Report No. 34177-NG Nigeria Health, Nutrition, and Population Country Status Report (In Two Volumes) Volume II: Main Report November 2005 Africa Region Human Development The Federal Ministry of Health The World Bank Group Nigeria Document of the World Bank Acknowledgements This Country Status Report (CSR) is ajoint product of the Federal Ministry of Health in Nigeria and the World Bank. The Marginal Budgeting for Bottlenecks section was undertaken in close collaboration with UNICEF. The Nigeria MOH team was headed by H.E. Professor Eyitayo Lambo (Minister o f Health), and was composed by Dr. Shehu Sule (Director o f Department o f Health Planning and Research), Dr.Tolu Fakeye (Head Division o f International Health, Dept.o f Health Planning and Research), Dr. M. Lecky (previous, Head Division o f International Health, Dept. o f Health Planning and Research), Dr. N. Azodoh (Dept. o f Health Planning and Research) and also included staff from the various M O H departments and agencies, as well as representatives from the private sector and universities. From the World Bank side, Christine Lao Peiia and Maria Eugenia Bonilla-Chacin coordinated the over-all CSR work. The World Bank CSR team i s composed o f Patrick Mullen, Anne Okigbo, Feng Zhao, Ulrika Enemark, and Edit Velenyi. Invaluable comments and suggestions were received from peer reviewers: Agnes Soucat, Christopher Walker, and Maureen Lewis as well as members o f the Nigeria country and the World Bank Human Development teams including Galina Sotirova, Eva Jarawan, Ok Pannenborg, Alexander Preker, Stephan Legros, Keith Hinchliffe, Jane Miller, and Victoria Kwakwa. Important inputs were received from Genevieve Begkoyian, Rudolph Knippenberg, James Patterson, Ephraim Kebede and Fagbemi Omoniyi. Abiodun Elifuoye and Therese Tshimanga provided very good logistical and document processing support. Laura Frigenti and Hafez Ghanemprovided over-all guidance. This report has been presentedand has benefitedfrom comments and suggestions in(1) a two-day CSR workshop organized in Nigeria in June 2005 which was attended by key MOH, federal health parastatal organizations, and donor agencies including DFID, WHO, UNICEF, CIDA, EU, and UNFPA; (2) a second two-day CSR workshop organized inNigeria inNovember 2005 which was attended by the FMOH, federal health parastatal organizations, and CIDA; and (3) discussions with the Minister and his Management Team in June 2005. Site visits and discussions were also conducted during the preparation o f this report in Abuja, Lagos, Jos, and Nasarawa. i Tableof Contents Introductionand Context .............................................................................................................. 1 Introduction ................................................................................................................................. 1 Context ........................................................................................................................................ 1 Chapter 1 . HealthOutcomes .................................................................................................. 5 ChildHealth ................................................................................................................................ 5 Childnutritional status .............................................................................................................. 10 Causes of child mortality and morbidity ................................................................................... 14 Maternal and Reproductive Health............................................................................................ 15 Women's Nutritional Status ...................................................................................................... 17 HIV/AIDS and other Sexually TransmittedDiseases ............................................................... 18 Other Diseases Prevalent inNigeria.......................................................................................... 19 Non-Communicable Diseases ................................................................................................... 20 the Millennium Development Goals ......................................................................................... Chapter 2 . Householdand Community FactorsandHealthService Utilization ............21 25 Access to improved sources of water and sanitation................................................................. 25 Householdbehavior and actions affecting child health and nutrition ....................................... 26 Maternal and ReproductiveHealth............................................................................................ 34 HIV/AIDS Knowledge .............................................................................................................. 39 Morbidity and Care Seeking Behavior ...................................................................................... 39 Reasons for not usinghealth services ........................................................................................ 42 Chapter 3 . ........................................................................... Health System Organizationand Governance........................................................................... The HealthSysteminNigeria 45 45 Health Service Delivery System................................................................................................ 51 Health Services and Programs................................................................................................... 73 Health Sector Strategy............................................................................................................... Chapter4 . The Role of the PrivateSector inHealthCareProvisioninNigeria .............92 85 Private health clinics and hospitals............................................................................................ 92 Pharmaceutical retail ................................................................................................................. 99 Purchasers o f private sector health services ............................................................................ 103 104 Key challenges. constraints and opportunities ........................................................................ Institutionalcontext ................................................................................................................. 109 Strategic options ...................................................................................................................... 110 Chapter 5 . HealthCareFinancinginNigeria .................................................................. 114 Economic Context ................................................................................................................... 114 Public Sector Health Spending................................................................................................ 114 123 Total Health Spending............................................................................................................. Private Sector Health Spending............................................................................................... 132 Health Financing Strategy ....................................................................................................... 134 Chapter 6 . ExtraResourcesNeededto Achievethe HealthRelatedMDGs ..................137 Applyingthe MBBtool inNigeria: Process andmethodology.............................................. 137 Examinationo f the health system............................................................................................ 138 138 Identification o f bottlenecks inthe delivery o f key health care interventions......................... Identification o f highimpact interventions............................................................................. 140 Formulation of a strategy to eliminate these constraints ......................................................... 143 Cost and impact....................................................................................................................... 149 Sources .......................................................................................................................................... ........................................................................................................................................ 153 Annex 161 ... 111 List of Tables, Figuresand Boxes Table 1.Neonatal, Infant and under five mortality across geopolitical regions, Nigeria, 1993-2003 .............7 Tables Table 2. Results from a Cox proportional hazards model o f infant and child survival in Nigeria using the N D H S 2003 ...................................................................................................................................,.9 Table 3. Children's nutritional status ina sample o f SSA countries ............................................................. 10 Table 4. Nutritional status across regions, Nigeria, 2003.............................................................................. 12 Table 5. Multivariate (probit) analysis o f determinants o f malnutritionusingNDHS 2003 ......................... 13 Table 6. Indicators o f maternal health inselected SSA countries ................................................................. 16 Figure 11. HIV/AIDS sero-prevalence over the years, Nigeria Table 7: HIV/AIDS sero-prevalence in 2003 .................................................................................................................. ..................... 18 Table 8. HIVIAIDS epidemic across the world........................................................... Table 9. HIV/AIDS epidemic scenarios inNigeria..................................................... Table 10. Meeting the MillenniumDevelopment Goals ...............................................................................23 Table 11. Treatment o f diarrhea, Nigeria, 1990-2003 (% o f children with diarrhea inprevious 2 weeks). ..28 Table 12. Immunization coverage, Nigeria, 1990-2003 (% children 12-23 months) Table 13. Percentage o f children taken to a health care provider across wealth qui Table 14. Percentage o f women who received professional antenatal and delivery care, and percentage o f women that delivered ina health facility, Nigeria....................................................... ......36 Table 15. Percentage o f people reporting an illness or injuryinthe last two weeks by age, Nig 04 .39 Table 16. Health service utilization by type o f provider, Nigeria, 2004 (% o f children and adults illor injured inthe previous two weeks who received care) (n= 7,028) ................. Table 17. Public sector health system service levels..................................................................................... 50 Table 18. Populationper health facility and proportion public sector, Nigeria, 2000..................................5531 Table 20. Functioning o f referral within wards, Nigeria, 2001 (% o f wards) (n=400 wards) ...................... Table 19. Health facility availability, Nigeria, 1999 (% o fhouseholds) ..................... ........................... Table 21. Registered pharmacies inNigeria, 2003 ................................,.................................,....................54 56 Table 23, Public sector PHC personnel, Nigeria, 2001 (per 100,000 population) ........................................63 Table 22. Professional situation o f registered doctors, Nigeria, 2003 ................................ Table 24. Traditional birthattendants and village health workers, Nigeria, 2001 (% o f villages) (n=674) ..64 Table 25. Availability o f essential drugs, Nigeria, 2001 (% o f facilities with different proportions o f essential drug package available) (n=674) ....................................................................................66 Table 26. Drug stock-outs, availability o f essential drugs list, and drug revolving funds in health facilities, Nigeria, 2001 (n=674) ................................................................................................................... 68 Table 27. Availability o f minimumpackage o f equipmentby region, Nigeria, 2001 (% o f facilities) (n=674 facilities) ......................................................................................................... Table 28. Availability o f minimum package o f equipment by level o f facility, facilities) (n=674 facilities) ......................................................... ..................................................70 Table 29. Availability o f obstetric and reproductive health equipment, Nigeria (% o f facilities) (n=4,503 facilities) ,........................................................................................,..............................,...............71 Table 30. Health management information system at the facility level, Nigeria (% o f facilities) (n=674 Table 31. Immunization schedule, Nigeria, 2005 ...............................................................................,.........75 facilities) .......................... ..............................................................................................................73 Table 32. Child health service availability, Nigeria, 2001 (% o f facilities) (n=674) .................................... Table 33. Participation innutritional programs, Nigeria, 1999-2002 (% children under-5) ........................76 76 Table 34. Reproductive and maternal health service availability, Nigeria, 2001 (% o f facilities) (n=674) ..78 Table 35. Quality o f antenatal care, Nigeria, 2003 (% o f women who received antenatal care) (n=2,462)..79 Table 36. Health sector component o f the 2004 NEEDS................................... .................................. 87 Table 37. Distribution o f CHAN member institutions by bed size (%) ........................................................ 94 Table 38. Categories o f services provided by CHANmember institutions (N=134) ........................ Table 39. Workload estimates (PDEiwork day) for health facilities inBenue state ...................... ...............97 Table 40. Comparison o f characteristics o f the not-for profit and for profit health care providers. ..............99 Table 41. GDP growth and GDP per capita, Nigeria, 2001-2005 ............................................................... 114 Table 42. Federal government health expenditures, Nigeria, 1998-2003 (in current and constant 2003 prices) ..........................................................................................................................................117 iv 119 Table 44. Implied average state health budgets by region, Nigeria, 2002.................................................. Table 43. State government health budgets. Nigeria. 2002 ......................................................................... 119 Table 45. Range o f estimates o fpublic sector health spending, Nigeria, 2002 .......................................... 122 Table 46. Estimates o f public sector health expenditures, Nigeria, 2003 ................................................... 122 Table 47. Health spending as % o f total household expenditures (household survey data) ........................ 125 Table 48. Health expenditure as percentage o f total household expenditure across expenditure quintiles, Table 49.Nigeria, 2004 ............................................................................................................................... 126 Health expenditure (excluding expenditure on inpatient care) as percentage o f total household Table 50.expenditure health expenditure across regions, Nigeria, 2004 (% o f households) ..................... across regions, Nigeria, 2004................................................................................... 126 Table 51. Catastrophic 127 Table 52. Average outpatient consultation expenditure across regions, Nigeria, 2004 (Naira) .................128 Average expenditure for outpatient consultations across regions, Nigeria, 2004....................... 128 Table 53. Average outpatient consultation expenditure across household expenditure quintiles, Nigeria, Table 54.2004 ............................................................................................................................................. Table 55.Average individual expenditure on health care services across regions, Nigeria, 2004 (Naira) .129 129 Average individual health care expenditure across population expenditure quintiles, Nigeria, 2004 (Naira) ................................................................................................................................ 130 Table 56. Percentage o f individuals that paid for vaccinations, post-natal, or pre-natal care across Table 57.household expenditure quintiles, Nigeria, 2004 .......................................................................... 130 Percentage o f children paying for last vaccination across places where the vaccination took place, Table 58.Nigeria, 2004 ............................................................................................................................... 130 Percentage o f individuals paying for post-natal care consultations across type o f facility, Nigeria, Table 59.2004 ............................................................................................................................................. 131 Table 60 Effective Neonatal, child, and maternal health interventions based on the Lancet Neonatal and .Estimates for private health expenditures, Nigeria, circa 2003-04 .............................................. 131 Child Survival Series and on a Cochrane review o f the literature on effective maternal health Table 61.interventions ................................................................................................................................ 139 Strategic Scenarios for Reaching the MDGs............................................................................... 144 Table 62. Highimpact interventions by service delivery mode ineach scenario........................................ 147 Table 63. Cost and Impact o f policy scenarios............................................................................................ 150 Table 64. Variables associated to the probability that a child o f less than six month o f age i s exclusively Table 65.breastfed associated to the probability that a child between 12 and 24 months i s immunized...161 (probit analysis) ........................................................................................................... Table 66. Variables 162 Variables associated to the probability that the child has received a supplementation o f vitamin A Table 67.in the last six months (probit analysis) ........................................................................................ 163 Variables associated with the probability o f seeking medical care in case a child has had diarrhea Table 68 .orVariables fevedcough inthe last two weeks (probit analysis) ................................................................ 164 associated with the probability that a women will go to more than two ante-natal care visits, be attended by trained personnel during delivery, and deliver in a health facility (probit analysis) ....................................................................................................................................... 165 Figures 3 Figure 2. Nigeria and global trend inunder-five mortality vs. GDP per capita ............................................. Figure 1.Relative poverty across regions ....................................................................................................... 5 Figure 3. Trends ininfant and child mortality, Nigeria................................................................................... 6 Figure 4. Infant and under five mortality across wealth quintiles, Nigeria, 1993-2003 .................................. 6 Figure 5. Pooresthichest ratio for infant and under five mortality .................................... Figure 6. Infant and under five mortality inurban and rural areas, Nigeria, 1993-2003................................. 8 Figure 7. Malnutrition across time and differences in percentages o f children underweight in rural and Figure 8. urban areas..................................................................................................................................... 11 Pooresthichest ratio o f the percentage o f children stunted and underweight in a sample o f SSA 11 Figure 9. Estimated Causes o f Mortality o f Children Under Five, Nigeria................................................... countries ........................................................................................................................................ 14 Figure 10. Reported causes o f maternal mortality, Nigeria, 1996................................................................. 17 Figure 11. HIViAIDS sero-prevalence over the years, Nigeria Table 7: HIViAIDS sero-prevalence in 2003 ............................................................................................................................................... 18 V Figure 12. Progress towards meetingthe MillenniumDevelopment Goals. Nigeria. 1990-2003 .................22 Figure 13 Source o fimproved water and sanitation across wealth quintiles. Nigeria. 2003........................ . 25 Figure 14. Access to improved water and sanitation facilities across regions and across rural and urban Figure 15.areas. Nigeria. 2003 ....................................................................................................................... 26 Figure 16 Effect o fmaternal education and region o fresidence inthe probability that a child is exclusively .Percentage o f children exclusively breastfed across years and months since birth...................... 27 breastfed ........................................................................................................................................ 28 Figure 17. Measles immunization coverage by GDP per capita (SSA countries with GDP per capita less Figure 18. Effect o f region o f residence and household wealth status on the probability that a child is than US$ 500) ................................................................................................................................ 30 Figure 19.immunized ..................................................................................................................................... 31 Figure 20 Effect o f maternal education and region o f residence on the probability o f receiving vitamin A .Immunization Concentration Curves ........................................................................................... 31 supplementation............................................................................................................................. 33 Figure 21. Current used o f modem contraception among currently married women across income, Figure 22.education, 35 Assistance during delivery (percentage o f births) across wealth quintiles................................... and across regions o f residence, Nigeria, 2003 ............................................................ 37 Figure 23. Effect o f women's education and household wealth on the probability o f more than two antenatal care visits, o f being assisted by skilled personnel during delivery, and o f delivering in a Figure 24. Knowledge o f methods to prevent HIV/AIDS across gender, income level, and place o f health facility, Nigeria 2003 .......................................................................................................... 37 Figure 25. Percentage o f people reporting and illness or injury in the last two weeks that visited a health residence........................................................................................................................................ 38 Figure 26.careprovider, by income quintile, Nigeria, 2004 .......................................................................... Figure 27. Place o f consultation and who was consultedby income quintile............................................... 40 Figure 28. Women's problems inaccessing health care across wealth status and place o f residence...........43 problems inaccessing health care across wealth status and place o f residence ...........42 Figure 29. Women's 44 Figure 30. Comprehensive Government funding flows to the health systeminNigeria........................................................ PHC centers and general hospitals per 100,000 population, Nigeria, 2001 .......46 52 Figure 31.Healthpostsiclinics/dispensaries per 100,000 population, Nigeria, 2001.................................... 55 Figure 32 Community health extension worker (CHEW) visits at least monthly, Nigeria, 1999 (% . Figure 33.households) .................................................................................................................................... 55 Figure 34.Primary health care (PHC) centers per 100,000 population, Nigeria, 2001................................. 56 Availability o f pharmacistsipatent medicine vendors, Nigeria, 1999 (% households living within Figure 35. Availability o f health service providers by socio-economic status, Nigeria, 1999 (??households 5 km) ............................................................................................................................................. 57 Figure 36.within 5 km) .................................................................................................................................. 58 Availability o f PHC facilities by socio-economic status and region, Nigeria, 1999 (% households Figure 37. Doctors and nursesimidwives per 100,000 population by GNI per capita, 2001 (countries with within 5 km) .................................................................................................................................. 58 Figure 38. Nursesimidwives per doctor by GNI per capita, 2001 (countries with GNI per capita under GNIper capita under US$1,500) ................................................................................................... 61 Figure 39.US$1,500) ...................................................................................................................................... 62 Figure 40.Gender distribution o fhealthpersonnel in5 states, Nigeria, 2003 (%) (n=930) ......................... 62 Health worker perceptions o f reasons for poor performance in 5 states, Nigeria, 2003 (% o f Figure 41. Government-supplied drug stocks, Lagos and Kogi States, Nigeria, 2002 (% o f facilities) (n= healthworkers) .............................................................................................................................. 65 Figure 42. 252) ................................................................................................................................................ 66 Causes o f patient dissatisfaction with public hospitals, Nigeria, 1999 (% o f treated cases o f illness inthe previous 4 weeks) ..................................................................................................... 67 Figure 43. Supplies and equipment in health facilities, Lagos and Kogi States, Nigeria, 2002 (% o f facilities) (n= 252) ........................................................................................................................ 70 Figure 44. Facility cleanliness, maintenance, equipment, and services, Lagos and K o g i States, Nigeria, Figure 45. Availability o f modem contraceptive method within 5 km, Nigeria 1999 (% o f households)....72 2002 (% o f facilities) (n= 252) ..................................................................................................... Figure 46. Availability o f qualified delivery assistance within 5 km,Nigeria, 1999 (% o f households) .....79 80 vi Figure 47. M e a ndistance to referral services for delivery complications. Nigeria. 1999 (km) ...................81 Figure 48. Comprehensive emergency obstetric care (EOC) services provided by public referral-level Figure 49.facilities. Nigeria. 2003 ................................................................................................................. 81 HIV/AIDS campaign incommunity inprevious year (% o f households). Nigeria. 1999...........83 Figure 50. Real GDP. total government revenues and government oil and gas revenues. Nigeria. 1999- Figure 51 Federal. state and LGA shares o f total government expenditures. Nigeria. 1999-2003............115 .. 2005 (million Naira at constant 1999 prices) ............................................................................... 116 Figure 52 Federal government health expenditures. Nigeria. 2001-2002 ................................................. 118 Figure 53. Per capita domestic public sector health spending and GDP per capita. 2002 (countries with GDP per capita less than US$1.000) .......................................................................................... 123 Figure 54. Private spending as a proportion o f total health financing and GDP per capita. 2002 (countries Figure 55.with GDP per capita less than US$5.000) .................................................................................. 132 Total health spending (as $US and as % o f GDP) and GDP per capita. 2002 (countries with GDP Figure 56. Total health spending and under-5 mortality. 2002-03 (countries with under-5 mortality over 40 per capita less than US$1.000) ................................................................................................... 133 134 Figure 57 148 Figure 58 Coverage frontier for clinical preventive maternal and neonatal care........................................ ..Coverage per 1.000)..................................................................................................................................... Frontier FamilyiCommunity Preventive Maternal and Neonatal Care ...................... 149 Figure 59. Change inaverage cost per personper year ............................................................................... 151 Figure 60. Changes in the average cost per person per year o f changes in HIViAIDS drugs and malaria combination treatment drugs ....................................................................................................... 152 Boxes B o x 1 The health components o f state poverty reduction strategies ................................. . ........................... 88 B o x 2. Classification o f health care interventions according to their delivery mode.................................. 140 B o x 3. Identification o f bottlenecks ina family/community based intervention......................................... 141 B o x 4. Identification o f bottlenecks ina population outreach activity ........................................................ 142 B o x 5. Identification o f bottlenecks ina clinical care intervention............................................................. 142 vii Introduction and Context INTRODUCTIONAND CONTEXT INTRODUCTION 1, TheHealth, Nutrition, and Population Country StatusReport (CSR)for Nigeria aims to contribute to the evidencebase of the Government'spoverty reduction strategy and health system reform efforts, as well as inform the Bank'spolicy dialogue with the Government. A major theme o f the CSR i s the analysis o f the health situation o f the poor and how the health system i s performing interms o f meeting the needs o f the poor. 2. The World Bank team collaborated with a Federal Ministry o f Health (FMOH)CSR working group, and consultedwith other government officials, international partners, and other stakeholders inthe health sector. 3. The report has six chapters: i)Health Outcomes; ii)Household Behavior and Community Factors Affecting Health; iii)Health System and Policy; iv) The Role o fthe Private Sector in Health Care Provision; v) Health Care Financing; and vi) Extra Resources Needed to Achieve the Health-RelatedMillennium Development Goals (MDGs). 4. The report benefits from several recent high-quality and representative household surveys, including a series of Core Welfare Indicator Questionnaire (CWIQ) surveys and the 2003 Nigeria Living Standards Survey (NLSS), as well as a 2003 Nigeria Demographic and Health Survey (NDHS). The report also draws on administrative data on health services and financing from the FMOH, the CentralBank o fNigeria (CBN) and other government sources. Suchinformation is sometimes dated and often incomplete, particularly with regard to the state and local levels. However, it i s supplemented by a variety o f surveys and studies o f different aspects o f the health system and financing, such as a large survey o f primary health care (PHC) services in2001by the National Primary Health Care Development Agency (NPHCDA), a 2003 FMOH study o f emergency obstetric care, a 2003 World Bank assessment o f state finances, and an in-depth World Bank study of PHC services inLagos and Kogi states in2002. Inaddition, numerous state or issue-specific studies were consulted, including several reports by programs supported by internationalpartners. Finally, national and global WHO and UNICEF reports were consulted on specific issues. CONTEXT 5. Thesheer size and complexity of the countrypresent large challenges to health policymakers. With a population of more than 130 million people, one out of every five people inSub-Sahara Africa i s Nigerian. Not only i s the population large, but it i s also very diverse: there are nearly 300 ethnic groups in the country, more than 500 languages, and two major religious groups (Islam and Christianity). This complexity i s mirrored by widely varying patterns o f health outcomes and health services. 6. Nigeria'sfederal system assigns different health system responsibilities to the three levels of government, each of which is largely autonomous in terms of management andfinancing. Administratively, the country i s organized as a federation with a federal government, 36 states and the Federal Capital Territory (FCT), and 774 local government areas (LGAs). For the purposes o f studies such as this, states are often grouped into six "geopolitical zones," but these do not have any legal existence (Figure 1provides an illustration). 1 Introduction and Context 7. Along with overall policy, the federal government i s mainly responsible for tertiary-level health services, state governments are responsible for secondary services and local governments are responsible for primary services. At the same time, a number o f programs and parastatal agencies, usually based at the federal level with state counterpart organizations, work on PHC services. Although national policies provide a certain measure o f standardization, each level o f government has large autonomy in financing and management of health services under its responsibility. 8. Weakgovernance has been an obstacle to improvement in public services, including health care, although the democratic government ispursuing vigorous reforms. The long years o f military rule weakened, politicized, and corrupted government bureaucracies. In2004, Transparency International ranked Nigeria as the thirdmost corrupt country inthe world. Similarly, the 2004 Africa Competitiveness report rankedNigeria's public institutions as the second to last country inAfrica, and the fifth to last inthe world, using an index that combines measures o f the degree of corruption and the rules o f contract and law. However, after taking office in 1999, the democratic government has taken several steps to address these systemic issues, including launching a high-profile anti-corruption campaign through the creation o f the Economic and Financial Crimes Commission; participating inthe Extractive Industries Transparency Initiative; raising civil service salaries; and embarking on civil service reform. 9. High dependency on oil revenue andpoor economic management has resulted in volatility and uncertainty in the level ofpublic resources. Nigeria's economy i s highly dependent on oil which inthe last decade represented about 95 percent o f total exports and 66 percent o f total government revenue (World Bank, 2004). State and local government budgets, which are used for the delivery o f primary and secondary health care, depend largely on oil revenue allocations since - with some exceptions - state and LGA revenue-generation capacity i s low. Although oil revenue could potentially be usedby the government for productive purposes, variations inoil prices have caused highvolatility and unpredictability ingovernment resources, negatively affecting the health system. 10. However, economic growth and better macroeconomic and$scal policies have improved the situation in thepastfew years. Government reforms and macroeconomic and fiscal policies have contributedto robust economic growth inthe past few years, notably inthe non-oil economy. For example, in2004 the federal and state governments were able to save some o f the revenue windfall from highoil prices, although there i s concern about the 2005 budget which would increase government spending by over 50% (IMF, 2005b). Both oil and non-oil GDP have grown significantly since 2003, keeping pace with and exceeding population growth. Real GDP growth was 10.9% in2003, 3.6% in2004, and a projected 7.4% in2005. In2005, the non-oil economy i s projectedto grow by almost 5% (IMF, 2005a). 11. Thegovernment has a large debt burden. By the end o f 2003 Nigeria's external debt was close to US$33 billion, more than half o f the country's gross domestic product (GDP). In2004 alone, US$ 1.8 billion were budgetedfor external debt payments. This situation i s not alleviated by official development assistance; inrecent years the country receivedabout US$2 per capita compared to the sub-Saharan Africa average o f about US$28 per capita. 12. Poverty is widespread and inequalities are large. Economic status i s one o fthe major determinants o f both health outcomes and health service utilization. Despite recent economic growth, most Nigerians remain very poor. In2005, projected GDP per capita i s US$ 582 (IMF, 2005a), less than the average for Sub-Saharan Africa. Non-oil GDP per capita, an indicator for 2 Introduction and Context 15. The vulnerablesituation of women affects not only their own health but also that of their children. Despite large regional differences, women ingeneral have less access to productive assets, have lower participationrates inincome-generating activities, and earn lower wages. All of these factors contribute to poverty and therefore to poor health. Additionally, women, particularly inthe northernregions, have less access to education than men. The 2003 NDHS found that 55% of primary school-aged girls attended school, compared to 65% of boys. Inthe NorthEast andNorthWest zones, about these proportionswere about 50% for boys and 3540% for girls. These low levels o f access to schooling for girls are worrisome as mother's education i s a consistently important determinant of family health status and health service utilization. 16. With a 4.4% urbanpopulation growth rate, cities in Nigeria are growingfaster than the average cities in Africa (UNFPA, 2004). Lagos, for instance, i s one o fthe biggest cities inthe world and its population i s expected to double between 1990 and the year 2015 (National Population Commission, 1998). This fast urbanization, particularly inLagos, presents many challenges interms o f provision o f health, education, water and sanitation services. 17. The ecology of the county exposes thepopulation to major infectious andparasitic diseases. For instance, almost the entire populationlives in areas o f endemic risk for malaria. Environmentalproblems, such as oil spills inthe Niger Delta, drought inthe north, and floods in the south, also havehadnegative effects on the population's health. Althoughmuch improvement has occurred inthe last decade, a large proportion o f the population does not have access to improved sources o f drinking water and adequate sanitation. 18. The county 's consolidation of democracy allows Nigerians to effectively confront these challenges and improve the health status of thepopulation. Since 1999 Nigerians have elected their political representatives. One o f the main pillars o f the government's poverty reduction strategy, the 2004 National Economic Empowerment Strategy (NEEDS), i s investment inhuman capital, including improving the health system and health situation of the population. 4 Chapter 1 CHAPTER 1. HEALTH OUTCOMES 1. This section will first present an analysis o f trends, internationalcomparisons, and regional, income, and rural and urban inequalities inthe health status o f the population. It will then present an assessment on the country's progress inmeeting the Millennium Development Goals (MDGs). Figure 2. Nigeria and global trend in under-fivemortalityvs. GDP per capita 250 -2003 0 0 0 50 0 100 1000 10000 100000 GDP per capita 2002 ($US) Sources for Nigeria: 2003 NDHS for under-5 mortalityestimate for the period 1998-2003and IMF for GDP per capita estimates fpr 1998and 2003. Sources for other countries: WHO for estimates of under-5 mortality rates for 2003 and World Development Indicatorsfor GDP per capita for 2002 CHILDHEALTH 2. Every year about one million children die in Nigeria. Many o f these children die from either preventable diseases or from diseases that can be treated at very l o w cost. At present, one out o f every ten children dies before his or her first birthday and one out o f every five before his or her fifth. Giventhe country's large population, this highunder-five mortality rate implies that each year about a million children younger than five die.3 3. Child mortality is higher than would be expected given Nigeria's per capita GDP. Figure 2 plots under-five mortality rates by GDP per capita inall countries. For Nigeria, the estimate for under-five mortality from the 2003 NDHS (201 per 1,000 live births)refers to the period 1998- 3 To estimatethe number of childrendying each year, we transformedthe birth cohort mortality rate over five years of 201 per 1,000 (NDHS 2003) to ayearly crude mortality rate. For this we usedthe formulabelow (Preston, Heuveline, and Guilliot, 2001), where n is the number of years inthe interval (5 years), rn is the observedcrude mortality rate for one year, and a is the averagenumber of years lived by those who died during the interval (here assumedto be 2). nm 5 4 0 1 (n-a)m + 5 Chapter 1 2003, a time when GDP per capita grew significantly. Nevertheless, Nigeria's under-five mortality rate i s above the trend even when usingthe 1998 GDP per capita estimate (around U S $ 270). Figure 3. Trends in infant and child mortality, Nigeria 250 1 k - 200 I100 5 O 50 ~ I 1989-1993 1994-1998 1999-2003 +Infant Under 5 Source: Nigeria DHS 2003. These are mortality ratesfor five-year periods precedingthe NDHS 2003 survey 4. Infant and under-jive mortality rates show a slight improvement after aperiod of stagnation. As seen inFigure 3, after two decades ofmilitary dictatorship and increases inpoverty, infant and child mortality rates didnot improve duringthe 1990s. This trend, however, has started to change and today bothrates show slight improvements. Figure 4. Infant and under five mortality across wealth quintiles, Nigeria, 1993-2003 350 i 293 300 250 200 150 100 50 0 Poorest Poorer Middle Rich Richest 0Infant 0Under Five Source: NDHS 2003 5. Inequalities across income groups remain very worrisome. Both infant and under-five mortality rates are more than 2.5 times higher among the poorest 20 percent o f the population than among the richest 20 percent (see Figure 4). While the infant and child mortality rates o f the richest endo fthe income distribution are only half sub-Saharan Africa averages, those o fthe poorest are more than 1.3 times as high. As seen inFigure 5, these differences are the greatest among all the SSA countries for which data are available. InNigeria, the poor have as many as three times more under-five children dyingthan the rich; in Senegal, the country with the second greatest difference, the poor to richratio i s about 2.5. Inthe case o f infant mortality, Nigeria i s once again the country with the greatest difference between rich and poor, followed closely by the Central African Republic. 6 Chapter 1 Figure 5. Poorestlrichest ratio for infant and under five mortality 3.5 UU5MR .IMR 2.5 2 1.5 1 0.5 0 Sources: World Bank HNP Stats and authors' calculationsfrom 2003 NDHS 6. Thereare also large regional and rural- urban inequalities in infant and child mortality. In general, children living inthe Northern regions fare much worse than children living inthe Southern ones (Table 1). Children born inthe South East and South West are more than twice as likely to reachtheir fifth birthday than children living inthe NorthEast and NorthWest. The South South and North Centralregions do not follow this North-South pattern. The very high infant mortality rates inthe South South region are mostly due to a very poor performance in rural areas of this region where the infant mortality rate, o f about 130per 1,000 live births,i s worse than in any other rural area o fthe country. Incontrast, the infant mortalityrate inurban areas inthis region i s lower than the rates innorthern urban areas. Infact, these large differences betweenrural and urban areas are found inthe entire country, with rural children faring much worse than urban children (Figure 6). Table 1, Neonatal, Infantand under five mortality across geopolitical regions, Nigeria, 1993-2003 neonatal infant under-five North Central 53 103 165 North East 61 125 260 NorthWest 55 114 269 South East 34 66 103 South South 53 120 176 South West 39 69 113 Source: NDHS 2003 7. There are a number o f possible reasons for why the NorthEast and NorthWest regions, as well as rural areas inthe South-South and elsewhere, fare worse than other parts o f the country. Poverty i s more widespread inthe north and inrural areas than inthe south and urban centers and this is certainly drivinga large part o fthe differences. Also, bothurbanareas andthe SouthEast and South West regions as a whole have better access to clean water and sanitation and to health care services. Additionally, women living inurban centers and especially inthe southern regions on average have more education, a well-documented determinant o f child survival, than their 7 Chapter 1 counterparts inthe rest o fthe country. To understandthe individual effects o f variables such as mother's education, place o f residence, or household wealth4 on infant and under five mortality we performed a multivariate analysis o f child survival. The results o f this analysis are presented inTable 2. Figure 6. Infant and under five mortality in urban and rural areas, Nigeria, 1993-2003 300 1 250 urban Hrural 200 150 100 50 0 Under five Infant Source: NDHS 2003 based on children born 10years before the interview 8. The child survival analysis uses data from the NDHS 2003 and includes all children born duringthe ten years before the interview. The analysis includes individual, household, and community variables that the international literature has identified as important inexplaining child ~ u r v i v a l . ~ 9. Mother's education has a large and signiJicant effect on reducing theprobability of children 's death. The children o f educated mothers are less likely to die than children whose mothers do not have any education. This effect increases with the mothers' level o f education. While children whose mothers have finishedprimary school die at a rate that i s more than 20% lower than the children o f uneducated mothers; children whose mothers have higher education die at a rate that i s more than 40% lower than children o f uneducated mothers. 10. Among household characteristics, the number of household members and the household's socio-economic status increase theprobability of child survival. However, once we account for all other variables, the effect o f the wealth index i s only statistically significant inincreasing the survival o f childrenbelongingto the richest 20 percent o f households. These children die at a rate that i s more than 20 percent lower than children belonging to the poorest group. 11. Once we accountfor mother's education, availability of health services, and wealth, the effect of region of residence on child survival almost disappears. Only for children living inthe South South region there i s a difference ininfant survival (with respect to the reference region) that cannot be explained by any o f the other variables included inthe analysis. Similarly, once 4The DHS survey used inthis analysis does not have informationon income or consumption. Instead a wealth index based household asset ownership i s used as an indicator for household socio-economic status. The analysis includes individual variables (age inmonths and gender o f the children), household variables (mother's parity and level o f education, gender o f head o f household, and the number o f people living inthe household), community variables (the region where the household lives and an indicator for urban areas). Finally, as there was no informationon household access to health facilities, the analysis includes the non-self community averages o f the number of children vaccinated against measles, the numbero fwomen that received more than two pre-natal care examinations, and the number o f women that gave birthin a health facility as variables that can act as indicators for availability and quality o f health services. 8 Chapter 1 we account for other variables, the difference inchild survival between rural and urban areas disappears. Table 2. Resultsfrom a Cox proportional hazards model of infantand child survival in Nigeria using the NDHS 2003 Children's Characteristics female 0.971 1.016 (0.040) (0.049) age 1.005 0.982 (0.001),** (0.001),** Birth order number 1.041 1.035 (0.008),** (0.009),** (0.133),* (0.107)*'* Household's characteristics Mother's education (none omitted variable) primary 0.783 0.752 (0.068),*' (0.069)*** higher 0.620 0.518 Wealth Index(poorest omitted) Poorer 1.186 1.205 (0.066),** (0.069),** Middle 0.939 0.949 (0.069) (0.070) Richer 0.951 0.959 (0.076) (0.083) Richest 0.687 0.715 (0.085),** (0.097)** Number of householdmembers (e6 omitted) 6-10 0.687 0.675 (0.035),** (0.036)*** > I1 0.684 0.665 (0.049)*** (0.051),** Female head 0.863 0.860 (0.082) (0.092) Region (SouthWest omitted) North Central 0.928 0.948 (0.116) (0.128) North East 1.159 1.215 (0.142) (0.158) NorthWest 1.124 1.181 (0.142) (0.157) South East 0.981 0.923 (0.131) (0.136) South South 1.263 1.259 (0.168), (0.181) urban (rural omitted) 0.954 0.973 (0.052) (0.058) Availability of services (non-selfcluster averages) measles vaccination 0.649 0.672 (0.107),,* (0.118),* delivery in healthfacility 0.638 0.710 (0.088),** (0.109)*' >2 pre-natalvisits 1.145 1.055 (0.139) (0.134) Observations 11391 11159 Hazard ratios are shown in first line robust standard errors in parentheses * significant at 10%; ** significantat 5%; *** significant at 1% 12. Finally, the variables that act as indicatorsfor the availability of health services significantly increase theprobability of child survival. The risko f childmortality i s statistically less significant in sample clusters, with higher proportions o f children vaccinated against measles and higher proportions ofwomen who deliveredina health facility. 9 Chapter I CHILDNUTRITIONALSTATUS 13. The nutritional status of Nigerian children ispoor, in line with countries with similar income per capita. Malnutrition contributes to infant and child morbidity, mortality, and also reduces children's future learningcapacity. Slightly more than one out o f three five year old children in Nigeria are stunted, almost one out o f every three i s underweight, and about one out o f every ten i s wasted.6 Although these percentages are high, they are not very different to those o f other countries in Sub-Saharan Africa (SSA) as can be seen inTable 3. Table 3. Children's nutritionalstatus in a sample of SSA countries Children's NutritionalStatus Stunted Wasted Underweight (low height-forage) (low weight-for-height) (low weight-for-age) Benin 30.4 7.5 22.8 Burkina Faso 36.8 13.2 34.3 Chad 40.1 14.1 38.8 Cameroon 26.0 2.9 15.1 Cote d'lvoire 25.2 7.8 21.2 Ethiopia 51.2 10.7 47.1 Ghana 25.9 9.5 24.9 Guinea 26.1 9.1 23.2 Malawi 49.0 5.5 25.4 Rwanda 42.4 6.8 24.5 SSA 31.O Tanzania 42.6 5.4 28.9 Uganda 38.6 4.0 22.5 Zambia 46.8 5.0 28.2 Zimbabwe 26.5 6.4 13.0 Source: DHS 14. There are signs of improvements in child nutritional status. The percentage o f malnourished children has decreased inthe last decade. Comparisons between the 1990 and 2003 N D H S surveys indicate that the number o f children stunted, a measure o f chronic malnutrition, decreased inrelativeterms bymorethan 10% since 1990, andthe number ofchildrenunderweight, an indicator for both chronic and acute malnutrition, decreased inrelative terms by almost 20%.7 The percentage o fchildren wasted didnot change duringthis period. This measure generally indicates a lack o f sufficient nutrients inthe periodjust before the survey, and it i s often the result o f an illness, especially diarrhea. While the other two nutritional measures are more likely to be affected by improvements inincome inthis last decade, wasting i s more likely to be affected by either lower prevalence o f childhood illnesses or better treatment. Childrenare consideredstuntediftheir height-for-agez-scoreis less than two standarddeviationsfrom the medianof theNCHSiCDCiWHO internationalreference population. Similarly, childrenare considered underweightiftheir weight-for-ageis less than two standarddeviationsfrom the median, andwasted iftheir weight-for-height is less than two standarddeviations fromthe median. 'Wedonot have the standarderrors of the 1990 estimatesbut the point estimatesof the percentageof children stunted andunderweight in this year do not fall inthe confidenceintervalsof the 2003 estimates. 10 Chapter 1 Figure 7. Malnutrition across time and differencesin percentagesof children underweight in rural and urban areas 40 35.3 30 30 - 31.8 20 20 - 22.2 Stunted Wasted Underweight I Nigeria 1990 Nigeria 2003 0Nigeria 1990 ENigeria2003 Source: NDHS 2003 Figure 8. Poorestlrichest ratio of the percentageof children stunted and underweight in a sample of SSA countries Zambia ~' Tanzania ~ Nigeria ~ I Kenya ~ Chad ~'~~ I Cameroon 1 I Burkina Faso 0 0.5 1 1.5 2 2.5 3 3.5 4 15. As was the casefor infant and child mortality rates, there are also large differences in the nutritional status of children across income levels, regions, and rural and urban areas. Children living inpoor households, children livinginthe Northernregions, and children inrural areas are much more likely to be wasted, stunted, or underweight than their richer, southern, and urban counterparts. The income differences are large; the percentages o f children underweight and stunted among the poorest fifth o f the population are more than 2.5 times larger than the percentages among the richest fifth. The poor-to-richratios inNigeria are among the highest in countries for which data are available (see Figure 8). Finally, the differences betweenrural and urban areas, although large, are not as large as wealth-based or regional differences. For instance, 31% o f children are underweight inrural areas, compared to only 22% incities. Moreover 43% o f children are stunted inrural areas, compared to 29% inurban areas. These differences have not changed much inthe last decade as can be seen inFigure 7. 11 Chapter 1 16. To better understandthe effects o fhousehold socio-economic status (usingwealth as a proxy), maternal education, and region o f residence, a multivariate analysis o f the determinants o f malnutrition i s done. The results o f this exercise are presented inTable 5. The model indicates that the higher the level o f maternal education, the lower the probability that a child i s malnourished. Moreover the greater the level o f household wealth, the lower the probability that a child i s chronically malnourished or stunted. This effect, however, i s only significant for children belonging to the richest 40 percent o f the population. The level o f household wealth, however, was not found to be a significant determinant o f underweight and wasting. Table 4. Nutritional status across regions, Nigeria,2003 Children Nutritional Status Stunted Wasted Underweight North Central 31% 8% 23% North East 37% 10% 34% North West 53% 13% 42% South East 23% 4% 8% South South 16% 13% 18% South West 23% 11% 20% Source: NDHS 2003 17. Some regional differences persist after other factors affecting malnutrition are accounted for. For instance, children livinginthe North West are much more likely to be malnourishedthan children inthe South West (the reference region), while children inthe South South are less likely to be stunted. Finally, children are also less likely to be malnourishedifthey live incommunities with a large percentage ofhouseholds with access to improveddrinkingwater and access to mass media. This i s not surprisingbecause malnutrition i s often not so much a consequence o f lack o f food but o f inadequate knowledge on how to diversify food intake or on how to treat diarrhea. 18. Aside from protein-energy deficiencies, micronutrient deficiencies can also limit children's health and development. The micronutrient deficiencies that represent major public health problems are: vitamin A deficiency, iodine deficiency, iron deficiency, and zinc deficiency. Vitamin A deficiency increases the risk o f children's death due to diarrhea, measles, malaria, and its deficiency can also cause child blindness. It i s estimated that Vitamin A deficiency leads to about 82,000 children deaths each year inNigeria (UNICEF, 2004). Iodine deficiency inthe early stages o f life can cause damage to the brain impairing the child's cognitive development. Serious iodine deficiency can create goitre among people living in iodine deficient areas. Iron deficiency i s a major cause o f anemia. Finally, there i s good evidence showing that zinc supplementation i s one of the major interventions that can prevent children's diarrhea and pneumonia (Lancet Child Survival Series, 2003); its deficiency can also cause growth retardation and mental lethargy. 19. A USAIDand UNICEFfood consumption and nutrition survey in Nigeriafound high levels of vitamin A and iron deficiency, as well as various degrees of iodine deJiciency in children under- five years of age (Maziya-Dixon, et al. 2004). For instance, the survey indicates that close to a quarter o f Nigerian children suffer from marginal vitamin A deficiency8, while 5% suffer from severe deficiency. Despite the large government success inmaking iodized salt available in almost all the country, about 4% o f children suffer from severe iodine deficiency, 9% from ~ ~~ Marginal deficiencywas definedas having serumretinol concentrationlevels<20ug/dl, while severe vitamin A deficiencywas definedas havinglevelslower than lOug/dl. 12 Chapter 1 moderate deficiency, and 15% from milddeficiencyg. Inthe case o f iron, almost 20% o f children were found to have a highlevel o f iron deficiency", while 8% were found to have already depleted their iron stocks. Finally, one inevery five Nigerianchildren under-five years o f age suffers from zinc deficiency. Table 5. Multivariate (probit) analysisof determinantsof malnutritionusing NDHS 2003 stunted wasted underweight Children's Characteristics Female -0.140 -0.012 -0.021 (0.052)*** (0.067) (0.050) Age 0.013 -0.008 0.005 (0.002)*** (0.002)*** (0.002)*** Birth order number -0.008 -0.004 -0.029 (0.012) (0.016) (O.OIl)** Household's characteristics Mother's education (none omitted variable) Primary -0.214 -0.255 -0.230 (0.077)'** (0.127)** (0.081)**' Higher -0.922 0.123 -0.492 (0.1go)*** (0.195) (0.165),** Wealth Index (poorest omitted) Poorer 0.004 0.052 0.113 (0.083) (0.096) (0.087) Middle 0.062 -0.035 0.079 (0.094) (0.115) (0.085) Richer -0.202 -0.001 0.046 (0.100)** (0.118) (0.095) Richest -0.279 0.062 -0.165 (0.123),* (0.157) (0.128) Number of household members (e6 omitted) 6-10 0.036 0.051 0.047 (0.057) (0.083) (0.059) > I1 0.289 -0.022 0.154 (0.078)*** (0.108) (0.079)* female head -0.093 0.039 -0.086 (0.103) (0.127) (0.090) Region (SouthWest omitted) North Central -0.075 -0.245 -0.188 (0.108) (0.132s (0.123) North East 0.092 -0.139 0.084 (0.139) (0.163) (0.148) North West 0.562 0.161 0.471 (0.1I,),** (0.144) (0.130),** South East -0.054 -0.268 -0.409 (0.152) (0.174) (0.172)** South South -0.426 0.157 -0.204 (0.1I,),** (0.143) (0.122)* urban (rural omitted) 0.057 0.019 0.092 (0.072) (0.091) (0.080) Availability of services (non-self cluster averages) Improved sanitation -0.121 -0.110 -0.102 (0.162) (0.152) (0.130) Access to mass media -0.564 -0.280 -0.685 (0.187)*** (0.207) (0.180)*'* Improved drinkingwater sources -0.187 -0.022 -0.156 (0.103)* (0.112) (0.111) Constant -0.078 -0.814 -0.103 (0.172) (0.I,,),** (0.175) Observations 4293 4414 4293 Standard errors in parentheses * significant at 10%; ** significant at 5%; *** significant at 1% Childrenwere consideredto suffer from severe iodinedeficiencyifthey hadmean iodine levels lower than 20ug/l, moderateifthe mean iodinelevelwas between20 and49, and mild ifit was between50 and 99. lo Irondeficiencywas definedas havingserumferritin values of less than lOng/ml. Childrenwere consideredto have depletedtheir stocks ifthey hadlevels lower than 20ng/ml. 13 Chapter I CAUSESOFCHILD MORTALITY AND MORBIDITY 20. Communicablediseases, often in association with malnutrition, are the major causes of mortality and morbidity among children in Nigeria. The major causes o f mortality and morbidity among children are malaria, neonatal related causes, diarrhea, pneumonia, measles, and AIDS (see Figure 9). Most o f these causes (malaria, diarrhea, pneumonia, and measles) are preventable or can be treated at low cost. Figure 9. Estimated Causes of Mortalityof Children Under Five, Nigeria Diarrhoea Neonatal 17% 27% Measles --. Source: background to Lancet series. Note: These estimates are based in models as there are no reliable data on causes of infant mortality in the country due to no reporting and to the weaknesses of the health system routine data collection. 21. Malaria causes the largest number of child deaths in Nigeria - around 300,000 annually. Although it potentially affects the entire population, the major risk for severe morbidity and mortality i s suffered by children under five and by pregnant women. Malaria represents about 30% o f estimated child deaths or about 300,000 children each year. It i s also very likely that a large proportion of the 32% o f children reporting fever and/or convulsions inthe last two weeks inthe 2003 NDHSwere suffering frommalaria. 22. More than 99% o fthe Nigerianpopulationlives under endemic risk o f malaria and 0.02% live under epidemic risk of the disease. The majority o f cases are due to Plasmodium fakiparum, the most malignant form o f the disease. Due to growing drug resistance, Nigeria has adopted artemisin combination therapy (ACT) as first-line treatment. 23, Diarrhea andpneumonia cause almost 400,000 child deaths annually in Nigeria. Pneumonia, the most serious acute respiratory infection (ARI),i s the thirdmajor cause o f mortality among children; it is responsible for an estimated 200,000 deaths each year in Nigeria. Diarrhea i s also a major cause o f child morbidity and mortality, causing more than 176,000 deaths o f children under five inthe country. Many o f these deaths could have been avoided, as both diarrhea and ARI can be treated at very low cost. Nevertheless, datafrom the NDHS 2003 indicate that in the last two weeks before the interview about 19%of children under-five had diarrhea and 10%had acute respiratory infection (AH) symptoms; these rates are lower than what isfound by such surveys in other SSA countries. 24. Finally, neonatal related causes represent an estimated 27% of all deaths among children underfive. Nigeria, with a neonatal deathrate o f 53 per 1000live births, i s the country with the fourth largest number o f estimatedneonatal deaths inthe world, representing close to 6% o f global neonatal deaths. The majority o f these deaths are due to asphyxia, premature birth, severe infection, tetanus, and congenital malformations. Invery highneonatal mortality settings as it i s the case inNigeria, it i s estimated that almost 50% o f death are due to severe infections, tetanus, 14 Chapter 1 and diarrhea (Lancet 2005, Neonatal Survival Series). Many o f these causes are the result o fpoor maternal care and inadequate access to delivery services. 25. Immunization coverage in Nigeria is among the lowest in the SSA region. This has resulted inoutbreaks ofotherwise preventable diseases. Among vaccine-preventable diseases, measlesis a major cause of morbidity inchildren. Inthe year 2001, Nigeria reporteda total o f 168,107 measles cases and inthe following year 42,00711 cases, for the year 2005 more than 500 death from measles has been already reportedI2.These numbers are nonetheless an underestimation as they only include cases that were reported to the health system. Similarly, there were outbreaks o f diphtheria and pertussis inthe year 2002, and neonatal tetanus continues to be one o f the main causes o f infant mortality inthe country. These last three diseases can be preventedby DPT immunization o f the child and the last one also by two doses o f Tetanus Toxoid (TT) for the mother duringpregnancy. 26. Nigeria is a reservoir for poliomyelitis, presenting a challenge to the international campaign to eradicate the disease. Political controversy over the polio vaccination campaigninsome states innorthernNigeria interrupted the eradication campaign in2003-04, leadingto aresurgence o f the disease and its spread to other countries. Vaccination has resumed and the incidence o f new cases up to end April 2005 has been at a slower rate than the previous year. Most cases are in northern states and among unvaccinated children under three years old. As the campaign overcomes this setback inNigeria, it estimates that it i s technically feasible to eradicate the disease by the end o f 2005. (WHO, 2005c) MATERNALREPRODUCTIVEHEALTH AND 27. There i s no recent reliable data on maternalmortality inNigeria. However, usinga statistical model, WHO, UNICEF,and UNFPA (AbouZahr and Wardlaw, 2001) estimated the maternal mortality ratio (MMR) at 800 per 100,000, equivalent to 37,000 maternal deaths per year. These estimates, however, are not precise and have a large range o f uncertainty around them (see Table 6). This uncertaintymakes the ranking o f countries, or analysis o f trends over time almost impossible. Nevertheless, as the Nigerianpopulation i s very large, it i s thought that the country has the largest number o f maternal deaths after India. 28. The major causes of maternal mortality and morbidity in the country are complications duringpregnancy and delivery. As shown inFigure 10, hemorrhage, sepsis, eclampsia, and cephalo-pelvic disproportion are the major delivery complications causing maternal death. Malaria i s a pregnancy complication responsible for a large share o fmaternal mortality inthe country. These complications can be better managed through intermittent preventive treatment (IPT) for malaria, the use o finsecticide-treatedbednets, as well as improved access to ante-natal care, birthattended by skilled personnel, and - above all - emergency obstetric care (EOC) services. 11WHO vaccine-preventabledisease monitoring system, 2003 global summary. l2IRINMarch21,2005. 15 Chapter 1 Table 6. Indicatorsof maternal health in selected SSA countries Total fertility to Meanideal Maternal Mortality no. of BMI % rate 15-49 childbearing children Estimate Range of uncertainty < 18.5 Benin 5.6 25.9 5.5 850 490 . 1200 10.5 Burkina Faso 6.4 19.6 5.7 1000 630 1500 13.3 Cameroon 4.8 19.6 6 730 430 g 1100 7.9 CAR 5.1 12.3 6.4 1100 670 1600 15.3 Chad 6.4 10 8.3 1100 620 1500 21.1 Cote d'lvoire 5.2 20.7 5.4 690 170 1300 7.4 Ethiopia 5.5 32 5.3 850 500 1200 26.0 Gabon 4.2 23.1 4.9 420 240 600 6.6 Ghana 4.4 35 4.3 540 140 1000 11.3 Guinea 5.5 20.9 5.7 740 420 1100 11.8 Kenya 4.7 53.3 3.8 1000 580 1400 11.9 Madagascar 6.0 38.1 5.3 550 310 780 20.6 Malawi 6.3 42.3 5 1800 1100 2600 6.5 Mauritania 4.5 19.1 6.2 1000 630 1500 8.6 Mozambique 5.2 16.9 5.9 1000 260 2000 10.9 Niger 7.2 9.6 8.2 1600 420 3100 20.7 Nigeria 5.7 18.3 6.7 800 210 1500 15.2 Rwanda 5.8 33.7 4.9 1400 790 2000 5.9 Togo 5.2 28.6 4.5 570 340 810 10.9 Uganda 6.9 38.5 4.8 880 510 1200 9.4 Zambia 5.9 35.5 4.7 750 430 1100 13.0 Zimbabwe 4.0 40.9 3.9 1100 620 1500 4.5 Source: AbouZahr and Wardlaw (2001) for the maternal mortality estimatesand DHSfor the other indicators. 29. Unsafe abortions have also been identi3ed as one of the major causes of maternal mortality that canpartly be avoided with better access to modern contraceptive methods. Abortion i s illegal inNigeria, except to save a woman's life. Nonetheless, many pregnancies end ininduced abortion, which are frequently performedby non-trainedhealth personnel posing serious risks to women's health. There are no official data on abortions; however a 1998 study estimated that about 610,000 abortions are performed every year inNigeria (Henshaw et al., 1998). In other words, about 18-25 abortions are practicedper 1,000 women aged 15-44 every year. The rate i s much higher inurban areas o fthe south than inthe rural and northern areas. 30. The risk of maternal morbidity is also high, of which vesico-vaginalJistula (VVF) is one of the most serious andprevalent in Nigeria, especially in the Northern regions. VVF i s usually caused by delivery complications and it i s an abnormal opening betweenthe bladder and the vagina that produces urine/feces incontinence. According to a USAID assessment, about 70% o f the estimated 200,000 to 400,000 cases o fVVF inNigeria in2001 occurred inthe North. 3 1. The largeprevalence of Female Genital Mutilations (FGM) in the country also increases the possibility of maternal related morbidity. As will be explained inthe next chapter FGMcan cause complications duringpregnancy and delivery. 32. Thefertility rate in Nigeria has decreasedsince 1990,but is still higher than the Sub- Saharan Africa average. The risko f death duringpregnancy and childbirth i s higher the more children women have, and the less time they have between births. Estimates from the 2003 NDHS indicatethat the total fertility rate (TFR) inNigeria i s about 5.7 children per women. This i s lower than the 1990 estimated rate o f 6.3 children per women but it i s still higher than the SSA average o f 5.2. As expected, women inrural areas have, on average, one child more than women inrural areas. Thislarge TFR is due to botha large demand for children(see Table 6) and an unmeet needfor contraception to space children and to avoid pregnancies. 16 Chapter I Figure 10. Reported causes of maternal mortality, Nigeria, 1996 Others CPD 5% Toxemia/clarn psia 11% Abortio 11Yo Anemia Malaria 11% 11% Source: FMOH as cited in NPC and UNICEF (2001) WOMEN'S NUTRITIONAL STATUS 33. Many women in Nigeria,particularly among thepoor, have low nutritional status. Table 6 compares the proportion o f women with low nutritional status across some Sub-Saharan African countries. The 2003 NDHS found that about 15% o f adult Nigerian women have a body mass index (BMI)13lower than 18.5 which indicates chronic energy deficiency among non-pregnant women. Chronic poor nutritional status inwomen not only lowers their overall well-being and productivity, but amongpregnant women, it can also result incomplications during child birth as well as low birth weight inchildren, predisposing them inturn to higher morbidity and mortality. Only countries with lower income per capita -Eritrea, Ethiopia, Madagascar, Chad, andNiger - have a higherproportion o f women with such a low BMI. There are also large regional and income disparities inwomen's nutritional status inNigeria. The 2003 N D H S indicates that among the poorest fifth o f the population as many as 22% o f women have l o w BMI, compared with only 9% amongthe rich. Similarly, inthe NorthEast about 23% o fwomen are considered malnourished, compared to only 8% of women inthe South East region. Finally, the 2003 NDHSalso reports that 6% o fadult women are obese andclose to 15% over-weight. Obesity can lead to diabetes and cardiovascular diseases. 34. Thereare also worrisome levels of micronutrient deficiency among women in Nigeria. Micronutrient deficiencies, especially duringpregnancy, can harm the health o fboth women and children. For instance, vitamin A deficiency can cause night blindness inpregnant women and increases their risk o f death. Serious iodine deficiency duringpregnancy can result instillbirth, child brain damage, cretinism, and it can also produce goiter. Finally, iron deficiency among pregnant women can result inmortality and also inpremature birthand low birthweight. A 2003 nutritional survey (Maziya-Dixon et al.,2004) found that about 10%o fpregnant women have vitamin A deficiency, about 20% suffer from iron deficiency, and 4% have severe iodine deficiency. 13BMIis calculatedby dividingthe weight inkilogramsby the squareofthe height inmeters. 17 Chapter I HIV/AIDS AND OTHER SEXUALLY TRANSMITTED DISEASES 35. Although HIV/AIDSprevalence in Nigeria has not reached the high levelsfound in other African countries, due to itspopulation size, it has the third largest number ofpeople infected with the virus in the world, after South Africa and India. In2003, between 3.2 and 3.8 million people were livingwith HIV/AIDS inthe country (FMOH, 2004). Nigeria also appears to be the country with the largest number o f HIV/AIDS orphans inthe world, with an estimated 1million children that have lost at least a parent as a consequence o f the disease (see Table 8). Figure 11. HlVlAlDS sero-prevalence over the years, Nigeria Table 7: HlVlAlDS sero-prevalence in 2003 7 - 6 - HIV prevalence 5 - 4 - 20-24 25-29 5.4 30-34 4.0 35-39 3.1 1990 1995 2000 2005 40-49 3.2 Source: FMOH 2004. HIVlAIDS sentinel survey Source: FMOH 2004, HlVlAlDS sentinel survey 36. The most recent national sentinel survey on HIV/syphilis, estimated the national adult prevalence as 5percent; the highest rates werefound among theyouth. This estimate represents a slight decrease from the estimated prevalence o f 2001. However, as the technical report on the last sentinel survey warns, this point prevalence i s not sufficient to conclude that the epidemic has stabilized or i s on a downward trend (FMOH, 2004). Additionally, this rate i s still almost three times the rate prevalent a decade ago (see Figure 11). 37. There are large regional variations inthe adult prevalence rates, with the largest median Age HIV prevalence prevalence in2003 found inthe North Central 15-19 4.0 region, and the lowest inthe SouthWest. There 20-24 5.6 are also large intra-regionalvariations, as the highest median prevalence rate, 12%, i s found in 25-29 5.4 Cross River State (South South region), and the 30-34 4.0 lowest, 1.2%, in Osun (South West region). 35-39 3.1 Similarly, there are variations across age groups. The highest prevalence rate (5.6%) was found 40-49 3.2 among people 20 to 24 years old. 18 Chapter 1 Table 8. HIV/AIDSepidemic across the world Numberof People Adults prevalence Orphans (o-14) Livingwith HIV/AIDS 15 to 49 years AIDS deaths 2001 South Africa 5,000,000 20.1 660,000 360,000 India 3,970,000 2.7 55,000 12,000 Nigeria 3,500,000 5.0 1,000,000 170,000 Kenya 2,500,000 15.0 890,000 190,000 Zimbabwe 2,300,000 33.7 780,000 200,000 Ethiopia 2,100,000 6.4 990,000 160,000 Tanzania 1,500,000 7.8 810,000 140,000 DRC 1,300,000 4.9 930,000 120,000 Zambia 1,200,000 21.5 570,000 120,000 Mozambique 1,I00,000 13.0 420,000 60,000 Cameroon 920,000 11.8 210,000 53,000 China 850,000 0.1 76,000 30,000 Malawi 850,000 15.0 470,000 80,000 Sub-SaharanAfrica 28,500,000 9.0 11,000,000 2,200,000 Source: UNAIDS 38. FMOH (2002) projections indicatethat by the year 2010 HTV/AIDSprevalence among adults 15-49 years old could reachmore than 9%, which represents as many as 7 million infected people. A more worrisome picture was given by a study by the U S National Intelligence Council (2002) which estimated that by the year 2010 10 to 15 million adults will be infected inNigeria. Table 9. HlVlAlDS epidemic scenarios in Nigeria 2005 2010 HIV prevalencefor 15-49year olds High Scenario 7.75% 9.24% Low Scenario 6.80% 6.73% Number of HIV positive High Scenario 5.5 million 7.4 million Low Scenario 4.9 million 5.5 million Cumulative deaths High Scenario 2.6 million 5.5 million Low Scenario 2.5 million 5.1 million Source: FMOH (2002). OTHER DISEASES PREVALENTINNIGERIA 39. Communicablediseases represent a major health burden in Nigeria. Among these diseases, malaria i s the main concern. However there i s a long list o f other communicable diseases that are endemic inthe country and that are major causes o f mortality and morbidity. These include tuberculosis (TB), measles, dysentery, typhoid, cholera, meningococcalmeningitis, yellow fever, and onchocerciasis. 19 Chapter 1 40. Nigeria has thefourth highest number of TB cases in the world. In2004, there were an estimated 293 new TB cases per 100,000 population (incidence) and an estimated 546 per 100,000 total cases (prevalence) (WHO, 2005a). This i s equivalent to about 675,000 people infected with the disease, so that Nigeria has the fourth largest number o f TB cases inthe world. Itis thought that about one quarter ofTB cases are co-infected withHIV. 41. Meningococcal meningitis i s another vaccine-preventable disease that affects Nigerians. The most recent significant outbreak o f the disease was in 1996, when almost 16,000 cases and more than 2,500 deaths were reported. This outbreak affected 12 states, mostly inthe north, which lies on the African "Meningitis Belt." (WHO, 1996). 42. Among the vector borne diseases, onchocerciasis (river blindness), guinea worm (dracuniasis), trypanosomiasis (sleeping sickness), schistosomiasis (bilharzias), and yellow fever represent an important healthburden for the population. NON-COMMUNICABLE DISEASES14 43. Although communicable diseases are major causes of mortality and morbidity in the country, there are reasons to believe that non-communicable diseases (NCDs) represent an increasing share of Nigerians ' burden of disease. WHO (2002) estimates that incountries inSub-Saharan Africa with mortality profiles similar to that o fNigeria, non-communicable diseases represent 25% o f total mortality, o f which 50% i s due to cardiovascular disease, 25% to cancers, and about 10%to respiratory disease. Inaddition, an estimated 7% o f total mortality i s attributed to injuries. Sickle cell disease i s the most common genetic disorder affecting Nigerians. Major NCDs inNigeria include: hypertension, diabetes mellitus, coronary heart disease, sickle cell disease, cancers, G6PD deficiency anemia, mentalhealth, roadtraffic injuries and violence, oral health, blindness, rheumatic heart disease, stroke, osteoporosis. 44. Datafrom a national survey on NCDs in Nigeria between 1990 and 1992 andpublished in 1997 identiJied hypertension as one of the main NCDs of concern in the country. This survey found that 11.2% (4.3 million) Nigerians over 15 years o f age have hypertension. O f these 66% (2.84 million) have mild hypertension, 20% (0.85 million) have moderate hypertensionand 14% (0.64 million) have severe hypertension. 12.5% (4.8 million) Nigerians, 15 years and above have borderline hypertension. The survey showed that hypertension i s more prevalent inurban than in the ruralareas. 45. This national survey also collected information on Coronary Heart Disease (CHD) in Nigeria, as well as riskfactors that promote arteriosclerosis and CHD. It was found that the mean total cholesterol (TC) i s (122.4 +42.0 mgper dl) low inthe general population, although urban men and women have higher levels than their rural counterparts. Glucose intolerance a n d or diabetes was found to be linked with excessive smoking, with hypertensionor with hypercholesterolaemia in0.07%, 0.7% and 0.09% respectively within the study population. About 4.14 millionNigerians over the age of 15 years smoke and 1.26 million people smoke 10 or more cigarettes a day. These persons are more at risk o f developing Coronary Heart Disease than the general population. 46. About 1.05million Nigerians who are at least 1.5 years of age are estimated to be Diabetic. O f these, 225,000 are aware that they have diabetes and 198,000 are on treatment. The risk o f diabetes increases threefold after the age o f 44 years. The male/female prevalence ratio i s 1.1to 1.O. Factors that appear to enhance the risko f diabetes inNigeria include urban living, overweight, physical inactivity, alcohol abuse and diabetes ina parent (1990-92 national survey o fNCDs inNigeria). l4This section i s based on information providedby Dr. M.E. Anibueze at the CSG!/NC NCDs at the FMOH. 20 Chapter 1 47. Halfof onepercent (240,000) of adult Nigerians has Sickle Cell disease (SCD) i.e. SS or SC. This figure, however, doesnot include the large paediatric populationof SCD patients not covered by the survey. Sickle cell trait (AS) accounts for 23.04% (10.9 million) o f adult Nigerians. 48. Datafrom the national cancer registry show that 100,000new cases of cancer are currently diagnosed every year in Nigeria. 1out o f every 6 Nigerians will develop cancer. The five most common cancers inNigeria are: cancer o f the cervix, primary cancer of the liver (Hepatoma), breast cancer, lymphomas, and prostrate cancer. 49. Thereare as yet no community-based data on Asthmaprevalence in Nigeria. However, Hospital based data indicate that: (i) asthma i s the most common chronic disorder in childhood; (ii) 6millionNigerians,mostlychildrenareasthmatic; and(iii) than5%ofasthmatics about less inNigeria receive appropriate medical care. 50. As at 2001, Nigeria ranked second on the weighted scale of countries with very high road traf$c crashes. (WHO Nigeriapublication on Road safety 2004). Similarly, accordingto data from the Federal Road Safety Commission (FRSC), over 7,000 Nigerians die every year from road traffic crashes, while over 26,000 injuries are recorded. From2000 - 2002, the annual death toll from road crashes inNigeria stood at more than 8,400 from about 17,000 road crashes. This i s an average o f one death inevery two crashes. Nationwide, a total of 208,361 cases of road traffic crashes were recordedby FRSC from 1990 - 2001. These resultedin 81,657 deaths and 238,573 injured people Between January and November 2003,4,5 14 cases of road traffic crashes were recordedinLagos state. A total o f 742 people lost their lives, while 1,903 were injured. THE MILLENNIUM DEVELOPMENT GOALS 51. InSeptember 2001at the UNMillennium General Assembly, the international community endorsed the Millennium Development Goals (MDGs) with the aim o f promotingpoverty reduction and human development. Five out of these eight goals are related to health: (i) eradicate extreme poverty and hunger inthe world; (ii) child mortality; (iii) reduce improve maternalhealth; (iv) combat HIVIAIDS,malaria and other diseases; and (v) ensure environmental sustainability. 21 Chapter I Figure 12. Progress towards meetingthe MillenniumDevelopmentGoals, Nigeria, 1990-2003 MDG No. 2: Halve the proportion of people who suffer from hunger (under 5 MDG No.4: Reduce by 213 the under 50 , chronic malnutrition) five mortality 250 40 . 200 30 150 20 5 100 lo 0 ~ 1980 1990 2000 2010 2020 1989-1993 1994-1998 1999-2003 2015 MDG No.5: Increase the proportion of birth MDG No.7: Halve the proportion of attended by skilled personnel people without access to safe water -4 0 8 (YOusing surface water as source) 50 40 ::t+, 60 30 20 0 lo` 0 1980 1990 2000 2010 2020 1990 1999 2003 2015 Sources: NDHS 1990, 1999,2003. Note: The unbroken line shows the trends in some of the MDGs indicators while the dashed line the path that the country should have followed in order to meet the goals by the year 2015. 52. As seen in the summary table (Table IO), Nigeria's MDG indicators have a mixed record when compared to other SSA countries. While progress has been made toward some o f the goals, much effort i s needed to get on track inothers. The percentage o f children suffering from chronic malnutrition (stunting) seems to have decreased inthe last decade, fi-om43.1% of under-five children measured by the 1990 NDHS to 38.3% measured by the 2003 NDHS. Basedon this indicator, the country i s achieving progress towards the hunger MDG (although still slightly o f f track) (Figure 12). Similarly, it seems likely that the country will reachthe MDG o f reducing by halfthe number o fpeople without improved sources o f water and sanitation. For example, in 1990,40% o fhouseholds drew water from rivers, lakes or ponds, and this had decreased to 21% by 2003. Nevertheless, despite this encouraging trend, the 2003 NDHS found that only 42% o f Nigerian households had a source o f clean water." Sources expectedto be relatively free o fdisease are pipedwater andprotectedwells. Access to these sources was measuredby the 2003 NDHS,but the 1990 NDHSand 1999NDHS did not distinguishbetweenprotectedand unprotectedwells. We have,therefore, limited analysis of trends to the proportionof householdsrelying on surface water as their (unsafe) source of drinking water. 22 Chapter 1 Table 10. Meeting the Millennium DevelopmentGoals Target ~~~ Indicator/ProxyIndicator Nigeria SSA Cameroon Benin Ethiopia MDG 1: Eradicate Extreme Poverty and Hunger Halve, between 1990 % population below $1 a day 49 33 82 and 2015, the proportion income per capita (constant 1995 us$) 254 675 414 116 of people who suffer from hunger Prevalenceof chronic malnutrition(under-5s) YObelow -3SD 9 11 26 % below -2SD 38 41 15 30 51 % infants < 6 mos. exclusively breastfed 17 % infants 2 mos. exclusively breastfed 26 19 63 79 MDG 4: Reduce child mortality Reduce by two-thirds, infant mortality rate (per 1,000 live births) 100 105 97 94 116 between 1990 and 2015, 201 164 155 158 173 the under-fivemortality under five mortality rate (per 1,000 live births) rate vaccinated against measles by 12 months of age 30 58 62 65 52 Yoof children with diarrheawho receive ORT 40 33 32 19 MDG 5: Improve maternal health Reduce by three- maternal mortality ratio (per 100,000 births) 800 920 730 850 871 quarters, between 1990 % deliveries by qualified attendant 36 43 56 60 10 and 2015, the maternal mortality ratio MDG 6: Combat HIVIAIDS, malaria, and other diseases Have halted by 2015 and HIV prevalence (YOadults) 5 9 13 4 8 begunto reversethe contraceptive prevalence rate all women 9 21 16 16 8 spread of HIV/AIDS Yo using condom during last high risk sexual act male 46 29 21 30 female 24 21 15 9 14 number of children orphaned by HIV/AIDS 1 M 2.2 M 210,000 34,000 990,000 Have halted by 2015 and % under-5 children sleeping under ITN 1 1 7 begun to reverse the TB incidence (per 100,000) 293 345 180 87 356 incidence Of malaria and other major diseases TB cases detected under DOTS (%) 18 86 94 36 MDG 7: Ensure environmental sustainability Halve, between 1990 % with improved water source 42 58 58 63 24 and 2015, the People % with improved sanitation 74 53 79 23 12 without sustainable access to safe drinking water Sources: Data for Nigeria come from NDHS 1999, NDHS 2003, FMOH HlVlsyphilis sentinel survey 2004, WHO (TB and MMR data). For other countries data come from: http://www.develoDmentnoaIs.org,WHO and DHS. Data on income per capita come from World Development Indicators. Note: all data correspond to the most recent year available. 53. The country is noticeably offtrack in reducing child and maternal mortality. Inthe last decade, child mortality rates have not changed much; significant improvement will be required to attain the fourth goal of reducing under-five mortality by two-thirds. Similarly, it i s very unlikely that maternal mortality will be reduced by three-quarters by the year 2015. For instance, the process indicator o f assuring that by the year 2015 all births will be attended by skilled personnel has not shown any progress inthe last decade. For Nigeria to be on track in attainingthis goal, more than 70% o f all births should be already professionally attended, which would entail doubling the current rate o f 36%. 54. Nigeria has also shown limitedprogress in controlling TB and malaria. The incidence o f TB inNigeria is lower than insome o fthe other countries inSSA; however, the percentage o f detected cases usingDOTS (18%) i s the smallest o f all o f the countries where data are available and very far from the global goal o f 70%. Inthe case o f malaria, the country adopted the Roll Back Malaria strategy in2000 and developed a strategic plan in2001, but the success o f this strategy will largely depend on improvements inprimary health care. Additionally, the coverage 23 Chapter 1 o f insecticide-treated nets (ITNs) i s extremely low, at around 1% o f households, although around 11% of households use untreatednets. 55. There are some encouraging data on combating HIV/AIDS. The most recent sentinel surveys suggests that prevalence among adults have not change much inthe past few years, and it i s currently estimated to be 5.0%. Although this i s not enough to conclude that the epidemic has stabilized, it i s nonetheless an encouraging development. Similarly, the proportion o fboth men and women using condoms during their last highrisk sexual act i s higher inNigeria than inother SSA countries (Table 10). 24 Chapter 2 CHAPTER 2. HOUSEHOLDAND COMMUNITYFACTORS AND HEALTHSERVICE UTILIZATION 1. The healtho f an individualis not only influencedby access to quality health services, but also by the knowledge, behaviors, and actions o f individuals and households. Community related factors such as the environment and infrastructure also influence health outcomes. This chapter analyzes the extent to which some o f these individual and household behaviors and actions, and community factors, have an effect on health inthe Nigerian context. It starts by analyzing access to improved sources o f drinking water and sanitation; it then discusses household knowledge and behavior that affect children's health and nutritional status, behaviors and actions that influence maternal and reproductivehealth, and morbidity and utilization patterns. 2. The chapter also analyzes household survey data on health service utilization, including barriers and utilization by type o f provider, Figure 13. Source of improvedwater and sanitationacross wealth quintiles, Nigeria, 2003 0.80 0.60 0.40 0.20 0.00 poorest poorer middle richer richest non-shared sanitation 0 sanitation Authors' estimatesfrom NDHS 2003 ACCESS IMPROVEDSOURCESOFWATER AND SANITATION TO 3, Only a relatively smallpercentage of Nigerian households have access to improved sources of drinking water and sanitation.'6 Having access to safe water and sanitation prevents a number o f diarrheic and parasitic diseases that are water-borne or are associated with inadequate sanitation. This i s specially the case for children, as diarrheic illnesses are one o f the major causes o f child morbidity and mortality inthe country. Only 42% o fNigerianhouseholds have access to safe sources o f drinkingwater and only 74% have access to an improved sanitation facility. Ifa more strict definition o f improved sanitation i s used, so that only households that do not share their facilities are included, only 37% have access to an improved sanitation facility." 4. Thepoorest 20% of households have almost no access to safe sources of drinking water and only 33% have access to an improved sanitationfacility. As can be seen inFigure 13, only about 6% o f the poorest households have access to an improved source o f water compared to more than 80% o f the richest households. Similarly, almost all richhouseholds have access to a flushtoilet l6Improvedsourcesof drinking water includepipedwater, public taps, andprotectedwells, while improvedsources of sanitationfacilities include flushtoilets and latrines. "Householdwithflushtoiletsandlatrinesarealsomorelikelytosharethemwithotherhouseholds,thereforethis definition shouldbe treatedwith caution. 25 Chapter 2 or to a latrine (although many share these facilities with another household) relative to only 25% o f the poorest households. Lack o f access to clean water renders the poor twice as vulnerable because they are more exposed to illness and they also have to spend more time fetching water from open sources. _ _ _ _ ~ Figure 14. Access to improvedwater and sanitationfacilities across regions and across rural and urban areas, Nigeria, 2003 1.oo 0.80 ' 0 ° 8 0 60 0.60 0 40 0.40 0 20 0.20 0 00 0.00 urban rural North North North South South South 0water Wnon-sharedsanitation 0sanitation Central East West East South West 0water W non-sharedsanitation 0sanitation Source: NDHS2003 5, While there are marked differences in access to improved water and sanitation between rural and urban areas, differences are evident across regions in terms of access to safe water but not in terms of access to sanitation. As shown inFigure 14, households inurban areas are more than twice as likely to have access to a safe drinkingwater source comparedto households inrural areas; similarly, urban households are more than 1.3 times as likely to have access to an improved source o f sanitation compared to rural ones. Regional differences seem to be less marked: households inthe Southernregionshave better access to safe drinkingwater than those inthe North, although there are no large regional differences inaccess to improved sanitation facilities. However, a closer look at the NDHS data shows that there are large differences inthe type o f improved sanitation facility across regions. For instance, more than 20% o f the households inthe Southern regions have access to flushtoilets compared with less than 10% o f those inthe Northern regions. HOUSEHOLD BEHAVIOR AND ACTIONS AFFECTING CHILD HEALTHAND NUTRITION 6. A review o f the literature on child survival identified a few preventive and curative interventions that have proven to be effective inreducing mortality and that are also feasible to deliver at highcoverage indeveloping countries (Jones, et. al., 2003). The study identified the following interventions as effective inpreventing diarrhea, pneumonia, measles, and malaria: breastfeeding, complementary feeding, insecticide-treated materials, water and sanitation, HiB vaccine, zinc, vitamin A, and measles vaccine. It also identified the following effective treatment interventions: oral rehydration therapy (ORT), antibiotics for pneumonia, antimalarials, antibiotics for dysentery, zinc, and vitamin A. Many o f these interventions depend directly on household knowledge and behavior and insome cases, on household demand (breastfeeding, health care, etc.). The following sections will examine Nigerianhouseholds' knowledge and practices regarding some o f these key interventions. 26 Chapter 2 Breastfeeding 7. Despite large increases in the last decade, veryfew children are exclusively breastfed during theirfirst six months of life. Although breastfeeding ingeneral i s a widespread practice in Nigeria (almost all children younger than 11months are breastfed), less than one fifth o f children are exclusively breastfed. This i s one o f the lowest exclusive breastfeeding rates among the SSA countries for which DHS data are availableI8. Nevertheless, as Figure 15 shows, the percentage o fbreastfed children has markedly increased ina decade. Figure 15. Percentage of childrenexclusively breastfed across years and months since birth 30.00 25.00 20.00 15.00 10.00 5.00 0.00 Nigeria 1990 Nigeria2003 Sources: NDHS 1990,NDHS 1999,and NDHS 2003, based on children breastfed in the last 24 hours. 8. The multivariate analysis indicates that theprobability that a child under six months of age is exclusively breastfed increases with the mother's level of education and signi$cantly varies with the region of residence. Once other variable^'^ are taken into account, the age o f the child, the education o f the mother, and the region o f residence are the factors found to have a significant correlation with the probability that a child i s exclusively breastfed. As can be observed inFigure 16, everything else heldequal, the education o f the mother increases the probability that the child i s exclusively breastfed. The effect increases with the mother's level o f education. Similarly, everything else constant, children from the North Central and South West regions are much more likely to be exclusively breastfed. '*Mainly poorer countries than Nigeria such as Chad, CAR, Niger, and Comoroshave lower rates o f exclusive breastfeedingamong children younger than 6 months. The multivariate analysis includes the following variables: child's age inmonths, child's gender, mother's educationalattainment, level ofhouseholdwealth, region or residence, a variable indicating urban residence, and variables indicating access to health services. See the analysis inthe annex. 27 Chapter 2 Figure 16. Effect of maternal education and region of residencein the probability that a child is exclusively breastfed 0 30 0.06 0 25 0.04 0 20 0.02 0.00 0 15 -0.02 0 10 -0.04 0 05 -0.06 0 00 -0.08 -0.10 J Note: The first graph shows results in relation to a motherwith no education, while the second shows results in relation to the South West region. Authors' estimates from a multivariate probit analysis using the NDHS 2003 Treatmentof Diarrhea 9. Diarrhea i s one o fthe major causes o f child mortality inNigeria. Oral rehydration salts (ORS) or any oral rehydration therapy (ORT) i s effective inpreventing dehydration due to diarrhea. These treatments can be provided directly by the households. As seen inTable 11, the use of ORSpackets or of any ORT (including recommended home remedies) has markedly increased in the last decade. Inthe case o f ORs, the relative increase has been almost 60%. The increase inusage has occurredmostly inrural areas, while the usage inurban areas has remained the same or has slightly decreased. Despitethis positive trend, Nigeria has still one o f the lowest ORT usage rates inall the SSA countries for which DHS data are available. Similarly, harmful home food practices when the child has diarrhea remainprevalent. For instance, data from the NDHS 2003 show not only that about 40% o f the children with diarrhea received fewer fluids, but also that about halfo fthem receivedless food. Table 11,Treatment of diarrhea, Nigeria, 1990-2003 (% of childrenwith diarrhea in previous 2 weeks). % difference 1990 2003 1990-2003 Urban rural overall urban rural overall overall treatment from health provider 40.3 22.6 25.1 30.3 18.8 21.5 -0.14 ORS packet 25.5 9.3 11.6 22.9 16.8 18.2 0.57 any ORT 57.1 29.6 33.5 49 37.5 40.2 0.2 Source: NDHS 1990,2003. 10. Poor children are much less likely to receive the recommended treatmentfor diarrhea. Data from the NDHS 2003 shows that children from the poorest 20 % o f households are half as likely to receive ORS or any ORT incase o f diarrhea as children from the 20 % richest households. Only 33% o fpoor childrenreceive the recommended treatment. 28 Chapter 2 Insecticide treated materials 11. The utilization o f insecticide treated bed-nets effectively prevents malaria which i s one o f the main causes o f child mortality and morbidity inthe country. Despite being a country were the disease i s endemic, data from the NDHS 2003 show that only about 10% o f households in Nigeria own a mosquito net and only 2% o f households own insecticide treated bednets. Almost no children under five years o f age sleep under an insecticide treated net (the estimate i s 1.2%). Utilization of child health services 12. The differences inchild health status that were detailed inthe previous chapter are partly the result o f differences inutilization o f bothpreventive and curative health services. Households differ interms o f knowledge and cultural norms regardinghealth care, they face different time and resource constraints inusinghealth services, and they also differ interms o f physical access to a health care providers. This section examines differences and some determinants o f utilization o f the following children's health interventions: immunization, vitamin A supplementation, and health care utilization duringepisodes o f diarrhea, fever and cough. Inthis section we will only analyze household and individual factors that affect the utilization o f health services. Inthe following chapters we will examine supply factors. Utilization and coverage of child immunization services 13. With the data available, it i s not possible to differentiate between the demand and supply constraints that have resultedin such a low level o f immunization inthe country.2o This section only examines some factors that mighthinder the demand for children's vaccinations. InChapter 3, we will revisit immunization by examining problems with the supply and distribution o f vaccines. Table 12. Immunizationcoverage, Nigeria, 1990-2003 (% children 12-23 months) Yodifference 1990 2003 1990-2003 urban rural overall urban rural overall overall measles 68.8 39.8 46.0 52.1 28.5 35.9 -22% DPT3 58.9 26.4 33.3 40.2 12.8 21.4 -36% all vaccinations 52.5 23.3 29.6 25.1 7.4 12.9 -56% no vaccinations 16.3 42.4 36.8 16.7 31 26.5 -28% Sources are 1990 and 2003 NDHS. 14. Immunization rates in Nigeria are extremelypoor and have declined significantly in the last decade. As shown inTable 12, the 2003 NDHS found that only 13%*' o f one-year-old children hadreceivedall recommendedimmunizations, while 26.5% had not received any vaccinations. Only 21% obtained three doses o f diptheria-pertussis-tetanus (DPT3) vaccine, often used as a performance indicator for routine immunization. Measles immunization, sometimes provided in 2o The NDHS 2003 does not have information on physical access and quality of health services. This limits multivariate analysis to household and individual variables. 2' Estimates from the 1999NDHSare similar, although slightly higher (around 5% higher for the different immunizationindicators). 29 Chapter 2 campaigns, i s a little higher, at 35.9%. Coverage o f each type o f vaccination (except for the first few doses o f polio vaccine) i s significantly lower than in 1990.22 15. Immunization coverage in Nigeria is among the lowest in Sub-SaharanAfrica (and the world). Figure 17 shows that Nigeria's measles immunization coverage i s comparable to those o f war-affected countries such as Sierra Leone or Somalia. The numerous points inthe upper left quadrant o f the graph represent countries such as Ghana and Uganda which have lower (or similar) GDP per capita but which have achieved higher immunization coverage than Nigeria. 16. There are large differences in immunization coverage across regions,place of residence, and across income groups. Rural areas and poorer regions have dramatically lower immunization coverage although urban areas have also experienced a decline, indicating a significant deterioration o f routine immunization inmany areas. Table 12 shows that immunization coverage inrural areas i s considerably lower than inurban areas. For example, only 13% o f one- year-old children inrural areas have receivedDPT3 and only 7% have obtained all recommended vaccinations, compared to 42 % and 25% inurban areas. The NDHS 2003 shows that regional differences inmeasles immunization coverage closely follow regional economic differences, as coverage inthe North-East and North-West i s only 16% to 23%, compared to 64% to 73% inthe South. Similarly, coverage o f DPT3 and all vaccinations are lower inthe North. These data indicate severe weakness inroutine immunization inthese regions as well as inpoorer rural areas inthe rest o fthe country. These largeregional differences inimmunizationpersists even after controlling for other variables such as mother's education, household income status, and an indicator for access to childhealth services. Figure 18 shows that, everything else constant, living ina region different from the South West (the reference region) decreases the probability o f being immunized. This i s especially the case for the Northernregions where immunization coverage i s the lowest, particularly for polio. Figure 17. Measles immunizationcoverage by GDP per capita (SSA countries with GDP per capita less than US$ 500) 0 Ghana 0 4: 0. 0 0 0 0 0 I Uganda- 0 e! O . Sierra Leone Central African Republic 1 0 100 200 300 400 500 GDP per capita ($) Sources are World Bank HNPStats (estimates are for 2001) and 2003 NDHS. 17. Poor children are much less likely to be immunized than their wealthier counterparts. Controlling for other variables, the wealth status o f the householdhas a large and significant effect on the probability that a childreceives a measles vaccine and receives three doses of DPT. The wealthier the family, the more likely a child has received those immunizations (see Figure 18). For instance, inthe case o f measles, children belonging to the richest quintile o f the wealth distribution are about 25 percentage points more likely to be immunized against measles than 22 The exception i s the proportion of children who had no vaccinations, which has declined due to polio vaccination campaigns. 30 Chapter 2 children belonging to the poorest quintile. Finally, maternal education has a significant effect in increasing the probability that a child will receive all vaccines (see analysis inthe annex). Figure 18. Effect of region of residenceand householdwealth status on the probabilitythat a child is immunized 0.05 030/ 0 00 025 0 20 -0 05 0 15 -0 10 0 10 -0 15 0 05 0 00 1 Poorer Middle Richer Richest -0 25 i- Drneasles mdpt3 Oallpolio Elmeasles Wdpt3 Source: Based on a multinomial probit analysis of determinantsof child immunization. Other variables in the analysis include: maternal education, children's characteristics,and indicator of urban areas, and an indicator for access to children health services. 18. To better illustratewealth inequalities inimmunizationrates, concentration curves were constructed for measles, DPT, polio, full immunization, and any immunization (Figure 19). These concentration curves are similar to inequality Lorenz curves. Inthe horizontal axis they plot the cumulative proportion o f childrenbetween 12 and 23 months o f age ranked inascending order o f their family wealth status. On the vertical axis they plot the cumulative proportion of children receiving a particular vaccine. The 45 degree line represents the line o f full equality, where there are no wealth differences inimmunization rates. The larger the distance betweenthe concentration curve and the 45 degree line, the greater the inequality. -igure 19. ImmunizationConcentration Curves 1 1 0 8 0 8 0 6 0 6 0 4 0 4 0 2 0 2 0 0 0 0 2 0 4 0 6 0 8 1 +Measles +4511ne DPTl +DPTZ +DPT3 +45line 0 8 0.8 0 6 0.6 0 4 0.4 0 2 0 2 0 0 0 0 2 0 4 0 6 0 8 1 --canny"&CI Iy) -4n- FulIInnU"l2SllO"+45line Authors' estimatesfrom NDHS 2003. 19. Figure 19 suggests that children from wealthier families are more likely to use health services and thus more likely to receive routine immunization services than their poor counterparts. 31 Chapter 2 Similarly wealthier children are more likely to be born in a health facility and therefore more likely to obtain the two required immunizations given at birth: B C G (not shown inthe figure) and OPVO. The largest wealth inequalities appear inthe continuity o f these services. As can be observed inthe case o f DPT, the first dose has a similar degree o f inequality as those vaccines that only require one dose (measles and BCG); the inequality slightly increases by the second dose23,and it i s muchhigher by the third dose. Polio i s an exception. As can be seen inthe figure the inequalities inpolio vaccination are very small. Only inthe case o f OPVO are these inequalities large. Incontrast to the other three doses o f polio, OPVO i s not given during campaigns but at birth, which might explain these differences. The polio eradication campaigns have successfully decreased wealth inequalities through active outreach activities. The other routine immunizations are offered inhealth centers only, suggesting that the health system has failed to reach the poor. This also explains the large differences inthe concentration curves for full immunizationand for any immunization. As aresult ofincludingthe polio vaccines the latter curve i s very close to the diagonal line o f full equality; incontrast, the full immunization curve i s far from it, showing a large degree o f wealth inequality. 20. Nigeria i s the second largest global reservoir o f the wild polio virus so that a major priority o f international donors inrecent years has been the polio eradication campaign. This has achieved some success interms o f surveillance and vaccination campaigns, but coverage remains insufficient, particularly for the second and thirddoses. The 2003 DHS found that among one- year-old children, 27% hadreceivedthe vaccine at birth(OPVO), 67% had receivedthe first dose (OPVl), 53% OPV2, and 30% OPV3. Incontrast to measles and DPTvaccination, there was no significant household wealth effect on the probability o f having received all polio vaccines (not shown inthe figure). Vitamin A 21, A review o f the literature (Jones et al., 2003) found clear evidence that vitamin A prevents mortality due to diarrhea and i s also an effective treatment for measles. The review also found limited evidence that vitamin A can prevent measles and malaria. Data from the NDHS 2003 indicate that less than half (43%) o f children younger than three years o f age consume fruits and vegetables rich invitamin A. Similarly, among childrenbetween six month and five years o f age only about 34% receivedvitamin A supplementation inthe six months preceding the survey. 22. To better understand which household and individual variables are correlated with the probability that a childbetween six months and five years o f age receives vitamin A supplementation, a multivariate analysis was performedusing the NDHS 2003 data.24 This analysis shows that maternal education, region of residence, household wealth, and access to children's health sewices (proxied by thepercentage of children in the community that had received measles immunization) significantly affect theprobability of receiving vitamin A supplements. Figure 20, indicates that, after controlling for the effect o f other variables, maternal education increases the probability that a child receives vitamin A supplementation. Even after controlling for householdwealth status, the region o f residence remains one of the more significant determinants o f vitamin A supplementation. The children living inthe North are less likely to receive this supplement compared to children living inthe South. 23The difference i s small and it i s probably not significant. 24See complete regression in annex. 32 Chapter 2 Figure 20. Effect of maternaleducationand region of residenceon the probabilityof receiving vitamin A supplementation - - 0.00 0 14 0 12 -0 05 0 10 -0.10 0 08 0 06 -0.15 0 04 -0.20 0 02 -0.25 0 00 primary higher -0.30 ' Authors' estimates from a probit analysisof determinantsof vitamin A supplementation using the NDHS 2003 23. Children belonging to households inthe richest two quintiles are significantly more likely to receive Vitamin A supplements than those belonging to the poorest quintile. The effect i s not statistically significant for children living inthe second and thirdquintiles o f the distribution. Utilization of health services in case of diarrhea or pneumonia 24. Veryfew Nigerian children receive medical treatment in case of diarrhea or in case of afever or cough. Less than one out o f every four children receives medical treatment 25 from a trained health care provider in case o f diarrhea and one out o f every three in case o f fever or cough. Whether or not a child i s taken to a health provider and the type o f healthprovider sought i s highly influencedby the wealth status o f the household. As seen inTable 13, while almost 70% of children with diarrhea inthe poorest quintile o f the population were not taken to any health provider, only 42% o f children inthe richest quintile were not. Similarly, while only 10% o f the poorest children received medical treatment, 35% o f the richest children did. N o t surprisingly, the rich are more likely to take their children to private providers; however, the rich are also more likely to use public facilities than the poor. Finally, more poor children are taken to pharmacies and traditional and spiritual healers than the richest children. A similar pattern i s also found in the case o f children with cough and fever. Table 13. Percentage of children taken to a health care provider across wealth quintiles Wealth index poorest poorer middle richer richest overall Diarrhea No Care 65.9 65.0 53.0 42.7 42.4 56.5 Public hospital 4.2 2.5 9.0 19.4 19.8 9.0 Health Center 2.0 3.2 8.8 12.9 3.8 6.0 Health post/ mobile clinic/CHW 1.o 4.3 1.7 1.4 1.7 2.2 Private hospitallclinic 1.6 0.8 0.8 4.4 8.8 2.4 Private doctor 0.0 0.0 0.5 0.0 0.0 0.1 Private other 0.6 0.2 0.0 0.0 0.5 0.2 Pharmacy 19.7 13.6 16.5 18.2 10.2 16.3 Traditional/spirituaIhealer/other 5.2 10.5 9.8 1.o 12.9 7.4 25 Excludes pharmacies, shops, traditional, and spiritual healers. 33 Chapter 2 Wealth index poorest poorer middle richer richest overall Coughlfever No Care 32.4 27.3 18.1 9.2 15.2 21.2 Public hospital 4.7 6.8 11.9 18.7 19.8 11.6 Health Center 4.3 7.5 9.1 15.2 6.7 8.5 Health post/ mobile clinic/CHW 3.4 4.3 4.1 1.9 1.I 3.2 Private hospital/clinic 2.8 2.4 3.8 7.0 23.5 6.8 Private doctor 0.2 0.4 1.2 0.2 0.9 0.6 Private other 0.7 0.4 0.5 0.0 0.0 0.4 Pharmacy 39.7 38.1 41.6 44.2 29.4 39.1 Traditionallspiritual healer/other 11.7 12.7 9.8 3.6 3.4 8.8 Source: NDHS 2003 Note: All cases were more than one responsewas given were not included in the analysis. This excluded 16 observations out of 845 in the case of diarrhea and 53 out of more than 1500 in the case of fever or cough. 25. When taking children to apublic sector healthproviderfor medical attention, about half of households bypassed lower levelfacilities such as health posts, mobile clinics, and community health workers and went directly to higher levelfacilities in searchfor care. It i s also worth noting that incases o f diarrhea, fever or cough, households rely significantly on pharmacies to seek treatment for children. Inparticular, treatment was sought from a pharmacy for about 25% o f children with diarrhea and for more than 33% o f children with fever or cough. 26. A multivariate analysis o fthe determinants o f utilization o f health care services incase o f diarrhea or fevedcough confirms that household wealth i s a significant determinant o f utilization. Similarly, maternal education and region o f residence are also relatedto variations inutilization. The higher the level o f maternal education the more likely the child will be taken to a health provider. Similarly, children from the South are more likely to be taken to a provider incase o f diarrhea or incase o f cough or fever (see complete analysis inthe annex). MATERNALREPRODUCTIVE AND HEALTH Contraception 27. The risk of maternal and child mortality and morbidity can be reduced ifthe mother has control of the number and spacing of births she has. Knowledge and access to modern methods o f contraception as well as decision making power within the householdinfluence the extent to which women have control over their pregnancies. 34 Chapter 2 Figure 21. Current used of modern contraceptionamongcurrently married women across income, education, and across regions of residence, Nigeria,2003 Richer Rich Middle Poorer Poorest Higher Secondary Primary No education South West South South South East North West North East North Central 0 5 10 15 20 25 Source: NDHS 2003 28. Knowledge of modern methods of contraception is increasing. In 1990 the NDHSreported that only 40% o f married women had any knowledge o f modem methods o f contraception, by 1999 this percentagehad increasedto 60%. In2003, 76% o f married women knew about at least one modem method o f contraception. 29. Despite this large increase in knowledge of modern methods of contraception, their current use is very limited. Data from the NDHS 2003 indicate that only 12% o f currently married women use any method o f contraception and only 8% use a modem one. The usage o f contraception largely varies with the level of women's education, their income level, the degree o f autonomy they have in family planning decisions, and access to family planning services. As can be seen inFigure 21, the percentage o f women using modem contraception i s 11times higher among women with higher education than among those without. The poor are also much less likely to use any modem methodthan the rich. Moreover women living inthe Northem regions, especially inthe NorthWest and North East are much less likely to use a modem method of contraception than those inthe south. 30. It i s very difficult for household surveys to capture the level o f empowermentwomen have inside the household. The NDHS 2003 asked the women interviewed whether they participated intaking decisions relatedto their own health, large householdpurchases, visits to family or relatives, children's health care and others. This surveyfound apositive correlation between the number of decisions where the women have a j n a l say and thepercentage of women using a 35 Chapter 2 modern method of contraception. Women livinginthe Northernregions and women with no education are less likely to participate inmany o f these decisions; they are also the least likely to use modem methods. 31. Despite their low utilization, access to contraceptives is relatively high. NDHS 1999 data show that modem contraceptives are available within 5 kmto 70-80% o f households inurban areas and to over 50% o f households inrural areas. Marriedwomen interviewed inthe NDHS 2003 cited their desire to have more children (36%), own or husband's opposition to their use (18%), and lack o fknowledge o f a method (8.4%), as reasons for not usingcontraceptives. Only less than 2% o f women reported not having access to any method, not knowing a source, or that their costs were high. 32. There is still some unmet needfor family planning to space and limit childbirth. The NDHS 2003 data report that, on average, 17% o f currently married women have expressed an unmeet need for family planning (12 percent for spacing childbirth, and 5 percent for limitingit). The unmet need for spacing i s largest inthe NorthCentralregion, while the unmet need for limitingi s highest inthe South South. Nonetheless this i s still one o f the lowest rates o f unmet need for family planning inall SSA countries for which DHS data are available. Table 14. Percentage of women who received professionalantenatal and delivery care, and percentage of women that delivered in a health facility, Nigeria antenatal care by delivery by professional professional delivery in health facility Urban 1990 85.2 60.8 58.2 2003 83.0 58.8 54.2 rural 1990 52.3 25.7 23.6 2003 50.6 27.1 23.8 overall 1990 59.3 33.0 30.9 2003 60.1 36.3 32.6 Notes: The denominatorfor antenatalcare is the most recent birth among women who had a live birth in the previous 5 years, while the denominatorfor delivery care is all live births in the previous 5 years. Health professionals are: doctor, nurse, midwife, auxiliary midwife, or community health worker. Sources are 1990 and 2003 NDHS. Utilization of pre-natal and delivery care 33. Utilization ofprenatal and delivery care, although consistent with other countries of similar incomeper capita, has not improved in the last decade. Table 14 shows that the proportions o f women who received antenatal care, whose children were deliveredby a health professional, and who deliveredina health facility, have remained largely unchangedbetween 1990 and 2003. Despite this, the level o fprofessional delivery care inthe country i s similar to other Sub-Saharan Africa countries with GDP per capita o f under US$500. 34. There are significant urban/rural, regional, and economic disparities in utilization and coverage of maternal health services. Coverage o f delivery by a healthprofessional inrural areas i s only 27%, compared to 59% inurban areas. As Figure 22 shows, the poor inboth urban and rural areas are more likely to receive care from a traditional birthattendant, a relative, or no one at all. Conversely, inboth urban and rural areas, the better-off are muchmore likely to receive delivery care from a nurse, midwife, or a doctor. These data also indicate that nurses and midwives are the most common care providers for all economic levels inurban areas, and among 36 Chapter 2 the better-offinrural areas. Finally, the NDHS2003 also shows enormous regional disparities, with less than 22% ofbirthsattended byhealthprofessionals inthe North-West andNorth-East, compared to 80% or more inparts o f the south. Figure 22. Assistance during delivery (percentageof births)across wealth quintiles Rural areas 0 9- Urban areas 0.8 - 0 8- 0 4 0 3 0 2 0 2 0 1 0 1 0 --Cdoctora1 Q2 q3 Q4 Q5 01 02 03 nurse/rnidwife nurse/midwife05 04 +doctor +CHW traditional birht and CHW birht attd relative/friend -0- noone --wCrelativelfnend --ttraditional no one Source: NDHS 2003 Figure 23. Effect of women's education and householdwealth on the probability of more than two antenatal care visits, of being assisted by skilled personnel during delivery, and of delivering in a health facility, Nigeria 2003 0 40 0 35 0 35 0 30 0 30 0 25 0 25 0 20 0 20 0 15 0 15 0 io 010 0 05 0 05 0 00 0 00 primary higher Poorer Middle Richer Richest More than 2 antenatal visits .Assisted delivery nMorethan2 antenatalvisits WAssisted ddivery OBirth inhealthfacility OBtrthinhealthfacility Authors' estimatesfrom probit analysis of data from NDHS 2003. 35. These income and regional differences persist after taking into account other variables that affect utilization of maternal health care services. A multivariate analysis o f some of the factors that affect utilization o f maternalhealth care services was performed (see complete analysis in annex). This analysis shows, that holding everything else constant, the probability o f going to 37 Chapter 2 more than two antenatal care visits, o fbeing attended by skilled personnel duringdelivery, and delivering ina health care facility increases with the level o f women's education and with the wealth status of the household (see Figure 23). The regional differences (not shown) also persist once the effect of all other variables are taken into account. Harmfultraditionalpractices 36. Harmful traditional practices such asfemale genital mutilation (FGM) are stillprevalent. Every year about 132 million acts o f FGMare practicedaround the world, almost a quarter o f these are inNigeria. Immediatehealthproblems associated with this practice are: severe bleeding, pain, shock, and infections. Among the longterm effects: bleeding, recurrent urinary track infections, incontinence, chronic pelvic infections, infertility, fistulae, sexual dysfunction, and problems inpregnancy and childbirth. Data from the NDHS2003 indicate that about 19% o f all women 15-49are circumcised, the majority living inurban areas and in the Southern regions o f the country. Only a very small percentage o f cases, about 4%, are infibulations, the most extreme form o f FGM. 37. There are other practicesprevalent in Nigeria with detrimental effects on the health of the mother and the child such as early marriage and short birth spacing. The NDHS2003 found that one inevery four teenage women (15-19 years) i s pregnant or already has a child. Inthe NorthWest and NorthEast regions about 45% o f teenage women have already begun childbearing, comparedto less than 5% inthe South East and South West. Additionally, the survey found that more than 60% o f women who gave birthinthe last five years had shorter birth intervals than the recommended three years. Figure 24. Knowledge of methods to prevent HIV/AIDS across gender, income level, and place of residence Women Men Rural Rural Urban Urban Richer Richer Rich Rich Middle Middle Poorer Poorer Poorest Poorest 0 50 100 0 50 100 0usingcondoms Elimitingsex 0both using condoms Elimiting sex 0 both Source: NDHS 2003 38 Chapter 2 HIV/AIDSKNOWLEDGE 38. Awareness of the epidemic is almost universal, but knowledge ofprevention is more limited. The NDHS 2003 indicates that about 86% o f all women and 97% of all menhave heard o f the disease. Despite this awareness, only 65% o fwomen and only 85% o f menbelieve that there i s a way to avoid HIV/AIDS. As seen inFigure 24, women are also much less likely to know o f any method to prevent the spread o f the disease than men. Poor women and women living inrural areas are especially vulnerable. For instance, only 25% o f the poorest women are aware that condom use limits the spread o f the disease, compared to about 66% o f the richest women and 40% o f the poorest men. These figures highlightthe needto increase household and individual information, especially among the most vulnerable groups o f the population, to prevent a further advance o f the disease. 39. Even though the situation i s worrisome, there has been a marked improvement inoverall awareness indicating that awareness campaigns may have had an effect, although the rural poor are less likely to be reached. The 1999NDHS found that less than 75 % o f women knew about the disease, compared to 86% in2003. Moreover it seems that this knowledge i s starting to be translated into practice, albeit slowly; the percentage o f condom use duringthe last highrisk sexual intercourse i s about 46% for men and 23% for women. Nevertheless, these percentages are very low, especially among women. MORBIDITY CARESEEKINGBEHAVIOR AND 40. The 2004 Nigeria Living Standards Survey (NLSS) estimated that 12%of thepopulation had an illness or injuly in theprevious two weeks (Table 15). 26 Table 15. Percentage of people reporting an illness or injury in the last two weeks by age, Nigeria, 2004 `' Age Total 0-5 16.2% 6-14 9.0% 15-24 8.0% 25-45 10.5% 46-69 15.3% r70 29.0% All 11.6% Authors' estimates from 2004 NLSS. 41. Less than 60% ofpeople reporting an illness or injury in the last two weeks visited a health careprovider; thepoorest were the least likely to seek care. Only 56% o fNigerians reporting an illness or injury visited a formal or informal health care provider. There are significant income differences inthe percentage o f people seeking care incase o f illness or injury. The richest 20 % o f the populationi s about 2 times more likely to seek care in case o f illness or injury than the poorest 20 % (Figure 25). 26 As this is a self reported measured of illness, the percentage of rich people reporting an illness is higher than those among the poor. This is relatively common using this type o f measure. Given that the Southern regions are richer than the Northern ones, more people inthe South reported an illness or injury inthe last two weeks. Even when restricting the measure o fillness and injuryto those that forced the person to stop their normal activities, ahigher percentage of rich people reported an illness or injury. *'There are measurement issues related to gender in the NLSS so the gender-disaggregated estimates should be viewed with caution. Incontrast to the DHS, there is a large difference in the female-to-male ratio across regions inthe NLSS, suggesting some underreporting of females in the Northernregions. 39 Chapter 2 Figure 25. Percentage of people reportingand illness or injury in the last two weeks that visited a health care provider,by income quintile, Nigeria, 2004 1 20.0% 1 0.0% ~ 1 Poorest II 111 IV Richest Authors' estimates from 2004 NLSS. 42. When also consideringpeople who went to a health servicefor preventive health care interventions2', the income differences are even larger. The poorest segment o f the population is much less likely to have used a health care provider inthe last two weeks than the richest segment. While 13% of the richest fifth of the population consulted a healthprovider inthe previous two weeks, only 4% o f the poorest fifth did. Inother words, the richest 20 percent o f the population are more than 3 times more likely to use a health care provider for bothpreventive and curative care than the poorest 20 percent. 43. Among medical providers, utilization is divided equally between hospitals and PHC services. The 2004 NLSS found that, among people o f all ages who visited a health care provider, 40% went to a hospital and 38% went to a PHC provider (Table 16). Table 16. Health service utilization by type of provider, Nigeria,2004 (% of children and adults illor injured in the previous two weeks who received care) (n = 7,028) urban rural overalI Hospital 51 32 40 Public 37 23 29 for-profit 13 a 10 non-profit 1 1 1 PHC provider 28 45 38 Public 11 25 19 for-profit 16 l a 17 non-profit 2 2 2 Other 21 23 22 Total 100 100 100 Authors' estimatesfrom 2004 NLSS data. 44. Among medical providers, around two-thirds of patients go to thepublic sector and one-third usesprivate sectorproviders. The 2004 NLSS found that among children and adults o f all ages '*The NLSS collectedinformation on visits to healthcare providersinthe last two weeks beforethe interview. Some of these visits were not linkedto an illness episode as i s the case of peoplereceivedapreventivehealthservice from these providers(e.g.vaccinations, pre-natalcare). 40 Chapter 2 who visited a health service, 48% went to a public sector facility, 27% went to a private for-profit provider, and 3% went to a private non-profit provider (Table 16). Three quarters o f hospitals consulted were inthe public sector while PHC providers consulted were evenly split between public and private. Ingeneral terms, excluding consideration o f pharmaciedpatent medicine dealers and other non-formal providers, public medical services are consulted about two-thirds o f the time and private providers about one-third o f the time. 45. Rural residents are less likely to use hospitals and more likely to use PHC services, but there seems to be little difference in the use ofprivate sector services. Among patients o f all ages, the 2004 NLSS found that 32% inrural areas went to a hospital, compared to 51% inurban areas; conversely, 45% inrural areas went to a PHC provider, compared to 28% inurban areas (Table 16). However, there does not seem to be a large urban-rural difference inthe use o f public versus private sector medical providers. Infact, the 2004 NLSS found that the proportion o f patients who used a public sector service was equivalent inrural and urban areas (48%). Similarly, proportions o f people who use non-formal private health providers, particularly patent medicine dealers, are similar inurban and rural areas, althoughpeople inrural areas may be slightlymore likely to go to a traditional healer. 46. Use of hospitals versus PHCproviders is similar across the county, but patients in the north are more likely to go topublic sectorproviders. The 2004 NLSS found that utilization o f hospitals i s for the most part similar across regions, ranging from 3 1to 52% o f patients, compared to a range o f 34 to 47% for PHC providers. However, with regardto both hospitals and PHC facilities, utilization o fpublic sector services i s higher inthe north than inthe south. 47. Utilization patterns are consistent with thefact that higher-level health services,particularly private ones, are more available in the south while lower-level PHC services are more available in the north. The availability o f health care facilities will be discussed inChapter 3. 48. Thepoor are less likely to use hospitals and more likely to go to PHCfacilities. Figure 26 presents estimates from the 2004 NLSS showing that utilization o f PHC providers decreases and hospitals increases as household income rises. The poorest households tend to obtain health care inPHCfacilities such as dispensaries, maternityhomes, M C Hposts, their homes, etc. Although the use of pharmacies and clinics i s fairly similar across all income levels; the use o f dispensaries (a low level PHC facility) decreases with income. About half o f people (44%) seek care from doctors. There are, however, large differences across income levels; the rich are two times more likely than the poor to see a doctor. Finally, the richest households are less likely to receive care from a Medical Assistant or a Traditional Healer. These findings are consistent with the 2003 NDHS data on utilization incases o f child illness presentedinTable 13. 41 Chapter 2 Figure 26. Place of consultationand who was consultedby income quintile 0.6 1 0.6- 0.5 - 0.4 - 0.3 - , , " I I I 1 1 2 3 4 5 1 2 3 4 5 +Traditional Healer -a- Doctor Nurseimidwife +Hospital - 4 s p e n s a Pharmacy Clinic +other Medical Assistant +Pharmacist +Other Authors' estimates from 2004 NLSS. The income quintiles are showed in the horizontal axis. The first quintile represent the poorest 20% of the populationwhile the fifth the richest 20 %. For example, the first graph show that on average among the 20% poorest individualsabout 20% use a hospital as a place of consultation while among the richest 20% more than 50% of them use a hospital as place of consultation. Note: In the left graph, others include Maternity homes, MCH post, Consultants home, patient's home and others. In the right graph, others include TBA, spiritualist, and others. REASONS FOR NOT USING HEALTHSERVICES 49. One in evey two women reports havingproblems in accessing health care services. The DHS2003 asked women between 15-49years ifany o fthe following couldprevent their access to health services when ill:(i) knowing where to go; (ii) permissionto go; (iii) the getting getting money needed for treatment; (iv) the distance to the health facility; (v) having to take transport; (vi) not wanting to go alone; and (vii) concern that there maynot be a female provider. Half o f the women indicatedthat at least one o f these issues represented a problem for them. The largest barrier women face i s that o f not having the resources to afford these services as one inevery three mentionedthat gettingthe money for treatment was a big concern. The second largest concern was that o fphysical access to these services as one inevery four women mentionedthe distance to the services and/or having to take transportation to reachthem. The other problems mentioned were the lack o fknowledge o f where to go in case o f illness (14%), the concern that there may not be a female provider (17%), or not wanting to go alone (14%). 50. Poor women, especially those living in rural areas, are more likely toface barriers to access health services. 75 % o f women belonging to the poorest fifth o f the population mentioned at least one problem in accessing health care when ill.As can be seen inFigure 27, poor women in both urban and rural areas are more likely to report problems in accessinghealth care when ill.In both urban and rural areas financial and knowledge barriers are reportedby a similar percentage o f women. Poor women inboth areas are equally likely to report having financial barriers to access services, as half o f the women belonging to the poorest fifth o f the population reported this problem. Poor women inboth areas are also equally likely to report not knowing where to seek 42 Chapter 2 care. However, physicalbarriers to access are a larger problem for women inrural areas, especially for the poor, as more than 40 % o f women livinginthe poorest two fifths o f the populationreport distance and/or transportation as a bigproblem inaccessing health services. 51. The largest difference in access to health care between rural and urban areas is due to differential 'householdpreferences and behaviors and not so much because of supply constraints, Inthe ruralareas women, especially the poorest, are muchmore likely to report not having permission to seek care, not wanting to go alone to a health provider, or having concerns that there mightnot be a female provider. Figure 27. Women's problems in accessing healthcare across wealth status and place of residence Rural Urban 0.80 0.70 0.70- 0.60 0.60 0.50 050 0.40 0 40 0.30 0 30 0 20 0.20 010 0.10 0 00 0.00 Poorest Poorer Middle Rich Richer Poorest Poorer Middle Rich Richer Owhereto go .Money Ddistance ntransport mother Inwhereto go .Money Odistance ntransport mother Source: NDHS 2003 52. There are also marked differences in women's barriers to health care across regions. According to the DHS 2003, women inthe South South region were the most likely to report a problem inaccessinghealth care, about 60% o f them reported at least one difficulty. Incontrast, women inthe SouthWest region were the least likely to do so, only 16% reported a problem. In all regions except the North West, the major problem inaccessing health services was having the money to pay for them followed by physical access to these services. Inthe North West, the major issue was a concern that these services would not have a female provider followed also by not having physical access to these services. 53. Poor women in all regions are the least likely to have access to health care services (see Figure 28). Thedifferences in access to services across household wealth quintiles are large in all regions but the South West. The South Southregion deserves special attention, as it was in this regionwhere the largestnumber o fwomen reportedbothfinancial andphysical barriers to access health care. Inthis region, about 47% o f the women reported not having the money to pay for these services and 35% reported distance to the services (and/or having to take transportation) as a problem. 43 Chapter 2 Figure 28. Women's problems in accessing healthcare across wealth status and place of residence 0 7 - 0 6 - F, 0 6 - 0 5 - *... . __ \ 0 5 - . 0 4 - \ \ b @- ~ 0 4 1 ---e 0 3 - li- - --a 0 2 - \.\ 0 1 - 0 1 Poorest Poorer Middle Richer Richest b l e s t borer Middle Rcher Rchest +where to go mney distance transport -#-other +where to go -.a- mney distance transport -m-other ::I:\- North East 0 7 - 0.7 0 6 - 0 5 - 0.5 0 4 - =-'.. 0 3 - 0 2 '... 0 1 -.. 0 Poorest horer Mddle Rcher Rchest horest horer Mddle Rcher Rchest twheretogo-C-mney distance transport -X- other t w here to go +mney distance transport -m-othei North West 0 7 South South 0 6 0 7 7 .#e. - - - e % 0 5 0 5 0 4 0 3 ;; 0 2 0 2 0 1 0 1 0 borest borer Mddle Rcher Rchest 0 4 Poorest Poorer Mddle Rcher Rchest --twheretogo--c-- rmney distance transport +other --ewheretogo I- rwney distance transport +other Source: NDHS 2003 44 Chapter 3 CHAPTER 3. THE HEALTHSYSTEMINNIGERIA 1. This chapter describes the organization o fthe health system inNigeria and characteristics o f the service delivery network, both public and private sector. Key components o f the system are assessed, inparticular facilities and equipment, human resources, drug supply, and support services. The health services and public healthprograms that are provided are described. Finally, government healthpolicy and strategy are discussed. HEALTH SYSTEMORGANIZATION AND GOVERNANCE Health System Organization 2. The health system is decentralized under afederal structure. Thefederal level is responsible for overallpolicy as well as tertiary services, the state level is responsiblefor secondary services, and local governments are responsiblefor primary services. Nigeria's Government i s a federation o f 36 States, with the Federal Capital Territory (FCT) and 774 Local Government Areas (LGAs). Governments at the different levels are involved in all o f the major health system functions: stewardship, financing, and service provision. The current (1999) Constitution mentions health only with regardto the responsibilities o f local governments, implyingthat responsibility for health services i s shared between the state and local levels. However, in practice, the division o fresponsibilities i s based on the 1988 National Health Policy and related directives. 3. Under the 1988 policy, the National Health Council, chaired by the federal Minister o f Health and comprising the Commissioners for Health o f all the states, has overall responsibility for national policy. The federal level i s responsible for policy guidance and technical support to the overall system, international relations on health matters, national health management information systems (HMIS), and government service provision by tertiary and teaching hospitals and national laboratories. State governments are responsible for government secondary (and some tertiary) hospitals and for regulation and technical support to primary health care (PHC) services. Operation o f public sector PHC services i s the responsibility of local governments, under which services are organized by district (ward) (with 7 to 10 districts per LGA). 4. For- and non-profit private health care services at the different levels o f care are regulated by the appropriate levels o f government. 5. Responsibilities are tied tofundingflows, leading topoor coordination and integration of the referral system. Figure 29 provides an illustration o fhow the government health system i s funded. Although the federal and state governments have tax revenues, the lion's share o f government resources comes from oil revenues to the Federation Account, which are shared between levels o f government according to an allocation formula. The transfers from the Federation Account to the states and local governments are not earmarked - that is, each state and local government decides how their funds are allocated between sectors. They are not required to provide budget and expenditure reports to the federal government. This considerably limitsthe effective influence that the federal Ministryo f Health (FMOH)may have over primary and secondary health services (except through special agencies and programs, discussed below), and that the state Ministries o fHealth (SMoHs) may have over primary health care services. This, combined with poor coordinationbetween levels o f government, has ledto limited integration o f the referral system, inparticular impeding the connectionbetween primary and first-referral services. 6. Parastatal agencies and verticalprograms, particularly at thefederal level, also intervene across the different levels of services. The FMOH, the SMoHs, and the Local Government 45 Chapter 3 Health Authorities, retain overall responsibility for health services at their respective levels. However, federal and state parastatal agencies have been createdto implement programs and manage services across the different levels. Inthe late 1980s, Hospital Management Boards (HMBs)were created at the federal and state levels to manage government tertiary and secondary level facilities respectively. Federal parastatals created inthe early 1990sincluded the National Agency for Food and DrugAdministration and Control (NAFDAC), National Primary Health Care Development Agency (NPHCDA), National Programme for Immunization (NPI), and the Population Activities FundAgency (PAFA). By 1999 at the federal level, there were 16 programs, at least 11agencies and departments, and 14 different policies or strategies to address particular issues. More recent organizations are the National Action Committee on AIDS (NACA) and the National Health Insurance Scheme (NHIS). Figure 29. Government funding flows to the health system in Nigeria State Governments secondaryhealth services Local Government Authorities 7. Inmany cases, the impetus behind these federal initiatives was to address problems, particularly at the PHC level, through direct provision o f resources and implementation o f programs. Inaddition, particularly ina context o f low domestic public fundingfor health inthe late 1990s, single issue initiatives were also driven by external donors. 8. The effectiveness o f federal healthparastatals and programs has been mitigated, particularly since the mid-1990s, by fluctuating funding levels, problems with management and political interference, and poor coordination with state and local governments. At the same time, even well-supported programs can contribute to fragmentation and duplication, with different programs operating inthe same LGA under different administrative and reporting modalities, all making different demands on the same health staff. More recent agencies have been created outside the FMOH for more technical reasons -HIV/AIDS requires a multi-sectoral response and health insurance i sjustifiably kept separate from government's regulatory and service provision functions. 9. Coordination mechanisms are being revitalized. Coordination structures put inplace inthe early 1990sincludedthe National Health Council whichjoins the FMOH, SMoHs, parastatal agencies, and other stakeholders, for an annual meeting. Similarly, State Health Councils which bringtogether the SMoH and Local Healthauthorities, were also created. Although many o f these structures became moribund by the end o f the 1990s, the National Health Council and some State Health Councils have recently started to meet more frequently. 46 Chapter 3 GovernanceandAccountability 10. Like other sectors, the health system is recoveringfrom aperiod ofpoor governance and corruption. The long decades o f military rule weakened, politicized, and encouragedrent- seeking ingovernment bureaucracies, and the health system was no exception. The culture and incentives became such that their own economic interest i s the first priority for many public employees. The present democratic government i s faced with an enormous and long-term challenge, and has taken steps to address this systemic problem, including launching a high- profile anti-corruption campaign, raising civil service salaries, and embarking on public service reform. Improvement inthe governance and accountability o f the public sector health system will depend to a large extent on such government-wide reforms. At the same time, a number o f actions specific to the health system can be taken, including more inclusive policy development and management mechanisms. Examples include revitalization o f the National and State Health Councils and involvement o f civil society representatives on Hospital Management Boards. 11. Governance and accountability isparticularly weak at the local level. Although reforms are required at the federal and state levels, and could be characterized as the "top-down" components o f a strategy to improve governance and accountability, health system administration at the local level presents a particular challenge requiring more "bottom-up" initiatives. Duringthe funding cuts and other disruptions o f the 1990s, public sector primary health care services seem to have fared somewhat worse than tertiary and secondary hospital services. Part o fthis i s due to capacity limitations at the local government level, exacerbated by the proliferation o f LGAs, which has spread capacity even more thinly. In 1988, when responsibility for PHC services was assigned to the local level, there were only 158 LGAs, comparedto 774 in 2005. 12. The financial resources available to LGAsmay have been insufficient for them to meet their responsibilities insome years, but this i s currently less o f a constraint given increased oil revenues. (Under the current allocation formula, LGAs are to receive transfers o f 20% o f the FederationAccount). However, there seems to be significant problems with commitment, governance, and accountability. A 1995 assessment o f PHC services inNiger State concluded that commitment to health services by LGAs was minimal and reflected in insufficient releases o f funds to their HealthDepartments (Niger State Ministryo fHealth, 2000). A similar assessment inBenue State foundthat LGAs"have yet to accept responsibility for the sustainability ofhealth services," particularly with regardto meeting non-salary recurrent costs. Ingeneral, information from such assessments suggests that although salaries are usually met, there i s little fundingmade available by the LGA for drugs, supplies, and maintenance (PATHS, 2003j). 13. There i s also evidence that even wages are not being paid in some areas, and that this i s not due to lack o fresources. A 2003 World Bank study o fPHC services in30 LGAs inLagos and Kogi States found that although most staff inLagos have beenregularly paid, 42% o f staff in Kogi reportednot receiving any salary for six months or more duringthe previous year. The study found little correlationbetween budget allocations andnon-payment, suggesting significant problems with "leakage" o f funds. This evidence suggests a problem o f accountability at the local government level, perhaps partly attributable to the fact that most or all o f LGA revenues derive from oil production and not taxation. That is, citizens may be less incited to hold government accountable for spending money that they have not contributed intaxes (Khemani, 2004). 14. Similar problems with wage payments for primary school teachers ledthe federal government to deduct teachers' salaries at source from local governments' transfers from the Federation Account. This led, in some cases, to the phenomenon o f "zero allocation," where the deduction at source totally consumes the transfer, leaving nothing for other local government expenses, including PHC services. Inthe past few years, however, highoil revenues have increased 47 Chapter 3 transfers to the LGAs while deduction at source has beenruled unconstitutional, increasingthe resources transferred to LGAs. CommunityRole 15. Community involvement at the local level has long been recognized as important, but is dij$cult to achieve. Measures can be taken for greater community involvement ingovernance o f the health system. Such initiatives could be seen as "bottom-up" components o f a comprehensive strategy which also involves reforms at the federal and state levels. The 1988 National Health Policy emphasized the community's role instrengthening PHC services, and its implementation included the creation o f PHC Management and Technical Committees at the LGA level, District (Ward) Development Committees, and Village Development Committees. It seems that although these committees have been put inplace inmost cases, many are not adequately functional. A 2001 survey of 202 LGAs intwo states in each o f the six geopolitical zones found that 89% had a PHC ManagementCommittee. Patterns were found to be similar across the country, indicating that implementation o f this aspect o f the 1988 Health Policy was similarly effective inbothNorth and South. For example, inthe NorthEast zone, 77% o f LGAs had a PHC Management Committee, not significantly lower than the national average. 16. However, only 27% o f PHC Management Committees had met inthe previous three months. Similarly, 75% o f LGAshad formed a PHC Technical Committee. This committee, focused on operational issues, i s more functional, as 44% were found to meet regularly. The same patterns were seen at lower levels o f governance. Among 403 Districts surveyed, 84% had a District Development Committee, but only 22% had met inthe previous three months. Of 698 villages surveyed, 78% had formed a Village Development Committee, but only 25% hadmet inthe previous three months (Adeniyi et al., 2001). 17. Some initiatives have been supported by internationaldonors, such as a four-state program supported by DFIDto implement the Bamako Initiative (particularly drugrevolving funds), and a project among six communities infour states which included community involvement in management and a self-financing scheme (Dare, 2000). More recent projects included one in Niger State involving tripartite agreements betweenthe project, the community, and the LGA, (Niger State Ministry o f Health, 2000) and a program inBenue State which worked at improving the linksbetween Village Development Committees and LGA PHC Management Committees (Benue Health FundProject, 2002). Lessons from these initiatives are that ensuring true participation i s not easy and takes time, requiring awareness-raising and capacity development among both community members and the local health administration. Volunteerism may also not be feasible as poor community members are unable to devote unpaid time to such activities. Adaptation o f community governance mechanisms to urban areas, where communities are less tightly-knitthan inrural areas, is also a challenge (FPC and UNICEF, 2001). 18. I n some instances, communities have a signijkant role infacility management but may not affect overall availability of resources. A 2003 World Bank study o f PHC services inLagos and Kogi States found that o f 145 facilities surveyed, 48 indicatedthat the Community Development Committee was the principal decision-maker inone or more o fthe following areas: supply o f drugs/medical supplies, acquiring and maintaining equipment, setting charges for drugs, and determining use o f facility revenues. (World Bank, 2003b) A regression analysis found a significant relationship between community involvement inthese decisions and staff productivity. Onthe other hand, no correlationwas found with equipment and drugavailability. As well, most o f the facilities with community involvement were inKogi State, where non-payment o f staff salaries was muchmore common. These patterns suggest that community involvement has more impact on facility operations than on overall availability o f resources (i.e. drug supplies and salaries). 48 Chapter 3 19. Community-based organizations are active in many areas, and increasing their involvement in local health services may be benejkial but is similarly challenging. A number o f assessments (PATHS, 2003b; PATHS, 2003g) have found that a variety o f community-based organizations (CBOs) are active inNigerian communities, including traditional and kinship institutions, community associations, occupational associations, women's groups, ethnic associations, faith- based groups, women's groups, and more institutionalizednon-governmental organizations (NGOs). A World Bank study inthe mid-1990s of CBOsin36 communities across Nigeria confirmed the rich variety of civil society organizations but also indicated that inmost cases they have little input into government programs (World Bank, 1996b). 20. The large presence o f CBOs inNigeriancommunities provides a basis for enhancement o f community involvement inhealth service administration, while experiences in some locations can provide examples and lessons. However, it should be emphasized that community involvement i s not a "magic bullet" for improving either financial sustainability or governance o f health services. Health services, even at the primary level, are expensive, and it i s clear from experience in Sub- Saharan Africa that inmost cases community financing should be complementary to stronger sources o f financial resources (Ekman, 2004). This means that "bottom-up" initiatives will be most effective inconjunction with "top-down" reform as well as a general context o f greater accountability. As well, efforts to strengthen the role o f civil society organizations inhealth system governance will confront a number o f limitations. First,the numbers and capacities o f civil society organizations are not evenly distributed. Infact, the World Bank study mentioned above found that the poorest communities and regions o f the country also tend to have fewer and weaker CBOs. Second, ingeneral, CBOs have significant capacity constraints, often characterized by a "missing middle," between their membership and a limitednumber o f capable leaders. Third, CBOs are just as vulnerable as other structures inNigerian society to factionalism, patronage, and corruption. Management 21, There are generalproblems withplanning, budgeting, andfinancial management at all levels. It i s widely acknowledged that there are deficiencies at all levels o f the government health systeminterms o fthe basic management functions o fplanning, budgeting, financial control, monitoring, and evaluation. An assessment at the federal level found a low level o f knowledge among FMOHstaff regarding standard procedures and regulations as well as current plans and budgets (DFID, 2000). 22. Recent studies at the SMoH level - inJigawa, Benue, and EnuguStates -have found similar problems (PATHS, 2003c; PATHS, 2003g; PATHS, 2003h). Strategic planning i s not done, and allocation o f resources (infrastructure, equipment, and staff) i s done inan ad hoc fashion driven bypolitical considerations. Although budgets are drawn, they do not correlate with expenditures and there are no financial performance reviews. Financialmanagement i s focused primarily on salaries. These assessments also looked at the local government level, findingeven less management capacity, and an absence o f effective planning, budgeting, and monitoring mechanisms. 23, At the facility level, a study o f secondary hospitals inJigawa State found that although hospital management committees exist, their responsibilities are unclear and their meetings irregular. Strategic planning and target-setting are not done, and budgeting i s not done inrelation to service delivery needs and goals. There i s no overall financial control, since accountingi s done separately by individual hospital departments (PATHS, 2003j). 24. Capacity-building in budgeting andfinancial management is needed, but systemic disincentives are the real problem. Training and other capacity-building activities, such as 49 Chapter 3 developing and redesigning budgeting and financial management systems or provision o f computer equipment, would clearly be beneficial at all levels o f the Government health system. However, the overall systemprovides very little incentive for staff to put their management skills into practice. At every level, from the FMOHto PHC facilities, lack o f control and uncertainty about fundingundermines rational planning and budgeting. Table 17. Public sector health system service levels admin. level of catchment facility level care population staffing specialized/teaching hospital Federal tertiary Federal hospital Federal tertiary general hospital State secondary 500,000 3 doctors 21-34 nurses 9-18 nurse attendants 4-7 x-ray and lab technicians 6-9 administrative staff 24-37 other support staff comprehensive primary LGA primary 50,000-100,000 1 doctor health care center 1clinical officer 2 nurses/midwives 1 environmental health officer 1 lab technician 3 CHEWs 8 Junior CHEWs (JCHEW) primary health care center Ward primary 10,000-20,000 1 clinical officer 1nurselmidwife 1 environmental health officer 1 lab technician 4 CHEWIJCHEW health clinic/health post community primary 500-3,500 1 CHEW and 2 JCHEW traditional birth attendants (TBAs) volunteer village health workers (VHWs) * One ward can have several PHC centers based on settlement areas, population, and existing facilities. 25. For example, withinthe FMOH, personnel budgets are not under the control o f individual department heads. At the same time, plans and budgets may be approved, but managers cannot have confidence that funds will actually be released, especially for non-wage expenditures. A similar lack o f control i s evident at the state level. One study found that inEnugu State, for example, managers h o w that resources are not allocated based on the plans they produce, so they do not work on strategic planning. The SMoH planning department instead concentrates on project planning geared to specific international donor programs (PATHS, 2003g). An assessment of local government Health Departments inJigawa State found that althoughthree-year rolling plans are produced annually, they are never carried out because the release o f funding i s not relatedto plannedbudgets. "There i s a pervading sense o f helplessness as managers have no control over personnel, funding, the erection o fnew facilities or purchases o f new equipment." (PATHS, 2003~). 50 Chapter 3 HEALTH SERVICE DELIVERYSYSTEM 26. Table 17 describes service levels o f the public sector health delivery system. The federal government operates a number of tertiary and specialized hospitals and there i s a federal hospital inmost states, although the level of services available inmany is oftencharacterized as more at the secondary level. Generalheconday hospitals are the responsibility o f state governments, and should have several physicians and at least 20 nurses. Local governments are responsible for primary health care facilities. Comprehensive primary health care centers should have a doctor and offer both PHC and a limited number o f secondary clinical care services. There shouldbe one comprehensive primary health care center per LGA. Within LGAs, there should be at least one primary health care center per ward, staffed by a clinical officer (medical assistant) and a nurse or midwife, and offering basic preventive and curative services. At the community or village level, health clinics or posts should be staffed by community health extension workers (CHEWS) and support birthattendants and other community health workers. Health Service Network 27. Data on health facility numbers are somewhat dated and perhaps incomplete, but the available information provides a good picture o f the patterns o f the network. Numbers o f registered health facilities, bothpublic and private, are available for 2000 from the the FMOH, although one state i s not included. A 2001 survey by the National Primary Health Care Development Agency (NPHCDA) sampled 202 LGAs,designed to be representative of each of the six geopolitical zones. The study collected information on the numbers and types o f health facilities ineach sampled LGA (Adeniyi et al., 2001). The 1999Nigeria Demographic and Health Survey (NDHS) collected data on distance to health facilities, providing information on the actual availability o f health services to household^.^^ Hospital Facilities 28. On average, Nigeria seems to have an adequate number of tertiary and secondary hospitals. At the tertiary level, the federal government operates 15 teaching hospitals, 8 psychiatric hospitals, and 3 orthopedic hospitals, as well as 27 Federal Medical Centers distributed among the states. There was reportedto be one private sector tertiary hospital inthe country in2000. The total o f 54 tertiary and specialized hospitals implies a population to facility ratio o f around 2.1 million. This suggests relatively good average availability o f higher-level services at the same time as representing an average catchment populationmore than large enough to generate the demand for a large hospital. 29. There were reportedto be 855 public sector secondary facilities inthe country in 2000, for a population to facility ratio o f around 135,000, which i s considerably better than the standard o f 500,000 noted inTable 17. Inaddition, there are a large number o f privately-operated facilities, bringingthe reportedtotal to 3,002 secondary facilities inthe country. 30. Overall averages mask considerable regional disparities, as there are signijkantbfewer hospitalfacilities in the north. Table 18 shows that there are considerable regional disparities in the distribution of secondary level services. Population to facility ratios are under 50,000 inthe North Central and southern zones, but over 150,000 inthe North West and North East zones. Table 18. Population per health facility and proportion public sector, Nigeria, 2000 29These data were collected from 399 communities and linked for this analysis to the survey data on 7,647 households. 51 Chapter 3 primary secondary population yo public population per facility per facility `0 public North Central 3,205 62% 49,729 57% North East 6,234 86% 162,355 95% NorthWest 7,170 91% 199,181 76% South East 5,437 35% 12,506 8% South South 6,854 67% 25,213 28% South West 5,421 54% 29,566 26% Nigeria 5,585 67% 38,383 28% Authors' estimates from FMOH data 31, This disparity is largely due to the much greater numbers ofprivate sector secondary facilities operating in the south, and to a lesser extent in the center of the country. Overall, private providers account for 72% o f secondary facilities, but only 5% inthe North East and 24% inthe NorthWest, comparedto over 90% inthe South Eastand over 80% inthe South Southand South West. Inthe North Central Zone, just over halfo f secondary facilities are private sector. The large number o f private secondary facilities inthe south accounts for the l o w population to facility ratios inthese regions (as well as for the low all-Nigeria average). At the same time, the north also has less public sector secondary hospitals. Population to facility ratios for public secondary facilities inthe North East and North West are more than double the ratios inthe South West, South South, andNorth Central zones. Figure 30. Comprehensive PHC centers and general hospitalsper 100,000 population, Nigeria, 2001 5 4 3 private non-profit 2 1 0 North North North South South South Nigeria Central East West East South West Source: Authors' estimates from data in Adeniyi et a/.(2001) 32. Facility survey data indicate that the availability of hospitals services in the South South zone is similar to the northern regions. Ingeneral, the regional patterns shown by the 2000 FMOH administrative data are confirmed by the 2001 NPHCDA survey, which puts inthe same category general hospitals and comprehensive PHC centers (which provide a number o f secondary inpatient services). Figure 30 shows that this level o f facility i s concentrated inthe South East, South West, and to a lesser extent inthe North Central zones. 30The number o f comprehensive PHC centers and general hospitals per capita inthe NorthWest and North East i s much lower. 30For ease of visual interpretation, the graphs present facilities per 100,000 population rather than populationper facility ratios. 52 Chapter 3 Private facilities are the majority inthe South East and South West, but there are also more public sector facilities per capita inthese regions than inthe north. Inthe southern part o f the country, the South Southregion, however, has ratios comparable to the northernregions. This result differs from Table 18 but i s consistent with the 1999 NDHS data presented inTable 19. 33. Private non-profit secondaryfacilities are important in the South East. The private non- profit sector, often church missions, does not operate a significant proportion o f secondary facilities in any region except for the South East, where they representjust over 10% of the sample. 34. However,privatefor-profit hospitals tend to be small and non-profit hospitals tend to be large. Facility numbers may be misleadingabout real service capacity. For example inBenue state, although 76% o f secondary facilities are private for-profit, they account for only 8% o fbed capacity. Conversely, private non-profit hospitals, runby church missions, account for 17% o f the total number o f secondary facilities inthe state but for 61% o f hospital bed capacity. Table 19. Health facility availability, Nigeria, 1999 (% of households) urban rural Overall PHC facility within 5 km 80 66 71 North Central 57 60 63 North East 80 63 65 North West 86 85 67 South East 79 58 84 South South 80 77 62 South West 87 82 83 hospital within 20 km 78 58 64 North Central 66 66 67 North East 60 32 45 North West 63 57 60 South East 85 79 79 South South 58 57 61 South West 77 53 70 Authors' estimates from 1999 NDHS data. Similar data not available form 2003 NDHS 35. Hospitals are most available in urban areas and least available in rural areas of the North. Regional patterns are confirmedby the 1999 NDHS data, which also provide information on urbadrural differences inavailability o f hospital services. Table 19 provides estimates o f the proportion o f households living within 20 kmo f a hospital (public or private). Overall, 64% o f the populationi s within this distance from a hospital, but the proportions are lower inthe North East (45%), North West (60%), and South South (61%), than inthe other parts o f the country. As would be expected, urban areas are better served, as 78% o fhouseholds are within 20 kmo f a hospital, compared to 58% inrural areas. Urban areas inthe North East, North West, and South South zones have the lowest proportions, around 60% within 20 kmo f a hospital. Inrural areas, availability i s by far the lowest inthe North East, with only 32% o f households within 20 kmo f a hospital, while availability i s highest inthe South East. 53 Chapter 3 Table 20. Functioning of referralwithin wards, Nigeria, 2001 (% of wards) (n=400wards) with operating with referral 2-way transport center referral for referral North Central 54 48 9 North East 45 27 13 North West 62 38 16 South East 45 42 3 South South 77 49 9 South West 61 57 10 Nigeria 58 44 11 Source is Adeniyi et a/.(2001). 36. The ward (district) referral system isfunctioning in around two-thirds of districts. An important component o fthe "classic" PHC strategy, adopted by a number o f countries inthe 1980s and early 1990s, i s the integrationo f primary and first-referral services under a single administrative structure incatchment areas (often called districts) o f around 100,000 or 200,000 people. InNigeria, the division o f responsibilities between local governments, responsible for primary health care, and state governments, responsible for secondary-level services, i s a structural barrier to full implementation o f the district strategy. Nevertheless, the ward system was implemented inNigeria, involving collaborative arrangements between states and local governments as well as between individual secondary (first-referral) hospitals and PHC facilities. The 2001 survey o f PHC services in202 LGAs collected information on referral in400 wards which provides indications o f how well the "district" model i s functioning. It found that 58% o f wards had a referral facility (Table 20). Two-way referral was functioning in44% o f wards, while only 11%had transport available for referral.31 Primary Health Care Facilities 37. Overall numbers of PHCfacilities suggest reasonable availability, with less regional disparities than is the case with hospitals. According to FMOHdata, in2000, there were over 13,000 public sector PHC facilities and almost 7,000 private PHC facilities, for a total o f about 20,000. The overall population to facility ratio o f around 5,500 suggests reasonable availability o f PHC services. Population-to -facility ratios are highest inthe North East, North West, and South South zones (Table IS), a pattern largely consistent with the data on secondary services. However the regional differences are not as great, and this i s due to the large number o f public sector PHC facilities inthe northern regions. Infact, when considering only public sector PHC facilities, population to facility ratios are better inthe norththan inthe south. 38. Data on availability o f PHC facilities from the 1999 NDHS provide further evidence. Table 19 shows the estimated proportion o f households living within 5 kmo f any type o f PHC facility. It indicates that, overall, 71% o fhouseholds are within 5 kmo f a PHC facility. Similar to the pattern o f facility numbers, the proportions are highest inthe South East and South West (over 80%), compared to the South South and the northernregions (around 65%). Inurban areas, 80% o f households are within 5 kmo f a PHC facility, compared to 66% inrural areas. Inurban areas, the North Central zone unexpectedly has the lowest proportion (57%), while availability in other regions i s similarly high. Inrural areas, the South East and South West zones have the highest proportions o f households within 5 kmo f a PHC facility, over 80%. Availability inrural areas o f the South South and northem zones i s comparable, around 65%. 3' This contrasts with the higher estimates for availability of transport discussed in the sections on equipment below - presumably only insome cases i s available transport used for referral. 54 N I Chapter 3 Figure 33. Primary health care (PHC)centers per 100,000 population, Nigeria,2001 7 6 5 private non-profit 4 3 2 1 0 North North North South South South Nigeria Central East West East South West Authors' calculations from data in Adeniyi et a/.(2001). 41. PHC centers, whichprovide higher-level services, are concentrated in the center and south. Figure 33, based on the NPHCDA facility survey, shows that incontrast with lower-level facilities, there are far fewer PHC centers per capita inthe northern regions than inthe south o f the country. Overall, the 2001 survey indicates that there are around 24,000 people per PHC center, which slightly exceeds the standard range o f 10,000 to 20,000 given inTable 17. The ratio i s around 15,000 inthe South East, South West, and NorthCenter, compared to almost 150,000 inthe North West and NorthEast. The ratio o f around 25,000 per PHC center inthe South South i s worse than inthe rest o f the south. Private for-profit providers operate a significant proportion o f PHC centers in the SouthWest and to a lesser extent inthe South East. There are some private non-profit PHC centers inthe South East. Table 21. Registered pharmacies in Nigeria, 2003 number population per pharmacy North Central 544 30,147 North East 114 280,702 North West 213 140,845 South East 348 99,425 South South 343 51,312 South West 1,189 48,360 total 2,751 42,421 Source is Pharmacists Council of Nigeria (2003). Pharmacies and Patent Medicine Vendors 42. Patent medicine vendors, a legally-recognized category of drug seller,far outnumber registered pharmacies. Registered pharmacies are licensed to fillprescriptions while patent medicine vendors are also a category recognizedand licensed by the government to sell over the counter medicines. Patent medicine vendors are far more numerous than registered pharmacies and are significant providers o f "informal" care. In2003, there were 2,75 1registered pharmacies, which represent significant growth from the 1,821 registered in 1992. Incontrast, there may be 36,000 or more patent medicine vendors inthe country. Inaddition, there are an unknown number o finformal drugsellers. More details onthe pharmaceutical retail sector are provided in Chapter 4. 56 Authors' estimatesfmrn 1999 NOHS data Siriiiiar data nut avaiiatilc from 2003 NDHS 57 Chapter 3 different types o f health facilities. These are, o f course, consistent with urbadrural and regional disparities, inthat rural populations are generally poorer than urbanresidents, and the North East, NorthWest, and South South zones are generally poorer than the rest o f the country. Figure 35 indicates that within urban areas, the availability o f health services, bothpublic (most PHC are public sector) and private (patent medicine vendors), i s significantly lower for the very poorest, but similar among the rest o fthe socio-economic scale. This is consistent with a situation where the poorest live inurban settlements/slums distant fi-om city centers. Inrural areas, differences in availability are evident along the socio-economic scale, increasing with each higher quintile. Figure 35. Availability of health service providers by socio-economicstatus, Nigeria, 1999 (% households within 5 km) loo 90 I urban rural 90 80 - 80 - E E Y In 70 - 7 0 - 60 - ZC 6 0 - '5 50 - g 5 0 - 40 - .c 40 - 3 30 7 e8 3 0 - 20 -PHCfacility 10 -hospital 0 1 2 3 4 5 1 2 3 4 5 quintile quintile Authors' estimates from 1999 NDHS data. Similar data not available from 2003 DHS Figure36. Availability of PHC facilities by socio-economicstatus and region, Nigeria, 1999 (% households within 5 km) rural South East 100 North West SouthWest North East 901 80 South South E Y 70 - North Central v) 'E C 60 - 5 50 - .c0 a 40 - s 30- 10 I *O 1 2 3 4 5 auintile Authors' estimatesfrom 1999 NDHS data. 47. The relationship betweenpoverty and low availability of health sewices holds in rural areas within each region. Figure 36 describes shows how availability o f PHC facilities changes by socio-economic status inrural areas within each region. It shows that the poor are less likely to be livingclose to a PHC facility inall regions. This shows that the relationship described inthe 58 Chapter 3 previous graph (Figure 35) i s not only due to differences inhealth services betweenregions but also due to socio-economic differences between communities and households. Inmost regions, availability o f PHC services increases steadily as socio-economic status rises. Inurban areas withineach region (not shown) the pattern is not clear). Summary o f Regional Patterns 48. The regional patterns shown by the different data sources are largely consistent and can be summarized as follows: North Central: The number o f secondary hospitals per capita inthis zone i s lower than in the south, but considerably higher than inthe North East and NorthWest, with about half o f the facilities inthe private sector. Urbadrural disparities inthe availability o f hospital services are not evident, as around two-thirds o f households live within 20 kmo f a hospital. The number o f PHC facilities per capita (mostly public sector) i s among the highest inthe country, and this i s true with regard to both lower-level facilities and PHC centers. Again, household survey data indicate that urbadrural differences are not large, although overall, the proportion o f households livingwithin 5 kmo f a PHC facility i s not as highas the reportednumbers o f facilities would suggest, perhaps indicating geographic clustering o f facilities incertain areas or states. Patent medicine vendors are readily available to more than halfthe population, and seem to be more available inrural areas than inurban. Like inall other zones, health services are less available to the poor inrural areas, although availability seemsto be somewhat similar amongthe highest three quintiles (Figure 36). North East: The number o f secondary hospitals per capita inthe North East i s much lower than inthe center and south o f the country, and there are large urbanhral disparities. Close to half o f higher-level services (comprehensive PHC and secondary hospitals) are providedby the private sector, while PHC facilities are mostly public sector. The number o f PHC facilities per capita inthis zone i s lower but comparable to the national average, but this i s mostly due to large numbers o f lower-level facilities, like health posts and dispensaries. Data on availability o f community health workers suggest that these basic services are most functional inurban areas. The per capita number of PHC centers, which provide higher-level services, i s clearly much lower than the national average. Urbadrural differences inavailability o f PHC services are not as pronouncedas with hospitals. About halfthe population lives close to a patent medicine dealer, whose availability i s significantly higher inurban areas. The rural poor are less likely to have health services closely available. NorthWest: Along with the North East, availability o f secondary hospitals inthe North West i s much lower than the rest o f the country. A large difference between urban and rural areas i s not evident, however. Also like the North East, the number o f PHC facilities per capita i s comparable to the rest o f the country, but this i s due to large numbers o f lower-level services. With regard to PHC centers, availability i s on par with the North East- much lower than rest o f the country. Urban-rural differences inthe availability o f PHC centers are not evident. The private sector i s not absent, but (along with the South South) accounts for among the lowest proportions o f services inthe country. Patent medicine vendors are within 5 kmo f about 60% of the population, and more available inurban areas than inrural. Like elsewhere, the ruralpoor are less likely to have access to all types o f health services. South East: The South East has among the highest number o f hospitals per capita inthe country, with more than half operated by the private sector. The region has the highest proportion o fprivate non-profit hospitals, and these likely account for a majority o f 59 Chapter 3 hospital capacity. Differences inavailability o f hospital services between urban and rural areas are not large. There i s a lower concentration o f lower-level PHC facilities than in the north, but much higherper capita number o f PHC centers. Most PHC centers are public sector, although about 10-15% are private. The difference between urban and rural areas inPHC facility availability i s larger than inmost other regions. Patent medicine vendors are closely available to about three-quarters o f the population, and more concentrated inrural areas. Health services are less available to the rural poor. South South: The different sources o f data show that the health facility network inthe South South zone i s generally comparable to the northernregions -that is, poorer than the rest o f the south and the North Central zone. The difference with the north i s that there are not as many lower-level PHC facilities and more PHC centers inthe South South. There are some private sector hospitals, but generally health facilities are managed by the public sector. About half the population lives close to a patent medicine vendor (comparable to the North East), but these are concentrated inurban areas. Health services are less available to the rural poor. e South West: Along with the South East, the SouthWest has the best health facility network inthe country. The difference with the South East, likely due to Lagos, i s that a majority o f hospitals and a large proportion o f PHC facilities are runby the private for- profit sector. Although present, the private non-profit sector i s less important than inthe South East. There i s a significant difference between urban and rural areas inthe availability o f hospital services, although this i s not the case with regard to PHC facilities. Patent medicine vendors are widely available, although somewhat less common inrural areas than in urban. Like inother regions, inrural areas health service availability i s worse for the poor. 49. Overall, considerable complexity is evident in the regional, urbanhral, and socio-economic distribution of health sewices in Nigeria. However, it i s clear that higher-level services are significantly less available inthe NorthEast and NorthWest, especially inrural areas. It seems that the network i s dominated inthe northby lower-level PHC services, and that these are more functional inurban than inrural areas. Availability o f health services inthe South South zone i s somewhat comparable to the northern regions. Urban-rural disparities in service availability are evident across the country, but not always. Patent medicine vendors are widely available, although somewhat less so inrural areas o f the NorthEast and South South. Inurban areas across the country, health services are less available to the poorest, but similarly accessible to the rest o f the population. Inrural areas inall regions, availability o f services steadily decreases as socio-economic status decreases. HumanResources 50. Thefederal government is responsiblefor training of doctors, states are largely responsible for nurse and midwife training, andprofessional associations are responsiblefor the registration of healthprofessionals. The federal government has the major responsibility for regulation and training o f physicians, and operation o f medical schools and teaching hospitals. State governments are responsible for schools o f nursingand health technologies. Professional associations register healthprofessionals. Government health workers are employed and paid by the level of government responsible for the service where they are working, that is, Federal for tertiary hospitals, States for secondary hospitals, and Local Government Authorities for PHC services. There are exceptions to this: in some States, professionals working inPHC facilities may be employed by the State Government, while skilled health staff employed by Federal parastatals and vertical programs may be working at the secondary and primary levels. 60 Chapter 3 Figure 37. Doctors and nurses/midwives per 100,000 population by GNI per capita, 2001 (countries with GNI per capita under US$1,500) . 1000.0 . doctors per 100,000 1000.0 nurseslmidwives per 100,000 . . 0 . 0 . 0 . 0 g- * . 8 100.0 '*. z 100.0 8 2 .. *; . L . 'rn Ba, & $ 8 F a Q C D '5 CD 92-0 g . e 0-0 10.0 $ 10.0 '0 W 2 C 1.o 1.o 0 500 1000 1500 0 500 1000 1500 GNI per capita 2001 ($US) GNI per capita 2001 ($US) Note: Countries of the former Soviet Union and former Yugoslavia are excluded. Data sources are World Development Indicators, Joint Learning Initiative (2004) and NPC and UNICEF (2001). 5 1. Large numbers of doctors and nurses are being trained, although growth slowed during the 1990s. Growthinthe number o fphysicians registered inNigeria was dramatic inthe 1970s and 1980s, withnumbers increasing from 5,300 in 1975 to 21,000 in 1985. The increase to 25,000 in 1995 was somewhat slower, although growth inthe number o f trained doctors i s continuing, as there were a reported 31,000 in 2001. There are 15 medical schools inthe country. A comparable pattern inthe training o f nurses and midwives i s evident, as their number per capita increased seven-fold between 1970 and 1986. The number o f newly trained nurses and midwives increasedmore slowly - slightly higher than population growth - over the next decade so that by 1997, there were reportedto be 95,000 registered nurses and 70,000 registered midwives.(NPC and UNICEF, 2001) 52. The overall number of doctors is in line with other countries of similar incomeper capita while the number of nurses exceeds what would be expected. Figure 37 plots doctors and nursedmidwivesper 100,000 populationby gross national income (GNI) per capita in counties with per capita income under $US 1,500. InNigeria in2001 (GNI$US 300), there were estimated to be around 24 doctors per 100,000 population, which puts it on the trend line, situating it between the ratio o f 12 inKenya (GNI $US 350) and 49 inIndia (GNI $460). The number o f nursedmidwives per capita inNigeria i s somewhat above the trend line, as there are an estimated 126 nursedmidwives per 100,000, compared to 80 inKenya and 58 inIndia. Separating the two categories, there are estimated to be 73 nurses per 100,000 and 53 midwives per 100,000 inNigeria. 53. The balance between nurses and doctors is also in line with most countries of similar income per capita. Figure 38 plots the ratio o fnurses to doctors against GNIper capita. It indicates that there may be a trend towards fewer nurses per doctors (i.e. more doctors) as income rises, but there are a number o f important exceptions. Inany case, Nigeria, with a ratio o f 5.3 nurses per doctor, i s on the trend line, similar to Kenya (6.8) and higher than India (1.2). 61 Chapter 3 Figure 38. Nurseshnidwives per doctor by GNI per capita, 2001 (countries with GNI per capita under US$1,500) nursesimidwives per doctor a .-m . E83 10.0 I k? C 5.0 0.0 0 500 1000 1500 GNI per capita 2001 ($US) Note: Countries of the former Soviet Union and former Yugoslavia are excluded. Data sources are World Development Indicators, Joint Learning Initiative (2004) and NPC and UNICEF (2001). 54. Most doctors are male and most nurses and midwives arefemale, and the majority of community health workers arefemale. A 2003 survey o f health personnel infive states (Ondo and Lagos inthe South West, Cross River inthe South South, Kano inthe North West, and Plateau inthe North Center) found that over 80% o f doctors are male and over 70% ofnurses and 90% o f midwives are female. Pharmacists are more evenly divided by gender. Two-thirds o f community health workers are female, which may improve cultural accessibility o f services. Most records clerks are female (Figure 39). Figure 39. Gender distribution of health personnel in 5 states, Nigeria, 2003 (%) (n=930) 8 0 20 40 60 80 100 doctors - 1 1 registered nurses registered midwives pharmacists commLn:ry nea tn enension worrters (CnEWs) medica recoros clerrts Source is Dare et a/.(2003). 62 Chapter 3 55. A large number of Nigerian doctors and a growing number of nurses are not working in Nigeria. Table 22 presents information about registered doctors in2003. It indicates that around 6,400 registered doctors, or 18% o f the total, are abroad. For example, in2003, there were 2,855 Nigerianphysicians registered with the American Medical Association (Dare et al., 2003). Table 22. Professionalsituation of registereddoctors, Nigeria,2003 number % Resident 3,228 9 public hospitals 3,562 10 private hospitals 19,571 55 Abroad 6,422 18 outside the profession 2,800 8 Total 35,584 100 Source is National Association of Resident Doctors, cited by Dare et a/.(2003). 56. There is also some evidence that a growing number o f Nigeriannurses are working abroad. Data from the nurses registry inthe UnitedKingdom indicates that there were 509 Nigerian nurses working inthat country in2003, compared to 179 in 1999 (Dare et al., 2003). 57. Forfinancial and career development reasons, the desire to migrate abroad is very common among Nigerian doctors. The 2003 survey o f healthprofessionals mentioned above found that in three states (Lagos, Cross River and Plateau) between 50 and 60% o f doctors expressed the desire to work abroad, while the proportion was 34% in Ondo and 24% inKano. Depending on the state, remuneration was a stated reasonby between 17 and 73%, while career development was cited by between 20 and 50% o f respondents (Dare et al., 2003). Similarly, a 1999 survey o f 105 interns intwo hospitals inLagos found that 69% hoped to practice abroad, mostly for financial and professional development reasons. (Odusanya and Nwawolo, 2001) Table 23. Public sector PHC personnel, Nigeria,2001 (per 100,000 population) Senior or Junior Community Medical Officer Health Community Nurse or Midwife Health Extension Worker North Central 0.3 4.9 12.3 33.7 North East 0.7 2.0 13.3 30.8 North West 0.2 1.8 3.7 23.9 South East 0.3 1.7 4.1 14.9 South South 0.6 3.1 7.3 18.0 South West 0.6 3.0 9.0 20.1 Nigeria 0.4 2.7 7.9 23.0 Authors' calculations from data in Adeniyi et a/.(2001). 58. There are more doctorsper capita in the south, but there are no large regional differences in the numbers of nurses. Incomplete data from 1994 indicate that, consistent with the regional disparities inhigher-levelhealth services discussed above, there are more doctors per capita inthe south of the country than inthe north. Although registeredmidwives are concentrated inthe south, particularly the South East, nurses are more evenly distributed across the country (FPC and UNICEF2001). 63 Chapter 3 59. Most doctors and nurses work in higher-level andprivate services. Table 22 indicates that most doctors work inhospitals (88% o f doctors practicinginNigeria) and most work inprivate hospitals (74% o f doctors practicing inNigeria). Only 12% o fpracticing doctors inNigeria work inprivate practice or inpublic sector PHC services. Table 23 presents data onpublic sector PHC personnel from a 2001 survey o f 202 LGAs intwo states ineach o f the six geopolitical zones. Overall, there were 0.4 medical officers (doctors) workmg inPHC services per 100,000 population; this can compare to the ratio o f all practicingdoctors to population o f 24 per 100,000. Similarly, the survey found that overall 7.9 nursedmidwives are working inPHC services per 100,000, which can compare to the ratio o f 126 per 100,000 for all nurses, indicating that most nurses work inhigher-level services. 60. There are generally morepublic sector PHCpersonnelper capita in the north than in the south. Regional differences inmedical officers (doctors) and community health officers (medical assistants) are not very evident, although there seem to be fewer o f these higher-skilled categories working inPHC services inthe North West and South East (Table 23). PHC nursedmidwives and community health workers, however, are generally more numerous inthe north than inthe south, consistent with the patterns o f facilities discussed above. Table 24. Traditional birth attendants and village health workers, Nigeria, 2001 (% of villages) (n=674) with trained VHW kits supervision traditional birth replenished TBA kits with trained TBAsn/HWs of attendants village health replenished (TBAs) regularly workers (VHWs) regularly taking place North Central 56 9 30 9 33 North East 60 10 31 14 35 NorthWest 63 13 65 13 33 South East 29 15 32 14 34 South South 56 4 50 4 27 South West 48 15 42 15 34 Nigeria 52 a 42 12 33 Source is Adeniyi et a/.(2001). 61. Traditional birth attendants and village health workers have been trained in a signijkant proportion of communities, but most do not receive regular supportfrom the health system. An important part o f Nigeria's PHC strategy inthe 1980s and early 1990s was to train and support community-based health workers. The 2001 survey in202 LGAs found that half o f surveyed villages had a trained traditional birthattendant (TBA) and around 40% had a trained village health worker (VHW) (Table 24). This shows progress inimplementing the strategy, but it appears that ongoing support from the health system-which has been shown inother countries as essential to the success o fthis strategy inimproving health outcomes - i s lacking. Inonly around 10% o f surveyed villages were TBA and VHW kitsreplenished regularly, although supervision was taking place ina third o f villages. Regional differences are not clear, although villages inthe North West zone are more likely to have both trained TBAs and trained VHWs. Ongoing support was similar inthe North West as inother regions, however. 62. Remuneration and working conditions draw skilled health personnel to urban areas and the private sector. A 2003 survey o f health staff infive states found that between 10 and 25% expressed the wish to move to another part o f the country. (Dare et al., 2003) A 1999 survey o f interns intwo hospitals inLagos found that, o f the minority (40%) who intended to work in Nigeria, two-thirds wanted to settle inLagos or Abuja, although most indicated a willingness to work inrural areas if"social amenities" were provided. (Odusanya and Nwawolo, 2001) At the 64 Chapter 3 same time, although it i s clear from the available data (Table 22) that skilled professionals such as doctors prefer to work inthe private sector, information on their perceptions and incentives inthis regard i s not available. 63. Human resource management in thepublic sector requires improvement. Federal and state ministries o f health and local health departments have little control over salary levels, yet the wage bill accounts for most o f their budgets. Rigidcivil service rules, althoughperhaps necessary inorder to reduce arbitrary and corrupt management, inhibit effective personnel management and lead to undesirable results. For example, a 2002 survey o f 700 health workers in252 facilities inLagos and Kogi States foundthat salaries basicallydepend on seniority, and have no relation to skills and qualifications, so that physicians, who tended to be younger, were usually paid less than other health staff. Highturnover rates were evident, particularly for physicians, who averaged only 3 months intheir current facility, compared to 2.7 years for all staff. (World Bank, 2003b) Figure 40. Health worker perceptionsof reasonsfor poor performancein 5 states, Nigeria, 2003 (% of health workers) 8o 713 . -1 0Lagos 2 60 e, 50 -m 3 5 40 c Y- 0 30 -s 20 10 0 Source is Dare et a/. (2003) 64. Lack of support undermines performance andpoor remuneration leads public sector health workers to supplement their incomes. A 2003 survey infive states asked health staff their views on the main reasons for poor performance. Figure 40 shows that the highestproportion o f health workers cited lack o f supplies and equipment as underlying poor performance. Poor remunerationi s stated as important in some states, but lack o f training and supervision and other factors are more important inothers. It i s common inmany developing countries for public sector health workers to provide services privately inorder to supplement their income, and Nigeria i s no exception. The 2003 survey in five states found that between 30 and 45% o f health workers reportedhaving a second source o f income. The 2002 study inLagos and Kogi states found that 42% o f staff reported that salaries had not been paid for 6 months or more inthe previous year, although non-payment o f salaries was not a significant problem inLagos. Analysis showed that the greater the extent of non-payment o f salaries the higher the likelihood that staff sell drugs privately and provide services outside the facilities. (Khemani, 2004) 65 Chapter 3 Drugsupply 65. Public sector drugprocurement is decentralized andfragmented. A 2002 FMOHassessment inthe six geopoliticalzones found that although state drugstores are generally well-organized for storage, they do not fill any central procurement function, as procurement i s done directly by individual hospitals and local government authorities. (FMOHand WHO, 2002) The lack o f central procurement prevents economies o f scale and top-down control over the system, although it may provide a certain safeguard of diversification against poor central management. Table 25. Availability of essential drugs, Nigeria,2001 (YOof facilities with different proportionsof essential drug package available) (n=674) 100% 7599% 50-74% 0-50% North Central 0.8 7.6 20.5 50.0 North East 1.5 13.3 25.9 40.7 North West 1.6 20.5 28.7 44.3 South East 2.5 12.7 24.1 45.6 South South 0.0 8.9 25.6 42.2 South West 0.0 2.6 13.9 51.7 Nigeria 1.2 11.0 17.1 45.8 ~ Source is Adeniyi et a/. (2001). 66. Drug supply is inadequate in public sectorfacilities, particularly PHC services. When resources are available to local governments, salaries are the first priority so that there i s often little left for other recurrent costs, particularly drugs, leading to "out o f stock syndrome." The 2002 FMOHassessment found that o f 28 PHC facilities surveyed, fully 18 reported that they had not received drugs from the Government inthe past two years and so staff were purchasing and selling stocks privately. (FMOH and WHO, 2002) The 2001 survey o f 674 facilities in202 LGAs describedpreviously found that 46% had less than halfo fthe essential drugs list available (Table 25). This varied little between regions, although the NorthEast and South Southregions had the lowest proportions, around 40%. The survey found that 54% o f facilities had experienced drug stock-outs inthe previous three months, andthat this differed little byregion, except for the North Central zone, where stock-outs were less frequent (Table 26). Similarly, a 2002 study o f 252 government facilities in Lagos and Kogi States found that less than halfo f the facilities had government-supplied stocks o f selected basic drugs. (Figure 41) Figure41, Government-supplieddrug stocks, Lagos and Kogi States, Nigeria, 2002 (% of facilities) (n = 252) 80 70 Source is World Bank (2003b). 66 Chapter 3 67. Hospitals have more reliable drug supply, but not suficient to meet their needs in many cases. Secondary and tertiary facilities tend to have more regular drug supply, likely related to the better overall fundingavailable at the federal and state levels. The 8 secondary-level facilities surveyed by the 2002 FMOHassessment were found to have regular drug supplies. However, a series o fhousehold surveys in 1999 and 2002 in 8 states found that among patients treated at public sector hospitals duringthe previous four weeks, lack o f drugs was a major cause o f dissatisfaction, ranging from around 5% inCross River state to 35% inLagos (Figure 42). Figure 42. Causes of patient dissatisfactionwith public hospitals, Nigeria, 1999(% of treated cases of illness in the previous 4 weeks) 40.0 - 35.0 - 30.0 - 25.0 - 20.0 - 100 5 0 0 0 Sources are CWlQS (1999 and 2002). 68. Drug revolvingfunds have been established widely, but have not been effective in ensuring reliable supply. Drug revolving funds were an important component o f the PHC strategy adopted bythe government inthe 1980s and 1990s andwere established widely, sometimes with external donor support. In 1997 the FederalGovernment attemptedto generalize drugrevolving funds through support fi-om the Petroleum Trust Fund. Poor management, political interference and corruption led to the discontinuationo f this program in 1999. The 2001 survey of facilities in202 LGAsfound that around 40% o f facilities had a drugrevolving fund inplace, andthat inmost regions, this proportion was over 60%, reaching 7585% inthe north. The lowest proportion (27%) was inthe South West, where presumably private sector drug supply i s ample, particularly inLagos. However, the drugrevolving funds donot seemto have ensuredregular supply, as over half of facilities experienced a stock-out inthe previous three months. The exception may be the North Central zone, where 86% o f facilities have a fund inplace and only 28% experienced a recent stock-out (Table 26). 69. The problems experienced by drug revolving funds include requirements by state and local governments that drug sale proceeds be centralized, which loosens accountability and control, so that funds are often used for purposes other than to replenish drug supply. Even when retainedby the facility, funds are often used to meet other costs, inparticular staffremuneration. More successful experiences are those where local governance and accountability have been strengthened overall, particularly with community involvement. 70. Strongprivate sector supply improves drug availability. There has been a strong private sector response to the shortcomings o f the public system's drug supply. This has been facilitated by the regulatory environment, which officially registers bothqualifiedpharmacists and "patent medicine dealers." The latter are not permittedto fillprescriptions, but inpractice this i s hardly 67 Chapter 3 enforced. Informal and itinerant "drug hawkers" also sell drugs, often obtaining their supplies from patent medicine dealers. Much supply i s imported, but there are also around 200 manufacturers inthe country. Table 26. Drug stock-outs, availability of essentialdrugs list, and drug revolving funds in health facilities, Nigeria, 2001 (n=674) :&+:: drug essential revolving months drugs list fund in place North Central 27.8 67.4 85.6 North East 50.4 32.6 74.8 North West 60.7 66.4 88.5 South East 57.0 32.9 84.8 South South 54.4 44.4 61.1 South West 57.8 32.8 26.7 Nigeria 54.2 47.2 42.7 Source is Adeniyi et a/.(2001). 71. As mentioned above, inmany instances government health workers have defacto privatized drugsupply inpublic facilities. The 2002 study o f 252 public facilities inLagos and Kogi States found that 38% had such privately-owned stocks o f chloroquine, 39% had paracetamol, 34% had antibiotics, 17% had oral rehydration salts (ORs), and 32% had multivitamins. This was more common inKogi than inLagos, where private drug retailers are more accessible. (World Bank, 2003b) 72. The 2002 FMOHassessment included a household survey which found that 56% o f respondents who were illinthe previous two weeks purchased drugs from a private seller, compared to 35% who obtained drugs from a public health facility. Of those who bought from private sources, 45% went to a drug seller, 30% to a private clinic, 14%to a pharmacist, 9% to a traditional healer, and 2% to a neighbor or family member. Similarly, a study o f 158 diarrhea cases in Oyo State found that 55% bought treatments from private sources, mainly drug sellers, compared to 33% who obtained drugs at health facilities. (Omotade, et al., 2000) 73. The cost to households o f drugs i s discussed inChapter 5. 74. There areproblems with rational use ofpharmaceuticals. A large number o f patients who receive medical advice from potentially unqualified drug sellers, raising concerns about inappropriate use o f drugs. For example, it has been observed that street drug hawkers often choose drugs for their customers on the basis o f the color and pictures on the packaging. (PATHS, 2003a) Problems have also beenobserved with the rational use o f drugs prescribed in public sector facilities. A hospital assessment inJigawa State found significant polypharmacy, withprescriptions inall facilities exceeding, on average, the 2-3 drugs recommendedbyWHO. The study also found that antibiotics and injections are over-used, as each are included in over 40% o f prescriptions, compared to recommended levels o f around 20% for antibiotics and around 10% for injections. (PATHS, 2003j) Consistent with this, the 2002 FMOH study o f 36 government facilities across the country found that the average number o f drugs per prescription was 4.7 and that 59% included antibiotics and 38% included injections. The study also assessed the appropriateness ofprescribed treatments, judging that only 6% o fprescriptionsfor diarrhea, 11% for acute respiratory infection (ARI),21% for pneumonia met accepted standards. The FMOHhas developed treatment guidelines, called "Standing Orders," but only 38% o f facilities had copies (FMOHand WHO, 2002). 68 Chapter 3 75. There are concerns about drug quality, but little empirical evidence is available. It has been estimated that up to 40% o f drugs incirculation inNigeria are fake. However, this term tends to be applied to drugs which have not been registeredby the National Agency for Food and Drug Administration and Control (NAFDAC), and for which there i s no evidence about quality one way or another. (PATHS, 2003a) A laboratory study o f the quality o f basic drugs from 35 registered pharmacies inthe Lagos and Abuja areas found that 48% o f the 581 samples contained amounts o f active ingredients outside o f standard limits. However, most were just outside the limits,andthe samples were as likely to containtoo higho fconcentrations as too low. The proportion which contained no active ingredient was 1%. (Taylor et al., 2001) However, this study didnot include patent medicine dealers or drughawkers, which supply a large proportion o f the drugs sold inNigeria. There i s little empirical evidence regardingthe quality o f the drugs they sell. 76. With regard to expired drugs, the FMOHassessment o f 36 facilities found that 7% o f essential drugs in stock were expired. 77. A regulatoryframework is inplace. NAFDAC i s responsible for regulation of pharmaceuticals and has registered a total o f 4,363 medicines, including 83 traditional medicines which have been proved safe. A National DrugPolicy was adopted in 1990. An Essential Drugs List was first publishedin 1986with 204 drugs and now contains 484. This inclusive list i s one reason why the 2002 F M O H assessment found that 90% o f reviewedprescriptions were for drugs on the list. Only 24% o f facilities, however, had a copy o f the list itself (FMOHand WHO, 2002). A much larger facility survey in2001 (Table 26) found that the list was available in47% o f facilities, ranging from 33% to 67% indifferent regions. Equipment 78. Most healthfacilities in all regions arepoorly equipped. The 2001 NPHCDAhealth facility survey in 202 LGAs measuredthe availability o f a minimumpackage o f equipment (defined by the researchers) ina sample o f public and private sector 674 health facilities. It found that, overall only around 25% o f health facilities had more than half o f the minimumpackage o f equipment, while 40% had less than a quarter o f the set o f equipment (Table 27). The proportion with more than halfo fthe equipment package differed little byregion. The proportion o f facilities with less than a quarter o f the equipment was higher inthe North East, NorthWest, and (surprisingly) South East, reaching45-50%. Table 27. Availability of minimum package of equipment by region, Nigeria, 2001 (% of facilities) (n=674facilities) more than less than one half of quarter of equipment equipment package package North Central 27 35 North East 24 44 NorthWest 23 49 South East 23 48 South South 27 33 South West 21 33 Nigeria 24 40 Source is Adeniyi et a/. (2001). 69 Chapter 3 79. As with drugs, healthpersonnelprivately supply some basic medical equipment and supplies. Figure 43 presents data from a 2002 survey o f facilities inLagos and Kogi, showing that a third o f the available blood pressure gauges and up to half o f the available antiseptics and sterile gloves were supplied by healthpersonnel themselves. Figure43. Supplies and equipment in healthfacilities, Lagosand Kogi States, Nigeria,2002 (% of facilities) (n = 252) 100 90 I privately-supplied 1 80 0Government-supplied 70 .--U =8 60 E 50 S 40 30 20 10 0 Source is World Bank (2003b). 80. A signiJcant number of higher-level facilities arepoorly equipped. Table 28 presents equipment availability by level o f facility, showing that as would be expected, the proportion o f facilities with less than half o f the equipment package decreases as the level o f facility increases. However, it i s still strikingthat only 30-40% o f PHC centers, comprehensive PHC centers and secondary hospitals have more than half o f the defined set o f equipment. Similarly, 30% o f PHC centers and 22% o f comprehensive PHC centers and secondary hospitals have less than one quarter o f the equipment package. Table 28. Availability of minimum package of equipment by level of facility, Nigeria, 2001 (% of facilities) (n=674 facilities) more than less than one half of quarter of equipment equipment package package Health postslclinicsldispensaries 19 47 PHC centers 31 30 ComprehensivePHC centers and general hospitals 38 22 Source is Adeniyi et a/. (2001). 81. Thesurveyfound, however, that morefacilities than might be expected had afunctioning vehicle, Overall, 58% o f facilities had a functioning vehicle. Nor are regional patterns exactly what would be expected, as the highest proportions (70-75%) were inthe South West and South South and the lowest (45-50%) inthe South East and North East. A 2002 study inLagos and Kogi states similarly found that around a third of facilities inboth states had access to a vehicle in an emergency (World Bank, 2003b). 82. About 66%ofpublic sector PHCfacilities and over 80% ofpublic secondary facilities have basic obstetric care instruments, but more complicated equipment is available in a minority of 70 Chapter 3 public sector hospitals. A large survey o f obstetric care in4,503 health facilities in 12 states in 2003 collected data on equipment (Table 29). As would be expected, few lower-level PHC facilities had more than basic instruments, although even these (such as fetal stethoscope and sphygmomanometer) were available in only around two-thirds. This i s consistent with the a survey inthe South West in 1998, which found for example that 69% o fPHC facilities had a fetal stethoscope, 59% had an oral thermometer, and 79% weigh scales for children (NF'C and UNICEF,2001). As would be expected, public sector secondary facilities have morebasic obstetric equipment than PHC facilities, with 80-90% possessing for example a fetal stethoscope. More complicated equipment, including hepatitis and HIV tests, are not as available, and are reported in only between a quarter and half o f the facilities. It i s significant that only 61% o f public secondary facilities have a sterilizer intheir delivery wards. Table 29. Availability of obstetric and reproductivehealth equipment, Nigeria (% of facilities) (n=4,503 facilities) public public private sector sector facilities s ~ [$::sector ~ ~ ~ $ ~ ~ fetal stethoscope 91 92 61 blood pressure gauge 93 79 67 artery forceps 89 88 72 Speculum 87 81 37 manual vacuum aspiration kit 54 30 9 Vacuum 43 19 10 Partograph 24 22 14 Sterilizer 82 61 bleach solution 74 53 hepatic test kits 23 24 HIV test kits 33 52 Source is Fatusi and ljadunola (2003). 83. Availability o f basic instrumentsi s similar inprivate facilities as inpublic sector secondary facilities, while some types o f more complicated equipment are more common inprivate facilities - except for HNtests. Significantly more private facilities (82%) than public (62%) have sterilizing equipment. Support Services 84. There are deficiencies in support services,particularly laboratories, although water and electricity supply is good in Lagos. Lack o f adequate funding for PHC services at the local level has underminedrepair and maintenance, as well as capital investment, in services such as ambulances, laboratories, electricity, cold chain, water supply, and environmental sanitation. Figure 44 provides information from a survey o f 252 health facilities inLagos and Kogi states. In Lagos, over 75% o f facilities had water and electricity supply, but around 50% suffer fi-om poor cleanliness and maintenance. InKogi, incontrast, 25% or less o f the facilities hadreliable water and electricity supply, but buildingmaintenance was generally better. Only a few health facilities have a functioning laboratory inboth states. 85. Blood supply and screening is largely ad hoc, but a number of centralized transfusion centers are being established. An assessment o fhospital and PHC services inJigawa State similarly found deficiencies inlaboratory services, electricity supply, and cleanliness. (PATHS, 2003j) In addition, the Jigawa assessment found that bloodtransfusions are mainly obtained from donations byrelatives and fi-om private sellers, andthat there are deficiencies inblood storage, mainly due to problems with electricity. However, all the hospitals visited were instituting HIV screening o f 71 Chapter 3 blood supplies. The government i s making efforts to address the blood supply situation in order to reduce the risks o f transmittinginfectious diseases, particularly HIV, through transfusion. It recently opened the first o f a planned seven centralized transfusion centers where screening will beroutine and standardized. (IRIN,2005) Figure44. Facility cleanliness,maintenance,equipment, and services, Lagos and Kogi States, Nigeria, 2002 (YOof facilities) (n = 252) 100 1 90 - 80 - 70 - .- -g 60 ._ + '3 50 40 30 20 10 0 Source is World Bank (2003b). HealthManagementInformationSystem 86. The health management information system is challenged by the decentralized character of the health system itselJ More centralized health systems have difficulty with disease surveillance and health management information systems, so it i s not surprisingthat these functions present challenges to Nigeria's decentralized and fragmented system. Intheory, information should flow upwards from healthproviders and facilities to the LGA health administration, and then to state ministries of health, and then to the federal ministry o f health. One considerable challenge i s that inmost cases, privatehealthserviceproviders are not includedinthis flow, missinga large proportion o f the relevant data. Another problem i s that the various vertical programs, unable to get the data they need from the existing system, are tempted to institute their own reporting requirements and systems. Consequently, a 1994 survey o f PHC services found no less than 34 different reporting formats for different vertical programs. 87. An information system is inplace, but poorly implemented at both thefacility and administrative levels. The National Health Management Information System was established in 1990 and forms and procedures have been developed and implemented. However, there are difficulties at all levels. Assessments inJigawa and Benue States, for example, indicate that reportingrequirements are complex and unclear, facility record-keeping i s incomplete, information i s poorly transmitted through the system, capacity for data analysis i s l o w within the SMoH, the collected information i s not usedby policy-makers, and there i s no feedback to facilities. (PATHS, 2003c; PATHS, 2003h) A 2001 survey o f health facilities in202 LGAs found that only 44% had the needed H M I S forms on hand and that only 55% had sent completed 72 Chapter 3 forms to the LGA inthe previous three months. Only 16% reportedreceiving feedback from the LGA.(Table 30) Table 30. Health managementinformationsystem at the facility level, Nigeria (% of facilities) (n=674 facilities) ~~ have sent forms receive have forms to LGA in past 3 feedback from months LGA North Central 44 66 13 North East 48 32 11 North West 53 69 16 South East 41 25 10 South South 49 57 12 South West 28 71 32 Nigeria 44 55 16 Source is Adeniyi et a/.(2001). 88. At the LGA level, 92% o f the 202 surveyed LGAshad forms on hand, but only 16% had sent completed forms to the state government. About 66% o f LGAs had compiled statistics for the previous year (Adeniyiet al., 2001). 89. At the state and federal levels, lacking timely and complete data from LGAs and facilities, there i s little possibility for analyzing and usinginformation for planning and policy development. The information relay between the states and the federal ministry i s also weak, as basic information such as state healthbudgets and expenditures or facility numbers i s difficult to obtain inthe capital. 90. Household surveys supplement the routine information system. Nigeria i s not alone inits difficulties in operating a reliable health management information system. For example, the UnitedStates, with its federal system, has similar challenges reporting, coordination and standardization. Insuch situations, facility and household surveys can often provide more reliable sources o f data on a number o f issues, and indeedthis report relies heavily on a number o f such studies. For example, the 2001 NPHCDA facility survey mentioned numerous times provides data not available elsewhere, and a repeat study would be very useful as Nigeriaputs more priority on improving its health system. The government's experience with population- representative household surveys has improved dramatically inthe past few years, as a Nigeria Demographic and Health Survey (NDHS) and a Nigeria Living Standards Survey (NLSS) were completed in2004. A census planned for 2005 will provide essential demographic data. HEALTH SERVICES AND PROGRAMS 91. Packages of essential health services have not been ofleially defined, but the range of services to beprovided by different levels of facilities is generally understood. Although policy work has been done on this by the FMOH, standard service packages have not been officially adopted across the country. However, it i s generally understood that PHC services comprise basic preventive and curative interventions, including health education, immunization, nutrition promotion and monitoring, management o f uncomplicatedmalaria, diarrhea, respiratory infection and other common illnesses, antenatal care, family planning, and some basic surgical care such as incision and drainage. Secondary services include delivery care and management of delivery complications, care for complicated and severe malaria, and surgical services including obstetric, orthopedic and radiological. Most secondary facilities also provide a range o f PHC services, such as antenatal care. The tertiary level i s focused on medical training and specialized services, 73 Chapter 3 including management o f cancer, radiological investigation, renal dialysis and advanced surgical care, (NPC and UNICEF, 2001) 92. Health service utilization i s analyzed in Chapter 2 while service availability i s discussedhere. Immunization 93. Theperformance of immunization services in Nigeria has risen andfallen based on domestic and international interest andfunding. The Expanded Program on Immunization (EPI) started in Nigeria in 1979. An assessment in 1983 explained low coverage by poor funding and vaccine supply, insufficient community mobilization, and over-reliance onhigh-cost mobile campaigns. These problems have continuedto hamper routine immunization inthe country. The program was modified inthe mid 1980s with substantial international donor support, and achieved progress, largely through the use o f campaigns. Coverage peaked in 1990 and then quickly deteriorated due to cuts in donor fundingnot compensated for by government support. For example, UNICEF fundingfor immunization was US$ 90 million in 1990, dropping to US$ 25 million in 1991 and to US$ 10 million in subsequentyears (NPC and UNICEF, 2001). 94. Inthe mid-l990s, the poor state o f immunization services again received attention, so that the National Program on Immunization 0 1 ) was created as a parastatal agency and available fundingincreased through the Family Support Program, headedby the wife o fthe president. Confusion between roles o f the MI,the Family Support Program, the structures o f the federal and state ministries o f health, and the local health authorities, hampered improvement inthe system. A 1997 assessment highlighted vaccine shortages, poor coordination, lack o f community mobilization, poor health staff training, and deficiencies inthe cold chain, particularly at the LGA level (FMOH, 1997). A 2001 facility survey in202 LGAs found that immunization services are offered by 76% o f facilities, but this presumably did not account for real functionality, since vaccine shortages were common. Only 12% o f facilities provided immunization on a daily basis (Table 32). Various small assessments have found that vaccine shortages are an important factor behind low coverage rates. For example, a 2002 assessment o f a USAID-supported project in Lagos, Abia, and Kano states found that despite material and training support to health services, vaccination coverage declined inproject areas over a two-year period due to shortages incentral supply (BASICS 11,2002). 95. Despite recent setbacks, thepolio eradication campaign has been well-supported and has achieved success in expanding coverage. Because o fNigeria's importance as a reservoir o fthe disease, the polio eradication campaign has received considerable international and domestic support inrecent years. Success inexpanding coverage was recently interruptedby controversy over vaccine quality in some northern states. Resurgence intransmission o f the disease has spread polio to other countries and set back the global eradication campaign. The political obstacles have now been resolved, and mass campaigns are being implemented inNigeria and other countries inorder to regainthe lost ground. 96. Due to the weakness of routine immunization, thepolio eradication campaign has dominated immunization efforts in recent years, introducing a number of distortions. The mass mobile campaigns are costly, both interms o f funds and interms o f divertinghuman resources from routine activities. Such campaigns have achieved success interms o f polio vaccine coverage, and can provide a means o f rapidly increasing coverage o f other antigens, but Nigeria's experience over the past two decades suggests that they cannot replace a sustainable routine immunization system, The polio campaignhas increasinglyrelied on monetary incentives to motivate staff to reach coverage goals. This may come to be expected by health staff with regard to improving coverage o f other types o f vaccinations. Finally, the political profile o f polio eradication, both positive and negative, may affect efforts to improve routine immunization. Positive political 74 Chapter 3 attention to polio, translated inparticular into international donor funding, has focused resources on this single antigen, possibly diverting limited capacity away from routine activities. Negative political attention, inparticular the recent controversy over polio vaccination in some northern states, may affect community acceptance o f other vaccines. 97. There is renewedfocus on improving immunization services, with thefederal government retaining a leadership role. Despite the overall division o f responsibilities for health services between the federal, state, and local levels, the federal government has retainedits strong role in immunization, for example specifying this as a federal responsibility inthe proposed Health Act. The NPIi s currently implementing a five-year strategic plan, including a five-year cold chain investment plan. After several years o f vaccine shortages, government fundinghas improved and vaccine procurementhas been handledby UNICEF. Cold chain and distribution, particularly at the LGA level, continue to poseproblems, although this has beenaddressed to some extent by increase investment and private sector involvement. GAVI approved in 2001 US$53.0 million for support to immunization services over five years and in2003 US$27.7 million for support to new and under-usedvaccines. Table 31describes the immunization schedule inNigeria. Table 31. Immunizationschedule, Nigeria, 2005 birth OPVO BCG HBVI 6 weeks OPVl DPTI HBV2 10weeks OPV2 DPT2 14 weeks OPV3 DPT3 HBV3 9 months measles yellow fever ViMl 15 months Via2 OPV oral polio vaccine; BCG Bacille Calmette-Guerin(tuberculosisvaccine); DPT diphtheria, pertussis, tetanus: HBV hepatitis-B vaccine; VitA vitamin A Source is NPI. 98. Divergence between administrative and survey-based estimates of coverageshould be recognized when assessingprogram performance. Government documents refer to successful achievement o f 80% coverage with all antigens by 1990, but this i s not at all evident inthe data collected by the 1990 Demographic and Health Survey, which found that around 30% o f children hadreceivedall recommendedvaccinations. This divergence between official and survey-based estimates o f immunization coverage remain wide even when accounting for the general observation inother countries that DHS estimates tend to be lower than facility-based data by about 10 to 20 percentage points, and i s likely due to N F T s reliance on incomplete administrative estimates from only public sector facilities. ChildHealthandNutritionServices 99. A largeproportion of PHC services address child health, but data are limited on service composition and quality. Of course, many o f the interventions routinely provided by health facilities address child illness and a large proportion o fpatients are children with common conditions, particularly diarrhea, respiratory infection, malaria, and malnutrition. Data on the utilization o f such services i s discussed inChapter 2. Detailed information on availability o f treatments for these particular conditions i s not available. A 2001 PHC facility survey found, however, that "general outpatient services," are available in7040% o f facilities (Table 32). Because this i s a basic service of PHC facilities, the rest o fthe facilities were presumably not functional. Data on service composition and quality are not available, but many o f the ingredients are discussed inthe previous sections (human resources, drugs, and equipment). 75 Chapter 3 Table 32. Child health service availability,Nigeria,2001 (YOof facilities) (n=674) North North North South South Nigeria Central East West East South service is offered general outpatient services 81 84 83 81 77 78 81 immunization 79 67 75 72 80 82 76 growth monitoring 55 44 38 41 59 71 51 school health services 28 21 15 22 19 37 23 service is offered daily general outpatient services 80 80 80 75 71 73 77 immunization 12 12 13 8 11 13 12 growth monitoring 16 7 6 6 10 16 10 school health services 8 3 5 3 4 10 5 Source is Adeniyi et a/. (2001). 100. IMCIhas been introduced but has not yet been scaled-up. The IntegratedManagement o f Childhood Illness (IMCI) program i s an important tool to improve the quality of child health services, involvingtraining o fhealthworkers inchildhood illness management as well as a community component. Supported by international partners (inparticular WHO, UNICEF, and USAID), IMCIwas introducedin 1997 and piloted insix LGAs. So far, the strategy has been introduced in 19 o f the 36 states as well as the FederalCapital Territory. The community component, promotingpractices such as hand washing, exclusive breastfeeding, and use o f oral rehydrationtherapy (ORT) and bednets, was introduced in2002. For the most part, IMCI implementation has been limited to the pilot phase. Table 33. Participation in nutritional programs, Nigeria, 1999-2002 (% children under-5) urban rural nutritional monitoring growth nutritional growth program monitoring program (weigh-in) (weigh-in) Kogi (North Central) 27.5 12.7 27.4 24.6 Yobe (North East) 8.3 11.0 3.0 5.8 Kebbi (North West) 2.1 2.1 5.1 7.1 Abia (South East) 25.2 52.0 22.2 43.3 Enugu (South East) 57.0 57.7 12.5 14.0 Cross River (South South) 57.0 57.7 12.5 14.0 Ekiti (South West) 4.7 51.9 0.0 50.1 Lagos (South West) 80.2 84.3 57.9 57.9 Sources are 1999and 2002 CWIQS. 101. A number of other initiatives relating to child health have been established over the years, usually under the impetus of international donors, including the Baby Friendly Initiative, the Child Survival Program, the Women and Children Friendly Health Services Initiative, and the Roll Bank Malaria program. Inaddition to such programs as well as the various departments of the FMOH, the NationalProgramon Immunization(NPI) andthe NationalPrimary Health Care Development Agency (NPHCDA) are also o f course working on child health services. 102. Policy work on nutrition has advanced. Growing recognitionthat highrates o f child malnutrition underlie much child morbidity and mortality ledto the creationin 1990 of the National Committee on Food and Nutrition, under the National Planning Commission, reflecting 76 Chapter 3 the multisectoral approach to nutrition. A National NutritionPolicy was adopted in 2002. The FMOHcontains a division onnutrition policy which works on standards andprotocols. 103. Nutritional sewices reachpart of thepopulation and there has beenprogress in particular areas such as breastfeedingpromotion and micronutrient supplementation. The 2001 facility survey found that halfo f PHC facilities offered growth monitoring, even though this was only available in 10% o f facilities on a daily basis (Table 32). Table 33 provides household survey data on the proportions o f under-five children indifferent States who participate in nutritional programs. This shows, first, that such programs are inplace inmany States, and second, that there are large differences between urban and rural areas and between States intheir coverage. Over half o f under-five children inurban areas o f many States inthe south participate ingrowth monitoring, while only 2% to 10%ofchildreninurbanandrural areas of some States inthe northare covered. 104. A 2002 World Bank study o f five LGAsin 12 Stateswith active nutrition programs found deficiencies inthe quality o f nutritional interventions at the PHC level. Only 40% o f sampled health workers had been trained inbreastfeeding promotion, 29% ingrowth monitoring, and 28% invitamin A supplementation. About 20% o f the workers had access to nutritional protocols or guidelines. The survey found that although 70% o f the children brought for consultation are weighed and 68% receive diet advice, only 55% are given information on vitamin A supplementation, and 37% are examined for anemia. (World Bank, 2002a) 105. Despite such deficiencies inquality, overall there has beenprogress inparticular areas such as breastfeeding promotion and micronutrient supplementation. Data fi-om the 2003 NDHS indicate that exclusive breastfeeding has become more common. A study in Osun state found that women were significantly more likely to exclusively breastfeed ifthey delivered inhealth centers where breastfeeding was promoted (Ojofeitimi et al., 2000). Salt iodization, advocated by UNICEF,is a preventive health success story inNigeria, so that the 2003 NDHS found that 97.2% o f households now use iodized salt. Vitamin A supplementation has also achieved a boost through distribution through vaccination campaigns. Population, Reproductive and Maternal Health Services 106. Building on groundwork in the 1980s and early I990s,policy development onpopulation and reproductive health has advanced in recent years. A National Policy on Population was adopted in 1989 and the National Population Commission and Population Activities Fund Agency were created, the latter under the impetus o f some international donors (including the World Bank). Along with the FMOHand the NPHCDA,the proliferation o f government bodies involved inpopulation and reproductive health issues contributedto duplication and poor coordination at the federal level. Like with nutrition and other areas, after a hiatus due to lack o f internationaland domestic funding duringthe 1990s, interest and policy work have been revived since 2000. The National Population Commission was reconstituted, with its mandate focused on demographic registration, enumeration, and data issues; the Population Activities FundAgency disbanded. 107. In2001the FMOHtook the leadwith anewNationalReproductiveHealthPolicyand Strategy that reflected the 1994 International Conference on Population and Development consensus inCairo, which expanded the focus beyond family planning to reproductive health, including emphasis on human rights. The policy and strategy encompass safe motherhood, in particular antenatal and delivery care, family planning, adolescent reproductive health, prevention and treatment o f sexually transmitted infections (STIs) and HIVIAIDS, and address harmful practices and reproductive rights. 77 Chapter 3 108. Like child health, a number o f different programs have been established to improve family planning, reproductive, and maternal health services. These include family planning programs, the Safe Motherhood Initiative, the Baby Friendly Initiative, Roll Back Malaria, and the national HIVIAIDS program. Table 34. Reproductive and maternal healthservice availability,Nigeria, 2001 (% of facilities) (n=674) North North North South South So:; Nigeria Central East West East South service is offered family planning 61 42 37 38 49 48 46 antenatal care 79 53 46 53 66 78 63 deliveries 71 52 31 48 60 78 56 post-abortion care 18 16 15 15 13 11 15 service is offereddaily family planning 39 26 10 17 33 35 27 antenatal care 37 16 10 8 27 28 22 deliveries 58 42 28 33 46 55 44 post-abortion care 7 9 9 6 8 4 7 Source is Adeniyi et al. (2001). 109. Availability of family planning services expanded in the 1990s,but social, cultural and economicfactors keep utilization low. Also like basic child health interventions, reproductive and maternalhealth services -particularly family planning and antenatal and delivery care - are integralto the PHC strategy implementedby Nigeria since the 1980s. By the mid 1990s, family planning services had expanded through bothpublic and private providers. Like with other aspects o f the health system, international and domestic fundingcuts, particularly affecting contraceptive supply, reducedthe availability o f family planning services by the end o f the decade. Nevertheless, the 1999 NDHS indicatedthat modem contraceptives were available within 5 kmto 70-80% o fhouseholds inurbanareas and to over 50% o fhouseholds inrural areas, even among the poorest quintiles (Figure 45). Consistent with this, a 2001 facility survey found that 46% o f facilities offered family planning services (Table 34). Utilizationo f modem contraceptives i s far lower (8.9% o f adult women in2003), suggesting that preferences are more important than availability o f services. This i s supported by an estimate from the 2003 NDHS that over 80% o f currently married women want more children. Although this may indicate that greater information and education efforts are required, structural social, cultural and economic factors are likely determinants. Nevertheless the 2003 DHS also estimates the unmet need for contraception o f about 17 % indicating that there may be still some supply issues incertain areas (i.e. it could be a function o f women not being able to readily obtain the type o f contraceptive they desire). 110. Antenatal care is a basic element of the PHC stratea, and the service is available in about two-thirds of healthfacilities. The 2001facility survey found that antenatal care i s available in 63% o f health facilities, although only 22% offered the service daily (Table 34). With regardto utilization, the 2003 NDHS found that around 60% o f women receivedantenatal care during their most recent pregnancy. 111. Thequality of antenatal care could be improved, particularly in rural areas. The 2003 NDHS collected data on the content o f antenatal care which provide indicators for service quality (Table 35). About 80% o f the women who receive antenatal care reported having their weight and bloodpressure measured. One would expect that such basic diagnostic interventions would be done in all cases, but there may be deficiencies ininstrument availability (discussed ina previous section). Only around 66% had urine or blood samples taken, while just over 50% were 78 Chapter 3 informed o f signs for pregnancy complications. Overall, this presents a picture o f poor quality care, Quality i s considerably lower inrural areas, as for example, around 50% of women inrural areas had a urine or blood sample taken, compared to over 80% inrural areas. Figure45. Availability of modern contraceptivemethodwithin 5 km, Nigeria 1999 (% of households) 90.0 70.0 Ln 60.0 - P 2 50.0 - ' w2 ::::I c 40.0 - 30.0 - , , , , , 0.0 1 2 3 4 5 economic status quintiles Authors' estimates from 1999 NDHS data. 112. Traditional birth attendants have been trained in about half of communities, butfew receive continuing support. Training o ftraditional birthattendants (TBA) i s also an important element o f the PHC strategy adopted by Nigeria inthe 1980s. A 2001 survey o f 674 villages in 202 LGAs across the country found that 52% o f villages had a trained TBA, but only 8% were regularly supplied with delivery kits (see Table 24 above), indicating poor follow-up support by the health system to TBA training programs. The 2003 NDHSfound that 20% o f deliveries are assisted by a TBA, 24% inrural areas and 12% inurban areas. Table 35. Quality of antenatal care, Nigeria,2003 (% of women who received antenatal care) (n=2,462) urban rural Nigeria informed of signs of pregnancy complications 66 48 55 weight measured 92 77 83 height measured 64 51 57 blood pressure measured 91 74 81 urinesample taken 81 53 64 blood sample taken 81 55 65 Source is 2003 NDHS. 113. QualiJieddelivety care is available to most urban households and to about two thirds of rural households, but much less available to thepoor in rural areas. The 1999 NDHS found that inurban areas, delivery services by aprofessional provider were available within 5 kmto 88% o f households. Overall inrural areas in 1999, 68% o f households had delivery services available within 5 km. Inurban areas, as could be expected, there is no clear patterno f differences in availability by socio-economic status. However, inrural areas, it i s clear that availability o f qualified delivery assistance i s much lower for the poor, risingmonotonically with socio- economic status. (Figure 46) The 1999 data also indicate that both inurban and rural areas, delivery services are most available from PHC facilities, which are mostly public sector. Inurban areas, private doctors are the second most available provider, while inrural areas it i s trained midwives. 79 Chapter 3 Figure 46. Availability of qualified deliveryassistance within 5 km, Nigeria, 1999(% of households) urban rural health posffcliniclcenter 70.0 health posffcliniclcenter i 8oo 70 0 80.0I 1 1 trained midwife uI 60.0 - private doctor 0 privatedoctor S g 50.0 hospital - S f0 40.0- a g 500 trained midwife I jo0! hospital s 30.0 - s 30.0 20.0 - 20.011 10.0 1 10.0 0.0 ` 0.01 1 2 3 4 5 1 2 3 4 5 economic status quintiles economic status quintiles Authors' estimates from 1999 NDHSdata 114. Utilization seems to be somewhat less than availability, likely due to cost andperhaps culturalfactors. The 2003 NDHS found that 36% o f deliveries are assisted by qualified personnel - 59% inurban areas and 27% inrural areas. Indeed, regression analysis o f data from the 1999 NDHS,controlling for householdeconomic status, indicates that inurban areas, availability o f delivery services i s not a significant determinant o f utilization. Inrural areas, however, controlling for economic status, the odds o f delivery by a health professional in community where services are available within 5 kmare 1.85 times the odds inother communities. 115. Most health workers have not received training toprevent neonatal mortality. Training in"life saving skills" is an important strategy for improvingneonatalresuscitation andcare. A 2003 study o f public sector services found that only 5-10% o f relevant health care workers had received this training. (Fatusi and Ijadunola, 2003) 116. Referral carefor delivery complications- a key determinant of maternal mortality - is least available to thepoor in rural areas. Data from the 1999NDHS on distance to the nearest referral services are described inFigure 47. Overall, inurban areas, the meanreported distance to such services i s 6.4 km, while inrural areas it i s 16.9 km. However, for households inthe poorest quintile inrural areas, the mean distance i s over 26 km, compared to 12 kmfor households inthe highest quintile. 117. Coverage of basic emergency obstetric care is insuflcient in most regions, particularly in the north. A 2003 study ofpublic and private health facilities in 12 states (two ineach geopolitical zone) estimated the coverage o f basic and comprehensive emergency obstetric care (EOC). Basic EOCinvolves managed deliveries, particularly administrationo f drugs and assisted delivery, while comprehensive EOC includes the ability to provide caesarean-sections. Coverage targets cited were 4 basic EOC facilities and 1 comprehensive EOC facility per 500,000 population. This study estimated that only inLagos the target for basic EOC coverage i s met. The states inthe North Center and North West had coverage o funder 1facility per 500,000 population. Inthe North East, Borno state had coverage o f under 1facility, while Taraba was higher, at almost 2 per 500,000 population. Similarly, inthe South Southzone, Cross River's coverage was under 1, while in Edo it was around 1.5 per 500,000 population. This poor 80 Chapter 3 coverage i s attributed to shortcomings inmidwifery training and lack o f equipment, particularly inPHCfacilities. Figure 47. Mean distance to referralservices for delivery complications,Nigeria, 1999 (km) 30.0 7 2 25.0 - .- $ - 20.0 - s2 E 15.0 - B .- 5E 10.0 - urban n E C 5.0 - 0.0 4 1 2 3 4 5 economic status quintiles Authors' estimates from 1999 NDHSdata. 118. However, coverage of comprehensive emergency obstetric care is deemed acceptable. With regardto comprehensive EOC, all surveyed states meet the target o fone facility per 500,000 population. The highest coverage i s inLagos (South West) with 18, while the lowest i s inSokoto (North West). The study found that most facilities providing this care are inurban areas. Figure48. Comprehensive emergency obstetric care (EOC) services provided by public referral-level facilities, Nigeria, 2003 100 1 1?:! North 1 North South South South Center East South ~ East ~ West ~ ~ Source is Fatusiand ljadunola (2003). 119. In all states,private sectorfacilities provide higher comprehensive EOC coverage than dopublic sectorfacilities. For example, inCross River state (South South) there is 1public and 4 private facilities per 500,000 which provide comprehensive EOC. Ingeneral, the majority o f public referral-level facilities do not provide comprehensive EOC, even though they should all normally provide such care. The study found that only 30% o fpublic referral facilities provided comprehensive EOC. Figure 48 shows that only inLagos does the proportion exceed 50%, while 81 Chapter 3 inmost states 30-40% ofpublic referral facilitiesprovide comprehensive EOC. (Fatusi and Ijadunola, 2003) 120. Quality of EOC is lacking in manyfacilities. Deficiencies inequipment are discussed in a previous section, which presents data showing that only a minority o f public sector hospitals have more than basic obstetrical equipment and instruments (Table 29). Similarly, assessments o f emergency obstetric care services inJigawa and Benue States have found deficiencies inthe quality o f such care. InJigawa State, an assessment o f two hospitals found shortages o f equipment and inadequate training o f staff. (PATHS, 2003d) InBenue State, a study o f 100 health facilities found that only 6 could provide a complete set o f basic emergency obstetric care services; for example, only 35 could perform cesearean sections and only 1had forceps for assisted delivery. The case fatality rate was found to be high, at 7% o f complicated deliveries (PATHS, 20030. HIV/AIDS Policy and Programs 121. Thepolicy and institutionalframework for combatingHIV/AIDS has been developed. A National AIDS Control Programwas established under the FMOHin 1986 and it i s responsible for the health sector's response to the epidemic. In2000, political commitment by the President led to the creation o f a multisectoralNational Action Committee on HIV/AIDS (NACA). State action committees have also been established. In2001, a three-year strategic plan focused on institutional development and mass awareness campaigns, as well as targeted interventionsfor high-riskgroups. 122. In2003, aNationalPolicyonHIVIAIDSwas adoptedwiththe followingcomponents which cover the range o f HIV/AIDS activities: 123. preventiono f H N / A I D S transmission: promotion o f safe sexual behavior, appropriate use o f condoms, bloodtransfusion safety, voluntary counselingand testing, prevention o f mother- to-child transmission, early treatment o f sexually transmitted infections (STIs) and youth-focused interventions; i) law and ethics: respect for the rights o f persons living with and affected by HIV/AIDS; ii) care and support to those infected or affected by HIV/AIDs: access to care for persons livingwith HIV/AIDS, home-based care, treatment o f opportunistic infections, access to anti-retroviral therapy, care for orphans and vulnerable children, support for people affectedby the disease, and certification o f traditional healers and other practitioners; iii) communication: improvinginformationandknowledge,andcatalyzingcommunity- basedresponses; iv) program development and management: institutional structure for multisectoral response, build capacity, implement programs, and monitoring and evaluation. 124. International partners are supporting the national strategy. The World Bank Multisectoral HIV/AIDS Program (MAP) i s supporting a five-year US$ 90 million project which started in2002. In2003, the Global Fundapproved grants totaling US$28.2 million over two years to promote the participation o f civil society organizations in combating HIV/AIDS and to support the expansion o f prevention o f mother-to-child transmission and anti-retroviral therapy (ART) services. A number o fbilateral donors support HIV/AIDS activities, inparticularthe US, which in 2004 had a budget o f US$55.5 million for prevention, treatment, and care interventions to be implementedby non-governmental actors. 125. Preventive interventions have expanded but more effort is needed to reach thepoor in rural areas. Data from 1999 indicate that about a thirdo f the population had been exposed to 82 Chapter 3 HIV/AIDS awareness campaigns inthe previous year, but the proportioni s lower (around a quarter) among the poor inrural areas (Figure 49). Community awareness has expanded since then, suggesting that such campaigns have had an impact. The 1999 NDHS found, for example, that only 17% o f women inurban areas and 7% inrural areas knew o f condoms as a preventive measure, while in2003, these proportions had increasedto 60% and 40% respectively. A number o f small initiatives, often involving non-governmental organizations, are targeting high-risk groups - an efficient strategy for reducing transmission o f the virus. A 2001 study o f 2,634 sex workers inseven cities inNigeria found that 84% hadbeen exposed to HIV prevention and condom promotion messages and that such exposure was significantly associated with condom use ina regressionanalysis controlling for other factors (Oladosu and Ladipo, 2001). 126. Voluntary counseling and testing has reportedly expanded but the authors do not have data on availability o f this service. Figure49. HIV/AIDScampaignin community in previous year (% of households), Nigeria, 1999 HIV/AiDS campaign 1 in previous year 10.0- -I 0.0 1 2 3 4 5 economic status quintile Authors' estimates from 1999 NDHS data. 127. Anti-retroviral treatment is available to a small number, but scaling-upfaces considerable challenges. In2002, the FMOHstarted implementinga government-financed anti- retroviral therapy program with the distribution o f drugs to 25 designated centers. In2003, there were around 13,000 people being treated under the program. There are few patients receiving treatment by the private sector due to the cost. A 2004 assessment o f 15 facilities providing treatment found that funding and drug supply problems as well as capacity constraints will need to be addressed inorder to effectively scale-up the government program. (Durgavich et al., 2004) 128. Anti-retroviral treatment to HIVpositive pregnant women inorder to prevent mother-to- child transmission has been introducedbut confronts similar challenges to scaling up. It i s reportedthat in2002, only 141HIV-positive pregnant women were treated with ARVs to prevent transmission. 129. A suweillance system is inplace and a recentprevalence suwey provides important data for program evaluation andplanning. A surveillance system based on testing o f antenatal care clients in sentinel sites has been established. In2004, the government completed a national prevalence survey which provides crucial data for assessing the status of the epidemic and informing strategies and programs. 83 Chapter 3 Malaria Strategy and Programs 130. Nigeria adopted the Roll Back Malaria strategy in 2000 and developed a strategicplan in 2001. The National Malaria and Vector Control Division o f the FMOHtakes the lead. The elements o f the strategy are: i)case management; ii)prevention; iii)information, education, and communication and community mobilization; iv) partnerships; v) operational research; and vi) monitoring and evaluation. 131. Improving case management will depend onprogress in PHC services, but Nigeria has also adopted a home-based treatment strategy. Dueto increasing resistance to standard anti- malarial drugs, Nigeria recently changed its drugpolicy to artemisin combination therapy (ACT) for first-line treatment o f uncomplicatedmalaria. A baseline assessment in 6 LGAs to inform the strategy found that 85% o f health facilities inrural areas had stock-outs o f anti-malarials during the previous three months, and that malaria treatment guidelines were not available inany facility. Achieving the strategy with regardto case management will be greatly dependent on improvements inthe primary health care system. The study found that public facilities were underutilized, with over 80% treating less than 15 patients per day. Consistent with this, it was estimated that 80% o f malaria cases are treated at home. Nigeria has therefore adopted a home- base care strategy, intending to make ACTSavailable directly to households through pharmaceutical retailers. 132. With regard toprevention, intermittentprevent treatment (IPT) ofpregnant women and insecticide-treated nets (InVs) are the main strategies. Data on availability o f IPT services are not available to the authors. With regardto ITNs, the national programreports that around 1.2 million nets were distributed in2002 and 1.5 million in2003, representing a fraction o f the total number needed to achieve significant coverage. A 2002 assessment found that local productioni s limited, and taxes on productionmaterials are as highas 25% (UNICEF, 2002). The tax on imported nets was reduceda few years ago from 25% to 5%. 133. Implementation of the strategy in some states is supported by the Global Fund. In2004, the Global Fundapproved a two-year grant o f US$41.5 millionto scale up the Roll Back Malaria strategy in 12 states and to introduce ACT in 6 states. 134. Meeting service coverage targets will require signiJicant effort. Nigeria hostedan international conference inAbuja in2000 and, by 2005, it committed to achieving 60% access to affordable treatment, 60% access to IPT for pregnant women, and 60% coverage o f children and pregnant women with ITNs. Service utilization data from the 2003 NDHS indicate that these targets are unlikely to be met: only 33.9% o f febrile children take an anti-malarial drug, 1.0% o f pregnant women receive IPT, and 1.2% o f under-five children and 1.3% o f pregnant women sleep under an ITN. TuberculosisStrategy and Programs 135. I n 2002, a strategicplanfor tuberculosis (TB) control was adopted,focusing on expansion of coverage of directly-observed treatment (short-course) (DOTS). The National Tuberculosis and Leprosy Control Program, established in 1991under the FMOH, takes the lead, with counterparts ineach o fthe states. 136. There has beenprogress in increasingDOTS coverage, but case detection is still very low. In2003,2,233 DOTS treatment centers had been set up and 477 microscopy centers were in operation. By 2004, DOTS hadbeen introducedin at least one LGA in each state. However, (smear-positive) case detection under DOTS i s estimated to be very low, at 18% in2003, little changed from 1997 and far from the global goal o f 70%. Part o f the reason for this i s the weakness o fpublic sector PHC services. InJigawa state for example, the state TB control 84 Chapter 3 program started in2003 and established 10 treatment centers and 5 microscopy centers within a year. However, the weakness o f establishedPHC services means that althoughTB activities are runninginPHC facilities, utilizationis low because most patients seek care elsewhere -in particular hospitals. (PATHS, 2003e) Recognizingthis, the national program i s working on involvingprivate clinics indelivery o f DOTS. 137. DOTS treatment success i s reportedto be 79%, close to the global goal o f 85%. 138. Governmentfinancial support has been limited, so that TB control is dependent on international donors. In2003, the nationalprogram's budget was US$ 12.6 million, o f which US$ 3.9 millionwas budgetedby the federal government, but only US$ 1.9millionreleased. A two-year grant o f US$9.8 million was approved by the Global Fundin2003, but withdrawn due to insufficient financial management arrangements and counterpart funding. (WHO, 2005) HEALTH SECTOR STRATEGY 139. Thepublic sector health system expanded dramatically in the 1970s and 80s. At independence in 1960, health services inNigeria were largely focused on curative care and centered inurban areas. Inthe 1960s and 70s, government considerably expanded the health care system, and its third national development plan for 1975-80 introduced a health sector development strategy focusing on primary health care. Duringthe early 1980s, the Basic Health Services Scheme (BHSS), inline with the 1978 Alma Ata Declaration, envisionedorganizing PHC services on a "district" model, by which one comprehensive health center, four health centers, and 20 clinics would serve a catchment population o f 150,000. A new cadre o f PHC workers was created and Schools o f Health Technology were to be established ineach state to train them. 140. The 1988National Health Policyfurther emphasizedprimary health care, andput in place the structures and division of responsibilities which characterize the system today. The policy and subsequent directives definedthe responsibilitieso f the federal, state, and local levels o f governments, and created the various state and federal programs and parastatal agencies responsible for specific aspects o fthe sector. 141. Inthe 1980sandearly 1990s,the strategy to improveprimaryhealthcareinvolved training PHCworkers, infrastructure investment, and setting up a governance structure which involved district and village development committees as well as LGA and state PHC Management Committees. The community-level component o f this strategy included training village health workers and traditional birth attendants (TBAs), as well as a householdregistration and records system. Resources and technical assistance were to be provided to the LGAsby states and the federal government, which established the National Primary Health Care Development Agency (NPHCDA) in 1992 for this purpose. LGAsreceived federal grants to establish model PHC services on the district model. A large-scale effort to institute Bamako Initiative-type drugrevolving funds was funded by a government oil revenue fund and external donors. The 1988National Healthpolicy specified user fees for curative services but significant subsidies for preventive interventions. Basic PHC services were broadly defined as including health education, adequate nutrition, safe water and sanitation, maternal and childhealth interventions, family planning, immunization, endemic and epidemic disease control, treatment o f common diseases and injuries, and provision o f essential drugs and supplies. Health education to promote individual responsibility was stressed. 142. Progress was achieved in the creation of the new institutional structures, particularly relating to the decentralization of responsibility for PHC to the LGAs. For example, as noted in previous sections, district and village committees are inplace and drugrevolving funds are operating inmany areas. Similarly, investment ininfrastructure, equipment, and training was 85 Chapter 3 considerable, while vertical programs such as immunization achieved substantial coverage. (World Bank, 1991) 143. However, institutional changes and investments were to a large extent notfollowed-up with sustained support, as the three levels of government, as well as external donors, reduced funding for health sewices during the military dictatorship of the 1990s. Political interference, increased corruption, and general deterioration o f public sector governance also severely affected the health system. Development o f the system became characterized by one-off investments, in infrastructureor drugs, determined by political criteria, and with little or no provision for sustainability. Ingeneral, institutions, such as the various coordinationcommittees at different levels, became non-functional, service availability and quality was reduced, and utilization declined. PovertyReductionStrategy and MDGprogram 144. Thegovernment democratically-elected in 1999started the difficultprocess of reversing the deterioration of the health system, includingpolicy development and increasing spending on health. Its 2004 Poverty Reduction Strategy Paper (PRSP), called the National Economic Empowerment and Development Strategy (NEEDS), includes a "Social Charter," focusing on human development, including a significant component on the health sector. The overall goal i s to improve the health status o f Nigerians through strengthening primary health care services and undertaking health sector reforms. The "policy thrusts" are: i) to improve the government's stewardship roles o f policy and regulation, resource mobilization, and monitoring and evaluation; ii) "to strengthen the national health system and improve its management;" iii) toimproveavailabilityandmanagementofhealthsystemresources; iv) to reduce the burden o f priority diseases, including malaria, TB, HIV/AIDS, and reproductive conditions; v) to improve physical and financial access to quality health services; vi) to increase consumers' awareness o f their health rights and obligations; and vii) to foster partnership with all actors inthe sector. 145. Table 37 lists the actions cited inthe NEEDS to achieve these objectives The more specific actions, such as establishing a National Hospital Services Commission and refurbishing teaching hospitals under an existing plan (VAMED), are related to federal responsibilities. More general intentions, such as improving health system resource management or providing a minimumpackage ofhealth services, are improvementswhich depend on all three levels o f government. More specifics about how these more general aspirations will be achieved are lacking, Similarly, rationalization o f the structures o f the FMOHand other federal institutions involved inhealthi s a specific goal, but, aside from the intentionto develop LGA capacity, changes inthe institutional structure at the state and LGA levels, largely outside o f federal control, are not mentioned. 146. With regard topolicy development, a range of important issues are to be addressed in a concrete manner, including national health accounts and insurance, reform of thefederal hospital management board, strengthening of drug regulation, and development of a National Health Act. Less specific are intentions to improve public-private partnerships and community participation. Although a strategy to improve the morale o f health workers i s mentioned, 86 Chapter 3 development o f a health humanresource strategy, addressingrequirements, training, and labor market incentives in a comprehensive way, i s not included. Table 36. Health sector component of the 2004 NEEDS Policy redefinition of the roles and responsibilitiesof the FMOH and other federal structures restructuring of the FMOH and other federal health institutions Review of existing health policies enactment of new National HealthAct strengthen FMOH policy capacity strengthen use of evidence and information establish National Hospital Services Commissionfor management of tertiary hospitals Burden of disease study study private healthsector and develop public-private partnership policy enhance coordination of developmentpartners Financing National HealthAccounts develop health care financing strategy National Health Insurance Scheme advocacy for increased budgetaryallocation to health at all levels hospitals and refurbish teaching hospitals other higher-level establish mechanismfor measuringperformance of tertiary hospitals Services refurbish National Drug Production Laboratory establish National Blood Transfusionsystem health system establish systems for efficient managementof health resources Resources strategy to improve health workers' attitude, morale and commitment establish reliablesupply system for drugs and medical supplies strengthen drug regulation (NAFDAC) enabling environment for local manufacture of 70% of essential drug and ARV needs primary health refurbish PHC facilities Care strengthen LGA capacity rapid and sustainable increase in routine immunization provision of a minimum packageof health services to all Nigerians increase antenatal, postnatal, and family planning services disease control strategic plans for malaria, TB, reproductivehealth, etc. develop health sector response to HIVIAIDS strengthen disease-specific initiatives (eg. guinea worm) stop polio transmission by end 2004 detection and response to epidemics outreach and increase consumers' awareness of personal obligations and rights to better health participation enhance community participation in provision and financing of health services campaign for eradication of harmfultraditional practices 147. A number of states have also developed StateEconomic Empowerment and Development Strategies (SEEDS). Box 1discusses examples from Lagos and EnuguStates. Inaddition, in some states, local strategies (LEEDS) are beingdeveloped. All o f these documents - federal, state, and local - include objectives and actions relatedto primary health care. This reflects less a lack o f clarity about the division o fresponsibilities - since it i s widely accepted that LGAs are responsible for PHC- and more the recognitionthat LGAsrequire significant support to fulfill theirresponsibility. Inrecent years, such support was largely not forthcoming. As Nigeria moves 87 Chapter 3 forward, the NEEDS and SEEDSshould provide a basis for the different levels o f government to work together to improve PHC services. 148. In 2004, thefederal government developed aprogram to achieveprogress towards the health-relatedMDGs in 14target states. The main areas o f action are routine immunization, improving basic child health care (IMCI), improving maternal care, malaria prevention, and HIV/AIDSinterventions. The document statesthat LGAsare primarily responsible for routine immunization and that there should be a "Charter" between the different levels o f government on the issue. Onmaternal mortality, there i s an innovative suggestion o f collaboration with the National Union o f RoadTransport Workers on access to emergency obstetric care. Otherwise, implementation modalities on issues which are largely local and state responsibilitiesare not described. The MDGprogram also describes the federal government's health sector reform goals, which mirror the actions includedinthe NEEDS, but also include recognition that health human resource strategy development i s needed. 88 Chapter 3 NationalHealth Act 149. Thefederal government introduced toparliament in 2004 a National Health Bill which generally codifies the structuresput in place by the 1988Health Policy, answering some of the implementation questions raised by the NEEDS, the MDGprogram, and other health sector strategy documents. Ingeneral, once it i s enacted, the new National Health Act will codify the structures put inplace by the 1988 Health Policy, particularly the division o f responsibility between the federal, state and local governments, with a significant role inprimary health care for federal vertical programs and a parastatal agency. 150. The objective o f the Act i s to regulate the national health system, including public and private providers, so that services are provided equitably to the population o f the country. The objective i s stated inrealistic terms, indicating that services will depend on available resources and that access will be realizedprogressively. The national health system i s defined as the federal and state ministries o f health, federal and state parastatal agencies under the ministries o f health, local government health authorities, ward (district) health committees, village health committees, private healthproviders, and traditional and alternative health providers. Inline with the 1988 Policy, the Act states that the federal government i s responsible for tertiary services, state governments are responsible for secondary services, and local governments are responsible for primary health care services. 151. TheAct strengthens thefederal National Primaly Health CareAgency (NPHCDA). The functions o f the federal Ministry o f Health include ensuring the provision of tertiary and specialized hospital services, as well as policy development and planning, health information systems, and provision o f technical assistance to the states. Inaddition, the Act suggests significant operational involvement o f the federal government inprimary health care services. There i s a clear provision that the federal level i s to ensure routine immunization, and more importantly the Act strengthens the National Primary Health Care Development Agency (NPHCDA). 152. TheAct creates a National Primary Health CareDevelopment Fund, apotentially important instrument to channelfederal and state resources to PHCsewices. The Act confirms the NPHCDA,first created in 1992. It i s to be governed by representatives from each o f the three levels o f government. The agency i s to define the minimumstandards o fprimary health care, carry out regular audits o f health care personnel, and produce, inconsultation with the states, periodic master plans for development o f PHC services. Inaddition, the agency i s to "inquire and advise" federal, state and local governments on funding for PHC. Most importantly, ina departure from the current situation, the Act creates the National Primary Health Care Development Fund, managedby the agency, intendedto finance investment and recurrent costs for PHC, including drugs and salaries. The fund i s to receive contributions from alcohol, tobacco and other taxes, grants from international donors and the federal government, and "counterpart" fundingby state and local governments. The Act specifies that states and local governments which do not contribute their counterpart funding will not receive grants from the Fund. 153. This mechanism has the potentialto bothchannel greater federal and state resourcesto PHC services. Along with this fund, the NPHCDA also will potentially increase federal involvement inPHC services. With this, the draft Act represents a serious effort to overcome the structural constraints to improving PHC services -particularly the lack of capacity and resources allocated to health at the local level. The amounts to be contributedby federal, state and local governments to the National Primary Health Care Development Fundare not mentioned, nor i s a process outlined for how these will be determined. 89 Chapter 3 154. Funds are to be disbursedthrough State PrimaryHealth Care Boards (also new under the Act) for distribution to local government health authorities. The Act states that 60% will be allocated "on the basis o f equality o f States" and 40% on the basis o f population size, child and maternal mortality, HIV prevalence, and other criteria that may be determined by the board o f the NF'HCDA. 155. Thefunctions of the state ministries of health outlined in the Act also suggest signijkant involvement in primary health care services. Inaddition to state-level policy and planning, the state ministries of health are to "coordinate" the fundingand financial management o f local government health authorities, provide technical and material support to the local level, and generally "facilitate and promote the provision o f comprehensive primary health services and community hospital services." 156. The Act creates State PrimaryHealthCare Boards, includingrotatingrepresentatives of local governments, to manage the funds which are to come from the National Primary Health Care Development Fund. These boards are to be "responsible for the coordination o fplanning, budgeting, provision and monitoring o f all primary health care services [...]." 157. At the same time, the Act specifies that local governments are "responsible for the coordination o f planning, budgeting, provision and monitoring o f all primary health care services [..,I," is exactly the same phraseusedwithregardto the State Primary Health CareBoards. This This kind of ambiguity, alongwith the fact that the Act recognizes the division o fresponsibility between the three levels o f government at the same time as creating federal and state structures to intervene inprimary health care, i s reminiscent o f the current situation. 158. TheAct confirms or sets up a number of governance structures. These include the National Health Council and State Health Councils which are to bringtogether federal, state, and local health authorities to advise on policy. Consultative bodies at the federal and state levels are to involve non-governmental stakeholders, including international donors. Consistent with the 1988 Health Policy, the Act confirms Ward (district) Health Committees, at the sub-LGAlevel, which are to coordinate PHC services ineach ward. A Village Health Committee i s to be created bylocal government authorities ineachcommunity without aPHC center, and the Act provides quite specific stipulations regarding its functions - including setting up a health post, and selecting, supervising, andremunerating village health workers and traditional birthattendants. 159. TheAct also confirms the existing structure of hospital management boards, setting out government power to approve new public andprivate hospital services. It establishes the National Hospital Services Agency, which i s to work on hospital policy and standards, and will consider applications and make recommendations to the FMOHon approvals (Certificates o f Need) for new public and private tertiary hospitals. State Hospital Management Boards are responsible for the administration and management o f state hospitals. 160. TheAct therefore sets out quite a complex structure, withfour sets of governance bodies. Firstare the federal and state ministries o fhealththemselves, along withthe local government health authorities, and ward and village health committees. Second i s the National Primary Health Care Development Agency and Fundand the State Primary Health Care Boards. Third are the National and State Health Councils and consultative bodies. Fourth are the federal and state hospitalboards. 161. The draft Act deals with a number o f other issues. A section sets out the rights and obligations o f health care providers and patients. Examples are patient rights to informed consent to treatment and to confidentiality, and the provider obligation to provide information on services and fees. 90 Chapter 3 162. Inrecognitionofthe importance ofthis issue, somewhatneglectedbypreviouspolicy documents, the draft Act specifies that the National Health Council will develop strategy, policy and regulation on healthhuman resources. 163. The draft legislation leaves openpossibilitiesfor different types ofpublic-private partnerships by stating that health authorities at any level of government may agree with a private or non-governmental provider "in order to achieve any object of this Act. " A significant section o f the Act i s devoted to regulationo f private health providers, setting out terms and procedures for obtaining "Certificates o f Need" from relevant federal and state agencies. A number o f criteria to be considered for approval are set out, including consistency with national, state and local planning, and the qualifications o f applicants. 164. The draft National Health Act is a signiJicant development in healthpolicy in Nigeria. It codifies existing structures which have been inplace since the 1980s, reflecting the accepted division o fresponsibilitiesbetween levels o f government and continuing the strategy o f decentralizing PHC management towards the LGA, ward, and village levels. However, at the same time, the federal and state roles inPHC delivery are strengthened, notably with the creation of national and state funds for PHC services which are to receive resources from the three levels o f government. This i s the government's strategy for addressing the key problem inthe sector - the weakness o f PHC services. Its success will depend on negotiation and coordination between the federal, state, and local government authorities, as well as avoidance o fthe problems which limited the effectiveness o f such parallel structures inthe past -particularly, political interference and funding cuts. 91 Chapter 4 CHAPTER 4. THE ROLE OFTHE PRIVATE SECTOR IN HEALTHCARE PROVISIONINNIGERIA 1. This chapter will: (a) explore the contribution ofthe private sector to overall health service delivery inNigeria; (b) analyze the institutional context inwhich it operates; and (c) examine potential policy instruments for enhancing its role inincreasing the efficiency, accessibility, and quality o f health services. Inparticular, this chapter will deal with the following issues: types o f services provided; quality and effectiveness o f service delivery; financial issues affecting its development; government's policies and regulation; and potential policy instruments for enhancing its role. The size, scope, and distribution o f the private provision o f services were already described inthe previous chapter. However, to better understand the importance o f the private sector inNigeria, this chapter will summarize some o f the issues discussed in detail in Chapter 3. 2. Private health careproviders in Nigeria are heterogeneous,rangingfrom patent medicine vendors, pharmacies, dental and medical clinics up to tertiary hospitals. Most o f them are registered, but there are also unregistered clinics, drug shops and numerous drughawkers. This chapter will focus mainly on formal health service providers; it will not address practitioners o f traditional medicine and traditional birthattendants. Inour discussion, private providers will be divided into two main groups: (a) health clinics and hospitals and (b) pharmaceuticalretailers. PRIVATE HEALTH CLINICS AND HOSPITALS 3. The Health Facilities Database, HFD, (FMOH,2000) has recorded 9,049 privately owned facilities in2000, corresponding to 38% o f all registered facilities. About 75% o f the private facilities are primary health care facilities, while 25% are secondary health facilities. There i s only one private tertiary health facility, which i s a mission facility. Private facilities comprise 33% o f all primary health care facilities and 72% o f all secondary health facilities. 4. The private health care sector consists o f bothnon-profit who are mainly faith-based service providers, as well as for-profit providers. Some private employers also provide health care through their own clinics. Non-profitproviders 5. NGOs, especiallyfaith-based ones, are very active in the Nigerian health sector and represent a large share of theprivate health providers. For instance, the Christian Health Association o f Nigeria (CHAN) i s the umbrella organization for church-sponsored health care programs. C H A N has about 400 registered member institutions throughout Nigeria that provide services through about 3,500 health facilities, ranging from hospitals to PHC programs and mobile and outreach programs3*(CHAN, 2001;CHAN2004a). InEnugu State it i s openly acknowledged that the mission facilities provide the majority o f health care services (PATHS, 2004a). C H A N institutions collaborate with Muslimorganisations in some areas, e.g. the Inter- faith Forum on HIV/AIDS and Sexual and ReproductiveHealth o f which the Federationo f MuslimWomen o fNigeria is also a member. 6. Under the Nigerian Supreme Council for Islamic Affairs there are also groups working on health promotion, mainly inrelation to sexual and reproductive health and HIV/AIDS, e.g. Aids Programme for Muslims (APMU) and the UMMAHsupport group. There are also other 32 For example the Evangelican Church o f Western Africa has 77 clinics in 11 districts; COCHIN has 75 facilities ranging from health posts to health clinics inPlateau State alone. 92 Chapter 4 independent Muslimorganisations that work inthe same areas, such as the MuslimStudent Society o fNigeria, the Professional MuslimSisters Organisations (Abuja) and Federation o f MuslimWomen inNigeria. Moreover some hospitals or healthcentres havebeenestablishedby Muslimcommunities, for example the MuslimSpecialist Hospital inZaria, Kaduna State. Through the Islamic Medical Association o f Nigeria (IMAN) which has branches inmost o f the Muslimdominated states inthe country, doctors render unpaid services. The Ahmadiyya is active inseveral places, butgenerally the facilities are establishedon anindividualbasis andarenotpart o f a wider organisational network. Information on the number and distribution o f these facilities and the services they provide are currently unavailable. 7. Inaddition to the faith-based NGOs, there i s only anecdotal information on smaller NGOs established by the NigerianDiasporato supply drugs and sometimes services. For-profitproviders 8. The majority ofprivate hospitals and clinics in Nigeria arefor- profit providers, representing about 20% of the total number of facilities in the country. The private for profit providers aim to generate an income for the owner(s), as well as sufficient surplus for replacement o f capital investments. The private for profit sector ranges widely from basic clinics, comprehensive health centres, small hospitals and maternity and convalescent homes, as well as a few diagnostic and specialised centres (eye, dental). Although the FMOHHealth Facilities Data Base didnot distinguishbetweenprivate for profit andnon-profit facilities, it is estimatedthat about 20% o f the 9,049 facilities were private for profit and only 15% were non profit33.Nevertheless, as will be pointed out later, the facilities inthe missiodnon-profit sector tend to be larger than those in the private for profit sector. 9. Limited evidence suggests that solopractices rather than grouppractices predominate in the private,for-profit sector (Ogunbekun et al., 1999;Alub, 2001). This i s a fairly common phenomenon in other countries and i s often attributed to the relatively low capital requirements to start operations at that level. Although there are large variations across states interms of the degree of private health sector activity, evidence from Benue state indicates that private for profit providers tend to be much smaller than the missions. For instance, while 76% o f secondary health care facilities are private for profit, they account for only 8% o f bed capacity inthe state. Incontrast, missions own 17% ofsecondary healthcarefacilitiesbutaccount for 61% ofhospital bed capacity inBenue (International Center for Gender and Social Research, 2004). A similar picture emerges for primary health care: private for profit providers o f primary health care have small shops, i.e. almost half o f all private for profit facilities are at the level o f healthposts and dispensaries. Separate diagnostic laboratories engage 6% o f the private for profit providers. Employer-basedproviders 10. Several large employers have health benefit schemesfor their employees and some also run their own health clinics, e.g. Julius Berger Construction Company, Dantata & Sawoe, Nigerian National PetroleumCorporation. These schemes are generally not set up for generating profit but as a benefit for employees and their dependents. 33 The 3,500 CHAN facilities represent about 40% o f the total private sector. Assumingthat the non-CHAN non-profit services represent a small proportion o f total service providers, we obtain the above mentioned percentages. 93 Chapter 4 Human resources: 11, As mentioned in Chapter 3 a large number of doctors and nurses work in theprivate sector, especially in the Southern states and the Central Belt. For example, inBenue State, there are 173 doctors, o f which 101 (58%) are working inprivate practice. 31o f these doctors are working in private hospitals, 70 doctors are spread across private or mission settings, (International Centre for Gender and Social Research, 2004 12. Many public sector healthprofessionals work in theprivate sector after regular office hours. Inaddition to full-time staffinthe private sector, publicly employedhealthprofessionals are allowed to work after hours inprivate health facilities. 13. Limited evidence suggests that as a consequence of large staff turnover, missionfacilities are more likely to have recently trained staffand a less efficient ratio of clinical to support staffthan publicfacilities. The clinical to support staff ratio o f 1.03 inmission facilities inBenue State appears to be significantly lower than inthe public sector (2.54). Onthe other hand, the percentage o f staff trained duringthe last year was generally muchhigher for mission facilities than for the public sector (PATHS, 2004b). Highstaff turnover i s a problem as it i s difficult to retain staff when government salaries increase and dwindling funds in the mission sector do not allow for similar increases. Lack of staff i s considered a larger problem inrural areas than in urban areas. 14. The limited available information shows large diversity in the state and availability of equipment and infrastructure acrossprivateproviders. Overall information i s unavailableto indicate diagnostic capacity interms o f X-ray and laboratory facilities inthe private sector. The private facilities visited,tend to rely on outsourcing or referral for diagnostics, although basic laboratory facilities were sometimes available. There are, however, wide variations as there are also privately owned hospitals with very advanced equipment, e.g. Abuja clinics offer CT-scan, fluoroscopy, color Doppler, digital ultrasound, dialysis etc. InBenue State, mission facilities were more likely to be well equipped than public facilities, but were still nevertheless in a poor situation. For example, only 48% had a functional fridge, 67% had the necessary OPD equipment, 50% had essential laboratory facilities, etc. (PATHS, 2004b). Mission facilities also have wide variations inthe state o f their equipment and infrastructure. While there are well equipped and well maintained facilities, there are also those where the operationtheaters are inpoor condition and rudimentarily equipped (CHAN, 2002). Similarly a survey o f 134 mission institutions found the state o f repair o f infrastructure to be poor or very poor in 7% o f urban facilities and 20% o f rural facilities (CHAN, 2001). Only 20 % o f the facilities were found to be invery good condition inboth urban and rural areas. Table 37. Distributionof CHAN member institutionsby bed size (%) Beds Urban Rural Total Less than 20 27 50 43 20 40 11 17 15 40 60 -- 9 7 7 More than 60 52 26 34 Total 100 (N=44) 100 (N=90) 100 (N=134) Source: (CHAN, 2001) 15. As mission hospitals tend to have a large number of beds, the capacity of theprivate sector to provide inpatient care might be larger than the number of facilities indicates. The capacity of the private sector depends on both the number o f facilities, as well as the number o fbeds in individual facilities. While no overall national figures are available, figures from Benue State suggest that the private sector could represent a larger source for service delivery than indicated 94 Chapter 4 by the number o f facilities. About 63% o f facilities are privately owned, but (mainly due to some large mission hospitals) account for 69% o f the state's total bed capacity (International Centre for Gender and Social Research, 2004). A survey o f a representative sample o f CHANmember institutions also found these institutions generally to be o f considerable size, especially inurban areas, see Table 44 above (CHAN, 2001). Drugs: 16. Availability of drugs in sufficient quantities and quality is aproblem in missionfacilities as elsewhere (CHAN, 2001; PATHS 2004b). Some mission facilities operate drugrevolving funds. These facilities score 63% on availability o f drugs compared to 46% for facilities without drug revolving funds. C H A N has its own drug supply system, CHANPharm, which was established in 1979 to ensure safe and affordable, mostly generic, drugs to member institutions.It operates as a revolving drug fund.It has been reorganised and i s now autonomous. Although still catering only to C H A N member institutions, it i s considering becoming a supplier to other private providers. CHANpharm also provides training in essential drugs management, sound pharmaceutical practices and rational use o f medicine. Some facilities still buy drugs inthe local market. It i s estimated that 61% o f urban facilities in2001 also sourced drugs from private companies and 11%also sourced from the open market (CHAN, 2001). Inrural facilities, 33% is estimated to have used private companies for purchase o f drugs and 34% also resorted to the open market. CHANphann has, however, beenreorganised since and i s now better able to meet the needs. 17. Anecdotal evidencesuggests that privatefor-proft providers obtain drugsfrom various sources. Facilities visited prefer to get them directly from companies inorder to avoid fake drugs or to rely on a trusted source. 18. TheFMOHprovidesfree drugs in some states. Irregular availability o f TB-drugs i s a particular problem instates where the drugs are not subsidisedby the government or donors. It means that patients have to pay and this results inmany defaulters, especially among the poor. The government also supplies vaccines free o f charge. Records must be submitted prior to receiving the next supply. According to some of the interviewees the process i s sometimes considered very cumbersome and the vaccines are often not available. Services 19. A wide range of curative services areprovided both in NGOfacilities andprivatefor proft facilities. The widest range i s probably for NGOs as one of the CHANmember institutions runs a teaching hospital. An illustration o f the range o f services provided by C H A N member institutions i s outlined in Table 6. A survey inBenue state to assess the capacity for emergency obstetric care found that 70% o fthe EOC was provided by the mission sector. Inthe private for profit facilities visited the set-up was to have a few in-house doctors, together with a whole range o f consultants with various specialties attached to the facility and on-call for specialised cases, thus enablingthe provision o f various primarily curative services. For example, one hospital visited inLagos had 5 inhouse doctors, and 8 specialists working as consultants according to need. 20. Of theprivate-for-profit facilities visited,preventive services comprise only a limited share of theirportfolio, whereaspreventionfigures higher on the agenda of the NGO-sector. For example, most o f the 358 C H A N member institutions are engaged inpreventive services, with 79% offering immunisation services, 50% offering HJY/AIDS prevention, information and counselling, 56% offering family planning services, and 75% offering growth monitoring (CHAN, 2001). The 1999NDHS also shows that basic preventive services like ANC, growth monitoring, immunisation and family planning are available from private doctors, clinics and 95 Chapter 4 hospitals, although to a lesser extent than for government facilities. Inparticular growth monitoring and immunisationsand family planning with around three quarters o f private providers offering them are relatively low compared to the public sector. Table 38. Categoriesof services providedby CHANmember institutions(N=134) Category % Description A 56 Inpatientand outpatient care, communityhealth programmes, primary health care, other speciaiised services such as medicalclinics, pediatrics,tuberculosis treatment, leprosy, obstetric emergencies. B 28 Inpatient and outpatient care, primary health care, including counselling, health education, family planning, immunisations,ante natal and post natal care, minor surgey and accident cases C 4 Outpatient care and communityhealth programmesrun by a general practitioner. Treatment of simple diseases, observationof cases, immunisations,ante natal and post natal care. D 13 Outpatient care run by matrons, nurses,community health worker. Primary health care including immunisations,ante natal and post natal care Note: Member institutions implementthrough a network of facilities, so while 56% as organisations provide extensive services, it does not mean that 56% of all mission healthfacilities providesuch services. Source: (CHAN, 2001) 21. An assessment o f the routine immunisation services inBenue state found that 13% o f immunised children were immunisedina private clinic and that although the private sector does not invest much inimmunisation services there are isolated cases o f consistent good performance (Change Agent Programme and Benue State Ministryo f Health 2004a). N o documentation o f sustained co-ordination o f immunisation activities to improve service delivery was found. There are no annual planning meetings or annual review meetings between the government and private sector to assess performance, lessons learnt and best practices (Change Agent Programme and Benue State Ministry o f Health 2004b). The private facilities offering immunisation are not necessarily part o fthe reporting system. 22. According to the NDHS 2003 the main source o f contraception i s the private sector (cited by 61% of respondents), followed by the public sector (24%) and 15% are other sources like shops, friends etc. Users o f pills and male condoms tend to go to the private sector, whereas users o f injectables and IUDsare more likely to choose the public sector. 23. Inview of the observation that private doctors tend to opt for small scale facilities, it i s not surprisingthat not allprivate providers offer diagnosticservices, butrely on private laboratories and X-ray centres or referral to public hospitals. 24. On HIV/AIDS the C H A N AIDS network and the organisations under the Council o f Islamic Affairs, do some work mainly inthe areas o f promotion, prevention and care. Patients are usually referred to public institutions for tests (CD4 count). Only a few patients can afford treatment in the private sector and are referredto the public sector for treatment (PHRplus and Policy, 2004). In1999 less that 50% ofCHANmember institutionsprovided services inthis area, increasingto about 57% in2003 (CHAN, 2004b). Quality and efficiency 25. There i s paucity o f data on outputs, especially for the for-profit providers, as they rarely file reports on services provided. Thus only little evidence on quality and efficiency o f private health care providers exist. The overall picture i s that irrationaluse o f drugs and a non-functional referral systempresents problems relatedto quality o f care and efficient resource use. 26. An assessment of the quality of the mission sector suggests a goodperformance despite wide variations acrossfacilities. CHANhas started assessing quality and efficiency issues usinga Peer and Participatory Rapid Health Appraisal and Action (PPRHAA) methodology, with a view to identifyinghow to improve the service delivery. Outreachvisits were generally regular and 96 Chapter 4 timely. International standard treatment guidelines were generally applied at lower levels o f care, butpractices and standards varied across doctors at higher levels of care. Availability o fNigerian standard protocols was very low. Most have none, while a few have all o fthem. Overall, staff were perceived to be friendly and patient flow was thought to be good; privacy was generally observed. Records systems especially for outpatients are generally good, although inpatient records need to be improved as a prerequisite for systematic quality monitoring and improvement. There i s also a need to develop systems for clinical audit, patient satisfaction surveys and reporting o f adverse incidents. 27. For example, 58% prescribed antibiotics, 41% prescribed smooth muscle relaxant and 29% prescribed IV fluids as first line treatment inmoderate cases. While 50% o f practitioners would prescribe sugar-salt solution, only 55% o fthem knew the correct recipe. A recent study among 48 private medical practitioners in Calabar on knowledge, attitude and practices on post abortion care indicate that 23% o f them routinely undertake abortions when requestedto do so, and 83% regularly treat women with complications to abortions (Etuk et al, 2000). However, only 18% used the standard procedures for abortions or abortion complications. 28. Irrational use of drugs appears to be a major problemfor bothfor-profit and non-proJit providers. Unnecessary use o f antibiotics for treatment o f diarrhoea and uncomplicated acute respiratory infections and insufficient use o f oral rehydrationtherapy have been reported. Too many drugs are prescribed and the use o f antibiotics and injectables are very high34 (CHAN, 2001). Prescription o f brand drugs i s preferred over generics, partly as a means to overcome problems related to fake drugs. For private providers inneed o f income, a desire to please patients combined with inappropriate patient preferences, often results ininappropriate treatment. 29. There is limited evidence of ineficiencies in theprivate sector based on low occupancy rates. The clinics visited generally had a fairly modest patient flow with average outpatient visits per doctor per day ranging from 2.5 to 10. Inefficiencies also exist inthe mission sector because only 3 out of 20 facilities had a number o f outpatients within the optimum range for OPD visits per capita intheir catchment population; similarly only 4 out o f 20 facilities were within the optimum range for inpatient visits (CHAN, 2001). Most o f the 20 facilities were found to be overstaffed based on a weighted workload index per health worker. Similar results have been found for Benue State (PATHS, 2004b). Furthermore, most hospitals had very low bed occupancy rates with recurrent cost per inpatient day steadily increasing over the past 5 years. While suchresults cannot be generalised, it mightbe worthwhile looking deeper into this issue inorder to determine whether strategies for downsizing facilities are needed. Table 39. Workload estimates (PDE/workday) for healthfacilities in Benue state Service level Publicsector Mission sector Secondary hospitals 2.3 1.1 PHC 9.7 9.7 Other 0.4 1.1 Note: Patient day equivalentsweighting 1 inpatientday at 3 outpatient visits have been calculated. The number of workdays is derivedfrom numberof health workers assuming an average of 15work days per month. Source: (PATHS,2004b) 34 Too many drugs are being prescribed during the OP visits (2-7.1 per prescription against a WHO standardof 1-2), low use of genericdrugs (4647%) and prescriptionsof antibiotics and injections being too high (in 18 out of 20 facilities more than 20% of prescriptions contained antibiotics and in 17 out of 20 facilities more than 10% of prescriptionsincluded injections, in many facilities the average proportion of prescriptionsincludingat least one antibiotic is 91%). 97 Chapter 4 30. Interviews indicated that cross referral between theprivate and thepublic sector is rare and largely dependent onpersonal relationships. There are no well-established practices and procedures for referral and when referral takes place, there i s usually no feedback. However, referral within the mission sector does take place. Private doctors can refer patients to government hospitals, but as they do not have admission rights, the patients have to go through all the procedures again. Insome places, patients do get referred for X-rays and return with X-rays to the private hospital, but this practice does not appear to be systematised, and seems to depend on personal relationships. Financing 31. TheNGO sector relies mainly on donations tofinance their services. Capital costs are almost purely financed through donations. Resource constraints due to declining charity funds for Africa as new target areas for fundinginEastern Europe and Asia emerged, have resulted inlack o f investment inequipment and infrastructure, to the extent that even minor investments cannot be carried out (for example, new boreholes to replace dried ones). 32. Whiledonationsfinance a major part of the recurrent expenditures, the collection of userfees alsoprovides a signijkant source of funds. The mission healthproviders, however, have a high level o f absconders. While some institutionshave funds based on donations to be used to exempt the very poor, these arerarely sufficient and the providersincur deficits. 33. Previousfederal governmentfinancial aid to the mission sector has been reduced or eliminated; however, some states offerfinancial aid. Inthe past the FMOHprovided direct subsidies inthe form o fpayments for the professionally qualified staff inmission hospitals, but thispractice stopped inthe 1990sbecause o fthe needto save on the wage bill.The federal government also used to provide tariff exemptions on imported drugs through CHANpharm but this practice was stoppedthree years ago. However, some state governments support mission facilities through monetary grants or through the payment o f salaries to some o f their staff (CHAN, 2001). It is not known how widespread this practice is. According to the CHAN HIV/AIDS directory 2004 (CHAN, 2004b) 10% o f the facilities infive states35,were receiving state funding for AIDS activities. 34. The establishment ofprivate for-profit clinics isfinancedfrom savings and loans. Access to commercial loans i s limited, but loans from relatives and others are common. There are no direct or indirect subsidies from the federal or state governments to establish or operate private for- profit health facilities. Several o f the clinics visited operate with a slidingfee scale relatedto the patient's ability to pay for services (and the need for care). They give credits and from time to time have to write o f f quite substantial bad debts. Some private clinics have retainership arrangements with large employers to service their employees or with an HMOto service its members. 35. Insome states, disease control programs provide vaccines for child immunizations and TB- drugs free o f charge to private providers who wants to participate inthe programs. Summary 36. Several sources do not provide disaggregated data for-profit and not-for-profit providers. Table 47 i s an attempt to summarise and compare the two types of private health care providers. ~ 35 Imo, Benue, Taraba, Plateau and Kogi States. 98 Chapter 4 Table 40. Comparison of characteristicsof the not-for profitand for profit health care providers. Characteristics Privatefor-profit Private non-profit Objectives Profit-maximising Break-even,social objectives Infrastructure Small scale activities(smallsize hospitals, Larger scale activities relatively more health posts) prevalent (large hospitals, PHC centers) Mostly urban or semi-rural Urban and rural Estimated 23% of all healthfacilities Estimated 15% of all health facilities Human resources Alongside own staff, often relies on personnel Rely on specialised personnel from own from public sector (dualpractice) hospitals Equipment Rely to large extent on referring patientsfor Refer to 'own' hospitals for more than diagnostics basic diagnostics Drugs Tend to prefer sourcingdirectly from drug CHANpharm companies. Govt. supplies vaccines Govt supplies vaccines Services Main focus on curativeservices, although Main focus curative, but also involved preventiveservices are available extensively in preventive and promotive activities. Also provides tertiary services and training Quality and efficiency Quality highly variable; indicationsof poor Quality perceived to be higher: measures quality; incentivesfor irrational use of drugs in place to improve quality, but funding is a problem: training in rational use of Very limited accessto training drugs Fairly low capacity utilisation Financing Access to loans is limited. Rely on charity -a diminishing source, Rely on private borrowing. and user payment Rely on user payment: insurance claims. Some states provide grants Organisation Weak (only 10%organised and organisation Fairly strong, except for Muslim NGOs. thinly resourced) PHARMACEUTICAL RETAIL 37. Pharmaceutical retail takes place ina number o f places. Pharmacies are licensed to dispense prescription drugs and patent medicine vendors (PMVs) are licensedto sell over the counter (OTC) drugs. Inaddition, drugs are soldby drugpeddlers, mobile street hawkers, at market stalls, etc. The following section focuses mainly on legal businesses, although the illegal sale o f drugs i s likely to be considerable inNigeria. 38. Pharmacies: The list o f registeredpharmacists and registered pharmaceutical premises i s updated annually. The total number of registeredpharmaceutical premises by December 312003 was 2,75 1(Pharmacists Council o fNigeria, 2003), a 51% increase from the 1,82 1registered pharmacies in 1992 (World Bank, 1996). 39. Patent Medicine Vendors: The exact number of PMVs is not known, but it is thought to be larger than the number ofpharmacies. A census from 1992 estimated that there were 22,640 patent medicine vendors inthe country, more than 12 times the number o f pharmacies at that time 99 Chapter 4 (World Bank, 1996). In2001, NAFDAC held a series o f meetings for registered PMVs, inwhich more than 36,000 PMVs (at least 13 times the number o f pharmacies) parti~ipated~~. 40. PMVs are more widely available especially in rural areas, thanprimary, secondary, or tertiaryfacilities. The situation inthree rural villages Idere (Oyo State), Ukehe (EnuguState) and Mbaugwu (Abia State) o f approximately 10,000 population each i s illustrative (Salako et al, 2001). Each community has a LGA facility that i s primarily staffed with community health extension workers and that often suffers from drug stockouts. These facilities are closedinthe evening. Furthermore, the communities are rather dispersed and the distance to a facility may be upto 20 km: There are privateclinics intwo o fthese villages; inIdere there is one 7 kmdown the road. Eachcommunity, however, has 4 to 8 PMVs that are open throughout the day and inthe evenings as well. Similarly, inBenue State there are 48 pharmacies and 1108 PMVs (International Centre for Gender and Social Research, 2004). While there i s one primary health care facility for every 10,000 persons and one healthpoddispensary for every 5,000 population, there i s one PMV per 3,250 persons. The study found that most public health facilities are concentrated inurban and semi-urban areas, whereas patent medicine stores and private dispensaries are closer to the rural population. 41. Drug hawkers: Thenumber of drug hawkers (unregistered and illegal vendors of drugs) is also thought to be large. A study undertaken inJigawa State suggests that it i s a widespread business intowns and villages with drug hawkers increasingly conductingtheir businesses in markets, motor parks and along the streets (PATHS, 2004a). Drug hawkers were present in almost all of the markets visited by the researchers and they were particularly prevalent in locations where there were no pharmaceutical chemists. Inputs 42. Interms o f inputsthe pharmaceutical retail sector i s as heterogeneous as the private health clinics. 43. Pharmacies: Thepharmacies visited varied greatly in terms of staflng, as well as scale and quality of infrastructure and equipment. A pharmacy must have a pharmacist incharge (not necessarily as owner). One pharmacy had four pharmacists and 9 support staff as well as a part- time doctor for consultation on prescriptions. At the other end was a pharmacy with one part- time pharmacist and 4 support staff. The quality o f infrastructure and equipment also varied, ranging from small congested and poorly ventilated shops to larger well-organised shops with air- conditioning. 44. Drug supplies are obtained through a variety of channels, including large retail and wholesalepharmacies in major cities andpharmaceutical companies or their representatives. In Jos, several pharmacies specialise inwholesale o f particular products and then buy from one another. Onitsha i s a less expensive place to buy wholesale drugs. There was a general concern about quality o f drugs, which resultedina variety o f coping strategies. Some choose to buy most drugs from one single trusted source, while others choose to buy directly fromthe drug companies and their representatives. One pharmacy used to import drugs directly from Europe, butwith newNAFDACregulations thispracticehas become difficult. 45. Patent medicine vendors:PMVs generally have smaller shops and have less education than pharmacy stafi but they also vary considerably. For example, one shop owner interviewed duringa site visit is a pharmaceutical technicianwho worked ina pharmacy for many years, 36Personal communication Ms.Ijeoma P.C. Nani, DeputyDirector, Technical Assistant to Director General, NAFDAC 100 Chapter 4 before setting up his own shop. The shop i s middle-sized, has air-conditioning to ensure the temperature does not get too high, and i s generally well organised. Incontrast, another PMV interviewedclaimed to be an auxiliary nurse, but was functionally illiterate. The shop was very small, exposed to heat and the dust from the street, and drugs were lying around disorganised. 46. Similar topharmacies, PMVs use several channelsfor obtaining drugs. Drugs can be purchased from a single trusted source as well as directly from the drug companies. It seems as if,PMVswho aremoreconcernedabout quality are also morelikely to choosethe more expensive sources. 47. Drug hawkers: Results from the earlier mentionedstudy inJigawa (PATHS, 2004a) suggest that almost all hawkers lack formal training on drug dispensing, and many o f them are illiterate. Drughawkers inJigawa mainlypurchase fromthe Sabon Gari market inKano (78%) or from local patent medicine stores. Services 48. Even thoughpharmacists are not allowed to diagnose, are only licensed to dispense drugs prescribed by a doctor, and to sell OTCdrugs, inpractice, many of them actually prescribe. The field visits showed that some pharmacies have examinations rooms and that they do diagnostics alongside dispensing o f medicine. This i s partly inresponse to a demand from the patients who do not want to go to the hospital for reasons o f convenience and finance. One pharmacy had a part-time doctor whom they could consult for diagnosis and treatment. One pharmacist divided their services into dealing with three categories o f clients: (1) those who would come with a prescription, (2) those who would present symptoms and ask for advice, and (3) those who diagnosed themselves and who think they know what they need. The pharmacies visited generally agreed on these categories, but while one claimed that almost all customers brought a prescription, another said that almost none had a prescription. 49. PMVs are licensed to sell OTCpatent medicines in original packages; however, they generally sell all types of drugs. For instance they sell generic drugs like paracetamol and chloroquin in large tins, i.e. not pre-packaged, and non-OTC drugs like antibiotic and psychotropic drugs that are outside the scope o f the PMVs' license. A survey among 46 PMVs in 1996 found that 75% stocked prescription drugs like antibiotics (Adikwuet al., 1996). Similarly, a study from Borno State found that PMVs sell prescriptiondrugs and some even prescribe and effectively runclinics (Igun, 1994). This was also observed duringthe field visit. 50. A study (Brieger et al, 2004) undertaken intwo communities inOyo State o f 720 customers in 149PMV shops showedthat --- most customers (79%) simply asked for specific medicines 24.7% presented an illness complaint or a problem only 9.0% had a prescription sheet, but customers inurban areas were twice as likely (13.8%) to present a prescription than their rural counterparts (6.1%) - in69.3% ofcases, the PMV soldthe drugrequested -- - in30.5% ofencounters, the PMV gave hisher own suggestions in 18.6% ofencounters, PMVs askedfor clarificationsandhistory communication about the drugs was observed for 20.8 % o f encounters (dosage, precautions and side effects) - - in4.4% of cases, the prescriptionwas filled 0.4% o f cases was referred to a health facility 101 Chapter 4 Quality and Efficiency 51. Pharmacies: Only anecdotal information i s available on the quality o f pharmacies in the country. For instance, all pharmacies visited claimedto check their shelves for expired drugs. However, at least one pharmacy did not use the last in-last out storage principle. The sales clerks were found generally to have little training, although some learn by experience. For example, in one case, the sales clerks had finishedsecondary school, and the proprietor- having several shops to look after could not be around the shop to give advice on a full-time basis. Such cases could - raise concerns about the quality o f informationand advice given. All pharmacists interviewed were concerned about the quality o f drugs and raised the issue voluntarily. They expressed general satisfaction with the improvements made after the management change o fNAFDAC. The pharmacies are customer-oriented and offer convenient opening hours, inone case from 7 a.m. to 23 p.m. 52. Patent Medicine Vendors: The educational backgrounds of PMVs appear to vaiy considerably. The PMV may have a background as an apprentice under a current license holder or ina pharmacy, be a former or current auxiliary health staff (Jimmy et al., 2000). A study undertaken in 1997-98 on the dispensing o f anti-malarials by PMVs in an LGA inRivers State interviewed40 out o fthe 50 PMVs inthe area (Jimmy et al., 2000). Only 8 had formal training, while 32 hadreceived non-formal or on-the-job training. Only 12 o f the PMVshad knowledge o f the correct dosage for Chloroquine; 23 ofthe PMVs suggested a lower dosage. Another study found that shop clerks and owners consider analgesics containing aspirin to be anti-malarial drugs (Brieger et al., 2002). Combined with poor compliance this could have serious consequences. Under-dosage and non-compliance with full course o f treatment could, however, also be due poor users not being able to afford the complete treatment. 53. Knowledge among PMVs about the medications or the common illnesses of customers is often limited. Ina study o f 720 encounters just under half o f the observed sales took place when a clerk or apprentice was selling (Brieger et al, 2004). While PMVs rarely ask questions about the illness, and vary widely inthe amount, accuracy and quality o f information they give on how to take the medicines, the actual license holders have been found to be better at communicating information and advice to customers than clerks or apprentices. 54. Drug hawkers: Resultsfrom Jigawa suggest that almost all hawkers lackformal training on drug dispensing (PATHS, 2004a). They learn about the effectiveness o f drugs through the prescriptionleaflets, from feedback from customers and from their peers. Irrational and indiscriminate dispensing o f drugs such as providing lower dosages o f antibiotics, wrongly perceivedaction of a drug, as well lack o f knowledge about the importance o f fluidreplacement indiarrhoea illustrate quality problems. However, 62% expressedinterest inreceivingtraining on how to diagnose, prescribe and dispense drugs. The drughawkers seem to recognise the need to protect their business, both interms o f avoiding arrest as well as interms o f maintaining a certain standard, such as not selling expired drugs; they appear to maintain a certain level o f professional self-regulation. As for the drugs themselves, only 13% were registered with NAFDAC. About 5% of drugs sold were expired. Financing 55, All types ofpharmaceutical retail are self;financed, relying solely onpayment by customers to cover operating cost,pay backlinterest on establishment cost and a surplus to theproprietor. There are no government subsidies or other support such as tax incentives. Basedon the interviews with pharmacies and patent medicine vendors it appears that it i s very difficult to 102 Chapter 4 obtain financing to open a pharmacy or a patent medicine shop and that the funds required for ensuring a quality environment for storage o f drugs are not insignificant. 56. It is very d@cult to get a bank loan, evenfor apharmacist wanting to establish a new pharmacy, so many resort to supplementing savings withprivate loans or helpfrom relatives. The financial situation worsens when it takes a long time from the planning and first approval o f the premises to the time o f final approval following a second site visit by licensing authorities. Duringthisperiod, no incomeis generated, but interest on loans would still haveto be paidand basic living costs o fthe pharmacist would still need to be covered. The approval procedure appears to be considerably longer for pharmacies than for PMVs. 57. Limited evidencesuggests that thepricing structure varies depending on the state. InAbuja, the association o f community pharmacist provides guidelines on price margins that range between 35% and 50% depending on distance to town; members generally follow these guidelines. InJos, there appeared to be no common agreement on the price structure, and the mark-up varied from 20-50% depending on type o f drug and from where it was procured. Some pharmacies practice cross-subsidisation, allowing the poorest customers to buy essential medicines at lower rates or on credit to be repaid ininstalments. Not all credit lines are repaid, but many are. 58. PMVs do not seem to have any agreedpricing structure, but theirprices are generally lower than in thepharmacies. Despite the lower prices, PMVs also have customers that find it difficult to pay. The strategies adopted include selling on credit, selling drugs one-by-one rather than as a full-course o f treatment, changing brandname drugs to generics, or adjusting the price to the customer's ability to pay. 59. Drug hawking requires a very small capital outlay and brings a relatively good return (PATHS, 2004a). One o f the reasons for the low capital requirement i s that drug hawkers can often obtain the drugs they sell on a credit basis. They themselves also offer credit to their customers. While a number o f drug hawkers, mainly see their business as an income source, others have started because there was an unserved need intheir community; some were even supported financially by village associations. PURCHASERSOFPRIVATESECTORHEALTH SERVICES 60. Individual consumers: Purchasers of health carefrom theprivate sector are mainly individual consumers. The main source o f income for the private health sector i s user fees, although the non-profit sector also relies to a large extent on donations. While some o f these expenditures relate to services inthe public sector, majority relate to the purchase o f services in the private sector. 61. Individual employers: A study from 1994 found that 4% o frural and 10% o f urban respondents claimed that their medical expenses were covered by their employers through employee medical benefits schemes (Ogunbekun et al., 1999). Some employers have direct arrangements with a specific local clinic or hospital through a retainership arrangement. Few other employers, mainly inthe larger cities, like Lagos and Abuja, engage the services o f health maintenance organisations. Suchbenefit schemes offer an opportunity for active purchasingby the employer on behalf o fthe employees through negotiations o f terms and conditions. However, from the interviews with a few clinics that hadretainership arrangements, such active purchasing did not appear to take place, except when they were engaged by HMOs. 62. HMOs/insurance: Inthe mid-90s, it was estimated that not more than 0.03% o f the populationwas covered by private health insurance (Ogunbekun et al. 1999). There is, however, an emerging market for health maintenance organisations. There are currently eleven managed care organisations operating inNigeria under the umbrella o f the Health Insurance and Managed 103 Chapter 4 Care Association o f Nigeria. Of these, 6 are currently functional, covering an estimated 200,000 memberso fpre-paidhealthplans. One example isTotal HealthTrustLtd,an HMOoperating managed health care services. Total HealthTrust Ltdpurchases services on behalf o f members and have bindingcontracts with a network o f 240 autonomous private clinics nationwide that have been accredited for service delivery to its members. Part o f the contract includes submission o f data for monitoring and evaluation and mandatory participation incontinuing education. Contracts are made with providers on a capitation basis and the basic package to be provided i s negotiated. 63. Government: Thefederal government is not currently buying sewicesfrom theprivate sector. Inthe past, the government had directly and indirectly (reduction induty tariffs on drugs) subsidised the private non-profit health care providers, but these subsidies have been removed. Nonetheless the F M O Hnever had an active purchasing role and no agreement on the services to be delivered seems to have been inplace. 64. Similarly, some state governments still subsidise theprivate non-profit sector, as do some LGAs. For example inJigawa state, LGAs support hospitals with staff, food items, ORT kits, diesel and maintenance (CHAN, 2004a). Inthese cases, however, the support does not seem to be relatedto active purchasing or existing contractual arrangements. INSTITUTIONALCONTEXT Government policy on the sector 65. TheFederal Ministry of Health is currently in theprocess of drafting a new Health Policy, which recognises the importance of theprivate health care sector. This new Health Policy also recognizes the needto develop Public Private Partnership inpursuit o f a service delivery system that utilises resources better, reaches more people, and provides better quality o f care. At present there i s only limited interactionor coordination o f activities between the public and private health sectors. 66. In general, coordination and collaboration throughplanning that is either undertakenjointly or that takes into consideration various stakeholders in thepublic sector, let alone theprivate sector, is lacking. Attempts towards comprehensive sector planning are, however, on-going in some states with support from DFIDthrough the PATHS project. The National H M I S policy and strategic plan from 1996 called for the State HMIS unit to coordinate information flows from boththe public and the private sector, which could enable more comprehensive planning. Compliance inthe private sector varies, but i s generally poor. Lagos State has linked licence renewal to submission o fbasic health information ina new initiative to strengthen data collection, but it is too early to measure its effects. 67. TheNational Health Insurance Scheme (NHIS) gives an important role to theprivate sector. NHIS i s supposed to be a publicly regulatedand private/public driven health scheme. The enabling law3' allows for public and private managed care organisations to use private and public health care providers. A tender for an external company to undertake accreditationassessments o f providers has recently been concluded, and the process i s expected to start soon. Inthe private sector there appears to be some uncertainty regarding who i s on the list to be assessed and why. It i s also not clear to what extent the private sector would be reimbursed. 37 The National Health Insurance Scheme based o n Decree 35 in 1999 is supposed to comprise a national pre-payment plan inwhichparticipants regularly pay a fixed amount for a pre-determined scope o f medical coverage and a social health insurance scheme based on the principles o fpooling of resources, cross- subsidisation, elimination o f adverse selection and solidarity. 104 Chapter 4 Regulation 68. The key roles that regulation can play inthe health sector include control o f quality (specifying standards), control o f quantity (number and distribution o f facilities), and control o f prices. The regulatory mechanism includes legislationas well as other formal or informal rules and may be supported by monetary or non-monetary incentives. 69. InNigeria, as inmost countries there i s basic legislation regarding market entry aimed at ensuring quality o f services. The legislationspecifies the physical characteristics requiredby a facility in order to be allowed to operate and the qualifications and characteristics o f its staff. Professional certification 70. Thequality of the human resources is ensured by semi-autonomous regulatory agencies responsiblefor professional certiJcation. Councils o f professionals are responsible for the regulation o f their respective professions, such as the Medical and Dental Council, the Nurses and Midwifery Council and the Pharmacy Council. The councils are responsible for setting the standards o f training (accreditation o f training institutions), for registration and licensing o f professionals (license to practice, not to open a clinic), for maintaining ethical standards and a few other tasks, e.g. development o f standardized prescription forms etc. The regulatory instruments available are related to the power to issue, renew and withdraw licenses. The license to practice i s given on an annual basis and there are plans to link the annual renewal to continuing education. The proposal i s to make renewal dependant on completion o f at least 30 credit hours o f continuing education (from an identifiedlist o f approved courses) within the precedingtwo years. Lagos state i s also proposingto make renewal dependent on submission o fkey H M I S data. 71, Based on complaintsfrom colleagues and thepublic aprofessional disciplinaryprocess may be initiated although they do not happen often. Firstan investigative panel will determine whether there mightbe a case, if so the case may be brought to a Tribunal (similar to a court) and iftheaccusedisfoundguiltythreelevelsofpenaltiesmaybeapplied:warning,suspensionfora period o f time and withdrawal o f the professional certificate. This does, however, not happen very often. The Medical and Dental Council estimate that there are about 38,000 licensed Nigerian doctors (not all may practice in-country, though). The annual production o f doctors i s approximately 1800. The estimated number o f complaintsreceived i s around 10per year. Inthe past, it has been difficult to put the Tribunal inplace and there has been a 5 year backlog. However, since 1999 the Medicaland Dental Council has been able to more or less cope and in 2003, it dealt with 20 cases (including backlog), o f which 11went to the Tribunal where 5 have been finalized. One or two o f these cases are pending as appeal cases inthe Supreme Court. Licensing private (for-profit) health clinics and hospitals 72. Anybody can apply to start their ownprivatepractice, but certain standards have to be met regarding thephysicalfacilities and qualiJications of the staff (which have to be certij?ed by the relevant Council). Applications are made to the Director o f Health Services at State level. At state level, a Private Health Institution Registration, Licensing and RegulatingCommittee is established with the powers to grant certificates o f registration and license, to inspect health facilities and to suspend, revoke or cancel a certificate o f registration and licensing where it i s deemed necessary and inthe public interest to do so. The Committee should ascertain that the facilities have adequate staff, that site, buildingand general amenities are satisfactory, that equipment are suitable and that the number o fbeds does not exceed the maximum granted. Inspectionshould take place at least twice a year. 105 Chapter 4 73. At state level there are stipulated minimum requirementsfor the registration and licensing of private hospitals and other health establishments.The requirementspertain to - General requirements, mainly regardingphysical facilities (e.g. number and size o f - rooms, water and sanitation, light, sterilizationfacilities) Specific requirements, mainly regardingphysical facilities and staff depending on type o f facility (Outpatient clinics, Inpatient facilities with less than 10 beds, Private hospitals with more than 10beds, Maternity homes, Convalescent homes, Physiotherapycenter, Eye clinic, Medical laboratory services, Radio Diagnostic center) 74. Ingeneral, facilities should be runby professionally certified personnel. There i s a limit as to how far away from the facility the person in-charge or a supervisor can live (10-25 km), as well as to the number of facilities (two) that a person can be incharge o f or supervise. 75. InFCT Abuja, 762 applications have been received since 1989, o f which 252 ended up being licensed. The rest didnot meet the requirements or the applicant changed hisher mind. The licensing procedure requires two inspections, the first to assess the appropriateness o f the building and environment and give the approval to go ahead and establish the facility, the second to assess the appropriateness o f the fully equipped premises and give the approval to hire staff. Finally, based on the appropriateness o fthe staff hireda license i s issued for display inthe facility. 76. The Committee aims to go on weekly inspections, but resources have been constrained. Complaints about the private service providers are receivedfrom the public, but mainly from the general hospitals (where mistreatedpatients end up).On average there are about 2-3 complaints o f unprofessional conduct per year. Television stations broadcast clinic closures. 77. Although there arepenaltiesfor not complying with regulations, both in terms offines and suspension and closure of facilities; thefines are not regularly adjusted and generally appear small compared to thepotential gainfrom running an illegal practice. Inpractice, the committees may not have sufficient resources to be able to adequately perform their functions. A fee i s collected for registration and licensing and there i s an annual renewal fee corresponding to 60-75% o f the registration and licensing fee. The fees vary with size and scope o f facility with OPD clinics payingthe lowest fee and private hospitals with more than 25 bedspayingthe most. 78. There is no regulation of numbers of facilities in a location, no consideration as to the distribution offacilities and noformal regulation on userfee levels. The Nigerian Medical Association, however, has issued a Guide to Standardized Medical Services and Approved MinimumTariffs. Itlists the approved minimumtariffs, below whichno private hospital or clinic must charge. Hospitals and clinics are at liberty to charge higher fees commensurate with their standards and status. However, inexceptionalcases apractitionermay charge fees to socially indigent persons or close family relations at hidher own discretion. The guide further stipulates the minimumcapitation fee and minimumnumber o f clients to be demanded when entering into agreements regardingmanaged care or the national health insurance system. Pharmaceutical retail licensing 79. Thepharmaceutical retail is regulated by the Pharmacy Law, which distinguishes between different types o f licenses. Only pharmacy shops (with trained pharmacists) can dispense prescriptiondrugs.Nigeria has a formal licensingprocedure for persons who want to sell over the counter drugs (OTCs). These PMVs are allowed to sell drugs intheir original pre-packaged form according to a list which include proprietary drugs that are considered safe to sell to the general public and which include common items like pain-killers, cough syrup, anti-malarials, and vitamins. It is, however, widely reportedthat PMVs sell and deal beyond OTC drugs (PATHS, 106 Chapter 4 2004d; Igun, 1994). A survey among 46 PMVs in 1996 found that 75% stocked prescription drugs (Adikwuet al., 1996). 80. According to the Pharmacy Law inNigeria the holder o f a PMV licence i s requiredto be at least 21 years old and to submit the names o f two references. There i s no specifiedrequirement regarding education, butby convention the minimumeducational attainment has been primary school (Brieger et al, 2004). All P M V licenses have to be renewed annually. 81. Patent Medicine Vendors previously had to register with the SMOH. The SMOH, however, had very low capacity and failed to enforce regulations. Some years ago, for purposes o f quality control the registration was transferred to an independent professional body, the Pharmacy Council, who was to be responsible for supervisory functions and capacity building, whereas the SMOHwas to undertake inspections. Now, the Pharmacy Council through State Committees is responsible for renewal o f licenses usingan inspection process and for ensuring compliance with the regulatory standards. Inorder to improve quality, entry requirements have been strengthened. Obtaining a license now requires passing a short test and an interview for literacy and basic knowledge. 82. Some PMVs welcome effective regulation inorder to strengthen the quality o f the profession, others have protested as they do not want to be monitored by the Pharmacy Council contending that the Council has conflicting interests, i.e. to reduce competition for community pharmacies. Court cases have been raised, inwhich the new guidelines were overruled, then new guidelines were developed, which were also challenged. The concern i s that responsibilitiesare not completely clear and the actual responsibility was only taken up in2003 (according to FCDA Abuja). InAbuja, this regulatory issue has resulted in a rapid decline inthe number o f registered PMVsin2003 and it has also createdproblems withregistrations and regulation inJigawa State. Inorder to increaseaccess, there is some pressure towards registeringmore PMVsinrural areas. 83. Thenumber, distribution ofpharmacies and PMVs, and theprices they charge are not regulated. Anybody who meets the requirements mentioned above will get a license. The Association of Pharmacists, however, has a guide on the price margin to be applied. InAbuja, the guidingprice marginvaries with the distance to the city center, with the higher percentage applied at the center. 84. NAFDAC is responsiblefor drugs and medical products registration, regulation, import and export and quality control. As o f January 2004, only products registered with NAFDAC can be sold Privatepracticeby governmenthealthstaff 85, Private practice within government premises i s presently not allowed. Similarly, publicly employedprofessionals are not allowed to own private health care facilities, but there are no restrictions on after-hours work inthe private sector as long as the work inthe public health facility does not suffer. Such dual practice appears to be common, but the exact magnitude i s not known. Enforcement 86. Lack of funds has resulted inproblems to undertake regulatoryfunctions. As already mentioned, it has beendifficult to put together the Tribunal. Moreover the Medical and Dental Council are under-resourced interms o f IT equipment to easily keep track o f the number o f doctors. There are plans to computerize the system to provide easily accessible data as well as to 107 Chapter 4 allow internet access to registration and paper work. There are also plans to create zonal offices to increase accessibility. 87. Similarly, the authority of the regulatory bodies is often not matched by the resources. For example, the Committee inFCT Abuja covering 252 registered private facilities has a chairman, a secretary, a personal assistant, and a messenger. The registration system and monitoring is kept manually as the Committee has no computer. A bus or a larger vehicle has to berequested from the FMOHvehicle pool when going on inspection tour, no separate allocation i s made for running costs. Inprinciple, i s has now been agreed that the office can keep 50% of the fees and fines collected, but it has yet to be practiced. Inorder to speed up the licensing process, the Association o f General and Private Practitioners of Nigeria i s offering to assist the SMOH in inspections. 88. The lack of enforcement results in a lack of respectfor the regulation. For example, a study on abortions foundthat the mainreason for nor undertakingabortions was personal conviction, very few respondents referredto the illegal status o f the operation (Etuk et al, 2000). For 6 communities inLagos State the situation was that "In theory patent medicine and chemist shops are expected to obtain government licenses, but experience has shown that this i s often not done and that the address of the license holder does not often correspond with the location of the shop. Furthermore small kiosks and market stalls that sell medicines are not registered" (Brieger et al., 2002). Incentives 89. At present there are nopositive incentivesfor adhering to regulations or to improve quality. The government does not contract with private health care providers and as such does not provide any rewards for good practice. Similarly, there i s no central policy on provisionof positive incentives for private providers to locate inunder-served areas 90. The emergingHMO sector is applying an accreditationpolicy and monitoring system that is stricter than the government system. An accredited provider will receive a capitationpayment and billingrights for enrolledpatients. The capitation paymentrepresents a secured income and as such it may act as a positive incentive for keeping up the standards for continued accreditation, Conclusionon the InstitutionalContext 91, The basic regulatory system is inplace but its enforcement is weak due tofundingproblems. Basedon the field interviews there appears to be awareness o f the basic regulation. Regulation o f the private health care sector, however, i s weak due to under-funding o fregulatory bodies. Regulation is mainly exercised through the punitive system, but fines appear l o w compared to the potential gains o f breaking regulations and licences are rarely revoked. Positive incentives are rarely, if ever, used. Moreover centralisation makes monitoring difficult and the complaints and disciplinary procedures seem cumbersome. As a consequence, adherence to the regulations may not always be high.While regulation may be weak, it i s not a problem particular to Nigeria, but rather a phenomenon that cuts across Sub-Saharan Africa. 108 Chapter 4 KEY CHALLENGES, CONSTRAINTSAND OPPORTUNITIES Increasing accessibility 92. Although not-for profit health careproviders tend to operate in remote areas and/or areas withpoor populations, thefor-profit providers of clinical care tend to serve relatively wealthier urbanpopulations. Private for profit or self-financing service providers can be expected to set up their shops only inplaces where there i s a sufficient market for sustaining their business costs. 93. A key constraint to using theprivate sectorfor expansion or continuation of services is the lack offinancial viability in rural areas. This issue i s also increasingly affecting the faith-based organisations. Moreover the absence o f a comprehensive policy, guidelines and plans for private health sector development i s a constraint. Finally, the lack o f information about the private health care sector limits effective planning. Increasing efficiency 94. Thereis no coordination between thepublic andprivate sectors. There i s a lack of integration o f service delivery and no strategy for how to use resources across sectors. It may, for example, be possible that existing private non-profit hospitals either purchase services or, as insome countries, use private infrastructure for outreach. The lack o f integration i s reflected inthe lack o f inclusion o f the private sector in sector planning, which to some extent has been made difficult by the lack of data collection on private sector activities. There is, however, some awareness o f this problem and comprehensive planning i s being developed ina few states. Inan attempt to improve data collection, Lagos state has decided to start linkingthe renewal o f licenses with submission o f activity data. 95. The lack of coordination is also reflected in the absence of an efficient and thought through referral policy and referral system. Private sector providers may refer their patients to the government hospitals because they do not have the diagnostic facilities required, but it often means that the patients will have to start over inthe system duplicating even the basic tests. The doctors working inthe private sector do not have admission rightsto the public hospitals. 96. Theprivate sector tends tofocus on curative services, which is where the market is and where profits can be made. Ifpublic healthneeds are to be addressed by private providers this issue may need special attention and possibly contracting arrangements with the State health services. 97. Irrational use of drugs isprevalent in theprivate sector, especially at the PMV level. It creates problems o f quality as well as efficiency. It i s important for optimal resource use and outcome that patients have the proper dosage o f medication appropriate to their condition, for an adequate period o f time and at the lowest possible cost. Ifthe effectiveness o fprivate services i s low, this creates risks o f increased disease transmission and drug resistance. Increasing quality 98. The quality of care offered by theprivate sector is not well-known, but appears highly variable both in the non-profit andfor-profit sub-sectors, in clinical services andpharmaceutical retail. Due to information asymmetries, patients are often not sufficiently qualified to assess the quality o f services. Quality i s often equated with availability o f drugs and short waiting times. Patient demand for drugs and injectables i s sometimes an important contributing factor to irrational drug use. The market for private health care i s atomised and consumers, even when they 109 Chapter 4 are aware o f poor quality, may not have muchpower to demand improved quality o f services. For example, in smaller and remote areas, they may not even have the choice to voice their dissatisfaction by obtaining their healthcare from other sources. The enforcement o f the regulatory system that shouldprotect the consumer and ensure a certain level o f quality i s weak, mainly due to lack o f institutional capacity. The basic regulatory framework i s inplace with fairly good regulation for entry, but it i s more difficult to monitor standards once entry i s gained. 99. Thesystemfor monitoring the activities ofprivate sector providers is notfunctioning. In principle the private sector i s supposed to report to the same HMIS as the public sector, but this rarely takes place. This i s a missed opportunity for getting information useful for planning, assessing service coverage and quality and identifying potential areas for collaboration. 100. Although reportedly increasing, theprivate sector is only to a limited extent, included in training initiatives to improve quality of services, updating treatment strategies and treatment guidelines. CHANhas been organising their own training activities for personnelinfacilities o f member institutions, and have also adopted the Peer and Participatory Rapid Health Appraisal and Action system (PPRHAA) to strengthen quality inclinical work as well as management. Unfortunately they have not been able to continue this due to lack o f funding. However, some states who undertake PPRHAA (withDFD support) have also includedmission hospitals as part o f the process, taking the opportunity to learn from one another. The fairly unorganisedprivate sector has even less access to training and has to mainly rely on pharmaceutical companies for training and guidelines on new developments in diagnosis and treatment. 101. Enforcement of regulation is a major challenge especially in markets with many small providers. A key constraint to improving quality i s the low capacity o f the regulatory bodies. A further constraint isthe focus onthe use o fcontrol to change behaviour andthe total absence of positive incentives. Finally, another great challenge i s the lack o f formal training for patent medicine vendors, who often serve as the frontline healthproviders inmany matters. Opportunities 102. Theplanned NHIS will, in principle, provide an opportunityfor increasing intra-sectoral collaboration. It i s an opportunity for purchasing services from bothpublic and private providers. However, much will depend on how the scheme i s defined and especially what support i s given to various stakeholders to meet the criteria for accreditation. N H I S will need time to develop, but there i s an opportunity to take advantage of the steps towards it to achieve a better understanding of the private sector and how gains inefficiency and quality can be made. 103. Collaboration with theprivate health care sector already exists to some extent in afew disease control areas and in afew states. There i s an opportunity to buildon the good experiences and positive attitude that has been developed inthis context. STRATEGIC OPTIONS 104. Like many other governments, the Nigerian government federal and state) hasfocused most resources on health care services in thepublic sector in order to improve the health of the population, The private sector is, however, significant interms o fnumber o fproviders and number of activities, as well as level o f expenditures. 105. Thechallengefor the government is how the resources available in theprivate sector can be channeled to work towards common objectives and strategiesfor a comprehensive health sector. Three basic issues to consider are whether to: (1) take advantage o f the private providers that already exist, (2) encourage the expansion o f the private sector interms o frange o f services 110 Chapter 4 offered, o f patients or areas served, and (3) shift public activities to the private sector for more efficient delivery o f services. The opportunities for pursuingsome commonly used strategies for private sector involvement inhealth care inNigeria are discussedbelow. Demand side strategies Improve consumer information 106. Quality of care is highly variable and while the consumers may be aware of this, they may not have sufficient knowledge to be able to assess the quality of the advice and treatment they receive. There i s also no easy channel for complaints. Patients are presenting demands, for example for injections, based on wrong perceptions o f effectiveness. Enabling consumers (for example, usingdifferent communication channels such as village meetings and various types o f media) to make more informed choices may improve care seeking behavior interms o f appropriateness and timing. 107. NAFDAC has already started carrying out public awareness campaigns about drug use, quality o f drugs especially the risko f fake drugs, and their potential adverse effects. Based on the concerns raisedby the clients o f the health care providers interviewed for this study about fake drugs, these campaigns appear to have been fairly effective. This work shouldbe continued and it could be explored whether experiences can be used for general health education and social marketingpurposes with a view to makingpatients better consumers. Empowering consumers 108. Consumers can be empowered by implementation of an accreditation system that transparently classipes the hospital or healthfacility according to certain standards. At least, where there are several providers to choose between, this can stimulate competition on quality standards and notjust on price. Currently, an accreditation process i s starting with the NHIS. It i s important that this process i s transparent and that clear criteria are applied. The accreditation process i s an opportunity notjust to provide information for NHIS members on whether certain providers qualify for the scheme, but to make a broader assessment available to the general public. 109. Other strategies include strengthening institutions to give consumers authority to challenge quality/consumer protection, implying afunctional complaints system. Complaints currently have to go through the regulatory bodies inLagos. Not only do these bodies have low capacity, the centralizationmakes it more difficult to access the system for patients who want to file a complaint. Decentralizationo f complaints and disciplinary processes inprofessional practices would therefore be desirable. 110. Strengthen consumer organizations to negotiate prices and quality and assist in individual or collective complaints. Such civil society organizations could obtain bargaining power, for example, by pooling o f funding and management and development o fnegotiating skills. Supply side strategies Training 111. Training and sharing o f information are intendedto improve quality o f case management skills among both clinical staff as pharmacy staff. 112. C H A N offers training for health service providers inPHC, including training o f village health workers and traditional birthattendants, training innutrition, immunization, mother and 111 Chapter 4 child health care, growth monitoring, water and sanitation issues, management o f STIs and HIV/AIDS as well as management training. The resources are, however, limited. 113. None o f the interviewedpharmacists and PMVs have been invited for training or workshops. NAFDAC has, however, started some orientationprogrammes for PMVs and other private sector providers, mainly related to quality o f drugs. Giventhe relative importance o f patent medicine vendors for all users and for the poor inparticular, it seems relevant to focus on strengthening this level o f service in order to improve accessibility o f quality care. Regulationby control 114. The capacity of the regulatory bodies is weak and there is a need to strengthen them in order to effectively enforce regulation. Limitingthe number o f licenses issued by geographical area, for example depending on population could be a way o f spreading private providers especially incombination with positive incentives. However, it could prove to be a challenge to impose such regulation. Regulationby incentivedresources 115. Regulationby incentives i s meant to positively induce change inbehaviour, thereby avoiding the information, administrative and political constraints related to regulation by control. Even inthe private for-profit sector other motives than profit-maximisation are at work. 116. To make locating inunderserved areas more attractive to private for profit providers reducingbusiness costs, increasing expected income, and other incentives are likely to be necessary, although they may be not sufficient conditions. 117. Reducingbusiness costs could be inthe form o f low interest loans for equipment and infrastructure, provision o f office space at a reasonable cost and support to strengthen management. Speeding up or fast tracking licensing procedures for businesses to be set up in underserved areas could also be explored. At the same time limitingthe number o f licenses by geographical areas would limit the number o f more attractive alternatives. 118. Increasingexpected income could be inthe form o f guaranteeing a minimumpayment, providing a tax exemption or giving a base capitationpayment for a clinic that i s set up inan area withno or limitedcoverage. Other incentives could include the provision o f good housing at a reasonable cost. This could be developed incollaborationwith local area councils who would like to work actively to attract providers to improve health services intheir community. Other non- financial incentives could include access to diagnostic and treatment services inpublic health facilities and access to training opportunities. 119. Faith-based organisations often operate inrural areas, and are willing to provide services inhardship areas, ifservice delivery canbe self-sustaining. Expansionor even continuationof NGO services inunderserved areas may also have to be stimulated insimilar ways by subsidising runningcosts, ifthe catchment area cannot sustain operations to break-even or to finance much neededcapital investment. 120. An accreditation system combines pre-defined standards and financial incentives to increase quality. The current accreditation process with N H I S i s an opportunity for developing an accreditation system useful to the whole sector. However, criteria should be reasonable (for example, many private providers inEurope would not be able to meet the criteria proposed inthe preliminary N H I S guidelines) and there shouldbe clear guidance regarding what needs to be strengthened for those who do not qualify inthe first round. Moreover the selection o f these providers should be transparent. Ideally, the accreditation system should eventually be open to all providers. 112 Chapter 4 Service contracts 121. Development of services contracts in which the government paysfor providersfor a specij?edservice of a defined quantity and quality at an agreedprice could be an optionfor mission hospitals especially those operating in undersewed areas. It would, however, probably require a fairly long development process o ftrust and consensus building. Contracting i s a business arrangement that may be more politically viable inorder to take advantage o f the resources that Nigeria has inits NGO network. Contractingwith private providers, for example to deliver services in areas o fpublic health concern, i s an option that could be part o f the basis for the above mentioned `guaranteed' basic payment inunder-served areas. Governance strategies 122. There is a needfor the government to take an increasing stewardship for the whole sector, to act as an active strategicpurchaser with clear priorities regarding where to go and how to use the combined resources available in the health sector. The FMOHi s working on increasing public-private partnership and consensus building.However, because o f the federal system, there i s a limit to how far the FMOHcan go. There i s a need also to strengthen the SMOHs to become active stewards o f the health sector and not see themselves as mainly providers incompetition with the private sector. The situation inthe states varies and some states have already started the process. However, this process often also requires a cultural/attitudinal change that takes time. It is, therefore, important to start as soon as possible, ensuring coordination with the FMOHinitiatives inorder to maximise synergy. 123. Representatives of theprivate sector providers are also important stakeholders. Well- organised professional organisations can play a very important role inself-regulation. One problem when it comes to more formally involving the private health care sector inrelation to contracting, developingincentive systems and changing the regulatory system i s the relative disorganization o f the sector. Insuch a situation it i s less clear with whom one should negotiate and a considerable amount o fresources would be used for negotiating with all individual providers. The private sector organisations inplace have very little capacity and do not have a very large membership. Thus professional organisations have to be strengthened so that they are empowered to effectively play their respective roles inpartnership with the public sector. 113 Chapter 5 CHAPTER 5. HEALTH CARE FINANCINGINNIGERIA 1. This chapter analyzes health careJinancing in Nigeria. Available data on federal, state, and local government spending on the health systemare discussed, as well as fundingby international donors. Recently available survey data on private spending, particularly householdout-of-pocket payments for health services, are analyzed. Overall conclusions on financing patterns are presented and the government's health financing strategy i s discussed. ECONOMIC CONTEXT 2. Nigeriafaces deep-rooted challenges, but government reforms, macroeconomic policy, an oil revenue windfall, and growth in the non-oil economy have contributed to overall economic growth. Nigeria faces considerable challenges from the legacy o f decades o f economic mismanagement, including corruption, poor infrastructure and weak institutions. However, government reforms to combat corruption and improve transparency, as well as macroeconomic policies to manage oil revenues, control inflation and stabilize the exchange rate, have contributed to robust economic growth inthe past few years. Notably, in2004 the federal and state governments successfully set aside a portion o f their windfall inoil revenues. Nevertheless, there i s concern about growth ingovernment spending, as the 2005 government appropriations bill implies a 52% increase infederal spending over the previous year. (IMF, 2005b) Table 41. GDP growth and GDP per capita, Nigeria,2001-2005 2001 2002 2003 2004 2005 real GDP growth (%) 3.3% 1.4% 10.9% 3.6% 7.4% real oil GDP growth (%) 1.4% -11.6% 26.5% 1.9% 12.9% real non-oil GDP growth (%) 4.3% 8.0% 4.4% 4.5% 4.9% real GDP per capita growth (%) 6.0% -13% 7.9% 0.8% 4.6% GDP per capita (US$) 362 341 415 485 582 non-oil GDP per capita (US$) 199 180 201 204 224 Figuresfor 2003 and 2004 are estimatesand those for 2005 are projections. Source is IMF (2005a). 3. Economic growth makes more resources availablefor the health sector. Both oil and non-oil GDPhave grown significantly since 2003, keepingpace with and exceeding population growth (Table 41). Overall estimated GDP per capita has increased from US$ 362 in 1999 to US$ 582 in 2005 (innominal terms). Estimatednon-oil GDP per capita, an indicator o f the resources directly available to ho~seholds,~~has also increased, but at a slower rate, from US$ 199 in 1999 to US$ 224 in2005, Intheory, both economic growth and better government policies and management make more overall resources available for the health sector - from both government and households. PUBLIC SECTOR HEALTH SPENDING 4. TheNigerian government at all levels is greatly dependent on oilproduction revenue. This has both benefits and drawbacks. Onthe benefit side, oil revenues allow the Government to control a significant proportion o f GDP without the difficulties o f collecting taxes from small f i r m s and households. This i s a benefit, o f course, if these resources are applied to productive uses. Among the drawbacks observed inthe case o f Nigeria are the apparent reduction in 38 This i s assumed in this case because much of the value of petroleumproduction i s translated into either revenues for the producersor royalty and tax revenues for the government. 114 Chapter 5 accountability to taxpayers, the greater opportunity for rent-seeking and other unproductive uses of government revenues, and the volatility o f oil revenues. (Sala-i-Martin and Subramanian, 2003). Figure 50 describes trends inreal GDP, total government revenue, and government oil and gas revenue between 1999 and 2005. A decrease inoil prices in2002 caused a decline in overall GDP as well as total government revenue (despite growth inthe non-oil economy), illustrating the volatility o f Nigerian government's main source o f resources. Figure 50. Real GDP, total government revenuesand government oil and gas revenues, Nigeria, 1999-2005 (million Naira at constant 1999 prices) -,b-5,000,000 I 1 total GDP - 4,000,000 Y 5 3,000,000 z 8 s" 2,000,000 other government revenue B g government oil and gas revenue 1,000,000 0 1999 2000 2001 2002 2003 2004 2005 Authors' calculations from IMF estimates. 5. GDP growth, particularly in the oil sector, has increased overall government revenue in real terms, at the same time that government revenue as aproportion of GDP has somewhat increased in thepastfew years. Since 2002, there has been steady growth inreal terms inboth total GDP and government revenue. Government revenue as a proportion o f GDP has fluctuated around 40%; it was 42% in2001, decreasing to around 36% in 2002-03 but increasing again to a projected 45% in2005. 6. TheFederation Account distributes oil and tax revenues between thefederal, state and local governments. Under Nigeria's federal system, the three levels o f government receive block transfers from centrally-collected revenues. The FederationAccount receives revenues from oil production royalties and a domestic value-added tax. Since 1999, oil revenues have accounted for around 80% o f the revenues o f the account. In2003, the allocation formula was 48.5% to the federal government, 24.9% to state governments and 20.6% to local governments. 7. The division of responsibilities for health sewices makes it dfjcult to compile data onpublic sector health spending. As described inChapter 3, under Nigeria's federal system, responsibilities for health services are divided between levels of government. The federal government finances public sector tertiary services, state governments finance public sector secondary hospital services, and local governments support public sector primary health care (PHC) services. The federal government also intervenes at the PHC level through a number o f vertical programs and parastatal agencies. Data on government health spending i s difficult to compile because states and local government authorities (LGAs) are not required to report budgets and expenditures to the federal level. FederalGovernmentHealthExpenditures 8. Although its share is decreasing, thefederal government accountsfor most government spending, while local governments accountfor the least. The large resources available to the federal government are an impetusfor federal involvement in basic sewices. Figure 5 1describes each level o f government's share o f total government expenditure based on data reportedby the Central Bank o fNigeria (CBN). It indicates that in2003, the federal government accounted for 115 Chapter 5 almost 50% o f total expenditures, state governments accounted for about 37%, and local governments for 14%. The federal government's share has decreased somewhat from 2000, mostly to the benefit o f the states.39The large resources available to the federal government are one o f the impetuses behind its involvement inbasic health and education over the years through various programs and parastatal agencies. Figure 51. Federal, state and LGA shares of total government expenditures, Nigeria, 1999-2003 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1999 2000 2001 2002 2003 Authors' calculations from CBN data. 9. Federal government health expenditures have increased in real terms in line with growth in GDP and growth in total government spending. According to CBNreports, federal government health spending has increased from the equivalent o f around US$ 141million in 1998 to the equivalent o f US$288 million in2003 (Table 42).40 Health spending as a proportion o f total federal spending declined between 1998 and 2000, but increased in subsequent years, reaching 3.2% by 2003. Most o f the real growth infederal health spending (around 100%between 1998 and 2003) seems to be the result o f overall growth inGDP and intotal government spending. 10. The three sources o f data on federal health spending available to the authors are similar except for the year 2002. The Central Bank o f Nigeria's figures, reproducedinIMF reports, are shown inTable 42. A National Health Accounts (NHA) study by the University o f Ibadan, supported by WHO, provides similar estimates for the period 1998-2001, but not for the year 2002 (Soyibo et al., 2004.). Expenditure data for 2001 and 2002 compiled by Fagbemi (2004) from reports from the Office o f the Accountant General (cash releases), the FMOH, and individual hospitals and agencies, are similar to the C B Nreports and the NHA study for 2001, but also diverge from the CBN for 2002. For 2002, the University o f IbadanNHA study estimates federal health expenditures at Naira 34,539 million, Fagbemi estimates Naira 36,052 million, and the C B N reports 63,171 million. This results ina range o f estimates for that year equivalent to US$250 to 450 million, or US$2.10 to 3.80 per capita. For 2003, the only source o f data i s the CBN, which reports federal health expenditures o fNaira 39,686 million, equivalent to US$288 million or US$2.35 per capita. 11. The2004 budget suggests aproportional increasefrom theprevious year in health spending but a slight decrease in real terms. The 2004 budget allocation to the FMOH andNational Population Commission accounted for 6.9% o f the total budget or Naira 35,300 million, equivalent to US$265 millionor US$2.10 per capita. However, the overall budget, including the 39IMFestimates indicate ahigher share for the federal government, although the trend is the same: 80% in 1999, 70% in2000,66% in2001,66% in2002, and 59% in2003. 40Parallel (unofficial) exchange rates are used because the divergence with official rates was so great before 2000 that resulting estimates o f health spending inUS$ would not reflect the real purchasing power of health expenditures. 116 Chapter 5 health budget, decreased from 2003 to 2004, so that inreal terms we can expect a slight decline in real health expenditures, even though data on budget releases in2004 are not available to the authors. Inother words, the US$2.10 per capita budgeted in 2004 (with some proportion probably not released) i s slightly lower than the US$2.35 per capita reportedto have been actually spent by the federal government on health in2003. Table 42. Federal government health expenditures,Nigeria, 1998-2003 (in current and constant 2003 prices) 1998 1999 2000 2001 2002 2003 Naira million at current prices 11,984 16,180 18,182 44,652 63,171 39,686 US$ million at parallel exchange rate 141 176 150 333 454 288 Annual growth in federal health 25 (15) 122 36 (37) expenditures in current $US Growth in federal health expenditures 104 in current $US 1998-2003(%) % federal government health 3.3 4.5 2.7 2.8 3.7 3.2 expenditures/ total federal expenditures Federal health expenditures at 2003 19,718 24,969 26,236 54,255 67,999 39,686 prices (naira million), based on the CPI Real annual growth rate in federal 27 5 107 25 (42) health expenditures (%) Real growth in federal health 101 expenditures 1998-2003 (%) Authors' calculations from CBN data. 12. Mostfederal health spending goes to teaching and specialized hospitals andfederal medical centers. Data on the allocationo f funds released for healthby the federal government in2001 and 2002 are presented in Figure 52. In2002, 77% o f federal health expenditures went to federal hospitals - 58% to teaching and specialized hospitals and 19% to the federal medical centers in state capitals. This i s somewhat o f an increase from 2001, when the proportion was 65%, with most o f the difference coming from federal parastatal agencies, whose proportion dropped from 28% in2001 to 14%in2002. These include the parastatal agencies most involved inprimary health care: the National Primary Health Care Development Agency (NPHCDA) and the National Immunization Program(NPI).The NPHCDA received 7% o f total federal health spending in2001and 5% in2002, while the NPIreceived 18% in2001and 6% in2002. Inthese years, the drugregulation agency accounted for around 1to 2% o f the total, while the National Health Insurance Scheme receivedaround 1%. 13, Actual releases of funds for capital spending are less than budgets, while releasesfor recurrent spending exceed budgeted amounts. Like inother countries, much budgeted capital spending does not occur, but recurrent needs are better met. Federal fundreleases for capital investment inthe health sector were 71% o f the budgetedamount in2001 and only 36% in2002. Incontrast, releases forrecurrentexpenditures were 216% ofthebudgetin2001and 116% in 2002 (Authors' estimates from budget data from the Ministry o f Finance and cashreleases data from the Office o f the Accountant General). These patterns indicate a situation were expenditures are determined more by resource availability (tied to inflows o f oil revenues) and the pressures from personnel and clients, to meet salaries and other recurrent costs, than by pre- determined budgets. Such "cash budgeting" has been observed at all levels o f government in Nigeria (World Bank, 2003a). Nevertheless, unlike some other countries, overall releases are not greatly inferior, and may exceed, budgets. Total releases were 129% o f the budget in2001 and 81% in 2002. 117 Chapter 5 Figure 52. Federal governmenthealth expenditures, Nigeria, 2001-2002 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2001 2002 Authors' estimates from data from Office of the Accountant General collected by Fagbemi (2004). 14. Mostfederal health spending is on salaries. In2001 and 2002, between 66% and 75% o f federal health expenditures were devoted to personnel costs, while around 5% covered other recurrent costs (suggesting that most drug costs are borne by patients). About a quarter o f the total was on capital expenditures (Authors' estimates from Office o f the Accountant General data on cash releases). StateGovernmentHealthExpenditures 15. Total state government expenditures are less thanfederal spending and greatly dependent on transfers from the Federation Account. As noted above, estimated total state government spending i s around a third o f total government spending.41 This compares to the federal government's share, which was around halfthe total in2003 (Figure 51). This i s because state governments receive around 25% o f Federation Account transfers while the federal government receives around 50%. Since 1999,60 to 70% o f total state government revenues have come from the Federation Account (CBN, 2005). Several states are exceptions; for example in2000 Lagos and Oyo derived 33% and 44% of their revenue respectively from in-state sources (World Bank, 2003a). 16. Total state government spending has increased dramatically since 1999,although growth has slowed more recently. Figures reportedby the C B N indicatethat total state government spending grew at an annual rate o f over 50% inreal terms between 1999 and 2002. Real growth in state spending has slowed to about 5% in2003 (Authors' estimates from C B N data). 17. Comprehensive data on state health spending are not available. The University o f Ibadan NHA study collected data on state government health expenditures from six states ineach geopolitical zone as well as Lagos and the Federal Capital Territory (FCT) (Soyibo et al., 2004). Extrapolating from this sample, the study estimated total state health expenditures at Naira 6,162 million in 1998, risingto Naira 20,660 million in2002. Another source o f information i s a World Bank study o f state finances which collected data on budget allocations to health in2002 (Table 43). Healthbudgets inthe 13 states, representing around 45% o f the country's population, totaled Naira 26,474 million, which already exceeds the NHA estimate for expenditures in all states. This may be due to differences betweenbudgetedamounts and actual expenditures, butmay also suggest that the NHA figure is an underestimate. 4'Whenreferringto states, the Federal Capital Territory (FCT) is included. 118 Chapter 5 Table 43. State government health budgets, Nigeria,2002 % of total recurrent capital total state budget North-West Kebbi 501 851 1,352 7% Sokoto 628 364 992 5yo North-East Bauchi 796 803 1,598 8% Yobe 118 375 493 3% North-Central Kwara 872 717 1,589 8% South-West Lagos 2,583 650 3,233 6% Ondo 912 1,011 1,923 6% OYO 1,784 960 2,744 10% South-East Imo 892 592 1,484 6% South-South Akwa-lbom 1,764 1,429 3,193 5yo Cross River 863 246 1,109 5% Delta 1,934 994 2,928 5% Rivers 2,162 1,675 3,837 6% total 15,807 10,667 26,474 Sources:World Bank staff estimates and World Bank (2003a). Table 44. Implied average state health budgets by region, Nigeria,2002 per capita per capita state state % of total health health state budget budget budget (Naira) ($US) North-West 277 1.99 5.6% North-East 281 2.02 6.8% North-Central 792 5.70 8.0% South-West 472 3.39 7.1% South-East 461 3.32 6.0% South-SoUth 761 5.48 5.4% Nigeria 505 3.63 6.3% Authors' estimates from data on 13 states in World Bank (2003a). 18. Available data suggest that state health spending represents a low proportion of total state spending. The population-weighted average for the proportion o f total state budgets represented by the healthbudget in2002 inthe 13 states for which data were collectedby the WorldBank study is 6.3%. This is a low proportion when considering that secondary health services are a significant part o f state government responsibilities. Applying this percentage to the CBN reported figure for total state expenditures inthat year o f Naira 724,537 millionresults in an estimate for total state health expenditures o f Naira 45,618 million. This figure can serve as an upper bound for estimated total state health spending in2002, with the University o f Ibadan figure o f Naira 20,660 serving as a lower bound. This range i s equivalent to US$ 150 to 330 million, or US$ 1.25 to 3.75 per capita, and i s o f comparable order o f magnitude to the range of 119 Chapter 5 estimates for federal spending inthat year. For 2003, applying the proportion o f 6.3% to total reported state expenditures results inan estimate for state health spending inthat year o f US$ 58,000 million, equivalent to US$420 million or US$ 3.50 per capita. Like federal spending, state health spending i s likely concentrated on the main area o f state responsibility, secondary hospitals, and is also likely mostly on personnel, although data on this are not available. 19. State government health spending seems to be lowest in the northern regions and highest in the North-Central and South-Southzones. The 2002 state healthbudget data also provide an indication o f regional patterns. Regional estimates impliedby the healthbudget data from 13 states collected by the World Bank state financing study (weighted by state population) are presented inTable 44. Consistent with various health outcome and service indicators, this suggests that state health spending i s lowest inthe northernregions. The highest state spending may be inthe North-Central and South-South zones, where health indicators are better than the north, but not as good as the other southern regions, but where the private sector i s not as strong as inthe South-East and South-West. LocalGovernmentHealthExpenditures 20. Total local government spending accountsfor I0-15% of total government expenditures, is very dependent on transfers from the FederationAccount, and has risen along with oil revenues. Central Bank figures on local government spending indicate that this level o f government accounts for around 10-15% o f total government expenditures (Figure 5 1). This i s in line with the facts that local governments receive around 20% o f transfers from the FederationAccount but do not have any significant source o f locally-generated revenue. Indeed, the share o f local government revenues accounted for by transfers from the Federation Account rose from 70% in 1998 to 80% in2003. Due to growth inthese transfers, based on increases in oil revenues, local government spending has increased inrecent years, for example by 80% inreal terms between 2002 and 2003. 21. Estimating health spending by the 774 local governments is a considerable challenge. The University o f IbadanNHA study collected information on LGA health spending from 7 states and the FCT (Soyibo et al., 2004). Extrapolating from this sample, the study estimates that in2002 local government health spending was Naira 5,013 million. This represents just 3% o f total local government spending reportedby the CBN. The only other data points available to the authors are from a World Bank study o f 30 LGAs inLagos and Kogi states which found that on average, LGAs inLagos allocated 26% oftheir budgets to healthin 1999 and 22% in2000. The averages inKogistate were 14%in 1999 and 12% in2000 (World Bank, 2003b). Applyingthe 2000 percentages to the reported2002 Federation Account transfers to local governments inthese states results in a figure o f Naira 1,850 million, which, for just two states, i s already equivalent to over a third of the NHA estimate for health spendingby all local governments. This suggests that the NHA figure i s an underestimate perhaps due to incomplete administrative data collected in state capitals. Unfortunately details on methodology used inthe NHA report are not available. 22. Inany case, taking the NHA estimate as a lower bound for a range o f estimates for local government health spending in2002, an upper bound could be obtainedby applying the lowest proportion (12%) observed by the Lagos and Kogi study to total reported local government spending. This results ina range o f Naira 5,013 to 20,378 million, equivalent to US$ 35 to 145 million, or US$0.30 to 1.20 per capita. 23. For 2003, the only data available are on total local government expenditures reportedby the CBN, which was Naira 326,400 million, or an almost 70% increase inreal terms fiom 2002. If we again assume that 12% was allocated to health, this results inan estimate o fNaira 41,568 million, equivalent to US$ 300 million or US$ 2.45 per capita. 120 Chapter 5 24. Like other levels of government, most health spending by local governments is onpersonnel. The 2002 study o f 30 LGAs inLagos and Kogi states found that inLagos, 65% o f local government health spending was on personnel, while the proportion inKogi was 78%. InLagos, 19% was for other recurrent costs and 16% for capital expenditure. InKogi, 15% went to other recurrent costs and 7% to capital investment (Khemani, 2004). This indicates that patients bear the bulk o f drug costs. 25. There is a general impression that inpast years, local governments did not receive suficient funding to meet their responsibilities but that the situation has improved recently. Inthe 1990s, local governments experienced the phenomenon o f "zero allocation" due to Federation Account transfers being deducted at source in order to ensure that primary school teachers were paid. The practice o f deduction at source has since been disallowedby the Supreme Court. Central Bank figures on total local government expenditures show that they were low inthe 1990s (even not accounting for deductions at source) but that they have increased dramatically inrecent years. For example, in 1998, reported total local government expenditures were Naira 44,056 million or the equivalent o f US$ 520 million. By 2003, this had increased to Naira 346,400 million or the equivalent o f US$2,5 10 million. This large increase inoverall resources available would suggest a corresponding improvement in local government financing o fprimary health care services. 26. Nevertheless there continues to be concern about governance and the commitment of local governments to basic services. Deduction at source came about because local governments were unable or unwilling to ensure regular salary payment for teachers. There i s evidence o f a similar lack o f commitment to primary health care services duringthis period. A 1995 assessment o f PHC services inNiger State concluded that commitment to health services by LGAs was minimal and reflected ininsufficient releases o f funds to their Health Departments (Niger State Ministryof Health, 2000). A similar assessment inBenue State found that LGAs "have yet to accept responsibility for the sustainability o fhealth services," with regardto meeting non-salary recurrent costs (PATHS, 2003j). There i s also some evidence on corruption at the LGA level. The 2002 study o f health facilities in30 LGAs inLagos and Kogi states found that salary delays were not evident inLagos but more than 42% o f staff had not been paid for more than six months inKogi. Regressionanalysis found that salary non-payment was not associated either withthe total level o f resources available to LGAs or to LGA reportedwage allocations, suggesting "leakage" o f funds intended for wages. (Khemani, 2004). InternationalDonor HealthExpenditures 27. International supportfor the health sector is growing and canpotentially have substantial impact on speciJic health issues. The University o f IbadanNHA study collected data on internationaldonor support to the health sector and estimated that it totaled around US$125 million in2002, which i s a significant increase since 1999 (Soyibo et al., 2004). Considering available information on plannedprojects and grants by the World Bank, Global Fundand large bilateral donors, particularly USAID and DFID, annual commitments by international donors to health and HIV/AIDS are currently reachingUS$200 million. 28. At an estimated US$ 1per capita annually in2002, international donor assistance to the sector i s equivalent to at least one third o f total federal government health spending, and to similar proportions or more o f state or local health spending. Donor-supportedprograms are therefore relatively large inrelation to any one level o f government, and can be influential because they are focused for the most part on specific health issues (such as HIV/AIDS, malaria, family planning or immunization). 121 Chapter 5 TotalPublicSector HealthExpenditures 29. Estimatesfor totalpublic sector health expenditures in 2002 rangefrom US$5to 10per capita, with thefederal government spending the most and local governments spending the least. As discussed above, the University o fIbadanNHA study's estimates for healthspending bythe three levels o f government in2002 are somewhat lower than CBN-reported federal health expenditures and plausible estimates for state and local health spending. Table 45 presents the range o f estimates for 2002. Total domestic public sector health spending i s estimated inthe range o f US$ 3.65 to 8.75 per capita. This range i s 2.9-5.8% o f total government spending and 1.2-2.3% o f GDP. Adding international assistance to the sector bringthe range o f total public sector health spending in 2002 to US$4.68 to 9.78 per capita. Table 45. Range of estimates of public sector healthspending, Nigeria,2002 Naira (million) US$ (million) US$ per capita Federal 26,474 - 45,618 34,539 - 63,171 250 - 450 2.10 - 3.80 State 150 330 0.30 - 1.20 1.25 - 3.75 Local 5,013 - 20,378 35 145 -- total domestic 66,026 - 129,167 435 925 - 3.65 - 8.75 international donors 17,104 123 1.03 Total 83,130 - 146,271 558 - 1,048 4.68 - 9.78 Sources are Soyibo et al. (2004) and authors' estimatesfrom World Bank (2003a) and CBN data. 30. Table 46 presents estimates for 2003, based on CBN-reported federal health expenditures, an assumed 6.3% (fkom 2002 data on the healthbudgets of 13 states) o f total reported state expenditures, and an assumed 12% (from 1999-2000 data on LGA healthbudgets inLagos and Kogi) of total reported local government expenditures. CBN-reported federal health expenditures in2003 are the equivalent of US$2.34 per capita. Applyingassumptions about state andlocal health spending results inan estimate of total domestic public sector health spending o f US$ 8.22 per capita, equivalent to 5.8% oftotal government spending and 2.1% o f GDP. Table 46. Estimatesof public sector health expenditures, Nigeria,2003 Naira US$ per sources and (million) (million) capita assumptions Federal 39,686 288 2.34 1 State 58,033 421 3.42 2 Local 41.568 301 2.45 3 total domestic 139,287 1,009 8.22 international donors 20.850 150 1.22 4 Total 160,137 1,159 9.44 1. CBN 2. 6.3% (from 2002 budget data from 13 states) of CBN reported total state expenditures in2003 3. 12%(from 1999-2000data on LGA budgetsin Lagos and Kogi states) of CBN reported total LGA expendituresin 2003 4. assumed increase from Soyibo et al. (2004) estimate ofUS$ 125 million in2002. Authors' estimates. 122 Chapter 5 31. Plausible estimatesfor government health spending are somewhat at odds with theprevailing impression of very low publicfunding for health services. An estimate o f total annual public sector health spending in2003 o f around US$ 8 i s higher than prevailing impressions. The authors suggest that the NHA figure o f around US$5 for 2002 (presented as the lower range in Table 45) mightbe an underestimate, due to unrepresentative and perhaps incomplete information about state and local government health spending. At the same time, growth intotal government spending in2003 and subsequent years, particularly at the state and local levels, suggests that health spending has grown accordingly. Ingeneral, impressions about government health spending may not have yet caught up with the increases in overall spending o f the past few years. 32. This level of domesticpublic sector health spending is consistent with Nigeria's economic growth andper capita GDP. Figure 53 plots estimated domestic public sector health spending and GDP per capita in2002 among countries with GDP per capita less than US$l,OOO. The US$ 8 per capita estimate for Nigeria in2003 i s situated on the graph (plotted against the IMF estimate o f GDP per capita o f US$415). Nigeria i s in fact slightlybelow the trend-line, indicating that in absolute terms this level o f domestic government spending on health i s consistent with the current size o f its economy. Figure 53. Per capita domestic public sector healthspending and GDP per capita, 2002 (countrieswith GDP per capita less than US$ 1,000) 0 200 400 600 800 1000 GDP per capita (US$) Authors' calculations from WHO 2005 and World Development Indicators 2005. 33. At the same time, it should be remembered that a largeproportion of total domesticpublic sector health spending is allocated to hospitals. One o f the reasons for the general impressionof low government spending for health services i s that international observers are largely focused on primary health care, while most o f the absolute growth in government health spending has gone to hospitals. Federal and state spending are mostly allocated to tertiary and secondary services, accounting for at least two-thirds o f total government health spending. Although total local government spending has increased dramatically, indicating that local government health spending on primary health care services has also increased, the plausible estimate for 2003 presented inTable 46 i s only US$ 2.45 per capita for total LGA health sector spending. PRIVATESECTOR HEALTH SPENDING 34. Privatehealth spending is composed o f direct support for health services byprivate non-profit organizations and for-profit firms, private insurance, and householdout-of-pocket payments. 123 Chapter 5 Private Organizations 35. Spending by private non-profit organizations is likely substantial in some regions. Given the importance o f private non-profit providers in some parts o f the country, particularly hospitals run by religious organizations, capital investments and subsidies financed by these groups are likely to be significant. The University o f IbadanNHA study estimated health expenditures by private non-profit organizations at Naira 6,018 million in2002, equivalent to about $US 45 million (Soyibo et al., 2004).42 36. Some largefor-profit firms andparastatal companies,particularly in Lagos and other large cities, directly provide orfinance health servicesfor employeesand theirfamilies. A survey in the mid-1990s found that 10% o f urbanresidents and 4% o f rural residents reported that medical expenses were covered by their employers. A survey o f 2,75 1private employers inthe 1980s found that employee medical benefits accounted for 6.5% o fpayroll costs. (Ogunbekun, Ogunbekun and Oroboaton, 1999). The University o f Ibadan study, which surveyed parastatals and private firms, estimated such spending at Naira 3,981 million in2002, or the equivalent o f around US$28 million.43 Private Insurance 37. Private health insurance has been very limited and vulnerable to increasing health care costs. A study inthe mid-1990s concluded that around 300,000 people (0.03% o f the population) were coveredby private medical insurance, mostly employee healthbenefit schemes. Increasing costs put pressure on these schemes duringthe last half o f the 1990s. The largest private insurance scheme, with 17,000 beneficiaries, collapsed in 1996. (Ogunbekun, Ogunbekun and Oroboaton, 1999). In2001, it was reportedthat only four private health insurance companies were operating inthe country, with the largest company covering around 18,000 people (Alubo, 2001). 38. However, it is reported thatprivate insurance has been growing in thepastfew years. A more stable economic and political climate may have facilitated a rebound inprivate health insurance as part o f employee benefits and catering to higher-income households. Companies have reportedly started to move away from fee-for-service arrangements in order to control costs, introducing some managed care measures, including capitationand negotiation with health providers on service packages and quality. The University o f IbadanNHA exercise, which surveyed private sector firms and insurance companies, estimated that spending by private health insurance, including employee pooling schemes, totaledNaira 13,836 million in2002, equivalent to aroundUS$ 100 million. The study estimates that spending by private health insurance has grown steadily since 1998, when expenditures were estimated to be around one-third the 2002 level inreal terms (Soybo et al., 2004). 39. Experience with community health insurance is mixed but holds potential. A numbero f experiments incommunity-based health insurance have been implemented inNigeria, including savings schemes through existing community-based organizations such as women's associations, faith-based organizations, and craft and trade groups. A review o f these found that their management capacity, revenue-raising, and negotiation with providers, could be improved, but in general they may have potential for sustainability (Atimet al., 1998). 42The University of Ibadanstudy treatedthis categoryunder internationaldonor spending, so it i s included in estimates discussedinthe previous section. Spendingon healthby non-profit organizationsfromdomestic sources,particularly churches,maybe substantial, but dataare not available. 43Information on methodologywas not providedinthe report. 124 Chapter 5 40. The National Health Insurance Scheme (NHIS) has supported pilot community health insurance programs in several locations and intends to adopt this strategy inthe future to extend insurance coverage beyond formal sector employees. Pilots include village-based mutual health insurance inNiger State and an urban scheme among an association o f shoemakers inAbia State. Inthese pilots, the NHISpays the premiumsfor under-five children, the elderly, and disabled, while association members pay a monthly premium. Inreturn, they receive care from providers who are paid by capitationfor a package o f services. Continuing external subsidies are likely necessary to sustain such arrangements. The government, supported by DFID, intends to undertake a comprehensive documentation and analysis o f existing community financing inorder to inform the N H I S strategy. HouseholdOut-of-PocketHealthExpenditures 41. Recent household survey data indicate that annual out-of-pocket spending on health services exceeds US$20per capita, representing around 9% of total household expenditures. The 2004 Nigeria Living Standards Survey (NLSS) collected data on household health expenditures from a representative sample of 19,159 households. The estimate from these data o f average annual per capita out-of-pocket spending on health i s Naira 2,999, equivalent to around US$ 22.50. The survey data indicate that this out-of-pocket spending on health services accounts for 8.7% o f total household expenditures. This health spending includes expenditure on outpatient care, transportation to health care facilities, and medication^.^^ 42. This is one of the largest shares of health expenditure out of total household expenditure in developing countriesfor which data are available. For instance, incountries such as Vietnam, Brazil, and Albania, householdhealth expenditure represents 7% or less o f total household expenditure (Table 47). InEthiopia, expenditure on healthrepresents only about 1% o f total expenditure; 45 Ethiopia i s however a muchpoorer country where almost all health services are offered by the public sector and where exemptions and waiver schemes exist. InNigeria, the large burden o f health expenditure i s due to highutilization o fprivate providers, to cost recovery bypublic facilities withno clear exemptions and waivers, and insufficient healthinsurance mechanisms. Table 47. Health spending as YOof total household expenditures (householdsurvey data) health spending as % of total GNI per capita household US$ expenditures Vietnam 1992/1993 6% 170 Nepal 1996 3% 210 Brazil 1996-97 5% 4,400 Albania 2002 7% 1,440 Ethiopia 2000 1Yo 110 Nigeria 2004 9% 350 Sources: Vietnam, Nepal, and Brazil come from (Deaton and Zaidi, 2000?), Albania from Albaina Poverty Assessment (World Bank, 2002), and Ethiopia from Ethiopia CSR (World Bank, 2004). The Nigerian estimates were obtained from the NLSS 2004. The GNI per capita for Albania, Ethiopia and Nigeria come from the WDI 2005. 44The estimates exclude data on household expenditures for hospitalization. The NLSS collected data on expenditures for hospitalizationduring the last two weeks. As hospital stays are rare events, it was not possible to annualize these data. 45This is likely to be an underestimation as the survey on which this percentage was calculated did not include some health expenditure. 125 Chapter 5 43. There are large income differences in terms of health expenditures. The richest households on average spend a larger share o f their total expenditure on healthcare than the poorest households (Table 48). This i s not surprising: as seen inprevious chapters, richer households have higher utilization rates than poor ones and they are also more likely to use private providers. On average, a household belongingto the poorest fiftho f the population spendsabout 6.9% o f its income on health care, or 530 Naira (equivalent to around US$4) per capita per year; incontrast, a household among the richest fifth spends about 13.4% o f its income on health care, or 9,200 Naira (equivalent to about US$70) per capita per year. Table 48. Health expenditure as percentageof total householdexpenditure across expenditurequintiles, Nigeria, 200446 Quintiles poorest II Iil IV richest total health expenditure as % of per capita 6.9% 6.7% 7.6% 8.8% 13.4% 8.7% household expenditure per capita annual health expenditure(Naira) 528 957 1,572 2,736 9,200 2,999 share of consultations 16.0% 20.4% 23.2% 21.6% 28.2% 22.3% share of transportation 5.5% 8.8% 8.5% 8.4% 10.1% 8.4% share of medication 78.5% 70.9% 68.4% 69.9% 61.7% 69.3% Authors' estimates from NLSS 2004 data. 44. Health expenditure also varies across regions. Regional differences, though, are smaller than income differences. Inmost regions, households spend 9% o f their total expenditure on health care, close to the national average (Table 49). The only exceptions are households inthe South East and North West. Households inthe South East spend the most on health (about 11%) relative to total expenditure; while households inthe NorthWest spend the least (about 7%). Again, this likely reflects a higher utilization of private providers inthe South East region and lower utilization of private providers inthe North West. Table 49. Health expenditure (excludingexpenditureon inpatient care) as percentageof total household expenditure across regions, Nigeria, 2004 regions South South South North North North South East West Central East West health expenditure as % of per capita household exp. 9.3% 10.9% 8.2% 9.7% 9.5% 6.7% per capita health expenditure (Naira) 3,338 5,488 3,170 2,764 2,426 1,928 share of consultations 20.7% 26.5% 13.3% 27.3% 24.7% 23.7% share of transportation 7.2% 8.1% 4.6% 7.5% 11.1% 11.4% share of medication 72.2% 65.4% 82.1% 65.2% 64.1% 64.9% Authors' estimates from NLSS 2004 data. 45. Expenditure on drugs represents the largest share of household health expenditure. About two thirds o f household per capita health expenditure inNigeria i s expenditure on drugs (Table 48). Pharmaceuticals are readily available throughout the country through pharmacies, patent medicine dealers and street vendors. Similarly, inmany instances pharmacies or patent medicine dealers are the first providers sought in case o f illness. For example, as shown inChapter 2, among children that were taken to a health facility incase of ART or diarrhea, the majority were ~ 46The consumption aggregate used to do these calculations was a modified version o f a preliminary one done by the Federal Office o f Statistics (FOS), the final consumption aggregate was not available at the time this report was produced. A different consumption aggregate might slightly change these results 126 Chapter 5 taken to a pharmacy or patent medicine dealer. Data from the NLSS 2004 further indicate that at least 11percent o f all health care consultations inthe last two weeks took place with a pharmacy or patent medicine dealer. 46. Visits to a health careproviderfor outpatient care represent about 20% of total per capita household health expenditure, and transportation to a health carefacility for outpatient care represents as much as 8%. These shares vary across household income levels and across regions (Table 48 and Table 49). The share o f consultations out o f total household expenditure increases with income quintiles. This reflects lower utilizationrates among poorer households and also lower use o fprivate providers. Table 50. Catastrophic health expenditureacross regions, Nigeria, 2004 (% of households) more than 25% more than 50% of total of total household household expenditures expenditures spent on spent on health health South South 12.9% 4.0% South East 14.5% 6.3% South West 10.3% 4.5% North Central 15.0% 4.8% North East 12.7% 4.3% North West 8.1% 1.8% Total 11.6% 3.9% Authors' estimates from NLSS 2004 data. 47. Many households couldfall intopoverty orfurther into it as a consequence of catastrophic health expenditure. InNigeria as many as 12% o fhouseholds had about a fourth o f their total expenditures on health care and as many as 4% o f households hadhalf o f their expenditures on health care (Table 50). Inthe South East region as many as 6% o f households incur catastrophic health expenditure (health care expenditures comprise more than 50% o f their total household spending). The situation i s alarming as the country i s poor, with non-oil per capita GDP estimated at US$224 in2005 (IMF, 2005a). One o f the main objectives o f a health system i s to assure financial protection against catastrophic health expenditure (Murray & Evans, 2003); in Nigeria the system does not seem to be achieving this goal. 48. On average, Nigerianspay about 655 Naira (equivalent to about US$5)per outpatient consultation; however, this average mash large regional differences in expenditure. The NLSS found that costs o f outpatient consultations inthe South East and South West regions are significantly higher than inany other region inthe country (Table 5 There are n o significant differences inconsultation payments betweenthe NorthEast and North West regions where the payments are the smallest. Regional differences are partly due to a higher utilization o f private providers inthe southern regions, especially inthe South East where 70% o f these consultations took place ina private facility. Incontrast, inthe NorthEast and NorthWest only about 25% o f the consultations took place in a private facility. 47 This difference was statistically significant at a 5% level. 127 Chapter 5 Table 51. Average expenditurefor outpatientconsultationsacross regions, Nigeria, 2004 averagefee Per yoat a private Region outpatient consultation for-profit (Naira) provider South South 618 48% South East 1,007 70% South West 891 54% North Central 665 34% North East 452 26% North West 440 25% Total 655 42% Authors' estimates from NLSS 2004 data. Table 52. Average outpatient consultationexpenditureacross regions, Nigeria, 2004 (Naira) South South South North North North South East West Central East West total Federal 826 1,733 990 652 1,162 1,124 Government (191) (341) (352) (174) (263) (138) State 790 1,328 573 630 409 656 Government (120) (180) (97) (153) (53) (52) Local 364 516 173 234 219 355 Government (84) (97) (55) (24) (43) (42) Private 650 977 1,165 439 390 745 (69) (73) (138) (58) (51) (40) Notel: The NLSS also gave information on consultations in religiousfacilities and others, however as there were not many observations the estimates were not precise and thus they were not included in the table. Note2: Standard errors in parenthesis. Authors' estimates from NLSS 2004 data. 49. Differences in expenditurefor outpatient consultations betweenpublic andprivate providers are not always large. Table 52 presents average outpatient consultationexpenditures across regions and across type o fprovider. It indicates that, except for federal facilities, there i s only a small difference in expenditure between public and private providers. Ingeneral, federal facilities are tertiary hospitals, state facilities are secondary level hospitals, and LGA facilities primary health care facilities. The level o f care available from the private providers i s not defined. There i s no statistical difference in expenditures on outpatient consultations between state facilities and private for-profit facilities. The only exception i s the South West region, where expenditure on outpatient consultations was lower inthe state-owned facilities than in private facilities. On average, expenditure on consultations ina local government facility was significantly lower than consultations in a private facility. The exception was the North Central region where there was no statistical difference inexpenditure on consultations betweenprivate and public facilities. The small differences in expenditure suggest that public facilities are not muchmore affordable than private ones and financial barriers to access are present throughout the country. 50. The higher the level of household expenditure, the higher the expenditure on outpatient consultations. Table 53 presents average outpatient consultationexpenditures by quintiles o f total household expenditures. Households inthe highest quintile spend considerably more than 128 Chapter 5 poorer households. This i s not surprisingas the higher the expenditure level the more likely the consultation took place at a privateprovider (Table 52). Data from the NLSS (not shown) also indicate that the higher the income level, the more likely the consultation took place ina hospital and not ina primary level facility. Table 53. Average outpatient consultationexpenditureacross household expenditurequintiles, Nigeria, 2004 quintile consultation average yoat a private fees (Naira) provider poorest 168 31% II 226 35% 111 303 40% IV 511 42% richest 1.257 49% total 655 42% Authors' estimates from NLSS 2004 data. 5 1. There are small regional differences in transportation expenditures to outpatient health care services and no regional differences in expenditures on medication. Differences intransportation expenditures are only statistically significant betweenthe South South and South West regions and all other regions (Table 54). With regardto medication, on average, Nigerians spend about 500 Naira a month on drugs (equivalent to about US$ 3.75), regardless of the region where they live. Table 54. Average individual expenditureon health care services across regions, Nigeria, 2004 (Naira) Transportation Hospital stay per day Medication South South 191 892 499 South East 341 1,270 507 SouthWest 306 1,378 511 North Central 203 698 415 North East 201 415 474 NorthWest 127 637 458 Total 217 894 482 Authors' estimates from NLSS 2004 data. 52, Thereare large regional differences in hospitalization expenditure. However, as there were not many observations, these estimates are not precise; it i s only inthe North East where the estimates are significantly different from any other region but the South South (Table 55). On average, the survey found that one day o f hospitalization costs 894 Naira, equivalent to about US$6.70. The South East and South West regions have the largest expenditure on hospitalization. The data do not allow for a disaggregation by type o f provider, but it i s likely that the highexpenditure inthese two regionsreflects a large utilization ofprivate hospitals. 53. Thereare large income differences in the average expenditure on transportation, hospitalization and medication; thepoor on average spend much less than the rich (Table 55). These differences are, in general, statistically significant. Although the data do not allow disaggregation, this i s likely the result o f richer households going to more expensive providers - 129 Chapter 5 both private and higher level public services. Inthe case o f average monthly expenditure on medication, these income level differences might also reflect more quantity and better quality o f the products purchasedby richer households. Table 55. Average individualhealth care expenditure across population expenditure quintiles,Nigeria, 2004 (Naira) quintile Transportation hospitalstay per dav Medication poorest 57 196 188 II 85 192 261 ill 103 270 336 IV 167 453 440 richest 412 1286 866 Total 217 894 482 Authors' estimates from NLSS 2004 data. Table 56. Percentageof individualsthat paid for vaccinations,post-natal,or pre-natalcare across household expenditure quintiles, Nigeria, 2004 last vaccination post-natalcare antenatal care poorest 33% 63% 82% II 36% 51yo 79% 111 35% 52% 83% IV 33% 58% 81'Yo richest 35% 54% 81% total 34% 55% 81% Authors' estimatesfrom NLSS 2004 data. 54. There are no uniform or clear exemption programs orfee waiver mechanisms across the countv, evenfor high-impact preventive services. Fee exemptions for specified groups inthe population can reduce financial barriers to care for the poor and reduce the risk o f catastrophic health expenditures. Fee waivers for some types o f health services can encourage utilization o f interventions with highpositive externalities, such as immunization, family planning, and tuberculosis treatment. A 2001 survey o f 202 LGAs found that only 22% o f LGA sampled had fee waiver mechanisms inplace (Adeniyi et al., 2001). Table 57. Percentageof children paying for last vaccinationacross placeswhere the vaccinationtook place, Nigeria,2004 percentage Paying health center 43% hospital 26% private clinic 79% mobile unit 15% school 5% home 10% Authors' estimates from NLSS 2004 data. 55. Data from the 2004 NLSS indicate L a t many patientspay for igh-impact child and maternal health interventions. Table 56 indicates that 34% o f parents paid for their child's most recent vaccination, 55% of mothers paid for post-natal care, and 81% paid for antenatal care - all services which are often officially free o f charge due to their public health impact. Further, there 130 Chapter 5 are no significant differences across household income (expenditure) quintiles, indicating that exemption mechanisms are not successfully distinguishingthe poor from the non-poor. Table 58. Percentage of individuals payingfor post-natalcare consultations across type of facility, Nigeria, 2004 type of facility Percentage oavina federal government 53% state government 53% local government 47% religious 79% industrial 42% private 87% other 51Yo Authors' estimates from NLSS 2004 data. 56. The NLSS collected information on where children had their last immunization. Table 57 indicates that vaccination offered through outreach (e.g. home, school, and mobile units), such as polio immunization, i s more likely to be free. However, immunizations at health facilities, most often routine vaccinations, are not always free o f charge. 57. About halfo f postnatal care consultations require payment to public providers while private providers charge fees for such services inalmost all cases. As seen inTable 58, public as well as private facilities charge for post-natal care. About halfo f the individuals taken to postnatal care consultations in a public facility (federal, state, or local) paid for this service. Inthe private sector the percentagethat paidis much larger, at 87%. Table 59. Estimatesfor private health expenditures, Nigeria,circa 2003-04 Naira US$ per capita Source million million (US$) private organizations 3,982 30 0.24 1 private insurance 13,836 104 0.83 1 private out-of-pocket 377,046 2,835 22.55 2 total private sector 394,864 2,969 23.61 total public sector 160,137 1,159 9.44 3 Total 555,OO1 4,128 33.06 1. Universityof lbadan NHA estimate for 2002 (Soyibo et a/.,2004). 2. Authors' estimatefrom 2004 NLSS data. 3. Authors' estimatefor 2003 from various sources (Table 46). Total PrivateHealthExpenditures 58. Private expenditures on health services are very large. Table 59 presents estimates for private health expenditures, usingfigures fiom the University o f Ibadan's NHA study for expenditures by private organizations and private insurance in2002, added to the 2004 NLSS estimate for per capita out-of-pocket payments in2004. It indicates that private health expenditures could be reaching almost US$ 3 billion per year inNigeria, or more than US$23 per capita. 59. Out-of-pocket payment accountfor 68% of total spending on health services in Nigeria. These estimates, combined with those for public sector health financing in 2003 presented in 131 Chapter 5 Table 46, suggest that private financing accounts for 71%, and out-of pocket payments for 68%, of total health financing inNigeria. Out-of-pocket payments by the poor are considered the most regressive form o f health care financing because the poor's burden o f payments for health services and risk o f catastrophic health expenditures are not shared by the rest o f the population. Figure 54. Private spending as a proportionof total healthfinancing and GDP per capita, 2002 (countries with GDP per capita less than US$ 5,000) 0 1000 2000 3000 4000 5000 GDP per capita (US$) Authors' calculations from WHO 2005 and World Development Indicators 2005. 60. Comparedto other countries, Nigeria is at the high end of the range with regard to the share of total healthfinancing accountedfor byprivate spending. Figure 54 indicates that the share o f total health financing accounted for by private spending decreases as GDP increases. The estimated proportion for Nigeria -71% -i s not inconsistent with other poor countries but it i s at the highend o f the range. TOTALHEALTH SPENDING 61. Total annual health spending in Nigeria is estimated at around US$30per capita, somewhat higher than other countries of similar GDPper capita and mostly made up ofprivate out-of- pocket spending. Followingthe discussion above, we have a lower bound estimate o f about US$ 5 per capita for domestic public sector spending from the University o f Ibadan's NHA study for 2002. An upper bound o f around US$ 8 comes from assumptions based on information about state and LGA healthbudgets from World Bank studies. An additional US$ 1to 1.50 can be attributed to internationaldonor support to the sector. Estimates for health spending by private organizations and insurance are made by the NHA study and total around US$ 1per capita. The 2004 NLSS provides a reliable and representative estimate for private out-of-pocket payments that i s equivalent to around US$22.50 per capita. 62. Thus, we have a lower-bound estimate for total health spending o f the equivalent o f around US$29.50 per capita and an upper-bound estimate o f aroundUS$ 33.00 per capita. These figures represent between 6.5 and 7.4% o f GDP. Compared to other countries, these estimated levels o f spending are somewhat higher than the trend, both in absolute terms and as a proportion o f GDP (Figure 55). 132 Chapter 5 80 - total health spending ($US) and 14 - total health spending (% GDP) and GDP per capita ($US) GDP per capita ($US) 70 - 12 - 0 0 . 0 5i 60 - 0 0 0 10- 0 0 E 3 0 . .-cn50 - 0 5 -2 C 0 & .- P 8 - 0 0 40 - Nigeria 0 . f U a, Nigeria 0 0. / Q 6 - 2 30- Gm 0 5v, -m 0 . 9 0 0 -m -s ,& 0 I 00or, .0 0 3 20 - $ 4 - O . . . 0 I 0 . . 0 . 0 O . 0 . 0 0 eo, *Go 00 0 10 `=r * (I 1 0.027 (0.043) femalehead 0.005 (0.043) Region (South West omitted) North Central 0.071 (0.064) North East -0.059 (0.035)* NorthWest -0.085 (0.048)* South East -0.077 (0.019)*** South South -0.069 (0.025)*** urban (rural omitted) -0.025 (0.032) Access to services (non-self cluster averages) measles vaccination 0.032 (0.083) delivery in health facility 0.081 (0.085) >2 pre-natalvisits -0.025 (0.078) Improveddrinkingwater sources 0.020 (0.048) Improvedsanitation -0.015 (0.055) Observations 6.32e+08 Note: Standard errors are in parenthesis. What is shown is not the coefficient but the change in probability produced by a unit change in the variable. 161 Table 65. Variables associated to the probability that a child between 12and 24 months is immunized. Measles DPT3 Ail polio All vaccines Any vaccines Children's Characteristics female 0.056 0.034 0.037 0.044 0.024 (0.048) (0.026) (0.046) (0.019)'* (0.034) age 0.019 0.007 0.012 0.004 0.004 (0.005)*** (0.004)' (0.006)** (0.002) (0.005) Birth order number 0.008 -0.003 -0.002 -0.004 0.000 (0.009) (0.007) (0,009) (0.004) (0.007) Mother's education (none omitted variable) primary 0.157 0.139 0.023 0.023 0.062 (0.069)'` (0.045)'** (0.072) (0.022) (0.044) higher 0.127 0.029 -0.025 -0.052 0.152 (0.110) (0.073) (0.115) (0.012)'** (0.049)*** Household's characteristics Wealth Index (poorest omitted) Poorer 0.098 -0.009 -0.072 -0.010 0.001 (0.066) (0.044) (0.059) (0.023) (0.046) Middle 0.182 0.029 -0.041 0.035 -0.003 (0.068)'** (0.053) (0.065) (0.038) (0.047) Richer 0.191 0.034 -0.039 0.012 0.010 (0.081)** (0.050) (0.081) (0.029) (0.055) Richest 0.254 0.168 -0.014 0.090 0.123 (0.089)*** (0.087)' (0.097) (0.051)' (0.063)' Number of household members (<6 omitted) 6-10 -0.019 -0.036 0.004 0.008 0.008 (0.046) (0.029) (0.044) (0.016) (0.041) > I1 0.077 0.011 0.139 0.021 0.029 (0.073) (0.044) (0.053)'** (0.030) (0.036) femalehead 0.049 0.031 0.000 0.006 0.019 (0.077) (0.041) (0.081) (0.029) (0.048) Region (South West omitted) North Central -0.024 -0.049 -0.201 0.005 -0.049 (0.078) (0.040) (0.100)'* (0.027) (0.128) North East -0.099 -0.091 -0.221 0.010 -0.059 (0.078) (0.041)** (0.090)'* (0.030) (0.134) North West -0.123 -0.145 -0.157 -0.011 -0.063 (0.081) (0.049)*** (0.095)' (0.030) (0.128) South East -0.043 0.016 -0.170 0.082 -0.233 (0.083) (0.060) (0.107) (0.048)' (0.167) South South -0.017 -0.087 -0.061 0.002 -0.021 (0.085) (0.036)'* (0.098) (0.028) (0.128) urban (rural omitted) -0.092 0.032 0.004 0.012 -0.052 (0.044)** (0.038) (0.046) (0.018) (0.042) Access to services (non-self cluster averages) measles vaccination 1.308 0.319 0.698 0.169 0.794 (0.I,,)'** (0.083)'** (0.I,,),,* (0.053)*** (0.133)'** delivery in health facility -0.116 0.146 0.070 0.030 0.027 (0.105) (0.072)** (0.131) (0.039) (0.100) >2 pre-natal visits 0.003 -0.029 -0.024 0.068 -0.073 (0.085) (0.079) (0.115) (0.038)' (0.091) Observations 1006 I010 1014 1012 1012 Note: Standard errors are in parenthesis. What is shown is not the coefficient but the change in probability produced by a unit change in the variable. * significant at 10%;** significant at 5%; *** significant at 1% 162 Table 66. Variables associatedto the probabilitythatthe child has receiveda supplementationof vitamin A in the last six months (probitanalysis) Children's Characteristics female -0.003 (0.054) age 0.009 (0.001),** Birth order number -0.001 (0.013) Mother's education (none omitted variable) primary 0.277 (0.090),** higher 0.374 (0.176)** Household's characteristics Wealth Index (poorest omitted) Poorer -0.197 (0.112), Middle -0.060 (0.103) Richer 0.286 (0.109),** Richest 0.284 (0.126),* Number of household members(c6 omitted) 6-10 -0.030 (0.068) >I1 0.017 (0.106) femalehead -0.080 (0.094) Region (SouthWest omitted) North Central -0.659 (0.123),** North East -0.661 (0.134)*** North West -0.912 (0.141),** South East -0.303 (0.190) South South -0.400 (0.120),** urban (rural omitted) -0.049 (0.069) Constant -0.865 (0.190)*** Observations 5148 Note: Standard errors are in parenthesis. What is shown is not the coefficient but the change in probability produced by a unit change in the variable. * significant at 10%;** significant at 5%; ***significant at 1% 163 Table 67. Variables associated with the probability of seeking medical care in case a child has had diarrhea or feverkough in the last two weeks (probit analysis) Diarrhea: medical treatment Feverlcough: medical treatment Children's Characteristics female -0.050 -0.019 (0.0287 (0.029) age -0.001 -0.002 (0.001) (0.001y* Birth order number -0.009 -0.020 (0.0057 (0.006)*** Mother's education (none omitted variable) primary 0.053 0.038 (0.049) (0.047) higher 0.405 0.119 (0.I,,),,* (0.136) Household's characteristics Wealth Index (poorest omitted) Poorer -0.017 -0.037 (0.052) (0.043) Middle 0.072 0.061 (0.057) (0.049) Richer 0.170 0.169 (0.074)** (0.063)*** Richest 0.041 0.233 (0.067) (0.079)*** Number of household members (<6 omitted) 6-10 -0.003 0.020 (0.036) (0.038) >I1 -0.003 0.058 (0.048) (0.051) femalehead -0.107 -0.039 (0.029)'** (0.050) Region (South West omitted) North Central 0.042 0.081 (0.074) (0.075) North East -0.217 -0.139 (0.062),,* (0.066)** North West 0.031 0.026 (0.078) (0.075) South East -0.055 -0.155 (0.069) (0.070)** South South -0.096 -0.166 (0.041)** (0.057)*'* urban (rural omitted) -0.006 -0.076 (0.052) (0.0447 Access to services (non-self cluster averages) 0.089 measles vaccination 0.091 0.089 (0.109) (0.100) >2 pre-natal visits 0.097 0.214 (0.095) (0.089)** ObseNations 912 1555 Note: Standard errors are in parenthesis. What is shown is not the coefficient but the change in probabilityproduced by a unit change in the variable. * significant at 10%; ** significant at 5%; *** significant at 1% 164 Table 68. Variables associatedwith the probabilitythat a women will go to more than two ante-natal care visits, be attended by trained personnelduring delivery, and deliver in a healthfacility (probitanalysis). More than 2 antenatal visits Assisted delivery Birth in health facility Woman's characteristics agebirth 0.002 0.001 0.000 (0.002) (0.002) (0.002) Education (none omitted variable) primary 0.115 0.191 0.212 (0.038)*** (0.035)*** (0.031)*** higher 0.174 0.273 0.375 (0.081)** (0.106),** (0.101)*** Household's characteristics Wealth Index (poorest omitted) Poorer 0.022 0.079 0.006 (0.030) (0.039)** (0.037) Middle 0.155 0.080 0.022 (0.033)*** (0.039)** (0.037) Richer 0.240 0.240 0.102 (0.034)*** (0.045),** (0.043)** Richest 0.250 0.317 0.158 (0.051),** (0.054),** (0.052)*** Number of household members (e6 omitted) 6-10 0.012 -0.015 -0.019 (0.029) (0.025) (0.023) >I 1 -0.001 -0.070 0.014 (0.040) (0.039)* (0.041) femalehead -0.040 -0.009 -0.014 (0.053) (0.041) (0.038) Region (South West omitted) North Central -0.100 0.075 0.076 (0.077) (0.045)* (0.031)** North East -0.156 0.082 0.071 (0.079)** (0.047), (0.034)** North West -0.222 -0.029 -0.005 (0.079)*** (0.048) (0.034) South East -0.236 0.264 0.151 (0.104)** (0.057)*** (0.039)*** South South -0.127 -0.017 0.011 (0.077) (0.045) (0.029) urban (rural omitted) 0.015 -0.072 -0.070 (0.036) (0.027)*** (0.020)*** Access to services (non-self cluster averages) measles vaccination 0.593 0.096 -0.001 (0.086)*** (0.051), (0.036) Birth delivered in health care facilities 0.246 1.005 1.081 (0.076)*** (0.061),** (0.048Y** Observations 3443 ' 5156 ' 5169 ' Note: Standard errors are in parenthesis. What is shown is not the coefficient but the change in probabilityproduced by a unit change in the variable. 165