W O R L D B A N K W O R K I N G PA P E R N O . 1 8 7 A F R I C A H U M A N D E V E L O P M E N T S E R I E S 54037 Improving Primary Health Care Delivery in Nigeria Evidence from Four States THE WORLD BANK W O R L D B A N K W O R K I N G P A P E R N O . 1 8 7 Improving Primary Health Care Delivery in Nigeria Evidence from Four States Africa Region Human Development Department Copyright © 2010 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First Printing: April 2010 Printed on recycled paper 1234 13 12 11 10 World Bank Working Papers are published to communicate the results of the Bank's work to the development community with the least possible delay. The manuscript of this paper therefore has not been prepared in accordance with the procedures appropriate to formally edited texts. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the International Bank for Reconstruction and Development/The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not 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 any judgment on the part of The World Bank of the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission promptly to reproduce portions of the work. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, Tel: 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, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, Fax: 202 522 2422, email: pubrights@worldbank.org. ISBN: 978 0 8213 8311 7 eISBN: 978 0 8213 8313 1 ISSN: 1726 5878 DOI: 10.1596/978 0 8213 8311 7 Library of Congress Cataloging in Publication Data has been requested. Contents Foreword .................................................................................................................................... ix Acknowledgments .................................................................................................................... x Acronyms and Abbreviations ................................................................................................ xi Executive Summary............................................................................................................... xiii Primary Health Care Delivery in Four States ..............................................................xiii Understanding the Performance of Primary Health Care in the States...................xvii Division of Responsibility among Government Levels .............................................xvii Citizens/Clients Policy Makers ....................................................................................xviii Policy Makers Providers.................................................................................................. xx Clients Providers ............................................................................................................. xxi Possible Ways Forward..................................................................................................xxii 1. Introduction............................................................................................................................ 1 Objectives............................................................................................................................. 2 Conceptual Framework...................................................................................................... 3 Methodology ....................................................................................................................... 5 2. Context .................................................................................................................................... 7 Health Outcomes and Access to Health Services in Nigeria......................................... 7 Context in States Included in the Study........................................................................... 9 3. Status of Primary Health Care Services........................................................................... 11 Organization of the Primary Health Care System........................................................ 11 Survey Results ................................................................................................................... 12 Infrastructure and Amenities .......................................................................................... 13 Services Available ............................................................................................................. 14 Equipment and Medical Supplies................................................................................... 15 Pharmaceuticals ................................................................................................................ 16 Health Personnel............................................................................................................... 17 Exemption and Waiver Programs .................................................................................. 20 Education and Promotion Activities of PHC Services ................................................. 30 Service Charges ................................................................................................................. 31 4. Division of Responsibilities among Government Levels............................................ 33 Laws and Policies Informing the Division of Responsibilities for the Delivery of Primary Health Care ................................................................................................. 33 Division of Responsibilities in Practice .......................................................................... 34 Policies and Guidelines .................................................................................................... 34 iii iv Contents Personnel Training............................................................................................................ 36 Health Care Personnel Management ............................................................................. 36 Infrastructure: Construction and Maintenance............................................................. 37 Procurement and Distribution of Pharmaceutical Products ....................................... 39 Supervision ........................................................................................................................ 41 Possible Ways Forward.................................................................................................... 41 5. Clients Policy Makers ........................................................................................................ 43 Local Government Revenues and Responsibilities ...................................................... 43 Public Financial Management ......................................................................................... 46 Local Government Civil Service ..................................................................................... 49 Local Government Health Expenditure......................................................................... 50 Local Government Accountability for Service Delivery.............................................. 54 Possible Ways Forward.................................................................................................... 55 6. Policy Makers Providers .................................................................................................... 61 Characteristics of Health Personnel ............................................................................... 62 Education Level................................................................................................................. 63 Incentives to Providers..................................................................................................... 64 Mechanisms to Reward and Discipline PHC Personnel.............................................. 67 Other Negative Incentives Faced by PHC Personnel................................................... 68 Health Personnel Coping Mechanisms.......................................................................... 70 Possible Ways Forward.................................................................................................... 71 7. Clients Providers ................................................................................................................. 74 Increasing Clients' Power ................................................................................................ 74 Survey Results ................................................................................................................... 75 Possible Ways Forward.................................................................................................... 77 References................................................................................................................................. 81 Appendixes............................................................................................................................... 85 Appendix A: Sample Size ................................................................................................ 85 Appendix B: Household Survey Sample Characteristics ............................................ 88 Tables Table 1. Primary Health Care Facilities Infrastructure across States and Facility Ownership (in %) ............................................................................................................xiii Table 1.1. Analysis of Survey Questionnaires ....................................................................... 6 Table 2.1. Health Outcomes and Health Care Utilization across Geopolitical Zones, Nigeria 2003 ........................................................................................................... 8 Table 2.2. Population, Poverty, and Inequality Indicators, Nigeria 2004......................... 10 Table 3.1. Health Facility Type by LGA Type ..................................................................... 12 Table 3.2. Basic Information from All States (in %) ............................................................ 13 Contents v Table 3.3. Primary Health Care Facilities, Infrastructure, and Amenities across States (in %) ...................................................................................................................... 14 Table 3.4. Percentage of Facilities Offering Basic Services across States.......................... 15 Table 3.5. Percentage of Facilities with Equipment and Medical Supplies across States.................................................................................................................................. 15 Table 3.6. Percentage of Facilities Having Basic Pharmaceuticals and Vaccines in Stock across States............................................................................................................ 17 Table 3.7. Average Staffing of PHC Facilities across States and Across Type of Ownership ........................................................................................................................ 18 Table 3.8. Average Staffing of Public Health Facilities across Facility Type ................... 19 Table 3.9. Average Staffing of PHC Facilities across LGA Type ....................................... 19 Table 3.10 Average Staff in Basic Health Centers across Type of LGA............................ 20 Table 3.11. Percentage of Facilities Offering Exemption and Waivers across States ......... 21 Table 3.12. Opening Hours across Facility Type and LGA Type (in %) .......................... 21 Table 3.13. Percentage of Facilities Offering Basic Services across Type of Facility and across Type of LGAs ................................................................................................ 22 Table 3.14. Percentage of Facilities with Basic Equipment across Type of Facility and Type of LGA ............................................................................................................. 23 Table 3.15. Percentage of Facilities with Basic Drugs and Vaccines in Stock across Type of Facility and Type of LGA ................................................................................. 23 Table 3.16. Basic Information on PHC Facilities across Public and Private Ownership (in %) ............................................................................................................. 25 Table 3.17. Percentage of Facilities with Basic Equipment across Public and Private Ownership........................................................................................................... 26 Table 3.18. Availability of Basic Health Services in Nearest Facility across States (in %).................................................................................................................................. 28 Table 3.19. Household Satisfaction with Nearest PHC Facility across States (in %) ...... 28 Table 3.20. Household Satisfaction with Nearest PHC Facility across Facility Ownership and across Type of LGA (in %).................................................................. 29 Table 3.21. Difference in Satisfaction with Nearest PHC Facility between Male and Female Heads of Households (in %) ............................................................................. 29 Table 3.22. Percentage of Households near a PHC Facility Visited by Facility Health Personnel across States, Type of Ownership, and Type of LGA .................. 30 Table 3.23. Reason for Health Facility Worker Visit across States (in %)......................... 31 Table 3.24. Percentage of Services with a Charge across States ........................................ 31 Table 3.25. Household Utilization of Nearest Health Facility across Type of LGA (in %).................................................................................................................................. 32 Table 4.1. Level of Government or Agency that Provided the Health Facility Building (in %) ................................................................................................................. 37 Table 4.2. Main Agency Responsibility for the Maintenance of Equipment and Buildings across States .................................................................................................... 38 Table 4.3. Main Supplier of Medical Consumables, Drugs, and Equipment to PHC Facilities across States...................................................................................................... 40 Table 5.1. Changes in the Actual Distribution of Federation Account Revenues across Three Government Levels (in %) ....................................................................... 44 vi Contents Table 5.2. Main Fiscal Trends for the Consolidated Government, 1999­2005, (billions of naira) .............................................................................................................. 45 Table 5.3. Budget Execution Rate across LG in Kaduna and Cross River (in %) ............ 46 Table 5.4. Capital Budget Execution Rate across LG in Kaduna and Cross River (in %).................................................................................................................................. 48 Table 5.5. Wage Bill in Different Sub Saharan Africa Countries, 2005............................. 49 Table 5.6. Percentage of Civil Servants out of Total Population in Sub Saharan African Countries............................................................................................................. 50 Table 5.7. Real Growth Rate of Kaduna LG Expenditures in 2003­04 and 2004­05 (in %).................................................................................................................................. 53 Table 5.8. Real Growth Rate of Cross River's LG Expenditures in 2005­06 (in %)......... 54 Table 5.9. Cross River Local Governments Expenditure 2005........................................... 59 Table 5.10. Kaduna Local Government Expenditure 2005................................................. 60 Table 6.1. Health Care Personnel Sampled across States ................................................... 61 Table 6.2. Health Care Personnel Sampled by Gender across States................................ 62 Table 6.3. Characteristics of PHC Personnel across States and across Type of Facility Ownership .......................................................................................................... 63 Table 6.4. Characteristics of PHC Personnel across Type of Personnel ........................... 63 Table 6.5. Highest Level of Education Completed by PHC Staff Interviewed (State Comparison) ..................................................................................................................... 64 Table 6.6. Average Salary of PHC Personnel across Type of Facility Ownership .......... 65 Table 6.7. Salary of Doctors and Nurses in Relation to GDP per Capita in Different Sub Saharan African Countries (in %) .......................................................................... 65 Table 6.8. Salaries and Fringe Benefits (State Comparison) (in %) ................................... 66 Table 6.9. Average Salary of Public PHC Personnel across Type of LGA ....................... 67 Table 6.10. Criteria for Promotion of Staff (in %) ................................................................ 68 Table 6.11. Negative incentives Faced by PHC Personnel across States (in %)............... 68 Table 6.12. Obstacles in Doing Job across Rural and Urban Areas (in %) ....................... 69 Table 6.13. Obstacles in Doing Job across Type of Facility Ownership (in %) ................ 69 Table 6.14. Percentage of Personnel Who Are Fulltime Employees and Supplement Their Salary ................................................................................................ 70 Table 6.15. Activities to Supplement Salaries of Health Staff across States (in %) ......... 71 Table 7.1. Percentage of Health Facilities with a Functioning Health Management/Development Committee and Gender of Committee Members across States, and across Facility Ownership ............................................................... 75 Table 7.2. Frequency of Meetings of Health Committees across States ........................... 76 Table 7.3. Actions of Community Health Management/Development Committees across States and Facility Ownership (in %) ................................................................ 76 Table 7.4. Final Decision on Health Facility Managerial Issues (in %)............................. 77 Table B.1. Employment and Occupation of Household Survey Respondents................ 88 Table B.2. Housing Characteristics of Household Survey Respondents.......................... 89 Table B.3. Household Survey Respondents' Proximity to Nearest Health Facility........ 90 Contents vii Figures Figure 1. Percentage of Facilities Having Equipment and Consumables across States.................................................................................................................................xiv Figure 2. Percentage of Facilities Having Basic Pharmaceuticals and Vaccines on Stock across States............................................................................................................ xv Figure 3. Household Satisfaction with Nearest Primary Health Care Facility...............xvi Figure 4. Accountability Relationships between Politicians/Policy Makers, Providers, and Citizens/Clients ...................................................................................xvii Figure 5. Clients/Citizen Policy Makers Relationship.....................................................xviii Figure 6. Relationship between Local Governments and Providers ................................ xx Figure 7. Accountability Relationship between Clients and Providers .......................... xxi Figure 1.1. Accountability Relationships between Politicians/Policy Makers, Providers, and Citizens/Clients ....................................................................................... 4 Figure 2.1. Socioeconomic Disparities in Health Outcomes and Basic Service Utilization, Nigeria, 2003 .................................................................................................. 8 Figure 2.2. Utilization of Outpatient Care across Population Consumption Quintiles and Type of Provider or Type of Provider Ownership ............................... 9 Figure 3.1. Percentage of PHC Facilities Offering Basic Services across Public and Private Ownership........................................................................................................... 25 Figure 3.2. Percentage of Facilities Having Basic Pharmaceuticals and Vaccines on Stock across Public and Private Ownership................................................................. 27 Figure 4.1. Government Agencies with Responsibilities in PHC in Cross River............ 35 Figure 4.2. Government Agencies with PHC Delivery Responsibilities in Kaduna....... 35 Figure 5.1. Average Ratio between Actual and Projected Internally Generated Revenues in Bauchi's Local Governments.................................................................... 47 Figure 5.2. Share of Total LG Expenditure Allocated to Health in Kaduna LG, 2005.... 52 Figure 5.3. Total per Capita Public Expenditure across Local Governments in Kaduna and Cross River, 2005 ....................................................................................... 52 Figure 6.1. Process to Discipline PHC Personnel ................................................................ 67 Figure B.1. Household Survey Respondents' Literacy Level ............................................ 88 Figure B.2. Household Survey Respondents' Education Status........................................ 88 viii Contents Boxes Box 1.1. Declaration of Alma Ata: International Conference on Primary Health Care, Alma Ata, (presently Almaty, Kazakhstan) 1978................................................ 3 Box 3.1. Proposed Health Manpower for a PHC Center to Provide the Minimum Health Care Package ....................................................................................................... 18 Box 5.1. Extract from Report of the Auditor General for Local Governments on the Accounts of the 23 Local Government Councils of Kaduna State. For the Year Ended 31st December, 2004. ............................................................................................ 49 Box 5.2. Citizen Report Cards: The Bangalore Experience ................................................ 57 Box 5.3. Participative Approaches in the Management of Education: Literacy Enhancement Assistance Project (LEAP)...................................................................... 57 Box 7.1. Kaduna: Example of Facility Health Committee Role in Improving the Condition of PHC Facilities ............................................................................................ 78 Box 7.2. Experience with Vouchers for Health Services..................................................... 79 Foreword E ach year many lives, especially children's, are lost in Nigeria. Communicable diseases, particularly malaria, pneumonia, and diarrhea, often linked with malnutrition, caused most of these premature deaths. Most interventions proven to prevent or treat these illnesses are primary health care interventions; some of them can be provided by the households themselves after some orientation from a health provider, either inside health facilities or through community outreach. Maternal mortality is also high in the country. Some of these deaths can also be prevented. Although access to referral care is essential to improve maternal survival; primary health care interventions can prevent some of the indirect causes of maternal deaths such as anemia, malaria, STI as well as the major factors underlying medical causes. Given the importance of primary health care services for the country to achieve the Millennium Development Goals, it is important to generate knowledge on the challenges faced in delivering PHC services that would allow authorities to design and implement programs to respond to these challenges. This report then aims precisely at understanding the performance of primary health care (PHC) providers in four Nigerian states, Bauchi, Cross Rivers, Kaduna, and Lagos and the variables driving this performance. The report also aims at offering policy options to improve the delivery of PHC services in these states. This report was prepared by the World Bank in partnership with the National Primary Health Care Development Agency, the Federal Ministry of Health, and the Canadian International Development Agency. The report was made possible thanks to the support received from the States Ministries of Health of Bauchi, Cross River, Kaduna, and Lagos and the financial support of the Canadian International Development Agency. Finally, the study also benefited from some financial support from the Bank Netherlands Partnership Program (BNPP). Agnes Soucat Advisor, Health Nutrition Population for Africa, Africa Region Human Development The World Bank ix Acknowledgments T his study was undertaken by the World Bank in partnership with the Canadian International Development Agency; The Federal Ministry of Health, Nigeria; and the National Primary Health Care Development Agency. The Nigeria team was headed by Mrs. Koleoso Adelekan (Executive Director of National Primary Health Care Development Agency) Dr. Shehu Mahdi (previous Executive Director of National Primary Health Care Development Agency), and Dr. Muhammad Pate (Executive Director of National Primary Health Care Development Agency at the time of the dissemination of the study). The team was composed by Prof. Akpala (Director of Planning, Research, and Statistics, NPHCDA), Dr. Iyabo Lewis (Consultant, NPHCDA), and Dr. O. Ogbe (Department of Planning, Research, and Statistics, NPHCDA). Dr. Tolu Fakeye (Head Division of International Health, Department of Health Planning and Research, Federal Ministry of Health) was also part of the team. The World Bank team was headed by M.E. Bonilla Chacin (Senior Economist, AFTH3) who coordinated the overall work and wrote the report. Ramesh Govindaraj (Senior Health Specialist, AFTH3) and Mrs. Anne Okigbo (Senior Health Specialist, AFTH3) also participated in the work. Ngozi Malife provided great support to the team. The study was done under the overall guidance of Lynne Sherburne Benz (Sector Manager) and Onno Ruhl (Country Director). This report is mainly based on quantitative surveys on health facilities, health care personnel, and households in their vicinity that were commissioned for this study. These surveys were designed and implemented by a consortium of the following firms: EPOS Health Consultants; Canadian Society for International Health; and Center for Health Sciences Training, Research and Development (CHESTRAD). This consortium also prepared a report on the results of the surveys that served as background for this report. The team also acknowledges the participation of Mr. Pierre Tremblay (Development Officer, CIDA), Mr. Martin Osubor (Development Officer, CIDA), and Mr. Bernard Heaven (Development Officer, CIDA) from the Canadian International Development Agency. We also gratefully acknowledge the support of the State Ministries of Health of the participating states: Bauchi, Cross River, Kaduna, and Lagos. In addition, the study benefited from invaluable suggestions and comments from: Dr. Kolawole Maxwell (PATHS), Dr. Stuti Khemani (Senior Economist, World Bank), Dr. Jeffrey Hammer (Lead Economist, World Bank), Dr. Oscar Picazo (Senior Economist, World Bank), Mr. Ismail Radwan (Senior PSD Specialist, World Bank) and Dr. Maureen Lewis (Advisor, World Bank). Finally, the study benefited from comments received during a workshop that took place in December of 2007 where the preliminary results of the survey were presented. This study was completed mainly with the financial support of the Canadian International Development Agency and also benefited from the Bank Netherlands Partnership Program. x Acronyms and Abbreviations ACT Artemisin Combination Treatment BASEEDS Bauchi State Economic Empowerment and Development Strategy BHC Basic Health Services BEOC Basic Emergency Obstetric Care CDC Center for Disease Control CHC Comprehensive Health Services CHEW Community Health Extension Worker CHO Community Health Officer CIDA Canadian International Development Agency CPS Country Partnership Strategy CSR Country Status Report DA Development Areas DFID Department of International Development DHS Demographic and Health Surveys DPHC Department of Primary Health Care EA Enumeration Area ESW Economic and Sector Work FA Federation Account FHC Facility Health Committee FMOH Federal Ministry of Health GDP Gross Domestic Product HIV/AIDS Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome HND Higher National Diploma HP Health Posts and Dispensaries IGR Internally Generated Revenue ITN Insecticide Treated Net JCHEW Junior Community Health Extension Worker LEAP Literacy Enhancement Assistance Project LEEMP Local Empowerment and Environmental Management Project LG Local Government LGA Local Government Area LGSC/LGS Local Government Service Commission/ Local Government Service B Board MDG Millennium Development Goals NAFDAC National Agency for Food, Drug Administration and Control NEEDS National Economic Empowerment and Development Strategy NLSS Nigerian Living Standards Survey NPHCDA National Primary Health Care Development Agency NYSC National Youth Service Corps OND Ordinary National Diploma xi xii Acronyms and Abbreviations ORS Oral Rehydration Salts PATHS Partnership for Transforming the Health System PEMFAR Public Expenditure Management and Financial Accountability Review PFM Public Financial Management PHC Primary Health Care SACI State Action Committee for Immunization SEEDS State Economic Empowerment and Development Strategy SMLG State Ministry of Local Government SMOH State Ministry of Health SRDC State Rural Development Commission SSA Sub Saharan Africa STI Sexually Transmitted Infections TB Tuberculosis UNFPA United Nations Population Fund UNICEF United Nations Children's Fund USAID United States Agency for International Develo VAT Value Added Tax WDR World Development Report WHO World Health Organization WHS Ward Health Services WMHCP Ward Minimum Health Care Package Executive Summary T his study aims mainly at understanding the performance of primary health care (PHC) providers in four Nigerian states and the variables driving this performance. The study is primarily based on quantitative surveys at the level of primary health care facilities, health care personnel, and households in their vicinity. These surveys were implemented in four states: Bauchi, Cross River, Kaduna, and Lagos. Primary Health Care Delivery in Four States The organization of the delivery of primary health care services largely varies across states. The role of the private sector in service provision is larger in the southern states, particularly in Lagos. The public PHC delivery system also varies significantly. For instance, many states have progressively eliminated health posts and dispensaries. These are the smallest PHC facilities offering only a limited set of services, mainly child health services. However, in the northern states, and particularly in Bauchi, they represent an important share of PHC facilities. The results of the health facility survey shows that often these facilities have decaying infrastructure, do not offer all basic services, and do not have all the health personnel, equipment, medical supplies, and pharmaceuticals needed to effectively offer services. There are, however, large differences across states, rural and urban local governments, and across public and private ownership. In general, the condition of the infrastructure of PHC facilities, particularly public facilities, is poor. As seen in table 1, about two out of every five facilities sampled in the survey have leaky roofs, broken windows and/or doors. Less than three out of every four facilities have waste disposals, electricity, fridge/icebox, or toilets. Table 1. Primary Health Care Facilities Infrastructure across States and Facility Ownership (in %) Bauchi Cross River Kaduna Lagos Private Government Infrastructure Taps with running water 22 26 27 80 78 16 Safe water 66 70 65 91 95 57 Electricity 44 60 31 95 94 38 Condition Leaky roof 65 43 52 11 15 57 Broken doors/window 61 40 56 12 15 57 Cracked floor 73 44 57 16 23 60 Clean 86 97 66 86 87 83 Source: Health Facility Survey (EPOS, CISH, CHESTRAD, 2007). xiii xiv Executive Summary Most health facilities offer child health services, however, maternal services and particularly family planning services are less likely to be offered. Type I facilities (health posts and dispensaries) are less likely to offer maternal services, including preventive services such as antenatal care. As most facilities in Bauchi are type I, these services are less frequently available in the state. Family planning and the control of sexually transmitted diseases are the services that are least available in PHC facilities, particularly in the northern states. A large share of PHC facilities do not have all the equipment needed to offer basic services to the communities they serve. PHC facilities are more likely to have medical consumables such as bandages, sterile gloves, and syringes (figure 1). Similarly, most facilities have some basic equipment such as thermometers, and stethoscopes. However, less than two thirds of PHC facilities have other basic equipment and supplies such as child weight scales, sharp containers, and antiseptics. Figure 1. Percentage of Facilities Having Equipment and Consumables across States Sterile gloves Disposable syringes Antiseptic for skin Stethoscope Sharps container Bandages Thermometer Child weight scale 0 20 40 60 80 100 Percent Bauchi Cross River Kaduna Lagos Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). A large percentage of facilities do not have basic pharmaceuticals on stock and only about half of the PHC facilities have vaccines on stock (see figure 2). Anti malaria drugs are the most frequently available drugs in the facilities. However, only two out of every three facilities have ORS sachets on stock. Micronutrient supplements are also in low supply. Despite the efforts and considerable improvements in immunizations, maintaining vaccines on stock remains challenging. Executive Summary xv Figure 2. Percentage of Facilities Having Basic Pharmaceuticals and Vaccines on Stock across States DPT Measles Co-trimoxazole Condoms Vitamin A ORS sachets Antibiotics ACT Chloroquine 0 10 20 30 40 50 60 Percent Bauchi Cross River Kaduna Lagos Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). There are major constraints in the referral system. Although most of the facilities refer patients, only about half of them have easy communication with the referral center. The average walk time to referral centers is 60 minutes and the drive time 20 minutes; nevertheless, the chances of encountering difficulties with transportation are considerable, since only 31 percent of the health facilities have access to transportation to deal with emergency cases. Most PHC facilities, with the exception of Lagos, are staffed by community health workers and nurses and midwives. Community health workers, including Community Health Officers (CHOs), Community Health Extension Workers (CHEWs) and Junior Health Extension Workers (JCHEWs), are unique to Nigeria. These cadres of health care personnel were introduced by the Basic Health Service Implementation Scheme (1975­1983). They have allowed the staffing of basic health facilities in the country. In relation to recommended national standards, most PHC facilities are understaffed. NPHCDA has established a minimum ward health care package to be provided by 2012. To provide this package, NPHCDA sets recommendations concerning the staffing of all PHC health facilities. However, on average, very few have this recommended number and skill mixed of staff. For instance, on average, the sampled clinics and health centers do not meet the proposed standard for clinics, let alone that of health centers, as they have less than 4 JCHEWs, less than 2 CHEWs, and less than 3 nurses/midwives on staff. xvi Executive Summary Facilities in all states offer exemptions and waivers but to a limited degree. Facilities in all states offer exemptions to some health services such as routine immunization, family planning, and antenatal care. Facilities in Cross River more frequently offered free services, while those in Lagos had the lowest percent of exemptions. However, these exemptions were not standard as most of them were offered less than 50 percent of the time. Concerning fee waivers for disadvantaged groups, most groups were generally asked to pay for services with the exception of clients with TB / leprosy and onchocerciasis. Lagos had the highest percent of people required to pay in all groups. Household Satisfaction with Services Provided by Nearest PHC Facility On average, households are satisfied with the availability of services in the PHC facilities, although there are large differences across states. Reflecting the results of the facility survey, households in Bauchi are the least likely to be satisfied with the availability of services in their nearest PHC facility, particularly regarding the availability of maternal services. However, satisfaction with the services provided by PHC facilities is low in all states. Less than 50 percent of households were satisfied with the availability of drugs, equipment, medical supplies, and staff. The pattern of satisfaction across states also mirrors the availability of the equipment and supplies in the health facilities across states. Households in Bauchi and Kaduna were the least satisfied, followed by Cross Rivers and Lagos (see figure 3). Satisfaction with waiting time, with information provided regarding disease control and care, and with information on facility management was highest in both Cross River and Lagos. The level of household satisfaction also varied with the gender of the household head, as women were more likely to be satisfied with the services. Figure 3. Household Satisfaction with Nearest Primary Health Care Facility Waiting Times Health information Availability of Equipment Attitude of Staff Availability of Supplies Drug Supply 0 10 20 30 40 50 60 70 80 Percent Bauchi Cross River Kaduna Lagos Source: Household Survey (EPOS, CISH, CHESTRAD, 2007). Executive Summary xvii The pattern of satisfaction with facility staff attitude was different. Households in Bauchi were the most satisfied with the attitude of health care staff, while those in Kaduna the least satisfied. This was in general the health service aspect that received the largest percentage of satisfaction. However, less than 60 percent of household heads were satisfied with the staff attitude. There are particular weaknesses regarding the education and promotion activities of PHC facilities, particularly in the two northern states. Only few households reported having access to both outreach and public health education activities in all states, but particularly in Kaduna and Bauchi. Similarly, the level of household satisfaction with the information on disease prevention and control is also very limited. In both Bauchi and Kaduna, less than 25 percent of households were satisfied with the information received. Understanding the Performance of Primary Health Care in the States This study follows the World Development Report (WDR) 2004 framework on service delivery to understand the performance of PHC services in Nigeria. This framework explains service performance through three accountability relationships: "voice" between citizens/clients and politicians/policy makers, "compact" between policy makers and providers, and "client power" between providers and clients. If any of these relationships is not working, the services provided will not meet the needs or expectations of the patients. Thus to improve service delivery community members have two different routes; a "long route" by exercising pressure to their elected officials for them to ensure that providers offer quality services, and a "short route" by increasing their power over providers. Accountability in this study is defined as the obligation to answer questions regarding decisions and actions (Brinkerhoff, 2004). Accountability would then imply both reporting information and justification for actions and decisions. It also implies the existence and application of sanctions. Division of Responsibility among Government Levels Figure 4. Accountability Relationships between Politicians/Policy Makers, Providers, and Citizens/Clients Federal Government State Government Local Government Clients/Citizens Providers Source: Adapted from WDR 2004 and World Bank 2006. xviii Executive Summary In Nigeria there are three "long routes" of accountability as shown in figure 4. The three levels of government, Federal, state, and local governments have some responsibility in the provision of health services. The three levels have relationships with citizens and with the PHC providers, in particular the states and LGs. Thus, as basic service delivery in the country is decentralized, to understand the performance of PHC facilities is also important to understand the relationships between the different levels of government regarding health services. The division of roles and responsibilities between the federal, state, and local government levels, particularly between states and LGs, is complex and varies across states. The local governments have the main responsibility regarding the management of PHC. However, there is no single level or a single agency in charge of financing, managing, and supervising these services; of recruiting, training, and promoting PHC personnel; of setting and paying staff salaries; building and maintaining facilities; and providing drugs and supplies. Often the three levels of government and various agencies within each level participate in these activities, creating duplication and gaps in provision. In addition, some states have created a subdivision of the LGAs, the development areas, which also have some responsibilities regarding PHC. These unclear lines of responsibilities have undermined the accountability relationships between citizen and policy makers, as it is not clear which level of government or agency within each level should fully answer the community on service delivery issues. The accountability relationship between providers and policy makers is also undermined, as there are many agencies with responsibility in the management of human resources, making sanctions for improper behavior difficult to implement. Citizens/Clients-Policy Makers Figure 5. Clients/Citizen-Policy Makers Relationship Federal Government State Government Local Government Clients/Citizens Source: Adapted from WDR 2004 and World Bank 2006. Executive Summary xix Although most levels of government and different agencies within each level share health care responsibilities, the local governments are the main level in charge of delivering basic services. To be fully accountable to citizens, local governments need to have the capacity to provide services, in other words, they need to have the financial and human resources required. Local Government Revenues and Responsibilities For many years there has been a debate on whether local governments receive enough resources to meet their responsibilities (World Bank, 2002). During the last military regime after many complaints for non payment of primary school teachers' salaries, the federal government started to deduct the salary of teachers from the LGAs (LGAs) Federation Account (FA) allocation. Many LGs complained that this deduction at source created such a large reduction of their total revenues that they were left with a "zero allocation" to fulfill their other responsibilities (World Bank, 2002). However, the local government revenues have increased considerably in the last years and thus the "zero allocation" phenomenon is not an issue at the moment. The LGs' share of the Federation Account, where oil revenues are centralized, has increased significantly since 1999. In addition, the total consolidated revenues of the entire government have also increased thanks to the increasing oil prices. Nevertheless, LGs face many limitations in the use of these resources. Some of these limitations are statutory, such as deductions at source; others are administrative, such as limitations to their autonomy in drafting and executing their budget or in personnel management (World Bank, 2002). For instance, in most states, LGs need clearances from the state government to spend resources above a threshold or to obtain a loan. These limitations can be large and vary across states. However, these limitations to the LGs autonomy and the little revenues they received in the past do not fully explain their service delivery record. Public expenditure management in LGs is weak: budgets are unrealistic, record keeping is poor, and irregularities in the use of funds are common. In addition, many local governments, despite having overstaffed civil services, have limited capacity in public financial management and other aspects linked to their service delivery responsibility. Local Government Health Expenditure Additionally, local government expenditure on health is low and varies largely across and within states. For instance, on average, local governments in Kaduna spend about US$2 per capita on health and local governments in Cross River spend about US$1.05 per capita in non salary health expenditure. Despite increases in total local government expenditure per capita in the last years, in the instances when health expenditure has increased, it has done so at a much lower rate. Most of this expenditure is on personnel remuneration, very little is set aside for other recurrent costs. In particular, very little is allocated to the maintenance of health facilities. Local Government Accountability for Service Delivery LGs' weak accountability towards health personnel concerning payment of salaries have been noted before (Khemani, 2005). In the states sampled in this study this does not seem to be an issue, although delays in salary payments are. Nevertheless, in Cross xx Executive Summary River, the state now manages the payroll of LGs, as in the past many LGs staff complained for salary non payments. The level of accountability of local governments towards other levels of government could also be measured by the amount of information sharing on budget process, and on activities or outputs. Very little of this is done. Information on local government budgets and expenditure is difficult to come by. LGs, however, are answerable to auditors general of LGs but this information is usually given with delays and the auditor general is often powerless to apply any sanction for irregularities. Local government accountability in relation to communities could also be measured by their responsiveness to communities. Information on rural local governments in nine states, including Bauchi, indicates that the level of responsiveness to communities is also low (Terfa Inc., 2005). Policy Makers-Providers Figure 6. Relationship between Local Governments and Providers Federal Government State Government Local Government Providers Source: Adapted from WDR 2004 and World Bank 2006 Policy makers aiming at providing quality services would not be able to achieve this goal if they cannot guarantee that providers will deliver these services. However, ensuring providers' compliance to offer quality services is not simple; it requires offering the right incentives and a close monitoring of their work. The Nigerian government has ensured the staffing of primary health care facilities by creating special types of PHC personnel, community health workers. Most health personnel working in primary health care facilities are CHEWs and JCHEWs, although there are also nurses. Often these workers come from the same area where they work, ensuring their integration in the community they serve. Nigeria does not have the acute lack of health personnel that is common in other countries in the region. The majority of these workers are women, with the exception of Bauchi state where the majority of PHC workers are men. Having women as PHC staff reduces a barrier to access services which is the concern of non availability of a female provider. Executive Summary xxi Despite these positive aspects in the recruitment of PHC personnel, there is still room for improvements as many factors determining health personnel motivation are missing. Most PHC personnel have received their salaries in the last year; however, a large share of them receives their salaries with delays. In relation to GDP per capita, when compared to other countries in the region, these salaries are relatively low. In addition, working conditions are difficult, particularly in rural areas. Health workers often do not have basic drugs and equipment to offer services; do not receive adequate training; and are poorly supervised. Finally, health care personnel are very unequally distributed across rural and urban areas, partly because the incentives to serve in rural and isolated areas are small. In addition, providers' accountability in relation to policy makers and clients is weak. Measuring providers' accountability to local governments and patients is difficult. Lewis (2006) includes as a key measure of provider's accountability the "authority to reward performance and discipline, transfer, and terminate employees who engage in abuses". In the four states surveyed, the management of PHC personnel is cumbersome and fragmented given the number of agencies involved. Similarly, the lines of responsibilities regarding personnel supervision and management are not always clear. This makes any measure to discipline or motivate health personnel difficult to implement. As a result, frontline providers face little consequence for non performance. Finally, their salaries are fixed and not linked to the provision of services; thus, they have little incentives to respond to the communities' demands. Many workers, in response to inadequate remuneration and working conditions, respond by developing different coping strategies (Van Lerberghe et al., 2002). Although the majority of PHC personnel work full time, a large percentage supplements their salaries, especially in the two northern states. Most do agricultural work; however, an important percentage also sells medicines or provides health care at home. Clients-Providers Figure 7. Accountability Relationship between Clients and Providers Clients/Citizens Providers Source: WDR 2004. When the "long route" of accountability is not properly working, increasing client's power can result in improvements in service delivery, but is not a panacea, as there are important market failures that affect health services and in particular clinical services (World Bank, 2003). There are information asymmetries between patients and health personnel, as the latter know more about the patients' diagnosis and treatment. In addition, without health education and communication, the demand for preventive services is usually low. These issues reduce the effect of the short route of accountability (see World Bank, 2003). xxii Executive Summary One mechanism to increase "client's power" is through their direct involvement in co producing and monitoring health services (World Bank, 2003). The Nigerian government has long recognized the importance of community participation in the delivery of basic health care services and has thus tried to involve the communities in the development of PHC along the lines of the Bamako Initiative. Indeed, the guidelines for the development of the PHC system establish the development of health committees to support activities at village and ward level. All these committees are involved in many needed health activities, although not necessarily in their management. The results of the facility survey show that half of all PHC facilities have or are linked to a community health development/management committee. These committees are present in two thirds of public facilities and in less than a third of privately managed ones. The majority of the members of these committees are men with exception of Lagos state where, on average, there is the same number of women and men in these committees. Most health committees meet at least once a month. In Bauchi, however, 30 percent of these committees only meet a few times a year. With exception of facilities in Bauchi, most public PHC facilities sampled in the survey worked closely with health committees that met at least monthly. However, the involvement of these committees in the management of facilities is rather limited, as most decisions are taken by either the facility head or by the LGA. This is not surprising as many of these committees were created to support health activities but did not have a strong mandate to participate in the facility's management. In particular, the community health development committees, as set up in the national guidelines, are not directly involved in the management of health facilities. The Ward Development Committees, in contrast, are supposed to oversee the functioning of the facilities in the Ward. Another mechanism to improve client's power in relation to providers is by making the provider's income depend on the demand of clients, particularly poor clients (World Bank, 2003). By paying for services, patients can exert their power to receive adequate services. If they are not satisfied with the service offered they can always go to another provider. This is what patients do in private facilities. In Nigeria, most services provided by public health facilities have fee charges. These charges, however, have not increased the power of clients, as the facilities and health personnel cannot retain these revenues and use them for improvements. These resources are sent back to the local government as they are considered part of their internally generated revenue. Possible Ways Forward There is an urgent need to clearly define the functions of each level of government and agencies within each level. Clearly defining who is responsible for what would avoid the existing gaps and overlaps. This is particularly the case for state governments. A larger participation of the state in the provision of these services, as intended in the Constitution, could improve the condition of these facilities and might decrease the fragmentation in the referral system. In particular, the state should be in charge of functions that have scale economies as is the case of the procurement of drugs and medical supplies and the training of personnel, both initial and in service training. A Executive Summary xxiii more clear division of responsibilities could also improve the accountability of policy makers in relation to communities and of providers in relation to policy makers as they will clearly know who they are answerable to. Linked to a clearer division of responsibilities, there is also a need for an institutional review of state agencies with health service delivery responsibilities. This will allow a better understanding of the structure of service delivery in each state and will provide needed information to prepare for any adjustment needed to eliminate redundancies and improve the delivery of services. Improving the performance and accountability of local governments and providers regarding service delivery often requires reforms that go beyond the health sector, in particular civil service reform. A comprehensive civil service reform that reduces the number of civil servants in the local governments and changes their skill mix will be needed. This reform is also needed to allow a more flexible and responsive mechanism to motivate and discipline frontline providers. Human resource management for health is fragmented, the LG and the LGSC or LGSB have the main responsibility, but other agencies also intervene. Performance based matching grants from the federal or state governments to local governments can be used as instrument to improve basic health service delivery. Both the federal and state level governments have shown interest in improving basic service delivery in the country. They have used different instruments to do so. The states regulate and control most of the activities of the LGs; they also deduct resources from the LGs allocation to ensure that some activities are carried out. Many of these instruments have not produced the intended benefits as the performance of services can testified. Matching grants conditional on performance can offer local governments the incentives to improve services, provided that they have flexibility and capacity to use these resources. The federal government has used this instrument to improve service delivery. The Office of the Senior Special Assistant to the President for the Millennium Development Goals has started a conditional grant mechanism intended to transfer funds to the sub national governments to improve basic service delivery and progress towards achieving the MDGs. The resources that fund this program come from debt relief. The Health Bill that is currently in the National Assembly would create a similar matching grant, the PHC Development Fund. These matching grants that the federal level is now providing and the future PHC Development Fund could be made conditional on performance, in particular, conditional on increasing the coverage of basic services, particularly population based services that are easy to monitor, such as vaccinations, pre natal and post natal care, and so forth. At the moment, the transfers from the MDG office are mainly transfers for capital projects. Similarly, the PHC Fund seems to be mainly focused on the joint financing of capital projects. These projects are needed given the large need for rehabilitation and equipment of facilities. But these resources could also be used for recurrent costs needed to improve the coverage of basic preventive services that remain low. In other words, the amounts of the transfers as well as their continuity could be conditional on performance measured in the increase in the use of services that can be easily monitored. xxiv Executive Summary For this performance based financing to be effective, providers need more autonomy in the use of resources and their remuneration should also be based on achieving results on the ground. At the moment, primary health care facilities only receive resources in kind from the different levels of government (for example, drugs and supplies). They collect some resources from fees but they cannot use these resources as they have to return them to local governments. With so little autonomy in the use of resources, it is hard to make these public providers accountable to improve service provision. By allowing facilities to retain the resources they obtain from the provision of services and by reducing the in kind financing of the facilities, they can be more responsive. For instance, if performance based transfers are used, facilities could receive funds also based on achieving a certain level of coverage. The community could offer oversight in the use of resources and can also help in monitoring results. However, for these conditional grant programs and performance based financing of providers to obtain the intended benefits, there is a need for systematic collection, analysis, and reporting of information (Bird, 2000). This information is needed to verify compliance with goals and to assist future decisions on whether or not to continue providing grants to sub national governments or providers. Information on service delivery is not just important for creating accountability from local governments to other levels of government but also to increase accountability of LGs in relation to clients. More information to the community on service delivery can increase accountability of local governments and also of providers. Monitoring the performance of government policies, through report cards can also work. These report cards have been used in different countries. In Nigeria, a scorecard assessment of rural governments in nine states, financed by the project LEEMP in 2005, was in essence a report card. Thus, publicizing broadly the results of the assessment and repeating it, could serve to monitor local governments performance. Information on service delivery is also important to increase the accountability of providers in relation to clients. Increasing information and community awareness on the services facilities provide and the resources they have to provide them and on the credentials and standard of services of providers can help. To ensure providers accountability towards the delivery of quality services, it is also necessary to ensure they do not face disincentives in their work. As describe before, often providers are paid with delays and work in difficult conditions. Providing them with the needed equipment, supplies, and in time remuneration could certainly help. Contracting out services to the private sector is also an option to explore. Contracts are difficult to monitor and enforce, in particular contracts for clinical services. However, it is possible to start by contracting out services that are easily to monitor and are highly cost effective such as social marketing of consumables (insecticide treated nets, ORS sachets, condoms) and population based services such as vaccinations, micronutrient supplementation, and so forth. Making these contracts based on performance, for instance based on achieving a pre specified coverage level would certainly align providers incentives with the achievement of these targets (see Loevinsohn, 2008). At the moment, some services in the country are contracted out to NGOs, as is the case of HIV/AIDS preventive services. As experience builds with the Executive Summary xxv design and monitoring of contracts, other services, including curative clinical services, could also be contracted. Given the difficulties involved in improving the "long route" of accountability, in the near future, improving client's power in relation to providers might have the largest results. Recent initiatives to revitalize health committees and to ensure their participation in the management of health facilities have already started to produce some effects. In Kaduna the SMOH, with the support of DFID financed Partnership for Transforming the Health System (PATHS), is implementing an initiative to build capacity in PHC committees, so that they can play a more prominent and proactive role in health and to ensure that the community voices "can be heard by health providers and the government" (Operation Manual for Health Facility Committees in Kaduna State). The Kaduna Facility Health Committee Strengthening Initiative centers the role of the Committee around the health facility, so that it can support the facility work and link it with the nearby community. PATHS has also supported similar initiatives in Ekiti and into less extent in Jigawa, Kano, and Enugu. The initiative in Kaduna is meant to increase "client's power" in relation to providers not only through the facility health committees' (FHC) participation in the management and monitoring of the facilities but also through encouraging clients complaints and redress mechanisms. The FHC in the states are encouraged to set up suggestion boxes, establish formal systems for client complaints, and undertake surveys of client satisfaction. The members of the revitalized FHC have also been trained to advocate in front of policy makers, in particular those that control the budgets, for issues affecting the performance of the PHC. Many states have started to implement programs to offer "free" services to women and children. This policy can provide an opportunity to make the income of providers depend more on the services they provide. The subsidy could be paid directly to the client through vouchers and not to the provider as has been done until now. In many urban and semi urban areas in Nigeria there are multiple providers, both public and private. By subsidizing the demand and giving patients a choice of providers, vouchers can create incentives among providers to improve service delivery. Vouchers are increasingly being used in many developing countries to improve access and quality of services; including some sub Saharan African countries such as Kenya, Tanzania, and Uganda. Finally, community insurance schemes can also increase the client's power in front of the providers. They can contribute to health care costs and increase utilization (Carin et al., 2005). These schemes buy services in bulk from the facilities, increasing thus the power of the community in relation to providers. There are already some functioning community based health insurance schemes in Nigeria, although at the moment they only cover a very small percentage of the population. CHAPTER 1 Introduction T he delivery of quality primary health care (PHC) services can have a large impact on the health of Nigerians. Many of the most cost effective health interventions to prevent and treat the major causes of mortality and morbidity in the country and progress towards the health Millennium Development Goals (MDGs) can be offered at this level of care. In addition, equity concerns draw attention to PHC as the poor in Nigeria are more likely to seek care in PHC facilities than the rich (FMOH & World Bank, 2005). The importance of primary health care in the country has long been recognized by the government. In 1975, three years before the Alma Ata conference on PHC, the Nigerian government started to put in place a PHC system in the entire country through the Basic Health Services Implementation Scheme (1975 1983). In 1992, the federal government created the National Primary Health Care Development Agency to assist states and LGAs to develop PHC. More recently in 2000, the government introduced the Ward Health Service System to ensure better community mobilization for health. One of the goals of the National Economic Empowerment and Development Strategy (NEEDS) is to improve the health status of the population as a mean to reduce poverty. To achieve this goal, NEEDS emphasizes the importance of continuing the focus on the strengthening of preventive and curative PHC services. The state governments have also recognized the importance of PHC. Accordingly, the State Economic Empowerment and Development Strategies (SEEDS) also aim at improving these services. The strengthening of basic health services has also been a major concern of donors. The World Bank and DFID Country Partnership Strategy (CPS) 2005­2009 aims at supporting the country on its progress to reach the MDGs. At the federal level, this strategy proposes analytical work to support the development of national strategies and policies for human development. In the lead states, the CPS proposes focusing on improving the availability, quality, demand, and utilization of basic health services. This is also a major concern for the Canadian International Development Agency (CIDA) in the states where it is currently supporting the health sector: Bauchi and Cross Rivers. This economic and sector work (ESW) aims to contribute to these efforts by filling some knowledge gaps. This study was jointly produced by the Federal Ministry of Health, the National Primary Health Care Development Agency, the Canadian International Development Agency, and the World Bank. More specifically, and in accordance to the CPS, the purpose of this study is three fold: (i) to contribute to the 1 2 World Bank Working Paper evidence base of the federal government's health system reform efforts; (ii) to inform the Bank's and CIDA's sector policy dialogue with the government; and (iii) to inform the current and eventual health support programs of both donors at state level. This study represents the second phase of the Nigeria Health, Nutrition, and Population Country Status Report (CSR). The first phase aimed at analyzing the health situation of the poor and how the health system was performing in terms of meeting their needs. This first phase identified PHC as the weakest chain in the entire health sector and the level of care the poor use the most. This second phase of the CSR is therefore focused on the analysis of the delivery of PHC services. In contrast to the first phase, this study is mainly based on primary data, data collected through facility, health personnel, and household surveys. This study follows a similar methodology used by a facility survey implemented in Kogi and Lagos in 2002 (Das Gupta, Gauri, and Khemani, 2003). However, this study is focused in the collection of information not previously available, such as detailed roles and responsibilities of the LGA and states and community perceptions of PHC services. As one of the purposes of this study is also to support on going or eventual health support programs of CIDA and the World Bank at the state level, the study was done on the states where CIDA is currently working, Bauchi and Cross Rivers, and in some of the World Bank lead states, Kaduna and Lagos. Objectives To better design and implement policies to improve service delivery for the poor it is necessary to generate the needed evidence and to understand the underlying relationships between the different actors involved in the delivery of health services. This study aims precisely at generating this information and helping us understand the variables affecting the performance of facilities and frontline providers. This information will help us generate policy recommendations on how to improve performance at this level. The specific objectives of this second phase of the CSR are to have a better understanding of: Performance of PHC personnel and facilities, both public and private, and the variables driving this performance. Flow of public funds to PHC facilities. Roles of the states and LGAs in the delivery of PHC. Although Primary Health Care services cover a broad range of interventions that can take place both in and outside health care facilities (see box 1.1); this study will be focused on those services that are currently being offered in PHC facilities or through community outreach done by health personnel based on these facilities. This study only looks at formal public and private PHC facilities; patent medicine vendors, traditional medicine practitioners, or pharmacies were not included in the study. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 3 Box 1.1. Declaration of Alma-Ata: International Conference on Primary Health Care, Alma-Ata, (presently Almaty, Kazakhstan) 1978 The Declaration of Alma-Ata defines Primary Health as "essential health care based on practical, scientifically sound and socially acceptable methods and technology made universally accessible to individuals and families in the community through their full participation and at a cost that the community and country can afford to maintain at every state of their development in the spirit of self-reliance and self determination. It forms an integral part both of the country's health system, of which it is the central function and main focus, and of the overall social and economic development of the community. It is the first level of contact of individuals, the family and community with the national health system bringing heath care as close as possible to where people live and work, and constitutes the first element of a continuing health care process." Following this declaration, PHC includes at least the following eight components: (i) education concerning prevailing health problems and the methods of preventing and controlling them; (ii) promotion of food supply and proper nutrition; (iii) an adequate supply of safe water and basic sanitation; (iv) maternal and child health care, including family planning; (v) immunization against the major infectious diseases; (vi) prevention and control of locally endemic diseases; (vii) appropriate treatment of common diseases and injuries; and (viii) provision of essential drugs. Conceptual Framework This study follows the World Development Report (WDR) 2004 framework on service delivery to understand the performance of PHC services in Nigeria. The framework of the WDR 2004, Making Services Work for Poor People, explains service performance through three accountability relationships: "voice" between citizens/clients and politicians/policy makers; "compact" between policy makers and providers; and "client power" between clients and providers. Clients, the patients in a PHC facility, have a relationship with providers, nurses, community health workers and others. For private providers, the user can held them accountable through their payments. If they are not satisfied with the service they can look for services somewhere else. However, for public services there is often no direct accountability of the provider to the client; this "short route" of accountability is often not working. However, there is often a "long route" of accountability by which the user can make the elected government accountable for the provision of quality services and the elected officials or "policy makers" can then influence providers to ensure that these services take place. In Nigeria there are three "long routes" of accountability as shown in figure 1.1. The three levels of government, federal, state, and LGAs, have some responsibility in the provision of health services. These three levels of government have relationships with citizens and with PHC providers, in particular the states and LGAs. If any of these accountability relationships "client power" or "short route" and "voice/compact" or "long route fails", service delivery would also fail (World Bank, 2003). For instance, if the incentives faced by policy makers are not aligned with those of communities regarding the delivery of health services, these services will not be a priority and not enough resources will be spent on them. However, even if policy makers are committed to improve PHC, if they are not able to generate a working "compact" with providers by, for instance, not assessing or rewarding performance according to whether they improve services, the delivery of services will also fail (World Bank, 2003). 4 World Bank Working Paper Figure 1.1. Accountability Relationships between Politicians/Policy Makers, Providers, and Citizens/Clients Federal Government State Government Local Government Clients/Citizens Providers Source: Adapted from WDR 2004 and World Bank 2006. Finally, "client power" or the short route of accountability is difficult to achieve in the short run, especially in the case of curative health services when there are large information asymmetries between patients and medical personnel. Nevertheless, "client power" can significantly improve service delivery if the bargaining power of the client is strengthened in relation to providers. Before analyzing the accountability relationships between policy makers, providers, and clients this study assesses the performance of PHC facilities and personnel and explains the institutional organization of the PHC sector in Nigeria. For this assessment, the study describes: (i) facilities' building conditions; (ii) access to water, sanitation, and electricity; (iii) access to an uninterrupted supply of drugs and medical supplies; (iv) access to equipment; (v) services offered; (vi) availability of qualified medical personnel; (v) characteristics of personnel; (vi) responsiveness of these services to the needs of women, men, girls and boys in the community: working hours, courtesy, waiting time for services, others. In addition, this section evaluates the differences between public and private (for profit and non for profit) providers. As shown in figure 1.1, due to the decentralization of service delivery in the country there are three different levels of government with some responsibility in the delivery of services. The study thus describes the institutional organization of PHC service delivery and the division of responsibilities across the three levels of government. The assessment of roles and responsibilities regarding PHC services is followed by an analysis of the accountability relationships that might explain services' performance. The focus of the study is on analyzing the policy maker provider and the provider client relationship. The client policy maker relationship is more difficult to analyze using surveys and changes in this relationship go beyond the health system; however, an analysis is done based on secondary data and information collected through interviews with state and local government officials in two states that participated in this study: Kaduna and Cross River. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 5 Clients policy maker relationship: The study looks at the accountability of the government and in particular local governments in relation to consumers regarding PHC services. The study examines local government revenues, public financial management, and health expenditure. Policy makers providers relationship: When policy makers have strong incentives to make PHC services work, how successful they are depends on how their commitment is "passed on" to providers (World Bank, 2003). This then depends, among other things, on how effective are the providers managed. The study evaluates the incentives faced by PHC personnel in both the public and private sector. In particular, the study looks at remuneration schemes; rewards and sanctions linked to performance in both the public and private sector; other non financial incentives faced by providers; and provider coping mechanisms when faced by inadequate incentives. Provider client relationship: To understand the reasons behind the strength or weakness of this accountability relationship, the study will assess community participation on the management and monitoring of health services: existence of functioning health committees with community participation, and availability of a functioning complaining mechanism. Methodology The study was based mainly on extensive quantitative survey work at the level of PHC facilities, health care personnel, and households in their vicinity. Three basic survey instruments for primary data collection were used: Health Facility Survey--This survey was administered to facility heads to obtain information on general facility characteristics and services provided. PHC Staff Survey--This survey of staff of health facilities included interviews of a sample of health facility staff from all facility occupations, and collected information on their general characteristics, working environment, and incentives. Household Survey--This survey of health facility clients (that is, households living near the care facilities) was used to collect data on their personal characteristics, facility usage, and satisfaction with services and care. Data collection was conducted in May 2007 in Bauchi, Cross River, and Kaduna, followed by Lagos in September. The study is then based on the Final Report submitted by the firms that implemented the surveys: EPOS Health Consultants; Canadian Society for International Health; and Center for Health Sciences Training, Research and Development (CHESTRAD). Information on the sample size determination and sampling procedures can be found in Appendix A. Table 1.1 shows the sample size accrued at the end of the study. 6 World Bank Working Paper Table 1.1. Analysis of Survey Questionnaires State LGA Health Facility Survey PHC Staff Survey Household Survey Ikara 25 25 134 Kauru 22 35 111 Kaduna Chikun 20 56 116 State 67 116 361 Calabar 26 101 135 Yakurr 25 102 137 Cross Rivers Yala 21 95 135 State 72 298 407 Bauchi 36 165 220 Ganjuwa 26 56 150 Bauchi Itas/Gadua 13 54 79 State 75 275 449 Epe 10 21 141 Ifelodun/Ajeromi 28 57 160 Lagos Surulere 48 114 99 State 86 192 400 Survey Total All States 300 881 1617 Source: EPOS, CISH, CHESTRAD, 2007. CHAPTER 2 Context W ith 140 million people, Nigeria is the ninth most populated country in the world, representing close to 20 percent of the population in sub Saharan Africa (SSA). It has a very diverse population with more than 300 ethnic groups and more than 500 languages spoken. Administratively, the country is organized as a federation with a federal government, 36 states, and 774 LGAs. The country is often sub divided in six geopolitical zones: North West, North Central, North East, South West, South South, and South East. Health Outcomes and Access to Health Services in Nigeria Each year many lives, especially children's, are lost in the country (FMOH and World Bank, 2005). One in every ten children dies before his first birthday and one in every five before his fifth. Child malnutrition rates are also very high (see table 2.1), although the percentage of children chronically malnourished has been decreasing over time. These outcomes are low not only in absolute terms but also when compared to other countries in Sub Saharan Africa. Communicable diseases, particularly malaria, pneumonia, and diarrhea, often linked with malnutrition, are the major causes of mortality and morbidity among children under five. These diseases can be prevented or treated at a very low cost, but the coverage of many of the health interventions needed to prevent and treat them is very low (FMOH and World Bank, 2005). For instance, in 2003 the Nigeria DHS showed that only 1 percent of children under five slept under an insecticide treated bed net; only 17 percent of children six month of age or younger were exclusively breastfed; only 34 percent of children under five received a vitamin A supplement; and only 13 percent of one year olds were fully immunized. There are signs of improvements in some health services, particularly in child immunization. The government has taken many measures to improve immunization and to eradicate polio. In the case of polio, 2007 was the year with the lowest polio incidence since 2002 and the lowest incidence ever of type 1 polio, the most virulent of all polio viruses. Despite these recent improvements, Nigeria is not likely to achieve the health related Millennium Development Goals. All of the interventions to address these issues are PHC interventions. Some of them can be provided by the households themselves after some orientation from a health provider, either inside health facilities or through community outreach. Maternal mortality is also thought to be high. Each year an estimated 59,000 women die from pregnancy related causes (WHO, UNICEF, UNFPA, World Bank, 2005). Some of these deaths as well as some neonatal deaths can also be prevented. However, data from the DHS 2003 shows that only 60 percent of pregnant women 7 8 World Bank Working Paper receive antenatal care, even fewer births (36 percent) are attended by skilled personnel, and access to emergency obstetric care remains limited. Although access to referral care is essential to improve maternal survival; PHC interventions can prevent some of the indirect causes of maternal deaths such as anemia, malaria, STI as well as the major factors underlying medical causes (for example, high fertility rate and low contraceptive use rate). Health outcomes and utilization of health care in the country are not only low but vary considerably across regions (table 2.1). In general, the North East and North West regions and rural areas fare considerably worse off than the rest of the country, a pattern that partly reflects regional income inequalities as the levels of poverty are higher in the north of the country. Table 2.1. Health Outcomes and Health Care Utilization across Geopolitical Zones, Nigeria 2003 Delivery in Under five % of children Total fertility BMI percent Full health mortality stunted rates < 18.5 immunization facility North Central 165 31 5.7 6.6 12.4 45.4 North East 260 37 7 23 6.0 17.1 North West 269 53 6.7 19.7 3.7 10.4 South East 103 23 4.1 8.2 44.6 84.1 South South 176 16 4.6 11.1 20.8 53.2 South West 113 23 4.1 16.7 32.5 77.6 Source: DHS 2003. There are also large income inequalities in both health outcomes and utilization of health care. As seen in figure 2.1, infant and child mortality rates among the poorest 20 percent of the population are 2.5 times higher than among the richest 20 percent. Full immunization rates among the poorest 20 percent of the population are 13 times lower than among the richest 20 percent. Similarly, the chronic malnutrition rate among the poor is about three times higher than among the rich; this is the highest rich poor difference reported in any SSA country where DHS data is available (FMOH and World Bank, 2005). Figure 2.1. Socioeconomic Disparities in Health Outcomes and Basic Service Utilization, Nigeria, 2003 80 350 90 child chronic malnutrition full immunization 70 women malnutrition 80 under-5 mortality per 1,000 300 ARI/fever treatment under-5 mortality 70 60 qualified delivery care % malnourished 250 60 50 Percent 200 50 40 150 40 30 30 100 20 20 10 50 10 0 0 0 lowest 2 3 4 highest lowest 2 3 4 highest quintile quintile Source: Federal Ministry of Health and World Bank 2005 based on DHS 2003 data. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 9 Data from the Nigerian Living Standard Survey (NLSS) 2004 also confirms the large inequalities in health service utilization. Among those ill or injured in the two weeks preceding this survey, only about three people in every five visited a health care provider. However, among the poorest 20 percent of the population, only 30 percent of people with an illness or injury visited a health provider. In contrast, 72 percent of people with an illness or injury in the richest end of the income distribution did. The NLSS data also showed that the poor are more likely to use PHC facilities for outpatient care and are less likely to use a private provider than the rich. As seen in figure 2.2, the higher the income level, the higher the proportion of people that used a hospital for outpatient consultation. As seen in the graph, the poor are more likely to use a PHC facility (that is, clinics, dispensaries, health posts, and others); while the higher the household expenditure level, the more likely the individual visits a hospital. Similarly, the use of formal private providers increases with household expenditure level. Figure 2.2. Utilization of Outpatient Care across Population Consumption Quintiles and Type of Provider or Type of Provider Ownership 0.7 0.9 0.8 0.6 0.7 0.5 0.6 0.4 0.5 0.3 0.4 0.3 0.2 0.2 0.1 0.1 0 0 Poorest II III IV Richest Poorest II III IV Richest Hospital PHC Pharmacy Other Public Private Source: Nigeria Poverty Assessment (World Bank, 2006), estimations based on NLSS 2003­04. Context in States Included in the Study The results of this study are not meant to be representative of the entire country, but of the four states sampled: Bauchi, Cross River, Kaduna, and Lagos. There are large differences across these states in terms of population, income levels, and economic activities (table 2.2). Information on health outcomes and access to health care at state level is scarce, and even though information at geopolitical zones can be indicative in some cases it can also be misleading. For instance, Kaduna is part of the North West geopolitical region, but it has the lowest poverty rate in the region. Bauchi state is a predominantly rural state located in the North East region of Nigeria. Agriculture is the main economic activity of the state. It is one of the poorest states in the country with about 77 percent of its population living under the poverty line (World Bank, forthcoming). Bauchi is also the state with the fourth largest income inequality index. Administratively, the state is subdivided into 20 LGAs. 10 World Bank Working Paper Table 2.2. Population, Poverty, and Inequality Indicators, Nigeria 2004 Geopolitical Population Poverty Poverty Poverty Gini zone State in millions head count gap severity index North East 67.6 Bauchi 4.7 77.0 33.1 14.1 43.6 South South 51.3 Cross River 2.9 55.0 23.7 9.5 40.7 North West 63.9 Kaduna 6.0 40.9 12.7 3.8 36.3 South West 43.2 Lagos 9.0 67.0 35.6 18.9 49.7 National 140.0 54.7 22.8 9 41.2 Source: population figures Census 2006, all other data from Nigeria Poverty Assessment. Cross River state is situated in the South South region. It has a population of about 2.9 million distributed into 18 LGAs. The poverty rate in the state is similar to the national average, indicating than more than half of the population lives under the poverty line. The majority of the state's population is engaged in subsistence farming. Kaduna is the third most populous state located in the North West of the country with an estimated population of 6.1 million people. In contrast to Bauchi and Cross River, a large share of Kaduna's population lives in urban and semi urban areas. As much as 20 percent of the population concentrates in two urban areas: Kaduna and Zaria. Despite the important share of the population living in both urban and semi urban areas, the main economic activity of the population remains agriculture. About two out of every five people in the state live under the poverty line; however, the poverty rate in the state is lower than the national average. Administratively, Kaduna is sub divided into 23 LGAs. Lagos state is the second most populous state in the country. It has the second most populated city in Africa after Cairo. According to the 2006 census, the state has 9 million people, the majority of them living in urban areas. The census estimated that there were close to 8 million people living in Metropolitan Lagos alone. About two thirds of the population lives in poverty and the states has the second largest inequality index in the country. Administratively, the state has 20 LGAs of which 16 are part of Metropolitan Lagos. CHAPTER 3 Status of Primary Health Care Services Organization of the Primary Health Care System A few years before the Alma Ata conference, the Nigerian government created the Basic Health Services Implementation Scheme (1975 1983) to ensure "the effective development of primary health care services in the country" (NPHCDA and FMOH, 2004). This innovative scheme introduced and developed different types of community health care personnel to staff PHC facilities. These PHC workers are unique to Nigeria. Currently, the community health workers have been streamlined into three, the Community Health Officer (CHO), Community Health Extension Worker (CHEW), and Junior Community Health Extension Worker (JCHEW). The scheme intended to build a Comprehensive Health Center, 4 Primary Health Care Centers, and 20 clinics and mobile clinics in each LGA. In December of 2000, the Ward Health Service (WHS) System was introduced to ensure that health districts coincide with political wards (serving 20,000­30,000 people). Each ward is subdivided into sections in urban areas and into groups of villages in rural areas. According to the WHS Operational Guide (NPHCDA, 2004), each section or group of villages should have a health post. Each ward should also have a Ward Health Center that should serve as first reference to the Health Posts in the same ward. The current organization of the PHC system reflects the WHS as well as previous government schemes. Public PHC facilities are then classified in the following four types: Type I: According to the 2004 PHC guidelines1 each community should have one type I facility which are health posts or dispensaries. Nevertheless, some states have progressively eliminated these facilities. Type II: A group of communities with about 2000 people should have a Primary Health Care Clinic or type II. Type III: Primary Health Care Centers or Ward Health Centers; there should be one PHC in each ward. Type IV: Comprehensive Health Centers were meant to be a referral for all PHC in the same LGA. They offer limited surgical services. In practice, many of these facilities have been upgraded and currently work as general or cottage hospitals. 11 12 World Bank Working Paper This classification does not apply to the private sector; rendering difficult the classification of these facilities. Survey Results The health facility survey sampled a total of 300 facilities; 179 (60 percent) identified as public, 118 (40 percent) private and 3 were of unknown ownership. The sample included 75 facilities in Bauchi, 72 in Cross River, 67 in Kaduna, and 86 in Lagos. Health Posts and dispensaries represented 30 percent of the sample; Health Centers, Health Clinics, and Maternities (BHC) represented 51 percent of the sample and Comprehensive Health Centers (CHC) 16 percent (table 3.1). The final 4 percent were hospitals. Government facilities made up 60 percent of the sample and private facilities the other 40 percent. The health posts were concentrated in rural and semi urban areas (78 percent). CHCs, on the other hand, were concentrated in urban areas (70 percent). Table 3.1. Health Facility Type by LGA Type Rural Urban Semi-Urban Total Health Post 31 19 35 86 Basic Health Center 29 68 56 152 Comprehensive Health Center 4 35 10 49 Hospital 2 6 3 11 Total 66 128 104 298 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Note: The classification of rural, urban, and semi urban LGA relates to government classification of the LGAs. However, rural LGAs might include urban centers, and urban LGAs rural areas. Thus, this classification is just a proxy for location of the facility. There are large variations in the organization of PHC delivery across states. As seen in table 3.2, while the majority of health facilities in Bauchi are health posts or dispensaries, there are no health posts in Lagos. In Cross River and Kaduna there is a similar distribution of health posts and health centers and clinics. In these states, about 30 percent of all facilities are health posts or dispensaries. There is also a large difference in the distribution of health facilities between public and private ownership. While the majority of the facilities in Lagos are privately owned, in the other three states, the majority of facilities are public. Most facilities can respond to urgent needs as they have staff on call 24 hours a day; however, only one in four is open 24 hours a day, seven days a week. There is a large difference across states in the number of facilities that remain open at all times. Facilities in Lagos and Cross River are more likely to open 24h a day 7 days a week. Facilities in the other two states are less likely to do so, especially in Kaduna where less than 50 percent of facilities are open 7 days a week and only 65 percent are open 24 hours a day, limiting access to health services. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 13 Table 3.2. Basic Information from All States (in %) Bauchi Cross River Kaduna Lagos Total Facility Type Health Post 59 33 26 - 29 Basic Health Center 29 42 59 73 51 Comprehensive Health Center 12 21 12 20 16 Hospitals -- 4 3 7 4 Ownership Government operated 83 78 67 21 60 Private for profit operated 16 22 28 73 37 Non-profit operated 1 -- 5 6 3 Hours of Operation open 5 days a week 24 19 50 9 24 open 7 days a week 74 74 48 86 72 open 24 hrs per day 55 82 65 87 73 24 hrs staff on call 88 87 91 88 89 Referral Refers patients 92 99 90 95 93 Ease of communication of 50 49 54 56 52 referral center with facility Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). There are major constraints in the referral system in all states. Although most of the facilities refer patients, only about half of them have easy communication with the referral center. The average walk time to referral centers is 60 minutes and the drive time 20 minutes; nevertheless, the chance of encountering difficulties with transportation is considerable, since only 31 percent of the health facilities have access to transportation to deal with emergency cases. The federal government through NPHCDA is trying to ensure the existence of a Ward Minimum Health Care package in all wards by 2012. This package includes the provision of a vehicle for referral in health facilities providing Basic Emergency Obstetric Services (BEOC). However, only 36 percent of all health centers have access to a motor vehicle and only 12 percent of health posts and dispensaries do. Infrastructure and Amenities Although there are large differences across states, in general, the infrastructure of PHC facilities is in very poor condition. As seen in table 3.3, most facilities do not have taps with running water and only about one in every four has access to safe water.2 More than three out of every five facilities do not have a toilet, waste disposal, sharp disposal, or sterilizing equipment. In addition, as many as two in every five facilities have leaky roofs and broken doors or windows. There are however significant differences across states. Facilities in Lagos fare considerably better than facilities in other states. This is partly due to the large differences in the PHC organization in Lagos. In this state most facilities are privately owned, located in urban areas, and are higher level health facilities. Facilities in Cross River fare better than those in Bauchi and Kaduna in terms of availability of equipment, but not in terms of the condition of the infrastructure. 14 World Bank Working Paper Table 3.3. Primary Health Care Facilities, Infrastructure, and Amenities across States (in %) Bauchi Cross River Kaduna Lagos Total Infrastructure Taps with running water 22 26 27 80 41 Safe water 66 70 65 91 72 Electricity 44 60 31 95 60 Amenities Lab 15 25 16 46 26 Phone 14 26 15 67 32 Waste disposal 25 56 56 96 60 Sharp disposal 39 81 69 95 72 Fridge/Icebox 68 77 47 74 67 Toilet 43 68 46 95 65 Sterilizing equipment 41 62 39 91 60 Condition Leaky roof 65 43 52 11 41 Broken doors/window 61 40 56 12 41 Cracked floor 73 44 57 16 46 Clean 86 97 66 86 85 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Services Available The ward minimum health care package (WMHCP) as defined in 2007 includes six groups of interventions: (i) control of communicable diseases (Malaria, STI/HIV/AIDS), (ii) child survival, (iii) maternal and newborn care, (iv) nutrition, (v) non communicable disease prevention, and (vi) health education and community mobilization (NPHCDA, 2007). This health care package includes priority interventions that should be provided in PHC centers on a daily basis at all times. The package refers mainly to interventions to be provided in ward health centers but it does not specifies if all or just a subset of interventions will be provided in health posts/dispensaries or in clinics. Previously, the 2004 PHC guidelines established a similar minimum package that did not include non communicable disease prevention. To determine the current availability of services, the facility survey asked the head of the facility about the services provided. The results are detailed below. Child and maternal care are the most readily available services. Child care is available in most facilities in all states. Maternal and newborn care services are not as readily available as childcare but as many as three out of every four facilities offer these services. In Cross River, Kaduna, and Lagos almost all facilities offer maternal health services; however, in Bauchi less than half of them offer these services, including many important preventive health services such as pre natal and post natal care. Family planning services are only offered in about two third of facilities. Family planning services are part of the country's PHC program as set in the Guidelines for the Development of Primary Health Care System. These services are also part of the maternal services included in the ward minimum health package. However, as seen in table 3.4, in Bauchi less than half of the facilities offer these services; in Cross River and Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 15 Kaduna about three out of every five facilities offer these services and in Lagos about 80 percent of them do. Table 3.4. Percentage of Facilities Offering Basic Services across States Bauchi Cross River Kaduna Lagos Total Antenatal 45 90 95 90 80 Postnatal 45 77 81 90 74 Family Planning 43 65 64 82 64 Childcare 85 87 86 90 87 Maternal Care 45 73 81 93 73 Adolescent/Youth 28 57 42 59 47 STI 39 58 33 67 51 Eye Care 34 15 20 14 21 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). The control of STIs is part of the ward minimum health care package; however, only half of the facilities offer this service. PHC facilities in Bauchi and Cross River are the least likely to offer these services. Equipment and Medical Supplies The facilities are also lacking some equipment needed to offer basic maternal and child services. As seen in table 3.5, there is a large shortage of basic equipment and medical supplies. Facilities were more likely to have basic medical consumables such as bandages, sterile gloves, and syringes. Similarly, most facilities have some basic equipment such as thermometers, spygnomanometers, and stethoscopes. Table 3.5. Percentage of Facilities with Equipment and Medical Supplies across States Bauchi Cross River Kaduna Lagos Total Generator 18 39 22 84 43 Refrigerator 27 39 23 95 49 Spygnomanometer 65 73 70 99 78 Child weigh scale 49 60 52 95 66 Microscope 16 30 28 51 32 Thermometer 65 90 72 93 81 Bandages 60 70 70 87 72 Sharps container 37 50 41 95 58 Stethoscope 68 78 83 95 82 Obstetric forceps 44 38 33 73 49 Vacuum extractor 19 15 21 51 28 Antiseptic for skin 44 59 50 86 61 Disposable syringes & needles 81 93 86 99 90 Sterile gloves 53 79 71 89 74 Malaria smear 15 26 20 42 27 Blood centrifuge 15 26 12 53 28 Urine test strip 29 48 32 71 46 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). 16 World Bank Working Paper However, other basic equipment and supplies such as children weigh scales, sharp containers, forceps, and antiseptics are in low or very low supply. For instance, child weigh scales which are needed to monitor child growth are only available in 66 percent of facilities. There are large differences across states in the percentage of facilities having the equipment and medical consumables needed to offer basic services. As seen in table 3.5, Bauchi is the state where facilities are the least likely to have the equipment and consumables. Facilities in Lagos, in contrast, are the most likely to have them. Not all equipment shown in table 3.5 will be needed in all facilities, for instance type I facilities are not likely to offer emergency obstetric care and will therefore not need a vacuum extractor. Nevertheless, even though only 29 percent of the facilities sampled were type I facilities; only 28 percent of facilities had a vacuum extractor and only 50 percent obstetric forceps. Vacuum extractors are part of the minimum ward health care package the government is trying to generalized; however, in the Ward Operational Guidelines of 2004, they were not included as part of the standard PHC equipment. Finally, type I facilities are not likely to offer lab test and thus are not likely to need a microscope or malaria smears. However, as seen in the table, very few facilities had microscopes (32 percent) or malaria smears (27 percent). As in the case of the vacuum extractors, microscopes and malaria smears are included in the WMHCP as defined in 2007, but they were not part of the standard PHC equipment as listed in the operational guidelines of 2004. Pharmaceuticals A large percentage of facilities do not have basic drugs, and micronutrient supplements on stock (table 3.6). For instance, even though malaria is the main cause of morbidity and mortality among children under five only 86 percent of facilities have an anti malarial drug on stock. Similarly, only 2 out of every 3 facilities have oral rehydration salts (ORS) sachets. Contraceptives are also not easily available as only about 50 percent of facilities have condoms or oral or injectable contraceptives. Indeed, only 63 percent of facilities have any type of contraceptive on stock. Similarly, micronutrient supplements for children and pregnant women are also in short supply. Only 59 percent of facilities have vitamin A supplements and only three of every four facilities have iron folate on stock. Lagos facilities in general are more likely to have pharmaceuticals on stock. Despite all efforts and considerable improvements in immunizations in the last years, maintaining vaccines on stock remains challenging. In the last few years, the Nigerian Government has increased efforts to improve children immunization rates. It has eliminated the large supply problems the country experienced in 2003 when only 14 percent of children were fully immunized (DHS 2003). These efforts have brought some success, particularly in polio immunization. However, large part of this success is due to the National and Regional Immunization days as routine immunization remains weak. Indeed, as seen in table 3.6, only about half of the facilities had vaccines on stock at the time of the interview. Partly this is due to the lack of capacity of many facilities to keep vaccines refrigerated as only 67 percent have a fridge or ice box. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 17 Table 3.6. Percentage of Facilities Having Basic Pharmaceuticals and Vaccines in Stock across States Bauchi Cross River Kaduna Lagos Total Chloroquine 77 85 67 66 73 ACT e.g. Coartem 54 69 51 82 65 Fansiddar 42 35 51 81 54 Paracetamol 82 89 81 88 85 Antibiotics 73 82 76 81 78 ORS sachets 67 69 55 69 66 Pregnancy test kit 33 46 38 71 48 Vitamin A 44 74 52 66 59 Iron/Folate 56 83 59 91 73 Condoms 29 71 35 49 47 Oral contraceptives 37 64 39 65 52 Injectable contraceptives 43 59 51 69 56 Benzyl benzoate 37 44 28 60 44 Co-trimoxazole 64 65 63 82 70 Vaccines BCG 41 44 42 72 51 Measles 45 44 40 72 52 DPT 45 45 42 72 52 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Nigeria has a large number of patent medicine vendors and thus some basic drugs can be easily found outside health facilities. However, this is not the case for vaccines which makes their lack of availability in the facilities a major challenge. All resources (drugs, supplies, and vaccines) face issues of re ordering and restocking. While facilities reported that pick up from supplier was the most frequent method of obtaining drugs and supplies, yet there were issues in receiving new supplies on time about 25 percent of the time. The reasons noted most frequently for these delays were inadequate transportation or out of stock at central store. Administrative issues in ordering or processing of requests were also reported as an issue to a lesser degree.3 Health Personnel With the exception of Lagos, most PHC services are staffed by community health workers (for example, CHOs, CHEWs, JCHEWS) and nurses/midwives. As seen in table 3.7, on average PHC facilities have about 11 people on staff, the majority nurses, midwives, and both CHEWs and JCHEWs. There are however large differences across the states. In the two states in the north, PHC facilities have about 7­8 people on staff, including attendants and security guards. In Lagos and Cross River, PHC facilities have a larger number of personnel working in the facility and are more likely to have a doctor on staff. 18 World Bank Working Paper Table 3.7. Average Staffing of PHC Facilities across States and Across Type of Ownership Bauchi Cross River Kaduna Lagos Public Private Total Total 7.8 11.8 6.9 15.4 9.5 13.0 10.9 Doctors 0.3 0.6 0.3 1.9 0.2 1.8 0.8 Community health officers 0.3 0.8 0.3 0.3 0.5 0.2 0.4 Nurse 0.8 1.1 0.7 3.1 0.5 3.1 1.6 Midwives 0.7 0.4 0.5 3.1 0.5 2.3 1.2 CHEW 0.7 1.7 1.3 0.2 1.4 0.2 0.9 JCHEW 0.8 1.3 0.4 0.2 0.9 0.3 0.7 Primary health worker 0.2 0.5 0.2 0.0 0.3 0.1 0.2 Community-based worker 0.3 0.3 0.1 0.1 0.3 0.1 0.2 Environmental health officer 0.1 0.0 0.1 0.3 0.2 0.0 0.1 Lab technician 0.2 0.2 0.2 0.6 0.2 0.7 0.4 Pharmacy technician 0.2 0.1 0.1 0.4 0.1 0.3 0.2 Medical records officer 0.1 0.3 0.3 0.5 0.2 0.4 0.3 Dental assistant 0.0 0.1 0.1 0.0 0.1 0.1 0.1 Attendant 1.7 2.6 1.1 2.9 2.2 2.0 2.1 Security guards 0.7 1.3 0.9 1.3 1.1 0.9 1.1 Other 0.7 0.6 0.2 0.4 0.6 0.3 0.5 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Note: The hospitals sampled are not included in this table. There are also large differences between public and private providers. Private providers are more likely to be staffed by nurses, midwives, and doctors than by community health workers. Most public PHC facilities are understaffed. The country is trying to provide a minimum heath care package in all wards by 2012. However, very few facilities have the proposed health manpower that will be needed to offer this package (see box 3.1). Health Post and dispensaries are the only PHC facilities that meet this proposed standard as they all have on average at least one JCHEW. In table 3.8, BHCs combine information from both health clinics and PHC centers. However, as seen in the table, on average the facilities sampled do not meet the proposed standard for clinics let alone that of health centers, as they have less than 4 JCHEWs, less than 2 CHEWs, and less than 3 nurses/midwives. Box 3.1. Proposed Health Manpower for a PHC Center to Provide the Minimum Health Care Package · Health Post: 1 Junior Community Health Extension Worker (JCHEW). · Primary Health Clinic: 2 Community Health Extension Worker (CHEW); 4 JCHEWs · Primary Health Care Center (Ward Health Center): 1 Community Health Officer (CHO); 1 Public Health Nurse; 3 CHEWs; 6 JCHEWs; 3 Nurse/midwives; 1 Medial Assistant (optional) Source: NPHCDA. 2007. Ward Minimum Health Care Package Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 19 Table 3.8. Average Staffing of Public Health Facilities across Facility Type HP BHC CHC Hospitals Total 4.9 10.6 18.7 100.9 Doctors 0.0 0.2 0.4 11.4 Community health officers 0.2 0.7 1.4 1.6 Nurse 0.1 0.5 1.2 24.0 Midwives 0.0 0.7 1.5 4.9 CHEW 0.9 1.5 2.7 0.0 JCHEW 0.7 0.9 1.7 0.0 Primary health worker 0.2 0.4 0.3 0.9 Community-based worker 0.4 0.2 0.2 0.4 Environmental health officer 0.0 0.3 0.6 0.7 Lab technician 0.0 0.1 0.4 9.0 Pharmacy technician 0.0 0.1 0.3 6.6 Medical records officer 0.1 0.2 0.4 8.1 Dental assistant 0.0 0.0 0.0 2.9 Attendant 1.1 2.6 4.3 13.4 Security guards 0.7 1.4 1.7 6.1 Other 0.2 0.7 1.5 10.9 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). On average, facilities in urban LGAs have more staff than those located in predominantly rural LGAs. As seen in table 3.9, PHC facilities located in urban LGAs, on average, have about 13 workers, almost twice as many as facilities located in rural LGAs. These facilities are more likely to have doctors, nurses, and midwives on their staff than facilities located in predominantly rural or semi urban LGAs. In contrast, facilities in rural LGAs on average are likely to be staffed by CHEWs. Table 3.9. Average Staffing of PHC Facilities across LGA Type Rural Semi-urban Urban Total 7.4 9.8 13.1 Doctors 0.2 0.7 1.3 Community health officers 0.2 0.4 0.5 Nurse 0.4 0.9 2.6 Midwives 0.4 1.4 1.6 CHEW 1.1 0.9 0.8 JCHEW 0.7 0.8 0.6 Primary health worker 0.2 0.3 0.2 Community-based worker 0.4 0.2 0.1 Environmental health officer 0.1 0.1 0.1 Lab technician 0.2 0.2 0.5 Pharmacy technician 0.1 0.2 0.3 Medical records officer 0.2 0.2 0.4 Dental assistant 0.0 0.0 0.1 Attendant 2.0 1.9 2.4 Security guards 1.0 1.1 1.0 Other 0.2 0.5 0.6 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). 20 World Bank Working Paper This difference in health personnel in rural and urban areas is partly because facilities in rural areas are more likely to be health posts or dispensaries, while facilities in urban areas are likely to be higher level facilities. However, as seen in table 3.10, when looking only at basic health care facilities (health clinics and health centers), this difference between rural and urban areas persists. Table 3.10 Average Staff in Basic Health Centers across Type of LGA Rural Semi-urban Urban Total 8.2 11.5 13.6 Doctors 0.4 1.1 1.4 Community health officers 0.2 0.4 0.5 Nurse 0.7 1.4 2.8 Midwives 0.8 1.9 1.9 CHEW 0.9 0.8 0.8 JCHEW 0.5 0.6 0.5 Primary health worker 0.3 0.2 0.1 Community-based worker 0.2 0.1 0.1 Environmental health officer 0.3 0.1 0.1 Lab technician 0.2 0.4 0.4 Pharmacy technician 0.0 0.3 0.3 Medical records officer 0.1 0.3 0.4 Dental assistant 0.0 0.1 0.1 Attendant 2.3 2.2 2.5 Security guards 1.3 1.2 1.1 Other 0.2 0.3 0.7 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Exemption and Waiver Programs Facilities in all states offer exemptions and waivers but to a limited degree. The cost of receiving health care is the main barrier to access health services in the country (FMOH and World Bank, 2005). To reduce these barriers and also to increase the utilization of services with large externalities, such as immunization, there are exemptions and waiver programs in all states. As seen in table 3.11 below, facilities in all states offer exemptions to some health services such as routine immunization, family planning, and antenatal care. Facilities in Cross River most frequently offered free services, while those in Lagos had the lowest percent of exemptions. However, these exemptions were not standard as most of them were offered less than 50 percent of the time. Concerning fee waivers for disadvantaged groups, most groups were generally asked to pay for services with the exception of clients with TB/leprosy and onchocerciasis. Lagos had the highest percent of people required to pay in all groups. This is not surprising as Lagos is the state with the largest percentage of private facilities. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 21 Table 3.11. Percentage of Facilities Offering Exemption and Waivers across States Bauchi Cross River Kaduna Lagos Services that are free Routine immunization 49 72 55 33 Family planning 18 35 28 22 Antenatal care 23 48 35 29 Other 45 41 29 90 Patients that must pay for services Disabled 57 33 52 80 TB/leprosy 21 9 40 82 Onchocerciasis 20 29 40 83 Elderly 57 44 73 79 Very poor 57 47 70 75 People under 18 yrs 59 72 75 82 Children under 5 yrs 53 69 73 80 Pregnant women 56 76 79 80 Important people 60 74 79 83 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Differences across Rural and Urban Areas and across Type of Facility There are large differences in the conditions of PHC facilities across states. As seen in the previous sections, facilities in Lagos fare considerably better than facilities in the other three states in terms of infrastructure, availability of equipment, medical supplies, and pharmaceuticals. This is partly explained by the very different organization of the PHC sector in that state when compare to the other three. Most facilities in Lagos are private facilities, are located in urban areas, and tend to be either health centers of CHC. As will be described in this and next section, facilities in urban areas, higher level facilities, and privately owned ones fare better than the rest. Table 3.12. Opening Hours across Facility Type and LGA Type (in %) Open 5 days Open 7 days Open 24 hours a day Type of facility HP 44 51 43 BHC 16 81 83 CHC 18 78 87 Location Rural 38 60 72 Urban 12 85 84 Semi-urban 30 64 60 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). 22 World Bank Working Paper Higher level facilities and facilities in urban areas are more likely to offer a larger variety of services seven days a week, 24 hours a day. Health posts offer a limited number of services and are usually staffed by a limited number of personnel. In consequence, they are not likely to be open all the time and are not likely to cover all the needs of the population. As seen in table 3.13, health post/dispensaries only offer a very limited set of services, mainly childcare. In contrast, most PHC services are provided in CHC. Similarly, BHCs (health clinics and health centers) also offer most PHC services with the sole exception of control of sexually transmitted diseases and adolescent youth care. Also, as seen in table 3.13, facilities in urban areas are likely to offer most PHC services. This is not the case in semi urban areas and particularly in rural areas. The services provided by PHC facilities depend on the level of care and thus health posts and dispensaries are the least likely to offer services outside childcare. However, even at the level of health centers and CHC not many facilities offer services outside maternal and child care and family planning. As a result, the difference in availability of PHC services in some states can then be explained by a larger proportion of HP services among their PHC facilities. Indeed as discussed before, while 59 percent of facilities in Bauchi are health post/dispensaries; there are none of these facilities in Lagos. Table 3.13. Percentage of Facilities Offering Basic Services across Type of Facility and across Type of LGAs HP BHC CHC Rural Urban Semi-urban Antenatal 52 92 92 75 87 75 Postnatal 41 86 87 63 82 70 Family planning 33 73 87 48 70 65 Childcare 85 87 91 83 88 89 Maternal care 48 81 89 67 78 72 Adolescent/youth 35 42 74 35 48 52 Sexual health & diseases 30 55 65 50 60 40 Eye care 19 18 24 17 27 15 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Facilities in urban areas as well as higher level facilities are more likely to be better equipped than other facilities. CHC which are concentrated in urban areas are more likely to be well equipped. About 90 percent of them have the equipment needed to offer child health services such as child weigh scale. More than 90 percent of them have needed medical supplies such as bandages, disposable syringes, and sterile globes. These facilities should be able to offer basic emergency obstetric care; however only 70 percent have obstetric forceps and only 48 percent have vacuum extractors. They are also not so well stock of other supplies such as antiseptics, urine test strips, and malaria smears. Not all equipment in table 3.14 will be needed in a health post or dispensary. However, basic equipment and supplies such as thermometers, antiseptics, stethoscopes, sharp container, and sterile gloves are in very short supply. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 23 Table 3.14. Percentage of Facilities with Basic Equipment across Type of Facility and Type of LGA HP BHC CHC Rural Urban Semi-urban Generator 4 55 61 23 58 36 Refrigerator 6 60 83 19 71 40 Spyghnomanometer 57 85 89 79 90 62 Child weigh scale 36 74 88 53 72 66 Microscope 6 33 59 15 46 25 Thermometer 67 85 88 74 87 77 Bandages 66 71 81 75 77 65 Sharps container 26 66 83 35 75 51 Stethoscope 67 85 94 88 89 69 Obstetric forceps 34 48 67 32 60 47 Vacuum extractor 7 30 44 11 37 28 Antiseptic for skin 38 66 77 48 79 47 Disposable syringes and needles 86 90 96 83 94 90 Sterile gloves 55 77 94 61 83 71 Malaria smear 33 212 188 10 40 20 Blood centrifuge 2 32 52 5 43 23 Urine test strip 11 55 70 18 67 38 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). The difference in availability of basic pharmaceuticals across type of facilities is much smaller than that of availability of equipment. As seen in table 3.15, the availability of certain types of pharmaceuticals is not very different across facility type. Table 3.15. Percentage of Facilities with Basic Drugs and Vaccines in Stock across Type of Facility and Type of LGA HP BHC CHC Rural Urban Semi-urban Chloroquine 75 75 69 58 74 83 ACT e.g. Coartem 53 65 80 68 66 63 Fansiddar 27 60 69 25 69 52 Paracetamol 83 84 90 75 87 90 Antibiotics 74 77 84 64 84 80 ORS sachets 63 64 73 48 72 70 Pregnancy test kit 19 56 67 17 70 41 Vitamin A 50 57 76 37 66 64 Iron/Folate 55 79 84 56 78 78 Condoms 31 45 78 35 52 47 Oral Contraceptives 38 52 76 24 61 58 Injectable contraceptives 44 57 75 27 65 63 Benzyl benzoate 30 46 53 24 56 41 Co-trimoxazole 61 73 74 57 76 70 Vaccines BCG 35 55 61 36 59 49 Measles 39 56 61 41 58 49 DPT 39 55 63 41 50 49 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). 24 World Bank Working Paper For instance, in the case of anti malarial drugs, HPs are more likely to have chloroquine than ACTs or Fansidar, while CHC are more likely to have ACTs. However, on average the percentage of facilities having any anti malaria drug on stock is not very different across type of facilities (from 85­87 percent). In general, CHC are better stocked of other pharmaceuticals but the differences are not as large. Similarly, the availability of pharmaceuticals is not very different between HP and BHC. The main different appears in pharmaceuticals and micronutrient supplements related to maternal health and family planning such as pregnancy test kit, iron/folate, and contraceptives. As shown before, HPs are less likely to offer maternal and reproductive health services than BHC and CHC; this thus explains the lower availability of these pharmaceuticals in these facilities. The difference between rural and urban areas in the availability of pharmaceuticals is large. Facilities located in rural areas are less likely to have any type of pharmaceutical, even the most common ones such as anti malaria drugs, ORS sachets, and vitamin A supplements. For instance, while 77 percent of facilities in rural areas have an anti malarial drug on stock, 87 percent and 92 percent of facilities in semi urban and urban areas respectively have. The differences in the availability of ORS sachets and vitamin A supplements are even higher. Finally, very few facilities in rural areas have contraceptives or pregnancy test kits or iron/folate supplements. The stock of vaccines in all facilities regardless of type or location is very low. CHC facilities were more likely to have vaccines. However, only about 63 percent of facilities had vaccines on stock at the time the survey was implemented. Private and Public Health Facilities Private facilities are more likely to be located in urban areas. About 56 percent of facilities sampled in urban areas were private; in contrast, only 12 percent of facilities in rural areas were privately owned as well as 37 percent in semi urban areas (table 3.16). In addition, while 43 percent of public health facilities were health posts or dispensaries, most private facilities were higher level facilities. Privately owned facilities are more likely to cover all needs of the population served. These facilities are more likely to be open seven days a week, 24 hours a day (table 3.16). They can also communicate better with the referral center and are more likely to have emergency vehicles available. They can thus respond to most needs of the community. Private facilities are also in much better condition than other facilities (figure 3.1). They are more likely to have access to safe water and sanitation, and to have a fridge/icebox and both sharp and waste disposals. The infrastructure is also in much better condition as only very few are in need of repair. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 25 Table 3.16. Basic Information on PHC Facilities across Public and Private Ownership (in %) Private Government Hours of operation Open 5 days a week 6 3 Open 7 days a week 90 72 Open 24 hours per day 91 62 24-hours staff on call 92 86 Amenities Communicate easily with referral center 57 50 Emergency vehicle available 58 15 Taps with running water 78 16 Safe water 95 57 Electricity 94 38 Lab 55 8 Phone 76 6 Waste disposal 91 39 Sharp disposal 88 60 Fridge/icebox 83 27 Toilet 91 47 Sterilizing equipment 86 43 Condition Leaky roof 15 57 Broken doors/window 15 57 Cracked floor 23 60 Clean 87 83 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Private facilities are slightly more likely to offer a larger set of services than public facilities (figure 3.1). Although public facilities are more likely to offer child health care services, private facilities are more likely to offer maternal health services such as postnatal and antenatal care as well as family planning services. Figure 3.1. Percentage of PHC Facilities Offering Basic Services across Public and Private Ownership Eye care Private Public Orthopedic Sexual health & diseases Adolescent/youth Maternal care Childcare Family planning Postnatal Antenatal 0 20 40 60 80 100 Percent Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). 26 World Bank Working Paper Private facilities are also better equipped and are more likely to have basic pharmaceuticals on stock (table 3.17). In some cases these differences are large. For instance, there is almost a 50 percentage point difference between the proportion of private and public facilities having a child weigh scale, sterile gloves, malaria smear, and urine test trips. Table 3.17. Percentage of Facilities with Basic Equipment across Public and Private Ownership Private Public Generator 81 17 Refrigerator 92 69 Spyghnomanometer 77 59 Child weigh scale 57 15 Microscope 95 72 Thermometer 97 56 Bandages 83 41 Sharps container 90 76 Stethoscope 70 35 Obstetric forceps 47 15 Vacuum extractor 92 40 Antiseptic for skin 98 85 Disposable syringes and needles 97 59 Sterile gloves 54 8 Malaria smear 59 8 Blood centrifuge 79 25 Urine test strip 82 27 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). A similar situation can be observed in the case of pharmaceuticals, private facilities are more likely to have these basic pharmaceuticals on stock than public facilities (figure 3.2). In the case of availability of vaccines, there is no difference between public and private facilities. Only about 50 percent of both public and private facilities had them on stock. Private facilities have less difficulty re ordering resources, generally having percentages of re order problems at less than half those of government facilities. Re order problems generally occurred in less than 20 percent of the private facilities compared to approximately 40 percent for government facilities. These differences were noted for drugs, supplies and to a lesser extent vaccines.4 Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 27 Figure 3.2. Percentage of Facilities Having Basic Pharmaceuticals and Vaccines on Stock across Public and Private Ownership Co-trimoxazole Benzyl benzoate Injectable contraceptives Private Oral Contraceptives Public Condoms Iron/Folate Vitamin A Pregnancy Test Kit ORS sachets Antibiotics Paracetamol Fansiddar ACT e.g. Coartem Chloroquine 0 20 40 60 80 100 Percent Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Households Satisfaction with Services To better understand whether PHC facilities respond to the needs of the communities a household survey was implemented for this study. This component of the study was designed to interview households in the same enumeration area where the facilities sampled were located with the aim of obtaining feedback on their performance. The household survey is not meant to be representative of the households in each state but of the households close to a PHC facility. In essence it was a client satisfaction tool, in addition to gathering some key information about the households. A total of 1613 households responded to the survey. The sample of households was distributed by state as follows: 28 percent in Bauchi, 25 percent in Cross River, 22 percent in Kaduna and 24 percent in Lagos. Thirty five percent of the sample was in rural LGAs, 25 percent in urban LGAs and 40 percent in semi urban LGAs. From the facility ownership perspective, 51 percent of households stated the nearest PHC was owned by the LGA, 20 percent by the state, 2 percent by the federal government and 26 percent by the private sector.5 A more detailed description of the household sample can be found in Annex B. As with the facility survey, access to core services largely varies across states (table 3.18). The sample of households was selected to ensure that the households had geographical access to services. However, not all services are offered in all facilities. In Bauchi, the percent availability was similar to the other states for child care, referrals, emergency, and general clinical services. Bauchi generally had a lower percent availability for all remaining services. Cross River, Kaduna, and Lagos had similar percent availability for antenatal, postnatal, child care, maternal care, referral and emergency services. Kaduna had lower percent availability for services such as STI 28 World Bank Working Paper control, and outreach. Lagos generally had the highest percent availability for most services, including outreach services. Table 3.18. Availability of Basic Health Services in Nearest Facility across States (in %) Bauchi Cross River Kaduna Lagos Antenatal 36 87 86 94 Postnatal 34 84 68 91 Family planning 24 65 43 84 Child care 76 83 90 94 Maternal care 36 74 83 84 Adolescent/youth care 13 57 63 65 Sexual health & diseases 11 41 17 51 Public health education 31 63 41 56 Referrals 63 56 52 75 Outreach 30 52 18 49 Eye care 24 34 13 38 General clinical 88 63 49 84 Source: Household Survey (EPOS, CSIH, CHESTRAD, 2007). Satisfaction with the services provided by PHC facilities is low in all states. Less than 50 percent of households were satisfied with the availability of drugs, equipment, medical supplies, and staff. The pattern of satisfaction across the states mirrors the availability of the equipment and supplies in health facilities across states. Households in Bauchi and Kaduna were the least satisfied, followed by households in Cross Rivers and Lagos. Satisfaction with waiting time and information provided in terms of disease control and care and information on facility management was highest in Cross River and Lagos. The pattern of satisfaction with facility staff attitude was different (table 3.19). Households in Bauchi were the most satisfied with the attitude of health care staff while those in Kaduna the least satisfied. This was, in general, the health service aspect that received the largest percentage of satisfaction. However, less than 60 percent of household heads were satisfied with the staff attitude. Table 3.19. Household Satisfaction with Nearest PHC Facility across States (in %) Bauchi Cross River Kaduna Lagos Total Drug supply 19 43 12 44 29 Availability of supplies 25 50 15 43 35 Availability of staff 23 50 19 61 37 Attitude of staff 70 51 42 62 57 Availability of equipment 11 25 7 51 21 Availability of diagnostic services 21 30 11 44 26 Information on diseases and care 21 38 22 52 31 Information on facility management 20 45 23 42 32 Waiting times 45 46 26 51 42 Source: Household Survey (EPOS, CSIH, CHESTRAD, 2007). Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 29 Although household satisfaction with private facilities is higher than with public ones, satisfaction with PHC facilities in general is low. As seen in table 3.20, household' satisfaction with both types of facilities is generally low, especially for availability of equipment, diagnostic services, and information on disease control and care. Households in general were slightly more satisfied with private providers, particularly regarding to drug and staff availability, attitude of staff and waiting time. Table 3.20. Household Satisfaction with Nearest PHC Facility across Facility Ownership and across Type of LGA (in %) Private Public Rural Urban Semi-urban Drug supply 59 23 14 40 32 Availability of supplies 47 31 25 35 43 Availability of staff 53 32 26 38 45 Attitude of staff 65 55 37 63 76 Availability of equipment 38 17 15 25 23 Availability of diagnostic services 36 23 19 31 26 Information on diseases and care 36 30 32 24 38 Information on facility management 40 30 26 30 36 Waiting times 60 38 36 58 33 Source: Household Survey (EPOS, CSIH, CHESTRAD, 2007). Reflecting the condition of the health facilities in rural and urban areas, household satisfaction with PHC facilities was lowest in rural areas. Satisfaction with different aspects of PHC facilities was lowest in rural areas with the sole exception of information on disease control and prevention. The largest differences in the level of satisfaction between rural, urban, and semi urban areas were related to drug supply and attitude of staff. The level of satisfaction among women was much higher than among men. The household survey interviewed household heads to obtain their opinion on satisfaction with the PHC facility nearest to the home. The majority of the women that answered the survey were located in either Lagos or Cross River where satisfaction with PHC facilities was in general higher. However, as table 3.21 shows, also in Lagos and Cross River female heads seem to be slightly more satisfied with the PHC facilities Table 3.21. Difference in Satisfaction with Nearest PHC Facility between Male and Female Heads of Households (in %) Total Lagos Cross River Male Female Male Female Male Female Drug supply 26 45 34 53 40 51 Availability of supplies 31 52 50 55 46 61 Availability of staff 30 62 50 70 47 61 Attitude of staff 57 58 55 67 51 54 Availability of equipment 17 39 48 53 23 33 Availability of diagnostic services 22 41 41 46 27 43 Information on diseases and care 27 50 42 60 35 44 Information on facility management 28 47 32 50 44 50 Waiting times 40 50 45 55 45 51 Source: Household Survey (EPOS, CSIH, CHESTRAD, 2007). 30 World Bank Working Paper in their localities than male heads. There were few women respondents in Kaduna; as there were no many observations, the results are not shown in the table. In any case, the pattern was similar to that in other states. The only difference was regarding satisfaction with drug supply and attitude of staff where women's satisfaction was lower in this state. Education and Promotion Activities of PHC Services There are particular weaknesses regarding the education and promotion activities of PHC facilities, particularly in the two northern states. One of the components of PHC as listed in the Alma Ata declaration is the education concerning prevailing health problems and the methods of preventing and controlling them. According to the PHC guidelines (NPHCDA, 2004), these activities are also included in the standard package of services that PHC facilities should provide. However, as seen in previous tables, only few households reported having access to both outreach and public health education activities in all states but particularly in Kaduna and Bauchi. Similarly, the level of household satisfaction with the information on disease prevention and control is also very limited. In both Bauchi and Kaduna, less than 25 percent of households were satisfied with the information received. Less than half of households were visited by a PHC provider; these visits are more likely to be done by public providers in rural areas, particularly in the two northern states (table 3.22). About 45 percent of the households were visited by a PHC provider. In Bauchi and Kaduna a larger percentage of households received a visit by a health provider, in contrast, in Lagos and Cross River less than 40 percent of households did. Most of these visits took place in rural areas and were done by public providers. Table 3.22. Percentage of Households near a PHC Facility Visited by Facility Health Personnel across States, Type of Ownership, and Type of LGA Visit Bauchi 58 Cross River 38 State Kaduna 52 Lagos 31 Rural 67 Location Urban 33 Semi-urban 40 Public 51 Ownership Private 30 Source: Household Survey (EPOS, CSIH, CHESTRAD, 2007). The majority of these visits are related to immunization followed by malaria prevention and control as can be seen in table 3.23. This is not surprising due to the large efforts of the government in the last years to improve immunization, mainly through national and regional immunization days that have focused attention in the northern states where immunization rates are low. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 31 Table 3.23. Reason for Health Facility Worker Visit across States (in %) Reason for visit Bauchi Cross River Kaduna Lagos TB dots 2 1 1 Malaria 40 16 12 10 Immunization 36 68 86 75 HIV/AIDS 1 1 2 5 Follow-up visit 21 14 9 Source: Household Survey (EPOS, CSIH, CHESTRAD, 2007). Service Charges The percentage of households that report that services are charged in the PHC facility varies largely across states (table 3.24). Bauchi and Cross River were the states where a lower percentage of households reported any charge to some basic services such as antenatal, postnatal, and maternal care. Kaduna was the state with the largest percentage of households reporting any charge for the services provided by the health facility. Even though most PHC facilities in Lagos are privately owned, a lower percentage of households reported a charge in this state than households in Kaduna with exception of general clinical services. Table 3.24. Percentage of Services with a Charge across States Bauchi Cross River Kaduna Lagos Antenatal 18 29 82 55 Postnatal 13 25 62 48 Family planning/contraception 9 11 20 25 Child care 31 39 79 56 Maternal care 13 16 73 35 Adolescent/youth care 11 27 35 29 Sexual health and diseases 4 5 7 12 Public health education 4 9 11 14 Referrals 10 7 14 18 Outreach 3 6 1 14 General clinical 24 30 33 43 Source: Household Survey (EPOS, CSIH, CHESTRAD, 2007). The households sampled were households in the vicinity of the PHC facilities sampled by the facility survey. However, households confronted with some choice do not necessarily visit their closest facility. Not surprisingly, households in urban areas were the least likely to use the nearest PHC facility. The main reasons given for not patronizing the nearest facility were the lack of equipment and the cost of the service (table 3.25). 32 World Bank Working Paper Table 3.25. Household Utilization of Nearest Health Facility across Type of LGA (in %) Rural Urban Semi-urban Households that Patronize nearest PHC facility 82 67 82 Reasons for not patronizing Not well equipped 38 29 21 No doctor 35 3 10 Service too expensive 9 39 34 Other 10 24 31 Source: Household survey (EPOS, CSIH, CHESTRAD, 2007). Notes General note: This chapter presents the results of the surveys of health facilities, health care personnel, and households in their vicinity. The chapter presents a summary of the Final Report of these surveys prepared by EPOS Health Consultants; Canadian Society for International Health; and Center for Health Sciences Training, Research and Development (CHESTRAD). 1 NPHCDA and FMOH (2004). Operational Training Manual and Guidelines for the Development of Primary Health Care System in Nigeria. 2 Safe water is defined here as water from the following sources: piped water, borehole, and protected well. 3 In interpreting this information, the reader should use caution as the responses to the questions on re ordering included a significant number of invalid or blank entries across different variables. 4 In interpreting this information, the reader should use caution as the responses to the questions on re ordering included a significant number of invalid or blank entries across different variables. 5 However, readers should keep in mind that there are some questions about the classification and coding of some health facilities in the household questionnaire, particularly in Lagos and Cross River States where PHCs were erroneously classified as hospitals either by respondents or enumerators. But since the correct classification (HP, BHC or CHC) was unknown, accurate coding could not be applied. The responses may have over represented the number of State facilities and the number of hospitals and under reported the number of LGA facilities and the number of primary care facilities, that is, Health Posts, Basic Health Centers and Comprehensive Health Centers. CHAPTER 4 Division of Responsibilities among Government Levels T he previous chapter described the state of PHC facilities in Bauchi, Cross River, Kaduna, and Lagos. Often these facilities have decaying infrastructure, do not offer all basic services, and do not have all the health personnel, equipment, medical supplies, and pharmaceuticals needed to effectively offer services. This chapter and the following ones assess the factors that explain this performance of PHC facilities following the framework of the 2004 World Development Report, Making Services Work for Poor People. The analysis is based on an evaluation of the relationships between service users, providers, and policy makers. However, as basic service delivery in Nigeria is decentralized, to understand the performance of PHC facilities is also important to understand the relationship between the different levels of government regarding health services. This chapter looks precisely at this relationship. This chapter is based not only on the results of the health facility survey commissioned for this study1 but also on secondary data from official and legal documents and on interviews with state and local government officials in two of the states that participated in the study: Kaduna and Cross River. Laws and Policies Informing the Division of Responsibilities for the Delivery of Primary Health Care The current Nigerian Constitution of 1999 makes reference to the division of health responsibilities among government levels only when establishing the functions of local government councils. The Constitution assigns the provision and maintenance of health services as a shared responsibility of the states and local governments (LGs). The National Health Policy of 1988 further defines this division of responsibilities. According to this policy, the federal government sets health policies and guidelines; monitors states and LGs health programs to ensure compliance; trains doctors; and provides tertiary and specialized health services. The state governments provide secondary health services; train nursing, midwifery and auxiliary health personnel; and assist LGs in managing PHC services. Finally, the LGs directly manage PHC services (see FMOH and WB, 2005). The Health Policy of 2004 follows the general division of health responsibilities established in the 1988 document. This policy, however, creates the State Primary Health Care Management Board to "provide technical support and supervision for the development and delivery of primary health care." These Boards are to be responsible 33 34 World Bank Working Paper for the coordination of planning, budgeting, provision and monitoring of PHC services. The Local Government Health Authority would be under the supervision of the State PHC Management Board to ensure that LGs are involved in the development and provision of health services. The successful implementation of the 2004 National Health Policy will greatly depend on the approval of the National Health Bill by the National Assembly as this Bill will provide the legal backing to this policy. The Bill delineates further the division of responsibilities across levels of government. Concerning PHC, this bill would give the responsibility of financing PHC services to the three levels of government through the creation of a National Primary Health Care Development Fund to be managed by NPHCDA. This fund would be financed with at least 2 percent of the total Federal Ministry of Health annual budget. However, for states and LGs to receive these funds they would have to contribute with 20 percent and 50 percent respectively of the total cost of the projects. This Fund will work as a conditional grant to the states and LGs, as NPHCDA would not disburse these resources if it is not satisfied with the use of the funds previously given and if states and LGs do not provide their counterpart funds. Many states have also drafted and often passed bills or regulations clarifying this division of responsibilities. For instance, some states have already established the State PHC Management Boards, often called State PHC Agency. Among the four states included in the survey, Bauchi has already created a State Agency and Kaduna has submitted a bill to the State Assembly that creates a State PHC Agency. Finally, Cross River has drafted a regulation that would give the state more responsibility in the delivery of these services. Division of Responsibilities in Practice In practice, the division of roles and responsibilities between the three levels of government and especially between the states and LGs is complex and varies across states. There is no single level or single agency in charge of financing, managing, and supervising these services; of recruiting, training, and promoting PHC personnel; of setting and paying staff salaries; building and maintaining facilities; and providing drugs and supplies. Often the three levels of government and various agencies within each level participate in these activities, creating duplication and gaps in provision. To illustrate this complexity, the diagrams below show all agencies affecting the delivery of PHC services in two of the states that participated in the survey: Kaduna and Cross River. The responsibilities of the agencies that appear in the diagrams will be detailed in the next paragraphs. Policies and Guidelines At the federal government, the Federal Ministry of Health (FMOH) and its parastatal agency, the National Primary Health Care Development Agency (NPHCDA), establish policies and guidelines regarding the provision of PHC services. For instance, both agencies designed guidelines for the development of the PHC system in Nigeria. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 35 NPHCDA designed an operational guide for the Ward Health System and it also established a ward minimum health care package. The FMOH has also drafted guidelines regarding specific health programs such as malaria, reproductive health, and others. Figure 4.1. Government Agencies with Responsibilities in PHC in Cross River NYSC NPHCDA FMOH Zonal SMOH SMLGA SRDC PHCDA DPHC LGSC PHC Facility PHC Department PHC LGA Service Desk Source: Authors. Figure 4.2. Government Agencies with PHC Delivery Responsibilities in Kaduna FMOH NPHCDA Zonal SMOH SMOLG PHCDA SACI DPHC DPHC PHC Department LGA PHC Facility LGSB Development Areas Source: Authors. 36 World Bank Working Paper The states can also generate policies and guidelines that affect basic health services. For instance, many states have implemented a policy to provide free maternal and child health care services. Among the states that participated in the survey, Kaduna has already started to pilot this policy. At the moment, all public hospitals offer these services but only a small set of PHC facilities do. In Cross River, the new Governor has also announced a similar policy; its implementation is supposed to start in the coming weeks. In addition and as mentioned before, some states have created or intend to create PHC agencies, like Bauchi and Kaduna. Cross River regulation to increase the responsibility of the state in the management of PHC is pending approval from the state executive. Some states have also generated administrative guidelines that affect the provision of PHC. For instance, the state ministries of local government (SMLG) often generate guidelines concerning local government budget planning and preparation. These guidelines can be very detailed. For instance, in Kaduna the SMLG for the preparation of the 2008 budget sent the LGs a list of "recommended/approved" health areas for inclusion in the budget. This list also includes "recommended/approved" minimum expenditure for some of these areas. Finally, the local government service board or commission (LGSB or LGSC) also sets guidelines regarding management of LG personnel, including PHC workers. Personnel Training The initial training of doctors and other university level health personnel is a responsibility of the federal government through the Federal Ministry of Education. The initial training of all other PHC staff, including nurses, midwives, CHEWs, and JCHEWs is a responsibility of the states that run schools of nursing and colleges of health technology where these personnel are trained. Many agencies participate in in service training of PHC personnel. NPHCDA offers training of PHC personnel, but also do the state ministries of health and the local government service board or commission. Indeed, a percentage of the Federation Account allocation going to the LGs is managed by the LGSB for the training of LG staff, including PHC. Health Care Personnel Management The management of PHC personnel is shared between the LGs and the local government service board or commission. The LGSBs are the agencies with the main responsibility regarding personnel management. They are in charge of hiring, promoting, transferring and firing LGA personnel. They delegate part of this authority to the LGs for personnel grade 1­6. For more senior level personnel, the LGSB retains all this responsibility. In the case of PHC, nurses, midwives, and doctors are grade level 7 and above; thus the LGs do not have complete control over them. The LGs, through their Departments of Primary Health Care, can initiate procedures for hiring, firing, and disciplining health providers, but the final decision is taken by the LGSB. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 37 Remuneration The LGs finance the salary of all PHC personnel. However, in the particular case of Cross River, the SMLG manages the payroll. The state deducts from the joint LG account, where the Federation Account allocation is deposited, the salary of all LG staff and pays them directly. This mechanism was implemented after many complaints from LG staff for non payment of salaries by the LGs. Finally, the National Youth Service Corps (NYSC), a federal agency, provides a one year service of recently graduated university students, including doctors and other health professionals. Cross River has used this program to ensure the presence of doctors in PHC facilities. The number of doctors and other health professionals varies from year to year, but it is usually small. There is less than one doctor per LGA; limiting thus the impact of this program. These health professionals are meant to give services in all the facilities in the LGA. They are usually based in either the LG headquarters or in the largest PHC facility. Infrastructure: Construction and Maintenance The results from the facility survey show that the construction of PHC facilities is done primarily by local governments (table 4.1). There is large variability in the role of LGs across the states, partly reflecting the presence of the private sector in the provision of these services and community participation. For instance, in Bauchi, Lagos, and Kaduna the percentage of facilities built by communities/individuals reflects the percentage of privately owned facilities. In contrast, in Cross River almost half of the buildings are provided by communities or individuals while only 22 percent of facilities are privately owned, reflecting a large degree of participation of the local communities in the construction of these facilities. Table 4.1. Level of Government or Agency that Provided the Health Facility Building (in %) Bauchi Cross River Kaduna Lagos Total Federal government 3 1 3 0 2 State government 1 8 12 7 7 LGA 77 39 45 14 43 Development partner 3 3 13 0 4 Community/individual 15 46 25 76 42 Faith-based organization 4 7 6 2 4 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). LGs are also the main provider of resources for the maintenance of buildings and equipment with exception of Lagos (table 4.2). Cross River has an important participation of the community in the maintenance of buildings and equipment but it is also the state where a larger percentage of facilities reported that this maintenance was not done. 38 World Bank Working Paper Table 4.2. Main Agency Responsibility for the Maintenance of Equipment and Buildings across States Maintenance equipment Bauchi Cross River Kaduna Lagos Total Facility funds 12.2 22.06 25.9 77.65 36.8 Federal government 4.1 7.35 3.5 3.5 State government 1.4 8.82 3.5 1.18 3.5 LGA/PHCMC 70.3 22.06 55.2 17.65 40.0 NGO/donor 2.7 1.18 1.1 Community 1.4 10.29 2.8 Individual 5.88 5.2 2.8 Staff 2.7 7.35 5.2 1.18 3.5 Not done 5.4 16.2 1.7 1.2 5.96 Maintenance of buildings Bauchi Cross River Kaduna Lagos Total Facility funds 12.2 19.12 23.7 23.73 35.7 Federal government 2.7 5.88 3.4 3.39 2.8 State government 1.4 5.88 6.8 6.78 3.5 LGA/PHCMC 73.0 25 57.6 57.63 42.3 NGO/donor 1.4 2.94 1.7 1.4 Community 2.7 20.59 3.4 1.69 5.9 Individual 10.29 3.4 3.39 3.5 Staff 1.4 0.4 Not done 5.4 10.3 3.4 4.55 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Due to the poor condition of PHC both the federal government and the states have increased their participation. NPHCDA has financed the construction of model PHC facilities in each ward. In addition, the states are increasingly participating in the construction and rehabilitation of facilities. For instance, Cross River applied and obtained funds in 2007 from the Office of the Special Assistant to the President for the MDGs. These funds were used for a major PHC rehabilitation program. A total of 130 PHC facilities were renovated (the work is still on going) and equipped, the facilities were provided with water, solar panels, and cold chain equipment. The state provided an even higher amount of funds than those obtained from the MDG office for this program. Cross River also applied this year for funds to continue the rehabilitation of PHC. This effort took place after the facility survey was implemented and thus is not reflected in the results. Indeed, the survey showed that Cross River was the state with the largest percentage of facilities where the maintenance of equipment (16 percent) and buildings (10 percent) was not done. In Kaduna, the Development Areas (DAs) also participate in the construction and maintenance of PHC facilities. The DAs were created by a state law in 2004 as a subdivision of the LGA. There are 46 DAs in the state. According to the DA operational guidelines, drafted by the Ministry of Local Governments, they have as functions the Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 39 provision and maintenance of dispensaries and clinics. They also provide some equipment and furniture. Lagos also has development areas with similar responsibilities. Procurement and Distribution of Pharmaceutical Products All levels of government participate in the procurement and distribution of drugs and medical supplies. At the Federal level, the FMOH procures and distributes to the states pharmaceuticals linked to some health programs such ITNs and ACTs from the Roll Back Malaria program, and contraceptives from the Reproductive Health program. NPHCDA also participates in the procurement and distribution of drugs by providing seed stocks for drug revolving funds in the model PHC facilities. The states also participate in the procurement and distribution of pharmaceutical products and consumables to the PHC facilities. For instance, in Cross River, to ensure the procurement of all pharmaceutical products in the essential drug list, the SMOH formed a partnership with a private firm, Worldwide Health Care Limited to ensure the functioning of a drug revolving fund. This program is in charge of procuring all drugs in the essential drug list in the state, facilities and LGs can only get pharmaceutical products from this agency which is part of the Essential Drug Program. This arrangement ensures that all drugs provided in health facilities are approved by NAFDAC, the Nigerian Agency for Food and Drug Administration and Control. This is also a more efficient way of procuring drugs by taking advantage of economies of scale. This program also manages the distribution of other drugs provided free of charge. For instance, it manages the distribution of ACTs and ITNs from the Roll Back Malaria program. In Kaduna, the state procures some of the drugs that are then distributed to the facilities; for instance, it has procured drugs for the free Maternal and Child health program that functions in some of the PHC facilities. To simplify the procurement process and to increase transparency and efficiency, the SMOH has also introduced a bill in the state assembly creating an autonomous agency, the Drug Management Agency to manage all procurements in the state. The LGs mainly store and distribute pharmaceutical products to the facilities. However, in some instances they have also procured drugs and medical consumables, even vaccines. For instance, in the Bauchi State SEEDS report (BASEEDS) an example is given of a local government council that procured large quantities of vitamin B that "would last for 20 years but only have a shelf life of only 4 years." In Kaduna, some LGs have also procured drugs and vaccines. Vaccines At the federal level, NPHCDA is the agency in charge of ensuring the availability of vaccines at national level. The international procurement of vaccines is done by UNICEF. All levels of government intervene in their distribution. The agencies involved are NPHCDA, the state MOHs, and the LGs PHC Departments. In Kaduna, an inter ministerial agency, the State Action Committee for Immunization (SACI), also participates in the distribution of vaccines. The results of the facility survey show LGs are the main supplier of drugs, consumables, and equipment to PHC facilities; followed closely by the facility's own funds (table 4.3). The presence of drug revolving funds in all states gives the facility's 40 World Bank Working Paper own funds a significant role in the provision of drugs and supplies. There are significant variations across the states in the roles of LGA, states, and facility's own funds. In Bauchi and Kaduna the LGs are the main provider of equipment, drugs, and supplies followed by the facility's own funds. However, as explained above, often the LGs also distribute drugs and medical supplies procured by other levels of government. The facility workers have a significant role in the provision of pharmaceuticals in Bauchi. In Lagos, due to the large share of facilities that are privately owned, the facility's funds represent the main source of procurement of pharmaceuticals and equipment. Cross River state follows a different pattern to that of the other three states. The results of the survey show that in this state LGs are the main suppliers to facilities but the state government plays a larger role than facility funds. However, as in this state the majority of the drugs available at the facility level are provided by the essential drug program revolving fund, it is likely that responses indicating that the state and LGs as main providers were referring to this program. Table 4.3. Main Supplier of Medical Consumables, Drugs, and Equipment to PHC Facilities across States Medical supplies Bauchi Cross River Kaduna Lagos Total Facility funds 11.1 20 25.9 79.76 36.6 Federal government 6.9 8.57 3.5 - 4.6 State government 1.4 21.43 6.9 2.38 7.8 LGA/PHCMC 68.1 34.29 43.1 13.1 38.4 NGO/donor - 1.43 1.7 2.38 1.4 Community - 4.29 - - 1.1 Individual 1.4 2.86 6.9 1.19 2.8 Staff 5.6 2.86 6.9 - 3.5 Not done 5.6 4.3 5.2 - 3.52 Drugs Bauchi Cross River Kaduna Lagos Total Facility funds 15.3 26.76 33.3 78.82 40.6 Federal government 4.2 5.63 3.3 3.1 State government 22.54 5.0 11.76 10.1 LGA/PHCMC 47.2 38.03 45.0 7.06 32.6 Individual 2.8 4.23 3.3 1.18 2.8 Staff 25.0 6.7 7.6 Not done 5.6 2.8 3.3 2.78 Equipment Bauchi Cross River Kaduna Lagos Total Facility funds 12.2 17.39 24.1 76.47 35.0 Federal government 6.8 7.25 3.5 2.35 4.2 State government 1.4 15.94 6.9 16.47 6.3 LGA/PHCMC 73.0 30.43 56.9 2.35 42.7 NGO/donor 5.8 1.7 2.5 Community 10.14 2.5 Individual 4.35 3.5 1.18 2.1 Staff 1.4 1.45 1.7 1.1 Not done 5.4 7.3 1.7 1.2 3.85 Source: Health Facility Survey (EPOS, CSIH, CHESTRAD, 2007). Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 41 Supervision As in other activities related to PHC, all levels of government participate in the supervision of PHC activities. NPHCDA through its zonal offices supervises and supports the PHC Department in the LGs as well as health facilities. The SMOH also supervises health programs. Finally, the LGs supervise health personnel. Possible Ways Forward Given the current situation there is an urgent need to clearly define the functions of each level of government and agencies within each level. Clearly defining who is responsible for what would avoid the existing gaps and overlaps. This is particularly the case for state governments. A larger participation of the state in the provision of these services, as intended in the Constitution, could improve the condition of these facilities and might decrease the fragmentation in the referral system. In particular, the state should be in charge of functions that have scale economies as is the case of the procurement of drugs and medical supplies and the training of personnel, both initial and in service training. Due to the poor condition of PHC services, the federal and state governments are increasingly participating in the delivery of these services. However, often these efforts have been fragmented and not well coordinated. The efforts to create PHC Management Boards or PHC Development Agencies at the state level intend to unify in one agency all activities linked to the management of PHC. For instance, in Kaduna the Bill that would create the PHC agency could offer great advantages if this agency concentrates the responsibilities currently shared by the SMOH, SMLG, SACI, and LGSB. This is the intention of the bill. However, if the new agency does not completely substitute all other state agencies but exists parallel to them, the new bill will not be able to improve the situation and will add more confusion and duplication. In addition, if this new agency is created, there will be a further need to clarify what would be the roles of the LGs and DAs in PHC. There will also be a need to clarify the relationship with NPHCDA. Finally, the creation of the agency would still not solve the fragmentation in the referral system as the links between hospitals and PHC facilities are not discussed in the bill. In Cross River, due to the poor performance of basic services in 2004, the state enacted a law by which the salaries of all LG civil servants were to be deducted from each local government Joint Account. In addition to salaries, this law also stipulated the following deductions: a 2.5 percent of the gross sum in the account is deducted for rural water supply and electrification; a 2.5 percent for a primary school rehabilitation program; and a 2.5 percent for PHC facility rehabilitation program. These resources were then centralized. Different state agencies needed to apply for these resources to implement the rehabilitation programs. This process proved complex for the sectoral ministries to obtain these resources. The Law was revised at the end of 2007; the revised law stipulates among other things a deduction of 9 percent of the LGs Joint Account for the State Rural Development Commission (SRDC). This Commission will be in charge of the primary school and PHC facility rehabilitation program. This Commission exists in parallel to all other agencies in the state that have some 42 World Bank Working Paper responsibilities regarding PHC service delivery (see figure 4.1). It is not clear what would be the division of responsibilities between this commission and the SMOH. Under these circumstances, there is also a need for an institutional review of state agencies with health service delivery responsibilities. This will allow a better understanding of the organization of service delivery in each state and will provide needed information to prepare for any adjustment needed to eliminate redundancies and improve service delivery. Bauchi has already started to do this institutional review with the support of the Canadian International Development Agency. Kaduna has done an institutional review of the State Ministry of Health with the support of DFID financed PATHS program which is a first step for an overall institutional review of the health service delivery architecture of the state. Notes 1 General note: The results of the facility survey discussed in this chapter come from the Final Report for this survey prepared by EPOS Health Consultants, Canadian Society for International Health (CSIH), and Center for Health Sciences Training, Research and Development (CHESTRAD). CHAPTER 5 Clients Policy Makers T he representatives of all levels of government in Nigeria, as in other democracies, are elected; the president, governors and local government chairmen as well as the representatives to the national, state, and local assemblies are elected. Through this election process the population could hold politicians accountable for the quality of basic services the government provides. However, this relationship does not always work. This chapter aims at looking at this accountability relationship between clients/citizens and policy makers concerning the delivery of PHC services. This chapter focuses the analysis on local governments as the main level of government in charge of managing PHC services. The chapter draws from different reports including the Nigeria Public Expenditure Management and Financial Accountability Review (PEMFAR, 2007); a study on State and Local Governance in Nigeria (2002); a report on a Scorecard Assessment of Rural LGs in nine states financed by the Local Empowerment and Environmental Management Project (LEEMP); and reports from the Auditor General of LGs of Kaduna (1999­2004) and Cross River (2005­2006). Finally, this chapter also draws from interviews that took place on April of 2008 with state and local government officials from two states that participated in the survey: Kaduna and Cross River. Accountability can be understood as having the obligation to answer questions regarding decisions and actions (Brinkerhoff, 2004). But for an agency or a level of government to be accountable for delivering services they also need to have the capacity to provide them, in other words, the financial and human resources needed to provide these services. This chapter first assesses the capacity of local governments to offer health services by examining local government revenues, public financial management, and health expenditure. Then the chapter looks more closely at the accountability relationship between local governments and clients and also between local governments and other levels of government. Local Government Revenues and Responsibilities Allocations from the Federation Account (FA) represent the largest share of local government revenues. The revenues from oil and gas are centralized in the FA. In the last years, these revenues have represented more than 80 percent of the total revenues of the consolidated government. In contrasts, internally generated revenues (IGR) from both states and LGAs have represented less than 5 percent of the total revenues. The remaining 5­10 percent comes from other revenues collected by the federal government such as value added taxes (World Bank, 2007). 43 44 World Bank Working Paper The FA revenue is distributed across the three levels of government following a predetermined and transparent allocation formula. This formula has changed considerably since 1999 in benefit of the sub national governments, who have seen their allocation increased considerably in the last years (World Bank, 2007). As seen in table 5.1, the share of LGAs increased from about 12 percent before 2000 to about 18 percent in 2005. Table 5.1. Changes in the Actual Distribution of Federation Account Revenues across Three Government Levels (in %) 1999 2000 2001 2002 2003 2004 2005 Federal 68.7 68.4 49.2 47.6 49.4 46.3 45.9 State, including FCT 19.7 19.8 31.6 31.9 31.6 35.3 35.8 Local 11.7 11.8 19.2 20.5 19.0 18.4 18.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: World Bank PEMFAR, 2007. The distribution across states and LGAs of this revenue is however very unequal. The 1999 Constitution reintroduced the mineral derivation rule by which 13 percent of all oil and gas revenues are deducted at source and distributed among the states where the resources are extracted. There are nine oil producing states in Nigeria; however, most production is concentrated in just four states, Akwa Ibom, Bayelsa, Delta and Rivers. These four states receive about 90 percent of all derivation oil payments or about US$2 billion in 2005, about 40 percent of the total funding available for FA distribution to all 36 states (World Bank, 2007). The LGAs also receive 10 percent of the state IGR revenues and they also collect own revenues, although their capacity to generate revenues is limited. For many years there has been a debate on whether local governments receive enough resources to meet their responsibilities.1 During the last military regime after many complaints for non payment of primary school teachers' salaries, the federal government started to deduct the salary of teachers from the LGAs FA allocation. Many LGs complained that this deduction at source created such a large reduction of their total revenues that they were left with a "zero allocation" to fulfill their other responsibilities (World Bank, 2002). However, the local government revenues have increased considerably in the last years. As shown table 5.1, the LGs' share of the Federation Account has increased significantly since 1999. In addition, the total consolidated revenues of the entire government have also increased considerably thanks to the increasing oil prices (see table 5.2). Finally, government expenditure has also increased, especially local government expenditure, which has experienced an increase of more than 400 percent since 1999. This increase is much higher than that experienced by the state and federal expenditure. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 45 Table 5.2. Main Fiscal Trends for the Consolidated Government, 1999­2005, (billions of naira) 1999 2000 2001 2002 2003 2004 2005 Consolidated government revenues gross 1,011.0 1,985.0 2,247.0 2,064.0 2,795.0 4,126.0 5,642.0 as % of GDP 29.4 42.4 42.1 36.6 37.1 43.1 43.4 Consolidated government revenues, US$ bn 11.0 19.4 20.1 16.9 21.4 30.9 42.4 Consolidated government expenditure 990.0 1,642.0 2,386.0 2,337.0 2,776.0 3,230.0 ,4274.0 as % of GDP 28.8 35.1 44.7 41.5 36.8 33.7 32.9 Subnational (state + local) expenditure 228.0 505.0 768.0 895.0 1,190.0 1,523.0 1,962.0 o/w local government expenditure 60.0 145.0 171.0 170.0 269.0 423.0 602.0 as % of GDP 6.6 10.8 14.4 15.9 15.8 15.9 15.1 Source: World Bank PEMFAR, 2007. Nevertheless, LGs face many limitations in the use of their revenues. Some of these limitations are statutory, such as deductions at source; others are administrative, such as limitations to their autonomy in drafting and executing their budget or in personnel management (World Bank, 2001). For instance, in most states, LGs need clearances from the state governments to spend resources above a threshold or to obtain a loan. These limitations vary from state to state. For instance, in Cross River and according to the state local government Law of 2007, the state government deducts from the LGAs' joint accounts (where the FA is deposited): (a) the salaries of entire staff in the local government Service; (b) 9 percent for the State Rural Development Commission; (c) 5 percent to the state electrification agency; (d) 2 percent to the State Joint Security Operations Fund; (e) 2 percent to the Ministry of Local Government; (f) 1 percent to the LGSC for staff training; (g) 2.5 percent to the state Joint Social Welfare Service; (h) 1 percent to the border Communities Development Fund; (i) 1 percent for sports development; and (j) 2.5 percent for environmental management and protection. Other states also withhold part of the LGs' FA funds for different purposes. In Kaduna, the Ministry of Local Government Affairs withholds part of LGA allocation to finance joint programs. In 2003, the Report of the Auditor General of LGs mentions that deductions were made at source for some of these projects without taking into consideration the needs of each LGA; for instance, resources were deducted in urban LGAs for agricultural projects. In addition, in this state, the Ministry of Local Government provides very detailed guidelines for the preparation of the LG budgets. In the case of PHC, a call circular for the preparation of the 2008 budget was sent to LGs listing the "recommended/approved" areas for inclusion, as well as some recommended minimum expenditures in particular areas. Finally, the LG budgets need to be approved by the state and any expenditure done by the LGA outside those related to salaries and overheads needs clearance from the state. During the last military government, a 5 percent of the LGs allocations were deducted to support traditional rulers. This requirement was suspended; however, most LGs continue the support to traditional rulers who are part of their payroll. 46 World Bank Working Paper Public Financial Management Limitations to local government's autonomy and the little revenues they received in the past do not fully explain the LGs service delivery record. For instance, public expenditure management in LGs is weak: budgets are unrealistic, record keeping is poor, and irregularities in the use of funds are common. The PEMFAR 2007 evaluated public financial management (PFM) practices in the federal and state governments, reporting weak PFM systems in the states. Between 2002 and 2004, on average three of the four states included in the PHC survey, Kaduna, Cross River, and Bauchi, had a 31 percent difference between their consolidated budget and their consolidated expenditure. The situation is similar at LG level where budgets largely differ from actual expenditure. As seen in table 5.3, in Kaduna, between 2003 and 2005 the execution rate of the LGs budgets on average was between 73 percent and 93 percent. In contrast, in Cross River, budget execution in 2005 and 2006 was higher than actual budget. Table 5.3. Budget Execution Rate across LG in Kaduna and Cross River (in %) LG 2003 2004 2005 LG 2005 2006 Birnin Gwari 27 Abi 111 129 Chikun 66 112 Akamkpa 109 81 Giwa 75 109 Akpabuyo Igabi 79 67 97 Bakassi 98 Ikara 76 84 Bekwara 120 104 Jaba 98 59 72 Biase 100 104 Jema'a 58 48 Boki 88 92 Kachia 109 65 100 Calabar Municipal 110 106 Kaduna North 188 62 Calabar South 100 Kaduna South 44 64 83 Etung 123 137 Kagarko 84 Ikom 122 123 Kajuru 85 Obanliku 92 125 Kaura 100 91 Obubra 91 100 Kauru 62 91 93 Obudu 134 118 Kubau 74 95 Odukpani 61 88 Kudan 140 116 92 Ogoja 107 136 Lere 102 92 Yakurr 91 89 Makarfi 136 Yala 90 116 Sanga 96 Cross River 103 109 Soba 83 Zangon-Kataf 114 45 94 Zaria 91 Zabon-Gari 63 83 Kaduna 91 73 93 Source: Authors estimates based on data from Kaduna and Cross River Auditor General of Local Governments Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 47 In 2005, a scorecard assessment2 of rural local governments in nine Nigerian states reported information on the budget execution rates of Bauchi's local governments. In the seven local governments for which information is available, on average, the execution rate was about 95 percent. There is no complete information on why there are such large discrepancies between the LGs budgets and expenditures, but some of the causes are likely to be similar to those found at state level, such as low capacity to project future revenues (see PEMFAR, 2007). Indeed, data from the Auditor General of Local Governments in Kaduna shows large discrepancies between estimated and actual revenues in the state LGs. Similarly, the scorecard assessment of LGs in Bauchi showed an average ratio between actual and projected internally generated revenues of these LGs of about 57 percent (see figure 5.1). Figure 5.1. Average Ratio between Actual and Projected Internally Generated Revenues in Bauchi's Local Governments 120 100 80 Percent 60 40 20 0 Ba ri Bo hi D oro m am o am a a m s Ka are m rfi au gi Ta a a ro ji ki ar Ita le az aw w ir w Za uc in ba gu Ki To is Sh ju fa W ka g a' N ar M am ta Al D Ja G G Source: Scorecard Assessment of Rural local governments in nine states of Nigeria Volume II (Terfa Inc., 2005) Most of the variation between budgets and actual expenditure is related to capital budgets, indicating very little attention to the financial management of projects. As seen in table 5.4, while in Kaduna LGs the execution rate of projects is in general very low, in Cross River LGs the actual expenditure is often many times higher than the budget. Limited information also highlights general weaknesses in other aspects of public financial management. A study on state and local governance in Nigeria (World Bank, 2002), recorded poor financial management in 13 LGs from the six states included in the study.3 In particular, the study described LGs budgets as just a list of needs, not an instrument to prioritize expenditures based on clear policies and procedures to identify clear goals. This study also indicated weak financial management capacity and also "willful" disregards for public financial management rules in some LGs. 48 World Bank Working Paper Table 5.4. Capital Budget Execution Rate across LG in Kaduna and Cross River (in %) LG 2003 2004 2005 LG 2005 2006 Birnin Gwari 25 Abi 209 269 Chikun 50 44 Akamkpa 101 76 Giwa 28 44 63 Akpabuyo Igabi 38 38 80 Bakassi 84 Ikara 40 68 Bekwara 68 111 Jaba 15 58 40 Biase 88 202 Jema'a 6 Boki 56 138 Kachia 23 100 Calabar Municipal 111 146 Kaduna North 40 68 Calabar South 129 Kaduna South 6 47 49 Etung 122 127 Kagarko 50 Ikom 126 154 Kajuru 12 64 Obanliku 115 209 Kaura 15 58 Obubra 71 104 Kauru 60 80 Obudu 162 398 Kubau 37 74 Odukpani 61 88 Kudan 614 44 82 Ogoja 124 215 Lere 35 80 Yakurr 98 81 Makarfi 80 Yala 108 131 Sanga 21 78 Cross River 107 161 Soba 84 56 Zangon-Kataf 21 25 85 Zaria 64 54 Zabon-Gari 45 56 Kaduna 77 40 69 Source: Authors estimates based on data from Kaduna and Cross River Auditor General of Local Governments Among the states that participated in the PHC study, the report of the Kaduna Auditor General of Local Governments can be indicative of weaknesses in public financial management in the LGs (see box 5.1). The situation in Lagos, as highlighted in the Lagos State Financial Accountability Assessment (World Bank, 2004), is similar. This assessment noted inadequate record keeping, poor supervision of revenue collection, unanswered audit queries, and long delays in audit reports. Finally, the scorecard assessment of rural local governments in nine states (Terfa Inc., 2005) looked at financial integrity of LGs through an index that considered the following aspects: recent audit reports, compliance with procurement procedures, appropriateness of borrowing, payment of advances to political and career staff, and fulfillment of reporting requirements. Local governments in Bauchi state had on average the highest score on overall financial integrity among the nine participating states. Nevertheless, in a scale from 1­100, on average, Bauchi's LGs score was only 56. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 49 Box 5.1. Extract from Report of the Auditor-General for Local Governments on the Accounts of the 23 Local Government Councils of Kaduna State. For the Year Ended 31st December, 2004. "...It is evident that the local governments are yet to make any meaningful departure from the past as far as record keeping is concerned. Problems such as missing payment vouchers, unvouched expenditures..., investing in dead or non-performing companies, non-remittance of third-party deposits among others, still persist." Note: This report is the report the Auditor General presents to the Kaduna State House of Assembly. Local Government Civil Service The size as well as the composition of local government civil services can also explain the weak performance of health services. In 2005, in Cross River's LGAs on average about 53 percent of the total expenditure went to personnel remuneration. In Kaduna, on average, about 23 percent of the LGs expenditure went to personnel costs. However, this average hides large differences across Kaduna's LGs; for instance, at least four LGs in the state, out of 23, spent more than 40 percent of their expenditure on personnel. Although there is no data available on the percentage of the personnel remuneration out of total expenditure in other sub national governments, information on national wage bills in countries in the region can be indicative of the problem. As seen in table 5.5 below, only Kenya has a larger wage bill than Cross River's LGs. Table 5.5. Wage Bill in Different Sub-Saharan Africa Countries, 2005 Country Compensation of employees (% of expense) Benin 43 Burkina Faso 41 Cote d'Ivoire 39 Kenya 60 Lesotho 37 Madagascar 41 Mali 33 Mauritius 39 Seychelles 37 South Africa 14 Uganda 13 Zambia 36 Source: World Bank Development Data Platform. This large wage bill is partly due to an overstaffed civil service. As explained before, Cross River state deducts a significant part of the LGs allocations for different purposes. This could partly explain such large wage bill. However, local governments in the state also have large civil services. In total, in 2008 there were 39,762 people in the Cross River LGs payroll, about 1.4 percent of the state population. Not all of them are civil servants.4 As seen in table 5.6, as percentage of the total population the LGs in Cross River are largely overstaffed. On average, countries in Sub Saharan Africa have 50 World Bank Working Paper civil services that represent about 1.5 percent of the population (1.7 percent in non Francophone countries). This 1.5 percent includes civil servants working in all levels of government, including both education and health employees. If we only include pensionable civil servants and primary school teachers, the total Cross River LGs personnel represent about 1.1 percent of the state population. Table 5.6. Percentage of Civil Servants out of Total Population in Sub-Saharan African Countries Sub-Saharan Non-Francophone Low income Middle income Africa Africa group group averageb averageb averageb averageb 1996­2000a 1996­2000a 1996­2000a 1996­2000a Civilian central government 0.30 0.38 0.46 0.59 Subnational government 0.30 0.38 0.46 0.59 Education employees 0.62 0.78 0.91 1.20 Health employees 0.29 0.20 0.62 0.70 General government total .. 2.67 2.27 4.26 General government excluding police and armed forces 1.50 1.73 2.45 3.09 Source: World Bank data base on public sector employment and wages. Large numbers of people under their payroll is not unique to Cross River's LGs. The state and local governance study (World Bank, 2002) recorded large variation in LGs staffing varying from 400 to over 1000 people on staff, not including primary school teachers. This is similar to the variation of pensionable civil servants in Cross River who vary from 254 in Bakassi, to 1039 in Boki. In response to this large payroll, many states have an embargo on recruitment of new local government staff. Despite the large number of personnel on their payroll, the LGs are also limited by personnel capacity constraints. For instance, a series of reports of the Kaduna Auditor General of LGs 1999­2004 can be indicative of the problem. These reports highlight the limited capacity of LGs treasures who lack book keeping skills and capacity to produce final accounts. Similarly, the Lagos State Financial Accountability Assessment (World Bank, 2004) also reports insufficient number of professionally qualified Treasurers and insufficient supporting staff in Treasury and Internal audit unit departments with relevant qualifications. Finally, as seen in a previous chapter, PHC facilities are understaffed, particularly those in rural areas. Local Government Health Expenditure The flow of resources to PHC facilities in the country is rather complex given the numerous agencies sharing responsibilities for the provision of services. The largest flow of resources is personnel remuneration which is financed by local governments. But as seen before, there are other flows of in kind resources going to these facilities from different levels of government and donors, although mainly from local governments. The following paragraphs assess health expenditure at local government level. This expenditure is mainly expenditure on PHC, although not all of it is. For instance, LGs in Kaduna include as health, expenditure on refuse collection. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 51 There is no consolidated account of LG expenditure. However, local government financial accountability is monitored by an Auditor General of Local Governments. The reports of the auditors general to the state assemblies present both budget and expenditure in all LGs. However, an estimate of actual expenditure across sectors is challenging. Recurrent and capital budget is presented separately, and while recurrent expenditure is presented in both administrative and economic classification, the capital budget is often presented in functional classification. Thus, while it is often possible to trace recurrent expenditure across administrative units (departments); it is not always possible to do the same with the capital budget. There are also large variations in budget presentation across states and within states, making comparisons difficult. Finally, the data presented in the Reports of the Auditors general is fairly aggregated making detailed analysis challenging. This section is based on partial data on LG expenditure in Kaduna and Cross River, two of the states that participated in the PHC study. The data are not fully comparable as in Cross River the data on personnel are not disaggregated across administrative departments. On average, local government expenditure on health is low. In 2005, Kaduna LGs spent on average about US$2 per capita on health; about 7 percent of the entire LG expenditure went to the sector. Partial information from LG expenditure on health in Cross River also indicates low expenditure on health. For instance, average overhead expenditure on health in 2005 was only about US$0.05 per capita, while capital expenditure was about US$1 per capita. Health expenditure varies largely across LGs. For instance, in Kaduna while health expenditure per capita in Zagon Kataf LG was only about US$1.34, in Jaba was about US$3.6. In Cross River, there is also large variation across LG. For instance, overhead expenditures per capita vary from US$0.01 to US$0.13 and capital expenditures per capita vary from US$4.6 to US$0.5 (see table 5.9 and table 5.10). These large variations are partly due to differences in shares of budget allocated to health. As shown in figure 5.2, in Kaduna LGs the share of total expenditure earmarked for health varies from as low as 3 percent to as high as 12 percent. In Cross River, there are also large variations in total expenditure allocated to health. For instance, overhead expenditure on health varies from 0.1 percent to about 6 percent of total overhead expenditure and capital expenditure on health varies from about 6 percent to about 12 percent of total capital expenditure. These differences in health expenditures are also due to large variations in total LG expenditure per capita. These variations mainly reflects the formula used to distribute the Federation Account revenues across states and LGAs as only 30 percent of the account is distributed according to population and, thus, in per capita terms states with less population receive more revenues. In 2005, on average, LGs in Kaduna spent about US$36 per capita while LGs in Cross River spent on average US$55. There are also large variations in total expenditures across LGs in each state as shown in figure 5.3. 52 World Bank Working Paper Figure 5.2. Share of Total LG Expenditure Allocated to Health in Kaduna LG, 2005 12 10 8 Percent 6 4 2 0 a i a ba So a Ka h o a u u n re Za S rfi on g a n- a i af ab ar iw ar na hi ut rk ur ur ba da bo ari a at Le Ja ng a n G Ig du Kac ga ak Ik Ka Ka G Ku Ku Sa Z -K M Ka Source: Authors estimates based on data from preliminary reports non audited data from the office of the Auditor General of Local Government of Kaduna. Information was not available on a few LGs. Figure 5.3. Total per Capita Public Expenditure across Local Governments in Kaduna and Cross River, 2005 Kaduna Cross River Sabon-Gari Yala Zaria Yakurr Zangon-Kataf Ogoja Sanga Odukpani Makarfi Obudu Lere Obubra Kudan Obanliku Kubau Ikom Kauru Etung Kaura Calabar Municipal Kagarko Kaduna South Boki Kachia Biase Jaba Bekwara Ikara Bakassi Igabi Akamkpa Giwa Abi 0 10 20 30 40 50 0 50 100 150 Source: Authors estimates based on data from Auditor General of Local Government of Kaduna and Cross River. Information was not available on a few LGs. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 53 There is also evidence that while total LG expenditure has increased significantly in the last years, health expenditure has increased only slightly. There are large variations in expenditure across LGs; however, health expenditure has actually decreased in many LGs and in those LGs where health expenditure has increased this increase has been lower than the growth in total expenditure (table 5.7). In only two LGs, Igabi in 2003­04 and Jaba in 2004­05, the growth rate of health expenditure was higher than that of total expenditure. Table 5.7. Real Growth Rate of Kaduna LG Expenditures in 2003­04 and 2004­05 (in %) Real growth rate 2003­04 Real growth rate 2004­05 Local government Total Health Total Health Chikun 93 20 Giwa 82 ­47 32 33 Igabi 22 31 110 ­54 Ikara 21 ­35 89 23 Jaba 200 ­65 ­8 222 Jema'a 35 ­2 Kachia 14 1 174 11 Kaduna North 32 ­33 Kaduna South 60 ­12 52 24 Kagarko ­4 1 179 ­15 Kaura 14 180 ­9 Kauru 18 2 114 1 Kudan 57 37 27 7 Zangon-Kataf 9 ­19 218 34 Sabon-Gari 126 22 Kaduna average 32 3 87 4 Source: Authors estimates based on data from reports of Kaduna's Auditor General of LGs 2003­04, and preliminary reports from LGs for 2005 (not yet audited). CPI data source: WB Development Data Platform. Data on expenditure in Cross River LGs show a similar pattern to that in Kaduna LGs. As seen in table 5.8, while the 2004­05 real expenditure in the LGs in Cross River grew on average 6 percent, both capital and overhead expenditure on health decreased. Similarly, on average overhead expenditure in the LGs decreased about 6 percent between 2005 and 2006 but the overhead expenditure on health decreased 45 percent. Most local government expenditure on health is on personnel remuneration, very little is allocated to overhead expenditure or capital expenditure. On average, recurrent expenditure on health in Kaduna's LGs represents more than 80 percent of total expenditure. Most of this expenditure is on personnel remuneration, representing on average about 64 percent of total recurrent expenditure on health. On average, in 20055 LGs in Kaduna spent US$0.76 per capita on non salary recurrent costs. Very little was spent on pharmaceuticals, medical supplies, or the maintenance of facilities. 54 World Bank Working Paper Table 5.8. Real Growth Rate of Cross River's LG Expenditures in 2005­06 (in %) Overhead Capital Overhead Capital Total expenditure on expenditure expenditure expenditure expenditure health on health Abi ­7 81 17 ­5 Akamkpa 3 11 ­1 ­66 ­44 Bekwara 6 60 13 ­71 Biase ­14 157 19 Boki ­15 74 12 ­70 14 Calabar Municipal 31 50 13 ­52 5 Etung 0 53 10 34 Ikom 3 54 8 24 Obanliku ­31 68 3 ­78 Obubra ­33 86 9 ­22 ­69 Obudu ­16 39 1 11 Odukpani 14 33 7 ­74 Ogoja ­7 59 12 ­21 Yakurr 4 37 5 54 Yala ­4 57 6 53 Cross River ­6 52 6 ­45 ­19 Source: Authors estimates based on data from reports of Cross River's Auditor General of LGs 2005 2006. CPI data source: WB Development Data Platform. In 2005, only two LGs in Kaduna (Ikara and Sanga), out of 17 for which data were available, had expenditure on the maintenance of health facilities. Similarly, only 11 LGs had expenditure on pharmaceutical products. Expenditure on drugs varied significantly across LGs from Naira 36,000 (US$275) in Sabon Gari to about 3 million naira (US$22,747) in Ikara. An important share of the total non salary recurrent costs in these LGs went to travel and transport costs, to the maintenance of office equipment and furniture, and in some LGs to "entertainment and hospitality" expenses. Most LGs included in their health recurrent expenditure the cost of refuse collection. The rest of the recurrent expenditure went to the logistical support of vertical programs such as TB control, HIV control, roll back malaria, and immunization. Information on health personnel expenditure in Cross River's LG was not available, but data on overhead expenditure show very little non salary recurrent expenditure (see table 5.9). Capital expenditure in Cross River LGs was about US$1 per capita. Local Government Accountability for Service Delivery Accountability connotes having the obligation to answer questions regarding decisions and actions (Brinkerhoff, 2004). It will imply both reporting information and justification for actions and decisions. It will also imply the availability and application of sanctions for illegal or inappropriate actions uncovered. The level of accountability of local governments could then be measured by the level of information sharing on budget process, and on activities or outputs. Very little of this is done. Information on Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 55 local government budgets and expenditure is difficult to come by. LGs, however, are answerable to auditors general of LGs but this information is usually given with delays and the auditor general is often powerless to apply any sanctions for irregularities. There is evidence of limited accountability of LGs in some states towards health service frontline providers. A study (Khemani, 2005) using data on LGs from Kogi state found very little accountability of the local governments reflected in the non payment of salaries of health workers despite available resources. The same study included data from Lagos were this problem was not found. Non payment of salaries of LGA staff have been reported in other states. For instance, in the last years Cross River state has been managing directly the LGs payrolls. The state took this decision after repeated complaints from LG civil servants for LG non payment of salaries. Local government accountability in relation to communities could be measured by their responsiveness to communities. The scorecard assessment of rural local governments (Terfa Inc., 2005), including all local governments in Bauchi, evaluated the responsiveness of local governments to the community by an index that included the following aspects: project implementation, project abandonment, overall councilors' responsiveness, councilor's meeting with community, consultation on budget issues, responsiveness to request assistance, performance of community outreach staff, and chairman's accessibility to members of the community. In general, the scorecard assessment found very poor responsiveness to communities among the participating local governments. In a ranking from 1­100, very few LGs in the nine states had more than 50 points in the ranking. Bauchi LGs, on average, had a score of 30 points. Possible Ways Forward Improving the performance and accountability of local governments regarding service delivery requires reforms that go beyond the health sector. A comprehensive civil service reform that reduces the number of civil servants and changes their skill mix will be needed. There is also a need for capacity building concerning public financial management. These reforms will increase the LGs capacity to provide services but they will not necessarily increase their accountability towards clients or towards other levels of governments. There is also a need to improve the accountability mechanisms at state level. For instance, auditors general monitor the financial accountability of local government. Despite limited resources, these auditors do a comprehensive work and present to the state assemblies detailed audit reports of local government finances. However, sanctions are often not imposed for uncovered irregularities. Conditional matching grants from the federal or state governments to local governments can be used as instruments to improve basic health service delivery. Both the federal and state level governments have shown interest in improving basic service delivery in the country. They have used different instruments to do so. As seen in this chapter, the states regulate and control most of the activities of the LGs; they also deduct resources from the LGs allocation to ensure that some activities are carried out. Many of these instruments have not produced the intended benefits as the performance of services can testified. Matching grants conditional on performance can 56 World Bank Working Paper offer local governments the incentives to improve services provided that they have flexibility and capacity to use these resources. The federal government has used this instrument to improve service delivery. The Office of the Senior Special Assistant to the President for the Millennium Development Goals has started a conditional grant mechanism intended to transfer funds to the sub national governments to improve basic service delivery and progress towards achieving the MDGs. The resources that fund this program come from debt relief. As discussed before, Cross River State has benefited from these conditional grants and has used these funds for a large PHC rehabilitation program in the state. The Health Bill that is currently in the National Assembly would create a similar conditional matching grant, the PHC Development Fund. However, for these conditional grant programs to obtain the intended benefits there is a need for systematic collection, analysis, and reporting of information. This information is needed to verify compliance with goals and to assist future decisions on whether or not to continue providing grants to sub national governments. The incentives provided by these grants will only improve performance if there is a real threat of funds withdrawal in case the performance is inadequate and this requires some standards or goals to be met and ways to measure whether these are met (Bird, 2000). In Kaduna the state and local governments' joint programs have the potential to create the incentives needed to improve performance. These joint programs in practice are matching grants for capital projects. These joint projects, however, have often worked not as incentives to the local government to perform, but as an imposition. So far they have been used mainly for investment projects, which might create a perverse incentive for the local government to finance part of these investments but not to maintain them. Nevertheless, this system could provide benefits if used as incentives for local governments that want to improve service delivery, not just construction and rehabilitation of facilities but also for some recurrent cost needed to provide services. Information on service delivery is not just important for creating accountability from local governments to other levels of government but more importantly to increase accountability of the LG in relation to clients. More information to the community on service delivery can increase accountability of local governments. Monitoring the performance of government policies, through report cards can work (see box 5.2). These citizen reports cards started in Bangalore, India but have been used in many different countries. In Sub Saharan Africa, South Africa, Ethiopia, Rwanda, and Mozambique are experimenting with these citizen report cards. The scorecard assessment of rural local governments in nine states was in essence a local government report card. However, both citizens and state and local government officials participated in the assessment. The objective of the assessment was to identify rural LGs for their inclusion as beneficiaries of LEEMP. However, publicizing broadly the results of the assessment and repeating it could also serve as a way to monitor LG performance. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 57 Box 5.2. Citizen Report Cards: The Bangalore Experience Citizen Report Cards (CRCs) are assessments of a municipality's public services from the point of view of its citizens who as users can provide useful feedback on the adequacy of the services and the problems they face in their interactions with providers. The resulting pattern of ratings, based on user satisfaction, is then converted into a "report card" on the municipality's service. These CRCs were developed in the city of Bangalore, India by the Public Affairs Center. The first report card was done in 1994 and only included a few municipal services such as water, electricity but they have since been extended to other cities and rural areas in India and have included health services. In the first report most public services received low ratings. Providers were rated and compared in terms of public satisfaction, corruption and responsiveness. The media publicity these results received and the public discussions that followed pressure public providers to improve services. When the second CRC was implemented in 1999 these improvements were reflected in better ratings and by 2003 the third CRC showed a large improvement of services. Public satisfaction has increased considerably and the incidence of corruption had declined perceptibly. Source: Samuel, 2004. A larger community participation in the local government budget planning process could also help improve service delivery. There is some experience in Nigeria in the education sector on these participative approaches (see box 5.3). Box 5.3. Participative Approaches in the Management of Education: Literacy Enhancement Assistance Project (LEAP) The LEAP project worked in Lagos, Nasarawa, and Kano and supported 9 local governments. This project was financed by USAID and implemented by the Research Triangle Institute (RTI) and the Education Development Center (EDC). "An essential aspect of the project was to encourage decision-makers and administrators to listen and respond to the opinions of stakeholders, or beneficiaries of the primary education system. Representatives of parents, teachers, general civil society and politicians at the local government level were facilitated to identify the most important problems that, in their collective opinion, impede the learning of mathematics and English for their children. These groups then brought their opinions to a workshop, where collectively they identified the priority problems and concurrent solutions for their local government. In Kano, the common themes identified for all three local governments working with LEAP, were (i) too many unqualified teachers; (ii) inadequate and irrelevant instructional materials; and (iii) a lack of school furniture. Solutions included the provision of summer training workshops for teachers, the development of school or classroom libraries and the construction of school furniture by parents. In Lagos, common themes included (i) the poor state of school infrastructure; and (ii) the limited interest of parents in their children's schooling. Solutions included a concerted lobbying effort of state authorities and private philanthropists to support classroom renovation (resulting in significant grants from an oil company to Lagos Island schools) and a program to provide ideas to parents on to how to use daily interactions to teach English and mathematics concepts (using billboards). In all the local governments, the stakeholders collaborated closely with the local government council and the education authorities to address and improve the learning of their children." Source: Destefano and Crouch, 2005. 58 World Bank Working Paper Notes 1 According to the 1999 Constitution, the local government councils are in charge of: the establishment and maintenance of cemeteries and burial grounds, slaughter houses, markets, motor parks, and public conveniences; construction and maintenance of roads, streets, street lightings, drains, and parks; provision and maintenance of sewage and refuse collection; control and regulation of advertising, shops, restaurants, and so forth; and participation with the state government in the provision of basic health and education services. 2 This assessment was done by the World Bank financed Local Empowerment Management Project (LEEMP). The objective of this scorecard exercise was to identify rural local governments in participating states where the "level of commitment to effective service delivery and responsiveness to rural communities justify their inclusion in the LEEMP project" (Terfa Inc., 2005). The participating states were: Adamawa, Bauchi, Bayelsa, Benue, Enugu, Imo, Katsina, Niger, and Oyo. This scorecard assessment was based on interviews with community representatives and with different state and local government officials, such as the Auditor General of LGs, the Chairman of the local government Service Commission, LGA Chairman, Councillors representing the communities visited, Director of Administration or Personnel, Treasurer or Director of Finance, Internal Auditor, and five heads of department. 3 Anambra, Bauchi, Nasarawa, Ogun, Rivers, and Sokoto. 4 Data from the Cross River Ministry of Local Government shows that as of April 2008 there were 20,540 primary school teachers; 11,717 pensionable civil servants; 3,196 non pensionable staff; 3,573 traditional rulers; and 736 political office holders. 5 This information comes from preliminary data from the Office of the Accountant General of LGs in Kaduna. These data have not been audited. Table 5.9. Cross River Local Governments Expenditure 2005 Calabar Abi Akamkpa Bakassi Bekwara Biase Boki Municipal Etung Ikom Obanliku Obubra Obudu Odukpani Ogoja Yakurr Yala Personnel cost 162,784,535 172,059,820 89,244,277 132,735,966 180,858,671 218,749,926 168,426,540 153,575,759 170,466,123 163,715,825176,076,176 196,685,727 193,611,482 152,412,661 190,441,345204,873,525 funding for primary education 159,134,496 178,009,212 50,581,751 121,643,839 194,897,624 262,977,178 192,945,751 121,398,682 268,788,141 201,809,563236,209,154 235,346,671 164,568,225 236,829,987 216,903,368238,828,146 pensions and gratuities 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 51,520,264 3,769,539 training fund 3,769,539 3,769,539 3,769,539 3,769,539 3,769,539 3,669,529 3,769,539 3,669,539 3,669,539 3,769,539 3,669,539 3,769,539 3,669,539 3,769,523 3,769,539 51,520,264 overhead cost 164,424,013 126,976,385 135,492,887 115,033,454 135,599,150 161,393,641 129,727,736 95,251,085 120,189,868 147,869,494130,774,257 129,439,343 130,662,041 116,566,486 147,266,809152,306,014 capital expenditure 218,386,872 350,377,771 181,372,136 174,563,915 127,558,222 292,330,719 205,089,468 189,197,147 237,084,270 174,362,046197,626,650 216,522,669 221,610,404 208,957,120 193,240,268191,243,099 Total 760,019,720 882,712,991 511,980,854 599,266,978 694,203,470 990,641,258 751,479,298 614,612,476 851,718,205 743,046,731795,876,041 833,284,213 765,641,956 770,056,041 803,141,593842,540,588 Health overhead expenditure 2,429,522 6,611,600 1,509,170 2,599,550 639,400 316,175 621,590 178,992 241,100 1,284,625 441,863 1,099,496 268,200 health capital expenditure 18,519,475 20,144,175 17,746,175 17,726,175 21,426,175 20,421,179 22,639,175 total health no personnel 22,573,697 19,235,345 24,025,725 20,600,171 22,907,375 Population 141905.96 148102.5 31737.3 103705.56 165799.34 182418.18 175804.16 78592.08 159135.34 108117.52 168995.12 156903.88 188595.12 168462.98 192521 206626.14 Remuneration as % of total expenditure 49.1% 37.4% 37.4% 51.0% 61.5% 53.8% 54.9% 53.1% 57.6% 56.1% 58.3% 58.0% 53.5% 57.2% 57.1% 53.1% capital exp. as % of total 28.7% 39.7% 35.4% 29.1% 18.4% 29.5% 27.3% 30.8% 27.8% 23.5% 24.8% 26.0% 28.9% 27.1% 24.1% 22.7% 59 health exp. as % of total overhead 1.9% 5.7% 0.9% 2.0% 0.7% 0.3% 0.4% 0.1% 0.2% 1.0% 0.4% 0.7% 0.2% health as % of capital 8.5% 5.7% 9.8% 6.1% 10.4% 10.3% 11.8% total per capita in nominal naira 5,356 5,960 16,132 5,779 4,187 5,431 4,275 7,820 5,352 6,873 4,709 5,311 4,060 4,571 4,172 4,078 total health no-personnel per capita in nairas 152 105 137 122 111 health overhead per capita in Nairas 16 64 8 15 8 2 6 1 2 7 3 6 1 health capital per capita in Nairas 131 136 559 97 122 121 110 total per capita in current US$ 41 45 123 44 32 41 33 60 41 52 36 41 31 35 32 31 total health no-personnel per capita in current US$ 1.2 0.8 1.0 0.9 0.8 health overhead per capita in current US$ 0.13 0.06 0.11 0.06 0.02 0.04 0.01 0.01 0.05 0.02 0.04 0.01 health capital per capita in current US$ 1.00 1.04 4.3 0.7 0.9 0.9 0.8 Source: World Bank estimates based on Report of the Auditor General of Local Governments 2005. Table 5.10. Kaduna Local Government Expenditure 2005 Kaduna Zangon- Sabon- 2005 Giwa Igabi Ikara Jaba Kachia South Kagarko Kaura Kauru Kubau Kudan Lere Makarfi Sanga Kataf Zaria Gari Total recurrent 824481923.1 1,261,056,554 806104679.5 705630132.7 929638993.2 1276535573 785558420.6 676113090.4 786472445.8 1000705409 636902765.1 965719608.7 706523741.8 678616906.4 854965404.8 1065982289 887442782 Recurrent education and social development 267641502 201977771 386908189 260877160.9 30915145.48 237793735.4 247042779.3 140469271.3 308208385.2 27683975.41 304760518.1 302981458 Recurrent Health 54868156.06 88,431,537 64030283.85 71882908.42 73267203.36 141098973.4 51786302.46 54721131.07 63944617.86 57392704.87 42553848.05 84667063.52 53322731 43971086.57 54597181.81 108485093.7 72,283,606 total capital 263046535.5 540438401 289957016.2 83588174.59 254161309 247322296.4 170988675.9 117545544.2 239885297.1 385693414.6 213199225 309716013.9 232170209 191172424.9 173765611.5 212485956.6 228001308 Capital education 30178673.5 76067964.72 41061908.47 13632000 36594014.14 20231659.3 2022727 8006770.6 65429245 30605828.4 56387863.05 14951584.15 20313977 15787390 32361150.25 11915000 Capital health 12866465 21768314.26 7649779.9 800000 9896849.58 52129689.55 5972625 181500 27365016.96 21705000 24500000 22174490 6232839.4 9900000 5542167.39 2797928.2 9422392 Total expenditure 1087528459 1801494955 1096061696 789218307.3 1183800302 1523857869 956547096.4 793658634.6 1026357743 1386398823 850101990.1 1275435623 938693950.8 869789331.3 1028731016 1278468246 1115444090 Total Health in Nairas 67734621.06 88,431,537 71680063.75 72682908.42 83164052.94 193228663 57758927.46 54902631.07 91309634.82 79097704.87 67053848.05 106841553.5 59555570.4 53871086.57 60139349.2 111283021.9 81705997.59 % of recurrent exp in total health expenditure 81% 62% 89% 99% 88% 73% 90% 100% 70% 73% 63% 79% 90% 82% 91% 97% 88% % of total capital expenditure that is health 5% 4% 3% 1% 4% 21% 3% 0% 11% 6% 11% 7% 3% 5% 3% 1% 4% Population 280698 421624 190047 152269 239389 394342 236124 218127 166608 276404 136212 324538 143334 146346 310043 400034 281134 Total exp. per capita 3874 4273 5767 5183 4945 3864 4051 3639 6160 5016 6241 3930 6549 5943 3318 3196 3968 Health exp. per capita in current Nairas 195 210 337 472 306 358 219 251 384 208 312 261 372 300 176 271 257 60 Total exp. per capita in current US$ 30 33 44 40 38 29 31 28 47 38 48 30 50 45 25 24 30 Health exp. per capita in current US$ 1.5 1.6 2.6 3.6 2.3 2.7 1.7 1.9 2.9 1.6 2.4 2.0 2.8 2.3 1.3 2.1 2.0 Health as % of total 6% 5% 7% 9% 7% 13% 6% 7% 9% 6% 8% 8% 6% 6% 6% 9% 7% capital as % of total 24% 30% 26% 11% 21% 16% 18% 15% 23% 28% 25% 24% 25% 22% 17% 17% 20% Total personnel cost 244290385.5 388694713.9 234830191.7 314861968.7 396997576.6 257321382.9 158382768 366528577.9 168445446.3 117526797.5 118699233 182160620.2 116586868 297540904.2 438143985.1 219497643.1 139755986.3 health personnel cost 33430730.68 53,157,485 37328752 47072276 49785019 90289696 36916105 40504616 50449019 35743467 14588017 65110141 34279661 29076328 40936495 73540076 34,897,426 health personnel as percentage of health recurrent cost 61% 60% 58% 65% 68% 64% 71% 74% 79% 62% 34% 77% 64% 66% 75% 68% 48% Health non-salary recurrent 39% 40% 42% 35% 32% 36% 29% 26% 21% 38% 66% 23% 36% 34% 25% 32% 52% Source: WB estimates based on data from preliminary reports non audited data from the office of the Auditor General of Local Government of Kaduna. Information was not available on a few LGs. CHAPTER 6 Policy Makers Providers O ften services fail communities and particularly poor communities if resources do not reach frontline providers; if these providers do not have the incentives to serve the community, especially the poor; and if they are not responsive to communities' preferences and demands (World Bank, 2003). However, ensuring providers' compliance to offer quality services is not simple; it requires offering the right incentives and a close monitoring of their work. To better understand the relationship between policy makers and primary health providers in Nigeria, this chapter first describes the characteristics of frontline providers in the four states sampled. This will be followed by an evaluation of the incentives these providers face to perform their work. This section is based on the survey on health facility personnel. In all facilities sampled, 25 percent of all types of personnel present in the facility at the time of the survey were interviewed (table 6.1). A total of 881 PHC workers were sampled1. Table 6.1. Health Care Personnel Sampled across States Bauchi Cross River Kaduna Lagos Total Medical officer 12 8 0 24 44 Community health officer 11 25 11 9 56 Public health nurse 14 40 12 37 103 Nurse 4 25 2 5 36 Nurse/midwife 15 14 16 32 77 CHEW 38 64 21 6 129 JCHEW 33 39 28 4 104 Environmental health officer 8 1 1 4 14 Lab technician 9 8 3 13 33 Pharmacy tech 6 5 2 8 21 Medical records officer 7 9 0 8 24 Dental assistant 1 1 1 0 3 Community health worker 8 7 6 0 21 Other (includes support staff such as 105 43 9 42 199 attendants, cleaners and security guards) Total 271 289 112 192 864 Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). The distribution of the sample by occupation reveals the majority of medical officers interviewed were from Lagos as was the case for nurses / midwives. Lagos also 61 62 World Bank Working Paper had a large sample of Public Health Nurses as did Cross River. The CHEW and JCHEW interviews were concentrated in Bauchi, Cross River, and Kaduna. The "other" occupation designation includes attendants, cleaners and security guards with a large number of the interviews of this group in Bauchi. Characteristics of Health Personnel The majority of the health personnel interviewed across states were women with the sole exception of Bauchi where about 68 percent of the sampled personnel were men. While the majority of doctors in all states were men, the majority of nurses were female. In the case of community health workers, in the two northern states the majority of CHEWs and CHOs were men while the JCHEWs female. The gender of the frontline providers often affects the demand of services. Services can also fail communities if they are not demanded. Lack of demand could be associated with the services not complying with the preferences of the community where they are located. The Nigeria 2003 DHS collected information among adult women on their perceived barriers to access health care. The main barrier to access services reported by women living in the North West region, where Kaduna is located, was the concern of non availability of a female provider. This was also an important concern among women living in the North East region, were Bauchi is located (table 6.2). Table 6.2. Health Care Personnel Sampled by Gender across States Bauchi Cross River Kaduna Lagos Male Female Male Female Male Female Male Female Medical officer 91.7 8.3 75.0 25.0 75.0 25.0 CHO 54.6 45.5 40.0 60.0 54.6 45.5 22.2 77.8 Nurse/midwives 15.2 84.9 8.9 91.1 16.7 83.3 2.7 97.3 other technical staff 83.3 16.7 45.8 54.2 71.4 28.6 36.4 63.6 Community-based health worker 100.0 0.0 57.1 42.9 16.7 83.3 CHEW 55.3 44.7 31.3 68.8 71.4 28.6 0.0 100.0 JCHEW 72.7 27.3 20.5 79.5 39.3 60.7 0.0 100.0 Other 80.0 20.0 58.1 41.9 66.7 33.3 26.2 73.8 Total 67.94 32.06 31.6 68.4 44.14 55.86 22.34 77.66 Source: Nigeria 2003 DHS. Note: This table collapses some of the categories of table 6.1. For instance, other technical staff includes environmental health officer, laboratory technician, pharmacy technician, medical records officer, and dental assistant. On average a PHC worker is 37 years of age and has about nine years of experience and five years working in the same health facility. Half of them are from the same area where they work, and at least three in every four live with their families. There are few marked differences across states. First, PHC workers in Bauchi and Cross River are more likely to come from the same area where they work. Second, PHC workers in Kaduna and Lagos, on average, have less experience than those in Bauchi and Cross River. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 63 There are however differences between public and private employees. Public PHC employees are more likely to be older and have about five years of experience more than private sector employees (table 6.3). Table 6.3. Characteristics of PHC Personnel across States and across Type of Facility Ownership Bauchi Cross River Kaduna Lagos Public Private Age 37.9 36.0 36.0 36.4 38.4 32.6 Years of experience 9.1 11.1 6.5 7.2 11.2 4.5 Years working in the PHC 8.6 9.5 8.4 8.8 10.7 4.5 Years working in facility 5.9 4.5 4.0 4.9 5.1 4.4 Indigene to the community 62% 57% 40% 32% 60% 32% Lives with wives and children 72% 70% 75% 69% 78% 55% Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). More qualified PHC workers are, on average, younger and much less experienced than less qualified workers. For instance, doctors have on average only about 5 years of experience, while CHEWs have almost 12 years. Medical officers are also less likely to come from the community (table 6.4). Table 6.4. Characteristics of PHC Personnel across Type of Personnel Medical Nurse/ Technical Officer CHO midwife staff CHEW JCHEW Age 34.9 40.5 34.8 35.2 37.3 31.7 Years of experience 4.7 14.0 7.6 6.1 11.6 7.0 Years working in the PHC 4.8 13.8 6.6 7.2 11.8 7.2 Years working in facility 4.4 4.1 4.8 4.2 4.3 3.3 Indigene to the community 32% 40% 42% 43% 45% 54% Lives with wives and children 55% 84% 64% 60% 77% 63% Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). Education Level Most health workers have an Ordinary National Diploma or a Higher National Diploma. These certificates are given in Schools or Colleges of Health Technology where most PHC personnel such as CHOs, CHEWs, JCHEWs, are trained. Community Health Officers receive four years of training; although in the past CHEWs with some years of experience and an extra year of training could also become CHOs. CHEWs receive two years of training and JCHEWs one year. Not surprisingly, most personnel with OND/HDN diploma are found in Cross River and Bauchi where most CHEWs and JCHEWs were interviewed. Most interviews with university graduates were conducted in Lagos, likely due to the concentration of medical officers in this state (table 6.5). 64 World Bank Working Paper Table 6.5. Highest Level of Education Completed by PHC Staff Interviewed (State Comparison) Bauchi Cross River Kaduna Lagos Primary school 41 22 9 16 Secondary school 44 59 17 39 OND/HND 85 150 53 35 University 10 23 1 30 Post graduate 7 18 3 12 Other 75 24 28 59 Source: Health Personnel Survey (EPOS, CISH, CHESTRAD, 2007). Incentives to Providers Bennet and Franco (1999) offer a conceptual framework to understand workers motivation. According to these authors, health workers motivation is a complex internal process that is determined by numerous individual, organizational, and socio cultural or environmental factors explained below. Workers individual needs, self concept, and their expectations for consequences affect their motivation for performance. Some factors that can influence workers to exert efforts in their performance might be more important than others.2 There are factors that can affect workers dissatisfaction by their presence or absence such is the case of salary, work conditions, job security, and interpersonal relations. Other factors such as achievement, the work itself, recognition, responsibility, advancement and growth can determine the level of motivation and satisfaction. However, without the first set of factors; known as "hygiene factors", it is very difficult to provide positive motivation to perform. The organization and structure of the health system also affect workers motivation by affecting the availability of inputs workers have to do their work (for example, drugs, equipment, and supplies); by affecting their autonomy in performing their tasks; and by providing the feedback and training needed to update and maintain the skills needed to perform. Finally, the environment in which the health workers provide services also affect their motivation. For instance, the integration of health personnel in the social environment where they work can affect their motivation to provide quality services to the community. Motivation is an internal process and thus not observable; however, some of the determinants of this motivation, and particularly some of the incentives the workers faced such as "hygiene factors" and some organizational and environmental determinants of their motivation can be observed. The next paragraphs give an overview to some of the incentives faced by PHC workers in Nigeria. Salary and Fringe Benefits There are large differences in the PHC personnel salaries paid by the public and private sector. Nurses and midwives in the public sector are better paid than their counterparts in the private sector. In contrast, medical officers are better paid in the private sector. These results are partly driven by large differences in years of Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 65 experience of nurses in public and private sector. On average, nurses in the public sector have about 10 years of experience; in contrast, nurses in the private sector have on average about 5 years. The sample of CHO, CHEWS, and other PHC personnel working in the private sector is too small to make any final conclusion. However, as in the case of nurses, on average the years of experience of these personnel in the public sector are higher than in the private sector. In the case of medical officers, the difference in experience is small (table 6.6). Table 6.6. Average Salary of PHC Personnel across Type of Facility Ownership Private Government Total Obs Mean Obs Mean Obs Mean Medical officer 22 78,773 12 38,342 35 64,674 Community health officer 8 10,500 45 29,566 53 26,688 Nurse/midwives 109 13,974 92 32,765 202 22,632 other technical staff 47 14,862 41 24,369 92 20,322 Community-based health worker 5 14,760 15 11,403 20 12,243 CHEW 7 8,500 109 20,810 117 20,203 JCHEW 21 8,724 75 15,527 96 14,039 Other 35 9,165 152 12,331 193 12,829 Total 256 18,433 552 20,978 808 20,485 Source: Health facility personnel survey(EPOS, CISH, CHESTRAD, 2007). Salaries in nairas. When compared with other countries in the region, these salaries in relation to GDP per capita are not high. The salary of nurses in the public PHC facilities, on average, is about 4 times GDP per capita, the salary of CHEWS about 2.4 times GDP per capita. In the case of nurses, these salaries are similar to the lower bound salaries for diploma nurses in other countries in the region (see table 6.7). CHEWS are a type of Table 6.7. Salary of Doctors and Nurses in Relation to GDP per Capita in Different Sub-Saharan African Countries (in %) Country General practitioner Diploma nurse Burkina Faso 7.3­23.5 4.2­13.5 Burundi 11 4.2­9.6 Cameroon 5 2 Chad 10.3­18.8 5­10.6 Congo, Dem. Rep. of 1.8­2.40 Ethiopia 18.3­30 11.7­27 Kenya 17 6,9 Mauritania 5.67­9.45 3.2­5.7 Niger 10.6­20.8 5.3­12.0 Zambia 23 54 Source: Most countries data are from WB Human Development, Africa Region, and Country Status Reports. For Zambia and Kenya: WDI, Country case study on health workforce financing and employment in Kenya (forthcoming) and Zambia report on human resources for health (forthcoming). 66 World Bank Working Paper PHC personnel that are Nigeria specific. As mentioned before, they are high school graduates with a two year training in schools of health technology. In other words, CHEWs have in some instances the same number of years of training of nurses in other countries in the area. Nevertheless, in GDP per capita terms their salaries are much lower than that of diploma nurses in other countries in the area. In the past there were complaints of non payment of salaries of PHC personnel. As explained before, in Cross River state after complaints for non payment of salaries by the local governments, the state now manages the payroll. In other Nigerian states similar problems have been reported (Khemani, 2005). At the moment, non payment of salaries does not seem to be a problem in the four states sampled, although there are delays in the payment. Indeed, most PHC employees have been paid for every month in the last 12 months (see table 6.8). However, there are delays in the payment of salaries especially in Cross River state where only 30 percent of the personnel sampled receives the salary at the end of the month. Not surprisingly, 37 percent of the health personnel sampled in Cross River indicated that getting paid constituted an obstacle to do their job. With the exception of Lagos, where the majority of the health care personnel sampled are employed by the private sector, less than a third of the personnel receives fringe benefits such as health care and housing from their employers. In addition, some health care personnel, particularly in Cross River and Kaduna receive housing benefits from the community. Table 6.8. Salaries and Fringe Benefits (State Comparison) (in %) Bauchi Cross River Kaduna Lagos Salary Paid every month for 12 months 94 83 90 83 Received by end of month 70 30 59 53 Employer benefits Healthcare 25 23 25 84 Medicine 22 18 14 82 Housing 8 27 33 26 Food items 3 13 8 11 Community benefits Housing 9 14 13 6 Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). As seen in table 6.9, the salary differences between types of LGAs are small. Salaries in urban areas are larger than those in rural and semi urban areas, while salaries in semi urban areas are the lowest. However, these differences mainly reflect differences in the characteristics of the personnel employed in urban and rural areas.3 Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 67 Table 6.9. Average Salary of Public PHC Personnel across Type of LGA Rural Semi-urban Urban Mean Mean Mean CHO 26,708 23,269 38,185 Nurse/midwife 33,771 27,500 34,128 CHEW 21,905 19,267 21,645 JCHEW 17,122 14,095 15,288 Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). Salaries in nairas. These small differences in the salaries of personnel across type of LGA show very little financial incentives for personnel to live in rural areas. As seen in a previous chapter this is reflected in the distribution of personnel across LGA, as facilities in urban LGAs have on average more workers than facilities in rural LGAs. Some states are aware of this issue and are providing or increasing a "rural allowance" or "rural posting" for their health personnel. For instance, the LGs in Kaduna state offer employees a rural allowance that represents about 30 percent of their basic salary, although only about 6 percent of their total salary. Mechanisms to Reward and Discipline PHC Personnel The process to discipline staff is very complex rendering almost impossible measures such as firing staff. Figure 6.1 shows the process needed for disciplining PHC staff in Kaduna state. The facility head can initiate the process by sending a request for revision to the Ward Focal person, who then sends it to the PHC Department in the LG. For staff grade level 6 and below, the complaint is then sent to the Jr. Staff Management Committee in the LGs who then takes the final decision, although the Figure 6.1. Process to Discipline PHC Personnel Local Government Service Board Sr. Staff Jr. Staff Management Management Committee Committee PHC PHC Coordinator Coordinator Ward Focal Ward Focal Person Person Facility Facility Head Head Staff Grade 7 and up Staff Grade 6 and below 68 World Bank Working Paper minutes of the meeting need to be forwarded to the LGSB. For Senior Staff, the final decision is taken by the Local Government Service Board after a request is sent from the Sr. Management Committee in the LG. A similar procedure is followed in other states. Given the complexity and the difficulties generated by this process to discipline personnel, many states are trying to change this procedure. In Kaduna, the draft bill creating the PHC Agency will take over the responsibilities of the LGSB. In Cross River, the draft regulation will give some of the LGSC responsibilities to the SMOH. Regarding staff motivation, the main criterion for the promotion of staff is simply the years of experience. Merit, performance, or obtaining additional qualifications are reported as the second main criteria. There is not much difference across states or even across private and public employees, with the sole difference that a lower percentage of private employees reported the number of years of service as the main criterion for promotion (table 6.10). Table 6.10. Criteria for Promotion of Staff (in %) Bauchi Cross River Kaduna Lagos Public Private Total Number of years of service 46.7 52.7 44.4 51.8 53.6 40.8 49.6 Recommendation from management 11.8 7.8 14.8 7.2 8.8 12.5 9.8 Merit/performance/additional qualifications 30.6 34.0 36.1 34.3 31.7 36.3 33.3 Value of my network 2.0 3.1 4.6 5.4 1.9 7.1 3.4 Combination of above 9.02 2.38 -- 1.2 4.03 3.33 3.89 Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). Other Negative Incentives Faced by PHC Personnel A large percentage of PHC staff reports obstacles in receiving supplies and equipment as well as training for their jobs. Almost 60 percent of health care personnel in Bauchi and Cross River reported obstacles in obtaining supplies and equipment and more than a third of the personnel in Kaduna and Lagos reported the same obstacles. More than half of the personnel in Bauchi, Cross River, and Kaduna also reported having obstacles in receiving training. In Lagos, only 20 percent of the personnel did. Transportation to the facility was also reported as an important obstacle in all states (table 6.11). Table 6.11. Negative incentives Faced by PHC Personnel across States (in %) Bauchi Cross River Kaduna Lagos Getting paid 23 37 19 24 Receiving supplies & equipment 56 59 46 31 Have enough work space 27 35 43 17 Receive training 53 54 59 23 Adequate supervision 35 37 38 15 Transportation 41 43 39 29 Time scheduling 37 40 23 20 Physical security 32 40 24 13 Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 69 Health staff in rural areas reported considerable more obstacles to provide services than staff in urban or semi urban areas. Only a very small percentage of health personnel in rural areas reported having adequate equipment to provide services, adequate toilets and water supply. In addition, the percentage of personnel reporting obstacles in providing services is much larger in rural areas than in both urban and semi urban areas where the differences are small. This is especially the case when reporting obstacles to receive supplies and equipment, supervision, enough space to work, and transportation (table 6.12). Table 6.12. Obstacles in Doing Job across Rural and Urban Areas (in %) Rural Urban Semi-Urban % with Adequate equipment 11 43 31 Assessment: Toilet-good 12 51 33 Water supply - good 19 38 36 Obstacles in doing your job: Getting paid 33 24 29 Receiving supplies & equipment 73 43 44 Have enough work space 43 24 29 Receive training 57 45 45 Adequate supervision 46 26 30 Transportation 53 33 38 Time scheduling 40 27 37 Physical security 37 23 35 Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). Government PHC employees are more likely to report facing obstacles when doing their work than private sector employees (table 6.13). There is a 40 percentage point difference between the percentage of public and private employees reporting obstacles in receiving supplies and equipment. There are also large differences between public and private employees reporting obstacles with transportation, time scheduling and physical security. Table 6.13. Obstacles in Doing Job across Type of Facility Ownership (in %) Private (n=271} Government (n=585) % Adequate equipment 64 17 Amenities: Toilet-good 66 24 Water supply - good 75 21 Obstacles in doing your job: Getting paid 19 31 Receiving supplies & equipment 24 64 Have enough work space 20 33 Receive training 34 54 Adequate supervision 18 38 Transportation 24 46 Time scheduling 16 40 Physical security 14 37 Source: Health facility personnel survey (EPOS, CISH, CHESTRAD, 2007). 70 World Bank Working Paper Finally, as mentioned before, the environment in which health workers provide services can also affect their performance. For instance, the degree of integration of the worker in the community where he serves can also affect their motivation. The desire to be appreciated and respected by their clients can be a powerful factor affecting the effort of the provider (Bennet and Franco, 1999). Workers coming from the same community or highly integrated into the community in which they serve are thus more likely to be motivated to offer quality services. As seen in table 6.3, workers in Bauchi and Cross River are more likely to be indigene to their community than workers in Kaduna. This could partly explain why households in Bauchi and Cross Rivers are also more likely to be satisfied with the attitude of personnel than households in Kaduna (see table 3.19). Health Personnel Coping Mechanisms In response to inadequate salaries and poor working conditions, many health care workers respond by developing different coping strategies; some, albeit not all, of these strategies might result in conflicts of interest or in taking time from their work in the PHC facilities (Van Lergerghe et al., 2002). The survey on PHC personnel collected information on these coping strategies. The results are detailed in the following paragraphs. The majority of the staff works fulltime in the health facility; however, a large percentage of these employees supplement their salaries with other economic activities (table 6.14). In the two northern states, more than two thirds of the staff supplements their salaries, while in the two southern states, only about a third or less of the staff do. Table 6.14. Percentage of Personnel Who Are Fulltime Employees and Supplement Their Salary State Fulltime employee Supplements salary Bauchi 93 67 Cross River 89 34 Kaduna 91 68 Lagos 98 25 Source: Health personnel survey (EPOS, CISH, CHESTRAD, 2007). A large number of fulltime workers who supplement their salaries with other activities work in a private facility or provides health care services at their house or in the house of patients (table 6.15). Although the most common activity to supplement their salaries is agricultural work in Bauchi, Cross River, and Kaduna and trade in Lagos; about 20 percent of staff supplements their salaries offering health services outside the facility in Bauchi and Cross River. In Kaduna, more than 40 percent of health staff offers services outside the facility and almost 30 percent of the staff that supplements its salary in Lagos does it by providing health care services outside. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 71 Table 6.15. Activities to Supplement Salaries of Health Staff across States (in %) Bauchi Cross River Kaduna Lagos Public Private Agricultural work 74 80 85 7 80 32 Trade 14 37 34 52 25 34 Private facility 11 11 3 24 5 33 Provides health care at home 8 10 41 4 9 20 Sells medicines 7 14 13 7 10 7 Source: Health personnel survey (EPOS, CISH, CHESTRAD, 2007). Personnel working in public facilities were more likely to supplement their salaries (52 percent) than those working in private facilities (35 percent). They were also more likely to supplement their salaries with agricultural work and trade. Possible Ways Forward The Nigerian government has ensured the staffing of PHC facilities by creating special types of PHC personnel. Often these workers come from the same area where they work, ensuring their integration in the community they serve. Nigeria does not have the acute lack of health personnel that is common in other countries in the region. However, there is room for improvements. Health care personnel are very unequally distributed across rural and urban areas and many basic factors determining health personnel motivation are lacking. Their salaries are delayed; they often do not have basic drugs and equipment to offer services; do not receive adequate training; and are poorly supervised. In addition, providers' accountability in relation to policy makers and clients is weak. Measuring providers' accountability to local governments and patients is difficult. Lewis (2006) includes as a key measure of provider's accountability the "authority to reward performance and discipline, transfer, and terminate employees who engage in abuses". In the four states surveyed, the management of PHC personnel is cumbersome and fragmented given the number of agencies involved. Similarly, the lines of responsibilities regarding personnel supervision and management are not always clear. This makes any measure to discipline or motivate health personnel difficult to implement. As a result, frontline providers face little consequence for non performance. Finally, their salaries are fixed and not linked to the provision of services; thus, they have little incentives to respond to the communities' demands. It is difficult to motivate personnel if basic factors such as in time payment of salaries are not present. Thus, to ensure providers compliance one of the first things needed would be to ensure that they are paid in time, and that they have a minimum set of equipment, drugs, and consumables needed to provide services. Beyond these basic or "hygiene" factors, policy makers have other options to ensure provider's compliance to offer quality services. Many of these options often escape the health sector. For instance, there is an urgent need for a civil service reform that allows a more flexible and responsive mechanism to motivate and discipline health providers. Human resource management for health at the moment is fragmented, the LG and the LGSC or LGSB have the main responsibility, but other agencies also intervene. This fragmentation also creates challenges for worker 72 World Bank Working Paper motivation. For instance, staff development and supervision are done by different government agencies, but despite this, too little is done. Inside the health sector there are also options to improve health workers motivation and ensure compliance. But these options are not simple to implement, especially in the case of clinical services. These services are difficult to monitor as they are discretionary and characterized by large information asymmetries between policy makers, providers, and clients (World Bank, 2003). However, not all services provided by PHC facilities are difficult to monitor. Preventive services offered to a target population such as immunization, micronutrient supplementation, and antenatal care, have been standardized and can be monitored. Policy makers could then monitor these services and offer public resources on the basis of increasing the coverage of these services. The conditional transfers that the Office of the Senior Special Assistant to the President for the MDGs is now providing and the future PHC Development Fund could be made conditional to increasing the coverage of these basic services. At the moment, the transfers from the MDG office are mainly transfers for capital projects. Similarly, the PHC Fund seems to be mainly focused on the joint financing of capital projects. These projects are needed given the large need for rehabilitation and equipment of facilities. But these resources could also be used for recurrent costs needed to improve the coverage of basic preventive services that remain low. In other words, the amounts of the transfers as well as their continuity could be conditional on performance, measured by the increase in outputs that can be monitored such as immunization rates, antenatal care coverage, and so forth.4 However, for this performance based financing to be effective, providers need more autonomy in the use of resources. At the moment, PHC facilities only receive resources in kind from the different levels of government (for example, drugs and supplies). They collect some resources from fees but they cannot use these resources as they have to return them to the local governments. With so little autonomy in the use of resources, it is hard to make these public providers accountable to improve service provision. By allowing facilities to retain the resources they obtain from the provision of services and by reducing the in kind financing of the facilities they can be more responsive. For instance, if performance based transfers are used, facilities could receive funds also based on performance in achieving a certain level of coverage. The community could offer oversight in the use of resources and can also help in monitoring results. Ensuring the provision of quality clinical care is more difficult. Empowering clients by strengthening their power in relation to providers could improve providers' responsiveness. Increasing information and community awareness on the services facilities provide and the resources they have to provide them and on the credentials and standard of services of providers could also help. Contracting out services to the private sector is also an option to explore. Contracts are difficult to monitor and enforce, in particular contracts for clinical services. However, it is possible to start by contracting out services that are easily to monitor and are highly cost effective such as social marketing of consumables (insecticide treated nets, ORS sachets, condoms) and population based services such as vaccinations, micronutrient supplementation, and so forth. Making these contracts Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 73 based on performance, for instance based on achieving a pre specified coverage level would certainly align providers incentives with the achievement of these targets (see Loevinsohn, 2008). At the moment, some services in the country are contracted out to NGOs, as is the case of HIV/AIDS preventive services. As experience builds with the design and monitoring of contracts, other services, including curative clinical services, could also be contracted. Notes 1 General note: The results of the health surveys presented in this chapter were taken from the Final Report for this survey prepared by EPOS Health Consultants, Canadian Society for International Health (CSIH), and Center for Health Sciences Training, Research and Development (CHESTRAD). 2 Hertzberg (1959) as quoted in Bennet S. and Franco (1999). 3 A regression on the determinants of salaries of nurses and CHOs showed that age, experience, ownership of facilities, and state where the main determinants of salaries, while type of LGA did not have a significant effect. 4 For a review of performance based incentives potential see Eichler, R., Levine R. and the Performance Based Incentives Working Group. 2009. Performance Based Incentives for Global Health: Potential and Pitfalls. Center for Global Development. Washington, DC. CHAPTER 7 Clients Providers T o improve service delivery community members have two different routes; a "long route" by exercising pressure to their elected officials for them to ensure that providers offer quality services, and a "short route" by increasing their power over the provider. The previous chapters described some of the shortcomings clients face to improve services through the "long route" of accountability (World Bank, 2004). As discussed previously, some of the reforms needed to improve the "long route" of accountability go beyond the health sector. For instance, there is an urgent need for civil service reform at local government level; a reform that will decrease the size of civil servants under the LGs payroll, change the skill mix of the personnel, and change the incentive structure faced by health providers. There is also a need to build capacity in public financial management at the local government levels. These reforms will take time and are difficult to implement. Therefore, to make significant improvements in PHC in the country, it is essential to improve the clients' "power" in the delivery of services. Increasing client's power can results in improvements in service delivery but is not a panacea, as there are important market failures that affect health services and in particular clinical services. There are information asymmetries between patients and health personnel, as the latter know more about the patients' diagnosis and treatment. This reduces the effect of the short route of accountability. Increasing Clients' Power One mechanism to increase clients' power is through their direct involvement in co producing and monitoring health services (World Bank, 2003). This chapter will look precisely at these existing mechanisms that community members have to exercise power in relation to providers. The Nigerian government has long recognized the importance of community participation in the delivery of basic health care services and has thus tried to involve the communities in the development of PHC along the lines of the Bamako Initiative. Indeed, the guidelines for the development of the PHC system (NPHCDA, 2004) establish the development of the following health committees to support activities at village and ward level: Villages/Community Development Committees, Ward Development Committees, and LG Development Committees. All these committees are involved in many needed health activities, although not necessarily in the management of facilities. According to the PHC guidelines, these are some of the roles and responsibilities of the Village/Community Development Committees regarding health facilities 74 Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 75 (mainly health posts and dispensaries): (i) determine exemption of drug payment and deferment; (ii) determine the pricing of drugs; (iii) supervise and monitor the quantity of drug supply; (iv) supervise all account books; and so forth. Among the roles and responsibilities of the Ward Development Committees the following affect the management of facilities: (i) take active role in the supervision and monitoring of Ward Drug Revolving Funds/ Bamako Initiative; (ii) supervise activities of Village Health Workers and CHEWs; (iii) monitor activities at both the health facilities and village levels; (iv) oversee the functioning of health facilities in the Wards; (v) monitor equipment and inventory of monthly intervals; and (vi) ensure the proper functioning of the health facilities using a maintenance plan. Some of these health committees have existed for many years; however, many are inactive. To assess the community involvement in the management of health facilities and in the monitoring of frontline providers the household survey collected information on the existence of these community management/development committees and their role in the functioning of PHC facilities. Survey Results1 Half of all PHC facilities in the country have or are linked to a community health development/management committee (table 7.1). These committees are present in two thirds of public facilities and in less than a third of privately managed ones. Most facilities in Kaduna and Cross River have management/development committees, while in Bauchi and Lagos not many have. Table 7.1. Percentage of Health Facilities with a Functioning Health Management/Development Committee and Gender of Committee Members across States, and across Facility Ownership Bauchi Cross River Kaduna Lagos Public Private Total Management/ development committee 40% 71% 75% 26% 67% 26% 51% Male members 9.6 6.3 16.4 4.3 11.6 4.8 10.2 Female members 4.1 4.3 3.2 4.5 4.2 3.1 3.9 Source: Health Facility Survey (EPOS, CISH, CHESTRAD, 2007). The majority of the members of these committees are men, with exception of Lagos state where, on average, there is the same number of women and men in these committees. Most of the members of these committees are selected by the community head or through an election in the community. Nevertheless, on average there are at least four women in these health committees. Indeed, according to the NPHCDA guidelines, a representative of women associations/groups should be a member of the community health committee. Most health committees meet at least once a month (table 7.2). In Bauchi, however, 30 percent of these committees only meet a few times a year. 76 World Bank Working Paper Table 7.2. Frequency of Meetings of Health Committees across States Bauchi Cross River Kaduna Lagos Total At least once a month 68.8 92.9 71.7 76.2 79.4 A few times a year 31.3 5.4 19.6 19.1 16.8 Once a year 0 1.79 8.7 4.76 3.87 Source: Health Facility Survey (EPOS, CISH, CHESTRAD, 2007). In all states but Lagos, community health committees only have a limited involvement in the facility management (table 7.3). Most of this involvement is in the request of vaccines and in the maintenance of facilities. Some of them also intervene in solving administrative and staff issues. In Lagos, even though only few facilities have a community management/development committee, these committees are very active and intervene in many different activities and decisions. Table 7.3. Actions of Community Health Management/Development Committees across States and Facility Ownership (in %) Bauchi Cross River Kaduna Lagos Public Private Total Action 33 28 38 90 34 61 41 Procurement of drugs 11 4 23 81 11 56 21 Fixed price of drugs 11 4 21 71 11 47 19 Fixed user charges 6 9 20 81 12 50 21 Requested more vaccines 47 52 60 76 58 51 56 Maintenance of facility 51 48 36 86 48 58 50 Provided fuel 0 23 15 86 15 56 24 Repaired equipment 9 27 26 90 23 58 31 New investment 3 23 21 76 16 56 25 Solved administrative issues 37 34 62 90 48 61 50 Solved staff issues 46 37 62 86 50 64 53 Source: Health Facility Survey (EPOS, CISH, CHESTRAD, 2007). Despite the large percentage of facilities with management committees and their involvement in some managerial issues, others make the final decision (table 7.4). In the facility survey, facility heads were asked who takes the final decision on hours of operation of the facility, new construction, use of IGR and others. Most facility heads responded that both the LGAs and facility heads were the main decision makers. For instance, the LGA was listed as the main decision maker for new construction, the acquisition of new equipment, the transfer of staff. The facility head was reported as main decision maker for the facility hours of operation, making drugs and supplies available, setting user charges, use of IGR and taking disciplinary actions. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 77 Table 7.4. Final Decision on Health Facility Managerial Issues (in %) Facility head LGA Facility hours of operation 73 19 New construction 38 61 Acquire new equipment 42 54 Make drugs available in facility 57 36 Making medical supplies avail 49 43 Setting charges for drugs 57 34 Setting charges for treatment 59 29 Use of IGR funds 56 37 Taking disciplinary action 52 46 Transfer of staff 30 58 Source: Health Facility Survey (EPOS, CISH, CHESTRAD, 2007). In summary, with exception of facilities in Bauchi, most public facilities sampled in the survey worked closely with health committees that met at least monthly. However, the involvement of these committees in the management of facilities is rather limited, as most decisions are taken by either the facility head or by the LGA. This is not surprising as many of these committees were created to support health activities in general but did not have a strong mandate to participate in the facility management. In particular, the community health development committees as set up in the national guidelines are not directly involved in the management of health facilities. The Ward Development Committees, in contrast, are supposed to oversee the functioning of the facilities in the Ward. But the guidelines do not specify what this oversight role implies and what power would these committees have to impose and enforce sanctions. Another mechanism to improve client's power in relation to providers is by making the provider's income depend on the demand of clients, particularly poor clients (World Bank, 2003). This is what patients do in private facilities. As seen previously, often in public facilities patients also pay for services. However, this does not always give patients power in relation to providers, especially if no other options or providers are available. Only when client's payments directly affect the income of the provider can these payments create the incentives for providers to offer quality services. When these payments are retained by the public provider and are reinvested in the facility or in the payment to frontline providers they can produce significant improvements in service provision. In Nigeria, most services provided by public health facilities have fee charges. These charges, however, have not increased the power of clients, as the facilities and health personnel cannot retain these revenues and use them for any improvements. These resources are sent back to the local government as they are considered part of their internally generated revenue. Possible Ways Forward Initiatives to revitalize health committees and to ensure their participation in the management of health facilities have recently started. In Kaduna, the SMOH, with the support of DFID financed project PATHS, is implementing an initiative to build capacity in PHC health committees so that they can play a more prominent and 78 World Bank Working Paper proactive role in health and to ensure that the community voices "can be heard by health providers and the government" (Operation Manual for Health Facility Committees in Kaduna State). PATHS has also supported similar initiatives in Ekiti and into less extent in Jigawa, Kano, and Enugu. The Kaduna Facility Health Committee Strengthening Initiative centers the role of the Committee around the health facility so that it can support the facility work and link it with the nearby community. In particular, the role of these committees are to: (i) support the health facilities to deliver services; (ii) increased access, particularly of the very poor to services; (iii) monitor the work of the facility; (iv) advocate for increased government support for the facility; (v) help build good relationship between facility and its catchment communities; (vi) and be the first point of contact for all services (Operation Manual for Health Facility Committees in Kaduna State). To support this initiative, the state has drafted detailed operational and training manuals for facility health committees. These committees have been revitalized in most of the state and some have started to produce results (see box 7.1). Box 7.1. Kaduna: Example of Facility Health Committee Role in Improving the Condition of PHC Facilities "Our local health facility lost some land when a dual carriageway was constructed. The state government planned to compensate local government for the loss. We lobbied the state government and asked to be paid the compensation directly. We wanted to avoid local government getting the cheque because there would be long bureaucratic delays in moving ahead with the building...The funds were released in September 2007. By early December 2007 we had renovated large parts of the health facility, built a new delivery ward, fenced the facility, installed a new water tank, and dug a pit latrine for patients. The new ward is bigger and better than what was originally in place. It was not easy to make the argument. We had to use impressive people for this... The Local Government PHC Co-ordinator assisted a lot. Everybody knew that if we got the money we could do a lot for the clinic. We submitted an expenditure report to local government in early December 2007 and have arranged for the Local Government Chairman to come and inspect the building work." Babban Dodo PHC Facility Health Committee, Kaduna Source: PATHS Technical Brief. The initiative in Kaduna is meant to increase "client's power" in relation to providers not only through the facility health committees' (FHC) participation in the management and monitoring of the facilities but also through encouraging clients complaints and redress mechanisms. The FHC in the states are encouraged to set up suggestion boxes, establish formal systems for client complaints, and undertake surveys of client satisfaction. The members of the revitalized FHC have also been trained to advocate in front of policy makers, in particular those that control the budgets, for issues affecting the performance of the PHC. Many states have started to implement programs to offer "free" services to women and children. This policy can provide an opportunity to make the income of providers depend more on the services they provide. The subsidy could be paid directly to the client through vouchers and not to the provider as has been done until now. Vouchers, as other demand side subsidies, can be costly. They need to be produced and distributed, providers need to be contracted, monitored and reimbursed, and so forth Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 79 (World Bank, 2005). Given this cost, it might only be possible to follow this policy in the cases where the benefit would be highest. This benefit will be highest when there is competition in the service provision, when there are multiple service providers and when the vouchers can be used in all available or accredited providers, including private providers (World Bank, 2005). In many urban and semi urban areas in Nigeria there are multiple providers, both public and private. By subsidizing the demand and giving patients a choice of providers, vouchers create incentives among providers to improve service delivery; otherwise patients might decide to go somewhere else for services. Vouchers are increasingly being used in many developing countries to improve access and quality of services (see box 7.2). Box 7.2. Experience with Vouchers for Health Services In Nicaragua vouchers have been used to increase access to sexual workers to STI treatment. Between 1996 and 1999 vouchers were delivered in six occasions and between 29 percent and 44 percent of these vouchers were used. The incidence of gonorrhea dropped 71 percent among voucher users. Also in Nicaragua between 2000 and 20002 in poor neighborhoods of Managua, a voucher program was used to increased adolescent access to reproductive health services. Perceived quality of these services as well as access to contraceptives increased. In Tanzania a voucher program started in 2000 to increased access to insecticide treated bed nets. By 2006, this program, called "Hati Punguzo," had achieved a 60 percent coverage of the targeted population and surpassed the key milestone of over one million redeem voucher. In Kenya, a pilot project has started to provide safe motherhood, family planning, and gender violence recovery services to disadvantaged people in three rural districts (Kisumu, Kiambu, and Kitui) as well as two urban slums in Nairobi. The program works through vouchers that can be redeem in certified public, private, and faith based facilities. This program started in 2006. By May 2007, more 38,000 vouchers have been purchased and 15,000 claims were purchased. In Uganda, a project has started to provide vouchers sold at a nominal fee to entitle the holder to the whole treatment associated with safe child birth or the treatment of STIs. 150,000 households in the greater Mbarara region in western Uganda are expected to benefit from this project. Source: Meuwissen et al. (2006a); Jones et al. (2006); Meuwissen, et al. (2006b); Weller, S.; and http://www.output- based-aid.net/. In places where patients have no choice of provider, vouchers could still offer some benefits as they give the existing provider an incentive to offer more services, although not necessarily to increased the patient's perception of quality of these services (World Bank, 2005). Vouchers are expensive to manage, but in the case of services that are relatively easy to monitor and quantified as it is the case of some standardized preventive services such as immunization and pre natal care and where the recipients of the vouchers are easy to identified, such is the case of demographic targeting (that is, pregnant women and children under five), the benefits from these vouchers might still outweighs their costs. Nevertheless, as seen in a previous chapter, most PHC facilities are in very poor condition and do not have the equipment, supplies, and drugs needed to offer services. In many areas, these facilities are the only provider households have access to. In this case, a demand subsidy would need to be complemented with a rehabilitation and equipment program in public facilities. 80 World Bank Working Paper Finally, community insurance schemes can also increase the client's power in front of the providers. They can contribute to health care costs and increase utilization (Carin et al., 2005). These schemes buy services in bulk from the facilities, increasing thus the power of the community in relation to providers. There are already some functioning community based health insurance schemes in Nigeria, although at the moment they only cover a very small percentage of the population. Note 1 General Note: The discussion on the results of the Health Facility survey comes from the Final Report for this survey prepared by EPOS Health Consultants, Canadian Society for International Health (CSIH), and Center for Health Sciences Training, Research and Development (CHESTRAD). References Bennet, S. and L. M. Franco. 1999. "Public Sector Health Worker Motivation and Health Sector Reform: A Conceptual Framework." Major Applied Research 5, Technical Paper 1. Bethesda, MD: Partnerships for Health Reform Project, Abt Associates Inc., January. Bird, R. 2000. "Transfers and Incentives in Intergovernmental Fiscal Relations." In S.J. Burki and G. Perry, eds., Decentralization and Accountability of the Public Sector, Annual World Bank Conference on Development in Latin America and the Caribbean, pp. 111­25. Washington, DC: The World Bank. Brinkerhoff, Derick W. 2004. "Accountability and Health Systems: Toward Conceptual Clarity and Policy Relevance." Health Policy and Planning 19.6: 371­379. Carrin, G, M. Waelkens, and B. Criel. 2005. "Community based Health Insurance in Developing Countries: A Study of its Contribution to the Performance of Health Financing Systems." Tropical Medicine and International Health 10(8): 779­811. Cross River State Auditor General of Local Governments. 2006. Report of the Auditor General of Local Governments on the Accounts of the Local Governments of Cross River State For the Year ended 31st December, 2005. Calabar, Cross River State, March. ------. 2007. Report of the Auditor General of Local Governments on the Accounts of the Local Governments of Cross River State For the Year ended 31st December, 2006. Calabar, Cross River State, March 2007. Das Gupta, M., V. Gauri, and S. Khemani. 2003. "Decentralized Delivery of Primary Health Care: Survey Evidence from the States of Lagos and Kogi." Africa Region Human Development Working Paper Series. World Bank, Washington, DC. Destefano, J., and L. Crouch. 2005. Education Reform Support Today. Equip 2 Project. United States Agency for International Development. Washington, DC. Federal Ministry of Health and World Bank. 2005. Nigeria Health, Nutrition, and Population Country Status Report. Washington DC: World Bank. Eichler, R., Levine R. and the Performance Based Incentives Working Group. 2009. Performance Based Incentives for Global Health: Potential and Pitfalls. Center for Global Development. Washington, DC. EPOS Health Consultants, Canadian Society for International Health and Center for Health Sciences Training, Research and Development. 2007. Study on Primary Health Care in Four States of Nigeria. December. Meuwissen, L., A. Gorter, and J.A. Knottnerus. 2006. "Perceived Quality of Reproductive Care for Girls in a Competitive Voucher Programme. A Quasi Experimental Intervention Study, Managua, Nicaragua." International Journal of Quality Health Care 18(1): 35­42. 81 82 World Bank Working Paper L. Jones, J. Quigley, and G. Foster. 2006. "Paupers, Princes and Paper: Vouchers Revisited--Can Small Enterprises Save Government Programs?" Small Enterprise Development Journal 17(4): 1­9. Federal Government of Nigeria. 1999. Constitution of the Republic of Nigeria. Abuja, Nigeria. ------. 2004a. Draft National Health Bill 2004. Draft, Abuja, Nigeria. ------. 2004b. Nigeria: National Economic Empowerment and Development Strategy (NEEDS). National Planning Commission, Abuja. ------. 2004c. Revised National Health Policy. Federal Ministry of Health. Abuja, Nigeria, September. Kaduna Auditor General of Local Governments. 1999. Report of the Auditor General of Local Governments on the Accounts of the 23 Local Government Councils of Kaduna State For the Year ended 31st December, 1999. Presented to the Kaduna State House of Assembly. ------. 2000. Report of the Auditor General of Local Governments on the Accounts of the 23 Local Government Councils of Kaduna State For the Year ended 31st December, 2000. Presented to the Kaduna State House of Assembly. ------. 2001. Report of the Auditor General of Local Governments on the Accounts of the 23 Local Government Councils of Kaduna State For the Year ended 31st December, 2001. Presented to the Kaduna State House of Assembly. ------. 2002. Report of the Auditor General of Local Governments on the Accounts of the 23 Local Government Councils of Kaduna State For the Year ended 31st December, 2002. Presented to the Kaduna State House of Assembly. ------. 2003. Report of the Auditor General of Local Governments on the Accounts of the 23 Local Government Councils of Kaduna State For the Year ended 31st December, 2003. Presented to the Kaduna State House of Assembly. ------. 2004. Report of the Auditor General of Local Governments on the Accounts of the 23 Local Government Councils of Kaduna State For the Year ended 31st December, 2004. Presented to the Kaduna State House of Assembly. Kaduna State Ministry of Health. 2007a. Operational Manual for Facility Health Committees in Kaduna: Primary Health Care. Manual prepared with the support of UK DIFD Partnerships for Transforming Health Systems Program (PATHS). Kaduna, Nigeria, compiled in December 2006 and updated in December 2007. ------. 2007b. Training Manual for Facilities Health Committees. Manual prepared with the support of the UK DIFD Partnerships for Transforming Health Systems Program (PATHS). Kaduna, Nigeria, November. Kaduna State Ministry of Local Government. 2004. Operational Guidelines for Development Area Management Committee. Kaduna State. Khemani, S. 2005. "Local Government Accountability for Health Service Delivery in Nigeria." Journal of African Economies. Oxford University Press. Lewis, M. 2006. "Governance and Corruption in Public Health Care Systems." Working Paper No. 78. Center for Global Development, Washington, DC. Loevinsohn, Benjamin. 2008. "Performance Based Contracting for Services in Developing Countries: A Toolkit." World Bank Health, Nutrition, and Population Series. World Bank, Washington, DC. Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 83 Meuwissen, L., A. Gorter, A. Kester, and J.A. Knottnerus. 2006. "Does a Competitive Voucher Program for Adolescents Improve the Quality of Reproductive Health Care? A Simulated Patient Study in Nicaragua." BMC Public Health 6: 204. National Population Commission, Nigeria and ORC Macro. 2004. Nigeria Demographic and Health Survey 2003, Calverton, Maryland. National Primary Health Care Development Agency (NPHCDA). 2007. Ward Minimum Health Care Package. Abuja, Nigeria. ------. 2004. Ward Health Service Operational Guide. Abuja, Nigeria, October. NPHCDA and FMOH. 2004. Operational Training Manual and Guidelines for the Development of Primary Health Care System in Nigeria. Abuja, Nigeria. Partnership for Transfoming Health Systems (PATHS), Health Partners International (HPI). Technical Briefs. HPI, Lewes, East Sussex, United Kingdom. Samuel, Paul. 2004. "Citizen Report Cards: An Accountability Tool." Development Outreach. World Bank Institute, March. Terfa Inc. 2005. Scorecard Assessment of Rural Local Governments in Nine States of Nigeria (Volumes I and II. Report submitted to The National Coordinator Federal Program Support Unit. Federal Ministry of Environment. Abuja, Nigeria, October. Van Lerberghe, W., Cl Conceicao, W. Van Damme, and P. Ferrinho. 2002. "When Staff Is Underpaid: Dealing with the Individual Coping Strategies of Health Personnel." Bulletin of the World Health Organization 80(7): 581­584. Weller, S. 2005. "Impact of Health System Reform on Reproductive Health in Public Private Associations." Initiative for Sexual and Reproductive Rights in Health Sector Reforms: Latin America. World Bank. 2002. State and Local Governance in Nigeria. Africa Region, Report No. 24477 UNI. Washington, DC: World Bank. ------. 2003. World Development Report 2004: Making Services Work for Poor People. Washington, DC: World Bank and Oxford University Press. ------. 2005. A Guide for Competitive Vouchers for Health. Washington, DC: World Bank. ------. 2006. Making Services Work for the Poor in Indonesia: Focusing on Achieving Results on the Ground. Washington, DC: World Bank. ------. 2007. Nigeria, A Fiscal Agenda for Change: Public Financial Management and Financial Accountability Review (PEMFAR). African Region Report No. 36496 NG in two volumes. Washington, DC: World Bank. ------. Forthcoming. Nigeria Poverty Assessment. Washington, DC: World Bank. World Bank Group and the Department of International Development. 2005. Country Partnership Strategy for the Republic of Nigeria (2005 2009). Report No. 32412 NG, Washington, DC. World Health Organization. Maternal Mortality in 2005: Estimates developed by WHO, UNICEF, UNFPA, and The World Bank. Geneva, 2007. Appendixes Appendix A: Sample Size Source: EPOS, CSIH, CHESTRAD (2007) Health Facility Survey Assumptions for Calculation The proportion of facilities with drugs and equipment was 20%, A level of significance of 5%, An estimate of 250 facilities per state An absolute deviation of 10% Sample Size Determination 1. Using software from the Center for Disease Control (CDC) in Atlanta (epi info 6), a minimum sample size of 49 facilities was arrived at. This was multiplied by 1.5 to arrive at a sample size of 75 facilities per state. A design effect of 2 was used for Lagos state, which gives 98 facilities in Lagos state. Sampling Procedures A stratified, multistage random sampling method was employed. In each state, a list of LGAs was prepared. The LGAs were then stratified into rural, urban and semi urban LGAs One rural, one urban and one semi urban LGA was chosen at random (using a table of random numbers or by balloting) In the selected LGAs, a list of all Primary Health Centers (PHCs) was compiled based on private and public sector stratification. The sample size of 75 is divided into three, based on the number of PHCs in the selected rural/ urban/semi urban LGAs, that is, probability proportional to size. Let the PHCs in the rural LGA=A , the PHCs in the urban LGA=B and the PHCs in the semi urban LGA=C No. of PHCs selected in the rural LGA = no of PHCs in A x 75 A+B+C Number of PHCs selected in the urban LGA = no of PHCs in B x 75 A+B+ C 85 86 World Bank Working Paper Number of PHCs selected in the semi urban LGA = no of PHCs in C x 75 A+B+ C The number to be selected in the private and public sectors in the rural, urban and semi urban LGAs was also determined based on the stratification by public and private sectors in each of the selected rural/urban/semi urban LGA. These PHCs were selected using simple random sampling by a table of random numbers. Justification for Sampling Method A stratified multistage random sampling method was employed based on the rural/ urban/semi urban and public/private sector stratification Probability proportional to size was used to divide the PHCs based on the rural/urban/semi urban and public/private sector division Sampling was done in 3 stages, hence the multistage approach. Household Survey Assumptions for Calculation Proportion of people using the nearest health facility = 10%, A level of significance of 5%, A population of 16500 is assumed for each LGA. An absolute deviation of 10% A non response rate of 15% A design effect of 2 Sample Size Determination Using CDC software, epi info 6, a minimum sample size of 137 households was arrived at. This was adjusted for non response (15%) and a design effect of 2. Therefore the sample size became 400 households per state. Sampling Procedures A stratified, multistage random sampling method was employed. In each state, a list of LGA was prepared The LGAs were then stratified into rural, urban and semi urban LGAs One rural, one urban and one semi urban LGA was chosen at random (using a table of random numbers or by balloting) In each stratum, that is, rural /urban/semi urban, a list of Enumeration Areas (EAs) was prepared The sample size of 400 was divided into three, based on the number of households in the selected EAs in the rural/urban/semi urban stratum: that is, probability proportional to size. that is, Let the EAs in the rural LGA=A, the EAs in the urban LGA=B, and the EAs in the semi urban LGA=C No. of households selected in rural EAs= no of households in A x 400 A+B+C Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 87 No. of households selected in urban EAs= no of households in B x 400 A+B+C No. of households selected in semi urban EAs= no of households in C x 400 A+B+C A list of the households in the selected EAs was prepared. The number of households was selected using simple random sampling by a table of random numbers. Since we were selecting 75 facilities per state, and 400 households, we required 5 households per selected facility. Justification for Sampling Method We defined coverage in terms of households within 10km radius from the selected facilities A stratified multistage random sampling method was employed based on the rural/ urban/semi urban and public/private sector stratification Probability proportion to size was used to divide the PHCs based on the rural/urban/semi urban and public/private sector ratios Sampling was done in 3 stages, hence the multistage approach. 88 World Bank Working Paper Appendix B: Household Survey Sample Characteristics Source: EPOS, CSIH, CHESTRAD (2007) Household Characteristics Seventy seven percent of household heads--the selected respondent for the survey-- were male. When the results were tabulated by gender, there was little difference in any of the key variables, suggesting that the household head consulted other family members in responding to the survey questions. The median age of the respondent was 37; 82 percent were married with 71 percent having one wife. Sixty one percent of respondents were self employed and 18 percent civil servants. The most frequent occupation was farmer (36 percent) followed by skilled worker (23 percent). Sixty percent of the respondents reported themselves to having good literacy skills, while 69 percent of respondents reported having some secondary education or less. Most respondents (70 percent) were indigene to their community. Both literacy and education percentages decline by age, as measured by age groups (20­44, 45 64, 65+) of the respondents. The largest household sizes were found in the 45­64 age group. Figure B.1. Household Survey Figure B.2. Household Survey Respondents' Literacy Level Respondents' Education Status Literacy Level of Education 22% 20% 16% 11% 60% 18% 23% 30% Some secondary none Some primary some No formal good Post-secondary Arabic/Koranic Table B.1. Employment and Occupation of Household Survey Respondents Employment % Occupation % unemployed 9 Farmer 36 self employed 61 Skilled worker 23 civil servant 18 Business 14 Informal employment 4 Petty trader 12 Improving Primary Health Care Delivery in Nigeria: Evidence from Four States 89 Housing Characteristics of Household Respondents The majority of households lived in basic housing conditions with most using all rooms but one as sleeping rooms. Thirty five percent of households did not have electricity, 22 percent had a toilet, and 57 percent used open dumping for refuse. Household water supply was from stream/river in 13 percent of the cases and from an unsanitary well in 14 percent of households. Table B.2. Housing Characteristics of Household Survey Respondents Electricity % Public source irregular 60 No Electricity 35 Other 5 Toilet Flush toilet 22 Latrine 54 No toilet- bush hole 24 Refuse Open dump 57 Controlled dumping 28 Water Stream/river 13 Well - sanitary 20 Well - unsanitary 14 Bore hole 29 Pipe borne 18 Household Survey Respondents' Proximity to Health Facilities As indicated earlier, respondents stated that their nearest health facility was owned by LGA (51 percent), state (20 percent), federal government (2 percent), and a private sector facility (26 percent). Eighty two percent of respondents were within a 30 minute walk of the facility with a further 15 percent between 30 60 minutes. In terms of facility type, 26 percent were health post, 42 percent basic health center, 9 percent comprehensive health center with the remaining 21 percent hospital. As noted above, the reader should be aware that the hospital percentage was over stated due to classification and coding errors. The majority of respondents (77 percent) stated they patronized their nearest health facility. Respondents who did not use their nearest facilities stated it was due to the facility "not being well equipped" (this reason decreased with facility type) or the facility being "too expensive" (this reason increased with facility type). 90 World Bank Working Paper Table B.3. Household Survey Respondents' Proximity to Nearest Health Facility Ownership % Local 60 State 15 Federal 3 Private 21 Distance <30 min walk 82 ½ -1 hour walk 13 > 1 hour walk 5 Type of Facility Specialist hospital 10 General hospital 11 Comprehensive Health Center 9 Basic Health Center 32 Dispensary/Health Post 26 Maternity 10 Patronize nearest PHC 77 Eco-Audit Environmental Benefits Statement The World Bank is committed to preserving Endangered Forests and natural resources. We print World Bank Working Papers and Country Studies on postconsumer recycled paper, processed chlorine free. The World Bank has formally agreed to follow the recommended standards for paper usage set by Green Press Initiative--a nonprofit program supporting publishers in using fiber that is not sourced from Endangered Forests. For more information, visit www.greenpressinitiative.org. In 2008, the printing of these books on recycled paper saved the following: Net Greenhouse Trees* Solid Waste Water Total Energy Gases 289 8,011 131,944 27,396 92 mil. *40 feet in height and Pounds CO2 6­8 inches in Pounds Gallons BTUs Equivalent diameter Improving Primary Health Care Delivery in Nigeria is part of the World Bank Working Paper series. These papers are published to communi- cate the results of the Bank's ongoing research and to stimulate pu- blic discussion. This paper, based on quantitative surveys at the level of primary health care facilities, health care personnel, and households in their vicinity, aims at understanding the performance of primary health care providers in four states in Nigeria. As possible ways to improve performance, the paper concludes that clearly defining lines of res- ponsibility, implementing performance-based financing of local go- vernments and providers, and collecting, analyzing, and sharing information are some options that can help realign incentives and improve accountability in the service delivery chain and service pro- vision. This working paper was produced as part of the World Bank's Africa Region Health Systems for Outcomes (HSO) Program. The Program, funded by the World Bank, the Government of Norway, the Govern- ment of the United Kingdom, and the Global Alliance for Vaccines and Immunization (GAVI), focuses on strengthening health systems in Africa to reach the poor and achieve tangible results related to Health, Nutrition, and Population. The main pillars and focus of the program center on knowledge and capacity building related to Human Resources for Health, Health Financing, Pharma-ceuticals, Governance and Service Delivery, and Infrastructure and ICT. More information as well as all the products produced under the HSO program can be found online at www.worldbank.org/hso. World Bank Working Papers are available individually or on standing order. This World Bank Working Paper series is also available online through the World Bank e-library (www.worldbank.org/elibrary). ISBN 978-0-8213-8311-7 WORLD BANK 1818 H Street, NW Washington, DC 20433 USA Teléphone: 202 473-1000 Site web: www.worldbank.org E-mail: feedback@worldbank.org SKU 18311