Human Development Network Health, Nutrition, and Population Series Systems for India's Poor Findings, Analysis, and Options David H. Peters Abdo S. Yazbeck- Rashbm-i R. Sharma G. N. V. Ramana Lant H. Pritchett Adamin AVagstaff ., I-,Y Health, Nutrition, and Population Series This series is produced by the Health, Nutrition, and Population Family (HNP) of the World Bank's Human Development Network. It provides a vehicle for publishing polished material on the Bank's work in the HNP Sector, for consolidating previous informal publi- cations, and for improving the standard for quality control, peer review, and dissemination of high-quality analytical work. The series focuses on publications that expand our knowledge of BNP policy and strategic issues that can improve outcomes for the poor and protect vulnerable populations against the impoverishing effects of illness. Best practice examples of both global and regional relevance are presented through thematic reviews, analytical work, and case studies. The Editor in Chief of the series is Alexander S. Preker. Other members of the Editorial Committee are Mukesh Chawla, Mariam Claeson, Shantayanan Devarajan, Gilles Dussault, A. Edward Elmendorf, Armin H. Fidler, Charles C. Griffin, Jeffrey S. Hammer, Peter E Heywood, Prabhat Jha, Gerard Martin La Forgia, Jack Langenbrunner, Ruth Levine, Maureen Lewis, Samuel S. Lieberman, Benjamin Loevinsohn, Elizabeth Lule, Akiko Maeda, Judith Snavely McGuire, Milla McLachlan, Thomas W Merrick, Philip Musgrove, Ok Pannenborg, Oscar Picazo, Mead Over, Juan Rovira, George Schieber, and Adam Wagstaff. Human Development Network Health, Nutrition, and Population Series Better Health Systems for India's Poor Findings, Analysis, and Options David H. Peters Abdo S. Yazbeck Rashmi R. Sharma G. N. V Ramana Lant H. Pritchett Adam Wagstaff THE WORLD BANK Washington, D.C. © 2002 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 All rights reserved. Published 2002 Manufactured in the United States of America 1 2 3 405 0403 02 The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is copyrighted. No part of this work may be reproduced or trans- mitted in any form or by any means, electronic or mechanical, including photocopying, recording, or inclusion in any information storage and retrieval system, without the prior written permission of the World Bank. The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, tele- phone 978-750-8400, fax 978-750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street, NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. ISBN 0-8213-5029-3 Library of Congress Cataloging-in-Publication Data has been applied for. Contents Foreword xv Preface xvii Acknowledgments xxi Acronyms and Abbreviations xxv Overview 1 The Indian Health System 3 Which Way Forward? 8 Options at a Glance 17 Part 1: Raising the Sights for India's Health System 23 1 A Crossroads 2 5 Objectives of the Health System and the Health Transition 27 The Current Policy Context 28 Historic Vision and Current Realities 29 Focusing on Four Critical Activities in the Health System 32 The Approach to Reform 32 2 Problems of the Public and Private Sectors 39 Current Structure of the Public Sector 39 iii iv * Better Health Systems for India's Poor Problems of the Public Sector 40 Structure of the Private Sector 44 Problems with the Private Sector 49 Notes 52 3 Policy Actions for Critical Health System Activities 53 Health System Oversight 55 Public Health Service Delivery 64 Ambulatory Curative Services 69 Inpatient Curative Services and Health Insurance 74 Concluding Remarks 83 Notes 84 4 Putting It Together: Raising the Sights of India's Health System 85 Different Choices for Different Parts of India 86 Choices for the Central Government 92 Choices for the States 96 Putting It All Together: The Case of Maternal Health 113 Specific Actions for Consideration across India 118 An Agenda for the Future 123 Concluding Remarks 124 Appendix 126 Notes 128 Part 2: Theory and Evidence 131 5 Health System Framework 133 The Descriptive Framework: Health System Actors, Functions, and Outcomes 135 Framework for the Consideration of Government Intervention 142 A Framework for Deciding How to Intervene: Make-Buy-Regulate-Inform 144 Notes 148 Contents * v 6 The Functioning of the Private Sector Market 151 Context: The Private Health Sector in India 155 Methods 157 Ambulatory Care 159 Inpatient Care 171 Concluding Remarks 186 Appendix 187 Notes 198 7 Setting National Health Care Priorities and Ensuring Equitable Delivery of Public Sector Services 201 Do Centrally Sponsored Schemes Meet India's Health Care Needs? 201 Households' Use of Personal Health Services 212 How Well Are Public Health Services Reaching the Poor? 217 How Does the Pattern of Public Spending Affect Health Outcomes? 229 Appendix 230 Notes 231 8 Financing Health 233 Health Sector Spending 233 Financing Mechanisms 240 Conclusions 255 Appendix 257 Notes 264 9 Health System Outcomes 265 Health Status 269 Financial Protection 286 Responsiveness to the Public 293 Concluding Remarks 298 Appendix 300 Notes 321 vi * Better Health Systems for India's Poor Appendixes A Studies Conducted for the Present Report 323 B Background Papers 327 C Major Recommendations of National Health Policy Reports since Independence 331 D Efforts to Address the Role of Private Providers in National Tuberculosis Control Programs 335 Bibliography 337 Tables 0.1 Major Indian States, by Stage of Health Transition and Institutional Capacity 8 0.2 Improving Health System Oversight 17 0.3 Strengthening Public Health Services 18 0.4 Strengthening Ambulatory Curative Services 19 0.5 Inpatient Care and Health Insurance 20 1.1 Examples of Health System Functions and Challenges in India 36 2.1 International Comparisons of Health Care Work Force and Hospital Beds, 1990-98 41 2.2 International Comparisons of Health Service Utilization and DALYs Lost, 1990-98 41 2.3 Health Care Work Force and Health Facilities in the Public and Private Sectors in India, Selected Years, 1981-98 46 2.4 Distribution of Outpatient and Inpatient Health Services across the Public and Private Sectors in India, 1986-87 and 1995-96 48 3.1 Summary of Actions for Critical Areas of Activity in the Health System 54 3.2 International Examples of Varying Public-Private Mixes in the Delivery of Tuberculosis Care 59 4.1 Selected Health Status Outcomes in India and Major Indian States, Selected Years, 1992-99 87 Contents * vi- 4.2 Selected Health Service Outcomes in Major Indian States and India Overall, Selected Years, 1995-99 88 4.3 Categorization of Major Indian States by Characteristics Influencing Fundamental Health System Choices 89 4.4 Local Factors to Consider at State Level when Prioritizing Health Systems Choices 91 4.5 Major Health System Choices Facing Indian States, by Stage of the Health Transition, and the Central Government 97 4.6 Pros and Cons of Actions for Improving Oversight 10l 4.7 Pros and Cons of Actions for Improving Public Health Services 10' 4.8 Pros and Cons of Actions for Improving Ambulatory Curative Care 1 07 4.9 Pros and Cons of Actions for Improving Inpatient Care and Health Insurance 1110 4.10 Allocation of Critical Health System Functions for Maternal Care between the Public and Private Sectors, by Type of Vision for Public-Private Partnerships 115 4.11 Recommendations for Government Action on Critical Functions of India's Health System 121 4A. 1 Key Gaps in Knowledge about and Experience in India's Health System, and Potential Uses of the Needed Information 126 6.1 Alternative Private Practitioners Receiving Informal Payments, Andhra Pradesh and Uttar Pradesh 161 6.2 Participation in National Health Programs by Private Practitioners, Andhra Pradesh and Uttar Pradesh 170 viii * Better Health Systems for India's Poor 6.3 Performance Indicators of Private Hospitals, Andhra Pradesh and Uttar Pradesh, 2000 180 6.4 Quality Assurance Standards in Hospitals, Andhra Pradesh and Uttar Pradesh 181 6.5 Private Hospitals Whose Managers Favor Various Procedures for Improving Hospital Quality, 2000 182 6.6 Proportion of Patients Satisfied or Very Satisfied with Overall Quality of Care at Public and Private Health Facilities, Andhra Pradesh, by Type of Facility and Wealth of Patient 183 6.7 Private Hospitals Offering Concessions to the Poor, Andhra Pradesh and Uttar Pradesh 183 6.8 Proportion of Private Hospitals that Have Received Public Benefits, Andhra Pradesh and Uttar Pradesh 185 6.9 Participation in National Health Programs by Private Hospitals, Andhra Pradesh and Uttar Pradesh 186 6A. 1 Distribution of Main Reasons Given by Alternative Private Practitioners in Andhra Pradesh and Uttar Pradesh for Becoming a Medical Practitioner 187 6A.2 Clinical Conditions Treated by Alternative Private Practitioners over Two Days, Andhra Pradesh and Uttar Pradesh 188 6A.3 Patient Fees for an Outpatient Consultation in the Private Sector, Alternative Practitioners Compared with Qualified Allopaths, Andhra Pradesh and Uttar Pradesh 189 6A.4 Allopathic Therapies Offered at Clinics of Alternative Private Practitioners, Andhra Pradesh and Uttar Pradesh 190 6A.5 Concessions Offered to the Poor by Alternative Private Practitioners, Andhra Pradesh and Uttar Pradesh 190 Contents * ix 6A.6 Alternative Private Practitioners Who Rate Government Health Programs as Good or Very Good, Andhra Pradesh and Uttar Pradesh 191 6A.7 Private Hospitals Experiencing Moderate or Severe Obstacles to Credit, Andhra Pradesh and Uttar Pradesh 191 6A.8 Private Hospitals Making or Planning Capital Investments, Andhra Pradesh and Uttar Pradesh, by Source of Funds 192 6A.9 Revenue Issues Affecting Private Hospitals, Andhra Pradesh and Uttar Pradesh 192 6A. 10 Labor Issues in Private Hospitals, Andhra Pradesh and Uttar Pradesh 193 6A.11 Public Infrastructure Issues Affecting Private Hospitals, Andhra Pradesh and Uttar Pradesh 19. 6A. 12 Regulatory Issues Affecting Hospital Managers, Andhra Pradesh and Uttar Pradesh 194 6A. 13 Ideal Job Characteristics Reported by Public and Private Sector Health Workers, Andhra Pradesh and Uttar Pradesh 195 6A. 14 Presence in Current Job of Ideal Job Characteristics, as Reported by Public and Private Health Sector Workers, Andhra Pradesh and Uttar Pradesh 195 6A. 15 Presence of Ideal Job Characteristics in Andhra Pradesh: Differences in Ratings between Public and Private Sector, by Characteristic of Health Worker 197 6A. 16 Average Number of Full-Time and Part-Time Nurses and Private and Government Doctors Working in Private Hospitals, Andhra Pradesh and Uttar Pradesh 198 7.1 Deaths and DALYs Lost Associated with Conditions Covered by National Health Programs, India, 1990 and 1998 206 x * Better Health Systems for India's Poor 7.2 Deaths and DALYs Lost Associated with Major Conditions Not Covered by National Health Programs, India, 1990 and 1998 207 7.3 Income Bias in Public Spending on Curative Care in India and Selected States 224 7A. 1 Infant and Child Mortality among Indian States: Effects of Public Hospitalization, Equity, and Other Factors, 1995-96 230 8.1 Market Failures, Consequences, and Responses in Financing Health Care 249 8A. 1 Health Expenditures and Cost Recovery in the Public Sector of Selected States, 1996 257 8A.2 Salient Features of Some Insurance Schemes in India 258 8A.3 Questions for Policymakers to Ask when Deciding on Resource Allocations and Purchasing 260 9.1 India's Share of the World's Health Problems 270 9.2 Top 10 Specific Causes of Death in India, 1998 271 9.3 Top 10 Specific Causes of DALYs Lost in India, 1998 272 9.4 Under-Five Mortality Rates in Indian States and in the World 280 9.5 Health Outcomes among Scheduled Castes, Tribes, and Rest of Population in India, 1998-99 283 9.6 Health Status Indicators-Comparison between the Poorest and Richest Quintiles of the Indian Population, 1992-93 284 9.7 Health Outcomes by Standard of Living, 1998-99 285 9.8 Indians Reporting an Illness within a 15-Day Period Who Did Not Seek Care, and Distribution of Reasons for Inaction, by Income Quintile, 1995-96 292 Contents * xi 9.9 Presence of Mechanisms of Consumer Redress at Public Health Facilities and Private Hospitals 294 9.10 Patients Satisfied or Very Satisfied with Health Services in Public and Private Facilities in Andhra Pradesh, by Sex and Income Level 297 9.11 Quality of Care Reported by Women after Their Most Recent Visit to a Health Facility, Public or Private, 1998-99 298 9A. 1 Distribution of DALYs Lost in World, India, and Countries Grouped by Income, by Condition, 1998 300 9A.2 Distribution of Deaths in World, India, and Countries Grouped by Income, by Cause, 1998 398 9A.3 Comparison of India and Other Countries on Selected Health-Related Indicators, Selected Years, 1992-99 316 9A.4 Causes of DALYs Lost in India, 1990 and 1998 317 9A.5 Infant Mortality and Total Fertility Rates in India and Major Indian States, 1981-97 318 9A.6 Underweight Children under Three Years of Age in Major States of India, 1998-99 319 9A.7 Reduction in Rates of Severe and Total Malnutrition (Weight for Age) among Children in India and Major States in India between 1992-93 and 1998-99 329 9A.8 Comparison of Female and Male Health Outcomes in India, 1998-99 321 Figures 0.1 Distribution of Public Expenditures in India on Curative Care, by Income Quintile, 1995-96 4 0.2 Percent of Hospitalized Indians Falling into Poverty from Medical Costs, 1995-96 5 * Better Health Systems for India's Poor 0.3 Public and Private Sector Shares of Delivery of Selected Health Care Services in India, by Income Status of Patients, 1995-96 7 3.1 Range of Interventions to Involve Private Practitioners 58 5.1 Descriptive Framework of the Health System 137 5.2 Decision Points for Government Financing and Provision in the Health Sector 143 5.3 Measurability and Contestability of Health Services (Product Markets) 145 5.4 Measurability and Contestability of Inputs (Factor Markets) in the Health Sector 146 5.5 Make or Buy Decisions, and Inform-Regulate Options to Increase Contestability and Measurability 147 6.1 Public and Private Sector Shares in Service Delivery across India, 1995-96 158 6.2 Types of Assistance Favored by Alternative Private Practitioners in Andhra Pradesh and Uttar Pradesh 171 6.3 Average Number of Beds per Full-Time- Equivalent Doctor and Nurse in Private Hospitals, Andhra Pradesh and Uttar Pradesh 179 7.1 Prevalence of Alcohol and Tobacco Use in India, by Income Quintile, 1995-96 210 7.2 Public and Private Sector Shares of Hospitalization, by Income Quintile of the Population 214 7.3 Public and Private Sector Shares of Inpatient Bed Days of Patients below the Poverty Line 217 7.4 Public Expenditures on Curative Care, by Income Quintile 219 7.5 Income Bias of Public Spending on Hospital and Primary Health Care Facilities 221 7.6 Income Bias of Public Spending on Immunizations 222 Contents * xiii 8.1 Public Sector Spending on Health in India, 1985-2000 236 8.2 Public Expenditures on Health in Selected States in India, Grouped by State and Distributed by Government Level, 1995-96 2 7 8.3 Private Spending on Health Services at Private and Public Facilities, 1995-96 238 8.4 Comparison of State Public and Private per Capita Health Spending, 1995-96 240 8A. 1 Structure of Health Care Sources and Uses 261 8A.2 Out-of-Pocket Payments and Household Income, 1995-96 262 8A.3 Out-of-Pocket Payments by Socioeconomic Group, 1995-96 263 9.1 Infant Mortality Rate in India, 1980-2000 273 9.2 Distribution of Total Out-of-Pocket Health Expenditures in India as a Proportion of Nonfood Expenditures, 1995-96 287 9.3 Sources of Financing for Private Expenditures on Hospitalization in India, by Income Quintile, 1995-96 289 9.4 Hospital Patients below the Poverty Line Who Financed Their Care in Public and Private Hospitals from Borrowing or Sale of Assets by State, 1995-96 29() 9.5 Reasons for Using the Private Health Sector, by Poverty Status 296f Boxes 1.1 Lessons on Health Reform in Wealthy Countries 34 2.1 A Proactive Public Health Provider 51 3.1 A Quality Assurance Initiative in Mumbai 6C xiv * Better Health Systems for India's Poor 3.2 Oversight Constraints and Opportunities: Reform of the Drug Control System in Uttar Pradesh 61 3.3 Effective Technologies to Combat Communicable Diseases 65 3.4 International Approaches to Spreading Risks and Subsidizing the Poor 75 3.5 Health Insurance for the Informal Sector: The SEWA Experience in Gujarat 79 4.1 Keys to Success 105 5.1 How Chronic Illness Makes People Poor 134 6.1 Empirical Findings and Policy Challenges 152 6.2 Working with Untrained Private Providers: The Janani Experience 168 7.1 Empirical Findings and Policy Challenges 202 7.2 Responding to Violence against Women 209 7.3 Equity of Public Spending on Health: International Experience 224 7.4 Adjusting Public Sector Hospitalization Rates for Income Bias, 1995-96 226 8.1 Empirical Findings and Policy Challenges 234 8.2 How Hospitalization Can Devastate 243 8.3 Innovative Health Insurance Schemes in India 247 8.4 International Experience with Health Insurance for the Informal Sector 250 8.5 Conclusions of the National Seminar on Health Insurance 252 8.6 Contracting Out Noncinical Hospital Services to the Private Sector: Experiences from Karnataka 254 9.1 Empirical Findings and Policy Challenges 266 9.2 "Achievement" Means Doing Well for Everyone-Not Just the Better-Off 276 WA. Foreword What type of health system should India have in the 21st century? That was the question posed by the government of India when it asked the World Bank to help analyze India's health system. The recognition was growing that conditions in India were changing rap- idly, that the health system needed to keep up with these charges, and that important aspects of the health system were being cver- looked by current approaches. Through a wide consultative process that included many of India's internal and external development partners, a set of topics was selected for study, with results reviewed as they emerged. Alore than a dozen Indian institutions, in partnership with the World Bank and the Government of India, conducted the research. The studies provided new data and analysis that have not been available before on a number of fundamental issues of health care in India: * The behavior of the private market in health * The prevalence of chronic disease risk factors * The distribution of benefits from different types of public and private health services * The degree of financial protection in health care xv xvi * Better Health Systems for India's Poor * The degree of protection of patients' interests * The laws and practices guiding health care. Better Health Systems for India s Poor fills many gaps in under- standing and synthesizes much about what can be learned from India's health system. Emerging from this analysis is an important set of principles for reform: * Look after the health needs of the poor and vulnerable sections of society * Prepare for the health transition with appropriate health financ- ing systems and programs * Harness the energy of the private sector while counteracting its failures * Focus on quality and accountability in health services. The key features of this work-the consultative processes, the sensitivity to variations in conditions among groups and regions, the clarity of analysis, and the tangibility of the alternatives for reform-make this study relevant to any country interested in mak- ing its health systems more effective, equitable, and accountable. Jo Ritzen Vice President Human Development Network The World Bank Preface A host of questions emerged as the study team, assembled a- the request of the government of India, considered the future o.^ the Indian health care system. But throughout the consultations and deliberations, two questions remained foremost: 1. How can India meet the health needs of the most vulnerable seg- ments of its population? 2. How can the roles of the public and private sectors be structured to better finance and deliver health services? The overall aim of the exercise was to help India answer these questions, by informing and facilitating a professional and public discussion on the future directions for India's health system. This report synthesizes the resulting detailed studies that addressed dif- ferent aspects of these questions. The report is intended for a vari- ety of audiences, including policymakers, health sector managers and workers, researchers, consumer advocates, staff in development agencies, and the public interested in the health system and devel- opment in India. Representatives of these stakeholders were involved in the selection, design, conduct, and dissemination of the research summarized in this report. In 1999, the research agenda for the studies was set through wide- spread consultations among many parties: central and state goviern- ment officials, private sector providers, health insurance companies, xvii xviii * Better Health Systems for India's Poor not-for-profit providers, legal experts, academics, research groups, consumer organizations, medical associations, international agencies (World Health Organization, United Nations Children's Fund, European Union, Department for International Development- United Kingdom, and U.S. Agency for International Development), and experts within the World Bank. Small working groups were established to collaboratively prepare terms of reference for each study that was identified. The studies themselves were conducted by Indian institutions. During 2000 and 2001, the same stakeholders were asked to go through the findings of the studies as they emerged and to formu- late and discuss options. International experts and those in India also provided advice on the individual studies and this report. The studies have produced data and analysis not previously avail- able on several key topics: (a) the behavior of the private market in health, (b) the prevalence of chronic disease risk factors, (c) the dis- tribution of benefits from different types of public and private health services, (d) the degree of financial protection in health, (e) schemes for the protection of patient interests, and (f) Indian law on health care. Appendix A lists the individual studies, the research organiza- tions, the policy questions addressed, and use of study results. The 21 reports from the individual studies are listed as background papers in appendix B. The studies are selective; they were not intended to cover all aspects of the Indian health system in detail, especially not aspects, such as public sector health services, on which considerable informa- tion was already available. In contrast, much research was conducted on the private sector, for which little information was available. Moreover, the scope of the overall study itself was selective. Important sectors that influence health but that are not part of the direct delivery of health care-such as education, sanitation, and water supply-are not examined in detail in the report. Within the health sector, important issues concerning health manpower devel- opment and pharmaceutical policy were not studied in detail on the belief that analysis of these topics should be influenced by other choices about how the health system would be shaped. Preface * xix The report focuses on four areas of the health system in which, in the judgment of the study group, reforms and innovations would make the most difference to the future of the Indian health system: oversight, public health service delivery, ambulatory curative care, and inpatient care, together with health insurance. Part 1 of the report contains four chapters that discuss current conditions and policy options. Part 2 presents the theory and evi- dence to support the policy choices. The general reader may be most interested in the overview chapter and in the highlights found at the beginning of each of the chapters in part 2. These highlights outline the empirical findings and the main policy challenges dis- cussed in the chapter. If reforms are to be carried out in India's health sector, the vision for change must come out of the discussions among the stakehold- ers in the health system. Therefore, the report does not set out to prescribe detailed answers for India's future health system. It does, however, have a goal: to support informed debate and consensus building, and to help shape a health system that continually st-ives to be more effective, equitable, efficient, and accountable to the Indian people, and particularly to the poor. Acknowledgments This report is a product of a wide range of consultations with poli- cymakers, professional associations, academics, private sector repre- sentatives, and nongovernmental organizations (NGOs). Key find- ings were discussed at six workshops: one in Calcutta, through the State Health Systems Development Projects annual review; one in New Delhi, organized by the Voluntary Organization in the Inter- est of Consumer Organization; two in Hyderabad, organized by the Administrative Staff College of India; one in Lucknow, organized by the Indian Institute of Management; and in New Delhi, a f nal, national seminar organized with the Ministry of Health and Family Welfare. Active participation was received, with appreciation, from state Secretaries of Health, researchers, and other representatives of national and state governments, NGOs, consumer groups, and development agencies. The teams that developed terms of reference for the background papers, and the authors of those papers themselves, made significant contributions to each others' work and to this final report; the authors of the papers included R. Baru, S. Chakraborty, G. Chel- laraj, R. Durvasula, A. Ferreiro, C. Garg, R. Govindaraj, R. Kutty, A. Mahal, P. Mahapatra, B. Misra, VR. Muraleedharan, S. Nandraj, M. Pearson, K. Prasad, R. Priya, I. Qadeer, K.S. Reddy, V Selvaraju, P. Srivastava, Aj. Syed, A. Thekkuveetti, and S.K Verma. The extensive collaboration of the Government of India's Mlin- istry of Health and Family Welfare, led by Mr. J. Chowdhury, Sec- xxii * Better Health Systems for India's Poor retary, is acknowledged. Mr. K.K. Bakshi and Ms. Shailaja Chandra, initiated the studies while at the Ministry of Health and Family Wel- fare. Ms. K. Sujatha Rao, Joint Secretary, Ministry of Health and Family Welfare, spearheaded the organization of stakeholders and the three workshops and numerous meetings that helped shape the studies. This report was prepared by a team led by David H. Peters. Abdo S. Yazbeck, Rashmi R. Sharma, and G.N.V Ramana were full-time team members. Other team members were Lant H. Pritchett and Adam Wagstaff. Major contributions to the development of the studies were made by Richard Feacham and Monica Das Gupta. Bruce Ross-Larson, Barbara Karni, Steve Kennedy, and Stephanie Rostron, all with Communications Development, assisted with the editing. For this publication, Gregg Forte edited the final product, with Janet H. Sasser serving as the production editor. Jan- mejay Singh, Pronita Chakrabarti, and Prasun Bhattacharjee pro- vided research assistance. Particular thanks go to Nina Anand and Katia Gomes Pinto Visconti for administrative support and help with production of the report. Richard Skolnik, in his capacity as Director, South Asia, Human Development, and Tawhid Nawaz, India Team Leader, worked closely with the team, which was also supported by the South Asia Health, Nutrition, and Population team and the India Country Team. The report is endorsed by Edwin Lim, India Country Direc- tor, and Joelle Chassard-Manibog, Country Coordinator. The team was advised by Alexander S. Preker, April Harding, Philip Musgrove, Davidson Gwatkin, and Robert Fryatt. Also pro- viding peer review were R. Radhakrishna, K.V Narayana, P.S. Vashishtha, R. Bhatt, C.A.K. Yesudian, and P. Srinivasan. A quality enhancement review of the studies was conducted by Alain Colliou, Susan Stout, Shanta Devarajan, and Maureen Lewis. Substantive comments were made by the following colleagues from the World Bank: Mukesh Chawla, Edgardo Favaro, Jeffrey Hammer, Clive Harris, Peter F. Heywood, Manuel Jimenez, Sanjay Kathuria, Ben- jamin Loevinsohn, Anthony Measham, Tawhid Nawaz, Richard Skolnik, and Roberto Zagha. Numerous colleagues from outside the Acknowledgments xxiii Bank provided comments and suggestions at various stages cf the work, including Cristian Baeza, Anne Bamasaiye, Victor Barbiero, Kevin Brown, Ken Grant, Indrani Gupta, Pradeep Kakkar, Tim Martineau, David Nicholas, J.-P. Poullier, Pravin Visaria, and M.S. Valiathan. The preparation of the background papers and the convening of several workshops were supported by the Ministry of Healt. and Family Welfare, World Health Organization, Department for Inter- national Development-United Kingdom, U.S. Aid for Interna- tional Development, the Dutch government, and the World Bank. Acronyms and Abbreviations AIDS Acquired immune deficiency syndrome AIIMS All Indian Institute of Medical Sciences ANM Auxiliary nurse midwives AP Andhra Pradesh APP Alternative private practitioner ASCI Administrative Staff College of India BAIF Bharatiya Agro Industries Foundation BMC Bombay Municipal Corporation CAG Citizen, Consumer and Civic Action Group CEHAT Centre for Enquiry into Health and Allied Themes CBHI Central Bureau of Health Intelligence CGHS Central Government Health Scheme CHC Community Health Centre CPR Centre for Policy Research CSS Centre for Social Studies DALY Disability-adjusted life year DFID Department for International Development DHS Demographic and health surveys DOTS Directly observed treatment short-course ESIS Employees State Insurance Scheme EU European Union GIC General Insurance Corporation GDP Gross domestic product GHIP Group Health Insurance Program xxv xxvi * Better Health Systems for India's Poor GP General practitioner HAP Health Action for the People HIV Human immunodeficiency virus HNP Health, Nutrition, and Population (The World Bank) ICDS Integrated Child Development Services IEG Institute of Economic Growth IHBAS Institute of Human Behavior and Allied Sciences IMR Infant mortality rate IHD Institute for Human Development IHS Institute of Health Systems IIPS Indian Institute for Population Sciences ISM Indian Systems of Medicine JSS Jagadguru Sri Shivarathreeshwara (College of Pharmacy) MBBS Bachelor of Medicine and Bachelor of Surgery MGRMU The Tamil Nadu Dr. M. G. R. Medical University MOHFW Ministry of Health and Family Welfare MTP Medical termination of pregnancy NCAER National Council for Applied Economic Research NGO Nongovernmental organization NIPFP National Institute of Public Finance and Policy NTP National Tuberculosis Control Programs NSSO National Sample Survey Organisation PHC Primary health center PRI Panchayati Raj Institution SC/ST Scheduled caste or tribe STD Sexually transmitted disease STEM Center for Symbiosis of Technology, Environment and Management TRIPS Trade-related aspects of intellectual property TB Tuberculosis TFR Total fertility rate UNICEF United Nations Children's Fund UP Uttar Pradesh USAID U.S. Agency for International Development WHO World Health Organization Overview India's health system is at a crossroads. Since the country's Indepen- dence, in 1947, India's health conditions have changed. A high pro- portion of the population continues to suffer and die from preventa- ble infections, pregnancy and childbirth-related complications, and undernutrition. At the same time, new health threats are stretching the capacity of the health system to respond. An estimated 3.5 million Indians are living with human immunodeficiency virus (HJV), an I the virus has now spread beyond highly susceptible groups to the general population in some states, threatening to erase much of the social, economic, and health gains since Independence. Also besetting the population now are noncommunicable diseases such as heart disease and mental illness, health problems associated with countries with a higher income than India. Building robust health systems reqlires building the capacity to do better on the "unfinished agenda." of health problems as well as meet the emerging realities and challenges. India is in the midst of a "health transition"-at varying rates depend- ing on the state and population group. The transition is demographic- a decline in mortality and fertility rates and an aging of the population; epidemiological-a shift in the pattern of ill health from malnutrition and communicable disease to the chronic diseases of adulthood; and social-rising capabilities and expectations of the population regarding health care. A high proportion of the population continues, however, to suffer and die from preventable infections, pregnancy and childbirth- related complications, and undernutrition-the "unfinished agenda" of 1 2 * Better Health Systems for India's Poor the health transition. The large disparities across India place the burden of these conditions mostly on the poor, women, and scheduled tribes and castes. The poorest 20 percent of Indians, for example, have more than double the mortality rates, malnutrition, and fertility of the rich- est quintile. Despite all this, the public remains uninformed about much of the health system. It knows little about whether health services are appropriate, who is benefiting from them, whether quality is suffi- cient, or whether people are getting good value from public and pri- vate spending on health. Equity, quality, and accountability are badly wanting in both the public and private health sectors. The time has come to reassess how the Indian health system should function and to retool it for the new millennium. This report is a product of extensive consultations and research conducted by more than a dozen Indian institutions. The report does not propose a blueprint for reform. The experience that emerged from its preparation strongly suggests that vision, broadly based political work, and a spirit of experimentation, rather than an abstract plan, are the key ingredients of any future improvements; plans and projects that do not emerge from a collaborative process animated by such ingredients will not be meaningful. The report clearly shows that reform is needed, however, and it outlines some broad principles for reform efforts, which should be led by govern- ment. The main message is that the government and the public should raise their sights in four ways: 1. Take responsibility for the needs of the entire population-by mak- ing the health system more pro-poor, gender sensitive, and client friendly, and by responding to the high burden of preventable dis- eases borne by the poor, scheduled tribes and castes, and women. 2. Look forward to the health transition-by preparing for the shift in disease burden and increase in health costs by developing health financing systems. 3. Remove the blind spot to the private sector-by harnessing its energy and countering its failures. Overviem * 3 4. Focus efforts-by emphasizing quality, efficiency, and accounta- bility of health services in both public and private sectors. Ultimately, changing the shape of the health system depends on political decisions at national, state, and local levels-decisions that will be shaped by expressions of political preferences, social expecta- tions, and the positions that leaders take at each level. International experience tells us that no single correct answer can be found for shaping a country's health system. In India, the priority of issues and choice of options should vary according to the conditions in the var- ious states and districts. An explicit approach to policy formulation and implementation will help to continually evolve the healtl. sys- tem-and to ensure that the health system is improving the health of all Indians in an accountable, equitable, and affordable manner. The Indian Health System Despite the establishment of a large public network of health providers, public spending on health is very low, stagnant at around 1 percent of gross domestic product (GDP). Such spending puts India among the bottom 20 percent of countries. It is lower than what most low-income countries spend, and it is far below what is needed to provide basic health care to the population. The :.arge variance in health financing among Indian states is increasing the gap in public resources for health between rich and poor states: and it threatens to expand existing gaps in health system outcomes. The states of Kerala, Punjab, and Tamil Nadu, for example, have dcuble the per capita public health spending of Bihar and Madhya Pradesh. As in other countries, public spending on preventive health serv- ices has a lower priority than curative care. And curative serrices themselves are highly pro-rich in distribution. About 3 rupees (Rs) is spent on the richest quintile for every Rs 1 spent on the poorest 20 percent (figure 0.1). Yet in three states (Kerala, Tamil Nadu, and Maharashtra), the distribution of public spending on health is nearly 4 * Better Health Systems for India's Poor Figure 0.1 Distribution of Public Expenditures in India on Curative Care, by Income Quintile, 1995-96 Percent 40- 30- 20 - 1 st 2nd 3rd 4th 5th (poorest) (richest) Income quintile Source: National Sample Survey Organisation (1998); Background Paper 18. uniform across income groups. Statistically, states with better equal- ity in their public spending have better health status outcomes. Private health spending presents a different story. It accounts for more than 80 percent of all health spending, one of the highest propor- tions of private spending found anywhere in the world. Nearly all the private spending in India is out-of-pocket at the point of service use, an inefficient way to finance health care that leaves people highly vulnera- ble. As in most developing countries, poorer households purchase less curative health care from the private sector than do richer households. Partly because of inability to pay and the lack of risk pooling, the poor are much less likely to be hospitalized. Across India, those above the poverty line have more than double the hospitalization rates of the poor. Hospitalization frequently results in financial catastrophe, espe- cially in the absence of risk-pooling mechanisms. Only 10 percent of Indians have some form of insurance, and most of the forms are Overview * 5 inadequate. Hospitalized Indians spent 58 percent of their total annual expenditures on health care. More than 40 percent of hospi- talized people borrow money or sell assets to cover expenses. One conservative estimate finds that one-fourth of hospitalized Indians were not poor when they entered the hospital but became so because of hospital expenses, a risk that, like many other elements of the Indian health system, varies greatly from state to state (figure 0.2). Even at public hospitals, which are intended to protect the poor from financial risks, the poor are vulnerable to health costs. Indeed, in some states (Uttar Pradesh, West Bengal, Madhya Pradesh, Rajaslhan, Haryana, and Bihar), the poor are more likely to borrow money when hospitalized in the public sector than in the private sector. Figure 0.2 Percent of Hospitalized Indians Falling into Poverty fro-n Medical Costs, 1995-96 Bihar Uttar Pradesh Rajas'than Gujarat Madhya Pradesh West Bengal Northeast states National average Maharashtra Orissa Haryana = Andhra Pradesh I _ I Karnataka _ Tarnil Nadu Kerala _ 0 5 10 15 20 25 30 35 40 Percent Note: Northeast states consist of Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland, and Tripura. Source: National Sample Survey Organisation (1998); authors' calculations. 6 * Better Health Systems for India's Poor The private sector accounts for the majority of curative care serv- ices in India, although, again, considerable variation exists across services and states (figure 0.3). Overall, the distribution of services in the private sector is even more skewed toward the rich than it is in the public sector. The poor still depend on the public sector for the majority of their health services, with one exception-the private sector provides 79 percent of outpatient care for those below the poverty line, much of which is of low quality and provided by untrained practitioners. An important group of providers are the alternative private prac- titioners, consisting of two groups: one, those who attempt to prac- tice Western, allopathic medicine without training in the discipline; and the other, those who practice Indian systems of medicine. The alternative private practitioners make up a larger segment of the pri- vate ambulatory care market than qualified allopathic doctors. Regardless of type of provider, quality assurance is a problem, with most care reflecting poor clinical practices and standards and inade- quate staffing. Many opportunities are, however, identified in this report for the government to work with or influence the practice of private providers, including simple measures to help improve the quality of the care from alternative private practitioners. The large variation between states-a major theme of this report- means that recommendations for the whole of India or even for groups of states run the risk of oversimplification (table 0.1 shows one cate- gorization of states used in this report). Health is a shared responsibil- ity of the central government and the states. How much discretion each state has over the policy options in any given area will vary. In some public health activities, the central government will likely take the lead through centrally sponsored schemes. But in curative services, states already exhibit enormous variations in what they do and in how well they do it; and vulnerable groups are affected differently in the various states. Such findings substantiate the need to look at solutions that are specific to different conditions in the states and that address several areas of the health system. The new areas of analysis show important differences and the particular vulnerability of the poor in: Overview * 7 Figure 0.3 Public and Private Sector Shares of Delivery of Selected Health Care Services in India, by Income Status of Patients, 1 995-96 Patient above poverty line Immunizations Prenatal care Institutional deliveries - U U U Hospitalization Outpatient care 0 1 0 20 30 40 50 60 70 80 90 100 Percent |1 Public sector * Private sector Patient below poverty line Immunizations Prenatal care Institutional deliveries Hospitalization Outpatient care I 1 T_TT__I_ 0 1 0 20 30 40 50 60 70 80 90 100 Percent |I Public sector M Private sector Source: National Sample Survey Organisation (1998); authors' calculations. 8 * Better Health Systems for India's Poor Table 0.1 Major Indian States, by Stage of Health Transition and Institutional Capacity INDIA'S STAGE OF TRANSIlTiON, POPULA1ON DEGREE OF CAPACfTY STATES (PERCENT) Middle to late transition, Kerala, Tamil Nadu 9.1 moderate to high capacity Early to middle transition, Maharashtra, Karnataka, Punjab, low to moderate capacity West Bengal, Andhra Pradesh, Gujarat, Haryana 39.1 Very early transition, Orissa, Rajasthan, Madhya Pradesh, very low to low capacity Uttar Pradesh 33.1 Special cases: instability, Assam, Bihar 13.3 high to very high mortality, civil conflict, poor governance Note: Major Indian states are those with a population of at least 15 million. The esti- mates were made before bifurcation, so Bihar includes the recently created state of Jharkhand, Madhya Pradesh includes Chatisgarh, and Uttar Pradesh includes Uttaranchal. * The behavior of the private market in health (chapter 6) * The prevalence of chronic disease risk factors (chapter 7) * The distribution of public and private health services (chapter 7) * The degree of financial protection in health (chapters 8 and 9) * The protection of patients' interests (chapter 9). Which Way Forward? The first four chapters of the report analyze the policy context and the choices for different activities, states, and programs. The focus is on strategies for four critical sets of activities selected from the larger set of functions of the health system. These strategies are summa- rized in the set of tables-"Options at a Glance"-at the end of this overview. Overview * 9 a. Health system oversight b. Public health service delivery c. Ambulatory curative services d. Inpatient care and health financing. Although changes will also be needed in the management of other inputs to the health system-such as human resource development, pharmaceuticals, technology development and assessment, anc. knowledge management-this report does not concentrate on them. But actions in those areas should support decisions made in the fou- critical areas of activity just listed. Oversight One of the main gaps in India's health system is in oversight. As a clear "public good" that benefits all persons, oversight is an area that requires public action. The analysis in this report demonstrates that increasing the measurability of health care providers and other actors is sorely needed. Example of activities that will increase meas- urability are collecting and analyzing information on price, type, and volume of cases and on service-quality indicators such as the use of technology and clinical outcomes. One of the most important con- siderations for the central government is whether it will spend mc re effort on overseeing the health sector, perhaps at the expense of directly managing public sector service delivery. The financial costs of strengthening oversight are small, and the benefits are considerable. Oversight activities are a prerequisite to improving quality, equity, and being able to use health insurance or strategic purchasing of health services. But public action is not smn- onymous with government action. One cannot assume that creating a government body with responsibilities and powers of oversight over the health sector will improve matters. Public oversight means empowering people who use the health system. Therefore, a wi der variety of potential actions to improve oversight are considered. 10 * Better Health Systems for India's Poor The growing prominence of the private sector makes the devel- opment of oversight by the public sector more urgent than ever. International and domestic experiences suggest, however, that rely- ing on only one approach to such oversight is unlikely to be suffi- cient. Considering the almost nonexistent formal collaboration to date between the public and private sectors and the public sector's limited enforcement of regulations, the most productive strategies are likely to be those that take a balanced approach. Each state ought to develop its own policies and strategies and then engage the pri- vate sector in developing a common agenda. Using independent organizations as honest brokers to measure performance may also be a better way to start than relying on government or provider organ- izations, because the problems of credibility and conflict may be larger at the beginning of these reforms. But such steps also depend on the development of trust between public and private sector players, which can also be enhanced through the involvement of professional organizations. Increasing the role of consumer advocates is another good approach to strengthening oversight, particularly to make health services more directly accountable to the public. Formal public sector regulation will still be needed, but the analysis suggests that these approaches have more potential to be more successful in regulating factor inputs to the health sector, such as drug quality control, than they do in regulating provider and hospital behaviors. In any case, given the limitations in public sector regulatory capacity, it is best to be selec- tive in starting actions to strengthen formal public regulation. States in the middle to later stages of the health transition and with moderate to high institutional capacity can use these strategies to institutionalize quality assurance procedures in inpatient and ambulatory services. States earlier in the transition and with less capacity may have to focus initial activities on expanding public information approaches to empower people to demand better health services-and on initiating steps to bring public and private sector actors together. As states gain experience and confidence, they can pursue more options. Overview * 11 Public Health Service Delivery Public health services considered here are the programs and institu- tions organized by society that directly protect, promote, and restore peoples' health through collective action. As with oversight activi- ties, these services are "public goods" or have significant benefits beyond the individual, giving strong justification for government involvement. Some of the main challenges facing public health are: * How can India best fight the HIV/AIDS (acquired immune defi- ciency syndrome) epidemic? * How can critical gaps be removed in planning, implementation, monitoring, technical leadership, and communications? * How can programs for existing public health priorities bz strengthened, while decentralizing responsibilities and resources to states? * When and how should effective programs be developed to address emerging public health priorities, such as the health con- sequences of smoking, mental illness, and injuries? Because India invests so little in public health, both in relative and absolute terms, the most relevant actions involve allocating more resources and effort on public health services. However, simply puzt- ting more money into public health will not be sufficient. Most of India is still in the early part of the health transition, sc it should redouble its efforts to attack the unfinished agenda as its highest public health priority. The success of public health progra.ns usually depends on more than funding the technical interventiois, as demonstrated in recent success stories in health in low-incomne countries. In the short term, intensive training and supervision in management, provision of supplies, and stronger and more profes- sional health education that strengthens people's ability to make healthy choices are important measures to make these progra.ms more effective. Continuing the current approaches across the coun- 12 * Better Health Systems for India's Poor try is unlikely to suffice, however, even with additional resources. The evidence from our studies makes a convincing case to involve private practitioners in these critical programs in most states because of their extensive reach and willingness to participate. Differences among states must be recognized and considered. Kerala and Tamil Nadu, being further along in the transition and having greater capacity, have more compelling reasons to focus on introducing and expanding public health services for cardiovascular disease, mental health, and injuries, as these are now the prominent conditions facing their populations. Other states are earlier in the transition but should also consider selective interventions in these areas. For example, Orissa and West Bengal have very high rates of tobacco use, particularly among their poor. They should consider putting more effort into public health campaigns against tobacco, and ensure that they are not using public funds to subsidize or encourage tobacco production and use. All states need to place additional effort on HIIV prevention because of the disease's potential for damage and the need for timely action, although the balance of interventions may be different from state to state because of differences in social conditions and transnis- sion patterns. The H[V epidemic and the emergence of other diseases have also exposed the weaknesses of the current public health system. In the medium to long term, investing in public health systems is a necessary strategy to provide the personnel and systems needed to carry out the current agenda and respond to future challenges. Ambulatory Curative Services The major challenges in service provision are fourfold: improving quality, increasing accountability, controlling costs, and promoting equity. These challenges raise several questions: * Can government effectively purchase services from the private sector? * Can performance and costs in the private and public sectors be properly monitored? Overview * 13 * Can government further extend coverage of health services to the poor by encouraging and perhaps formalizing pro-poor measures now being provided by the private sector? * Can government effectively monitor health gains by the poor? * How can India balance the dual challenges of expanding public spending on critical health services and improving regional equity in resource allocation? * How can the experience of stronger states be replicated in poorly performing states to improve health status and reduce polariza- tion? Some clear choices face governments in the area of ambulatory care. Economic analysis indicates that outpatient care is the area in which the rationale for direct government provision of services is weakest. Two countervailing facts, however, are that (a) many pre- ventive health programs depend on the opportunities and infra- structure used for outpatient curative care, and (b) outpatient care is the area in which the public has the most contact with the health sys- tem and where most private money is spent. Outpatient care is therefore a public concern, at least from an oversight perspective. The options considered here all involve relatively more resources than oversight and public health actions, although room may exist for better use of the large amount of private money already being spent on ambulatory care. Given the large presence of the private sector, all states clearly need to focus more attention on services provided privately. Sup- porting the oversight options can help to do this. But it also means delivering more credible ambulatory care services through both public and private sectors. The report explores many options for doing so. Influencing private practice by financing them and focusing Fub- lic delivery on areas where the private sector does not reach should be the most logical choices for most states. In states with low insti- tutional capacity, high infant mortality, and low coverage of basic 14 * Better Health Systems for India's Poor public health measures (such as for immunizations and nutrition), a reduction in publicly provided ambulatory curative care would allow a reallocation of effort into public health activities. The most expen- sive option is to try to revitalize the entire network of public facili- ties. Following such a strategy without a several-fold increase in resources may avoid difficult political decisions, but it would lead to little improvement in public services. The resources are spread too thinly, and efforts to improve availability and performance of staff and other resources have made slow progress. Even if a windfall of resources were to become available, strategies would still be needed to deal with the much larger private sector that is preferred by the majority of the Indians. Inpatient Care and Financing For inpatient care and financing, some of the same challenges iden- tified for ambulatory care are also relevant when considering qual- ity, costs, accountability, and equity. * How can India as a whole learn from its better-performing states and address regional equality in resource allocation? * How can the distribution of public expenditures be improved so that a greater fraction reaches the poor? * How will India take on the task of moving from out-of-pocket, fee-for-service financing to risk-pooling mechanisms in a weak regulatory environment? * How can public and private health facilities be made more responsive to the needs of their clients? * Can client perceptions of providers' manners and skills be incor- porated in training and supervision programs to make health providers more responsive? * Can governments intervene in areas that would make the biggest difference in motivating health workers, notably training oppor- tunities? Overview * 15 Some unique challenges also face this segment of the health sys- tem. Inpatient care is particularly expensive and risky. A "two-tier" system may emerge in which the poor are effectively left out of qual- ity care in the public sector (due to low access and quality) and the private sector (due to cost). Plans for care or insurance that attempt to provide the poor with exactly the same quality and quantity of inpatient care as the rich receive run the risk of cost escalation. Because of the close relationship between financing and high- cost-per-episode inpatient care, we consider sets of options for financing and provision of inpatient care together. The reforms pro- posed are not cheap, but the costs of not reforming are even greater. Given the high levels of private financing and debt currently caused by inpatient care, methods that are able to capture some of this funding while preventing the hardships of the current situation wilt have a greater chance of long-term success. India needs to set its sights on developing a more efficient ancl equitable health financing system. This means a financing system that has compulsory membership, a socially acceptable and afford- able package of benefits, pre-payment, and risk pooling for people with different incomes and health status. The main questions ar2 when and how to get there. Progress is likely to come from those states with the greatest administrative capacity and whose govern- ments are able to learn from the experimental and smaller-scale community financing schemes being tested by some NGOs. Big experiments are particularly needed to see how to cover the large informal sector, and to gain positive experience in managing health insurance. Relying on private health insurance is likely to play some role :n the short term, since the insurance market has been liberalized, ar.d this consumes little additional public resources. But the central gov- ermient will probably have to play a more aggressive role in reg- lating private insurance. International experience has shown that thle cost of not tightly regulating private voluntary health insurance is that regulating it later on will become more necessary because of escalating health costs and increased inequalities, yet more difficult 16 * Better Health Systems for India's Poor to do because private interests will have become more entrenched and practices more firmly established. Pursuing options for reform of hospital provision are comple- mentary to those oriented around financing. For states earlier in the transition and with less capacity, the most reasonable approach would be for governments to spend more effort on improving the quality of existing public first-referral hospitals, but to do so in a selective manner that promotes more equitable distribution of pub- lic resources. Experience has shown that such approaches can lead to improved quality of care and greater efficiencies, although only after several years. In conclusion, now is the time to conduct big experiments through- out India's health care system, particularly since the status quo is leading to a dead end. Either the central government or individual states could take the initiative, but new experience is needed to build on what has become an outmoded health system. The opportunity exists now for governments to reform the way they work-and to take on critical oversight functions. Governments also need to con- sider new public health services and how to implement them in ways that improve on current approaches. New ways of managing ambu- latory and inpatient services should also be tried, along with a con- certed effort to develop a more efficient, equitable, and risk-reduc- ing health financing system. The new experiments need flexibility if they are to deal with the highly varied conditions found across India, and they must include intensive monitoring and evaluation if lessons are to be learned. Clearly, no single choice is best for India. But the broadest opportu- nity exists now-to travel with eyes open, consciously deciding which fork in the road to take. Those who seek the right way must consult, analyze, debate, and forge a consensus on key issues and be willing to try new options. In the long run, this approach will best enable India's health system to meet the health needs of its people. Options at a Glance Table 0.2 Improving Health System Oversight ACTION DESCRIPTION 1. Develop partnerships with private sector to Build networks of private and public providers; seek cooperation on information, materials, share information, improve quality, and training; and use tools such as subsidies and contracting. These activities could have cooperate on service provision relatively low financial costs and offer the possibility of improving efficiency and increasing service coverage without adding staff to the government. Contracting offers the prospect of increasing accountability and creating efficiencies that could not be attained with direct government provision of services. 2. Support independent organizations to Provide an honest broker capable of better resisting unwonted pressures from the public and measure performance in public and private sector. The broker might also be more efficient. private sectors 3. Facilitate health-advocacy organizations Stimulate new organizations to become more active representatives of people's concerns in independent of both public providers and the health sector. Such organizations hold the promise of greater participation by people in private associations to work as consumer their health care, and could help raise accountability and performance of health services at advocates relatively low costs. 4. Support professional self-regulation Use professional bodies to more actively regulate practitioners through measures such as continuing medical education, accreditation of providers and facilities, and providing means for redress of patient complaints. Self-regulation is likely to be more attractive to service providers than regulation by the public sector. 5. Strengthen formal regulation of health Strengthen government agencies to regulate health inputs such as pharmaceuticals. Such inputs (drug quality) organizations exist but are underfinanced, poorly organized, and poorly functioning. 6. Formalize public sector regulation of private Create or strengthen government agencies responsible for the regulation of health care providers providers to set standards, license providers, receive complaints against providers, inspect premises, and generally take responsibility for bringing the private sector under public sector control. Table 0.3 Strengthening Public Health Services ACTiONS DESCRIPTION 1. Concentrate effort on programs for the "unfinished agenda" by: a. Increasing funding Allocate more resources to programs that combat the conditions of the unfinished agenda, namely diseases of childhood and maternity, malnutrition, tuberculosis, and malaria. and b. Increasing effectiveness Focus activities on supervision, monitoring results, increasing public accountability, using professional communications strategies, strengthening logistics systems, training, decentraliza- tion, improvement of public health systems, and partnerships with the private sector. 2. Initiate and strengthen programs in non- Accelerate preventive interventions against cardiovascular and tobacco-related illnesses and communicable diseases in states well develop more comprehensive programs to prevent injury and disability from major mental advanced in the epidemiological transition illness. 3. Reduce centrally sponsored schemes, turn Refocus functions of the central health ministry on developing policies and plans, allocating over the resources to the states, reassign funds strategically, and sharing experiences and technical expertise, while devolving functions of the central Ministry of Health management of programs and institutions to the states and autonomous institutions. Centrally and Family Welfare sponsored schemes might still be used for experimental activities (for example, the introduc- tion of a public health insurance program), for programs of national interest in which public and political awareness is insufficient to ensure adequate attention (HIV control), and for states experiencing special hardships. 4. Reinvest heavily in public health systems Invest in public health and health management training broadly, health information systems, generally disease surveillance, public health monitoring, and health promotion activities, as proposed by the Bajaj Commission (Bajaj 1996). Table 0.4 Strengthening Ambulatory Curative Services ACTION DESCRIPTION 1. Focus public resources strategically by a. Revitalizing the network of public sector Be selective and focus on raising the quality of public sector facilities in poor areas and facilities in disadvantaged areas where NGO and qualified private sector providers have not been forthcoming; do not try to cover the whole country or all rural areas with rigid input- and population-based norms. and b. Purchasing curative care from the private Move critical functions such as prenatal care and family planning to public procurement via sector when possible, limiting public private sector providers, at least where quality private sector providers exist. Subsidize care provision from primary health centers for the poor. (PHCs) and subcenters in those areas 2. Expand coverage for the poor through the Build on existing mechanisms whereby private providers provide, or claim to provide, use of demand-side mechanisms that give discounts to poorer customers. Offer direct reimbursement to certified private providers for the poorest access to publicly subsidized services provided to poor clients, preferably through mechanisms that allow clients a choice discounts at either public or certified among providers. private providers or 3. Revitalize the entire network of public Put money into public sector ambulatory care, notably care delivered at subcenters and sector facilities to raise the quality of PHCs across-the-board and make organizational and management changes to improve publicly provided services efficiency and public responsiveness. Table 0.5 Inpatient Care and Health Insurance ACIlONS DESCRIMON Financing-Oriented Alternatives: 1. Encourage multiple insurance pools. Use competing private and social insurance systems to reach all with compulsory purchose of Facilitate private insurance to pay private insurance, but with choices and public subsidies and equalization funds to cover the poorest. providers in a regulatory environment that Takes advantage of multiple revenue raising mechanisms; but depends on development of encourages voluntary employer- or quality assurance and information systems to distinguish good quality care from bad and union-based insurance schemes, with strong ability to monitor and regulate. public insurance schemes for the poor or 2. Initiate compulsory purchase of "single Develop publicly accountable universal health insuronce that raises revenues publicly and payer" insurance coverage maintains public and private provision and patient choice but also uses quality assurance and health information systems. Offers the ability to control costs, is easier to administer and regulate than other health insurance, and is a powerful way to align clinical practices with public priorities. or 3. Rely on the introduction of private Continue the present course, whereby the wealthy obtain private insurance and health costs voluntary health insurance and inequities grow. Richer households are using public hospitals less and relying on private hospitals more, a trend that is fueling demand for private health insurance targeted primarily to wealthier households. As richer households abandon the public system, political support for public financing erodes. This could lead to a vicious circle in which low and deteriorating quality in the public sector drives middle- and high-income households into the private market, undermining public hospitals. Provision-Oriented Alternatives: 4. Invest in quality of care, especially in Introducing quality improvements in underutilized public secondary hospitals increases their public sector secondary hospitals serving efficiency while providing services largely to the poor. This approach is popular among state rural areas, but increase cost recovery bureaucracies and politicians, so ownership is high and implementation relatively good. By with or without prepayment mechanisms providing a small but important amount of funds for improving quality, cost recovery at such to improve equality institufions becomes more politically and socially acceptable. 5. Make public hospitals autonomous, and Separates public sector "provision of insurance" from "provision of curative care." The fund their services publicly revenues for these autonomous hospitals could be based strictly on reimbursement for care provided via a public sector financing mechanism-but with no explicit insurance premium, as costs would be paid from general revenues. Would also require careful monitoring of quality and social mandates to provide care to the poor. PART 1 Raising the Sights for India's Health System 7- CHAPTER1 A Crossroads India's health system is at a crossroads. Its ability to fight infant mor- tality, communicable disease, and malnutrition is being stretched a: the same time that it faces emerging demands for better service ancl more attention to the chronic diseases of adulthood. India's under- funded public sector and its extensively used but largely unaccount- able private sector cannot hope to meet the country's enormous, growing, and shifting health needs. If India continues on its presen- path, the mismatch between its health system and its health problems will become only more severe. The present moment is a decisive one because the government of India is now seeking to define a better health system for the country, one that can take better advantage o§ the capacity of the private sector and deliver better service and out- comes for all regions and socioeconomic groups. The overall state of health in India has been improving-life expectancy at birth rose from 49 years in 1970 to 63 years in 1998. But its historic health problems-among them high infant mortality, child malnourishment and its associated diseases, and high fertil- ity-remain unresolved. For example, from the 1950s to 1990, infan- mortality was halved, to about 70 deaths per 1,000 live births; but: that rate is still too high and has moved little in the past decade. Also childhood malnutrition and maternal health problems are still wide- spread-nearly half of children under five years of age are malnour- ished, while anemia afflicts about three-fourths of children under 25 26 * Better Health Systems for India's Poor three and about half of women of reproductive age. And although India's national program for population control, launched in 1951, was one of the first in the world and met with some success, the 1999 fertibty rate of 3.3 remains higher than in most other Asian coun- tries. These problems-along with the communicable diseases of tuberculosis, malaria, and leprosy-constitute the unfinished agenda of the post-Independence health system. Now added to the unfinished health agenda is a vast new threat. The human immunodeficiency virus (BIV) and its product, the autoimmune deficiency syndrome (AIDS), is spreading fast in India. With 3.5 million Indians already living with HIM; and with the virus in some states having spread beyond the most susceptible groups, the epidemic threatens to erase much of the social, economic, and health gains India has made since 1947, much as the disease is already turning back the clock in Sub-Saharan Africa. This study finds that India's past success and current struggles together suggest new choices for its health system, both for govern- ment and for the private sector. * Where should government focus its efforts-on the poor? On selective diseases, programs, or functions? On developing health financing systems? * In what direction should the private sector be guided-toward outpatient care? Toward hospital care? Toward public service? Although progress has stalled, the very social, economic, and health gains India has realized in the past 50 years have helped pre- pare the ground for the shift in health conditions that is emerging in India-the incipient "health transition." The health transition encompasses three specific and interrelated shifts: (a) demo- graphic-a decline in rates of mortality and fertility and an aging of the population, (b) epidemiological-a change in the dominant pat- tern of disease, from malnutrition and the communicable diseases of childhood to the chronic diseases of adulthood, and (c) social-a gen- eral rise in knowledge and expectations of the health system and a A Crossroads * 27 greater ability to care for oneself. The health transition presents chal- lenges unfamiliar to India, to be sure. But their solution may point the way to a new and more successfil health system, one that can solve old problems while attending to the new. If India builds a health sys- tem that renews progress on the traditional health agenda, it must do so in a way that will also attend to the new health conditions and expectations that will come even faster with such success. Objectives of the Health System and the Health Transition A health system has three main objectives: * Improve the health status of the population by lowering mortalite and morbidity rates * Protect the population against the financial risks of health prob - lems * Respond to citizens' demands and needs. The relative importance of these objectives shifts as the health sta- tus and other conditions of a country improve and evolve. That shift constitutes the health transition that is now emerging in India even as it struggles with the stagnation of its health status. When mortality is high and morbidity widespread, the primar7 objective is to improve life expectancy and health status by reducing avoidable health losses from malnutrition and from communicable and readily prevented or treated diseases of maternity, birth, and childhood. As progress is made on these conditions and in the basiz infrastructure of public health services, as is happening in some areas of India, the three elements of the transition-demographic, epi- demiological, and social-will alter the types of care required of thz health system, raise the financial burden of disease, and increase thz sensitivity of the system to social demands. The demographic and epidemiological transitions: as rudimentary pub- lic health (water, sanitation, nutrition) improves and low-cost-per- 28 * Better Health Systems for India's Poor episode conditions (diarrhea, respiratory infections, malaria) are bet- ter treated, the population ages and an increasing fraction of the dis- ease burden and of health expenditures will come from high-cost- per-episode diseases. Besides presenting medical challenges, the new risk factors and diseases raise the financial vulnerability of the popu- lation and thereby challenge the health system's ability to protect individuals in that dimension as well. The social transition: as the severity of pre-transition health conditions diminishes, the responsiveness of the system to citizens' demands and needs comes into sharper focus. The quality of health services-both technical and perceived-tends to become increasingly important as basic health conditions are addressed and expectations are raised. Of particular concern in India is ensuring that, in pursuing each of the three objectives, the interests of the poor are protected. This concern does not mean that public health services ought to be made available only to the poor; in fact, the emerging "two-track" health care system, in which the poor can afford to use only the public sec- tor hospitals of lower quality while the rich can afford to choose the better-performning hospitals, which are usually in the private sector, may be the least effective way to protect the poor. Neither the pub- lic nor the private system can be considered acceptable until it addresses the large gaps in outcomes between the poor and rich on each of the three objectives. The Current Policy Context India's current health policy has its origins in the nation-building activities at the time of Independence and in the thinking embodied in the Bhore Committee report (Bhore, Amesur, and Banerjee 1946). The Bhore Committee focused on primary health care (at that time seen as simple curative and preventive care that could be provided in a clinic or home setting); it laid down the principle that access to pri- mary care is a basic right and thus not contingent on ability to pay-or on any other socioeconomic condition. The commission established A Crossroads * 29 primary health care as the foundation of the national health care sys- tem and developed the first system for primary health care facilities and health personnel in the public sector. Building on this thinking. India became a strong supporter of the Alma Ata Declaration of 1978. in which it committed itself to attaining "Health For All" on the basis of the primary health care approach. The Bhore Committee clearly modeled its vision for a public national health service on the one adopted by the United Kingdorr (Bhore, Amesur, and Banerjee 1946). The early planners focused or. a public national health service in part because the private sector involved with Western medicine was very small at that time. The vision was similar to the existing nationalization ideology of thc United Kingdom. The approach in India was unlike that of high-income former British colonies such as Canada and Australia. By the end of World War II, those countries had large private sec- tors; they were able to guarantee universal access to health services through more pluralistic health systems. Health policy in India has thus paid little attention to the private sector, though in recent years it has started to give it more consider- ation (see appendix C for a digest of national policy reports since Independence). By the time that India adopted its first forma'. national health policy, in 1983, the central government had recog- nized the need to "cooperate" with the private sector. Since then, however, efforts in that direction have been limited. The time has surely come to examine more specifically and seriously what can be done with the private health sector in India. Historic Vision and Current Realities Plans and policy options should be ambitious, but they should also be informed by the realities of the present. The vision of a universal, vertically integrated, publicly provided health care system has a utopian appeal. It holds out the promise of capturing synergies by linking preventive and promotional activities through ambulatory 30 * Better Health Systems for India's Poor and inpatient clinical care; of improving quality and generating cost savings from an integrated referral chain that prevents routine cases from being treated in high-cost facilities; and of providing universal and equal coverage for all. Indeed, after the Second World War, the comprehensive, univer- sal approach was pursued in many countries in Europe and in for- mer colonial areas of Africa and Asia. Over time, only countries that have had the political and social support for such systems have been able to maintain the high levels of public investment in health (more than 5 percent of gross domestic product [GDP]) necessary to sus- tain them. Despite relatively high funding, most European countries introduced market-oriented reforms in the 1980s and 1990s (Salt- man and Figueras 1998). Countries that have invested little in their health systems (less than 2 percent of GDP) have not been able to frilly deliver health services through the public sector, even if the rhetoric of a universal public provider has continued. In India, the state of Kerala proves that achieving good health outcomes, particularly low infant and child mortality rates, is possi- ble even at India's levels of economic development. But even in Ker- ala, the vision of complete public provision is not a reality-health spending in both the public and private sectors is greater there than in other states. Whereas Kerala may have better public clinics than most states, 69 percent of outpatient clinical visits are in the private sector. The distribution of public spending in Kerala is remarkably equitable, but the result is due in part to the fact that many of the richer households in Kerala use only private hospitals. In Kerala, as elsewhere in India, the emergence of a two-track health care system is of growing concern (see chapter 7). The experience of other states (such as Bihar and Uttar Pradesh) also reveals troubling realities that are not conducive to the rhetorical vision of a universal, vertically integrated, publicly provided system: * India's state governments have not provided the funds necessary to make that vision a reality. In contrast to other countries that have had more success with public health systems, India's fiscal effort has been much smaller. As chapter 8 shows, India's public A Crossroads * 3 1 spending on health is among the lowest in the world, whereas its proportion of private spending on health is among the highest. * The states have not demonstrated the capacity to provide curative health services of sufficiently high quality to attract users, despite lower direct costs to consumers. As shown in chapter 6, users pre- fer the private sector in its various forms for most curative care. * In most states a strong pro-rich bias exists in the distribution of the benefits of public curative care. Chapter 7 shows that it is not, uncommon to find a ratio of 3 rupees (Rs) spent on the richest 20 per- cent of the population for every Rsl spent on the poorest 20 percent. Endeavoring to correct present imbalances by allocating more public spending to the health sector has its own complexities. The political risk taking and bureaucratic commitment required to real- locate the funds from other uses has been too great to be attempted: any proposal for a significant expansion of the very low level of pub- lic spending on health would instead have to show how the addi- tional spending would be financed. The mandatory purchase of P~~~ 4.: Mh an i O B Mother~~~~~~i an chl (POORP BY CUR CANMR.H OL AKPoLBAY 32 * Better Health Systems for India's Poor insurance is one option, with the poor covered from public revenues; another option is the combination of mandated benefits and manda- tory insurance. Some cost recovery could be accomplished at the point of service, but the level of recovery would be significant only where it is least important-in those ambulatory care goods and services for which private providers are the most viable. These realities are not immutable, but one cannot simply assume that they will disappear. They can, however, be changed through institutional reforms and strong political commitment. Viable options for the health system must be technically feasible, consistent with public sector capacity, and capable of commanding sufficient social and political support to be sustainable. Focusing on Four Critical Activities in the Health System The focus in this report is on four critical sets of activities selected from within the larger set of functions of the health system: a. Oversight of the health system b. Delivery of public health services c. Delivery of ambulatory curative services d. Delivery and financing of inpatient care. Other health care functions-human resource development, phar- maceuticals, technology development and assessment, and knowledge management-must also be improved, but this report does not con- centrate on those functions. Options in those areas ought to support decisions made in the four critical areas of activity just listed. The Approach to Reform A number of pitfalls await health care reformers; among them are making ill-advised international comparisons, choosing a fixed A Crossroads * 33 model rather than clear goals that encourage experimentation, and striving for uniformity where variation is needed. International Comparisons International comparisons of health care systems, although useful. should be made with care (box 1.1). Debates about the latest policy changes in North America or Europe cannot directly inform the next steps for India, whose demographic and institutional realities are so different from those of high-income countries. Even if a health system with universal availability through a vertically inte- grated, public sector monopoly is the chosen vision-and whether such a vision is the right one is hardly clear-immediate expansion of the public sector is not likely to be the best way to implement it. Better understanding and use of current resources in the private sec- tor as well as in the public sector would be a more realistic approach. The Needfor Experimentation Because reforms need to emerge from continuing analysis, public discourse, and experimentation, this report does not propose a sin- gle vision for India's health sector. Indeed, any one of many differ- ent pathways could lead to improvements in health, in health-related financial security, and in the responsiveness of the health care sys- tem. More important than the attempt to define the best path in detail is the attempt to articulate the criteria that broadly define a better health system. A better health system would do the following: * Keep people well informed about the choices that affect their health and health care * Provide to everyone a minimum level of high-quality, high- impact health services-such as immunizations, safe-motherhood programs, communicable disease control, family planning, and first-referral clinical services * Pool health funds to reduce the financial risk from catastrophic illness 34 * Better Health Systems for India's Poor Box 1.1 Lessons on Health Reforn in Wealthy Countries The health svstenim; of many high-income countries have been reformed in the past few decades. The main motiva- tion of such reforms has often been runaway costs, and the main challenge has been to control costs vl'hilc maintaining or broadening access to high-q uali t care for all members of society. A number of the lessons learned are germane to the Indian agenda: * Access and infrastructure are easier to expand than to cut back. * The preferred stratepg for controlling costs is to empower consumers and tie payn1enus to services pro- vided to patients within the context of an o%-erall health budget. * Reforms in one part of the health sector, such as insur- ance mechanisms for selected services or populations or subsidies for certain services or drugs, require careful monitoring of access to services and of expenditures in other parts of the health sector. * The payment systems least successful in controlling costs or in promoting equity are those without a single paver or single set of rules, such as the fragnienited fee-for-serv- ice systems in the United States and India. * Carefully implemented strategies to provide insurance through full or partial risk sharing along with certain types of managed care have been best able to control costs without compromising quality and access. * Payment mechanisms used in high-income countries require sophisticated information systems and adminis- trative structures. Source: OECD (1992, 1994). A Crossroads * 3 5 * Maintain services at an affordable and socially sustainable level. In identifying options for India's future health system, we look at the full set of tools available to government, including oversight functions, financing, managing nonfinancial inputs, and service delivery options (table 1.1). The diversity of the Indian population and the large variations among the states are major themes of this report. We consider these differences with special concern for the most vulnerable segments of Indian society and with the understanding that recommendations for the whole of India or even for groups of states run the risk of oversimplification. Health is a shared responsibility of the central government and the states. The degree of discretion enjoyed by states in any aspect of health policy varies by state and by issue. In some public health activities, the central government will likely take the lead through centrally sponsored schemes; but for curative serv- ices, states already differ enormously in what they provide and in how well they provide it. Such differences, along with variations between the states in social expectations and in the political positions that leaders take, also affect the choices to be made at the state and national levels. International experience tells us that there is no single correct way to shape a country's health system. Within India, priorities and options should vary according to the conditions prevailing in the various states and districts. If the country approaches the task of reforming its health care system in the collaborative, wide-ranging manner of these studies and with vigilance for the welfare of the weakest segments of society, the health system may well become able to continually evolve in a manner that is accountable, equitable, and affordable for all Indians. Table 1.1 Examples of Health System Functions and Challenges in India FUNCTION EXAMPLES CHALLENGES Oversight Policy setting * National Health Policy (Government 1983) * How can health needs and interventions be prioritized? * National Population Policy (MOHFW 2000b) * How can policy intentions be translated into decisions on * State health policy allocation of resources? * How can the needs of vulnerable populations be addressed? * How can a meaningful framework for private sector participation in health be provided? Regulation and * Regulation of drug quality, private nursing homes * How can regulation be done positively to influence behavior of setting standards * Promoting quality assurance systems le.g., public and private providers? accreditation of hospitals, licensing of providers) * How can sanctions be enforced effectively? * Health insurance regulation * How can market failures of private health insurance be ameliorated? CN Providing * Subsidizing health providers to work in * How can subsidies be structured so they can be monitored? incentives remote areas * How can direct subsidies be directed toward public objectives? * Duty exemptions, free land for hospitals * Providing materials, drugs, and training to providers to follow good clinical practices Developing * Developing networks of providers who offer good * How can providers who contribute to national objectives be partnerships quality, who contribute to national programs distinguished from those who do not? * Training NGOs and for-profit providers in * What are the mechanisms to develop meaningful partnerships? national guidelines. * Shoring information on disease surveillance Providing infor- * Disclosing good- and poor-quality health * How can information to consumers, providers, financiers, and mation and providers and products government be used to improve accountability, quality, use, and advocacy * Communicating standards for care and costs of health services? pricing to the public and to industry * Should government buy or directly produce its information services? Monitoring and * Measuring use of health services by identified * How can the capacity to develop and monitor health system evaluation groups (e.g., poor, women, SC/ST) functions and outcomes be strengthened? * Measuring health outcomes * Who should conduct the monitoring and evaluation activities? * Monitoring efficiency of health services * How can monitoring results be fed into planning and budgeting? and programs Financing Raising revenues * User fees (out-of-pocket) * What is an adequate level of public funding? * Tax revenue and user charges * How can private spending on heath be better collected and used? * Insurance premiums Pooling resources * Health insurance (private for-profit, nonprofit * What should be the level of prepoyment? community financing, social insurance) * What should be in the package of benefits and who will provide the services? * How will the risks be pooled? * How will the poor be subsidized? Purchasing services * Contracting for clinical and nonclinical services * When and how should government contract for clinical and * Salaries and grants for public providers nonclinical services? and institutions * What is the best way to pay for publicly provided services? * Fee-for-service payments * Are there ways to move away from user fees (which raise health costs) and toward global budgets and capitation? Managing nonfinancial inputs Providing * Producing drugs and vaccines * Should government produce or purchase these services? drugs, health * Training medical professionals * Should government change the standards for these inputs (e.g., professionals, * Developing health information systems remove PHC inpatient beds) or the way it purchases them (e.g., capital, etc. * Building hospitals and clinics rational drug procurement)? * Whot types of health professionals and skills are needed? lTab!c continuc, cn thc fof lowing pogo. Table 1.1 (continued) FUNCTION EXAMPLES CHALLENGES Service provision Delivery of public * Communicable disease control programs * What is the best way to organize public programs? health services * Behavior change communications (e.g., AIDS * How should programs be decentralized to states and districts? prevention, family planning promotion) * How can programs be more equitable, efficient? * Should government buy or directly produce its information services? Ambulatory * Unqualified allopathic practitioners solo practice * How should public provision of ambulatory curative care be clinical care * Nonallopathic private practitioners continued? * Private solo practitioners * Can private nonallopathic and unqualified allopathic be used to * Public subcenters, PHCs, hospital outpatient provide health services of public priority to extend coverage? departments Inpatient care * Private nursing homes * Should public hospitals be organized to be more efficient? * Trust hospitals * How can catastrophic costs of hospitalization be ameliorated? * Public hospitals Note: We do not explicitly outline demand function interventions. Interventions in information and advocacy, regulation, and standard setting are largely intended to influence demand and correct for market failures. MOHFW is Ministry of Health and Family Welfare; SC/ST is scheduled castes and tribes; PHC is primary health center. CHAPTER 2 Problems of the Public and Private Sectors Before proceeding to policy options, we briefly sketch the structures and problems of the public and private health sectors in India. Exist- ing research on Indian health care is at its weakest with regard to the private sector; this study is the first to synthesize the literature on that sector. In broad terms, the public sector is vast, but it is sorely under- funded and not nearly large enough to meet the current health needs of the country. Moreover, it is overly centralized and rigid in its planning, politically manipulated, and poorly managed. The private sector is growing quickly but it is undirected and unregulated. It is without standards of care, is populated by many unqualified practi- tioners, and likely provides far too many inappropriate treatments. Whether seen in the public or private sector, patients finance much of their care out-of-pocket. Current Structure of the Public Sector The public sector has been organized largely to finance and deliver curative care, although it also implements a number of centrally sponsored programs for family welfare and disease control. These programs are almost exclusively delivered through an enormous array of underfunded public institutions.1 39 40 * Better Health Systems for India's Poor Internationally comparable data on manpower and facilities are weak, but what are available suggest that the number of medical per- sonnel and hospital beds in India's public sector-although huge in absolute numbers-is, in per capita terms, well below the comparable ratios in other low-income countries. Combining the data for public and private sectors, the per capita number of physicians in India is about average for low-income countries, whereas the ratios for nurses and midwives and for hospital beds are well below average (table 2.1).2 The rate of outpatient visits and hospitalizations are poor indicators of disease levels in a country, and differences in definitions and data col- lection methods between countries require that such data be treated with caution. The data do suggest, however, that hospital and outpa- tient utilization in India in the public and private sectors (total) is lower than in most countries, including low-income countries (table 2.2). Under the Indian Constitution, the responsibility for public health is shared by the central, state, and local levels of government, but the delivery of public sector health services is effectively a state responsibility. State and local governments account for about three- fourths of public spending on health, but states vary widely in the size of their health budgets (see chapter 8). Decentralization of state authority also varies widely by state. At the local level, only large cities have a significant financial authority. In some states, however, local bodies have a significant responsibility for managing services and implementing national or state government programs. Problems of the Public Sector As documented in many previous studies, the delivery of health serv- ices in India's public sector is rife with problems (World Bank 1995, 1996, 1997b, 2000c; Mukhopadhyay 1997). High levels of poverty lead to, and are exacerbated by, poor health conditions, and poor governance creates a weak environment for reform. The public sec- tor health system also suffers from poor management, low service quality, and weak finances. Weak management and the low quality Problems of the Public and Private Sectors * 41 Table 2.1 International Comparisons of Health Care Work Force and Hospita' Beds, 1990-98 (per 1,000 persons) COUNTRY PHYSICIANS NURSES MIDWIVES HOSPITAL BEDS India Public sector 0.2 - 0.2 0.4 Total 1.0 0.9 0.2 0.7 All countries, by income Low income 0.7 1.6 0.3 1.5 Middle income 1.8 1.9 0.6 4.3 High income 1.8 7.5 0.5 7.4 All 1.5 3.3 0.4 3.3 - Not available. Note: Data are the most recent available in the time period. Income is unweighted per capita GNP in 1999 U.S. dollars: Low income, less than $755; middle income, $756-$9,265; high income, more than $9,265. Source: World Development Indicators (World Bank 2000d), except for India: CBHI (vari- ous years) and MOHFW (2000a}; and nurse and midwife data (WHO 1999). Table 2.2 International Comparisons of Health Service Utilization and DALYs Lost, 1990-98 INPATIENT ADMISSIONS AVERAGE OUTPATIENT DALYs LOST PER CAPITA LENGTH OF VISITS (PER (PER 1,000 PER YEAR INPATIENT STAY CAPITA PERSONS COUNTRY (PERCENT) (DAYS) PER YEAR) PER YEAR) India Public sector 0.7 14 0.7 - Total 1.7 12 3.90 274 All countries, by income Low income 5 13 3 256b Middle income 10 11 5 256b High income 15 16 8 119 All 9 13 6 234 - Not available. Note: Data are the most recent available in the time period. DALYs, disability-adjusted life years. Income is unweighted per capita GNP in 1999 U.S. dollars: Low income, less than $755; middle income, $756-$9,265; high income, more than $9,265. a. Includes all visits to health providers, regardless of system of medicine. b. Estimated for low-income and middle-income countries combined. Source: World Development Indicotors (World Bank 2000d), except for India utilization data (Nationol Sample Survey Organisation 19981 and DALYs (WHO 19991. 42 * Better Health Systems for India's Poor of services are related problems that include structural and institu- tional issues as well as constraints on processes and skills. Public health management in India suffers from overly centralized and inflexible planning and control of resources; high levels of political interference in staff postings and transfers in some of the larger states; a failure to integrate programs devoted to family welfare, nutrition, and disease control and different levels of care; and the neglect of approaches that would encourage the private sector to meet public policy objectives. An example of inflexible planning is that staffing norms for auxil- iary nurse midwives are based on a standard population coverage, although birthrates vary widely across the country. As a result, the workload to deliver immunizations to children in high-fertility states like Uttar Pradesh and Bihar is more than double that in a low-fer- tility state like Tamil Nadu (Satia 1999). Managers have neither the authority nor the information neces- sary for accountable decisionmaking. Human resource systems offer little by way of monitoring, staff incentives, or in-service training, and the result is an undisciplined, poorly performing staff. An inap- propriate mix of skills is one of the most critical issues, as large num- bers of key posts remain vacant, particularly in rural areas. Accord- ing to the established staffing norms for existing subcenters, primary health centers, and community health centers, the shortfalls range from 17 percent for auxiliary nurse midwives, to 28 percent for doc- tors, to 47 percent for male multipurpose workers and nurse mid- wives (Ministry of Health and Family Welfare 2000a). The problems are not simply that the staff norms are too ambitious, that the selection of staff is inappropriate, and that too few health workers are being trained. The problems extend to insufficient pay in the public sector, particularly in comparison with the private sector, unsatisfactory living conditions in rural areas, and limnited professional opportunities. However, questions of staff motivation and incentives have not been well studied in the health sector and need a more sys- tematic assessment (chapter 5). The quality of health services is not well monitored in either the public or private sectors because meaningful standards and quality Problems of the Public and Private Sectors * 43 Patiently waiting at a health cliniC (PHOTOGRAPH By GEETANJALI CHoPRA/HEMANT MEHTA/ THE WORLD BANK) assurance systems are absent. Hence, little is known about clinical outcomes, clinical quality, management quality, or quality from the perspective of the user. Public sector health services are largely underutilized in rural areas-according to STEM (2000), bed occu- pancy rates of rural inpatient facilities in Uttar Pradesh are around 30 percent-and one reason is the perceived poor quality of service. The public sector is further constrained by staffing limitations, particularly in poor and remote areas that are also not served by the formnal private sector, and is more hampered by weaknesses in super- vision, maintenance, dr-ugs, and supplies. Despite the establishment of a large public network of health providers, public spending on health has stagnated at levels of around 1 percent of GDP, far below what is needed to provide basic health care to the population (World Bank 1997b; Mahal, Srivastava, and Sanan 2 00 1). The bulk of public spending on primary health care has been spread too thinly to be effective, while the referral linkages to 44 * Better Health Systems for India's Poor secondary care have also suffered (Tulasidhar 1996; Mukhopadhyay 1997). As in other countries, preventive and promotive health services take a back seat to curative care. Yet preventive care is almost exclu- sively provided through the public sector: an estimated 90 percent of immunizations and 60 percent of prenatal care is provided through the public sector (1IPS 2000; Background Paper 18). The states, which bear between 75 percent and 90 percent of the burden of public health spending, have their funds largely tied up in "nonplan" salary expen- ditures (Duggal 1997; Reddy and Selvaraju 1994). The disparity between rich and poor states is apparendy increasing, while expendi- tures are not reaching the implementing bodies, particularly the more geographically remote ones (Rao, Ramana, and Murthy 1997). Structure of the Private Sector In India, the private health sector is commonly understood to refer to private, for-profit, medically trained providers. Their range of prac- tice varies from solo practices and small nursing homes (inpatient facilities with usually less than 30 beds) to large corporate hospitals. However, the set of nongovernment actors involved is much broader and includes nonprofit entities and providers of Indian systems of medicine such as auyervedic and unani. Many untrained providers offer a combination of systems of medicine, although Western medi- cine (allopathy) tends to dominate. The private sector also offers ancillary services such as diagnostic centers, ambulance services, and pharmacies. In addition, a large number of private actors provide services or manage other inputs to the sector (construction compa- nies, consultancy firms). A few private companies and community organizations finance health services for their members, but overall the private sector's formal role in health financing has been limited. The number of studies concerning the private health sector has increased in recent years, but the present work is the first to attempt a systematic synthesis of the literature on the private health sector. Three Indian institutions have created a single database on private Problems of the Public and Private Sectors * 45 health sector studies in India; the studies point toward the rapid growth of private sector health provision, particularly by for-profit and nonqualified providers.3 Despite this growth, relatively few stud- ies have been conducted on the for-profit private sector in India. Studies on the nonprofit sector tend to have small coverage, and many have weak methodologies. Also, some key innovations that have been taking place in India in the last few years lack documenta- tion; these innovations have been in areas such as the contracting of services in the public sector; partnerships between public and private sectors; payment systems; and the use of subsidies. The data on health financing, also limited, suggest that, through- out the country, health financing is predominantly private and paic! out-of-pocket from individual consumers in a fragmentary way to many different types of service providers. The lack of a clear health policy framework and inadequate implementation mechanisms a+- national, state, and local levels toward private sector health are citec. in Background Papers 6-8 as major reasons why the research anc information base for planning and evaluation of the private sector ir. India is very limited. Although the most recent data are quite weak, they suggest that the private provision of health is growing rapidly and is the major source of outpatient and inpatient health care across India. At the time of Independence (1947), the private sector involved in allo- pathic medicine was quite small. Only about 8 percent of all medical institutions in the provinces were operated by private agencies, and another 5 percent in the nongovernment sector were receiving gov- ernment grants-in-aid (Bhore, Amesur, and Banerjee 1946). By 1995, government publications estimate that private hospitals represented more than two-thirds of all hospitals and nearly 40 percent of the hospital beds (see table 2.3). However, a census of private facilities undertaken in Andhra Pradesh in 1993 found that the actual number of hospitals was 3.8 times larger than the official number, and the actual number of hospital beds was 10.5 times larger. In more recent estimates, private hospitals represent 93 percent of all hospitals and 64 percent of all hospital beds nationwide. In addition to these allo- 46 * Better Health Systems for India's Poor Table 2.3 Health Care Work Force and Health Facilities in the Public and Private Sectors in India, Selected Years, 1981-98 INDICATOR AND MEASURE VALUE Doctors Total number (1998) (includes all systems) (CBHI) 1,109,853 Population per doctor 880 Percentage of doctors in rural areas (11981) (census) 41 Percentage of all doctors in private sector (estimated) 80-85 Nurses Total number (1996) 867,184 Population per nurse 976 Doctors per nurse (1996) 1.4 Hospitals Total number (1996) 15,097 Population per hospital 56,058 Percentage of hospitals in private sector 68 Estimated total number of hospitals 71,860 Estimated population per hospital 11,744 Estimated percentage of hospitals in private sector 93 Hospital Beds Total number (1996) (CBHI) 623,819 Population per hospital bed 1,357 Percentage of beds in rural areas 21 Percentage of beds in private sector 37 Estimated total number of beds 1,217,427 Estimated population per bed 693 Percentage of beds in private sector 64 PHCs Total number 22,975 Rural population per PHC 27,364 Note: PHCs, primary health centers. The estimate for manpower is based on medical coun- cil lists. The estimate for the number of hospitals and beds are based on the extent of underestimation in government (Central Bureau of Health Intelligence [CBHI]) data found in Andhra Pradesh in a 1993 census of all hospitals by the Director of Health Services and the Andhra Pradesh Vaidya Vidhan Parishad; they found 2,802 hospitals and 42,192 hos- pital beds in the private sector in Andhra Pradesh as against only 266 hospitals and 11 ,103 beds officially reported by the CBHI in that year. Thus, compared with the official (CBHI) data, the number of private hospitals was larger by a factor of 10.5, and the num- ber of beds by a factor of 3.8. Source: Estimates are by Duggal (2000) and Nandraj (Background Paper 6). Background Paper 6 drew on CBHI (various years), Census Commissioner of India (1981), and Ministry of Health and Family Welfare (2000a). Problems of the Public and Private Sectors * 4? pathic facilities, an estimated 2,800 hospitals (and 46,000 hospita: beds) are operating under the Indian systems of medicine, the vast majority of which are in the private sector. Data on the health workforce in the private sector are hard tc come by. Between 400,000 and 470,000 allopathic doctors were estimated to have been in practice in 1997 (Planning Commission 1998), with about 80-85 percent of them in the private sector (Dug- gal 2000). However, many doctors employed in the public sector also work in the private sector, with one study in Delhi showing 85 percent of public sector doctors also practicing in the private sector (Chawla 2000). Of the 120,000 doctors estimated to have been prac- ticing Indian systems of medicine in 1981, about 85 percent were in the private sector. Although information on the numbers of private doctors is limited, estimates about the numbers of other medical and paramedical professions in the private sector are not available. Estimating the number of informal providers is even more prob- lematic, since they are not registered, and many work part-time. Conservative estimates put the number of nonqualified rural medical practitioners at 1.25 million; almost all are solo practitioners located in outpatient settings (Rohde and Viswanathan 1995). A census in three districts in Andhra Pradesh found about one non-MBBS (Bachelor of Medicine and Bachelor of Surgery) doctor per 2,000 population (Rao, Ramana, and Murthy 1997), which would extrapo- late to about 500,000 nonqualified medical practitioners. A number of other studies have examined the role of traditional practitioners in its various dimensions (BAIF 1997; Kumar and Patel 1992; Chand 1988; Yesudian 1994). Broadly, these studies reveal that the majority of qualified solo practitioners practice in urban areas. Untrained practitioners, faith healers, traditional birth attendants, priests, and local medicine women and men largely cater to the rural areas. In rural as well as urban areas, however, the allopathic treattnent is the dominant type of care provided. Population surveys on the use of health services indicate an increas- ing use of health services through the private sector. Between the 42nd Round of the National Sample Survey in 1986-87 and the 52nd 48 * Better Health Systems for India's Poor Round in 1995-96 (NSSO 1992, 1998), the proportion of people using care outside the public sector increased (table 2.4). The vast majority of people both in urban and in rural areas (more than 80 per- cent) use the private sector for outpatient curative services as a first line of treatment. As already mentioned, the qualifications of practitioners and the systems of medicine used for outpatient care vary widely. Indigenous and folk practitioners, along with traditional providers, are particularly used as a first tine of outpatient treatment in rural areas (Rohde and Viswanathan 1995). For inpatient care, the majority of people are now using the private sector for hospitalization. States dif- fer a great deal in the extent to which their populations use private services as well as in the level of poverty and type of service provided. The majority of private hospitals are small (less than 30 beds); they are usually each owned by one person, a practicing doctor. Although some private hospitals are well known, very few of the pri- vate hospitals or private hospital beds are in the tertiary sector, com- prising roughly 1 percent of the total number of institutions, whereas charitable hospitals cater to about 4 percent of hospitalized patients (NSSO 1998). Although partnerships own a number of hos- pitals, relatively few are corporate, public limited, or trust hospitals. Most of the nursing homes are owned by a doctor entrepreneur and Table 2.4 Distribution of Outpatient and Inpatient Health Services across the Public and Private Sectors in India, 1986-87 and 1995-96 (percent) 1986-87 1995-96 TREATMENT OF AILING PERSONS RURAL URBAN RURAL URBAN Not treated 18 11 17 9 Treated as outpatients Public 26 28 19 20 Private 74 72 81 80 Treated as inpatients Public 60 60 44 43 Private 40 40 56 57 Source: National Sample Survey Organisation (1992, 1998). Problems of the Public and Private Sectors * 49 provide general curative medical and maternity services. The large private hospitals are either owned by trusts or are corporate enter- prises and offer more specialized services. In the typical staffing pattern, small hospitals have fewer than four physicians working at the hospital and depend on visiting consult- ants. For example, in Muraleedharan's study of private hospitals in Chennai (1999b), the consultant physician averaged just over three hours per day in a hospital and visited at least two different practice localities. Muraleedharan also found that about two-thirds of private hospitals had on average about two government doctors on their panels of consultants, averaging about two government doctors per private hospital. The presence of government doctors in private hos- pitals has been reported in other parts of the country as well, even in states where private practice by government doctors is prohibited. Problems with the Private Sector One of the main problems with the private health sector is that it has grown in an undirected fashion, with virtually no effective guidance on the location and scope of practice, and without effective stan- dards for quality of care or public disclosure on practices and pric- ing. Quality of health care in the private sector has become a major concern in the popular press. But few reports systematically examine the quality provided in the private sector; hence, generalizations about the care provided by such a large and heterogeneous private sector are difficult to make. The available studies are quite limited in scope. For example, a study in two districts of Maharashtra found a large number of prac- titioners practicing modern medicine without being qualified to do so. It also found several hospitals that were operating without any licenses or registration and did not have even the basic infrastructure and personnel to carry out their functions (Nandraj and Duggal 1996). More-recent studies of private medical hospitals in Calcutta and Bombay also indicate that private sector facilities are in poor 50 * Better Health Systems for India's Poor condition and are frequently used to perform medically unnecessary procedures (Nandraj, Khot, and Menon 1999). Studies that have examined provider behavior with respect to spe- cific diseases, such as tuberculosis and diarrhea, have documented significant deficiencies among both qualified and untrained practi- tioners (Balambal, Faggarajamma, and Rahman 1997; Bhandari 1992; Uplekar and Shepard 1991). Whether clinical management is better in the public sector is not clear, as these studies tend to com- pare private sector treatment behavior with standard treatment pro- tocols and not with how patients are actually treated in the public sector. Data on the performance of private hospitals are especially diffi- cult to obtain. Most hospitals, including many of the large hospitals, do not have patient records or information systems to report on per- formance. Of the studies that have examined work volume, Homan and Thankappan's study in Kerala (1999) showed that private city hospitals had higher occupancy rates than public hospitals. The available evidence also suggests that private hospitals provide more intensive and expensive services, using more x-rays and laboratory tests per patient than the public sector (Homan and Thankappan 1999). This finding may suggest good care or unnecessary expense; nonetheless, simple diagnostic tests such as x-rays and laboratory tests have been shown to be vastly underused in the public sector (World Bank 1997b). Other evidence of overutilization of procedures and diagnostic services in the private sector more clearly accounts for an increase in health care expenditures in the private sector (see box 2.1). An exam- ple is the very high rates of cesarean deliveries in a number of com- munities, often more than 40 percent (Pai and others 1999; Muraleedharan 1997; Kannan and others 1991). Although the appro- priate proportion of deliveries that should be done by cesarean section is widely debated, the rates found in India are far beyond what could be considered acceptable (WHO 1985). In numerous countries, the influence of the private sector as well as fee-for-service payments to doctors has been shown to be associated with increased rates of Problems of the Public and Private Sectors * 51 Box 2.1 A Proadive Public Health Provider Shyam is a Harijan, a member of the scheduled caste, living in southern Uttar Pradesh. When his 5-year-old son fell ill with vomiting and diarrhea, Shyam took him to an unquali- fied private practitioner in the nearest town. Shyam did not consult the nearby (public sector) primary health center, as he had heard from neighbors that it had no medicines. The boy received injections and medicines of substances unknown to his illiterate parents, but he failed to improve. He was then taken to a private nursing home, where he was admitted and given an intravenous solution (locally known as "bottles") for two days. The child recovered, but the total cost of his treatment was Rs 500. Meanwhile, the child's 12-year-old brother had devel- oped the same symptoms. This time the family consulted the government hospital. One of the health workers at the government hospital took the child to his residential private practice, where he gave two bottles for Rs 200 and medi- cines from the market, which cost another Rs 300. Very soon the third son, aged 14, also developed diarrhea and vomiting. All three children eventually recovered, but the family had spent more than Rs 1,500 on their children's medical expenses. All of the money was borrowed from neighbors and Shyam's employer. The Medical Officer in charge of Shyam's primary health center heard about the case and visited the village to investi- gate the outbreak. He took samples from the family's source of drinking water-an open water tank-and had them tested at a government laboratory. The tests showed bacterial con- tamination. The Medical Officer returned to the village to treat the water and counsel the villagers on the importance of boiling or treating potentially contaminated water. Source: World Bank (2000a). 52 * Better Health Systems for India's Poor cesarean sections (Cai and others 1998; Stafford 1990; De Regt and others 1986). Similar detrimental effects on medical practice arise from the marketing practices of the pharmaceutical industry. A num- ber of studies in India have pointed to medically inappropriate treat- ment by private providers who are linked to incentives provided by pharnaceutical companies and salesmen to increase sales of their products (Shah 1996; Thaver and others 1998; Phadke 1998; Green- halgh 1986). In this report, we build on these findings and examine in more detail how the private sector functions. Notes 1. As of 1999, the public infrastructure included about 137,000 subcenters, 28,000 dispensaries, 23,000 primary health centers (PHCs), 3,500 urban family welfare facilities, 3,000 community health centers (CHCs, which are 30-bed secondary hospitals), and an additional 12,000 secondary and tertiary hospitals. In rural areas, the public sector work force in 1999 included 29,000 doctors, 18,000 nurse midwives, 134,000 auxiliary nurse midwives, 73,000 male mul- tipurpose workers, 21,000 pharmacists, and 60,000 paramedics plus nontechnical workers (Ministry of Health and Family Welfare 2000a). Data on the size of the public sector workforce in urban areas are not available. 2. The number of hospital beds in the private sector in India is likely more than double the number recorded in government esti- mates, so the public-private total for India is therefore about 1 bed, instead of 0.7 bed, per 1,000 population and thus about two-thirds, instead of one-half, of the average number for low-income countries. 3. Background Papers 6-8, published together as "Private Health Sector in India: Review and Annotated Bibliography," are available elec- tronically at the following address: http://wblnO018.worldbank_org/ SAR/India/HealthESW/AR/cover.nsf/HomePage/1 ?OpenDocument. CHAPTER 3 Policy Actions for Critical Health System Activities This chapter outlines several feasible public actions (table 3.1) for improving the four critical health system activities-oversight of the health sector, public health services, ambulatory care, and inpatient care and financing. For each activity, we present the justification for public intervention. We also present the broad advantages and dis- advantages ("pros" and "cons") of each proposed action. Because of the variation in state conditions, not all options will have the same relevance or attraction to the different states. In chapter 4, we will examine how to put some of these options together, by assessing which actions are more important for states under different condi- tions and by examining programs that cut across the main activities discussed in this chapter. We provide our judgments of the possible pros and cons of each action as an aid to citizens, planners, and policymakers in addressing change. Some caveats are in order: The list of advantages and disad- vantages is of course not comprehensive; others might be realized instead. Moreover, advantages realized may not turn out to have been sufficient to justify the action, and disadvantages realized may be manageable. Our basic approach is to examine options for change, not for maintaining the status quo. Maintaining the health system in its present form will become untenable in India. The health transition is polarizing health conditions while bringing higher health care 53 54 * Better Health Systems for India's Poor Table 3.1 Summary of Actions for Critical Areas of Activity in the Health System AREA OF ACTIVITY ACTION Health system 1. Develop partnerships with private sector to share information, oversight improve quality, cooperate on service provision 2. Support independent organizations to measure performance in public and private sectors 3. Facilitate health-advocacy organizations that are independent both of public providers and private associations to work as consumer advocates 4. Support professional self-regulation 5. Strengthen formal regulation of health inputs (drug quality) 6. Formalize public sector regulation of private providers Public health 1. Concentrate efforts on programs for the "unfinished agenda" by: service delivery a. Increasing funding and b. Increasing effectiveness 2. Initiate and strengthen programs in noncommunicable diseases in states well advanced in the epidemiological transition 3. Reduce centrally sponsored schemes, turn over the resources to the states, reassign functions of central Ministry of Health and Family Welfare 4. Reinvest heavily in public health systems generally Ambulatory 1. Focus public resources strategically by: curative services a. Revitalizing the network of public facilities in disadvantaged areas and b. Purchasing curative care from the private sector when possible, limiting public provision from primary health centers (PHCs) and subcenters in those areas 2. Expand ambulatory curative care coverage for the poor through the use of demand-side mechanisms that give the poorest access to publicly subsidized discounts at either public or certified private providers or 3. Revitalize the entire network of public sector facilities to raise quality of publicly provided services Inpatient care 1. Encourage multiple insurance pools with strong regulation, using and health employer or union schemes, community financing, and public insurance insurance or 2. Initiate compulsory purchase of "single payer" insurance coverage or 3. Rely on the introduction of private voluntary health insurance 4. Invest in quality of care, especially in public sector secondary hospitals serving rural areas, but increase cost recovery with or without prepayment mechanisms to improve equality 5. Make public hospitals autonomous, and fund their services publicly Policy Actions for Critical Health System Activities * 55 costs, new technologies, and rising expectations. Under such condi- tions, low levels of public investment in health, overly centralized planning, inefficient public services, heavy reliance on private, out- of-pocket financing, little public oversight of the expanding private sector, and a loosely regulated health insurance market are not likely to be sustained. The demand for change is increasing, not diminish- ing. The question is whether the changes will involve tinkering at the margins or more substantial reforms; stopgap reactions or planned, publicly considered improvements. Health System Oversight Rationale for Public Sector Intervention Oversight, a key function of the health system, affects all other activ- ities. As a clear "public good" that benefits all individuals, oversight is an area that requires public action. But, as Amartya Sen has repeatedly emphasized, public action is not synonymous with gov- ernment action (Dreze and Sen 1995). In its half-century of Inde- pendence, India's experience with the "inspectorate raj" demon- strates the folly of assuming that creating a new government body with oversight powers and responsibilities will improve matters. At its best, public oversight means empowering the people who usc the health system. The goal is not to increase just the quality of services but also the accountability of the system. Health care is a complex and technical field, and health care providers will always have some infor- mation that their clients do not. Although empowerment does not mean learning to practice medicine, it does mean learning and having options. Any reforms will be far more effective in curtailing low-qual- ity, fraudulent, and extortionist behaviors if people have the informna- tion, tools, and options they need to make better health choices. Current and Emerging Policy Challenges The private sector is a major provider of curative health services. both ambulatory and inpatient. The most plausible future for the 56 * Better Health Systems for India's Poor health care system in India is that it will continue to be a mixed sys- tem of public and private institutions. Given that scenario, this study highlights a number of key policy challenges in the area of oversight (chapters that address them in more detail are in parentheses): * What formal and informal steps can government take to improve performance of the private sector? Can government create a pos- itive environment for the private sector as opposed to merely cre- ating an inspectorate raj? (chapter 6) * Can the quality and efficacy of health services be improved by increasing accountability for quality in both the private and the public sector? (chapter 7) * Can the performance of the public sector be improved? Can the respective roles and resources of the center and the states be redistributed? Are there ways of organizing public sector services to be more efficient, equitable, and accountable? (chapter 7) * Can consumer rights be promoted in ways that will boost public awareness of health and establish mechanisms for increasing the voice of the poor? (chapter 9) * Can new and more responsive mechanisms be put in place to pro- tect consumer interests and increase the social accountability of the health system? (chapter 9) Actions for Improving Oversight One of the most important considerations for government is whether it will spend more on overseeing the health sector, perhaps at the expense of public sector service delivery. One of the main gaps in India's health system, one that will not be closed by the private sector, is oversight. In particular, activities that will increase the measurability of health care providers and other actors in the health sector are sorely needed. These activities involve collecting and ana- lyzing information on price, type, and volume of cases, and on serv- ice-quality indicators, including use of technology and clinical out- Policy Actions for Critical Health System Activities * 57 comes. It is also critical that such information be made available to the public and other providers. Measurement is a prerequisite for greater public accountability, wider use of health insurance, and the strategic purchasing of health services. The potential benefits thus include more efficient health services and greater public accountability. One of the main difficul- ties in moving to a more measurable system is the uneven availabil- ity of technical expertise. Among the main problems to be managed in a more measurable health system are the risk of conflict with providers and the risk of manipulation of results. The remainder of this section considers various actions for improving oversight, with an emphasis on actions that improve measurability and accountabil- ity (table 3.1). Action 1: Develop partnerships with the private sector to build networks oJ private and public providers; to seek cooperation on information, materials, training; and to use tools such as subsidies and contracting. This action covers a wide range of options for dealing with the private sector (figure 3.1), and much international experience is available on how well the partnerships work (table 3.2). * Pro. These options offer the possibility of improving efficiency and increasing service coverage at relatively low cost and without adding staff to the government. Contracting offers the prospect of increasing accountability while creating efficiencies that could not be attained with direct government provision of services. * Con. As long as services are difficult to measure and social man- dates and other expectations are unclear, the ability to contract effectively and to use subsidies is limited. Monitoring contracts requires expertise, the availability of which has not been well tested. Without goodwill from the public and private sectors, partnerships would not work well under existing circumstances. Another consideration is that many private providers from both for-profit and nongovernmental organizations (NGOs) do not work together except in their capacity as visiting consultants. 58 * Better Health Systems for India's Poor Figure 3.1 Range of Interventions to Involve Private Practitioners Provide tools ond guidelines for private practitioners Outlaw management of Involve in program specific conditions or planning drugs in the private sector Mandatory internships Contract out in public programsprogram (e.g., TB control) implementation Tv~~~~~~~ Identify and punish Offer in-practice poor quality of care by education privote practitioners Offer recognition and Modify medical rewards for curricula positive contribution Source: Adopted from WHO (2001). Bringing groups of providers together into networks for patient referral and the sharing of clinical responsibilities and informa- tion could create conflicts that might make some partnerships unworkable. Action 2: Support independent organizations to measure performance among public and private providers. * Pro. Such organizations would be honest brokers and best able to resist unwanted pressures from the public and private sectors. The broker might also be more efficient. * Con. Maintaining independence would be difficult, and staff members could be confronted by personal threats. Action 3: Facilitate the creation of "health advocacy" organizations inde- pendent of both public providers and private associations that would work Policy Actions for Critical Health System Activities * 59 Table 3.2 International Examples of Varying Public-Private Mixes in the Delivery of Tuberculosis Care STRATEGY AND INTERVENTION EXAMPLE EXPERIENCE Public system (excludes private providers) Restrict antituberculosis drug sales Chile Successful Oman Successful Mandate referrals Syrian Arab Republic To be evaluated Parallel system (independent public and private delivery systems) Ignore Many countries Unsatisfactory Compete for and attract tuber- Morocco, Peru Successful culosis cases Collaborative system (includes private providers) Educate, inform providers Many countries To be evaluated (long-term aim) Collaborate in delivery 1. Public health support services to Netherlands Successful private providers Jamnagar (India) To be evaluated 2. Incentives to individual providers China Successful for performance of specific tasks 3. Private agencies responsible Chennai (India) Successful for delivery of care to defined Hyderabad (India) Successful populations Bangladesh Promising Manila Promising Haiti To be evaluated Source: WHO (2001). as consumer advocates. This action would involve stimulating new organizations to become active representatives of people's concerns in the health sector. * Pro. Such organizations hold the promise of greater participation by people in their health services and could help raise accounta- bility and performance of health services at relatively low costs. * Con. Few organizations currently have the capacity to advocate effectively, so that considerable time and effort would be needed 60 * Better Health Systems for India's Poor to develop them. Preventing health advocacy groups from becoming "captured interests" of government or other political groups might also prove difficult. Consumer groups would tend to represent the middle class and elites before taking on the inter- ests of the poor. Action 4: Support professional self-regulation. This action would involve professional bodies in more active self-regulation through measures such as continuing medical education, accreditation of providers and facilities, or providing means for redress of patient complaints (box 3.1). Box 3.1 A Quality Assurance Initiative in Mumbai A Health Care Accreditation Council was recently formed in Mumbai1 to help set quality standards for health care facilities. Uniquely, the council includes a range of stake- holders-representatives of private hospital owners, profes- sional bodies, consumer organizations, and NGOs. The council is being registered as a nonprofit body, with the initial funds for establishing the body raised from the founding members. The council is pursuing methods of financing for its continued work. Earlier attempts at setting up accreditation bodies for private sector health facilities have been unsuccessful, as they did not adequately involve key stakeholders. This initiative is an attempt to create a more positive environment for the private sector by involv- ing it more meaningfully with other stakeholders in a qual- ity assurance mechanism. The council is focusing on small private hospitals, devel- oping standards that will cover structural design, equipment, essential drugs, maintenance of medical records, and waste management. It is also addressing methods of rating facilities and how frequently to assess them. The council plans to develop standards for other health facilities in the near future. Policy Actions for Critical Health System Activities * 61 * Pro. Self-regulation is likely to be more attractive to service providers than is regulation by the public sector. * Con. The professional associations have done little self-regulation in the past and might not have the confidence of consumers or gov- ernment, who might see these bodies as representing their own interests. Their technical capacity for self-regulation is untested. Action 5: Strengthen fo rmalpublic sector regulation of health factor inputs. The inputs include factors such as drug quality control. Such regu- latory organizations exist in India, but they are underfinanced, poorly organized, and poorly functioning (box 3.2). Box 3.2 Oversight Constraints and Opportunities: Reform of the Drug Control System in Uttar Pradesh Drug control in Uttar Pradesh and many other states is generally known to be seriously defective. Substandard and fake drugs are on the market, and manufacturing and retail- ing establishments have poor standards. The harm from such a chaotic system can be as subtle as prolonged illness, medical complications, and higher rates of hospitalization, or they can be as dramatic as the 1974 Kanpur glucose scandal, in which contaminated glucose solution caused serious illness and death. The reasons for the problem are numerous: a frag- mented drug control organization with weak management, insufficient funds and material resources, poorly trained inspectors, inadequate sampling and laboratory capacity, and a reporting system that almost encourages corruption. By law, manufacturers and retailers should be inspected twice yearly, but many pharmacies have not seen an inspec- tor for 10 years. Currently, reports on manufacturers rarely go to the Drug Control Authority; manufacturers need deal only with local officials in their effort to continue the pro- duction of substandard or spurious drugs. 62 * Better Health Systems for India's Poor Box 3.2 (continued) More than 50 stakeholders of the drug control system (manufacturers, pharmacists, doctors, government adminis- trators, NGOs, and political organizations) were interviewed about a reformp program for the drug control system. The program would include establishment of a Drug Control Administration that would be independent of the Depart- ment of Health and have strengthened testing and inspection capabilities. More than 90 percent of those interviewed were in favor of reforms, but 80 percent also felt that they did not have the power to effect such reforms and that spurious drug manufacturers and corrupt traders and officials would con- tinue to pose major problems. About 43 percent felt that cor- ruption was widespread in the government bureaucracy, and 38 percent felt that corruption was widespread among drug manufacturers and wholesalers. The government of Uttar Pradesh is now testing reforms of its drug control system. The reforms include instituting a direct state-level authority over inspections and supervision, with instantaneous, transparent computer- ized reporting. At the same time, the government is pursu- ing an expansion of its laboratories to test drug quality along with the option of using accredited private laborato- ries. The scheme would be financed out of modest fees charged to industry-fees that are expected to be lower than the bribes currently paid. Sources: Dukes (2000); Basu (2000). Policy Actions for Critical Health System Activities * 63 * Pro. Tackles important areas, such as pharmaceuticals, where low quality can place many Indians at risk. Input markets are more measurable than provider behaviors and hence easier to regulate. * Con. The capacity of most states to regulate is quite weak, so thar there is a substantial risk of not being able to enforce regulations, particularly as more costs are involved. Dealing with wealthv industries adds significant potential for corruption. Action 6: Establish formal public sector regulation of private provider' tbrough legislation and legal enforcement. The private sector has many highly qualified and dedicated professionals but also many unquali- fied practitioners who pose a danger to public health.2 One approach to reform would be to create a government agency responsible for the regulation of health care providers. The agency would set stan- dards for, license, and receive complaints against providers; it would also inspect premises and otherwise generally take responsibility for bringing the private sector under public sector control. * Pro. The current practice of responding only to complaints is a weak mechanism for disciplining abuses, especially given the overall weakness and slow pace of resolution of legal disputes. * Con. First, in most states the capacity for regulation is quite weak. Second, to a large extent the expansion of the private sector has been facilitated by the general public's dissatisfaction with the qual- ity of services in the public sector; proponents of regulation would thus have to explain why the public sector would nevertheless have the capacity to regulate the private sector. Third, formal regulationl would run the risk of recreating an inspectorate raj in the health sector. As a result of the adversarial relationship thus created, more providers might be pushed outside the formal system of provision and regulation. Finally, experience in many other countries and with other activities demonstrates the risk of "regulatory capture," in which government regulators would form a coalition with exist- ing providers to limit new entrants into the industry and use qual- ity standards as a mechanism to limit competition and raise prices. 64 * Better Health Systems for India's Poor Public Health Service Delivery Rationale for Public Sector Intervention The rationale for public sector intervention in the delivery of public health services is strong. By definition, these services are organized in the public interest and take on issues of national concern such as family planning and HIV prevention. They address health problems with large externalities, particularly in communicable disease con- trol, immunization, and safe motherhood. Current and Emerging Policy Challenges Some of the main challenges facing the delivery of public health services are expressed in the following questions. * How can India best fight the HIV/AIDS epidemic, which has the real potential to undermine the health and development gains India has made since its Independence? (chapter 1) * How can critical gaps be removed in planning, implementation, monitoring, technical leadership, and public communications? (chapter 7) * When and how should effective programs be developed to address emerging public health priorities, such as the health con- sequences of smoking, mental illness, and injuries? (chapter 7) * How can programs for existing public health priorities be strengthened while responsibilities and resources are decentral- ized to the states? (chapter 7) Actions for Improving Public Health Services Action 1: Concentrate effort on programs to address the "unfinished agenda." This action would entail increasing the funding for, and effectiveness of, programs to combat the conditions of the unfinished agenda, namely, the diseases of childhood and maternity, malnutrition, tuber- culosis, and malaria. There is no single way to improve effectiveness, Policy Actions for Critical Health System Activities * 65 but the steps needed include increasing supervision and monitoring, invoking greater public involvement to increase accountability, enhancing communications strategies, and strengthening logistics systems and training. General efforts to improve public health capac- ity (action 4), greater decentralization of centrally sponsored schemes in some states (action 3) or greater partnerships with the private sec- tor would also improve effectiveness. Pro. The conditions of the unfinished agenda still cause th largest burden of disease in India, particularly among the poor and in the states and districts with high mortality. Highly effectiv and inexpensive technologies exist to combat many of these con- ditions (box 3.3), and programs already exist to combat these con- ditions-they needn't be invented, just improved. Box 3.3 Effective Technologies to Combat Communicable Diseases The following drugs and technologies are highly effective when used correctly: * Antibiotics for treating pneumonia are 90 percent effec- tive. Cost: Rs 15 per course * Oral rehydration for treating dehydration due to diar- rhea is highly effective. Cost: Rs 20 per dose * Measles vaccine is 85 percent effective in preventing measles. Cost: Rs 12 per vaccination * Tuberculosis medicines are 95 percent effective in curing TB. Cost: Less than Rs 500 for a six-month course * Antimalarials are 95 percent effective. Cost: Rs 6 per course * Bednets, by reducing mosquito-borne malaria, can reduce child deaths by 25 percent. Cost: Rs 200 each Plus: * Latex condoms are highly effective at preventing IIV Cost: Rs 650 for a year's supply Source: WHO (2000a). 66 * Better Health Systems for India's Poor * Con. These programs need more than money and improvements to be effective-they need to be reorganized. Although cost-effec- tive technologies exist, money spent on them will be wasted with- out a rise in political commitment, a more functional underlying delivery system, better management, and a greater capacity to innovate locally and develop partnerships across public and private sectors. This is more true among those states with weak capacity, where the need is greatest. Existing programs would also need to fill gaps in their design, such as targeting better nutrition during the weaning period and focusing on neonatal health care, which may be difficult to achieve. Action 2: Initiate and strengthen programs addressing injuries and non- communicable diseases in states well advanced in the epidemiological tran- sition. The large and growing burden of disease requires the acceler- ation of preventive interventions against cardiovascular and tobacco-related illnesses, along with more comprehensive programs to prevent mental illness and injury. * Pro. Noncommunicable diseases and injuries are already a major and growing part of India's disease burden, yet public health interventions to deal with them are still in their infancy. Because a long lead time exists between the development of interventions and the resulting improvements in the public's health, an early start on prevention is warranted for most such conditions. Pre- ventive interventions in these areas have been shown to prolong healthy life and reduce expensive hospitalizations. Data suggest that risk factors and disease are more prevalent among the poor than the rich (chapter 7). * Con. In most states, the addition of new programs would divert attention from the unfinished agenda and add complexity to the task of trying to integrate the numerous special programs. Given limited public health capacity and resources, efforts would need to focus on common diseases of childhood and maternity, tuber- culosis, and malaria control, which impose a larger burden on the Policy Actions for Critical Health System Activities * 67 poor. Implementing these programs in the traditional manner, as centrally sponsored schemes, might make it more difficult to undertake other reforms in the national Ministry of Health and Family Welfare and more difficult to organize programs in a state-specific manner. Action 3: Reduce centrally sponsored schemes and turn over the resources to the states. The central health ministry functions could be refo- cused on developing policies and plans, allocating funds, and shar- ing experiences and technical expertise, while management of pro- grams and institutions was devolved to the states and autonomous institutions, as proposed by the Committee on Restructuring the Ministry of Health and Family Welfare (Centre for Policy Research 1999). Centrally sponsored schemes might still be usec for experimental activities (for example, the introduction of a pub- lic health insurance program), for programs of national interest where public and political awareness is insufficient to ensure ade- quate attention (for example, HIV control), and for states experi- encing special hardships. * Pro. Most centrally sponsored schemes have outlived their pur- pose. Turning responsibility and resources over to the states would encourage adaptation and innovation in the states, which would be better able to integrate planning and implementatior. of these schemes with other programs at district and panchayat- (a local government administrative body) levels. * Con. Some states might not follow through on implementatior of national-priority programs even if the central governmen: retained the ability to monitor progress and influence resource allocation (through the Ministry of Health and Family Welfare and the Planning Commission). Some states might not have the technical capacity to design and implement their own programs, and the central government currently is not organized to provide the type of technical support needed. Resistance to change would be likely within the central bureaucracy, since change 68 * Better Health Systems for India's Poor would alter roles and reduce central control of resources and influence. Some states also might not want to take up additional responsibilities, as they can now blame insufficient or delayed resources, inappropriate planning and norms, and other inade- quacies on the national government (Centre for Policy Research 1999). Action 4: Reinvest heavily in public health infrastructure and systems. This action would require greater investment in public health and health management training, health information systems, disease surveil- lance, public health monitoring, and health promotion activities, as proposed in the Bajaj Commission Report (Bajaj 1996). • Pro. Revitalizing public health infrastructure, systems, and human skills would provide a stronger base for public health planning and implementation, areas that are not being filled by the private sector and that have been relatively neglected by the public sec- tor. In the medium to long term, this will be a necessary strategy for providing personnel and systems that are needed to carry out the public health agenda. * Con. With a constrained budget, greater spending on public health might mean lower funding for curative care facilities, a politically sensitive move. Also, some in the public sector would fight measures that empower public health practitioners working in the trenches, wanting instead to preserve planning and infor- mation systems that centralize power. For example, shifting the responsibility of the Directorate of Medical Services away from direct administration of hospitals and toward the development of public health policy, the dissemination of health information, and the monitoring public health would likely result in a smaller budget, less purchasing power, and at the least the perception of diminished authority and prestige. More generally, within the medical profession, public health has a lower status than other specialties, and increasing its role might cause friction within existing hierarchies. Policy Actions for Critical Health System Activities * 69 i i i SS14FA';- ilars r ~ ~ ~ ff1 rrr Going to work at a public hospital (PHOTOGPAPH By GEETANJALI CHoPRA/HEM.ANT MEHTA/THE WORLD BANK) Ambulatory Curative Services Rationale_for Pzblic Sector Intervention Visits for ambulatory curative care account for the great bulk of con- tacts with the health system, making this an area for public concern. The vast majority of such visits are to the private sector, which includes everything from large, sophisticated hospitals to individuals with no training who dispense dubious or even harmfuil remedies (chapters 2 and 6). The rationale for public sector intervention in ambulatory cura- tive services is weaker than in any other area of health care. The case is particularly weak for services that are not related to the treatmnent of childhood pneumonia or malaria or other such priority programs. Although some externalities exist for nearly any sort of curative care, they are generally not related to pubic health, and in most instances 70 * Better Health Systems for India's Poor the benefits are almost entirely private. Since ambulatory curative services have a low cost per episode, the issue of insurance does not loom large. With the increasing burden of chronic disease, however, ambulatory care will become more expensive. Thus, the public sector arguably needs to play a role in ambula- tory curative care because of its supply-side connections with other health activities. For example, curative visits for minor conditions could be used for preventive or promotional activities, and ambula- tory curative care visits are part of the chain of referral into more complex care. Credible public preventive services, moreover, may depend on the provision of quality outpatient and inpatient services. However, none of these arguments creates a compelling case for direct public sector production of these health services. A final argument might be that citizens are entitled to universal availability of a minimal level of curative services. However, such an entitlement is not fulfilled by the current system, nor is it clear why it would be better accomplished through direct production by the public sector versus "demand-side" financing that would allow indi- viduals to choose providers. Current and Emerging Policy Challenges The major challenges in service provision are threefold: improving quality, increasing accountability, and promoting equity. Some of the main questions brought out in studies of those challenges are as follows: * Can government effectively purchase services from the private sector? Can the necessary measurement of performance and costs in the private sector or public sector be created? (chapter 6) * How can government take advantage of opportunities to work with private providers? Are there opportunities for the public sec- tor to network with alternate private providers by overcoming legal constraints on untrained medical practice? (chapter 6) * Can government further extend coverage of health services to the poor by encouraging and perhaps formalizing pro-poor measures Policy Actions for Critical Health System Activities * 7 now being provided by the private sector? Can it effectively mon- itor health gains by the poor? (chapter 6) * How can India balance the dual challenges of expanding public spending on critical health services and improving regional equity in resource allocation? (chapter 8) * How can the experience of stronger states be replicated in poorly performing states to improve health status and reduce polariza- tion? (chapter 9) Actions for Improving Ambulatory Curative Care Action 1: Focus public resources strategically in two ways: a. Revitalizing public facilities in disadvantaged areas b. Purchasing curative care from the private sector when possible. la. Revitalizing the network ofpublic sectorfacilities-but only in disad- vantaged areas. Rather than trying to cover the country or all rural areas with rigid input- and population-based norms, government would focus on raising the quality of public sector facilities only where nongovernmental and qualified private sector providers have not been forthcoming. * Pro. This option offers some of the advantages of saving on scarce public sector management capacity while at the same time not withdrawing essential services for which no substitute exists. Under the current approach, in which the government tries to achieve universal coverage following rigid norms, the remote and rural areas already end up with low-quality care, while in urban and densely populated areas the public clinics simply compete, often unsuccessfully, with other providers. * Con. This option has some of the political disadvantages of any attempt to reallocate public sector staff. Stiff political resistance could be expected from health care workers who prefer living in urban areas. The option would be seen as a retreat from a com- mitment to provide universal basic care. 72 * Better Health Systems for India's Poor lb. Purchasing ambulatory curative care from the private sector where possible, and limit its provision from primary health centers (PHCs) and subcenters in those areas. This option would involve moving critical functions such as prenatal care and family planning to public pro- curement via private sector providers, at least where private sector providers exist and could be demonstrated to be of good quality. * Pro. The main benefit would lie in conserving scarce public sec- tor resources by closing or contracting out underutilized facili- ties. Because at present only a small fraction of curative care vis- its are made to government-run facilities, individuals would be able to find other sources of care in nearly all instances of such closures. Most important would be the gain in reorienting the attention and efforts of existing personnel to core public health functions, freeing them from responsibility for curative care while taking advantage of the large private sector in ambulatory care. * Con. The major drawback would be that in some areas-particu- larly in remote and rural regions-it might be difficult to identify private providers of good quality. Curtailing ambulatory curative care would obviously be a politically unpopular option, one likely to face stiff opposition from public health workers. Action 2: Expand ambulatory curative care coverage for the poor through the use of "demand-side" mechanisms that give publicly subsidized discounts to the poorest patients at either public or certified private providers. Chap- ter 6 shows many private sector providers claim to give a variety of discounts to poorer customers. One way to ensure access for the poor is to build on such ad hoc mechanisms by allowing some direct reimbursement to private providers for services provided to poor clients, preferably through mechanisms that allow clients a choice among providers. * Pro. First, reimbursement schemes provide a voluntary linkage between public and private providers. To be eligible for reim- bursement, private providers would have their quality of care cer- Policy Actions for Critical Health System Activities * 73 tified. Second, reimbursement would help redirect poorer house- holds from the lowest-quality providers (untrained private providers) toward higher-quality providers by reducing the costs of the higher-quality private providers. Third, in areas having a large number of certified providers, the public sector could move away from provision altogether, allowing it to focus direct public provision on areas where the poor now have few or no health care options. Con. The reimbursement option would require the public sector to develop significantly greater administrative capacity to link with the private sector. The public sector would have to be able to assess the quality of private sector providers, create criteria for judging which households were eligible for subsidy and for deter- mining which services were eligible for reimbursement (and at what rate), and avoid corruption (in the form of collusion to over- report services provided).3 Another major question is how the nationally sponsored programs would interact with private and alternative providers in the various states. In some states, centrally sponsored schemes would be carried out through the system of public clinics, whereas in another the same activities might be contracted out to private sector providers if the state had chosen to eliminate universal coverage of facilities. Action 3: Invest in raising the quality of the entire network ofpublicly pro- vided ambulatory services, reorganizingfor greater efficiency. This option would mean putting money into public sector ambulatory care, notably care delivered at public health centers and subcenters across the board. Because "more money" is not the solution, revitalization would require organizational and management changes to improve efficiency and public responsiveness. * Pro. The benefit would be better care in the health system. If the quality of public facilities were higher and costs were lower than in the private sector, then presumably people would be attracted to the public alternative. In turn, if the public facilities were 74 * Better Health Systems for India's Poor attracting more visitors, they could better carry out their preven- tive and promotional functions. Con. Revitalizing public health facilities would be enormously expensive. First, since the public sector currently accounts for only 20 percent of visits, enormous increases in facilities or in uti- lization rates would be required to increase the frequency and proportion of contacts between the poor and the public sector. Second, although money is not the only problem, substantial investments would almost certainly be required to upgrade the quality of care sufficiently to enable public centers to attract new patients. Perhaps even more important, such an effort would attract scarce public sector managerial capacity and human resources away from other parts of the health system: public health, quality assurance, and whatever inpatient role the public sector will play. This strategy would do nothing about the main source of outpatient care provision in the public sector. Inpatient Curative Services and Health Insurance Rationale for Public Sector Intervention Even though inpatient curative care has linkages with oversight, public health services, and ambulatory curative care, it is possible to consider its policy options separately. Treatments having a high cost per episode make the expansion of inpatient care dependent upon some mechanism for pooling risks and thereby redistributing finan- cial costs. In most countries, governments have tried to spread risks and subsidize the poor, though the methods tried have been quite different (box 3.4). The need to reduce the high levels of risk of financial ruin from serious illness makes inpatient care and health insurance an issue of public concern. Another public sector justification is to ensure that hospitalizations are equitably distributed. The expectation that risk pooling can help to strengthen quality assurance, public accounta- Policy Actions for Critical Health System Activities * 75 Box 3.4 International Approaches to Spreading Risks and Subsidizing the Poor To make health financing fair and efficient, countries have gradually developed a variety of systems to pool risks. By pooling risks, the healthy subsidize those who are sick, and the rich subsidize those who are poor. In the examples shown below, cross-subsidies based on risk (a person's level of illness) and income can occur among members of the same pool or through government subsidies to single or multiple pool arrangements. All systems involve prepay- ment and a separation of contributions from the use of health services. FINANCING RISK SUBSIDIZING COUNTRY SYSTEM POOLING OF THE POOR Colombia Multiple pools: Intrapool via non- Intropool and competing social risk-related contri- interpool: salary- security organi- butions; interpool related contribu- zations, municipal via a central risk tions plus explicit health systems, equalization fund. subsidy paid to the Ministry of Health Minimum benefits insurer for the poor packages are to join social mandatory for all security; supply- members of all side subsidy via pools. no-charge services provided by Ministry of Health and municipal health services Netherlands Multiple pools: Intrapool via non- Risk equalization largely competing risk-related contri- fund, excluding the private social butions; interpool rich insurance organi- via central risk zations equalization fund. Korea, Two main pools: Intrapool via non- Salary-related con- Republic of national health risk-related contri- tribution plus sup- insurance (covers bution. A single ply-side subsidy of up to 30% of a benefit package Ministry of Health member's health for all members. services; notional expendituresl, health insurance Ministry of Health from government allocations. 76 * Better Health Systems for India's Poor Box 3.4 (conlinued) FINANCING RISK SUBSIDIZING COUNTRY SYSTEM POOUNG OF THE POOR Public subsidy for insurance for the poor and farmers. Zambia Single formal Intropool, implied Intropool via gener- pool: Ministry of single benefit al taxation. A sup- Health/Central package for all ply-side subsidy via Board of Health in the system. no-charge services provided by Ministry of Health. Canada Single pool in Intrapool, Intrapool via each province single-benefit general taxation. (with portability package for across provinces) all citizens. Source: WEHO (2000c). bility, and purchasing of health services adds to the justification for public involvement in health insurance. The market for health insurance is especially prone to the prob- lem of "adverse selection": if an insurer offers a policy to all who wish to buy it, then those who elect to buy the insurance will be those who think they have more than the average risk. The resulting loss experience of the insurers will cause them to price the policies "too high" (worse than actuarially fair) for a person of typical risk. For insurers to remain financially viable, they must either force con- sumers to buy as a "pool" (for example, everyone in a given area, everyone in a given occupation, everyone working for the same firm), screen potential buyers, or price their policies "too high." Governments often address the insurance problem by providing hospital care at less than full cost. As chapter 9 shows, even with public hospitals, many consumers are forced into debt and poverty by the costs of hospitalization. Moreover, chapter 8 shows that increased cost recovery in public hospitals cannot cover their budget Policy Actions for Critical Health System Activities * 77 without imposing even greater financial burdens on consumers. But nearly everywhere in the developing world, and in most states of India, the low-cost provision of public hospital care causes the ber- efits of public expenditures to flow mainly to richer households (chapter 7). Current and Emerging Policy Challenges Some of the challenges identified for ambulatory care are also rele- vant when considering questions of quality, accountability, and equity for inpatient care-and are particularly relevant when cor.- sidering how India as a whole can learn from its better-performing states and address regional equality in resource allocation. Sorne specific challenges also face this segment of the health system. -A "two-tier" system may emerge in which the poor are effectively left out of quality care in the public sector because of low access and low quality and out of the private sector, as well, because of cost. Plans for care or insurance that attempt to provide the poor with exactly the same quality and quantity of inpatient care as the rich receive run the risk of cost escalation. The questions raised in stud- ies of such challenges include: * How can public and private health facilities be made mo-e responsive to the needs of their clients? Can client perceptions of providers' manners and skills be incorporated into training ar.d supervision programs to make health providers more responsive? (chapter 6) * Can governments intervene in areas that would make the biggest difference in motivating health workers, notably training oppor- tunities? (chapter 6) * How can the distribution of public expenditures be improved 3o that a greater fraction reaches the poor? (chapter 7) * How will India take on the task of migrating from out-of-pocket, fee-for-service financing to risk-pooling mechanisms in a weak regulatory environment? (chapter 8) 78 * Better Health Systems for India's Poor * How can India strengthen its financing systems to reduce the large financial risks faced by Indians, particularly the poor, when they become ill? (chapter 9) Actions for Inpatient Care and Health Insurance Action 1: Encourage multiple insurance pools. Facilitate the creation of private insurance to pay private providers in a regulatory environ- ment that encourages replication of voluntary employer-based or union-based insurance schemes such as the Self-Employed Women's Association (boxes 3.5 and 8.3). Other community financing and public insurance could also be added to increase cov- erage. The ultimate objective here would be to reach compulsory purchase of insurance, but with choices for all and public subsidies for the poorest. * Pro. An incremental approach that promises universal coverage might be technically feasible and politically popular. This option would preserve the patient's ability to choose insurer and provider and potentially reduce patient hospital costs markedly and lessen inequity. * Con. For this option to work, information systems, quality assurance procedures, administrative systems, and appropriate marketing strategies would need to be developed, all of which would require investments that the government has not made and capacities that it has not had. The costs would likely be substantial. Explicit decisions about the minimum package of benefits, amounts of public funding, and level of cross-subsi- dization might expose the Ministry of Health to additional political risks and strain its capacity. From the beginning, increased attention would be needed to get the regulatory framework right lest entrenched interests block subsequent reform. India has had little experience with large-scale non- profit health insurance, and government has little capacity to develop or administer its own public insurance. Partitioning of Policy Actions for Critical Health System Activities * 79 the risk pools along employment lines would leave government to cover those with the greatest risks and least resources, thus requiring more complicated subsidy schemes among insurance plans. Box 3.5 Health Insurance for the Informal Sector: The SEWA Experience in Gujarat The Self-Employed Women's Association (SEWA), a trade union of more than 2 million women in the informal sector, is providing health insurance to its members as part of an integrated insurance scheme. The union developed the plan after SEWA members identified illness-their own and that of family members-as the key stress in their lives and the major cause of indebtedness. The SEWA health insurance scheme functions in co- ordination with the Life Insurance Company of India and the new India Assurance Company. Members pay an annual premium and receive coverage for maternity benefits, hos- pitalization for a wide range of diseases, occupation-related illness, and diseases specific to women. Several problems emerged in the implementation of the plan (administrative snags, the rejection of valid claims, dis- honest claims), but SEWA members are joining the scheme in increasing numbers, setting aside their limited earnings well in advance to pay for the annual premium. SEWA mem- bers have no expectation of "free" insurance or subsidies. The major benefit of the insurance scheme has been to provide security to poor women and their families in times of crisis. In addition, it has enhanced health-seeking behav- ior and has helped workers to increase their savings and plan for the future. 80 * Better Health Systems for India's Poor Box 3.5 (continued) Insurance and the concept of risk pooling was initially an unknown concept, but SEWA members were quick learn- ers. Women barely familiar with the written word quickly learned some key procedures such as the preservation of bills, certificates, medical cards, and so on. "Normally, when any of us in the family is sick or hospi- talized, it is a dark time for us," said a SEWA member. "This time, when I was hospitalized for cerebral malaria, my family was worried but we did not go into debt forever. I had paid my premium so we recovered most of our costs. Who would have dreamed it was possible?" Action 2: Move toward "single payer" insurance with compulsory purchase of insurance that would reimburse public and private providers neutrally. * Pro. Universal coverage, the ability to control costs, ease of administration compared to other insurance, and ease of regula- tion compared to other health insurance-all of these are advan- tages. The collectivist philosophy behind single-payer insurance and the ability to maintain private provision and patient choice of providers might resonate well in India. This type of financing could strengthen efforts to distinguish good-quality providers from bad and could provide a way to influence clinical practices in the direction of public priorities. * Con. Health systems in India are not ready for compulsory single- payer insurance because public and private providers lack the infor- mation systems, quality assurance procedures, and administrative systems to operate such a system. Its initial costs would therefore be very high, depending on the benefit package proposed. The public might lack confidence in its administration and be unwilling to pay additionally for it. The rich might prefer private health Policy Actions for Critical Health System Activities a 8: insurance to subsidizing the poor, and the healthy might not wan: to contribute if they did not benefit. Introduction would require political leadership, yet the political risks of failure might be quite high. The problem of controlling costs in a predominantly fee-for- service system has yet to be addressed in India. Action 3: Rely on the introduction of private, voluntary health insuranct. Richer households are using public hospitals less and relying mor on private hospitals, a trend that is fueling demand for private healta insurance targeted primarily to wealthier households. As richer households abandon the public system, political support for public financing erodes. This could lead to a vicious circle in which low and deteriorating quality in the public sector drives middle- and higher- income households into the private market, which would then fur- ther undermine fiscal support of public hospitals. Such a develop- ment would effectively exclude the poor from quality care, either Ly pricing them out of the private market or by deteriorating quality and weakening support for public services. * Pro. The benefit of private, voluntary health insurance is that by essentially ignoring the development of private hospitals the pub- lic sector would be able to focus resources on basic public health. This is also the path of least change for the existing bureaucracy and directorate, one that would avoid opposition from that quarter. * Con. Inequalities would grow, and costs would escalate, because the insurance that develops in the private sector for richer clier.ts would be impossibly expensive to generalize. The presence ol: a private market in one regulatory environment would cre-ete entrenched interests that resist reform. Action 4: Strengthen quality of care especially in public secondary hospitals in rural areas but increase cost recovery, with or without prepayment mech- anisms to improve equality. In addition to the financing-based options discussed above, hospital systems and management can also be strengthened in a complementary way. 82 * Better Health Systems for India's Poor * Pro. The existing rural secondary hospitals are underutilized. Experience shows that improving the quality of these public hos- pitals increases their efficiency while they continue largely to pro- vide services to the poor (Institute of Health Systems 2000). This approach is popular among state bureaucracies and politicians, so ownership is high and implementation relatively good. Cost recovery at such institutions-when accompanied by improved quality-has provided modest but important amounts of money for facility operations, making them more politically and socially acceptable. * Con. Improving quality in rural secondary hospitals can be expensive and requires continued public financial commitment to operation and maintenance-as well as diligence to ensure that appropriate staff are placed at such facilities. Without such efforts, these approaches could not be sustained. Departments of health have no clear advantage in directly administering hospitals, and in the long run the public agencies might be better off supporting such institu- tions financially but having them managed by private professional groups under the guidance of a public board. User-fee exemptions for the poor are in place, but it is not known how much of a barrier to care such fees have been to the very poor. Prepayment mecha- nisms hold a promise of reducing catastrophic costs, but they have not yet been tested. Action 5: Make public hospitals autonomous and fund their services pub- licly. This type of institutional reform would separate public sector "provision of insurance" from "provision of curative care." The rev- enues for these autonomous hospitals could be based strictly on reimbursement for care provided through a public sector financing mechanism-but with no explicit insurance premium, as costs would be paid from general revenue. * Pro. Autonomous organization would offer public hospitals the opportunity to be managed more professionally. More explicit Policy Actions for Critical Health System Activities * 83 purchasing of services could make hospitals more accountable anJ improve their performance. Con. Without changes in the legal environment, the freedom to dis- charge low-performing workers, more measurable social mandates (such as free care for the poor), and sufficient budgets, making pub- lic hospitals autonomous might not achieve the savings, efficiencies, or social mandates expected of them. Because governments woulc be unlikely to allow public hospitals to be closed, market exposure of these institutions would be limited, and an element of "moral hazard" (insulation from the consequences of poor or irresponsible performance) for these hospitals would likely remain. Professional hospital management has not been well developed in India, so that expected management improvements might not be achievec. Bureaucratic opposition to relinquishing control over hospit1 administration might make such reforms difficult. Opposition frora labor unions and doctors could also be anticipated, particularly if the loss of employment or benefits appeared likely. Concluding Remarks This chapter outlined several feasible actions for improving four critical activities of the health sector: oversight of the sector, publ.c health services, ambulatory care, and inpatient services and health insurance. Although we presented the broad advantages and disad- vantages of each option, we did not analyze them in terms of their applicability to individual states or health care programs. In chapt-r 4, we examine how to put some of these options together by assess- ing which actions are more important for states under different co'1- ditions and by examining programs that cut across the main actiii- ties discussed in this chapter. To stimulate debate, we propose which of the measures ought to be taken forward in the short and medium terms and outline some of the main policy questions needing further investigation. 84 * Better Health Systems for India's Poor Notes 1. S. Nandraj, personal communication, 2001. 2. The unqualified offer treatments that are ineffectual or con- traindicated and, by unmonitored overuse of medicines such as antibiotics, further harm public health by increasing drug resistance (Background Papers 6 and 8). 3. However, all of these are skills that the public sector would need to develop in any case if it were to be involved in regulating providers and insurance. W CHAPTER 4 sAi:L Putting It Together: Raising the Sights of India's Health System We now look at ways to pull together the various choices addressed in the preceding chapter. First, we examine the choices facing dif- ferent states. Second, using maternal health care as an example, w2 see how different visions for the health system affect health services that cut across the sets of activities discussed in the previous chapter. We then offer some specific policy and operational recommenda- tions for ways to think about what short- and medium-term steFs can be taken. We conclude by looking how to take forward the po:- icy and research agenda. Health system reform should generally improve outcomes in an efficient, equitable, accountable, and sustainable manner. On thIs basis, we propose in this chapter some broad principles for the types of the reform needed in India's health sector. Government ought to play a leading role in reforming the health sector by raising its sights in four ways: 1. Oversee the needs of the entire population by making the health system more pro-poor, gender sensitive, and client friendly ar.d respond to the high burden of preventable diseases borne by the poor, scheduled tribes and castes, and women. 2. Look forward, by preparing for the challenges already posed by the health transition-a shift in the burden of disease, a rise in 85 86 * Better Health Systems for India's Poor the costs of treatment, and the call to develop a health financing system. 3. Remove blind spots by grappling with the challenges of the devel- opment of the private sector so the health system can be consid- ered in its entirety. 4. Focus on improving the quality, efficiency, and accountability of health services, both in the public and private sectors. Different Choices for Different Parts of India Overall, in our review a recurrent pattern has emerged in which southern and western states tend to have better health outcomes, higher spending on health, greater use of health services, and more equitable distribution of services than other parts of India, particu- larly the poor north-central states. A summary of health outcomes for the major Indian states is in table 4.1. Large differences in health service outputs also exist among the major states (table 4.2). In ranking outputs, we considered higher coverage of prenatal care, higher rates of institutional deliveries, and full immunization as desirable. On the other hand, a higher rate of hospitalization is not necessarily desirable or appropriate.1 Nearly all states have experienced some success in immunization coverage in at least one district (Rajasthan is the sole state in which no district has at least 70 percent coverage) (table 4.2). To a lesser degree, most states also have at least one district with good prenatal care and insti- tutional delivery coverage. Southern and western states each per- form above the Indian average in all three indicators, whereas the performance of poor north-central states tends to be below average for all three areas. Orissa, alone among the poor north-central states, has above-average levels for immunization coverage and pre- natal care. These differences suggest that lower-performing districts and states may be able to learn from their better-performing counter- Putting It Together: Raising the Sights of India's Health System * 87 Table 4.1 Selected Health Status Outcomes in India and Major Indian States, Selected Years, 1992-99 LIFE NEONATAL INFANT EXPECTANCY MORTALITY, MORTAITY UNDER-FIVE UNDER- AT BIRTH, 1998-99 RATE, 1998 MORTALITY TOTAL WEIGH- AVERAGE (PER 1,000 (PER 1,000 RATE, FERTILITY CHILDRENl, FOR 1992-96 LIVE LIVE 1998-99 RATE, 1997 1998-99 AREA (YEARS) BIRTHS) BIRTHS) (PERCENT) (PERCENT) (PERCEN1 India 61 43 72 95 3.3 47 Andhra Pradesh 62 44 66 86 2.5 38 Assam 56 45 78 90 3.2 36 Bihar 59 47 67 105 4.4 54 Gujarat 61 40 64 85 3.0 45 Haryana 64 35 69 77 3.4 35 Karnataka 63 37 58 70 2.5 44 Kerala 73 14 16 19 1.8 27 Madhya Pradesh 55 55 98 138 4.0 55 Maharashtra 65 32 49 58 2.7 50 Orissa 57 49 98 104 3.0 54 Punjab 67 34 54 72 2.7 29 Rajasthan 60 50 83 115 4.3 51 Tamil Nadu 64 35 53 63 2.0 37 Utar Pradesh 57 54 85 123 4.8 52 West Bengol 62 32 53 68 2.6 49 Note: Major Indian states are those with a population of at least 15 million. Bihar includes Jharkhand, Madhya Pradesh includes Chatisgarh, and Uttar Pradesh includes Uttaranchal. Neonatal mortality is of those less than one month of age; infant mortality is of those less than one year of age; under-five mortality is of those less than five years of age; total fertility rate is lifetime births per woman aged 15-49; underweight children are those under three years of age whose weight is statistically low for their age (that is, more than 2 standard deviations below average). Source: Registrar General (1999); IIPS (2000). parts. Differences among states in the degree of equality in the dis- tribution of health services (chapter 7) should also provide good learning opportunities. Two factors have a major influence on the health systems choizes facing each state: (a) the state's position in the health transition and (b) capacity of the state's public health sector. The position of the major states on these two scales is measured according to infant mortality, child mortality, total fertility, and irmmunization coverage (table 4.3).2 Table 4.2 Selected Health Service Outcomes in Major Indian States and India Overall, Selected Years, 1995-99 PUBUC PRIVATE HOSPITAU- HOSPITAU- FULL PRENATAL INSTITUTIONAL FULL IMMUNIZATION, 1998-99 ZATIONS ZATIONS CARE, 1999 DELIVERIES, 1999 (PERCENT OF CHILDREN (PER (PER (PERCENT OF BIRTHS) (PERCENT OF BIRTHS) I TO 3 YEARS OLD) 100,000 100,000 PERSONS PERSONS STATE LOWEST HIGHEST STATE LOWEST HIGHEST STATE LOWEST HIGHEST PER YEARI PER YEAR( STATE AVERAGE DISTRICT DISTRICT AVERAGE DISTRICT DISTRICT AVERAGE DISTRICT DISTRICT 1995-96 1995-96 Andhra Pradesh 62 41 82 51 28 88 75 51 91 442 1,153 Assam 24 4 82 24 6 76 47 4 95 200 522 Bihar 10 3 43 15 5 44 22 7 71 539 1,172 Gujarat 36 16 65 46 13 68 58 21 76 905 1,946 Haryana 21 15 37 26 17 43 66 47 90 2,206 270 Karnataka 59 27 88 50 18 83 72 25 95 2,944 4,536 cx Kerala 85 68 92 97 88 100 84 60 97 581 448 Madhya Pradesh 17 2 53 22 7 62 48 11 90 792 1,727 Maharashtra 49 30 81 57 16 93 80 59 94 1,168 223 Orissa 32 17 61 23 7 57 58 28 80 1,162 158 Punjab 18 14 41 41 25 58 73 52 94 530 1,092 Rajasthan 15 5 30 23 7 37 37 12 60 669 336 Tamil Nadu 75 46 95 79 55 99 92 77 100 848 1,290 Uttar Pradesh 11 4 67 16 5 43 44 3 82 440 565 West Bengal 33 16 53 39 20 91 52 29 83 1,088 353 All states in India 28 2 95 34 5 100 54 3 100 726 928 Note: Major states are those with a population of more than 15 million. Bihar includes Jharkhand, Madhya Pradesh includes Chatisgorh, and Uttar Pradesh includes Uttaranchal. Source: IIPS (1998-99); National Sample Survey Organisation (1998). Putting It Together: Raising the Sights of India's Health System * 89 The ability of governments to oversee the health sector or to inform and influence the public has not been the dominant feature of govern- ment actions in the public sector, so it is not explicitly measured for the ranking of state capacity. The capacity and scope of the private sector also influences the range of options available to the state. At this poin:, however, we can distinguish only between the levels of involvement cf the private sector in the states and not between the differences in qua - ity of services or cohesiveness of organization.3 As chapter 7 note3, high levels of private sector hospitalization are associated with a more pro-poor distribution of public hospitalization, a fact suggesting that states are better able to concentrate public resources on deliverirg services to the poor when the private sector is more active.4 States differ in their position within the health transition. A stave such as Kerala already faces the burden of how to deal with high-cost-per-episode health care. Whatever capacity Kerala has developed to deliver public services, it must now meet the new ch2.l- Table 4.3 Categorization of Major Indian States by Characteristics Influencing Fundamental Health System Choices PERCENT OF INDIAS POPULATIOr CHARACTERISTIC OF STATE STATES (2001) A. Middle to late transition, Kerala, Tamil Nadu 9.1 moderate to high capacity B. Early to middle transition, Maharashtra, Karnataka, Punjab, low to moderate capacity West Bengal, Andhra Pradesh, Gujarat, Haryana 39.1 C. Very early to early transition, Orissa, Rajasthan, Madhya Prodesh, very low to low capacity Uttar Pradesh 33.1 D. Special cases of instability: Assam, Bihar 13.3 high to very high mortality plus civil conflict or very poor governance, or both Note: Major states (those having a population of more than 15 millionl were ranked according to rates of infant mortality, child mortality, total fertility, and full immunizaticn. The estimates were made before bifurcation, so Bihar includes jharkhand, and Madhya Pradesh includes Chattisgarh, and Uttar Pradesh includes Uttaranchal. 90 * Better Health Systems for India's Poor lenges without having developed the type of systems needed to face them-systems such as risk-pooling financing systems, information systems on provider performance, and mechanisms to take full advantage of networks of private and public providers. Indeed, as shown in part II, these systems are lacking throughout India. Ker- ala's public health and preventive programs also need to respond to a shifting burden of disease and rising public demands. States like Orissa or Madhya Pradesh, on the other hand, are at the other end of India's health transition. They must focus on pre- transition diseases by implementing the public programs needed to prevent and diminish these conditions. In such states, building the government's capacity to measure outpatient and hospital perform- ance and to develop health financing systems are important, but they are less important right now than improving the quality and reach of the current priority public programs. Although states in the pre- transition phase (categories C and D in terms of table 4.3), like those further along, need partnerships between the public and private sec- tors, the focus of such efforts must be less on hospitalization and financing systems and more on basic public health services.5 States in category B, by contrast, must adjust to the health transition by shifting their emphasis on systems, capacities, and content of pro- grams; their challenge is to decide when and how to start doing so. Many other factors also influence the choices to be made between and within states, some of which are outlined in table 4.4. The large urban municipalities need to simplify the overlap and confusion of public providers, systematically work with the dominant private sec- tors, reduce the harmful effects of pollution, and develop more extensive services for the urban poor. Some of these steps have been taken in projects such as the Calcutta Urban Slums Project, but none has focused on making urban services more comprehensive and coherent. Geographic factors distinguish conditions in mountainous states such as Uttaranchal and Himachal Pradesh. These states have been quite diligent and creative in providing outreach health services, but their problems with transport and access to primary curative and first Putting It Together: Raising the Sights of India's Health System * 91 Table 4.4 Local Factors to Consider at State Level when Prioritizing Health Systems Choices LOCAL FACTOR EXAMPLES Lifestyle differences * Nonsmoking tobacco use is 25 times greater in Orissa than in Haryana * Smoking rates are 3.4 times greater in West Bengal than in Maharashtra * Alcohol use is 5 times greater in Madhya Pradesh than in Haryana Poverty differentials * Large differences between northeast Karnataka and south Karnataka; and in Maharashtra State between Mumbai and rural areas Natural risks * Flooding in Ganges delta, drought in Rajasthan * Cyclones in Orissa, Andhra Pradesh, and West Bengal * Earthquakes in Gujarat and Uttaranchal Physical environment * Slums and pollution around megacities * Indoor air pollution (fuel combustion) in rural households * Mountain isolation in Uttaranchal and Himachal Pradesh Political outlook * Communist and collectivist philosophies in Kerala and West Bengal * Greater decentralization and stronger local bodies in Kerala and Madhya Pradesh Social capital * Large numbers of NGOs ond community groups in Gujarat referral care are more daunting than elsewhere and justify a higher priority for public infrastructure in such remote areas. The island territories also have special needs. Some states are especially vulner- able to natural risks such as flooding and cyclones, making prepared- ness for such disasters a more important factor than elsewhere in shaping health systems. Large differences in health risks arising from lifestyle patterns should also be considered. According to the 1995-96 National Sa.m- ple Survey (NSSO 1998), which surveyed people aged 10 years or older, regular use of nonsmoking tobacco ranges from less than 3 percent in Haryana and Uttar Pradesh to more than 40 percent- in Orissa.6 On the other hand, the prevalence of tobacco smoking is I ess than 8 percent in Maharashtra and Punjab and goes as high as 27 percent among the northeast states and 24 percent in West Bengal. The prevalence of regular alcohol consumption is as high as 11.5 92 * Better Health Systems for India's Poor percent in the northeast and 8.6 percent in Madhya Pradesh, com- pared with less than 2 percent in Uttar Pradesh and Haryana. Men and women also differ in their lifestyle risks (men have much higher rates of use of tobacco and alcohol), as do people at different income levels (see chapter 7). For those below the poverty line in India, the prevalence of regular use of nonsmoking tobacco is 37 per- cent higher than for those above the poverty line, 8 percent higher for smoking, and 28 percent higher for alcohol consumption. The poor are therefore more likely than those with higher incomes to suf- fer the negative health consequences of these behaviors. In states where the risks are higher, considerations of efficiency as well as equity suggest that preparation for dealing with these high-cost ill- nesses should come sooner rather than later. Choices for the Central Government The central government has important options for deploying its resources and for arranging its relationships with the states in address- ing programs of national importance such as health care. Yet a high- powered advisory committee from the Centre for Policy Research (CPR), entrusted to consider "Restructuring the Ministry of Health and Family Welfare," noted that no previous attempt had been made to define the appropriate role for the Health Ministry in the Indian federal system (Centre for Policy Research 1999). The CPR report held that the central government has taken a role in health care that has gone beyond that envisioned in the Constitution. In doing so, the central government has assumed overly centralized control of health and family welfare programs and of executive and regulatory functions, created overly rigid planning procedures and standards for its pro- grams, and weakened policymaking and innovation at the state level. Rather than provide options, the CPR report emphatically rec- ommended that the core functions of the central ministry be defined around a set of national responsibilities for policy, planning, and monitoring, while the responsibilities for the allocation of resources Putting It Together: Raising the Sights of India's Health System * 93 7 - U ,. ,.; -I PHTGAHByGEETANJALI CHOPRA/THE WORLDBA qK and executive functions are delegated to state governments and fully autonomous organizations. For the miinistry to fulfill the proposed core functions, the report recommended a specific and challenging list of reforms, many of which would strengthen the central govern- ment's oversight capacity while hastening the prescribed devolution to the states.7 If the central government were to take on this ambi- tious agenda, it would have little time for other activities in the short term. In the long run, however, the payoff could be quite large: the miinistry would be more cohesive and efficient, and many of the new 94 * Better Health Systems for India's Poor functions proposed are currently neglected but vital to the role the central government will need to play in the future. The importance of centrally sponsored schemes to the health profile of India (see chapter 7) and the problems with the current implementation arrangements in many of the programs mean that decentralization of these programs will be difficult, even if not taken to the extent advocated in the CPR report. Some steps have already been taken in the Leprosy Elimination Program, which features state plans that vary on the basis of the local prevalence of the dis- ease and the ability of the state to provide supporting systems such as information technology, planning and budgeting, and logistics. Programs also exist for reproductive and child health and for the control of AIDS, tuberculosis, and malaria. As in the case of the lep- rosy program, these efforts must consider the variations in health status in the states and the feasibility of integrating the programs with supporting systems and general health services. The central govemment's role in setting service norms also deserves reconsideration. For decades, centrally mandated national norms for public health infrastructure have dominated policy dis- cussion and public effort (Duggal 2000; Centre for Policy Research .1999). Yet rigid input norms have been unrelated to workload, local epidemiological conditions, and the presence of a private sector that is the main provider of outpatient and inpatient care. The norms thus create large distortions and inefficiencies, and the centralized approach is less relevant today than ever before. Seven states have attempted to rationalize their public service norms (for services, staff, drugs, and equipment) to better match their capacity to deliver with the presence of the private sector, and different needs across the state.8 These efforts have largely focused on much neglected secondary services and have been part of broader investments to improve referral systems and the quality of services. This type of exercise needs to be continually updated as experience is gained to ensure that high-quality services are being sustained and that the public and providers are satisfied with the scope and per- formance of health services. The approach is also worthwhile extending across all levels of care and in other states and could pro- Putting It Together: Raising the Sights of India's Health System - 95 vide a basis for using standards in the private sector as well. The cen- tral government could facilitate this process by bringing together the lessons learned across the country. Guidance from the central government may now be more impor- tant than ever as the need for new areas of government interventior. becomes clear. For example, guidance on quality standards anc. processes in health care is relevant for both the public and private sectors, but it is sorely lacking. Sharing of information and analyses of health system financing, outputs, and outcomes across states and districts is also needed if lessons are to be learned and successful practices emulated. Existing areas of involvement in oversight func- tions also need to be strengthened. Examples include the use of th2 Medical Certificate of Cause of Death and the Survey of Cause cf Death. Expanded investment in health information systems by the central government should be considered even if it means reducing the curative services the central government finances. Whether the central government takes on these oversight tasks itself or finances others to do them, it will have to acquire for itself new skills, systems, and behaviors. Refocusing its efforts presumab:y implies a reduction in the other activities that currently take up i:s energies, such as the operation of national institutions, advancement of the Central Government Health Scheme, the management of centrally sponsored schemes, and the recurrent high profile healh emergencies that are raised in Parliament or the media. Chapter 8 points out that many options also exist for determining the future of central financing of health. Although central spending on health is currently relatively minor (about 23 percent of pub.ic spending on health), it provides important funds for priority pro- grams, in some cases replacing what states might otherwise spend on them. However, questions remain as to whether central funds shoti.d be given equally to states (for example, on a per capita basis); used to equalize public spending among states; used to provide extra furds for states with special needs (for example, states with poor hea.th conditions, poor health infrastructure, or those having a disaster); or provided to those states that perform better or take up important innovations (for example, states that implement a centrally spon- 96 * Better Health Systems for India's Poor sored scheme particularly well or introduce a new scheme, such as hepatitis B vaccine or public health insurance). The central government also has choices to make about how it wishes to intervene in particular states, especially those identified as having special needs (category D in table 4.3). The options for inter- vention in such states represent a special case for the central govern- ment in part because of the high level of health needs in those states; the size of their population; and the breakdown in governance, includ- ing the public health sector. The Action Plan for the National Popula- tion Policy (Ministry of Health and Family Welfare 2000b) established an Empowered Action Group to provide the necessary resources and organizational force to improve health conditions in Bihar and four other poorly performing states (Madhya Pradesh, Orissa, Rajasthan, and Uttar Pradesh). Although building local capabilities through the central government is necessary in the long run, other options for the financing and management of programs should be considered for the short to medium term in states such as Bihar and others having special needs. One option is to have a greater role for direct central govern- ment financing and management of health and family welfare pro- grams. This option may be realized by pay to have the management and delivery of services handled by a third party such as a new society, a non- profit NGO, a private company, or an international agency.9 Choices for the States The differences among states should mean that states have different priorities. In table 4.3, position in the health transition and the state government's capabilities were proposed as major points of differen- tiation among states, along with other features specific to the vari- ous states (in table 4.4). Recalling the options for the four critical areas of health care activity discussed in chapter 3, we now propose a set of options that are facing the states in different categories as weli as the central government (table 4.5). Since international experience suggests that the need to evaluate Table 4.5 Major Health System Choices Facing Indian States, by Stage of the Health Transition, and the Central Government HEALTH SYSTEM PUBLIC HEALTH AMBULATORY STATE CATEGORY OVERSIGHT SERVICES CURATIVE CARE INPATIENT CARE HEALTH FINANCING A. Middle to late * How to instill * How to introduce * How to build * Whether and how * How to introduce health transition, quality assurance and expand networks with to move from insurance with universal moderate to for public and programs for private providers public provision coverage high capacity private sectors heart disease, * How to "contract" to public * How to raise more * How to measure injuries, mental with private insurance resources for health. and disseminate health, and HIV sector * How to "con- * How to test demand-side performance of * Whether to tract" with private financing for priority private and reduce direct sector curative care (ambulato- public sectors public provision, * Whether and ry and inpatient) * Types of partner- more selectivity how to strengthen ship with private in services pro- publicly run sector that can be vided (e.g. pre- hospitals serving implemented (con- natal care) or rural areas tracting, coopera- focus on back- tion, etc.) ward areas B. Early to middle * How to inform and * When and how * How to build * Whether and how * When and how to intro- transition, low to empower people to introduce pro- networks with to move from public duce health insurance by moderate capacity to demand better grams for heart private providers provision to experimentation health services disease, injuries, * Whether to refocus public insurance * How to test demand-side * Types of partner- mental health, public outpatient * How to "contract" financing for priority ship with private and HIV care only in back- with private sector curative care sector con be imple- ward areas or * Whether and how mented (contracting, across state to strengthen cooperation, etc.) publicly run hos- (T,7hte rontinues on the followinq paqe.l Table 4.5 (continued) HEALTH SYSTEM PUBLIC HEALTH AMBULATORY STATE CATEGORY OVERSIGHT SERVICES CURATIVE CARE INPA;lENT CARE HEALTH FINANCING pitals serving rural areas C. Very early to * How to inform * How to better * Whether and * How to put * Whether to test small early transition, and empower inform and em- how to work with appropriate demand-side financing very low to low people to power people untrained balance on for priority curative care, capacity demand better to live more practitioners primary, secon- especially where the health services healthfully, * How to rejuvenate dary, and private sector is large -How to bring focusing on pre- public facilities in tertiary care public and private transition con- backward areas sector actors ditions and HIV or across state together to work on common interests D. National * How to refocus * How to devolve * Whether and how * Whether and how * Whether to raise more government on national over- centrally sponsored to develop stan- to provide exam- funds for health (e.g. sight issues, schemes to states dards and accredi- ples for how to through general rev- promoting quality and local bodies to tation and reorganize large enues or tobacco or assurance in public facilitate better licensing schemes hospitals alcohol taxes) and private sectors; implementation * Whether and how * Whether and how * Whether to use central national-level infor- * How to intervene to develop guide- to provide incen- funds to counteract inter- mation, education, in very poorly lines, training, and tives for develop- state differences in public and communica- performing states patient education ment of provider financing, health needs, tions; information (e.g. Bihar) or materials networks or performance on health system states in special * Whether and how to performance; drug circumstances stimulate public health quality control ]ammu and Kashmir) insurance Putting It Together: Raising the Sights of India's Health System * 99 the health system and improve it is continuous, the suggestions outlined in table 4.5 should be considered a work in progress. The purpose ol presentation here is not so much to limit the options for specific states but to provide additional substance to the debate and suggest which set of options may be tackled first by states having different conditions. Health System Oversight In the long run, all states ought to be able to deliver on each of thL actions in table 4.6 for improving oversight. Each state ought to develop its health policy to be more comprehensive in its scope an l initiate steps to improve measurability in the private sector. The dif- ference between what states can do and should do may be more dependent on opportunities that are available, leadership that emerges in the states, and the urgency of the issues. States in categories A and B in table 4.3 (especially mid-late transi- tion, moderate-high capacity) need to institutionalize quality assurance procedures in inpatient and ambulatory services through a balanced approach of options involving providers and consumers (actions 1 through 6). Perhaps the best way to start is with action 2, using ind- pendent organizations to measure performance because the problems of credibility and conflict may be larger at the beginning of these reforms. However, action 2 may require the development of consider- able trust between public and private sectors, which could be accoin- phshed through the promotion of partnerships between the public and private sectors (action 1). Increasing the role of consumer advocates is another good approach to strengthening oversight (action 4). For other states, focusing initial activities on empowering people to demand better health services and initiating steps to bring together actors from the public and private sectors may be more f-a- sible (actions 1 and 3). As experience and confidence is gained, more actions could be pursued. Public Health Services Clear differences exist between the states in the area of public health services. States in category A have a more compelling reason to focus Table 4.6 Pros and Cons of Actions for Improving Oversight ACTION PROS CONS 1. Develop partnerships with the private * Potential for improving information, * Difficult with low measurability of services and sector to share information, improve efficiency, and coverage of services. unclear social mandates. quality, cooperate on service provision. * Contracting has potential for * Requires capacity in public sector to manage increased accountability and efficiency. contracts. * Potential for increased conflict among competing providers. * Shortage of qualified providers in rural areas may persist. 2. Support independent organizations * Credibility of honest broker. * Difficult to maintain independence. to measure performance among public * Potentially efficient way to address * Needs goodwill from public and private sector. and private providers. information and quality assurance needs. * Potentially threatening to public and private sectors. 3. Support professional self-regulation. * Likely to gain support from private sector. * Little credibility to consumers or government as an honest broker. *Technical capacity has not been tested. 4. Facilitate consumer health advocacy * Potential to increase people's participation. * Few organizations have capacity. organizations independent of both public * Low-cost way to raise accountability and * Requires investment in their development. providers and private associations. performance of the health system. * Risk of advocacy groups being "captured" by political and other interest groups. * Consumer groups typically represent middle and upper classes; may not take on interests of the poor. 5. Establish or strengthen formal * Addresses major problems with drug * Little capacity in public sector to effectively regulation of health inputs such quality control. regulate. as pharmaceuticals. * Current system underfinanced. * Risks creating inspectorate raj. 6. Establish formal public sector regulation * Addresses problems of untrained * Same as in action 5 above, but more difficult of private providers through legislation. practitioners and harmful medical to measure objectively practices. * Risk of "regulatory capture," limited * Gives teeth to current weak system. competition, and higher prices. * Creates adversarial relationship between regulators and providers. Table 4.7 Pros and Cons of Actions for Improving Public Health Services ACTION PROS CONS 1. Concentrate effort on programs * Addresses major burden of disease * Existing programs are overly centralized and have for the "unfinished agenda" by * Highly cost-effective interventions exist. weak management capacity, and do not have the increasing funding and increasing * Programs already exist. political support to make them work as well as they effectiveness should. * The reforms needed in these programs may not be forthcoming. * Costs of these programs may require tradeoffs which may reduce funds for curative care and raise political risks and opposition from staff. 2. Initiate and strengthen * Will address a burden of disease that is * May distract attention from the "unfinished agenda," programs in noncommunicable large and growing especially in states that are in early stages of the diseases in states well advanced * Will help develop programs that are smcll health transition and have little capacity to develop in the epidemiological transition and underfunded and implement new programs * Effective prevenfive interventions will prolong * May be implemented as traditional centrally sponsored healthy lives and reduce expensive hospitali- scheme, making it more difficult to undertake other zations in the future. reforms in the national MOHFW or organize pro- * For states well into the health transition, grams in a state-specific manner the poor will benefit disproportionately from such measures. 3. Reduce centrally sponsored * Reduces problems with national norms and * Some states may not follow through on implementation schemes, turn over the resources dependence on center of national priority programs. to the states, and reassign functions * Increases the role of states, so that planning * Some states do not have technical capacity to design of national MOHFW and accountabilities are closer to the level and implement their own programs. of implementation * Some states may not want additional responsibilities, * Should encourage adaptation to local needs since they can now blame national government for and innovation by states inadequacies. * Should encourage greater integration of * National MOHFW is not organized to provide the public programs technical support needed by states. * Allows national MOHFW to focus on policy, * Risk of resistance by national bureaucracy and information sharing, technical guidance, politicians, since change entails loss of control of and issues of national importance, which resources and direct influence and reduces currently receive insufficient attention opportunities for promotion * Untested mechanisms for national MOHFW to appropriately monitor progress and influence resource allocation under new arrangements 4. Reinvest heavily in public * Leads to stronger skills base * Costs may involve tradeoffs which may reduce funds health systems-institutions, * More effective health planning, for curative care and raise political risks and management information systems, implementation, and monitoring opposition from staff. surveillance * Private sector is not addressing this area, * Little current capacity in training institutions and it has been neglected by public sector. * Risks opposition in directorates of health, as powers shift to public health, with increased information flow and less direct control over hospital budgets and purchase of supplies * Potential opposition from non-public-health medical specialties Note: MOHFW, Ministry of Health and Family Welfare. 104 * Better Health Systems for India's Poor on introducing and expanding public health services for cardiovascular disease, mental health, and injuries, as these are now the prominent conditions facing their populations (action 2 in table 4.7). States in cat- egory B may be in the position of considering when and how to intro- duce these programs, as these threats loom large in the near future. States in table 4.3 in category C, such as Orissa, in which tobacco chew- ing is widespread, may need to consider putting more effort into pub- lic health campaigns related to lifestyle-related health problems; but in the main, states in categories C and D will want to focus on improving implementation of the existing public health programs that tackle the conditions of the "unfinished agenda," notably reproductive and child health, nutrition, and tuberculosis control (action 1) along with invest- ments in sanitation and water. These options could be implemented through action 3 (provide resources to the state and convert the role of the central government's Ministry of Health and Family Welfare). All states need to put additional resources into HIV prevention, although the mix of interventions may vary from state to state. Each state and the central government may find that action 4 (reinvest in general public health systems) is also an important means of imple- menting any public health strategy in the medium to long term. In any event, the success of public health programs usually depends on more than funding the technical interventions, as demonstrated in an analysis of recent "success stories" in low-income countries (box 4.1). Ambulatory Care Using the options for ambulatory curative care discussed in chapter 3 (shown in this chapter in table 4.8), the higher-capacity states (cate- gories A and B) may wish to move toward something like a combina- tion of action 1 (revitalizing public facilities in disadvantaged areas and purchasing services from private providers) and action 2 (demand-side mechanisms for the poor); such an approach suggests itself because the health sector in the category A and B states is sufficiently developed so that the presence or absence of a public facility would make a differ- ence only in remote areas, and these states would have the ability to form linkages with the private sector to protect the poor. Putting It Together: Raising the Sights of India's Health System * 105 Box 4.1 Keys to Success Many low-income countries have shown that if they use the available tools both widely and wisely, health outcomes can be improved dramatically. Many countries have success sto- ries in spite of poverty. Malawi is set to eliminate measles despite the fact that only 3 percent of the population has access to adequate sanitation, and Bangladesh has reduced neonatal tetanus fatalities by more than 90 percent even though most mothers in the country do not have access to a clean delivery environment. The keys to these successes seem to be political commitment, the development of public- private partnerships, a willingness to innovate, health edu- cation, and measurability in programs. Several of these factors seem to have been at work in the successes being realized by Malawi and Bangladesh; additional country examples are given in the following elaboration of these keys to success. Political Commitment Efforts to reduce the burden of disease have been driven by a firm political commitment. Examples include Uganda and Thailand, where political leadership has been critical in the fight against HWV/AIDS. Another example is Peru, where the government has established the control of tuberculosis as a social, political, and economic priority. Parnerships Success has involved partnerships with the private sector and NGOs for the social marketing of condoms in Uganda and for malaria control in Azerbaijan. In some countries, governments are providing health services and commodi- ties outside the formal sector in an effort to broaden access to health care. In Senegal, mosques throughout the country 106 * Better Health Systems for India's Poor Box 4.1 (continued) are a focal point for HiIV prevention efforts, counseling, and support. In Tanzania, a school-based program has improved the health of children with intestinal worms, and in Kenya, employers are supplying bednets to their work force through payroll purchasing schemes. Innovation Innovation, born of a pragmatic approach to achieving results, has made a difference. In Nepal, accommodations in hostels are provided to TB patients from remote moun- tain areas to encourage their compliance with treatment. In Thailand, the government worked with brothel owners to spread the use of condoms-despite the fact that prostitu- tion remains illegal. Health Education Health education and training of extension health care workers have been key elements for success. In Sri Lanka, high female literacy rates and midwifery training for health care workers have both been instrumental in preventing maternal deaths. Sex education for children and adolescents has been an integral part of successful HIV prevention pro- grams in Thailand, Senegal, and Uganda. Measurability Measuring outcomes is central to the development of suc- cessful programs. In Senegal, Thailand, and Uganda, sys- tems of disease surveillance and monitoring have been essential to tracking the course of the HIV/AIDS epidemic and to monitoring the effectiveness of interventions. Source: WIHO (2000a). Table 4.8 Pros and Cons of Actions for Improving Ambulatory Curative Care ACTION PROS CONS 1. Focus public resources * Realizes some savings in scarce public * Some political opposition if seen as a retreat from strategically by funds and management capacity. universal care. (a) Revitalizing the network of * Improves quality of public services where * Potential for nontransparent manipulation for selection public sector facilities-but they are most important. of disadvantaged areas. only in disadvantaged areas * Focuses public resources on areas of * Opposition from public sector staff who may need to where private sector and need and reduces inflexible application be moved from urban areas and hospitals to rural NGO alternatives are of norms. primary care facilities. not available. * Politically unpopular if seen as a retreat from public (b) Limiting the provision of * Saves resources by closing underutilized commitment to health. ambulatory curative care by facilities. * Public sector workers likely to oppose change in roles PHCs and subcenters wherever * Potentially reorients staff back to core and potential loss of jobs. such care is available from the public health services. * Some rural areas may have no quality private private sector. -Takes advantage of large private sector. providers available. a Difficult to identify private providers of good quality. 2. Expand ambulatory curative * Builds on a voluntary linkage between * Requires greater commitment of public sector to build care coverage for the poor through public and private providers. administrative capacity to link with the private sector. demand-side mechanisms that give * Provides an entree for meaningful quality * Capacity needed to assess quality of private providers, the poorest patients access to assurance, if providers need to have create criteria for household eligibility, define service publicly subsidized discounts from "certified" quality of care to be eligible packages and reimbursement rates, and avoid public or certified private providers. for reimbursement. corruption (collusion and overreporting of services). * Helps redirect poorer households from This has not been tested. untrained practitioners to higher-quality * Priority centrally sponsored schemes requiring clinical providers. services may have very different approaches with -Allows public sector to more easily move public and private clinics across states, testing the out of provision in areas where there are flexibility of the central government good private alternatives. * In some rural areas no quality private providers may be available. (Table continues on the followihig pig&.) Table 4.8 (continued) ACTION PROS CONS * Allows public sector provision to be concentrated in areas where poor have fewer options. * Because many private providers already provide discounts to the poor, option would build on current experience. 3. Invest in raising quality of the * Potential for increased quality of care. * High costs; needed funds have not come in the past. entire network of publicly * Could attract more poor people currently * The needed organizational reforms have always been provided ambulatory services, using low-quality private care to use limited. reorganizing for greater efficiency. public facilities. * Scarce public sector managerial capacity would be * More opportunities to carry out preventive diverted from other parts of the health system: public and promotional services. health services, oversight, quality assurance, inpatient * Maintains current balance of interests, care, and development of new financing systems. preventing opposition from those who may not gain if more selective improvements are made. * The private sector would still provide the bulk of services nationwide and to the poor; the effect of these actions upon the poor is probably limited. Note: PHCs, primary health centers. Putting It Together: Raising the Sights of India's Health System * 109 In the low- and very-low-capacity states (categories C and D) that still have high rates of infant mortality and low coverage of basic pub- lic health measures such as immunizations and nutrition programs, a different approach would make sense. These states would de-empha- size publicly provided ambulatory curative care, emphasizing instead action la (reducing public ambulatory curative care at non-hospital facilities) and action lb (concentrating public facilities in disadvan- taged areas) to allow a reallocation of effort to public health activities These actions may be more viable in states with greater private pro- vision of outpatient services, although the fear of political repercus- sions may prevent such approaches, especially for action lb. If a wind- fall of resources were to become available, action 4 (revitalize the entire network of public facilities) would become more attractive. The states with medium capacity and medium health perform- ance (category B) face the largest strategic issue: because the privatz sector may still be underdeveloped and the treatable conditions cf the "unfinished agenda" may constitute a large component of mor- tality, they might want to invest in the creation of a well functioning public ambulatory curative care system (action 3). On the other hand, since the private sector delivers most ambulatory curative care, they may want to skip that effort entirely and move to the cre- ation of more links with the private sector (action 2). Inpatient Care and Health Insurance The states in categories A and B must pay considerable attention -o inpatient care and health insurance because high-cost care consumes a large part of the health expenditures in these states. Forward-looK- ing states in categories A and B should be concerned about choosing between action 1 (multiple public and private health insurers) and action 2 (universal public insurance) as outined in table 49. Nonetheless, no state is in a position to introduce major health insuLr- ance without first doing the careful groundwork that is needed to make these efforts successful. The groundwork includes public ccn- sensus-building and policy decisions; addressing the level of prepay- ment for insurance, the mechanisms for pooling funds, and the man- Table 4.9 Pros and Cons of Actions for Improving Inpatient Care and Health Insurance ACTION PROS CONS 1. Encourage multiple insurance pools. * An incremental approach that holds * The necessary information systems, quality assurance Facilitate the creation of private insurance out the promise of universal care may procedures, administrative systems, and marketing to pay private providers in a regulatory be technically feasible and politiccily strategies are undeveloped. environment that encourages replication popular. * Will require investments that the government has not of voluntary employer or union-based * Preserves choice of provider and mode in the past, and the costs are likely to be insurance schemes (such as the Self- insurer and has the potential to substantial Employed Women's Association) along reduce catastrophic costs and * Explicit decisions about the minimum package of with public insurance inequality if done well benefits, amounts of public funding, and level of cross- subsidization may expose the national MOHFW to additional political risks and overtax its capacity. * The regulatory framework is not yet in place, so that interests may become entrenched and block subsequent reforms. * India has little experience with nonprofit health insurance of sufficient scale, and government has little capacity to develop or administer its own public insurance. * Partitioning of the risk pools along lines of employment leaves government to cover those with the greatest risks and least resources, requiring more complicated subsidy schemes among insurance plans. * Most of the population work in the informal sector and is difficult to include in these schemes. 2. Move toward compulsory purchase * Universal coverage, ability to control * Public and private providers do not have the of "single payer" insurance that costs, ease of administration necessary information systems, quality assurance reimburses public and private compared to other insurance options. procedures, and administrative systems. providers neutrally * Collectivist approach and ability to * The costs are likely to be initially very high, depending maintain private provision and patient on the benefit package proposed. choice may give this popular appeal. *The public may lack confidence in administration of * Can be used to improve quality scheme and be unwilling to increase payment for assurance and influence clinical public insurance. practices in line with public priorities * The rich may prefer private health insurance to subsidizing the poor, and the healthy may not want to contribute if they do not benefit. * Requires political leadership, yet the political risks of failure may be quite high. * The problem of controlling costs in a predominantly fee-for-service system has not yet been addressed. 3. Rely on the introduction of * Public sector may be able to focus * Escalation of the cost of health services to consumers private voluntary health insurance resources on public health by * Increased inequalities: If richer households use the spending less attention on hospitals. public system less, there will be even less political * Path of least resistance in health care support for public financing of public hospitals and a financing reform vicious circle of declining quality and fewer middle- and upper-income patients. * The private market is difficult to regulate; creates entrenched interests that resist reform. 4. Strengthen quality of care, * Would take advantage of under- * Requires considerable resources and continued especially in public sector secondary utilized public hospitals financial commitment for operation and maintenance, hospitals serving rural areas, but * Positive experience in state health none of which was available in the past increase cost recovery with or systems' development projects * Staff and politicians resist reforms in staff placement. without prepayment mechanisms * Approach is popular with state * State departments of health may not have the capacity to improve equality politicians and bureaucracy, so to directly administer hospitals. commitment is relatively high. * User-fee exemptions for the poor are in place, but it is * Cost recovery at these institutions not known how much of a barrier to care such has been associated with quality fees have been. (Table continues on the following page.) Table 4.9 (continued) ACTiON PROS CONS improvements and provided important * Approach is insufficient to address issues of reducing flexible funds for operations, making catastrophic costs to patients, as no prepayment and fees socially and politically feasible. risk pooling have been attempted. 5. Make public hospitals autonomous, * Public hospitals can be managed more * Without changes in the legal environment and ability and fund their services publicly professionally. to release low-performing workers, savings may not be * Explicit purchasing of services can realized. make hospitals more accountable and * Unless social mandates (such as free care for the poor) improve performance. are made more measurable and sufficient budgets are provided, social mandates and efficiencies may not be realized. * Moral hazard: Because it is unlikely that governments would allow public hospitals to be closed, market exposure of such institutions will be limited. * Professional hospital management has not been well developed in India, so expected management improvements may not be achieved. * Bureaucratic opposition to relinquishing control over hospital administration may make such reforms difficult. * Opposition from labor unions and doctors may be anticipated, particularly if losses of employment or benefits appear possible. Note: MOHFW, Ministry of Health and Family Welfare. Putting It Together: Raising the Sights of India's Health System * 113 ner in which the poor and the sick will be subsidized; the package of health services to be covered; and the payment mechanisms to be used. The health information systems, clinical networks, and quality assurance procedures that health insurance will depend on will also require considerable effort to develop. Relying on private health insurance (action 3) is likely to play some role in the short term because the insurance market has been liberalized and its use consumes little in the way of additional public resources. However, it would still be desirable for the central government to play a larger role in actively regulating private insurance, although states could contribute to the monitorng and regulating of health insurance. International experience has shown that the cost of not aggressively regulating private voluntary health insurance is escalating health costs, increased inequalities, and greater difficulty in regulating it later on. Options for the reform of hospital operations are complementary tc those oriented around financing. For states earlier in the transition and with less capacity (category C), governments might reasonably spenc more effort on improving the quality of existing public first-referra'. hospitals but do it in a selective manner that would promote more equi- table distribution of public resources (action 4). Experience with proj- ects to develop state health systems has shown that such approaches can improve quality of care and efficiency, although several years of worlc are required before many of the benefits are seen. Many countries hav2 tried to make the type of changes in hospital organization addressed i l action 5, but few have been able to obtain the full set of benefits intended at the start of reforms. Nonetheless, given the degree of inef- ficiencies and discontent in public hospitals, some states may find t attractive to make hospitals more autonomous in the hope that they can be managed more professionally and made more accountable. Putfing It All Together: The Case of Maternal Health In the previous section, we examined in broad terms how different states have different priorities and opportunities to make different 114 * Better Health Systems for India's Poor choices for their future health system. In this section, we take a look at how to put the choices together for a particular program. We use the example of maternal health because it is a national priority, its range of activities span all aspects of the health system, and it illus- trates the variety of interrelationships among the pieces of the health system. Below are four visions of how maternal health serv- ices could be directed in India. Table 4.10 summarizes how these visions would be implemented for a detailed set of maternal health activities. Vision 1: A revitalized, vertically integrated, publicly provided system. Vision 1 is for all maternity services to be delivered and financed through the public sector. The public sector would provide the facilities and staff for universal prenatal, delivery, and postnatal care. A benefit of this vision, if realized, is the expansion of oppor- tunities to provide child health and health education services. A large increase in public finances and a change in the way the public service works (especially with respect to staff placement and disci- pline) would be needed to attract patients from the private sector. Because all service would be rendered free of direct charge to patients, health insurance would not be required. Vision 2: A private sector system with public financing for the poor and public oversight. In this vision, the private sector would provide all services, and the public sector would provide financing through vouchers of insurance for services provided to the poor or other identified groups. Few savings would be realized from a reduc- tion in staff and capital for prenatal and postnatal care unless all ambulatory care were also privatized. Referral networks to handle complicated cases would need to be strengthened within the private sector, and systems of accrediting quality providers would need to be put in place. Incentive schemes would also be needed to attract pri- vate sector and NGO providers to offer services in remote and poor regions. Provider associations could be encouraged to promote safe motherhood. Table 4.10 Allocation of Critical Health System Functions for Maternal Care between the Public and Private Sectors, by Type of Vision for Public-Private Partnerships FUNCTION SPECIFIC COMPONENTS VISION I VISION 2 VISION 3 VISION 4 Oversight * Health management information Public sector responsibility for financing and possibly implementation. Private firms systems on access and utilization or independent agencies may be contracted to design and implement some of of services by poor, cesarean these tasks. Goodwill with the private sector needs to be developed to extend delivery rates, etc. coverage to private sector * Placement and incentive policies * Quality assurance on clinical standards, cesarean deliveries, blood banks, and training of clinicians Public health Behavior change communications on: Public sector responsibility for financing and possibly implementation. Privote firms services * Age at marriage may be contracted to design, implement or monitor these tasks. * Birth spacing * Early registration for prenatal care * Signs of high-risk pregnancies * Use of referral services Ambulatory * Prenatal care (blood pressure Public Private Private Shared care measurement, abdominal palpation, implementation implementation implementation implementation blood and urine test, iron folic acid and financing Public Private Networking supplements, tetanus toxoid) financing financing or between public * Home visits Iprenatal and postnatal) for the poor public and private *Timely identification of high risk financing sectors * Referral transport for the poor * Mapping of referral facilities (Table continues on the following page.) Table 4.10 (continued) FUNCTION SPECIFIC COMPONENTS VISION I VISION 2 VISION 3 VISION 4 Inpatient care * Prenatal complications Public Private Shared, Shared * Delivery (normal and assisted) implementation implementation public implementation * Emergency obstetric care Public funding of Public insurance * Blood transfusion financing for public for poor or single * Care of newborn (normal and sick) the poor hospitals payer for all * Postnatal complications (vouchers or insurance) Financing * Payment (out-of-pocket; free care; Public Shared Shared Shared (public insurance financing (public funds (public funds funds used for poor, focused on focused on or for those with the poor) poor) complications, or for all) Note: For definitions of visions, see text. Putting It Together: Raising the Sights of India's Health System * 117 Vision 3: Public sector health promotion, private sector prenatal visits and routine deliveries, and public sector emergency care. In this vision, the public sector focuses on advancing awareness of good health practices such as sufficient birth intervals, the use of pre- natal care and safe abortions, and institutional deliveries; the imple- mentation of communication programs may be done through private sector agencies. Prenatal care and routine deliveries would be done primarily through the private sector and emergency obstetric care primarily through public sector hospitals. Vision 4: Public sector health promotion, private and public prenatal care, insurance for institutional deliveries in either public or private sector. In this vision, the public sector would again be responsible for the promotive health communications. The current mix of public and private prenatal care would continue or could be enhanced through quality assurance procedures and net- working that would involve both public and private providers. Insur- ance could cover the higher costs of institutional deliveries and any complications through a number of approaches, such as (a) volun- tary insurance with public insurance for the poor, (b) mandated pri- vate purchase of insurance policies of choice and public financing oF insurance for the poor, and (c) publicly funded single payer insur- ance. Emergency obstetric care could be provided through either public or private hospitals, and financed through insurance. The implementation of any one of these four visions for matern.l services would have its own set of challenges and risks. Vision 1 may fit best with the traditional rhetoric of the health system. However, the private sector is currently providing large portions of the ambu- latory and inpatient care for maternity services. In the absence of considerable increases or shifts in public spending and a feasible plin to replace the private sector, vision 1 may be the least realistic of the four visions. Vision 2 may be the most politically difficult of the fo-ir because it envisions the largest reduction in the public sector's direct delivery of services. 118 * Better Health Systems for India's Poor Visions 3 and 4 rely more on shared approaches; they have the advantages of being susceptible to implementation in steps and of using the private sector while giving greater attention to the needs of the poor. However, vision 3 would entail a difficult and at times arbitrary reduction in public services for prenatal and routine deliv- eries while requiring the public sector to expand its emergency care. Opposition from the labor force and politicians could be expected, and efficiency gains in the public sector are hard to imagine if it must maintain and expand its emergency obstetric services. Vision 4 takes most advantage of the current situation and may offer less political resistance, although it depends on the development of meaningful partnerships between public and private sectors that have so far eluded the health system. It also depends on insurance schemes that have yet to be tested on a significant scale. Specific Actions for Consideration across India Up to this point, we have assessed options without prescribing spe- cific recommendations for the reform of India's health sector. In this section we advance such concrete proposals. The intention is not to prescribe detailed steps but to further the discussion and decision- making about India's health system. These proposals for translating the options into policy and action are offered in that spirit. Health Policy National health policy, last articulated in the 1983 Health Policy and the National Population Policy (Ministry of Health and Family Wel- fare 2000b), is being revised and preparations are being made for the 10th Five-Year Plan as this is written. We argue that in the updating of its policies and plans, India must give more attention than it has in the past to the monitoring of health system outcomes and to improv- ing the distribution of services among different groups of Indians, particularly the poor. Given the great variation in conditions among the states, revisions in national policy also ought to allow greater Putting It Together: Raising the Sights of India's Health System * 119 specificity and flexibility in dealing with the various states. In particu- lar, the new plans must discover ways that national priority programs can be productively decentralized to states and local bodies and ways that the central government can provide relevant support to states facing different issues. The questions raised in this chapter about cen- tral government allocations to health should also be clarified. Shortcomings in policy are even greater at the state level. No state has a comprehensive policy or strategy for dealing with the health sector. All of them should have plans that encompass the issues confronting the national government, but in addition, the states ought to address how they intend to integrate centrally spon- sored schemes and state health programs and how they can be mon- itored. Thus, both the national and state-level plans must be sensi- tive to the existence of the other. The studies supporting this report make it clear that national an i state health policies must take a far more comprehensive and oper- ational approach to dealing with the private sector if these policies hope to accomplish the following: * Improve the quality of services in the private sector and build sys- tems to assure quality (for example, accreditation systems, focused regulations, and public disclosure) * Reduce inequity in charges in the private sector and build trars- parency in the pricing of private health services (for example, through publicly disclosed pricing guidelines) * Define the rights and obligations of the public with respect to health services provided by the private health sector (for example, a charter of patient rights) * Outline the requirements for sharing of information with the pri- vate sector (for example, on clinical care, outcomes, and costs) * Create tools to control costs and encourage appropriate utiliza- tion in the private sector (for example, mandatory review for high-cost procedures and hospitalizations) 120 * Better Health Systems for India's Poor * Define what government agencies will offer to private providers under service agreements (for example, information, materials, immunizations, and drugs for national programs such as tuberculosis) * Find opportunities to increase access to health services through the private sector (for example, through joint ventures, incentives for service in remote areas) * Determine what types of subsidies will support public policy objectives (for example, land in rural areas for clinics) * Lay ground rules for public-private partnerships in health services (for example, by contracting and by extending provider networks and referral systems to the private sector) * Develop strategies for dealing with practitioners without medical qualifications * Place boundaries on the roles of various types of organizational structures in the health sector (for example, public oversight boards, large inpatient facilities, for-profit corporations, and local bodies). Operational Steps In addition to articulating health policies that would deal with these neglected areas of the health system, it is also important to find means to translate them into action. Without repeating the rationale for each of the various options discussed in the preceding chapter, we now pro- pose a number of important options that can be taken up by govern- ment (table 4.11). Steps to be taken in the near to medium term are the more feasible or are prerequisites to larger reforms. For example, experimentation in health financing and the development of health information systems and quality assurance are proposed as steps that need to be taken before establishing a long-term health financing sys- tem. The proposals outlined are also selected because they allow the government to follow a particular vision it may develop for the sector, thereby leaving open as many options as possible. Table 4.1 1 Recommendations for Government Action on Critical Functions of India's Health System FUNCTON SHORT AND MEDIUMTERM MEASURES LONG-TERM MEASURES Health system * Update national health policy and specify roles for * Evaluate private and public provider performance oversight private sector * Publicly disclose pricing of health services * Develop state health policies - Accredit hospitals based on quality * Invest in drug quality control * License or relicense health providers * Invest in disease surveillance and control * Implement human resource development plan * Invest in health and management information systems * Revise mechanisms for consumer redress in health care for public and private sector * Monitor distribution and levels of public utilization and subsidies * Expand and share knowledge base about health system performance * Establish public-private forums on health care * Develop human resource development plans at national and state levels, including new capacities Public health * In early health transition states, focus on injuries, and * Expand public health programs for tobacco control, services improved implementation and decentralization of mental health (short- to medium-term action in some centrally sponsored schemes states) * Expand health promotion of unfinished agenda and HIV control * Provide public information on health service performance and rights of public Ambulatory * Eliminate inpatient beds at public primary health centers * Promote mixed public-private provision through greater curative care * Begin measuring quality, costs, efficiency, and social networking, contracting, and shared training and mandates in private and public sectors information and with clear public accountabilities * Experiment with greater contracting of services with NGOs * Promote greater specialization and professional and privul. ,ector manogement of public sector logistic support systems (Table continues on the following page.) Table 4.11 (continued) FUNCTION SHORT- AND MEDIUM-TERM MEASURES LONG-TERM MEASURES * Invest on improving quality rather than expansion (e.g., drug and equipment procurement and management) of publicly provided services for minor and inpatient care * If no increase in public financing, pay for oversight and * Separate health providers from civil service (for both out- public health activities through reductions in public patient and inpatient care) funding of ambulatory curative care Inpatient care * Rationalize services and develop dynamic need and * Promote mixed public-private provision through greater performance-based service norms at state level networking, contracting, and shared training and infor- * Develop policy and plans for dealing with long-term mation and with clear public accountabilities inpatient stays, currently dominated by the rich * Decentralize locol management of public and private * Improve measurability of quality, costs, efficiency, and hospitals through local governments and public boards social mandates in private and public sectors * Experiment with increased autonomy of public hospitals * Develop service agreements with hospitals run by NGOs and by private firms * If no increase in funds, limit public spending at tertiary hospitals and focus resources on fewer tertiary institutions and rural secondary hospitals Health financing * Experiment with prepayment risk pooling of sufficient * Establish publicly accountable administration of health scale (e.g., at least one district) insurance with universal coverage (single payer or * Actively regulate private, voluntary health insurance multiple plans) (see chapter 7 for details) * Increase revenues for health through general taxation and special taxes on health-harming products (e.g., tobacco and alcohol). Putting It Together: Raising the Sights of India's Health System * 12 3 These recommendations are supported by the four principles advanced at the outset of this chapter for raising the sights of India's health system: 1. Oversee the needs of the whole population 2. Look forward to new challenges of the health system 3. Remove a large blind spot by making better use of the private sector 4. Focus on improving quality, efficiency, and accountability of health services. The initial steps include measures to broaden the scope of the health policy framework while making it more operational, increase "measurability" in the health sector, add more experimentation and flexibility in approaches, and rely more extensively on public-private partnerships. An Agenda for the Future As India's health system continues to evolve, an ongoing process of analysis, discussion, and informed decision is needed to meet the challenges. The need to better understand the main actors, finc- tions, and outcomes of the health system will continue. And the wide variation in conditions across India demand that local investigation be carried out so that solutions can be custom-made to particular states and districts. If India's health system is to address the needs of the entire pop- ulation in an effective, efficient, and accountable manner, peop .e from all segments of society must become engaged in decisions about the future direction of the health system, and the health sys- tem must find ways of facilitating such broad engagement. If India is to face the challenge of the health transition, then different approaches to health financing and new public health programs will be needed. If India is to take advantage of all its resources in tne 124 * Better Health Systems for India's Poor health sector, the energies of the private sector need to be harnessed and market failures counteracted. The public sector will also need to better oversee the health sector on behalf of all citizens, especially the poor and the vulnerable. Important gaps still remain in our knowledge about and experi- ence in the Indian health system. Appendix table 4A. 1 outlines these gaps and indicates how the information filling those gaps could be applied. Here are some of the most pressing areas for more detailed analysis and consideration of options: * Working effectively with the informal sector of private health providers within the boundaries defined by law. * Testing new health financing systems. * Analyzing pharmaceutical policy in the context of a new interna- tional trade regime and the challenge of emerging diseases. In particular, examining how HIIIV drugs can be made affordable in India. * Analyzing options for urban health care. * Testing innovative approaches to delivering services for the unfin- ished agenda, particularly neonatal health and malnutrition. * Developing strategies for health manpower development, in par- ticular how to strengthen public health and nursing capacity, and how to reform medical education. * Understanding how to maximize the benefits from health research and development. Concluding Remarks This report began by observing that India's health system is at a crossroads. The study process generated a considerable amount of new information about the workings of the Indian health system and revealed many options for reform; it also showed that no single Putting It Together: Raising the Sights of India's Health System * 125 choice or path is best for all of India. Particular conditions and par- ticular states require their own mixes of policy and timing. So those seeking new approaches must travel with eyes open, consciously deciding which fork in the road to take for each situation. Discover- ing the right way requires experimentation, consultation, analysis, debate, and the forging of a consensus on key issues. Now is the time to conduct big experiments in India's health sec- tor, particularly since the status quo is leading to a dead end. Either the national government or individual states could take the initiative, but new experience is needed to build on what has become an out- moded health system. The opportunity exists now for governments to reform the way they work and take on critical oversight functions. Governments also need to consider new ways to implement new and existing public health services. New ways of managing ambulatory and inpatient services should also be tried, along with a concerted effort to develop a more efficient, equitable, and risk-reducing health financing system. The new experiments need to build in flex- ibility to deal with very different conditions in India and to provide a basis for learning through intensive monitoring and evaluation. In the long run, this approach will better enable India to meet the health needs of its people. Appendix Table 4A.1 Key Gaps in Knowledge about and Experience in India's Health System, and Potential Uses of the Needed Information GAPS IN KNOWLEDGE AND EXPERIENCE USE OF INFORMATION Role of the public * How to empower people to better manage their health * Improved programs to strengthen people's ability to manage their health Role of public and private sectors * Understanding the constraints, means of operation, and * Design and implement interventions to influence the effectiveness and performance of the private sector in health in local markets equity of the private sector in health, and improve accountability * Feasibility of health networks involving NGO and private * Models for providing comprehensive health care to more people providers * Evaluation of contracting mechanisms in health * Appropriate expansion of good management practices Financing functions * Feasibility of health insurance schemes (social insurance, * Working model of health insurance or community financing schemes community prepayment) * Unit costs for health services provided in different settings * Means to improve the efficiency of health services, and better negotiate with private sector * Analysis of alternative financing schemes for NGOs- * Mechanisms to improve selection and performance of health NGOs feasibility studies Input management * How to prepare for the new TRIPS regime in pharmaceuticals. * Plans for making new pharmaceuticals affordable and available in India How to develop, finance, and make available affordable drugs for HIV and other emerging conditions * How to reorganize and strengthen health manpower, * Renewed education and training programs for key health professions particularly in key areas of public health, nursing, management Service delivery * Innovative approaches to addressing the "unfinished agenda" * More effective and locclly responsive priority programs * How to reorganize public hospitals * New ways to improve the efficiency, accountability, and social mandates of public hospitals * How well can responsibilities in the public health sector * How to strengthen capacity of states and local bodies in health delivery, be decentralized including options for integrating health programs * How to work effectively with untrained medical practitioners * Find ways to eliminate harmful practices and to use untrained practitioners to extend health services in underserved areas *How to take successful but small community health programs * Scaling up of useful public health programs. to scale * How to appropriately integrate various health programs at state * More efficient and effective health services and local levels Health systems outcomes * Obtaining data through special surveys and building systems * Better monitoring, evaluation, and planning of health resources to collect and use information on cause of death, financial and services impact of illness and health services, and satisfaction with health services, with special focus on vulnerable groups Note: TRIPS, trade-related aspects of intellectual property. 128 * Better Health Systems for India's Poor Notes 1. Higher use of hospitalization does not necessarily imply that people are sicker. Chapter 2 (table 2.2) showed that levels of hospi- talization and outpatient visits are lower in India than in other low- income countries. Rates of inpatient and outpatient use were pro- gressively higher in higher-income countries. 2. These classifications are somewhat arbitrary; they are intended to be only indicative, partly because state capacity is not directly measured. Full immunization coverage rates were used as an indica- tor of public service capacity since these are largely delivered through the public sector. 3. Adding the state private sector hospitalization rate changes the rank order of some of the states, but they cluster within the same four categories of states. 4. But this may also indicate the existence of a two-tier delivery system in which the poor rely on the public sector and the better-off use the private sector. 5. As noted later in this chapter, this is still a simplification of the type of choices faced by these states. Institutional deliveries and first-referral hospital care for diseases such as pneumonia and tuber- culosis are critical to the success of these public programs. 6. National Sample Survey rates are based on self-reported regu- lar use and are not adjusted for differences in the age structures of state populations. 7. These recommendations include: merging of the Departments of Health, Family Welfare, and Indian Systems of Medicine; full autonomization of institutions of national importance and the Cen- tral Government Health Scheme (CGHS); handing over the admin- Putting It Together: Raising the Sights of India's Health System * 129 istration of hospital facilities in the capital region to the Delhi gov- ernment; devolving responsibility for most of the 19 centrally spon- sored schemes to the states; establishing an independent organiza- tion to ensure medical education standards; developing more operational policies for working with the private sector; and com- pletely restructuring the Directorate General of Health Services to become a health policy and technical support organization for health information and monitoring, thereby terminating its duties as an administrator of health services, purchaser of drugs and supplies, and executive of drug quality control. 8. The states are Andhra Pradesh, Karnataka, Maharashtra, Orissa, Punjab, Uttar Pradesh, and West Bengal. 9. Several reasons exist for the central government to spend more on health services in states experiencing civil conflict, but this report does not assess options particular to conditions of civil conflict. PART 2 Theory and Evidence ?IW1T7~CHAPTER 5 Health System Framework Health status is widely recognized as both an input and an outcome of broader social and economic developments. As an outcome, health gains arise out of improvements in the general standard of living rather than simply from improvements in the health sector alone.1 As an input, good health improves educational attainment and fos- ters economic growth and political participation. By the same token, ill health and poor health services are increasingly recognized a s major causes of poverty, so efforts to combat poverty ought to cor.- sider the role of health (box 5.1). In describing health systems, the international literature tends to focus on institutions and services whose primary purpose is to pro- tect and improve health.2 Other factors that influence health are usu- ally considered as part of the health system's external environment, that is, outside its boundaries. Even though such an approach inevitably involves some ambiguous or arbitrary distinctions (for example, deworming of children may be part of a school program rather than a health service), defining these boundaries is still usefal in conceptualizing a health system and in examining the choices that are possible for its development. More specifically, here is our approach to the description of India's health system and to the analysis of how to improve it: We adapt a model used in the World Health Report 2000 (WHO 2000c) to describe the actors, functions, and outcomes of a health system. 133 134 * Better Health Systems for India's Poor * To assess whether government should finance or provide certain health services, we use an economic framework that relies both on traditional public economics and a concern for equity. * To address the more difficult questions of the extent to which gov- ernment should intervene and the extent to which markets should play a role, we utilize a framework borrowed from the field of institutional economics. That framework clarifies the areas in which direct government production is needed and those in which other choices should be made for financing or contracting for goods and services. * We also recognize that policymaking is not a linear process, that choices of action are determined by social and political prefer- ences, and that such preferences are influenced by the experiences and interests and values of different groups. Box 5.1 How Chronic Illness Makes People Poor Poor people see good health as a major asset. Illness is dis- abling and costly, especially when a breadwinner or other active adult is struck down. A commonly encountered sce- nario starts with iUness or injury causing an immediate loss of income from work. The afflicted person either goes without treatment, or treatment is received but the costs impoverish the family. Assets are sold and debts taken on. Thus begins a downward spiral from which the family may never recover. Food becomes scarce. Children may become malnourished, and they are withdrawn from school to save money and to work. The poor household becomes perma- nently poorer. When an active adult dies, the ratio of dependants to adults jumps up. The problem worsens when an active adult becomes permanently disabled: the person can no longer work but must still be fed. Health System Framework * 135 A case in point is the family of Padma, a 30-year-old woman who lives in Geruwa, a village on the outskirts of Tatanagar in south Bihar. She has four daughters, the eldest of whom is seven years old. Padma's husband used to work in a dairy, cleaning buffaloes, but he suffers from diabetes and can no longer do labor-intensive work. To raise money for her husband's ongoing treatment, Padma sold her house and land to another resident of the village for Rs 1,300, although the actual value was more than Rs 20,000. She knows she was underpaid but feels indebted to the buyer because he has allowed her to use a small room in the house to shelter her children and ailing husband. The daughters do not go to school, and she is reluctant for them to do so. Padma has taken over the sup- port of the family by carrying wood fuel on her head for a distance of about 10 kilometers every other day. She and her family live hand to mouth, as her earnings are just enough to purchase 2 kilograms of rice a day. Source: Narayan and others (2000). The Descriptive Framework: Health System Actors, Functions, ancd Outcomes Three elements constitute our description of the health system actors (people and institutions), functions (the things they do), anc outcomes (the results of what they do). Actors In our simplified framework of the health system (figure 5.1), we examnine three broad categories of actors: people, the state, and pri- vate sector actors. 136 * Better Health Systems for India's Poor * People are placed at the top of the health system because we con- sider the promotion, maintenance, and recovery of people's health to be the defining characteristics of a health system. * The state is represented by the ministries and departments of health and family welfare at central, state, and local levels, govern- ment entities whose primary activities are most closely related to the health sector. Other important public bodies that affect the health system include the judiciary, the ministry of finance, plan- ning commissions, and sectoral ministries such as those for women and child development, education, and water and sanitation. * Private sector actors include both for-profit and nonprofit health providers, practitioners of allopathic (or Western) and other sys- tems of medicine, and untrained, informal providers. Beyond the private health providers are the private sector actors involved in financing of health care and in the management of other inputs to the health sector. The health system also responds to demand, which is created by people. Demand for health services in turn creates the main health sector market-a point worth emphasizing. The health sector encompasses many types of ancillary markets as well-for example the markets for pharmaceuticals, medical manpower, and diagnos- tics, and these markets do not necessarily compete with each other. These markets are part of the context for the functions of the health system. Functions In our framework, the functions of the health system (shown in the ovals in figure 5.1) are financing, management of nonfinancial inputs, health service delivery itself, and oversight.3 Responsibilities for the first three sets of functions considered- financing, management of inputs, and service delivery-are cur- rently shared between private sector and public actors. Health System Framework * 137 Figure 5.1 Descriptive Framework of the Health System Health System Actors, Functions and Outcomes People inancing Health statUS (chapter 8) Revenue generation Service (chapter 9 Risk pooling delivery Allocation and (chaptr 6 ad 7) Financial status purchasing Public health (chapter 9 | No services (catr9 rPut Ambulatory care management Inpatientrcare Responsiveness to Pharmaceuticals Technology publ(c Consumables Capitchpter9 Oversight / ~~~~~~~~(chapters 3 and 41 4 Policy seeing ~Information, disclosure and advocacy < Develoing parterships Regulation and standard setting J ~_oring an evaluatin Strategic incentives/ Private sector actors The state (chapter 6) (chapters 1, 3, and 4) For-profit Central government Nonprofit State governments Nonformal Local bodies Allopothic Indian Systems Source: Adopted from WHO (2000cl. The main financing functions are the collection of revenues through taxes and insurance premiums, the pooling of funds through insurance, and the allocation of revenues through purchase of services or budget transfers to health providers. The management of nonfinancial inputs is the gathering and use of assets such as human resources, knowledge and software, drugs, phys- ical capital such as medical equipment and buildings, and supplies. 138 * Better Health Systems for India's Poor Service delivery functions are divided among public health serv- ices, ambulatory clinical services, and inpatient, hospital-based, clin- ical services. The nature of the services may be preventive, diagnos- tic, therapeutic, rehabilitative, or palliative. The fourth main function, oversight, is largely a responsibility of the state. The concept of oversight goes beyond the conventional idea of regulation-setting and enforcing rules-to other functions such as developing policy and providing strategic direction to the health system. Oversight may also require mediation between dif- ferent actors in the health sector to set a level playing field (negoti- ating between health providers or financiers) or facilitating the improvement of performance (such as promoting or mandating pro- fessional self-regulation). Other oversight roles are the development of partnerships or networks among health service providers and fin- anciers and the strategic use of incentives to promote public policy objectives. Acquiring and disseminating information about performance, qual- ity, or pricing is another oversight function, one intended to guide the health-related decisions of the public, policymakers, and providers. Even if government does not undertake the monitoring and evaluation itself, it has a primary interest in ensuring that monitoring and evalu- ation occurs and that the information is used for policy and imple- mentation responses. Examples of nongovernmental bodies providing oversight functions are consumer organizations disclosing good med- ical practice and professional bodies engaging in self-regulation. The distinctions made between some of these functions may be somewhat arbitrary because the functions are often closely related. People's demand for, say, healthy deliveries, creates markets for more particular services (fetal ultrasound testing, for example) and products (perhaps new drugs). Oversight activities influence each of the other actors and functions. The management of inputs is tied to the types of services delivered; how they are paid for affects their quality, quantity, and distribution. Public health is also divided among several functions. In its basic definition, public health involves the programs and institu- Health System Framework * 139 tions organized by society to protect, promote, and restore peo- ple's health. By its nature, public health involves the maintenance and improvement of health through collective or social actions. These public health functions are largely divided into oversight activities and the public health programs. The oversight functions are primarily directed at private sector and government actors on behalf of the citizenry. Public health programs are aimed directly at the public, either as individuals, communities, or larger popula- tions. In some cases, such as the control of tuberculosis and malaria, public health services also overlap with ambulatory care and hospital services. Outcomes The health system has three types of outcomes (boxes on the right side of figure 5.1): health status, financial status, and consumer responsiveness.4 The outcomes begin with traditional measures of health status, such as rates for mortality, nutrition, fertility, illness, and disability, which are the most important considerations of a health system. For the sake of simplicity, figure 5.1 does not include the important underlying determinants of health outcomes that fall outside the health sector, such as education, income, and use of san- itation facilities and safe water. The second health system outcome, financial status, is a measure of the financial loss due to illness, which can include direct costs o: health care and earnings lost because of illness. The concern is noi only about how costs affect equity in access, but also the risk of loss of income and assets. Financial protection is particularly importan' to the poor, as the costs of ill health push people into poverty and deepen the levels of poverty. Consumer responsiveness of the health system is measured byr how satisfied the public is with various aspects of health services. I- also includes consideration of whether health services treat people with respect and whether they are provided with protection against malpractice and exploitation.5 Here again, the poor are just as con- 140 * Better Health Systems for India's Poor cerned with dignity and with being treated respectfully by the health system as are those who are wealthier and more powerful. How can India's health system meet its fundamental objectives in an equitable, effective, accountable, and affordable manner? This question, which is also considered in the analysis, raises a second order of objectives relevant for the health system, though not shown in figure 5.1. This level of intermediate objectives may include the following parameters: 6 * Equity. Some minimum of health care should be accessible to all citizens in accordance with their needs, at least in services publicly financed. Equity can also be considered in other terms, such as an equal distribution of public expenditure on health, equal use of health services, or equal health outcomes. * Efficiency and quality. Health services should provide the optimum combination of good outcomes-good health, financial protec- tion, and consumer satisfaction (allocative efficiency), with costs minimized for a given output (cost and technical efficiency). Another way of stating this is that quality of health services should be optimized, which can be considered in terms of technical qual- ity of services (how well the interventions provided work), mana- gerial quality (how well outputs are maximized given the level of inputs), and perception of quality (how well patients are satisfied with services). * Macroeconomic efficiency. Health expenditure should consume an appropriate proportion of GDP. * Consumer choice. The public should have a sufficient choice of providers in both public and private sectors. * Provider autonomy. Doctors and health providers should have the maximum freedom compatible with the attainment of other health system objectives. The overall levels of health system outcomes are important, but so too is the distribution of the results among different geographic Health System Framework * 141 _ __ oA _ !- lAk_{t, r - T areas and populations, and particularly for various vulnerable groups such as the poor, scheduled castes and tribes, women, and the young. Distributional aspects are exam~ined in detail in this report. Although health status is the most important health outcome, we also argue that all three health system objectives are important in India and should be addressed systematically with an eye to the implications for the poor. As we discuss later in this report, some of these objectives can be pursued at little cost and do not necessarily take away resources from activities that direcdly affect health status. This report examines why the health transition that is occurring in India lends urgency to the reform of India's health system financing and to making it more responsive. Dealing with financing and consumer responsiveness can also contribute to the improvement of targeted health outcomes through gains in quality access, accountability, and efficiency of health services. 142 * Better Health Systems for India's Poor Framework for the Consideration of Government Intervention The descriptive framework oudined above provides a good basis for understanding the various health system functions and for orienting its key actors toward a fundamental set of objectives. To fuilly assess the options for government intervention, we need to build on this understanding so we can delve further into the alternatives and the question of how to choose among them. A first-level question is whether the public sector should inter- vene in the health sector. Public economics theory provides one set of answers to this question. The basic rationales for intervention are to: (a) ensure provision of public goods or services with large exter- nalities,7 (b) provide a safety net to alleviate poverty, and (c) correct for market failure, either when access to appropriate health services is limited or when the insurance market has failed. The studies included in this report help to clarify these choices by showing how the poor are affected by health services, identifying the beneficiaries of public sector services, and spotlighting areas in which the private market is failing. Figure 5.2 shows decision points for government intervention that are based on a combination of traditional public economics (public goods, externalities, demand, catastrophic costs) and a concern for equity. This chart makes clear that while the health system may have one set of objectives, the various activities in the health system are not fungible or homogeneous goods and services. If one is concerned with "public health" services that are not individualized curative care serv- ices, then the decision process is much different from that if the con- cern is for curative care. Within curative care, we distinguish services involving "catastrophic" cost, a level of expense we used above to dis- tinguish between ambulatory and inpatient care.8 Public finance economics provides a robust justification for pub- lic oversight of the health system (regulation of private providers and health insurance), for public financing of health services when large externalities exist (communicable disease control, reproductive health programs, health promotion, training), and for targeted pub- Figure 5.2 Decision Points for Government Financing and Provision in the Health Sector Examples Yes Public good? a * Regulation * Health education ... ............. .......... ieducatir . .Examples: * Health *o Disease Ye Sr.rf.cani externalite?N information . control i systems * Reproductive: * Salt & child heolth' iodization. . ............................... Adequate demand? Yes -- Carh cs No -~ II + I I r ,,, , , .. ....................................~~~~~~~~~~~~~~~~~~~~~~~~~~~~. ... .. .. .. .. .. .. .. .. .. No Insurance appropriate? YesI Example: Beneficiaries poor? No Examples: * Hospitali- I * ~~Minimum z iz E .zation --- 1 - .standards and. 4- No Yes :;Examples: nQualiy Cost effective? * ICDS . assurance and Yes: assuracean ; Ar . * Rural I . ., accreditation hospitals * Price No Yes .......................... transparency Ilnsurance leave to regulation Do not provide Finance or 4 - Piublic? | vate? guided * Drug quality provide publicly private control m arket L.................. Note: ICDS, Integrated Child Development Services. Source: Adapted from Musgrove (19991. 144 * Better Health Systems for India's Poor lic programs for the poor (rural health facilities, financing fee exemptions for the poor). However, this approach also has its limi- tations: that is, one must account for historical context, social pref- erences, and political processes and also move beyond the question of whether government should intervene to the question of how gov- ernment should intervene. A Framework for Deciding How to Intervene: Make-Buy-Regulate-Inform A useful new framework for looking at how government can intervene in the health sector takes advantage of tools of institutional economics (Girishankar 1999; Preker, Harding, and Travis 2000). Using two basic characteristics of goods and services, contestability and measurability, the framework lays out the market conditions for the inputs (factors of production) as well as the services (products) in the health sector. Contestability is the extent to which a market can be entered and exited freely. Markets that are highly contestable have low barriers to entry by new providers and mechanisms (such as bankruptcy pro- tection) that allow the easy exit of existing providers. Markets that have low contestability have high barriers to entry by new providers because, for example, entry requires large sunk costs or existing providers wield monopoly power or have geographic advantages. Measurability is the extent to which it is easy to measure important elements such as inputs, processes, outputs, and outcomes. The framework allows policymakers to examine different aspects of the health sector and assess the degree to which direct govern- ment provision may be necessary ("make") and the conditions under which purchasing may be used instead ("buy"). Figures 5.3 and 5.4 give examples of the application of this framework. Figure 5.3 shows that the difficulty of measuring health outputs explains where diffi- culties may lie in dealing with the product markets of health service delivery. Measurement issues also help explain why contracting out for laundry and catering services, whose outputs are quite measura- Health System Framework * 145 Figure 5.3 Measurability and:Contestability of Health Services (Product Markets) Contestability High Medium Low Type I Type 11 Type III I - Type IV Type V Type VI 2. E Nonclinical activities *Clinical interventions *Monitoring/evaluation _ -Management support * High-tech diagnostics o E -Laundry & cotering s _ *Routine diagnostics Type VII Type VIII Type IX 3 *Ambulatory care oPublic health interventions *Policymaking o -Medical *lntersectoral action -Nursing *Inpatient care -- Dental Source: Preker, Harding, and Travis (2000). ble and whose markets are highly contestable, makes more sense than contracting out for ambulatory care, whose market may be highly contestable but whose outputs are very difficult to measure. The situation in the factor market, which comprises the inputs tc the health system, is quite different. Figure 5.4 shows that its meas- urability in factor markets is less of an issue than for services but thas contestability for the factor markets is more important than it is for services markets. This framework can be used not only to outline the nature ot goods and services in the health sector but also to identify policy levers to improve the functioning of the markets. Figure 5.5 indi- cates that buying services is a more feasible option when measura- bility is high, and buying inputs is more feasible when contestabilit7T is high. 146 * Better Health Systems for India's Poor Figure 5.4 Measurability and Contestability of Inputs (Factor Markets) in the Health Sector Contestability High Medium Low Type I Type 11 Type III * Production of consumables * Production of equipment * Production - * Retail of * Wholesale - Pharmaceuticals I - Drugs & equipment - Drugs & equipment - High technology - Other consumables - Other consumables * Large capital stock * Unskilled labor * Small capital stock - Type IV Type V Type VI o E * Basic training * Research ', ._ - Skilled labor - Knowledge a E * Higher education - High-skilled labor Type VIl Type VIII Type IX Source: Preker, Harding, and Travis (2000) Figure 5.5 also shows how various policy tools can improve the functioning of the markets. Disclosing data concerning quality, prices, and effectiveness of health services increases both measura- bility and contestability. In some cases, the availability of such infor- mation makes market provision of goods and services more feasible or improves the potential for government to purchase services. Policies that increase the availability of information, enable health providers and organizations to use such information, and increase patients' understanding of illness and health services have positive effects on contestability and measurability in the health system. Monetization of social benefits, such as measuring and financing services provided to the poor, can also help improve contestability and ensure that these benefits are provided in both public and pri- Health System Framework * 147 Figure 5.5 Make or Buy Decisions, and Inform-Regulate Options to Increase Contestability and Measurability Contestability High < Medium Low Leave to morket Option to vate sectors. Regulation can be used to increase measbuy inputs Option to 0 buy services vate sectors. Reguilation can be used to increase measurability by changing reporting requirements and accountability mechanisms. Regulation can also be used to reduce barriers to entering the mar- ket, such as by reducing the use of noncompetitive bidding to pur- chase equipment and drugs or by blocking inappropriate political interference in public production and purchasing. Contracts can alsc be designed to use quantifiable results to trigger performance- related payments. Other factors besides measurability and contestability are impor- tant in understanding how markets in the health sector work, how tc level the playing field for public and private providers, and how tc get the most out of public and private sector production. Motivatior. and incentives of providers, financiers, and patients are important; 148 * Better Health Systems for India's Poor so are transaction costs, innovation, and experience with regulation and various types of partnerships. In the analysis of the private sec- tor (chapter 6), these factors are examined in detail; they are used to understand the advantages and disadvantages of various arrange- ments and to provide an empirical basis for choosing options rooted in the Indian experience. Notes 1. This relationship has been described in the Poverty Reduction Strategy Sourcebook (World Bank 2000b). 2. Some of widely used descriptions of health systems and health system reform are found in Roemer (1991), OECD (1992), Frenk (1994), Cassels (1995), and Berman (1995). 3. The four critical activities on which we focus our proposals for action (chapter l)-delivery of public health services, delivery of ambulatory care services, delivery and financing of inpatient care, and oversight-can be seen as largely overlapping these functions. 4. In chapter 1, we described the objectives of the health system in terms of making progress against these three types of outcomes. 5. The concept of responsiveness used in this report is somewhat different than the definition used in the World Health Report 2000 (WHO 2000c). That report refers to nonmedical aspects of care and excludes satisfaction per se, which includes medical aspects. We also include legal protection and redress, which were not considered in that report's framework. 6. These are adapted from Barr (1990). Health System Framework * 149 7. In this context, an externality is a benefit that accrues to others (who do not pay for it) when an individual pays for and receives a serv- ice or good. Under such circumstances, private entities may not pro- vide as much of the good or service as would be socially beneficial. 8. Even though several of the decision paths flow through a con- sideration of "cost-effectiveness," activities coming from differen: paths should not be compared head-to-head. That is, one should nob compare a "public health" intervention with a "catastrophic cost" intervention using a cost-effectiveness ratio. Particular programs must be considered only within sets of items that are reasonably comparable (for example, public health campaigns for different dis- eases). (iWr W; CHAPTER 6 The Functioning of the Private Sector Market In this chapter, we use new tools of inquiry to examine the charac- teristics and performance of private health care, and we use Andhra Pradesh and Uttar Pradesh as the "laboratory" for our inquiry (see box 6.1). The object of our attention will be two private health serv- ices that, for the most part, do not compete with each other: Ambu- latory (outpatient) care and inpatient (hospital-based) care. Product markets, such as those for diagnostic services, are not examined here, nor are factor markets, such as those for pharmaceuticals, medical equipment, private medical colleges, and other suppliers of technol- ogy and capital. Using the theoretical model described in the preceding chapter. we organize the analysis of the private market around three consid- erations: * Contestability * Measurability * Interactions with the public sector. To explore contestability, we concentrate on factors that affec- the ease of entering or exiting the marketplace. In the cases of ambulatory and inpatient care such factors include reasons for set- ting up a practice and the manner in which it is set up, including the organizational form chosen for the practice. Access to credit, 151 152 * Better Health Systems for India's Poor Box 6.1 Empirical Findings and Policy Challenges Health System Actors, Functions, and Outcomes People Financing Health status _ Service _ | Financial status | f \ I el~~~~~~~eiver, Input management deliver' |Responsiveness to public| C_i verigt Private sector actors The state Empirical Findings * Private health markets differ significantly from state to state. In the two states studied-Andhra Pradesh and Uttar Pradesh-differences were found among private practitioners in the types of cases they see as well as in their therapies, consultation charges, incomes, costs of doing business, concessions offered to the poor, and interest and participation in national health programs. * Alternative private providers (untrained allopathic doc- tors and practitioners of nonallopathic medicine) make up the largest segment of the private health care market. Most clinics run by these providers lack expensive equip- ment, such as microscopes and operating theaters, but are equipped with basic tools, such as thermometers, stethoscopes, and blood pressure gauges. The Functioning of the Private Sector Market * 153 * Problems with measurability of private sector perform- ance are large, limiting the ability of the private sector to meet health sector objectives. Constraints to entering the market and transaction costs for conducting business are greater for inpatient facilities than for outpatient facilities but are smaller than issues of measurability. * Irrespective of type of provider, quality assurance in staffing and clinical practice is a problem. * Although the factors that motivate health workers in the public and private sectors are similar, important differ- ences exist in motivating factors on the job. Public sector workers perceive the lack of training opportunities and the presence of corruption as drawbacks. * Nearly all private facilities (98 percent) report making concessions to the poor, but they do not document the activity. * Private sector providers received significandy higher rat- ings from patients regarding quality and satisfaction than did public sector providers. * Access to private sources of credit for starting and run- ning new service facilities is relatively easy, and few pri- vate hospitals took advantage of public subsidies or exemptions. * Most private hospitals face problems with public infra- structure such as electricity and drainage, particularly in Uttar Pradesh. Policy Challenges * How can public and private health facilities be made more responsive to the needs of their clients? Can client perceptions of providers' manners and skills be incorpo- rated into training and supervision programs to reverse poor treatment? 154 * Better Health Systems for India's Poor Box 6.1 (confinued) a What steps can government take to improve perform- ance of the private sector? What are the most effective ways of introducing quality assurance systems? Can gov- ernment create a positive environment for the private sector, as opposed to creating an inspectorate raj? * Can govermnent effectively purchase services from the private sector? Can the necessary measurement of per- formance and costs in the private and public sector be created? * Can government further extend coverage of health serv- ices to the poor by encouraging and perhaps formalizing pro-poor measures being claimed by the private sector? Can it effectively monitor health gains by the poor? * How can government take advantage of opportunities to work with private providers? Are there opportunities for the public sector to network with alternate private providers, even though there are legal constraints on untrained medical practice? * Can governments intervene in areas that would make the biggest difference in motivating health workers, notably training opportunities? the role of public infrastructure, and other business constraints are also examined, including an assessment of transaction costs related to issues of labor, revenue collection, and regulation. Incentives and other motivating factors, although not strictly matters of con- testability, play a critical role in shaping the behavior of the private sector, and that role is also explored here. We treat motivation along with contestability largely because it is important in deter- mining whether an individual will choose to work in the private sector and how that individual will behave, and because motivation The Functioning of the Private Sector Market * 155 is not considered a measurable indicator of quality or output of health service. Some of the most important issues of measurability in the pri- vate sector concern performance and pricing. Quality is an impor- tant performance factor in health care, so we look at dimensions of quality of care related to staffing, clinical practices, and manage- ment of quality as well as patients' perceptions of quality. We also examine the measurability of services provided, outputs, and equity questions, particularly how the poor are treated in private practice. The studies shed light on the area of private sector interactions with the public sector. They show the degree to which the private sector is already working with the public sector and reveal where greater partnership may be possible. Participation in national pro- grams, opportunities to improve the quality of care or coverage of services, and other areas for collaboration are explored. Our findings give a better understanding of how the private sec- tor works and provide a basis for governments to make decisions tc make-buy-regulate-inform, as described in the preceding chapter and to provide better oversight of the health sector generally. Sucl. information can reveal areas in which the private sector might ben- efit from further support as well as areas in which it can or should be left alone. It also generates options for government to overcome pri- vate market failures, to form strategic partnerships with the private sector, and to adapt ideas and ways of doing things from the private sector to the public sector. Context: The Private Health Sector in India Levels of private financing of health in India are among the highest in the world, meaning that much of health care is already exposed to market conditions. The analysis of health financing in chapters 8 and 9 shows that reliance on individual private payment for health is not only inefficient and less accountable than other methods cf financing; it also drastically increases vulnerability to poverty. 156 * Better Health Systems for India's Poor Health care in India is also largely provided through the private sector and is highly diverse in methods and quality. The Indian sys- tems of medicine, consisting of the ancient methods of ayurveda, unani, siddha, amchi, naturopathy, and yoga, as well as homeopathy, have always been dominated by private practitioners and private financing, at least in ambulatory care. Doctors trained in Indian sys- tems, about one-fourth of formally trained doctors in India, con- tinue to provide a significant part of the private ambulatory care market. At the time of independence, most allopathic (Western) medicine was provided through government facilities. Private allopathic med- icine was limited to a few missionary hospitals and private industries and thus accounted for only a small part of India's formal health sys- tem. Now, however, more than 80 percent of qualified allopathic doctors are based in the private sector. The picture is complicated by the fact that many public sector doctors also practice privately after hours, either in their own clinics or as consultants in private hospitals. The legality of private practice for public sector doctors differs from state to state, and the fre- quency of this practice is unknown. In the consultations associated with these studies, most participants familiar with the health sector believed that the majority of public doctors also practiced privately, a belief supported by the few small-area studies that have looked at the question (Chawla 2000). Most health statistics are limited to the formal sector (the realm of trained practitioners). A reliable estimate of the number of untrained providers does not exist because such practitioners are not registered-the Supreme Court has held their work to be illegal and has labeled them "quacks." Nonetheless, the largest number of health providers in India, estimated at well over 1.25 million, are untrained and are believed to usually practice allopathic medicine or a mix of systems. Unlike the formally trained allopathic doctors, who concentrate in urban areas and larger towns, most untrained providers are found in rural areas, where they are the first source of ambulatory care for the rural poor (Rohde and Viswanathan 1995). The Functioning of the Private Sector Market * 157 The distribution of private and public services also varies by loca- tion and type of service, an indication that the markets are already quite segmented. The private sector offers a relatively small share of preventive services (less than 10 percent of immunizations deliverec. and 40 percent of prenatal care) but takes a much larger portion o:: curative services, particularly dominating the provision of ambulatory care (figure 6.1). Results from the 52nd Round of the National Sam- ple Survey reveal that the private sector is generally much more pro- rich in its distribution than the public sector (Background Paper 18). (Chapter 7 provides further analysis of the distribution of differen: types of health services in different states and across services and poverty groups.) The poor depend largely on the public sector for services, except for ambulatory care, for which all groups overwhelm- ingly depend on the private sector. Thus a larger private sector is asso- ciated with a more pro-poor distribution of public health services. This chapter addresses the question of how the different seg- ments of the private health sector actually work. Methods The illegality of some types of private practice and the general neg- lect of the private sector by researchers and policymakers have left a gap in the theoretical framework, research tools, and data on the functioning of the private health market. Because no comprehensive registry or sampling frame of private providers exists, and because admission statistics and patient records are rarely available for exam- ination, less desirable methods had to be used for sampling an] examining performance in the private sector. The selection cf providers, for example, was based on a stratification of districts anl random selection of hospitals, which were then used as locators tn randomly select qualified private allopathic practitioners-those having an M.D. (Doctor of Medicine) or MBBS (Bachelor of Mec.- icine and Bachelor of Surgery) degree-and the two types of private practitioners that we term "alternative":1 158 * Better Health Systems for India's Poor Figure 6.1 Public and Private Sector Shares in Service Delivery across India, 1995-96 . I I I I II I II Immunizations Prenatal care m ll Institutional deliveries Hospitalization U Outpatient care 0 1 0 20 30 40 50 60 70 80 90 100 Percent El Public sector * Private sector Source: National Sample Survey Organisation (1998); authors' calculations. * Practitioners of Indian systems of medicine (those practicing non- allopathic systems of medicine, most of whom have formal quali- fications in a nonallopathic system). * Untrained allopathic practitioners (those who usually practice allopathic medicine but have no formal training or qualifications in the system of medicine they are practicing). One consequence of these methods is that the sample selected is more representative of markets having high concentrations of inpa- tient facilities, which tend to serve urban areas and nearby rural pop- ulations. Second, the likelihood of selection is increased for those providers who do more business. This means that the sample tends to represent more successful private providers and to underrepresent providers who work part-time or who are not competing well. The studies may therefore be underestimating some of the problems of The Functioning of the Private Sector Market * 159 .~ ~~~~ B v A private clinic (PHOTOGRAPH By GEETANJALI CHOPRA/THE WORLD BANK) contestability and are likely to be examining providers who have bet- ter quality and measurability. One indicator of the latter point is that among the alternative private providers selected, the proportion of untrained practitioners was lower than in previous studies conductedi in rural Uttar Pradesh. In the sample used by Rohde and VTiswanatha.- (1995), just over half the providers had no training, whereas only one- third had no trairing in the present Uttar Pradesh sample. Ambulatory Care The study sampled 71 clinics of qualified allopathic providers ir.1 Andhra Pradesh and 86 clinics in Uttar Pradesh. It also sampled 1 56 alternative private practitioners in Andhra Pradesh and 84 in Uttux Pradesh. Of the alternative practitioners in Andhra Pradesh, 44 per- cent claimed to practice allopathic medicine, compared with 30 per- 160 * Better Health Systems for India's Poor cent in the Uttar Pradesh sample. Homeopathy was the most com- mon system of medicine used in Uttar Pradesh (57 percent of all sampled alternative practitioners) and the second most common sys- tem after allopathy in Andhra Pradesh (35 percent). Ayurveda was the third most frequent system of medicine used in both states. Prac- ticing more than one system of medicine is relatively common (34 percent in Andhra Pradesh, 19 percent in Uttar Pradesh), usually a combination of allopathy and ayurveda. An additional fifth of those claiming to practice nonallopathic medicines actually provide allo- pathic therapies to their patients (see p. 164). Contestability Among the qualified allopathic practitioners in both states, nearly all were established as proprietary, for-profit clinics, averaging 13 years of practice. The alternative private practitioners also had well-estab- lished practices, averaging 14 years duration in Andhra Pradesh and 13 years in Uttar Pradesh. About 80 percent of alternative private practitioners in both states worked full time as medical practitioners, almost always alone. Likewise, the qualified allopaths practiced ambulatory care in solo clinics. In both states, the main motivation stated by alternative private practitioners for entering practice was to provide a service for those in need. In Andhra Pradesh and among nonallopathic practitioners in Uttar Pradesh, carrying out a family tradition was the next most common reason, whereas the untrained practitioners in Uttar Pradesh cited a good income as the second most common reason (appendix table 6A. 1). The capital needed to operate a clinic is often minimal-only a room and a limited amount of equipment and drugs are needed. Low-cost items such as thermometers, stethoscopes, and blood pressure equipment were nearly always present in such clinics. Patient consultations are the main source of income for nearly 90 percent of all alternative private practitioners. This pattern suggests that the traditional practice of charging little or nothing for a con- sultation but building a margin into medicines may have changed The Functioning of the Private Sector Market * 161 over the past decade (Rohde and Viswanathan 1995). In any case, the incomes of most alternative private practitioners we sampled are now comfortably middle class. Among the alternative providers, the nonallopathic practitioners earned more than Rs 8,000 per month on average, compared with less than Rs 5,000 per month on average for the untrained allopathic practitioners. Only a few alternative private practitioners received money for referring patients to a facility (table 6.1). Pharmaceutical representa- tives have extensively penetrated the alternative private practitioner market in Andhra Pradesh, where nearly all alternative private prac- titioners benefited from free samples of drugs. More than one-thirc' of the untrained practitioners received compensation from drug companies that was tied to their patients' use of pharmaceutica. products, compared with about one-tenth of the nonallopathic prac- titioners. In Uttar Pradesh, where the association of pharmaceutica. companies with alternative private practitioners is not as great, the practice of kickbacks for use of a pharmaceutical product was more common and indicated a more aggressive marketing strategy by the pharmaceutical companies. The data presented here suggest that barriers to entry are quite lovr in the ambulatory care market. Given the absence of any licensing o- accreditation, and without enforcement of regulations concerning Table 6.1 Alternative Private Practitioners Receiving Informal Payments, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH UNTRAINED UNTRAINED ALLOPATHIC NONALLOPATHIC ALLOPATHIC NONALLOPATHIC TYPE OF PAYMENT PRACTITIONERS PRACTITIONERS PRACTITIONERS PRACTITIONERS Cash for patient referrals 7 8 8 14 Free drug samples 86 80 60 56 Rewards for patients' use of productsa 35 13 52 27 a. For Andhra Pradesh, statistically significant at p < 0.01; for Uttar Pradesh, at p < 0.05. Source: Background Papers 20 and 21; authors' calculations. 162 * Better Health Systems for India's Poor educational standards or physical standards, nearly anyone can set up shop as a doctor. The capital requirements for ambulatory clinics are also minimal, and so is the need for access to credit or infrastructure. As a result, the ambulatory care market is often quite competitive. Although their stated motivations are frequently altruistic, practition- ers are subject to influence by pharmaceutical companies. Although few steps have been taken to alter the contestability of the ambulatory care market, it could be done by encouraging quali- fied practitioners to serve markets in which greater needs exist (for example, through special subsidies) or by restricting the entry of low-quality providers through rigorous licensing or accreditation. Nonetheless, the main constraints to influencing the ambulatory care market lie in the area of measurability. Measurability Here we present previously unavailable measurements of the work of outpatient care providers. We first look at information on the workload of private providers, including the number of patients and type of clinical conditions seen. We then examine the prices charged for outpatient consultations in different settings. We conclude by discussing what can be learned from measurements of quality of care and of the relationship of ambulatory care practice to the poor. Service data. Although data are now available on the rates of outpatient consultations in the private sector (see chapter 7), little is known about the workload of private practices, or about the type of patients seen by the private health provider. Data gathered by the present studies show that qualified allopathic practitioners in Andhra Pradesh saw an average of 14 patients per day, compared with 18 patients per day in Uttar Pradesh. These rates are similar to those for alternative private practitioners in the two states: 18 patients per day in Andhra Pradesh and 14 patients per day in Uttar Pradesh, with no statistically significant difference between the untrained allopathic doctors and the nonallopathic doctors. Although both allopathic and nonallo- The Functioning of the Private Sector Market * 163 pathic practitioners treat a high proportion of common ailments such as fever, cough, diarrhea, pain, skin conditions, and sexual problems, some important differences exist. In both states, the nonallopathic practitioners saw more chronic cases, such as diabetes, and in Uttar Pradesh, they had a larger share of the market for skin disease, hyper- tension, and sexual problems. The untrained practitioners also had a bigger portion of the market for cases of fever, diarrhea, and deliver- ies in Andhra Pradesh (appendix table 6A.2). Pricing. Untrained allopathic practitioners charged far less than any other type of practitioner in either state (appendix table 6A.3). Ir. Andhra Pradesh, the nonallopathic practitioners charged about the same per consultation as allopathic generalists but less than specialists. In Uttar Pradesh, the allopathic generalists charged about three times as much as the nonallopathic doctors in an outpatient setting. Among qualified allopathic practitioners, the generalists charged less than the specialists. We also compared the prices of consultation with the fees charged for outpatient visits to small and large hospitals. In both states, the specialist doctors charged considerably more than anyone else and were more expensive at larger facilities. However, in Andhra Pradesh, generalist doctors charged about the same amount per con- sultation at small and big hospitals, which was about one-third higher than the rates charged by the same category of provider at outpatient clinics. Similarly, generalist doctors did not attract high fees at the big hospitals in Uttar Pradesh, although the large hospitals tended to rely on the more expensive specialist doctors. These data suggest that market forces have already differentiated the types of providers in the two states. The untrained allopathic practitioners are placing their fees at a much more affordable level for the poor, whereas other providers can attract higher fees an:c incomes. The fact that nonallopathic doctors charge the same fees as generalist allopathic doctors in Andhra Pradesh, but less in Uttar Pradesh, suggests that these markets have developed quite differ- ently, either because of differences in the public's expectations c r because of differences in the type, quality, or availability of services. 164 * Better Health Systems for India's Poor Similarly, the markets in both states are allowing for considerably higher fees for specialists and hospital-based providers. Quality of care. Without good medical records or the ability to observe patients being cared for in an ambulatory setting, measure- ments of the quality of care in private outpatient clinics are difficult. One way to examine quality is to look at what types of medicines are used. Some 10-20 percent of the homeopaths and ayurvedic doctors used allopathic medicine despite their being unqualified to do so (appendix table 6A.4). Among the untrained allopathic doctors in both states, the majority used drugs, such as injectable antibiotics and steroids, that should require a prescription. Contraceptives and condoms were relatively uncommon. We were able to study 1,800 randomly selected prescriptions given by qualified allopathic doctors in Uttar Pradesh and two other states (Tamil Nadu and Kamataka). The average prescription con- tained more than 3.6 different formulations, compared with 3.0 in the public sector. These results are a crude indicator of overuse of drugs in both private and public sectors. No exact standards exist for the limitation of polypharmacy (the concurrent use of several drugs), especially in the absence of data on the patient's clinical conditions; nonetheless, in most cases no more than two drugs are needed to treat a condition. More drugs are usually unnecessary, difficult to administer, and likely to lead to unwanted drug interactions. Pre- scribing multiple drugs does, however, raise the income of the phar- macist; it also raises the doctor's income if the doctor has a stake in profits from the drugs prescribed. The conclusion of irrational drug use in both the public and pri- vate sectors is supported by the further analysis of prescriptions given for diarrhea. Every one of the prescriptions for diarrhea in each state included antibiotics or antimotility drugs, although the use of such drugs in the common acute diarrhea seen in outpatient settings is not usually indicated (WHO 1995). The data support the view that rational use of drugs is an important area for improving the quality of care by both public and private providers. The Functioning of the Private Sector Market * 165 Other indicators showed that patient safety is a problem among some alternative private providers. No sterilizer was found in a considerable proportion of clinics that had an operating theatre (20 percent in Andhra Pradesh, 29 percent in Uttar Pradesh) nor in an even larger proportion of clinics that offered antibiotic injections (35 percent in Andhra Pradesh, 70 percent in Uttar Pradesh). Not surprisingly, practitioners in these clinics were not trained to con- duct surgery or give injectable antibiotics. If nothing else, these conditions place patients at risk for diseases such as hepatitis B and HIV; Given such practices, treatment guidelines, provider educa- tion, and education of the public may be useful in improving the safety of the public. The appropriate use of diagnostic services such as radiology, microbiology, or hematology is another important consideration in the quality of ambulatory care that is difficult to measure in India. In our studies, a large difference in the use of diagnostic centers was found among the different types of providers in Andhra Pradesh. Nearly all (91 percent) of the untrained doctors referred patients to diagnostic centers, whereas just over half (56 percent) of the nonal- lopathic doctors referred patients to diagnostic centers. In Uttar Pradesh, about three-fourths of both types of alternative private practitioners referred patients to diagnostic centers. More detailed studies are needed to examine whether the particular use of diag- nostic centers is appropriate or whether it merely generates more revenue for physicians. However, a systematic lack of referral to diagnostic centers by some practitioners suggests that nonreferring providers do not have a strong basis for making a diagnosis or mon- itoring progress among their patients, since most clinics do not have a wide range of diagnostic equipment to do the tests themselves. The poor and the private market. The evidence shows that across India the poor overwhelmingly rely on the private sector for ambulatory curative care (chapter 7). Virtually all providers stated that they offered some concessions to the poor (appendix table 6A.5). For the alternative private practitioners, giving such concessions is consis- 166 * Better Health Systems for India's Poor tent with the reasons they gave for entering medical practice. Nearly one-third (29 percent) of all patients seen by alternative private prac- titioners in Andhra Pradesh received some kind of concession because of poverty; the rate was lower in Uttar Pradesh (17 percent). In both states, free care was the most common type of concession, followed by the provision of free samples in Andhra Pradesh and dis- counted prices in Uttar Pradesh. Traditional mechanisms for mak- ing concessions to the poor, such as deferred payment or payment in kind, were quite rare, as was the offer of less expensive care. The only statistically significant difference between types of practitioners in Andhra Pradesh was that nonallopaths were more likely to offer free samples of medicines than the untrained allopathic practition- ers, whereas in Uttar Pradesh, the nonallopaths were more likely to offer free care. However, records on concessions given to the poor were not actually maintained by these providers. As for the qualified allopaths, nearly all offered concessions to poor patients. However, the pattern of their concessions is different between the two states and in contrast to the pattern of alternative private practitioners' concessions. Among qualified allopaths in Uttar Pradesh, free samples of medicine and free care are much more commonly used than any other method; discount prices and deferred payment were other common methods in Andhra Pradesh. These data show that the private market is sensitive to the economic condition of the poor and may help explain why the poor rely on the private sector for ambulatory care. Other studies have suggested that the convenient clinic hours, helpful attitudes of providers, availability of medicines, and knowledge that the provider is a longstanding mem- ber of the community are often reasons why the private sector is pre- ferred over the public sector (Rohde and Viswanathan 1995). Within a given market area, these factors and the ability to give concessions to the poor may well give a provider an edge over competitors. Whether giving concessions to the poor is considered a profit- maximizing behavior or a commitment to the community is a diffi- cult question to test or monitor, especially because treatment for the poor is not well measured. The alternative private practitioners The Functioning of the Private Sector Market * 167 studied did not keep records on the type of concessions given to the poor. As for the qualified allopathic clinics, less than 10 percent of them in Andhra Pradesh had records on assistance given to the poor. while none of them in Uttar Pradesh had such records. Interaction with the Public Sector Private sector participation in various national health programs ir. the two states is quite variable. Broadly viewed, however, the level of reported participation in national programs is much lower than the levels of expressed desire to participate; participation usually occurs through the Pulse Polio campaigns or by referring patients for fam- ily planning (table 6.2). Given the Supreme Court ruling requiring state governments to crack down on medical quackery, careful thought is needed when exploring the partnerships that may be pos- sible with untrained medical practitioners. Nonetheless, among the alternative private practitioners, the disparity between thei. expressed desire to participate and their actual participation in national programs suggests a number of potentially useful opportu- nities for partnerships (see box 6.2). In Andhra Pradesh, the alternative private practitioners also reported having a high opinion of public health and family planning programs, particularly in contrast to their views of government inpa- tient and outpatient services (appendix table 6A.6). However, alterna- tive private practitioners in Uttar Pradesh have a much poorer opinion of govermnent programs. This suggests that in Andhra Pradesh therz is a greater opportunity to increase the coverage of national prograrrs by further involving alternative providers. In Uttar Pradesh, there may be a larger credibility gap with the alternative private practitioners. More than one-half of the alternative private practitioners in Andhra Pradesh belong to professional associations (and about 40 percent in Uttar Pradesh); the majority of members say they find the groups to be useful. They say that training and the setting of guide- lines by the associations would be particularly welcome (figure 6.2)-both would provide practical ways to induce alternative pri- 168 * Better Health Systems for India's Poor Box 6.2 Working with Untrained Private Providers: The Janani Experience Most health care to the rural poor is provided through alterna- tive private practitioners (APPs), a group that consists of non- allopathic practitioners and untrained allopathic providers. In the states of Bihar and Jarkhand, the number of APPs is esti- mated at between 200,000 and 250,000. Janani, a nonprofit social marketing and reproductive health services organization, decided that the only way to reach the rural poor with repro- ductive health services was through the huge arny of APPs. To do so, Janani launched a novel project to franchise reproductive health services in Bihar through a network of APPs. The rural providers, however, were not interested in sell- ing condoms and oral contraceptives, the only two methods they are legally allowed to sell, as the returns were too low. Janani estimated that in a village of 2,000, the monthly income from providing all the contraceptive needs would be about Rs 2, not enough to buy a cup of tea. Janani's strategy is to bundle services to make the partici- pation of rural providers into a viable network. Janani's pro- gram attempts to offer the entire range of reproductive health services, and uses a franchisee network of urban clin- ics run by qualified doctors for referral. APPs are trained to identify reproductive-tract infections, do pregnancy tests, and offer referral to qualified urban-based allopaths who are also part of the network. One qualified doctor provides back- up for 20 APPs. Advertising of the plan is built around the "Butterfly" logo that is displayed at the franchisee's place. Clients pay a fixed fee for all services. For every client referred to the urban doctors, who are franchised by Janani under a "Rising Sun" logo, the rural providers earn an attrac- tive commission. In exchange for participating in the Janani "Butterfly" network and deriving benefits from the aggressive marketing and promotional campaign, the rural providers have to sell condoms and oral contraceptives at all times. The Functioning of the Private Sector Market * 169 To have a significant effect on reproductive health (and be financially rewarding to the rural providers) the pro- gram has to be accommodating to women clients. Toward that end the Janani network requires that each rural provider invited to join the Janani network be, or include, a woman (the partnerships are typically among family members). The women of the community gain the comfort of being able to deal with another woman regarding repro- ductive health; and the provider partnership gains more business and more income by being able to offer that accommodation. Through its aggressive advertising and promotions, the Janani program creates the impression that the "Butterfly" network of rural providers is prestigious and economically rewarding to its members. All rural providers pay a yearly fee of Rs 500. But to encourage adherence to the strictly enforced norms of the network, the rural providers can earn discounts on the fee (of up to 50 percent) according to how closely they operate their clinics tojanani's prescribed standards. The success of the program hinges on an effective man- agement structure, and the current challenge of the Janani program is not establishing its networks but sustaining them. Janani's plan is to keep the organization thin and stream- lined. Its role is increasingly becoming one of oversight rather than implementation. Almost all the fieldwork is out- sourced, and competition among rival entrepreneurs is used as the way to monitor and implement. With regard to mon- itoring, most of the quantifiable indicators are tied to money to create strong pressures within the management system. The program has had much success since its beginning in 1996. From the outset, Janani intended to include two APPs per panchayat, or about 24,000 providers. Janani has already recruited 16,000 APPs and is expected to cover all the pan- chayats by the end of 2001. During 2000, the network pro- vided contraception for 1 million couple years. 170 * Better Health Systems for India's Poor Table 6.2 Participation in National Health Programs by Private Practitioners, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH UNTRAINED QUALIFIED UNTRAINED QUAIJFtED PARTICIPAllON ALLOPATH NONALLOPATH ALLOPATH ALLOPATH NONALLOPATH ALLOPATH Desires to participate in national programs 81 64 31 48 39 61 Does participate in particular national programs Family planning program 42 10 27 24 27 51 Polio eradication program 38 24 31 36 44 69 Malaria control program 25 13 21 8 9 16 Blindness control program 21 8 17 8 2 5 HIV/AIDS control program 19 13 21 8 10 48 Tuberculosis control program 18 13 17 9 0 41 Leprosy control program 12 10 10 8 7 1 Source: Background Papers 20 and 21; authors' calculations. vate practitioners to participate in national programs, to improve the quality of the care they give, to better network with the public sec- tor, and to refer patients as needed. Overall, the present studies point out that despite substantial vari- ation among market locations, the public sector has many opportu- nities to work in partnership with private ambulatory care providers. The kinds of partnership activities that are welcome cost little money, but they take time and effort. The benefits that may result include improved quality of care, greater coverage for health inter- ventions of national priority, and better communications and shar- ing of information among providers (see appendix D for a list of efforts to address the role of private providers in national tuberculo- sis control progams). The Functioning of the Private Sector Market a 171 Figure 6.2 Types of Assistance Favored by Alternative Private Practitioners in Andhra Pradesh and Uttar Pradesh Percent 80 70 60 w - 30 20 i__ 10 ff Nonallopathic APPs Untrained APPs AnhaPradesh Uttar Pradesh Andhra Pradesh Uttar Pradesh | OTraniing * Treatment guidelines O Equipment O Patient materials| Source: Background Papers 20 and 21; authors' calculations. Inpatient Care The studies examined two types of private inpatient facilities, using public sector hospitals for comparison where possible: * Small hospitals (facilities having fewer than 50 beds), commonly called nursing homes. * Large hospitals (those having at least 100 beds). The sample covered 69 small hospitals and 10 large hospitals in Andhra Pradesh, and 64 small hospitals and 12 large hospitals `n Uttar Pradesh. The number of private sector health personnel inter- viewed was 331 in Andhra Pradesh and 421 in Uttar Pradesh. The number of private patients interviewed was 1,17 5 in Andhra Pradesh and 1,620 in Uttar Pradesh.2 172 * Better Health Systems for India's Poor Contestability Issues of contestability generally loom larger for the inpatient mar- ket than for ambulatory care: the investment and maintenance costs are much greater, hospitals have a higher public profile, and inpa- tient operations are more complex. In this section, therefore, we go into more detail in examining how hospitals have been established and expanded over time. We take a more comprehensive look at the issues of costs of credit, labor problems, constraints of public infra- structure, and the regulatory environment. Finally, we examine the questions of motivation of workers in more detail than we did in the previous section. In both Andhra Pradesh and Uttar Pradesh, most private hospi- tals were established in the 1980s. Nearly all the small hospitals sam- pled in both states are incorporated as proprietary companies, largely to provide their owners with a facility in which to practice medicine. In both states, half of the large hospitals were nonprofit hospitals incorporated as a trust or society. Financial factors. To get established, most hospitals in both states required a start-up loan, usually from a commercial bank or other private source. Access to finance was relatively easy for most hospi- tals (appendix table 6A.7). The most significant obstacles to credit cited by hospital owners and managers were the amount of paper- work required for loans and high interest rates. The private hospitals sampled, particularly in Uttar Pradesh, tend to continue to invest in their business and expand (appendix table 6A.8). Virtually all the private hospitals have grown in num- bers of beds since they were established, although the floor-space per bed is considerably smaller than that found in the public sector. Another indication of growth is that, at the time of the sampling, the majority of private hospitals in Uttar Pradesh had invested in buildings during the past year, and about four-fifths had invested in equipment. More than half expected to make capital investments in the next year as well. As was the case with initial financing, gov- ernment played a negligible role in providing financing for invest- The Functioning of the Private Sector Market * 173 ment in private hospitals in both states, with private sources pre- dominating. Regarding cash flow, we found that the vast majority of private hos- pitals sampled are meeting their operating costs from their revenues (appendix table 6A.9), the largest share of which almost always comes from patient fees. More than three-fourths of facilities report rising revenues. These results are in part related to the study design, which biased selection toward facilities with more patients and away from failing facilities. Nonetheless, most small hospitals complained of sig- nificant competition from other hospitals, which may be desirable from a public policy perspective. In Andhra Pradesh, nonpayment was a frequent problem for nearly two-thirds of small hospitals and for one-third of large hospitals, whereas hospitals in Uttar Pradesh were more likely to complain of having too few paying patients. The stud- ies have also demonstrated that private hospitals are able to reduce costs by using significantly smaller buildings than are used in the pub- lic sector and by keeping down the number of paramedical and sup- port staff, factors that may also reduce the quality of care. Business environment. On the question of labor constraints, the stud- ies found that low productivity, an absence of skilled workers, and the high cost of skilled employees are the most significant problems with labor in private hospitals (appendix table 6A.10). Surprisingli, problems between management and unions, regulations on working conditions, and restrictions on laying off workers-all issues that often constrain management action in the public sector-are rela- tively minor concerns in the private sector. Private hospital owners in both states view problems with the public infrastructure, especially drainage, telecommunications, and electricity, as more important than labor issues (appendix tatle 6A. 11). As would be expected given their different levels of develop- ment, infrastructure problems are larger in Uttar Pradesh than in Andhra Pradesh. General regulatory issues are a much more significant problem for private hospitals in Uttar Pradesh than in Andhra Pradesh 174 * Better Health Systems for India's Poor (appendix table 6A.12). In both states, the largest constraint is reported to be high taxes. In Uttar Pradesh, problems with the judi- cial system are cited as the next most common regulatory constraint, followed by tax administration, whereas in Andhra Pradesh, obtain- ing government clearances is ranked second. The relatively low level of difficulty with quality standards and price controls reflects the absence of intervention in these areas by government or professional bodies. These findings show that contestability issues are important for the private hospital sector in the two states. The data suggest that most private hospitals are succeeding financially and are expanding. Among the private hospitals studied for this report, credit has not been a major constraint. In any case, few private hospitals needed financial help from the government. Instead, as demonstrated by their continued investment in equipment, private hospitals tended to rely on banks and their own private resources to expand their size or services. Unless one takes the view that steps are needed to encour- age more private hospitals to enter the market, expanded access to credit is unlikely to improve contestability in the private hospital sector. On the other hand, the most important business constraints appear to be related to public infrastructure and the regulatory envi- ronment, particularly in Uttar Pradesh. Improvements in these areas have greater potential for improving contestability in the private hospital sector. Motivations of Providers Before turning to questions of measurability, we will examine some of the important motivators of participants in the public and private sectors. People who practice in the public health sector are often assumed to be motivated by factors other than those motivating pri- vate health practitioners. In particular, money is said to motivate those in the private sector, and security those in the public sector.3 Rather than dwell on these stereotypes, we examined the extent to which the motivating factors reported by health providers were The Functioning of the Private Sector Market * 175 actually present in their work. Seventeen types of motivating factors were studied; they were grouped into the categories of work envi- ronment, professional fulfillment, work relationships, and personal benefits. The study results reveal relatively few differences between public and private sector health workers in the motivations they say are important (appendix table 6A.13). In Andhra Pradesh, having good working relationships with colleagues, good physical working con- ditions, and challenging work that offers a sense of accomplishment are the most important motivating factors for public and private sec- tor providers. In Uttar Pradesh, good working relationships with colleagues is most important for public and private sector providers, although private sector providers consider good physical working conditions and training opportunities as the next most important factors, whereas public sector workers rank availability of tools for work and career advancement as next most important. Notably, hav- ing a good income or job security is not rated highly by either pub- lic sector or private sector workers. Nonetheless, some important differences exist in the factors tha': motivate public and private sector health workers. Using multivari- ate regression analysis to examine the differences between public and private sector health workers, and adjusting for the effects of profession, type of health facility, age, and sex, four significant dif- ferences were found among the motivating factors in the Andhra Pradesh sample. Compared with private sector health worker;, public sector health workers attach significandy higher importanc e to employment benefits, opportunities to advance to a better jo', and having a superior who recognizes their work. Private sector workers rate the importance of being respected and trusted by clients significantly higher than do public sector workers. No sig- nificant difference emerged between the two groups in the impcr- tance of income or job security, which were rated lower than other common factors. An examination of the degree to which health workers believe that motivating factors are actually present in their work revealed 176 * Better Health Systems for India's Poor greater differences than were found in their stated ideals (appendix table 6A.14). The first observation is that ways to improve working conditions can be discovered from the fact that the rating for the actual presence of each motivating factor is much lower than its desired presence. There are also many more differences between the public and private sectors. Using the same types of multivariate models, we found that the public and private sectors differ signifi- cantly on eight parameters (appendix table 6A.15). In Andhra Pradesh, private sector workers rate their physical working condi- tions as significantly better than those in the public sector and report less political interference. Curiously, public and private sec- tor workers report a similar level of need to pay bribes to accomplish things. Private sector workers also report having more tools and materials to do their work, feel they have greater respect from their clients, and are more likely to be based in a desirable location and have a good income. Public sector workers have higher ratings for having a superior recognize their work and having good employ- ment benefits. The results also pointed out some relevant gender differences in motivating factors. Women rate having sufficient time for personal or family life significantly higher than men do, whereas men rate having challenging work that offers a sense of accomplishment and being respected by clients significantly higher than women do. With regard to actual working conditions, the differences between men and women are even greater. Men rate their working conditions bet- ter than women do on five parameters, while women did not rate any conditions more highly than men did. Men reported having bet- ter training opportunities, more tools to do their job, more respect from clients, more independence from interference by superiors, and better incomes. We also examined the level of unmet needs by looking at the dis- parity between reports of ideal job characteristics and the degree to which those characteristics were reported to be present in the cur- rent job. This analysis may be used to design interventions to influ- ence health worker behavior. The first observation is that the dis- The Functioning of the Private Sector Market * 177 parity between ideal and actual job characteristics among private sector workers is significantly smaller than among public sector workers. High income and job security rate relatively low as unmet needs, possibly because they are found to be satisfactorily present among 43 percent of private employees and 48 percent of public employees. Among the top five motivating factors in both the pub- lic and private sectors, only training opportunities are a top source of disparity in both Andhra Pradesh and Uttar Pradesh. Otherwise, the largest disparity is among factors that are rated as less important. However, what is notable in the public sector in Uttar Pradesh is that the largest disparity involves the problems of corrupt practices and political interference in decisionmaking. These findings suggest that those who work in the public and pri- vate sectors have similar ideals but that large differences exist in how well employment in the public and private sectors satisfies those ideals. Reassuringly, the results also point out that income is not among the most important factors in meeting worker expectations, indicating that other interventions may have a greater effect on sat- isfaction than raising personal incomes. Training opportunities are among the most straightforward and highly sought-after interven- tions that would satisfy health workers in both sectors. Measurability As they do for the ambulatory care market, questions of measurabil- ity loom larger than those of contestability. Here, we look at some of the main issues of measurability related to services, prices, qua.- ity, and dealings with the poor. Service data. In Andhra Pradesh, small private hospitals tended zo provide a more limited range of clinical services than did small pu :)- lic hospitals, whereas large private hospitals were more comprehea- sive than large public hospitals. For example, family planning serv- ices were provided at 76 percent of small private hospitals and )1 percent of large private hospitals; the same services were provided at all small public hospitals but at just 44 percent of large public hos?i- 178 * Better Health Systems for India's Poor tals. Twenty-four-hour emergency services were offered at 60 per- cent of small private hospitals and 82 percent of large private hospi- tals; the same services were offered at 94 percent of small public hos- pitals and 67 percent of large public hospitals. Diagnostic services such as biochemistry, microbiology, ultrasound, and computerized tomography scanning are all more commonly available at private hospitals. Patterns in the range of services offered at private facilities in Uttar Pradesh were similar to those found in Andhra Pradesh. Staffing patterns also affect the type and quality of services pro- vided, and they vary between the two states (appendix table 6A.16). In Andhra Pradesh, where private practice by government doctors is legal, the number of part-time government doctors (almost all spe- cialists) reported to be working in large hospitals is large. In Uttar Pradesh, where private practice by government doctors is not per- mitted, relatively few are reported to practice in private hospitals. General medical duty officers nearly always work filll time at hospi- tals and are rarely government doctors. The number of beds covered by doctors and nurses varies consid- erably between the states and by size of hospital (figure 6.3). Hospi- tals in Andhra Pradesh have a much higher number of beds per doc- tor and per nurse than do those in Uttar Pradesh. The number of nurses is exceptionally low in both states, raising serious questions about the ability to provide quality nursing care. Accounting for three shifts of nurses in a day would mean that one qualified nurse would be expected to care for patients in more than 80 beds at one time in large hospitals in Andhra Pradesh, a ratio that allows for less than 6 minutes of care per patient per day. Regarding surgery, most patients are in a poor position to assess whether surgery is needed or to know who provides the best surgery. Surgery can also be a major source of revenue for a hospital. High rates of surgery may indicate greater specialization by a hospital, but they can also suggest supplier-induced demand for unnecessary serv- ices. Without being able to assess the type of cases seen, their sever- ity, and their clinical outcomes, judgments about the appropriate- ness of care or its quality cannot easily be made. The Functioning of the Private Sector Market * 179 Figure 6.3 Average Number of Beds per Full-Time-Equivalent Doctor and Nurse in Private Hospitals, Andhra Pradesh and Uttar Pradesh 30- 25- 20 - 1 5 10 5- 0 Andhra Pradesh Uttar Pradesh Small hospitals Large hospitals Small hospitals Large hospitals l oPer doctor * Per nurse Note: Small hospitols are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. Cesarean sections (births delivered surgically) constitute one o.^ the few areas in which some judgment can be made on the appro- priateness of surgery in the absence of full clinical information. Although no international standard exists, hospitals with cesarean section rates above 15-20 percent are likely to be overusing the pro- cedure. Yet in the present studies, nearly one-third of all hospital deliveries were cesarean (table 6.3). Such a rate is especially likely to represent overuse because the hospitals examined are not tertiary care institutions, to which women with high-risk pregnancies would have been referred and where higher rates of cesarean section wou-ld therefore be expected.4 Other common measures of hospital performance include bed- occupancy rate and average length of stay. Confident interpretations of such data would require further information on the mix of cases seen or on the outcomes of care. Nonetheless, low bed-occupancy rates 180 * Better Health Systems for India's Poor Table 6.3 Performance Indicators of Private Hospitals, Andhra Pradesh and Uttar Pradesh, 2000 ANDHRA PRADESH UTrAR PRADESH SMALL LARGE SMALL LARGE INDICATOR HOSPITALS HOSPITALS HOSPrIALS HOSPITALS Bed occupancy rate (percent) 40 64 41 42 Average length of stay (days) 6.2 5.5 3.9 5.5 Cesarean section rate (percent) 29.7 29.3 30.1 36.4 Major surgeries per specialist doctor (annual) 36.7 107.4 55.1 92.9 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. show that the private hospitals are operating far below their capacity and that, in general, they are not keeping beds full through long stays. Bed-occupancy rates are slightly higher in small private hospitals than in public hospitals in the same areas, and stays are marginally shorter (STEM 2000). Among large hospitals in Andhra Pradesh, the public sector has a higher bed occupancy than the private sector; among large hospitals in Uttar Pradesh, the opposite is the case. Quality. Available information about the appropriateness of staffing levels and the physical size of facilities provides some indication of limitations in the quality of private hospital services. In addition, the stucly specifically examined quality assurance processes. Overall, lit- tle recording and analysis of health information occur at private health facilities. Few private facilities use clinical protocols, although considerably more claim to maintain medical records, review deaths, and use a patient referral policy (table 6.4). In providing more com- prehensive services, the private sector did relatively well. Twenty- four-hour emergency and ambulance services are nearly always pro- vided at private hospitals in Uttar Pradesh and at large hospitals in Andhra Pradesh; they are less likely to be found in small hospitals in Andhra Pradesh. The Functioning of the Private Sector Market * 181 Table 6.4 Quality Assurance Standards in Hospitals, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH SMALL LARGE SMALL HOSPITALS LARGE HOSPITALS HOSPITALS HOSPtTALS INDICATOR PUBLIC PRIVATE PUBLIC PRIVATE PRIVATE PRIVATE Uses clinical protocols 57 23 62 50 31 33 Keeps medical records for more than 3 years - - - - 66 75 Has procedure to review deaths - 54 - 100 66 75 Has referral policy 91 78 82 70 84 75 24-hour emergency service 57 67 73 80 91 100 Ambulance service 36 23 91 80 83 100 Cesarean section rate of less than 20 percent - 27 - 29 25 10 - Nor available. Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. Hospital managers were quite attuned to the need to improve the quality of care (table 6.5). In particular, managers preferred less inva- sive forms of improvement, such as continuing medical education. hospital quality assurance, doctor and hospital registration, and vol- untary accreditation by a nongovernmental organization. Compul- sory accreditation through the government was not seen as a gooc. option. Patient satisfaction is another important dimension of qualityr assurance. As a routine tool for quality assurance, patient satisfaction surveys are still not commonly used in the private sector, although they are being introduced in the public sector. Some interesting findings on overall satisfaction with services emerged in Andhra Pradesh (table 6.6). Satisfaction was much higher among users cf private sector facilities. The private sector seems most able to satisfy the richest quintile of patients. Public facilities failed to satisfy users from the wealthiest 60 percent of the patients. 182 * Better Health Systems for India's Poor Table 6.5 Private Hospitals Whose Managers Favor Various Procedures for Improving Hospital Quality, 2000 (percent) ANDHRA PRADESH UTTAR PRADESH SMALL LARGE SMALL LARGE PROCEDURE HOSPrTALS HOSPITALS HOSPITALS HOSPITALS Hospital registration 89 88 75 92 Renewing hospital registration 73 88 55 82 Registering doctors 92 88 88 92 Renewing doctors' registration 77 88 50 73 Voluntary hospital accreditation by NGO 62 83 75 83 Compulsory hospital accreditation by government 44 57 30 58 Hospital quality assurance programs 84 88 89 100 Continuing education for doctors 95 100 97 100 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Paper 20; authors' calculations. The poor and the private market. In their dealings with the poor, pri- vate inpatient facilities show results similar to those for private providers of ambulatory care. Nearly all private hospitals claim to make concessions for poor patients (table 67). However, the pattern of concessions differs between the states and by practitioner. Dis- count pricing, rather than free care, is the most common type of concession among hospitals. Free samples of medicine are less com- mon in Andhra Pradesh than Uttar Pradesh, and less expensive care is a more likely option in Uttar Pradesh hospitals than it is for any other private provider. These differences in methods for dealing with the poor make standard approaches and measurement of assis- tance more difficult. Despite these concessions, the clients served by the private sector are generally not the poor. In both Andhra Pradesh and Uttar Pradesh, the public sector sees significantly higher proportions of poor patients than the private sector, although the private sector also sees more patients (see chapter 7). In Andhra Pradesh, for example, The Functioning of the Private Sector Market * 183 Table 6.6 Proportion of Patients Satisfied or Very Satisfied with Overall Quality of Care at Public and Private Health Facilities, Andhra Pradesh, by Type of Facility and Wealth of Patient (percent) PRIVATE-PUBLIC PUBLIC FACIUTIES PRIVATE FACILrITES RATIO, FACILITY OR INCOME QUINTILE (1) (2) (21) Clinics and primary health centers 13 30 2.2 Small hospitals 16 23 1.4 Large hospitals 10 29 2.8 Income quintile 1 (lowest) 20 24 1.2 2 16 14 0.9 3 9 20 2.2 4 13 22 1.8 5 (highest) 8 33 3.9 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. Table 6.7 Private Hospitals Offering Concessions to the Poor, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH SMALL LARGE SMALL LARGE TYPE OF CONCESSION HOSPITALS HOSPITALS HOSPITALS HOSPITALS Any 99 80 100 100 Free core 74 50 81 75 Free samples of medicine 62 30 81 92 Discount prices 71 60 88 92 Deferred payment 33 20 52 50 Less expensive care 26 20 53 83 Payment in kind 9 10 11 17 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. the proportion of patients seen at small hospitals in the public sec- tor decreases with each rise in wealth quintile, whereas it rises in thz private sector. In practice, the private hospitals are catering to a wealthier consumer than the public sector. 184 * Better Health Systems for India's Poor The data shown here indicate that the measurement of activities, prices, and performance is particularly weak in the private hospitals studied. * Most hospitals are quite casual about the setting of fees, allowing doctors to set their own. * Tariffs are frequently not published, making it more difficult for the consumer to compare prices or to know what to expect. * Most patients believe that the billing at private facilities is fair. * No formal system exists for determining exemptions or conces- sions to the poor; the doctor is nearly always the one who decides whether a patient should be given a concession. * Few facilities keep records on the concessions they make to the poor. * The majority of hospitals claim to maintain medical records for at least three years, but the quality of those records is questionable. * Few private hospitals actually collate records on service outputs, diseases and conditions, or clinical outcomes, suggesting that internal clinical quality assurance mechanisms are largely nonex- istent and that vital information is not available to consumers and govermment planners. Inproving the measurability of the private sector is clearly important if public intervention is to make the health sys- tem more accountable, efficient, and pro-poor. The Functioning of the Private Sector Market * 185 In the next section, we examine interactions between the public and private hospital sectors and explore some opportunities for action. Interaction with the Public Sector Overall, private hospitals seem to receive little by way of public sub- sidies (table 6.8) despite the high profile of the debate over this issue. The practice of public subsidy seems most notable in Uttar Pradesh, where half of large hospitals have received some tax exemptions as nonprofit organizations, and one-third have received discounted land, but few small hospitals in either state have received subsidies. Compared with other private providers, private hospitals, partic- ularly the big hospitals, had a higher level of current participation in national health programs (table 6.9). Nearly all large hospitals have been involved in polio eradication and the family planning program. However, the desire to participate in national programs was consid- erably lower in Andhra Pradesh than in Uttar Pradesh. The levels of interest expressed again signal an opportunity for government to work more closely with private hospitals, particularly in Uttar Pradesh. Table 6.8 Proportion of Private Hospitals that Have Received Public Benefits, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH PUBUC BENEFIT SMALL HOSPITALS LARGE HOSPITALS SMALL HOSPITALS LARGE HOSPITAU Tax exemptions 9 20 9 50 Duty exemptions 1 10 2 25 Reduced utility charges 0 0 0 8 Discounted or free land 1 0 2 33 Low interest loan 4 0 2 0 Note: Small hospitals are those with fewer than 50 beds; lorge, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. 186 * Better Health Systems for India's Poor Table 6.9 Participation in National Health Programs by Private Hospitals, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH PARTICIPATION SMALL HOSPrrALS LARGE HOSPITALS SMALL HOSPITALS LARGE HOSPITALS Desires to participate in national programs 54 40 89 75 Does participate in particular national programs Family planning 62 80 53 83 Polio eradication 67 90 80 92 Malaria control 28 50 16 50 Blindness control 25 60 11 58 HIV/AIDS control 42 40 14 50 Tuberculosis control 35 70 30 58 Leprosy control 15 50 11 42 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; autlors' calculations Concluding Remarks This chapter revealed new information about the way the private health sector behaves in India, with particular attention given to seg- ments of the market in Andhra Pradesh and Uttar Pradesh. Wide variations were found in the behavior of different market segments and in different geographic areas. Private providers face many obsta- cles not encountered by the public sector and for the most part have dealt with those obstacles without assistance from government. At the same time, the private sector appears to be interested in working more closely with the public health sector and recognizes the need for greater government oversight and involvement in quality assur- ance and regulation. The private sector tends not to measure its performance very closely or to provide much information about its pricing. Otherwise it appears to be more responsive to patient satisfaction than the pub- lic sector. The Functioning of the Private Sector Market * 187 As discussed in part 1, inforrnation about the factors that motivate health personnel, transaction costs, the regulatory environment, measurement of performance, and use of partnerships creates options for government intervention. Because the private sector plays such an important role in providing health services and has been ignored by government for so long, a more active and strategic engagement with the private sector should allow it to contribute more fully to improving Indians' health and protecting their finan- cial well-being. Appendix Table 6A.1 Distribution of Main Reasons Given by Alternative Private Practitioners in Andhra Pradesh and Uttar Pradesh for Becoming a Medical Practitioner (percent) ANDHRA PRADESH UTTAR PRADESH UNTRAINED UNTRAINED REASON ALLOPATH NONALLOPATH ALLOPATH NONALLOPATH Serve people 46 33 48 51 Family tradition or obligation 29 27 12 25 Professional ambition 15 20 8 14 Good income 9 20 32 10 Total 100 100 100 100 Number of providers 85 66 25 59 Note: Andhra Pradesh: chi2 = 4.75, 3 degrees of freedom; p = 0.2; Uttar Pradesh: chi' = 6.98, 3 degrees of freedom; p = 0.07. Source: Background Papers 20 and 21; authors' calculations. 188 * Better Health Systems for India's Poor Table 6A.2 Clinical Conditions Treated by Alternative Private Practitioners over Two Days, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH UNTRAINED NON- UNTRAINED NON- CONDMON TREATED ALLOPATH ALLOPATH p VALUE ALLOPATH ALLOPATH p VALUE Fever 95.6 85.2 0.03 76.0 89.8 0.1 Cough 94.1 94.3 1.0 68.0 81.4 0.2 Pain 92.6 86.4 0.2 80.0 81.4 0.9 Diarrhea 79.4 60.2 0.01 64.0 74.6 0.3 Skin disease 67.6 76.1 0.2 44.0 72.9 0.01 Hypertension 60.3 61.4 0.9 44.0 67.8 0.04 Diabetes 30.9 70.5 <0.001 16.0 39.0 0.03 Tuberculosis 32.4 38.6 0.4 8.0 30.5 0.02 Vaginal discharge 32.4 37.5 0.5 4.0 35.6 0.001 Sexual dysfunction 30.9 39.8 0.2 12.0 44.1 0.03 Penile discharge 29.4 28.4 0.9 8.0 16.9 0.2 Pregnancy-related problem 32.4 23.9 0.2 16.0 22.0 0.5 Insect bite 27.9 27.3 0.9 12.0 3.4 0.2 Fracture 817.6 18.2 0.9 8.0 6.8 0.8 Delivery 19.1 8.0 0.04 4.0 8.5 0.4 Snake bite 16.2 8.0 0.1 4.0 0.0 0.3 The Functioning of the Private Sector Market * 189 Table 6A.3 Patient Fees for an Outpatient Consultation in the Private Sector, Alternative Practitioners Compared with Qualified Allopaths, Andhra Pradesh and Uttar Pradesh (rupees except as noted) ANDHRA PRADESH UTTAR PRADESH RATIO OF RATIO OF CHARGE TO CHARGE TO THAT OF THAT OF LOWEST-COST LOWEST-COST PROVIDER CHARGE PROVIDER CHARGE PROVIDER Altemative practitioner Untrained allopath 10 1 14 1 Nonallopath 36 3.4 21 1.5 Qualified allopoth Clinic Generalist 33 3.2 66 4.5 Specialist 58 5.6 - - Small hospital, outpatient department Generalist 45 4.3 54 3.7 Specialist 64 6.1 104 7.2 Large hospital, outpatient department Generalist 50 4.8 37 2.6 Specialist 82 7.9 170 11.8 - Not available. Note: Small hospitals are those with fewer than 50 beds; large, at least 1 00 beds. Source: Background Papers 20 and 21; authors' calculations. 190 * Better Health Systems for India's Poor Table 6A.4 Allopathic Therapies Offered at Clinics of Alternative Private Practitioners, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH UNTRAINED NON- UNTRAINED NON. THERAPY ALLOPATH ALLOPATH p VALUE ALLOPATH ALLOPATH p VALUE Paracetemol 84 19 <0.001 92 12 <0.001 Oral antibiotic 79 18 <0.001 76 10 <0.001 Injectable antibiotic 71 11 <0.001 68 10 <0.001 Oral rehydration fluids 68 21 <0.001 64 29 0.003 Intravenous fluids 71 13 <0.001 52 17 0.001 Steroids 25 5 <0.001 32 5 0.002 Oral contraceptives 38 8 <0.001 28 12 0.08 Condoms 28 8 0.001 12 9 0.6 Surgery 10 6 0.3 24 5 0.02 Source: Background Papers 20 and 21; authors' calculations. Table 6A.5 Concessions Offered to the Poor by Alternative Private Practitioners, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH UNTRAINED NON- QUALIFIED UNTRAINED NON- QUALIFIED TYPE OF CONCESSION ALLOPATH ALLOPATH ALLOPATHS ALLOPATH ALLOPATH ALLOPATHS Any 99 99 97 100 98 100 Free care 85 84 82 76 93 95 Free samples of medicine 43 59 65 20 22 99 Discount prices 43 33 69 40 44 21 Deferred payment 13 13 35 8 14 1 Less expensive care 9 9 23 12 12 26 Medical camps for poor 4 5 .. 12 4 Payment in kind 3 6 7 5 0 0 .. Negligible. Source: Background Papers 20 and 21; authors' calculations. The Functioning of the Private Sector Market * 191 Table 6A.6 Alternative Private Practitioners Who Rate Government Health Programs as Good or Very Good, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH UNTRAINED UNTRAINED PROGRAM ALLOPATH NONALLOPATH ALLOPATH NONALLOPATH Family planning 90 78 48 51 Public health 81 75 27 48 Government hospitals 66 61 39 44 Government outpatient services 47 51 22 46 Source: Background Papers 20 and 21; authors' calculations. Table 6A.7 Private Hospitals Experiencing Moderate or Severe Obstacles to Credit, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTrAR PRADESH SMALL LARGE SMALL LARGE OBSTACLE HOSPITALS HOSPITALS HOSPITALS HOSPITALS Too much paperwork for loan 43 25 48 25 Interest rate too high 23 25 40 45 Excessive collateral required 26 25 38 25 Need for connection with bank officials 15 25 32 25 Problems with letters of credit 16 0 14 8 Corruption of bank officials 14 0 5 0 Lack of supplier credits 10 25 12 1 8 Problems with money transfers 6 25 10 0 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. 192 a Better Health Systems for India's Poor Table 6A.8 Private Hospitals Making or Planning Capital Investments, Andhra Pradesh and Uttar Pradesh, by Source of Funds (percent) ANDHRA PRADESH UTTAR PRADESH SMALL LARGE SMALL LARGE INVESTMENT AND SOURCE HOSPiTALS HOSPITALS HOSPITALS HOSPITALS Building investment in past year 32 50 55 75 Bank loon 40 67 33 50 Government 0 0 3 0 Personal resources 50 33 57 25 Donor 5 0 3 25 Equipment investment in past year 57 50 78 92 Bank loan 45 40 57 44 Government 3 0 2 22 Personal resources 56 20 42 0 Donor 6 40 0 33 Capital investment planned 23 10 56 58 Financial partner welcomed 44 0 22 57 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. Table 6A.9 Revenue Issues Affecting Private Hospitals, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UITAR PRADESH SMALL LARGE SMALL LARGE REVENUE ISSUE HOSPITALS HOSPITALS HOSPITALS HOSPITALS Too much competition from other hospitals 64 38 60 36 Nonpayment by patients 65 33 36 25 Too few paying patients 27 22 40 27 Nonpayment by government 7 0 16 30 Nonpayment by employers under contract 19 17 8 11 Donations reduced 2 0 2 0 Revenues exceeding operating expenses 87 88 91 92 Income rising 77 86 79 83 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 2 1; authors' calculations. The Functioning of the Private Sector Market * 193 Table 6A.10 Labor Issues in Private Hospitals, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH SMALL LARGE SMALL LARGE ISSUE HOSPITALS HOSPITALS HOSPITALS HOSPITALS Low productivity of labor 41 33 48 25 High cost of skilled employees 27 50 45 17 Scarcity of skilled employees 40 33 48 27 Missed work due to illness 17 50 18 0 Absenteeism 34 33 19 17 Lack of skilled management 24 33 13 0 Regulations on working conditions 14 33 19 8 Restrictions on laying off workers 12 17 10 0 High staff turnover 11 33 14 0 Union activities/restrictions 6 17 2 0 Seasonal shortages of unskilled labor 9 0 8 8 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21 and authors' calculations. Table 6A.1 1 Public Infrastructure Issues Affecting Private Hospitals, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH SMALL LARGE SMALL LARGE ISSUE HOSPITALS HOSPITALS HOSPITALS HOSPITALS Electricity breakdowns 62 50 94 83 Poor drainage 45 50 72 55 Telecommunications breakdowns 45 25 63 67 Inadequate water supply 39 53 64 33 General waste disposal problems 25 25 63 50 Biomedical waste disposal problems 23 13 61 50 Availability of transportation 23 0 27 0 Access to land 22 25 21 9 Availability of office space 22 0 12 0 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. 194 * Better Health Systems for India's Poor Table 6A. 12 Regulatory Issues Affecting Hospital Managers, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH SMALL LARGE SMALL LARGE ISSUE HOSPITALS HOSPITALS HOSPITALS HOSPITALS High taxes 26 40 76 50 Judicial system 14 40 66 50 Tax administration 15 20 59 42 Government clearances 22 40 57 64 Unfair competition from other providers 20 60 46 42 Government official corruption 20 40 46 50 Police corruption 14 20 40 42 Labor union regulation 12 50 39 40 Quality standards 1 8 20 24 17 Price controls 8 0 33 25 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 2 1; authors' calculations. The Functioning of the Private Sector Market * 195 Table 6A.13 Ideal Job Characteristics Reported by Public and Private Sector Health Workers, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH PUBlIC PRIVATE PUBLIC PRIVATE JOB CHARACTERISllC SECTOR SECTOR SECTOR SECTOR Work environment Good physical working conditions 90.9 90.5 91.7 94.2 Knowing what you are expected to do and achieve at work 83.8 72.2 90.5 89.6 Freedom from political interference in decisionmaking 80.6 79.5 76.9 85.9 Not needing to pay bribes to get what you want 73.1 75.9 87.7 87.9 Knowing you can keep your job as long as you want 57.8 52.9 69.1 73.3 Professional fulfillment Training opportunities to improve skills or learn new skills 90.5 82.5 94.2 93.4 Challenging work that offers a sense of accomplishment 88.8 84.4 91.5 91.5 Tools and materials to use skills fully on the job 83.0 84.4 95.7 93.0 Good opportunities to advance to a better job 75.1 64.1 94.6 77.7 Work relationships Good working relationship with colleagues 93.8 91.0 98.2 97.8 Superior recognizes good work 74.7 62.6 93.6 68.8 Respected and trusted by clients 69.7 74.5 94.1 93.2 Independence from interference by superiors 59.2 50.7 54.7 61.6 Personal benefits Desirable location (e.g., one with good schools) 83.5 73.6 93.9 84.3 Sufficient time for personal or family life 78.5 75.7 93.7 85.5 Good employment benefits (e.g., pension, housing) 78.5 54.8 93.9 65.8 Good income 76.3 66.2 84.6 73.4 Source: Background Papers 20 and 21; STEM (2000). 196 * Better Health Systems for India's Poor Table 6A. 14 Presence in Current Job of Ideal Job Characteristics, as Reported by Public and Private Health Sector Workers, Andhra Pradesh and Uttar Pradesh (percent) ANDHRA PRADESH UTTAR PRADESH PUBUC PRIVATE PUBLIC PRIVATE JOB CHARACTERISTIC SECTOR SECTOR SECTOR SECTOR Work environment Good physical working conditions 53.3 61.1 42.6 78.9 Knowing what you are expected to do and achieve at work 61.8 61.6 64.3 79.5 Freedom from political interference in decisionmaking 69.3 79.4 35.3 89.9 Not needing to pay bribes to get what you want 69.9 74.6 15.3 89.1 Knowing you can keep your job as long as you want 44.5 50.4 48.2 71.1 Professional fulfillment Training opportunities to improve skills or learn new skills 42.9 29.9 21.7 49.0 Challenging work that offers a sense of accomplishment 61.6 62.1 51.8 74.0 Tools and materials to use skills fully on the job 35.5 52.3 40.1 75.5 Good opportunities to advance to a better job 37.1 33.9 30.7 41.4 Work relationships Good working relationship with colleagues 83.5 79.5 80.5 90.1 Superior recognizes good work 50.8 44.2 55.5 43.4 Respected and trusted by clients 57.3 67.4 76.0 91.5 Independence from interference by superiors 39.4 39.8 38.5 49.9 Personal benefits Desirable location le.g., one with good schools) 51.0 53.6 36.7 78.7 Sufficient time for personal or family life 35.7 36.0 50.0 55.0 Good employment benefits le.g., pension, housing) 45.8 23.4 42.7 22.2 Good income 36.1 38.1 42.8 45.2 Source: Background Papers 20 and 21; STEM 120001. The Functioning of the Private Sector Market * 197 Table 6A.1 5 Presence of Ideal Job Characteristics in Andhra Pradesh: Differences in Ratings between Public and Private Sector, by Characteristic of Health Worker CHARACTERISTIC OF HEALTH WORKER THAT PREDICTED SIGNIFICANTLY HIGHER RATINGS OF PRESENCE OF JOB CHARACTERISTIC JOB CHARACTERISTIC Work environment Good physical working conditions Private sector; small hospitals; specialists Freedom from political interference Private sector; specialist and general doctors in decisionmaking Professional fulfillment Tools and materials to use skills fully Private sector; small hospitals; specialist and on the job general doctors; males Work relationships Superior recognizes good work Public sector; nurses; younger age Respected and trusted by clients Private sector; clinics or primary health centers; males Personal benefits Desirable location (e.g., one with Private sector good schools) Good employment benefits Public sector (e.g., pension, housing) Good income Private sector; large hospitals; specialist and general doctors; males Note: Multivariote models are built as follows: Motivating factor = bintercept + bpublic + bprofession + bfcaciliy size + bse. + b.g. + error. The comparison group for profession is nurs- es (versus specialists and general allopoths), and for facility is clinics (versus small hospi- tals, large hospitals, and diagnostic centers). Characteristics are ordered from largest to smallest proportion of variance explained in the model. Small hospitals are those with fewer than 50 beds; large, at least 100 beds. 198 * Better Health Systems for India's Poor Table 6A.16 Average Number of Full-Time and Part-Time Nurses and Private and Government Doctors Working in Private Hospitals, Andhra Pradesh and Uttar Pradesh ANDHRA PRADESH UTrAR PRADESH PUBLIC PRIVATE PUBUC PRIVATE PRACTITIONER SECTOR SECTOR SECTOR SECTOR Certified nurses Full time 2.1 20.2 7.0 27.1 Part time 0.1 0.0 0.1 0.0 Private doctors Specialists Full time 1.3 8.0 2.4 17.9 Part time 1.4 6.4 3.6 9.1 Generalists Full time 0.9 7.7 0.9 3.7 Part time 0.1 0.0 0.2 0.0 Government doctors, part time Specialists 0.6 8.0 0.2 1.0 Generalists 0.0 0.0 0.0 0.0 Note: Small hospitals are those with fewer than 50 beds; large, at least 100 beds. Source: Background Papers 20 and 21; authors' calculations. Notes 1. The two types of providers were combined because they were sampled together and later distinguished based on their answers in the questionnaires; allopaths were given a different questionnaire. 2. Patient exit interviews come from both outpatients and inpa- tients at hospitals. 3. The researchers were unable to distinguish the results of those who practice in both the public and private sectors, as respondents were categorized according to their work location. The Functioning of the Private Sector Market * 199 4. No universal agreement exists as to the optimal cesarean sec- tion rate, but some authors have argued that a rate of about 6 per- cent to 8 percent would be an appropriate response to the common medical indications for the surgery (Francome and Savage 1993). During a consensus-building exercise by the World Health Organi- zation, a rate of 10-15 percent was considered appropriate (W7HO 1985), although the rate was defined somewhat arbitrarily. The U.S. government, in its Healthy People 2000 strategy, set a goal of reduc- ing the cesarean section rate to 15 percent by 2000 (U.S. Depart- ment of Health and Human Services 1991). Where the population is impoverished and general and reproductive health are poor (con- ditions leading to high-risk pregnancies), some have argued that a rate of 20 percent may be appropriate. CHAPTER 7 piL Setting National Health Care Priorities and Ensuring Equitable Delivery of Public Sector Services This chapter considers four aspects of public health care spending: (a) the extent to which centrally sponsored schemes meet India's health care needs, (b) the division of service utilization between the public and private sectors, (c) the degree of equity in current public health care spending, and (d) the implications of current patterns of public health care spending for health outcomes (see box 7.1 for a summary of the empirical findings and policy challenges discussed in the chapter). Do Centrally Sponsored Schemes Meet India's Health Care Needs? National priority programs, or centrally sponsored schemes, have traditionally been justified on one or more of the following grounds they provide a public good; significant externalities exist; the magni- tude of the health problem is enormous; beneficiaries are predomi- nantly poor; and affordable, effective intervention is available. The health issues addressed by these programs have been deemed to be of national importance. Rapid population growth and the emerging 201 202 * Better Health Systems for India's Poor Box 7.1 Empirical Findings and Policy Challenges Health System Actors, Functions, and Outcomes People Financing Health status _ Service Financial status | deliver \nput management delivery Responsiveness to public Private sector actors The state Empirical Findings * National programs continue to address much of the dis- ease burden in India. * As India's disease profile changes, important public health problems, including cardiovascular disease, men- tal illness, injury prevention, and the risks associated with tobacco, will need to be addressed. * The poor in India smoke more and use nonsmoking tobacco and alcohol more than the population as a whole, placing them at higher risk for developing certain noncommunicable diseases. Findings from the analysis of household data from the 52nd round of the National Sample Survey: Setting National Health Care Priorities * 203 National Level * The private sector accounts for 82 percent of outpa- tient care, 56 percent of hospitalizations, 46 percent of institutional deliveries, and 40 percent of prenatal care visits. It provides only 10 percent of immunizations. * As in most developing countries, richer households purchase more curative health care from the private sector than do poorer households. * Public spending on curative services is highly pro-rich: nationwide Rs 3 is spent on the richest 20 percent for every Rs 1 spent on the poorest 20 percent. * The distribution of public outpatient services are pro- poor, particularly at the primary health center level. Inpatient services are less likely than outpatient serv- ices to reach the poor. * The public sector provides most health services to Indians living below the poverty line, accounting for 93 percent of immunizations, 74 of prenatal care, 69 percent of institutional deliveries, and 60 percent of hospitalizations. The exception is outpatient care, where the private sector accounts for 79 percent of the care for the poor. Regional Level * The use of public services is more equitable in urban areas than in rural areas. * In Kerala, Tamil Nadu, and Maharashtra, the distribu- tion of public spending on health is nearly uniform across income groups. In six states, however, more than Rs 4 goes to the richest quintile for every Rs 1 that reaches the poorest. * Health status outcomes are better in states in which public spending on health care is more equitable. 204 * Better Health Systems for India's Poor Box 7.1 (confinued) Policy Challenges * How can implementation of national public health pro- grams be strengthened by addressing critical gaps in pro- gram management, technical leadership, and communi- cation? * How and when should effective programs that address emerging public health problems-cardiovascular dis- ease, mental illness, injuries, and the health conse- quences of smoking-be developed? * How can the distribution of public spending be improved so that a greater share of spending reaches the poor? * How can the quality and accountability of public and pri- vate health services be improved? HIV/AIDS epidemic, for example, pose significant threats to India's future economic and social prospects. Widespread health problems with significant externalities are addressed by programs of disease control, such as those for tuberculosis, leprosy, and malaria, and by safe motherhood and child health interventions such as immuniza- tions. Curing one case of tuberculosis, for example, can prevent the spread of infection to as many as 15 people (WHO 2000a). Iodizing salt and draining swamps for malaria control are public goods whose benefits are shared across the population. Integrated Child Devel- opment Services, a program that addresses malnutrition, is intended to target benefits to the poor. Many national programs have received additional support in recent years because they have introduced new cost-effective tech- nologies. Intraocular lenses are now implanted as part of blindness control. Tuberculosis control now uses directly observed treatment, short courses. Multidrug therapy is used for leprosy control, and a Setting National Health Care Priorities * 205 new mix of case management and prevention techniques is used to control malaria. In 1990, centrally sponsored schemes addressed conditions affecting more than 40 percent of the deaths and 46 percent of the DALYs lost in India.' By 1998 the number of deaths and DALYs lost associated with these conditions had fallen 15 percent, but they still accounted for 34 percent of all deaths and 42 percent of DALYs lost (table 7.1). Most centrally sponsored programs are available to all Indians, but the programs are expected to ensure that basic services are available to the poor. Program activities for leprosy control and malaria control target the poorest states and they also target the poorest districts within states in which these diseases are most prevalent. These pro- grams increase the welfare of the poor. An analysis of employability and earnings of people with leprosy in Tamil Nadu, for example, found that eliminating the deformity would raise the probability of gainful employment from 42 percent to 78 percent and increase the annual earnings per employed person 119 percent (World Bank 2001). The combined effect of increased employment and increased earnings would triple the annual earnings of all people with leprosy. The large burden of disease, large externalities, poverty dimen- sions, and cost-effectiveness of the interventions covered by the cur- rent national priority programs suggest that they remain a good fit for public intervention. The main challenge is improving their implementation. Because India's disease profile is changing, several major health con- ditions that account for a large burden of disease are not covered under current national health programs (table 7.2). The proportion of deaths associated with tobacco use, for example, rose to more than 2 5 percent in 1998 (risk factors other than tobacco use contribute to these diseases as well). Since these conditions largely affect older people, they repre- sent a much smaller, albeit still significant, burden of DALYs lost (8 percent). Major psychiatric diseases (unipolar and bipolar disease, psy- choses, self-inflicted injuries) account for 6.6 percent of DALYs lost- more than either tuberculosis or HFV Injuries, particularly from traf- fic accidents, fires, and falls, also represent a large (8.9 percent) and Table 7.1 Deaths and DALYs Lost Associated with Conditions Covered by National Health Programs, India, 1990 and 1998 DEATHS DALYS LOST 1990 1998 1990 1998 NUMBER PERCENT OF NUMBER PERCENT OF NUMBER PERCENT OF NUMBER PERCENT OF PROGRAM (THOUSANDS) TOTAL (THOUSANDS} TOTAL (THOUSANDS} TOTAL (THOUSANDS) TOTAL Reproductive and child health 2,962 31.6 2,492 26.7 109,955 38.2 91,431 34.0 Safe motherhood 775 8.3 737 7.9 32,772 11.4 31,207 11.6 Immunization 513 5.5 429 4.6 18,328 6.4 14,463 5.4 Control of diarrhea, disease, and acute respiratory infections 1,624 17.3 1,285 13.8 52,124 18.1 38,893 14.5 Vitamin A supplementation 21 0.2 16 0.2 746 0.3 565 0.2 Anemia control 29 0.3 26 0.3 5,985 2.1 6,302 2.3 Integrated Child Development Services 69 0.7 53 0.6 5,076 1.8 3,734 1.4 Protein energy malnutrition 69 0.7 53 0.6 5,076 1.8 3,734 1.4 Universal salt iodization 6 0.1 5 . . 378 0.1 280 0.1 Communicable disease control 780 8.3 621 6.6 15,380 5.3 13,973 5.2 Tuberculosis control 752 8.0 421 4.5 13,763 4.8 7,577 2.8 HIV/AIDS control 1 .. 179 1.9 236 0.1 5,611 2.1 Malaria control 26 0.3 20 0.2 1,195 0.4 577 0.2 Leprosy 1 . . 1 . . 186 0.1 208 0.1 Blindness control 0 . . 0 . . 3,038 1.1 3,732 1.4 Total national programs 3,811 40.7 3,166 33.9 133,449 46.4 112,869 42.0 Total 9,371 100 9,337 100 287,739 100 268,953 100 . . Negligible. Source: Murray and Lopez (1996); WHO (1999). Table 7.2 Deaths and DALYs Lost Associated with Major Conditions Not Covered by National Health Programs, India, 1990 and 1998 DEATHS DALYS LOST 1990 1998 1990 1998 NUMBER PERCENT OF NUMBER PERCENT OF NUMBER PERCENT OF NUMBER PERCENT OF PROGRAM (THOUSANDS) TOTAL (THOUSANDSI TOTAL ITHOUSANDS) TOTAL {THOUSANDS} TOTAL Respiratory and circulatory illnesses 1,878 20.0 2,360 25.3 18,347 6.4 21,282 7.9 Trachea, bronchus, lung neoplasm 35 0.4 79 0.8 387 0.1 921 0.3 Mouth and oropharynx cancers 80 0.9 100 1.1 1,100 0.4 1,313 0.5 Ischemic heartdisease 1,175 12.5 1 471 15.8 10,131 3.5 11,697 4.3 Cerebrovascular disease 448 4.8 557 6.0 4,235 1.5 4,814 1.8 Chronic obstructive pulmonary disease 140 1.5 153 1.6 2,494 0.9 2,536 0.9 Mental health 107 1.2 7 1.4 14,821 5.2 17,726 6.6 Unipolar major depression 1 . . 0 . . 8,063 2.8 9,679 3.6 Bipolar affective disorder 2 . . 2 . . 2,305 0.8 2,746 1.0 Psychoses 5 0.1 5 0.1 1,650 0.6 1,964 0.7 Self-inflicted injury 99 1.1 124 1.3 3,337 1.2 3,337 1.2 Injuries 344 3.7 401 4.3 21,821 7.6 23,824 8.9 Road traffic injuries 174 1.9 217 2.3 5,992 2.1 7,204 2.7 Fire 124 1.3 135 1.4 5,647 2.0 5,723 2.1 Falls 46 0.5 50 0.5 10,182 3.5 10,898 4.1 Negligible. Source: Murray and Lopez (1996); WHO (1999). 208 * Better Health Systems for India's Poor growing share of India's burden of disease. Violence against women has been neglected in India and many other countries, though new approaches are being developed to deal with these situations (box 7.2). Prevention of tobacco-related disease is a classic public health issue for which a mix of social marketing activities (such as education and publicity) and regulation is appropriate. International estimates suggest that tobacco control interventions cost $20-$80 per DALY saved (World Bank 1999a), a reasonable level of cost-effectiveness. Whether cost-effective interventions can be provided in India for tobacco-related diseases and the other problems not covered by national programs depends to a large extent on the implementation capacity of the agencies involved.2 The poverty dimension to noncommunicable diseases is often overlooked in low-income countries. Such diseases become more prominent as mortality levels fall and national incomes increase; as countries undergo a health transition, the burden of such ailments is sometimes assumed to fall disproportionately on the higher social classes. The scant data available in India, however, suggest that peo- ple in rural Rajasthan with low education levels and unskilled jobs are already suffering higher rates of coronary heart disease and hyperten- sion than the more educated and skilled (Gupta, Gupta, and Ahluwalia 1994). Analysis of risk factors for cardiovascular disease shows that tobacco use is higher among illiterate men and unskilled workers in urban and rural settings in India than among the literate and more skilled (Gupta, Gupta, and Ahluwalia 1994; Reddy and oth- ers 2000; Prabhakaran and others 2000). National Sample Survey data show that throughout India the prevalence of tobacco and alcohol use is higher among the poor than among the nonpoor, which puts the risk of cardiovascular disease among the poor, other things being equal, at a higher level than it is for others (figure 7.1). In turn, these higher risks may lead to higher rates of cardiovascular disease, cancer, liver disease, and injuries among the poor than among the nonpoor. The degree to which programs addressing serious mental condi- tions affect the poor has not been well studied in India. In other countries the poor and the most vulnerable are more likely to be Setting National Health Care Priorities * 209 Box 7.2 Responding to Violence against Women In Mumbai, a city of 14 million people, the Bombay Munic- ipal Corporation (BMC) manages a network of 26 hospitals with a staff 17,000 health workers. In two of its hospitals located in the poorer sections of the city, the BMC is pilot- ing One Stop Crisis Centers to assist women injured by domestic and sexual violence. In addition to providing medical and psychological care, the One Stop Crisis Centers will consolidate in one place key related services offered by other agencies such as the police, the welfare department, legal aid, and shelters. The hospi- tals' Emergency Department would run the center and be responsible for establishing links with the other agencies. The initiative is a collaboration among the BMC, an NGO (Center for Enquiry into Health and Allied Themes [CEHAT]), and the Ford Foundation. CEHAT is training the BMC staff who will run the center. It is also providing support for two experts to be located at the center for its first two years for supervising, coordinating with other agencies, and setting up systems. The project includes a research com- ponent to document the prevalence of domestic and sexual violence and develop new forms for record taking. BMC and CEHAT designed the One Stop Crisis Center concept on the basis of lessons learned from similar crisis centers established extensively in hospitals in Malaysia and the Philippines. Some key lessons incorporated into the Mumbai design are that: * The hospital is a good venue for providing crisis center services. * Government agencies and NGOs can work together effectively on targeted interventions. * Through community outreach, crisis centers can provide prevention as well as treatment programs. The need for One Stop Crisis Centers is large, and the BMC is seeking to implement a low-cost model that can be replicated in other hospitals. 210 * Better Health Systems for India's Poor Figure 7.1 Prevalence of Alcohol and Tobacco Use in India, by Income Quintile, 1995-96 Income quintile 5th (richest) 4th- 3rd 2nd- 1 St - I - - (poorest)_ 0 5 10 15 20 Percentage regularly using substance eSmoking *Tobacco (nonsmoking) oAlcohol Source: National Sample Survey Organisation (1998); authors' calculations. affected by serious mental conditions (Dohrenwend and Dohren- wend 1969), and major depression has been linked to loss of employ- ability (Eaton, Day, and Kramer 1988). In India, major psychosis is likely to be most common among the most destitute of people. Treatment of affective disorders and psychosis involves supervised community outreach workers, pharmaceuticals, and structured pro- grams to keep people living safely and working in their communi- ties. Like other disease control programs, these programs rely on an effective referral system. No estimates exist of the feasibility of such treatment in India or of their cost-effectiveness. Studies conducted in other parts of the world have estimated the cost of community- based mental health programs at $250-$999 per DALY saved Setting National Health Care Priorities * 211 (World Bank 1993), making these programs much more cost-effec- tive than most hospital-based interventions. The government may also have to consider providing support to other types of programs that the private sector is unlikely to finance. These include interventions whose effectiveness and efficiency are difficult to quantify but that are nevertheless important to India's health system. Examples include public education on health, health information systems, disease surveillance, pharmaceutical quality control, and other types of regulatory interventions. In addition to determining which kinds of programs to offer, pol- icymakers need to decide how best to implement programs. Should government provide services directly, finance them, or both? As we document below, the vast majority of outpatient visits are to private providers; can centrally sponsored programs work more effectively with the private sector? For services the public sector provides, what type of institutional arrangements are most effective? Spurred in part by the passage of the 73rd and 74th Amendments to the Constitution, administrators are increasingly devolving the management of health programs to state and local bodies. The current dilemma is that the central government is not able to manage centrally sponsored programs without greater reliance on the states and local bodies, but states and local bodies have little capacity to manage these programs themselves. Recent reviews of the Reproductive and Child Health Program and other centrally sponsored programs have suggested that India continues to suffer from critical gaps in its capacity for planning, implementing, and monitoring the outcomes of these programs. Many of the difficulties are tied to underlying weaknesses of the general health systems on which these programs depend. Shortages exist in technical leadership, public health management, and the capacity to plan and manage public communications. The hierarchi- cal administrative processes at the central and state levels continue under the guise of financial and administrative accountability, lengthening delays in implementation and weakening the respon- siveness of programs to local needs. 212 * Better Health Systems for India's Poor Interestingly, when states are given more flexibility, many prefer not to exercise it fully. Under the Reproductive and Child Health Program, for example, the Ministry of Health and Family Welfare has offered to decentralize procurement, but so far only one state (Tamil Nadu) has responded. What's needed is more accountability for program outputs and outcomes, but the emphasis is often on administering inputs and expenditures. The current challenge is to address the increasing need to decentralize and integrate health pro- grams while strengthening skills, creating systems, and building local demand. For some programs, separate program management and financing will remain the best alternative. Examination of the motivation and behavior of workers in the public health sector may also provide insights about alternative ways of providing or financ- ing health interventions, a subject explored in the next chapter. Households' Use of Personal Health Services Unlike the use of community-based public health services, which is driven by government, the use of personal health services is deter- mined by individuals and households. People choose whether or not to seek out medical care, which type of providers to consult (and in what order), and whether to comply with the recommended therapy. These choices are constrained, especially for the poor, by lack of income, time, knowledge of the disease, and information about the efficacy of various types of providers. This section examines how these choices affect the pattern of consumption of health services provided by the public and private sectors. National Findings As noted in the preceding chapter, the distribution of consumption of services provided by the public and private sectors varies by type of service (figure 6.1). The National Sample Survey data distinguish five types of services: outpatient care, inpatient care (excluding deliveries), Setting National Health Care Priorities * 213 _v_ W~~~~~ - X4 A day at the immunization clinic (PHOTOGRAPH BY RAY WITLIN/THE WORLD BANK PHOTO LIBRARY) institutional deliveries, prenatal care, and immunizations. Outpatient care is dominated by the private sector, which accounts for 83 percent of all outpatient visits. Hospitalizations and institutional deliveries are split about equally between the public and private sectors. The public sector plays the larger role in preventive services, delivering 60 per- cent of prenatal visits and 90 percent of immunization doses. The high percentage of outpatient services provided by the pri- vate sector is similar across household income and expenditure lev- els, urban and rural settings, gender, and caste and tribe affiliation. The poor do, however, use a different class of provider than do the rich, with the poor relying more on untrained practitioners and the rich using qualified practitioners. Both public and private hospitalizations increase with income (figure 7.2). Relative to people in the bottom income quintile, peo- ple in the top income quintile have almost six times as many hospi- talizations-about twice as many hospitalizations in the public sec- tor and about 11 times as many private hospitalizations. The poor depend on public hospitals more than the rich; for the poorest quintiles, 61 percent of hospitalizations are in public hospi- 214 * Better Health Systems for India's Poor Figure 7.2 Public and Private Sector Shares of Hospitalization, by Income Quintile of the Population Income quintile sj5th h e 33% (richest) 4th 45R. 3rd 5o 2nd 58% 1 St (poorest) 6 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 Hospitalizations per 100,000 population I oPublic hospitals uPrivate hospitals Note: Percentages represent the proportion of hospitalizations in the public sector for that quintile. Source: National Sample Survey Organisation (1998); Background Paper 5. tals. However, even those in the poorest quintile of the population use the private sector for nearly 40 percent of hospitalizations, even though the reported hospital charges per inpatient day are eight times higher in the private sector (Rs 48 per day in private facilities versus Rs 6 per day in public facilities). Institutional deliveries reveal a similar pattern. Individuals in the top income quintile are three times more likely to use the public sec- tor for institutional deliveries than the poorest quintile and 20 times more likely to use the private sector. Poor women who deliver in an institution are most likely to do so in public facilities, which account Setting National Health Care Priorities * 215 for 73 percent of institutional deliveries among women in the bot- tom income quintile. An important dimension to hospitalization is the length of stay (number of inpatient bed days). Long-term hospitalizations of the rich account for a large proportion of hospital days in the public sec- tor. Stays of more than 90 days account for more than one-fourth of all public inpatient bed days, and 40 percent of those stays are by the richest quintile. The fact that the rich account for a disproportion- ate share of inpatient bed days raises questions about the need for appropriate policies to improve equity and efficiency. For example, a long-term care policy may include higher fees for richer Indians having long-term stays or the establishment of a different level of facility to keep costs down. As with the income quintile findings, people below the poverty line are more likely than people above the poverty line to use the public sector. Two types of services-inpatient bed days and prena- tal care-show the greatest variation. For people living below the poverty line, 66 percent of inpatient bed days take place in public facilities; for people living above the poverty line, only 44 percent of bed days are in public facilities. Similarly, 74 percent of prenatal vis- its by women below the poverty line take place in public facilities, while only 52 percent of visits by women above the poverty line are provided by the public sector. A striking finding of the National Sample Survey is the gross underutilization of inpatient beds at primary health care facilities. While an estimated 20 percent of all public sector inpatient beds in India are at primary health centers, less than 5 percent of inpatient bed days take place at these centers. Moreover, these centers are not particularly pro-poor. The inability of these facilities to provide staff and ensure supplies contributes to their low quality and utilization (Mukhopadhyay 1997). Exercises conducted as part of State Health Systems Development Projects to rationalize public health services and design referral systems have shown that inpatient beds at pri- mary health facilities are not needed, suggesting that the govern- ment should no longer invest in these beds. In states in which budg- 216 * Better Health Systems for India's Poor ets are allocated on the basis of the number of beds, the budget allo- cation practice would have to be changed. State-Level Findings The distribution between private and public provision of personal health services varies considerably across states. To simplify the presentation, we discuss only inpatient curative care and institu- tional delivery services here. Inpatient curative care is included because hospitalization accounts for the largest outlay of both pub- lic and private resources; institutional deliveries are included because they are a determinant and leading indicator of maternal ill- ness and death. National data show a relatively even distribution of inpatient bed days between the public and private sectors. State data show much greater variation. The use of public hospitals is much higher than average in several states, including Himachal Pradesh (92 percent), Orissa (89 percent), West Bengal (81 percent), the Northeast states (77 percent), and Rajasthan (74 percent). Elsewhere-in Haryana (24 percent), Punjab (34 percent), and Maharashtra (36 percent)- the use of public facilities is lower than average. Large variations across states are also evident in the use of public versus private facilities by those below the poverty line (figure 7.3). In West Bengal and Orissa, the poor are heavy users of public facil- ities. In contrast, the poor in Punjab and Bihar make very limited use of the public sector. The use of public and private facilities for institutional deliveries also differs across states. Nationwide, the public sector share of bed days for deliveries is 50 percent. Utilization of public facilities is sig- nificantly higher than the national average in Orissa (89 percent), Uttar Pradesh (81 percent), and the Northeastern states (77 per- cent). States exhibiting less reliance on the public sector include Haryana (24 percent), Punjab (34 percent), and Maharashtra (36 percent). The same basic distribution between public and private facilities across states appears among women living below the poverty line. Setting National Health Care Priorities * 217 Figure 7.3 Public and Private Sector Shares of Inpatient Bed Days of Patients below the Poverty Line Punjab l l llll Bihar ' - Northeast states - - Gujarat - I - Uttar Pradesh ' l l l - - Karnataka - - Andhra Pradesh Kerala - - Maharashtra _ Nahonal average E Tamil Nadu - - Madhya Pradesh I Haryana- Rajasthan Orissa West Bengal 0 10 20 30 40 50 60 70 80 90 100 Percent 0 Public sector * Private sector ] Note: Northeast states consist of Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland, and Tripura. Source: National Sample Survey Organisation (1998); Background Paper 5. How Well Are Public Health Services Reaching the Poor? In the historical context of building a socialist nation, India's early plans for a public national health service that would provide serv- ices to all-a system similar to that being created simultaneously in the United Kingdom-were easy to justify. Public funds sufficient to maintain such a comprehensive health care system were neve- forthcoming, however. In the absence of an adequate public sys- tem, the private sector has grown dramatically. By the time of th 2 National Health Policy in 1983, the concern for equity-particu- larly the need to provide health services to the poor-had becom2 the main rhetorical justification for publicly provided curative health services. (The other rationale was the need to train doctors 218 * Better Health Systems for India's Poor and other medical professions. This rationale is the foundation for public involvement in tertiary health care.) India's Constitution laid most of the responsibility for providing curative care in the hands of the states.3 The analysis presented here estimates the relative size of the financial subsidy provided to people at different income or expendi- ture levels through government health care services. It is based on data from the study on the benefit incidence of health services (Background Paper 5), which in turn derived its information from budget and cost data for public health services and data from the 52nd round of the National Sample Survey (1995-96), which cov- ered more than 121,000 households.4 National Findings For the nation as a whole, the poorest 20 percent of the population captured only about 10 percent of the total net public subsidy from publicly provided clinical services (figure 7.4). The richest quintile received more than the three times the subsidy received by the poor- est quintile, indicating that publicly financed curative care services are unambiguously pro-rich. In part, this bias reflects the fact that the better-off use public facilities more than the poor, but it also reflects the fact that they pay more than the poor for each unit of uti- lization. For example, the top two quintiles (the richest 40 percent) pay 87.6 percent of the collected fees for inpatient care. The better-off pay higher fees because they receive higher-qual- ity service-a fact not captured by the benefit incidence data. Net cost recovery (fees relative to incremental costs of services) may therefore well differ from that implied by the amounts presented. Moreover, except in three states, cost recovery is low, with fees never exceeding 5 percent of costs. These amounts suggest that a better system for ensuring that the poor receive at least their fair share of public health expenditures needs to be put in place. An alternative approach to assessing the distributional perform- ance of the health sector is to use poverty measures based on a poverty line for rural and urban areas in each state. This approach Setting National Health Care Priorities * 219 Figure 7.4 Public Expenditures on Curative Care, by Income Quintile Percent 40 30 20 _ l st 2nd 3rd 4th 5th (poorest) (richest) Income quintile Source: National Sample Survey Organisation (1998). reveals that while 36 percent of the nation's people live below the poverty line, they receive only about 24 percent of public financing of curative health services.5 The pro-rich distribution of public health care expenditures in India reflects several factors: * For a given health condition, the rich are more likely to seek care, and when they do they are more likely to go to a higher-level facility. The richest quintile of the population is thus six times more likely than the poorest quintile to have been hospitalized ir either the public or private sector (3,447 versus 563 per 100,000) * The richest quintile accounts for 38.5 percent of inpatient bec. days, while the poorest quintile accounts for just 6.6 percent. * Because hospital costs represent by far the largest share of cura- tive health costs (87 percent) and curative health costs dominate 220 * Better Health Systems for India's Poor aggregate health care spending, aggregate health expenditures are dominated by hospital expenditures.6 The health care spending gap between rich and poor is even wider in the private sector. Thus, without public sector spending, the gap between rich and poor in service utilization would be even larger. Private sector facilities accounted for 67 percent of hospitals visits by the richest quintile. For the poorest quintile, such facilities accounted for 39 percent of hospital visits. The pro-rich bias varies across categories and types of health spending (figure 7.5). A concentration curve below the diagonal line indicates that the rich receive a more than proportional share of public health spending; a concentration curve above the diagonal line indicates a pro-poor bias. Outpatient care in primary care facil- ities shows a slight pro-poor bias, while spending on both inpatient and outpatient hospital care is biased toward the rich. Preventive care, immunizations, and prenatal visits show a more equitable distribution than most types of curative care. The public subsidy for immunizations is pro-poor, with the poorest two quintiles receiving 47 percent of all publicly provided doses (figure 7.6). Some, but not all, of the pro-poor distribution can be explained by the fact that poor households have more children. Prenatal care services pro- vided at primary care facilities also appear to be pro-poor, with the bottom two quintiles accounting for 46 percent of visits. A major limitation of the National Sample Survey data is that they do not allow deeper exploration of hospital-based services. The data do not distinguish between large urban tertiary hospitals and small rural secondary hospitals. A recent summary of international evi- dence argues that the poor are more likely to use primary care facil- ities than secondary care facilities, and more likely to use secondary care facilities than tertiary facilities (Yaqub 1999). Moreover, analy- sis of facility-based data in several Indian states shows that the poor use proportionately more secondary hospitals, particularly those located in poorer rural areas (STEM 2000; Institute of Health Sys- tems 2000; Blackstone Ltd. 2000). This may mean that the use of Setting National Health Care Priorities * 221 Figure 7.5 Income Bias of Public Spending on Hospital and Primary Health Care Facilities Cumulotive percentage of benefits 100 80 Pro-poor 60 ~ ~ ~~ distribution >9 60- 20 0 1 st 2nd 3rd 4th 5th (poorest) (richest) Cumulative percentage of population, by income quintile Hospital inpatient - Hospital outpotient A Primary care facility outpatient Note: Primary care facilities are primary health centers, subcenters, and dispen- sories. Values below the diagonal line indicate a pro-rich distribution of spending. Source: National Sample Survey Organisation 11998); Background Paper 18. secondary hospitals is not as pro-rich as that of all hospitals taken together. Segmenting the data by region shows that rural residents receive a lower public subsidy than urban residents: although 75 percent oi India's population live in rural areas, rural residents capture only 67.6 percent of the net benefits from curative care. The concentra- tion curve for the urban public subsidy is almost diagonal (that is, benefits are neither pro-poor nor pro-rich), whereas the concentra- 222 * Better Health Systems for India's Poor Figure 7.6 Income Bias of Public Spending on Immunizations Cumulative percentage of publicly provided immunizations 100 80 80 ~~Pro-poor distribution/ 60- 40 - Pro-rich / / ~~~~~~~~distribution 20 - 20 0 1 st 2nd 3rd 4th 5th (poorest) (richest) Cumulative percentage of population, by income quintile Note: Values above the diagonal indicate a pro-poor distribution of services. Source: Background Paper 5. tion curve for the rural population shows a pro-rich bias. Two fea- tures may explain this difference. Among the bottom income quin- tile, hospitalization rates (in both public and private facilities) are much higher in urban than in rural areas (1,266 versus 471 per 100,000). In the top income quintile, utilization rates are about the same (3,656 versus 3,269 per 100,000). Moreover, because private hospitals are more available in urban areas, the public sector utiliza- tion rate by the richest quintile is lower in urban than in rural areas. Setting National Health Care Priorities * 223 These findings are consistent with those of other studies. A World Bank study (1997b) found that use of secondary care public hospitals, the most common type of hospital in rural areas, is low, largely because of poor quality and the lack of private sector alter- natives. The study found that improving the quality of rural public hospitals would increase public sector utilization in rural areas. Members of scheduled castes or scheduled tribes represent 29 percent of India's population and capture 28.4 percent of the subsidy benefits for curative care services. This suggests a fairly equal distri- bution of public services along caste and tribe affiliation lines. State-Level Findings Since curative care is primarily a state responsibility, one should expect health status outcomes and expenditure distributions to vary across states. Using the same National Sample Survey data, we repeated the utilization and benefit incidence analysis for 15 of the largest states and regions (which account for 97 percent of India's population); we based the analysis on income quintiles and poverty lines created for the urban and rural populations in each state (table 7.3). To simplify the presentation, we constructed a concentration index for each concentration curve to measure the level of inequal- ity in subsidy benefits. The possible values of this index range from -1 (all benefits accrue to the poorest) to 1 (all benefits accrue to the richest); an index of 0 indicates that benefits are distributed evenly across income groups (Kakwani, Wagstaff, and van Doorslaer 1997). In four states (Kerala, Gujarat, Tamil Nadu, and Maharashtra) cur- ative care services are distributed nearly equally across income levels, with concentration indexes near zero. All other states reveal a pro-rich pattern of spending on curative care services, with Rs 2 or more going to the richest quintile for every Rs 1 that reaches the poorest quintile. In Bihar and Rajasthan Rs 5 or more of public health care spending is used by the top quintile for every Rs 1 that benefits the poorest quin- tile. The range of inequalities found in Indian states is similar to that found across developing countries (box 7.3). 224 * Better Health Systems for India's Poor Table 7.3 Income Bias in Public Spending on Curative Care in India and Selected States SUBSIDY, STATE CONCENTRATION INDEX' RATIO OF RICHEST TO POOREST QUINTILE Kerala -0.041 1.10 Guicrat 0.001 1.14 Tamil Nadu 0.059 1.46 Maharashtra 0.060 1.21 Punjab 0.102 2.93 Andhra Pradesh 0.116 1.85 West Bengal 0.157 2.73 Haryona 0.201 2.98 Karnataka 0.208 3.58 National average 0.214 3.28 Northeast states 0.220 3.16 Orissa 0.282 4.87 Madhya Pradesh 0.292 4.16 Uttar Prodesh 0.304 4.09 Rajasthan 0.334 4.95 Bihar 0.419 10.30 Note: Northeast states consist of Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland, and Tripura. a. Possible values range from -1 (all benefits accrue to the poorest) to 1 (all benefits accrue to the richest); zero indicates equal distribution between income groups. Source: Background Paper 18. Box 7.3 Equity of Public Spending on Health: International Experience International experience with measuring equity of public spending on health in developing countries indicates that: * Public spending is most pro-poor at basic levels of social services. * Discrimination by race, gender, caste, and minority sta- tus plays a role in utilization of publicly provided financed services. * Some evidence supports the finding that public spending in the health sector appears to be more pro-poor in urban settings than in rural settings. Setting National Health Care Priorities * 225 * Socialist countries have a better pro-poor record than nonsocialist countries. These four findings are consistent with the new utiliza- tion and benefit incidence results from India. The figure below compares the results of benefit inci- dence studies for India and for selected states in India with findings from other countries. Variations across countries in methodologies and sources of data make international comparisons difficult, but as the figure shows, the bulk of the findings point to public spending patterns in health that are not very pro-poor. International Comparison of Inequalities in Health Sector Subsidies Concentration index 0.50 0.40 * 0.30 HHj 0.20 E 0.10 [J[ 0.00. . . . . -0.10 -0.20 O- 2 2 -O E 2 > m C Lz >. > 0 o ': D Note: Possible values range from -1 (all benefits accrue to the poorest) to 1 (all benefits accrue to the richest); zero indicates equal distribution between income groups. The shaded bars are values for India and selected states in India. 226 * Better Health Systems for India's Poor Inpatient care. Because inpatient care is the most costly health care service, analyzing state-level inpatient bed use by the poor provides the best way of comparing equity across states (box 7.4). In just two states, Maharashtra and Kerala, did the poor and the nonpoor spend about the same number of days in inpatient facilities. In all other states, people below the poverty accounted for a relatively small per- centage of inpatient bed days, with the largest differences occurring in Bihar, Uttar Pradesh, Orissa, and the Northeastern states. These differences vary widely between urban and rural populations, with urban areas far more equitable than rural areas in all states. Box 7.4 Adjusfing Public Sector Hospitalizafion Rates for Income Bias, 1995-96 The number of public sector hospitalizations varies greatly across states in India, ranging from 2 per 1,000 in Bihar to 29 per 1,000 in Kerala. The pro-rich bias in hospitalization also varies across states, with Bihar having the most pro- rich distribution and Kerala being the only state to have a pro-poor distribution. Thus, some states that perform less well in terms of number of hospitalizations perform better when the number is adjusted to account for a pro-poor bias (an adjustment that produces a measure called "achieve- ment"-see chapter 9, box 9.2). As shown in the figure below, Tamil Nadu, for example, has a lower public sector hospitalization rate than Orissa, but hospitalization in Tamil Nadu is much more concentrated among the poor. As a result, its achievement in public sector hospitalization is higher than Orissa's. Of course, in many Indian states, a sizable proportion of hospitalizations are in the private sector. In Tamil Nadu, for example, only 40 percent of hospitalizations are in the pub- Setting National Health Care Priorities * 227 lic sector (in Orissa the proportion is nearly 90 percent). Taking into account the role of the private sector changes how states compare in terms of achievement. The achieve- ment of Tamil Nadu's small but relatively equitable public sector is reinforced by a fairly large private sector, while the lack of achievement of Orissa's large but inequitable public sector is compounded by the fact that there is only a very small private sector. Annual Hospitalization Rates and Achievement Indexes across Major Indian States Bihar… Madhya Pradesh Rajasthan Uttar Pradesh Orissa Andhra Pradesh West Bengal Northeast states National average Karnataka Pun jab Gujarat Ik I Tamil Nadu Maharashtra Haryana Kerala 0 10 20 30 4050 60 70 80 [=Public average -- Public achievement _-- Overall achievement Note: Northeast states consist of Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland, and Tripura. Source: National Sample Survey Organisation (1998); authors' calculations. 228 * Better Health Systems for India's Poor Outpatient curative care. The National Sample Survey distinguishes between hospital-based outpatient care and care at primary health care facilities, but it does not disaggregate care provided by second- ary and tertiary hospitals. On average, outpatient care is more equi- table than inpatient care, but spending on hospital-based outpatient care is less equitably distributed than spending on outpatient care at primary care facilities (primary health centers, subcenters, and dis- pensaries). The national average hides several important differences across states. In Rajasthan, Bihar, Haryana, and Madhya Pradesh, both hospital-based outpatient care and outpatient care at primary care facilities appear to favor people above the poverty line. These find- ings are consistent with anecdotal evidence that the poor join other groups in bypassing a failing primary care system. In West Bengal, Karnataka, Punjab, and the Northeastern states, hospital-based out- patient care favors people above the poverty line, while primary care favors people below the poverty line. The opposite is true in Andhra Pradesh and Gujarat, where hospital-based outpatient care favors people below the poverty line while primary care facilities favor those above the poverty line. Only in Maharashtra and Kerala do both types of outpatient care favor people below the poverty line. This variability in the outpatient utilization data presents an important opportunity for policymakers to explore the determinants of equity performance. Important roles are perhaps being played by supply-side factors, such as placement of facilities, budget alloca- tions, the level of facilities (secondary versus tertiary hospitals), and human resource and other input factors to improve quality. On the demand side, literacy and empowerment may influence poor house- holds' likelihood of using public facilities. Immunizations. The data reveal considerable variability across states in the use of preventive services such as immunization-a finding that is consistent with the National Family Health Surveys. Although public sector delivery of vaccines is pro-poor, not all of the poor are served: 37 percent of all unimmunized children are from Setting National Health Care Priorities * 229 the poorest income quintile, and Rajasthan, Bihar, and Uttar Pradesh have the highest proportions of children without immu- nizations. With the exception of Kerala and Tamil Nadu, which have few unimmunized children, and Uttar Pradesh, which has many, children below the poverty line are more likely not to be immunized than children above the poverty line. Gender differences. Nationally, no clear bias against females is evident in public sector utilization. In fact, males account for 46.6 percent and females for 54.4 percent of total benefits from curative care services, even though males represent a higher percentage of the sample and curative care services included no pregnancy-related services. These national levels mask large variations across states, however, especially in rural areas. For example, in rural Gujarat, Punjab, and Haryana, men get more than 55 percent of the outpa- tient care, but in the Northeast states, they consume less than 45 percent. How Does the Pattern of Public Spending Affect Health Outcomes? Without some form of public action, the poor will be left without critical health services. Relying on private provision alone will not be equitable-it will not provide all Indians with a basic standard of health care nor will it allocate health care spending proportionally to all income groups. While some states have ensured that public financing is not skewed to the rich, many more states are doing too little to ensure that the poor have access to and use health services. In these states, overall public financing of health services is not equi- table, and the public health system tends to respond more to the demands of richer Indians than to the poor. Alongside the question of whether the distribution of health serv- ices and health financing is fair is the question of whether fairness in public financing of health makes a difference in other health out- comes. We examined the relationship of state-level rates of infant 230 * Better Health Systems for India's Poor and child mortality to the concentration index derived from public hospitalization. After adjustment for other factors influencing mor- tality-state per capita income, literacy levels, and per capita public spending on health-the analysis showed that a pro-poor distribu- tion is significantly associated with reduced infant and child mortal- ity (see appendix table 7A.1). The analysis also suggests that the dis- tribution of public resources may be more important in influencing mortality than the amounts of public health spending, at least over the range of public health spending provided by states in 1995-96 (amounts ranging from Rs 57 per capita in Bihar to Rs 132 in Ker- ala). Although we don't fully understand the mechanisms that explain the relationship between more equitable public health serv- ices and better health outcomes, the results highlight the importance of equity in improving health outcomes and the need to further study and monitor equity and health system outcomes. Appendix Table 7A.1 Infant and Child Mortality among Indian States: Effects of Public Hospitalization, Equity, and Other Factors, 1995-96 INFANT MORTALTrY CHILD MORTALUTY COEFFICIENT t STATISTIC COEFFICIENT t STATISTIC INDICATOR (STANDARD ERROR) (p VALUE) (STANDARD ERROR) (p VALUE) Intercept 88.26 n.a. 24.47 n.a. State public hospitals 92.08 2.19 44.48 3.20 Concentration index (42.04) 1.05) (13.91) (.008) State per capita income (rupees) -0.001 -.36 -0.0009 -.79 1.003) (.72) (.0011 (.401 State per capita public -0.10 -.89 -2.97 -.17 health spending (rupees) (.121 (.391 (17.63) (.90) State literacy rate (percent) -43.32 -.81 -0.08 -2.01 (53.3) (.43) (.0391 (.07) R-squared .75 .83 n.a. Not applicable. Note: Infant mortality is for those up to one year of age; child mortality is for those age one year to just under age five. Setting National Health Care Priorities * 231 Notes 1. These data do not imply that the programs could prevent the same proportion of deaths and DALYs (disability-adjusted life years) lost. We count only the direct deaths and disability related to these conditions and do not estimate their effects as risk factors for other diseases. For example, vitamin A deficiency and protein energy mal- nutrition contribute to a larger loss of life than is directly attributed to them because they also contribute to pneumonia, diarrhea, and measles. 2. An effective tobacco control program also involves taxing tobacco products, restricting advertising, and providing smoking cessation interventions. Dietary advice and exercise promotion are also low-cost interventions that are part of related programs aimed at reducing cardiovascular disease. The prevention of injury involves public education; engineering solutions, such as improving road safety; and regulatory measures, such as requiring the installation and use of seatbelts and enforcing speed limits and drunk driving laws. Fire-related injuries and deaths are concentrated among women of reproductive age; this condition reflects their exposure to domestic violence, as well as to kitchen fires, and requires interven- tions tailored to address violence against women in traditional and modern settings. 3. Since the Fifth Five-Year Plan, national resources have been used to supplement rural health infrastructure (particularly for sub- centers, primary health centers, and community health centers); some urban clinics; and rural health personnel (particularly auxiliary nurse midwives). Successive World Bank-supported India Popula- tion Projects have facilitated this process, as states received 90 per- cent of these credits. In part, these investments were intended to support centrally sponsored programs and provide the first level of curative care more generally. 232 * Better Health Systems for India's Poor 4. The benefit incidence is calculated in the following manner. First, rank all individuals (or households) from poorest to richest by the chosen measure of current welfare. Then identify which indi- viduals use each type of publicly provided service and calculate the average unit cost of providing each type of service (net of cost recov- ery fees). Then multiply the utilization figures by the government's unit cost of provision (net of fees) to derive the amount of public spending on the good or service going to each group. Mathemati- cally: Xj = i Uij# jS , iS X. = health sector subsidy enjoyed by groupj li.. = utilization of service i by groupj V:= utilization of service i by all groups combined Si = government net expenditure on service i e, = groupj's share of utilization of service i 5. The share of the population reported here as living below the poverty line is based on the 16 largest states and regions. The 34 percent figure reported in chapter 4 is based on the entire popula- tion. 6. The analysis measures costs per type of visit using standardized unit-cost data rather than actual expenditures. That is, if a primary health center had only one patient per year, the analysis would not attribute all the cost of inpatient beds to that one patient. Using budget data to distribute expenditures to different levels of facilities produces similar results (Background Paper 5). rx 1 CHAPTER 8 Financing Health In this chapter we address three financing mechanisms: revenue rais- ing, resource intermediation (including pooling), and resource allo- cation and purchasing.1 As a prelude, we give a detailed picture of the amount and sources of financing for private and public elements of the health sector and the unique structure of financing responsi- bilities in the public sector. As in the other chapters, the dominant questions are: How much are the poor protected? And what are the appropriate roles for the public and private sectors? See box 8.1 for a summary of the empirical findings and policy challenges discussed in the chapter. Health Sector Spending In 1996, India spent about 4.5 percent of its GDP on health, less than the average of 5.6 percent for low- and middle-income coun- tries.2 India also has one of the highest proportions of private health financing anywhere in the world, about 82 percent. Only five coun- tries (Cambodia, the Democratic Republic of the Congo, Georgia, Myanmar, and Sierra Leone) have a higher dependence on private financing in the health sector (WHO 2000c). The low spending (about $18 per capita) and the dominance of private sources of financing make India unique (World Bank 1997b; WHO 2000c). 233 234 * Better Health Systems for India's Poor Box 8.1 Empirical Findings and Policy Challenges Health System Actors, Functions, and Oukomes People C D~~emand Financing Health status _ Service Financial status deliver t Input management ) eIiv' |Responsiveness to publicl Private sector actors The stote Empirical Findings * Overall health spending in India is estimated at 4.5 per- cent of GDP, less than the average of 5.6 percent for low- and middle-income countries. * Public spending on health in India is 0.9 percent of GDP, among the lowest in the world . * Out-of-pocket private spending dorninates health expen- ditures, with 82 percent of all health spending from pri- vate sources. * Hospitalization frequently results in financial catastro- phe, especially in the absence of risk pooling. About 10 percent of Indians, most of them employees in govern- ment or elsewhere in the formal sector, have some form of health insurance. * The gap in public financing for health between rich and poor states is widening and threatening to expand the gaps in outcomes. Financing Health * 235 Policy Challenges * How will India expand public spending on basic health services to improve quality of services and the health of the poor? * How can regional inequalities in health sector spending be decreased given decentralized financing and differ- ences in implementation capacities? * How will India be able to shift from predominantly pri- vate out-of-pocket health financing to risk-pooling mechanisms, particularly when incomes are low and most people belong to rural informal sectors? * How can India develop the regulatory environment and information systems to oversee new health financing mechanisms? India's public expenditures on health are even further below inter- national levels than its overall spending- 0.9 percent of GDP, com- pared with an average of 2.8 percent for low- and middle-income countries and a global average of 5.5 percent (World Bank 1997b).3 The government's fiscal effort, measured as the proportion of total government expenditure spent on health, again identifies India as a low performer-13th from the bottom when compared with all other countries (WHO 2000c). Over the 1985-2000 period, India's public spending on health rose in inflation-adjusted per capita terms, but it fell as a share of GDP (figure 8.1). Even if the share of GDP had remained at 1.1 percent, the level at the start of the period, it would still rank among the lowest in the world. Given India's size and decentralized political structure, patterns of public spending on health must be examined at the state level. In fact, the financing responsibility for public spending is primarily at the state level with some overlapping responsibilities with the cen- tral govermnent on a series of centrally sponsored schemes (chapter 236 * Better Health Systems for India's Poor Figure 8.1 Public Sector Spending on Health in India, 1985-2000 Rupees Percent 100 1.2 95 Share of GDP 7 1.1 5(_right scale) 1.0 90' 0.9 85 ~~~~~~~~~~~~~~~~~0.8 85' Per capita expenditure 0.7 80 (left scale) 0.6 0.5 75. 0.4 '0 N CO 0 0 C - 04 C'3 '0 '0 N m 0' 0 In 10 1 co os ol N n LO 1 so KCO 04 00 0' CO tDOL OL O' O' 0' 0' O.0O' 0' 0' O Source: Central Statistical Office; Ministry of Finance; Reserve Bank of India. 7). Several facts are evident from the distribution of public spending by state and source (figure 8.2): * States vary widely in overall per capita public spending on health- in Kerala and Punjab, for example, spending is double that in Bihar, Madhya Pradesh, and Uttar Pradesh. * Public spending per capita is lowest in the poorest states (Bihar, Madhya Pradesh, and Uttar Pradesh). * States typically account for about 75 percent of total public spending per capita for health and thus are the major determi- nants of the variation in such spending. * The central government's share of health spending is consider- ably more equally distributed than is the state share, but is slightly tilted away from the poorer states.4 Although data on per capita spending by state do not address the important dimension of how resources are spent, the size of the dif- Financing Health * 237 Figure 8.2 Public Expenditures on Health in Selected States in India, Grouped by State and Distributed by Government Level, 1995-96 Kerala - Punjab j a Tamil Nadu Ra jasthan Assam . Maharashtra_ Karnataka Gujarat . Haryana . West Bengal_ Andhra Pradesh Orissa _ Uttar Pradesh Madhya Pradesh Bihar _ _ _ 0 25 50 75 100 125 150 Rupees El Central government * State government Source: Selvaraju (2000). ferences and the pro-rich distributions have serious long-term implications. With health financing a major responsibility of state governments, spending per capita will inevitably be lower in the poorer states than elsewhere. The challenge for the national gov- ernment is to consider a variety of policy instruments, including centrally sponsored schemes, equalization funds, and other fiscal incentives to address the inequalities perpetuated by the current financing arrangements. State variations in private spending on health services (figure 8.3) are equally impressive: * In Kerala and Punjab, per capita private spending is almost four times larger than in Rajasthan and Bihar. 238 * Better Health Systems for India's Poor * VVhile there is some consistency between the high a low level of private and public spending (Bihar and Orissa low, and Kerala and Punjab high), some states have different rankings between the two. * Private facilities receive the lion share of out-of-pocket expendi- tures. Here again, the amount of money paid for health services does not capture individuals' choices of type of provider or quality of care pur- chased (both issues addressed in chapters 6 and 7). Policymakers should, however, be concerned with the large state-level variations, Figure 8.3 Private Spending on Health Services at Private and Public Facilities, 1 995-96 Rajasthan Bihar Karnataka Orissa Northeast states Gujarat Tamil Nadu Madhya Pradesh West Bengal National average m Andhra Pradesh Maharashtra Uttar Pradesh Haryana Punjab Kerala _ _ . 0 100 200 300 400 500 Rupees El Public facilities * Private facilities Note: Northeast states consist of Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland, and Tripura. Source: Background Paper 1 8. Financing Health * 239 i1P s i j I ,=_7a _ |R -- - . 1t 1 Lining up to receive drugs at a dispensary (PHOTOGRAPH By GEETANJALI CHOPRA/ HEMANT MEHTA/THE WORLD BANK) which may lead to increasing the gaps in outcomes for the three types of health system objectives (improved health status, financial protec- tion, and consumer satisfaction), addressed in the next chapter. The determiinants of private spending include the relationship between income and demand for services, the imipact of education on health- seeking behavior, the availability of alternative providers, and the qual- ity (actual or perceived) of both private and public sector providers. One determinant of high private spending on health in India is low public spending on health. The state-level relationship between public and private spending appears, however, to be more complex (figure 8.4). In Bihar and Orissa, both public and private spending on health appear to be relatively low, while in Kerala, Punjab, and Maharashtra, spending is relatively high for both. The size of spend- ing appears to move in different directions for Rajasthan and Taril Nadu (high public, low private) and for Uttar Pradesh and Andhra Pradesh (high private, low public). 240 * Better Health Systems for India's Poor Figure 8.4 Comparison of State Public and Private per Capita Health Spending, 1995-96 Public spending on health 140 KE 120 TN *K 100 vR K A MH Median public spending *HA 80 OR V B AP *U 60 *BI M 40 Median private spending 20 0 100 200 300 400 500 Private spending on health Note: AP = Andhra Pradesh, BI = Bihcr, GJ = Gujarat, HA = Haryana, KA = Karnataka, KE = Kerala, MH = Maharashtra, MP = Madhya Pradesh, OR = Orissa, PJ = Punjab, RJ = Rajasthan, TN = Tamil Nadu, UP = Uttar Pradesh, and WB = West Bengal. Source: Background Paper 1 8. Financing Mechanisms One way of thinking about the basic functions of financing the health sector is to segregate them into three categories: * Mechanisms for raising resources for the sector. * Mechanisms for organizing clusters of funds as resources flow from sources to providers of care. * Mechanisms for resource allocation that influences which providers receive the relative shares of available funds. Financing Health * 241 Structure of the Flow of Funds A useful first step in analyzing the different financing mechanisms is to understand the architecture of the health care financing system. Several state-level health account activities provide a strong founda- tion for exploring the structure of financial flows (Background Paper 1; Sharma, McGreevey, and Hotchkiss 2000). In Rajasthan (appendix figure 8A.1), about 71 percent of total health spending is from private sources, a level smaller than the national average but one that still dominates Rajasthan's spending picture. Another critical feature is the shared responsibility between the central government and the state governments. Detailed analy- sis found that over the past two decades about 73 percent of total public resources for health have been provided directly by state gov- ernments (World Bank 1997b). Recent estimates show the share of states may have increased to account for 75-78 percent of total pub- lic sector health spending. For private financing, the out-of-pocket share of the flow of funds dominates and for the most part is not pooled through insurance mechanisms. Here again, this dominant form of financing has strong negative implications for equity and increases the risk that vulnera- ble groups will slip into poverty. Revenue Generation As is the case in other low-income countries, the government in India has an intractable problem in mobilizing adequate resources when much of the population is poor. Taxation and the pooling of funds for health insurance is made difficult by the low rate of partic- ipation in the formal labor market and the large, informal rural labor market. Cost recovery accounts for a very small share of public sec- tor expenditure; the bulk of resources come from general revenues.5 Only three states-Kerala, Punjab, and Haryana-have cost recov- ery ratios close to 10 percent, while for all other states the ratio is less than 4 percent (appendix table 8A.1). The bulk of the fees for public facility care (more than 80 percent) were paid by the richest 242 * Better Health Systems for India's Poor 40 percent of the population. The progressive cost recovery system was consistent for both urban and rural populations. That the rich pay more may reflect the fact that the rich use services more (chap- ter 7) or that the quality of care provided to those who pay may be higher than for those who do not pay. Assessments of the cost recovery system in several states have uncovered weaknesses (World Bank 1997b): * The states have no appropriate institutional framework for reviewing user charges. * The level of cost recovery is minimal because of low fees and inadequate collection mechanisms. * Mechanisms for exempting the poor from user charges are diffi- cult to target and implement. * No adequate mechanisms exist to ensure that recovered funds would be used at the point of collection-and thus serve as an incentive to operate efficiently-rather than being sent to the government treasury. Those in the richest 20 percent of the population spent, on average, five times more than did those in the poorest 20 percent (appendix fig- ure 8A.2). Similarly, those above the poverty line spent three times more than did those below it (appendix figure 8A.3). Individuals not belonging to a scheduled caste or tribe spent, on average, Rs 100 more than others. Urban residents spent Rs 100 more than rural residents. An important consequence of these differences in out-of-pocket health spending is that when combined with the low use of "free" public services by the most vulnerable groups, the access to health care services appears to be severely limited. A second concern is the hardship created by the mode of resource generation by the private sector (fee for service paid directly by the patient) and the limited availability of resource pooling; this is discussed in more detail in the next section. Out-of-pocket spending depletes savings and assets and forces borrowing to cover medical bills, especially inpatient bills. Financing Health * 243 Intermediation and Pooling Mechanisms One argument for pooling resources in the health sector is that it spreads risks across insured populations- an important considera- tion in India because of the current huge financial risks involved with hospitalization (box 8.2). In one year, about one-quarter of all hospitalized Indians may fall into poverty because of the costs of medical care (chapter 9). Pooling becomes increasingly important as the epidemiological transition takes place in India and health serv- ices provision changes from low cost per episode to high cost per episode. As was shown in chapter 3 (box 3.4), many countries are taking steps to pool risks and subsidize health care for the sick and the poor. Box 8.2 How Hospitalization Can Devastate Krishna is 30 years old and lives in Katkol, a rural village in Karnataka. Six months ago Krishna fell while harvesting coconuts at a plantation. He lost consciousness, and his left leg started to bleed. The plantation owner immediately took Krishna to the primary health center, where the Med- ical Officer gave him first aid and asked them to go to the city of Mysore because the nearby hospital did not have facilities to attend to him. Krishna's employer took him to Mysore in his car; with the Medical Officer's referral letter, Krishna was immedi- ately admitted into the government hospital. He was given saline solution and after two days regained full conscious- ness. Krishna was told that his back had been injured and that he would receive surgery the next day. 244 * Better Health Systems for India's Poor Box 8.2 (continued) After the operation, Krishna stayed in the hospital for almost four months. During that time his wounds became worse, his legs swelled, and big bedsores formed on his back, which was full of pus. The hospital provided some medicines, but he was also asked to purchase more injec- tions and glucose "bottles." Krishna described his hospital experience as being very unpleasant. Basic amenities like drinking water were not available. Paramedics and other staff were hostile and would shout because Krishna's family could not give money when they demanded it. His elder brother and his wife came to look after him and had to sleep on the floor and eat in a hotel. The doctors would come to see him once every two days. Eventually he was told that he would not live for more than two months and that the hospital could not do any- thing more for him. He was asked to leave because of a shortage of beds, and the family hired a car to take Krishna home. For Krishna's hospitalization, the family incurred expenses of Rs 15,000-on transportation, medicines, food, and tips to health staff and doctors. The family has an annual income of about Rs 6,000. The funds for hospital- ization were raised by selling the half-acre of land they had and also by taking a loan at the rate of 10 percent per month. Krishna now sleeps with his back facing up, and all his wounds on his back and legs are oozing. He has become pale and weak, and the family expects him to die at any moment. Source: Administrative Staff College of India (1995). Financing Health * 245 Insurance mechanisms, however, are not without risks of market failure (Background Paper 2). Several characteristics of the health care market may lead to cost escalation: * Information gaps between providers and patients may create incentives to provide more care than may be medically appropri- ate, especially in a regime of pure indemnity insurance. The information gaps make the patient, and insurers for that matter, less willing to question the doctor's recommendations (Arrow 1963). The problem will be greater in situations in which the patient can choose his or her doctor and treatment freely and then present the bill to the insurer for reimbursement. * Health insurance reduces the incentive of individuals to guard against poor health and thus tends to drive up demand for more expensive care. The covered individual may know that he or she is taking more risks, but the insurance company may not (h e insurance term used for this information gap is "moral hazard"). * Another information gap is that patients are likely to know much more about their health status and future needs than insurer;. Thus, people expecting to incur significant health expenditures in the near future will figure disproportionately among those wh: choose to get insured (the insurance term used for this informa- tion gap factor is "adverse selection"). * Without minimum capital reserves and incomplete epidemiolog- ical information about the population, which is the case in most developing countries, insurance companies risk long-term failur Z by guessing wrong and charging premiums that are too low in comparison to the benefits offered in a competitive environment. * Empirical evidence exists that health spending per capita is posi- tively associated with the proportion of the population covered by private insurance. The contribution of an insurance scheme, whether public or pri- vate, to improving the quality of health care depends on whether the 246 * Better Health Systems for India's Poor scheme is able to influence the licensing of medical personnel and facilities and the entry of highly skilled individuals into the health sector. For equity, the main concern is that private insurers will try to sell only to low-risk customers. But an expansion of private health insurance could eventually improve access to public facilities by those unable to get private insurance. The insurance schemes in India can be sorted into four broad groups: mandatory, voluntary, employer-based, and NGO-based (Garg 2001). The four types of scheme together cover roughly about 10 percent of India's population (appendix table 8A.2). Mandatory health insurance schemes. The mandatory schemes consist of the Employees State Insurance Scheme (ESIS), for certain low- income employees of the organized industrial sector, and the Cen- tral Government Health Scheme (CGHS), mainly for central gov- ernment employees. Both these schemes are principally financed by the contributions of beneficiaries and their employers and from taxes. The ESIS receives some contribution from state governments, whereas the CGHS is mainly financed from central government revenues. ESIS covered 35.4 million beneficiaries in 1998 and CGHS covered 4.4 million beneficiaries in 1996. Providers are mainly salaried, and hospitals work under global budgets. Voluntary health insurance schemes. Voluntary schemes are for indi- viduals and corporations, and are available mainly through the General Insurance Corporation (GIC) of India-a government- owned monopoly-and its four subsidiaries. These schemes are financed from household and corporate funds. GIC offers one pol- icy for groups and mostly nonpoor individuals and another mostly for poor individuals and families. These policies have had only lim- ited success in India; in 1996 they covered only 1.7 million people. With the passage of the Insurance Regulatory and Development Authority Act of 1999 and the liberalization of insurance, more pri- vate voluntary health schemes are expected to be available in the near future. Financing Health * 247 Employer-based schemes. Public and private sector companies offer health insurance through their own employer-managed facilities by way of lump sum payments, reimbursement of employees' health expenditures, or coverage under one of the GIC policies. Workers buy health insurance through their employers. Ellis, Alam, and Gupta (2000) estimate that roughly 30 million people are covered under employer-based schemes. Community-based insurance scbemes. Primarily for the informal sectcr, community-based schemes tend to offer coverage for all availabl.e services but emphasize primary health care (box 8.3). Most of these schemes are financed from patient collections, government grant;, donations, and miscellaneous items such as interest earnings cr employment schemes. Most NGOs have their own facilities or mobile clinics to provide health care. Total coverage is estimated to be about 30 million people (Ellis, Alam, and Gupta 2000). Table 8.1 summarizes the main market failures in the financing of health care, the consequences, and the policy measures for correct- ing them. Box 8.3 Innovative Health Insurance Schemes in India Up to 10 percent of the people in India, most of them in government and other industries in the formal sector, are covered by some form of health insurance, but the policy benefits are limited and claims service is poor. A few inno- vative schemes have been tried out across the country to address the needs of the larger segment of self-employed and informal workers. Among them are the following plans: * Ambikpur Health Association, Orissa: Free outpatient care and limited hospitalization are provided to about 75,000 individuals on a voluntary basis. Among the innovative features of the plan are screening at the time of enrol- ment to avoid adverse selection of participants. However, 248 * Better Health Systems for India's Poor Box 8.3 (continued) premiums cover only 1-2 percent of the plan's outreach costs. * Mallur Milk Co-operative, Karnataka: Covering a popula- tion of 7,000 spread across three villages, the scheme provides preventive and curative health care (both out- patient and inpatient) to all eligible community mem- bers. Participation is mandatory. Income from an endowment fund covers all expenses. * Sewagram, Maharashtra: Free primary care, drugs, refer- rals, and hospitalization for nonchronic conditions are provided to a population of more than 14,000 spread across 12 villages (75 percent mandatory attendance within a participating village). Sliding-scale premiums are employed to promote equity. * Meko Milk Co-operative, Gujarat: Outpatient consulta- tion, discounted drugs, and diagnostic services are pro- vided by Aga Khan Health Services. Enrollment is mandatory for all co-op members. Among the complex issues and problems these organiza- tions face, two are of particular importance: (a) arranging for cost-recovery without excluding the poor and (b) deal- ing with moral hazard and adverse selection. These issues have been addressed by these organizations with varying degrees of success. In addition to the protection from finan- cial shock that they provide to their members, these plans have sometimes also helped to enhance the allocative effi- ciency of health spending by developing low-cost treat- ments or by increasing the utilization of preventive care. Source: Ranson (1999). Table 8.1 Market Failures, Consequences, and Responses in Financing Health Care MARKET FAiLURE CONSEQUENCES MEASURES TO CORRECT FAILURE Demand side Moral hazard Overuse of services by patients Deductible, co-insurance, co-payments, etc.; gatekeepers, waiting lines Adverse selection Little risk pooling Tax subsidy, compulsory universal coverage. No insurance market will exist; only some insiured Lifetime enrollment Underuse of health care Underuse of services with lumpy costs by poor Education, information, and communication; and also underuse of preventive care and free or subsidized care of care for diseases with externalities Supply side Supplier-induced demond Increased demand by patients; raises costs of care Use provider payment mechanisms such as salaries, global budgets, and case payments Risk selection (skimming) No insurance for disabled, sick, poor, and elderly Open enrollment, community ratings,risk-adjusted premiums for individuals, compulsory or social insurance Skimping Deny benefits to the sick Social insurance, redress procedures Exclusions Exclude pre-existing conditions and certain diseases Lifetime and compulsory insurance, guaranteed for stipulated period or life of the contract renewability Monopoly or insurance cartel Excess profit, poor quality products, underproduction Antitrust laws 250 * Better Health Systems for India's Poor Linking health insurance to employment in an organization will not provide extensive coverage in India because most people are self- employed, or do not have a formal employer or steady employment. Many of the poor are excluded from access to high quality health care and health insurance because of inability to pay, lack of knowl- edge, or other factors related to geography or discrimination. Grow- ing international experience with insuring the informal sector shows some possibilities and limitations that can complement India's own experience (box 8.4). Box 8.4 International Experience with Health Insurance for the Informal Sector Providing health care and financial protection for people who do not work in the formal sector-those who do not have regular salaried jobs with social security or employ- ment benefits and taxed incomes-are major challenges in most low-income countries. In surveys worldwide, workers in the informal sector regularly single out health insurance as their greatest insurance need. Experiments in China and Tanzania are two examples of how countries have dealt with this challenge: In China, rural health insurance covers hospital and pri- mary health care costs through private and public contribu- tions. Premiums paid by beneficiaries are supplemented through a village public welfare fund and government sub- sidies. In Tanzania, a pilot project provides health insurance through five mutual associations of workers in the informal sector. A key feature of successful contributory insurance schemes for the informal sector is their organization around an association based on trust and mutual support (profes- sional group, village) and the administrative capacity to col- lect contributions and provide benefits. Financing Health * 251 Some of the lessons learned from health insurance schemes in the informal sector: * Risk pools that cover only a few communities or only the poor often do not have sufficient economies of scale to be financially viable and are not able to provide cross- subsidies and withstand high demands for catastrophic care. * The success of a risk pooling scheme is dependant on the quality and range of health services provided and the degree of accountability to the community. * The collection and management of revenues is often inefficient and thereby reduces the funds available for investment or for providing services. * Exclusion from an insurance scheme can be a big prob- lem; preventing unreasonable exclusion requires the capacity to regulate (for example, making participation mandatory, making contributions non-risk-related) or to provide financial incentives (for example, provide subsi- dies for the poor to join). Source: World Bank (2000e); WHO (2000c). As the health sector in India reacts to liberalization in the insurance market, the policy challenges are immense. The large and predomi- nantly fee-for-services nature of private sector spending will continuc to dominate in the short and medium term. But the transformation into more pooling is inevitable; it will bring with it the potential for improvements in efficiency and equity in the health sector as well as risks of escalating costs and of widening the gap between rich and poor. Essential tasks for policymakers to meet these challenges include laying the groundwork for appropriate policy measures and building the institutional capacity to implement them. 252 * Better Health Systems for India's Poor Box 8.5 Conclusions of the National Seminar on Health Insurance In anticipation of the passage of the Insurance Regulatory and Development Authority Act of 1999, the government of India held a seminar for senior policymakers on health insurance. The objectives of the seminar, which was held on November 16-17, 1999, were to understand the potential risks and benefits of private health insurance and to identify what the government should do to ensure that social objec- tives are met with the liberalization of health insurance. The approach proposed by the seminar was to support voluntary insurance rather than expand existing social insurance schemes. The likely impact on the poor was not clear. They could benefit from an expansion of quality in the private sector, or the gap in access to quality care might increase; the risk also remains of subsidizing the wealthy. Pro-Poor Recommendations * Reduce the public subsidy to the wealthy by charging the full cost of service to the insured who use private insur- ance, finance the regulatory agency through premiums, and reduce or eliminate tax incentives for private insur- ance, particularly indemnity based insurance. * Define a minimum package of covered services that includes preventive, maternity, and catastrophic cases in order to prevent such cases from being relegated to the public sector. * Encourage informal community financing schemes that have less regulation and lower capital deposit require- ments (such as managed care schemes through NGOs) and assess other financing options for the poor. Financing Health * 253 Health Systems Recommendations * Establish a specialized regulatory agency for health insurance that would define benefits packages, ensure the transparency and comparability of packages, define treat- ment protocols, ensure guaranteed renewal of policies, reduce ability to deny coverage on the basis of pre-exist- ing conditions, establish conflict resolution mechanisms, promote community financing, and monitor the per- formance of different schemes. * Develop quality assurance procedures in health care. Resources and Allocation and Purchasing Mechanisms The third set of financing mechanisms relates to resource decision, in the public and private health sectors. Recent work on resource allocation and purchasing has identified a number of questions tha: can help public sector policymakers fine-tune their decisions ancl improve the oversight of both the private and public sector (Preker Harding, and Travis 2000). The questions include what the resources are being spent on and for whom; the nature of the allo- cation and purchasing mechanisms; and institutional arrangement, governing resource allocation and purchasing transactions (appendix table 8A.3). The health sector in a number of states, as well as through cen- trally sponsored schemes, has experimented with different forms of contracting with the private sector (for-profit and not-for-profit providers). These forms ranged from contracting for nonmedical services (box 8.6) to purchasing medical services such as cataract sur- geries and leprosy rehabilitation. Other examples have included the purchase of laboratory services and media and sensitization cam- paigns. Policy research should pay attention not only to the effi- ciencies of various purchasing and contracting arrangements but also to the institutional requirements for their success. 254 * Better Healt Systems for India's Poor Box 8.6 Contracting Out Nonclinical Hospital Services to the Private Sector: Experiences from Krnataka The Karnataka Department of Health has been contracting out a set of nonclinical services for 82 secondary level hos- pitals in hopes of improving the maintenance of their facil- ities in a cost-effective way. The pilot initiative began in 1997 and is part of the Karnataka Health Systems Develop- ment Project, which is supported by the World Bank. The services contracted include building cleaning and maintenance and waste management. The contracts were let in a competitive bidding process, and monthly payments to the vendors are based on satisfactory performance. The experience has been generally encouraging to the Depart- ment of Health. The government is no longer recruiting unskilled labor for the tasks being contracted out, and it plans to expand the scheme. Displaced staff members have been transferred or retrained as ambulance or operating theater assistants. Overall, the contract payments have been less than the salaries they replaced. Some findings from a recent evaluation of the pilot scheme: * Patients and hospital staff reported that the level of cleanliness in the hospitals had improved and recom- mended continuation of the pilot scheme. * Performance was most satisfactory for lower-skilled and visible tasks such as maintenance of corridors, wards, and toilets and for those that required less financial input, such as minor repairs, replacements, and maintenance of exteriors. * Performance was less satisfactory in areas requiring tech- nical competence or larger financial inputs or both, such as waste management, providing supphes, maintaining safety (repairing leaks), and displaying information. Financing Health * 255 One of the main recommendations was that better per- formance and compliance is contingent on strengthening the capacity of hospital staff to supervise the work and training those working in more complex areas of waste management. Experience from several countries indicates that the early phases of contracting often involve problems of adjustment and that benefits tend to go up over time as the managers of contracts get more experience. Another consistent finding is that the price of the con- tract needs to be sufficient to provide adequate service. If the price is too low, then the service will never be adequate, whether it is provided in-house or contracted out. Under some of the catering contracts at hospitals in other state health system development projects in India, food quality and quantity reportedly declined after being contracted out because the services were priced too low. Conclusions While the consultative process that launched the studies for this report focused attention in the area of health finance on the chal- lenges of insurance, the data presented here have raised other chal- lenges. By international standards, India's commitment to public financing of critical health and public health services appears to be very low (0.9 percent of GDP for all public spending on health). The weakness of public funding may have contributed to the dominance of private financing, which overwhelmingly is out-of-pocket fee-for- service payments by patients; such payments put the poor at higher financial risk and skew the provision of critical health services to the richest. The inequities and distortions will become more apparent in 256 * Better Health Systems for India's Poor chapter 9, where we pursue the issues of financial risks and fairness in health in greater detail. Another striking finding relates to the astonishing variations across states in both public and private spending, with some states spending up to four times more per capita than others. A critical challenge for India is to find ways to improve the distribution of resources, especially for priority public health activities and services, to ensure that poorer states do not lag behind as much in terms of resources and capacity. The state-level analysis in this chapter is but one dimension of a needed countrywide public expenditure review in the health sector. Given the large variations in state-level commitments to public spending, differences are to be expected in how the resources are spent. But as shown in this report, the challenge facing India is not only a question of increasing public spending on health but of find- ing the mechanisms to raise resources, pool risks, and purchase care that will protect the poor and the sick and be more efficient in driv- ing the health system in the future. Financing Health * 257 Appendix Table 8A.1 Health Expenditures and Cost Recovery in the Public Sector of Selected States, 1996 (100,000 rupees except as noted) TOTAL TOTAL PERCENTAGE STATE EXPENDITURES USER FEES OF COSTS RECOVERED Andhra Pradesh 47,898 735 1.53 Bihar 22,130 230 1.03 Gujarat 33,564 438 1.31 Haryana 11,957 1,137 9.51 Himachal Pradesh 11,621 784 0.67 Karnataka 42,614 1,180 2.77 Kerala 31,226 4,952 15.86 Madhya Pradesh 36,218 578 1.60 Maharashtra 45,893 2,127 4.63 Northeast states 22,695 386 1.70 Orissa 19,093 183 0.96 Punjab 17,693 1,888 10.67 Rajasthan 43,161 398 0.92 Tamil Nadu 55,984 1,238 2.21 Uttar Pradesh 84,308 2,726 3.23 West Bengal 50,802 1,063 2.09 Note: Northeast states consist of Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland, and Tripura. Source: Background Paper 5. Table 8A.2 Salient Features of Some Insurance Schemes in India MANDATORY SOCIAL INSURANCE SCHEMES VOtUNTARY PRIVATE EMPLOYER-BASED COMMUNITY-BASED INDICATORS ESIS CGHS INSURANCE-MEDICLAIM SCHEME INSURANCE/NGOS Beneficiaries Factory sector em- Employees of * Individuals and Public and People in the ployees with central government- groups with private sector communities income of less current and retired, persons ages employees than Rs 6,500 some autonomous 5 to 75 years. per month (their and quasigovern- * Children between dependants are mental organizations, 3 months and 5 also covered) MPs, judges, free- years covered dom fighters, journa- with parents. lists r-J Coverage About 35.3 million About 4.4 million * 1 .7 million bene- About 30 million * About 30 million x0 beneficiaries in beneficiaries in ficiaries beneficiaries in beneficiaries in 1999 1998 1996 * Urban poor and 1999 * Normally quarter of groups more likely eligible populations to purchase policy targeted Benefits eMedical benefits, *All outpatient * Hospitalization Categorized under Mainly preventive care; cash benefits, facilities, preven- according to * GHIP also ambulatory and * Preventive and tive and promotive benefit level * Reimbursements inpatient care. promotive care, care available in * Exclusions and * Lump-sum and health dispensaries waiting period payments education. * Inpatient facilities * Maternity * Own facilities available in govern- benefits allowed ment hospitals and with extra in approved private premium hospitals on being referred Premiums * 4.75 % of * Vary from Rs 15 Premiums based Based on benefits * Financed by patient (financing employee wages to Rs 150 per on age and collection, government of scheme) paid by employers month according benefit level grants, and donations * 1.75 % of to salary * Premiums depend on employee wages * Mainly financed the scheme-flat rate or paid by employees by central income-based * 12.5 % of total government expenses paid by state governments Provider * Mainly salaries * Salaries for Indemnity type- * Salaries to Mainly fee for service payments for physicians in doctors insured pays to providers in dispensaries and * Providers not the provider who own facilities referral hospitals allowed is later reimbursed * Fee for service * Hospitals have private practice according to the by patients, global budget * Treatment in benefit level. covered partly financed by ESIS private or wholly by through state hospitals is the company governments. reimbursed on case basis subject to actual expen- diture and pre- scribed ceilings. Administrative About 21% of reve- *Direct administrative *Generally high Depends on the Generally low (3-5% costs nue expenditure for costs including *Low claim-premium scheme imple- depending on the scheme) employee wages, travel expenditure, ratio because a mented by the administering cash office expenses- large proportion of company (will benefits, revenue 5% of total funds are spent on be highest for recovery, and imple- expenditure administration or own facilities) mentation in new area ' Part of salaries can kept as profits also be charged to administrative costs. Note: ESIS, Employees State Insurance Scheme; CGHS, Central Government Health Scheme; NGOs, nongovernmental organizations; GHIP, Group Health Insurance Program. Source: Background Paper 1. 260 * Better Health Systems for India's Poor Table 8A.3 Questions for Policymakers to Ask when Deciding on Resource Allocations and Purchasing ISSUE QUESTIONS Core policy * Demand (for whom to buy?) characteristics * Supply (what to buy, in what form, and what to exclude?) * Prices and incentive regime (at what price and how to pay?) Organizational * Organizational forms (what is the degree of economies of characteristics scale and scope and what are contractual relationships?) * Incentive regime (what is the degree of decision rights, market exposure, financial responsibility, accountability, and coverage of social functions?) * Linkages (what is the degree of horizontal and vertical integration or frogmentation?) Institutional * Stewardship (who controls strategic and aperational characteristics decisions?) * Governance (what are the ownership arrangements?) * Insurance markets (what are the rules on revenue collection, pooling, and transfer of funds?) * Factor and product markets (from whom to buy, at what price, and how much?) Source: Preker, Harding, and Travis (2000). Financing Health * 261 Figure 8A.1 Structure of Health Care Sources and Uses Sources .- Donor Government of India Corporate sector, Households -. agencies (Disbursements) charitable hospitals (Grants and (out-of-pocket) X, loans) -. State Government ' ( - (including municipalities) (donations) _~ ~iJ L1 1 Nonpian - ----- - 4, 4r ," \ t Consultation ^ Medical Public Family \ Medicine (plus ESIS) health welfare Tests r t r \ * ~~~~~~~~~~~~~Hospitalization Co | Government health care facilities c; Teaching hospitals Private hospitals ls District hospitals Nursing homes 5 Subdivisional hospitals Private practitioners i Community health centers Traditional birth attendants X Primary health centers NGOs Subcenters/dispenseries Capital, recurrent personnel,- - equipment, drugs, supplies, etc. 2 Note: As sources, the State Government and municipalities spend their own revenues on health as well as passing through disbursements from the central government. 'Plan' refers to funds coming from Ninth Plan allocations for devel- opment purposes; CSS, centrally sponsored schemes; ESIS, Employees State Insurance Scheme; NGOs, nongovernmental organizations. Source: Adopted from Sharma, McGreevey, and Hotchkiss (2000). 262 * Better Health Systems for India's Poor Figure 8A.2 Out-of-Pocket Payments and Household Income, 1995-96 Rupees 600 - 500 400 300 200 100 -, - 1 st 2nd 3rd 4th 5th (poorest) (richest) Income quintile El Public facilities * Private facilities Source: Background Paper 18. Financing Health * 263 Figure 8A.3 Out-of-Pocket Payments by Socioeconomic Group, 1995-96 Rupees 350 300 -5- 150-. .- 100 U ... 50 U.. Poor Not Scheduled Not in Rural Urban poor caste or scheduled tribe caste or tribe Source: Background Paper 1 8. 264 * Better Health Systems for India's Poor Notes 1. Data and analysis of resource allocation within the public sec- tor and by households in the private sector are also presented in chapter 7. 2. The most recent source of detailed private spending on health is the 1995-96 National Sample Survey (National Sample Survey Organisation 1998). 3. In keeping with standardized accounting methods for national health accounts, government expenditures on health-related areas of food distribution, water, and sanitation have been excluded from these estimates. Central and state government spending on health and family welfare is included. 4. Centrally financed services address different elements of the health sector (chapter 7). 5. Taxation in India is widely assumed to be progressive (Garg 2001), but detailed analytical work on the question is not available. r - CHAPTER 9 Health System Outcomes This chapter reviews the current state of health systems outcomes in India according to the health system objectives laid out in chapter 1: * Improve the health status of the population by lowering mortal- ity and morbidity rates * Protect the population against the financial risks of health problems * Respond to citizens' demands and needs. Improving the health status of the population is often considered to be the most important objective of the health system, but the pre- viously neglected areas of financial protection and health system responsiveness are also important; indeed they become critical for India as it moves into the health transition. Therefore, we analyze outcomes for all three objectives. We hope that in doing so we will stimulate a careful and transparent monitoring of the results of the health system, better planning for future interventions, and greater efficiency, equity, and public accountability in the health sector. See box 9.1 for a summary of the empirical findings and policy chal- lenges discussed in the chapter. 265 266 * Better Health Systems for India's Poor Box 9.1 Empirical Findings and Policy Challenges Health System Actors, Functions, and Outcomes People t D~~emand Financing Health status Srvice FinanciTalstatu-s /- ~~~~ \ (del~~eiver t Input management delive | Responsiveness to public| C ~~~Oversight Private sector actors The state Empirical Findings Health Status * Despite large gains in health status since independence, the poor continue to suffer widely from readily curable or preventable conditions-"the unfinished agenda" in world health. The poorest quintile of Indians have more than double the mortality rates, malnutrition, and fertil- ity of the richest quintile. * Disadvantaged groups in India-scheduled tribes and castes-have consistently worse health outcomes than other groups. Females also do worse in many health out- comes, as do people living in rural areas. * Because the country bears a large share of the world's dis- ease burden, progress toward global targets for diseases Health System Outcomes * 267 of the unfinished agenda depends on increased action in India. * With rising life expectancy, changing lifestyles, and progress in addressing the unfinished agenda, the epi- demiological profile of India will shift increasingly toward diseases with higher costs per episode. The top causes of premature death and disability in India reflect this transition: (a) acute lower respiratory infections, (b) diarrheal diseases, (c) ischemic heart disease, (d) injuries from falls, and (e) major depression. Persistently high rates of maternal mortality and morbidity are linked to high fertility levels. * There is no unique "Indian" health status but rather a diversity of state-level situations. Health conditions in Kerala are comparable to those in upper-middle-income countries such as Trinidad, Argentina, and Mauritius. Most Indian states, however, are comparable to lower- middle-income countries such as Brazil, Egypt, and Peru. Conditions in the poorest performing states are similar to those in low-income countries such as Sudan, Nigeria, and Tanzania. Financial Protection * The lack of prepayment systems for health care has put Indians at great financial risk in the event of hospitalization, and most of their total expenditures are in fact for hospital- ization. About one-fourth of hospitalized Indians fall below the poverty line as a result of their hospital stays. The use of public hospitals reduces this risk only marginally. * Financial risk from serious illness affects nearly all income groups in India, with more than 40 percent of hospitalized patients depending on loans and the sale of assets to pay for hospitalization. 268 * Better Health-Systems for India's Poor Box 9.1 (continued) * Cost remains a significant barrier to the use of health care, particularly for the poor, who cannot afford the level or quality of care they need. Cost is a greater bar- rier than physical access to health providers. Responsiveness to the Public * India has a comprehensive set of laws, but mechanisms for enforcement of those laws and for the redress of patients' complaints function poorly, whether at the facility level, in consumer forums, or in the civil and criminal courts. These mechanism are in any case largely inaccessible to the poor, who require intermediaries for protection and meaningful voice. * Users of the public hospitals are less satisfied than those in the private sector. Women who use public health facil- ities express concerns about lack of privacy, lack of female doctors, lack of confidentiality, and long waiting times to see a doctor. In Andhra Pradesh and Uttar Pradesh, specific areas of dissatisfaction include the cost of services (particularly unofficial payments for nomi- nally free services and drugs), lack of cleanliness of facil- ities, and behavior of health staff. Policy Cballenges * How can the health system be made more pro-poor, gen- der-sensitive, and client-friendly to respond to the high burden of preventable disease borne by the poor, women, and scheduled tribes and castes? * How can India strengthen financing systems to reduce the large financial risks faced by Indians, particularly the poor, when they become ill? As the burden of disease Health System Outcomes * 269 shifts toward diseases with a high cost per episode, this challenge becomes more pressing. * How can the experience of stronger states be replicated in weaker performing states to improve health status and reduce polarization? * How can new and more responsive mechanisms be put in place to protect consumer interests and increase social accountability of the health system? In addition to better enforcement of existing laws, can other approaches be used, such as promoting consumer rights, increasing public awareness of health and consumer issues, and enhancing the voice of the poor? Health Status We begin the analysis of health status outcomes by comparing India's health burden with that of the world and then examine India's trends in mortality, fertility, malnutrition, and illness. We also focus on disparities in health status within India, looking at differences between states, sexes, the poor, and scheduled castes and tribes. India's Share of the World's Health Problems India carries a large burden of the world's disease, as one would expect given its large population and high levels of poverty (table 9.1 and appendix table 9A.1). India has a major portion of the world's child and maternal deaths as well as a disproportionate amount of the disease burden due to tuberculosis, leprosy, and immunizable diseases. Worldwide progress against these conditions will depend on India's achievements. Also, although India's burden of HIV infec- tions is not disproportionate (14 percent of the world's cases), India will need to play a central role in preventing the further spread of the HIV/AIDS pandemic. 270 * Better Health Systems for India's Poor Table 9.1 India's Share of the World's Health Problems (percent) PEOPLE UNDER- LIVING FIVE IN POVERTY MORTALITY POPULATION (LESS THAN TOTAL (DEATHS PER MATERNAL US$1 /DAY) DEATHS 1,000 LIVE BIRTHS) DEATHS 1 7 36 1 7 23 20 DEATHS PREVENTABLE WITH CHILDHOOD TUBERCULOSIS LEPROSY DALYS LOST VACCINATIONS HIV CASES CASES CASES 20 26 1 4 30 68 Note: DALYs, disability-adjusted life years. Source: World Bank (2000d); WHO (2000c). Although India's share of the world's deaths (appendix table 9A.2) is about equal to its share of the world's population (17 percent), the country's contribution to pre-health transition diseases affecting younger people is disproportionately high (table 9.2 and appendix table 9A.3). Compared with its share of the world's population, India has high levels of deaths due to traditional childhood infectious dis- eases such as acute lower respiratory infections (pneumonia), diar- rhea, measles, and tetanus. These conditions, as well as tuberculo- sis-another disease with a high death rate in India-are common to low-income countries at the early stages of an epidemiological tran- sition. However, India's top cause of death, ischemic heart disease (which results in heart attacks), is a condition of adulthood that becomes an increasing burden in countries already undergoing an epidemiological transition. Heart disease deaths are already rela- tively more common in India than in the rest of the world, despite the younger age structure of the country's population. Road traffic injury is the sixth most common cause of death in India and is also more frequent in India than in the rest of the world. A fuller measure of the burden of disease counts not only deaths but also years of healthy life lost due to disability. A consolidated measure of losses from death and disability is disability-adjusted life Health System Outcomes * 271 Table 9.2 Top 10 Specific Causes of Death in India, 1998 INDIA INDIAS PERCENTAGE OF WORLDWIDE CAUSE OF DEATH NUMBER (THOUSANDS) PERCENT CASES Ischemic heart disease 1,471 15.8 19.9 Acute lower respiratory infections 969 10.4 28.1 Diarrheal diseases 711 7.6 32.1 Cerebrovascular disease 557 6.0 10.9 Tuberculosis 421 4.5 28.1 Road traffic injury 217 2.3 18.5 Measles 190 2.0 21.4 HIV/AIDS 179 1.9 7.8 Tetanus 165 1.8 40.3 Chronic obstructive pulmonary disease 153 1.6 6.8 Total deaths 9,337 100.0 17.3 Total population 982,223 100.0 16.7 Source: WHO (1999). years (DALYs) lost. India's share of the world's total burden of DALYs lost was 19.5 percent, moderately larger than its 16.7 per- cent share of the world's population (table 9.3). Except for depres- sion, India's contributions to all leading causes of DALYs lost are greater than its proportion of the world's population. Diseases char- acteristic of the pre-health transition still figure prominently in India, particularly childhood infections (pneumonia, diarrhea, and measles), as well as tuberculosis and anemia. However, other condi- tions are also significant. Injuries from falls, road traffic accidents, and fires each account for relatively large portions of DALYs lost in India-and in comparison to the rest of the world. Ischemic heart disease and depression also make up a large burden of disease in India, ranking third and fifth respectively among specific causes of DALYs lost in India. The World Health Organization (WHO 2000c) ranks India 134th of 191 countries in disability-adjusted life expectancy at birth. Although India's infant mortality and total fertility rates are near the average for low-income countries, it has relatively high levels of childhood malnutrition (appendix table 9A.3). Fortunately, India is not yet suffering from the low levels of disability-adjusted life 272 * Better Health Systems for India's Poor Table 9.3 Top 10 Specific Causes of DALYs Lost in India, 1998 INDLA NUMBER OF INDIAS PERCENTAGE CAUSE OF DALYS LOST DALYS LOST PERCENT OF WORLD Acute lower respiratory infections 24,806 9.2 30.1 Diarrheal diseases 22,005 8.2 30.1 Ischemic heart disease 11,697 4.3 22.5 Falls (injuries) 10,898 4.1 40.3 Unipolar major depression 9,679 3.6 16.6 Tuberculosis 7,577 2.8 26.9 Road traffic injuries 7,204 2.7 18.5 Measles 6,474 2.4 21.4 Anemia 6,302 2.3 25.5 Fire-related injuries 5,723 2.1 47.8 All causes 268,953 100.0 19.5 Population (thousands) 982,223 100.0 16.7 Note: DALYs, disability-adjusted life years. Source: WHO (1999). expectancy seen across Africa (including middle-income countries like South Africa), where the HITV epidemic is further advanced. Yet in a comparison of country performance of health outcomes between 1960 and 1990 (Wang and others 1999), India performed worse than predicted by its income and education levels in the areas of mortality of those under the age of five years, female adult mor- tality, and total fertility. India performed better than expected for male adult mortality. National Health Status Trends India's life expectancy has shown remarkable improvement, rising from 49 years in 1970 to 63 years in 1998. Similarly, infant mortality dropped from 146 deaths per 1,000 births in the 1950s to 70 in 1999, while the total fertility rate fell significantly, from 6.0 in the 1960s to 3.3 in 1999 (Registrar General 1999a, 2000). Little progress has been made since the mid-1990s, however, in the area of malnutrition. Although Indians no longer suffer from the waves of famine that marked earlier periods in the nation's history, recent surveys show that Health System Outcomes * 273 47 percent of all children under three are underweight, down from 5 2 percent six years previously (IIPS 2000). In addition, nearly three- fourths of children under five are anemic, a condition that signifi- cantly harms cognitive development (IIPS 2000). Despite the marked long-term reduction in infant mortality, the pace of India's infant mortality reduction, like its progress on malnu- trition, has slowed during the 1990s (figure 9.1). Since the infant mor- tality rate is a sensitive indicator that responds to many underlying causes, including general socioeconomic conditions and the use of health services, the reasons for the slowdown are not obvious. One explanation lies in the coincident slowdown in poverty reduction ove:- the same period (Kathuria and Hanson 2000). More immediate Figure 9.1 Infant Mortality Rate in India, 1980-2000 Deaths per 1,000 live births 120 110- 100-\ 90 80 \ . 70 . 60 -,,, O -0C C) L O V0 N X 0' 0 - CN cn s ) 0 N a) 0 X X co co m co m m co o o. ol ol o. ol ol o. o os ol os soooso s ol ovo vo oo so Note: Infant mortality is death before one year of age. Source: Registrar General (various years). 274 * Better Health Systems for India's Poor causes are related to the stubbornly high levels of malnutrition and disease; the latter is due in part to persistently high levels of exposure to disease combined with languishing rates of immunization coverage and low use of health services for safe motherhood and common dis- eases of infancy (Claeson, Bos, and Pathamanathan 1999). The introduction of child survival initiatives in the 1980s-such as universal immunization and control of diarrheal disease and acute respiratory infections-may have made important contributions to the reduction of postneonatal mortality rates (from age one month to one year). Neonatal deaths now comprise the majority of infant deaths and may be attributed to poor maternity care, maternal mal- nutrition, and a high risk of neonatal infections (Ministry of Health and Family Welfare 2000b). National surveys show that about half of all women of reproductive age were anemic in 1998-99 and that only one-third of births were institutional deliveries (IIPS 2000). Each of these conditions contributes to high neonatal mortality. Although the decline in adult mortality accelerated through the 1990s compared with previous decades, the rate of decline of child mortality was faster in the 1980s than it was in the 1970s or the 1990s. The annual rate of decline in mortality for children under five was 3.6 percent between 1981 and 1991, before dropping to 2.0 percent between 1991 and 1996. The same rate dropped by 2.1 per- cent per year between 1971 and 1981. Because India's population is young, reductions in overall death rates for India also slowed during the 1990s compared with the 1980s. Overall death rates and DALYs lost appear to have dropped somewhat in the last decade (appendix table 9A.4), although part of the change may be due to differences in the measurement of mor- tality. Shifts in the composition of causes of death and disability appear to have accelerated over the 1990s. Chronic diseases of adulthood, notably heart disease and depression, as well as injuries, are playing an increasingly important role in India's burden of dis- ease. HIV is also increasing at rapid rates. At the same time, India continues to have a high burden of readily preventable and treatable conditions due to childhood communicable diseases, tuberculosis, Health System Outcomes * 275 malnutrition, and maternal illness. These illnesses are concentrated in poorer states and among the poor. Health Status Disparities We have seen that the causes of death and disability are shifting rap- idly in India while the overall levels of mortality and malnutrition are stabilizing at relatively high levels. In this section we will demonstrate how health outcomes are becoming increasingly polarized, first bv providing some international comparisons of health outcomes and then by comparing health outcomes according to geographic region, gender, membership in scheduled castes and tribes, and poverty level. International Comparisons Both the average levels and inequality in health between the pooI and better-off vary considerably from country to country (box 9.2) All countries show variations in health status-some people die ir childhood while others survive to old age. Compared with many countries, however, India displays a high degree of variability ir health status across its population. For example, India ranks 153rd oi 191 countries in variability of child mortality (WHO 2000c). Esti- mates point to more variability in India than in China (ranked 101, and Bangladesh (ranked 125). India's variability of child mortality is similar to other low-income countries, like Ghana (ranked 149), Indonesia (ranked 156), and Nepal (ranked 161), but it is greater than that of C6te d'Ivoire (ranked 181), Pakistan (ranked 183), and Nige- ria (ranked 188). Health outcomes for women are generally worse than for men throughout the world; in India the disparity is more pronounced (World Bank 2000d). Although females are in the majority in most countries, only 48.4 percent of India's population is female, the eighth-lowest proportion in the world. Life expectancy at birth is usually considerably higher for females than males1-4.1 years higher worldwide. However, the difference in India is only 1.5 years. Among 65 low- and middle-income countries for which data are available, 276 * Better Health Systems for India's Poor Box 9.2 "Achievement" Means Doing Well for Everyone- Not Just the Better-Off Countries-and states within countries-typically vary both in average health and in the degree of inequality in health between the poor and better-off. For example, according to demographic and health surveys between 1990 and 1998, India had a lower under-five mortality rate than Bangladesh (119 compared with 128 per 1,000), but the inequality between the poor and the better-off was higher in India (Gwatkin and others 2000). A country's or state's "achievement" index captures both these considerations. Higher mortality rates among the poor push the achieve- ment index (or, more correctly in this case, the nonachieve- ment index) above the sample mortality rate. The bigger the inequality by wealth, the greater the proportional "wedge" between achievement and the sample mean. The simplest achievement index is the mean multiplied by the complement of the concentration index of inequality (Wagstaff 2002). The concentration index ranges from -1 (all mortality concentrated among the poorest), through 0 (all children have the same rate), to +1 (all mortality con- centrated among the richest) (Kakwani and others 1997). To calculate the achievement index for under-five mortality, we take India's concentration index for under-five mortality, which is -0.17; find its complement, which means subtract- ing it from 1 (that is, 1 - (-0.17) = 1.17); and multiply the result by the under-five mortality rate (1.17 x 119 = 139). The final result is the achievement index for under-five mortality-139 per 1,000. By contrast, Bangladesh's con- centration index is only -0.08, so its achievement index is 108 percent of 128, or 139 per 1,000-the same as India's, despite its higher average under-five mortality rate. Health System Outcomes * 277 Under-five mortality rate perl ,000 live births 160- <---r- rr------- - --- 140- ' _ _ _ 120 X 1 li II_ 80- 1 IIIIlIIIVi\- = 60 l ll ll l 20 - - ° : s Q > t' ° .0 oc2 E 2 C - a - 00 N m c c, CL. :~ -o >- _- -0 c c cn c N N =Mean -- Achievement Note: Values for 'achievement" are the mean values adjusted to reflect degree of concentration of deoths among the poor. Source: Gwatkin and others (2000), which used demographic and health surveys for various years, 1990-98; and authors' calculations. India has the sixth-largest difference between female and male chilc. mortality rates-1 3 deaths per 1,000 births. Geographic Disparity Disparities in India are also reflected in the large and growing inter- state differentials in infant mortality (appendix table 9A.5). In 1998, infant mortality rates were as high as 98 per 1,000 live births in Orissa and Madhya Pradesh, 85 in Uttar Pradesh, and 83 in Rajasthan (Registrar General 1999b). At the other end of the spec- 278 * Better Health Systems for India's Poor trum, Kerala had a remarkably low rate (16), followed by Maharash- tra (49), Tamil Nadu (53), Punjab (54), and Kamataka (58). Within states, large differences appear between districts, with the worst-off districts found in states with the poorest overall mortality rates. Urban areas consistently have better health outcomes than rural areas, although these figures probably conceal the extent of poor health in urban and periurban slums, where many migrants live and where few organized primary health services are available. Fertility rates also vary widely among the states (see appendix table 9A.5). These differences are significant not only because they prompt different strategies for fertility reduction in different states but also because population shifts have important implications for changes in the division of public resources and political representation. Only two of the fifteen major states have reached so-called replacement levels of fertility (where the total fertility rate is no greater than 2.1 births per woman aged 15 to 49). Another five major states have total fertil- ity rates of between 2.1 and 3. However, in eight major states the rate is 3 or more; these states comprise about 44 percent of India's popu- lation and include the poorest states with the highest rates of infant mortality. The differences among states stem largely from poverty, illiteracy, and inadequate access to health and family welfare services (Ministry of Health and Family Welfare 2000b). Interstate differences in malnutrition are also noteworthy (appen- dix table 9A.6). In a pattern consistent with the synergistic relation- ship between illness and malnutrition, the states known to have poor mortality outcomes also have higher levels of malnutrition. Even in the states with the lowest levels of malnutrition (Kerala and Punjab), more than one-fourth of children below three years of age are still malnourished. While the situation in Punjab and Haryana is no doubt improved by fertile and irrigated land, Assam has relatively low levels of malnutrition despite its relative lack of the endowments enjoyed by Punjab and Haryana. To assess changes in state levels of malnutrition over time, we exam- ined the results of two rounds of the National Family Health Survey (IIPS 2000). During the six years between the two rounds (1992-93 Health System Outcomes * 279 $P -Xtam - ) -V ~ ~ ~ ~ ~ ~ ~~~ -L .i4w A refreshing shower (PHOTOGRAPH By RAY WITLIN/THE WORLD BANK PHoTo LIBRARY) and 1998-99), levels of severe and total malnutrition (6.4 percent) have declined ontly marginally among children under three in India (to 2.6 percent and 6.4 percent respectively), according to measures of weight for age (appendix table 9A.7). Malnutrition levels are estimated to have increased between the rounds in Rajasthan, Haryana, and Madhya Pradesh; all other states have shown varying amounts of improvement. The largest gains in child nutrition status were observed in the states of Punjab, Assam, Tanmil Nadu, Andhra Pradesh, and Kamataka, all of which improved by more than 10 percentage points. 280 * Better Health Systems for India's Poor The high level of diversity in under-five mortality between and within states can make national measures highly misleading from a policy point of view. Some states compare well with middle-income countries, whereas others fare much worse (table 9.4). Kerala is a clear outlier-its under-five mortality rate is comparable to that of upper-middle-income countries such as Trinidad, Argentina, and Mauritius. The rate in most Indian states, by contrast, is compara- ble to that in lower-middle-income countries such as Brazil, Egypt, and Peru. The poorest-performing states compare with poor coun- tries such as Sudan, Nigeria, and Tanzania. The Indian average rate Table 9.4 Under-Five Mortality Rates in Indian States and in the World UNDERFIVE MORTALITY RATE BY COUNTRY INCOME GROUP (DEATHS PER 1,000 LIVE BIRTHS) INDIAN STATE INTERNATIONAL COMPARISON Upper-middle-income countries less than 20 Kerala Trinidad, Uruguay, Argentina, Mauritius Lower-middle-income countries 40-50 Himachal Prodesh, Goa Brazil, Tunisia, Peru, Dominican Republic 50-65 Mizoram, Delhi, Manipur, Guatemala, Egypt, Morocco Maharashtra, Tamil Nadu, Nagaland 65-80 West Bengal, Karnataka, Papua New Guinea, Bolivia Sikkim, Punjab, Haryana Low-income countries 80-100 Jammu and Kashmir, Benin, Yemen Gularat, Andhra Pradesh, Assam, Arunachal Pradesh 100-140 Orissa, Bihar, Rajasthan, Sudan, Nigeria, Zimbabwe, Meghalaya, Uttar Pradesh, Tanzania Madhya Pradesh Greater than 140 No Indian State Angola, Cameroon, Ethiopia, Malawi, Mali, Uganda, Zambia Note: Under-five mortality is measured from birth up to five years of age. Source: Data for Indian states are from IIPS (2000). Data for countries are from World Bank (2000d). Health System Outcomes * 281 for under-five mortality is now 94.9, a level considerably higher than the average of 79 for low- and middle-income countries. Gender Disparity As noted above, gender disparity in health outcomes is particularly prominent in India. India has approximately 933 females to every 1,000 males and has had that low ratio for more than 30 years.1 The largest gender disparities in India are found in the northern states, notably Haryana and Punjab, despite their relative prosperity Among Indian states, only Kerala had more women than men in 200 1. The low ratio of women to men is usually attributed to a pref- erence for sons, discrimination against girls (which results in lowe- female literacy, among other things), female feticide, and higher mortality levels among females (Ministry of Health and Family Wel- fare 2000b). Girls start out having lower mortality rates than do boys during the first month of life (the neonatal period), which accounts for their lower rates of infant mortality (that is, death in the first year of life) (appendix table 9A.8). However, death rates in the postneonatal period (age one month to one year) and in the whole period up to age five (under-five mortality), are higher for girls (IPS 2000). Girls have higher childhood mortality despite the fact that boys are reported to have a higher prevalence of acute respiratory infections and similar levels of diarrhea and anemia-major causes of child- hood death. This paradox may be explained by the fact that boys are more likely to receive health care: 66.5 percent of boys with acute respiratory infections are taken to a health provider, compared with 60.8 percent of girls (IIPS 2000). Girls also have marginally higher rates of malnutrition, which places them at higher risk of severe ill- ness and death. In contrast to childhood illnesses, the prevalence of medicall-y treated tuberculosis is significantly lower among females than males (appendix table 9A.8), which is also the case in most of the workc. The possible reasons are that males are more likely to come int) contact with tuberculosis or simply have a higher susceptibility, 282 * Better Health Systems for India's Poor though it has also been suggested that their higher smoking rates may also contribute (IIPS 2000). The relative neglect of women's health is also reflected in poor reproductive health indicators: maternal mortality is estimated at over 407 deaths per 100,000 live births in India (Registrar General 2000), compared with an average of 350 among low- and middle-income countries (World Bank 1997a). A major reason for the poor maternal health outcomes are the high levels of malnutrition among women. In 1998-99, 52 percent of all women of reproductive age were found to be anemic, and 36 percent were chronically malnourished (HPS 2000). Low levels of access to, and utilization of, safe motherhood services could also contribute to higher maternal mortality. Scheduled Castes and Tribes Data on health outcomes among scheduled castes and tribes show consistently that these groups are at a disadvantage (table 9.5). Of all disadvantaged groups in India, scheduled tribes tend to have the highest rates of infant and child mortality, malnutrition, and mor- bidity, followed by scheduled castes and then by other disadvantaged (or "backward") classes. Total fertility rates are highest among scheduled castes. Poverty Disparity Large-scale studies that assess specific causes of illness and death by level of poverty are not available in India. Yet India continues to suf- fer disproportionately from communicable diseases, malnutrition, and maternal conditions, which constitute the "unfinished agenda" of the health transition (Murray and Lopez 1996; World Bank 1997b). These conditions are concentrated among the poor (Gwatkin and others 2000). Compared with the poor, the rich are sick less, become sick at an older age, and suffer more from non- communicable diseases than from communicable diseases. Examination of health status outcomes for Indians grouped accord- ing to wealth shows that the poor have much higher levels of mortal- Health System Outcomes * 2 8 3 Table 9.5 Health Outcomes among Scheduled Castes, Tribes, and Rest of Population in India, 1998-99 OTHER SCHEDULED SCHEDULED DISADVANTAGED REST OF OUTCOME CASTES TRIBES CLASSES POPULATION Infant mortality (per 1,000 births) 83.0 84.2 76.0 61.8 Under-five mortality (per 1,000 births) 119.3 126.6 103.1 82.6 Total fertility rate 3.15 3.06 2.83 2.66 Children underweight (percent) 53.5 55.9 47.3 41.1 Children with anemia (percent)a 78.3 79.8 72.0 72.7 Children with acute respiratory infection, during two-week period (percent)a 19.6 22.4 19.1 18.7 Children with diarrhea, during two-week period (percentla 19.8 21.1 18.3 19.1 Anemia among women (percent) 56.0 64.9 50.7 47.6 Note: Infant mortality is of those less than one year of age; under-five mortality is of thosE' less than five years of age; total fertility rate is lifetime births per woman ages 15-49; underweight children are those under three years of age whose weight is low for their age (that is, statistically below normal by more than 2 standard deviations). a. Children under age three. Source: IIPS (2000). ity, malnutrition, and fertility than do the rich (table 9.6). Compared with the richest quintile, the poorest quintile generally had double the risks of infant and child death, malnutrition, and high fertility. Although data from the most recent round of the National Fam- ily Health Survey are not yet available for direct comparison of trends among poverty quintiles (as was done with the 1992-93 sur- vey), some early results show similar trends. Indians classified as hav- ing a low living standard suffer worse health outcomes than do those 284 * Better Health Systems for India's Poor Table 9.6 Health Status Indicators-Comparison between the Poorest and Richest Quintiles of the Indian Population, 1 992-93 POOPEST OUINTILE RICHEST GUINrILE RISK RAirO INDICATOR (1) 12) (1) / (2) Infant mortality (per 1 000 births) 109 44 2.5 Under-five mortality (per 1,000 births) 155 54 2.8 Underweight children (percent) 60 34 1.7 Total fertility rate 4.1 2.1 2.0 Note: For definitions of indicators, see general note to table 9.5. Source: Gwatkin and others (2000). classified as having a medium or high living standard (table 9.7). In particular, those from the low-standard-of-living group had more than double the rates of infant and child mortality and childhood malnutrition than the high-standard-of-living group. They also had 60 percent higher fertility rates, 34 percent more childhood acute respiratory infections, and 24 percent more childhood diarrhea. The smallest difference was for children with-anemia: all groups had high levels, ranging from 67 percent to 79 percent. Many pretransition diseases are concentrated among poor states and among poorer districts within states. For example, the poorest states in India account for 70 percent of the leprosy cases but for only 46 percent of the total population.2 Equity analysis in these states confirms that the most socially vulnerable groups (scheduled castes and women below the poverty line) shoulder the highest burden of the disease (World Bank 2001). In Bihar, for example, prevalence rates for leprosy were 50 percent higher than the state average for men and women below the poverty line. The preva- lence rate for scheduled castes is even higher-twice the state average. In Uttar Pradesh and Orissa, prevalence rates for sched- uled castes are 2 to 2.5 times that of the state average. In West Bengal, prevalence rates for women below the poverty line were 4 times the state average. Malaria has also been shown to be more prevalent in rural dis- tricts that have higher levels of illiteracy and larger populations of scheduled tribes (World Bank 1997c). Tuberculosis rates are highest Health System Outcomes * 285 Table 9.7 Health Outcomes by Standard of Living, 1998-99 STANDARD OF LIVING RATIO, OUTCOME LOW MEDIUM HIGH LOW TO HIGH Infant mortality (per 1,000 births) 88.8 70.3 42.7 2.08 Under-five mortality (per 1,000 birthsl 130.0 94.6 51.5 2.52 Total fertility rate 3.37 2.85 2.10 1.60 Children underweight (percent) 56.9 46.8 26.8 2.12 Children with anemia (percent)l 78.7 73.5 67.3 1.17 Children with acute respiratory infection, during two-week period (percent)a 21.0 19.4 15.7 1.34 Children with diarrhea, during two-week period (percent)a 19.9 19.7 16.1 1.24 Anemia among women (percent) 60.2 50.3 41.9 1.44 Note: For definitions of outcomes, see notes to table 9.5. a. Children under age three. Source: IIPS 12000). in poorer households living in crowded conditions. Maternal death-s are also found more frequently in poorer states (Registrar General 2000). Yet to label some diseases as diseases of the poor and others as diseases of the rich would be simplistic. Although communicable diseases, malnutrition, and maternal conditions are concentrated among the poor, they are not the only diseases that beset the poJr. Chapter 7 reports that poorer Indians already have higher rates of ischemic heart disease and hypertension than more affluent Indians, and have higher prevalence of alcohol and tobacco use, importmnt risk factors for noncommunicable diseases. The poor at all ages have a higher risk of getting virtually all diseases and of having worse c ut- comes (Frank and Mustard 1994). 286 * Better Health Systems for India's Poor Financial Protection Health care financing matters for two reasons. First, it influences how much and what type of health care people receive when they fall ill, which in turn influences their ability to maintain and improve their health. In other words, health financing affects access to and use of health services. Second, health care financing influences the ability of a household to maintain its living standards when one of its members needs health care. This is the issue of financial protection. Consultations with the poor in India (Narayan and others 2000) revealed that, after illiteracy and unemployment, spending on health care was the greatest precursor to poverty among poor households and the greatest impediment to continued household solvency. This section examines how financial protection in health operates in India, focusing on differences among Indians at different income levels. More specific assessments of health financing functions and how they influence financial protection are examined in chapter 8 and in the discussion of policy options in chapter 3. In addition to examining the issue of financial protection, this section also exam- ines the other side of the coin-the effect of health financing arrangements on access to and use of health services. Financial Protection and Progressivity On average, about 5.3 percent of annual household expenditure in India is spent on health care, or about 13.7 percent of nonfood expenditures. As a proportion of nonfood expenditure, richer Indi- ans spend marginally more than poorer Indians on health care, so the distribution of out-of-pocket health expenditures may be con- sidered progressive (figure 9.2).3 One reason is the linking of fees to a patient's income. According to the 1995-96 NSSO survey, fees paid at govermnent health facilities have been very progressive, with the richest quintile of Indians paying 73 percent of the fees and the poorest quintile only 1.4 percent (Background Paper 5). However, these fees do not represent a significant portion of the actual costs of care, and not all the direct costs of illness are captured at public facil- Health System Outcomes * 287 ities. Moreover, most out-of-pocket expenditures from a hospital- ization do not go to the public hospital; diagnostic tests, drugs, materials, and other items often have to be purchased separately. In the private sector, too, fees are often charged on a progressive slid- ing scale, with the poor paying less. In Andhra Pradesh and Uttar Pradesh, nearly all private providers offer some level of free or dis- counted care to the poor (chapter 6). The other main source of health revenue is taxation, which is assumed to be highly progressive in India. Overall, then, payments for health care are progressive. Viewed in this way, the fairness of financing might not seem to be a major issue in India. However, we also need to examine the effect of health financing on poverty and access to services. Figure 9.2 Distribution of Total Out-of-Pocket Health Expenditures in India as a Proportion of Nonfood Expenditures, 1 995-96 Percent 16 14 12 m 2- ~~ - 4 2 1 st 2nd 3rd 4th 5th (poorest) (richest) Income quintile Source: National Sample Survey Organisation (1998). 288 * Better Health Systems for India's Poor Financial Protection and Its Effect on Poverty The fact that out-of-pocket payments are progressive simply means that they do not worsen the distribution of income. It says nothing about their effect on poverty. Spending even 4 percent of household income (or 12 percent of nonfood expenditures) on out-of-pocket payments represents a large drop in living standards for a low- income household and may make the difference between being just able to manage and falling into poverty. The evidence suggests that the very large out-of-pocket payments in India do indeed leave people at great risk of financial catastrophe in the case of serious illness. Analysis of the NSSO data shows that the cost of a hospitalization for nearly all people is extremely high when compared with their total annual expenditures, averaging 58 percent. Although the richest 20 percent of Indians pay the highest proportion, they are not the most financially vulnerable. The rich have more resources to pay for their health care, whereas the poor lack savings, assets, income, and the ability to borrow at low interest to pay for health care, forcing them deeper into poverty when they get seriously ill. An analysis of sources of financing for hospitalization shows that large proportions of all people borrow money or sell assets to pay for hospitalization (40 percent), but that doing so is more common among the bottom four income quintiles than among the richest, who are better able to use their current income and savings to finance hospitalization (figure 9.3). Further analysis shows that these trends hold for those hospitalized in both public and private hospi- tals, though the overall rate of borrowing and selling assets is about 5 percent lower in public hospitals. The major states of India show large differences in the degree to which people below the poverty line finance hospitalization from borrowing or selling assets (figure 9.4). The data suggest high levels of borrowing for hospitalization in both the public and private sec- tors. Even though fees for public hospital admissions are minimal or nonexistent, hospitalization is still costly because patients often have to pay for diagnostic services or drugs, or because bribes are Health System Outcomes * 289 demanded. The data also indicate that the differences between states in the level of borrowing are larger than the differences between public and private sector hospitalizations within a state. That find- ing demonstrates that the failure to provide financial protection to the poor for the costs of hospitalization is significant across the country, even in the presence of public sector hospitals that provide nominally free care. To learn how the costs of medical care might contribute to poverty levels in India, we used a conservative approach to calculate how many more people fell below the poverty line levels of consumption Figure 9.3 Sources of Financing for Private Expenditures on Hospitalization in India, by Income Quintile, 1995-96 5th (richest) E u.... 4th ffffff 3rd 2nd I St I I I I (poorest) 0 10 20 3040 50 60 70 80 90 100 Percent a*Debt *Sale of assets *Savings oCurrent income oOther I 290 * Better Health Systems for India's Poor Figure 9.4 Hospital Patients below the Poverty Line Who Financed Their Care in Public and Private Hospitals from Borrowing or Sale of Assets by State, 1995-96 Northeast states Andhra Pradesh I Tamil Nadu Karnataka . i Bihar Kerala l l_ _ Maharashtra l National average Haryana Uttar Pradesh West Bengal Gujarat Madhya Pradesh Rajasthan 0 10 20 30 40 50 60 70 80 Percent oPublic *Private Note: Northeast states consist of Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland, and Tripura. Source: National Sample Survey Organisation (1998); authors' calculations. after medical expenditures were eliminated from their consumption estimates. This analysis showed that direct out-of-pocket medical costs pushed 2.2 percent of Indians into poverty in one year.4 Another way of looking at the same question is to determine the proportion of hospital patients who would fall below the poverty line after direct hospitalization costs were deducted. By this analysis, at least 24 per- cent of all people hospitalized in India in a single year fell below the Health System Outcomes * 291 poverty line because they were hospitalized. These estimates have limitations, as it is not clear how much people would consume if they did not have to pay for hospitalization. We nonetheless believe that the estimates may underestimate the effects of hospitalization on poverty because a large proportion of hospital spending was financed from borrowing, which is less likely to occur in the absence of serious illness.5 Furthermore, the estimates include only the direct costs of medical care, which are less than the real costs of serious illness if lost earnings are also included.6 These quantitative estimates are reinforced by qualitative evi- dence (Narayan and others 2000). Households with sick and elderly people are reported to be invariably on the brink of ruin because of heavy expenditures for medical treatment. Lost wages and treatment expenses mean that poor groups are hit doubly hard by ill health. Health Finance and Access to Health Services The other side of the coin is the issue of whether the manner of health financing in India impedes access to and use of health serv- ices, and the evidence suggests that it does. The poor use health services less and do not have the same access to the facilities in the public or private sectors that are used by the better-off. (The ques- tion of who benefits from public and private health services is dis- cussed in detail in chapter 7.) Furthermore, because the poor rely so widely on untrained health practitioners. the medical care they receive is of much lower quality. The poor are less likely to seek care when ill (National Sample Survey Organisation 1998). On average, the poorest quintile of Indi- ans is 2.6 times more likely than the richest to forgo medical treat- ment when ill. Aside from cases in which people believed that their illnesses were not serious (which comprised more than half of all cases), the main reason for not seeking care was cost, particularly for the poor (table 9.8). By contrast, physical access to medical facilities was a much less common reason for not seeking care, although it too was strongly 292 * Better Health Systems for India's Poor Table 9.8 Indians Reporting an Illness within a 15-Day Period Who Did Not Seek Care, and Distribution of Reasons for Inaction, by Income Quintile, 1995-96 (percent) RATIO OF FIRST TO BEHAVIOR AND FIRST FIFTH FIFTH REASONS (POORESTi) SECOND THIRD FOURTH (RICHEST] ALL QUINTILE Did not seek care when ill 24.3 20.9 18.1 17.8 9.2 16.7 2.6 Distribution of reasons for not seeking care (percent of those not seeking care) Illness not considered serious 42.4 52.2 54.7 57.3 59.8 52.7 0.7 Cost 32.9 23.0 21.0 21.9 15.2 24.0 2.2 Medical facility not available in area 11.1 10.0 7.2 5.1 3.3 7.8 3.4 Other 13.6 14.4 16.6 15.2 21.7 15.6 0.6 Source: Background Paper 5; National Sample Survey Organisation (1998). associated with poverty. The aggregate figures mask variations across and within states, as well as gender differences, but the over- all relative importance of financial barriers to health care is similar across India. Conclusions on Financing, Protection, and Access The information on financial protection in India suggests that whereas overall financial contributions to health appear to be pro- gressive, financial risk from serious illness affects nearly all income groups in India, with people from the four poorest quintiles depend- ing on loans and sale of assets to pay for hospitalization. The lack of prepayment systems for health care has put Indians at great financial risk in the event of hospitalization, and most of their total expendi- Health System Outcomes * 293 tures are in fact spent on hospitalization. The use of public hospitals reduces this risk only marginally. Cost remains a significant barrier to the use of health care, particularly for the poor. Cost is a greater barrier than physical access to health providers. Responsiveness to the Public The final health system objective for which outcomes are exam- ined in this chapter is that of responsiveness to the public, that is, how well the health system meets the expectations of the public (WHO 2000c).7 Those expectations may include protection from fraud and abuse, respect, and satisfaction with service. Respon- siveness is becoming increasingly important to politicians and public officials as the public comes increasingly to hold them accountable for public services. In the private sector, success depends in part on responsiveness to customers. For our analysis, we assessed responsiveness by looking at how consumers' interests are protected and how satisfied consumers are with the health services they receive. Protecting Consumer Interests The Indian Law Institute compiled rulings by courts and con- sumer forums on health issues to examine how well the law pro- tected the health of citizens. It concluded that India has a com- prehensive set of legal instruments. The courts have held that health is a fundamental right, as described in the Indian Constitu- tion,8 and have been active in defining the boundaries of medical negligence. The law is much stronger on paper than in practice, however, because of weak enforcement and long delays in judicial proceedings. In part to respond to those problems, "consumer forums" were established under the Consumer Protection Act of 1986 to provide a quicker and less formal method to resolve public complaints. A 294 * Better Health Systems for India's Poor study of the forums highlighted many inadequacies (Background Paper 10). The forums were rarely approached in medical negli- gence cases (except by well-educated men from forward castes) and were technically ill-prepared to deal with such cases. Ninety percent of cases required more than a year to reach a judgment, despite the legal requirement to pass judgment within 90 days. The present study examined mechanisms of consumer redress at health facilities and found them to be poorly developed in both the public and private sectors (table 9.9).9 Although most facilities claimed to have mechanisms such as complaint boxes, these often were absent or hard to find. Large public sector institutions tended to have formal mechanisms, whereas the private sector tended to do things more informally. Among the facilities that had systems for consumer complaints, the types of complaints most commonly received had to do with hospital amenities or billing issues rather than quality of clinical care, suggesting that consumers have little knowledge about medical services and low expectations of the care they receive.10 Table 9.9 Presence of Mechanisms of Consumer Redress at Public Health Facilities and Private Hospitals (percent) PRIVATE FACILITIES PUBUC FACIUTIES 30 OR MORE FEWER THAN PRIMARY MECHANISM BEDS 30 BEDS HOSPITALS HEALTH CENTERS Procedures or guidelines for receiving and processing complaints 25 10 20 0 Unit or individual responsible for settling disputes 79 43 67 67 Complaint box or book claimed 71 33 58 40 Complaint box or book observed 33 10 4 0 Source: Background Paper 10. Health System Outcomes * 295 Client Satisfaction Studies of client satisfaction with health services have only recently been introduced in India. One of the rare sources of national-level data on patient satisfaction comes from the NSSO 52nd round (NSSO 1998). The data suggest that the private health sector responds better to patient interests than does the public sector, a difference that may explain why the private sector is used more fre- quently than the public sector for most curative services and why it is growing. Across India, 44 percent of people using the private sec- tor for outpatient care chose it because private doctors were more accessible than government doctors. Another 36 percent stated that they were less satisfied with treatment in the public sector, and 7 percent noted that medicines were not available in the public sec- tor. As shown in figure 9.5, these opinions did not vary strongly when disaggregated by income. Studies that directly measure patient satisfaction with health care are becoming an increasingly important tool for learning how health services might better respond to people's needs. Under the state health systems development projects, more such surveys are being used to monitor performance. To date they have been used for the most part to identify poorly performing hospitals and prob- lem areas within hospitals. Before a state health systems project was launched in Uttar Pradesh, a random-sample interview of more than 8,300 patients using public health services there showed fairly uniform levels of satisfaction (STEM 2000).11 The item getting the lowest marks was the financial burden of health care, followed by the cleanli- ness of the facility. Access to the facility was quite satisfactory (96 percent satisfied), but satisfaction with availability of staff, per- ceived technical quality of care, and behavior of doctors and nurses was considerably lower. Interestingly, no major differences in opinions emerged between women and men or among castes. The widest differences in satisfaction were between the wealthi- est and poorest quintiles regarding financial burden (19 percent- 296 * Better Health Systems for India's Poor Figure 9.5 Reasons for Using the Private Health Sector, by Poverty Status Percent 60 - 50 - 40 -l _ 30 - i __ 20 - | _1 I a _I 1 Below poverty line Above poverty line M Access * Quality C Medicines E Other Source: Based on Background Paper 5; National Sample Survey Organisation (I1998), 52nd round. age point difference), availability of staff (12 percentage points), and behavior of nurses (9 percentage points). This disparity sug- gests not only that the poor in Uttar Pradesh are more bothered about the financial burden of health care in public facilities, but also that those with higher incomes are likely to be more satisfied with the staff and services, possibly because they are able to pay for better service. Studies of satisfaction with private sector services explored dif- ferences in patient satisfaction between parts of the private and public sectors. In Andhra Pradesh, users of private facilities were consistently more satisfied with them than were users of public sector facilities, with the exception of nursing services used by the Health System Outcomes * 297 Table 9.10 Patients Satisfied or Very Satisfied with Health Services in Public and Private Facilities in Andhra Pradesh, by Sex and Income Level (percent) DIMENSION OF SATISFACTION WAITING DOCTOR'S DOCTOR'S NURSE'S NURSE'S EXPLANATION OVERALL PATIENT/FACILITY TIME MANNER SKILLS MANNER SKILLS OF CARE VISIT Male Public 8.4 25.0 24.6 14.4 13.9 12.2 14.9 Private 12.7 42.6 41.2 16.3 15.3 16.7 28.1 Female Public 7.3 27.6 26.8 12.0 11.2 10.0 13.2 Private 16.6 46.3 43.7 14.0 13.1 16.0 25.1 Poorest quintile Public 10.0 23.8 24.4 21.2 21.6 14.6 20.4 Private 11.7 43.0 45.4 16.2 16.2 13.0 23.5 Richest quintile Public 5.5 30.3 24.8 9.3 8.4 17.4 8.4 Private 19.9 45.7 47.0 15.8 16.7 19.7 32.6 Total Public 7.8 26.4 25.8 13.2 12.5 11.0 14.0 Private 14.9 44.7 42.6 14.9 14.0 16.3 26.0 Source: Background Paper 20; authors' calculations. poor (table 9.10).12 Among users of private sector services, the richest quintile of Indians were more satisfied overall and more satisfied with waiting times than were the poorest quintile. Among users of public sector services, the rich tended to be more dissatis- fied than the poor. Ratings by males and females were broadly similar. Similar issues have emerged from a national survey of about 3 7,000 women asked about the quality of care on their most recent visit to a health facility (table 9.11). Public sector facilities were consistently rated lower than private facilities; the greatest amount of dissatisfaction involved the indicators for cleanliness, politeness, and privacy in the public sector facilities. 298 * Better Health Systems for India's Poor Table 9.1 1 Quality of Care Reported by Women after Their Most Recent Visit to a Health Facility, Public or Private, 1998-99 (percent) INDICATOR OF CARE PUBUC SECTOR FACILIlY PRIVATE SECTOR FACIUTY Staff spent enough time 90.3 97.5 Staff talked nicely to them 62.7 78.4 Privacy was respected 68.2 83.9 Facility was very clean 52.1 75.3 Source: IIPS (2000). Concluding Remarks India bears a disproportionate amount of the world's disease burden. Moreover, India's disease profile is changing. High levels of pre- transition diseases persist as reductions in infant mortality and mal- nutrition slow their pace; at the same time, noncommunicable dis- eases and injuries are playing a growing role in the death and disability of Indians. Our analysis of disparities in health outcomes showed a large polarization of conditions within India. Interstate differences appear to be widening, with wealthier states and states in the south improv- ing at more rapid rates than poor and northern states. Individuals in scheduled tribes and castes are worse off, and women have worse outcomes than men in some respects. Finally, we showed that poverty is associated with many different types of poor health across India. Our investigation of the responsiveness of the Indian health sys- tem to public concerns showed significant gaps in consumer pro- tection and satisfaction. The poor appear to be at a disadvantage in both respects. The public and private sectors respond to consumers differentially, a subject examined in detail in chapter 6. Many improvements in responsiveness would not necessarily cost much money, but management attention and a public spotlight would be Health System Outcomes * 299 required to bring them about. Such improvements would likely contribute to more sustainable health services and improved uti- lization. The general lack of some social means of insuring against the costs of health care in India needs to be corrected. The lack of insurance exacerbates inequities in health services and leaves the seriously ill financially vulnerable, even if they use the public health system. Addressing financial protection will require structural changes in the way health services are financed (chapter 8). How- ever, such changes must also lead to improvements in quality and equity, which will contribute in turn to better health. Appendix Table 9A.1 Distribution of DALYs Lost in World, India, and Countries Grouped by Income, by Condition, 1998 INDIA PERCENTAGE HIGHINCOME LOW- AND MIDDLE- LOW- AND WORLD COUNTRIES INCOME COUNTRIES TOTAL MIDDLE- NUMBER NUMBER NUMBER NUMBER PERCENTAGE INCOME CONDITION (THOUSANDS) PERCENT (THOUSANDS) PERCENT {THOUSANDS) PERCENT (THOUSANDS) PERCENT OF WORLD COUNTRIES Memo: Distribution of population 5,884,576 100 907,828 15.4 4,976,748 84.6 982,223 100 16.7 19.7 All conditions 1,382,564 100 108,305 100 1,274,259 100 268,953 100 19.5 21.1 1. Communicable diseases, maternal and perinatal con- ditions, and nutri- tional deficiencies 565,528 40.9 7,834 7.2 557,694 43.8 135,263 50.3 23.9 24.3 A. Infectious and parasitic diseases 323,993 23.4 2,994 2.8 321,000 25.2 67,619 25.1 20.9 21.1 1. Tuberculosis 28,189 2.0 142 0.1 28,047 2.2 7,577 2.8 26.9 27.0 2. STDs excluding HIV 17,082 1.2 416 0.4 16,666 1.3 4,909 1.8 28.7 29.5 a. Syphilis 4,967 0.4 10 0.0 4,957 0.4 1,449 0.5 29.2 29.2 b. Chlamydia 7,150 0.5 354 0.3 6,796 0.5 1,982 0.7 27.7 29.2 c. Gonorrhea 4,955 0.4 46 0.0 4,909 0.4 1,479 0.5 29.8 30.1 d. Other STDs 10 0.0 5 0.0 5 0.0 0 0.0 0.0 0.0 3. HIV/AIDS 70,930 5.1 1,022 0.9 69,907 5.5 5,611 2.1 7.9 8.0 4. Diarrheal diseases 73,100 5.3 359 0.3 72,742 5.7 22,005 8.2 30.1 30.3 5. Childhood diseoses 56,855 4.1 396 0.4 56,459 4.4 14,463 5.4 25.4 25.6 a. Pertussis 13,226 1.0 179 0.2 13,047 1.0 2,692 1.0 20.4 20.6 b. Poliomyelitis 213 0.0 0 0.0 213 0.0 63 0.0 29.4 29.4 c. Diphtheria 181 0.0 0 0.0 181 0.0 75 0.0 41.1 41.1 d. Measles 30,255 2.2 188 0.2 30,067 2.4 6,474 2.4 21.4 21.5 e. Tetanus 12,979 0.9 29 0.0 12,950 1.0 5,160 1.9 39.8 39.8 6. Meningitis 4,725 0.3 154 0.1 4,571 0.4 1,191 0.4 25.2 26.1 7. Hepatitis 1,700 0.1 55 0.1 1,645 0.1 300 0.1 17.7 18.3 8. Malaria 39,267 2.8 0 0.0 39,267 3.1 577 0.2 1.5 1.5 9. Tropical diseases 10,984 0.8 6 0.0 10,977 0.9 3,204 1.2 29.2 29.2 a. Trypanosomiasis 1,219 0.1 0 0.0 1,219 0.1 0 0.0 0.0 0.0 b. Chagas disease 589 0.0 0 0.0 588 0.0 0 0.0 0.0 0.0 c. Schistosomiasis 1,699 0.1 3 0.0 1,696 0.1 0 0.0 0.0 0.0 d. Leishmaniasis 1,710 0.1 3 0.0 1,707 0.1 1,141 0.4 66.8 66.9 e. Lymphatic filariasis 4,698 0.3 0 0.0 4,698 0.4 2,063 0.8 43.9 43.9 f. Onchocerciasis 1,069 0.1 0 0.0 1,069 0.1 0 0.0 0.0 0.0 10. Leprosy 395 0.0 1 0.0 393 0.0 208 0.1 52.6 52.8 11. Dengue 558 0.0 0 0.0 558 0.0 353 0.1 63.2 63.2 12. Japanese encephalitis 503 0.0 0 0.0 502 0.0 66 0.0 13.1 13.1 13. Trachoma 1,263 0.1 8 0.0 1,255 0.1 32 0.0 2.5 2.6 14. Intestinal nematode infections 4,279 0.3 4 0.0 4,275 0.3 797 0.3 18.6 18.6 a. Ascariasis 1,292 0.1 1 0.0 1,290 0.1 163 0.1 12.6 12.6 b. Trichuriasis 1,287 0.1 0 0.0 1,287 0.1 102 0.0 7.9 7.9 c. Hookworm disease 1,698 0.1 2 0.0 1,695 0.1 532 0.2 31.3 31.4 d. Other intestinal infections 2 0.0 0 0.0 2 0.0 0 0.0 0.0 0.0 (Table continues on the following page.) Table 9A.I (continued) INDIA PERCENTAGE HIGH4NCOME LOW AND MIDDLE- LOW- AND WORLD COUNTRIES INCOME COUNTRIES TOTAL MIDDLE- NUMBER NUMBER NUMBER NUMBER PERCENTAGE INCOME CONDMON (THOUSANDS) PERCENT (THOUSANDS} PERCENT (THOUSANDS) PERCENT (THOUSANDS) PERCENT OF WORLD COUNTRIES 15. Other infectious diseases 14,163 1.0 430 0.4 13,734 1.1 6,325 2.4 44.7 46.1 B. Respiratory infections 85,085 6.2 1 488 1.4 83,597 6.6 25,556 9.5 30.0 30.6 1. Acute lower respiratory infections 82,344 6.0 1 355 1.3 80,990 6.4 24,806 9.2 30.1 30.6 2. Acute upper respiratory c2 infections 975 0.1 50 0.0 924 0.1 274 0.1 28.2 29.7 3. Otitis media 1,766 0.1 84 0.1 1,683 0.1 475 0.2 26.9 28.2 C. Maternal conditions 32,250 2.3 398 0.4 31,852 2.5 7,891 2.9 24.5 24.8 1. Hemorrhage 3,833 0.3 25 0.0 3,807 0.3 902 0.3 23.5 23.7 2. Sepsis 5,965 0.4 49 0.0 5,916 0.5 1,338 0.5 22.4 22.6 3. Hypertensive disorders of pregnancy 1,882 0.1 18 0.0 1,865 0.1 441 0.2 23.4 23.6 4.Obstructed labor 7,040 0.5 250 0.2 6,790 0.5 1,601 0.6 22.7 23.6 5. Abortion 5,498 0.4 18 0.0 5,479 0.4 1,704 0.6 31.0 31.1 6. Other maternol conditions 8,032 0.6 37 0.0 7,995 0.6 1,905 0.7 23.7 23.8 D. Perinatal conditions 80,564 5.8 2,020 1 .9 78,544 6.2 23,316 8.7 28.9 29.7 E. Nutritional deficiencies 43,636 3.2 935 0.9 42,701 3.4 10,881 4.0 24.9 25.5 1. Protein-energy malnutrition 14,931 1.1 122 0.1 14,810 1.2 3,734 1.4 25.0 25.2 2. Iodine deficiency 1,078 0.1 23 0.0 1,055 0.1 280 0.1 26.0 26.6 3. Vitamin A deficiency 2,801 0.2 8 0.0 2,793 0.2 565 0.2 20.2 20.2 4. Anemias 24,746 1.8 773 0.7 23,973 1.9 6,302 2.3 25.5 26.3 5. Other nutritional disorders 80 0.0 9 0.0 71 0.0 0 0.0 0.0 0.0 11. Noncommunicable conditions 595,363 43.1 87,732 81.0 507,631 39.8 88,657 33.0 14.9 17.5 A. Malignant neoplasms 80,837 5.8 16,257 15.0 64,580 5.1 8,754 3.3 10.8 13.6 1. Mouth and oropharynx 4,473 0.3 446 0.4 4,027 0.3 1,313 0.5 29.4 32.6 2. Esophagus 4,180 0.3 398 0.4 3,782 0.3 681 0.3 16.3 18.0 3. Stomach 8,156 0.6 1,049 1.0 7,107 0.6 615 0.2 7.5 8.6 4. Colon/rectum 5,191 0.4 1,818 1.7 3,373 0.3 307 0.1 5.9 9.1 5. Liver 7,878 0.6 391 0.4 7,486 0.6 176 0.1 2.2 2.4 6. Pancreas 1,761 0.1 669 0.6 1,092 0.1 102 0.0 5.8 9.4 7. Trachea/bronchus/ lung 11,176 0.8 3,122 2.9 8,054 0.6 921 0.3 8.2 11.4 8. Melanoma and other skin cancers 611 0.0 264 0.2 347 0.0 1 8 0.0 3.0 5.3 9. Breast 5,202 0.4 1,643 1.5 3,560 0.3 711 0.3 13.7 20.0 10. Cervix 3,183 0.2 199 0.2 2,985 0.2 836 0.3 26.2 28.0 11. Corpus uteri 705 0.1 199 0.2 506 0.0 42 0.0 6.0 8.3 12. Ovary 1,545 0.1 403 0.4 1,142 0.1 206 0.1 13.3 18.0 13. Prostate 1,551 0.1 673 0.6 878 0.1 87 0.0 5.6 9.9 14. Bladder 1,392 0.1 429 0.4 963 0.1 77 0.0 5.6 8.0 (Table continues on the following poge.) Table 9A.1 (continued) INDIA PERCENTAGE HIGH4NCOME LOW- AND MIDDLE- LOW- AND WORLD COUNTRIES INCOME COUNTRIES TOTAL MiDDLE- NUMBER NUMBER NUMBER NUMBER PERCENTAGE INCOME CONDITION (THOUSANDS) PERCENT (THOUSANDS) PERCENT RTHOUSANDS) PERCENT (THOUSANDS) PERCENT OF WORLD COUNTRIES 15. Lymphoma 3,419 0.2 759 0.7 2,659 0.2 360 0.1 10.5 13.5 16. Leukaemia 4,828 0.3 630 0.6 4,198 0.3 429 0.2 8.9 10.2 17. Other cancers 15,586 1.1 3,164 2.9 12,421 1.0 1,874 0.7 12.0 15.1 B. Other neoplasms 4,032 0.3 880 0.8 3,152 0.2 238 0.1 5.9 7.6 C. Diabetes mellitus 11,668 0.8 3,131 2.9 8,537 0.7 1,981 0.7 17.0 23.2 D. Nutritional/endocrine disorders 5,804 0.4 1,217 1.1 4,588 0.4 96 0.0 1.7 2.1 E. Neuropsychiatric disorders 159,462 11.5 25,414 23.5 134,048 10.5 22,944 8.5 14.4 17.1 1. Unipolor major depression 58,246 4.2 7,029 6.5 51,217 4.0 9,679 3.6 16.6 18.9 2. Bipolar affective disorder 16,189 1.2 1,768 1.6 14,421 1.1 2,746 1.0 17.0 19.0 3. Psychoses 14,265 1.0 2,280 2.1 11,984 0.9 1,964 0.7 13.8 16.4 4. Epilepsy 5,147 0.4 488 0.5 4,659 0.4 936 0.3 18.2 20.1 5. Alcohol dependence 18,292 1.3 4,739 4.4 13,553 1.1 1,074 0.4 5.9 7.9 6. Alzheimer and other dementias 8,510 0.6 2,983 2.8 5,527 0.4 922 0.3 10.8 16.7 7. Parkinson's disease 1,109 0.1 489 0.5 621 0.0 138 0.1 12.4 22.2 8. Multiple sclerosis 1,530 0.1 221 0.2 1,308 0.1 234 0.1 15.3 17.8 9. Drug dependence 6,326 0.5 1,544 1.4 4,782 0.4 89 0.0 1.4 1.9 10. Post traumatic stress disorder 2,174 0.2 278 0.3 1,896 0.1 369 0.1 17.0 19.4 11. Obsessive-com- pulsive disorders 11,566 0.8 1,504 1.4 10,062 0.8 1,947 0.7 16.8 19.4 12. Panic disorders 5,429 0.4 719 0.7 4,710 0.4 882 0.3 16.2 18.7 13. Other neuro- psychiatric disorders 10,678 0.8 1,370 1.3 9,308 0.7 1,964 0.7 18.4 21.1 F. Sense organ disorders 12,542 0.9 158 0.1 12,385 1.0 3,701 1.4 29.5 29.9 1. Glaucoma 3,070 0.2 85 0.1 2,985 0.2 698 0.3 22.8 23.4 2. Cataracts 9,182 0.7 68 0.1 9,114 0.7 3,001 1.1 32.7 32.9 3. Other sense organ disorders 290 0.0 5 0.0 286 0.0 2 0.0 0.7 0.7 G. Cardiovascular diseases 143,015 10.3 19,518 18.0 123,497 9.7 26,932 10.0 18.8 21.8 1. Rheumatic heart disease 6,576 0.5 180 0.2 6,396 0.5 1,793 0.7 27.3 28.0 2. Ischemic heart disease 51,948 3.8 9,501 8.8 42,447 3.3 11,697 4.3 22.5 27.6 3. Cerebrovascular disease 41,626 3.0 5,219 4.8 36,407 2.9 4,814 1.8 11.6 13.2 4. Infammatory cardiac disease 10,509 0.8 722 0.7 9,787 0.8 2,071 0.8 19.7 21.2 5. Other cardiac diseases 32,356 2.3 3,896 3.6 28,460 2.2 6,556 2.4 20.3 23.0 H. Respiratory diseases 61,603 4.5 8,050 7.4 53,553 4.2 5,833 2.2 9.5 10.9 1. Chronic obstruc- tive pulmonary disease 28,654 2.1 2,449 2.3 26,205 2.1 2,536 0.9 8.9 9.7 2. Asthma 10,968 0.8 1,208 1.1 9,760 0.8 1,525 0.6 13.9 15.6 (Table continues on the following page.) Table 9A.1 (continued) INDIA PERCENTAGE HIGH-INCOME LOW- AND MIDDLE- LOW- AND WORLD COUNTRIES INCOME COUNTRIES TOTAL MIDDLE. NUMBER NUMBER NUMaER NUMBER PERCENTAGE INCOME CONDITION (THOUSANDS) PERCENT (THOUSANDS) PERCENT (THOUSANDS) PERCENT (THOUSANDS) PERCENT OF WORLD COUNTRIES 3. Other respiratory diseases 18,392 1.3 1,303 1.2 17,089 1.3 3,352 1.2 18.2 19.6 1. Digestive diseases 41,111 3.0 4,365 4.0 36,746 2.9 5,618 2.1 13.7 15.3 1. Peptic ulcer disease 2,637 0.2 241 0.2 2,395 0.2 853 0.3 32.4 35.6 2. Cirrhosis of the liver 12,813 0.9 1,638 1.5 11,175 0.9 2,628 1.0 20.5 23.5 3. Appendicitis 1,446 0.1 35 0.0 1,411 0.1 313 0.1 21.7 22.2 4. Other digestive diseases 24,216 1.8 2,451 2.3 21,765 1.7 1,823 0.7 7.5 8.4 J. Diseases of the genito-urinary system 15,576 1.1 1,220 1.1 14,356 1.1 2,036 0.8 13.1 14.2 1. Nephritis/ nephrosis 8,429 0.6 470 0.4 7,959 0.6 1,578 0.6 18.7 19.8 2. Benign prostatic hypertrophy 2,150 0.2 239 0.2 1,911 0.1 366 0.1 17.0 19.2 3. Other genito- urinary system diseases 4,997 0.4 511 0.5 4,486 0.4 92 0.0 1.8 2.0 K. Skin diseases 1,619 0.1 136 0.1 1,482 0.1 114 0.0 7.0 7.7 L. Musculoskeletal diseases 21,464 1.6 4,512 4.2 16,952 1.3 1,710 0.6 8.0 10.1 1. Rheumatoid arthritis 3,682 0.3 991 0.9 2 692 0.2 197 0.1 5.4 7.3 2. Osteoarthritis 15,513 1.1 3,046 2.8 12,468 1.0 1,482 0.6 9.6 11.9 3. Other musculo- skeletal diseases 2,269 0.2 476 0.4 1,793 0.1 31 0.0 1.4 1.7 M. Congenital abnormalities 28,147 2.0 1,915 1.8 26,232 2.1 7,454 2.8 26.5 28.4 N. Oral diseases 8,483 0.6 959 0.9 7,524 0.6 1,247 0.5 14.7 16.6 1. Dental caries 4,720 0.3 432 0.4 4,288 0.3 783 0.3 16.6 18.2 2. Periodontal disease 295 0.0 36 0.0 258 0.0 86 0.0 29.2 33.3 3. Edentulism 3,351 0.2 485 0.4 2,866 0.2 356 0.1 10.6 12.4 4. Other oral diseases 118 0.0 7 0.0 111 0.0 22 0.0 18.8 19.9 111. Injuries 221,673 16.0 12,739 11.8 208,934 16.4 45,032 16.7 20.3 21.6 A. Unintentional 156,184 11.3 8,972 8.3 147,213 11.6 39,716 14.8 25.4 27.0 1. Road traffic accidents 38,849 2.8 4,556 4.2 34,293 2.7 7,204 2.7 18.5 21.0 2. Poisoning 6,364 0.5 280 0.3 6,085 0.5 988 0.4 15.5 16.2 3. Falls 27,021 2.0 1,397 1.3 25,624 2.0 10,898 4.1 40.3 42.5 4. Fires 11,967 0.9 261 0.2 11,706 0.9 5,723 2.1 47.8 48.9 5. Drowning 14,896 1.1 280 0.3 14,616 1.1 2,703 1.0 18.1 18.5 6. Other uninten- tional injuries 57,088 4.1 2,198 2.0 54,890 4.3 12,201 4.5 21.4 22.2 B. Intentional 65,489 4.7 3,768 3.5 61,721 4.8 5,316 2.0 8.1 8.6 1. Self-inflicted 21,511 1.6 2,416 2.2 19,095 1.5 3,337 1.2 15.5 17.5 2. Homicide and violence 21,573 1.6 1,210 1.1 20,363 1.6 1,847 0.7 8.6 9.1 3. War 22,405 1.6 142 0.1 22,264 1.7 132 0.0 0.6 0.6 Note: STD, sexually transmitted disease. Source: WHO (1999). Table 9A.2 Distribution of Deaths in World, India, and Countries Grouped by Income, by Cause, 1998 INDIA P'ERCENTAGE HIGH-INCOME LOW- AND MIDDLE- LOW- AND WORLD COUNTRIES INCOME COUNTRIES TOTAL MIDDLE- NUMBER NUMBER NUMBER NUMBER PERCENTAGE INCOME CAUSE {THOUSANDS) PERCENT (THOUSANDS) PERCENT ITHOUSANDS) PERCENT (THOUSANDS) PERCENT OF WORLD COUNTRIES Memo: Distribution of population 5,884,576 100 907,828 15.4 4,976,748 84.6 982,223 100 16.7 19.7 Total Deaths 53,929 100 8,033 100 45,897 100 9,337 100 17.3 20.3 I. Communicable diseases, maternal and perinatal con- ditions, and nutri- tional deficiencies 16,447 30.5 510 6.3 15,937 34.7 3,944 42.2 24.0 24.7 A. Infectious and parasitic diseases 9,802 18.2 122 1.5 9,680 21.1 2,121 22.7 21.6 21.9 1. Tuberculosis 1,498 2.8 18 0.2 1,480 3.2 421 4.5 28.1 28.4 2. STDs excluding HIV 181 0.3 1 0.0 180 0.4 55 0.6 30.4 30.6 a. Syphilis 159 0.3 1 0.0 159 0.3 47 0.5 29.6 29.7 b. Chlamydia 13 0.0 0 0.0 13 0.0 5 0.1 38.7 38.8 c. Gonorrhea 8 0.0 0 0.0 8 0.0 3 0.0 33.9 33.9 d. Other STDs 1 0.0 0 0.0 0 0.0 0 0.0 0.0 0.0 3. HIV/AIDS 2,285 4.2 32 0.4 2,253 4.9 179 1.9 7.8 8.0 4. Diarrheal diseases 2,219 4.1 7 0.1 2,212 4.8 711 7.6 32.1 32.2 5. Childhood diseases 1,650 3.1 10 0.1 1,640 3.6 429 4.6 26.0 26.2 a. Pertussis 346 0.6 3 0.0 342 0.7 71 0.8 20.7 20.9 b. Poliomyelitis 2 0.0 0 0.0 2 0.0 1 0.0 29.7 29.7 c. Diphtheria 5 0.0 0 0.0 5 0.0 2 0.0 41.1 41.1 d. Measles 888 1.6 5 0.1 882 1.9 190 2.0 21.4 21.5 e. Tetanus 410 0.8 1 0.0 409 0.9 165 1.8 40.3 40.4 6. Meningitis 143 0.3 4 0.1 139 0.3 36 0.4 25.3 26.1 7. Hepatitis 92 0.2 4 0.1 88 0.2 16 0.2 17.0 17.7 8. Malaria 1,110 2.1 0 0.0 1,110 2.4 20 0.2 1.8 1.8 9. Tropical diseases 106 0.2 0 0.0 106 0.2 30 0.3 28.2 28.3 a. Trypanosomiasis 40 0.1 0 0.0 40 0.1 0 0.0 0.0 0.0 b. Chagas disease 17 0.0 0 0.0 17 0.0 0 0.0 0.0 0.0 c. Schistosomiasis 7 0.0 0 0.0 7 0.0 0 0.0 0.0 0.0 d. Leishmaniasis 42 0.1 0 0.0 42 0.1 30 0.3 70.8 70.9 e. Lymphatic filariasis 0 0.0 0 0.0 0 0.0 0 0.0 f. Onchocerciasis 0 0.0 0 0.0 0 0.0 0 0.0 10. Leprosy 2 0.0 0 0.0 2 0.0 1 0.0 35.7 36.2 I11. Dengue 15 0.0 0 0.0 15 0.0 10 0.1 63.2 63.2 12. Japanese encephalitis 3 0.0 0 0.0 3 0.0 1 0.0 23.3 23.3 13. Trachoma 0 0.0 0 0.0 0 0.0 0 0.0 0.0 14. Intestinal nematode infections 17 0.0 0 0.0 17 0.0 3 0.0 17.3 17.4 a. Ascariasis 8 0.0 0 0.0 8 0.0 1 0.0 14.2 14.2 b. Trichuriasis 5 0.0 0 0.0 5 0.0 0 0.0 9.3 9.3 c. Hookworm disease 4 0.0 0 0.0 4 0.0 1 0.0 35.0 35.0 d. Other intestinal infections 0 0.0 0 0.0 0 0.0 0 0.0 0.0 0.0 15. Other infectious diseases 478 0.9 46 0.6 432 0.9 209 2.2 43.7 48.4 B. Respiratory infections 3,507 6.5 . 309 3.9 3,198 7.0 987 10.6 28.1 30.9 (Table continues on the following page.) Table 9A.2 (continued) INDIA PERCENTAGE HIGH-INCOME LOW- AND MIDDLE- LOW- AND WORLD COUNTRIES INCOME COUNTRIES TOTAL MIDDLE- NUMBER NUMBER NUMBER NUMBER PERCENTAGE INCOME CAUSE (THOUSANDS) PERCENT (THOUSANDS) PERCENT (THOUSANDS) PERCENT (THOUSANDS) PERCENT OF WORLD COUNTRIES 1. Acute lower respiratory infections 3,452 6.4 306 3.8 3,146 6.9 969 10.4 28.1 30.8 2. Acute upper respiratory infections 34 0.1 3 0.0 31 0.1 10 0.1 28.3 31.0 3. Otitis media 20 0.0 0 0.0 20 0.0 9 0.1 42.1 42.6 C. Maternal conditions 493 0.9 2 0.0 491 1.1 125 1.3 25.3 25.4 1. Hemorrhage 123 0.2 0 0.0 122 0.3 30 0.3 24.4 24.5 2. Sepsis 74 0.1 0 0.0 74 0.2 20 0.2 26.9 27.0 3. Hypertensive disorders of pregnancy 62 0.1 1 0.0 61 0.1 15 0.2 24.3 24.5 4. Obstructed labor 38 0.1 0 0.0 38 0.1 10 0.1 26.6 26.7 5. Abortion 66 0.1 0 0.0 66 0.1 19 0.2 28.0 28.1 6. Other maternal conditions 131 0.2 0 0.0 130 0.3 31 0.3 24.0 24.0 D. Perinotal conditions 2,155 4.0 53 0.7 2,102 4.6 612 6.6 28.4 29.1 E. Nutritional deficiencies 490 0.9 23 0.3 467 1.0 100 1.1 20.4 21.4 1. Protein-energy malnutrition 281 0.5 7 0.1 274 0.6 53 0.6 19.0 19.5 2. Iodine deficiency 16 0.0 0 0.0 16 0.0 5 0.0 27.8 27.8 3. Vitamin A deficiency 78 0.1 0 0.0 78 0.2 16 0.2 20.2 20.2 4. Anemias 110 0.2 15 0.2 95 0.2 26 0.3 23.6 27.4 5. Other nutritional disorders 4 0.0 1 0.0 3 0.0 0 0.0 0.0 0.0 II. Noncommunicable conditions 31,717 58.8 7,024 87.4 24,693 53.8 4,470 47.9 14.1 18.1 A. Malignant neoplasms 7,228 13.4 2,020 25.1 5,209 11.3 653 7.0 9.0 12.5 1. Mouth and oropharynx 352 0.7 41 0.5 312 0.7 100 1.1 28.4 32.1 2. Esophagus 436 0.8 49 0.6 387 0.8 62 0.7 14.3 16.1 3. Stomach 822 1.5 143 1.8 679 1.5 51 0.5 6.2 7.5 4. Colon/rectum 556 1.0 243 3.0 313 0.7 25 0.3 4.5 8.1 5. Liver 609 1.1 46 0.6 563 1.2 16 0.2 2.6 2.8 6. Pancreas 214 0.4 99 1.2 115 0.3 9 0.1 4.3 8.0 7. Trachea/bronchus/ lung 1,244 2.3 422 5.3 822 1.8 79 0.8 6.3 9.6 8. Melanoma and other skin cancers 55 0.1 25 0.3 30 0.1 1 0.0 2.4 4.4 9. Breast 412 0.8 160 2.0 252 0.5 47 0.5 11.4 18.6 10. Cervix 237 0.4 17 0.2 220 0.5 57 0.6 24.2 26.1 11. Corpus uteri 73 0.1 27 0.3 46 0.1 4 0.0 5.7 9.1 12. Ovary 122 0.2 45 0.6 76 0.2 14 0.2 11.7 18.6 13. Prostate 239 0.4 115 1.4 124 0.3 14 0.1 5.9 11.3 14. Bladder 158 0.3 56 0.7 101 0.2 8 0.1 5.3 8.3 15. Lymphoma 248 0.5 91 1.1 157 0.3 22 0.2 8.9 14.0 16. Leukaemia 253 0.5 65 0.8 188 0.4 18 0.2 7.3 9.8 17. Othercancers 1,199 2.2 373 4.6 825 1.8 124 1.3 10.4 15.1 B. Other neoplasms 109 0.2 39 0.5 69 0.2 5 0.1 4.6 7.2 C. Diabetes mellitus 600 1.1 161 2.0 439 1.0 102 1.1 17.0 23.2 (Table continues on the following page.} Table 9A.2 (continued) INDIA PERCENTAGE HIGH4NCOMEE LOW- AND MIDDLE- LOW- AND WORLD COUNTRIES INCOME COUNTRIES TOTAL MIDDLE- NUMBER NUMBER NUMBER NUMBER PERCENTAGE INCOME CAUSE (THOUSANDS) PERCENT RHOUSANDS) PERCENT (THOUSANDS) PERCENT (THOUSANDS) PERCENT OF WORLD COUNTRIES D. Nutritional/endocrine disorders 147 0.3 50 0.6 96 0.2 2 0.0 1.4 2.1 E. Neuropsychiatric disorders 720 1.3 225 2.8 495 1.1 104 1.1 14.4 21.0 1. Unipolar major depression 0 0.0 0 0.0 0 0.0 0 0.0 .. .. 2. Bipolar affective disorder 16 0.0 1 0.0 15 0.0 2 0.0 15.7 16.3 3. Psychoses 54 0.1 14 0.2 40 0.1 5 0.1 8.7 11.8 4. Epilepsy 68 0.1 8 0.1 60 0.1 13 0.1 18.8 21.2 5. Alcohol dependence 59 0.1 16 0.2 42 0.1 5 0.1 9.0 12.5 6. Alzheimer and other dementias 216 0.4 105 1.3 111 0.2 22 0.2 10.1 19.5 7. Parkinson's disease 63 0.1 33 0.4 30 0.1 6 0.1 9.8 20.9 8. Multiple sclerosis 26 0.0 6 0.1 20 0.0 3 0.0 12.5 16.1 9. Drug dependence 11 0.0 3 0.0 7 0.0 0 0.0 2.3 3.4 10. Post traumatic stress disorder 0 0.0 0 0.0 0 0.0 0 0.0 .. .. 11. Obsessive-compulsive disorders 0 0.0 0 0.0 0 0.0 0 0.0 12. Panic disorders 0 0.0 0 0.0 0 0.0 0 0.0 .. .. 13. Other neuropsychi- atric disorders 208 0.4 38 0.5 170 0.4 47 0.5 22.7 27.8 F. Sense organ disorders 20 0.0 0 0.0 20 0.0 0 0.0 0.2 0.2 1. Glaucoma 6 0.0 0 0.0 6 0.0 0 0.0 0.0 0.0 2. Cataracts 6 0.0 0 0.0 6 0.0 0 0.0 0.0 0.0 3. Other sense organ disorders 7 0.0 0 0.0 7 0.0 0 0.0 0.5 0.5 G. Cardiovascular diseases 16,690 30.9 3,592 44.7 13,098 28.5 2 820 30.2 16.9 21.5 1. Rheumatic heart disease 383 0.7 22 0.3 361 0.8 86 0.9 22.4 23.7 2. Ischemic heart disease 7,375 13.7 1,884 23.5 5,492 12.0 1,471 15.8 19.9 26.8 3. Cerebrovascular disease 5,106 9.5 893 11.1 4,213 9.2 557 6.0 10.9 13.2 4. Inflammatory cardiac disease 548 1.0 74 0.9 474 1.0 100 1.1 18.2 21.1 5. Other cardiac diseases 3,277 6.1 719 9.0 2,558 5.6 606 6.5 18.5 23.7 H. Respiratory diseases 2,995 5.6 391 4.9 2,604 5.7 284 3.0 9.5 10.9 1. Chronic obstruc- tive pulmonary disease 2,249 4.2 280 3.5 1,969 4.3 153 1.6 6.8 7.7 2. Asthma 144 0.3 24 0.3 120 0.3 21 0.2 14.8 17.7 3. Other respiratory diseases 602 1.1 87 1.1 515 1.1 110 1.2 18.2 21.3 I. Digestive diseases 1,783 3.3 322 4.0 1,461 3.2 240 2.6 13.4 16.4 1. Peptic ulcer disease 174 0.3 33 0.4 141 0.3 41 0.4 23.4 28.9 2. Cirrhosis of the liver 775 1.4 122 1.5 653 1.4 144 1.5 18.6 22.1 3. Appendicitis 48 0.1 2 0.0 47 0.1 11 0.1 22.3 23.0 4. Other digestive diseases 786 1.5 165 2.1 621 1.4 44 0.5 5.6 7.1 (Table cutlifiou- -ll i1lt7 UU-lvtvii pvigp.) Table 9A.2 (continued) INDIA PERCENTAGE HIGH-INCOME LOW- AND MIDDLE- LOW- AND WORLD COUNTRIES INCOME COUNTRIES TOTAL MIDDLE- NUMBER NUMBER NUM3ER NUMBER PERCENTAGE INCOME CAUSE {THOUSANDS) PERCENT (THOUSANDS) PERCENT {THOUSANDS) PERCENT (THOUSANDS) PERCENT OF WORLD COUNTRIES J. Diseases of the genito- urinary system 765 1.4 139 1.7 626 1.4 102 1.1 13.4 16.3 1. Nephritis/nephrosis 554 1.0 90 1.1 464 1.0 89 1.0 16.1 19.3 2. Benign prostatic hypertrophy 33 0.1 4 0.1 29 0.1 11 0.1 31.9 36.7 3. Other genito- urinary system diseases 178 0.3 45 0.6 133 0.3 2 0.0 1.3 1.7 K. Skin diseases 44 0.1 13 0.2 30 0.1 2 0.0 5.4 7.7 L. Musculoskeletal diseases 100 0.2 35 0.4 65 0.1 3 0.0 2.5 3.8 1. Rheumatoid arthritis 17 0.0 10 0.1 7 0.0 2 0.0 10.6 25.9 2. Osteoarthritis 0 0.0 0 0.0 0 0.0 0 0.0 0.0 0.0 3. Other musculo- skeletal diseases 83 0.2 25 0.3 58 0.1 1 0.0 0.9 1.2 M. Congenital abnormalities 515 1.0 36 0.5 478 1.0 153 1.6 29.8 32.1 N. Oral diseases 2 0.0 0 0.0 2 0.0 0 0.0 18.7 23.0 1. Dental caries 0 0.0 0 0.0 0 0.0 0 0.0 2. Periodontal disease 0 0.0 0 0.0 0 0.0 0 0.0 0.0 0.0 3. Edentulism 0 0.0 0 0.0 0 0.0 0 0.0 4. Other oral diseases 2 0.0 0 0.0 2 0.0 0 0.0 18.8 23.0 I. In juries 5,765 10.7 498 6.2 5,266 11.5 923 9.9 16.0 17.5 A. Unintentional 3,493 6.5 327 4.1 3,166 6.9 723 7.7 20.7 22.8 1. Road traffic accidents 1,171 2.2 142 1.8 1,029 2.2 217 2.3 18.5 21.1 2. Poisoning 252 0.5 14 0.2 238 0.5 32 0.3 12.5 13.3 3. Falls 316 0.6 77 1.0 239 0.5 50 0.5 15.9 21.0 4. Fires 282 0.5 11 0.1 271 0.6 135 1.4 47.7 49.6 5. Drowning 495 0.9 13 0.2 482 1.1 92 1.0 18.5 19.0 6. Other unintentional injuries 977 1.8 70 0.9 907 2.0 199 2.1 20.3 21.9 B. Intentional 2,272 4.2 172 2.1 2,100 4.6 200 2.1 8.8 9.5 1. Self-inflicted 948 1.8 130 1.6 818 1.8 124 1.3 13.1 15.2 2. Homicide and violence 736 1.4 38 0.5 698 1.5 72 0.8 9.8 10.3 3. War 588 1.1 4 0.0 584 1.3 4 0.0 0.6 0.6 Negligible. Note: STD, sexually transmitted disease. Source: WHO (1999). Table 9A.3 Comparison of India and Other Countries on Selected Health-Related Indicators, Selected Years, 1992-99 TOTAL MALNUTIRMION . LIFE FERTILITY OF GNP PER POPULATION ILLITERATE EXPECTANCY, INFANT RATE, 1998 CHILDREN CAPITA, BELOW US$1 ADULT 1999 MORTAULY, (LIFETIME BIRTHS UNDER POPULATION, 1999 PER DAY, FEMALES, ( {DISABILITY- 1998 PER WOMAN AGE 5, 1999 (U.S. 1992-98 1998 ADJUSTED (PER 1,000 AGES 1992-98 COUNTRY (MILLONSI DOUARS( (PERCEN7l (PERCENT UFE YEARS( POPULATION) 15 TO 49) (PERCENTIO India 998 450 44.2 57 53.2 72 3.3 53a Low-income countries 2,417 410 36.0 49 68 3.1 36 Nigeria 124 310 70.2 48 38.3 76 5.3 39 Bangladesh 128 370 29.1 71 49.9 73 3.1 56 Pakistan 135 470 31.0 71 55.9 91 4.9 38 Indonesia 207 580 15.2 20 59.7 43 2.7 34 Middle-income countries 2,667 2,000 8.2 20 31 2.5 12 China 1,250 780 18.5 25 62.3 31 1.9 16 Sri Lanka 19 820 6.6 12 62.8 16 2.1 38 South Africa 42 3,160 11.5 16 39.8 51 2.8 9 Brazil 168 4,420 5.1 16 59.1 33 2.3 6 High-income countries 891 27,730 0.1 .. 6 1.7 United Kingdom 59 22,640 .. .. 71.7 6 1.7 United States 273 30,600 .. .. 70.0 7 2.0 1 World 5,975 4,890 24.0 32 54 2.7 30 .. Negligible. Note: Blank spaces denote that calculations have not been done on regional disability-adjusted life expectancy. a. Weight more than 2 standard deviations below average for age; in 1999, proportion of children under age 3 who fit this definition was 47 percent. Source: World Bank (2000d); WHO (2000c). Table 9A.4 Causes of DALYs Lost in India, 1990 and 1998 DEATHS PER 1,000 POPULATION DALYs tOST PER PROPORTION OF PROPORTION OF DEATHS (PERCENT) 1,000 POPUlATION DALYs LOST (PERCENT) CHANGE CAUSE 1990 1998 (PERCENT) 1990 1998 CHANGE 1990 1998 1990 1998 Communicable diseases, maternal and perinotal conditions, and nutritional deficiencies 5.6 4.0 -28.6 51.0 42.2 -17.3 191.0 137.7 56.4 50.3 Noncommunicable conditions 4.5 4.6 2.2 40.4 47.9 18.6 98.2 90.3 29.0 33.0 Injuries 1.0 0.9 -10.0 8.6 9.9 15.1 49.3 45.8 14.6 16.7 Total 11.0 9.5 -13.6 100.0 100.0 0 338.5 273.8 100.0 100.0 Note: DALY, disability-adjusted life year. Source: Murray and Lopez (1996); WHO (2000c). 318 * Better Health Systems for India's Poor Table 9A.5 Infant Mortality and Total Fertility Rates in India and Major Indian States, 1981-97 INFANT REDUCTION REDUCTION MORTALITY RATE, IN RURAL TOTAL IN RURAL POPULATION, 1998 [MR, FERTILITY TPR, 1999 (PER 1,000 1981-97 RATE, 1981-97 STATE (MILLIONS) LIVE BIRTHS) (PERCENT) 1997 (PERCENT) India 981.3 72 34 3.3 24.5 Group A: TFR of 2.1 or less Kerala 32.0 16 64 1.8 37.9 Tamil Nadu 61.3 53 40 2.0 40.5 Group B: TFR of more than 2.1 and less than 3.0 Karnataka 51.4 58 14 2.5 35.7 Andhra Pradesh 74.6 66 17 2.5 28.2 West Bengal 78.0 53 38 2.6 28.6 Maharashtra 90.1 49 30 2.7 37.5 Punjab 23.3 54 35 2.7 22.0 Group C: TFR of at least 3.0 Orissa 35.5 98 26 3.0 28.3 Gujarat 47.6 64 44 3.0 25.0 Assam 25.6 78 47 3.2 19.0 Haryana 19.5 69 31 3.4 26.9 Madhya Pradesh 78.3 98 29 4.0 20.0 Rajasthan 52.6 83 25 4.2 23.7 Bihar 98.1 67 36 4.4 20.7 Uttar Pradesh 166.4 85 44 4.8 16.4 Note: IMR, infant mortality rate; TFR, total fertility rate. Within groups, states are listed in order of TFR. IMR refers to infants less than 1 year of age; TFR is lifetime births per woman ages 15-49. Major Indian states are those with a population of more than 15 million. Bihar includes Jharkhand, Madhya Pradesh includes Chatisgarh, and Uttar Pradesh includes Uttaranchol. Source: Registrar General 12000). Health System Outcomes * 319 Table 9A.6 Underweight Children under Three Years of Age in Major States of India, 1998-99 (percent) STATE PERCENT Madhya Pradesh 55 Bihar 54 Orissa 54 Uttar Pradesh 52 Rajasthan 51 Maharashtra 50 West Bengal 49 Gujarat 45 Karnataka 44 Andhra Pradesh 38 Tamil Nadu 37 Assam 36 Haryana 35 Punjab 29 Kerala 27 Note: Underweight children are those whose weights are statistically low for their ages (that is, more than 2 standard deviations below average). Major Indian states are those with populations of more than 15 million. Bihar includes Jharkhand, Madhya Pradesh includes Chatisgarh, and Uttar Pradesh includes Uttaranchal. Source: IIPS (2000). 320 * Better Health Systems for India's Poor Table 9A.7 Reduction in Rates of Severe and Total Malnutrition (Weight for Age) among Children in India and Major States in India between 1992-93 and 1998-99 (percent) AREA SEVERE TOTAL India 6.40 2.60 Andhra Pradesh 11.40 5.30 Assam 14.40 5.40 Bihar 8.20 5.60 Gujarat 5.00 1.40 Haryana 3.30 -1.10 Karnataka 10.40 2.90 Kerala 1.60 1.40 Madhya Pradesh 2.30 -2.00 Maharashtra 4.60 3.70 Orissa -1.10 2.00 Punjab 17.20 5.40 Rajasthan -9.00 -1.60 Tamil Nadu 11.50 2.70 Uttar Pradesh 7.30 2.70 West Bengal 8.10 2.10 Note: Negative values denote an increase. Malnutrition in 1992-93 was measured among children under age 4; in 1998-99, under age 3. Severe molnutrition is weight more than 2 standard deviations below average for age; total malnutrition is weight more than 3 standard deviations below average. Major Indian states are those with popula- tions of more than 15 million. Bihar includes Jharkhand, Madhya Pradesh includes Chatisgarh, and Uttar Pradesh includes Uttaranchal. Source: IIPS (2000). Health System Outcomes * 321 Table 9A.8 Comparison of Female and Male Health Outcomes in India, 1998-99 INDICATORS FOR CHILDREN UNDER AGE 3b TUBERCULOSIS UNDER MORTAUTY PER ACUTE MEDICAL 1,000 LIVE 8IRTHS, RESPIRATORY TREATMENT INFECTION DIARRHEA (PER POST- UNDER- IN PAST IN PAST 100,000 SEX AND NEO- NEO. UNDER WEIGHT 2 WEEKS 2 WEEKS POPULA- RATIO NATAL NATAL INFANT 5 (PERCENT) (PERCENT) (PERCENT) TION) Female 44.6 26.6 71.1 105.2 48.9 17.9 18.9 357 Male 50.7 24.2 74.8 97.9 45.3 20.7 19.4 502 Female-male ratio 0.88 1.10 0.95 1.07 1.08 0.86 0.97 0.71 a. Mortality rates are for the 10-year period preceding the survey. Neonatal denotes birth to less than one month of age; post-neonatal, one month to less than one year; infant, birth to one year; under five, birth to under five. b. Underweight is weight that is more than 2 standard deviations below average for age. Source: IIPS (2000). Notes 1. The Census of India 2001 provisional population estimates report sex ratios of between 927 and 934 in each of the four censuses between 1971 and 2001 (table 10 of the results posted at wwwcen- susindia.net/ on April 26, 2001). 2. The states are Bihar, Jharkhand, Uttar Pradesh, West Bengal, Orissa, Madhya Pradesh, and Chatisgarh. 3. Private health spending as a proportion of total expenditure is distributed more progressively. For example, the richest quintile spends 6.6 percent of its total expenditure on health, compared wit. 3.8 percent for the poorest quintile. 4. Estimates of the true effect of hospitalization on poverty level; may require more intensive studies involving prospective analysis of hospitalization, income, and consumption. 322 * Better Health Systems for India's Poor 5. We assumed conservatively that all the money borrowed for hospitalization would not have been spent if people were not ill. Including borrowing raises the level to 35 percent. 6. An additional 3.3 percent of hospitalized Indians fall below the poverty line if indirect medical expenses, such as transport costs, are also included. 7. The definitions and methodologies for this area of inquiry are the least developed of the three health system outcomes and are least amenable to international or interstate comparisons. We use a dif- ferent definition of responsiveness than does the World Health Organization's World Health Report 2000, which refers to the non- medical aspects of care and excludes satisfaction; in contrast, we look specifically at satisfaction and the legal protection of the public from negligence and redress. 8. Equal access to health care is listed under the Directive Princi- ples of the Constitution, but not explicitly as a fundamental right. The courts, however, have held that the fundamental right to life includes health care (Background Paper 9). 9. More detailed analysis and specific recommendations for improving patient redress are provided in Background Paper 10. 10. The alternative hypothesis-that people are receiving high quality medical care and therefore do not complain-is highly unlikely, according to reviewers of quality of the private sector (Background Papers 6-8). 11. The study used an adapted and translated version of a 51-item patient satisfaction questionnaire. 12. This study used an adapted and translated version of an 11- item visit satisfaction questionnaire. - :APPENDIX A Studies Conducted for the Present Report 323 STUDY RESEARCH ORGANtZATiON POUcY QUESIIONS ADDRESSED USE OF slUTy RESUlTS 1. Privote Health Sector Market a. Indian Institute of How can India take * Better understonding of constraints, Analysis Technology; Center for advantage of the private incentives, and subsidies and of how a. Systematic Review of Enquiry into Health and sector to meet social the private sector functions are to Knowledge Allied Themes ICEHAT); goals? influence policy proposals and pro- b. Field Analysis in Andhro Jawoharlal Nehru University ject design Pradesh (AP) and Uttar (funded by WHO) * New partnerships in service delivery Pradesh (UP) b. Institute of Hecith Systems and financing of health services (AP); Indian Institute of * New approaches to regulotion and Manogement-Lucknow (UP) quality assurance * New public accountabilities * Benchmarks for standards 2. Consumer Protection in Health: a. Indian Law Institute How can consumers become * Better understanding of role of Legal Framework and Current b. Voluntory Organization more empowered over consumer laws Practices in Interest of Consumer health issues? * Better public advocacy a. Legal framework for health Education b. Mechanisms of consumer redress 3. Health Financing Oplions: a. Institute of Economic How can health insurance * Insurance regulations to minimize Health Insuronce Growth be used to improve equity harm to poor a. Critical anolysis of b. National Council for ond efficiency of health * Increased copacity of technical stoff Indian experience Applied Economic services and how can to deal with insurance issues b. Prospect of insurance Research India minimize the * New experiments with health and regulation in India c. Fereirro (Chile) negative effects of heahh insuronce for the poor c. Effects of private heolth insurance liberalization? insurance on the poor and the health system- international experience 4. Pharmaceuticals Analysis a. Benaras Hindu How can safe, effective, * Revise design af new projects and a. Pharmaceuticals sector University (UP); Kilpak affordable drugs be improvement of supervision of analysis in three states Medical College accessible to Indians, pharmaceuticals lending b. Pharmaceuticals industry (Tamil Nadu); JSS whether through the * Revised drug policies; new efforts on analysis College of Pharmacy private or public sector? quality assurance and regulation of (Karnataka); IHBAS private sector (Delhi); * Preparation for TRIPS b. Adminstrative Staff College of India (AP) 5. Quality of Health Services National Quality Assurance How can India systema- * New institutions and networks for Council formed from eminent tically improve quality quality assurance physicians. Subpanel work assurance in health at * New modalities and systems for convened by AIIMS, CEHAT, national and state levels? quality assurance HAP, MGRMU. * Hospital accreditation scheme 6. Distribution of Health National Council on Applied How well do public and Better problem identification, monitoring, Benefits and Costs Economic Research (with NIPFP private health services and advocacy to ensure the focus of and IEG) reach the poor? HNP efforts is on the poor 7. Critical Issues in Decentralizing National Institute of Rural How can India best reor- * Identification of how to strengthen Health Responsibilities Development- Hyderabad ganize its programs capacity of states and PRIs in health (DFID funding) around districts and sector local bodies? * Options for streamlining and integrating health programs 8. Options for Reorganizing Institute for Health Systems How can hospitals and New experiments in hospital Public Hospitals Development (U.K.) plus health facilities be reor- organization (DFID funding) local counterparts ganized to become more efficient, equitable? Note: WHO, World Health Organization; JSS, Jagadguru Sri Shivarathreeshwara; IHBAS, Institute of Human Behavior and Allied Sciences; TRIPS, trade-related aspects of intellectual property; AIIMS, All Indian Institute of Medical Sciences; HAP, Health Action for People; MGRMU, The Tamil Nadu Dr. M. G. R. Medical University; NIPFP, National Institute of Public Finance and Policy; IEG, Institute of Economic Growth; HNP, Health. Nutrition and Population (The World Bank); DFID, Department for International Development; PRI, Panchayati Raj Institutions. APPENDIX B Background Papers These papers, prepared in support of the present report, are avail- able on the web site of the World Bank, at http://wblnOO1 8.worldbank.org/SAR/India/HealthESW/AR/cover. nsf/HomePage/1 ?OpenDocument 1. Garg, C. "Implications of Current Experiences in Health Insurance in India." 2. Mahal, A. "Private Entry into Health Insurance in India: An Assessment." 3. Ferreiro, A. "Private Health Insurance in India: Would Its Implementation Affect the Poor?" 4. Nandraj, S. "Accreditation System for Hospitals in India." 5. Mahal, A.,J. Singh, F Afridi, V Lamba, A. Gumber, andV Sel- varaju. "Who 'Benefits' from Public Sector Health Spending in India? Results of a Benefit Incidence Analysis for India." 6. Nandraj, S. "Contracting and Regulation in the Health Sec- tor: Concerns, Challenges, and Options." 327 328 * Better Health Systems for India's Poor 7. Muraleedharan, VR. "Private-Public Partnership in Health Care Sector in India: A Review of Policy Options and Challenges." 8. Baru, R.V., I. Qadeer, and R. Priya. "Critical Review of Stud- ies on the Private Sector in Health." 9. Indian Law Institute. "Legal Framework for Health Care in India: Experience and Future Directions." 10. Misra, B., and P. Kalra. "The Regulatory Framework for Con- sumer Redress in the Healthcare System in India." 11. Pearson, M. "International Experience of Hospital Auton- omy." 12. Pearson, M. "Overview Paper: Hospital Autonomy in India." 13. Administrative Staff College of India. "The Indian Pharma- ceuticals Industry." 14. Govindaraj, R., and G. Chellaraj. "Pharmaceuticals Sector in India: Issues and Options." 15. Kilpauk Medical College, Department of Community Medi- cine. "Pharmaceutical Study on Drug Policy: Tamil Nadu." 16. Benaras Hindu University. "Drug Policy Assessment Study: Uttar Pradesh." 17. JSS College of Pharmacy. "Drug Policy Assessment Study: Kamataka." 18. Mahal, A., A. Yazbeck, D.H. Peters, and G.N.V. Ramana, "The Poor and Health Service Use in India." Background Papers * 329 19. Peters, D.H., A. Yazbeck, G.N.V Ramana, and R. Sharma. "Public-Private Partnerships in Health, Background Paper: Issues and Options." 20. Institute of Health Systems. "Private Health Sector Market Analysis in Andhra Pradesh." 21. Indian Institute of Management-Lucknow. "Private Health Sector Market Analysis in Uttar Pradesh." APPENDIX C Major Recommendations of National Health Policy Reports since Independence Bhore Committee, 1946 * No individual should lack access to medical care because of inabil- ity to pay for it. * Special emphasis should be placed on preventive methods and on communicable diseases. * Health services should be as "close to the people as possible in order to ensure the maximurn benefit to the community to be served." * All facilities for diagnosis and treatment should be available in the public health services when it is fully developed. * One primary health unit per 10,000-20,000 population with 75 beds, 6 doctors, and 6 public health nurses * One bed per 175 population; one doctor per 1,600 population and one nurse per 600 population * One 650-bed hospital at taluka level (300,000 population) anc one district hospital of 2,500 beds * No patents in pharmaceutical products * 15 percent of govermnent expenditure to be devoted to health care. 331 332 * Better Health Systems for India's Poor Mudaliar Commiltee, 1961 * Strengthen primary health centers (PHCs) * One PHC per 40,000 population that lacks hospital services * One bed per 1,000 population and one doctor per 3,000 population * One 50-bed basic specialty hospital for each taluka and one 500- bed district hospital * Central government to control communicable diseases * One medical college per 5 million population * Only process patents for 5-10 years for drugs * No integration of systems of medicine Jain Commiltee, 1966 * One bed per 1,000 population * One 50-bed hospital at taluka level * Enhance maternity facilities at each level * Health insurance for a larger population coverage * Charge for health access to augment resources Kartar Singh Committee, 1974 * Integration of all health programs and health workers: retrain health workers as multipurpose workers * A team of one male and one female worker at subcenter level (3,000 population) * One PHC per 50,000 population * One health supervisor for every four health workers Srivashova Committee, 1975 * One male and one female health worker per 5,000 population * One health assistant per two health workers * One additional doctor and nurse at PHC for maternal and child health services Major Recommendations of National Health Policy Reports * 333 * Increase PHC drug budgets * Compulsory national service of two years at PHC by every doc- tor between fifth and fifteenth year of career * Establish medical and health education commission * Integration of various health systems Indian Council for Medical Research-Indian Council for Social Science Research Joint Panel, 1980 * A village health unit per 1,000 population with one male and one female health worker * One subcenter per 5,000 population with one male and one female health worker * One 30-bed community health center per 100,000 population with 6 general doctors and 3 specialists * A district health center for every 1 million population and a spe- cialist center for every 5 million population * No further expansion of medical education and drug production but only their rationalization and reorientation * 6 percent of GNP must be ultimately spent on health care serv- ices National Health Policy, 1983 * Provision of universal, comprehensive primary health care serv- ices * Involvement of private practitioners and NGOs to expand cover- age of and access to services * Train village-based workers in simple skills * Evolve a decentralized system of health care and establish a refer- ral systems * Establish a nationwide chain of epidemiological stations * Encourage private investment in health sector to reduce govern- ment burden * Selected health and demographic targets to be achieved by 2000 334 * Better Health Systems for India's Poor National Population Policy, 2000 * Seek a mix of sociodemographic and health goals for 2010 with the primary aim of bringing the total fertility rate to replacement level * Increase outreach and coverage of comprehensive package of reproductive and child health services by government in partner- ship with NGOs and the private sector * Create one-stop, integrated service delivery at the village level * Expand public health infrastructure by increasing numbers of subcenters, primary health centers, and community health centers * Decentralize planning and program implementation with high involvement of the Panchayati Raj Institutions (PRIs) and com- munity groups * Promote intersectoral approach among key government depart- ments * Establish a national commission on population with equivalent structures at the state level * Set up a National Technical Committee with medical experts and government representatives * Double the annual budget of the Family Welfare Department * Create incentives to promote the small-family norm -- APPENDIX D Efforts to Address the Role of Private Providers in National Tuberculosis Control Programs LOCATION PRACTICES Congo, Dem. National Tuberculosis Control Programs (NTP) provide team training Rep. of to a doctor, a laboratory technician, and a nurse from Kinshasa city hospitals and polyclinics. Drugs are subsidized. Patients are man- aged according to guidelines. Egypt, Arab Prominent private chest physicians on NTP board. Pilot projects are Rep. of started with five university hospitals adopting directly observed treat- ment short-course (DOTS). Continuing tuberculosis (TB) education for in-practice chest physicians initiated; modifications to TB education in medical curricula planned. In another pilot, private laboratories report results of all sputum tests to the NTP. India A few running and evolving models: * A private nonprofit hospital runs a DOTS project for patients referred by private general practitioners; DOT done in neighbor- hood centers located in private nursing homes, clinics, and private and nongovernmental organization dispensaries. * A voluntory organization acts as an interface between private providers and NTP to facilitate referrals; and DOT by private providers. * NTP treatment supervisors assign diagnosed patients to their preferred private practitioner agreeing to do DOTS, maintain records, and report default. * Local association of doctors tries out graded involvement of pri- vate providers ranging from referral to running a DOTS program. 335 336 * Better Health Systems for India's Poor Kenya Anti-TB Association provides subsidized drugs to private hospitals and chest physicians in Nairobi who in turn follow NTP guidelines, notify cases, assist in defaulter retrieval, and maintain and submit records. Morocco Two successive yearly surveys show good TB management practices of private practitioners. Forty percent of patients referred to NTP are from private sector. Probable reasons for good management prac- tices of private doctors: undergraduate medical curricula provide substantial time for training in TB, and all postgraduates have to work within NTP before getting license to practice. New York, Upgrading and improving the clinical services offered by chest New York, clinics located throughout the city. State-of-the-art and confidential United States services including DOTS provided free of cost to suspects and patients, including treatment for latent infection to high-risk individu- als, social services, and HIV counseling and testing. Result: a four- fold increase in referrals from private sector. Obligatory for labora- tories to report results of sputum smears and those of drug susceptibility testing. Korea, Rep. of NTP surveys private providers' TB management practices and treat- ment outcomes and shares results with the providers. Improved per- formance demonstrated in a subsequent survey. Syrian Arab Rep. Dissatisfied by private physicians' poor response to persistent and varied approaches to involve them, the NTP manager persuaded the Minister of Health to ban sale of anti-TB drugs in private pharmacies. Effectiveness yet to be evaluated. Netherlands Involvement of private providers at all levels including representation on TB Control Policy Committee. Clarity and consensus on roles to be played by the public and private sectors in managing each patient. Philippines NTP supports two projects: a university hospital and an expensive private hospital in Manila, which run effective DOTS clinics. Note: NTP, National Tuberculosis Control Programs; DOTS, directly observed treatment short-course; TB, tuberculosis. Source: WHO (2001). Bibliography The set of 21 background papers prepared in support of the present report are listed in appendix B. The word processed describes informally reproduced works that may not be com- monly available through libraries. Administrative Staff College of India. 1995. "Beneficiary Assessment for the Karnataka State Health Systems Development Project." Processed. Antia, N. H. n.d. "Voluntary Organizations and Health Care in India." Foundation for Research in Community Health, Mumbai. Arrow, Kenneth J. 1963. "Uncertainty and the Welfare Economics of Medical Care. American Economic Review 53:941-69. Ashtekar, S., and D. Mankad. 2001. "Who Cares? Rural Health Practitioners in Maharashtra." Economic and Political Weekly, February 3 -10, 448-53. BAIF (Bharatiya Agro Industries Foundation). 1997. "Traditional Medicine in Rural Tribal Areas in India." BAIF Development Research Foundation, Pune, Maharashtra. Bajaj, J. S. 1996. "Report of the Expert Committee on Public Health System." Ministry of Health and Family Welfare, New Delhi. Balambal, R., K. Faggarajamma, and R Rahman. 1997. "Impact of Tuberculosis on Private For-Profit Providers." Tuberculosis Research Centre, Chennai, Tamil Nadu. Processed. Barr, N. 1990. "Economic Theory and the Welfare State: A Survey and Reinterpretation. Welfare State Programme." Discussion Paper No. 54. London School of Economics and Political Science, London. 337 338 * Better Health Systems for India's Poor Baru, R. V 1998. Private Health Care in India: Social Characteristics and Trends. New Delhi: Sage Publications. Basu, S. 2000. "Policy for the Reform of the Drug Control Authority: A Stakeholder Analysis." Processed. Bennett, S., and V. R. Muraleedharan. 2000. "'New Public Management' and Health Care in Third World." Economic and Political Weekly, January 8, 59-68. Berman, P. 1995. "Health Sector Reform: Making Health Development Sustainable." In P. Berman, ed., Health Sector Reform in Developing Countries. Boston: Harvard School of Public Health. . 1996. "Health Care Expenditure in India." In M. Das Gupta, L. Chen, and T N. Krishnan, eds., Health, Poverty and Development in India. New Delhi: Oxford University Press. . 2000. "Organization of Ambulatory Care Provision: A Critical Determinant of Health System Performance in Developing Countries. Bulletin of the World Health Organization 78:791-802. Berman, P., and M. E. Khan, eds. 1993. Paying for India's Health Care. New Delhi: Sage Publications. Bhandari, N. 1992. "The Household Management of Diarrhoea in the Social Context: A Study of a Delhi Slum." Ph.D. diss. Jawaharlal Nehru University, New Delhi. Bhat, P. N. 1995. "Maternal Mortality in India: Estimates from Regression Model." Studies in Family Planning 26:217-32. Bhat, R. 1996a. "Regulating the Private Health Care Sector: The Case of the Indian Consumer Protection Act." Health Policy and Planning 11:265-79. . 1996b. "Regulation of the Private Health Sector in India." International journal of Health Planning and Management 11:253-74. Bhore, J., R. A. Amesur, and A. C. Banerjee. 1946. Report of the Health Survey and Development Committee. Vol. I. Government of India, New Delhi. Blackstone Ltd. 2000. "West Bengal Health Systems Development Project: Quality Assurance Programme." Draft Report. Processed. Cai, W W, J. S. Marks, C. H. Chen, Y. X. Zhuang, L. Morris, and J. R. Harris. 1998. "Increased Cesarean Section Rates and Emerging Patterns of Health Insurance in Shanghai, China." American Journal of Public Health 88:777-80. Bibliography * 339 Caldwell, J. C., and G. Santow. 1989. "Introduction." In J. C. Caldwell and G. Santow, eds., Selected Readings in The Cultural, Social, and Behavioural Determinants of Health. Canberra: The Australian National University. Cassels, A. 1995. "Health Sector Reform: Key Issues in Less Developed Countries. journal of International Health Development 7:329-49. CBHI (Central Bureau of Health Intelligence). Various years. Health Information of India. Annual. Ministry of Health and Family Welfare, Directorate General of Health Services. New Delhi. Census Comnmissioner of India. 1981. Census of India. Ministry of Home Affairs. New Delhi. Centre for Policy Research. 1999. Report on the Restructuring the Ministty of Health & Family Welfare. New Delhi. Chaix-Couturier, C., I. Durand-Zaleski, D. Jolly, and P. Durieux. 2000. "Effects of Financial Incentives on Medical Practice: Results from a Systematic Review of the Literature and Methodological Issues. International Journalfor Quality in Health Care 12:133-42. Chand, S. K. 1988. "The Traditional Herbal Medicine System of Chotanagpur: A Study of Its Present Status and Future Prospects." Xavier Institute of Social Service, Ranchi, Jharkhand. Processed. Chawla, M. 2000. "Private and Pubic Markets for Physician Services in Developing Countries: Evidence of Inter-Linkages." Processed. Chollet, D. J., and M. Lewis. 1997. "Private Insurance: Principles and Practice." In G. Schieber, ed., Innovations in Health Care Financing. Proceedings of a World Bank Conference, March 10-11, 1997. World Bank Discussion Paper 365. Washington, D.C. Claeson, M., E. R. Bos, and I. Pathamanathan. 1999. Reducing Child Mortality in India: Keeping Up the Pace. Washington, D.C: World Bank. De Regt, R., H. L. Minkoff, J. Feldman, and R. H. Schwarz. 1986. "Relation of Private or Clinic Care to the Cesarean Birth Rate." New England Journal of Medicine 3 15:619-24. Dohrenwend, B., and B. Dohrenwend. 1969. Social Status and Psychological Disorder: A Causal Inquiry. New York: John Wiley and Sons. Dreze,J., and A. Sen. 1995. India: Economic Development and Social Opportunity. New Delhi: Oxford University Press. 340 * Better Health Systems for India's Poor Duggal, R. 1997. "Health Care Budgets in a Changing Political Economy." Economic and Political Weekly, May 17-24, 1197-1200. - 2000. The Private Health Sector in India: Nature, Trends, and a Critique. New Delhi: Voluntary Health Association of India. Duggal, R., S. Nandraj, and A. Vadair. 1995. "Health Expenditures across States, Part II: Regional Disparity in Expenditure." Economic and Political Weekly, April 22, 901-08. Dukes, M.N.G. 2000. "The Reform of the Drug Control System in Uttar Pradesh." Processed. Eaton, W W, R. Day, and M. Kramer. 1988. "The Use of Epidemiology for Risk Factor Research in Schizophrenia: An Overview and Methodologic Critique." In M. T Tsuan and J. C. Simpson, eds., Handbook of Schizophrenia. Vol. 3, Nosology, Epidemiology and Genetics. Amsterdam: Elsevier Science Publishers. Ellis, R. P., M. Alam, and I. Gupta. 2000. "Health Insurance in India: Prognosis and Prospectus." Economic and Political Weekly, January 22, 207-17. Francome, C., and W Savage. 1993. "Caesarean Section in Britain and the United States, 12 % or 24%: Is Either the Right Rate?" Social Science and Medicine 37:1199-218. Frank,J. W, and J. F Mustard. 1994. "The Determinants of Health in a Historical Perspective. Daedalus 123(4):1-17. Frenk, J. 1993. "The Public-Private Mix and Human Resources for Health. " Health Policy and Planning 8:315-26. . 1994. "Dimensions of Health Sector Reform." Health Policy 27:19-34. Garg, C. 2001. "Punjab State Health Accounts." Processed. Ghosh, A. 1998. "EEC for Promoting Behaviour Change in the Population, Health, and Nutrition Sector, A Review." Paper presented at the workshop on the World Bank's Role in the Health System of India, New Delhi, April 2-3, 1998. Processed. Girishankar, N. 1999. "Refonning Institutions for Service Delivery: A Framework for Development Assistance with an Application to the HNP Portfolio." Policy Research Working Paper 2039. World Bank, Washington, D.C. Government of India. 1998-99. "National Health Policy." New Delhi, India. Greenhalgh, T 1986. Drug Marketing in the Third World: Beneath the Cosmetic Reforms. Lancet 1(8493):1318-20. Bibliography * 341 Gribble, J. N., and S. H. Preston, eds. 1993. The Epidemiological Transition. Washington, D.C.: National Academy Press. Gupta, R., V P. Gupta, and N. S. Ahluwalia. 1994. "Educational Status, Coronary Heart Disease, and Coronary Risk Factor Prevalence in a Rural Population of India." British Medical journal 3099:1332-36. Gwatkin, D., S. Rutstein, K. Johnson, R. P. Pande, and A. Wagstaff. 2000. "Socio- economic Differences in Health, Nutrition, and Population in India." HNP Publication Series, HNP/Poverty Thematic Group. World Bank, Washington, D.C. Homan, R. K., and K. R. Thankappan. 1999. "An Examination of Public and Private Sector Sources of Inpatient Care in Trivandrum District, Kerala (India) 1999." Kerala: Achuta Menon Center for Health Services. Processed. IIPS (Indian Institute for Population Sciences). 1995. National Family Health Survey (MCH and Family Planning), India, 1992-93. Bombay. 1998-99. Reproductive and Child Health Surveys. Bombay. 2000. National Family Health Survey Summary, India 1998-99. Bombay. Institute of Health Systems. 2000. "APVVP Patient Satisfaction Survey 2000." Report Series. Hyderabad. Kakwani, N., A. Wagstaff, and E. van Doorslaer. 1997. "Socioeconomic Inequalities in Health: Meaurement, Computation, and Statistical Inference." Journal of Econometrics 77:87-103. Kannan, K. P., K. R. Thankappan, V R. Kutty, and K. P. Aravindan. 1991. "Health and Development in Rural Kerala." Kerala Sastra Sahitya Parishad, Thiruvananthapuram. Processed. Kathuria, S., and J. Hanson, eds. 2000. India: Reducing Poverty, Accelerating Development. New Delhi: Oxford University Press. Krishnan, T. N. 1995. "Access and the Burden of Treatment: An Inter-State Comparison." Centre for Development Studies. Thiruvananthapuram, Kerala. Processed. Kumar, D., and R. B. Patel. 1992. "Study of Knowledge, Assessment and Practice of ISM Practitioners and Health Functionaries in the Context of Delivery of MTP Services in Bihar and Maharashtra." Operations Research Group, Baroda, Gujarat Processed. 342 * Better Health Systems for India's Poor Last, J. M. 1995. A Dictionary of Epidemiology. New York: Oxford University Press. Mahal, A., V. Srivastava, and D. Sanan. 2001. "Decentralisation and Public Service Delivery in Health and Education Services: Evidence from Rural India." In J.-J. Dethier, ed., Governance, Decentralization and Reform in China, India and Russia. Boston: Kluwer Academic Publishers. Management Sciences for Health. 1997. Managing Drug Supply: The Selection, Procurement, Distribution, and Use of Pharmaceuticals, 2nd ed. Bloomfield, Conn.: Kumarian Press. Mathiyazhagan, K. 1998. "Willingness to Pay for Rural Health Insurance through Community Participation in India." International Journal of Health Planning and Management 13:47-67. Ministry of Health and Family Welfare. 2000a. "Bulletin on Rural Health Statistics in India." Rural Health Division, New Delhi. - 2000b. The National Population Policy 2000. New Delhi. Mukhopadhyay, M., ed. 1997. Report of the Independent Commission on Health in India. New Delhi: Voluntary Health Association of India Press. Muraleedharan, V. R. 1997. "Hospital Services in Urban Tamil Nadu: A Survey of Maternity Services in Madras City and Chidambaram/Cuddalore Region." Report prepared for Citizen, Consumer and Civic Action Group, Chennai. . 1999a. "Availability and Distribution of Medical, Dental, Nursing and Pharmaceutical Professionals in Tamil Nadu : A Preliminary Assessment." Report submitted to the Tamil Nadu Dr. M. G. R. Medical University, Chennai. . 1999b. "Characteristics and Structure of the Private Hospital Sector in Urban India: A Study of Madras City." Small Applied Research Paper 5. Partnerships for Health Reforn Project. Bethesda, Md.: Abt Associates. Murray, C. J. L., and A. D. Lopez. 1996. The Global Burden of Disease. Cambridge, Mass.: Harvard University Press. Musgrove, P. 1999. "Public Spending on Health Care: How Are Different Criteria Related." Health Policy 47(3):207-223. Nandraj, S., and R. Duggal. 1996. "Physical Standards in the Private Health Sector." RadicalJournal of Health 2(2/3):141-84. Nandraj, S., A. Khot, and S. Menon. 1999. Accreditation of Hospitals: Breaking Boundaries in Health Care. Mumbai: CEHAT. Bibliography * 343 Narayan, D., R. Chambers, M. K. Sha, and P. Petesch. 2000. V'oices of the Poor: Crying Out for Change. New York: Oxford University Press. National Sample Survey Organisation (NSSO). 1992. "Morbidity and Utilisation of Medical Services: NSS 42nd Round (uly 1986-June 1987)." Sarvekshana 15(4):50-75, S131-S571. . 1998. "Morbidity and Treatment of Ailments: NSS Fifty-second Round (July 1995-June 1996)." Calcutta. OECD (Organisation for Economic Co-operation and Development). 1992. The Reform of Health Care: A Comparative Analysis of Seven OECD Countries. Paris. . 1994. The Reform of Health Care Systems: A Review of Seventeen Countries. Paris. Osborn, D., and T Gaebler. 1993. Reinventing Government. Reading, Mass.: Addison Wesley. Pai, M., P. Sundaram, K. K. Radhakrishnan, K. Thomas, and J. P. Muliyil. 1999. "A High Rate of Caesarean Sections in an Affluent Section of Chennai: Is It Cause for Concern? National Medical journal of India 12(4):156-58. Pauchari, S., ed. 1994. Reaching India's Poor: Non-governmental Approaches to Community Health. New Delhi: Sage Publications. Phadke, A. 1998. Drug Supply and Use: Towards a Rational Policy in India. New Delhi: Sage Publications. Planning Commission. 1998. Ninth Five-Year Plan 1997-2002. Vol. H. New Delhi. Prabhakaran, D., P. Shah, U. Shrivastava, A. K. Prabhakar, B. Shah, V K. Bahl, S. K. Puri, A. Bhaniani, M. Joshi, and K. S. Reddy. 2000. "Tobacco Consumption in North Indian Males Is Inversely Related to Professional Status: Results of Three Cross Sectional Surveys." Abstract 225/14. 1 lth World Conference on Tobacco or Health, August 6-11, Chicago. Preker, A. S., A. Harding, and N. Girishankar. 1999. "The Economics of Private Participation in Health Care: New Insights from Institutional Economics." Paper submitted to the International Social Security Association. Processed. Preker, A. S., A. Harding, and P. Travis. 2000. "'Make or Buy' Decisions in the Production of Health Care Goods and Services: New Insights frorr Institutional Economics and Organizational Theory." Bulletin of the Worlc Health Organization 78:779-90. 344 * Better Health Systems for India's Poor Purohit, B. C. 1995. "Private Voluntary Health Sector in India." Asian Economic Reviezv 37:297-311. Ranson, K. 1999. "The Consequences of Health Insurance for the Informal Sector: Two Non-Govermnental, Non-Profit Schemes in Gujarat." London School of Hygiene and Tropical Medicine, Health Policy Unit. Rao, K S., G. V. N. Ramana, and H. V. V. Murthy. 1997. "Financing of Primary Health Care in Andhra Pradesh: A Policy Perspective." Administrative Staff College of India, Hyderabad. Processed. Rao, V M., N. Alakh, R. Sharma, U. Shrivastava. 1999. Voices of the Poor: Poverty in People's Perceptions in India. New Delhi: Institute for Human Development. Reddy, K. N., and V. Selvaraju. 1994. Health Care Expenditure by Government of India: 1974-75 to 1990-91. New Delhi: Seven Hills Publications. Reddy, K. S., D. Prabhakaran, P. Shah, U. Shrivastava, A. K. Prabhakar, B. Shah, V K. Bahl, S. K. Puri, A. Bhaniani, and M. Joshi. 2000. "Tobacco Consumption in North Indian Males Is Inversely Related to Educational Level: Results of Three Cross Sectional Surveys." Abstract 225/12. 1 1h World Conference on Tobacco or Health, August 6-11, Chicago. Registrar General. Various years. Sample Registration System Bulletin. Ministry of Home Affairs, New Delhi. 1995. "Sample Registration System." Ministry of Home Affairs, New Delhi. 1998. Sample Registration System Bulletin. October. Ministry of Home Affairs, New Delhi. - . 1999a. Compendium of India's Fertility and Mortality Indicators 1971-1997: Based on the Sample Registration System (SRS). New Delhi. - 1999b. Sample Registration System Bulletin. April. Ministry of Home Affairs, New Delhi. . 2000. Sample Registration System Bulletin. April. Ministry of Home Affairs, New Delhi. Robinson, M. 1997. "Physician-Hospital Integration and Economic Theory of the Firm." Medical Care Research and Review 54:1. Roemer, M. I. 1991. National Health Systems of the World. 2 vols. New York: Oxford University Press. Bibliography * 345 Rohde, J., and H. Viswanathan. 1995. The Rural Private Practitioner. New Delhi: Oxford University Press. Saltman, R. B., and 0. Ferroussier-David. 2000. "On the Concept of Stewardship in Health Policy. " Bulletin of the World Health Organization 78:732-39. Saltman, R. B., and J. Figueras. 1998. "Analyzing the Evidence on European Health Care Reforms. Health Affairs (Millwood) 17(2):85-108. . eds. 1997. European Health Care Reform: Analysis of Current Strategies. Eurpoean Series 72. Copenhagen: WHO Regional Office for Europe. Satia, J. 1999. "Institutional Assessment: Strengthening Routine Imnunization- India." World Bank, New Delhi. Processed. Selvaraju, V 2000. "Public Expenditures on Health in India." Background paper for National Council of Applied Economic Research. Processed. Sen, P. 1997. "Community Control of Health Financing in India: A Review of Local Experiences." Bethesda, Md.: Abt Associates. Shah, G. 1996. "Public Health-Urban Society Interface: A Study of Pneumonic Plague in Surat." Centre for Social Studies, Surat, Gujarat. Processed. Shariff, A. 1997. "Human Development Profile of Rural India: Inter-State and Inter-Group Differentials: A Summary." NCAER, New Delhi. Sharma, S., W McGreevey, and D. Hotchkiss. 2000. "Financing Reproductive and Child Health Care in Rajasthan." USAID Report, Indian Institute for Health Management Research, The Policy Project, The Futures Group. New Delhi. Stafford, R. S. 1990. "Cesarean Section Use and Source of Payment: An Analysis ol Canadian Hospital Discharge Abstracts." American Journal of Public Healti 80:313-15. STEM (Center for Symbiosis of Technology, Environment and Management). 2000. "Uttar Pradesh Health Systems Development Project: Baseline Study." Project Report. Bangalore, Karnataka. Thaver, I. H., T. Harpham, B. McPake, and P. Garner. 1998. "Private Practitioner, in the Slums of Karachi: What Quality of Care Do They Offer?" Social Science and Medicine 46:1441-9 Tulasidhar, V B. 1992. State ' Financing of Health Care in India: Some Recent Trend,-. New Delhi: National Institute of Public Finance and Policy. 346 * Better Health Systems for India's Poor 1996. "Government Health Expenditures in India: Public Financing for Health in India: Recent Trends." Supported by the International Health Policy Program. Processed. Upleker, M. W, and D. S. Shepard. 1991. Treatment of Tuberculosis by Private General Practitioners in India. Bombay: Foundation for Research in Community Health. U.S. Department of Health and Human Services. 1991. Healthy People 2000. DHHS Publication No. (PHS) 91-50212, Washington, D.C. Wagstaff, A. 2002. "Inequality Aversion, Health Inequalities and Health Achievement." Policy Research Working Paper 2765. World Bank, Policy Research Department, Washington D.C. Wang, J., D. T.Jamison, E. Bos, A. S. Preker,J. Peadbody. 1999. Measuring Country Performance on Health: Selected Indicators for 115 Countries. Health, Nutrition, and Population Series. Washington, D.C.: World Bank. WHO (World Health Organization). 1985. "Appropriate Technology for Birth." Lancet 2(8452):436-3 7. . 1995. "The Treatment of Diarrhoea: A Manual for Physicians and other Senior Health Workers." WHO/CDR/95.3. Geneva. 1999. World Health Report 1999: Making a Difference. Geneva. 2000a. Health a Key to Prosperity: Success Stories in Developing Countries. Geneva. . 2000b. "Research into Action." Report on Tuberculosis Research. Regional Office for South-East Asia. .2000c. World Health Report 2000: Health Systems: Improving Performance. Geneva. 2001. Involving Private Practitioners in Tuiberculosis Control: Issues, Interventions, and Emerging Policy Framework. Geneva. World Bank. 1993. World Development Report 1993: Investing in Health. New York: Oxford University Press. . 1995. "India: Policy and Finance Strategies for Strengthening Primary Health Care Services." Report 13042-IN. Washington, D.C. Bibliography * 347 . 1996. "Staff Appraisal Report: State Health Systems Development Project II." Report 15106-IN. Washington, D.C. . 1997a. "Health, Nutrition, and Population: Sector Strategy." Health, Nutrition, Population Department. Washington, D.C. - 1997b. "India: New Directions in Health Sector Development at the State Level: An Operational Perspective." Report 15753-IN. Washington, D.C. . 1997c. "Project Appraisal Document: Malaria Control Project." Report 16393-IN. Washington, D.C. . 1999a. Curbing the Epidemic: Governments and the Economics of Tobacco Control. Development in Practice Series. Washington, D.C. . 1999b. "India Second State Health Systems Project, Mid-Term Review Report." South Asia Region. Washington, D.C. .2000a. "Confronting Poverty: The Challenge of Uttar Pradesh." South Asia Region. Washington, D.C. . 2000b. Poverty Reduction Strategy Sourcebook (CD-ROM). Poverty Reduction Economic Management, Poverty Unit. Washington D.C. Also available at http://www.worldbank.org/poverty/strategies/sourctoc.htm . 2000c. "Project Appraisal Document. Immunization Strengthening Project." Report 19894-IN. Washington, D.C. 2000d. World Development Indicators. New York: Oxford University Press. 2000e. World Development Report 2000/2001: Attacking Poverty. New York Oxford University Press. 2001. "Project Appraisal Document. Second National Leprosy Elimination Project." Report 21751-IN. Washington, D.C. Yaqub, S. 1999. "How Equitable is Public Spending on Health and Education." Background paper to World Bank (2000c). Yesudian, C. A. K. 1994. "The Behaviour of the Private Sector in the Health Service Market of Bombay." Health Policy Planning 9(1):72-80. I R - ~~~~~~~~- Inclia.s health care system is .at a croSSr'O.lCl5. 1an111v Indinils colntilnUC to stiffel rlncl clie fiomi preventable infectiois, un1clel-lLutr-ition, anld cmiupliCations fi-om i egnaicy and Childbirth- thle "LuInfillishlCdI 'agenda." At the same timiec, new health threats, suiclh as the AIDS epidemic, are cihalleigirin, the abilitv ol the health so,ystem to respond. Ind(lia's health caIrC systemii n'eeds to build capacity to complete the unfinishcl lClagcnda anild to iiect emenging r leinenainls. \Vith the Indian government, the \'Vorld Bink .issenibled a stucdy te.am of Indlia' s inter-lxil andl external ldevelopment pa tners to amnlyze options for the futur-e couise of the healthi cire system. BMIttr I-lHlth .ytems for- Indias Poor : Fiingds, -I Analjids: aid Optionis is tle restldt. I-he report addresses two main qLiuestionis: * - low can Iliia imeet the heailthi care neecIs of its poor. * HiOW Ca.1 the 1ubflic a11d pl ivate heailthi sectors be struCtUred to bettel finalnl Ce a1ntI deliver services: ThL book presents new empirical iresearleb, and nap;ll)S out poliCly option6s for impr-ovilng heallth outcomlies for all Indians. TIhI1 fir-st section presents cui rrent con(litionis and policy opitions. Thle seconid section dlescrlibcs the theory and evidence to SupI)Iport thc policy1 choices. AlthoLgHll aCddrI1essing the spccific needIs of Indlil, this report is relevant to anly country inter-estecl in making its heallthi systems imor-e effective, equithle. and acCoL untable. lAMB ISBN 0-821 3-5029-3