Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu February 2019 World Bank Team Sheena Chhabra Owen Smith Ajay Tandon Valerie Ulep CMCHIS Team T.S. Selvavinayagam, Government of Tamil Nadu Sukeshini R., United India Insurance Co. Ltd. Sainath Iyer, MDIndia Health Insurance TPA Pvt. Ltd. Ajitha Menon, Vidal Health Insurance TPA Pvt. Ltd. T. V. Ramesh, Mediassist Health Insurance TPA Pvt. Ltd. Table of Contents Acronyms v Preface vii Foreword viii Acknowledgements x Summary xi Chapter 1: Introduction 1 Chapter 2: Health in Tamil Nadu 4 Chapter 3: Covering People 12 Chapter 4: Defining Benefits 16 Chapter 5: Managing Money 21 Chapter 6: Improving Supply 29 Chapter 7: Strengthening Accountability 35 Chapter 8: Transition to PM-JAY 37 Chapter 9: Summary and Pending Agenda 43 Table of Contents  iii ANNEXES Annex A: Key Population Health Outcomes 46 Annex B: Universal Health Coverage 47 Annex C: Burden of Disease and Risk Factor Trends for Tamil Nadu and India 48 Annex D: Key Health Financing Indicators 50 Annex E: Tamil Nadu District-Level Variables: Population and Enrollment 51 Annex F: Tamil Nadu District-Level Variables: Facilities and Beds 52 Facilities that Provide the Maximum of Packages Across 10 Specialties and  Annex G:  53 8 High-End Procedures Annex H: District-Level Utilization Rate of CMCHIS 58 Annex I: District-Level Utilization Rate of CMCHIS, by Specialty 59 Annex J: Claims Volume and Value by Age Group 60 iv  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Acronyms AIDS Acquired Immunodeficiency Syndrome ANC Antenatal Care BMI Body Mass Index BPL Below Poverty Line CGHS Central Government Health Scheme CHC Community Health Center CMCHIS Chief Minister’s Comprehensive Health Insurance Scheme DALY Disability-Adjusted Life Year DPT Diphtheria, Pertussis, and Tetanus EDC Empanelment and Disciplinary Committee ENT Ear, Nose, and Throat ESIS Employee State Insurance Scheme EQUAS External Quality Assessment Scheme GDP Gross Domestic Product GNI Gross National Income GSDP Gross State Domestic Product HIV Human Immunodeficiency Virus IHME Institute for Health Metrics and Evaluation Acronyms  v IMF International Monetary Fund IPHS Indian Public Health Standards IRDA Insurance Regulatory and Development Authority LMIC Lower-Middle-Income Country MFP Modern Family Planning MIOT Madras Institute of Orthopedics and Traumatology MMC Mortality and Morbidity Committee MOU Memorandum of Understanding MRI Magnetic Resonance Imaging NABH National Accreditation Board of Hospitals NFHS National Family Health Survey NSSO National Sample Survey Organization NQAS National Quality Assurance Standards OOP Out-of-Pocket PM-JAY Pradhan Mantri Jan Arogya Yojana RSBY Rashtriya Swasthya Bima Yojana SAN Sanitation SDG Sustainable Development Goal SECC Socio-Economic Caste Census SRS Sample Registration System TNHSP Tamil Nadu Health Systems Project TOB Non-Tobacco Use TPA Third-Party Administrator UHC Universal Health Coverage UNICEF United Nations Children’s Fund VAO Village Administrative Officer VHN Village Health Nurse WDI World Development Indicators WHO World Health Organization vi  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Preface Progressive realization of universal health coverage (UHC) is one of the important aspects of the United Nations’ Sustainable Development Goals. The Government of India is fully committed to attaining UHC as reflected by its policies and institutional mechanisms to improve coverage and access to healthcare. India is a highly diverse nation, and states have different approaches towards achieving UHC. Tamil Nadu is one of the pioneer states in implementing such reforms and accumulated rich experiences worth documenting and sharing. The Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS), which initially began implementation in 2009 in Tamil Nadu, provides state-financed noncontributory coverage to over 42 million poor, near-poor, and vulnerable individuals. The organizational and design features, expenditures, potential impacts, innovative practices, and challenges of the scheme were examined during the comprehensive study of the scheme. The integration of the scheme with Ayushman Bharat Pradhan Mantri Jan Arogya Yojana (PM-JAY), a centrally-sponsored scheme introduced in 2018, was also documented. The implementation of PM-JAY, given its magnitude, can potentially change the landscape of India’s Health System. This publication is timely and relevant. In a setting constrained by low public spending for health, significant number of poor and vulnerable, and complex governance and accountability issues, the Government of Tamil Nadu has introduced a mechanism to manage public funds for health and deliver healthcare services to the poor and vulnerable populations. The architecture of CMCHIS and the lessons learned in the last 6 years provide important insights not only for other Indian states, but also for the central government, which is currently at the crossroad of health financing reforms. Other countries can learn as well especially those questing for universal healthcare and implementing pro-poor health insurance schemes with large presence of private sector. Preface  vii Acknowledgements The case study on the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS) was prepared by a joint team from CMCHIS-TNHSP and South Asia Region (SAR), Global Practice, Health, Nutrition and Population (HNP), World Bank. The team comprised of Dr. T.S. Selvavinayagam (Additional Director, Public Health, Government of Tamil Nadu (GoTN)), Ms. Sukeshini Ramaswamy (DGM, United India Insurance Company (UIIC)), Mr. Sainath Iyer (Project Head, MDIndia Health Insurance TPA Pvt. Ltd.), Ms. Ajitha Menon (Project Head, Vidal Health Insurance TPA Pvt. Ltd.), Dr. T. V. Ramesh (Project Head, Mediassist Health Insurance TPA Pvt. Ltd.), Ms. Sheena Chhabra (Senior Health Specialist, World Bank), Mr. Owen Smith (Senior Economist, World Bank), Mr. Ajay Tandon (Lead Economist, World Bank) and Mr. Valerie Ulep (Consultant, World Bank). We will like to express our gratitude to Dr. J. Radhakrishnan, Principal Secretary, Department of Health and Family Welfare, GoTN, Ms. P. Uma Maheshwari, Project Director, Tamil Nadu Health Systems Project, GoTN, and Ms. Rekha Menon, Practice Manager, SAR, World Bank for providing constant encouragement and guidance for documenting the case study. We also appreciate the keen interest and enthusiasm of Dr. Indu Bhushan, CEO, National Health Authority (NHA) and Dr. Dinesh Arora, Deputy CEO, NHA, for the case study to draw lessons for PM-JAY and promote practitioner to practitioner learning among states. The authors are also deeply appreciative of the extensive contributions and support provided by a large number of officials affiliated with CMCHIS. Special gratitude to Dr. Amanda, Dr. Senthil, Dr. Kaarthika, Dr. Parameswari, Dr. Anusha, and Dr. Devasena from TNHSP; Mr. K Shiva Kumar, Mr. S Raman, Mr. Dinesh Reddy, Mr. Varun Kumar Sukhla, and Prema Mukilan, from UIIC; Mr. M Balamuralikrishnan, and Mr. S Valaguru from Vidal Health Insurance TPA Pvt. Ltd.; Mr. T Ambedkar, Mr. C Dinesh, and Ms. Pratheema from Mediassist Health Insurance TPA Pvt. Ltd.; and Ms. B K Deepa Rani, Mr. B Yuvaraj, Mr. G Ramchandran, and Dr. S Janane from MDIndia Health Insurance TPA Pvt. Ltd. for rich inputs and engaging discussions. The team is indebted to Mr. Sainath Iyer for excellent coordination especially of numerous requests for data. We will also like to thank Ms. Aparnaa Somanathan, Dr. Somil Nagpal, Mr. Aakash Mohpal, Mr. George Schieber, Mr. Adrien Dozol, Mr. Christoph Kurowski, Mr. Rajesh Jha, and Mr. Aloke Gupta from the World Bank for the excellent peer review of the document. The World Bank technical support has been provided under the Programmatic Advisory Services and Analytics for Universal Health Coverage in India with financial support by the Bill & Melinda Gates Foundation Trust Fund. x  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Summary This publication summarizes the architecture Although some elements of the scheme’s of the Chief Minister’s Comprehensive Health architecture are similar to those being implemented Insurance Scheme (CMCHIS) which initially began across several states in India, certain modalities of implementation in 2009, in the state of Tamil Nadu,1 CMCHIS are also quite different in important and in India. CMCHIS provides state-financed notable ways, serving to provide valuable lessons noncontributory coverage to over 42 million poor, for other states and countries implementing near-poor, and vulnerable individuals—more than similar programs. For instance, CMCHIS half of Tamil Nadu’s population—for `100,000 includes coverage for diagnostic packages that (~US$1,429) and up to `200,000 (~US$2,858) beneficiaries can also avail of in empaneled health for certain procedures per family per year for care facilities. These diagnostic packages cover an explicitly defined inpatient-focused benefits investigative procedures even if they do not lead package that can be availed of at empaneled to admissions, potentially helping diagnose health government and private facilities. CMCHIS uses a conditions before they become severe enough to ‘mixed’ implementation modality; a public sector require imminent hospitalization. Follow-up care insurance company is paid an annual premium of for up to one year post-hospitalization is also `699 (~US$10) per family enrolled, which forms provided for a preidentified range of conditions. the pool of resources from which providers are Payments under CMCHIS to the public health reimbursed using a case-based method, while insurance company are for a minimum of four years top-up claims to providers for selected high-end and are front-loaded: 95 percent of premiums paid procedures with a higher beneficiary coverage cap for by the government on behalf of beneficiaries are paid for using an ‘assurance’ modality, that is, are advanced on an annual basis at the beginning they are paid directly by the scheme-administering of the year, reducing uncertainty on the timing and agency. regularity of financing for the scheme. Almost all claims are settled within seven days of submission, helping bolster provider confidence in the scheme. 1 This publication has been prepared by a team comprising CMCHIS is also notable for implementing strong members from CMCHIS (T.S. Selvavinayagam, Sukeshini R., Sainath Iyer, Ajitha Menon, T. V. Ramesh) and from the World processes for ensuring provider accountability Bank India team (Sheena Chhabra, Owen Smith, Ajay Tandon, and for preventing, detecting, and deterring fraud and Valerie Ulep). Summary  xi through engaging multiple TPAs and shifting gross state domestic product (GSDP), financing responsibilities over time, preauthorizations and remains thin and the scheme is operating on the claims reviews. In addition, CMCHIS mandates that edge with regard to its overall claims ratio, raising all empaneled private health facilities hold monthly concerns regarding sustainability, especially as outreach ‘health camps’ to provide screening, and when utilization rates increase and to ensure detection, and referrals. CMCHIS has pioneered continued engagement with the for-profit private the implementation of tiered reimbursements for sector. private hospitals, aimed at incentivizing quality of care, in addition to implementing regular reviews In looking to the future, CMCHIS can serve as an of morbidity and mortality outcomes related to important platform to consider options by which admissions at both government and empaneled some of the gains made by Tamil Nadu’s health private hospitals. systems could be sustained and expanded as it faces new challenges and opportunities. As has As CMCHIS gets integrated with the national been the experience of many countries, a longer- Pradhan Mantri Jan Arogya Yojana (PM-JAY) term agenda could be to assess the feasibility of scheme, there remains a pending agenda related the scheme as a mechanism to expand coverage to systematic monitoring and evaluation of the for a larger set of health needs, not just those scheme’s impact on health outcomes and on around relatively infrequent hospitalizations, and financial protection as well as of finding ways to integrate primary and specialist outpatient care for enhancing flexibility for course correction more systematically and holistically to enhance during implementation. Despite best efforts, access to all levels and to provide continuum of there remain concerns that private providers are care, a key consideration as the state undergoes its balance-billing some beneficiaries; furthermore, epidemiological transition toward a predominance a related issue regarding increasing knowledge of non-communicable diseases (NCDs): improving and awareness among beneficiaries of the full health outcomes, reducing financial barriers to range of entitlements provided by the scheme accessing care for the poor and vulnerable, and remains an ongoing challenge, especially reducing the health system’s overall dependence for new additions. At 0.1 percent of the on financing from out-of-pocket (OOP) sources. xii  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu 1. Introduction This publication summarizes the architecture of million, Tamil Nadu is southernmost among India’s the Tamil Nadu state’s government-sponsored 29 states: it is one of the four southern states health insurance scheme. The scheme—currently in India, bordering Karnataka, Kerala, Andhra referred to as the Chief Minister’s Comprehensive Pradesh, and the union territory of Pondicherry. Health Insurance Scheme (CMCHIS)—was initially Tamil Nadu is also geographically close to the launched in 2009. With a population of 79.3 island country of Sri Lanka (Figure 1). In terms Figure 1: Tamil Nadu State in India Chapter 1: Introduction  1 Figure 2: Per capita income (2017) and economic growth (2000–2016): Tamil Nadu versus comparators of population, it is the seventh-largest state in it akin to a lower-middle-income country (LMIC) India comparable in size to countries such as Iran (Figure 2).2 Within India, its income is above the and Turkey, both of which are among the top 20 national average and it ranks 11th among states in largest countries in the world. Tamil Nadu’s per 2 The World Bank currently classifies countries as low income if capita income of about US$2,300—similar to the gross national income (GNI) per capita is less than US$995; LMICs are those with GNI per capita greater than US$995 and that of Nicaragua and Vietnam and comparable less than US$3,895; upper-middle-income countries (UMIC) to that of several other Indian states such as as those with GNI per capita greater than US$3,895 and less than US$12,055; countries with GNI per capita greater than Kerala, Punjab, Gujarat, and Maharashtra—makes US$12,055 are classified as high income. 2  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu terms of per capita gross state domestic product scheme’s architecture are similar to those being (GSDP).3 Per capita economic growth rates in implemented across several state schemes in the state have been close to 7 percent over the India, certain modalities of CMCHIS are also quite past 15 years—both before and after the 2009 different in important and notable ways, serving launch of CMCHIS—making Tamil Nadu one of to provide valuable lessons for other states and the fastest growing regions of the world. Strong countries implementing similar programs. In economic growth has resulted in a decline in describing CMCHIS, the focus is on assessing key poverty rates to about 12 percent of the population elements of its architecture, highlighting innovations in the state.4 Tamil Nadu is notable for being one and ‘good-practice’ elements where evident, of the most literate states in India (80 percent of while also noting areas where there remains a the population is literate).5 It has 32 districts and is pending agenda for improving the performance of highly urbanized.6 the scheme as it gets integrated with the national Pradhan Mantri Jan Arogya Yojana (PM-JAY) for CMCHIS currently provides coverage to more continued implementation. than half of Tamil Nadu’s population. CMCHIS provides state-financed noncontributory coverage The remainder of the publication is organized to over 42 million poor, near-poor, and vulnerable as follows. The framework draws on a recent individuals—more than half of Tamil Nadu’s World Bank publication, Going Universal: How 24 population—for up to `100,000 (~US$1,429) and Developing Countries are Implementing Universal up to `200,000 (~US$2,858) per family per year Health Coverage Reforms from the Bottom for an explicitly defined inpatient-focused benefits Up, which identified the ‘nuts and bolts’ of pro- package that can be availed of at both government poor universal health coverage (UHC) reform and empaneled private facilities. CMCHIS uses a implementation globally across five broad policy ‘mixed’ implementation modality: a public sector areas: covering people, defining benefits, managing insurance company is paid an annual premium of money, improving supply, and strengthening accountability.7 Before doing so, the next section `699 (~US$10) per family which forms the pool of summarizes health-relevant background for the resources from which providers are reimbursed state of Tamil Nadu to help position the state in using a case-based method, while top-up claims a national and global context and as a prelude to to providers for selected high-end procedures introducing CMCHIS. Subsequent sections outline with a higher coverage cap are paid for using an (a) how CMCHIS covers people, including eligibility ‘assurance’ modality, that is, they are paid directly and enrollment; (b) how benefits are defined and by the scheme-administering agency. how they evolved over time; (c) how resources are CMCHIS includes several innovative design and managed and expended; (d) how CMCHIS includes elements for improving supply of health services, implementation elements that can potentially both for government and empaneled private provide useful lessons for other states in India as hospitals; and (e) how accountability is being well as globally. Although some elements of the strengthened under the scheme. The publication 3 Reserve Bank of India. 2018. Handbook of Statistics of Indian ends with an overview of integration of CMCHIS States. Mumbai: Government of India. with PM-JAY as well as a summary of key lessons 4 Reserve Bank of India. 2018. Handbook of Statistics of Indian States. Mumbai: Government of India. that the architecture of CMCHIS potentially offers 5 In Tamil Nadu, the literacy rates among women and men were to other states in India under PM-JAY, as well as 73 percent and 87 percent, respectively; in India, on average, the literacy rates were much lower for both genders: 65 percent for globally. women and 81 percent for men (Reserved Bank of India, 2018). 6 In Tamil Nadu, almost half of the population lives in urban areas, 7 Cotlear, D., S. Nagpal, O. Smith, A. Tandon, and R. Cortez. 2015. significantly higher than less than one-third for India on average; Going Universal: How 24 Countries Are Implementing Universal Ministry of Statistics and Program Implementation. 2017. Selected Health Coverage Reforms from the Bottom Up. Washington, DC: Socio-economic Statistics. New Delhi: Government of India. World Bank. Chapter 1: Introduction  3 Health in Tamil Nadu 2.  Tamil Nadu’s population has undoubtedly average ever since. Tamil Nadu has already met become healthier over the past several decades. the child and maternal health 2030 United Nations Life expectancy at birth has steadily increased Sustainable Development Goals (SDGs) of an to 71 years in 2015, up from 64 years in the late under-five mortality rate of less than 25 per 1,000 1990s and close to 50 years in 1970 (Figure 3). The live births and a maternal mortality ratio of less infant mortality rate has declined from over 100 than 70 per 100,000 live births. Unlike the average in the 1970s to 50 in the late 1990s to less than for India, both under-five mortality and maternal 20 per 1,000 live births in 2016. Whereas Tamil mortality ratios are far better than what might be Nadu’s health outcomes were similar to the Indian expected for Tamil Nadu’s income level (Figure 4). average in the early 1970s, it is notable that the Tamil Nadu’s total fertility rate of 1.6 is among the state has consistently outperformed the national lowest in India, comparable to that of much richer, upper-middle-income countries (UMIC) such as China and Thailand (see Annex A). Figure 3: Key population health outcomes in Tamil Nadu, 1970–2016 Tamil Nadu is committed to attaining UHC for its entire population. UHC—a policy commitment that is also part of the 2030 SDGs—is about ensuring that all people can use the promotive, preventive, curative, rehabilitative, and palliative health services they need, of sufficient quality to be effective while also ensuring the use of these services does not expose the user to financial hardship. The global monitoring framework of the World Health Organization (WHO)-World Bank recommends tracking a mix of preventive and promotive treatment service coverage and financial coverage indicators to assess progress toward UHC. Recommendations under preventive and Source: Sample Registration of India and Reserve Bank of India promotive coverage include monitoring access 4  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Figure 4: Under-five mortality rate and maternal including the proportion of households mortality ratio relative to income, 2015 that spend more than 10–25 percent of their income or consumption on health and the proportion of households that are impoverished as a result of high OOP spending.8 It is notable that recommended UHC indicators do not include the proportion of population that has insurance coverage as that is a means and not an end in itself: the focus is on effective coverage of services and of financial protection when seeking care. One reason for Tamil Nadu’s relatively good performance on population health outcomes is due to high coverage rates for key UHC-related service coverage interventions. Tamil Nadu’s coverage indicators for key health services such as access to modern family planning methods, ANC, skilled birth attendance, immunization, and tuberculosis treatment rates are greater than 80 percent, higher than India’s and the average for LMICs and, in some cases, better than high- performing countries such as Sri Lanka and Turkey (see Annex B). The areas where Tamil Nadu shows some weakness are in access to sanitation and for financial protection indicators because, as also Source: Sample Registration of India and Reserve Bank of India discussed later, OOP remains the largest source of financing for the health sector. Other to modern family planning methods; antenatal factors that contribute to Tamil Nadu’s success in care (ANC) for pregnant women; skilled birth health include training and deployment of village attendance; diphtheria, pertussis, and tetanus health nurses, investments in a strong network of (DPT3) immunization coverage; non-prevalence 2,226 primary health care centers (of these, 416— of tobacco smoking; access to improved water almost 20 percent—are open for 24 hours), and sources; and access to improved sanitation, among development of institutions such as the autonomous others. Recommended treatment interventions Tamil Nadu Medical Services Corporation that have include treatment rates for tuberculosis, been critical to ensure a reliable supply of essential hypertension, diabetes, and antiretroviral therapy coverage for those with HIV. Recommended 8 WHO (World Health Organization) and World Bank. 2017. financial coverage indicators include those derived Tracking Universal Health Coverage: Global Monitoring Report. from high levels of OOP health expenditures Geneva: WHO. Chapter 2: Health in Tamil Nadu  5 medicines.9 A range of non-health system factors lost due to morbidity and premature mortality in also explains why Tamil Nadu has experienced 2016 (Figure 5).11 Ischemic heart disease was also better health outcomes than other states in India. the leading cause of premature mortality in Tamil Tamil Nadu enjoys a relatively higher level of Nadu. Other NCDs such as diabetes have more education and women’s empowerment: the state than tripled their share of the disease burden over has been a pioneer in the introduction of programs 1990–2016 (see Annex C). New risk factors related to enhance school enrollment, particularly among to urbanization and lifestyle factors are growing. girls. Even before the National Right to Education Ageing is one contributory factor—7 percent of Act was passed in 2009, the state made major the population of the state is over 65 years of age legislative efforts for the universalization of and this share is growing rapidly—although the education in the early 1990s. Along with sustained prevalence of NCDs among younger age groups programs in education and health, Tamil Nadu has in Tamil Nadu is also increasing. Unhealthy diets, relatively good infrastructure (for example, roads high fasting plasma glucose, and hypertension and transportation). Tamil Nadu was one of the are prominent among the top 10 risk factors first states to achieve almost universal access to contributing to the overall disease burden in the electricity (for example, the electrification rate in state. The share of high body mass index (BMI) rural Tamil Nadu is 85 percent compared to 40 as contributor to DALYs lost has increased more percent in rural Uttar Pradesh). Tamil Nadu has a than fivefold and high total cholesterol has more long history of democratized administration and than doubled over 1990–2016. Alcohol and drug this has helped improve the responsiveness of the use in Tamil Nadu is also rising in its contribution health system.10 to the overall burden of disease. Despite recent reductions, the prevalence of smoking remains Non-communicable diseases (NCDs) now high, particularly among men, with 32 percent of account for almost two-thirds of the burden men currently using tobacco products. of disease in Tamil Nadu, higher than their overall average share of 55 percent for India Despite notable progress on key population health (Figure 5). Whereas in 1990 only about 38 percent outcomes, several challenges remain especially of morbidity and mortality in Tamil Nadu was with regard to malnutrition, adult survival rates, due to NCDs, by 2016, this number had risen to NCDs, and inequalities in outcomes. Although 65 percent (Figure 5). This trend is expected to the infant mortality rate in some districts such as continue in coming years. Ischemic heart disease Kanchipuram and Erode was less than 10 per 1,000 was responsible for the largest share of the overall live births, the rate was more than double in other disease burden in Tamil Nadu, causing 14.3 districts such as Thoothukudi and Dharmapuri. percent of all disability-adjusted life years (DALYs) Infant mortality rates were also sharply lower among higher socioeconomic groups. Stunting rates in 9 Balabanova, D., A. Mills, L. Conteh, et al. 2013. “Good Health at Low Cost 25 Years On: Lessons for the Future of Health Systems children remain high, having barely improved from Strengthening.” Lancet 381: 2118–2133. 30 percent in 2005 to 27 percent in 2015. Despite 10 Muraleedharan V., U. Dash, L. Gilson. 2011. “Tamil Nadu 1980- 2005: A Success Story in India.” In Good Health at Low Cost 25 substantial efforts, screening for NCDs remains Years On: What Makes a Successful Health System? edited by D. Balabanova, M. McKee, and A. Mills. London: London School of Tropical Medicine and Hygiene; Harmer, A. 2011. “Improving 11 Whereas in India as a whole, ischemic heart disease is also the Lives of Half the Sky: How Political, Economic, and Social responsible for the top slot, the remainder of the disease burden Factors Affect the Health of Women and their Children.” In Good is quite different. For example, tuberculosis remains in the top 10 Health at Low Cost 25 years On: What Makes a Successful Health for India but not for Tamil Nadu, diabetes is the second-largest System? edited by D. Balabanova, M. McKee, and A. Mills. cause of disease burden in Tamil Nadu but is not yet in the top 10 London: London School of Tropical Medicine and Hygiene; Sinha, for India, and stroke is high for India but does not appear in the D. 2016. Women, Health, and Public Service in India. New York: top 10 for Tamil Nadu. Similar differences in rankings are notable Routledge. for risk factors (see Annex C). 6  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Figure 5: Burden of disease in Tamil Nadu and India, 1990–2016 12 months were advised to quit smoking; in comparison, in the neighboring state of Karnataka, 80 percent of the smokers who visited a doctor in the past year were advised to stop smoking.13 Tamil Nadu has mixed public- private provision of health services and dual practice is legal. Tamil Nadu is one of the states with the highest levels of human resources for health relative to its population: there are an estimated 1.6 physicians as well as 4.1 nurses and midwives per 1,000 population, almost double the national average of 0.8 physicians and 2.1 nurses and midwives per 1,000.14 There are approximately 71,000 government hospital beds (1 per 1,000 population) in the state, significantly higher than the Indian average (0.5 beds per 10,000 population).15 When asked where household members went when they were sick, almost two-thirds reported using government facilities, higher than the proportion on average for India (Table 1). As with the nationwide average, 13 The prevalence of tobacco use among adults in Karnataka is low—less than 30 percent of individuals 30 years 23 percent, slightly higher than in Tamil Nadu (20 percent); Tata Institute of Social Sciences and Ministry of Health and Family or older are screened annually for hypertension Welfare. 2017. Global Adult Tobacco Survey GATS 2 India and diabetes and around 20 percent of women 2016–17. New Delhi: Government of India. 14 Ministry of Health and Family Welfare. 2018. National Health aged 30 years or older are screened for cervical Profile. New Delhi: Government of India. and breast cancer.12 Only slightly more than half 15 Number of hospital beds includes beds in primary care health centers; reliable data on the total number of private hospital of the smokers who visited a doctor in the past beds are not available; industry estimates indicate total beds in India to be roughly 1 per 1,000 population. Ministry of Health and Family Welfare. 2018. National Health Profile. New Delhi: 12 National Family Health Survey (NFHS-4) (2015–2016). Government of India. Chapter 2: Health in Tamil Nadu  7 Table 1: General utilization-related indicators Where do people generally go for treatment when sick? (%) Government Private Tamil Nadu Bottom 40% 82.7 17.1 Top 10% 20.5 79.3 All 63.3 36.5 India Bottom 40% 48.1 46.3 Top 10% 27.7 71.3 All 44.9 51.7 Source: NFHS-4. the bottom 40 percent in Tamil Nadu are far more higher among the poor, whereas the share of NCDs likely—by a wide margin—to self-report utilization (such as cardiovascular diseases and cancer) was at government facilities compared with the top 10 slightly higher among the rich.18 Notably, public percent of the population.16 inpatient utilization rates were higher among the poor versus richer quintiles in the state. Outpatient Almost 6 percent of the population reported using utilization rates in Tamil Nadu are also double the inpatient care services (excluding childbirth) national average, with the poorest quintiles more in the previous year. As is commonly observed likely to use public facilities; richer people were far globally, self-reported illnesses in the past 15 days more likely to use private facilities when seeking were much higher among the rich relative to the outpatient care. poor (Table 2).17 Although they are significantly higher than the national average (3.7 percent), At less than 3 percent of the state’s economy, inpatient care utilization rates vary significantly total expenditures on health are relatively low in across economic quintiles: household survey data Tamil Nadu. Recent estimates indicate levels of per estimates indicate that the poor had significantly capita spending of only about `4,101 (~US$57), lower rates of hospitalization compared to the about 2.8 percent of the state’s economy. Health rich. Infectious diseases (such as tuberculosis, spending has remained low despite increasing from diarrhea, sexually transmitted infections, malaria, `1,703 (~US$24; 2.1 percent of the economy) a diphtheria, and whooping cough) were the most decade earlier. Per capita health spending levels are common reason for hospitalization, accounting for generally higher in the states of Kerala, Maharashtra, almost a fourth of all admissions; this was followed Punjab, and Karnataka when compared with Tamil by injuries and cardiovascular diseases. The share Nadu. This low level of spending on health is of hospitalization related to infectious diseases was even more stark when compared to countries at similar levels of income. In Ghana and Vietnam, 16 Among those not using government facilities in Tamil Nadu, the for example, total health spending was more than most common reason given was that ‘waiting time was too long’ followed by ‘poor quality of care’, ‘no nearby facility’, and ‘facility 5 percent of the gross domestic product (GDP). timing not convenient’. 17 See: Tandon, A., C. J. L. Murray, J. A. Salomon, and G. King, Globally, LMICs on average spend US$121 per 2003. “Statistical Models for Enhancing Cross-Population capita on health, about 5.4 percent of GDP (see Comparability.” In Health Systems Performance Assessment: Debates, Methods and Empiricism, edited by C. J. L. Murray and D. B. Evans. Geneva: World Health Organization. 18 NSSO 71st round. 8  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Table 2: Self-reported illness and utilization rates, Tamil Nadu versus India (%) Tamil Nadu India Poorest Second Middle Fourth Richest All Self-reported illness (last 15 days) 12.8 14.3 16.0 18.8 23.4 16.9 10.0 Outpatient (last 15 days) 11.3 12.6 13.4 17.0 22.6 15.2 8.4 Public 7.2 5.4 5.7 6.1 4.0 5.7 2.3 Private 4.9 8.1 8.8 12.6 20.2 10.7 6.7 Inpatient (last 365 days) 4.4 4.1 5.7 6.5 7.7 5.6 3.7 Public 2.3 2.2 2.0 2.1 1.1 1.9 1.5 Private 2.0 1.8 3.7 4.4 6.4 3.6 2.2 Source: National Sample Survey Office (NSSO) 71st round. Annex D).19 Tamil Figure 6: Public financing for health as share of Nadu’s attainment of GDP versus income good health outcomes is notable despite relatively low levels of financing for health. OOP remains the primary source of financing for health in Tamil Nadu. OOP financing for health accounted for 68 percent of total health expenditures, one of the highest shares in the world, comprising, in large part, expenditures on drugs and diagnostic services.20 This was followed by central Source: MOHFW and World Development Indicators. and state government expenditures, accounting for 25 percent of total over the last 10 years—the overall composition health expenditures. As in India, on average— of health financing has remained relatively despite increases in levels of health expenditures unchanged in the state (for example, OOP was 70 percent of health expenditures in 2005). At 19 Tandon, A., J. Cain, C. Kurowski, and I. Postolovska. 2018. “Intertemporal Dynamics of Public Financing for Universal Health less than 1 percent of the state’s economy, public Coverage: Accounting for Fiscal Space Across Countries.” HNP financing for health in Tamil Nadu is among the Discussion Paper, World Bank, Washington, DC. 20 Tandon et al. 2018. lowest in the world. Recent estimates indicate Chapter 2: Health in Tamil Nadu  9 Table 3: Average OOP spending per hospitalization and outpatient visit in Tamil Nadu versus India (`) Tamil Nadu India Poorest Second Middle Fourth Richest All Outpatient 323 (11%) 347 (6%) 581(9%) 595 (8%) 730 (6%) 552 (8%) 721 (11%) Public 137 (5%) 100 (2%) 127 (2%) 175 (2%) 189 (3%) 140 (3%) 557 (10%) Private 677 (20%) 533(9%) 847 (13%) 769 (11%) 830 (7%) 763 (10%) 831 (12%) Inpatient 9,545 (20%) 11,491 (18%) 14,846 (20%) 19,789 (23%) 36,675(23%) 20,397 (23%) 20,381 (21%) Public 2,067 (6%) 2,311 (6%) 2,686 (3%) 9,200 (3%) 5,372 (3%) 3,879 (5%) 8,615 (11%) Private 18,213 (36%) 22,215 (32%) 21,175 (28%) 24,742(27%) 42,056 (26%) 29,047 (28%) 27,674 (26%) In parenthesis: the average OOP spending as a share of total consumption expenditure. Those who did not utilize were excluded from Note:  the mean. It is possible that an individual may have utilized both inpatient and outpatient services. In such cases, the expenditures for inpatient and outpatient visits were calculated separately. Source: NSSO 71st round. per capita public financing for health in the state to use toward more productive activities. In Tamil to be `1,025 (~US$15), despite having tripled in Nadu, the high levels of OOP result in almost a real per capita terms over the past decade, largely quarter of households spending more than 10 because of strong economic growth (Figure 6); percent of consumption expenditure.21 OOP public spending on health as a share of GDP and payments encourage people to deter and delay as a share of the state government expenditure care, making it also an inefficient source of has remained largely unchanged over the past financing for health because health problems are 10–15 years. addressed later once they have advanced rather than earlier when they can be prevented. OOP Reducing OOP and improving financial financing also constrains the redistributive capacity protection, especially for the poor and vulnerable, of health financing systems and reduces the ability by increasing and effectively targeting public to implement pooled purchasing of health services financing for health are key for making progress to reduce and control costs. toward UHC. Household survey data estimates indicate average OOP spending per hospitalization With this backdrop, CMCHIS was introduced of `20,397 (~US$291): `3,879 (~US$55) for public by the state to provide quality health care and and `29,047 (~US$415) for private (Table 3). Overall reduce financial hardship for poor and near- OOP payments for inpatient care were similar in poor families. The stated objectives of CMCHIS magnitude in the state to those in India; however, are to “…provide quality health care to eligible OOP payments for outpatient utilization in the persons through empaneled government and public sector tended to be much lower in Tamil private hospitals and to reduce the financial Nadu. Interestingly, OOP payments for inpatient hardship for enrolled families and move towards and outpatient care incurred in the public sector UHC by effectively linking with the public health as share of total expenditure were also much lower system.”22 As noted earlier, CMCHIS targets in Tamil Nadu compared to the national average. poor and near-poor families and provides them OOP financing connects access to ability to pay for services and is, therefore, inherently inequitable. 21 These numbers are likely to be inflated when compared with global estimates as total household consumption was underestimated OOP financing is a risk factor for impoverishment, in the 71st round of the NSSO. pushing households into poverty while reducing 22 Government of Tamil Nadu. 2017. Tender Document to Select Insurance Company to Implement: Chief Minister’s the availability of scarce resources for households Comprehensive Health Insurance Scheme. 10  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu with noncontributory ‘cashless’ coverage—that Figure 7: Organizational framework for is, coverage without any beneficiary premium or CMCHIS co-payments—for an explicitly defined inpatient- focused benefits package in the government and empaneled private hospitals. The Tamil Nadu Health Systems Project (TNHSP), a registered society under the purview of the state’s health department that was initially set up to implement a World Bank project, is the scheme-administering agency for CMCHIS. From 2009–2012, the scheme—referred to then as the Chief Minister Kalaignar’s Insurance Scheme for Life Saving Treatments—contracted a private insurance company for implementation. In 2012, the name of the scheme was changed to CMCHIS and a public sector insurance company — the United India Insurance Company Limited—was contracted.23 The public sector insurance company, in turn, has contracted three third-party administrators (TPAs): Vidal Health Insurance TPA, Mediassist Health Insurance TPA Private Limited, and MDIndia Health Insurance TPA Private Limited (Figure 7).24 Settlement of claims to providers is managed by the insurance company and the TPAs; however, where there is a public purchasing agency that settlements for selected high-end procedures purchases services directly from providers; the over and above an augmented beneficiary cap former modality uses an insurance company as are paid for by the scheme-administering agency intermediary to purchase services from providers). directly to providers from a corpus fund which sets aside 27 percent of reimbursements to Additional details regarding the architecture government hospitals. In doing so, CMCHIS can be of CMCHIS are provided under five different characterized as having a ‘mixed’ implementation subheadings: covering people, defining benefits, modality—that is, a combination of both insurance managing money, improving supply, and and assurance models (the latter modality is one strengthening accountability. 23 United India Insurance Company Limited is a public general insurance company headquartered in Chennai. 24 All three TPAs are licensed by India’s Insurance Regulatory and Development Authority (IRDA). Chapter 2: Health in Tamil Nadu  11 3. Covering People CMCHIS targets families whose annual income they are not denied coverage even if they do not is below `72,000 (~US$1,029). In addition, have Aadhar linking in the scheme’s database. In irrespective of income, families who are members other terms, this requirement is for financing of the of 26 welfare boards constituted by the state scheme but not for determining entitlement: the government including agriculture, construction, onus is on the insurer to ensure Aadhar linking, not manual laborers, auto rickshaw drivers, artists, the beneficiary.25 goldsmiths, tribal persons, differently abled persons, orphans, and refugees from Sri Lanka Administrative data indicate that the scheme are also eligible for coverage under the scheme. provides coverage to more than half of Tamil Families must procure an income certificate from the Nadu’s population. A total of 15.7 million village administrative officer (VAO) or from revenue families are currently enrolled under the scheme, authorities along with a family ration card to enroll more than half the total number of estimated in the scheme. VAOs generally have knowledge families in the state; of these, to date, 14.7 million and awareness on the eligibility of people dwelling have also been Aadhar linked. Initially, in 2012, in the village and their backgrounds. For those almost 13.4 million families were enrolled in the families who are members of eligible welfare scheme; this number rose to over 15 million in boards, membership cards must be produced at 2016 (Figure 8). Of all enrolled, two-thirds are the time of enrollment. Migrants from other states estimated to be those living below the income who are certified by the Labour Department that threshold of `72,000 per year; an additional 5 they have resided in the state for more than six million families are members of the agricultural months are also eligible for coverage. At the time welfare board; the remaining 100,000 or so of enrollment, details shared by the beneficiary are families are Sri Lankan refugees and members verified with the database of the Food and Civil of other welfare boards. The total number of Supplies Department. As of 2017, Aadhar card individuals —as opposed to families—covered is linking is also required and premiums are paid to not known precisely; the average family size is the insurance company based on the number of estimated under the scheme to be 2.7, implying families whose enrollment in the scheme has also 25 Aadhar is the unique 12-digit identity number that can be been Aadhar linked for at least one member of the obtained by all residents of India based on their biometric and demographic information; as of July 2018, the Government of family. Nevertheless, as long as a family is eligible, India has issued an estimated 1.2 billion Aadhaar cards. 12  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Figure 8: CMCHIS enrollment trends, 2012–2018 The unit for CMCHIS eligibility and enrollment is the family and not the household. Coverage is extended to the head of the family, the legal spouse, dependent parents, and dependent children (the latter if they are unemployed, unmarried, or below the age of 25, whichever is earlier) as long as they are listed on the family ration card.26 Children who are employed, married, or above the age of 25 are not considered dependents and must demonstrate income eligibility and register separately under the scheme as a different family (even if they continue to live in the same household). There Source: CMCHIS are no restrictions or caps on family size, but—as explained later— the Figure 9: Distribution of enrolled families by annual coverage expenditure cap district, 2018 of `100,000 (~US$1,429) and up to `200,000 (~US$2,858) for certain specified procedures is determined at the family level and is the same regardless of family size. Enrollment awareness under the scheme happens through information and education campaigns and during monthly outreach camps. The insurer is responsible for awareness generation activities, in consultation with the scheme-administering agency, and is incentivized to do so given that each new Aadhar-linked enrolled family implies additional premium transfers. Monthly health camps, mandated for private Source: CMCHIS providers, are another important part of raising awareness regarding coverage for approximately 42 million individuals, the scheme; the providers have an incentive to more than half of Tamil Nadu’s current estimated population. The distribution of enrollees across 26 This is how most government-sponsored health insurance the state varies (Figure 9 and Annex E). schemes have implemented their coverage. Chapter 3: Covering People  13 Figure 10: Specimen of CMCHIS ‘smart card’ hold these camps for demand generation.27 These selected biometric information (Figure 10).29 health camps are also meant to provide free The card also clearly displays a number of a call basic primary care services to attract attendance center, which is run by TPAs, that provides 24/7 by beneficiaries, in addition to identifying and access to information and grievance redressal referring cases for hospitalization. In addition to services regarding the scheme. CMCHIS allows health camps, some beneficiaries have reported for beneficiaries to download an e-card from finding out about the scheme from Village Health the public portal of the scheme, searching the Nurse (VHN). From a purely financial perspective, scheme’s online database using their family both the insurer and provider have incentives to card number, name, village, district, and taluk ‘cream-skim’—that is, select beneficiaries who are (that is, subdistrict). The smart card also allows at low risk for utilization (for insurers) and to select beneficiaries (and their providers) to access their beneficiaries for high-margin procedures (for medical history and claims records online. providers)—but there is no evidence to suggest that this is happening under the scheme. Only a Recent survey estimates confirm administrative relatively small number of new enrollees appear to data of CMCHIS coverage numbers.30 Household have been adversely selected, that is, they enrolled survey data estimates indicate 57 percent of all just before seeking care. Despite efforts and household reporting coverage under CMCHIS, improvements, awareness in the population about similar to the estimated coverage numbers from eligibility and exact nature of benefit entitlements CMCHIS administrative data. This number rises remains an ongoing challenge.28 to 64 percent when other schemes—including the Employee State Insurance Scheme (ESIS) and the On enrollment, beneficiaries are provided with Central Government Health Scheme (CGHS)—are a ‘smart card’. The card has a microprocessor included (compared to 29 percent nationally). chip displaying a photograph of the beneficiary Based on economic deciles estimated from family, a unique identification number that is linked household asset ownership, CMCHIS coverage to the family’s ration card, and—in some cases— was generally higher among lower economic 27 Some government facilities also participate in these camps, 29 Even if family members are not listed in the CMCHIS database, although it is not mandatory for them to do so. as long as other proof of family identification can be provided, 28 Karan, A., A. Chakraborti, H. Matela, et al. 2017. Process Evaluation they are eligible to receive treatment. Report: Chief Minister's Comprehensive Health Insurance Scheme, 30 This number is based on NFHS-4 which was conducted in 2015– Tamil Nadu. New Delhi: Public Health Foundation of India. 2016; estimates from the NSSO 71st round in 2014 are lower. 14  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu deciles (Figure 11).31 Less than 2 percent of relatively wealthy in India, so large-scale CMCHIS card-holders reported owning a car and mistargeting does not appear to be an issue. having an air conditioner, two markers of being Figure 11: Household survey data estimates of health insurance/scheme coverage by economic deciles 31 Economic deciles are derived based on ownership of assets including mattress, color television, Internet, computer, washing machine, car/tractor, and air conditioner. 4. Defining Benefits The CMCHIS benefits package emphasizes committee to determine procedure package rates. coverage for hospitalization for secondary It is usually benchmarked against market prices and tertiary care. The scheme’s benefits have and other government health schemes. evolved over time. Pre-2012, under Phase I of the scheme, 647 procedures under 14 specialties Most providers where beneficiaries can avail were covered and the beneficiary cap was benefits are in the private sector. Currently, 600 private facilities are empaneled under the `100,000 (~US$1,429) per family over the course scheme along with 225 government providers, of four years. From 2012–2016, under Phase II of and coverage is portable in that beneficiaries the scheme, the number of procedures covered can go to any empaneled government or private almost doubled: high-end procedures including provider within the state that offers a given implantation surgeries and organ transplantation procedure.32 An Empanelment and Disciplinary (such as renal, liver, and cochlear surgeries) were Committee (EDC) is responsible for empanelment included, as were a number of diagnostic and of providers; all government hospitals—including follow-up procedures. In addition, the beneficiary a few Community Health Centers (CHCs)—are cap was raised to `100,000 (~US$1,429) and up included in the scheme. The number of empaneled to `200,000 (~US$2,858) for certain procedures private facilities is capped with adjustments made per family per year. Restrictions were added such over time, allowing for volume gains to be realized that, to prevent abuse, some procedures such as for those empaneled; there is usually a waiting hearing aids and hip/knee replacements could only list of private facilities wanting to be empaneled be availed of in government hospitals or by using under the scheme. Government and empaneled government referrals. Post 2017, under Phase III of private facilities had a combined number of beds the scheme, additional procedures such as heart exceeding 120,000—2.6 beds per 1,000 enrolled and lung transplantations have been added and under CMCHIS—with almost two-thirds in the the number of diagnostic and follow-up procedures private sector. However, there are wide variations have been expanded (Table 4). The addition and across districts, both in terms of number of facilities, amendment of benefits over time appears to have been done based largely on demand and 32 Reliable information on the number of private facilities in the state is not available; estimates suggest about one-fifth of eligible supply considerations. CMCHIS has an in-house private providers are empaneled under the scheme. 16  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Table 4: Evolution of the CMCHIS benefits package over time Benefits Package Pre-2012 2012–2016a Post–2017b Secondary/tertiary care Total number of procedures 647 1,016 1,027 Procedures reserved for government facilities 56 86 Number of specialties 14 33 43 High-end tertiary 5 8 Follow-up Number of procedures 113 154 Diagnostic 23 38 Coverage cap per family `100,000 every `100,000 per year `100,000 per year 4 years a. High-end tertiary and follow-up procedures had a higher annual cap of `150,000. Notes:  Follow-up procedures had a higher annual cap of `200,000 per family per year; high-end tertiary procedures include renal, liver, b.  lung, heart, bone marrow/stem cell transplants, and cochlear and auditory brain stem implants and are financed on a case-by- case basis from a corpus fund once the beneficiary cap of `200,000 is exhausted; although the tender document states that 158 procedures are reserved for government facilities, in reality only 86 are reserved. beds, and the government-private split of providers Diagnostic procedures can also be availed where benefits can be availed. The districts of of at private outpatient facilities. The 38 Viluppuram, Tiruvannamalai, Ariyalur, and the diagnostic procedures—including metabolic Nilgiris had generally the lowest numbers of facilities screening, magnetic resonance imaging (MRI), and available for the scheme, less than 1 bed per 1,000 mammograms—can also be availed at 238 private enrolled population. Ariyalur, Nagapattinam, diagnostic facilities but only if these procedures are Ramanathapuram, and the Nilgiris had some of referred by a government hospital (see Annex G the lowest penetration rates of empaneled private for distribution of private diagnostic facilities by facilities in the state (see Annex F). district); these diagnostic procedures are covered even if they do not lead to hospitalization, and Given the relatively large scope of benefits in these are separate from diagnostic procedures the CMCHIS package, not all procedures and that are included as part of hospitalization.33 specialties can be availed of at all government Follow-up coverage, subject to a higher annual and empaneled private facilities. Some cap of `200,000, is provided from the sixth day up empaneled private providers are single-specialty to a post-discharge period of one year—including facilities, for example, only for ophthalmology, for consultations, follow-up investigations, and orthopedics, genitourinary surgery, oncology, and medicines—for 154 procedures. Pre-existing nephrology, among others. Among multispecialty conditions are covered, and coverage begins on providers, 183 government and private facilities the day of premium payment. provide a maximum of packages across 10 specialties and 8 high-end procedures (see About 3 percent of all enrolled families utilized Annex G). It is notable that several big corporate CMCHIS benefits in 2017. In absolute numbers, hospital chains—for example, Apollo, Fortis, Madras Institute of Orthopedics and Traumatology 33 In 2017, over three-fourths of the 160,941 diagnostics claims (MIOT), and Manipal—are also empaneled under did not lead to hospitalization; the rates of conversion to hospitalization were slightly higher at government than in private the scheme. facilities. Chapter 4: Defining Benefits  17 Figure 12: Utilization-related trends, 2012–2017 Source: CMCHIS this represented utilization of benefits by a total utilization rate under CMCHIS is comparable of 403,890 families (a little under 1 percent of the in magnitude to other government-sponsored total enrolled population) and overall processing schemes across India.35 of 641,000 unique claims. This was more than three times the volume of utilization and claims in District-level variations in CMCHIS utilization 2012 (Figure 12). On average, claim volumes have rates were notable, including in terms of the increased steadily by about 13 percent every year government-private split and the share of following an initial big jump of 77 percent over families utilizing out-of-district services. Districts 2012–2013; some of the increase in the utilization such as Kanyakumari and Chennai had some of rate in 2017 is due to expansion of the benefits the highest utilization rates under the scheme: package described earlier. Despite the relatively more than 4.5 percent of enrolled families used it large number of empaneled private facilities and in 2017; in the Nilgiris, on the other hand, less than beds under the scheme, more than half of all 35 Utilization rates under selected government-sponsored health utilization occurred in government facilities (partly insurance schemes in India: 2 percent under Karnataka’s also because some packages are reserved for Yeshavini Cooperative Farmers Health Scheme and <1 percent under Vajpayee Arogyashri Scheme; 2.5 percent under Rashtriya government facilities).34 In general, the claims Swasthya Bima Yojana (RSBY); <1 percent under Andhra Pradesh’s Rajiv Aarogyasri Community Health Insurance 34 As expected, given the relatively low rates of utilization and program; and <1 percent under Himachal Pradesh’s RSBY Plus. the nature of the benefits package, about 13.1 million (slightly Utilization rates were calculated using data from:La Forgia, G., less than 85 percent of all enrolled families) have not yet used and S. Nagpal. 2012. Government-Sponsored Health Insurance CMCHIS benefits. in India: Are You Covered? Washington, DC: World Bank. 18  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Figure 13: Claim volume shares in government versus private facilities, 2017 Source: CMCHIS 2 percent used the scheme. Chennai had the highest and medical oncology. Other specialties such as utilization rate in government hospitals, whereas orthopedic trauma, genitourinary surgery, general Theni’s private utilization rates were the highest in surgery, general medicine, and neonatology the state (see Annex H). About 40 percent of families each accounted for less than 3–5 percent of used the scheme in districts other than the ones they the total claim volume in 2017, the first full year were officially residing in. As expected, some of the when the latest iteration of the benefits package districts with the lowest out-of-district utilization was implemented.36 Annex I summarizes family rates were those that had relatively good supply utilization rates by specialty across districts. One of services: Chennai, Coimbatore, Kanyakumari, to of the biggest differences between the share of name a few. In districts with relatively poorer supply claims volume between the government versus of hospitals such as Tiruvallur and Ariyalur, more empaneled private providers was in nephrology than three-quarters of utilization occurred outside (which includes dialysis, with each visit counting of the district (see Annex H). as a different claim). It was the largest share More than half the number of claims volume 36 Note: this refers to the volume of claims and not the value of under CMCHIS is from diagnostics, nephrology, claims; the latter is discussed in a subsequent subsection. Chapter 4: Defining Benefits  19 of the volume of claims for private providers, radiation oncology, and interventional cardiology followed by diagnostics. Among public providers, (Figure 13). Claims utilization varied by age and the largest volume of claims was diagnostics, sex of beneficiary. Claims peaked among those followed by medical oncology. Neonatology; ear, who were 51–60 years old for men and 41–50 nose, throat (ENT); general medicine; cardiology; years for women. Surprisingly, across all age and hepatology appear among the top 10 in categories, claims volumes were much lower for government facilities. The top 10 claims volume female beneficiaries than for male beneficiaries among private providers included genitourinary (see Annex J).37 surgery, cardiothoracic surgery, ophthalmology, 37 There are a range of factors—for example, health-seeking behavior, benefit design, disease pattern— that might explain higher utilization and spending among men. In developed countries, on average, women account for higher health spending than men; while overall health spending is skewed toward women, men tend to account for higher proportion of spending on hospital care. OECD (Organisation for Economic Co-operation and Development). 2016. Focus on Health Spending: Expenditure by Disease and Gender. Paris: OECD. 20  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu 5. Managing Money CMCHIS is financed entirely from own-state incentive to enroll as many families as possible general government revenues. The state while reducing unnecessary utilization of benefits. government pays a ‘premium’ on behalf of enrolled However, profit and loss stipulations are tightly families to the insurer out of general government regulated. Premium payments are front-loaded revenues (that is, there was no earmarked tax for with a stop-loss provision. It was agreed that financing the scheme): this per-family premium 95 percent of the premiums would be advanced rate was `469 (~US$6.70) pre-2012, `497 at the beginning of the year with the remaining (~US$7.10) over 2012–2016, and `699 (~US$10) 5 percent paid at the end of the year, removing a thereafter.38 Although the benefits packages are key source of uncertainty for the insurer.40 If the not strictly comparable, government-subsidized claims ratio—that is, the ratio of claim payments premium payments for the poor and near-poor to total premiums received—goes above families in other countries running similar schemes 110 percent, 50 percent of the loss amount above were notably higher—ranging from US$22 in 110 percent is shared equally between the insurer the Philippines, US$62 in Vietnam, to US$77 and the scheme-administering agency, that is, in Indonesia; in each of these countries, most there is sharing of financial risk of larger losses inpatient care is included in the benefits package to accord some degree of financial protection as is outpatient primary care. for the insurer. On the flip side, if the claims ratio falls below 90 percent, the insurance company is The contract period with the insurer has obligated to return a proportion of this amount to typically been four years, with the possibility the scheme-administering agency. of a one-year extension.39 As such, with such an arrangement, the insurer has a financial Preauthorization and final claims settlement are conducted by three TPAs contracted by 38 Assuming an average family size of 2.7, this implies an implicit premium of `258.89 (~US$3.70) per capita; in addition to the 40 Section 64VB of the Insurance Act 1938 requires advance payment per-family premium payment, the state government also pays of premium to the insurer. However, in most government health applicable taxes to the insurer which has not been factored into insurance schemes, premiums are paid in quarterly installments. the calculations. Often, a situation arose where the claims exceeded the quarterly 39 Most state schemes do not have long-term pricing contracts with premium installment creating a cash flow deficit with the insurer, insurers; most contracts are annual. La Forgia, G., and S. Nagpal. which delayed claims settlement. Tamil Nadu became the first 2012. Government-Sponsored Health Insurance in India: Are You state to recognize this problem and initiated advance payment of Covered? Washington, DC: World Bank. 95 percent premium. Chapter 5: Managing Money  21 Figure 14: TPA district assignments, 2012–2015 (left) versus current (right) Source: CMCHIS the insurer. Each of the three TPAs—MDIndia, insurance coordinator who serves as the first point Mediassist, and Vidal—has responsibility for of contact for beneficiaries when they seek care. designated districts within the state, and district TPA assignments are rotated over time (Figure 14 Providers are paid on a bundled case basis. shows the assignments for 2012–2015 versus Package reimbursement rates were fixed ex current). At present, Vidal has 14 districts covering ante by the scheme-administering agency; they 5.8 million families, Mediassist covers 10 districts were the same for both government and private with 4.5 million families, and MDIndia covers hospitals and were set based on previous rates, 8 districts with 4.1 million families. All claims consultation with experts as well as stakeholders, are subject to preauthorization and claims need and were supposedly reflective of adjustments for to be settled within seven days of their receipt; inflation.41 Analysis of morbidity and mortality data more than 95 percent of claims are indeed settled was also conducted to estimate the incidence rate within seven days of receipt under the scheme. of different conditions and to help arrive at tariffs. The insurer/TPAs employ a team of project Nevertheless, as with many other schemes in India officers, vigilance officers, and medical officers as well as globally, many private providers are of at the district level to oversee implementation. In 41 As discussed subsequently, payment rates for private facilities addition, each of the empaneled hospitals has an varied depending on hospital supply-side readiness. 22  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Table 5: Market price versus CMCHIS tariffs in selected private facilities for tracer interventions Package Private hospital A Private hospital B Private hospital C CMCHIS tariff Maintenance hemodialysis for 27,200 12,000 10,400 8,000 chronic renal failurea Surgical correction of long-bone 300,000 80,000 100,000 22,250 fracture Percutaneous transluminal 250,000 150,000 135,000 63,000 coronary angioplastyb Ureterorenoscopic lithotripsy 80,000 70,000 35,000 22,000 Total knee replacement 200,000 160,000 90,000 65,000 Coronary bypass surgery 250,000 225,000 225,000 90,000C Note: a. Eight dialysis including seropositive; b. With stent; c. range 90,000-116,000. the view that the tariffs are low and do not cover diagnostics and follow-up procedures generally the full cost of providing care.42 As can be seen had the lowest tariffs. In general, over 90 percent in Table 5, CMCHIS tariffs for many packages of procedures had tariffs lower than `100,000 are significantly lower than the market prices for (~US$1,429). Tariffs were meant to cover (as the same packages in private facilities. The flip needed) bed charges in a general ward, nursing side of this argument is that the providers need and boarding charges, surgeons and anesthetists, to view this as reimbursements of marginal costs medical practitioners, consultant fees, anesthesia, and as a shift from a ‘low-volume high-margin’ to blood, oxygen, operating theater charges, cost of a ‘high-volume low-margin’ regime. To date, there surgical appliances, medicines and drugs, cost of have been no systematic assessments of private prosthetic devices, implants, x-rays and diagnostic providers to assess the impact of the scheme on tests, follow-up medicines, and food. Beneficiaries facility-level financial performance. The fact that a were paid a nominal amount—to cover public large majority of private providers have remained transport costs—when using care; this amount empaneled with the scheme—and there is a waiting was subsumed under the package rate and was list of others wanting to join—is indicative that given to beneficiaries by providers. Diagnostics up participation in the scheme has been beneficial, to one day before admission and five days post- either financially or otherwise.43 discharge were also covered under the scheme; diagnostic procedures require prescription The average tariff across all procedures was by a doctor at a government hospital and are about `30,000 (~US$429). Surgical procedures reimbursed separately even if they do not lead to had the highest average tariffs (~`35,000; admission. Multiple procedures during the same US$500), followed by medical procedures admission have tiered payments: 100 percent for (~`22,000; US$314) and diagnostics (~`4,000; the first diagnosis, 50 percent for the second, and US$57). Bariatric surgery and hematology 25 percent for the third. were the specialties with the highest tariffs; In 2017, aggregate CMCHIS outlays amounted 42 Karan et al. (2017). 43 In some cases, private providers appear to have appealed to to `8.6 billion (~US$122.4 million). These outlays additional philanthropic sources to augment revenues, using represented about 8 percent of the state’s overall empanelment with CMCHIS as a signal of corporate social responsibility. health spending, 0.4 percent of the state’s overall Chapter 5: Managing Money  23 Table 6: Key financing indicators for CMCHIS, 2009–2017a 2009–2011b 2012 2013 2014 2015 2016 2017c Premium per family (`) 469 497 497 497 497 497 699 Number of families (millions) 12.9 12.9 13.0 13.9 15.4 15.7 14.2 Amount transferred to insurer 11.7 6.4 6.4 6.6 7.6 7.8 8.6 (`, billions) Share of state health expenditure (%) 16 13 11 9 9 9 8 Share of state government 0.8 0.6 0.5 0.4 0.5 0.4 0.4 expenditure (%) Share of state GDP (%) 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Health share of state government 5.0 5.0 5.0 5.0 5.0 4.0 5.0 expenditure (%)  or the ratios, calculations are based on calendar year 2012 corresponding to fiscal year 2012–2013 and so on for each of the Note: a. F following years. This period is pre-CMCHIS, when the scheme was referred to as the Chief Minister Kalaignar’s Insurance Scheme for Life Saving b.  Treatments and contracted a private insurance provider for implementation.  s mentioned earlier, as of 2017, premiums have been paid only for the number of families that have also been Aadhar seeded: this c. A number totaled 13.4 million families (an estimated 36.2 million beneficiaries). expenditure, and 0.1 percent Figure 15: Health financing flows in Tamil Nadu45 of the GSDP (Table 6).44 CMCHIS resources remain a relatively small 2 percent Sources Schemes Providers share of the overall resources in Tamil Nadu’s health financing system as Central Central Ministry General Government of Health & Family Hospitals summarized in Figure 15. (5%) Welfare (5%) (34%) The average claim amount across all procedures was State State Ministry Specialist about `25,000 (~US$357). Government of Health & Family Hospitals The average value of a claim (16%) Welfare (16%) (34%) was higher in private versus government facilities. Across Private Firms CMCHIS Ambulatory Care all specialties, government (34%) (2%) (13%) facility claims averaged `11,490 (~US$164) in 2017 and the corresponding Pharmacies External Other Schemes number for private facilities (29%) was `16,340 (~US$233). In 44 The numbers do not exactly match the premium rate times the number Households Voluntary Health Other (21%) of families because of pro-rated and (39%) Insurance (1%) other minor adjustments. 45 The overall health financing picture for Tamil Nadu is similar to that of India OOP (68%) on average. 24  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu mirroring tariffs, claims amounts were highest for cardiothoracic surgeries that accounted for the surgical procedures (`37,502; US$536), followed largest share of claims among private providers. by medical procedures (`12,792; US$183) and Nephrology and interventional cardiology were diagnostics (`2,253; US$32). more prominent shares of claims among private providers as opposed to government facilities. Although diagnostics constituted the largest General surgery and diagnostics was a higher share of the volume of claims, cardiothoracic share among government providers versus surgeries claim the largest share in terms of the private ones. Time series analysis of claims over value of claims. Other specialties that accounted 2014–2016 indicate that—although there was an for more than 5 percent of the total claims value increase in claims value across all specialties, an in 2017 included orthopedic trauma, nephrology, increase annually of about 6 percent—greater than interventional cardiology, genitourinary surgery, 10 percent increases were evident for diagnostics and general surgery. As with claim volumes, and nephrology across both government and claim value differences are notable between private facilities; for cardiology, neurosurgery, government and private providers (Figure 16): and orthopedic trauma in government facilities; whereas orthopedic trauma was the largest value and for cardiothoracic surgery in private share of claims among government facilities, it was facilities (Table 7). Figure 16: Claim value shares in government versus private facilities, 2017 Source: CMCHIS Chapter 5: Managing Money  25 Table 7: Specialties with large changes in claim value, 2014–2016 Specialties with > 10% increase in claims value, 2014–2016 Government Private All Diagnostics Diagnostics Diagnostics Nephrology Nephrology Nephrology Cardiology Cardiothoracic surgery Cardiology Neurosurgery Cardiothoracic surgery Orthopedic trauma In some cases, reimbursements are lower than to-claims ratio—the difference between what the those paid for OOP by patients who are not hospital claimed versus what was reimbursed by covered by CMCHIS in private facilities. There the scheme (following validation)—was 66 percent are concerns that scheme reimbursements are in 2017–2018: 78 percent for private hospitals and low relative to the ‘market price’ for many of the only 51 percent for government facilities. There packages; crude analysis of household data does is some volatility in the month-to-month paid-to- suggest this to be so for some cases, although claims ratio, especially for government facilities this is to be expected given CMCHIS’ monopsony and for medical claims, less so for surgical claims. power. For example, using latest NSSO data, the The ratio seems to dip for the months of January average OOP payment for heart disease (surgical) and February—toward the end and beginning of was reported to be `132,125; the tariff for the same new year of the scheme, respectively—due to ranged from `4,100 to `200,000 with average reconciliation of claims (Figure 17).46 A relatively claims paid amounting to `86,611: 35 percent small number (4,304; less than 1 percent) of all lower (Table 8). A similar lower magnitude was claims were rejected in 2017: two-thirds of claim observed for accidental injury (surgical) compared rejections were from government facilities and the with average claims for orthopedic trauma. On the remainder from empaneled private facilities. other hand, for gastric and peptic ulcer (surgical), the reported OOP payment amounts at private With claims ratios in recent years ranging facilities were comparable to the average claims between 100 percent and 110 percent, the paid under CMCHIS for gastroenterology surgery. scheme is operating at the edge. Since 2012, the claims ratio under CMCHIS has never exceeded On average, two-thirds of what was claimed 110 percent; in 2017, the claims ratio was 102 was paid out, and this ratio was higher for percent, indicating a small loss for the insurance private versus government hospitals. The paid- company even before factoring in administrative Table 8: Comparing OOP payments in Tamil Nadu with claim reimbursements for selected interventions (`) Category for inpatient admission OOP at private CMCHIS package Average claim Heart disease (surgical) 132,125 Cardiothoracic surgery 86,611 Gastric and peptic ulcer (surgical) 36,707 Gastroenterology surgery 36,108 Accidental injury (surgical) 51,730 Orthopedic trauma 38,297 Source: NSSO 71st round. 46 The increase in March–April for medical claims for government facilities is due to payment of backed-up and pending reimbursements. 26  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Figure 17: Paid-versus-claims ratio, 2017–2018 costs. A slight increase in utilization rate could be catastrophic for the scheme: for example, at current average claim levels, an increase in family utilization—say from a current rate of 3 percent to 3.25 percent—would increase the claims ratio to 110 percent. CMCHIS is ‘cashless’ in that there are no beneficiary co-payments until the annual family cap is reached. Any amount under the annual cap that is not used by any family member by the end of the year cannot be carried over to the following year. Once the annual family cap is reached, the beneficiary must bear the full cost of coverage until the end of the year. Annually, a relatively small proportion— about 2–3 percent of those families that use the scheme—reached the cap of `100,000: for example, only almost 12,000 families reached this cap in 2017, 80 percent of which did so while availing benefits from private providers; only Source: CMCHIS 998 families exceeded the higher cap of `200,000 (~US$2,857). Although there is no widespread reporting of balanced billing by providers even when the beneficiary cap has not been reached, there are sporadic reports that this is occurring to some extent but not being captured through normal grievance mechanisms. CMCHIS was mandated to create a corpus fund to help needy beneficiaries. The corpus fund was used for beneficiary payments over the cap, for example, when coverage amounts exceeded the Chapter 5: Managing Money  27 beneficiary cap of `200,000 for selected high-end Under CMCHIS, the insurer reimburses up to procedures. Payments over this capped limit are `200,000 for these selected procedures; any made from a corpus fund on a case-by-case basis amount above this is paid for directly to the by the scheme-administering agency directly. All provider by the scheme-administering agency government hospitals are required to set aside 27 making CMCHIS a ‘mixed’ model in using both the percent of the value of claims for this corpus fund. insurance and assurance modality. The number of These funds cover top-up reimbursements for high-end claims handled through the assurance liver transplantation, renal transplantation, bone mode is low: only 1,213 high-end procedure claims marrow transplantation, cochlear implantation, were reimbursed in 2017 for a total amount of stem cell transplantation, heart transplantation, `0.3 trillion (~US$3.3 million), less than 3 percent lung transplantation, combined transplantation of the total amount reimbursed by the insurer in procedures, and auditory brainstem implantation. the same year (Figure 18). Figure 18: Volume and value of high-end procedures, 2012–2017 Source: CMCHIS 28  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu 6. Improving Supply Elements of CMCHIS have been consciously as part of implementation of UHC schemes. Some designed to instigate supply-side improvements. of these elements for improving supply are also Across countries, UHC schemes are often designed notable under CMCHIS. and implemented in ways to serve as instruments not only to improve access and financial CMCHIS, by empaneling private providers in protection but also to nudge reforms in health care addition to government hospitals, expands provision.47 For example, flexibility to retain and access of health services for the poor and use demand-side scheme reimbursement funds vulnerable. Almost all government-sponsored for salary incentives, infrastructure improvements, health insurance schemes in India have empaneled operating costs, and to bypass traditional public private providers, so this is not unique to CMCHIS. financial management rigidities in public systems Hospitals under the scheme are required to have a have been noted in many countries. In Thailand, minimum of 30 beds; in addition, they need to be additional funds from the UHC scheme have equipped with a microbiologically-safe operation been used to hire contractual staff in hospitals theater and have minimum staffing levels of to cope with additional demand without violating doctors, nurses, and other human resources to the government’s zero-growth policy for civil be eligible under the scheme.48 CMCHIS’ contract servants. In Argentina’s Plan Nacer Program that with the insurer requires the latter to make sure provided coverage for maternal and child services that a minimum number of private providers are to vulnerable groups, reimbursements to facilities empaneled. In the district of Chennai, the insurer from the insurer have been retained to provide staff is required to empanel a minimum of 50 private incentives. In addition, insurance schemes have hospitals, 30 each in districts of Coimbatore and actively been used to integrate private provision Madurai, and 10 each in other districts of the to help ease the burden on government providers. state. The EDC is charged with the process of Other supply-side improvements—for example, empanelment and de-empanelment for private through use of accreditation for empanelment and providers.49 Some facilities in neighboring states are requirements for tightening referral chains—have 48 In some cases, exemptions may be given, for example, for single- also been used as mechanisms to improve supply specialty hospitals and for those serving remote communities. 49 The EDC is composed of officials from the Tamil Nadu Health 47 Cotlear et al. 2015. Systems Project and United India Insurance Company Limited. Chapter 6: Improving Supply  29 also empaneled. For example, 13 private hospitals private facilities in the state are not empaneled in neighboring states and union territory—3 in under the scheme; they either do not meet the Karnataka, 4 in Kerala, 1 in Andhra Pradesh, and minimum standards, do not offer packages offered 12 in Puducherry—are also empaneled under the under CMCHIS, or have chosen not to empanel scheme as in some cases these may be closer to because the reimbursements are too low. About beneficiaries than in-state facilities. These out-of- 40 percent of empaneled private facilities reported state private hospitals have the same tariffs and having initiated operations less than 10 years other stipulations to provider coverage under the ago—about the time the scheme was launched in scheme. In addition, beneficiaries have access to 2009—and 15 percent began operations less than 238 in-state private diagnostic outpatient services, 5 years ago (Figure 20). Clearly, a lot of new private although the distribution of these facilities varies hospitals have entered Tamil Nadu’s health sector; widely across the state; more than 15 such facilities however, without additional follow-up research, it are empaneled in districts such as Chennai, is not possible to assess to what extent market Madurai, and Kanyakumari but only one each was entry of new private providers was a direct result empaneled in the districts of Ariyalur, Perambalur, of CMCHIS. and Thiruvannamalai. Some facilities only provide coverage for a What sets CMCHIS apart from many other subset of benefits in the package, and the entire schemes is that private providers are mandated benefits package is not available at all facilities. to hold monthly outreach camps that provide Nevertheless, many of the same benefits can be free primary care services. Government facilities availed of at multiple providers —both government are also encouraged to hold these monthly camps and private—which introduces an element of but are not mandated to do so. These monthly choice for beneficiaries and of competition among camps also provide information and education providers, although the impact of the latter on about the scheme and help identify potential provider behavior and outcomes has not been candidates for hospitalization. Except for some assessed systematically. This multiplicity of relatively isolated instances and in a few districts, options is particularly evident in urban areas.51 these health camps do not seem to have been a vehicle for large-scale immediate demand creation CMCHIS proactively incentivizes quality or for supply-induced care. For example, less than improvements among private providers. All 1 percent of admissions in government hospitals private providers under the scheme are required and less than 2 percent of admissions in private to have ‘entry-level’ accreditation by the National hospitals in 2017 were a direct result of camp- Accreditation Board of Hospitals (NABH) within referred hospitalizations of CMCHIS beneficiaries. 12 months of empanelment.52 At present, 90 percent of all empaneled private facilities have Access to private facilities has increased 51 All districts have at least one hospital that offers the complete under CMCHIS. Despite some de-empanelment set of specialties covered under the scheme. However, the total number of such facilities varies across districts. The following over time, the total number of private facilities are districts with the largest number of hospitals with complete contracted by CMCHIS has steadily increased specialties: Tirunelveli (19), Chennai (15), Tiruchirappalli (11), and Coimbatore (11). These districts are also among the most over time, more than 200 additional facilities have populous in Tamil Nadu. signed-on for the scheme from the number in 52 The NABH has developed pre-accreditation entry-level certification standards as a stepping stone for enhancing 2012 (Figure 19).50 However, some of the smaller quality of patient care and safety; once entry-level certification is achieved, the health provider can then prepare and move to full accreditation status. Hospitals need to attain 167 domain 50 This amounts to more than 8,000 private beds added under the objectives (mostly process quality indicators) to get entry-level scheme since 2012. accreditation. 30  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Figure 19: Number of empaneled private facilities with graded from a high of ‘S1’ to CMCHIS, 2012–2018 a low of ‘S2’.53 A hospital with more than 100 beds is given 3 points compared to 1 point to hospitals with less than 30 beds. Hospitals meeting other quality standards are also given a higher score. Hospitals meeting the Indian Public Health Standards (IPHS) of doctor-to-bed ratio of 1:6 and nurse-to-bed ratio of 1:2 are given 3 points. The same is the rating for hospitals with quality and professional certifications. Likewise, a hospital laboratory service with External Quality Assessment Source: CMCHIS Scheme (EQAS) certification, a microbiology lab headed by Figure 20: Distribution of years of operation for a microbiologist with master’s empaneled private hospitals degree, and a pharmacy with a licensed pharmacist all have correspondingly higher points. A hospital can get a maximum of 100 points, and this assessment is conducted annually.54 Higher-tiered private hospitals are entitled to higher reimbursement rates for some procedures. Table 9 shows reimbursement differences to private hospitals by grade for selected conditions. As can be seen, tariffs for some Source: CMCHIS procedures—for example, dengue hemorrhagic fever, this accreditation. In addition, payment rates for private providers are tiered based on availability 53 From 2012–2016, CMCHIS used a different set of criteria as basis of grade level. The criteria focused mainly on the availability of of infrastructure and expertise: multispecialty limited set of hospital equipment. In 2017, CMCHIS adopted new private hospitals are graded from a high of ‘A1’ to criteria with a more extensive list of hospital equipment including various quality assurance and professional certification. a low of ‘A6’; single-specialty private hospitals are 54 Additional details are available at the CMCHIS website. Chapter 6: Improving Supply  31 diabetic ketoacidosis, Figure 21: Distribution of private providers by grade, 2018 and acute severe asthma with ventilation—are almost half for ‘A6’ facilities relative to ‘A1’ facilities. In other cases, 11% the differences are not 19% 19% that large. For some 5% 29% procedures, such as total knee replacement, 8% there are no differences 23% 26% in reimbursements 15% by grade. In 2018, 19 16% percent of all private 16% 13% providers were graded ‘A1’ and 29 percent of private beds were graded ‘A1’ (the facilities graded ‘A1’ also tended to have more beds). Source: CMCHIS Although 84 percent of all private facilities Government hospitals are automatically that have been in the scheme for multiple years graded in the highest category but are required did not change grade over 2012–2016, 12 percent to use a portion of reimbursements for quality moved down at least one grade in the past five improvements. NABH accreditation is not years (only the remaining 4 percent moved up at required for government hospitals, although the least one grade) (Figure 22). This needs further process of introducing this is under way, and they study to better understand the dynamics behind are currently automatically accorded a grade of this: factors other than reimbursement levels could ‘A1’. However, unlike private facilities, government equally well explain the observed intertemporal hospitals are required to set aside 17 percent of grade transitions. all CMCHIS reimbursements for ‘institutional Table 9: Variations in CMCHIS tariffs by hospital grade for selected procedures Procedure A1 A2 A3 A4 A5 A6 S1 S2 Dengue hemorrhagic fever 18,000 16,200 14,580 13,122 11,810 10,629 18,000 13,122 Diabetic ketoacidosis 30,000 27,000 24,300 21,870 19,683 17,715 30,000 21,870 Acute severe asthma with 51,200 46,080 41,472 37,325 33,592 30,233 51,200 37,325 ventilation Coronary bypass surgery 100,000 100,000 90,000 90,000 90,000 90,000 100,000 90,000 Total knee replacement 65,000 65,000 65,000 65,000 65,000 65,000 65,000 65,000 Hearing aid 8,000 8,000 8,000 8,000 8,000 8,000 8,000 8,000 Total hip replacement (cemented) 66,750 60,075 60,075 60,075 60,075 60,075 66,750 60,075 Source: CMCHIS 32  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Figure 22: Grade transitions for private providers, 2012–2016 Source: CMCHIS development/hospital upgradation’ (in addition, 40 CMCHIS stipulates that continued provider percent can be used for consumables; 15 percent empanelment is dependent on quality metrics. as incentives for operating teams; 27 percent for There is a contractual stipulation that states that a corpus fund to finance high-end procedures; quality parameters such as hospital mortality and 1 percent for information, education, and rates and infection rates may be considered communication activities).55 In 2017, on average, as parameters for continuation of providers in total reimbursements per bed in government the scheme. As mentioned above, at present, hospitals were `21,000 (~US$290). However, as government hospitals are not required to undergo expected, it appears that CMCHIS reimbursements NABH accreditation but are expected to undergo were a more significant source of financing in a ‘facility assessment’ and follow National Quality smaller government hospitals compared to bigger Assurance Standards (NQAS). To date, almost government hospitals (Figure 23). For example, 300 private providers have been suspended under hospitals in the bottom quintile in terms of bed the scheme. None of the government facilities size (number of beds between 11 and 56) received have been de-empaneled. De-empanelment reimbursements per bed of `38,000 (~US$510), has occurred for a variety of reasons, the most much higher than the top quintile of hospitals that prominent being low performance especially on had beds above 200 (`15,400; ~US$210). the requirement that facilities conduct monthly outreach camps. A few private providers were 55 Government Order No. 331 of 2014. also de-empaneled because they did not receive Chapter 6: Improving Supply  33 NABH accreditation within the Figure 23: CMCHIS reimbursement per bed in government requisite 12-month period. A facilities few private facilities voluntarily de-empaneled themselves from the scheme. 56 CMCHIS adopts a simplified protocol for medical and surgical packages to ensure health providers do not overprovide or underprovide care. Although these protocols are not very detailed, they specify for each package, the bare minimum signs and symptoms that indicate need for admission, the minimum and types of tests and treatment that should be rendered (including, in some cases, the minimum number Source: CMCHIS of days of admissions). of every month, is composed of consultants from Compliance is monitored and enforced through various medical colleges and hospitals in the the preauthorization approval and claims payment state. The review process of the committee is process, and noncompliance may lead to reduction standardized; guidelines are in place to ensure of claim payment. To preauthorize coronary bypass the investigations are conducted in a transparent surgery, the patient needs to have symptoms fashion. If the committee finds the hospital to have of angina and hospitals are required to perform ‘unacceptable’ mortality or morbidity outcomes, an angiogram. If the surgery is conducted, the they can be given warnings, show-cause notices, full settlement amount is paid. If the surgery is or even suspension. Hospitals, however, can abandoned (that is, initiated but not completed), appeal against the decision of the committee.57 only 25 percent of the tariff is settled. Empaneled hospitals are also required to A special committee regularly monitors quality submit information on notifiable diseases. As of care. The Mortality and Morbidity Committee part of the state’s effort to strengthen an integrated (MMC) was established under CMCHIS to disease surveillance system, CMCHIS mandates investigate all deaths as well as a sample of all government and empaneled private hospitals morbidity cases. The goal of the committee is to to report on 24 notifiable infectious diseases that ensure quality standards are being met, deaths are may pose serious public health concern. These reported, and morbidity cases do not occur due are mostly outbreaks of infectious diseases (for to unsafe clinical practice and negligence. The example, AH1N1) and re-emergence of infectious committee, which convenes on the last Wednesday diseases (for example, malaria and tuberculosis). 56 About 6-7 private facilities have voluntarily de-empaneled to date; the primary reason for voluntary de-empanelment appears to have 57 In 2017, out of 108 cases submitted for possible investigation, been the relatively low levels of reimbursements under CMCHIS. 35 were acted upon by the MMC. 34  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu 7. Strengthening Accountability Changes introduced through CMCHIS include scheme. As noted earlier, part of the financial a fundamental alteration in the roles and risk for the scheme is borne by the insurer. The responsibilities of key stakeholders. As in other contract between the scheme-administering states and countries, expansion of government- agency and the insurance company specifies the financed coverage in Tamil Nadu was not simply overarching financing envelope for the scheme a case of doing ‘more of the same’; CMCHIS for four years and benefits to be provided. The introduced different roles and responsibilities for contract also includes actions and metrics by the Department of Health, insurance providers, which both parties must adhere and be held TPAs, and health care providers. Instead of accountable. The insurance company is chosen purchasing services from government hospitals through a bidding process and, under current through traditional line-item financing without guidelines, only public sector insurance companies a clear separation between the payer (the state can bid for the scheme. Currently, the insurance government) and providers, CMCHIS introduced company only signs service agreement contracts several arm’s-length separations with explicit with private providers, and there are plans to do responsibilities. Public financing is now being the same with government providers. In addition, used to purchase some health services from both the insurer separately contracts three private TPAs government and empaneled private providers on that manage the nuts and bolts of implementation behalf of beneficiaries using insurance and TPAs from enrollment to claims processing to day-to- through case-based bundled payments. The day management of the scheme. Each district in decision to use an insurance company appears the state is handled by one TPA only, and TPA to have been made to protect the integrity of district assignments are rotated over time, and this the process from political and other forms of appears to have fostered a healthy competitive- interference, something that would likely have been collaborative relationship between them. far more difficult to realize had Tamil Nadu chosen to use a trust model with the state government Guidelines and memorandums of directly administering the scheme. understanding (MOUs) are also used to specify accountabilities. For example, Legal contracts specify accountability provision of services in government hospitals arrangements among the larger actors in the is outlined in MOUs with the insurer. The Chapter 7: Strengthening Accountability  35 VAO or revenue authorities are responsible for their claims to be processed.58 In addition, a for verifying the income-related eligibility of sample of claims is subject to a comprehensive beneficiaries. Although the Aadhar card is not post-claim audit. CMCHIS employs a team of used for verification as such, the system links the doctors and retired police officers (also known beneficiary’s Aadhar number if this is available as district vigilance officers) to handle fraud and, as of 2017, premium payments have been management at the district level. When fraud is tied to Aadhar-linked families. suspected, the EDC directs retired officers to conduct the initial investigation. The insurer can de-empanel private providers resulting from poor performance or if evidence CMCHIS uses several mechanisms for routine of fraud is found. Fraud by providers is not clearly monitoring of the scheme. The scheme- defined but is specified as “…where any fraudulent administering agency (TNHSP) holds a weekly claim, negligence, not rendering cashless open-door meeting with stakeholders where treatment, not following the norms and guidelines anyone with a grievance or query can walk in or related to implementation of scheme including poor connect through Skype. In addition, as mentioned performance, etc. becomes directly attributable to above, the scheme has a toll-free number, the hospital.” As noted earlier, almost 300 private website, district review meetings, and regional hospitals have been suspended by the scheme review meetings for logging complaints and to date. All claims are subject to preauthorization grievances. Complaints from beneficiaries can to prevent fraud. Once admitted, the hospital lead to de-empanelment from the scheme. The submits documents for preauthorization to the call center logs about 500 calls per day; analysis TPA; preauthorization is typically completed by reveals that calls registering complaints such as the TPA within 24 hours. Once the beneficiary has unsatisfactory treatment or denial of benefits are been discharged, the provider submits a claim to relatively rare.59 There is regular monitoring of the insurer; claim settlement needs to occur within facilities, including surprise visits. Once admitted, seven days following which the insurer makes an the liaison officer of the scheme is expected to electronic payment to the provider. Providers are visit the beneficiary patient at least once per required to submit paperwork, upload photographs day; in addition, the officer also monitors bed- and videos, and provide detailed information occupancy rates. 58 The IT system of CMCHIS has built-in triggers that flag potentially fraudulent claims. The following are some examples of triggers used: claims coming from family with the same illness, ‘anatomically-impossible claims’ (for example, claims from an adult submitting pediatric claims), sudden surge of claims in a particular district, and so on. 59 In 2017, less than 1 percent of calls were registered as complaints. Emergency intimation and queries on diagnostics, enrollment, benefits, and about the scheme in general were the top calls registered. 36  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu 8. Transition to PM-JAY India is implementing several major reforms for on—has been the expansion of noncontributory progressively realizing UHC for its 1.3 billion coverage for the poor using general government population. Ayushman Bharat or ‘Long Live India’— revenues, either through bolstering traditional the umbrella term for health sector reforms in the supply-side financing for public facilities or country—comprises two primary components: through insurance-style demand-side financing (a) creation of ‘health and wellness centers’ that will that provides access to both government and provide diagnostic tests, free essential medicines, empaneled private facilities. Similar to India, many and other comprehensive primary care services at countries—including China, the Philippines, and sub-health centers that cater to a catchment area Turkey—initially expanded demand-side coverage population of 5,000 in rural plains areas versus only for inpatient care, later expanding benefits 3,000 in hilly/desert/tribal areas and are generally to include outpatient primary and specialist the first point of contact for India’s public sector care services. Nevertheless, what sets India’s health system and (b) implementation of PM- reforms apart from reforms in other countries is JAY that was launched in September 2018 to the sheer scale of population coverage under a provide government-sponsored health insurance single program: with an estimated 500 million coverage for a package of inpatient secondary and beneficiaries, PM-JAY is often referred to as the tertiary care that can be availed of at government largest government-sponsored health insurance and empaneled private hospitals to 100 million scheme in the world. poor and near-poor families (an estimated total of 500 million individuals, roughly 40 percent Over the past decade, states such as Tamil of the country’s population) up to a maximum Nadu have amassed significant experience annual limit of `500,000 (~US$7,143) per family. with implementing government-sponsored In launching PM-JAY, India is following in the health insurance schemes similar in design footsteps of many other developing countries that to PM-JAY. Health is a state subject in India. have recently implemented and expanded pro- PM-JAY—which almost all states have agreed to poor UHC reforms. A key common element across implement—is a centrally sponsored scheme that many of these UHC reforms—in China, Indonesia, is co-financed by the center and the states for a Mexico, the Philippines, Thailand, Vietnam, and so minimum standard benefits package of secondary Chapter 8: Transition to PM-JAY  37 Figure 24: Overlap of benefits packages: PM-JAY versus CMCHIS and tertiary health care services.60 However, states 2008, several additional states have implemented can expand own-financed coverage—in terms of the central government-sponsored RSBY that who they cover, what they cover, and how much provided inpatient secondary care coverage they cover—beyond what is stipulated under PM- to Below Poverty Line (BPL) families up to a JAY should they choose to do so. States can also maximum annual amount of `30,000 (~US$429) choose different implementation modalities: they per year. Subsequently, many states have financed can contract private or public sector insurance additional coverage and benefits beyond what firms to implement the scheme (the so-called was stipulated under RSBY or implemented their ‘insurance’ mode), use a state department or own schemes in place of RSBY. As a result, many state implementation agency (‘assurance’ mode), states have amassed knowledge and experience or use a mix (‘mixed’ mode). In its design, PM- in designing and implementing a variety of JAY builds upon several central and state-level government-sponsored health insurance schemes: government-sponsored health insurance schemes experiences which can and should be tapped to that have been implemented in the country over provide lessons for maximizing the effectiveness the past decade. For example, in 2007, one of the of PM-JAY. first-ever government-sponsored health insurance schemes in India was launched in the state of As with other government-sponsored health Andhra Pradesh (the Rajiv Aarogyasri scheme) that insurance schemes in India, CMCHIS is merging provided inpatient tertiary care coverage to over with PM-JAY. There is significant overlap between 20 million poor and near-poor families.61 Since the benefits packages of CMCHIS and PM-JAY. In many ways, this makes the transition to PM-JAY 60 The center-state co-financing split is 60:40 except for eight north- eastern and three Himalayan states where it is 90:10; for union easier for a state such as Tamil Nadu as compared territories with legislatures (Delhi and Pondicherry) co-financing to introduction of the scheme in ‘greenfield’ states is 60:40 and for other union territories it is entirely financed from the center. such as Uttar Pradesh, Madhya Pradesh, and 61 Nagpal, S. 2013. “Expanding Health Coverage for Vulnerable Bihar which do not have existing government- Groups in India.” Universal Health Coverage Studies Series No. 13, Washington, DC: World Bank. sponsored health insurance schemes. There are a 38  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu total of 1,342 secondary and tertiary care packages Tariffs for overlapping packages are, for the under PM-JAY, including one that is classified most part, similar across CMCHIS and PM-JAY. as ‘unspecified’ which can be reimbursed at a There is no systematic tendency for rates to be recommended tariff of `100,000 (~US$1,429): of higher for CMCHIS (which are range depending these, roughly 72 percent are already covered on supply characteristics for private facilities) under CMCHIS; this leaves roughly 28 percent of versus PM-JAY. In some cases—laparoscopic the total number of packages (424) that are covered cholecystectomy, hernia (epigastric), and temporal under PM-JAY but are reserved for government bone excision/resection—CMCHIS tariffs are higher facilities. In addition, CMCHIS covers additional than those proposed under PM-JAY; in others, the diagnostics and high-end packages that are not reimbursements for CMCHIS are at the lower end covered by PM-JAY (Figure 24). from those proposed under PM-JAY (Figure 25). Figure 25: Tariff comparisons, CMCHIS versus PM-JAY Source: CMCHIS Chapter 8: Transition to PM-JAY  39 Table 10 summarizes some of the ways in which stipulations, the benefits package and tariffs are CMCHIS will transition to PM-JAY, under the as per the national package list are applicable. The same broad themes of covering people, defining payment of claims to hospitals is to be made by the benefits, managing money, improving supply, and trust or insurance company that is implementing strengthening accountability. PM-JAY in the home state to which the beneficiary belongs. Processing of claims and payment PM-JAY allows for national portability, that is, to hospitals need to be made within 30 days of beneficiaries from any part of the country are receiving the claim. In the case of any disputes, entitled to seek care at empaneled hospitals standard grievance redressal processes are to be anywhere in the country. Under portability followed. Table 10: CMCHIS-PM-JAY transition CMCHIS PM-JAY guidelines PM-JAY-CMCHIS Covering Primary target: families with Primary target: Poor families identified Primary target: CMCHIS people annual income less than in 2011 Socio-Economic Caste enrollees plus any `72,000. Census (SECC) based on deprivation additional families criteria in rural areas and occupational identified under SECC categories in urban areas. not currently enrolled in CMCHIS. Number of families: Number of families: 7,770,986. Number of 15,724,432. families:15,724,432. Enrollment required. Smart Entitlement based on validation, no Enrollment required. card issued on enrollment. enrollment recommended. Smart card issued on enrollment; for all families an additional e-card may be provided on validation. Defining 1,027 packages, no 1350+ packages, including one 1,451 packages, benefits ‘unspecified’ package. ‘unspecified’ package. States can including one add additional packages. ‘unspecified’ package: including 533 common packages, 494 packages exclusive to CMCHIS, and 424 packages covered under PM-JAY but not under CMCHIS for all. 38 stand-alone diagnostics No stand-alone diagnostics packages CMCHIS stand-alone packages. included in minimum. diagnostics packages will continue for all. 154 follow-up packages. No specific follow-up packages CMCHIS follow-up included in minimum. packages will continue. 8 high-end packages. No high-end packages included in CMCHIS high-end minimum. packages will continue. 40  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu CMCHIS PM-JAY guidelines PM-JAY-CMCHIS Managing Premium: `699, own-state Premium varies by state; 60:40 Premium: `699, 60:40 money financed. co-financed between center and center-state co-financed state except for northeastern for 7,770,986 families states, Himalayan states, and and own-state financed Union Territories where the center for 7,953,446. co-finances 90–100% of the premium.62 Annual family cap: `100,000 Annual family cap: `500,000/family. Annual family cap: per family; cap of `200,000 `500,000/family for all. per family for 154 follow- up packages; corpus- funded reimbursements for >`200,000 for 8 high-end packages. ‘Mixed’: Public sector Public or private insurance mode, or CMCHIS ‘mixed’ mode insurance company with assurance mode, or ‘mixed’, with or will continue. multiple TPA support, without TPAs. assurance mode for reimbursements above `200,000 for 8 high-end packages. Four-year contract, with Recommended minimum of three CMCHIS practices will one-year extension. years contract. continue. Provider payment: Provider payment: Case based, tariffs CMCHIS tariffs for Casebased, tariffs set recommended by the center. overlapping benefits; by state government; PM-JAY recommended 27% of reimbursements tariffs for non-CMCHIS to government hospitals packages. mandated for corpus fund. Improving Network of in-state National portability for empaneled 424 PM-JAY packages supply government facilities and government and private facilities. to be provided only by empaneled private facilities, empaneled government some empaneled facilities facilities for now. in neighboring states. Private providers No such recommended mandate. CMCHIS practices will mandate to conduct continue. monthly outreach health camps for provision of free primary care; for information, education, and communication activities. 62 In 2018, the central government imposed a 4 percent ‘cess’ on payable income tax (that is, on total tax liability). This 4 percent ‘cess’ is made up of a 2 percent education ‘cess’, 1 percent is for a senior secondary education ‘cess’ (a total 3 percent education ‘cess’), and a new 1 percent for a health ‘cess’. For formal salaried employees, the tax liability and ‘cess’ are deducted automatically from payroll. For non- formal workers (non-salaried) workers, the tax liability and ‘cess’ are reflected in annual income tax returns. The government is anticipating raising US$1.5 billion ((`107 billion), which will be used to finance PM-JAY. Chapter 8: Transition to PM-JAY  41 CMCHIS PM-JAY guidelines PM-JAY-CMCHIS Six-tiered reimbursements 10–15% over and above the CMCHIS reimbursements for multi specialty private recommended PM-JAY package will continue. providers depending on tariff depending on the level NABH infrastructure and staffing; accreditation, whether serving two-tiered reimbursements ‘aspirational’ districts, or whether for single-specialty private offering postgraduate courses. facilities. Total number of empaneled No specific guidelines for total CMCHIS practices will private hospitals is number of private hospital enrollment. continue. controlled. Regular review of mortality No specific recommendations. CMCHIS practices will and morbidity outcomes. continue. For government hospital No specific recommendations. CMCHIS practices will reimbursements: 17% for continue. infrastructure/upgradation, 15% for staff incentives. Strengthening Legal contracts, Legal contracts, guidelines, and CMCHIS practices will accountability guidelines, and MOUs MOUs with different actors with continue. with different actors with explicit accountability arrangements. explicit accountability arrangements. Strong anti-fraud Recommendations outlined in anti- CMCHIS practices will orientation through 100% fraud guidelines. continue. pre-authorization, claims review, and sample post- claim audits. 42  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu 9. Summary and Pending Agenda Several innovative design elements in significantly helped reduce premium implementation of Tamil Nadu’s CMCHIS payments under CMCHIS. Advancing of are notable. As India embarks on nationwide payments reduced a significant source of financial protection reforms under PM-JAY, there uncertainty and could be encouraged when are important lessons to be learned from similar contracts are made with insurance agencies reforms implemented in states such as Tamil or even with providers. Nadu over the past decade or so. Some notable ™™ Use of multiple TPAs. Use of multiple aspects of the experience from CMCHIS include TPAs, including cross-mixing district the following: allocations over time, has led to a spirit of ™™ Inclusion of diagnostics. One of the most competitive collaboration for day-to-day important characteristic of CMCHIS that sets implementation of the scheme. TPAs can it apart from PM-JAY is on the diagnostic have the opportunity to learn from each benefits front, namely inclusion of a separate other and this can stimulate efficiency and diagnostics package—which can be availed encourage experimentation with alternatives of also at private primary care facilities on to implementation; competition among TPAs an outpatient basis—is an important step can act as a check, while enabling exploitation forward toward expanding the reach of of scale economies where possible. Shifting benefits beyond narrowly defined secondary district responsibilities every couple of years and tertiary inpatient care, something that removes incentives for entrenchment and PM-JAY may consider encouraging states to reduces the potential of fraud. incorporate. Although diagnostics accounted ™™ Incentivizing quality. CMCHIS proactively for more than one-fifth of the overall volume uses financial levers to incentivize quantity of claims under CMCHIS, they represented and quality of health care; although less than 5 percent of the total claim values. preliminary analysis suggests this is not ™™ Flexibility in advancing payments. having the intended effect on improving Longer-term contracts with and flexibility quality, the attempt to do so is what sets the in advancing payments to the insurer scheme apart from PM-JAY. The scheme’s Chapter 9: Summary and Pending Agenda  43 MMC is another mechanism by which ™™ Primary health care outreach camps. quality enhancements are implemented, with CMCHIS mandated monthly proactive feedback loops and regular assessments outreach camps by providers, both to raise of clinical quality of care. Setting aside 17 awareness among beneficiaries as well percent of reimbursements under the scheme as to serve as a first point of contact for for improving supply of government facilities routine primary health care and for helping is notable; these resources have been set beneficiaries identify needs and navigate the aside for infrastructure improvements and system if further interventions are needed; are separate from the resources for financing although in some ways this is not an ideal consumables.63 arrangement—some form of seamless ™™ Strong governance and fraud inclusion of primary health care and specialty management. Despite lack of an explicit outpatient services under the scheme would fraud management policy or guidelines, obviate the need for such camps—it can CMCHIS appears to have an extremely robust and does serve to improve access to health fraud prevention, detection, and deterrence services for vulnerable population groups. mechanism. There is strong leadership This element of CMCHIS is potentially and governance of the scheme, vigilant efficiency-enhancing, as it can help the monitoring systems in place, and a very strong system catch diseases early, before severe technological infrastructure implemented symptoms appear that might necessitate by a highly committed team comprising the immediate hospitalization. scheme-administering agency, the public In looking forward, as CMCHIS is implemented insurer, and the three TPAs. in conjunction with PM-JAY, it is an opportune ™™ Corpus funds for needy beneficiaries. time to flag some pending agenda issues and Beneficiary welfare improvements, over challenges for the state. Systemically, high and above what is provided by the scheme, levels of OOP spending remain in both Tamil Nadu have been implemented by helping provide and nationally and are unlikely to be influenced top-up financing for high-end procedures, significantly by PM-JAY: the scheme covers accessories, and follow-up procedures. interventions that are of fairly low frequency; over Although this has come at the expense of time, the scheme could consider adding more lowering payments to government facilities, benefits to the package; expansion of population the principle is laudable. coverage will help make a dent in moving the dependence of the health financing system on 63 In developed countries, more sensitive indicators of hospital quality (for example, readmission rates and infection rates) OOP spending.64 At present, CMCHIS amounts are routinely collected and used as basis for incentivizing to only 2 percent of all health resources in the performance. We attempted to estimate readmission rates of specific conditions: pneumonia, COPD, and hip and knee system. With a relatively small resource outlay, replacement. For pneumonia cases, 30 hospitals had their CMCHIS is operating at the edge: even small patient readmitted at the same facility or other facility within 90 days after discharge. Of the 30 hospitals, 24 were government increases in utilization will render the scheme to facilities. The share of readmitted cases to total pneumonia cases were <1 percent to 9 percent. For COPD cases, 16 hospitals had become financially unsustainable. Lack of systemic their patients readmitted at the same facility or other facility with integration of primary care with the scheme is a key 90 days after discharge. Of the 16 hospitals, 11 were government facilities. The shares also vary from <1 percent to 11 percent. bottleneck, especially critical given the continuum- No readmission case was noted for knee and hip replacement. These estimates, however, should be taken with a grain of salt. 64 This is even more critical as global experience has shown that Standard method of readmission rate estimation uses ICD10 to sometimes when similar schemes are introduced they lead to classify conditions and includes all hospitals cases. Here, only increase in both utilization and OOP payments, at least in the CMCHIS patients were used. beginning. 44  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu of-care needs of NCDs and of integrated care facilities that do not empanel, understanding why more generally.65 A strong primary care system they do not do so; assessing impact on beneficiary that works in conjunction with—not in parallel to— health outcomes and financial protection; examining CMCHIS will go a long way to ensure sustainability. and reducing hospitalizations for ambulatory- In addition, the two-tiered treatment of facilities sensitive conditions (such as diabetes and asthma, could be slowly removed. Government hospitals for example); emphasizing and incentivizing should be required to adhere to the same quality prevention over curative services; and tracking guidelines and metrics that private providers are referrals from lower levels of care, among others. required to adhere to. The scheme should also CMCHIS can serve as an important platform to move beyond focusing on traditional monitoring consider options by which some of the gains made and evaluation metrics to those that provide by Tamil Nadu’s health systems could be sustained additional information about systemic weaknesses and expanded as it faces new challenges and that could be addressed: tracking private opportunities to do so in coming years. 65 Approximately 13 percent of claims submitted to CMCHIS were from patients referred from lower-level facilities. There is no significant difference in the referral rates between government and private hospitals. Chapter 9: Summary and Pending Agenda  45 Annex A Key Population Health Outcomes Country/state Population TFR Life Adult Under-five Infant Maternal Stunting expectancy survival mortality mortality mortality (%) Andhra Pradesh 47.9 1.7 70 71 37 34 74 31 Bangladesh 164.9 2.1 72 76 34 28 176 36 Brazil 209.2 1.7 76 79 15 14 44 7 Chhattisgarh 28.9 2.5 65 49 39 173 38 China 1,397.0 1.6 76 86 10 9 27 8 Ghana 29.0 4.0 63 62 59 41 319 19 Goa 1.6 76 8 20 Gujarat 68.6 2.2 70 75 33 30 91 39 India 1,334.2 2.3 69 71 43 35 174 38 Indonesia 265.3 2.4 69 72 26 22 126 36 Karnataka 68.5 1.8 69 73 29 24 108 36 Kerala 36.3 1.8 75 82 11 10 46 20 Meghalaya 3.5 67 40 30 44 Maharashtra 127.2 1.8 72 78 21 19 61 34 Malaysia 32.5 2.0 75 81 8 7 40 21 Philippines 107.4 2.9 69 70 27 22 114 33 Punjab 31.0 1.7 73 76 24 21 122 26 Russia 144.0 1.8 72 70 8 7 25 13 South Africa 57.4 2.5 63 55 43 34 138 27 Sri Lanka 21.7 2.0 75 80 9 8 30 17 Thailand 69.2 1.5 75 79 12 11 20 11 Turkey 81.9 2.1 76 84 13 11 16 10 Tamil Nadu 79.3 1.6 71 75 19 17 66 27 Vietnam 94.6 2.0 76 81 22 17 54 25 West Bengal 102.4 1.6 71 77 27 25 101 33 Lower-middle- 2,969.5 3.2 68 70 41 31 194 27 income country Sources.Countries: Population (2018) using International Monetary Fund (IMF), total fertility rate (2016), life expectancy (2016), under-five mortality rate (2016), infant mortality rate (2016), adult survival rate (2016), and maternal mortality rate (2016) using World Development Indicators; Stunting rates: United Nations Children’s Fund (UNICEF)/WHO/World Bank joint malnutrition rates (latest available years), no data provided for Russia but estimate by Lunze et al. (2015) was 13 percent; States: Population (2018) using authors’ projection using five rounds of decadal census, TFR (2015), life expectancy (2014), under-five mortality rate (2016), infant mortality rate (2016), adult survival rate, and maternal mortality rate (2016) using Sample Registry System (SRS) infant mortality rate and under-five mortality rate for Meghalaya using NFHS 2015–2016; life expectancy for Meghalaya and Goa using Institute for Health Metrics and Evaluation (IHME) estimates. Sample Registry System (SRS) only provides life expectancy for big states; stunting rates using NFHS 2015–2016. 46  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Annex B Universal Health Coverage UHC tracer indicators (%) Country/state MFP ANC DPT SAN TOB CHI HIV TBT PPE CAT10 CAT25 Andhra Pradesh 93 76 89 54 80 91 38 92 22 71 83 Bangladesh 74 31 97 47 77 42 13 93 28 86 95 Brazil 88 90 96 86 86 50 57 71 72 74 97 Chhattisgarh 77 59 91 33 61 78 12 92 42 88 94 China 95 74 99 75 75 79 41 94 68 82 95 Ghana 43 87 88 14 96 56 28 85 64 97 100 Goa 56 89 94 78 90 — 46 93 — 84 91 Gujarat 78 71 73 64 75 79 29 93 47 83 93 India 72 45 87 44 88 77 44 72 35 83 96 Indonesia 81 84 78 68 61 75 10 85 50 96 100 Karnataka 83 70 78 58 77 92 48 84 48 74 86 Kerala 74 90 90 98 87 — 33 89 26 57 80 Meghalaya 45 50 74 60 53 76 19 82 — 92 98 Maharashtra 84 72 75 52 73 89 54 84 40 77 89 Malaysia 53 80 99 100 78 87 26 78 64 99 100 Philippines 54 84 60 75 75 64 27 91 46 99 99 Punjab 80 69 95 82 87 92 40 93 21 67 82 Russia 73 78 97 89 59 83 28 71 64 95 99 South Africa 84 87 75 73 79 65 49 81 92 99 100 Sri Lanka 74 93 99 94 86 58 23 85 62 97 100 Thailand 91 93 99 95 79 83 61 79 88 97 99 Turkey 60 89 97 96 72 85 28 87 83 97 100 Tamil Nadu 83 81 85 52 80 89 57 80 34 76 87 Vietnam 77 74 97 78 76 81 43 92 57 — — West Bengal 73 77 93 51 67 77 13 90 — 67 82 Lower-middle- 61 68 84 62 80 63 34 85 49 91 98 income country Sources. Countries: World Bank and WHO UHC Monitoring Report (2017); States: MFP using estimates of New et al. (2017); ANC, DPT, SAN, CHI using NFHS 2015–2016; TOB using GATS India Survey (2016); HIV estimates using data from National Aids Council (2018); TBT using data from India TB Report (2018). OOP using data from India’s Ministry of Health and Family Welfare CAT10 and CAT25 estimates using NSSO 71st round. Note. MFP: Demand satisfied for modern family planning; ANC: at least four prenatal care visits; DPT: DPT3 vaccination; SAN: basic sanitation; TOB: non-tobacco smoking; CHI: care-seeking behavior of children with pneumonia; HIV: people living with HIV/AIDS under anti- retro viral therapy; TBT: tuberculosis treatment success rate; PPE: prepaid/pooled expenditures; CAT10: households not spending 10 percent of total expenditures; CAT25: households not spending 25 percent of total expenditures. CAT10 and CAT25 for states and countries should compared with caution. In NSSO 71st round, the survey used monthly reported expenditures, which grossly underestimates the total household expenditures and the potential reason for higher catastrophic expenditure estimates. Annexes  47 Annex C Burden of Disease and Risk Factor Trends for Tamil Nadu and India Tamil Nadu Rank in DALYs lost share (%) Top 10 diseases/conditions in 2016 2016 1990 2000 2010 2016 1 Ischemic heart disease 6.6 8.8 12.2 14.3 2 Diabetes 1.4 2.4 3.9 4.9 3 Self-harm or violence 3.4 4.3 4.3 4.1 4 Chronic obstetric pulmonary disorder 2.5 3.3 3.4 3.7 5 Dietary iron deficiency 2.4 3.0 3.4 3.6 6 Diarrheal diseases 10.8 6.2 3.8 2.8 7 Falls 1.8 2.1 2.5 2.4 8 Migraine 1.2 1.7 2.1 2.4 9 Neonatal prenatal birth 6.8 4.8 3.3 2.2 10 Lower respiratory tract infection 6.9 5.0 3.1 2.1 DALYs (millions) 33.6 29.1 27.0 25.4 DALYs per 100,000 55.9 44.1 36.4 33.5 Source: Institute for Health Metrics and Evaluation (IHME) (2016) India Rank in DALYs lost share (%) Top 10 diseases/conditions in 2016 2016 1990 2000 2010 2016 1 Ischemic heart disease 3.7 4.8 7.0 8.7 2 Chronic obstetric pulmonary disorder 3.1 3.7 4.0 4.8 3 Diarrheal disease 12.4 9.2 6.2 4.1 4 Lower respiratory tract infection 9.8 7.9 6.1 4.4 5 Stroke 2.0 2.4 3.0 3.5 6 Dietary iron deficiency 2.1 2.7 3.2 3.5 7 Neonatal preterm birth 5.5 5.0 4.3 3.4 8 Tuberculosis 5.0 4.5 3.7 3.2 9 Sense organ 1.4 1.8 2.4 2.9 10 Road injuries 1.5 2.0 2.6 2.9  DALYs (millions) 537.4 527.3 495.1 466.3  DALYs per 100,000 62.2 50.6 40.4 35.4 Source: Institute for Health Metrics and Evaluation (IHME) (2016) 48  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Tamil Nadu Rank in Top 10 risk factors in 2016 DALYs lost share (%) 2016 1990 2000 2010 2016 1 Dietary risk 7.7 9.6 12.7 14.4 2 High fasting blood glucose 4.9 8.4 10.6 13.0 3 High systolic blood pressure 5.9 7.9 10.7 12.3 4 High BMI 1.5 3.5 6.3 8.1 5 Child and maternal nutrition 31.1 19.3 11.9 8.0 6 Air pollution 9.6 9.1 8.0 7.2 7 High total cholesterol 3.2 4.4 6.2 7.2 8 Tobacco 4.9 5.7 6.2 5.6 9 Impaired kidney function 2.7 3.5 4.1 4.6 10 Alcohol and drug use 1.8 2.8 3.5 3.8 DALYs (millions) 18.5 14.6 13.2 12.2 DALYs per 100,000 31.2 22.2 17.9 16.1 Source: Institute for Health Metrics and Evaluation (IHME) (2016) India Rank in Top 10 risk factors in 2016 DALYs lost share (%) 2016 1990 2000 2010 2016 1 Child and maternal nutrition 35.5 28.1 20.9 14.6 2 Air pollution 11.1 10.5 10.1 9.8 3 Dietary risk 4.5 5.4 7.4 8.9 4 High systolic blood pressure 3.9 5.0 7.0 8.5 5 High fasting plasma sugar 2.3 3.6 4.8 6.0 6 Tobacco 4.4 4.8 5.6 5.9 7 Unsafe water, sanitation, handwashing 12.5 9.5 6.3 4.7 8 High total cholesterol 1.7 2.3 3.3 4.1 9 High BMI 0.8 1.5 2.5 3.6 10 Alcohol and drug use 1.7 2.4 3.1 3.6 DALYs (millions) 292.1 271.8 244.6 219.0 DALYs per 100,000 33.8 25.9 20.0 16.6 Source: Institute for Health Metrics and Evaluation (IHME) (2016) Annexes  49 Annex D Key Health Financing Indicators Country/state Total health Total health Public health Public health OOP spending expenditure expenditure expenditure expenditure share of per capita share of GDP per capita share of GDP total health (US$) (%) (US$) (%) spending (%) Andhra Pradesh 58 4.3 9 0.7 78 Bangladesh 32 2.6 6 0.5 72 Brazil 780 8.9 338 3.9 28 Chhattisgarh 49 3.6 14 1.0 58 China 426 5.3 254 3.2 32 Ghana 80 5.9 41 3.1 36 Goa Gujarat 48 2.1 16 0.7 53 India 63 3.9 16 1.0 65 Indonesia 115 3.4 38 1.5 50 Karnataka 68 3.0 15 0.7 52 Kerala 106 4.5 19 0.8 74 Meghalaya Maharashtra 70 3.0 12 0.5 60 Malaysia 377 3.9 199 2.0 36 Philippines 127 4.4 41 1.5 54 Punjab 81 4.1 14 0.7 79 Russia 524 5.6 320 3.4 36 South Africa 471 8.2 252 4.4 7.7 Sri Lanka 118 3.0 64 1.6 38 Thailand 220 3.8 166 2.9 12 Turkey 455 4.1 355 3.2 17 Tamil Nadu 64 2.8 16 0.7 66 Vietnam 117 5.7 51 2.5 44 West Bengal Lower-middle- 132 5.7 71 3.0 40 income country Sources. Countries: Global Health Expenditure Database (2015). States: Ministry of Health and Family Welfare (2017) 50  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Annex E Tamil Nadu District-Level Variables: Population and Enrollment District Populationa Families enrolled Individuals enrolledb Enrollment ratec(%) Ariyalur 813,014 181,073 488,896 60.1 Chennai 5,404,072 663,083 1,790,323 33.1 Coimbatore 3,918,339 705,207 1,904,059 48.6 Cuddalore 2,873,026 606,491 1,637,526 57.0 Dharmapuri 1,720,502 395,922 1,068,989 62.1 Dindigul 2,357,113 474,778 1,281,901 54.4 Erode 2,503,447 624,535 1,686,245 67.4 Kancheepuram 4,787,249 794,061 2,143,964 44.8 Kanyakumari 2,038,957 446,722 1,206,149 59.2 Karur 1,159,651 298,239 805,245 69.4 Krishnagiri 2,153,888 447,147 1,207,296 56.1 Madurai 3,409,010 649,342 1,753,223 51.4 Nagapattinam 1,739,912 349,737 944,290 54.3 Namakkal 1,898,602 409,806 1,106,477 58.3 Perambalur 622,054 145,533 392,940 63.2 Pudukkottai 1,808,974 417,416 1,127,023 62.3 Ramanathapuram 1,467,461 301,208 813,262 55.4 Salem 3,889,354 829,456 2,239,531 57.6 Sivaganga 1,458,741 305,199 824,037 56.5 Thanjavur 2,619,248 555,318 1,499,359 57.2 The Nilgiris 801,674 166,300 449,011 56.0 Theni 1,364,486 298,489 805,920 59.1 Tiruchirappalli 3,016,328 572,328 1,545,286 51.2 Tirunelveli 3,350,490 605,186 1,634,002 48.8 Tiruppur 2,837,502 539,000 1,455,299 51.3 Tiruvallur 4,509,658 753,096 2,033,358 45.1 Tiruvannamalai 2,697,961 527,796 1,425,050 52.8 Tiruvarur 1,367,726 318,954 861,176 63.0 Tuticorin 1,867,315 311,614 841,358 45.1 Vellore 4,387,467 82 0,721 2,215,945 50.5 Viluppuram 3,820,189 811,590 2,191,293 57.4 Virudhunagar 2,145,860 399,088 1,077,539 50.2 Tamil Nadu 79,263,612 15,724,432 42,455,968 53.6 Source: Families enrolled from CMCHIS. Note: a. Estimated using decadal census b. Estimated, assuming average family size of 2.7. c. Estimated individuals covered divided by estimated population. Annexes  51 Annex F Tamil Nadu District-Level Variables: Facilities and Beds Government Private Beds per District Facilities Beds Facilities Beds Diagnostics 1,000 enrolled Ariyalur 2 222 0 — 1 0.5 Chennai 16 6,255 54 6,413 16 7.1 Coimbatore 11 2,309 67 8,398 12 5.6 Cuddalore 9 2,525 17 6,335 10 5.4 Dharmapuri 5 1,297 10 362 3 1.6 Dindigul 5 1,014 28 1,965 15 2.3 Erode 7 1,661 42 2,434 10 2.4 Kancheepuram 8 476 27 6,195 8 3.1 Kanyakumari 5 1,179 30 4,542 22 4.7 Karur 5 699 11 444 10 1.4 Krishnagiri 4 791 12 1,394 4 1.8 Madurai 7 1,807 53 5,408 21 4.1 Nagapattinam 10 1,260 3 93 3 1.4 Namakkal 6 1,020 17 836 6 1.7 Perambalur 1 105 6 626 2 1.9 Pudukkottai 8 875 7 579 1 1.3 Ramanathapuram 10 1,156 3 177 8 1.6 Salem 6 1,954 52 5,539 2 3.3 Sivaganga 7 1,260 10 603 8 2.3 Thanjavur 6 2,690 24 2,077 3 3.2 The Nilgiris 4 742 3 136 9 2.0 Theni 4 284 12 666 5 1.2 Tiruchirappalli 11 2,261 35 3,479 3 3.7 Tirunelveli 19 1,786 15 1,068 1 1.7 Tiruppur 4 969 23 1,119 3 1.4 Tiruvallur 8 510 16 1,442 10 1.0 Tiruvannamalai 5 286 5 395 6 0.5 Tiruvarur 4 1,099 5 268 14 1.6 Tuticorin 10 1,036 8 484 5 1.8 Vellore 10 2,292 18 1,443 4 1.7 Viluppuram 7 440 10 568 4 0.5 Virudhunagar 11 1,310 15 509 9 1.7 Tamil Nadu 235 43,570 638 65,997 238 2.6 Source: CMCHIS 52  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Annex G Facilities That Provide the Maximum of Packages Across 10 Specialties and 8 High-End Procedures District Type Name of Facility Chennai Government Government Royapettah Hospital Chennai Government Government Stanley Medical College Hospital Chennai Government Government Kilpauk Medical College Hospital Chennai Government Government Kasturba Gandhi Hospital Chennai Government Madras Medical College Coimbatore Government Coimbatore Government Medical College Kanchipuram Government Chengalpattu Medical College Kanyakumari Government Kanyakumari Medical College Hospital Madurai Government Government Rajaji Hospital Madurai Private Meenakshi Mission Hospital and Research Center Salem Government Government Mohan Kumaramangalam Medical College Sivagangai Government Government Sivagangai Medical College and Hospital Theni Government Theni Government Medical College and Hospital Tirunelveli Government Tirunelveli Medical College Hospital Trichy Government KAP Viswanathan Government Medical College Vellore Government Government Medical College Hospital Cuddalore Private Mahatma Gandhi Medical College Cuddalore Private Manakula Vinayagar Medical College Cuddalore Private PIMS Cuddalore Private Jipmer Hospital Kanyakumari Private Sri Mookambika Institute of Medical Science Hospital Madurai Private Velammal MC Hospital and Research Institute Trichy Private Chennai Medical College and Hospital Cuddalore Private A G Padmavathys Hospital Pvt Limited Kanyakumari Private C.S.I Mission Hospital Kanyakumari Private Dr.Jeyasekharan Medical Trust Kanyakumari Private Holy Cross Hospital Kanyakumari Private Siva Hospital Kanyakumari Private M L Hospital Karur Private Amaravathi Hospital Annexes  53 District Type Name of Facility Madurai Private Guru Multispeciality Hospital Madurai Private Meenakshi Mission Hospital and Research Centre Madurai Private Saravana Hospital Madurai Private SKG Hospital Madurai Private VMC Speciality Hospital Madurai Private BGM Hospital Madurai Private Bala Hospitals Madurai Private Harshitha Hospitals Pudukottai Private Muthu Meenakshi Hospitals Sivagangai Private Senthil Hospital Sivagangai Private Kauvery Medical Centre Thanjavur Private Meenakshi Multispeciality Hospital Thanjavur Private KG Multi Specialty Hospital Tirunelveli Private Shifa Hospital Tiruvarur Private Thiruvarur Medical Centre Trichy Private Dr. G. Viswanathan Specialty Trichy Private Kavery Medical Centre and Hospital Trichy Private A. J. Hospital Trichy Private G V N Hospital Trichy Private KMC Specialty Trichy Private Maruti Hospital Trichy Private Apollo Hospitals Trichy Private GKM Hospital Trichy Private Gitanjali Medical Centre Coimbatore Private Abinand Hospital Coimbatore Private Alwa Hospital Coimbatore Private Arun Hospital Coimbatore Private Ashwin Polyclinic Coimbatore Private Deepam Hospital Coimbatore Private K.G.M. Hospital Coimbatore Private K.R. Health Care Coimbatore Private Kalpana Medical Centre Coimbatore Private Karpagam Hospital Coimbatore Private KG Hospital Coimbatore Private Kongunad Hospital Coimbatore Private Kovai Medical Center and Hospital Coimbatore Private Kurinji Hospital Coimbatore Private Manu Hospital 54  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu District Type Name of Facility Coimbatore Private N.G. Hospital Coimbatore Private Pills Hospitals Coimbatore Private PSG Hospitals Coimbatore Private Ram Polyclinic Coimbatore Private Royal Care Super Specialty Hospital Coimbatore Private S.P.T Hospital Coimbatore Private Shree Sivaa Hospital Coimbatore Private Sree Abirami Hospital Coimbatore Private Sree Resmika Hospital Coimbatore Private Sri Bala Medical Centre Coimbatore Private Sri Lakshmi Hospital Coimbatore Private Sri Ramakrishna Hospital Dharmapuri Private DNV Polyclinic Dharmapuri Private K.S. Hospital Dharmapuri Private Om Sakthi Hospital Erode Private Care 24 Medical Centre and Hospital Erode Private City Hospital Erode Private CK Hospital Erode Private Erode Medical Centre Erode Private KGR Surgical Nursing Home Erode Private KMCH Specialty Hospital Erode Private Lotus Hospitals Research Centre Erode Private Maaruthi Medical Center and Hospitals Kerala Private DM Wayanad Institute of Medical Science Krishnagiri Private ARK Nursing Home Krishnagiri Private Arogya Seva Hospital Krishnagiri Private Chandrasekara Hospital Krishnagiri Private P.E.S.I.M.S.R Hospital Krishnagiri Private Vijay Hospital Nilgiris Private Ashwini Gudalur Tribal Hospital Nilgiris Private Nankam Hospital Salem Private Aishwaryam Speciality Hospital Salem Private Appu Venkatachalam Medical Research Centre Salem Private Aravind Hospital Salem Private Deepam Hospital Salem Private Dharan Hospital Salem Private Dr. Suraksha Specialty Medical Centre Salem Private Dr. Thiru Neuro Multi Specialty Annexes  55 District Type Name of Facility Salem Private Dr. Sundararajan Neuro Hospital Salem Private Kamala Hospital Agraharam Salem Private Karthik Media Centre Salem Private Kiruba Hospital Salem Private Kurinji Hospital Salem Private Manipal Hospital Salem Private Nathan Super Specialty Hospital Salem Private Neuro Foundation Three Roads Salem Private Nitish Ge and Multispeciality Hospital Salem Private Priyam Speciality Hospital Salem Private Salem Polyclinic Salem Private Saravana Hospital Salem Private Shanmuga Hospital Cancer Institute Salem Private Shanthi Nursing Home Salem Private SIMS Chellum Hospital Salem Private SKS Hospital Salem Private SPMM Hospital Salem Private Sri Gokulam Hospital Salem Private Sri Shellappa Hospital Salem Private Sundaram Multispecialty Hospital Salem Private Universal Cancer Hospital Salem Private Vinayaga Mission Kirubananda Medical College Salem Private Vinyaga Mission Hospital Hi-Tech Tirupur Private Aresta Medical Centre Tirupur Private Deepa Hospital Tirupur Private L.G. Medical Centre Tirupur Private Malar Priya Medical Centre Tirupur Private Rams Multispecialty Centre Tirupur Private Revathi Medical Centre Tirupur Private Sri Kumaran Hospital Tirupur Private Sri Saran Hospital Tirupur Private Sugan Sugaa Medical Centre Vellore Private Arun Hospital Vellore Private Christian Medical College and Hospital Vellore Private Indira Nursing Home Vellore Private Kafeel Emergency Care Vellore Private Sri Narayani Hospital and Research Centre Chennai Private Billoroth Hospitals 56  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu District Type Name of Facility Chennai Private Chennai National Hospital Chennai Private Dr Mehtas Hospitals Chennai Private Faith Multispecialty Hospital Chennai Private Kavery Hospital Chennai Private Kumaran Hospital Chennai Private Medway Hospital Chennai Private Murugan Hospital Kilpauk Chennai Private New Hope Medical Centre Chennai Private Prashanth Hospital Chennai Private Public Health and Welfare Society Chennai Private Vijaya Hospital Kanchipuram Private Chettinad Super Specialty Hospital Kanchipuram Private Dr Kamakshi Memorial Hospital Kanchipuram Private Hindu Mission Hospital Kanchipuram Private Karpaga Vinyaga Institute Kanchipuram Private Meenakshi Medical College Kanchipuram Private Saveetha Medical College Kanchipuram Private Sri Balaji Hospital Kanchipuram Private Sri Ramachandra Medical Center Kanchipuram Private SRM Medical College Hospital Kanchipuram Private St. Thomas Hospital Kanchipuram Private Tagore Medical College and Hospital Namakkal Private Aravinth Hospital Namakkal Private Thangam Hospital Namakkal Private Vivekanandha Medical Care Hospital Perambalur Private Dhanalakshmi Srinivas Hospital Theni Private NRT Hospital Thiruvallur Private K.V.T. Specialty Hospital Thiruvallur Private Sugam Hospital Villupuram Private E. S. Nursing College Hospital Villupuram Private E. S. Hospital Source: CMCHIS Annexes  57 Annex H District-Level Utilization Rate of CMCHIS Families Utilization rate (%) Out-of-district utilization District utilizing Total Government Private share (%) Ariyalur 4,327 2.4 1.1 1.3 76.6 Chennai 23,930 3.6 2.7 0.9 8.0 Coimbatore 15,071 2.1 0.9 1.2 7.9 Cuddalore 11,788 1.9 0.8 1.1 65.8 Dharmapuri 9,718 2.5 1.1 1.3 37.8 Dindigul 11,516 2.4 0.9 1.5 52.8 Erode 13,518 2.2 0.8 1.4 31.0 Kancheepuram 18,758 2.4 1.4 1.0 48.9 Kanyakumari 18,225 4.1 2.1 2.0 3.8 Karur 5,811 1.9 0.8 1.1 53.1 Krishnagiri 8,221 1.8 0.8 1.1 44.2 Madurai 18,017 2.8 1.4 1.4 13.7 Nagapattinam 7,734 2.2 1.3 0.9 58.5 Namakkal 12,447 3.0 0.9 2.2 46.9 Perambalur 5,797 4.0 1.8 2.2 37.8 Pudukkottai 8,709 2.1 0.9 1.2 60.5 Ramanathapuram 10,300 3.4 1.9 1.5 41.6 Salem 17,828 2.1 1.0 1.2 24.9 Sivaganga 8,805 2.9 1.7 1.1 41.4 Thanjavur 11,048 2.0 1.2 0.8 26.5 The Nilgiris 2,684 1.6 0.7 0.9 68.9 Theni 11,462 3.8 1.4 2.4 23.1 Tiruchirappalli 11,898 2.1 0.9 1.2 29.0 Tirunelveli 17,165 2.8 1.7 1.2 25.2 Tiruppur 11,796 2.2 0.7 1.5 61.7 Tiruvallur 21,460 2.8 2.0 0.8 89.7 Tiruvannamalai 14,509 2.7 1.8 0.9 69.7 Tiruvarur 6,106 1.9 1.1 0.8 55.4 Tuticorin 11,762 3.8 2.4 1.3 36.5 Vellore 20,589 2.5 1.5 1.1 42.5 Viluppuram 19,769 2.4 1.1 1.3 59.3 Virudhunagar 13,122 3.3 1.6 1.7 42.5 Tamil Nadu 403,890 2.6 1.3 1.2 40.3 Source: CMCHIS 58  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu Annex I District-Level Utilization Rate of CMCHIS, by Specialty Medical (%) Surgical (%) Diagnostics (%) District Government Private Government Private Government Private Ariyalur 0.4 0.3 0.5 0.6 0.3 0.5 Chennai 1.0 0.3 1.0 0.5 1.0 0.3 Coimbatore 0.4 0.4 0.3 0.8 0.3 0.2 Cuddalore 0.4 0.4 0.3 0.6 0.3 0.4 Dharmapuri 0.5 0.2 0.3 0.7 0.4 0.6 Dindigul 0.5 0.4 0.3 0.8 0.2 0.5 Erode 0.2 0.3 0.2 0.8 0.4 0.5 Kancheepuram 0.5 0.3 0.6 0.5 0.5 0.3 Kanyakumari 0.5 0.4 0.5 0.6 1.3 1.5 Karur 0.4 0.2 0.3 0.7 0.2 0.4 Krishnagiri 0.3 0.2 0.2 0.5 0.3 0.6 Madurai 0.6 0.7 0.6 0.6 0.4 0.3 Nagapattinam 0.6 0.2 0.5 0.4 0.2 0.5 Namakkal 0.4 0.3 0.2 1.1 0.3 1.1 Perambalur 0.5 0.7 0.5 1.0 0.9 0.9 Pudukkottai 0.5 0.2 0.3 0.5 0.2 0.7 Ramanathapuram 1.2 0.5 0.4 0.5 0.4 0.8 Salem 0.4 0.3 0.4 0.9 0.3 0.1 Sivaganga 0.5 0.5 0.6 0.5 0.8 0.3 Thanjavur 0.5 0.2 0.5 0.5 0.2 0.3 The Nilgiris 0.4 0.2 0.3 0.5 0.1 0.3 Theni 0.5 0.3 0.5 0.8 0.5 1.7 Tiruchirappalli 0.4 0.3 0.4 0.7 0.2 0.4 Tirunelveli 0.8 0.2 0.6 0.5 0.5 0.7 Tiruppur 0.3 0.4 0.2 0.9 0.2 0.5 Tiruvallur 0.8 0.2 0.8 0.4 0.7 0.3 Tiruvannamalai 0.6 0.2 0.8 0.6 0.6 0.3 Tiruvarur 0.5 0.2 0.5 0.3 0.1 0.4 Tuticorin 1.0 0.2 0.7 0.5 1.0 0.9 Vellore 0.7 0.2 0.5 0.5 0.4 0.5 Viluppuram 0.4 0.4 0.5 0.7 0.4 0.3 Virudhunagar 0.7 0.3 0.5 0.6 0.5 1.1 Tamil Nadu 0.5 0.3 0.5 0.6 0.4 0.5 Source: CMCHIS Annexes  59 Annex J Claims Volume and Value by Age Group Source: CMCHIS Source: CMCHIS 60  Nuts and Bolts of the Chief Minister’s Comprehensive Health Insurance Scheme (CMCHIS): A Case Study from Tamil Nadu