing, implementing, monitoring and evaluating the POW-II. The specific questions to answer in this regards are; · What is the extent to which GOG ensured quality of preparation, implementation, monitoring and evaluation of the POW-II? · How well the GOG complied with all the covenants and agreements of the Development Financing Agreement toward the achievement of development outcomes? · How effective were the key strategies developed and implemented during the POW- II? Were the organization and management arrangements effective? Could the constraints have been recognized and solved earlier? 3. The appropriateness of Bank assistance · The extent to which services provided by the Bank ensured quality at entry and supported effective implementation through appropriate supervision toward achievement of development outcomes · The attention to fiduciary responsibilities and safeguards, including procurement reviews, disbursements, review of budget and expenditure information and audits, environmental impact assessments, monitoring of progress of key indicators and attention to results 1.1.3 Methodology To achieve the objectives of the assignment a literature survey methodology in addition to interviews with key stakeholders was adopted. Concerning the literature survey, several documents specific to the assignment were reviewed (list can be found in the list of references).This was done to find background information as well as the relevant information to measure performance. In addition to this, the opinion of 4 key stakeholders were also sought (health economist at the country office of the World Bank in Ghana, the country lead of health sector development partners in Ghana and one official each from finance and procurement unit of MOH respectively). Names of specific individuals are attached at the end of the report. 2.0 Programme Outcome This section of the report looks at the implementation outcome of the Health Sector Programme Support Project (HSPSP) via the POW II. Specifically the issues looked at include the development impact of HSPSP, the effectiveness of GOG/MOH in implementing the POW II and the appropriateness of the Bank's assistance. 2.1 Development Impact of HSPSP II The overall development objective of the POW-II was to improve the health status of the population while reducing the geographical, socioeconomic and gender inequalities in health outcomes. The pursuance of this objective is based on several reasons; lessons learnt from the implementation of the first five year medium-term health strategy; POW I, the Ghana Poverty Reduction Strategy (GPRS) and the Millennium Development Goals (MDGs). 60 These as well as government's own priority areas concerning health informed the strategic objectives that were set for the POW II. The development objectives include; 1) Increase Assess (concern was on increasing geographical, financial as well as socio- cultural access). 2) Improve Quality (improving health worker performance, and improving response to client needs). 3) Improve Efficiency (ensuring cost effectiveness, improving planning, management and administration) 4) Improve partnership and collaborations (concerned with improving partnership with households and community, private and public sector providers, ministries, departments and agencies as well as expanding relations with development partners) 5) Improve financing for the health sector (focus on reducing budgetary burden of healthcare for the poor and increase public expenditure on the poor and vulnerable). Various strategies were pursued in the course of implementation to achieve programme targets of the development objectives of the POWW II. The issues discussed in this section focuses on the extent of achievement of the development objectives of the POW II, risk that could other wise have imposed some limitations on the achievement of the development objectives as well as the factors that engineered the current level of success. 2.1.1 Extent of Achievement of development objectives Though baseline indicators and targets existed in the Project Assessment Document (PAD), that was as at 2001 and so in determining the extent of achievement, data from the report of the independent reviewers as at 2006 was used. The rational being that, it gives data that coincide with the programme period (2002 ­ 2006). Data used for the analysis of the development impact of HSPSP II is contained in table 1 in annex 1. The table analyses performance by looking at the level of change and the extent of achievement of the target set for the five broad objectives of the HSPSP II; (Health status, Access, Quality, Efficiency, Partnership and Financing). It is also important to state that these are basically the same as the development objectives that were outlined in the POW II and supported by the World Bank through HSPSP II. Health Status As part of the objectives of the POW II/HSPSP II was the improvement of health status of the Ghanaian population. To measure improvements, four development indicators were defined; infant, under-five and maternal mortality as well as under-five that are malnourished. From the data: · Infant and under-five mortality deteriorated below the baseline by 25% and 2.7% respectively over the programme period. · Under-five who are malnourished (U-5MC) however improved by 28% over the baseline and exceeded the programme target by 40%. · Unfortunately data is currently not available for maternal mortality to be able to measure performance. Access to Healthcare Another objective of POW II/HSPSP II was improvements in access to healthcare. The PAD contains several indicators for tracking improvement in access to healthcare; population to doctor ratio (PDR), population to nurse ratio (PNR), out-patient visit per capita (OVC) and number of specialist outreach carried out (NSC). 61 In terms of performance, all the indicators with the exception of hospital admission rate (deterioration of 6.6% below baseline) recorded positive changes based on base line data. Amongst the five, the most significant was the doctor patient ratio that recorded about 41.4% improvement over the baseline with the programme target being exceeded by 327%. Though the other indicators recorded improvement, they were all 50% or less of the programme target. Quality Improvements Amongst the development objectives of POW II/HSPSP II was improvement in the quality of service. The indicators for measuring quality improvements include; case fatality rate of malaria for under-five (UCFRM), Availability of tracer drugs (ATD) and maternal audits to maternal deaths (MAMD). In all, the indicators recorded improvements with the exception of availability of tracer drugs (13% below the baseline). The data suggest that in spite of the improvements in the other two indicators, MOH could not meet the programme target. The output constituted 59%, 94%, of the programme target for UCFRM and MAMD respectively. Efficiency Improvements Another critical area that the POW II sought to address is the issue of efficiency. To measure efficiency improvements some indicators were selected; (HIV/AIDS prevalence rate, TB cure rate, Number of guinea worm cases, ANC coverage, PNC coverage, Supervised deliveries, EPI coverage; that is DPT3 and measles coverage, Bed occupancy rate, Family planning acceptors and AFP non polio rate. TB cure rate exceeded the programme target by 65% with two of the indicators deteriorating below the baseline (ANC coverage, 5.6% and Bed occupancy, 22%). The rest all recorded improvement from the base but unable to achieve the programme target. With the exception of HIV/AIDS prevalence which achieved 62% of the target the rest were all below 46% of the programme target. Partnerships The POW II also sought to stimulate the level of partnerships especially between the public and private health sector. This was expected to be measured by one indicator, which is the percentage of re-current budget of GOG and health fund used by the private sector, NGOs CSOs and MDAs. Unfortunately, data does not exist currently to be able to measure this indicator Financing The fifth and final objective of POW II was aimed at improving financing to the sector by focusing on reducing budgetary burden of healthcare for the poor and increase public expenditure on the poor and vulnerable. To measure the performance of MOH in achieving this objective, six indicators were put forth with three (% of GOG budget spent on health, % of GOG re-current budget spent on health and % of donor funds earmarked) being discussed in this report because data is not available for the remaining three. All the three indicators recorded improvements over the baseline with the most significant being % of GOG budget spent on health (93.5% increase over the baseline and exceeding programme target by 53%). The remaining two, though exceeded the baseline (% of GOG re-current budget spent on health, 21.7% and % of donor funds earmarked, 2.1%), they were still below the programme target by 29% and 94% respectively. Overall Programme Outcome The figures contained in the analysis suggest that MOH/GOG has performed above average in- terms of the outputs recorded for the POW II/HSPSP II. Responses from stakeholders during the 62 interview session, also creates the impression that performance is satisfactory. For instance, representatives of two DPs scored the performance of MOH/GOG in the implementation of the POWII/HSPSP II at 3.5 on a scale of 1 to 5 with one being the lowest and five being the highest. Not withstanding the above, it is also important that the current figures are interpreted with caution. This is because the Ministry is still confronted with numerous challenges in dealing with a couple of the development objectives of POW II. For instances technical presentations at the just ended annual MOH summit revealed serious issues that border on quality of service at the institutional level. Additionally, the issue of productivity and efficiency is another, which the interview respondent perceived to be very low in the health sector. It may be premature to use these perceptions to invalidate the picture painted by the indicators. Probably, a higher-level analytical work may be needed to establish very reliable levels of productivity and efficiency in the sector. Again, it appears that some of the indicators were computed from facility level data with it attendant errors and biases and thus could be confusing. Perhaps the 2008 demographic and health survey may give a better picture. Notwithstanding the caution raised, the general thinking among stakeholders is that MOH/GOG performed creditably in executing the POW II. 2.1.2 Programme Risk As part of its programme appraisal practice, the World Bank carried out a pre-implementation appraisal and identified about 10 risk factors which they believed at the time posed substantial 3 risk to the achievement of the development objectives of the POW II. With the benefit of hindsight, one cannot but agree with those who did the appraisal that the issues raised could easily constrain the achievement of the development objectives of the POW II. However, MOH/GOG may have done it work well such that the issues raised could not pose the kind of risk that was anticipated. For instance, the first two risks identified included government commitment and support to the health sector. In the analysis section, it is realised that government demonstrated tremendous commitment to the sector by even exceeding the target set in-terms of GOG spending to health. Other areas raised include financial and procurement management, the brain drain and staff attrition and the sector re-organisation. Interview responses suggest that in the area of financial and procurement management programmes were introduced to build the relevant capacity. The analysis also shows that MOH/GOG exceeded it target for the population to doctor ratio, suggesting that significant work was done in this area to reduce the brain drain phenomenon. Perhaps the only area some respondent anecdotally believed constituted a risk and possibly slowed the implementation process was the MOH/GHS split. 2.1.3 Programme Success Factors Respondents of the interview as well as stakeholders in the health sector in general are unanimous in their conviction that the singular most crucial success factor to the implementation of the POW II is the implementation of the National Health Insurance Scheme (NHIS). To them evidence exist that the implementation of NHIS has increased access and that coverage for the poor has 3 In the project appraisal document page 36 and 37 63 improved. Not withstanding the above, the analysis on the management of the anticipated risk also seem to suggest that, GOG's commitment in terms of resource allocation to the sector, the commitment of both MOH and DPs to improve on the financial and procurement management of the sector. Also, strategies to address the brain drain and related human resources crisis, might have all worked together to reduce the impact of risk factors identified and therefore made way for a greater level of success. However, it has also increased the wage bill beyond what is affordable under the budget. It is also important to note that the influence of the donor group at the MDBS forum at the national level could also be responsible for the level of success since at that level they are able to engage the government on allocation of resources to the health sector. Finally, there is also the thinking that the alignment of MOH planning process with the government's budget cycle, as well as consultation with civil society to guide the policy development process, from a curative to preventive approach may have also contributed immensely to the current level of success. 2.2 Effectiveness of GOG/MOH in POW II Implementation This section looks at the effectiveness of GOG/MOH in the preparation of the POW II, its implementation, monitoring and evaluation. The specific questions answered in this section includes the quality of work done by GOG/MOH with respect to the preparation implementation, monitoring and evaluation of the POW II, the level of compliance with covenant agreements and finally the effectiveness of key strategies pursued as well as the CMA. 2.2.1 Quality of the POW II implementation process The MOH is responsible for the preparation of the 5YPOW as well as the Annual Programs of Work (APOW). Preparation is done through broad stakeholder consultation to solicit inputs from agencies of the MOH, DPs CSOs, NGOs etc. Additionally the annual health summit which has broad stakeholder representation is also used as a forum to thoroughly discuss the POW before finalization. It is also said that in recent times the parliamentary sub-committee on health is also engaged in the process by meeting MOH to discuss the health budget. Implementation of the POW is the responsibility of the different agencies of MOH in addition to the private sector. The effectiveness of these bodies in implementing the POW can in general be viewed from the performance of GOG/MOH as depicted by the indicators. The results are clear that good improvements have been made, notwithstanding the fact that the sector still faces some challenges. In the area of monitoring and evaluation MOH carries out it own monitoring as well as joint monitoring exercises with DPs. Additionally the performance of MOH is reviewed every year by the MOH and independent external reviewers, whose reports are made public and also discussed by MOH and DPs and other stakeholders in different fora especially the health summit. There is also the belief among some of the respondents that, the quality of plans, it implementation and monitoring was initially affected by the MOH/GHS split. However, the current thinking is that the situation has improved. 2.2.2 Compliance with agreements The covenants and agreements relating to development financing provided to GOG/MOH from the Bank and other Partners include MOH maintaining financial management systems that are acceptable, preparing budgets and financial reports on a timely basis. Not much information was 64 found with this issue but at least available information suggests that GOG/MOH has duly discharged it obligations. 2.2.3 Effectiveness of key strategies and CMA To achieve it development objectives, the POW II pursued different set of strategies. As to whether these strategies have been effective or not, can best be answered by the output indicators recorded. The table in annex 1 shows that about 26 indicators were measured. On rate of effectiveness of GOG/MOH in executing the POW II, 4 (% of U-5 malnourished, Population to doctor ratio, TB cure rate and % of GOG budget spent on health) out of the 26 indicators had their targets exceeded, 17 recorded improvement over the baseline without achieving the set target with 7 of them deteriorating below the baseline. The details of the individual indicators can be checked from the table. It is important to note that this can only be a preliminary means of assessing the effectiveness of GOG/MOH. It will however be appropriate that a more detailed analytical work is done to ascertain the real level of effectiveness. Another crucial area very relevant to the effectiveness assessment is the implementation of the management arrangement that accompanied the POW II (Common Management Arrangement, CMA). The performance of GOG/MOH in implementing or adhering to the issues in the CMA has been reviewed and the findings contained in the CMA review report as well as the report of the external review team. In both reports, the common issues that were flagged for attention include; 1. Role difficulties between MOH and some of it agencies especially the GHS, which is believed to be the result of the poor management of the transition process (i.e. resulting from the split between the MOH and GHS) 2. The capacity of MOH to perform it supervisory duties in respect of it agencies; a. Coordinating the work of the various agencies b. Performance monitoring and evaluation. In this regard one of the issues raised in both reports is the inability of MOH to sign performance agreements with the agencies as a basis for evaluating their performance 3. Other areas of challenge in addition to the above raised by the CMA review report boarders on poor coordination of earmark funds and delays in the preparation and submission of financial reports at all levels. On the whole the review team of the CMA was of the view that not withstanding the above challenges the management arrangements that accompanied the POW II was implemented by the MOH. As to whether the challenges observed above, could have been recognized earlier and resolved is a fairly difficult question to answer. The last two issues are those that MOH has been dealing with for a long time and therefore may seem normal to expect the Ministry to excel in them. However the first issue; management of the transition process (which seem to be mentioned in several reports) is fairly a difficult one. Evidence in many of the reports that capture this issue seem to suggest that the issue of role definition is quite clear on "paper", however the difficulty has been with the way the actors have interpreted what is written. Perhaps one may not be wrong to admit that all the challenges have served as a good learning experience for the Ministry and it agencies, of course this being the first 65 time in Ghana. That none of the actors in "the drama" regret it because it has opened up new opportunities to cooperate and in commitment push the agenda of the sector forward. 2.3 The Appropriateness of Bank Assistance The bank has had a long standing relationship with GOG/MOH in-terms of it assistance to the sector both financial and technical. In this part of the report, the importance of the assistance offered GOG/MOH in implementing POW II as well as the commitment of the bank in honoring it fiduciary responsibility is assessed. 2.3.1 The importance of Bank's assistance Evidence from the financial statement of MOH indicate that over the five year period of the implementation of the POW II, the cedi equivalent of about US$ 1.5 billion was spent by MOH. Out of this amount, the cedi equivalent of about US$ 163.8 million was contributed by the Bank (i.e. through the World Bank and IDA) and represents about 10.7% of overall spending by MOH for the period. The document also indicate that over the same period, the cedi equivalent of about US$ 419.8 million was contributed by all the donors out of which the World Bank contributed about US$ 163.8 million representing 39% of donor contributions. From an initial contribution of about US$ 7.3 million in 2002, the contributions of the bank rose to the cedi equivalent of about US$ 73.9 million in 2005 and dropped to a negligible figure in 2006 (i.e. Bank's funds sent through MDBS). In terms of actual funds contributed, the figures are clear that the contributions from the Bank were quite substantial and in no doubt assisted GOG/MOH in implementing the POW II. Aside the funds provided, the evidence gathered also indicate that the bank provided technical assistance to GOG/MOH in several areas (policy formulation, implementation, monitoring and evaluation) during the implementation of POW II. One area that staff of MOH interviewed believes could be a reference point for the Bank's assistance to the sector is in the area of procurement. The Bank's assistance has been very key in building and continuously strengthening the capacity of the procurement unit to carry out it functions in the implementation of the POW II. The evidence gathered suggests that the bank was very instrumental in making funds available for the procurement of commodities to support the implementation of the POW II. Additionally the Bank also assisted the Ministry to establish a procurement unit, put in place the right structure and systems, trained staff at all levels and assisted in the development of a procurement manual which became the official document for procurement in the health sector. To some of them, the procurement capacity within the MOH constituted the pilot and the basis for the drafting of the national procurement law which is currently in force. 2.3.2 Fiduciary responsibility As part of the HSPSP II/POW II the Bank had a fiduciary responsibility together with other donors for ensuring that safeguard measures are put in place to reduce risk of poor implementation outcomes through procurement review, disbursements, review of budgets and expenditure information, audits, environmental impact assessments, progress monitoring of key indicators and attention to results. For the purpose of this report the performance of the bank in fulfilling it responsibility is structured financial, procurement and programme performance. In the area PFM, the bank's work was more in the review of budget out-turns and ensuring that financial and audit reports are submitted on time. Due to the on-going financial management reforms most of the things done by the finance unit to implement the POW II was not necessarily 66 as a result of donor demand but rather a product of improvements in national level financial management procedures. This is not however to suggest that the efforts of DPs was irrelevant or not important in this direction. The Bank is also said to have assisted in the procurement of vehicles to support the routine monitoring exercise of the finance unit. In the area of procurement, the Bank's responsibility was more in the area of building the procurement capacity of MOH as well as monitoring procurement outcomes. The evidence above indicates that the Bank discharged this responsibility especially in the area of capacity building. With programme performance the Bank's responsibility together with other donors in terms of monitoring performance of GOG/MOH was also fulfilled to a large extent. This was done through inputs made by them through participation in meetings, reviews, joint donor monitoring programmes and direct technical assistance to the Ministry. 3.0 Challenges of the Implementation Process In spite of the relative success of the implementation of POW II/HSPSP II challenges still exist that are of great concern to some stakeholders. The thinking is that, addressing these challenges will go a long way to improve the performance of the sector and perhaps outcomes in general. One of the issues that seem to cut across is the issue of productivity in the sector. There seem to be the suggestion that current levels of outputs do not match the level of resources injected into the sector, particularly in the light of recent increases in salaries. There is therefore the need for analytical work to establish the current levels of productivity and efficiency in the sector and map out the right strategies to enhance them. There is also the concern that currently; policy planning discussions are more centered on the budget, reducing discussions more to budget issues than to real planning and development issues. There is also the thinking that the current review process lacks analytical rigor and could be taken a step further by incorporating more rigorous analytical work in the review and preparation of documents such as the annual plans. There is also the concern that inconsistencies have existed between the trend of resource allocation and policy objectives of the ministry and therefore the need to take a critical look at this phenomenon. It was also suggested that the current dialogue between the Ministry and DPs could be taken another step forward by ensuring a more open process that is accommodating to the interest of all stakeholders. The urge is that both parties must continue to work together to deepen the dialogue and make it more open as possible. 4.0 Conclusion The implementation of the POW II/HSPSP II has been an insightful experience to the GOG/MOH. If the views of the respondents and the output indicators recorded at the end of the programme period are any thing to go by, then one may not be wrong in concluding that GOG/MOH crossed the pass mark and therefore needs to be encouraged. However, it will be responsible to admit that in spite of the relative success, the Ministry is still faced with great challenges. In addition to those enumerated above, gaps still exist (i.e. between the rich and the poor, the south and the north etc) and therefore making the objectives of POW II/HSPSP II still relevant. A lot more is expected to be done if the sector is supposed to register 67 the desired quantum leaps. Additionally it will also be important that more sustainable strategies are investigated into and used for the implementation of development programmes. Interestingly, respondents are unanimous in their optimism that the sector will continue to improve with the right levels of commitment from all stakeholders and continuous good leadership. 68 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders No comments were received from cofinanciers or other partners, though the ICR was circulated and partners were encouraged to submit comments. 69 Annex 9. List of Supporting Documents Aduonum-Darko, L., A. Nkrumah, P. Nomo. 2006. Review of the Common Management Arrangements for the Implementation of the Health Sector Five Year Programme of Work 2002 ­ 2006. Draft. Benning, Anang, and Partners. 2004. Ghana Health Sector Support Program Final Procurement Audit Report, September, Accra, Ghana. Draft. Benning, Anang, and Partners. 2007. Ghana Health Sector Support Program Final Procurement Audit Report, September, Accra, Ghana. Draft. Binak F.N., A.A. Bawah, J.F. Phillips, A. Hodgson, M. Adjuik, B. MacLeod. 2007. Rapid achievement of the child survival millennium development goal: evidence from the Navrongo experiment in Northern Ghana. Tropical Medicine & International Health 12 (5): 578-583. Campbell, S. M. et al. 2000. Defining quality of care. Soc. Sci. Med. 11: 1611-1625. Christian Health Association of Ghana (CHAG). 2006. Annual Report: Witnessing Christ in the Healing Ministry II, June 2005 ­ May 2006, Accra, Ghana. Coulombe H., Q. Wodon. 2007. Poverty, Livelihoods, and Access to Basic Services in Ghana: An Overview. The World Bank, unpublished document, April 23, Washington, DC. DANIDA. 2006. Semi-Annual Report. Health Sector Support Office, June, Accra, Ghana. Development Partners. 2004. Health Partners Memorandum Joint Response to the 2003 Health Sector Review, May 27, Accra, Ghana. Development Partners. 2005. Health Partners Memorandum Joint Response to the 2004 Health Sector Review, April 21, Accra, Ghana. Ernst & Young/The Auditor General. 2004. Management Letter and Audited Financial Statements, December 31, Minister of Health, Accra, Ghana. Ernst & Young/The Auditor General. 2005. Management Letter and Audited Financial Statements, December 31, Minister of Health, Accra, Ghana. External Review Team. 2003. Report of the External Review Team on the Ministry of Health Programme of Work 2002, May, Accra, Ghana. External Review Team. 2004. Report of the External Review Team on the Ministry of Health Programme of Work 2003, May, Accra, Ghana. External Review Team. 2005. Report of the External Review Team on the Ministry of Health Programme of Work 2004, April, Accra, Ghana. 70 External Review Team. 2006. Report of the External Review Team on the Ministry of Health Programme of Work 2005, March, Accra, Ghana. External Review Team. 2006. Review of the Exemption Policy: A Report of the Annual Health Sector Review 2005, March, Accra, Ghana. External Review Team. 2007. Report of the External Review Team on the Ministry of Health Programme of Work 2006, June, Accra, Ghana. Fedelino, A. G. Schwartz, and M. Verhoeven. 2006. Aid Scaling Up: Do Wage Bill Ceilings Stand in the Way? IMF Working Paper, WP/06/106, International Monetary Fund, Washington, DC. Filmer, D. & L. Pritchett. 2001. Estimating wealth effects without expenditure data--or tears: an application to educational enrollments in states of India. Demography, 38, 115-32 Ghana Health Service. 2007. Volta Regional Health Directorate 2006 Performance Review Report, Volta, Ghana. Johnson, K., S. Rutstein, P. Govindasamy. 2005. The Stall in Mortality Decline in Ghana: Further Analysis of Demographic and Health Surveys Data. Calverton, Maryland: ORC Macro. Ministry of Health. 2001. Partnerships for Health: Bridging the Inequalities Gap. The Second Health Sector Five Year Programme of Work: 2002-2006. SWAp II. Government of the Republic of Ghana. Ministry of Health, Republic of Ghana. 2001. A Bulletin of Health Information: Information for Action, Vol. 1(1). Ministry of Health, Republic of Ghana. 2002. The Ghana Health Sector Programme of Work. Ministry of Health, Republic of Ghana. 2002. A Bulletin of Health Information: Information for Action, Vol. 1(2&3). Ministry of Health, Republic of Ghana. 2003. The Ghana Health Sector Programme of Work. Ministry of Health, Republic of Ghana. 2004. The Ghana Health Sector Programme of Work. Ministry of Health, Republic of Ghana. 2005. The Ghana Health Sector Programme of Work. Ministry of Health, Republic of Ghana. 2006. The Ghana Health Sector Programme of Work. Ministry of Health, Republic of Ghana. 2007. The Ghana Health Sector Programme of Work. Ministry of Health and Development Partners. 2002-2007. Aide Memoires. Ministry of Health and DESERV ­ JM Associates. 2006. Republic of Ghana Health Sector 2002- 2006 Capital Investment Programme Review, Accra, Ghana. 71 Morrison & Associates. 2006. The Republic of Ghana Health Sector 2002-2006 Capital Investment Pgoramme Review (Financial Analysis). Multiple Indicator Cluster Survey. 2007. Monitoring the Situation of Children and Women: Findings from the Ghana Multiple Cluster Survey 2006, Preliminary Report, February. Nyonator F.K., J.K. Awoonor-Williams, J.F. Phillips, T.C. Jones. 2005. The Ghana Community- based Health Planning and Services Initiative for scaling up service delivery innovation. Health Policy Planning 20(1) 25-34. Peters, D.H., A. Garg, G. Bloom, D.G. Walker, W.R. Brieger, M.H. Rahman. 2007. Poverty and Access to Health Care in Developing Countries. Annals of the New York Academy of Science (in press). Republic of Ghana. 2002. Financial Report, December 31, Ministry of Health. Republic of Ghana. 2003. HIV Sentinel Survey 2002 Report. Accra, Ghana: Ghana Health Service. Republic of Ghana. 2004. Financial Report, December 31, Ministry of Health. Republic of Ghana. 2004. HIV Sentinel Survey 2003 Report. Accra, Ghana: Ghana Health Service. Republic of Ghana. 2005. HIV Sentinel Survey 2004 Report. Accra, Ghana: Ghana Health Service. Republic of Ghana. 2006. Financial Report, December 31, Ministry of Health. Republic of Ghana. 2006. HIV Sentinel Survey 2005 Report. Accra, Ghana: Ghana Health Service. Republic of Ghana. 2007. Financial Report, December 31, Ministry of Health. Republic of Ghana. 2007. HIV Sentinel Survey 2006 Report. Accra, Ghana: Ghana Health Service. Republic of Ghana. 2007. Ho Municipal Mutual Health Insurance Scheme, Ho, Ghana. UNICEF. 2004. Report of the Review of the Accelerated Child Survival and Development Programme in the Upper East Region of Ghana, November 2004, Accra, Ghana. Vujicic M., E. Addai, S. Bosomprah. 2006. Methodology for Measuring the Productivity of the Health Workforce in Ghana. Draft unpublished document, December 6, 2006. Witter S. & S. Adjei. 2007. Start-stop funding, its causes and consequences: a case study of the delivery exemptions policy in Ghana. Int J Health Planning Management. 22: 133-143. World Bank. Implementation Status Reports, 2002 ­ 2007. 72 World Bank. 2003. Project Appraisal Document for a Health Sector Program Support Project II, P073649. Washington, DC. World Bank. 2003. Development Financing Agreement for the Second Health Sector Program Support Project between Republic of Ghana and International Development Association. World Bank. 2006. Ghana 2006 External Review of Public Financial Management, Volumes I & II. Washington, DC: World Bank. World Bank. 2006. Implementation Completion and Results Report Guidelines, OPCS, unpublished document, August, Washington, DC. World Bank. 2007. Project Appraisal Document for a Nutrition and Malaria Control for Child Survival Project, P105092, Washington, DC. World Bank. 2007. Project Appraisal Document for a Health Insurance Project, P101852, Washington, DC World Bank. 2007. Project Performance Assessment Report for Ghana: Second Health and Population Project and Health Sector Support Project. 73 Annex 10. Sector-wide Indicator Comparison Table PAD POW-II (2002-2006) POW-2006 Review of POW-2006 Infant mortality rate reduced from IMR per 1,000 reduced to 50 IMR per 1,000 reduced to 50 IMR per 1,000 reduced to 50 57 to 50 per 1000 live births with significant reduction in disparities across regions U5 mortality rate reduced from U5M per 1000 reduced to 95 U5M per 1000 reduced to 95 U5M per 1000 reduced to 95 108 to 95 live births with significant reduction of disparities across regions Maternal mortality is reduced from MMR per 100,000 reduced to 150 MMR per 100,000 reduced to 150 MMR per 100,000 reduced to 150 240 to 150 per 100,000 live births with significant reduction of disparities across regions U5 who are malnourished to 20% U5 who are malnourished to 20% Life expectancy at birth at least Life expectancy at birth at least maintained at 58 years maintained at 58 years Per capital health expenditure on Per capita spending on health health is increased up to about (US$) US$12 in 2006 and public expenditure benefit more to the poor HIV prevalence reduced to 2.6% HIV prevalence reduced to 3% HIV prevalence reduced to 2.6% HIV prevalence reduced to 2.6% TB cure rate increased from 43% TB cure rate TB cure rate increased to 65% TB cure rate increased to 65% to 60% TB case detection rate Use of bednets increased from % use of ITNs for <5 and pregnant 10% to 56% women increased to 55% Antenatal coverage increased Antenatal coverage increased Antenatal coverage increased to Antenatal coverage increased to from 47% to 70% from 47% to 70% 99% 99% Prenatal coverage increased to Prenatal coverage increased to 65% 65% Supervised deliveries increased Supervised deliveries increased Supervised deliveries increased to Supervised deliveries increased to from 44% to 50% from 44% to 50% 60% 60% 50% maternal deaths audited 50% maternal deaths audited Guinea worm disease eradicated Guinea worm disease eradicated Guinea worm disease eradicated Guinea worm disease eradicated 75 PAD POW-II (2002-2006) POW-2006 Review of POW-2006 DPT3 and measles coverage EPI coverage for Penta3 85%; EPI EPI coverage for Penta3 85%; EPI raised and maintained at 80% or coverage for measles 90% coverage for measles 90% more Polio free certification obtained AFP non polio rate less than 1% AFP non polio rate less than 1% % of FP acceptors raised from % of FP acceptors raised to 40% % of FP acceptors raised to 40% 14% to 40% U5 malaria case fatality rate U5 malaria case fatality rate U5 malaria case fatality rate U5 malaria case fatality rate reduced reduced to 1% reduced to 1% Hospital admission rates Hospital admission rates per Hospital admission rates per 1,000 1,000 Outpatient visits per capita Outpatient visits per capita Bed occupancy rates are 80% Bed occupancy rates are 80% No. of specialized community No. of specialized community outreach services carried out outreach services carried out 80% tracer drugs available 80% tracer drugs available 95% tracer drugs available 95% tracer drugs available 80% drugs purchased from CMS 85% transactions meeting agreed standards % of clinical, pharmaceutical, and laboratory practice meeting international quality standards raised from 66% to 80% New incentives and performance management systems in place; 80% staff appraised HR budgets and management decentralized; 100 of BMCs 70% nurses/med assist/doctors doctor to population ratios by doctor to population ratios by posts filled by category and by regions regions regions nurse to population ratios by nurse to population ratios by regions regions 80% graduating at first sitting 50% increase in intake in medical 50% increase in intake in medical and nursing schools and nursing schools 80% of staff receive in-service 80% of staff receive in-service training training 76 PAD POW-II (2002-2006) POW-2006 Review of POW-2006 70% of core staff continuing work 70% of core staff continuing work in Ghana 3 years after graduation in Ghana 3 years after graduation interregional and interdistrict 35% of health professionals with distribution of staff improved short-term placements to the north/deprived areas 40% change in interregional and inter-district distribution of key staff in favor of deprived regions and districts 40% change in intra-district rural/urban distribution in favor of rural areas Community-based health centers 40% of new facilities developed in established in deprived areas. 4 deprived areas 40% of all new facilities developed in 4 deprived regions. Mapping of physical assets 30% of district health facilities completed and infrastructure rehabilitated investment-diversification plan developed to reduce duplication Value of new investments are more than 3% of current level 40% of hospital facilities brought in line with strategy Average transport fleet age Average transport fleet age reduced to 5 years reduced to 5 years 70% of vehicles available at lower levels 80% of vehicles are serviceable 85% of functioning equipment at value 70% of BMCs using private sector 70% of BMCs using private sector maintenance maintenance 80% households have improved health and safety knowledge; 50% households have improved health and safety practices 77 PAD POW-II (2002-2006) POW-2006 Review of POW-2006 40% districts in each region are implementing District Health Plans (DHPs) 80% households have knowledge of consumer rights and responsibilities 80% of CHPS districts are No. of functional CHPS zones No. of functional CHPS zones operational (target 400) (target 400) 4 funded community initiatives funded in each district Ombudsman's office established Ombudsman's office established and 80% of complaints handled and 80% of complaints handled successfully successfully; 60% patient satisfaction level The reorganization of public agencies (MOH/GHS split) completed. MOU signed with mission sector All mission institutions and 20% of and converted into service other NGOs have MOUs and agreements and contracts agreements signed 50% of non-essential services 50% of non-essential services contracted out contracted out 0.5% of the health budget 0.5% of the health budget allocated to the innovation fund allocated to the innovation fund % of Program recurrent funds 2% of Program recurrent funds Recurrent budget from GOG and Recurrent budget from GOG and used by private sector, NGOs, used by private sector, NGOs, health fund allocated to private health fund allocated to private CSOs, and other MDAs CSOs, and other MDAs sector, CSOs, NGOs, and other sector, CSOs, NGOs, and other MDAs (target 2%) MDAs (target 2%) 10% of facilities at district level are private facilities % of GOG budget spent on health % of GOG budget spent on health % of GOG budget spent on health (target 15%) (target 15%) % of GOG recurrent budget spent % of GOG recurrent budget spent % of GOG recurrent budget spent on health on health (target 15%) on health (target 15%) % of pooled donor funds to total % donor funds earmarked (target % donor funds earmarked (target donor funds (per partner) 40%) 40%) 78 PAD POW-II (2002-2006) POW-2006 Review of POW-2006 % of IGF coming from pre- % IGF from pre-payment and % IGF from pre-payment and payment and community community insurance schemes community insurance schemes insurance scheme (target 20%) (target 20%) % of capital and recurrent proportion of non-wage recurrent proportion of non-wage recurrent expenditure by level, region and budget spent at district level budget spent at district level by source (target 43%) (target 43%) total amount spent on exemption % Recurrent budget spent on % Recurrent budget spent on by exemption category exemptions (target 8%) exemptions (target 8%) National agenda for herbal and alternative medicine developed Guidelines for quality assurance developed and disseminated 6000 Traditional and alternative medicine practitioners trained 79 Annex 11. Sector-wide Indicators 1998 ­ 2006 Indicator 1998 1999 2000 2001 2002 2003 2004 2005 2006 Targets Infant Mortality Rate per 1,000 live births 61 57 64 71 50 Under five mortality Rate per 1,000 live births 110 111 95 Maternal Mortality Ratio per 100,000 live births 214 560 150 Under five who are malnourished4 24.9 22.1 17.8 20 HIV sero-prevalence (%) 2.9 3.4 3.6 3.1 2.7 2.9 2.6 % EPI coverage (DPT3/Penta3) 76.3 78.0 76.0 75.0 85.0 84.2 80 % EPI coverage (measles) 67.4 71.0 81.5 82.4 83.7 79.0 78.0 83.0 85.1 80 AFP non polio rate (%) 2.8 1.9 1.3 1.5 1.68 1.55 >1 Guinea worm cases 4733 5545 8290 7275 3992 2,968 0 % Family planning acceptors 14.5 13.7 11.6 24.9 21.0 22.6 24.3 22.6 26.8 40 % Antenatal coverage 94.7 92.2 96.5 98.4 93.7 91.2 89.2 88.7 88.4 70 % Prenatal coverage 37.4 43.1 46.3 52.5 53.6 55 53.3 52.7 55.9 65 % Supervised deliveries 40.8 43.5 50.2 50.4 32.0 55.0 53.4 54.1 44.5 60* % Maternal deaths audited 10.0 84.0 85.0 55.9 75.6 52 50 Outpatient per capita 0.38 0.4 0.45 0.49 0.49 0.5 0.52 0.53 0.52 0.6 Hospital admission rates per 1000 population 27.5 27.9 32.1 34.9 35.3 36 34.5 36.5 32.6 40 Bed occupancy rates (%) 78.1 76.5 58.9 61.6 65.5 64.1 62.7 58.4 50.9 80 Under five malaria case fatality rate (%) 1.7 3.7 3.6 2.7 2.4 2.7 n/a Tuberculosis Cure Rates (%) 44.9 53.8 63.9 65.0 67.6 .. 60 Number of specialized outreach services 141 158 175 158 164 170 200 % Tracer drug availability 79.0 75.0 70.0 85.0 85.0 87.5 84.7 73.8 80 Doctor to population ratio 1:20,036 1:18,274 1:16,759 1:17,615 1:10,380 1:10,700 1:16,500 Nurse to population ratio 1:1,728 1:1,675 1:1,649 1:1,510 1:1,508 1:1,587 1:1,500 No. of functional CHPS zones 19 39 55 84 186 270 400 % GOG budget spent on health 8.5 8.4 8.7 9.3 9.1 8.2 15.0 18.0 15 % GOG recurrent budget spent on health 10.2 11.5 11.2 11.9 14.5 14.0 15 % GOG recurrent health on non-salary (items 2+3) 8.1 5.9 6.9 5.4 6.6 7.0 % GOG spending on districts and below (items 2+3) 50 42 40.9 35.4 37.9 36.0 40.0 43 4Malnourishment is defined as weight for age. The reference population is the WHO/CDC/NCHS reference, representing the percentage of children who score more than 2 standard deviations below the mean. 80 Indicator 1998 1999 2000 2001 2002 2003 2004 2005 2006 Targets % Donor funds earmarked 62.3 32.8 39.5 26.3 40.0 61.0 40 % IGF from pre-payment and insurance schemes 3 10 % recurrent funds from GOG and HF allocated to CSOs 1.2 3.1 2 % recurrent budget spent on exemptions 3.6 8.0 3.0 8 Per capita spending on health ($US) 6.3 8.1 10.5 13.5 19.0 25.4 81