Report No. AAA51 - BJ Republic of Benin Health, Nutrition and Population Health & Poverty Analytical Report May 2009 The World Bank Group Africa Region, Human Development & Ministry of Health, Republic of Benin 1 TABLE OF CONTENTS ABREVIATIONS .........................................................................................................................................3 ACKNOWLEDGEMENT ............................................................................................................................5 EXECUTIVE SUMMARY...........................................................................................................................6 INTRODUCTION.......................................................................................................................................12 Chapter 1: BACKGROUND & METHODOLOGY...................................................................................13 1. Background .........................................................................................................................................14 2. Methodology of the sector review.......................................................................................................27 Chapter 2 : HEALTH INDICATORS .........................................................................................................30 1. General Mortality ................................................................................................................................31 2. Child Health ........................................................................................................................................32 3. Maternal Health...................................................................................................................................54 4. HIV/AIDS ...........................................................................................................................................71 Chapter 3 : COMMUNITIES & HOUSEHOLDS: PRACTICES & POLICIES ........................................75 1. Nutritional practices and policies ........................................................................................................76 2. Hygiene behaviors and access to clean water and to waste disposal equipment.................................83 3. Practices and preferences regarding fertility .......................................................................................87 4. Utilization of health care system .........................................................................................................92 5. Policies for community-based health ..................................................................................................93 Chapter 4 : HEALTH SYSTEM GOVERNANCE & STEWARDSHIP ....................................................95 1. Governance..........................................................................................................................................96 2. Stewardship of the health system ......................................................................................................103 Chapter 5 : HEALTHCARE SERVICES..................................................................................................108 1. Assessing performance of healthcare services ..................................................................................109 2. Explaining performance of healthcare services.................................................................................121 Chapter 6 : HUMAN RESOURCES .........................................................................................................132 1. What is the HRH situation in Benin? ................................................................................................133 2.What are the characteristics of the health labor market in Benin? .....................................................141 3. How to explain these results on HRH ? ............................................................................................150 4. What is the environment for HRH?...................................................................................................157 Chapter 7 : DRUGS, VACCINES AND BLOOD PRODUCTS ..............................................................161 1. Drugs .................................................................................................................................................162 2. Vaccines ............................................................................................................................................176 3. Blood products ..................................................................................................................................182 Chapter 8 : PRIVATE SECTOR ...............................................................................................................185 1. A quickly expanding sector...............................................................................................................186 2. An often illegal and low quality sector .............................................................................................187 3. A legal framework that is difficult to enforce and thus not enforced................................................189 4. The development of the private sector still remains poorly promoted ..............................................190 Chapitre 9 : HEALTH SECTOR FINANCING........................................................................................193 Funding the Benin health system: key aspects......................................................................................194 1. Government funding is still largely oriented towards health care supply .........................................197 2. Household funding of health expenditures without any significant risk pooling mechanisms .........209 3. A highly volatile donor funding ........................................................................................................212 Chapitre 10 : RECOMMENDATIONS ...................................................................................................215 1. More resources for the health system................................................................................................216 2. A better organization of the health system........................................................................................221 Bibliography..............................................................................................................................................228 ANNEXES ................................................................................................................................................231 2 ABREVIATIONS In French In English ACE Agent Contractuel de l’Etat Contract-based Civil Servant APE Agent Permanent de l’Etat Permanent Civil Servant BGE Budget General de l’Etat General Government Budget BM Banque Mondiale WB / World Bank CD Crédits délégués Budgetary allocation to health districts CHD Centre Hospitalier Départemental Regional Hospital Center CHNU Centre Hospitalier National Universitaire National Teaching Hospital Center CNLS Conseil National de Lutte contre le Sida National HIV-AIDS Council CNS Comptes Nationaux de la Santé NHA / National Health Accounts COGEA Comité de Gestion d’Arrondissement (CSA) Management Committee for CSA COGECS Comité de Gestion de Commune (CSC) Management Committee for CSC CPN Consultation Pré Natale Antenatal visit CSA Centre de Santé d’Arrondissement First level primary care center CSC Centre de Santé de Commune Second level primary care center CTB Coopération Technique Belge BTC / Belgian Technical Cooperation CAME Centrale d’Approvisionnement en Central Drug Purchasing Agency Médicaments Essentiels DDS Direction Départementale de la Santé Regional Health Authority DDZS Direction du Développement des Zones Directorate for the Development of Health Districts Sanitaires DIEM Direction des Investissements et des MoH Directorate for Investment and Medical Equipements Médicaux / MS Equipment DMS Durée Moyenne de Séjour ALOS / Average length Of Stay DNPS Direction Nationale de la Politique Sanitaire National Directorate for Health Policy DPP Direction de la Programmation et de la MoH Directorate for Planning and Prospective Prospective / MS DRFM Direction des Ressources Financières et du Directorate for Financial and Physical Resources Matériel DSF Direction de la Santé Familiale / MS MoH Directorate for Family Health EDS Enquête Démographie et Santé DHS / Demography and Health Survey EMICOV Enquête Modulaire Intégrée sur Les Integrated and Modular Survey on household Conditions de Vie des ménages livelihood ENIIEB Ecole Nationale des Infirmiers et Infirmières National School for Registered Nurses diplômées d’Etat du Bénin ENSFEB Ecole Nationale des Sages-femmes du Bénin National School for Midwives FMI Fonds Monétaire International IMF / International Monetary fund FNUAP Fonds des Nations Unies pour la Population United Nations Population Fund FSI Fonds Sanitaire des Indigents Health equity Fund FSS Faculté des Sciences de la Santé Medical School IDA International Development Agency IEC Information, éducation et communication Information, education and communication IHP+ International Health Partnership + International Health Partnership + INSAE Institut National de Statistique et National Statistic Agency d’Administration Economique IREEP Institut de Recherche Empirique en Economie Institute for Empirical Research in Political Politique Economics JLI Joint Learning Initiative Joint Learning Initiative JNV Journées Nationales de Vaccination National Immunization days 3 LNCQ Laboratoire National de Contrôle Qualité National Laboratory for Quality Control MBB Marginal Budgeting for Bottlenecks Marginal Budgeting for Bottlenecks MCZS Médecin Coordonnateur de Zone Sanitaire District Health Officer ME Médicament Essentiel ED / Essential drug MF Ministère des Finances MoF / Ministry of Finance MII Moustiquaire Imprégnées d’Insecticide ITN / Insecticide Treated Net MS Ministère de la Santé MoH / Ministry of Health CDMT Cadre de Dépenses à Moyen Terme MTEF / Medium Term Expenditure Framework ODM Objectif de Développement du Millénaire MDG / Millennium Development Goal OMS Organisation Mondiale de la Santé WHO / World Health Organization ONG Organisation Non Gouvernementale NGO / Non Governmental Organization PAA Plan d’Actions Annuel Annual Action Plan PAR Paiement Aux Résultats RBF / Result Based Financing PIB Produit Intérieur Brut GDP / Gross Domestic Product PIP Programme d’Investissement Pluriannuel Multi-year Investment Program PMA Paquet Minimum d’Activités Essential Care Package PNDS Plan National de Développement Sanitaire National Health Development Plan PNLS Programme National de Lutte contre le Sida National Program against HIV-AIDS PPTE Pays Pauvres Très Endettés HIPC / Highly Indebted Poor Countries PTFs Partenaires Techniques et Financiers Local donors PTS Poste de Transfusion Sanguine Blood Transfusion local facility RGPH Recensement General de la Population National Population Census RHS Ressources Humaines de Santé HRH / Human Resources for Health SCRP Stratégie de Croissance pour la Réduction de Growth Strategy for Poverty Reduction la Pauvreté SDTS Service Départemental de transfusion Regional Blood Transfusion Department Sanguine SG Secrétaire General / MS MoH / General Secretary SIGfip Système Intégré de Gestion des Finances Integrated System for Public Finance Management Publiques SNIGS Système National d’Informations et de Health Information System Gestion de la Santé SONUB Soins Obstétricaux (et Néonataux) d’Urgence BEmOC / Basic Emergency Obstetrical Care (ou SOUB) de Base SONUC Soins Obstétricaux (et Néonataux) d’Urgence CEmOC / Comprehensive Emergency Obstetrical (ou SOUC) Complets Care ORS Sachets de Réhydratation Orale Oral Rehydration Salt TMM Taux de Mortalité Maternelle MMR / Maternal Mortality ratio UEMOA Union Economique et Monétaire Ouest WAEMU / West Africa Economic and Monetary Africaine Union ZS Zone Sanitaire Health District 4 ACKNOWLEDGEMENT This report was prepared by a team composed of Mr. Alphonse Akpamoli (Deputy Director of the MoH Directorate for Planning and Prospective), Mr. Valère Goyito (Deputy Permanent Secretary / Ministry of Health) and Mr. Christophe Lemiere (World Bank / AFTH2 / Task Team Leader). Statistical analyses were made by Ms. Yuko Okamura (Consultant / AFTHD), Mr. Marco Alfano (Consultant / AFTHD) and Mr. Ramsès Zang (Consultant / AFTH2 / Cerdi). The development of the Marginal Budgeting for Bottlenecks was financed by UNICEF and carried out by Mrs. Mariam Sylla (UNICEF / New-York). The pharmaceutical sector analysis was done by HDNHE (Aissatou Diack and Andreas Seiter) and financed by the Norwegian Trust Fund. The team built on the work done by six (06) thematic groups respectively chaired by Mr. Pascal Dossou-Togbé (Governance and Stewardship), Mr. Alfred Koussémou (Financing), Mr. Vincent Faby (Human Resources), Mrs. Julienne Brahi-Dossouvi (Health Care Services), Mrs. Marie- Rose Nago and Mr. Alfred Dansou (Drugs, vaccines and blood products) and Mr. Jacques Hassan and Mr. Joseph Hessou (Environment and nutrition). These groups especially benefitted from analyses and comments by: Pr Eusèbe Alihonou, Mr. Mathias Finoudé, Mr. Lucien Dossou-Gbeté, Mrs. Evelyne Ezin, Mrs. Béatrice Radji, Mr. Alain Amoussouga, Mr. Marius Chanhoun, Mr. Célestin Ganssé, Mr. Vence Agassoussi, and Mr. Septime Hessou. The 6 groups were led and coordinated by a team of national consultants: Mr Emmanuel Guidibi, Mrs. Marie-Paule Guidibi and Mr. Marcellin Koba. Dr Kessilé Tchala, Minister of Health until October 2008, played a key role for the involvement of the various ministries in this approach. Similarly, as he had just taken office, Pr Issifou Takpara, Minister of Health since November 2008, insisted on chairing the final workshop for the adoption of the report (November 2008, Grand Popo). Our appreciation to them for their commitment. Technical and financial partners have largely supported the process, especially by sharing their studies, and more importantly by allowing their staff to contribute to group discussions. Specific contributions were made by: Mr Comlan Comlanvi, Mr Barthelemy Semegan (WHO), Mrs. Dominique Robez-Masson, Mr. Jacques Hassan (UNICEF), Mr. Pascal Zinzindohoue (USAID), Mr. Cyrille Callens, Mr. Christian Abonnel (France), Mrs. Vicentia Glélé, Mr. Remo Meloni, Mr Pierre Lebrun and Mr. Christophe Dossouvi (Belgium). The report had input from: Mrs. Eva Jarawan, Mrs Tonia Marek, Mr. Ayite Fily D’Almeida, Mr. John May, Mr. Vincent Turbat, Mr. Menno Mulder-Sibanda, Miss Chloé Fèvre, Mr. Ibrahim Magazi, Mr. Djibrilla Karamoko (AFTH2), Mrs. Kate Tulenko (AFTHD), and Mr. Boubou Cissé (WBIHD). This study was conducted with the support of the Global Alliance for Vaccines and Immunization (GAVI), UNICEF and the French Ministry of Foreign and European Affairs (MAEE). 5 EXECUTIVE SUMMARY 1. Background (chapter 1) 1. According to the most recent data, poverty rate in Benin is estimated between 37% and 40% of the population. Even though certain methodological precautions are usual practices here, this poverty rate seems to have been stabilized over the last few years. 2. At the demographic level, Benin still has one of the highest fertility rates in the sub region (5.7 children per woman), particularly due to a low prevalence of contraception, which is stagnant since 2001. Thus, in 2001 and 2006, only 17% of the women interviewed were using a contraceptive method. Under such circumstances, the population of Benin may double by 2025. 3. This demographic pressure first of all bears on economic growth by limiting its impact on the average per capita income. This is all the more the case as despite the significant efforts made in terms of economic management (especially in public finance), the GDP growth rate remains below the UEMOA average over the last 5 years. 4. The demographic situation also creates an increasing pressure on the health system, especially as for mother and child health. Among the various Millennium Development Goals for health, children and maternal mortality are indeed the most worrying. 2. Health indicators (chapter 2) 5. Even though there was a significant drop in children mortality during the last few years (probably due to the significant progress in immunization in Benin), it is still on the high side due to 3 factors: malaria, malnutrition and neonatal causes. 6. The high prevalence of children malaria seems to be explained mostly by the low use of bed nets. It is therefore possible that massive distributions of free bed net (with the support of the Malaria Booster of the World Bank and the Presidential Malaria Initiative (PMI) of the United States) will have a significant impact. 7. In terms of children malnutrition, Benin made significant progress, especially in terms of underweight, low birth weight and emaciation. The prevalence of these 3 problems has indeed significantly gone down during the last few years. Nevertheless, several concerns still remain: (i) the large difference between regions in terms of malnutrition, the region of Atakora being in the worst situation; (ii) the still very high and increasing prevalence of growth retardation, probably due to lack of micronutrients and (iii) the high prevalence of anemia. As opposed to malaria, there is a need to develop and more importantly implement a strong policy on malnutrition, a policy which should be multisectoral and community-based. 8. Within children mortality, neonatal mortality is the one which dropped the least over the recent period. This situation is related to the same stagnation noted on maternal mortality. This issue is detailed in the next part. 9. The rate of skilled staff assisted deliveries is usually the main factor driving maternal mortality. This rate is actually high in Benin (78% according to the DHS). With such a rate, 6 Benin should normally have a much lower maternal mortality rate. But this rate is still at the level of 400 deaths for 100,000 births. Other factors such as the quality of care for deliveries obviously play a key role. This poor quality has to do with staff competence and incentives as well as supply in drugs and equipment. 10. In addition to the need to improve the quality of obstetrical care, an increase in skilled staff assisted deliveries still remains an important objective for the reduction of maternal mortality among the poorest. This requires a better financial accessibility for obstetrical care and a better distribution of midwives across the country. 11. Since 2006, there is a particularly relevant policy for the reduction of maternal and neonatal mortality. This policy deserves a massive support. 12. On the other hand, the HIV/AIDS epidemic seems to be under control, at a relatively low level for the region (1.2%). This success is attributable to the very vigorous policies implemented in this area, more particularly at regional level. 13. More generally, the health results obtained in Benin can be explained by two factors: (i) the practices of communities and households, especially in terms of nutritional and hygiene habits, and (ii) health system performance. For these two factors, the main outcomes of the analysis are presented below. 3. Community and household practices and community health policy (chapter 3) 14. We have seen that though nutritional indicators are satisfactory enough in Benin, there remain two major problems with children: (i) the prevalence of growth retardation and (ii) the prevalence of anemia. One may reasonably assume that the first is related to the lack of breastfeeding and zinc supplementation. The second one is due to the poor consumption of food with high iron content, especially by pregnant women. 15. A particular attention should therefore be given to these different points for any multisectoral nutritional policy, policy yet to be defined. One condition for this policy to be developed is to establish a coordination body for nutritional actions in each ministry. 16. In Benin like in most other African countries, 17% of infant mortality is caused by diarrhea, which is closely related to hygiene and the quality of sanitation. But the situation in Benin on that aspect is a serious concern. As a matter of fact, though significant progress have been made in terms of hand washing, the situation seriously deteriorated (i) in terms of access to the water supply system (especially in rural areas where this rate decreased from 28% to 14% between 2001 and 2006 and (ii) in terms of waste disposal (Benin has the poorest results in the region, with Chad, Burkina Faso and Niger). 17. Generally, community-based actions are still not well developed in Benin, apart from the control of some specific disease (the Guinea Worm disease). Each program or department finances and organizes its community-based activities. In addition to the limited integration and irregularity of community-oriented actions resulting from this situation, some activities are relatively neglected (e.g. nutrition), due to lack of support from a program or a department able to finance them. This situation is all the more a pity as a large number of high impact interventions on children mortality can be implemented only as community-based interventions. 7 4. Health system components 18. In the area of institutional governance (chapter 4.1), although Benin was one of the pioneers of primary health care, the country engaged late enough in the deconcentration of its health system (officially in 1995 but actually started only in 2000). The major part of resources is still used by central level (Ministry of Health / MOH) and regional level (RHA), even though deconcentration in favor of health districts is in progress every year. This evolution will clearly be accelerated by the implementation of a really decentralized planning and by a better representation of management committees for health facilities (COGECS). Administrative decentralization should make things easier. 19. Another less usual problem confronted by the MoH is the lack of coordination between the activities of the national directorates and the decentralized levels. An audit conducted in 2008 has thus proposed a new organizational chart for the MoH. 20. As for the stewardship function (chapter 4.2), despite some weaknesses, the health information system is particularly efficient. However, the country is still suffering from a lack of a decentralized planning and insufficient coordination with external partners (Technical and Financial Partners). In addition, within the national health strategy, objectives are poorly prioritized and more importantly do not tally with their forecast impact on health. The capacity of the MoH to advocate with the Ministry of Finance and Donors is therefore seriously limited. 21. The level of utilization of public healthcare services (chapter 5) is still insufficient. Increase in ambulatory service utilization rate is slow and irregular, estimated today (2006) at 44% only. Better rates are recorded for newborns (83%) and pregnant women (91% for the first antenatal care visit). However, the already low hospitalization rate has been decreasing steadily since 2001. These alarming developments are essentially related to the low level of progress achieved in quality of care. The low hospitalization rate also explains the limited technical efficiency of the hospital system, which does not exceed 62%. This means that 38% of the resources (staff and beds) are “wasted”. The healthcare system remains inequitable, especially for in-patient services (in-patients stay, including deliveries). 22. The poor performance of public healthcare services is due to several factors. First of all, there is limited consistency in the healthcare pyramid as illustrated by the weaknesses in the referral system. Facilities are also suffering from imbalances in geographic distribution of health personnel. The high drug prices (in the facilities) do not make these facilities very attractive either. However, in terms of physical infrastructures (i.e. existing buildings), Benin is in a favorable situation, as over 90% of the population live less than 5km away from a healthcare facility. 23. Finally and most importantly, despite a strong commitment of the Ministry to contracting, arrangements for healthcare service management remain unclear. Currently, few facilities have a clear view of the objectives assigned to them. Similarly, payment systems for healthcare services do not relate to funding and actual workload of health facilities. Finally, management autonomy at facilities’ level (especially in terms of financial and human resources management) is still modest. 24. It is therefore desirable to reorganize public healthcare services in order to (i) better manage facilities in terms of results and (ii) grant them a greater autonomy in management. 8 25. Overall, there is a lack of health staff in Benin. However, this is relatively limited. Moreover, the productive capacity of schools can help filling the gap quickly. The Ministry of Health (MoH) can recruit only a small number of these graduates. As a consequence, many graduates end up working in the private sector, which should be fully recognized and integrated into the health care system. 26. Like in most African countries, the health staff in Benin is very unevenly distributed across regions. This situation is worsened by the quasi-absence of mechanisms to improve this distribution. 27. The third and probably most worrying problem is human resources performance, especially low in the public sector. Many factors account for this situation, but the most important ones are obviously the lack of reform in the civil service and the slow pace of fiscal deconcentration. 28. As for drugs (chapter 7), the supply system can be considered as working quite well. The availability rate of essential drugs at health facilities is slightly above 80%, which is still unsatisfactory, but in line with the results achieved by neighboring countries. Disruptions in supplies seem more related to weaknesses in inventory management from health facilities than to the system of procurement and distribution of CAME. This is indeed what has been observed (see chapter 2) for stock outs of obstetrical drugs. 29. In fact, the main obstacle to access to drugs lies in their affordability, as drug prices are obviously very high in Benin and weighing heavily on households (76% of national expenditure on medicines is borne directly by households). Here too, it appears that the problems emanate more from the very high margins (sometimes higher than those of the private sector) charged by public health facilities and less from the prices charged by CAME. This situation is inevitably fuelling a very big illicit market, which has been so far fought very timidly by the government. It is also likely that drugs price at health facilities will not decrease without a radical change in the financing mechanism of such facilities. 30. For vaccines, Benin is one of the model countries in the sub-region, with coverage rates close to 90% and this for many years now. The main challenge of this sub-sector lies in the very weak sustainability of its funding, still largely dependent on Donors and especially on GAVI. Yet it is clear that the efforts of Benin as far as immunization is concerned had a major impact on children health and explains probably much of the decline in children mortality in recent years. 31. Conversely, the blood transfusion sub-sector is poorly organized. The development of a genuine national policy is a first step essential for this activity strengthening. 32. The importance of the private sector (chapter 8), both commercial and faith-based, is growing in Benin, especially in the outpatient sector where it is believed to provide more than 60% of consultations. The faith-based sector (mainly hospitals) is relatively homogeneous and plays an essential role in the Northern part of the country, role which has been anyway recognized by the Government, as it has granted six faith-based hospitals the status of regional hospitals. Conversely, the commercial sector (mostly ambulatory) includes both excellent practitioners and real quacks. The existing legal mechanism to regulate the private sector is not much applied. This can be partly explained by the fact that many health care civil servants practice illegally a private activity (on a dual-job basis), which does not encourage a serious control of this activity. 33. Most of the understaffing and under-performance problems in the public sector could yet be resolved through a better integration of the private sector. In this respect, little progress has been made, except for the above-mentioned public-private partnerships with faith-based hospitals. 9 34. Healthcare system financing (Chapter 9) by the Ministry of Health (MoH) represents currently 9% of the national budget, showing a strong decrease compared to the level of the 90s. However, this level is still significant. The MoH expenses seem to prioritize the most effective and the most pro-poor interventions (prevention and ambulatory). Similarly, significant efforts have been made (and should continue) so as to devote an increasing share of the health budget to health facilities. 35. However, two problems are of particular concern: (i) the low execution rate, a problem plaguing the government of Benin as a whole and more specifically the Ministry of Health (including in its investments) and (ii) the biased use of the Health Equity Fund to the benefit of the richest patients. 36. The financing of the system by households (52% of healthcare spending) remains essential, like in other African countries. Risk-pooling mechanisms are almost non-existent. Despite a long tradition of community health care plans (“mutuelles”), they currently cover only 1% of Benin population. This explains why, each year, more than 2% of households in Benin are confronted with ‘’catastrophic’’ health care expenses (over 40% of their income) and swing into poverty. 37. Finally, external financing (Donors) is quite low (16% of health care spending) and relatively unstable. 10 5. Recommendations (chapter 10) (i) More Targeted Strategies 38. In light of MDGs and the "burden of disease", three health problems seem top priorities in Benin: (i) child malaria, (ii) maternal and neonatal mortality and (iii) child malnutrition. For child malaria, a consistent policy already exists and is being intensively implemented, with substantial support from external partners. For maternal and newborn health, a relevant policy exists also, but its implementation is still very modest. Finally, much remains to be done on child malnutrition, where there is no real and consistent multi-sector policy, or a strategy for developing the community component of the health system, a component yet essential in the fight against malnutrition. (ii) Strengthening the basic clinical and community components of the health system 39. Having appropriate strategies won’t be sufficient. A huge financial effort is required (about 687 billion FCFA over 7 years), as shown in the estimates produced by the MBBs. This effort requires first that the existing budget allocations of the Ministry of Health be redirected towards the three priorities mentioned above. 40. Practically speaking, the administrative machinery of the public sector should be streamlined, so as to clear resources for basic clinical services (i.e. health facilities) and community actions. More specifically, for the health system to be strengthened, it is important to identify aspects of this system that are most in need, namely: (i) geographic distribution and clinical competence of the health staff, (ii) funding of health facilities and drugs, (iii) blood transfusion and (iv) promotion of healthy behaviors (particularly at community level). Finally, such a program funding inevitably involves an increase in external partners support. The preparation of a Compact1, scheduled for 2009 in Benin, is the essential prerequisite to this process. (iii) A better governance 41. Even with more consistent strategies and budget allocations redirected towards clinical and community services, Benin will find it difficult to improve its health indicators if all health stakeholders (i.e. health facilities and health staff) are not more empowered or encouraged to provide quality services to the population as a whole. Therefore, two major transformations in the system governance are to be provided: • In terms of institutional governance : Through the decentralization of the health system, the needs of basic health facilities must be better taken into account (hence the need for decentralized planning) and especially become more autonomous both on financial and administrative aspects. • In terms of individual governance : Health workers should be strongly encouraged to improve their performance (competence, productivity and respect for patients). Given the limited success of inspection strengthening measures and other « top-down» type controls, this incentive can only come from patients, either directly (i.e. bonuses based on cost recovery), or, preferably, indirectly with a mechanism of payment for results (i.e. the number of patients cared for) funded by the Government and possibly by the partners. 1 Under the International Health Partnership (IHP), a Compact is a national strategy document for the entire health sector, approved by national stakeholders and external partners. To date (January 2009), only Ethiopia and Mozambique have produced a Compact. 11 INTRODUCTION 42. The 2008 review had two main objectives. 43. First of all, this review is a situation analysis. In its development strategy, Benin has given importance to the health of its population. This effort is part of the long term vision of the country, described in the Alafia 2025 document. Health improvement, particularly for the poor population, is one of the 8 strategic orientations of this vision. Similarly, on a more operational level, this objective is consistent with the current Growth Strategy for Poverty Reduction (SCRP 2007-2009). Benin is particularly committed to Millennium Development Goals where 3 are related to the health sector. As a “Demographic and Health Survey” (DHS) was conducted in 20062, it was thus particularly relevant to analyze the data available in order to measure the progress achieved by the country, particularly regarding the attainment of these MDGs. This review was also an opportunity to analyze further the constraints related to the health system, in consistency with the new Health Nutrition & Population strategy of the World Bank, a strategy adopted in 2007. 44. However, this exercise did not only seek to be analytic. It also aimed at enriching the policy dialog between stakeholders of the health sector, on the one hand, and on the other hand the World Bank and other development partners. This effort involves more specifically certain topics related to governance3, private sector involvement and harmonization of the efforts of partners (referred to as Technical and Financial Partners or Donors in Benin). From this perspective, this review wants to contribute to Benin efforts toward a progress in the IHP+ Initiative (International Health Partnership Plus). It should be recalled that this initiative is nowadays the main tool for the implementation of the Paris Declaration. Practically, the review made efforts to contribute to the attainment of a consensus between the Ministry of Health and the Donors on the diagnosis of the heath system and on the change required for its strengthening. In addition, the review was in response to a specific request from the Ministry of Health. As a matter of fact, the latter organized in November 2007 a General Conference of the health sector, a meeting gathering more than 600 actors in health sector, to assess the situation of the health system. Several strong orientations emerged from this discussion, especially in the area of governance. Upon request of the Minister, the sector review was expected to use its diagnosis to formulate specific recommendations enabling to implement the orientations adopted during this General Conference. 2 The previous survey dates back to 2001. 3 Here governance means the following. Governance refers the consistency of the incentives of the different actors to achieve objective assigned to them. Incentives include both (i) the objectives assigned, (ii) the powers attributed to achieve these objectives and (iii) the mechanisms for verifying that the assigned objectives have been achieved. This concept of governance applies both to institutions and individuals. 12 Chapter 1: BACKGROUND & METHODOLOGY 1. Background .........................................................................................................................................14 1.1. Physical geography.......................................................................................................................14 1.2. Political situation and administration organization ......................................................................15 1.3. Population.....................................................................................................................................17 1.4. Macroeconomic situation .............................................................................................................21 1.5. Poverty Situation ..........................................................................................................................25 2. Methodology of the sector review.......................................................................................................27 2.1. Analytical framework...................................................................................................................27 2.2. The preparation process...............................................................................................................28 13 1. Background 1.1. Physical geography The Republic of Benin is a West African country, located entirely in the inter-tropical zone between the Equator and the Tropic of Cancer. Stretched in latitude, Benin extends over a surface area of 114 763 square kilometers, bounded in the South by the Atlantic Ocean, in the West by Togo, up North by Burkina Faso and Niger and in the East by Nigeria. Benin’s landscape does not have big differences in level. The average altitude is 200 meters. Only the Atacora chain, with modest dimensions in the North-West of the country is rugged, with an altitude ranging between 400 and 700 meters. Figure 1: Map of Benin (with landscape) Source: http://www.lib.utexas.edu/maps/africa/benin.gif 14 In view of its location in the inter-tropical area, Benin has a hot and wet climate; temperatures and steadily high with an average of 25°C for the whole country. It is in March that they are highest and in August they are lowest. This climatological and geo-morphological profile has a strong influence on the epidemiological profile of Benin. As a matter of fact, it is characterized by a variety of tropical pathologies, with a predominance of endemic and epidemic affections, especially malaria, cholera and encephalo- spinal meningitis. 1.2. Political situation and administration organization Benin is one of the oldest democracies on the continent. Independent since August 1, 1960 under the name of Dahomey, the country adopted its current name in 1975. Ruled for more that fifteen years by a Marxist-Leninist regime, Benin faced in the late 80s a harsh political and economical crisis, particularly as a result of the collapse in the banking system and a drop of Government incomes. This crisis led, in 1990, to a deep policy and administrative reform characterized by the adoption of economic liberalism and a democratic structure. Since December 11, 1990, the country has a Constitution which made provision for a presidential regime and an assembly elected through universal suffrage. Moreover, this constitution warrants human rights and gradual construction of a rule of law. The administrative organization is being decentralized. At administrative level, Benin includes 12 regions since January 15, 1999, in accordance to Law No 97-028 relating to organization of the Republic of Benin. They are: Alibori, Atacora, Atlantique, Borgou, Collines, Couffo, Donga, Littoral, Mono, Ouémé, Plateau, and Zou. These regions are divided in 77 communes among which three have a particular status including: Cotonou, Porto-Novo and Parakou. The 77 communes are subdivided in 546 arrondissements composed of 3743 villages and neighborhoods, the village being the smallest administrative unit in a rural constituency exactly as the quarter is for the urban areas. A process for decentralizing the overall public administration is currently ongoing. 15 Figure 2: Administrative map of Benin Source: Health Statistic Directory 16 1.3. Population The Population in Benin is growing fast. When compared to other countries in West Africa (see graph below), Benin is still presenting one of the highest natural growth rates in the sub- region (+2.96% per year as against 2.64% for the average in West-Africa). This situation is largely related to the high fertility rate of the population. This rate is estimated to be 5.7 children per woman (2006 DHS). Even though this rate is lower than the one recorded in 1996 (6.3) and 1982 (7.1), it is however almost at the same level as the one of 2001 (5.6). We may then wonder whether the demographic transition in Benin was not “stopped” since the last years. This high fertility situation also explains the peculiar shape of the population pyramid in Benin, which is very extended at the base (cf. graph below). Figure 3: Natural growth rate and fertility in various West-African countries 8.00 7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 RI e RI S T N e ia Se E de G al au E d' o na Bi r G n ia au r ia SA ’OU a a o li e AF oir on g ni rk Ma ne as N M QU bi an er g ne Nig Cô To er be ha ss HA E EN ne Be am Iv Le ig SU DE ui Si a F rit V G Li e ra a- in te ap L er C Bu ui G UE UE B- Q RI Taux de Taux de fécondité AF croissance IQ FR naturelle A Source: World Development Source 2008 Figure 4: Population pyramid in Benin (2002) 17 This high fertility is likely to generate a faster growth of the population, thereby imposing a serious economic and social burden on the development of the country. According to UN forecasts (see graph below), the lack of reduction in the fertility rate (an hypothesis consistent with existing data) may induce a very fast growth of the population, moving from approximately 8.4 million in 2008 to more than 16 million in 2025 (date retained by vision Alafia). Considering the lack of adequate educational and health capacities, one cannot see clearly how the country can sustain a doubling of the population and therefore of the demand. Figure 5 : Population forecasts for Benin 45000 40000 Population totale (en milliers) 35000 Medium variant 30000 High variant Low variant 25000 Constant-fertility variant 20000 15000 10000 5000 0 50 55 60 65 70 75 80 85 90 95 00 05 10 15 20 25 30 35 40 45 50 19 19 19 19 19 19 19 19 19 19 20 20 20 20 20 20 20 20 20 20 20 Source: UN population division If nothing is done to reduce the fertility rate, one of the highest in the sub-region (5.7 children per woman), Benin may witness a doubling of its population by 2025, which will necessarily limit the capacity of the country to develop. Firstly, it is not sure that the ecosystem of the country will be able to accommodate a growing concentration of the population. Secondly, the educational system is subject to an increasing pressure, while it is facing nowadays difficulties to assume its basic mandates. For example, just to mention the children aged between 6 and 11 years, their number would increase from 1.49 million in 2007 to almost 3 millions in 2025. Before accommodating an increase (recommendable) of schooling rates, we can see that the capacities of the educational system (with regard to the number of classrooms and the number of teachers) should double during the 25 coming years. Similarly, such a population growth will put pressure on the health system to allocate more and more resources for children and pregnant women. To take the example of deliveries assisted by qualified staff, it is likely that Benin has to double the capacity of its health system, which specifically means a doubling of the number of midwives by 2025. It is undeniable that the productivity of the staff is quite low today (as will be seen in chapter 6). Ways to improve this productivity are also to be explored. In all cases, the population burden will push Benin to increase considerably its investments in terms of human capital and to improve radically the efficiency of the current systems. These various challenges seriously limit the capacity of the 18 country to reduce poverty. Chapter 3 explores in detail the causes explaining the high fertility level in Benin and the solutions that may help reducing it. When compared to other West-African countries, Benin is characterized by a very high population density and an urbanization rate near the average. According to INSAE’s forecasts (base on the 2002 General Population Census), the Benin population in 2006 was 7,840,000. As indicated in table below, the UN estimates are rather close to 9,300,000. The population is slightly denser than the average for other countries (see graph below), as the population is largely concentrated along the coast (see graph below). Table 1: Demographic data for various West-African countries % % de la de la Pourcentage Population Taux de population population Esperance de la total Taux de migration âgée âgée de vie population en milliers croissance nette Taux de de moins de de 65 ans et a la en milieu (2008) naturelle pour 1,000 fécondité 15 ans plus naissance urbain Densité AFRIQUE 967,049 2.3630 -0.56 4.886 41.36 3.38 53.93 38.04 32 AFRIQUE SUB- SAHARIENNE 809,115 2.47 -0.39 5.373 43.38 3.07 50.47 35.16 33 AFRIQUE DE L’OUEST 290,826 2.64 -0.78 5.708 44.41 3.23 51.01 41.72 47 Benin 9,309 2.96 1.00 5.730 44.20 2.70 55.50 40.78 83 Burkina Faso 15,213 2.99 -0.90 6.235 46.10 3.10 50.70 16.30 56 Cape Vert 503 2.49 -2.10 3.500 37.84 6.01 70.95 58.67 125 Côte d'Ivoire 20,677 2.37 -1.50 4.900 40.30 2.40 51.60 48.20 64 Gambie 1,559 2.69 0.50 5.130 42.00 3.30 58.00 53.90 138 Ghana 23,947 2.15 -0.40 4.300 40.15 3.50 58.50 47.76 100 Guinée 10,302 2.85 -6.30 5.710 45.90 3.05 53.70 29.60 42 Guinée-Bissau 1,746 3.08 0.20 7.085 47.50 3.00 44.90 30.00 48 Liberia 3,942 3.13 3.15 6.770 46.90 2.20 45.70 58.10 35 Mali 12,716 3.34 -2.70 6.570 47.70 3.60 56.40 30.50 10 Mauritanie 3,204 2.66 1.40 4.830 40.30 3.60 60.40 40.00 3 Niger 14,731 3.14 -0.45 7.100 49.00 3.40 56.90 16.70 12 Nigeria 148,071 2.52 -0.35 5.850 44.90 3.20 46.90 46.70 160 Sénégal 12,688 2.95 -1.60 5.280 43.80 4.20 62.35 40.80 64 Sierra Leone 5,450 2.52 -3.80 6.100 41.70 4.40 48.40 36.80 76 Togo 6,761 2.77 -0.15 5.085 43.30 3.10 58.00 40.00 119 Source: World Development Source 2008 Figure 6: Urbanization and density in various West-African countries 180.00 160.00 Pourcentage de la population 140.00 en milieu urbain 120.00 Densité (habitants au km2) 100.00 80.00 60.00 40.00 20.00 0.00 E T N ES a i re so ia ia er l ria na n u de a e go al ga N bi ni ne sa on an er oi M ig Fa IE U be ha To er am ne Be Iv N is ui ig Le ri t O AR V G d' Li B N Se ina G L’ G au a- e ra e AH DE ap M rk ôt ne er Bu C C -S Si ui E B G U SU R IQ E AF U IQ R AF Source: World Development Source 2008 19 20 Figure 7: Population density (2002) Source: Bollig et alii 1.4. Macroeconomic situation Since 2006, there is a slight acceleration of the economic growth rate, although this is still insufficient to attain the Millennium Development Goals (MDGs). The real GDP growth rate increased from 3.3% in 2003-2005 to 4.5% in 2006-2008 (see graph 1.8 below). Due to these good performances, Benin experienced over this period an economic growth rate higher than the average in UEMOA member countries (see graph below). However, the global financial crisis would produce significant negative impacts on the economic growth of Benin over the period 2009-2010. As a matter of fact, according to the latest forecasts of the Government of Benin and IMF services, the GDP growth rate was expected to be 3.8% in 2009 as against 5.7% initially while it was expected to be only 3.0% in 2010. Considering the relatively high level of the population growth rate, the good economic results obtained in the past years remain very insufficient to improve significantly the per capita income and thus to reduce poverty. 21 Figure 8: Benin: GDP Growth rate, 2004-2010 Sources: Benin authorities and IMF estimates. Figure 9: Average GDP growth rate between 2005 and 2008 within the UEMOA area 8.0 7.0 6.0 5.0 (%) 4.0 3.0 2.0 1.0 0.0 A i re so er l n au go al ga O ni oi M ig Fa To iss né EM Bé Iv N d' B a Sé U e- in e rk ôt né Bu C ui G Sources: IMF estimates Economic growth remains very closely related to two activities highly influenced by external factors: cotton production and trade with Nigeria. Despite the efforts made to diversify the economy, cotton still represents 40% of total exports. Meanwhile, the prices of this raw material were very volatile and sharply decreased in the past years. Trade with Nigeria mainly involves transit activities from the port of Cotonou to Nigeria. This trade represented approximately 7% of the GDP and unavoidably depends on the economic situation of Nigeria. While the inflation rate was controlled as requested by UEMOA convergence criteria (3%) in the past years, inflationist pressures associated with the high increase in oil and food commodities prices led to a general increase in prices. Thus, the inflation rate increased from 1.3% in 2007 to 7.9% in 2008. However, the drop in the prices of raw materials, the global economic slowdown and to a lesser extent the cautious monetary policy conducted by BCEAO were expected to bring back inflation around 4% in 2009. 22 Figure 10: Benin: Inflation rate, 2005-2008 10.0 8.0 6.0 (%) 4.0 2.0 0.0 2005 2006 2007 2008 Sources: Benin authorities and IMF estimates The budget situation of the Government of Benin continues to be strengthened particularly due to an improved tax performance and a better control of expenses. The measures taken by the authorities in order to strengthen governance in the financial departments, intensify controls and limit exemptions have helped increase the ratio of total incomes in relation to the GDP from 16.4% in 2004 up to 20.6% in 2007 (see graph below). Over the same period, the total expenses increased from 20.1% of the GDP to 22.0% of the GDP. This control of expenses can be explained (i) by a strict control of recurrent expenses (for example there is a ceiling for staff expenses) and (ii) by an insufficient capacity to spend the funds allocated for investments. This last point translated in a public investment level being systematically below the average of UEMOA. The low investment expense execution rate particularly affects the Ministry of Health. Because of the low capital expenditure disbursement level and the increase in tax incomes, the budgetary balance was improved significantly in the past years (see figure below). Figure 11: Evolution of the Government revenues, 2004-2007 Recettes Recettes 25 fiscales totales (en pourcentage du PIB) 20 15 10 5 0 2004 2005 2006 2007 Sources: Benin authorities and IMF estimates 23 Figure 12: Evolution of public investment expenses, 2006-2007 Sources: Benin authorities and IMF estimates Figure 13: Evolution of gross public investment, 2003-2007 (in % du PIB) Sources: Benin authorities and IMF estimates Figure 14: Overall budgetary deficit (excluding grants) 4.0 (en pourcentage du PIB) 3.0 2.0 1.0 0.0 2004 2005 2006 2007 Sources: Benin authorities and IMF estimates 24 In spite of considerable efforts, governance level remains low. The CPIA rating for Benin practically stagnate since 2003, around 3.6 (out of 6). However, this rate is slightly higher than the African average (3.2). Since 2003, some specific indicators have improved (for example corruption control), while others deteriorated (i.e. compliance with law and government efficiency). Furthermore, corruption in the public sector remains a major concern for the citizens, as pointed out by the 2007 GAC survey. Figure 15: Perception by the population of “very serious” problems Source: GAC 2007 1.5. Poverty Situation In recent years, poverty incidence seems to have stabilized in Benin. According to the recent EMICOV survey (INSAE 2008), conducted in 2006, monetary poverty incidence reached 37.4% in Benin. Another index - that of non-monetary poverty4 - is slightly higher, i.e. 40.2%. The index of perceived poverty is much higher since 50.6% of households consider themselves poor in Benin. It is difficult to measure exactly the evolution of income poverty. The latest survey (CWIQ 2002) does not use the same methodology as the EMICOV 2006 Study. Their results are not therefore comparable. One can assume that poverty has leveled off or even declined. Indeed, between 2002 and 2006, income per capita has increased slightly, while income inequality (Gini 4 This index is calculated according to a methodology similar to that of DHS, since EMICOV produces a non-monetary poverty index based on the 19 variables reflecting living conditions and assets of households. 25 index) reduced. The other poverty measures give quite divergent results. If non-monetary poverty appears to have reduced (from 43% in 2002 to 40.2% in 2006), perceived poverty has increased (37.2% in 2002 against 38.7 in 2006). Poverty is still largely rural, even if it is spreading rapidly in urban areas. In terms of monetary poverty, the two poorest departments are Couffo (40.6%) and Alibori (43%). In terms of non-monetary poverty, the two poorest departments are Atacora (72%) and Mono (59.5). These departments are heavily rural. Conversely, the two most urban (Cotonou and Porto-Novo) departments are the least affected whatever poverty indicator is used. When we compare urban and rural households, differences in terms of monetary poverty incidence are low: 35.04% in urban areas are poor as opposed to 38.82% in rural areas; same for perceived poverty. In contrast, the approach in terms of non monetary poverty brings up a considerable gap, if only 18% of urban people are poor in terms of assets and living conditions, this rate reached 54.2% in rural areas. 26 2. Methodology of the sector review 2.1. Analytical framework All analyses and reflections made during the review follow a reasoning illustrated by the following chart. Policies from non-health sectors Household resources (education, macroeconomics, (revenues and assets, transport…) education level, cultural contraints…) 1. Environmental “burden of disease” and community factors MDGs (child mortality, maternal mortality, malaria and HIV- Financing AIDS) 2. Effective coverage Health care delivery Stewardship & Gouvernance Other causes of for high-impact (organization and mortality interventions resources) Inputs (human Health outcomes resources) Inputs (equipments and buildings) Inputs (drugs, vaccines and blood products) Health policy Health system As we can see on the chart above, health indicators have been analyzed, then the various constraints that may explain shortcomings in the health indicators (going from the left side to the right side of the chart): 1. The analysis of the « burden of disease» helps first to confirm that MDGs reflect most of the major health problems in Benin. 2. We selected the four most common health problems in Benin in the light of MDGs, namely: child mortality due to malaria, child malnutrition, maternal mortality, and HIV/AIDS. For each of them, we tried to analyze bottlenecks, both with respect to (i) environmental and community factors and (ii) coverage of the so-called ‘’high-impact’’ interventions (for example childbirth by trained personnel, for maternal mortality). 3. For ‘’high impact’ health interventions, after the identification of bottlenecks, we sought to connect them to the various components of the health system (for example, the low impact of assisted delivery is related to the poor quality of the staff) 27 4. For each of these components, we tried to understand these specific problems (i.e. how to explain the low quality of the health staff performance? Is it a problem related to training or to staff management?). Chapters 2 («Health indicators») and 3 («Practices of communities and households») deal with Points 1 and 2 for each of the four issues convened. Chapters 4 to 9 focus on the analysis of different components of the health system, with regards to the bottlenecks previously identified. Let’s finally add that this review was conducted with the information available at the Ministry of Health and Donors. Apart from a study on governance (at the health staff level), no primary data has been collected. The objective of the review was first to show how information routinely available can lead to a thorough analysis of the situation prevailing in the health sector. 2.2. The preparation process Given the objectives assigned to this review (see introduction), the approach chosen has been highly participatory. This has not slowed down the work, which began in June 2008 and was completed in October 2008, during a workshop focused on the discussion of results. Six thematic working groups were set up (see chart below), under the guidance of a steering committee. Participants included both (i) representatives of the public sector and private sector (private commercial and private associative) and (ii) representatives of FTPs. About 70 people took an active part in the preparation of this review. The Steering Committee has regularly reported to the Minister of Health and the Monitoring Committee of the General Conference. Support was provided both on (i) methodology (by a Beninese consulting firm -Afrique Conseil) and (ii) technical aspects by the Ministry of Health (including the Department of Planning and Prospective), the World Bank, WHO and UNICEF. 28 Monitoring Committee for the General Conference Steering Committee for the Health Sector Review Methodological and technical support Group 1: Group 4: Stewardship and Human Resources Cabinet governance for Health Afrique Conseil World Bank Group 2: Group 5: Health system Supply chains (drugs, financing vaccines and blood bags) WHO Group 3: Group 6: UNICEF Health care services Nutrition and environment Health sector stakeholders 29 Chapter 2 : HEALTH INDICATORS 1. General Mortality ................................................................................................................................31 2. Child Health ........................................................................................................................................32 2.1. Child Mortality and Morbidity .....................................................................................................32 General trends of child mortality.....................................................................................................32 Children mortality per age group ....................................................................................................33 Child Mortality Factors ...................................................................................................................34 2.3. Main health problems of children and their factors......................................................................35 2.2.1. Child malaria .........................................................................................................................37 Incidence and lethality.................................................................................................................37 Factors of child malaria...............................................................................................................38 2.2.2. Children malnutrition ............................................................................................................44 Malnutrition Indicators................................................................................................................44 Malnutrition factors.....................................................................................................................49 3. Maternal Health...................................................................................................................................54 3.1. Maternal mortality........................................................................................................................54 3.1.1. Maternal mortality slightly decreased over the last 10 years ................................................54 3.1.2. The clinical causes of maternal mortality are unchanged......................................................56 3.1.3. Maternal mortality is rather caused by the low quality of obstetrical services than by their insufficient utilization rate...............................................................................................................56 a. The increase of the Rate of Assisted Delivery explains only partly the decrease in maternal mortality ......................................................................................................................................58 b. Quality of obstetrical care plays an important role to explain the high maternal mortality in Benin ...........................................................................................................................................63 3.2. Nutritional status of the women ...................................................................................................70 4. HIV/AIDS ...........................................................................................................................................71 4.1. Prevalence in the general population is stable since 2002............................................................71 4.2. HI prevalence within the high-risk populations is decreasing......................................................73 4.3. Risk factors...................................................................................................................................73 30 After a short analysis of general mortality in Benin, this chapter focuses on the problems related to the Millennium Development Goals (MDGs) in relation to health. For each of these MDGs, we will address the recent progress of their indicators and especially the possible causes of this progress. 1. General Mortality In Benin, life expectancy (at birth) is noticeably higher than the average in Sub-Saharan Africa, but it has not improved since the last ten years. According to the United Nations (UN 2006), life expectancy at birth is estimated in Benin around 54.4 years. This level is noticeably higher than the other countries of the region, as is shown in diagram below. Despite this, the country seems to have been also affected by a tendency of stagnation (or even a set-back) in life expectancy. Figure 16: Variation of life expectancy at birth (Benin, Sub-Saharan Africa and West Africa) Esperance de vie a la naissance 60 55 50 45 Benin 40 Afrique Sub-saharienne Afrique de l’Ouest 35 30 1950- 1955- 1960- 1965- 1970- 1975- 1980- 1985- 1990- 1995- 2000- 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Source : United Nations, Population Division The causes of mortality are those of a country in the process of epidemiological transition, with a strong prevalence of infectious diseases. Many of these diseases are related to the MDGs. According to the last “burden of disease” estimate produced by the WHO in 2004, mortality in Benin remains marked by infectious pathologies, even if cancerous and cardiovascular pathologies play an increasing role, which shows that epidemiological transition is in progress. Figure 17 below shows (in dark) that most deaths are caused by problems that are targeted by the MDGs. These problems represent more than 54% of the mortality of the Beninese population. The commitment of the country to the MDGs is therefore particularly relevant. 31 Figure 17: Distribution of the causes of mortality (“burden of disease”) in Benin (2004) Taux de mortalite pour 100,000 habitants 250.0 200.0 150.0 100.0 50.0 - at nai lles al t h s r an s nn A nn s in i ge A) sc me e rc s ia es e tra s d ter s VI ees ira rs rd alu s C ll e be elle nc nt t ive nt SID ou oid ma oire os at c a P se re s d SID ph Fai olo esp c e r s ne e ai M T ( eon m sa rr h ul di s s ul t - ba H s rs io io n ho D s va e i re Tu um e te te io s u ie s to is in y x bl e gie es ad S ira Pa die n s M no al sp a p s al re s ie re M ad ns su Bl al io es ie M ct Bl fe In As Source : WHO 2004 For this reason, the remaining part of the chapter focuses on the issues of the four health MDGs: - infant mortality, notably that caused by malaria (part 2.1) and malnutrition (part 2.2) ; - maternal mortality (part 3) ; - mortality caused by HIV-AIDS (part 4). 2. Child Health 2.1. Child Mortality and Morbidity General trends of child mortality Since 1990, children mortality has been decreasing by about 3.5% per year in Benin; unfortunately, this tendency does not allow achieving the MDG no 4. The latest Demographic and Health Survey (DHS/EDS 2006) revealed undeniable progress in the field. Indeed, the children mortality rate dropped from 160 deaths for 1,000 births in 2000 to 125 in 2006. Henceforth, when compared to other West African countries, the average child mortality in Benin is clearly lower. Figure 18: Children mortality rate in West African countries (year 2006) 300 250 200 moyenne 150 100 50 0 au re G ina ad Bu ria da G a Li r G al ria Rw a S e na e G di e te nin pe go Ni u a- so a i ge al i ne rd on a bi M Ivoi n an g ge n e Fa Ca To M rk Ch an be ha ss Cô Be ne ru Ve am Ni Le ui r it Bi d' ra er Bu Si ui 32 Source : WHO 2006 Unfortunately, with a simple extrapolation of the current trends (see Figure 19 below) it appears that the annual rate of decrease of child mortality must double (i.e. increase from 3.5% to 6.1%) in the coming years if Benin wants to achieve the MDG no 4, which is a rate of 61 child deaths for 1000 births in 2015. Figure 19: Evolution of children mortality in Benin 200 185 180 166.5 160 Current trend 160 -3.5% 140 125 120 100 80 -7.1% Required trend 60 for reaching MDG 4 61 40 1985 1990 1995 2000 2005 2010 2015 2020 Source : DHS 2006 Children mortality per age group The drop of children mortality is basically that of post-neonatal mortality. Since 1996, the decrease of post-neonatal mortality (children aged between 1 month and 1 year) has resulted in more than 50% of the overall decrease of infant mortality. Child mortality (children aged between 1 and 5 years) also dropped significantly. These two favorable evolutions are probably the result of the progress achieved in vaccination (see chapter 7) and in breastfeeding (see chapter 3). On the contrary, the decrease in neonatal mortality (children aged less than one month) has been very low over the period (less than 0.6% of annual average decrease versus 2.08% for child mortality and 1.46% for postnatal mortality). The low progress recorded in that age group is the result of the stagnation of maternal mortality. Indeed, both indicators are closely linked to the degree of development of obstetrical care services (see part 3 of this chapter on maternal health) 33 Figure 20 : Child Mortality per age group 180 160 140 72.6 70.9 120 Mortalite juvenile (1-4 ans) 100 58 Mortalite post-neonatale (29-365 jours) 80 55.8 50.7 Mortalite neonatale (0-28 jours) 60 35 40 38.2 38.4 32 20 0 1996 2001 2006 Source: DHS1996, 2001 and 2006 Child Mortality Factors Over the past ten years, the decrease of child mortality (less than 5 years) seems to be closely linked to birth spacing and breast feeding. Benin is one of the rare countries where three “Demography and Health” (DHS) surveys have been conducted at regular intervals (every five years). This availability of data allows analyzing finely the factors that can explain the decrease of child mortality. Table 2 : Factors for survival of under-5 (1996, 2001 et 2006) (Cox Hazard model) Variables 1996 2001 2006 Biological factors Child gender is male 0.921 1.128 1.080 Mother age at child birth 1.120 0.930 0.934 Squared mother age at child birth 0.999 1.002 1.002 Behavioral factors Space between two births is between 24 and 48 months 0.395*** 0.580** 0.756 Space between two births is more than 48 0.264*** 0.555** 0.641* Health system factors At least two shots for tetanus 1.376 1.162 1.278* Antenatal care 0.982 0.985 1.029 Delivery assisted by qualified staff 1.542 1.084 0.530 Delivery with c-section 1.057 3.137*** 0.955 Length of breastfeeding 0.844*** 0.877*** 0.878*** Breastfeeding immediately after birth 0.963 0.977 1.184 Postnatal care 1.020 Socio-economic factors Education level of the mother Primary 1.097 1.074 1.077 (base : no education) Secondary and more 1.010 0.936 0.916 Wealth index of household Poorer 0.943 0.842 1.063 (base : poorest) Average 0.716 1.250 1.083 Richer 0.906 0.690 0.828 Richest 0.493 0.447 0.525* Environmental factors 34 Size of household 0.925 1.041 0.926 Urban residence 1.454 1.453* 0.718* Access to clean water 1.114 0.910 1.312* Access to quality toilets 1.206 0.364 Constant 0.00979** 0.435 0.119 Number of observations 2915 3583 12958 2 Pseudo R 0.308 0.158 0.163 Sources: DHS 1996, 2001 and 2006 and World Bank analyses Note 1 : The figures presented here are mortality ratios (hazard ratio). A ratio below 1 positively contributes to children survival. On the contrary, a ratio above 1 corresponds to a factor that reduces survival. Note 2 : « *** » is significant at 1%, « ** » at 5% and « * » at 10%. The table above (and the diagram below) shows that the two most significant factors are birth spacing (especially when the interval is more than 48 months) and breast feeding. Unexpectedly, the impact of the level of education of the mother and the level of wealth is not significant (and this, despite the important size of the samples). The same is true of environmental aspects (access to clean water for example) or antenatal care. These results can be illustrated with the following diagram. Figure 21: Main factors of the survival of children under 5 (1996, 2001 and 2006) 16 E s pacement des nais s ances (24 14 mois et plus )*** Allaitement maternel (duree et 12 precocite)*** Accouchement as s is te par des 10 profes s ionnels 8 S oins pre-nataux 6 Age de la mere a la nais s ance 4 Niveau de riches s e (appartenance au quintile plus riche) 2 Niveau d’education de la mere 0 1996 2001 2006 Sources: DHS 1996, 2001 and 2006 2.3. Main health problems of children and their factors Children mortality is still mostly caused by malaria, malnutrition, and neonatal factors. The causes of children mortality in Benin are more or less similar to those observed in the African region. Neonatal causes, respiratory infections, and diarrhea explain 58% of children mortality. Similarly, according to the recent Profiles analysis (2004), about 39% of children deaths occur in a context of malnutrition; and that rate is comparable to those of the other 35 African countries. However, Benin is characterized by a children mortality which is mostly caused by malaria (27% of the deaths, first cause of child mortality). Table 3: Causes of children deaths (2000-2003) Benin Region Malaria 27 17 Neonatal Causes 25 26 Respiratory infections 21 21 Diarrheas 17 17 HIV-AIDS 2 7 Measles 5 4 Injuries 2 2 Others 1 6 100 100 Source : WHO estimates 2006 36 Figure 22: Causes of children deaths (2000-2003) 30 25 Bénin 20 Région 15 10 5 0 A e s s es e es es ID tr e m re ol é al ss is -S ge su Au rh at d ba H u es r lu on ia Ro VI Pa Bl s D né ire to s se ira u sp Ca re n io ct fe In Source : WHO estimates 2006 The rest of this section focuses on malaria and malnutrition. Neonatal mortality is addressed in the section on maternal mortality. 2.2.1. Child malaria Incidence and lethality remains high The incidence of malaria is still high, especially among children under 11 months. The table below shows that the cases of malaria are very frequent and mostly affect children in the neonatal and postnatal periods. This is probably the cause of excess mortality for this age group. In addition, it seems that this incidence has not decreased over the past years. Table 4: Incidence of malaria among children (2004-2006) Simple malaria (cases for 1000 inhabitants) 2004 2005 2006 0-11 months 502 491 490 1-4 years 218 197 212 Serious malaria (cases per 1000 inhabitants) 2004 2005 2006 0-11 months 76.7 79.7 131.0 1-4 years 40.0 39.7 66.0 Source: Statistic Directory 2006 37 The lethality also remains very high. It even worsened for children under one year old. Figure 23: Lethality of malaria among children 35 30 25 Population générale 20 1-4 ans 0-11 mois 15 10 0-11 mois 5 1-4 ans 0 2002 Population générale 2003 2004 2005 Source: Statistic Directory 2006 Factors of child malaria As experiences of many countries (i.e. Eritrea, Rwanda…) showed, two interventions are likely to have a major impact on the mortality caused by child malaria. The first is preventive and concerns the use of insecticide-treated mosquito nets (ITN). The second is curative and concerns the treatment of serious malaria cases. As we shall see, the coverage of these two interventions hardly progressed in Benin (at least, up till 2007). This widely explains the insufficient results of the country as regards infant malaria. Use of impregnated/infused mosquito nets Why are mosquito nets insufficiently used in Benin? We developed the causal tree below in order to identify the possible “bottlenecks” concerning the use of ITNs. First, the use of ITNs can be limited due to an insufficient number of ITNs in households (“supply-side” constraint) or because families are not convinced that it is necessary to use these ITNs for children and prefer using them for friends or old people (“demand-side” constraint). While the availability of ITN appears as the main bottleneck (which is the case of Benin, as we shall see), a second analysis has to wonder about the causes of this weak availability in the households. Is it because the demand coming from the households is low (due to a limited acceptability and/or a low affordability of the ITNs) or because villages are not sufficiently supplied in ITNs to answer the demand? Data of the 2006 DHS allow partly answering these fundamental questions for the development of an effective strategy for distribution and promotion of ITNs. 38 Figure 24 A tree of bottlenecks for the use of the mosquito nets 5.1. AFFORDABILITY 4.1. AVAILABILITY OF 4.2. AVAILABILITY OF ITNs IN COMMUNITIES COMMUNITY HEALTH WORKERS if ITNs are sold, how many communities had no ITN stockouts are they affordable ? in the last 6 months ? how many districts have a shop or a CHW for distributing ITNs ? 5. DEMAND FOR ITNs 4. COMMUNITY SUPPLY BY HOUSEHOLDS FOR ITNs 5.1. ACCEPTABILITY do knowledge and cultural norms limit 3.1. SUPPLY OF HOUSEHOLDS demand and utilization of ITNs by households ? IN ITNs how many ITNs are in households ? 6. WILLINGNESS OF HOUSEHOLDS 3. AVAILABILITY OF ITNs TO USE ITNs FOR CHILDREN IN HOUSEHOLDS 2. UTILIZATION OF ITNs how many U5 slept under an ITN during the last night ? 1. IMPACT ON MALARIA-RELATED CHILD MORTALITY The low use of the ITNs seems to be caused more by the insufficient availability of these ITNs in the households (“supply-side” constraint) than by the cultural resistances or by the unwillingness of households to use them for children (“demand-side” constraints). In 2006 (DHS), the proportion of children who slept under a mosquito net (ITN or not) was 46.5%, versus 32% in 2001. As far as ITNs are specifically concerned, the rate was only 20% in 2006, but the 2001 rate is unknown. These improvements appear all the more modest as the inequalities regarding the socio-economic status among the households are particularly deep. Thus, for the ITNs, the use rate varies from 9.3% with the poorest to 33.9% with the richest. The other gradients (according to the place of residence or according to region) are less strong. In order to better understand the reasons that explain this weak use of ITNs, a logistic regression has been produced (see table below). This analysis shows that the use of ITNs (for children) is specifically explained by (i) the number of available ITNs within the households (marginal effect at 0.24) and, (ii) in a less clear way, by socio-economic status. If the richest (Q5) have an obviously higher probability (odds-ratio) to use the ITNs, the probabilities are not that different for the other quintiles. Similarly, level of education seems to play a limited role in the explanation of the probability to use ITNs. On the contrary, this probability is strongly linked to the number of MIIs existing within the households. These results suggest that when households have a significant number of ITNs, they do not have strong reluctances (reluctances linked to the level of education or the socio-economic status) to use them. And the “supply-side” constraint seems to be more 39 important than the “demand-side” constraint. It would be of particular interest to attempt to identify the factors likely to explain this low availability of the ITNs within the households. 40 Table 5: Logistic regression and marginal effect of the use of insecticide impregnated bed nets by the children under 5 years in Benin. U5 slept under a ITN Odds Ratio Standard P>|z| dy/dx* Error Number of ITNs in households (0-6) 5,13 0,167 0,000 0,24 Socio-economic status (Ref. :Poorer) Q2 1,33 0,116 0,000 0,04 Q3 1,64 0,140 0,000 0,08 Q4 1,65 0,145 0,000 0,79 Richer 2,19 0,226 0,000 0,13 Age of child (0-5years) 0,85 0,015 0,000 -0,02 Region (Ref. : Alibori) Atacora 2,15 0,355 0,000 0,13 Atlantique 1,42 0,232 0,031 0,06 Borgou 1,80 0,293 0,001 0,09 Collines 3,51 0,576 0,000 0,24 Couffo 3,52 0,552 0,000 0,24 Donga 1,76 0,312 0,001 0,09 Littoral 1,89 0,324 0,000 0,10 Mono 3,64 0,602 0,000 0.25 Queme 3,15 0,478 0,000 0,21 Plateau 2,39 0,425 0,000 0,15 Zou 3,17 0,487 0,000 0,21 Education received by the mother (Ref. : no education) Primary 1,47 0,097 0,000 0,060 Secondary and higher 1,48 0,153 0,000 0,065 Antenatal Care Visit 2,72 0,432 0,000 0,112 Total number 9 884 Prob> chi2 0,000 Pseudo Rsq 0,36 Area under the ROC curve 0,9 * Marginal effect after the logistic regression of the probability that a child under 5 sleep in an insecticide- impregnated mosquito net = 0.18 Source: DHS2006 In fact, despite recent improvements, availability of bed nets remains low: less than 25% of Benin households have at least one ITN. More than 56% of the households have a mosquito net (ITN, impregnated or other); this is a real progress as compared to 2001 when the proportion was only 40%. However, many points are still worrying. Firstly, the average number of mosquito nets (all types) is only 1, which is obviously insufficient considering the average size of the 41 households (3 children). Secondly, when only ITNs are considered (as they are really effective against mosquitoes), fewer than 25% of households have such bed nets. Lastly, inequalities regarding the availability of ITNs are considerable. The most important gradient is clearly that of the socio-economic status. The average number of the ITNs varies from 0.2 for the poorest to 0.7 for the richest. The differences are less considerable in view of the place of residence (0.5 in urban area and 0.3 in rural area) and in view of the region of residence (the rate varies between 0.2 and 0.5, except the worrying situation of Alibori, which is at 0.1). Availability of bed nets is mainly explained by the utilization of prenatal services and by the socio-economic status of households. The regression below shows that two factors play a major role to explain the availability of bed nets within households. The first factor is the utilization of prenatal services. This link is easily explained by the fact that prenatal visits are currently the main channel of distribution of ITNs (which are not free of charge, and cost around 500FCFA). But, given that utilization of prenatal services is already relatively high (more than 80%), one can wonder if the solutions for improving availability of ITNs should use this channel. The other important factor is the socio-economic status of households. Indeed, it may be possible that the main obstacle for obtaining bed nets is their cost (and not their availability). This would tend to justify the current policy for massive distribution of free mosquito nets. Table 6: Regression of the availability of insecticide impregnated mosquito nets for children under 5 years in Benin Number of the ITNs in the households Coefficient Standard P>|t| Prenatal visits 0,16 0,024 0,000 Level of education of the mother 0,12 0,018 0,000 Area of residence 0,029 0,02 0,16 Number of children under 5 in the household 0,06 0,009 0,000 Socio-economic status 0,16 0,008 0,000 Health facility reportedly too far -0,04 0,018 0,021 Region (ref. : Plateau) Alibori -0,27 0,044 0,000 Atacora 0,16 0,048 0,001 Atlantique -0,31 0,041 0,000 Borgou -0,06 0,048 0,189 Collines 0,12 0,051 0,014 Couffo 0,19 0,050 0,000 Donga 0,22 0,059 0,000 Littoral -0,24 0,056 0,000 Oueme 0,006 0,047 0,888 Mono 0,05 0,046 0,303 Zou 0,04 0,045 0,427 Constant -0,28 0,070 0,000 Number of observations 9884 Prob>F 0,00 Rsq 0,12 Treatment of severe malaria cases Though malaria is the first cause of outpatient visit and hospitalization among children, utilization rates of health care services (for malaria cases) are still too low. According to the 42 health statistic directory of Benin (2006), malaria is the cause of 44.3% of children visits and 39.8% of children hospitalizations. However, the use of health care services for children malaria cases seems to be limited. There are no specific data on malaria to appraise this, but DHS 2006 provides data on the use of health care services for cases of fever (those which are not caused by a respiratory infection). On average, when a child suffers from fever, only 37% of the cases are treated in hospitals or with other health care providers. While this average hardly varies according to the residence area or across regions, it is strongly influenced by the level of education of the mother (34% for the mothers without instruction and 62% for those who have at least the secondary level) and by the socio-economic status (24.8% for the poorest versus 55.6% for the richest). Malaria policy in Benin As early as in the 50s, the country implemented some actions in order to eradicate and then control malaria. The current policy is defined by the National Program for the Fight against Malaria (PNLP) 2006-2010. This strategy is particularly relevant in so far as it highly focuses on the two high impact interventions we mentioned earlier: (i) use of ITN and (ii) medical treatment of malaria, especially for the severe cases. As far as the utilization of ITNs is concerned, the PNLP plans to reach a use rate of 80% (versus 20% in 2006). In order to reach this result, the mosquito nets distribution channels have been diversified. While the antenatal visits remain the usual channel, other channels have been set up: vaccination campaigns, child visits and especially the massive and free distributions. The first of these distributions took place in October 2007. The other main action axis is the distribution of artemisinin-based combinations therapies (ACT), so as to make up for the growing resistance to the usual drugs. This effort of rolling out ACT is in progress. We should mention that the policy against malaria is highly supported by external partners, especially the World Bank (Malaria Booster Program) and the United States (Presidential Malaria Initiative). 43 2.2.2. Children malnutrition Malnutrition Indicators How can we measure malnutrition and what is its impact on children mortality? Malnutrition is usually measured through two types of indicators: (i) anthropometric data (underweight, stunting, wasting and low birth weight) and (ii) micronutrients deficiency (basically vitamin A, iron, zinc, and iodine). Most of these indicators have an impact on children mortality, even if this impact may considerably vary, as is shown in the figure below. We can notice that chronic malnutrition and severe malnutrition have the strongest impact on children mortality, because they cause 19 to 21% of that mortality on average. Underweight (weight by age) is used as main indicator for the MDG no 1. This indicator is actually the result of two indicators: stunting and wasting. Stunting (size by age) corresponds to a chronic malnutrition. It reflects the long term cumulative effects of inadequate dietary intakes and bad health conditions. It is also closely linked to zinc deficiencies. On the contrary, wasting (weight by size) corresponds to a severe and recent malnutrition. Below, we shall see that these two indicators move in the opposite direction according to the age of the children. Source: Bhutta 2008 44 Underweight, stunting and wasting Underweight becomes less frequent in Benin and affects only 18% of the children. The prevalence of underweight decreased from 23% in 2001 to 18% in 2006. This favorable evolution took Benin out of the countries where underweight is considered to be high (i.e. countries where the prevalence of underweight is over 20%). However, the improvements achieved are unequal according to the regions and the socio-economic status. In fact, the geographical spread of the underweight prevalence rate is very wide, going from 10% in Littoral (Cotonou city) to 30% in the northern departments (Atakora and Alibori). These two are the only regions where this prevalence worsened between 2001 and 2006. The socio-economic status of the households is also a very important gradient, since the prevalence of underweight varies from 10% in the richest quintile to 25% in the poorest. Figure 25: Prevalence of underweight per region (2006) Insuffisance ponderale (underweight) (EDS 2006) 35% 30% 25% 20% 15% 10% 5% 0% fo a u u u i é o es a n l or ra ng Zo go m ea tla on uf or ib lin tto ué ac Co Do or A at M Al ol Li Pl O B At C Source: DHS 2006 The prevalence of stunting, which is an indicator of chronic malnutrition, has increased for many years and affects 43% of the children today. Between 2001 (date of DHS-2) and 2006 (DHS-3), prevalence of stunting has increased from 31% to 43%. As a reminder, WHO considers that there is a very serious crisis when stunting affects more 40% of the children. This situation is homogeneously spread over the territory, except the city of Cotonou (where the prevalence is only 26%). Even some relatively urbanized regions like Atlantique and Ouémé record some rates higher than 37%. Regarding the socio-economic status of households, prevalence of stunting is also comparable in the three poorest quintiles (between 49 and 47%). The richest households (Q5) record a clearly lower prevalence, around 28% (which is still high). 45 Figure 26: Prevalence of stunting per region (2006) Prevalence du retard de croissance (stunting) (EDS 2006) 70% 60% 50% 40% 30% 20% 10% 0% fo a ou u i u é es a n o l or ra ng Zo m ea on la uf or rg ib lin tto At ué ac Co Do at M Al Bo l Li Co Pl O At Source: DHS 2006 Wasting rate, which corresponds to severe malnutrition, tends to stagnate around 8%. In fact, this rate increased very slightly between 2001 and 2006, but this evolution is not statistically significant. When we consider the WHO5 criteria, this national average rate is not considered to be particularly alarming. In addition, the difference between the urban and rural areas is very small (less than 1%). On the other hand, the prevalence of wasting is higher than 15% in two rural regions (Atacora and Donga). The gradient of the socio-economic status is relatively important, as the wasting prevalence varies from 10% with the poorest to 5% with the richest. Figure 27: Prevalence of wasting per region (2006) Prevalence de l’emaciation (wasting) (donnees EDS 2006) 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% o ff o ga i a s au u u or on e Zo go r co ib lin ou n te M Do r Al a a ol Bo C At Pl C Source: DHS 2006 Micronutrients deficiency More than three Beninese children out of four still suffer from anemia (iron deficiency). Between 2001 and 2006, the prevalence of anemia decreased only slightly, going from 82% to 5 According to WHO, a low rate is below 5% and a high rate is over 15%. 46 78%. Once again, except the region of Littoral6 (where the prevalence is only 60%), the deficiency is spread over all the departments, with a maximum of 90% observed in Alibori. According to Hotz (2004), when the anemia rate is higher than 40%, it is considered to be a serious public health problem. Despite this, a comparison between Benin and other African countries shows that the situation in Benin is within the average of the region (see diagram below). It should be mentioned that anemia is hardly linked to the socio-economic status of the households. The first four quintiles record a comparable prevalence (varying from 83% to 75%). Figure 28: Prevalence of infant anemia in Africa 100 90 80 70 60 Mild 10.0-10.9 g/dl) Moderate (7.0-9.9 g/dl) 50 Severe (below 7.0 g/dl) 40 Any anemia 30 20 10 0 5 03 05 5 01 06 03 ille 4 1 6 06 00 00 0 00 00 20 20 20 20 20 20 20 )2 l2 i2 i2 na a so er n n on ga al al ne ni ni ig Fa M M ha ro av Be Be ne ui N go m e G zz Se G a in ra a rk C (B Bu on C Source: various DHS Deficiencies in zinc and iodine are not measured in Benin. For zinc, we can assume that the deficiencies are relatively high, as it has been shown (Hotz 2004) that they were strongly related to stunting (43% in Benin) and to anemia (78%). For iodine, we ignore the prevalence of its deficiency in Benin, but on the other hand we know from now on (DHS 2006) the practices regarding the consumption of that micronutrient element. They are presented in chapter 3. Deficiencies in vitamin A seem to be relatively high. The only data available for the children date back from 1999 (WHO 2007). They show that prevalence of the deficiency in vitamin A around 82% in the north of the country (Atacora and Borgou) and 64% in the south (Atlantique, Mono and Ouémé). The first rate is clearly above the regional average (67%). Birth underweight More than 12% of the new-born babies suffer from underweight (below 2,500 g), a rate close to the average of the African continent. As is the case for anemia, all the regions of 6 i.e. the City of Cotonou. 47 Benin are affected by birth underweight, its prevalence varying only from 11% to 17%. Alibori is once more the most affected region. 48 Malnutrition factors How can malnutrition be explained in a given country? The factors of malnutrition are numerous and strongly interdependent. Today, most of the international organizations use an analytical framework which strives to take into consideration at the same time all malnutrition factors and their articulation. 49 This framework implies that malnutrition is linked to three types of causes that are strongly interlinked. Firstly, malnutrition is the result of two immediate causes which are (i) the nature of food taken (in quantity and quality) and (ii) the state of health. Of course, these two causes are interdependent. Secondly, these two causes are themselves affected by three underlying causes: (i) access to food (a cause which is very rare and occurs only in situations of food insufficiency), (ii) dietary practices, especially as regards breastfeeding and the consumption of food rich in micronutrients and (iii) access to drinkable water, sanitation and health services. Lastly, these different underlying causes are linked to social, economic, religious and political dynamics under implementation in each country (or region). These dynamics are called fundamental causes. In terms of analysis, of course, it is difficult to rigorously examine the fundamental causes, unless we conduct very comprehensive studies (for example, a study on the cultural behaviors in such and such a region). They are however mentioned in this analytical framework so as to remind that, for example, the same malnutrition situation in two regions of the same country can have very different causes, especially because of the economic conditions or the cultural features of each of the regions. The underlying causes are by definition more operational; and as such they can guide us in our effort to explain the nutritional situation of the Beninese children. These causes are shortly described below and examined in detail in the following chapters of the report. Source : Rokx 2000 1. Insufficient access to food This underlying cause does not seem to affect Benin. Even if some communities are sometimes in difficult situations (especially for small farmers in the south of the country, as well as for fishers and for small income families in urban and suburban areas), Benin does not experience structural food insecurity. 2. Dietary practices Feeding of new-born babies and young children is still not adequate. Child malnutrition can be significantly reduced if each child receives the following three practices: (i) breastfeeding from birth (ii) exclusive breastfeeding until 6 months and (iii) some supplements between 6 months and 2 years. According to DHS, less than the third of the Beninese children were in that situation in 2006. There is less concern as regards consumption of micronutrients, except for iron and zinc. As we have seen earlier, the nutritional situation of children is getting better when they take appropriate quantities of micronutrients: iodine, vitamin A, iron and zinc. These last two micronutrients are particularly a source of problem in Benin, as is shown by the high level of anemia (which affects three children out of four). These different practices (and the policies aiming to improve them) are examined in detail in the following chapter (chapter 3). 3. Access to water, sanitation and health services In terms of access to water and to adequate sanitation infrastructures, Benin is in a very worrying situation. Prevalence of diarrhea drops significantly when three conditions are met: (i) systematic hand washing (especially by the mothers after cleaning their children’s stools), (ii) 50 easy access to clean water and (iii) infrastructures for waste disposal, more particularly human waste (i.e. toilets). While hand washing habits have progressed in Benin in the last years, the same is not true of the other two conditions. On the contrary, access to clean water has deteriorated since 2001. Frequency of defecation in open air put Benin among the most backward countries in Africa. Only Niger and Chad are in a more unfavorable situation. These different points are analyzed in detail in the following chapter (chapter 3). Utilization of the health services to manage severe malnutrition cases is still low. For treatment of severe malnutrition cases, access to quality health services is fundamental. There is no specific data on the treatment of severe malnutrition cases. However, there is some data on diarrhea management. The 2006 DHS survey shows that while 71% of the women know the Oral Rehydration Therapy (ORT), only 30% of children suffering from diarrhea receive such a treatment, and a lower proportion (20%) use health services. This low use is remarkably homogeneous within the country. It seems to be caused neither by an insufficient geographic access (i.e. the differences between the regions and between the urban and rural areas are very low), nor by financial obstacles (i.e. the poorest women use the health services in only 15% of cases while the richest women use them in only 29% of cases, which is still low). These data suggest that the women do not see any interest in bringing their children suffering from diarrhea to health centers, either because they do not perceive the risk linked to malnutrition (i.e. insufficient efforts for the promotion of health), or – and this is more probable – because they consider that health services do not offer a good quality service (i.e. low quality of the health care). The different problems of accessibility and quality of the health care services are described in detail in the following chapters, especially in chapter 5 (health services). Child health: main observations Even if children mortality dropped significantly these last years (probably with the important progress of Benin with regard to vaccination), it is still at a high level, because of 3 factors: malaria, malnutrition, and neonatal causes. The high prevalence of children malaria – a particular feature of Benin – seems to be caused to a large extent by a low use of impregnated mosquito nets. This low use itself is linked to insufficient supply of the mosquito nets and to their cost, which is still too high. It is therefore probable that the ongoing massive distribution of free mosquito nets (effort that is strongly supported by the Malaria Booster project of the World Bank and the Presidential Malaria Initiative of the United States) would have an important impact. With regard to children malnutrition, major improvements have been achieved by Benin, more particularly as regards underweight, birth underweight, and wasting. The prevalence of these three problems has significantly decreased over the last years. However, some concerns still remain: (i) the high inequality between regions in terms of malnutrition, the region of Atacora being in the most difficult situation; (ii) the prevalence which is still very high and the increase of stunting, probably linked to micronutrient deficiencies (especially in zinc and vitamin A) and (iii) the high prevalence of anemia. Unlike malaria, there is a need to develop and especially to implement a strong policy on nutrition; that policy should be multisectoral and community- oriented. 51 Within children mortality, neonatal mortality is the one that knew the lowest decrease over the recent period. This situation is probably linked to the same stagnation recorded for maternal mortality. This point is examined in the following section. 52 In conclusion, about children health, this section of the chapter allowed us to better identify: (i) the main causes of children mortality, (ii) the coverage of interventions with high impact on these mortality causes and (iii) the components of the health system that are the major bottlenecks for the generalization of these high impact interventions. This logic is illustrated in by the diagram below. The following chapters of the report aim to better explain the strengths and weaknesses of each of the components of the health system and more particularly the ones that have been identified as bottlenecks. How strengthening the entire health system can reduce children mortality in Benin? HEALTH SYSTEM GOVERNANCE & STEWARDSHIP OF THE SYSTEM (chapter 4) HEALTH FINANCING (chapter 9) MAIN ENVIRONMENTAL COMMUNITY-BASED PREVENTION CLINICAL CAUSES HIGH IMPACT FACTORS COMPONENT COMPONENT COMPONENT OF CHILD INTERVENTIONS (chapter 3) (chapter 3) (chapter 7) (chapters 5 to 8) DEATHS Supply and distribution Bed nets of bednets Malaria Clincial services Treatment (buidlings, staff of acute cases and drugs) Antibiotics Respiratory infections Imunization Vaccines (measles) supply and distribution Treatment of acute malnutrition cases Micronutriments Distribution of Malnutrition consumption micronutriments Exclusive breastfeeding Health promotion Handwashing Investments Diarrheas Water supply in sanitation and Sanitation waste management Neonatal causes (cf. maternal health) 53 3. Maternal Health 3.1. Maternal mortality7 3.1.1. Maternal mortality slightly decreased over the last 10 years In Benin, maternal mortality has decreased by at least 5% over the last 10 years, but it is still at a high rate. It is unanimously recognized that the maternal mortality ratio (MMR) (number of the maternal deaths for 100,000 births) is very difficult to measure. In this case, for Benin, we have two data sources. The first source results from the estimates made by WHO (with UNICEF, UNFPA, and the WB) (Hill 2007). These estimates result in a MMR of 840 in 2005 (last estimate produced), with a slight decrease over the last 10 years. The other data source is provided by the DHS. The last DHS conducted (2006) estimated the MMR at 397 in Benin. Regardless of the estimate we consider, the diagram below suggests that the MMR has a downward tendency, even if the extent of decrease is uncertain (-20% according to DHS and -5% according to WHO). It is also clear that the MMR is very high today. According to Hill (2007), the average in Sub-Saharan Africa is 905, which is not that different from the Beninese rate (840). Figure 29: Evolution of the maternal mortality ratio in Benin Sources: WHO and DHS The current downward rate of maternal mortality will not allow Benin achieving the MDG in 2015. Even if the DHS method (the most favorable) is used, the current tendency will not allow Benin to achieve the MDG no 5, which is about 190 deaths for 100,000 births. In fact, in order to achieve this goal, Benin should have its MMR decreased by 15% per year, while the current tendency is only 10% per year. Obviously, a major reinforcement of the maternal health policy is necessary. The analysis (below) of the causes of maternal mortality enables to identify the priorities of such a policy. 7 While this section specifically addresses maternal mortality, all the conclusions also concern neonatal mortality, which is affected by the same factors. 54 55 Figure 30: Evolution of the maternal mortality ratio in Benin in comparison to the MDGs 600 550 569 498 500 Deaths/100,000 live births 450 397 400 -10.1 350 300 Benin MDG 250 -15.2 200 189.57 150 0 2 4 6 8 0 2 4 6 8 0 2 4 9 9 9 9 9 0 0 0 0 0 1 1 1 19 19 19 19 19 20 20 20 20 20 20 20 20 Sources: WHO and DHS 3.1.2. The clinical causes of maternal mortality are unchanged It seems that the clinical causes of maternal mortality have not changed since 2001. Since 2001, no study concerning the causes of maternal mortality has been conducted. However, a specific study allows having a recent perspective. Conducted in 2003 (Saizonou 2006), it was based on only 4 reference maternity hospitals, which nevertheless accounted for more than 52% of the maternal deaths in hospital in Benin. The clinical causes of these deaths were the following: Table 7: Clinical causes of maternal deaths (2006) N % Postpartum haemorrhage 55 23% Infections 54 23% Hypertension (including eclampsia) 36 15% Dystocia 18 8% Other obstetrical causes 16 7% Indirect causes (anemia, etc…) 60 25% TOTAL 239 100% Source : Saizonou 2006 This distribution is not much different from the one observed in 2001. It is not different from the one recorded elsewhere in the African continent either (WHO 2003). 3.1.3. Maternal mortality is rather caused by the low quality of obstetrical services than by their insufficient utilization rate 56 Generally, maternal mortality is highly influenced by the level of utilization and the quality of the intra-partum case management, i.e. the Basic Emergency Obstetrical and Neonatal Cares (BEmOC). In fact, the recent review produced by Lancet (Campbell 2006) has clearly shown that intra-partum case management (by a qualified staff) was the intervention with the strongest impact. This means that antenatal care, which has been promoted for a long time, presents only a limited interest (at least regarding to the reduction of maternal mortality). We can therefore explain a maternal mortality rate with the following analysis tree: How to explain maternal mortality in a given country? Available supply Available Maternal mortality qualified staff can be explained by… 1.1. Utilization of BEmOC Functional facility can be explained by… Adequate demand 1. Utilization and quality Poverty of BEmOC 1.2. Quality of BEmOC is related to three factors Education Skills & knowledge of staff Availability of drugs and equipment Accessibility of CEmOC 2. Other factors, Prenatal care less essential Postnatal care Vitamin A With regard to utilization of the BEmOC facilities, we must mention that the notion of BEmOC goes beyond the simple idea of assisted delivery (by a qualified staff, present 24 hours a day, preferably a midwife or a doctor). It also assume a minimal quality of this assisted delivery, which means that (i) clinical staff is really competent, (ii) they have the necessary equipment and drugs, and (iii) access to the BEmOC facilities is possible within a reasonable time period. According to the logic presented in the tree above, we tried to explain maternal mortality in two steps. Firstly, we examined the relationship between (i) maternal mortality rate and (ii) utilization and quality of the BEmOC facilities. Secondly, we analyzed the factors that can explain the utilization rate and levels of quality of BEmOC facilities in Benin. As far as this first level of analysis is concerned, we adopted the Rate of Assisted Delivery as proxy for the level of 57 use of the BEmOC facilities, because there was no available data to analyze the relationship between maternal mortality and the quality indicators of BEmOC facilities. We have not examined the relationship between maternal mortality and other factors (apart from the BEmOC facilities) either, because data was not available and because this link is considered to be weak in several different international studies. a. The increase of the Rate of Assisted Delivery explains only partly the decrease in maternal mortality The Rate of Assisted Delivery (RAD) is monitored by the National System for Health Information (NSHI). Thus, we have data for each year and each health district. Data is rather of good quality, as the DHS reaches very similar results. The latter shows a rate of assisted delivery close to 78%, whereas the NSHI data are at 83%. 1. The increase of the RAD does not fully explain maternal mortality within the general population A first possible analysis is of a transversal nature. It consists in measuring the link between maternal deaths (in hospital) and the RAD, and this for all the health districts in Benin. The analysis, made on 2006, shows a weak or even a non-existent relationship (Adjusted R2=0.147). Figure 31: Relationship between the assisted delivery rate and (intra-hospital) maternal mortality in the 34 health districts of Benin (2006) Source : Statistic directory 2006 It is true that the two variables are questionable. On the one hand, as far as the MMR is concerned, only maternal deaths in hospitals are taken into consideration. Some deaths occur after the stay in the health center or before (for women who do not deliver in health centers). On the other hand, as far as the RAD is concerned, it does not really reflect the use of maternity hospitals by the local population. Many urban maternity hospitals (especially in Cotonou) receive a large number of rural women who come to deliver in town. As a result, urban health 58 districts record at the same time very high RADs8 and mortality rates (because the women sent to these hospitals often have some complications). A second possible analysis is of a longitudinal nature and consists in examining trends for the two indicators. We apparently observe a negative relationship between the assisted delivery rate and the maternal mortality ratio, but the small number of observations does not allow having a significant result. 8 These RADs are often above 100% (for example, the districts of Cotonou 1&4 record a RAD of 253%), because the health centers of these districts receive many patients who do not live in the said districts. 59 Figure 32: Compared evolution of the assisted delivery rate and maternal mortality Source: DHS In total, the available data do not allow to conclude that the increase of the assisted delivery rate had a favorable impact on the MMR. A final analysis, this time in terms of international comparisons, shows that in Benin, the link between the RAD and maternal mortality is unusually weak. On the diagram below, we can see that Benin records a maternal mortality rate excessively high as compared to its assisted delivery rate (one of the highest in Africa). This suggests that the problem is less that of the assisted delivery rate than that of the quality of provided services. This essential point is explored further below. Figure 33 : Assisted delivery rate and maternal mortality in the world BENIN Source: A Review of the Evidence, 2001: Vincent de Brouwere & Kim van Lerberghe 2. Increasing the RAD remains nevertheless an important objective for the reduction of maternal mortality with the poorest and it goes through the reduction of financial obstacles and the improvement of the geographical distribution of midwives 60 In any circumstances, increasing the RAD cannot be the only strategy for achieving the MDG No 5, even if this must remain a priority in order to reduce inequalities. When we reconsider the (above) relationship between the assisted delivery rates and the MMR, a RAD of 105% (!) should be reached to achieve the MDG No 5. However, while increasing the assisted delivery rate (RAD) may not be key for the general population, it is certainly necessary for the poorest households. In fact, disparities of the RAD, especially according to the socio-economic status, remain considerable. While the richest quintile reaches 97.4% of assisted deliveries, the rate is only 55.8% for the poorest quintile (DHS 2006 data). Disparities per region are comparable. In the whole population of Beninese women, the decision to give birth with the assistance of qualified staff is henceforth a question of socio-economic status or education. Indeed, availability of qualified personnel plays a limited role. In order to explain the decision of assisted delivery, we made a regression of the probability of that decision on a set of explanatory variables. The results of the (logistic) regression are the following: Table 8: Logistic regression of assisted delivery on various factors Assisted deliveries Odds Ratio Standard p>|z| dy/dx* Error BEmOC services availability index 0,28 0,023 0,000 -0,160 BEmOC services availability index and 1,25 0,033 0,000 0,028 density of midwives (interaction) Socio-economic status (Ref.: Poorest) Q2 1,75 0,130 0,000 0,063 Q3 2,56 0,204 0,000 0,099 Q4 4,73 0,451 0,000 0,146 Richest 13,76 2,489 0,000 0,191 Education of the mother (ref. No education) Primary 1,63 0,156 0,000 0,056 Secondary and higher 3,56 0,958 0,000 0,111 Region (Ref. : Alibori) Atacora 0,28 0,034 0,000 -0,223 Atlantique 17,43 3,731 0,000 0,183 Borgou 0,98 0,104 0,839 -0,003 Collines 4,49 0,659 0,000 0,122 Couffo 0,28 0,046 0,000 -0,219 Donga 0,40 0,058 0,000 -0,149 Ouémé 1,22 0,194 0,209 0,024 Plateau 0,52 0,082 0,000 -0,102 Zou 5,82 0,862 0,000 0,140 Birth order 0,95 0,011 0,000 -0,006 Total number 8975 Chi2 of Hosmer-Lemeshow 1170 (0.0436) Area under the ROC curve 0.8259 * Marginal effects after logistic regression of the probability that a woman be aided during her childbirth = 0,84987 Source: DHS 2006 data, Ministry of Health and WB analysis We notice that, regarding the general population of Benin, supply-side factors (i.e. availability of BEmOC services and density of qualified staff) play a very limited role. It is firstly the socio-economic status and education that have an influence on the decision to deliver with the assistance of the qualified staff or not. Given the importance of the socio-economic status effect, one may think that increasing the assisted delivery rate within the general 61 population requires a reduction of the cost paid by patients (either by improving efficiency of maternity hospitals so as to lower their costs, or by establishing a mechanism for subsidizing payment of the expenses; this may be a health insurance mechanism or a program for free childbirth). As far as the poor population (the poorest quintile) is concerned, supply-side factors play a much more significant role. Indeed, the same regression, when it is limited to the poorest population, shows that supply-side factors are more important than in the previous regression. This is particularly true for the density of midwives. In other words, if the poorest have a relatively low assisted delivery rate (55.8%), it is firstly because there are no sufficient midwives in their region. It is also probably because they do not have enough revenues to be transferred to an urban region to deliver thereto. Table 9: Logistic regression of assisted delivery on various factors (within the poorest quintile) Assisted deliverys Odds Ratio Standard error p>|z| dy/dx* Density of the midwives 4,75 0,545 0,000 0,388 BEmOCs availability index and density of 1,39 0,087 0,000 0,081 midwives (interaction) Education of the mother (ref. no education) Primary 2,01 0,293 0,000 0,166 Secondary and higher 12,36 13,35* 0,020 0,406 Region (Ref. : Alibori) Atacora 1,31 0,17 0,043 0,067 Borgou 0,58 0,09 0,001 -0,133 Collines 1,77 0,34 0,003 0,136 Couffo 5,37 0,98 0,000 0,344 Donga 1,72 0,38 0,013 0,131 Ouémé 10,91 2,04 0,000 0,433 Plateau 2,17 0,39 0,000 0,182 Zou 0,86 0,18 0,488 -0,036 Total number 3792 * Marginal effects after logistic regression of the probability that a woman be aided during her childbirth = 0,53 Source: DHS 2006 data and WB analysis Overall, the simple increase of the assisted delivery rate, which is already high in Benin, will probably have a limited impact on maternal mortality in the general population. However, this quantitative objective must be pursued in order to reduce inequalities and to improve maternal health within the poorest. It is essential to increase the assisted delivery rate within the poorest and this goes through the increase of the density of midwives in the poorest regions. It is probable that a policy focused on demand-side factors (for example, making childbirths free) will also have a significant effect on maternal health of the poorest. 62 b. Quality of obstetrical care plays an important role to explain the high maternal mortality in Benin We saw that despite a high rate of assisted deliveries, Benin records a maternal mortality which is still high. This logically leads us to think that the problem lies in the quality of the childbirth aid. The example in the box below gives a good illustration of this issue. Box 1 - The death of Mrs. K, a pregnant woman. On December 18, 2006, Mrs K went to the maternity hospital of Y to deliver. She had three prenatal visits, the first being in September 2006, i.e. in the last quarter. When she reached the maternity hospital of Y, neither the midwife (in vacation), nor the nurse (absent for unknown reasons) were present. At 09:05, Mrs K is suddenly seized by eclamptic convulsions. The auxiliary nurse could not succeed in making that diagnosis, but she decided to quickly send the patient to the district hospital of S. Unfortunately, the director of the hospital had used the ambulance to go to a meeting and the ambulance was therefore not available for the transfer. At 09:10, after the prescription of diazepam to Mrs K, she was taken in a taxi to the district hospital. She got there at 10:07. A midwife was present. The gynecologist was absent for unknown reasons. Up till then, the eclampsia diagnosis was not made. And there was no drug for emergency situations anyway. The staff asked the husband to go and buy the drugs. He came back after the childbirth, which took place at 10:17. The child was stillborn. Mrs K did not get any postnatal assistance (she did not even receive the appropriate drugs). She died at 10:25, from respiratory distress. Still, the district of S is considered to be well staffed, and adequately equipped in ambulance, drugs and health facilities. Source: Audit report on the death of Mrs K, January 2007, Departmental Direction of Health (DDS) Z. Measuring quality of obstetrical services is particularly difficult. However, a good proxy indicator is the obstetrical case fatality rate 9. This rate was estimated at 3.5% in 2003, while the standard is 1%. If one is to examine the causes of this low quality more thoroughly, one can rely on the particularly enlightening study of Saizonou (2006), even though it is an old study. As mentioned above, in 2003, the 4 main obstetrical centers of Benin audited their maternal deaths10. The results of this analysis are presented in the table below: 9 This is the relationship between (i) the number of maternal deaths linked to obstetrical complications and (ii) the number of all pregnant women with at least one of the 7 obstetrical complications (recognized as such at international level). 10 52% of the maternal deaths of the country occurred in their centers. 63 Table 10: Causes of maternal deaths in 4 Beninese hospitals (2003) Deaths Deaths due to a bad care Main causes of non-quality N % N % Postpartum hemorrhage 55 23% 32 58% - clinical errors at primary care level (in diagnosis and treatment of high-risk situations) - blood shortages Infections (majorly linked to 54 23% 34 63% - antibiotic therapy is not administered because of intentional abortion) shortage in medicine (or financial incapacity of the families to buy the medicines) Hypertension (including 36 15% 15 41% - clinical protocols were not respected at primary eclampsia) care and referral care levels Dystocia (prolonged labor) 18 8% 7 39% Other direct obstetrical causes 16 7% 12 75% Anemia 35 14% 24 68% - blood shortages Other indirect causes 25 10% 19 76% TOTAL 239 100% 143 60% Source : Saizonou 2006 We can also notice that 60% of the deaths are linked to a low quality of care. Among these causes of non quality, two stand out: (i) low competence of staff and (ii) shortage in drugs and blood bags. For a further analysis, we selected three proxies of quality of obstetrical care: (i) competency level of clinical staff, (ii) availability of drugs and equipment and (iii) access to Complete Emergency Obstetrical and Neonatal Care (CEmOCs11) services, measured through the utilization rate of c-sections. 1. Relatively low competence of the staff The knowledge and capacity tests show that skills of clinical staff are still insufficient. In Benin, the best study on staff skills is the one conducted in 2003 by the Quality Assurance Project (Gbangbade 2003a). The main results are given below: Table 11: Level of skill and knowledge of clincial staff in the field of obstetrics Source: Gbangbade 2003a We notice that staff skills are very low on the whole. When we come back to the example of eclampsia (see the death of Mrs K), it appears that only 54% of the medical staff fully know 11 CEmOC services include the BEmOC services as well as two essential interventions for the survival of pregnant woman: (i) blood transfusions and (ii) caesareans. In Benin, only the Departmental Hospital Centers (CHD) and the District Hospitals (as well as some private hospitals) are entitled to provide CEmOC services. 64 about diagnosis and treatment of this situation. Yet, this is a staff working at a referral level. It is therefore probable that the knowledge of staff at primary level would be lower. The same study also analyzed the reaction time of the clinical staff to diagnose and treat the obstetrical complications. For example, on average, four hours passed between the diagnosis of eclampsia and the administration of anti-hypertensive drugs. Another example - which is more worrying - is the time period between the prolonged labor and the caesarean section, which was 8 hours. These results are confirmed by an assessment of BEmOC and CEmOC services in Benin (Gbangbade 2003b). Only 7% of the potential BEmOC12 facilities are actually able to provide BEmOC services. Out of the 6 functions necessary to be considered a real BEmOC facility, two functions were generally missing: the womb evacuation operation and the delivery by fetal vacuum extractor or forceps. In more than 80% of the cases, the interviewed staff recognized that not only they did not have the necessary equipment but also they were not competent enough to perform these two procedures (although these procedures are quite basic ones). 2. Drugs, equipment and especially blood are rarely available for emergency cases A certain number of basic equipment is often missing. The Gbangbade study (2003b) showed that there is often a poor supply in some simple equipment (and not necessarily the most expensive ones). This is particularly the case of thermometers, tensiometers, and cupping glasses. Given that these equipments are not very expensive, one can wonder if these shortcomings are not linked to a mismanagement of the equipment inventory rather than to a lack of resources to purchase them). Figure 34: Availability of basic equipment in the potential BEmOC and CEmOC facilities in Benin Source: Gbangbade 2003b 12 These are health centers (especially the Commune Health Center and the Arrondissement Health Center) supposed to provide BEmOC services. 65 Similarly, basic drugs for treatment of obstetrical emergency cases are not available enough. The same study (Gbangbade 2003b) shows that there is a shortage of important drugs such as ocytocin (for delivery hemorrhages) or diazepam (for eclampsia cases) in the delivery rooms of half of the BEmOC facilities centers and even in CEmOC ones. Strangely, in the pharmacies of these facilities, the study found higher availability rates. For example, ocytocin is available in 80% of the pharmacies of the potential BEmOC facilities, but in only 30% of the delivery rooms of the same facilities. How could these medicines get lost between the pharmacy and the delivery room, if not because of a poor inventory management or because of drug pilfering by health workers ? Figure 35: Availability of basic drugs in the potential BEmOC and CEmOC facilities of Benin Source: Gbangbade 2003b The situation is even more catastrophic for blood bags. According to Gbangbade (2003b), less than 37% of the potential CEmOCs had blood supply. 3. Access to the CEmOC services is still limited The accessibility of CEmOC services is a multifaceted notion and is therefore difficult to measure. If a CEmOC service is to be accessible, (i) the staff at primary level care must quickly and accurately make the diagnosis justifying the transfer, (ii) an ambulance (with qualified care if possible) must be available and (iii) the CEmOC facility itself must be operational. The data presented above already gave a first description of this accessibility of the CEmOC services. However, in absence of national data on the pregnant women transferred to CEmOC level, we decided to use the c-section rate as a proxy indicator for the accessibility of CEmOC services. The caesarean rate is still low in Benin. According to DHS-3, the c-section rate was 3.6% in 2006. This is low if we refer to the recommendations of UNFPA (5% minimum). This rate is also highly variable. Across regions, it varies from 1.3% (Alibori) to 10.5% (Donga). As we shall see, the variation is also strongly depending on the level of education and socio-economic status of households. 66 Utilization of c-section is strongly correlated to the socio-economic status of pregnant women. We carried out a logistic regression with the same factors as for the assisted delivery rate. The results of the regression show that the socio-economic status of pregnant women plays a fundamental role, while the supply-side factors are much weaker. The low rate of c-sections is therefore a demand-side problem, i.e. this intervention is too expensive for a part of the population. This tends to justify the current policy for implementing free c-section. 67 Table 12: Logistic regression of caesarean birth on various factors Caesarean Odds Ratio Standard p>|z| dy/dx* Error Index of the availability of CEmOC services 0,64 0,115 0,013 -0,010 Index of the availability of CEmOC services 1,06 0,018 0,001 0,001 and density of doctors (interaction) Socio-economic status (Ref: Poorest) Q2 1,97 0,500 0,008 0,020 Q3 1,45 0,387 0,158 0,010 Q4 2,25 0,557 0,001 0,026 Richest 4,45 1,118 0,000 0,063 Education of the mother (ref. No education) Primary 1,67 0,245 0,000 0,015 Secondary and higher 1,80 0,328 0,001 0,018 Birth order 0,90 0,029 0,002 -0,002 Total number 8947 Chi2 of Hosmer-Lemeshow 948(0.9997) Area under the ROC curve 0.7358 * Marginal effects after a logistic regression of the probability that a woman delivers by caesarean =0.025 Benin’s policy on the reduction of maternal and neonatal mortality Benin prepared a policy for maternal and neonatal mortality in 2006 and officially adopted it in 2007. This policy specifically targets the 3 possible obstacles to the improvement of maternal and neonatal health, which are: (i) availability of BEmOC and CEmOC services (with very pragmatic proposals on task shifting to the benefit of midwives and on partnerships with the private sector), (ii) geographical accessibility (i.e. reinforcement of the referral system) and financial accessibility (i.e. establishment of an innovative tool for subsidizing health care demand) of these services, and (iii) quality of care (i.e. clinical training and supply in drugs). In addition, this policy identifies the health districts that must receive a priority support. It is therefore a model of that kind. The implementation of that policy started in 2008 with the support of UNICEF, UNFPA, and USAID. Maternal health: main observations The assisted delivery rate (RAD) is normally the main factor to explain maternal mortality. This RAD is high in Benin (78% according to DHS). Yet, with such a rate, Benin should normally record a clearly lower maternal mortality. Other factors such as quality of obstetrical care obviously play an important role. This insufficient quality concerns staff skills as much as availability of drugs and equipment. However, the increase of the assisted delivery rate remains an important objective for the reduction of maternal mortality within the poorest households. This requires increasing financial accessibility of obstetrical care and better distributing midwives across urban and rural areas. Since 2006, there has been a particularly relevant policy for the reduction of maternal and neonatal mortality. It must be supported in a massive way. 68 69 3.2. Nutritional status of the women “Small size” is rare with Beninese women. As regards pregnant women, a small size has a clear impact on maternal mortality13 and on neonatal mortality. In Benin, DHS-3 considered that the threshold of small size is 145 cm. The proportion of Beninese women whose size is under 145 cm is only 1.4%. The poor are more affected (2.5% in the first quintile), as well as some departments (like Donga with 5%). On the whole, the prevalence of small size is still low. In most of the cases, the body mass index (BMI) of the women is normal. About 9.2% of the women (15-49 years) have a low BMI (lower than 18.5). This rate is close to the African average. The factors explaining the BMI are similar to those of the small size. In this case, the residence region and the socio-economic status play an essential role. For instance, low BMI rate increases to 14.7% for the poorest. On the other hand, anemia affects a great number of Beninese women. In Benin, more than 61% of the women (15-49 years) suffer from anemia, of which 1% with a severe form. This rate is 75% with regard to the pregnant women (73% in 2001). The rate is 59% with the nursing women. It seems that the situation has not known any major improvement since 2001 when the anemia rate was at 64%. When we analyze the anemia factors, the region of residence sometimes plays a significant role, as well as the education level. However, there is no clear link between anemia and the socio-economic status. Table 13: Regression of the anemia rate with the women on various factors Level of hemoglobin (Anemia) Coefficient Robust standard error p>|t| Wealth index 0,0000036 0,00000356 0,039 Education of the mother Primary 0,683 0,697 0,327 Secondary and higher 11,523 2,287 0,000 Number of children born -0,279 0,122 0,040 Pregnant woman -13,468 0,770 0,000 Region (except Alibori) Atacora 5,502 1,097 0,000 Atlantique 2,897 1,012 0,004 Borgou 7,189 1,017 0,000 Collines 3,990 1,161 0,001 Couffo 4,496 1,053 0,000 Donga 4,274 1,458 0,003 Littoral -0,026 1,303 0,984 Mono 3,908 1,111 0,000 Ouémé -4,693 1,114 0,000 Plateau 0,111 1,431 0,938 Zou 3,935 2,057 0,056 Constant 112,650 0,975 0,000 Total number: 4750 Prob > F = 0,000 Source: DHS2006 13 The pelvis is narrower and therefore the complications of the childbirth are more frequent. 70 4. HIV/AIDS 4.1. Prevalence in the general population is stable since 2002 Prevalence of HIV-AIDS has stabilized around 1.2%. There are two sources of data concerning prevalence of HIV in Benin. The first is that of the “sentinel surveillance” system, which collects data transmitted by the prenatal consultation services. Set up in 1990 in the urban maternity hospitals, it has been extended starting from 2002, especially in the rural area. The data are produced on a yearly basis and processed by the National Program for the Fight against Aids (PNLS). The second source is the DHS. Only the last DHS (2006) included a measurement of the prevalence of the HIV-AIDS. By design14, the sentinel surveillance system tends to overestimate the HIV prevalence. In fact, it is estimated at 2% in 2006, while the DHS reaches 1.2%. In order to estimate the trends of HIV prevalence, we only have the data of the PNLS. We can notice, starting from 2002 at least, that the prevalence rate is stable, around 2% (PNLS). Figure 36: Evolution of HIV-AIDS prevalence Prévalence pondérée du VIH (données système sentinelle / Bénin) 2.2 2.1 2 1.9 1.8 1.7 1.6 1.5 2002 2003 2004 2005 2006 Source: CNLS, 2008. The epidemic is clearly influenced by gender. Indeed, prevalence is 1.5% (DHS) with the women versus only 0.8% (DHS) with the men. Unlike other countries, the young are less affected by HIV. The diagram below shows quite well that prevalence reaches its maximum around 40 with the women (2.5%). The tendency is less clear with men, even if in that case too, the elders have the highest prevalence rate (between 1.6 and 1.8%). 14 As its data come from prenatal services, the sentinel surveillance system includes neither men (who generally have a lower prevalence), nor women using condoms. 71 Figure 37: HIV prevalence rate according to age and gender (2006) Source: DHS 3 - 2006 The differences in prevalence per region and per type of residence are relatively important, at least for women. Figure 38: Prevalence of HIV-AIDS according to regions and gender Source : DHS-3, 2006 72 With regard to women, like in many other countries, prevalence is higher in urban area (2.2% according to DHS) than in rural area (1% according to DHS). The maximum is reached in Cotonou, with 2.8% (DHS). However, a rural region (Donga) has the maximum with 3.5% (according to DHS). We also find an urban-rural gradient with men, but with much smaller geographic differences. 4.2. HIV prevalence within the high-risk populations is decreasing Prevalence among sex workers (SW) seems to be decreasing. The PNLS monitors only one high-risk population: sex workers (SW). Within that population, the prevalence of HIV remains very high (25.5% in 2006 according to PNLS). However, it seems to have decreased during the recent years. There is no impact study to carefully explain this very good result. We can however assume that the Corridor project has been particularly effective, especially because it precisely targeted the lorry drivers and the prostitutes working all along the Lagos-Abidjan road. Figure 39: Prevalence of HIV-AIDS with sex workers Prévalence VIH chez les TS 60% 50% 53% 49% 40% 41% 30% 25.50% 20% 10% 0% 1993 1995-1996 1998-1999 2006 Source: CNLS, 2008. 4.3. Risk factors Education level plays a role in the prevalence among women. Women having a primary education level are the most affected. We do not find such a connection with men. It is interesting to observe that socio-economic status has no significant connection with the level of prevalence. Though HIV in Benin is basically transmitted through sexual intercourses (92% of the infections according to CNLS), the DHS results do not show any clear connections between the level of prevalence and high-risk sexual behaviors (extramarital sexual intercourse or a partner who does not live under the same roof, non use of condoms, etc…) HIV-AIDS: main observations 73 The HIV-AIDS epidemic seems to be brought under control today, and this, at a level which is relatively low for the region (1.2%). This success can be attributed to the vigorous policies conducted in that field, especially in coordination with neighboring countries. 74 Chapter 3 : COMMUNITIES & HOUSEHOLDS: PRACTICES & POLICIES 1. Nutritional practices and policies ........................................................................................................76 1.1. Micronutrients consumption.........................................................................................................78 Vitamin A........................................................................................................................................78 Zinc..................................................................................................................................................79 Iron ..................................................................................................................................................79 Iodine...............................................................................................................................................81 1.2. Breast-feeding and supplement foods ..........................................................................................81 2. Hygiene behaviors and access to clean water and to waste disposal equipment.................................83 Hand washing......................................................................................................................................83 Access to clean water ..........................................................................................................................83 Elimination of Human Defecation ......................................................................................................84 3. Practices and preferences regarding fertility .......................................................................................87 Practices and knowledge related to fertility ........................................................................................87 Unmet needs in terms of family planning ...........................................................................................91 4. Utilization of health care system .........................................................................................................92 5. Policies for community-based health ..................................................................................................93 75 By « Community and household practices », we mean three types of practices or behaviors likely to have a major impact on population health. They include (i) food habits (ii) hygienic practices, conditions of access to potable water and waste disposal equipment, and (iii) utilization of healthcare system. This last point is dealt with in chapters 5 (health services) and 7 (medicines and vaccines). The last part of the chapter reports on the community component of the health system. 1. Nutritional practices and policies Nutritional policy in Benin: situation analysis 1. An institutional position that is still fragile In Benin like in many other countries, nutrition is still connected to food security issues. That is why the Ministry of Agriculture has been (and is still) the very first ministry to get involved in nutrition. Thus, it was at the origin of the creation of the Dahomean Service for Food and Applied Nutrition (SDANA) in 1962, which became DANA in 1974. On the field, its main activity was to create household gardens. They were assisted by two horticulture centers (Ouando and Pabegou) and by rural development coordinators, trained in nutrition. From 1974 to 1984, the control of DANA – together with the Benin nutritional policy – became an institutional quarrel between the Ministry of Agriculture and the Ministry of Health. DANA was successively linked to these two ministries. The results obtained during the previous period were largely lost. DANA was definitely attached to the Ministry of Agriculture only in 1984. The Ministry of Family and Social Affairs developed nutritional activities from 1987, especially with Social Promotion Centers. The strong involvement of the Ministry of Health in nutrition came up late, as it was only in 1994 that a service in charge of nutrition was created within the Ministry. Similarly, the involvement of the Ministry of Health in the preventive aspects of nutrition was not a reality before 1997, when the Essential Care Package (PMA/NUT) had been introduced at the USAID’s instigation. The coordination of activities remains very limited. A National Committee for Food and Nutrition (CNAN) was created in 1994 but was unable to carry out its mandate especially because of lack of regular budget for its operations. The future creation of a HCAN (Haut Conseil pour l’Alimentation et la Nutrition / High Council for Food and Nutrition) should put an end to several decades of quarrels over the institutional positioning of nutrition. 2. A multisectoral policy yet to be developed What remains to be done shall be the development of a multisectoral nutritional policy, which is not yet available in Benin. A National Action Plan for Food and Nutrition (PANAN) was developed in 1995 but proved to be too ambitious and most importantly, could not be linked to any significant financing. Its implementation then remained limited. Currently, only the Ministry of Health has a nutritional policy (around the PMA/NUT). It is advisable that the other ministries involved should also prepare similar documents to be integrated into a national and multisectoral strategy. There is also an opportunity in the fact that for the first time, nutrition is henceforth integrated as a Government objective in the Poverty Reduction Strategy Paper (DRSP 2007-2009). As a whole, actions related to nutrition seem to almost entirely depend on foreign financing. This situation tends to feed conflicts among the three ministries involved, each of them striving to obtain specific financing from one foreign partner or another. 76 3. Some previous experiences were often successful, but remain to be integrated, sustained and replicated Often times in nutrition, several experiences are conducted, especially in the community area, and are led (together with foreign partners) without planning for their sustainability. It is clearly the case with the first nutrition project (“Family Gardens”) which quickly stopped after 1974 when DANA became a major stake among many ministries. Many other important projects were implemented by the Ministry of Family and Social Affairs, viz: the National Food Program (PAN) from 1987 to 1997 and the Community Based Food and Nutritional Program (PBC) from 1997 to 2005. Those two projects allowed improvement of the nutritional monitoring of 60.000 children through CPS and Community relays. Food distribution was one of the main activities of these projects; and of course, this encouraged women to better monitor the nutritional situation of their children, but led to negative consequences on the other hand. Apart from usual problems of implementation (i.e. distributed food that are sold by mothers or given to adults rather than to children), these projects have contributed to develop the idea that nutrition is first of all a matter of food availability (and then of economic growth). The community component of the Food Security Local Response Project (FSLRP) is another successful project implemented from 1996 to 2000. As for the PBC, community health workers were strongly involved and Village Nutrition Committees (CVN) - in charge of monitoring Project’s activities – were set up. The PILSA also included income generating activities for mothers; this allowed them to diversify their children’s food. Owing to this project, 20,000 children’s nutritional status improved. It was also one of the really multisectoral nutritional projects in Benin. The Ministry of Health is striving to maintain two programs. The first one is the community component of the Integrated Management of Childhood Illnesses (IMCI). It includes many important nutritional activities. The second program is the one of PMA/NUT. Unfortunately, both programs are not managed in an integrated way in spite of their strong potential synergies and despite the fact that they are controlled by the same directorate of the Ministry. Moreover, these two programs seem to be under financed. Brief background of the institutions involved in nutritional policy and activities in Benin Services nationaux de nutrition Politique Instance Instances Ministère de Ministère de Ministère de la Famille nationale de coordination de coordination l’Agriculture la Santé et des Affaires Sociales en nutrition nationale départementales SDANA 1962 Années 60-70 DANA 1974 CDMAN 1984 Direction Années Enfance et 80-90 Adolescence PANAN CNAN CDAN Service nutrition 1995 1994 1994 ? DSF 1994 Programme HCAN Aujourd’hui National 2009 ? Multisectoralisé 2009 ? CDAN: Comité Départemental pour l’Alimentation et la Nutrition CNAN: Comité National pour l’Alimentation et la Nutrition CDMAN: Comité Départemental Multidisciplinaires pour l’Alimentation et la Nutrition HCAN: Haut Conseil pour l’Alimentation et la Nutrition PANAN: Plan d’Action National pour l’Alimentation et la Nutrition SDANA: Service Dahoméen d’Alimentation et de Nutrition Appliquée Sources : Agbato 2008 et Mulder-Sibanda 2004 77 1.1. Micronutrients consumption Vitamin A A large part of children regularly consume vitamin A enriched food or vitamin A supplements. A regular consumption of vitamins allows the reinforcement of the immune system of children. It also helps prevent twilight blindness. Among 6 to 35-month-old children, 70% consumed vitamin A enriched food (within the 24 hours preceding the DHS survey). Also, 60% of 6 to 59-month-old children have been given vitamin A supplements. These relatively positive indicators are linked to the existence since 2001 of vitamin A supplements distribution campaigns (coupled sometimes with vaccination days). The table below shows the great efficiency of these campaigns15. Compared with the gap related to the education level of mothers and the one related to the socio-economic status of households, the gap linked to the regions of residence is more important. Across regions, consumption of vitamin A enriched food varies from 83% (Plateau) to 42% (Alibori). At the same time, as regards vitamin A supplements, the rates run from 84% (Collines) to 41% (Plateau). Table 14: Percentage of children from 6 to 59 months old supplemented in vitamin A during vaccination national days Years Sessions 1st 2nd 2003 98% 94% 2004 89% 106% 2005 92% 94% 2006 94% 88% Source : Ministry of Health On the other hand, only 30% of women usually consume vitamin-A-enriched foods, whereas 41% have been provided a vitamin A dose within the two months following their childbirth. As for this vitamin A dose, gaps across regions are important, running from 56% in Atlantique to 18% in the north. Vitamin A: which type of policy is in place in Benin? For 6 months and more old children As previously indicated, the supplementation of children of more than 6 months old in vitamin A has been put in place in 2001 and is still being implemented adequately. Although the supplemented children’s rate is different from one year to the other (see table below), it remains high (around 90%) and constitutes an important experience for the Benin health policy. Additional policies are being developed, especially for food16 fortification and food diversification. For young children (0-5 months) When they are breastfed, these children are of course, exclusively dependant on their mother’s milk concentration in vitamin A. So far, results obtained in Benin are not better than those for 6-month-old children. As a matter of fact, 15 This scheme is a replication of a pilot launched in 1997 by USAID. 16 Agronomic tests are ongoing for fortifying manioc. 78 the supplementation of women in vitamin A remains low, probably because of the weak postnatal monitoring in general (see chapter 5 for details). Zinc Zinc consumption is not known. Zinc deficiencies lead to growth retardation of children. They also cause diarrheas and pneumonias. Unfortunately, no information is available on zinc consumption in Benin. Nevertheless, one can hypothesize (Hotz 2004) that its consumption is comparable with the one of iron, given that food with high concentration of iron and zinc are basically the same. Iron Consumption of iron enriched food remains low, with only 53% of children being regularly fed with iron enriched food. Anemia is prevented through (i) breast-feeding (if the mother is sufficiently provided with iron) and (ii) through regular consumption of iron enriched food (particularly for children that are more than 6 months old). The consideration of this last point is not satisfactory at all for Benin. A very low rate of children is provided with iron enriched foods. And the anemia rate has not substantially improved, decreasing from 82% in 2001 to 78% in 2006. As in the case of vitamin A, geographical differences are more important than the differences related to the education level of mothers or to socio-economic status. At a regional level, the rate of 6 to 35-month-old children having consumed iron enriched food runs from 76% (Littoral) to 23% (Atacora). Less than 25% of women have an iron enriched diet whereas only 53% have been provided with iron supplements during more than three months of pregnancy. This last rate was of 40% in 2001, which shows a substantial increase. Yet, as it was indicated (see Chapter 2), the anemia rate of pregnant women has not really changed since 2001. Such a situation can be due to the high rate of malaria. As for iron supplements, the differences between regions are once again very important, varying from 91% at Cotonou to 22% in Alibori. At the same time, in the poorest quintile, only 32% of women have consumed iron supplements. Iron: which policy in Benin? For 6 months and older children As we saw earlier, the consumption of iron enriched foods by children remained limited. Many actions have been carried out in this area without really positive results. The very first action was the development of strengthened flour by DANA (owing particularly to “family gardens”). Their use by the populations is not well known and is probably low. The second action is food diversification. This relates to the consumption of more animal origin foods, enriched with iron. It seems that this principle is not usually applied in households due to many reasons such as poverty, ignorance and cultural sluggishness (see Agbota 2008). There are no policies of iron supplements distribution to children. 79 For young children (0-5 months) As for vitamin A, children are seriously dependant on their mother’s diet. And as in the case of vitamin A, the results are very low. Yet, iron tablets are distributed during prenatal visits; which means that as in the case of vitamin A, the problem may be related to the lack of postnatal monitoring. 80 Iodine Consumption of adequately iodized salt is insufficient, following an unusual urban-rural gradient. Less than 60% of children from 6 to 59 months old are living in a household provided with adequately iodized salt. This rate is unexpectedly higher in rural areas (61%) than in urban areas (53%). For instance, this rate runs from 30% at Cotonou (Littoral) to 91% in Borgou region. At the same time, the gradient is inverted when we consider the education level or the socio-economic status of households. What is more worrying is the fact that the rate of households consuming adequately iodized salt is decreasing due to the laxity of health policy in this area. Many campaigns to promote consumption of iodized salt have been organized from 1995 to 2000 with the support of UNICEF. But obviously, these efforts have not been pursued, since, according to DHS data, from 2001 to 2006, the percentage of households consuming iodized salt substantially decreased from 72% to 55%. Such a drop is more important in the northern part of the country (i.e the rate dropped from 82% to 40% in Borgou region). Iodized salt: which policy in Benin? Two policies supported by UNICEF have been set up in 1994 but their implementation has slowed. The first policy aimed at requiring traders to sell only iodized salt either through direct importation of iodized salt or by iodizing locally-produced salt. Today, about 80% of salt consumed in Benin comes from importation and is then already iodized. On the other hand, as for locally produced salt, producers are still reluctant regarding iodination, particularly because of the additional transportation costs. The two iodination units (at Ouidah and Comè) are indeed far from the production sites. The second policy is more usual and consists in campaigns of promotion for the consumption of iodized salt. 1.2. Breast-feeding and supplement foods In order to maximize growth and development of children, UNICEF and tWHO urge that children be (i) breast-fed from the very first hour after their birth (ii) exclusively breast-fed (without any supplement food) till the age of 6 months (iii) breast-fed with supplement foods till the age of two years. Slightly more than 50% (54%) of children are breast-fed from their birth. The rate is a little higher with children born with the assistance of qualified health staff (58% vs. 35%). Considerable progress has been made regarding exclusive breast-feeding of children who are less than 6 months old which reached 43% in 2006. The rate was only of 10% in 1996. 81 Then, it increased to 38% in 2001. Such a remarkable progress shall be attributed to the national strategy relating to the Feeding of Baby and Young Infant. Fifty percent of less than two years old children enjoy a continuous breast-feeding. The average duration of breast-feeding for half of these children is 21.4 months. One must stress the fact that this duration has been slightly reduced since 2001 dropping from 22.3 months to 21.4 months. Moreover, supplement foods are not provided systematically. For example, one third of 6 to 8 months old children do not receive any supplement foods. As a whole, less than one third (32%) of 6 to 23 months old children are fed according to the three recommended practices. This rate drops to 15% in the northern departments. Yet, even in the urban area of Cotonou, the rate is not more than 45%. On the other hand, the differences related to the education level or the socio-economic status of mothers are relatively low. Policy promoting breast-feeding and food supplements in Benin. Various actions have been implemented during the second part of the 1990s. Three of them had a direct impact on the exclusive breast-feeding rate: (i) the Exclusive Breast-feeding Promotion Strategy (since 1998), (ii) the national strategy relating to the Feeding of Baby and Young Infant and (iii) the Baby-Friendly Hospital Initiative. Nutritional behaviors and policies: main findings Although some nutritional indicators were satisfactory in Benin, two major problems remain to be solved regarding children: (i) the prevalence of stunting and (ii) the prevalence of anemia. We can reasonably suppose that the first problem is related to breast-feeding and zinc supplementation deficiencies. The second problem is due to the low consumption of iron enriched foods, including by pregnant women. These various points should draw a particular attention that a multisectoral nutritional policy needs to be developed. One of the conditions to the elaboration of such policy is to set up a coordinating unit of nutritional actions from each ministry. 82 2. Hygiene behaviors and access to clean water and to waste disposal equipment The impact of hygiene and sanitary practices on diarrhea prevalence We now know that good hygienic practices (hand washing), access to clean water and adequate disposal of human defecation have a considerable impact on diarrhea prevalence. The table below shows the most recent estimates of such impact: Area to improve % of diarrhea prevalence reduction Source Clean water supply17 (more to wash than to drink) 63% Cairncross (2006) Promotion of hygiene (Hand wash with soap) 43-48% Curtis (2003) Sanitation 36% Esrey (1991) Hand washing In spite of the important progress made, less than 28% of mothers wash their hands before feeding their children and after disposing of their feces. This rate reaches 42% at Cotonou and at most 1% in Atakora region. Nevertheless, one must notice that this national rate was 4% in 2001, showing that progress are consistent. Access to clean water Despite it was already low, Benin households’ access rate to drinkable water has decreased since 2001 particularly in rural areas. On average, less than 38% of households are connected to a water supply network. Such access rate was 46% in 2001. It is in rural areas that the rate has considerably dropped to half, from 28% to 14%. Today, this percentage runs to 92% at Cotonou but comes down to 20% in the northern regions. Access to a « modern »18 source of drinkable water supply is a reality only for 35% of the population. This already low rate hides considerable differences as it reaches 98% at Cotonou and 17% in Atakora region. 17 Clean water is provided by an « improved » source. See next footnote. 18 The Joint Monitoring Program for Water Supply and Sanitation (JMP) of WHO and UNICEF defines an « improved » source for clean water as the following: tap water, protected well or rain water. Other sources of water are susceptible to be contaminated by germs. 83 Elimination of Human Defecation Benin is one of the African countries that are most behind regarding waste disposal and particularly fecal waste. The charts below show that out of 24 African countries, Benin has a very high rate of open air defecation, i.e. 68% (the worst method of fecal waste disposal) and one of the lowest rates as for the use of traditional latrines (15%). Only Niger, Chad and Burkina Faso are in a worse situation. And what’s the more, the situation in Benin is improving very slowly. In 2001, the rate of open air defecation was still around 72% whereas the rate of use of modern latrine dropped slightly from 17% to 15%. Figure 40: Human waste elimination modes in 24 African countries Flush toilet coverage VIP latrine coverage by country by country 50% 46% 35% 45% 29% 40% 30% 27% 36% 25% 35% 31%31% 25% 23%22% 30% 21%21% 25% 20% 18% 25% 18% 15% 20% 14%13%13% 15% 12% 13%12% 11% 15% 10% 10% 9% 8% 10% 8% 10% 6% 6% 5% 4% 3% 3% 3% 2% 4% 4% 5% 2% 2% 2% 2% 2% 2% 1% 1% 1% 1% 5% 3% 3% 3% 2% 2% 2% 2% 1% 1% 0% 0% 0% 0% DRC CAR otho o frica d ya an in pia on r ire Mali bia ri a na ea nda nda n a) ique we ia ibia ia wi l Nige ega Fas Cha er oo Be n z an r azz ri tan Co k ina F s Ke n R C o ad Sud Ma e nya ia as o Guin ica Les n dan er Gha Mala Gab a ana t e d nin i re Na ria a bia Zim oon li a) da Nige d’Ivo we Tan nia ia ue Eth wi Ethio ia bab Zam o ot h and Mo Guine bo CA DR Ma Na m Uga Rw a iop Les mib th A Nig Bu omor azz zan Sen a mb Ch an la e ’Ivo biq A fr ina bab Zam Be Tan ta Ca m Gh Su Ga go (B Ni g r Mau Zim Ma K me Ug Rw u ri (Br Cote za m Sou Burk uth Moz C Ca ngo Con So r Co Traditional latrine coverage Open defecation by country by country 90% 90% 81% 80% 79% 79% 80% 76% 75% 80% 70% 67% 72%70% 66%64% 68% 70% 62% 70% 59% 59% 58% 62% 60% 53% 51% 60% 57% 48% 49% 50% 44% 50% 47%45% 41% 39% 43% 40% 35% 34% 33% 31% 40% 35% 30% 24% 28%28%27% 30% 25%25% 15% 15% 22%21% 20% 18% 10% 20% 15%14%14% 8% 7% 13%12% 10% 10% 7% 10% 3% 2% 0% 0% 0% Br o s C Ni g R S e t ho ad Na s o a za m on er ia ia i ca li Ta nda d’ Iv a Zi m e nin da Za n Rw ea me i a a) ue e rk in bwe ia bia i l ia DRC ga l aw ny na ya d DR CA Ma nda n in an on iopia ibia ia er oi r o ea eria a r oo l Mali a nia ) wi bia S o thi op mb o on oire we s Ni g ega er ique ng om or an tan azz zza a Cha Ch fric in Co Gha Fas Mo Gab and an b iq oth zan A fr Ben oro so Nig Ken mi ne Gha Sud Guin Gab aF Ke Mala a Zam ero ba rita B Ma Gu Uga bab Nam nz Nig Ug (Bra d’Iv Sen uri th A Le Eth Les Rw Com a mb Tan ina uth E Cam Mau C o( Zim Ca Ma te Burk Cote Sou go Moz Bu Con Co Source : Africa Infrastructure Country Diagnostic, 2008 Rural areas, particularly in the North of Benin, are the less equipped in sanitation. DHS data of 2006 show the difference between Cotonou (where the rate of open air defecation is only 13%) and the rural areas (where this rate reaches 89%). The difference is particularly important in the northern regions: Atakora/Donga and Borgou/Alibori. 84 Table 15: Modes of elimination of Human defecation in Benin (2006) Cotonou Other cities Rural Flush toilets 9.4 0.8 0 Improved Latrines 49.3 21.7 2.4 Traditional Latrines 27.8 27.3 5 Open air 13.4 47.9 89.2 Non defined 0 2.3 3.4 Source : DHS 2006 Figure 41: Human defecation Elimination modes in Benin (2006) 100% 90% Undetermined 80% 70% Open air 60% 50% Traditional 40% latrines 30% 20% Improved latrines 10% 0% Flush toilets Cotonou Autres Rural villes Source : DHS 2006 Only 26% of mothers get rid of their children’s feces in a hygienic way (in toilets or latrines). Unsurprisingly, this rate reaches 75% at Cotonou but falls to only 12% in rural areas. The socio-economic gradient is particularly important, given that the rate is 74% for the richest quintile and only 5% for the poorest one. Whereas Benin is making relatively large investments to improve its sanitation equipments, the efficiency of these investments seem to be small. Considering the sample of 24 African countries, Benin is one of those that invest most in sanitation. Yet, it is obvious that the results are still very insufficient. On the charts below, one can see that Mozambique and Malawi have the same results as Benin although they invest six times less than Benin. 85 Figure 42: Annual Investment (per capita) in sanitation VS annual increase of human waste disposal methods Source : Africa Infrastructure Country Diagnostic, 2008 Hygiene and sanitation: main findings In Benin as in most other African countries, 17% of children mortality is due to diarrhea, a pathology which is strongly related to hygiene and to quality of sanitation. Benin situation is really worrying. In fact, although obvious progress has been made regarding hand washing, the situation is worsening (i) as for access to water supply networks (especially in rural areas where such rate has fallen from 28% to 14% between 2001 and 2006) and (ii) regarding human waste disposal (as Benin shows one of the worst results in Africa with Chad, Burkina Faso and Niger). 86 3. Practices and preferences regarding fertility Practices and knowledge related to fertility Despite the high level of information on contraception methods, their use remains very low. According to DHS 2006, 90% of women know at least one modern method of contraception. The level of knowledge is higher with men (96%). There is a gradient linked to education and socio- economic status but it is not strong. For instance, in terms of education level, 87% of illiterate women know one modern method of contraception Vs 99% of women having reached at least secondary school level. As for wealth, 89% of poorest women (1st quintile) know at least one modern method of contraception Vs 98% for wealthier women (5th quintile). Yet, the modern method use rate has stagnated around 7% from 2001 to 2006. One must note that, contrary to the knowledge level, the use rate (modern methods) varies according to the education level (4% for illiterate women Vs 19% for most literate) and the wealth level (2.4% for the poorest Vs 13.2% for the richest). Considering only the rate of use of modern contraceptive methods (the most effective ones), Benin is lagging behind many countries, coming just before Sierra Leone and Niger. The worst situation is that Benin is now one of the rare African countries where the use rate of modern methods of contraceptive has not increased during the last years. Figure 43: Use rate of modern methods of contraceptive Sub-Saharan Africa Ghana Togo Liberia Senegal Gambia, The Burkisa Faso Mauritania Nigeria Côte d'Ivoire Western Africa Mali Guinea Guinea-Bissau Benin Niger Sierra Leone 0 2 4 6 8 10 12 14 16 18 Source: Population Reference Bureau (2008): 2008 African Population Data Sheet. Washington, DC: PRB 87 Figure 44: Evolution of the use rate of modern contraceptive methods 35 Kenya Malawi Zambia 25 Tanzania Ghana Uganda 15 Ethiopia Rwanda Senegal Nigeria Benin Mali 5 1985 1988 1991 1994 1997 2000 2003 2006 -5 Source: USAID 2008 Data from the health information system confirm that the contraceptive methods prevalence does not progress much or even at all. The NSHI follows the number of women « accepting » to use a modern contraceptive method. The chart below shows that even by disregarding the temporary increase related to the PSI project, the activities of public health centers (FHC) and services of the Benin Association for Family Planning have fallen considerably since at least 2002. 88 Figure 45: Number of women accepting to use a modern contraceptive method 300,000 PSI ABPF 250,000 FSP 200,000 150,000 100,000 50,000 - 2001 2002 2003 2004 2005 2006 2007 Source: MoH / Health statistic directory Half of the women refusing to use modern contraceptive methods do not have appropriate information. In the case of DHS 2006, women who are not using contraceptive methods – i.e. the majority of women – have been questioned about the reason motivating their decision (see the table below). Putting aside those who think they do not need such methods (barrenness, non frequent sexual intercourse), the various answers can be classified into two equal categories (i) women who wish to have many children and then don’t plan the use of any contraceptive methods (33.2%) and (ii) those who, on the contrary wish to use a contraceptive method but either, lack information (27.6%), or have financial difficulties or others (3.8%). As a whole, we must note that problems of access to Family Planning services play a minor role. Conversely, it is probable that a better way of informing women would highly increase the use of contraceptives. Table 16: Reasons that prevent women from using contraceptives 1. No need in FP, because no desire or capacity to have children 31.8 Sexual intercourse not frequent 7.5 Menopauses or hysterectomy 5.6 Infertility 18.7 2. Wants children and refuses FP 33.2 Wants as many kids as possible 12.9 Woman refuses 13.1 Spouse refuses 3.2 Other refuse 0.2 FP not allowed by religion 3.8 3. Need in FP, but concerns about the methods 27.6 89 Do not know any method 3.7 Fearing health issues 23.9 4. Need in FP, but cost or accessibility problems 3.8 Too far 0.1 Do not know about any provider 2.4 Too expensive 1.3 Other 3.6 Source : DHS 2006 Regarding information for women, the family planning system seems not strongly efficient. Considering, for instance, the exposure level of populations to various messages on family planning (radio, TV or newspapers), it is observed (DHS 2006) that 48% of women have not heard of any family planning message during the last three months. The gradient related to education and socio-economic levels is relatively high. For example, in terms of socio-economic status, 66% of the poorest women have never heard of such message Vs 26% for the richest ones. What is more worrying is that 90% of women who are non-users of contraceptives have neither received the visit of a health agent nor ever discussed of family planning when visiting a health centre. Among those who went to a health centre, less than a quarter received information on family planning. It is interesting to note that those rates are almost identical whatever the education or wealth level of women. This observation points out a major deficiency in connection with the supply of family planning services. 90 Unmet needs in terms of family planning For about a decade, overall needs (i.e. met and unmet ones) in terms of family planning (FP) have remained stable. During the DHS 2006, 27% of women stated they did not want children anymore (birth limitation) whereas 38% wish to space out their next birth by at least 2 years (birth spacing out). The proportion of women in need of family planning is then of 65%. These needs are almost the same as in 1996 and 2006, which prove that household preferences in terms of fertility have not been impacted by the existing policy. We must also note that the needs in family planning do not change much according to the various social categories of the populations. For instance, the need in birth limitation (an average of 27%) was varying from 25% in rural areas to 30% in urban ones. This rate was 26% for illiterate women and 34% for the most literate. In short, the rate is not different among socio-economic quintiles (except for the wealthiest). Unmet needs in family planning have increased since 2001. Almost 30% of women consider their needs in family planning as unmet. The rate is increasing as it was 27% in 2001 but is relatively consistent within the population. As a matter of fact, differences are not perceptible from urban to rural areas or among socio-economic quintiles. On the other hand, it changes appreciably according to the education level, dropping from 30% for illiterate women to 19% for the most literate women. With regard to other West African countries, Benin has one of the highest ratios of unmet family planning needs (30% in Benin Vs 23% in the sub-region). Figure 46: Rate of women with unmet needs in family planning (2006) % de femmes avec des besoin en PF non satisfaits Niger Nigeria Guinea Western Africa Sub-Saharan Africa Guinea-Bissau Côte d'Ivoire Burkisa Faso Benin Mali Senegal Mauritania Ghana Liberia Togo 0 5 10 15 20 25 30 35 40 45 Source : Population Reference Bureau (2008): 2008 African Population Data Sheet. Washington, DC: PRB 91 Family planning in Benin: key points and recommendations Classically, there are two types of fertility determinants. A first group is the one of indirect (intermediary) determinants, that is to say, socio-economic determinants. They are the most powerful in terms of impact but they are dependant of long term evolution and are not easily modifiable. One of the most important determinants is women education, for example. Direct (or immediate) determinants – or biological and behavioral determinants - make the second group. Although their impact is low, they are more easily activated by demographic policies. Among these direct determinants, are included the contraceptive methods as well as information and education programs (IEC) to change perceptions and behaviors. From that point of view, despite the adoption of a Population Policy Statement (PPS) in 1996, progress in the area of family planning is still slow in Benin. We already mentioned that the rate of use of contraceptive methods is stagnant at very low levels and that unmet family planning needs have increased since 2001. It is true that birth control does not seem to be a national priority. Interestingly, none of the 16 targets in the PPS (reviewed in 2006) includes some reduction of fertility. On the contrary, the PPS puts an exclusive stress on the need of social and economic development to face demographic growth… Consequently, fertility reduction requires the implementation of a set of measures according to the following order: 1) Update and improve demographic simulations, taking into account the whole set of cases, including the absence of drop in fertility (which is the current situation); 2) Review the PPS to include the necessary control of demographic growth; 3) Advocate nationally on the demographic burden, so that its consequences will be taken into account in the Government strategic documents (particularly the Growth Strategy for Poverty Reduction); 4) Review national policies (particularly the National Plan for Health Development), in order to better take into account the demographic burden; 5) Prepare and implement a plan to strengthen information, education and communication (IEC) actions to change behaviors; 6) Improve training of health staff (public sector and AFSB) in terms of reproductive and family planning health; The main objective is to get an annual increase of at least 1.5% for the use of modern contraceptive methods, a rate which is nowadays at 7%, one of the lowest in Africa. 4. Utilization of health care system While it is obvious that poor people are more often sick, the type of health services used by households depends particularly on the (urban/rural) area and the region of residence. A survey (Ouendo 2005) gives the opportunity to have a better understanding of health care seeking behaviors in Benin. Among the 9,554 persons included in the survey, about 20% stated having suffered from sickness at least once during the last 12 months. This rate was significantly higher (25%) among the poorest19. On the other hand, the type of health care service used first depended mainly on the area of residence. The figure below shows clearly that health care seeking patterns are almost identical in rural areas, whether the person is rich or poor. In such areas, self-prescription of modern drug is the first health care seeking decision. The situation is significantly different in urban areas, where financial inequalities play a heavy part. In cities, poor people’s first health care seeking decision is self-prescription, whereas wealthier persons go to health centers. 19 In the Ouendo survey (2005), the poor were identified by community leaders. They do not necessarily correspond to the poorest quintile defined by the DHS. 92 Figure 47: Health care services utilization (1st contact) M H T N R P R N U P U N Source : Ouendo 2005 5. Policies for community-based health Community-based health policies have a formalized organization and identified volunteers. At the Ministry of Health, there is a department in charge of community health (SSC), responsible for developing and leading the community health workers (CR) networks. Each commune has village volunteers (VV) supervised by coordinators (CA). These community health workers are normally supervised by management committees at the commune and arrondissement levels. However, the number of community health workers seems to be very low, as it is estimated to 2000 persons, that is to say, one community health worker for 3,500 inhabitants. Most routine community-based activities are focused on diseases that are considered as « neglected » diseases, such as the eradication of Guinea Worm. Obviously, and despite the efforts of the SSC department, there is no significant national financing for integrated community activities that could cover all the interventions that can be implemented at community level (see Table below). The community activities are mainly performed by vertical programs, such as the eradication of Guinea Worm. To date, it is the program which has trained the largest number of community health workers. The SSC department developed a national training program, covering most of the interventions described in the table, except, once more, nutritional activities. POTENTIAL COMMUNITY-BASED INTERVENTION Intervention in Community Health IMCI20 ? Workers trained on the topic Exclusive breastfeeding (0 – 5 months) IMCI Additional feeding (6 months – 2 years) IMCI 20 IMCI: Integrated Management of Childhood Illness 93 Supplementation in micronutrients IMCI Growth monitoring Elimination of human waste and hand washing IMCI YES Bed net use IMCI YES HIV prevention IMCI YES ORT and liquid feeding for diarrheic children IMCI Signs of danger learnt by mother IMCI Infectious diseases care IMCI Child immunization IMCI YES Diagnosis and treatment of Guinea worm infection YES Diagnosis and treatment of onchocercosis YES Diagnosis and treatment of leper YES Diagnosis and treatment of trypanosomiasis YES Other community-based activities exist for malaria and young children’s health but they are implemented on a relatively occasional way or without a real integration. For instance, in 2003, the National Program for Malaria Control (PNLS) trained more than 55,000 women for malaria treatment at home. In the same way, the Directorate for Family Health (DSF) usually organizes campaigns for promotion of breast-feeding. Community-based interventions: main findings Overall, community-based activities in Benin are not much developed, except for some very specific diseases (i.e. Guinea Worm). Each program or directorate finances and organizes its own community-based activities. Besides the limited integration and the irregularity of community- based activities resulting from such a situation, some interventions are relatively left aside (i.e. nutrition) for lack of financial support by a program or a directorate. This situation is especially damaging given that many high impact interventions on children mortality are community-based. 94 Chapter 4 : HEALTH SYSTEM GOVERNANCE & STEWARDSHIP 1. Governance..........................................................................................................................................96 1.1. A central level that is both divided and powerful.........................................................................96 1.2. An institutional system which is still not sufficiently and consistently deconcentrated ..............97 Regional Health Authorities (RHA or DDS) have limited powers and resources for controlling districts, but quite large staff numbers ............................................................................................99 Governance of health districts (HD or ZS) is spread among several stakeholders .......................100 District Health Officers (MCZS) have increasing power, but still very limited resources ...........101 Health committees are poorly representative of their communities ..............................................101 COGECS are also poorly representative and have limited decision rights ...................................101 In spite of their strong legitimacy, local councils remain insufficiently involved in the management of health facilities.....................................................................................................103 2. Stewardship of the health system ......................................................................................................103 2.1. Health information system ....................................................................................................103 2.2. Health policy formulation .....................................................................................................106 95 1. Governance In this chapter, we analyzed particularly problems of governance from the health system institutional point of view. For each level and stakeholder in the system, we tried to understand the consistency between (i) the objectives assigned to them and (ii) the decision powers and resources to achieve these objectives. Individual aspects of governance are treated in the chapter “Human resources”. 1.1. A central level that is both divided and powerful The Ministry of Health (MoH) has kept its control on most of resources. One can see in chapter 9 that the major part of the health budget remains spent and controlled by the MoH at central level. With regards to human resources, in the same way, the MoH included 727 personnel in 2006, of which 66 doctors (representing close to 15% of the public sector’s doctors). Finally, there is still no decentralized planning system in Benin. Directorates in the MoH decide on most of the activities to be implemented without ensuring consistency with activities planned by healthcare facilities. In spite of its important power, the MoH functions more as a group of relatively autonomous directorate than as a coordinated entity. The MoH consists indeed of 14 directorates (of which 3 have cross-cutting responsibilities), as the organizational chart below shows. It is clear that these directorates operate in a relatively isolated way, notably due to their dedicated funding (derived from budgets-programs and external partners). One of the main organizational deficiencies is the absence of a strong structure to coordinate the achieved activities. The units in charge of this coordination are normally the General Secretariat (GS) and the Planning and Prospective Directorate (DPP), but they do not have adequate tools and mechanisms (procedures, commissions, staff, etc…) to assume this role, yet essential. A coordination body also exists for the implementation of health policies: the National Committee for supervision of the projects/programs execution and assessment of the sector (CNEEP). This entity includes the MoH as well the other ministries involved in health sector, the private sector and donors. Based on its composition, the CNEEP seems to be an adequate coordination body. But it does not have any decision power to harmonize the interventions of the different directorates and those of the Donors Figure 48 : Organizational chart of the MoH(2008) MINISTRE Secrétariat particulier Assistant du Ministre Cellule de DIVI=IGM Communication Directeur de Cabinet Secrétaire Général Assistant Directeur de Cabinet Secrétaire Général Adjoint Adjoint DRFM AC CT CT CT CT CT DPP DRH OST DNPS DPM DSF DHAB DRS DSIO 96 DNPEV DH DIEM DEDTS DDZS DDS DDS DDS DDS DDS DDS Atacora/Donga Atlantique/Littor Borgou/Alibori Mono/Couffo Ouéme/Plateau Zou/Collines al 1.2. An institutional system which is still not sufficiently and consistently deconcentrated Box 2 - Administrative zoning and Health Districts in Benin. 21 Since its independence in 1960 until 1999, Benin was divided into 6 regions and 77 sous-préfectures . Each sous- préfecture comprises of several municipalities (or communes), which are subdivided into villages. As for health issues, the principle (not always applied) was that each municipality should have a health center and a maternity hospital (with theoretically a medical doctor). Partly under the influence of international donors and with the launching of the Bamako initiative (1987), the idea of creating “health districts” was approved during a large round table organized in 1994. Like in other countries, the principle was to create health districts comprising of health centers (without doctors) referring their patients to a district hospital and coordinated by a district doctor. The implementation of this idea took a lot of time, due to reluctance of medical specialists (who did not accept the idea of being under the supervision of a district coordinator, who is a public health doctor and not a clinician) and of Members of Parliaments (worrying about potential disputes between municipalities and sous-préfectures over the headquarters of the district hospital, a facility where doctors would hence be based). Reflection on the creation of districts started in 1995, but was implemented only in 1997. The coverage of districts was based on health centers’ catchment areas and not on the existing administrative zoning. This explains the choice of the term “district”. The French word “district” relates to a purely administrative notion and is geographically distinct from other administrative division (i.e. generally every health district includes 4-5 sous-préfectures). At the same time, with several laws passed in 1999-2000, Benin implemented an important decentralization process for all its public administration (not only in the health sector). The sous-préfectures have become municipalities or communes (still 77 in number), including 508 arrondissements (former municipalities) and villages. Moreover, 6 additional regions were created, bringing the current number of regions to 12. The lack of consistency between health zoning (dated 1997) and administrative zoning (dated 1999) caused a certain number of problems. Particularly, as already said, each health district was supposed to have a referral hospital (i.e. a district hospital). As expected, the choice of a site to build this hospital resulted in relatively violent disputes among municipalities, each one desiring to host the district hospital. These conflicts left some scars. For instance, certain district hospitals have very low attendance, either because the population has never been happy with its location or because the choice of its location was based more on political considerations than of population’s habits to use healthcare services. We can analyze an institutional system through (i) the level of deconcentration22 and (ii) the consistency of this deconcentration. On this last point, we will actually see that, if real deconcentration effort were indeedmade, they were not always consistent. Clearly speaking, certain levels have been assigned a wide scope of objectives, without the appropriate decision rights and resources. We also come across the opposite situation, where abundant resources are no longer attached to strong and clear objectives (example of the Regional Health Authorities). 21 Respectively called « provinces » and « districts » during the Marxism-Leninism period (1974-1989). 22 Following a French tradition in public administration, we make here an important between deconcentration and decentralization. The first notion (deconcentration) happens when within the Government administration, more power is granted to local Government authorities, while the second notion (decentralization) refers to transfers of power outside the Government administration, towards local (elected) councils. 97 98 Currently, the institutions of a health district can be understood as follows: Figure 49 : Health districts institutional architecture (2008) RHA/DDS (regional health authority) CHD (regional hospital) MCZS District health Management EEZS/DHMT officer Health committee Health distrcit team committee District hospital Management commitee Commune-level Commune-level health center health center COGECS COGECS (management (management committee) committee) Légende: Supervisory Arrondissement-level Arrondissement-level entity health center health center COGECS COGECS Representative (management (management body committee) committee) HEALTH DISTRICT Health facility Regional Health Authorities (RHA or DDS) have limited powers and resources for controlling districts, but quite large staff numbers Existing regulations give very little power to RHA/DDS to manage regional health systems. The RHA/DDS are supposed to represent the Ministry at the regional level. As such, they are - theoretically - the coordinators of regional health systems. In fact, whereas the RHA/DDS were relatively powerful (in comparison to the health district level) before 2004-2005, they are now mostly destitute of influence levers on the health district (HD). The table in annex shows that RHA/DDSs have been deprived of their responsibilities to the benefit of the HDs and the communes. Overall, the RHA/DDS can above all (i) finance the training of HD staffs and (ii) carry out their supervision. Again, it is necessary to indicate that RHAs do not have any decision rights to reward or to punish HD staffs (including the District Health Officer or MCZS). The impact of the above-mentioned supervisions by the RHA/DDS is therefore very limited. In total, they especially assume specific missions (described in the table below), of which one can wonder if most of them could not be delegated to the HD level. RHA/DDSs have limited financial resources, a situation that reduces even more their potential influence on HDs. Since 2000, delegated credits (CD) are not allocated anymore to HDs by the RHA/DDS. HDs receive directly a part of the CD. The table below (for the regions 99 of Ouémé/Plateau) shows that these amounts are directly allocated to HDs. The other regular resource received by the RHA/DDS is a share of the cost-recovery amounts. These amounts remain nevertheless small and are mostly used to finance travel costs of the RHA/DDS to carry out their audits and supervisions. On the other hand, RHA/DDSs’ number of staff remains quite important. As a matter of fact, a typical RHA/DDS has about a hundred agents, including about thirty senior staffs (doctors, midwives, statisticians…), as showed in the example below. Although this staffing seems justified for some essential activities (blood transfusion, maternal deaths audits), one may wonder why the other activities require so many staff, as they seem to overlap with the activities of the MoH and its vertical programs. The organizational chart of many RHA/DDS is a replication of the MOH’s one. Table 17 : Staff and functions of DDS: an example. Units Functions Staff Including qualified staff Research & analysis Carry out various studies 6 4 Health promotion Carry out prevention (HIV, malaria) and immunization 60 10 campaigns Human resources Manage HR of DDS 3 1 Financial resources Manage resources of DDS 5 1 Family health Carry out family planning activities 6 5 Nurse and obstetrical care Support maternal care activities (including maternal deaths 4 4 audits) Equipment and maintenance 4 4 Administration 8 1 Diagnosis and drugs Supervise labs and pharmacies 3 3 Blood transfusion sanguine Supply health facilities with blood ?? ?? TOTAL 99 33 Source: WB analysis Moreover, it seems that the annual action plans (AAP) prepared by the RHA/DDS are not complied with by the vertical programs nor by the central directorates of the Ministry. As mentioned in a document presented during the health sector review in September 2007, “the most frequent case is that lower administrative levels (intermediate and peripheral) are asked by the central level to take part in the execution of their activities during the very same periods for which the lower levels have programmed their activities". Obviously, though the objectives of the programs are well integrated at the national level (PNDS) that is not the case for the implementation phase, where the AAP prepared by the RHA/DDS are generally not respected. Once again, one must underline the dramatic absence of decentralized planning. Governance of health districts (HD or ZS) is spread among several stakeholders A large number of stakeholders are involved in the strategic and operational management of healthcare facilities within each district. With the decree n2005-611 of September 28, 2005, no fewer than 5 bodies were established (not to mention the MCZS) to coordinate the health district, namely: (i) a Health Committee which is supposed to be the "Parliament" of the district, (ii) the Health District Team (EEZS), that is supposed to be a sort of district cabinet (but is in 100 fact very parliamentary), (iii) District Hospital Management Council, (iv) Management Committees of health facilities or COGECS and (v) Village Health Committees. The management of each healthcare facility is controlled in theory by the COGECS, but in fact – in view of its weak prerogatives – more by the HD Health Committee2324, the EEZS, the MCZS, and even the Mayor. District Health Officers (MCZS) have increasing power, but still very limited resources The MCZSs have limited but progressively increasing power. As one can note it on the table in the annex, decision rights held by the MCZS have increased little by little. They remain nevertheless very weak as regards to management of human resources. For instance, performance assessment of MCZSs by the RHA/DDS and the MOH remains anyway very limited. In total, the MCZSs neither have the tools, nor the incentives to manage their districts effectively. Their resources are insufficient. Budgetary allowances for HDs remain very limited, in spite of recent efforts in this sense. Moreover, the fact that the MCZSs partially operate (and the RHA/DDS) in collecting a percentage of the cost-recovery amounts (collected by health facilities) does not seem a healthy practice (and even less equitable). It should be replaced by an increase of the CD. Finally, one can wonder, as many stakeholders do in Benin, about the educational background of the MCZS. Appointing to these jobs young public health doctors without any clinical experience does not allow giving them enough legitimacy vis-a-vis staff on the ground. This situation is worsened by the fact that they do not carry out any clinical supervision (anyway, do they have the expertise to do it?). The supervisions by the MCZS are focused on the control of resources coming from cost recovery, which can naturally be explained by the fact that the MCZSs receive a portion of these resources. Health committees are poorly representative of their communities Health committees in principle include the following people: (i) representatives of the RHA/DDS, (ii) a representative of the Management Committee of health facilities (COGECS), (iii) the mayors, (iv) two representatives of Donors, (v) two representatives of NGOs, (vi) a representative of a nonprofit private sector, (viii) a representative of the for-profit private sector and (ix) two health facilities staff representatives. We notice that the representation from the COGECS is very weak. COGECS are also poorly representative and have limited decision rights Just like many African countries, Benin has created bodies for community representatives within health centers, called COGECS. COGECS were created in 1994, as a follow-up of decree n90-346 of November 14, 1990. The operating procedures of these bodies have been defined by decree n3090 of February 14, 1995 (for COGECS) and decree n6723 of July 30, 2004 23 COGECS is represented within the HD health committee, but only for a small minority. 24 whose composition and attributes are defined by the order n10196 dated 19 October 2005. 101 (for COGEAs). Besides, it is from 2004 that COGECS began truly to function once the operating procedures have been defined. The current composition of the COGECS does not reflect the diversity of their communities. And their motivations are not always altruistic. Each COGECS normally consists of 8 members, including (i) two women representatives, (ii) two for the youth, (iii) one for the local politicians, (iv) one for locally elected authorities, (v) one for NGOs and (vi) the health center director. In fact, many are appointed and not elected. Moreover, composition rules are not always complied with. An analysis of a sample of COGECS (Zébra survey 2006) shows that the members generally have (71%) an educational level at least equal to high school level, are civil servants (50%), notably from the teaching or health sector, and often cumulate other representative functions25. All in all, these people do not really represent their communities, which are much poorer and less educated. Moreover, the combination of these functions with their salaried activity leads to a relatively high rate of absenteeism. The table below shows also that, if the majority of the members of the COGECS claims they want to contribute to the development of health in their commune (84%), a significant number (31%) sits in the COGECS for financial interest (whereas these functions in principle are on a voluntary basis). Table 18: Expectations of COGECS members (2006). Expectations Number Frequency Improving health situation in the community 101 84,9 Managing adequately resources of the health center 99 83,2 Obtaining training on financial management 84 70,6 Obtaining financial bonuses 37 31,1 Getting free care 32 35,3 Getting a better social status within the community 29 24,4 Source : Enquête Zébra 2006 Regarding management of the Health Centers, COGECS have a very limited role. The table in annex 4.2 clearly shows that the COGECS does not have any control on the 3 main types of operational decisions (HR, equipment and budget). The situation is therefore far from a genuine community-based management (as it exists for example in Mali). Unsurprisingly, members of the COGECS are not fully involved in the management of the health centers, apart from the aspects related to budget and especially to drugs sales. The above-mentioned survey of 2006 endeavored to measure this involvement of the members. Reading the table in annex 4.2, one notes that the members are mostly present for voting the budget and for controlling financial resources from drug sales. One can think that this strong attention to drugs is linked to the fact that it is the main resource, which has to be preserved against numerous robberies by some health workers26. An analysis (Credesa 2006) of the minutes of 28 COGECS meetings showed indeed that all COGECS were confronted with this practice. 25 A similar survey (Credesa 2006), conducted on 28 COGECS, produced comparable results. 26 Resources from cost recovery are essentially used for the recruitment of unqualified staff (with several convenience recruitments) and also for (more or less legal) allowances paid to the President of COGECS. Conversely, the staff receives no or little allowance from these cost recovery amounts. Stealing or embezzling drugs from the center to sell them on the informal market seems to be a more lucrative activity. This explains the frequent disputes between healthcare facilities staff and COGECS. 102 Apart from these two points, day-to-day involvement of the COGECS is weak. Particularly, it can be noted that only 3% of the members of the COGECS are present to decide a fee exemption for the poor, which is evidence that this fee exemption mechanisms are not of any interest for the majority of COGECS members. In spite of their strong legitimacy, local councils remain insufficiently involved in the management of health facilities Responsibilities of local councils in health are legally very limited. They are solely responsible for the construction and the maintenance of basic health infrastructures. It is indeed anticipated, since 97-029 law of January 15, 1999, that "the council has the responsibility for construction, equipment purchase and repairs of health public centers buildings, of the public youth infrastructures, sport and leisure, at the arrondissement level, the village or the area of the city. Besides, it assures the maintenance of these centers and infrastructures. To this effect, the Government transfers the necessary resources". Some local councils tried to invest further in the health sector, notably in recruiting some staffs or in providing bonuses to retain qualified health workers. For instance, Malanville, on the border with Niger, has complemented the remuneration of a surgeon, to attract him for a long time in this remote city. On the other hand, the Ministry of Health always opposed the recruitments of staff by local councils (an example in Ouidah). Other local councils intervene in a more casual way, by granting drugs and equipments or by supporting communication campaigns. Generally, no decision right over health centers management has been granted to the local councils, which explains their lack of involvement. Due to lack of transferred resources, the local councils spend very little on health. An analysis (Credesa 2006) showed that less than 3% of the investment budgets of the local councils are dedicated to health. Relationships between COGECS and local councils remain limited. Most COGECS are not involved in the development of the Local Development Plans (LDP). It is also true that the local councils are not explicitly represented within the COGECS. 2. Stewardship of the health system To lead a health system, at least two elements must be present and more importantly they have to be efficient: 1. A health information system, to identify the problems and to measure progress achieved in the implementation of health policies; 2. A formulation mechanism for these policies and which enables to take into consideration the opinion and needs of all stakeholders, while assuring an indispensable coordination between these stakeholders. 2.1. Health information system Despite the existence of a national system, many information subsystems have proliferated. There is an information system called National Health information and Management System 103 (NHIMS) that uses routine data managed by the Directorate of Planning and Prospective and an alert system managed by the National Directorate of the Health Protection. In addition to these subsystems, others are managed by vertical programs (including PNLP, PNLS, PNT) and some central directorates (DFPR, DRH, DH and DIEM). 104 Table 19: The various health information systems in Benin Utilisation Avènements Maladies à Médicaments, Nom du Information Ressources Equipement, Evènements des des potentiel contraceptifs, système financière humaines construction vitaux services maladies épidémique vaccins Système de reportage basé SNIGS X X X X X X X sur les services de routine Surveillance SIMR X X X X épidémiologique Système de vaccination PEV X X X Système spécial TB PNT X X X X Système spécial paludisme PNLP X X X X Système spécial VIH/SIDA PNLS X X X X Système spécial SMI DSF X X Système d'information DH X X X X X X hospitalière Système d'information à base communautaire Système administratif DRFM X (finance) Système administratif DIEM X (infrastructure, équipement) Système administratif DRH X (ressources humaines) Enregistrement état civil X Source: DPP/MS 2008 The information system has several overlaps and an insufficient quality control. It is not rare that the very same data are collected together by the HIS and by vertical programs (i.e. malaria, HIV-AIDS…). Moreover, these multiple data collection processes do not permit to improve data quality. The same indicator can lead to different measurement even if these measurements are made by the same persons. Data quality control is carried out at the central level. But no data analysis is made at the local level although it would obviously help to assess reliability and validity of collected data before forwarding it to the higher levels. A certain number of stakeholders are not included in the information system, both due to their lack of transparency and to an ignorance of these stakeholders by the MoH. Private facilities (in particular those in the commercial sector) provide few data. A large number of them refuse to report data to the MoH, because these structures are illegal (cf. chapter on the private sector). But those that wish to provide some data are rarely contacted. The not-for-profit private sector is more transparent, because of partnership policies already put in place. It remains nevertheless very secretive about its financial information. Concerning donors, the information provided to the MOH, especially for financial data, appears very limited. The information system also ignores some key indicators. The epidemiological situation of the country and the activities of health facilities are quite well measured. However, patients referrals are not monitored, which is surprising in a system where the concept of health pyramid is supposed to play an essential role. In the same way, very little information is collected on the socioeconomic profile of health system users. This situation is not specific to Benin and is easily explained by the high cost of collecting this type of data. Nevertheless, in a country where there is a Health Equity Fund, this weakness makes it difficult to ensure correct monitoring and evaluation of this mechanism. Finally, information on quality of care is rare. 105 2.2. Health policy formulation The formulation of the health policy is the subject of a fairly large dialogue, at least at the central level. The preparation of National Health Development Plan (NHDP 2007-2016) illustrates well the wide range of involved stakeholders. This document has been prepared with not only all the directorates of the MOH, but also the representatives of the other ministries, unions and Donors. On the other hand, little consultations were held with the private sector (commercial). In particular, the Office for PPP, which was supposed to assure the representation of the private sector, has been mostly forgotten during the discussions on the NHDP. Decentralized planning is not operational. With the support of USAID (PISAF project), two regions got involved recently in a process of decentralized planning. On this occasion, a procedures manual of decentralized planning has been elaborated. Officially, the MOH has committed itself to carry out this process nationwide. However, few efforts are put in this sense. According to the MOH, “the NHDP is supposed to be broken down at regional and district level, meaning that once the strategy document is adopted, the health districts have to use it to elaborate their own development plan”. This view is exactly a "top-down" one and is precisely in contradiction with the principle of decentralized planning. There is therefore a long way to go. The relationship with the donors is essentially limited to policy dialog (of good quality). Harmonization efforts are low. The MOH meets every 6 months with health Donors. Donors appoint a coordinator. A sector-based review is conducted every year and allows the MOH to assess the implementation of its health policies. As seen already, the dialogue between the MOH MoH and Donors works relatively well. One could hope however for a little stronger harmonization and especially more proactive. No sector-based approach has been attempted for the meantime. In the same way, the yearly sector-based review could serve as an opportunity to invite donors to announce their projects of interventions (with their amount) and to explain their consistency with the health policy of the country, as it is done by some countries (e.g. Ethiopia, Tanzania, Niger…). Benin is still far from this approach. The preparation of the health strategy is not based on a rigorous method of prioritization, which does not facilitate MoH’s advocacy with the Ministry of Finance and Donors. The current PNDS includes more than 30 sub-programs that are not prioritized. Moreover, the methodology used to prepare these recommendations does not seem clear. In particular, it does not depend on a prioritization-setting tool, such as the Marginal Budgeting Bottlenecks (MBB). In these conditions, the PNDS looks like a list of actions to be taken (and especially to be financed by donors) without any clear priorities, and more importantly without the health impact of the various actions being clearly estimated. This explains to a large extent the unwillingness of donors to support this health plan. One can hope that the current training efforts on the MBB (financed by UNICEF) would produce results quickly. Finally, it is necessary to indicate that the MTEF is rarely used today, which can easily be understood given the preparation conditions of the National Health Development Plan. 106 Institutional governance and health system stewardship: Main findings While being one of the pioneer countries for primary healthcare, Benin has been involved quite late into the deconcentration of its health system (officially in 1995 but in fact in 2000). The central (MoH) and regional levels (DDS) still control most of the resources, even though the deconcentration of resources to health districts is progressing every year. It is clear that this evolution could be accelerated by the implementation of a real decentralized planning process and by a better representativeness of the COGECS. Another problem, less usual, is the insufficient coordination between the activities of national directorates and local health authorities. In spite of some weaknesses, the health information system is very effective. As indicated above, the stewardship function is still suffering from the lack of decentralized planning and limited coordination with donors. The more important concerns are the weaknesses of the national strategy. Its objectives are poorly prioritized and mostly with little connection with their foreseeable impact on health status. These two problems seriously limit the capacity of the MoH in its advocacy with the Ministry of Finance and with donors. 107 Chapter 5 : HEALTHCARE SERVICES 1. Assessing performance of healthcare services ..................................................................................109 1.1. Effectiveness ..............................................................................................................................109 A public health services system that is still irregularly used.........................................................109 Utilization of outpatient services...............................................................................................109 Hospital inpatient utilization .....................................................................................................114 A technical and organizational quality of healthcare that did not significantly improve..............116 Non-clinical functions are poorly evaluated..................................................................................117 1.2. Efficiency ...................................................................................................................................118 1.3. Equity .........................................................................................................................................119 1.4. Financial protection....................................................................................................................119 2. Explaining performance of healthcare services.................................................................................121 2.1. Does the proposed range of healthcare services correspond to health care needs ?...................122 Mix of services ..............................................................................................................................122 Relationships between levels of care (referral system) .................................................................123 2.2. Are there adequate resources for health services?......................................................................124 a. Equipment and buildings ...........................................................................................................124 b. Human Resources......................................................................................................................125 c. Drugs .........................................................................................................................................125 2.3. Do existing accountability mechanisms foster performance of health facilities? ......................126 Accountability mechanisms are very weak ...................................................................................126 Management autonomy is limited .................................................................................................127 Capacities for management are insufficient ..................................................................................128 2.4. Do health services payment systems contribute to efficiency? ..................................................129 108 1. Assessing performance of healthcare services Performance of healthcare services can be analyzed through 4 dimensions: a) Effectiveness (i.e. the improvement of health indicators due to the effectiveness of health services) b) Efficiency (i.e. the relationship between the used inputs and the achieved results) c) Equity (i.e. is it organized to prioritize services to the neediest patients ?) and d) Financial protection (i.e. regardless of patients’ income, is healthcare financing set up so as to reduce the final costs that patients bear?) Considering the rarity of data on the private sector, it is necessary to note that only the public sector is analyzed here. The chapter 8 contains a detailed description of the private sector in Benin. 1.1. Effectiveness Health services contribution to health indicators is generally difficult to measure. It is true that a country’s healthcare system is one factor among others to improve population health. It is also probably one of the least important. Besides this, it is exceptionally difficult to measure this impact. To measure the effectiveness of health system, we have retained two proxy indicators: (i) populations’ level of utilization of the system (i.e. outpatient and inpatient services utilization rates) and (ii) level of technical and organizational quality of care. Unfortunately, no reference indicator is available for these proxies. It can only be compared to those achieved by other countries or to their evolution be observed as time goes on. A public health services system that is still irregularly used Utilization of outpatient services For the general population, public healthcare facilities utilization rate (i.e. curative outpatient visits) remains low and quite variable. In 2006, the ratio of the number of new outpatient visits out of the population (i.e. utilization rate27) was only 44%. The figure below seems to show a brutal slump in 2002 and later a slow and irregular rise. Considering WHO’s benchmark, whereby one must attain a rate of 100 % (i.e. every inhabitant has at least a contact per year with the health system), the current rate still remains low, in spite of the obvious progress. 27 This rate refers to utilization of curative outpatient services by new patients (i.e. excluding chronic patients) for all age (i.e. including children) in primary health care facilities. 109 Figure 50: Utilization rate for outpatient visits (2001-2006) Taux de frequentation (consultations curatives) 70 60 59% 50 40 44% 39% 38% 37% 30 35% 20 10 0 2001 2002 2003 2004 2005 2006 Source : NHIS On the other hand, the outpatient utilization rate for children under one year is higher and is improving. The gap between regions is considerable. Indeed, rates vary from 51% (Couffo) to 132% (Borgou). It is not possible to determine whether this variation is due to the lack of healthcare infrastructure, to difference in households’ income levels or to other factors. Figure 51: Evolution of outpatient utilization rates for under-1 children (2001-2006) Taux de frequentation (enfants de moins d’1 an) 120 100 112% 99.3% 80 80.6% 83% 77% 60 60.7% 40 20 0 2001 2002 2003 2004 2005 2006 Source : NHIS Figure 52: Outpatient utilization rates for under-1 children across regions Taux de consultations pour les enfants de moins d’un an (2006) 140 120 100 80 60 40 20 0 f fo u ga u u ri a ue es o e l ra o Zo o m ea on or ou n rg ib lin iq t to ue ac Do at M Al nt Bo C l Li Co Pl O At la At 110 Source : NHIS For several years, the outpatient utilization rate for children aged between 12 and 60 months is stagnant at a low level. This rate is about 58% for the last 6 years, without any improvement what so ever. If regions where most referrals are made (Littoral and Oueme) are not taken into account, the gaps between regions are less important. The rate then ranges from 40% (Dongo) to 66% (Mono). Compared to other African countries (see figure below) Benin records the lowest utilization rate for treatment of acute respiratory infections (35%). In the same way, the utilization rate for treatment of diarrhea does not go beyond the 20% rate (see table below). Overall, it seems that the utilization of health services for the treatment of children is not common. Figure 53: Outpatient utilization rate for children aged 1 to 5 (2001-2006) Taux de frequentation (enfants de 1 a 5 ans) 62 61 61% 60 59 59% 59% 58 58% 58% 57 56 55 55% 54 53 52 2001 2002 2003 2004 2005 2006 Source : NHIS Table 20: Health services utilization rate for children aged 1 to 5 with respiratory infection symptoms % of children (1-5 years) % of children (1-5 years) with ARI symptoms with diarrhea and taken to a health care center and taken to a health care center Bénin 2006 35.7 20.8 Guinée 2005 42.0 26.9 Liberia 2007 62.2 49.3 Mali 2006 38.1 17.8 Niger 2006 47.2 17.2 Sénégal 2005 47.2 20.2 Source : DHS On the other hand, prenatal care utilization remains very high and stable. The prenatal consultation rate is defined as the ratio between new consulting women and expected pregnancies28. According to the NHIS, this rate remained very stable, around 91% in 2006, 28 WHO defines the number of estimated pregnancies as the number of expected birth multiplied by 1.15 . 111 whereas DHS 2006 estimated this rate to 88%. The difference is therefore small and shows as once again the high reliability of information produced by the NHIS. 112 Figure 54: Evolution of the prenatal care utilization rate in Benin (2001-2006) Taux de CPN 100 90 92% 91% 91% 91% 90% 80 70 76% 60 50 40 30 20 10 0 2001 2002 2003 2004 2005 2006 Source : NHIS (NB : data for 2004 are missing) Table 21: Prenatal care utilization rates in several African countries 2006 Bénin 88.0 Congo (Brazzaville) 86.8* Guinée 82.2* Mali 37.2 Niger 46.1 Sénégal 87.4* * : year 2005 Source : DHS Behind this rather satisfactory national rate, there are some significant differences across regions (from 61% in the Alibori to 99% in the Atlantic), across levels of education (from 84% for the women without education to 99% for those having a level at least secondary education) or across socio-economic status (from 74% for the poorest quintile to 98% for the richest). The highest difference is in rural areas, where the pregnant women rarely consult. If the majority of women actually use health services at least once during their pregnancy, an important number (close to 40% according to DHS 2006) do not comply with the recommendation of 4 prenatal care visits. Besides, only 42% (DHS) of women do their first visit in the first 4 months of pregnancy. It is especially the case of women living in rural areas. One of the reasons put forward is cultural. In the northern regions, the first months of pregnancy are considered as strictly confidential. "It is necessary to wait for the belly to come out before going to a health centre” (Kpatchavi 2005). However, this problem has to be put in perspective. For several years, evidence has been found that prenatal care has actually a relatively limited impact on maternal mortality, as maternal health depends more on intra-partum care (cf. chapter 2 for details). Utilization of postnatal care remains insufficient. According to the NHIS, the rate of postnatal utilization visits stagnates around 41% for several years. The data from the DHS 2006 are more optimistic, since they estimate a rate of 68%. This last rate is probably closer to the reality, not only because of the highest quality of DHS data, but also because utilization of postnatal care 113 appears bound extensively to the rate assisted delivery. Nevertheless, even with a rate of 68%, the coverage in postnatal care remains insufficient and contributes to the high neonatal mortality rate. Utilization of postnatal care is highly related with poverty prevalence and with the region of residence. Whereas on the average 32% of women having given birth did not benefit from postnatal care, this rate is 50% for the poorest quintile and more than 46% in the two Northern regions. Hospital inpatient utilization The overall hospitalization rate has dropped since 2001. Roughly, one can estimate the effectiveness of a hospital system by analyzing the evolution of hospital cases in relation to that of population (i.e. hospitalization rate). Benin moved from an already low rate29, of 2,454 admissions for 100,000 inhabitants in 2001 to 1,644 in 2006. Even without taking into account population growth, one observes an absolute reduction of hospital cases. This trend is observed at all levels of the hospitals’ pyramid (national, departmental and district), but it is especially strong for district hospitals. The first question concerns the causes of this trend. The stability (or sometimes the small increase) of outpatient activity (cf. supra) suggests some substitution effect (i.e. hospital inpatient care being replaced by outpatient care). Another potential explanation could be the longer average lengths of stay (ALOS). For a given capacity, it is necessary to reduce the number of cases if the ALOS increases. However, the table below shows that the ALOS has either diminished or remained stable. Overall, one can advance the hypothesis that healthcare demand remains high, but that the public hospital system is less and less able to satisfy it, which could be explained by a deterioration of quality of its services and the reduction of resources allocated to hospitals. Figure 55: Hospitalization rate for 100,000 inhabitants (2001-2006) 3,000.00 Zone (HZ) Departemental (CHD) Nombre de sejours pour 100,000 habitants 2,500.00 National (CHNU) 2,000.00 1,500.00 1,000.00 500.00 - 2001 2002 2003 2004 2005 2006 Source : NHIS 29 According to OECD (the only international organization monitoring this indicator), hospitalization rates in developed countries ranged from 8,450 (Turkey) to 27,100 (France). 114 115 Figure 56: Average length of stay in hospitals (2001-2006) 12.00 10.00 DMS CHNU 8.00 DMS CHD DMS HZ DMS moyenne 6.00 4.00 2.00 0.00 2001 2002 2003 2004 2005 2006 Source : NHIS A technical and organizational quality of healthcare that did not significantly improve Some indicators show some progress of the technical quality of health care, at least for prenatal care. There are very few consensual indicators to measure quality of health care. One can use the data related to the treatment of some specific clinical cases. We have previously seen (chapter 3) the low quality of obstetric care in intra-partum. We can also measure quality of health care through the content of prenatal care visits. According to the DHS 2006, almost all women benefited – during prenatal care visits - from measurement of weight (99%) or of size (96%), of a verification of arterial blood pressure (99%), of an abdominal palpation (99%) and of a urine sampling (92%). On the other hand, blood samples are less frequent (40%), as well as nutritional advice (45%) or information on the signs of pregnancy complication (39%). Nevertheless, it is necessary to notice a real progress, since these statistics were a lot lower in 2001: 19% for information on signs of pregnancy complication and 29% for the blood sample. But organizational quality remains low. By organizational quality, we mean the aspects linked to responsiveness of staff to patients. Without being the most important obstacles in access to health care, these problems remain significant. The table below shows the frequency of the problems faced by women having used health care services (DHS 2006). More than 23% receive a bad welcome and 30% had to wait a long time before being taken care of. Again, DHS data is for public and private health care structures. It is likely that these indicators are less favorable in the public sector. 116 Table 22 : Obstacles mentioned by women when accessing healthcare services Frequency Rank Problems reported % Collect cash to pay for services 73,9 1 Cost too high 56,7 2 Distance 38,1 3 Need to find a car or motorbike 36,6 4 Waiting time 30,4 5 Staff absent or late 29,2 6 Medical care not effective 24,2 7 Low responsiveness of staff 23,6 8 Go by oneself 21,6 9 Know where to go 18,3 10 Lack of female staff 16,0 11 Need authorization to go 14,9 12 Source : DHS 2006 Overall, in spite of some encouraging signs, it is unlikely that quality of healthcare has improved generally. The government created in 2007, a National Committee for Quality Assurance in the Health Sector (CNAQSS) and endeavors to formulate a national policy for improvement of healthcare quality. While waiting for a real nationwide effort, quality is currently improved through individual initiatives30 and foreign financial sponsors (Switzerland and United States). But the impact of these efforts is probably very limited. The stagnation (or even the decrease) of the utilization rate in the public health care system and the continuous development of the private sector, bring indeed to question the progress achieved until now. Non-clinical functions are poorly evaluated. One should also assess the effectiveness of the healthcare system (particularly at the higher levels of the pyramid) regarding (i) pre-service and in-service training of staff and (ii) supervision. The supervision system is particularly poor. There is no data available on the number and type of supervision missions conducted. It is even likely that there is very little information to collect. According to health system stakeholders, supervision missions are very irregular and they depend essentially on the availability of funds. Moreover, it seems to be an exercise aiming at personal enrichment. "We conduct a supervision to increase our monthly income with the per- diems" (cited in Kessou 2008). There are no technical recommendations for these supervision activities. One facility is rarely supervised by the same people. This would normally enable to assess progress achieved between two supervision missions. But “everybody should normally benefit from supervision per diems” (cited in Kessou 2008). Finally, these supervision missions are more administrative than clinical. For example, when there is a supervision mission to a healthcare facility, supervision teams are more interested in collecting financial data, especially data concerning cost-recovery resources31. There is therefore very limited clinical supervision. 30 For example, the mother and child hospital of Cotonou (LAGUNE-HOMEL) obtained a certification ISO 9001 in three clinical sectors. 31 HD teams receive a portion of these financial amounts. 117 With their education background in public health, HD teams will anyway have difficulties in carrying out this kind of supervision. 1.2. Efficiency The overall efficiency of the hospital system (public and private) remains limited, with an efficiency index of 62%. To estimate the efficiency of the Beninese hospitals, a Data Envelopment Analysis (DEA) method has been used (see annex 5.5 for the details). It included about 80% of the hospital facilities in the country. This analysis reveals an index for technical efficiency of only 62%. In other words, 38% of hospital inputs (i.e. staff and beds) are wasted as they do not contribute to production. This 62% index is relatively low, especially when compared with other countries of the region (for example, the 73% index of Senegal). Figure 57: Technical efficiency and scale economies in Benin hospitals (2006) 1.8 1.6 efficience (CRS) 1.4 effet d'échelle 1.2 1 0.8 0.6 0.4 0.2 0 Lo titi n ) TH EH U o) de é (P ero Z uè ch lte) -O A L ié HZ u P mb ll e ai n) lF o é ki kou k è o u ék r i é ) de uc Sa ) dé g o tal HZ é ou ey- i la D è a Ad ah ué ur un Ap ndi (M lav u ré Ba ME m H ané IE SD m n ok rd ngb go om ar t nv b ti Kl aste ho ké vi BE AT i ta Mo om as s N ara éré a L Po ué d n ho ho Lé on an S D a Ka ko an (B M Zi ua H Oui E H HO C an ro nt ap C a n la jo én ôp D- u ni B P al ou no de de Ko Z ue Na lE D Z M ou H H H S Z Z s an AR ha ou Be (S ot H ui H C H ix Z u ôp Z Z Z Z C Je Z H S ro Tc - S se p O A H H 5 H Z Ab 3 H Z C H nt H et H ô a ai Z i ta jou opi on (H ik ph ll ie iq 2 H .S ta li n D ot ou H u pi ôp C de C Jo Z on Hô (H H Z ot n H nt ea a C D ai ét t-J Z lS Z ui H H in ng Sa ôp Ta Z H Z H H Source : MoH and WB analysis NB : Histograms reflect efficiency index (from 0 to 1, with 1 being the maximum) for each hospital, as estimated by the DEA algorithm. Curves reflect scale economies for each hospital. When a point is higher than 1, the corresponding hospital is in a situation of decreasing returns. In other words, its size (i.e. number of beds or staff) is too large. Conversely, if a point is below 1, such a hospital is in a situation of increasing returns (i.e. its size is too small). Unsurprisingly, referral facilities are the least efficient. The two regional hospital Centers (CHD) as illustrated in the diagram above and the national mother-child hospital (HOMEL) have an efficiency index varying from 41 and 62%. It is also among them that one finds the two hospitals experiencing severe decreasing outputs (HOMEL and CHD Ouémé). This situation is 118 frequent in many countries, but the relatively recent creation of the CHDs may explain the case of Benin. Indeed, the new CHDs seem to have difficulties to attracted enough patients and secure an adequate “market share”. Private facilities, on the other hand, do not seem to be more efficient than public hospitals. In the figure above, the private hospitals are indicated by light colored bars. The average efficiency of the two sectors (public and private) oscillates around the same index of 62%. Note however that data concerning for private facilities was less reliable than for the public. 1.3. Equity The poorest households have difficulties in accessing health care services, especially for maternal care. In the absence of adequate information on living conditions of households (EMICOV survey), it was not possible to analyze the equity of access to the health system for all the services. Only the services included in the DHS 2006 are treated therefore. The results are given on the figure below. One can note that the income gradient is particularly strong for deliveries and postnatal care. Figure 58: Utilization of various health services across wealth quintiles (2006) P P A R R P P H H D A C H Source : DHS 2006 1.4. Financial protection Usually the financial protection of the population (i.e. protection against the risk of catastrophic health expenditures) is highly related to the health financing system (health financing) and not to the health care delivery (health care delivery). Yet, the experience of some countries (i.e. Kirghizstan) showed that the out-of-pocket expenses of patients could be reduced significantly through (i) an efficient health care delivery system and (ii) a better price control system. 119 Considering the difficulties (presented farther) to implement risk-pooling arrangements (i.e. creation of “mutuals” or of health insurance plans) in Benin, it is therefore worthwhile to explore how the organization for the health care delivery system can have an impact on financial protection. In the commercial private sector, a national fee scale has been elaborated by the association of faith-based health facilities (AMCES), even though it is not always complied with. In the public sector, the price-setting process for clinical services is quite confusing. In addition, set prices do not reflect the real costs. There is a price list, but it has not been updated for many years and notably not since the devaluation of the FCFA in 1994. A legal provision nevertheless allows some adjustments of prices at local level. These adjustments are made by boards of directors in hospitals (or their equivalent in PHC facilities, that is to say the COGECS) or by HD teams. The diagram below shows that, for the same service (normal delivery), the charged fees can vary widely. The fees partially take into consideration the real costs of services. But tools for management accounting are usually absent. The possible gains of efficiency achieved by some organizations hence cannot be translated in the reduction of fees and therefore cannot improve financial accessibility of care. Figure 59 : Fees (in FCFA) for normal delivery in 4 groups of providers (2005) 5,000 4,000 3,000 2,000 1,000 CHD CSC & CSA HZ PNL Source : Benaudit report 2006 120 2. Explaining performance of healthcare services. The results achieved by a health care delivery system on the 4 previously described dimensions can be explained by several factors: • Does the range of proposed healthcare services match health care demand? The question can be broken down the following way: o Are the provided services consistent with health needs? o Are the relations between levels of care (from the primary to the tertiary) operational (i.e. referral system)? • Are the resources of the system adequate, in terms of facilities, human resources and drugs? • Do accountability mechanisms foster performance? This question can also be broken down into three. o Are the objectives assigned to healthcare facilities clear; and is progress toward these objectives being monitored? o Is management autonomy of these facilities sufficient and aligned correctly with the objectives? o Are the management capacities of these facilities adequate? - • Are healthcare facility payment systems related to performance? These different points have a specific impact on the 4 performance criteria: Effectiveness Efficiency Equity Financial protection Range of services Mix of 99 9 9 services Referral 99 9 9 system Resources of the 99 99 9 health care delivery system Accountability Objectives 9 9 9 mechanisms Management 99 99 autonomy Management 9 99 capacity Payment systems 9 99 99 99 These factors are explored below. 121 2.1. Does the proposed range of healthcare services correspond to health care needs ? Mix of services One of the factors that influence performance of a health care delivery system is the consistency of proposed services with health demand. A system can be less effective if some services are not offered (which implies an unmet demand and/or the costly transfer of patients abroad). Some health care services on the contrary (for example, normal delivery care) can be provided by many healthcare levels in the same region ("overlap"), which leads to inefficiencies and to a decrease in quality of health care. Despite the apparent consistency of the system, some important services are not provided in Benin, while the provision of some basic services leads to unnecessary competition between some neighboring facilities. As illustrated in the table below, Benin has structured its health care system as a sort of a three-level pyramid. The national level is represented by the CHNU-KHM. This hospital does not provide all the services one can normally expect from a tertiary level hospital, in spite of its capacity and resources. That explains the importance of “health transfers" abroad (about 5% of the health ministry budget). Therefore, one can better understand the Government’s wishes to build a new tertiary level hospital, with a private management and a staff coming (partly) from the Benin medical diaspora. Conversely, the allocation of services between regional and district levels is not always clear, especially because some regions do not even have a regional hospital. If district hospitals are the only facilities with inpatient services, their outpatient activities sometimes duplicate those of health centers (and vice versa). It is true that the health district strategy has been applied recently. That explains why the “health pyramid” actually looks like a mere accumulation of various administrative layers, each one created at a very different period (and not streamlined so far). 122 Table 23: « Health pyramid » (2006) Source : Statistic Directory 2006 Relationships between levels of care (referral system) The effects of the insufficient consistency within the health pyramid are worsened by some weaknesses of the referral system. Although the notion of health pyramid appears in most health documents and reports, no monitoring is in place for the referral system, at least at the national level. One ignores therefore what is the percentage of children or pregnant women referred to the higher levels. Similarly, at the higher levels of the pyramid, the proportion of patients that could have been treated at a lower level is not known. The limited set of available data in some districts suggests that these referral rates are very low. This impression is reinforced by the weaknesses of the referral system: (i) clinical training (to identify the cases to refer) seems 123 rare; (ii) financial incentives for referral are non-existent, (iii) vehicles (i.e. ambulances) for transferring patients are usually either not available or non functional. In a general manner, unavailability of data on the referral system illustrates the low priority given to this topic, yet essential. Box 3: Referral system in the Klouekanme – Toviklin – Lalo (KTL) health district While there is no data on referrals at national level, a relatively old (2004) but still useful survey on a health district (KTL) can provide some views on the referral system. This survey shows that, for the curative consultations, the referral rate (i.e. patient sent out from the health center to a higher care level) oscillates between 2.1 and 2.7%, whereas one would normally expect a rate close to 10 or 15%. Most of these references were for children visits. Referrals generally took place for emergency (92% of referrals), often for cases of anemia caused by malaria. The survey also showed that referral actually took place in only 42% of the cases, because of the costs which families could not cope with. The results are similar concerning maternal health. About 5% of pregnancies (as seen during prenatal care) were referred, whereas WHO estimates that 15% of pregnancies are at risk and therefore justify a referral to a secondary level. Also, only 64% of the referred women actually went to the district hospital. Source: Discussion on referrals in the health district of KTL, CTB, 2004 2.2. Are there adequate resources for health services? a. Equipment and buildings In terms of "availability" of healthcare facilities, the Benin territory is overall well covered. If one defines availability as the existence of a health center in at least every district32, 90% of the districts benefit from the existence of at least a health center (MoH 2006). Given the population size of each district, one can therefore say that 90% of the population lives less than 5 km away from a health center. It is necessary to specify however that the coverage rate estimated at the time of the DHS 2006 is significantly low. About 76% of women, according to DSH 2006, are living less than 5 km away from a health center. In terms of travel time, the situation is more favorable, since more than 84% of the women estimate to be at less than 30 min from a health center. It is necessary to note nevertheless that, in the DHS 2001, this rate was 89%. Household perception is basically that of degradation. 32 According to the national norms, each arrondissement must have at least a health facility. The coverage rate is therefore the ratio of the number of arrondissements with at least one health facility over the total number of arrondissements. 124 Figure 60 : Rate of population coverage (by health care centers) across regions. 100.00 90.00 80.00 70.00 60.00 50.00 Taux de couverture 40.00 sanitaire en % 30.00 20.00 10.00 - u es l é o ou ue ga u i ra a ffo or ea m on Zo or lin tto rg iq on ib ou ué at ac M nt ol Al Bo Li D C Pl O At C la At Source: 2006 Health Statistics Directory / MoH Benin On the other hand, when one takes into account the “functionality” of health facilities, the coverage rate is much lower. When one takes in account "functionality" (that means the availability of staff to operate a facility), the coverage rate falls to 63% (cf. figure in annex 5.2). Worse, when one choose a less demanding definition of functionality (but yet reflecting the potential benefits of a facility for the population), that is the capacity to provide the minimum package of services (PMA), the coverage rate by facilities is only of 46% (cf. picture in annex 5.2). One can conclude therefore that the priority must be to post adequate human resources to the most destitute structures. To invest in the construction of new health centers does not seem urgent. b. Human Resources Health workers are distributed unequally across the territory and their performance (i.e. expertise, quality and productivity) is very limited. These two points are explored in detail in the following chapter. c. Drugs Performance of facilities is strongly limited by the insufficient availability of drugs and their very high prices. One will see in the next chapter that low availability of drugs is a problem both at facility level and at central warehouse (CAME) level. At facility level, stockouts seem especially related to the inability of many administrators of health centers to order drugs in adequate quantity and frequency. But CAME is not always able to deliver the whole order requested by the facilities. It is also true that drug theft in health facilities inventories is very frequent (and often done by health workers). In the same way, drug prices are especially high at facility level, which does not make them affordable. But it is also clear, as one will see it, that, along the supply chain, it is the facilities that apply the strongest margin. This situation can be explained by the fact that an important part of their revenues depends on drug sales. 125 2.3. Do existing accountability mechanisms foster performance of health facilities? Accountability mechanisms are very weak Box 4: Funding and controlling health care services: two possible models A key factor to improve performance of an organization (as it is for health care services, including hospitals and health centers) is to specify an accountability mechanism, that is (i) to spell out clearly the objectives of this organization, (ii) to measure regularly and objectively the achieved results, and (iii) to make sure the necessary changes are implemented. In spite of the great simplicity and the triviality of this accountability concept, one can observe that it is often very badly put to work, particularly in the health sector. Indeed, it is not rare to see some countries where hospitals, for example, do not have clear strategic objectives and even less mechanisms to measure the achieved results. When a government gets involved in an accountability policy of the health care services, it can choose between two models: one of state-controlled nature (supply-side model), the other based on markets (demand-side model). The supply-side model requires the Government to become the coordinator of the system, which implies that it has to specify its expectations with regards to heath care services and has to monitor the achieved results. Such a model requires the implementation of several mechanisms: • A planning process for the services to be provided, a process that should be regular, discussed at local level and comprehensive (i.e. also taking in account the private sector as well as all levels of health care), • A contracting process between the Government and health care facilities, so that every facility is not only endowed with clear objectives (production and quality), but also with adequate resources, and • A governance framework where the Government really fulfils its coordination role (for instance, it is strongly represented within the board of directors of every hospital). This model also assumes that health care services are essentially financed by the Government. The annual budget of each facility must be then consistent with the objectives assigned to every facility. Conversely, the demand-based model is based on the hypothesis that demand for health care (preferably subsidized by a health insurance system) will express itself spontaneously and that it constitutes an acceptable guide for the organization of a health care system. Contrary to popular wisdom, this model does not correspond in any way to a non-interventionism policy, as it requires the implementation of very specific mechanisms: • A mechanism for health insurance for the poorest and/or for catastrophic expenses, • Financing of health care services directly by the insured patients, which implies that the hospitals are not financed by global budget, but on fee-for-service (or case) basis, • Governance arrangements essentially assured by the patients or by the representatives of communities. It is necessary to note that there is no evidence today to consider that one of the two models is superior to the other. The problems rather emerge from the inconsistency in implementing one or another model, an inconsistency which is often linked to the lack of an explicit choice between the two models, or merely to the absence of a real willingness to control hospitals and health care centers. Although Benin chose (more or less implicitly) a supply-side model, health care services are very little controlled. Just like many African countries, Benin made the choice, influenced by its history, to give the role of coordinating healthcare services to the Government. However, the mechanisms of this accountability remain very weak. With regard to the planning process, one has seen that there exist today relatively sophisticated mechanisms, with a national plan of development, a 3-year plan of development, action plan and budgets-programs. But the planning process is suffering from two problems. In the first place, these plans are mostly about ad-hoc 126 activities (for example, immunization campaigns) and not enough about routine clinical activities (for example, c-sections). Inevitably, this type of planning leads to a low consideration given to routine clinical activities. In second place, these numerous plans are not complied with. A lot of stakeholders complain about interferences generated by the activities imposed by the Ministry of Health and not planned. On the contrary, commitment to contracting seems strong. So far, very few contracts have been signed between the Ministry and health care facilities, but political will is here very clear. In particular, experimentation has been thrown in 2007 in 3 health districts in order to elaborate and to implement performance contracts. These contracts are consistent with the supply-side model, since they define the clinical objectives to reach and the resources to allocate in the districts. Finally, financing of health care services is, as one will see it farther, quite hybrid and not always consistent. For health facilities, the proportion of their revenues (excluding remuneration of civil servants) coming from the Ministry of Health is estimated between 39 and 56% of total facility budget (i.e. delegated credits). Here the mechanism is a mix between the supply-side and the demand-side models. The major problem of this financing mechanism is its discretionary feature. The yearly amount of the credits delegated (to all health facilities) is discussed between the Ministry of Health and the RHA/DDS (regional health authorities), without using any specific formula to take into account the evolution of health care demand that the facilities face. There are also accountability mechanisms based on demand, but they are not fully developed. The other major source of financing for health facilities is constituted by revenues from cost-recovery, which means the revenues coming from drug sales (user fees). In a surprising enough way, this direct financing by the patients did not generate a culture of patient- focused management. A possible reason is these revenues are captured by the COGECS, for recruiting non qualified staff. Contrary to what can be observed in Senegal (for example), health workers get a very limited portion of these revenues (i.e. in Benin, the bonuses are rarely based on cost-recovery revenues, whereas in Senegal, this so-called "profit-sharing bonus" may constitute more than half of the salary). The other demand-side mechanisms are nearly non- existent. With regard to health insurance, the few mechanisms in place (community based health insurance and health equity fund) cover a very small proportion of the population (see chapter 5). With regard to the quality, one has seen that a system of quality insurance does not exist, nor a mechanism for accreditation. Management autonomy is limited Box 5: Management autonomy for health care facilities: what is it about ? Experiences in many countries have clearly shown that health care services performance can be improved with enhanced management autonomy. But it is also clear that this management autonomy can only succeed with two conditions: (i) if this autonomy is accompanied by a strengthening of accountability mechanisms, from the Government or by the market (as we have seen it previously); (ii) if this autonomy is consistently implemented across the different domains of health care management (i.e. to make hospitals autonomous as regards to the governance without giving them some autonomy on the financial aspect does not lead to any positive impact). 127 Particularly, one can distinguish four main domains for management autonomy, that have to move at the same pace, if one wants to avoid putting health facilities in difficult situations: - Logistical autonomy: it is about allowing health facilities to buy drug and supplies for the quantity of their choice and by the service providers of their choice; it is also about letting them free to invest according to their wishes (or rather according – hopefully – to their strategy); - Human resources management autonomy: it consists in having health facilities (i) free to decide which staff to hire, (ii) free to reward or sanction (even to dismiss) health workers according to their performance, and (iii) free to pay staff (in determining the level of the bonuses for example); - Governance autonomy: it corresponds to the creation of elected boards of directors and giving them decision rights to define a strategy for their health facility and to control its implementation (including changing the director, for example); - Financial autonomy: it exists (i) when a health facility receive (from the Government) non discretionary revenues but rather case-based revenues, (ii) when these facilities can determine their prices and (iii) when they are free to manage their expenses as they want. One immediately observes that management autonomy can only lead to chaos if it does not go with a radical strengthening of accountability mechanisms. Obviously, management autonomization is way different from privatization. In Benin, management autonomy of healthcare facilities is relatively low. If healthcare facilities are autonomous enough with regards to logistics (i.e. notably for ordering drugs and supplies), this autonomy remains low concerning management of human resources. It is linked to the fact that most qualified health workers are civil servants. This implies that their career is managed entirely at the central level. But even for the contract-based staff, their recruitment must be authorized by the district health authority. With regards to governance, autonomy has made some progress, with the recent creation of the COGECS (for health centers) and of the management committees (for district hospitals). However, these bodies are not elected and are therefore not representative. They have few powers. Finally on financial aspects, one has seen that delegated credits are not linked to the actual needs of healthcare facilities. With regard to the revenues from cost recovery, their use is decided by the COGECS, on the basis of a proposal from the HD. Overall, healthcare facilities, whatever is their situation in the health pyramid, are very little autonomous. It contributes to explain their low performance, none of the stakeholders really having an incentive or capacity to improve services provided to the patients, since it will not change their financing or remuneration. Capacities for management are insufficient Unsurprisingly, this low empowerment and low accountability come with a low managerial capacity of staff in charge of managing healthcare services. A strong autonomy requires that stakeholders are trained on strategic and financial management. It is rarely the case in Benin, not only by inadequate training, but also and especially because the low autonomy does not induce stakeholders to request any training. 128 2.4. Do health services payment systems contribute to efficiency? In Benin, revenues for health facility come essentially from 4 sources: (i) in-kind revenues (i.e. health workers paid directly by the Government and posted in health facilities), (ii) cost recovery (user fees), (iii) delegated credits, transferred by the Ministry of Health, and (iv), more punctually, some support from technical and financial partners. Each of these sources poses several problems. Health workers posted in health facilities remained paid at the central level. Just as in many African countries, civil servants (permanent and contractual staff) are paid directly by the Treasury. Considering the difficulties in cash management that could arise at decentralized levels, this solution is probably the most adequate, for the meantime. On the other hand, one can wonder why health facilities are not "owners" of their budgetary allocations for paying staff. In practice, when a staff decides to transfer to another facility, he or she leaves with the financing of his/her own salary. Replacing this staff requires therefore that the facility enters in a long and uncertain lobbying to get the appointment and the financing of a new staff. Performance of the health care services would be improved extensively if, as in an increasing number of African countries, budgetary allocations for staff were affected directly to the facilities, thus allowing them to keep funding for replacing staff. The delegated credits seem to represent an increasing part of the financing of health facilities, but amounts are still not related to actual workload faced by these facilities. It was not possible to analyze in details budgets of health facilities, but a survey in progress (Zou 2008) estimates that these delegated credits represent today between 40 and 60% of the facility revenues (excluding civil servants). The reality is probably close to the low hypothesis (40%), as one could verify it in other regions33. The annex contains a description of these amounts for 2006. One will see in the chapter 9 that funding for delegated credits seems to increase, which is positive. On the other hand, a formal criterion to determine the amount of these credits does not exist yet. The mechanism remains again very discretionary, which does not encourage performance. Cost recovery revenues are the main resource of facilities (there again, civil servants excluded), to the detriment of the fairness and without any gain in performance. Revenues from cost recovery account today for more than 50% of health facilities revenues (civil servants excluded). Most (69%) of these cost-recovery revenues come from drug sales. One will see that the proposed drug prices are often high, which fosters the expansion of the illicit market and constrains access to drugs by the poor. According to Alihonou (2008), "in Benin, the initiative of Bamako has been completely misled". Drug prices (at facility level) correspond indeed by no means to real costs, as one can measure it when comparing the structure of revenues with the structure of expenses. 33 An analysis of 2006 data about delegated credits and communuty revenues shows that they have became the main funding source (between 50% and 80% of revenues, excluding civil servants). 129 Table 24: Cost recovery revenues and expenditures (2006) In total, in spite of the size of cost recovery revenues, this financing mechanism does not reward efficiency. The volume of financing related to post civil servants is mostly linked to past lobbying efforts. The delegated credits also do not follow any formal criteria allowing to link the amount of these credits with the caseload of facilities. Finally, while cost-recovery revenues, by design, are related to the caseload of facilities, they are not accompanied by any mechanism for funding demand. Worse, the cost recovery system does not reward efficiency since it is essentially constituted by drug sales and since the structure of fees or prices is not related by any means to the real costs. 130 Health care services in the public sector: main findings The utilization rate of the public health care services remains insufficient. The utilization rate for outpatient services increases slowly and irregularly and has reached today (2006) only 44%. The rates are better with regard to newborns (83%) and pregnant women (91% for the first prenatal visit). On the other hand, the hospitalization rate, already low, is in constant reduction since 2001. These troubling evolutions can be explained notably by the low progress in the quality of health care, even though data are rare on this point. This low hospitalization rate contributes to explain the limited technical efficiency of the hospital system, which does not go beyond 62%, meaning that 38% of resources (in staff and in beds) are wasted as they do not contribute to production. The health care system remained also very inequitable, particularly for hospital services (hospitalization cases, including deliveries). The low performance of public health care services can be explained by several factors. In the first place, consistency of the health pyramid is limited, as the demise of the referral system illustrates it. Health facilities also endure an uneven distribution of health staff. The reduced availability of drugs (at facility level) and especially their very high prices do not make these facilities very attractive. Conversely, in terms of physical structures (existing buildings), Benin is in a favorable situation, since more than 90% of the population live less than 5 km away from a health center. Finally and especially, in spite of a strong commitment of the Ministry on contracting policies, coordination of health care services remained fuzzy. Few health facilities have today any visibility on the objectives that are assigned to them. In the same way, financing mechanisms for health care services do not permit to reconcile funding and caseload. Also, management autonomy of facilities (notably regarding financial and human resources management) remains modest. A reorganization of the public sector is therefore desirable both (i) to manage health facilities more in line with results and (ii) to give them stronger autonomy in management. 131 Chapter 6 : HUMAN RESOURCES 1. What is the HRH situation in Benin? ................................................................................................133 1.1 Availability of health workers.....................................................................................................133 1.2. Geographical distribution of the HRH .......................................................................................136 1.3. Performance of health workers...................................................................................................138 2.What are the characteristics of the health labor market in Benin? .....................................................141 2.1. Flows coming into the labor market (inflows) ...........................................................................141 Production .....................................................................................................................................142 Production of doctors ................................................................................................................142 Production of midwives.............................................................................................................143 Production of nurses..................................................................................................................145 Production of the other categories of health personnel .............................................................146 Recruitments..................................................................................................................................147 2.2. Level of participation of qualified health workers in the labor market ......................................148 2.3. Outgoing flows from the labor market .......................................................................................148 Premature deaths of health workers ..............................................................................................148 Retirement .....................................................................................................................................149 Migration abroad ...........................................................................................................................149 3. How to explain these results on HRH ? ............................................................................................150 3.1. Explaining deficits in the number of health workers..................................................................150 Flows entering on the labor market (inflows) ...............................................................................150 Production of health workers ....................................................................................................150 Recruitments in the public and private sectors..........................................................................150 Outflows on the labor market........................................................................................................151 3.2. Explaining the unequal geographical distribution of health workers .........................................151 Disparities between the urban and rural areas...............................................................................151 Distribution between public and private sectors ...........................................................................151 3.3. Explaning the low performance of health workers.....................................................................152 Pre-service and in-service training ................................................................................................152 Pre-service training....................................................................................................................152 In-service training .....................................................................................................................154 Capacities, motivation and accountability.....................................................................................155 4. What is the environment for HRH?...................................................................................................157 Wages and bonuses ...........................................................................................................................157 Civil service system...........................................................................................................................158 Decentralization ................................................................................................................................158 Healthcare care delivery system........................................................................................................159 132 1. What is the HRH situation in Benin? One can classically analyze the situation of Human Resources for Health (HRH) regarding three aspects: 1. Availability: is there, nationwide, a sufficient number of health workers and for each category? 2. Geographical distribution: are health workers densities consistent with population densities across regions ? 3. Performance: are health workers competent, productive and respectful of the patients? 1.1 Availability of health workers A significant proportion of health workers (particularly doctors) are employed in the private sector. In Benin, a majority of health workers are civil servants or Government- employed agents on a contractual basis. For these agents, the most recent data are as follow: Table 25: Number of health workers in the public sector HRH category Number Doctors – general practitioners 183 Doctsor – specialists 208 Other medical cadres (dentists, pharmacists…) 31 Sub total medical 422 Nurses 1988 Midwives 828 Lab technicians 391 Imaging technicians 78 Other paramedical staff 2583 Sub total paramedical 5868 Sub total administrative and technical 3033 Grand Total 9323 Source: HRH Development Plan 2007 (PDRHS 2007) In 2003, specific data were collected on Human Resources in the private health sector. However, only qualified personnel were considered during the investigation. Moreover, since a great number of health personnel have dual jobs (i.e. they illegally work in the private sector, in addition to their Government job), it is likely that the reported number of "health workers in the private sector" is only for full-time private workers and therefore corresponds to an important underestimation. , for qualified HRH, current total manpower is as follow: Table 26: Total number of qualified health workers (all sectors considered) Public Private Total Ratio for sector sector existing 1,000 habitants Doctors 494 594 1088 0.14 133 Nurses 2738 825 3563 0.47 Midwives-femmes 642 357 999 0.13 Total 3874 1776 5650 0.75 Source: Statistic directory One can note that most doctors (55%) work in the private sector, though this figure may be over- estimated, given the fragility of the data available on the private sector. For the other categories of health workers, the proportion working in the private sector is also significant (23% for nurses and 36% for midwives), but closer to the average observed in the other countries in sub-Saharan Africa. The doctors/nurses ratio is very different between the public sector and the private sector. In the public sector, the ratio today is one doctor to 5.54 nurses. There is no reference point to judge the relevance of this level. Indeed, such a ratio depends a lot on the kind of training the nurses go through, the tasks they are given and the environment within which they work (health centers, hospitals…). In the private sector, the ratio is largely inferior, with a doctor to 1.39 nurses. This difference can be easily explained by the fact that majority of the private doctors work individually, while public doctors work mostly in facilities. Moreover, this result confirms that such a ratio cannot be analyzed without reference to the health system context. Few foreign doctors practice in Benin. For example, unlike in many of the West African countries, there are no Cuban doctors in Benin. This is probably because Benin has already an adequate number of doctors, which will be shown further in this chapter. One can, however, come across some Japanese, Chinese and Egyptian (specialist) doctors made available as a result of bilateral cooperation. With the exception of traditional practitioners, the number of "informal" HRH is very limited. According to an investigation carried out in 2006, approximately 7,500 traditional practitioners are active in Benin. The number of Community health workers is not known and is probably low. Besides, Benin does not have a major community program in the health sector. Overall, the total number of health workers (qualified or not) in Benin has regularly increased, at least over the last few years. Since 2004, the total number of health workers has approximately increased by 8.7% each year (cf PDRHS 2007). Although Benin suffers from some insufficiency of qualified health workers, this deficit is probably limited. To evaluate Benin needs in the number of health workers, three reference points are available. The first is the traditional benchmark produced by WHO. This reference defines minimum ratios of qualified HRH per capita. A second - and more recent - benchmark was worked out by the "Joint Learning Initiative (JLI) on HRH" and was used by WHO in its 2006 annual report. This benchmark is also expressed in terms of HRH per capita. The third and last reference is that produced by Benin itself. As in many of West African countries, the requirements in HRH are estimated based on the minimal manpower needed for each type of health facility (health centre, district hospital and referral hospital). 134 Table 27: Current number and needed number of qualified HRH Needed number of HRH Existing number of HRH (according to three benchmarks) Public Private Total Ratio JLI-WHO WHO National sector sector number for 1,000 recommandation recommandation recommendations habitants for 1,000 for 1,000 habitants habitants Doctors 494 594 1088 0.14 0.55 0.10 0.12 Nurses 2738 825 3563 0.47 0.20 0.59 Midwives 642 357 999 0.13 0.10 0.22 Total 3874 1776 5650 0.75 2.02 - 2.54 0.40 0.92 Total pop 7,560,930 Source: WHO 2006 report, PDRHS Bénin and WB analysis Among these three reference points, the JLI one is by far the most ambitious. It indeed amounts to 5 times the traditional WHO reference. Moreover, Benin, in its HRH plan, estimates its health workers needs at a quite lower ratio (0.92 instead of 2.3). Given the unusual (and highly unrealistic) results produced by this JLI reference, we considered an intermediate reference in this document, namely that estimated by Benin itself. With this reference, one can see that when doctors working in the private sector are taken into account (though with the risk of a possible double counting), Benin has an adequate number of doctors 3435. On the other hand, the deficit is significant for the midwives and especially for the nurses. Deficits in the number of nurses and midwives could be quickly filled in by the Benin education system, given its high production level. This production is described further in the chapter. Nevertheless, one can already show that Benin could make up for this deficit in a few years. Table below indicates that if the production of nurses and midwives had not been suspended in 2007, the deficits would have been made up for in 2009 for nurses. For midwives however, the deficit will be made up for only in 2012. Table 28: Forescasts for HRH production Needs Existing Estimated annual stocks of HRH (according to national norms) 2006 2007 2008 2009 2010 2011 2012 DOCTORS New graduates (+) 80 80 91 105 140 80 80 Retiring workers (-) 9 9 9 9 9 9 9 Stock at end of year 907 1088 1159 1242 1338 1469 1540 1612 MIDWIVES New graduates (+) 115 115 115 115 115 115 115 Retiring workers (-) 18 18 18 18 18 18 18 Stock at end of year 1663 999 1096 1192 1289 1385 1482 1578 NURSES New graduates (+) 417 417 417 417 417 417 417 Retiring workers (-) 16 16 16 16 16 16 16 34 Nevertheless, an issue remains about medical specialists, for whom the deficit is probably significant. But no easy reference is available to assess these needs. 35 If the number of private doctors is not taken into consideration, there would be a deficit of about 600 doctors, a gap that would be covered in less than 5 years, considering the existing production level of medical schools. 135 Stock at end of year 4461 3492 3893 4294 4694 5095 5567 5968 Source: analysis WB 1.2. Geographical distribution of the HRH Geographical distribution of qualified health workers is strongly skewed in favour of the large cities (Cotonou and Porto-Novo). As illustrated on the graph below, a great number of HRH are concentrated in Cotonou ("Littoral" region) and in Porto-Novo ("Ouémé" region). Whereas these two regions gather only 21% of the population, they have 62% of the doctors, 31% of the nurses and 45% of the midwives. Compared to other African countries, the situation of Benin is of serious concern. The graph below illustrates the degree of iniquity in geographical distribution of health workers. One can observe that Benin is located in the higher part. It moreover reaches a level close to the Sahel countries (Mali and Niger).The absence of policy aimed at reducing this gap thus appears even more astonishing. Figure 61: Index of geographical concentration of health workers in some African countries 0.6 0.5 0.4 Index Doctors 0.3 Nurses 0.2 0.1 0 C er al n a e i al qu ni ny DR g g M Be ne Ni Ke bi Se am oz M country Source : Lemiere 2008 136 Figure 62: Density of health workers per region (public and private sectors) Densities of health workers per region in Benin (number of public and private health staff for 1,000 inhabitants) Littoral Oueme Mono Zou Borgou Midwives - total Nurses - total Collines Doctors - total Atlantique Atacora Alibori Couffo Plateau Donga - 0.200 0.400 0.600 0.800 1.000 Source: Statistics directory 2005 Inequalities in geographical distribution are especially strong for doctors and midwives. For doctors, the ratio between the region with the strongest density and the one with the weakest is 26.5. For midwives, the ratio is 11.7. On the other hand, it is much weaker for nurses, with "only" 4.35. In the private sector, qualified health workers are even more concentrated in urban areas. As one could predict, private doctors are mostly installed in urban areas (79% against 58% for those in the public sector) and mainly in the South. Moreover, 70% of the midwives in the private sector (against 45% in the public sector) serve in urban areas. As to the nurses, 52% of them (against 31% for their colleagues in the public sector) are in urban areas. 137 Figure 63: Densities of health personnel by region (private sector) Densities of health workers per region in Benin (number of private sector HWs for 1,000 inhabitants) Littoral Atlantique Oueme Zou Midwives - private Donga Nurses - private Atacora Doctors - private Plateau Borgou Collines Couffo Mono Alibori - 0.100 0.200 0.300 0.400 0.500 0.600 Source: Statistics directory 2005 / MS 1.3. Performance of health workers Performance of HRH can be analyzed through three aspects: - technical quality (i.e. competence of health workers); - organizational quality (i.e. behaviors of health workers with regard to the patients); - productivity (i.e. ratio between manpower and output). Health workers’s level of qualification is probably insufficient. There is no national survey on the knowledge and skills of health personnel. However, some surveys have explored this field, in particular as regards to maternal health. One of the best surveys is an assessment of obstetric skills and knowledge carried out by Quality Assurance Project. Doctors, midwives and nurses of four Benin hospitals (2 referral hospitals and 2 district hospitals) were assessed as regarding their knowledge and skills on obstetrical care. This same methodology was used in 3 other countries. 138 Table 29: Level of obstetric skills and knowledge in 4 countries (including Benin) Source: Quality Assurance Project 2003 Comparisons with other countries lead to very mixed results. The Beninese personnel are clearly more efficient in term of skills, but their knowledge seems insufficient. More generally, one can only be concerned because of the very weak scores obtained, in Benin and elsewhere. For example, only 54.8% of the personnel know how to correctly manage an eclampsia situation, whereas it is a very frequent obstetric problem and a major cause for maternal mortality. Worse, this survey was carried out on the second and third level of the medical pyramid. It is therefore possible that skills and knowledge of personnel are even weaker on the first level (health centers and maternity hospitals). In the same way, data now available suggest that behaviors of personnel with regard to patients could improve significantly. There is no regular and national survey on patient satisfaction. One can only use case studies. For example, a rather old study (2000) was carried out in the Borgou region. Unsurprisingly, a majority of the surveyed patients complained about the health personnel, about their absenteeism, of their racketeering practices on patients, and their rudeness. Other studies available suggest that this observation is not isolated and especially has not improved ever since. In fact, the observation on racketeering practices was even confirmed by a recent national study (2005), whose results are illustrated on the graph below. 139 Figure 64: Direct payments (official and informal) in FCFA made by health services users (by type of service). O U G P G P I O Source: GAC 2007 Table 30: Direct payments in FCFA (official and informal) made by health services users (by region). O I R N C S O Source: GAC 2007 It is interesting to note that informal payments are frequent both in the public sector and in the private sector, and this, on a comparable level. Whichever the chosen sector, patients will thus have to pay approximately 18% more in the form of informal payments. Box 6A shortage organized on purpose to facilitate racketeering of patients: the radiology department at X hospital "Shortage is sometimes systematically and methodically organized. That is the case, for instance, with the almost weekly breakdowns of equipment within the Radiology department at X hospital. Many technicians, who receive little gifts from the patients of this department, denounce this practice of artificial breakdowns of installations, breakdowns that are triggered by those they call their "bosses". These breakdowns lead to the referral of patients to private radiology centers, where these same "bosses" are employed as “experts” or from which they perceive a percentages out of the generated revenues. Only one sub-department is preserved from these breakdowns, that of the emergencies that workers call "hot line", because patients come always in numbers. Informal payments made there are significant and frequent. De facto, the "hot line" is perceived as a highly lucrative department in this hospital, i.e. a department with strong corruption revenue". Source: Blundo 2003 The politicization of the health system and particularly of its health workers seems significant. According to one of the working groups that prepared this report, “a previously discrete politicization is currently openly practiced: health workers use their political 140 connections to obtain scholarships which they are not entitled to 36 , to obtain decision-making appointments which should benefit to others, or to be maintained in a position, in violation of existing rules”. 2.What are the characteristics of the health labor market in Benin? In order to get a better grasp of the causes of a possible deficit in health workers, it is necessary to analyze the dynamics of the labor market in the Health sector. For this purpose, we used the simple model below. Many analyses are indeed limited to estimations (sometimes in a very sophisticated way) of (i) the HRH needs, (ii) the existing HRH and (iii) the production capacity. The labor market model provides a broader perspective, as it also analyzes both inflows and outflows in the system. It allows identifying a possible pool of under-employed health workers or, on the contrary, significant leakages to foreign countries. The graph below illustrates such a labor market analysis for the Beninese doctors. Figure 65: A labor market analysis of doctors in Benin Under Employed (part time Health care private sector sector) (public) 400 ? Training 1,000 in domestic schools Pool Full time 2,000 High of trained Employed MOONLIGHTING School doctors Training 1,100 ? 3,000 abroad Health care sector 1,000 (full time private) OUTMIGRATION 100 Abroad 1,500 2.1. Flows coming into the labor market (inflows) To better understand inflows of the health labor market, we look here at: (i) the capacity of the secondary education system to "produce" candidates for the various health schools; (ii) the 36 Civil servants should practice for at least four years before obtaining a scholarship for specialization. Some will obtain their scholarship for specialization just after their first year of experience, in using their political connections. 141 production capacity of these schools themselves, and finally (iii) the ability of the public sector to recruit these graduates. Production In Benin, cohorts of secondary school graduates are largely sufficient to "feed" the health schools in candidates. Whereas these health schools admit approximately 650 students per year, secondary schools "produce" approximately 10,000 graduates per year. Clearly, there are no bottlenecks at this level. Table 31: Production of secondary schools graduates 2000-2001 2001-02 2002-03 2003-04 2004-05 2005-06 Graduates from secondary schools 6,020 6,825 6,912 6,467 9,721 10,689 Source: World Bank 2008 Production of doctors In Benin, two medical schools produce approximately 80 doctors every year. The main medical school (Faculté des Sciences de la Santé / FSS) is located in Cotonou and was created in 1971. Since then, it has produced approximately 1,100 doctors (2005 data). Although the number of students has varied wildly, the Cotonou Medical School envisages stabilizing its yearly production around 80 doctors. Figure 66: Number of medical graduates since 1990 (Medical school of Cotonou) Annual number of medical graduates (Cotonou) 90 80 70 60 50 40 30 20 10 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Source: Statistics Directory 2006 / MoH In a bid to accommodate the growing number of medical students and to encourage young doctors to settle in rural areas, a second school was opened in Parakou (northern part of the Country) in 2001. It would produce 11 doctors in 2008, then 25 in 2009 and 60 in 2010. This last 142 number would be the normal production capacity of the school. In all, the production of doctors in Benin will be approximately 140 per year in 2010. Most of the specialist doctors are trained by the FSS in Cotonou.On average, since 1984 (date from which the main medical and surgical specialties were offered by the FSS), approximately 30 specialists graduate every year. The four principal specialties offered are (i) general surgery, (ii) gynecology-obstetrics, (iii) internal medicine and (iv) pediatrics. Table 32: Number of specialist doctors trained in Benin Medical specialties available at FSS Year of Total number of graduates since creation Average annual creation Beninese Citizens from Total number of citizens other countries graduates since creation Anesthesia ND 9 14 23 Biology 2006-2007 0 0 0 0 Cardiology 2006-2007 0 0 0 0 General surgery 1984 52 08 60 6 Dermatology ND 1 2 3 Gyn-obs 1984 84 85 169 17 Internal medicine 1985 9 6 15 2 Ophthalmology 2000 08 10 18 3 ENT 2000 08 10 18 3 Pediatrics 1984 43 18 61 6 Psychiatry 1985 10 5 15 1 Occupational medicine 1998 20 14 34 TOTAL 244 172 416 Source: Statistics Directory 2006 / MoH A significant number (but not known exactly) of specialist doctors have been trained abroad, particularly in cardiology and biology. Production of midwives Benin produces approximately 115 midwives per year. All midwives are trained in only one school, based in Cotonou (Ecole Nationale des Sages-Femmes diplômées d'Etat du Benin / ENSFEB). This school was created in 1965. The annual production has strongly varied these recent years, as the graph below shows it. It has now stabilized around 115 per year. 143 Figure 67: Annual production of midwives by the ENSFEB Nombre de sages-femmes diplômées chaque année (ENSFEB) 200 180 160 140 120 100 80 60 40 20 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Source: Statistics Directory 2007 / MoH 144 Production of nurses Approximately 220 new nurses are trained every year in Benin, but the scope of their assigned tasks and of their training are not homogeneous. First and foremost, we exclusively consider here nurses with 3-year training (state registered nurses). For this category, there is only one school: the Ecole Nationale des Infirmiers et Infirmières diplômées d'Etat du Benin (ENIIEB), created in 1963. Once again, the production level has strongly varied. Figure 68: Annual production of registered nurses by the ENIIEB. Nombre d'infirmières diplômées chaque année (ENIIEB) 500 450 400 350 300 250 200 150 100 50 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Source: Statistics Directory 2007 / MoH Since 2000, the average annual production is approximately 226 registered nurses. For our simulation (cf. infra), we considered a lower level (167), given that this is the most recent production (167) and that the Ministry of Health considers there is an overproduction of registered nurses. Moreover, it should be noted that for ENSFEB (midwives) and ENIIEB (registered nurses), Government decided in 2007 not to admit any students for two years. The principal reason for this suspension is, officially, the declining quality of candidates, and, unofficially, the overproduction of these categories of HRH (an overproduction confirmed by the nurses and midwives’ trade unions). Another school also produces so called “qualified nurses”, but they are recruited at a lower level. This school is located in Parakou. The production has strongly varied and has stabilized today around 250. 145 Figure 69: Annual production of qualified nurses (School of Parakou) Nombre d'infirmières brevetées chaque année 500 450 400 350 300 250 200 150 100 50 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Source: Statistics Directory 2007 / MoH Production of the other categories of health personnel Several health schools in Benin train an important number of health workers for other categories. The levels of annual production, very unstable, are illustrated below: Figure 70: Annual production of other categories of health workers in Benin 80 70 Sanitary technicians 60 Lab technicians 50 Social workers 40 Biomedical engineers (biology) Biomedical engineers (imaging) 30 Hospital administrators 20 Public health specialists 10 ICU nurses and midwives 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Source: Statistics Directory 2007 / MS 146 In Benin, auxiliary nurses do not have any formal training. Inevitably, the qualification level of this cadre is very low. The problem is worsened by unofficial task delegations, which one can be observed in many health centers. In practice, the distribution of responsibilities between nursing auxiliaries and qualified nurses (and even registered nurses) becomes increasingly fuzzy. Quality of health care can only suffer from that. There is no other pre-service training school in the health sector in Benin. In particular, Benin does not train any mid-level cadre, as one can notice in Ghana (medical officers) or in Niger (general practitioners with some training in surgery). Recruitments With the exception of year 2004, the number of health workers recruited as civil servant has been falling for several years. Today, the Ministry of Health annually recruits approximately 15 doctors (including 10 general practitioners), 24 nurses (all categories) and 11 midwives. Figure 71: Number of health workers recruited each year by the Ministry of Health TOTAL PERSONNEL MEDICAL 250 TOTAL PERSONNEL PARAMEDICAL 200 TOTAL AUTRE PERSONNEL 150 100 50 0 2001 2002 2003 2004 2005 2006 Source: PDRHS 2007 Table 33: Annual recruitments of qualified health workers by the Ministry of Health Cadres 2001 2002 2003 2004 2005 2006 Doctors – general practitioners 19 8 24 30 13 9 Doctors - specialists 0 4 3 8 4 6 Other medical staff 0 0 0 0 0 3 TOTAL MEDICAL STAFF 19 12 27 38 17 18 Lab technicians 12 9 10 18 4 5 147 Imaging technicians 4 2 4 7 1 Sanitary technicians 0 0 0 0 7 8 Physical therapists 0 0 0 4 3 2 State registered nurses ("diplômée") 61 30 34 46 19 11 Qualified nurses ("brevetée") 92 29 43 58 15 13 Midwives 39 32 36 48 16 11 Nursing auxiliaries 0 28 0 0 0 Other paramedical staff 0 14 0 9 21 8 TOTAL PARAMEDICAL STAFF 208 144 127 190 86 58 TOTAL OTHER CADRES 135 43 43 77 13 7 TOTAL GENERAL 362 199 197 305 116 83 Source: PDRHS 2007 2.2. Level of participation of qualified health workers in the labor market Having measured incoming flows in the health labor market, it is necessary to estimate the level of participation of qualified health workers, namely the level of absenteeism and the degree of under-employment. Very little is known about the labor market participation of health workers, even if many data sources suggest that the situation in Benin is hardly different from those observed in other countries in the region. Absenteeism is widespread, as well as the dual-job holding practice (for doctors and nurses). Given the level of absenteeism, unofficial task delegations are frequent. For instance, many deliveries are carried out by auxiliary nurses. A significant number of anesthesias are carried out by unqualified personnel. Under-employment of health workers is probably significant in the private sector. We observed earlier that the private sector has become a major component of the health system in Benin. A majority of the young doctors work today in the private sector. In a survey carried out in 1999 on the private sector in Cotonou (World Bank 2000), a majority of the surveyed private doctors explained their decision to settle in the private sector by the recruitment freeze of in the public sector. Given (i) that this freeze is still in effect as we said earlier, (ii) that the production capacity of health workers is not decreasing (it is even increasing for doctors) and (iii) that the private sector remains largely uncontrolled, it is therefore very likely that a majority of the young doctors still settle in the private sector in a bid to avoid unemployment. It is also likely that their workload be small. They are therefore in a situation of under-employment. 2.3. Outgoing flows from the labor market There are several types of "outflows" in the health labor market: (i) premature deaths (sometimes related to an occupational disease or accident), (ii) retirement and (iii) migration abroad. Premature deaths of health workers 148 The premature death rate of health workers is probably low. Although no specific study was carried out, nobody in Benin (neither Ministry of Health nor trade unions) mentions any problem on the matter. It is also true that the HIV/AIDS prevalence rate remains low (1.2% according to DHS 2006). Retirement The number of retirees should not increase significantly in the next years. As indicated in the table below, the number of health workers (public sector) who have to retire between 2008 and 2012 are very variable, but no rising trend has been observed. Nevertheless, this observation must be qualified by the fact that, according to the Ministry in charge of the Civil Service, the average age of civil servants is in rise, which is explained very simply by the impact of the past recruitment freeze. Table 34: Estimates of the number of health workers going on retirement each year. 2008 2009 2010 2011 2012 Total Medical Personnel 0 1 18 21 5 Total Paramedical Personnel 48 56 96 126 69 Total Administrative Personnel 11 13 20 31 15 Total 59 70 134 178 89 Source:PDRHS 2007 Figure 72: Estimates of the number of health workers going on retirement each year Total Personnel Médical Total Personnel Para-Médical 140 Total Personnel Administratif 120 100 80 60 40 20 0 2008 2009 2010 2011 2012 Source:PDRHS 2007 Migration abroad The "medical brain drain" is significant in Benin, although much less than usually argued. As for the migration of health workers, some data are available, but they are not necessary consistent. The latest international research on the matter (Clemens 2007) estimates that 35% of doctors born in Benin work abroad today. Whatever the selected denominator, it would 149 correspond to approximately 300 doctors. Of course, this estimate in terms of "country of birth" is questionable. Some doctors were born in Benin, and then emigrated in France to study there. Except when these studies were financed by scholarships from the government of Benin, it is difficult to view these doctors as part of the "brain-drain" phenomenon. Another widely cited figure (in particular within the Beninese Diaspora) is that "2,000 Beninese doctors are working outside Benin". There are serious doubts about the validity of this assertion. Given that (i) the medical school of Cotonou has trained only 1,100 doctors since its inception and (ii) that today approximately the same number of doctors exerts in the country, this would imply that nearly 2,000 other doctors have been trained outside Benin, which is obviously a serious hypothesis. 3. How to explain these results on HRH ? 3.1. Explaining deficits in the number of health workers In Benin, health workers’ deficits are primarily for nurses and midwives. In the light of our previous analysis of the labor market, what are the bottlenecks ? Flows entering on the labor market (inflows) Production of health workers As previously mentioned, the production capacity of health workers is clearly not a bottleneck in Benin. Whereas 80 doctors (and soon 140) are produced each year, the Ministry of Health recruits only 10 of them. The overproduction is even worse for the other categories (nurses and midwives), so much so that the Ministry of Health had to suspend admissions in paramedical schools for two years. Consequently, the problems related to the training of health workers will be considered more in the performance section (related to training quality) and not in the section relating to the quantity of produced health workers. Moreover, the Ministry of Health also considers that the low quality of health workers’ training is the most significant problem. Recruitments in the public and private sectors We also indicated that recruitments by the Ministry of Health are very limited. However, the number of new recruits remains higher than that of the retirements. That explains (partly) why the number of health workers has increased in the public sector over the last years. Consequently, a majority of the young graduates are under-employed and try to start a private activity. The private sector is probably saturated. According to a recent report on this sector (Scen Afrik 2006), approximately 90% of the private centers are individual cabinets. Two reasons explain the chaotic character of the private sector: (i) the legal environment has never been clarified (consequently, many centers are illegal today), and (ii) in spite of official statements, there is no incentive to public-private partnerships, apart from contracts with some faith-based hospitals. The chaos that results from the absence of a legal environment leads to a 150 wild competition between private health workers, a competition which increasingly makes difficult the absorption by the private sector of the overproduction of doctors. Outflows on the labor market Although far below the levels of Anglophone Africa, migration of Beninese doctors abroad is not negligible. But we also observed that, in spite of this migration, the deficit of doctors in Benin is not high. As for premature deaths and retirement, they play a negligible role in terms of "leakages" from the HRH labor market. Overall, the major constraints on HRH in Benin are primarily (i) the low budgetary capacity of the Ministry of Health to recruit additional health workers and (ii) the chaotic organization of the private sector. 3.2. Explaining the unequal geographical distribution of health workers Disparities between the urban and rural areas Inequalities in the geographical distribution of health workers are caused by the absence of a policy to make rural jobs more attractive. As in many other countries (not only African ones), jobs in rural areas in Benin are not very attractive. For example, many health centers in the Northern part of the country are insufficiently equipped. The population there has low incomes, which reduces the potential for a private activity by a health worker. Career opportunities for husbands or wives, as well as education capacities (for children), are also very limited. These problems, relatively frequent in the region, are worsened in Benin by the absence of any policy that could make jobs in rural areas more attractive. For example, contrary to many other countries of the region, there is no incentive for young doctors (or midwives) to serve a few years in rural areas. There was, in the Nineties, a system that granted health workers a financial bonus for settling in rural areas, but it was abandoned, given that it was difficult to enforce (i.e. verifying the actual presence of recipients in rural areas). The only factor that could possibly help improving the geographical distribution of health workers is the existence of two regional schools in Parakou (one for doctors, created in 2001 and the second, for nurses, created in 1973). However, several incentive measures should come along with these creations (i.e. bonuses, preferential admission for students with a rural background, etc…) Distribution between public and private sectors An analysis of the positive and negative impacts of the development of the private sector tends to portray this sector as a potential outlet for HRH and one that plays a more significant role. The very significant private sector in Benin is probably not a problem. First and foremost, given the poor quality of healthcare in the public sector and the population's strong utilization of the private sector, one cannot be sure that the quality of care is significantly lower in the private sector. Besides, from this point of view, we shall see (chapter 8) that this sector is very heterogeneous. Moreover, we noticed that prices charged (informal payments included) in the private sector are not appreciably higher than those in the public (for ambulatory services). One could object that the lack of regulation in the private sector generates a strong competition 151 between actors of the sector and that this competition generates, at least partly, an induced demand for healthcare services. It is however difficult to estimate the impact of this induced demand. One can nevertheless think that considering the low quality in the public sector and its rather high cost, it would be advantageous, for the population, that the private sector be better organized. 3.3. Explaning the low performance of health workers Pre-service and in-service training Technical quality (i.e. competence) of health workers is, indeed, strongly related to the intensity and quality of their training. More precisely, competence levels of health workers can improve only through a high-quality pre-service training and regular in-service training programs. Pre-service training Doctors Data available on quality of medical teaching are not consistent. There is no study that assesses competences of lecturers and students. Two types of proxy indicators are nevertheless available. (i) The first one is the number of foreign students coming to study in Benin. One can view this number as a proxy of the attractiveness and thus of the quality of the provided training. As indicated in the table below, this number is significant, particularly for medical specialties. Foreign students indeed account for 41% of the graduates in anesthesia and gynecology- obstetrics. (ii) Another type of information available is the ratio of students to lecturer (only available for the Cotonou Faculty of Medicine). It is described below. Table 35: Number of students and lecturers at the Faculty of Medicine in Cotonou Beninese lecturers Foreign lecturers TOTAL Tenured Professeurs Maitres Assistant professors agrégés assistants professors Interns Number 23 25 31 28 1 27 135 Ratio students / lecturers 28.70 26.40 21.29 23.57 660.00 24.44 4.89 Source: Statistics Directory 2006 / MS The total ratio appears adequate (4.89 students to 1 lecturer). However, some officials of the Ministry of Health consider that the Faculty of Medicine has a capacity for 30 students, whereas it accommodates 110 today. De facto, when one looks at the recent evolution of the ratio students / lecturers, there is obviously deterioration. 152 Figure 73: Number of medical graduates VS number of medical lecturers Nombre annuel de médecins diplômés (Cotonou) VS Nombre d'enseignants en médecine 140 120 100 80 60 40 20 0 2000 2001 2002 2003 2004 2005 Source: Statistics Directory 2005 Overall, no firm conclusion can be drawn from the available data. The Faculty of Medicine of Cotonou is obviously still very attractive to foreign doctors. But one can legitimately wonder whether the quality of training has not declined these recent years. Paramedical staff Quality of training for paramedical health workers is considered weak. A detailed assessment of INMES and ENIIAB was carried out in 2005. As indicated previously, INMES includes 5 schools, for registered nurses, midwives, laboratory technicians, social workers and medical technicians. ENIIAB is located in Parakou and trains qualified nurses. The final report of this study clarified some interesting points, namely: (i) there is an overproduction of paramedical staff in Benin (i.e. the number of trained paramedical staff is higher than the number that can be recruited in the public sector and even probably in the private sector); (ii) although those who graduated recently have not been evaluated, it is likely that their knowledge and skills are weak, given that the admission policy in these schools authorizes the enrolment of candidates without a secondary school diploma and given the degradation of the ratio students/lecturers. In fact, the two phenomena are related. Technical quality of graduates has decreased because of the uncontrolled increase in admissions, an increase which puts an enormous pressure on the limited resources of these schools (in terms of the number of lecturers, internship possibilities, equipment, etc…). This increase seems due to the reduction in Government financing. To compensate for the fall of their revenues, the schools have apparently strongly increased the admissions of paying students. Regarding the quality of lecturers, the ratio above mentioned specifies that only 11% of the lecturers are permanent and only 10.4% have the necessary qualifications to teach. As shown in the graph below, the ratio of students to lecturers is three times higher in 2005 than in 2000, moving from 12.1 students to 1 lecturer to 37.5 in 2005. 153 Figure 74: Paramedical graduates VS lecturers Dipômés VS Enseignants a l'INMES 1200 35 30 Nombre d'enseignants permanents 1000 25 Nombre de diplômés 800 20 600 15 400 Diplômés paramédicaux 10 (INMES) 200 Enseignants permanents 5 (INMES) 0 0 2000 2001 2002 2003 2004 2005 Source: Statistics Directory 2005 Following the dissemination of this ratio in 2006, the Government decided to suspend the admissions in these two schools, and this, for two years. This transitional period is to be used (i) to change the admission policy (a priori, only secondary school graduates can now be candidates), (ii) to strengthen training of lecturers, (iii) to recruit new ones and (iv) to improve financial autonomy of the two schools. In-service training In-service training is a frequent activity in Benin, but perhaps more in a bid to obtain additional income than to improve one’s knowledge and skills. No study has been carried out on the volume and nature of in-service training in the health sector in Benin. On the other hand, a majority of the actors complain about a phenomenon called "race to per-diems". A phenomenon very much spread in Benin, it explains the highly absenteeism of health workers, who attend seminars organized by the Government, NGOs or external partners, seminars which come with "per-diem" being paid to each participant. For example, during a recent supervision of a district hospital, out of the 6 midwives assigned there, only one was present, another one was on leave and the 4 others at a workshop. This example is not isolated and similar cases are observed almost everywhere. This suggests that a significant proportion of the personnel’s working days are in fact devoted to various workshops. Here too, no study has been carried out on the contents of these seminars. Moreover, no health center has a training plan. It is thus likely that attendance to workshops is decided more according to their potential of bringing in additional incomes than for the need to improve one’s competences according to a well defined training plan. A study recently undertaken revealed that revenues related to per-diems could be equivalent to 6 months of salary. It should be stressed that this study was carried out after the 2007 decree (which rationalized to some extent the “per-diems” practice) and in a district without any support from a major external partner. It is therefore clear that the collected data underestimates reality. 154 Table 36: Incomes and time related to "seminars" in a health district (2008) FUNDING SOURCES Funding from Funding from NATIONAL Total (in K Equivalent in EXTERNAL partners budget FCFA) salary months STAFF TYPES Number Amount per Number of Amount per of days day days day District health 22 24 6 45 798 3 months officer EEZS Paramedical 7 20.5 3 35 248 2 months Administrative 1 24 3 45 159 ?? Manager 0 24 3 15 45 ?? Doctor 6 24 12 21 396 1 month HZ Paramedical 4 20.5 10 10 182 ?? Administrative 0 24 3 15 45 ?? Chief doctor 28 24 8 35 952 6 months Paramedical 15 20.5 3 30 397 3 months CS Administrative 0 16.5 2 2.5 5 ?? Source: World Bank 2008 Capacities, motivation and accountability Quality, both technical and organizational, of health workers is limited by lack of equipment, of supplies and drugs. No detailed assessment of the work environment was carried out recently. However, an excellent study on obstetrical services is available, a study carried out in 2003 and in all the regions of Benin. The key results are presented below. Table 37: Availability of equipment, supplies and drugs in health facilities offering Basic Emergency Obstetrical Care (2003) EQUIPMENT – Thermometers in labor rooms 70.1% EQUIPMENT – Tensiometers in labor rooms 56.8% SUPPLIES – Catheters in labor rooms 17.9% SUPPLIES – Oxygen in labor rooms 13.7% DRUGS – Ocytocin 30.8% DRUGS –Magnesium sulphate 2.6% Source: Gbangbade 2003 Given that obstetrical care is already supported by specific programs in Benin, there are few chances that the situation would be better for other healthcare services. Worse, the same problems can be observed at the higher level (referral hospitals). These various elements suggest that, even if health workers had regular and adequate in-service trainings, they would still not have the equipment and supplies that necessary for their work. The three performance dimensions37 are affected by the low motivation of health workers in Benin. In 2005, a sample of health workers was questioned on their motivation and the factors 37 Technical quality (i.e. competence), organizational quality and responsiveness (with patients). 155 likely to improve it. Regarding their motivation level, a majority of these health workers estimated that it was average, as the graph below shows it. Figure 75: Self-assessment by the health personnel of their level of motivation Source: Mathauer 2006 In the mentioned study, researchers observed that motivation was higher in the private sector. Two-thirds of health workers in the private sector (14/21) considered their motivation as "very high", "high" or "rather good", whereas only a quarter of health workers in the public sector (11/41) gave similar answers. The surveyed personnel also indicated the factors likely to increase their motivation (see graph below). Three factors seemed essential: (i) the working environment (situation of buildings and availability of equipment and supplies), (ii) the wages and bonuses (cf. infra) and (iii) the opportunity to attend trainings/seminars. Figure 76: Factors likely to improve the motivation of the health personnel Source: Mathauer 2006 Inconsistencies of incentive systems (both negative and positive incentives) also affect the performance of health workers. As seen previously, motivation (and thus performance) of the personnel in the public sector is lower than that in the private sector. Even if wage differences clearly play a role here, many reports also insist on the better supervision ensured in the private sector. Conversely, in the public sector, many health workers are civil servants and have only 156 few incentives to improve their performance. Promotions are still awarded primarily through a seniority-based rule and seldom on actual results. Individual assessments seem extremely rare. Bonuses are not related to performance. Sanctions are very rare. Many officials of the public sector complain about the fact that, even in cases of obvious offence, disciplinary actions are generally cancelled, because of pressures from unions and politicians. 4. What is the environment for HRH? Wages and bonuses Wage levels of the health workers appear relatively low in Benin, even if it is possible that the bonuses partly compensate for this situation. Compared below are the beginning and end of career wages in 4 countries. It is clear that wage levels are lower in Benin. This point is reinforced by the analysis of the ratio between (i) wages of health workers and (ii) the average national income per capita (approximately 19,000 FCFA per month in 2007). This ratio is indeed 8.75 for a general practitioner and 4.10 for a state-registered nurse or a midwife. Though not catastrophic, these ratios place Benin in the low fringe of African countries. One could nevertheless draw more reliable conclusions only after an analysis of the bonus system. But it is almost impossible today to obtain precise data on these bonuses, which proliferated in an anarchistic way in all countries and constitute a significant share (or even majority) of health workers revenues. Figure 77: Beginning and end of career wages for the health personnel in 4 West African countries. 600000 Medecins generalistes 500000 Infirmiers (IDE) Sages-femmes 400000 300000 200000 100000 0 debut fin debut fin debut fin debut fin BENIN MALI NIGER COTE D’IVOIRE Source: Human Resources in the Health seminar in Cotonou (2008) Though wage level is probably insufficient, its structure seems more adequate. By “wage structure”, we imply (i) their evolution with seniority and (ii) their evolution according to responsibilities. One can see on the previous graph that wages in Benin progress rather quickly with seniority, which is a possible factor for motivating staff. On the second point, the situation is less positive. The wage ratio between a specialist doctor and a trainee administrative worker is 157 about 5.80, which seems rather low. This strong compression of wages is unfortunately very frequent in Africa and is largely explained by past wage freeze. The bonuses system remains opaque and not linked to performance. As previously said, it is still difficult to precisely assess the bonuses system in Benin. Among the known bonuses (production bonus, exceptional motivation premiums, risk allowance…), one can note that they are (i) either completely independent from actual staff performance, (ii) or dependent on this performance, but only in a totally virtual manner. They are indeed often based on staff ratings, whereas it is well-known that these ratings (in Benin) are not linked to actual performance (cf Grandvoinnet 2006). The very few bonuses that are linked to performance are those that are financed out of user fees, but they are very low, unlike what can be observed in other countries (for example in Senegal). Civil service system Even if wage levels are probably insufficient in the public sector, the lack of a civil service reform has led to a system of generalized impunity, where high performance is seldom rewarded and low performance never sanctioned. As Grandvoinnet (2006) indicates, "the civil service legal framework today is still that of 1986, i.e. it is former to the political and economic liberalization of the nineties". This status still includes promotions mainly based on seniority. It also includes an automatic recruitment of university graduates. This provision, introduced in 1972, was however suppressed in the Eighties, due to the quasi-bankruptcy of the Beninese State. But it still seems to weigh on the mindsets of civil servants. Generally, it is difficult to find an example of public administration whose history and culture are so far from any idea of performance. Following the failure of the 1998 reform, the Government tried to bypass the civil service legal framework by massively recruiting contract-based staff. In the public sector, health workers may indeed be recruited through three channels: (i) the "Social Measures " (HIPC financing), (ii) community financing (financing resulting from user fees), and (iii) the other funds resulting from investments by government or external partners. Overall, in 2007, more than 65% of the health workers (generally the least qualified) were contract-based. However, this evolution does not seem to have improved general performance of health workers, given the low qualification level of these contract-based workers, their not very rigorous recruitment and the absence of an adequate supervision and incentive system. In fact, in 2008, very strong pressure from unions led to the progressive transfer of these contract-based staff in the category of civil servants No existing policy is there to support a better distribution of staff across regions (Grandvoinnet 2006). There is no obligation of mobility in the career of civil servants. There is no regionalization of recruitments and postings. There is no bonus to reward staff in rural areas. The only significant factor is the creation of the medical school of Parakou, an interesting factor but probably not efficient, given the overproduction of doctors in Benin. Decentralization Even if there were HR procedures to assess and reward performance, they should be decentralized, which is not the case today. Managers in charge of health facilities (from 158 primary health centers to hospitals) have today only very little decision rights in regards to human resources management. All key decisions are made at the central level, either by the Ministry of Health (recruitment and posting), by the Ministry of Finances (wages and bonuses), or by the Ministry in charge of the Civil Service (disciplinary sanctions and promotions). Moreover, this centralization of the decision chain benefits the very powerful unions, which therefore systematically block any attempt by local executives to have their management autonomy increased. As regards working environment (equipments and supplies), we saw that fiscal deconcentration still remains very limited. That also contributes to stockouts. Healthcare care delivery system As we saw previously, many of the human resources problems are related to constraints that are cross-sectoral and not specific to the health sector. However, HR problems are worsened in the health sector by the absence of any link with performance. Improving pre-service training of health workers is a difficult task today, given the little control from the Ministry of Health on training institutions. All the health schools (medical and paramedical) are under the Ministry in charge of Higher education. There is no formal procedure (i) to ensure that curricula are consistent with health priorities of the country, (ii) to make it possible for the Ministry of Health to recruit and appoint lecturers in these schools or (iii) to organize internships in medical centers for students. Lack of equipment and supplies is worsened by the very centralizing approach in planning. We saw previously that planning mechanisms are still not decentralized, which makes difficult to fully take into account health facilities needs. Links between the public and private sectors are still limited. We shall see that in spite of some progress in the integration of private hospitals (non lucrative), the private sector (particularly the ambulatory sub-sector) is still not fully integrated in the health system. Human resources for Health: Main findings Benin suffers a shortage of health staff, but it is relatively limited. Moreover, the production capacities of health schools allow to quickly making up for the deficits. The Ministry of Health can recruit only a small proportion of these young graduates, but many of them could (and can) work in the private sector, which stands to gain a lot in being fully recognized and integrated in the healthcare system. On the other hand, like many the West-African countries, Benin suffers from a very uneven geographical distribution of its health workers. This situation is worsened by the quasi- absence of mechanisms aimed at improving this distribution. We also saw that Benin has a very significant private sector, which must be seen more like an opportunity than like an anomaly. 159 The third and probably more worrying problem is that of the performance of health workers, particularly low in the public sector. Many factors explain this situation, but the most significant are the absence of a civil service reform and the slow fiscal decentralization. 160 Chapter 7 : DRUGS, VACCINES AND BLOOD PRODUCTS 1. Drugs .................................................................................................................................................162 1.1. Strategic management of the pharmaceutical system.................................................................163 1.2. Drug Selection............................................................................................................................163 1.3. Quantification.............................................................................................................................164 1.4. Acquisition & quality control.....................................................................................................166 1.5. Local production.........................................................................................................................166 1.6. Distribution & storage ................................................................................................................167 1.7. Use of drugs ...............................................................................................................................167 1.7.1. Geographical distribution of the structures (or individuals) that provide drugs..................167 Public sector ..............................................................................................................................167 Private sector (for profit)...........................................................................................................168 Illicit market ..............................................................................................................................169 1.7.2. Availability..........................................................................................................................170 1.7.3. Price.....................................................................................................................................171 1.7.4. Prescription and dispensation..............................................................................................174 2. Vaccines ............................................................................................................................................176 2.1. Achived results on vaccination...................................................................................................176 2.2. Strengths and weaknesses of the immunization system .............................................................177 2.2.1. Selection, supply and quality of vaccines............................................................................177 2.2.2. Supply, storage and distribution..........................................................................................178 2.2.3. Dispensation ........................................................................................................................178 2.2.4. Monitoring...........................................................................................................................180 2.2.5. Financing.............................................................................................................................180 3. Blood products ..................................................................................................................................182 3.1. Estimate of needs and planning..................................................................................................182 3.2. Donations, imports, analyses and treatment ...............................................................................182 3.3. Distribution and financing..........................................................................................................183 161 1. Drugs The private sector is the first supplier of drugs to the Beninese population. As depicted on the graph below, approximately 44% of the drugs are bought from the private sector (pharmacies and depots). The public sector comes immediately after. The share of the illicit market (14%) is significant, but not as much as envisaged by the local actors (their estimates go from 25 to 36%)38. Figure 78: Market shares of the three major drug distribution sectors (2006) P P I Source : Abdoulaye 2006 et analyse BM To these three pharmaceutical sectors in Benin correspond three supplying channels, relatively independent between them. As depicted in the graph below, the three channels ((i) public and semi-public, (ii) private and (iii) illicit) are rather independent, each one having its suppliers, its wholesalers and its transportation arrangements. We shall further see that the two essential links are (i) the possibility (seldom used) by the private sector to buy from CAME and (ii) the fact that some health centre personnel in the public sector are involved in the illicit market. 38 As a matter of fact, our analysis probably underestimated the importance of the illegal market. It is actually based on statements from a sample of the population of Cotonou (see. Abdoulaye 2006), regarding the frequency of use of the illegal market, (« have you ever had, at least once, used the illegal market? »). It is essentially based on the assumption that the people interviewed buy the same number of drugs every time they have used one of the three markets (even though their frequency of using one market or another is obviously different). Considering the low prices on the illegal market, it is likely that at every recourse to the illegal market, the volume purchased would be more important. 162 Figure 79: Description of the three major drug distribution channels (2006) Purchased drugs Purchased drugs Donations in Benin abroad Illegal (NGOs (30% (70% importations and donors) of public sector of public sector (e.g. from Nigeria) Origin purchases) purchases) CAME (60% Supply of the annual volume 5 wholesalers and distribution of drugs consumed in Benin) Publis sector Private sector (hospitals (pharmacies, depots Illicit market Delivery and health centers) and private clinics) 15 of demand 42% of demand 44% of demand Source : Abdoulaye 2006 et analysis WB 1.1. Strategic management of the pharmaceutical system The pharmaceutical strategy in Benin was revised very recently, in 2008. The former strategy ("National pharmaceutical policy 2000-2004") was evaluated in 2007 and subject to a complete revision in 2008. Few pharmacists work for the Ministry of Health. Out of the 230 pharmacists registered in Benin, less than 27 work in the public sector, including 7 for the Ministry of Health. Even worse, the Direction of Pharmacies and Drugs has only 3 pharmacists (of which the director). It is obviously difficult to attract and preserve pharmaceutical competences to control Government’s policy in this field. The inspection role over the pharmaceutical facilities is particularly impacted by this shortage of qualified staff. 1.2. Drug Selection There is a list of the Essential Drugs (EM), although it has not been updated since 2003.The first EM list in Benin was prepared in 1989. In theory, this list must be updated every other year. Actually, the current list was prepared in 1993. The number of EM registered on this list is relatively high. The image below illustrates the number of EM to be available at each level of the health care system. This number is a little excessive for the tertiary level (CHNU) (429 EM against an average of 300-400 in other countries) and the primary level (CSC) (335 EM against 150-200 in average in other countries). 163 Figure 80: Number of EM on the list of health facilities (2006) Source: USAID 2006 The EM list seems little used. A study, certainly very partial (carried out in a single health district, that of Ouidah (Somda 2006)) found that all the surveyed health centers selected the EM to be ordered on the basis of past consumption and not on basis of the EM list. Out of the 10 facilities surveyed, only one had a copy of the EM list. 1.3. Quantification In the public sector, the quantification carried out by health facilities is still of poor quality, which makes difficult and thus less efficient supplying functions by the CAME. In theory, as it would be preferable in a supply chain, end-users should be responsible for assessing their needs and thus for placing the orders (pull or demand-driven supply chain). In reality, though the health facilities indeed prepare their drug orders, quantification is not particularly rigorous. The quantification guidelines developed by CAME and the Ministry of Health are seldom used. According to Somda (2006), 70% of the health facilities do not have a regular periodicity for ordering drugs (this periodicity seems to depend more on their cash situation than on inventory situations). None of the studied facilities used a formula to determine the volumes to be ordered. In particular, levels of safety inventory are only seldom calculated. Overall, it is the CAME - and not health facilities - that in fine carries out the quantification. This situation creates additional costs 39, even if the satisfaction rate of orders (placed by health facilities and fulfilled by CAME) is nevertheless quite high (cf infra). In health facilities, the process for placing an order is relatively long. As described by Somda (2006), "once prepared, orders are checked by the person in charge of drug management and the president of COGEC/COGEA, and then transmitted to the head of health district (MCZS) for approval (that is to say 3 visas for only one order). The orders are then centralized at commune level by the accountant in charge of purchasing and distributing the EM. They are finally transmitted to CAME for valuation of every centre’s orders. Once the amount is known, the amount is paid to the commune’s accountant for the purchase". Despite this very cumbersome 39 According to Adeya (2007), about 5% of orders fulfilled by CAME are emergency orders. 164 process, very few health centers prefer buy directly from the CAME (i.e. without passing through their commune). 165 1.4. Acquisition & quality control In the public sector, acquisition is centralized at the level of a national semi-public entity, the Centrale d'Achat des Médicaments Essentiels et Consommables Médicaux, commonly called CAME. Created in 1992, it became administratively and financially autonomous in 1996, when it changed its legal status and became a not-for-profit entity. It is bound by a contract with the Government, a contract which came to an end in 2007 and has still not been renewed40. CAME is essentially financed by drug sales. As an indication, revenues from those sales of drugs and consumables reached approximately 5 million FCFA in 2007. CAME is authorized to sell some drugs to the private sector (with a 20% margin). The purchasing process of CAME seems efficient. CAME carries out its procurement through international tenders, with bids launched every 2 years. The selection of a supplier takes approximately 1.5 months (from the day the tender for bids is published). The first delivery requires 6 to 8 months, between the order date and the effective delivery of the products. All drugs introduced into the country must in theory be registered with the Direction of Pharmacies and the Drug (DPM). Within the Ministry, it is the SESCQ under the DPM which ensures that drugs are registered. The legal texts relating to this registration procedure are precise and consistent with international standards (USAID 2006). The committee in charge of registrations must give a response within the 4 months following a request. Registration is valid for 5 years. A tax of 250,000 FCFA is paid for each registration, but the corresponding revenues are paid to the Treasury and not to the SESCQ.Naturally, this procedure cannot apply to illicit importations, the volume of which is relatively significant, as we said earlier. The process for drug quality control is still fragile. For a long time, Benin contracted out its quality control to foreign countries, particularly Niger. A drug quality control laboratory was finally set up in 2000. It now has a capacity to test approximately 30 (simple) molecules per day. For a significant number of molecules, the tests cannot be carried out by the laboratory and are sent abroad (mostly to Tunisia). The laboratory is not yet certified by WHO. The private sector providers primarily obtains its drugs from the 5 authorized wholesaler- distributors and marginally from CAME. Private health facilities can buy their drugs from CAME, especially for generic drugs. But the bulk of their supply comes from wholesaler- distributors. 1.5. Local production Local producers are very few in Benin. For many years, only three Beninese companies have been producing drugs or medical supplies. The major one is Pharmaquick, which produces approximately 72 generic drugs in the form of tablets. Second is Bio-Benin, which provides especially massive aqueous solutions for perfusion. The last company is SOPAB, specialized in bandages (of which a part goes into exportation). These producers have quite high production 40 Which creates difficulties for CAME, as it no longer enjoys custom privileges. 166 costs. That explains why CAME imports more than 70% of its products and why certain external donors established in Benin also resort to foreign companies, including for basic products (like impregnated mosquito nets). 1.6. Distribution & storage In the public sector, distribution and mass storage are ensured by CAME. At CAME, a considerable share of inventories are lost by obsolescence. According to the 2007 annual report of CAME, the rate of loss because of expired period was less than 1%, which appears to be a good result, to be confirmed by an independent audit. Two factors can explain this expiration problem at CAME. First, CAME’s storage capacities are more and more limited. CAME has a central warehouse in Cotonou and two regional warehouses, in Parakou and Natitingou. These three warehouses seem now saturated. Secondly, it seems that the lack of coordination between CAME and international agencies brings about expired supplies. CAME also stores and manages drugs provided by external agencies (WHO, WB, etc…) without charging any management fees. CAME is indeed considered as a relatively efficient entity and is seldomly bypassed by the multilateral agencies. But CAME is not allowed to ask for the reimbursement of the expenses thus created, a situation that jeopardizes its financial health. At health facilities level, conditions of drug storage are inadequate and probably explain significant losses. For example, in the health district of Ouidah (Somda 2006), less than 40% of the health facilities have adequate storage capacity to protect drugs from the sun’s heat and direct light. The purchasing process is relatively long. The delay from placing an order (by the commune with CAME) to the beginning of distribution in the communes is on average 11 days, as for the communes sampled by Somda. In the private sector, five wholesaler-distributors ensure supply and distribution. They are: GAPOB, SOPHABE, PROMOPHARMA, UBPHAR and GBPHARM. 1.7. Use of drugs41 1.7.1. Geographical distribution of the structures (or individuals) that provide drugs Public sector Geographical accessibility to the public entities that distribute drugs is relatively satisfactory. In Benin, all hospitals and health facilities (CSC and CSA) must provide drugs. Further, we shall see that drug availability, including at health facilities, is within the average of 41 This part includes 4 sub-parts, each one corresponding to the 4 possible obstacles for accessing drugs: (i) geographical accessibility, (ii) availability, (iii) affordability and (iv) quality of prescription. 167 other countries in the sub-region. Geographical accessibility to these structures is 86% (i.e. 86% of the population lives within 5 km from a public health facility). Private sector (for profit) Conversely, accessibility to private pharmacies is very limited, given that they are concentrated along the Coast. Benin has approximately 161 private pharmacies, but 60% of them are installed in the Atlantic region (Cotonou and Porto-Novo). Table 38: Distribution of pharmacies by region (2006) Number of Number of pharmacies Area served (in sq km) per pharmacies / 10,000 habitants pharmacy Alibori 3 0.050 8747 Atacora 1 0.016 20499 Atlantique 18 0.192 180 Borgou 9 0.109 2873 Collines 4 0.067 3483 Couffo 2 0.034 1202 Donga 2 0.052 5563 Littoral 86 1.219 1 Mono 3 0.076 535 Oueme 27 0.334 47 Plateau 3 0.067 1088 Zou 7 0.107 749 Source : Annuaire statistique 2006 Figure 81: Distribution of pharmacies by region (2006) N Source : Statistics directory 2006 Consequently, according to 2001 data, approximately 80% of the urban population had access to a private pharmacy, as against only 15% of the rural population. 168 To increase coverage of the Benin territory in terms of private pharmacies, the Government conducts an active, but not always consistent policy. In order to guarantee a minimal revenue to private pharmacies, their creation must be authorized by the Ministry of Health, according to legal texts that define maximum densities of pharmacies per region. In the same way, the Government allowed retired doctors to create and manage pharmaceutical depots. These depots must be linked to a pharmacy, both for their supply and their supervision (which must be conducted by a pharmacist). However, this controlled installation policy is not implemented in a rigorous way (i.e. there are actually many illegal creations of depots 42, managed by traders, without any supervision by a pharmacist). Especially, as will be described below, the price control system in the private pharmaceutical sector heavily reduces profitability of private pharmacies that is located far away from the main sources of drug supply. This largely explains the quasi-absence of private pharmacies in rural areas. Illicit market Informal vendors who sell drugs imported in an illicit way constitute a third type of significant supplier. We previously said that the share of drugs bought on the illicit market is estimated between 14% and 36%. It is likely that this market share of illicit drugs is even higher in rural areas, because of the quasi-absence of private pharmacies and the greater ease of communication with bordering countries. The exact number of these informal vendors is not known, but a study carried out in the largest Beninese cities listed approximately 6000 of them (often women). Half of them worked as itinerant salesmen and women (selling door to door). The other had stalls in markets. The illicit market is mainly supplied by Nigeria, but also by some health workers. It seems that a majority of the drugs sold on the illicit market are clandestinely conveyed from Nigeria, having been either imported through Nigeria, or manufactured by clandestine Nigerians laboratories. Even if no quantitative estimates are available, many Beninese actors also confirms that a significant number of health workers (public sector) sell, through middlemen, drugs taken or stolen from the health facilities’ inventories, generally drugs whose expiry date has been reached for a few days. Government started to address the problem only lately and not with a very firm will. The necessary legal framework was enacted in 1999. A spectacular crackdown operation was launched the same year, but without durable effect. It is only recently (2004) that the repression policy was abandoned in favor of public awareness campaigns on the dangers of drugs bought from illegal retailers. An assessment (cf Abdoulaye 2006) of the campaign carried out in Cotonou in 2004 suggests that its impact was very significant, even if it is still difficult to quantify. Overall, one can nevertheless question the strength of Government's will to fight the illicit sector. In any case, as we shall further see it, high prices of drugs in Benin remain a major obstacle to the reduction of the extent of this illicit market. The Abdoulaye (2006) study on the illicit market in Cotonou highlighted that the low price was the principal factor mentioned by 42 Officially, there are 270 depots, but a more realistic estimate is close to 1,000. 169 households for using this market. 86% of the surveyed persons put this factor in first position.A majority (82%) also acknowledged that these illicit drugs are dangerous. But it is not sure that the population really has the possibility to turn away from the illicit market, given the prohibitive prices charged in the public sector (see below). 1.7.2. Availability At CAME, the availability of EM is probably correct. Though CAME has selected 40 EM that it strives to make always available, there is not regular monitoring. The total availability ratio was 92% in 2002 and 90% in 2007. Table 39: Availability of some tracer drugs at CAME At health facilities’ level, although the availability of EM is not monitored in a regular way, several recent studies suggest that this availability is relatively satisfactory. One can only use two recent studies. The first is that of Somda (2006), in the health district of Ouidah (n=10). The second is that of Gbangbade 2006, carried out in 4 departments (n=40). According to these two surveys, the average EM availability rate in the health facilities was 81% overall (cf table below), and even 83% if we do not take into account condoms (whose availability ratio is generally weaker than for the EM). Table 40: Availability of 13 tracer products at health facilities level. Study from Gbangbade (2006) Study from Somda (2006) TOTAL Data 2004 Data 2005 Data 2006 (n=40) (n=40) (n=10) Antimalaria drugs Chloroquine 92 85 99 92 Sulfadoxine + Pyriméthamine N/A N/A 87 87 Antibiotics Cotimoxazole 90 83 86 86 Amoxicillin 90 80 N/A 85 Ciprofloxacine N/A N/A 80 80 Metronidazole 82 80 N/A 81 Pénicilline N/A N/A 52 52 Doxycycline N/A N/A 82 82 Analgesics Paracétamol 90 83 91 88 Nutritional Fumarate fer 90 88 98 92 complements Acide folique N/A N/A 98 98 ORS N/A N/A 77 77 170 Condoms Condoms 69 43 N/A 56 TOTAL 86 77 85 81 Sources:Gbangbade 2006 and Somda 2006 This 81% availability rate is hardly different from those found in the neighboring countries (86% in Mali according to Niangaly 2001 and 89% in Burkina according to Ridde 2005). It is possible that this 81% rate corresponds nevertheless to an improvement compared to the previous period. In Benin, this availability ratio was indeed 70% in 1996 (according to Nkuzimana 1996) and 75% in 2001 (according to the 2004 World Bank sectoral review). Drug availability problems at health facility level appear related to weaknesses more in inventory management at facility level than at CAME level. We earlier saw that the availability ratio of EM at CAME is 90%, but only 81% at health facility level. Somda (2006) found that the fulfillment rate (by CAME) of orders placed by health facilities was however close to 99%. In its 2001 annual report, CAME estimated this rate at 90%. Table 41: Fulfillment rate of orders received by CAME. Rate of order fulfillment (%) Cotrimoxazole 100 Chloroquine 100 Sulfadoxine 100 + Pyriméthamine AAS 100 Paracétamol 95 Peni 100 Iron 100 Folic acid 98 ORS 100 Ciprofloxacine 100 Doxycycline 100 Source: CAME 2007 This suggests that the limited availability of drugs is not related to the supply (quantification and acquisition) and distribution processes, which are managed by CAME. The bottleneck seems actually located at the health facilities level, as they obviously do not order their EM according to their needs or their consumption forecasts. 1.7.3. Price It is likely that drug prices are particularly high in Benin. Although the country did not take part in the 2004 REMED comparative survey, some facts suggest that drug prices have become very high in Benin. A comparative study carried out by PHR plus (2004) and based on a sample of 25 National Health Accounts reveals that Benin is the 2nd country with the highest proportion of health expenditure devoted to drugs (76% of the households health expenditures). 171 The principal factor to explain these high prices of drugs is the absence of price control in the public sector. Several recent studies estimated that the multiplying coefficient (i.e. margin rate) in health facilities and public hospitals was between 2.5 and 5. By way of comparison, in the private sector, where, unexpectedly, the prices are strictly controlled, the coefficient is only 1.7843. In public health facilities, prices are set, not always very explicitly, by the COGECS, which then manage revenues from drug sales44. Thus, there is a conflict of interest here, as the COGECS has an incentive to set prices at very high levels so as to maximize their revenues. The table below illustrates the situation in the Tanguieta district and shows that patients are better off getting their drugs from the illicit market, given the high prices in the health facilities. Even when the COGES/COGEC set prices at a reasonable level, it is not rare that health workers require an unofficial supplement from patients. They can also pretext a stock out so as not to sell some drugs and to lure patients into buying these drugs either directly through them, or on the illicit market (that they largely supply). Just as some health actors in Benin said, "the principles of the Bamako Initiative have gone astray in Benin". Whereas cost recovery was to be used only to ensure a better availability of drugs, this source of income has become the financial "cash cow" of health facilities. Interestingly, while drugs sales account for 69% of cost recovery revenues of health facilities, it corresponds only to 36% of their expenditures (2006 data). In other words, 48% of the average price of drugs is used to cover other expenditures than those necessary to the replenishment of drug inventories. One can also interpret this result as an indication that drugs are 48% too expensive in the public sector. It is interesting to note that the same drugs sold on the illicit market generally have a price 50% lower than those sold by health facilities, which can suggest that a 50% reduction in public drug prices would make it possible for households to easily face this expenditure. Table 42: Selling prices of some products in the district of Tanguiéta (2005) Products Unit Price at Price Price at Price at Margin rate applied CAME at district health illicit by health centers center market Ibuprofen Tablet 6 11 17 10 2.83 Paracetamol Tablet 2.5 4.5 12 6 4.8 Chloroquine en vrac Tablet 3.15 4 7 5 2.2 Chloroquine en blister Tablet 5 6 9 5 1.8 Source: KPATCHAVI 2005 Box 7A recent attempt at controlling drug prices in the public sector In October 2006, Benin adopted a decree on the harmonization of drugs’ prices in health facilities. One of the major objectives of this decree was to limit the margin rate applied by health facilities. An evaluation of this decree is envisaged in 2009. Some results are however available in the health district of Zogbodomey - Bohicon - Zakpota. They show that, though the maximum margin rate is complied with, it seems that health facilities compensate for this drop in their incomes by an over prescription. In other words, the decline in 43 That makes new pharmacies outside the capital city not profitable. As an example, to send a pack of paracetamol 400km away from Cotonou (sold in Cotonou for 85 FCFA), this pack should be sold at about 300 FCFA, to cover shipment costs. Price control mechanisms do not allow such an increase. 44 According to Somda (2006), in 70% of cases, prices are set by COGECS following non written rules. In other cases, prices are set according to those observed in neighboring health centers. 172 unit prices is offset by an increase in quantities.It is therefore probable that drug prices in the public sector will not drop significantly unless the financing system of these health facilities undergoes a radical reform. Despite price freedom in the public sector, prices are relatively uniform from a region to the other. A study carried out in 2006 measured these price differences across all regions (for public and faith-based sectors). We selected 12 tracer products (cf figure below). It appears the average price variation coefficient (between the departments) is 0.24, that is to say the average price variation ranges between +/-24%. Figure 82: Drug prices across regions (2006) 70 ATLANTIQUE 60 LITTORAL 50 MONO 40 COUFFO OUEME 30 PLATEAU 20 ATACORA-DONGA BORGOU ALIBORI 10 ZOU COLLINES 0 ) 0 iV ne e e 0 e le e ne e ol 50 at 40 id in ch illin id zo én m cli ci ac ar qu ac e ta da xa e (p cy yc um in ro ol ce ue ue ni lo xy ox m ne az lo ra rf of ro q liq ha do am illi yli ox pa fe pr ét fo ét ic lic m m ci én rim sa tri lp co py yl hy ét e- ét ac xin ym do ox lfa én su ph Source: Benaudit 2006 Figure 83 Variability in drug prices per region (2006) 60 0.6 Moyenne 50 0.5 Coefficient Var 40 0.4 30 0.3 20 0.2 10 0.1 0 0 e le 0 ol e ) e e ne 0 e ne iV at id in 50 cin 40 m zo id én ar illi qu cli ac ac ta da xa e e um yc cy (p ce ro in ol ue ni fl o ue ox xy m lo az ra ne rf ro liq ro liq ch ha am do pa ox fe ét illi fo c ip icy ét m m ic rim én al tr i yls co lp py hy ét e- ac ét xin ym do ox lfa én su ph Source: Benaudit 2006 173 Price differences between the public and the faith-based sectors are not significant. Using again the previously-mentioned data (with the same tracer products), one notes that the average price observed is 19 FCFA in the public sector and 18 FCFA in the faith-based sector. The difference is not significant. It was not possible to carry out a comparison with the private for- profit sector. Figure 84: Drug prices by sector (2006) 60 Public 50 Prive non-lucratif 40 30 20 10 0 ue t e le 0 e ol ) f lo ne ne e xy 00 r f ine ox d e ha éniV id 50 ch illin a py illin a m zo ui o x a ci 4 ar ac i cl ac da cip roq e e et um yc cy p x in ol ni e rim e ( x in én rac m lo az qu ro liq ro am do pa fe ét li fo icy ét m m ic al tr i yls co do ylp ét e- th ac s u mé y ox lfa én ph Source: USAID 2006 1.7.4. Prescription and dispensation Beninese people often resort to self-medication, which probably reduces quality of prescriptions. According to Abdoulaye (2006), only 56% of the questioned people consulted a doctor before buying drugs. In health facilities, drug dispensation is often done not in a rigorous way. Although a majority of health facilities have process charts, they are seldom used. Clerks in charge of selling drugs are obviously insufficiently qualified45 and trained. As an example, during the Somda investigation (2006), all the surveyed clerks claimed that they deliver only drugs that do not need prescription, although they were not able, of course, to provide a list of these “over-the-counter” drugs (it does not exist). Hospitals are not really involved in rationalization of prescriptions. There are no drug committees in the hospitals. Some of them however have committees in charge of analyzing prescriptions. Very little is known about their operation and their impact. 45 They are usually recruited by COGES/COGEC without any requirement (degree or experience). 174 The process charts prepared by the Ministry of Health are not disseminated in the private sector. 175 2. Vaccines 2.1. Achived results on vaccination Results as regards immunization coverage seem overall favourable, but remain prone to interpretation. According to the data from the Ministry of Health, the immunization coverage has strongly progressed since 2003. This encouraging trend is illustrated by the graph below, even if it is still incomplete (since it does not measure the "number of completely vaccinated children"46). Figure 85: Evolution of immunization coverage according to the Ministry of Health (2003-2007) 140 120 100 Polio 80 BCG DTC3 60 VAR / VAA VAT2+ 40 20 0 2003 2004 2005 2006 2007 Source : Ministère de la Santé Data resulting from the DHS-2006 survey suggest that these reported results have to be qualified. The table below shows that the observed results (for the year 2006) are obviously weaker than reported. More seriously, data from the last three DHS investigations (1996-2001- 2006) suggest a deterioration of immunization coverage (i.e. proportion of completely vaccinated children). This proportion indeed went up from 56% in 1996 to 59% in 2001, and then fell to 47% in 2006. It is possible that this degradation be related to the success of the national vaccination days (JNV). These campaigns may have encouraged mothers to await the JNV events rather than spontaneously taking their children to health centres. 46 According to WHO, a child is completely vaccinated if he/she received BCG, three shots for polio, three DTC shots and measles. These vaccines must be received before a child turns 12 months. 176 Table 43: Comparison of immunization coverage data between Ministry and DHS (year 2006) 2006 Data from MoH Data from DHS Children aged 12-23 Children aged less than months 12 months BCG 100 88.3 87.9 DTC3 93 67 64.5 Polio 100 63.9 61.6 Measles (VAR) 89 61.1 51.1 All vaccines ??? 47.1 40.2 Source: Ministry of Health and DHS 2006 Inequalities of immunization coverage are still strong, particularly in relation to socio- economic status and residence. Inequalities of immunization coverage between regions are quite limited, as shown in the graph below. Moreover, their directions are rather unusual. For instance, the Ouémé/Plateau region is the least well covered, despite its urban character and its proximity with Porto-Novo, the second largest city of the country. On the other hand, the inequalities are very marked according to areas of residence (43% of vaccine coverage in rural areas as against 55% in urban areas) and according to socio-economic status (34% for the poorest and 65% for richest). Figure 86: Coverage in completely vaccinated children across region Source : DHS 2006 2.2. Strengths and weaknesses of the immunization system 2.2.1. Selection, supply and quality of vaccines The EPI (Expanded Program on Immunization) ensures the greater part of vaccines supply. In Benin, vaccines are bought by 3 organizations: EPI (through UNICEF), DNPS (National Directorate in charge of Health Policies) and private wholesaler-distributors. 177 The purchased vaccines are consistent with Benin’s epidemiologic profile. The EPI uses 9 antigens, which correspond to the major diseases prevailing in Benin. The DNPS takes charge for ensuring supply of vaccines against epidemics (i.e. meningitis). Quality of vaccines is adequately controlled. When tenders for bids of vaccines are launched, the LNCQ controls batches based on samples. 2.2.2. Supply, storage and distribution Availability of vaccines seems adequate. Despite the absence of precise statistical data on vaccine availability, the latter seemingly reaches a high level, as can testify the good results as regards immunization coverage (cf above). This adequate availability is largely related (i) to the use of rigorous need assessment methods (UNICEF’s "Forecast" model) (ii) to a secure financing and (iii) to a very regular supply to the regions. It should be emphasized that the whole process is computerized. The cold chain is functional, despite a worrying ageing of some equipments. At all levels of the health pyramid, storage premises are complying with cold chain47 standards. These premises are regularly supplied in spare parts. Maintenance personnel are available in several regions (5 out of 12) and in some health districts. An equipment inventory is carried out frequently. However, data obtained during the latest inventory depict a rather significant ageing of equipment. In fact, 92% of the cold chain equipment will need to be renewed before 2011 (when one considers an 8-year depreciation period). Also, vehicles are often already fully depreciated (63% of the 2-wheel vehicles and 48% of the 4-wheel vehicles). A renewal plan was prepared in 2006. 2.2.3. Dispensation The EPI uses a broad range of strategies to ensure adequate immunization coverage of the population. Four immunization strategies are regularly implemented in Benin. (i) The first and most classic, is that of vaccination sessions in health centres. It should be mentioned that the private and non lucrative medical centres are involved in this strategy. (ii) This is supplemented by an outreach strategy, for towns or villages located at more than 2 km of health centres. This strategy is strongly based on Community health workers and uses a financial incentive system (vaccination bonuses) (iii) The organization of immunization campaigns is a third strategy, used very often (especially for the oral polio vaccine, the VAR, the anti-amaril and the VAT). (iv) Last, - and fourth strategy - additional sessions are organized in some villages. Despite these efforts, some areas in the country are still not adequately covered. At the moment, it is still not very clear why some towns/villages present a low immunization coverage. Some supply-side factors are obviously at work, given the weaknesses of resources available in certain districts. Nonetheless, factors related to demand-side also probably contribute to this 47 In particular, all equipments have internal thermometers or devices for recording temperatures. Vaccines have their own control system. 178 situation. Efforts on communication towards the populations are still insufficient. Despite the existence of a policy on the matter, resources for communication are still primarily positioned at the regional level. Few districts have an EPI communication unit. 179 2.2.4. Monitoring The epidemics surveillance system is still fragile. A monitoring system was set up in 2003, under the aegis of the DNPS. A network covering the whole territory has been put in place by regional and local committees for epidemics management. Focal monitoring points have been set up in some districts. But the arrangements are still partially functional. Notification of diseases under monitoring is weak. That is related (i) to the lack of resources allocated to the surveillance system (training, resources for communication, computer equipment…) and (ii) to the absence of a mechanism to involve communities and the health workers (in particular Community health workers). 2.2.5. Financing Financing of vaccination programs is still strongly dependent on external resources. In 2005, Benin spent approximately 10.5 million USD for its immunization program. This amount is almost equally shared between the routine vaccination and additional vaccination campaigns. As depicted in the graph below, Government funds approximately 26% of the routine expenditure. Today, the bigger share of the financing (48%) comes from the GAVI funds, which supports Benin since 2002. Other funders are (i) households, through community financing (10%) and (ii) the HIPC funds (14%). The financial sustainability of the EPI (especially after 2010, when GAVI's support would be reduced by the mechanism of co financing) is thus rather weak. Even as the budget necessary to the immunization program will be weaker in the next years (approximately 6 million USD per annum), it is likely that only half of this amount could be provided by the Government (2006 WHO report). Figure 87: Structure of financing for routine vaccination (2005) 180 Source: 2006 WHO report 181 3. Blood products 3.1. Estimate of needs and planning There is no strategy on blood transfusion yet. A master development plan on blood transfusion exists for the 2008-2012 period, but it is more an estimate of the budgetary needs. A comprehensive strategy is still needed. Requirements in blood products are probably not met, even if they are still to be better evaluated. The Ministry of Health estimates annual requirements of blood bags (complete blood) between 120,000 and 130,00048. This estimate seems based on a study carried out in 2004 (Association of blood donors of Benin) which observed that only 50% of demand was satisfied. Given that blood donations oscillate between 50,000 and 60,000 pockets per annum (cf table below), one can assume that this figure was simply multiplied by two in order to reach an estimate of the total needs. Finally, it should be noted that blood requirements are strongly related to malaria outbreaks. This seasonal variability makes adequacy between supply and demand even more difficult. Figure 88: Recent evolution of blood donations in Benin (2003-2006) Dons du sang (en nombre de poches) 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 2003 2004 2005 2006 NB: The 2006 data are incomplete, as one of the regions did not submit all data. Source: Ministry of Health 3.2. Donations, imports, analyses and treatment For its supply, Benin depends entirely on voluntary blood donations, and efforts to encourage donations are still very limited. Benin does not officially import blood bags. This does not prevent some patients from getting blood bags from the neighbouring countries. Also, some blood bags cross the borders in direction of neighbouring countries. The policy in favour of blood donation is very little followed. There are certainly mobile sampling teams, but they are very few and inadequately equipped. The Association of blood donors does not receive any support from the Government. Inevitably, communication actions in favour of blood donation are 48 According to the Association of blood donors, the most frequent indications for blood transfusion in Benin are the following: anemia and malaria among children (50%) and postpartum hemorrhage (30%). 182 very sporadic. Lastly, the practice of auto-transfusion is still rather rare (less than a hundred bags per annum), although it is internationally recognized as a practice contributing to transfusion safety. It is truly not encouraged in Benin. Biological tests are carried out on blood bags before distribution. Despite some stockouts of reagents, the 6 Regional Services of Blood Transfusion (SDTS) make sure all the bags are tested and this for the main risks (HIV, hepatitis B and C, syphilis). These tests lead to rejecting approximately 15% of bags. Processing capacities (i.e. centrifuging and distillation) are small. The scarcity of centrifuges leads to a production being primarily made in the form of whole blood. Red cells and platelets concentrates are very few. This situation clearly results in a significant wasting of collected blood. 3.3. Distribution and financing Distribution of blood bags is carried out according to requests made by health centers. Given the small amount of available blood bags, no safety inventory is held in health centers. These are re-supplied according to their requests by the 12 regional blood banks (BDS) and the 34 blood transfusion post (PTS), supplied in their turn by the SDTS. Financing of blood transfusion primarily rests on the Government and external partners. The total blood transfusion expenditures can be estimated around 1 billion FCFA per annum. More than 80% of this amount is financed, in equal shares, by the Ministry of Health and external partners (particularly CTB). The rest is borne by the patients, who pay approximately 2,000 FCFA per bag and per transfusion. Contrary to the situation existing in many countries, hospitals do not financially take part in the transfusion system, apart from a very weak support to the blood donations. 183 Drugs, vaccines and blood products: main findings With regard to drugs, the supply chain functions relatively well. The availability ratio of essential drugs at health centers level is slightly higher than 80%, which is still unsatisfactory, but in line with the results reached by neighbouring countries. Stock outages seem linked to weaknesses more in stock management by health facilities than in CAME’s purchasing and distribution arrangements. This is what was observed (cf chapter 2) concerning drug stockouts for obstetrics. The principal obstacle for accessing drugs actually resides in prices, which are obviously very high in Benin and weighs strongly on households (76% of the national expenditure on drugs are directly borne by the households). There too, it seems the problems come more from the very high margins (sometimes higher than those of the private sector) applied by the health centres and less from prices obtained by CAME. This situation inevitably feeds a very significant illicit market, market which was fought timidly by government. It is also likely that drug prices at health centre level cannot drop without a radical revision of the financing mode of these centres. As for vaccines, Benin is one of the success stories in the sub-region, with immunization coverage rates close to 90% and this, for several years. The main challenge of this sub-sector lies in its very fragile financial sustainability, as funding is still largely dependent on donors and especially on GAVI. It is however clear that the efforts of Benin as regards immunization have had a major impact on children’s health and are probably the major reason for the drop in infant mortality over the last years. Conversely, the sub-sector of blood transfusion is still very poorly organized. The development of a national policy would constitute a first step towards achieving results. 184 Chapter 8 : PRIVATE SECTOR 1. A quickly expanding sector...............................................................................................................186 2. An often illegal and low quality sector .............................................................................................187 3. A legal framework that is difficult to enforce and thus not enforced................................................189 4. The development of the private sector still remains poorly promoted ..............................................190 185 The private sector is particularly diversified in Benin. It includes not only individuals (physicians, midwives, nurses…) but also health facilities of various sizes (i.e. from small pharmacies to hospitals, or group-owned clinics). There also exist two main types of legal status: faith-based entities and for profit entities, to which must be added the NGOs sector, consisting sometimes of fake NGOs working mainly for a profit-making purpose. We have excluded from this chapter the pharmacists, who are addressed in chapter 4.4 of the report. 1. A quickly expanding sector More than 1,400 private health entities exist in Benin. Data on private centres are mainly collected by the RHA/DDS on a relatively irregular basis and without any common methodology. Data are particularly fragmented for the most urbanized regions (Atlantic and Littoral), i.e. those where the private sector is the more widespread. Despite this deficiency, the latest estimates (see table below) show a very fast increase of the private sector (almost 4% annually). One can note that this expansion of the private sector has accelerated mainly in rural areas, where it still remained low. Table 44: Private health centres across regions 1998 2005 Average annual growth rate Atacora / Donga 23 54* 18.61% Borgou / Alibori 98 132 4.35% Mono / Couffo 66 150 12.44% Ouémé / Plateau 187 287 6.31% Zou / Collines 93 260 15.82% Atlantique / Littoral 526 526** N/A TOTAL 993 1409 4.47% Source : Scen Afrik 2006 report * Data 2003 ** The 2005 census being very partial in the Atlantic/Littoral Department, we have by caution merely taken the 1998 data, even if they probably underestimated the current situation. The private sector has probably a predominant importance in the outpatient sector. Though it is very difficult to obtain reliable data, the private sector is estimated to have ensured about 4,900,000 outpatient visits in 2006, i.e. 63% of the consultations carried out in the country. We have to specify that faith-based facilities (members of AMCES) perform half of those private visits. Health services providers are however little involved in vaccination operations (only 24% declare to be used to practicing it), but a little more are involved in family planning (47%). The contribution of the private sector in the hospital sector is more limited, except in the north of Benin and for some specific surgical interventions. In the obstetrical field, the private sector performs only 16¨% of child deliveries and 14% of c-sections. In fact, this activity is mostly carried out done by the AMCES hospitals (faith-based), most of them being located in the north of the country. So their regional weight is likely to be more important. Also, the private sector has a quasi monopoly for some surgical interventions, such as cataract, eye angiography or prostate endo-surgery. 186 2. An often illegal and low quality sector Private entities are usually illegal. According to the Scen Afrik 2006 survey, less than 9% of existing centres had received a formal opening authorization. It is possible that such authorization was never been requested, as it would not have been granted for two main reasons. First, 76% of the entities are opened by unqualified persons (and thus not authorized to undertake such a private activity). For instance, in the Ouémé/Plateau department, 69% of private centres have been opened by nurse’s aides, 11% by auxiliary nurses and 4% by first-aid workers and traditional birth attendants. It means that only 16% of centres were opened by qualified personnel. But - it is the second reason accounting for the non-authorization of the centres- even in the rare cases where a centre is created by a qualified personnel, there is little chance that the opening of the centre would lead to a request for authorization, as most of the personnel creating those entities are very often civil servants (who are forbidden to have a second job). To preserve their official work, those qualified personnel rapidly train some auxiliary nurses, who will later be responsible with health care provision in the private centre. Box 8: Types of private health entities in Benin The 2006 survey reported a great number of different health centres, not always consistent with existing regulation (NB: illegal entities are indicated below by one*). Overall, one can find: - Entities managed by physicians: o Either individually (doctor’s office) o Or in group: ƒ Either a private medical or specialised clinic, providing also inpatient services, ƒ Or a group-owned doctor’s office. - Entities managed by nurses - Nursing clinics (managed by a nurse, performing mostly treatments prescribed by a medical doctor) - Healthcare offices* (managed by an auxiliary nurse, creating confusion with nursing clinics as auxiliary nurses do not have the same skills as a nurse) - Entities managed by midwives: - Maternity clinics, for uncomplicated deliveries (in principle, only for performing prenatal surveillance, easy deliveries, postnatal surveillance and gynaeco-obstetrical care) - Entities managed by other health professionals : - Biomedical laboratory (managed by a biologist physician, a biologist pharmacist or a biologist veterinary); - Centres for radiology (radiologist physician); - Centres for physiotherapy (physiotherapist) - Health offices* (nursing aides ) - Maternity offices* (nurse aides and girl-maids) 187 Quantitatively, the most frequently encountered entities are those operated by physicians, nurses and midwives, or, more often, their illegal copies. Table 45: Distribution of private health entities (per type of entity) Medical Nursing and Maternity Hospitals Health and Traditional Labs office health care social health clinics clinic centres Atacora/Donga 4 27 3 2 6 10 0 Atlantique/Littoral Borgou/Alibori 14, 84 7 4 23 -- 0 Mono/Couffo 2 122 2 7 12 3 0 Ouémé/Plateau 8 176 97 2 4 0 0 Zou/Collines 30 200 7 11 2 2 2 58 609 116 26 47 15 2 Total Source : Scen Afrik 2006 report Besides the low qualification of many private health care providers, the sector is characterised by a low level of compliance with equipment and facilities standards. We saw previously that many private providers are not qualified. In addition, in-service training is scarce. In 2006 (Scen Afrik), about 43% of private centres did not benefit from any training programme. Furthermore, premises are generally inadequate. One-third of private centres do not have a washbasin for washing hands. 42% do not decontaminate their work tools. This situation is observed particularly in the illegal health centres. Box 9: Three examples of illegal health entities ¾ A retired nurse who carries out hernia surgery at home. He is established in a very dusty and tiny room that he uses as an operating ward, lit up by an incandescent bulb and ventilated by a fan placed at the centre of the ceiling. He operated about 70 cases in 2004. Physically diminished due to sickness, and as he can no longer operate by himself, he passed on to his son who learned the art by working with his father and who is pursuing today the practice of ‘’family transmitted hernia surgery’’ ¾ A healthcare office created by a nurse aide working in the surgery section of a regional hospital, and held during office hours by an unschooled female ‘’nursing aide apprentice’’ who, in the same room, combines health care services with the selling of cooked foods. . ¾ A “gynaeco-therapy” office created and managed by a nurse retired for 22 years and who claims to heal infertilities by a combination of modern drugs and traditional products. According to local practitioners, that person is suspected to be a frequent operator of voluntary abortion, and many young girls are daily consulted by him. As done by most of illegal practitioners, he is training one of his sons for taking over the office’s activities in the future. 188 3. A legal framework that is difficult to enforce and thus not enforced The approval system for practising (or opening) private activities is incomplete and mostly inconsistent. The act 97-020 of 17th June 1997 does not mention any minimum distance between providers, while it is expressly provided for (and applied) for pharmacists. The act does not also mention any standards for premises and equipment to be used by specialists. What is worse, the act indicates that laboratory tests can be proposed privately only by biologists (biologist physicians, biologist pharmacists or biologist veterinaries) or possibly only under their supervision. Given the fact that these categories of specialists are extremely scarce in Benin (below one dozen) and that laboratory tests account for a major source of income for private providers, one cannot really see how this provision can be applied. The case for drugs (which in principle can be sold only by pharmacists) is similar. In general, the 1997 act is not enforced. We previously saw that many private providers do not have an opening authorization. But this situation seems to be linked less to a so-called will of fraud than to a legal mechanism that is not really applicable, and thus little enforced. For example, the minimum timeframe for granting an opening authorisation is 5 to 8 months. According to the DNSP, a large part of the opening applications was processed (¾ were rejected), but neither the applicants nor the professional associations are notified the final decision. Finally, the sanctions for non-compliance with the act are generally theoretical, in spite of the 200 inspections conducted annually. In early 2006 (i.e. after 8 years of implementation of the act), not a single case of sanction was reported. This legal framework favours unfair competition, not only by NGOs but also and mainly by civil servants. The non-for-profit health entities (NGOs and faith-based health centres) should enjoy various benefits ensured by the act. They can be supplied with drugs from CAME and particularly be exempted from taxes. We can question the relevance of those benefits since the NGO status of some centres is just theoretical and they operate as purely commercial entities. Many private providers take advantage of the legal vacuum to turn their health centre into a for- profit NGO. In principle, civil servants can not have a private activity (article 21 of the decree 73/88), apart from exceptional situations, i.e. when the lack of specialists in a location is proven (article 9 of the law of 1997). Undoubtedly, the exception has become the rule. Not only the decree and the act are bypassed, but there is no compensation planned for private practitioners (who could possibly work in the public sector in case of lack of public physicians49). On balance, one may think that the law of 1997 had for objectives (i) to officialize the dual jobs of many civil servants (which likely reduced the pressure for wages increase) and (ii) to put in place a framework for regulating (in theory) the private practicing of medicine. In the law, it is difficult to identify any provision aiming at promoting the development of a high quality and fully integrated private sector. 49 This is the option applied, with success, by Mali. 189 4. The development of the private sector still remains poorly promoted In 2002, Benin prepared a private sector development policy, whose objectives were particularly relevant: (i) better integrate the private providers into the health system; (ii) strengthen partnerships representing the private sector and (iii) improve health care quality by private providers. Unfortunately, this policy was implemented only partly, in spite of a substantial support from the European Union. The private sector remains little integrated into the national system. Less than 43% of private centres provide regular statistics to the Ministry of Health (Scen Afrik 2006). Once again, incentives to do so are weak. These entities do not get any feedback on the information processed. It is also interesting to observe that entities that send regularly their data are generally private hospitals having signed contracts with the Government (see below) and which benefit from a subsidy. Private entities do not always have access to the same suppliers as the public system. If the private health providers have from now on the possibility to get supplied with vaccines from the EPI, they remain excluded from access to generic drugs distributed by CAME. The only major progress noted in this integration policy is that 6 private hospitals have been granted the role of district hospitals. This success is largely due to the experience in decentralized management at the Hospital of Menontin (Cotonou) and to the lobbying efforts from the Association representing faith-based hospitals in Benin (AMCES). These hospitals henceforth receive a financial support of the Government and subsidies for extending or renovating their premises. But, apart from the case of Mènontin, no private provider has been granted the full management of a public health entity. Also, contracting out of logistic functions (i.e. catering, laundry) is still rare. The only example is that of the Hospital Mère-Enfant of Cotonou. So the potential is huge here. Finally, it must be mentioned that private providers remain excluded from the health equity fund, while their contribution to provide care to the poorest is probably very significant. Private sector professional associations are rarely supported. In principle, it was planned that the private sector professional associations would be supported for their inspection activities and for issuing opening authorisations. This support is still obviously lacking. The OPP (Office for public-private partnership) is mostly neglected. An office for consultation between private sector and public sector was set up several years ago. It is headed by a committee whose chairman is the representative of the independent union of private sector physicians (SAMSEP). Although a budget line is normally allocated to the OPP, in 2007, the allocation was withdrawn by the MoH for other purposes. Presently, the OPP has no financial resources to even pay its rent. The building that it occupied had to be given back to its landlord. So far, overall, the Government has poorly supported the private sector in its development and the improvement of its quality of health care. In terms of pre-service training, for example, paramedical private staff has been systematically excluded. On the other hand, private physicians have benefited from this type of training, namely in gynaecology. Those medical 190 trainings are a little bit outside the framework of strengthening the private sector; since the private physicians accepting to have their training financed by the Government have to agree working in a public health center for 5 years. Concerning in-service training programmes, available budgets (about 60 million FCFA annually) are disbursed only very partially. The DNPS estimates that about 300 workers of the private sector have enjoyed this type of training (generally as seminars or workshops), which still remains very low as compared to the 1,400 existing health entities. Finally and as already seen, support to inspection operations (to check compliance with the quality standards) and subsidies for purchases of equipment remain very low. 191 Private sector: main findings The (commercial and faith-based) private sector plays an important role in Benin, particularly in the outpatient sector where it ensures over 60% of visits. The faith-based sector is relatively homogeneous and plays an essential role in some regions in the north, a role that has been acknowledged by the Government in granting 6 of the faith based hospitals the status of district hospital. On the other hand, the outpatient sector includes both excellent practitioners and real quacks. It is true that the existing legal framework for controlling the private sector is little applied (as in many other countries). This can be explained in part by the fact that many civil servants practice illegally a private activity (on a dual-job holding basis), a fact which does not push to adequate control of that activity. A large part of the problems of understaffing and underperformance of the public sector could however be solved with a better integration of the private sector. From this point of view, little progress was achieved, except for the public-private partnerships with the faith based hospitals (see above). 192 Chapitre 9 : HEALTH SECTOR FINANCING Funding the Benin health system: key aspects......................................................................................194 Resource mobilization.......................................................................................................................195 Risk pooling arrangements................................................................................................................195 Purchasing health care.......................................................................................................................196 1. Government funding is still largely oriented towards health care supply .........................................197 1.1. Funding of health care supply by the Government ....................................................................197 (i) Is the amount of Government funding sufficient?....................................................................198 (ii) Is Government funding beneficial to the poor? .......................................................................198 (iii) Are resources allocated according to health priorities?..........................................................200 Do the most populated regions receive a more significant portion of the funding?..................200 Is the portion of expenditures consumed by the central level (and more generally by the health administration function) reasonable? ........................................................................................201 Does the Government finance the “best-buys” (the most effective interventions) ?.................202 (iv) Is the management of financial flows well-performing? ........................................................204 1.2. Funding of health care demand by the Government: free c-sections .........................................205 1.3. Funding of health care demand by the Government: the Health Equity Fund ...........................205 Fonds Sanitaire des Indigents (FSI): A Health Equity Fund in Benin..........................................205 Is the Health Equity Fund Fund used for funding the best-buys? .................................................205 Is the Health Equity Fund really favourable to the poor? .............................................................205 2. Household funding of health expenditures without any significant risk pooling mechanisms .........209 2.1. Health care funding directly by households ...............................................................................209 2.2. Health care funding by community health insurance schemes (“mutuelles”)............................210 2.3. Health care funding by private insurance companies.................................................................211 3. A highly volatile donor funding ........................................................................................................212 Is donor funding sufficient or on the contrary too important? ......................................................212 Is external financing favourable to the poor? ................................................................................213 Is donor funding consistent with health priorities? .......................................................................214 Is management of donor funding efficient?...................................................................................214 193 Funding the Benin health system: key aspects Benin allocates only 4.6% of its GDP to the health sector expenditures. According to the National Health Accounts (prepared in 2006, with data of 2002 and 2003), total health expenditure reached 96 billion FCFA in 2003 which corresponded to a per capita expenditure of 13,742 FCFA, i.e. 4.6% GDP. As compared to its neighbouring countries, this rate is higher than Niger’s, but lower than Burkina Faso’s and Togo’s. Figure 89: Evolution of health expenditures (in % GDP) in the 8 WAEMU countries 35.00 30.00 25.00 Benin Burkina Faso Côte d'Ivoire 20.00 Mali Niger 15.00 Senegal Togo 10.00 Moyenne 5.00 - 2001 2002 2003 2004 2005 Source: WAEMU 194 Resource mobilization The main flows of this financing system can be described as follows: Figure 90: Main financial flows in the Benin health system Households Goverment 49,962 29,572 External funding Resource mobilization (52%) (31%) 15,840 (16%) Households Government 49,461 34,192 Ext. Fund Risk pooling 11,318 Drugs Health care services Other Public Health functions 38,169 29,382 health health Health acre Other health functions services administration (purchasing) 8,793 4,478 expenditures 4,976 9,938 Source: USAID 2006 (CNS 2006 data) Note: - All amounts are in million of FCFA; - By ‘’government’’ we mean here all ministries (not only the Health Ministry); Unsurprisingly, households are the primary funder (52%) of the health system. The Government funds about 31% of the system expenditures, the rest being covered by international donors (16%). Other stakeholders (local communities, state-owned and private enterprises, NGOs) play an insignificant role (less than 1% of the system’s resources). Risk pooling arrangements Given that households’ contributions mainly consists (99%) of direct payments, the level of progressivity (and thus of equity) of the health system is very low. Direct payments of households are mainly composed of drug purchases and fees for visits and hospital cases (see part 2 for details). Those payments amounts do not vary in relation with patients’ incomes. This observation made by the NHA is also confirmed by the situational analysis of community based insurance schemes, which cover less than 1% of Benin population. There exists today no real risk pooling mechanism for households. This absence of risk pooling mechanisms is only lowly compensated by the subsidizing mechanisms of demand financed by the Government. The implementation of the Health Equity Fund (since 2005) has not improved risk pooling, given the very low number of beneficiaries (8,276 people in 2006, i.e. a little more than 1% of the population). Another subsidizing mechanism is free c-section care, but it has not been put in place yet. 195 Purchasing health care As explained in annex, health system equity can also be addressed by the “purchasing” function. In other words, although there no risk pooling mechsnisms in Benin, it remains possible that public expenditures are beneficial to the poor. The Government and international donors seem to prioritize services that are the most likely to be used by the poorest. In light of the figure above, the Government spends 16% of its budget on ambulatory care and 22% on prevention. Only 12% of expenditures go to hospitals, whose inequitable nature is well-known. 30% of expenditures go to constructions/renovations and equipment purchases (but it is more difficult to conclude here that those expenditures really have a pro-poor nature). As for donors, they dedicate 67% of their expenditures to ambulatory care and 17% to prevention. Only a ‘’benefit incidence analysis’’ would enable to conclude that the Government (and possibly the international donors) is subsidizing adequately health care services used by the poor. However, the available data suggest that pro-poor expenditures are not neglected. Table 46 : Financial flows in million of FCFA) between funding sources and health functions (2003) FUNDING SOURCES Ministry Other Local Priv. Pub. of Health Ministries councils CNSS Insur OOP Comp Donors H F Hospital care 3,698 881 23 4 5 3,843 27 502 8,983 9.36% E U Outpatient care 4,944 325 79 251 4 2,511 20 7,627 15,761 16.43% A N Nursing care 429 429 0.45% L C Imaging and lab tests 1,144 380 6 6 8 2,607 43 1 4,195 4.37% T T Drugs 244 3 4 5 37,881 27 1 38,165 39.78% H I Prevention 6,836 1,956 8,792 9.17% O Health care management 3,248 1,229 4,477 4.67% N Other health expenditures 1,601 218 3 3 5 3,117 24 1 4,972 5.18% S Investments 9,520 400 16 9,936 10.36% Education and training 219 219 0.23% 31,639 2,448 130 268 27 49,959 141 11,317 95928.5 32.98% 2.55% 0.14% 0.28% 0.03% 52.08% 0.15% 11.80% 95928.5 Source : NHA 2006 (2003 data) Table 47 : Expenditures of funding sources per health function (2003) Government Households Donors (MoH and other Min) Hospital care 48% 8% 4% Outpatient care 29% 5% 67% Nursing care 1% 0% 0% Imaging and lab tests 19% 5% 0% Drugs 10% 76% 0% Prevention 22% 0% 17% Health care management 10% 0% 11% Other health expenditures 14% 6% 0% Investments 46% 0% 0% Education and training 1% 0% 0% TOTAL 100% 100% 100% Source : NHA 2006 (2003 data) In analysing here the way health expenditures are financed, Benin health system appears particularly inequitable, given the potential for catastrophic expenditures. By catastrophic 196 expenditures, we mean health expenses that, due to their high amount and/or their unpredictable nature, can push households into poverty. Hospital care (i.e. hospital stays) and drugs are usually considered as catastrophic expenses. They should be as far as possible paid/subisidized by the Government (or possibly by international donors although this solution is not to be maintained in the long run). In Benin, we notice (table below) that 43% of hospital care expenses are financed by households and particularly that 99% of drug expenditures are financed by these same households. It is thus obvious that the health system is not only inequitable but also poverty- generating. Table 48 : Financing of health functions per funding source (2003) Government Households Donors (MoH and other Min) TOTAL Hospital care 51% 43% 6% 99% Outpatient care 33% 16% 48% 98% Nursing care 100% 0% 0% 100% Imaging and lab tests 36% 62% 0% 98% Drugs 1% 99% 0% 100% Prevention 78% 0% 22% 100% Health care management 73% 0% 27% 100% Other health expenditures 37% 63% 0% 99% Investments 100% 0% 0% 100% Education and training 100% 0% 0% 100% Source : NHA 2006 (2003 data) 1. Government funding is still largely oriented towards health care supply Benin Government funds the health system in several ways. Mostly, the Government finances directly the supply of public health care services (more exactly, it finances its inputs). It finances also indirectly demand for health care50, through the Health Equity Fund and the forthcoming programme of free c-sections. We will examine these 3 mechanisms successively. 1.1. Funding of health care supply by the Government In the light of the above table, analyzing the funding of supply of health care by the Government consists in asking the following 4 questions. (i) is the funding level sufficient?, (ii) is the health funding system favourable to the poor? (i.e. does the Government subsidize as a priority the services usually used by the poor?), (iii) does the Government finance the “best-buys” (the most effective services) ?, and (iv) is the Government funding efficient (execution rate of expenditures, sources of wastes and leakages)? 50 In fact, with the Health Equity Fund and the free c-section program, the Government continues to finance health care supply, given that these 2 programs are reimbursing health facilities for services provided to the poorest. However, with these mechanisms, it no longer finances inputs but rather outputs performed for the benefit of some categories of the population (i.e. the poor, children below 5, and pregnant women). 197 (i) Is the amount of Government funding sufficient? Though the budget of the Health Ministry (MoH) has increased by about 10% each year since 1997, its portion in the general budget has seriously decreased, going down from 15- 16% to 8-9% over the same period. It has remained since then below 10% and fell to 8% in 2006. So we are far from the 15% recommended at the Abuja Conference. However, the aggregate amount of health expenditures has reported a regular increase. Table 49 : Evolution of the Health Budget (1999-2006) 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Budget for health (in millions FCFA) 25764 30590 30821 33548 40509 38328 39430.0 45670 46855 48917 % of general budget 15.15% 16.15% 11.13% 9.98% 9.78% 8.02% 9.81% 8.36% 8.41% 8.00% % of GDP 1.95% 2.11% 2.01% 2.00% 2.21% 1.96% 1.91% 2.14% 2.04% 1.99% Source : CSE/DPP data of the MoH Figure 91 : Evolution of the health budget (1999-2006) in million FCFA 60000 18.00% 16.00% 50000 14.00% 40000 12.00% 10.00% 30000 8.00% Budget sante 20000 % du budget de l’Etat 6.00% % du PIB 4.00% 10000 2.00% 0 0.00% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Source : CSE/DPP data of the MoH (ii) Is Government funding beneficial to the poor? The few available data suggest that the Government dedicates a little less than half of its resources to finance services favorable to the poor. It is usual to evaluate the pro-poor nature of Government expenditures by analysing the subsidy level of the services most used by the poor (Benefit Incidence Analysis). It is not possible today to carry out such analysis because the levels of utilisation of the services by quintile of wealth are not known and there is no study on the subsidy level of health care services (i.e. for instance, we do not know today the cost of one curative outpatient visit and its subsidy level by the Government). In the absence of those data, we can only analyze the distribution of Government expenditures per type of providers and per type of health intervention. In principle, we consider that the poor 198 should benefit first from outpatient visits (health centres) and vaccination programs. On the contrary, hospitals are generally little favourable to the poor, when taking into account their costs and their concentration in geographical areas little accessible to the poor. Data from the NHA (above) show that about 40% of Government expenditures are dedicated to outpatient visits and to prevention activities. Even if hospital expenditures remain relatively limited (20% of expenditures for providers and 11% for functions), we think that there exists significant room for improving the distribution of Government funding for the benefit of the poor. 199 Table 50 : Total expenditures (operating and capital) of the MoH per health care providers (2003) Amount % (in 000 FCFA) out of the MoH budget Teaching hospital (CHNU) 1,954,828 6.2 General hospitals 6,201,083 19.6 Specialized hospitals 1,155,708 3.7 Health care centers 17,256,865 54.6 Administration 4,375,762 13.8 Other health care institutions 134,000 0.4 Other providers 487,785 1.5 Source : NHA 2006 (2006 data) (iii) Are resources allocated according to health priorities? A Government must in principle finance the most essential health care services, that is to say interventions that are effective and well targeted. Achieved results on this matter can be assessed through several analyses: (i) Do the most populated regions receive a more significant portion of the funding? (ii) Is the portion consumed by the central level (and more generally by the health administration function) reasonable ? (iii) Does the Government finance the “best-buys” (the most effective interventions)? Do the most populated regions receive a more significant portion of the funding? Funding allocated across the regions remains very inequitable, even if recent progress has been made. We considered here as financial flows to regions the following expenditures: (i) delegated credits, (ii) the grant to CNHU51 (Atlantic and Littoral Region) and (iii) the expenses linked to the Government’s permanent staff (permanent civil servants, contract civil servants--- APE ACE) posted in regions. The figure below highlights that variation across regions goes up to 4.08, that is to say the best funded regions (the Atlantic/Littoral region, where are located the two main towns in the country) receives 4.08 times more budget than the least financed region (Borgou/Alibori). It is mainly the distribution of Government helath workers that accounts for this inequality. Conversely, an obvious effort has been made with delegated credits, namely for the Atacora/Donga region. It is also true that the MoH has a better control of those credits (than of credits for paying civil servants). 51 Ideally, it must also be added the grant to the COGECs, but its allocation per region is not known. 200 Figure 92 : Government health budget allocated to regions per number of inhabitants 2500 FCFA par habitant (credits en personnel) FCFA par habitant 2000 (credits delegues) 1500 1000 500 0 ATACORA- MONO- ZOU- OUEME- BORGOU- ATLANTIQUE- DONGA COUFFO COLLINES PLATEAU ALIBORI LITTORAL Source: Statistics directory MoH / 2006 Is the portion of expenditures consumed by the central level (and more generally by the health administration function) reasonable? The portion of expenditures consumed by the central level does not seem excessive. Not accounting for expenditures whose destination could not determined, we come up with a ratio of 13.8% of the total expenditures dedicated to the central level (in 2006), which does not appear excessive. However, the portion of expenditures still controlled by the central level remains very high. Indeed, it is important to differentiate two notions: the level at which the credits are truly spent and the one at which the amount of those same credits is decided. On this last point, decentralization still appears embryonic in Benin, since over 61% of the expenditures of the MoH (operating and capital i.e. investment) remain controlled by the central level. As for the control on health expenditures, the progress made is difficult to be assessed. Indeed, in terms of expenditures, the portion of the central level is decreasing; but this seems to be particularly due to a drop in the execution rate of the PIP (Pluri-annual Investment Programme). Table 51 : Evolution of Government expenditures per level of the health pyramid 2002 2003 2004 2005 2006 52 Central level + PIP on domestic resources 14,503 14,984 15,274 17,778 13,696 Regional level. (DDS+CHD) 4,220 4,911 3,322 2,560 2,450 District level 1,550 2,188 4,580 5,434 6,123 TOTAL 20,273 22,083 23,175 25,771 22,270 Percentage controlled by central level 71.54% 67.85% 65.90% 68.98% 61.50% Source : MoH/DRFM 52 Including purchase of vaccines. 201 Figure 93 : Evolution of Government expenditures per level of the health pyramid 30,000 25,000 20,000 Niveau Péripherique (Zones Sanitaires) 15,000 Niveau Départ. (DDS+CHD) Niveau Central + PIP sur 10,000 ressources interieures 5,000 - 2002 2003 2004 2005 2006 Source : MoH/DRFM Does the Government finance the “best-buys” (the most effective interventions) ? The portion of resources allocated to investments has tended to decrease since 1999. As shown by the graph below, the portion of investments (PIP) dropped from 58% of the budget in 1998 to 45% in 2006. Obviously, the sudden drop in 1999 led ministries to sacrifice investment expenditures for the sake of current expenditures. This observation is probably underestimated, because the PIP includes operating expenses, namely staff bonuses. Thus the actual proportion of budget allocated to investments is even lower than 45%. Figure 94 : Evolution of Government health expenditures per nature (operating and investment) 35000 70% Fonctionnement 30000 PIP 60% 58% 55% Ratio 54% 25000 51% 50% 50% 46% 45% 20000 41% 40% 38% 36% 15000 30% 10000 20% 5000 10% 0 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Source : CSE/DPP data of the MoH 202 Currently, operating expenditures of the MoH consist mostly of (i) staff costs, (ii) transfers to decentralized entities and (iii) health transfers abroad. The table below illustrates the structure (after re-processing53) of the MoH expenditures in 2005. Staff costs can be estimated to draw close to 43% of the total, given the fact that several budget lines for grant or for transfer correspond actually to staff cost. The second significant budget item (22%) is the one of delegated credits, such as credits transferred to decentralised entities [mainly the RHA/DDS (Regional Health Authorities), but also the CHD (Regional Hospital) and the ZS (Health districts)]. Finally, an important line item (14%) is allocated for health transfers outside Benin (i.e. for patients seeking specialized health care abroad). Table 52 : Structure of the Government health expenditures (after re-processing) RECURRENT EXPENDITURES (in millions FCFA) 2005 Expenditures at central level (MoH) 3,181 12% Human resources (salaries) 905 3% Other recurrent expenditures 2,276 9% Expenditures at local level (regions + teaching hospital) 13,877 53% Human resources 8,073 31% Expenditures related to contract-based staff (APE-ACE) 4,335 17% Subsidies to COGEC 2,188 8% Subsidies to CHNU (teaching hospital) 1,550 6% Other recurrent expenditures 5,804 22% Grants to regional and districts health authorities 5,804 22% Non allocated expenditures 8,794 34% Social aid * 3,536 14% Maintenance 1,260 5% Grant to PEV (immunization program) 950 4% Miscellaneous recurrent expenditures 617 2% Other expenditures (goods and services) 439 2% Support to health sector ** 430 2% Allowances for missions abroad 55 0% Travel expenses for students 10 0% Allowances for missions in-country 9 0% Grant to onchocercosis committee 3 0% Small investments 217 1% Grants to health centers *** 1,486 6% Miscellaneous investments (BESA) 217 1% TOTAL 26,069 100% Source : SIGFIP, AStat 2006 and WB re-processing NOTE : * The ‘social aid’’ line is in principle credits dedicated to poverty alleviation. This line includes three types of expenditures: (i) purchase of drugs to be distributed for free in health centres (for the poor --- in 2003, 461 MFCFA), (ii) health transfers abroad (212 MFCFA in 2003) and (iii) subsidy to haemodialysis patients at CHU (285 MFCFA in 2003). ** The “support to the health sector” line aims at enabling to face health disasters. Actually, those credits are most of the time used for vaccination campaigns. Their use is controlled by the Ministry of Finance. 53 This reprocessing was made according to a method similar to that of the 2003 PER. It mainly consisted in ‘’identifying’’ the various staff expenses, hidden under other headings. For example, several budget lines ((i) recurrent expenditures at central level (for over 50% of the allocation), (ii) grant to COGECs and (iii) grant to CNHU) actually consist mostly of staff costs, despite their names. 203 *** This line finances equipment purchases for the health centres recently built or renovated. The corresponding funds are assigned to the RHA/DDS, according to a decision taken by the DIEM (MoH Directorate for Equipment). (iv) Is the management of financial flows well-performing? A funding system, while targeting the “best-buys” interventions and the needy populations, can be totally inefficient if funds are not fully disbursed as planned. Two problems can explain such a situation: (i) weaknesses in procurement (weaknesses that can thus explain an insufficient budget execution rate) and (ii) funds leakages. The execution rate of the health budget has deteriorated for a few years, mainly because of investments. Indeed, the graph below shows that the execution rate of the health budget fell from 92% in 2003 to 78% in 2006. This regression is almost entirely due to the difficulties faced in executing PIP budget. Figure 95 : Evolution of total execution rate of the health budget and of amounts committed but not disbursed for the PIP and drugs 8,000 95% 7,000 92% 90% 6,000 89% 5,000 85% 85% Infrastructures (Constructions+équipements) 4,000 Médicaments 81% 80% 3,000 Taux d'execution global 78% 2,000 75% 1,000 0 70% 2002 2003 2004 2005 2006 Source : DRFM of the MoH As mentioned before, even though it is impossible to estimate the extent of leakages, they are likely to be significant. Leakages can take various forms: absenteeism of the personnel paid by the Government, neglecting equipment maintenance in favor of a policy of systematic purchase (i.e. it seems more profitable for some people to invite for bids for equipment purchase rather than to put in place a real maintenance policy), perishing and stealing of drugs,…. etc . 204 1.2. Funding of health care demand by the Government: free c-sections The principle of free c-sections was adopted in 2008 and should be implemented in the course of year 2009. Targeting this health service appears very timely. As seen in chapter 2 of the present report, the main obstacle faced by pregnant women to obtain a caesarean is financial. The details of this free c-section program remain to be specified. 1.3. Funding of health care demand by the Government: the Health Equity Fund Fonds Sanitaire des Indigents (FSI): A Health Equity Fund in Benin By « Health Equity Fund », we mean a fund (i) which is used to reimburse health centers for their patients that were unable to pay, (ii) which is managed at local level and (iii) whose identification of beneficiaries is independent from the Government. In Benin, a first mechanism was established in 2000 and was allocated about 1 Bn FCFA each year. It seems to have been led astray. The new mechanism (FSI) put in place in 2005 tries to limit abuses via 3 measures: (i) While the FSI covers most of primary and secondary care services, it excludes very clearly the most costly health services (i.e. treatment of chronic diseases, implants, haemodialysis, health transfers abroad…) ; (ii) For each service, a maximum reimbursable amount is set (i.e. an outpatient visit is reimbursed within the limit of 1 000 FCFA, a hospitalization in the limit of 1 750 FCFA per day and for a maximum of 10 days, a delivery in the limit of 15 000 FCFA); (iii) Identification of the poor (i.e. issuance of a poverty certificate) is done by a local identification committee created within each health centre. This committee is composed of (i) the head of the health centre, (ii) the President of COGEA/C, (iii) one representative of the locally elected people (mayor or head of arrondissement) and (iv) and one representative of social services (Centre of Social Promotion). A social investigation is carried out and its results are submitted to the identification committee, who decides to include (or not) the beneficiary within the FSI. On balance, the FSI of Benin partially fulfils the requirements of a Health Equity Fund. Though truly managed locally, identifying the beneficiaries remains influenced by the Government and in particular by its health civil servants. Is the Health Equity Fund Fund used for funding the best-buys? The scope of services covered by the HEF/FSI is relatively wide and corresponds a priori to health care services that are the frequently used by the poor. Is the Health Equity Fund really favourable to the poor? 205 Budget allocations between health districts are not based on poverty prevalence. A study was carried out to measure possible correlation between (i) the amount of allocation (2006) to the FSI of each health district and (ii) the number of households considered as poor in the district. The table below shows that the relation is significant for the two poorest quintiles, but it is unfortunately negative. In other words, the analysis indicates that the more there are poor households in a health district, the lower is its FSI allocation. Obviously, the criteria used for allocating the FSI funding to health districts should be seriously reviewed. Table 53 : Correlations between (i) poverty prevalence and (ii) FSI credit, FSIs expenditures and number of FSI beneficiaries (2006) Allocated credits Disbursed credits Number of FSI for FSI FSI beneficiaries Richest 0.2432 0.4436*54 0.4061* Richer -0.1088 0.1729 0.1387 Average -0.1504 -0.0580 -0.0934 Less poor -0.4263* -0.5020* -0.5281* Poorest -0.4424* -0.5166* -0.5288* *significant at the level of 5% Despite significant sensitization efforts, the number of beneficiaries remains low and, once more, ill-correlated with poverty prevalence. An assessment of the Fund management, achieved by the DRFM in February 2006, shows a very low use of the Fund in 2005 (6.7% of available credits) and a net progression in 2006 (42%), though very variable according to districts, going from 0 to 100% (see annex 9.2). A very rough estimation55 indicates that less than 1% of the poor population is covered. What is worse, the correlation is negative between (i) on the one hand the expenditures of the FSI or the number of FSI beneficiaries and (ii) on the one hand the number of people belonging to the last quintile of wealth (the poorest). Namely, the more there are poor households in a health district the fewer beneficiaries and also the fewer FSI credit or disbursement there are. We even find a correlation, positive this time, between the number of households belonging to the richest quintile. Of course, in the absence of studies on the socio-economic profile of FSI beneficiaries, we cannot conclude with certainty that the rich are the prime beneficiaries. These results seem to confirm however an observation made by Blundo in 2003: ‘’ a (….. ) variant of the operation of health equity fund concerns the negotiation of the poor status for having access to free medical care. Indeed, there exist, in each hospital, assistance funds to the poor patients (FSI). But access to the poor status is not automatic. It is the social workers of hospitals that are the only persons competent for analysing cases and, if they identify them as poor, for engaging the procedure of their free medical support by the hospital. Hence the negotiations are often long and include doubtful practices around the decision of including somebody in the category of “poor”. It is frequent to see health civil servants’ parents to be identified as poors”. We have seen that the FSI was reformed in 2005. The observations made in 2003 are perhaps less relevant. Yet, the results found in 2006 are very 54 Credit consumptions are here correlated significantly (at 5%) with the number of households included in the highest quintile of wealth. 55 Given that 31% of the population is considered as poor and that the average utilisation rate of health services is 37% (i.e. each person has 0.37 contact per year with the health services), it can be estimated that each year the poor population should have about 800 000 contacts with the health system. But the FSI funded only 7 000 of those contacts in 2006. 206 concerning and require, at least, to review radically the identification method of the poor used by the FSI. 207 Figure 96 : Correlation between HEF/FSI disbursements/beneficiaries and poverty prevalence 120000 100000 Number du dernier quintile Number du premier quintile 90000 effectifs of households Effectifs of households in the poorest quintile le plus quintile 100000 80000 de pauvrete in the poorest riche 70000 80000 60000 60000 50000 40000 R2 = 0.0178 40000 R2 = 0.069 30000 20000 20000 10000 0 0 0 10000000 20000000 30000000 40000000 50000000 60000000 0 10000000 20000000 30000000 40000000 50000000 60000000 consom m ation des credits FDI HEF disbursements (in FCFA) Consom m ation des credits FDI HEF disbursements (in FCFA) 110000 180000 Effectif du premier quintile le Effectif duof households inde 100000 160000 Number of households Number dernier quintile in the poorest quintile 90000 140000 the poorest quintile 80000 120000 plus riche pauvrete 70000 100000 60000 80000 50000 60000 40000 40000 R2 = 0.0182 30000 R2 = 0.0746 20000 20000 0 10000 0 500 1000 1500 2000 2500 0 50 100 150 200 250 300 350 400 Effectif des beneficiaires Number of beneficiaries effectifs beneficiaries Number of des beneficiaires 208 2. Household funding of health expenditures without any significant risk pooling mechanisms 2.1. Health care funding directly by households Households constitute the main financier of the health system, even if this situation seems to improve gradually. The latest reliable data are of year 2003 and indicate that Benin households finance about 52% of health expenditures. Despite its very high level, this rate remains slightly inferior to the average observed in UEMOA/WAEMU (about 54%). Moreover, according to the WHO estimates, this rate seems to be decreasing over the past years. Figure 97 : Portion of health expenditures financed by the households. 75.00 70.00 65.00 Benin 60.00 Burkina Faso Côte d'Ivoire 55.00 Guinea-Bissau Mali 50.00 Niger 45.00 Senegal Togo 40.00 35.00 30.00 2001 2002 2003 2004 2005 Source : WHO The large portion of health expenditures paid by households increases the risks that the latter might face catastrophic expenditures. The notion of “catastrophic health expenditures’’ is defined as being health expenditures higher than 40% of disposable income (outside food) of a household (see Xu 2003). This type of expenditures can propel a household into poverty. In 2006 (EMICOV data), about 2.2% Benin households were facing catastrophic health expenditures. Even if a longitudinal study would provide better evidence, such results suggest that over 2% of households became poor in 2007 because of health expenditures and because there exists no adequate mechanism to pool health risk. 209 Figure 98 : Proportion of health expenditures as compared to total expenditures of households 1 par rapport aux depenses totales hors alimentation Proportion des depenses de sante 0 .2 .4 .6 .8 0 1.00e+07 2.00e+07 3.00e+07 4.00e+07 Depenses totales hors alimentation par menage Sources : EMICOV 2006 and WB 2.2. Health care funding by community health insurance schemes (“mutuelles”) Community health insurance schemes (CHIS) cover today only 1% of Benin population and probably not the poorest. While only 9 CHIS were operational in 1997, between 80 and 1000 were listed in 2005 (Senelle 2005). In spite of this increase, CHIS cover only 70 000 people, i.e. less than 1% of the population. The table in annex 9.1 presents the main existing CHIS. Some CHIS target explicitly the poor (PROMUSAF and ASSEF), but other target mostly middle-class people (e.g. MUSANT). Total membership is however negligible. The low attractiveness of CHIS appears linked neither to the amount of premiums nor to their payment deadlines. As shown in the table in annex 9.1, the monthly average premium (i.e. contribution for a family) is generally inferior to 1000 FCFA and never exceeds 2000 FCFA. An available study on ‘willingness to pay’’ in urban areas (Ceda 2006) shows that ‘’82% of persons surveyed would be ready to contribute an amount of 1000F/family/month and 64% of persons up to 2000F/family/month’’. Yet, a great number of households (namely in rural areas) face a strong problem of income irregularity. This problem has however been taken into account by several CHIS, through an annualization of premiums. All CHIS are however limited by the low quality of care services and by the illegal practices of some health personnel. To ensure quality, most CHIS have entered into contracts with some 210 health centres, often private and not-for-profit. The success of those contracting experiences is mixed. The best recognised CHIS (and considered the most creditworthy), such as the MUSANT and the MSSC/P had obviously no difficulties to enter into a contract. The other CHIS find it more difficult to have their contracts signed and enforced (Senelle 2005). All members of these CHIS find it difficult to obtain drugs in the contracted health centres, particularly in the public sector. It seems that many health personnel are involved in illicit selling of drugs. Any drug distributed for free to a CHIS member corresponds to a heavy revenue loss for these personnel. So they prefer refusing free care to the CHIS member or mentioning a drug shortage as an excuse. As drugs purchase represents a large portion in the households’ health expenditures, it is clear that the attractiveness of CHIS suffer from that. On balance, one can reasonably think that the development of the CHIS is very constrained by the low level of quality of care and of governance in the public health sector. Attempts at combining CHIS with micro-credit are not convincing so far. Various experiences were made in this field. For instance, PROMUSAF proposes a mechanism of micro credit (available only for members having no arrears in their premium payments), which aims to help members to develop their income generating activities. This micro-credit should also facilitate the payment of premiums to the CHIS. This option was based on the objective of PROMUSAF of targeting the poor populations. Of course, one can wonder if people do not join this CHIS only for benefiting from this micro credit opportunity. The low rate of claims tends to confirm this hypothesis. On the contrary, AssEF was built on its clientele of micro-borrowers and then developed and proposed a health plan for them. Health coverage is thus a ‘’bonus’’ to borrowers. This CHIS can also rely on an existing base of members. In return - and it is what happened-, this CHIS is very dependent on the continuity of the micro credit mechanism. 2.3. Health care funding by private insurance companies Involvement of private insurers in health in Benin is very low. There exist five important insurers in Benin. Only 3 offer health insurance: Nouvellle Société Interafricaine d’Assurances (NSIA-BENIN) Africaine des Assurances(AA), Federale des Assurances (FEDAS). Data collected during the preparation of the NHA in 2006 show that the sums paid in 2003 by NSIA-B and AA (FEDAS having not completed the questionnaire) totalled about 707 million FCFA, i.e. below 1% of total health expenditures. The significant size of the informal sector (85% of the population) in Benin can likely explain for this situation. 211 3. A highly volatile donor funding Is donor funding sufficient or on the contrary too important? On average, donors finance 16.5% of health expenditures. In 2003 (date of the latest NHAs), the portion of health expenditures covered by donors (16.5%) was slightly superior to the WAEMU average (15%). As it will be seen further, this portion strongly decreased in 2004, before going up in 2005. Figure 99: Portion of health expenditures financed by international donors in 9 countries. 40.00 35.00 30.00 Benin Burkina Faso 25.00 Côte d'Ivoire Guinea-Bissau 20.00 Mali Niger 15.00 Senegal Togo 10.00 Moyenne 5.00 - 2001 2002 2003 2004 2005 Source: WHO Like many African countries, Benin suffers from a high volatility of external funding. The graph below takes into account the main donor funds, as reported by the MoH56 . We can note that from one year to another, external funding can vary between 16 and 12 billion of FCFA, i.e. a yearly variation of 20-30%. We can henceforth understand the difficulty of the Health Ministry to plan coordinated actions with donors. This can help account for the low execution rates of the PIP. 56 This assumes that some funds from donors are not included in the analysis, as the MoH is not always aware of all donor funds. The major partners (IDA, FED, USAID and Belgium) are however present in the analysis. 212 Figure 100 : Evolution of health expenditures financed by international donors in Benin RESSOURCES EXTERIEURES 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 2001 2002 2003 2004 2005 2006 Source: Health Statistic Directory / MoH Most of external funding is provided by a group of 9 international donors. As shown in the graph below, 9 donors represent over 93% of external funding. We can observe that the drop in external funding is mainly due to the end of an IDA project in 2002 (and to the withdrawal of several small bilateral donors). On the contrary, other donors have increased their support, namely Belgium, the Global Fund, and WHO. Figure 101 : Structure of external funding, by donor 9000 8000 7000 OMS SUISSE 6000 UNICEF 5000 BADEA USA 4000 FED FONDS MONDIAL 3000 BELGE 2000 IDA 1000 0 2001 2002 2003 2004 2005 2006 Source: Statistics directory MoH Is external financing favourable to the poor? Foreign funding likely contributes to reducing the inequity of health system funding. As previously seen, donor financing is mainly oriented towards ambulatory care (67% of donor financing) and prevention (17%), two health functions of which the poor are generally the main beneficiaries. 213 Is donor funding consistent with health priorities? External funding is evenly shared between vertical programmes and support to health districts. Since the disengagement of the European Union, there exists no support to the health system at national level. The table in annex 3 shows that in 2005 (latest available data), the international donors’ support was dedicated, for 51%, to vertical programmes (mainly HIV/AIDS, malaria and vaccination) and for 38%, to strengthening health districts (each donor choosing one or several districts). Only the European Union was financing a support programme to the health system at national level. This program (8th EDF) came to an end in 2006. Therefore, while many donors recognised that the main problem of the health system is its weak governance, it exists no national support program in this field. As in other countries, donor funding allocated to some diseases are not consistent with their epidemiological significance. So while HIV prevalence is very low in Benin, over 12% of external funding was dedicated to it in 2005. Also the 4% allocated to tuberculosis and leprosy appears relatively generous, given the low epidemiological importance of those diseases. Is management of donor funding efficient? A large part of external financing flows outside the circuit of the Health Ministry. According to the NHAs, over 71% of external resources (except for NGOs) bypass the Health Ministry. Serious efforts remain to be made by the Ministry and by donors to take real control of the donor assistance. Health Financing : Main Observations Funding by the Health Ministry represents today 9% of the general budget, which shows a strong drop as compared to the amounts of the 90s, but remains significant. The MoH expenditures seem to give priority to the most effective and most favourable interventions to the poor (preventive and ambulatory). Also significant efforts have been made (and should continue) to allocate an increasing portion of the budget to health centres. However, two problems are particularly worrying: (i) the low execution rate, which is a problem that affects the whole Benin government but more specifically the Health Ministry (namely in its investments) and (ii) the diversion of the Health Equity Fund for the benefit of the richer. Funding by households (52% of health expenditures) remains important, as in other low-income countries. Risk pooling mechanisms are almost inexistent. Despite a long experience of community health insurance schemes, they cover today only 1% of Benin population. Finally, external financing is fairly low (16% of health expenditures) and relatively volatile. . 214 Chapitre 10 : RECOMMENDATIONS 1. More resources for the health system................................................................................................216 1.1. What are the high impact interventions (to be implemented primarily)?...................................217 1.2. What are the main bottlenecks preventing the increase of effective coverage for these high impact interventions? ........................................................................................................................219 1.3. What are the costs of increasing coverage for the high impact interventions? ..........................220 2. A better organization of the health system........................................................................................221 2.1. A health system more consistent and more responsive to local needs .......................................221 2.2. A fully recognized and integrated private sector........................................................................222 2.3. Better distributed and performing health workers......................................................................223 2.4. A more efficient and more equitable financing system..............................................................224 2.5. More autonomous and more efficient health services ................................................................224 2.6. More available, less expensive and better prescribed drugs.......................................................225 2.7. A vaccination coverage extended to all regions and social categories.......................................226 2.8. Blood products accessible to everyone.......................................................................................226 215 As we have seen, it is necessary to increase the health system resources if we want to reach the MDGs in terms of health. The first part of this chapter presents the preliminary results of an estimate of these additional resources. It was done by UNICEF with the relevant method of reference: Marginal Budgeting for Bottlenecks (MBB). Nevertheless, it is also clear that the simple addition of resources will be either insufficient or very inefficient. It is necessary to thoroughly reorganize the whole system. Some reforms are already ongoing. Others still have to be discussed with key stakeholders. These changes are presented in the second part of the chapter. 1. More resources for the health system The Marginal Budgeting for Bottlenecks method (MBB) Jointly developed by UNICEF and the World Bank, the MBB is a quite rigorous process of prioritization and budgeting of the actions to be undertaken in order to improve maternal and child health. This logic consists of the following steps: 1. identify the different causes of infant and maternal mortality ; 2. select the health interventions (community-based, preventive and curative) likely to have a strong impact on these mortality causes57 ; 3. determine the optimal coverage level of these interventions, so as to maximize their impact while remaining realistic ; 4. for each of these interventions, identify and measure the bottlenecks that have to be addressed in order to reach the expected coverage rates ; 5. budget the various costs related to the elimination of the bottleneck; This logic is the one the report has made an effort to follow (cf. part 2.1 of chapter 1). The MBB has been used in more than twenty African countries. Till date, it is the most precise tool (i) to prioritize the interventions for addressing child and maternal health and (ii) to estimate the cost of that effort. It must be kept in mind that the MBB simulation presented below is still provisional. As a matter of fact, it uses the DHS data as a coverage target of the high impact interventions. These data have been updated early 2009. In its current version, this MBB simulation of Benin is based on the following objectives (to be reached in 3 phases aiming year 2015): reduction of neonatal mortality by 41%, of post-neonatal and juvenile mortality by 50% and of maternal mortality by 48%. The table below describes the estimated impact for each phase and for each component of the health system. 57 This selection is based on a (regularly updated) review of international scientific literature on the health impact of these various interventions. 216 Table 54 : MBB estimated impact of of interventions on child and maternal deaths (2009-2015) 2009-2015 Service delivery mode Impact in mortality reduction Under Neonatal five Maternal 1. Family oriented community based services 0.2% 13.5% 0.0% 1.1 Family preventive/WASH services 0.3% 2.1% 0.0% 1.2 Family neonatal care 0.0% 0.0% 0.0% 1.3 Infant and child feeding 0.2% 4.6% 0.0% 1.4 Community illness management 0.0% 8.9% 0.0% 2. Population oriented schedulable services 6.2% 8.6% 10.4% 2.1 Preventive care for adolescents & adults 0.0% 0.0% 0.3% 2.2 Preventive pregnancy care 6.1% 1.6% 1.5% 2.3 HIV/AIDS prevention and care 0.0% 1.3% 0.0% 2.4 Preventive infant & child care 0.0% 6.0% 0.0% 3. Individual oriented clinical services 36.7% 40.3% 39.5% 3.1 Maternal and neonatal care at primary clinical level -6.2% -1.6% -14.6% 3.2 Management of illnesses at primary clinical level 2.9% -5.8% 0.0% 3.3 Clinical first referral care 12.0% 15.6% 12.1% 3.4 Clinical second referral care 31.5% 34.4% 38.2% Total Family & comm. Outreach & Clinical care 40.9% 50.5% 48.3% Source : UNICEF MBB 2008 1.1. What are the high impact interventions (to be implemented primarily)? 1.1.1. Child health The interventions on child health are presented in three groups: (i) community, (ii) preventive clinical and (iii) curative clinical. a. Community package According to the MBB analysis (see details in annex 10.1 table), three community interventions should have a major impact on the MDG n.4: 1. Hand-washing ; 2. Breastfeeding for infants under 6 months ; 3. Oral rehydration. These three interventions should therefore be at the centre of the community-based component of health system strengthening. 217 b. Clinical preventive package On the clinical preventive package, three interventions can play an important role in child mortality reduction: 1. Prenatal consultations ; 2. Anti-smallpox immunization ; 3. Intermittent prophylaxis for child malaria. The detailed analysis is presented in the annex 10.2. c. Clinical curative package Finally, the most promising clinical curative interventions are: 1. Distribution of zinc (for diarrhea) ; 2. Care and support for seriously ill children based on IMCI protocols. The detailed analysis is presented in the annex 10.3. table. 1.1.2. Maternal health As often the case, the community-based package of interventions has rather a low impact on maternal health. Only the two other packages are presented here. a. Clinical preventive package Unsurprisingly, interventions like anti-tetanus immunization or prenatal consultations have a limited impact on maternal mortality58. On the other hand, two interventions seem to be important: 1. Detection and treatment of urinary infections ; 2. HIV prophylaxis for women and men. A detailed analysis is given in the annex 10.4. table. b. Clinical curative package It is with this package that the efforts will have more impact. The three most important interventions are indeed: 1. Basic Emergency Obstetrical Care (BEmOC) at primary care level ; 2. BEmOC at referral level ; 3. Complete Emergency Obstetrical Care (CEmOC) at referral level. The MBB thus confirms that the current Beninese policy of maternal mortality reduction is relevant as it includes these interventions. In conclusion to this part, it is interesting to note that the « MBB » estimates largely overlap with the analyses in the rest of this report, namely: 58 This does not mean that these interventions are useless for infant health. 218 - the reduction of child mortality presupposes that major efforts should be made on (i) malaria (prevention through mosquito-nets and anti-malarial treatment) and (ii) malnutrition (zinc insufficiency among others) ; - the reduction of maternal mortality implies a considerable strengthening of the health care system, especially at the referral level. 1.2. What are the main bottlenecks preventing the increase of effective coverage for these high impact interventions? Each of these above-mentioned interventions is confronted to different bottlenecks. As an illustration, we have simply retained the example of two of the most important interventions: (i) oral rehydration and (ii) BEmOC at primary care level. Example 1: oral rehydration The following figure (produced by the MBB tool) shows that, while the utilization rate of the ORS (Oral Rehydration Salt) is about 35% in Benin, supply-side factors do not seem to be a bottleneck. Supply of ORS and availability of community health workers (in charge of their distribution) are superior to 60%. This result remains to be better understood, but we can presume so far that women are probably ill-informed on diarrhea cases and their treatment (through ORS) and/or that the community health workers are not sufficiently active. A major effort in terms of IEC is thus necessary and has to be budgeted in consequence. Figure 102 : Bottlenecks for distribution of oral rehydration salt (ORS). A SRO A F F ORS ORS Source: MBB Bénin 2008 Example 2: BEmOC at primary care level 219 Equally, the MBB approach enables to identify bottlenecks on deliveries at primary level. We observe, below, that maternity hospitals generally benefit from adequate drug supply and staff availability. What is more, they are relatively accessible from a geographical point of view. In sum, the ratio of deliveries with assistance by qualified staff is overall high. However, as we have seen previously, a still too small number of deliveries are preceded by adequate prenatal monitoring. Newborns are not always weighed on a regular basis. Last but not least, obstetrical skills of staff members seem to be insufficient. Figure 103 : Bottlenecks for deliveries A A G D D D BE OC Source: MBB Bénin 2008 1.3. What are the costs of increasing coverage for the high impact interventions? According to the MBB model, removing bottlenecks of all the high impact interventions requires an additional investment of 687 billion FCFA (1.3 billion USD) over a period of 7 years, that is on average 98 billion FCFA per year (196 million USD). Table 55 : Estimated costs for achieving MDGs 4 and 5 (MBB estimates) ANNUAL Phase I Phase II Phase III TOTAL AVERAGE Types of interventions 2009 2010 2011 2012 2013 2014 2015 1. Community-based interventions 7,910 10,966 11,867 21,554 20,617 43,975 33,353 150,241.42 21,463.06 2. Preventive interventions 2,992 5,329 8,045 5,566 6,212 9,535 11,766 49,444.44 7,063.49 3. Clinical (curative) interventions 14,610 55,237 135,208 156,456 211,669 235,944 289,378 1,098,502.08 156,928.87 Management of health districts 1,005 1,324 1,644 1,569 1,764 3,691 3,877 14,873.63 2,124.80 Management of health regionss 3,804 4,983 6,161 5,907 6,693 13,535 14,308 55,391.11 7,913.02 Technical support 558 661 765 453 516 1,459 1,519 5,932.20 847.46 Total (000 US Dollars) 30,878 78,500 163,690 191,505 247,472 308,139 354,201 1,374,385 196,341 Total (000 FCFA) 15,439,221 39,249,799 81,845,002 95,752,374 123,735,852 154,069,604 177,100,588 687,192,439 98,170,348 Source: MBB Bénin 2008 220 This estimation is a considerable one, since it would imply at least tripling the annual Government health budget. If, as we said previously (Chapter 9), a significant increase of the health budget seems necessary (at least to return to the level of the years 2000-2001), it is probably not possible for the Beninese Government to reach the increase recommended by the MBB. It is thus necessary (i) to refine the MBB analyses by targeting the key priorities and (ii) facilitating a rise in external financing sources, in order to help the implementation of a strategic plan that takes into account the recommendations of the MBB. The preparation of a Compact within the framework of the International Partnership for Health (IHP) would enable to progress on that track. 2. A better organization of the health system 2.1. A health system more consistent and more responsive to local needs Regarding institutional governance, the central point of the analysis is the delay of Benin in deconcentrating its health system. Ideally, we should have an institutional governance system where the healthcare facilities are both (i) strongly autonomous in their management processes, and (ii) strongly controlled on their results (which must be defined in a realistic way through decentralized planning). It is thus better to speed up this process by: 1. Drastically increasing the management autonomy of districts and hospitals, which implies not only an increase of their resources (especially through delegated credits, but also additional decision rights and a rigorous monitoring of their performance (see further); 2. Deconcentrating staff budget at districts’ level (i.e. payment will still be made at the central level, but the district will keep its budget when a staff is transferred out of the district, so that the district can start replacing this staff); 3. Putting in place a truly decentralized planning, essentially through the DPP, which would be in charge of collecting and harmonizing the annual work plans of health districts with these of the directorates of the MoH and these of the vertical programs ; 4. Putting in place a mechanism of public review and of classification of healthcare facilities. 5. Reinforcing the Result-Based Management Program. This deconcentration equally presupposes that the national directorates (within the MoH) put their acivity plans in phase with these of the health districts. A reinforcement of the function of coordination – ensured today by the DPP – is thus indispensable. It could be achieved with the following changes : 6. Creating within the MoH a two-fold coordination structure. The first level will have to plan and validate the 3-year strategies (i.e. High Health Council, as requested by the General States) and the other will be more operational and will have to plan and validate the annual work plans of the Directorates and Programs, and the HD. This second entity will implement the decentralized planning. The two structures will be attached to the DPP ; 221 7. Reinforcing the attributions and resources of the Prospective and Planning Department (DPP), a department that is normally in charge of the synthesis of the plans developed within the MoH. Aside from the conclusions of the ongoing institutional audit at the MoH, some points deserve to be rapidly clarified. At the regional level, though the RHA/DDS should be able to control the MCZS, their responsibilities are wide and often not consistent. The RHA/DDS do not have much decision rights. At the local level, management committees for health care facilities are not representative and their attributions are very restricted. In contrast, communes, which are from now on the only legitimate representatives of communities, only have a feeble influence on the management of healthcare facilities. It is thus suggested that: 8. Responsibilities and resources of the DDS be reviewed (i.e. either by reducing their resources and allocating them to the districts , or by reinforcing their responsibilities) ; 9. The representativeness of the local management committees be enhanced, replaced by Boards, where the elected authorities will be present ; 2.2. A fully recognized and integrated private sector Though it plays an essential role in outpatient activities as well as in hospital activities, the private sector is not sufficiently recognized, as far as the business sector is concerned. An orderly development of this sector would then help resolving the issues of the public sector’s low staff and poor performance. A certain number of actions to be taken are suggested below. On the one hand, it is necessary to normalize the legal situation of private providers by: 10. Authorizing civil servants’ private activity only within the public structures (concept of « open clinic ») 11. Streamlining and clarifying the procedure for granting authorizations for such dual-job holding practices; 12. Supporting private sector’s professional associations, especially in their disciplinary and training functions ; 13. Allowing private health workers to work in public health facilities; 14. Reinforcing the institutional framework for the development and monitoring of public- private partnerships. On the other hand, the private sector should benefit from strengthened support, by: 15. Creating a specific budget line to subsidize private sector’s development ; 16. Offering trainings, financial guarantees for equipment purchase (eventually shared with the public sector) and full access to drugs coming from the CAME. To benefit from these advantages, private recipients will have to submit regularly information to the NHIS. 222 2.3. Better distributed and performing health workers Before a policy for human resources can be implemented, it is beforehand necessary to a clear and precise picture of the existing workforce. While data on the public sector HRH is rather adequate, the situation is very different for the private and traditional sectors. It is thus indispensable to: 17. Set up an HRH observatory (for data collection and analysis on HRH in the public sector, the private sector, the informal sector and traditional medicine) ; We have also seen that Benin suffers from a very inequitable geographic distribution of its health workers. This situation does not prevail only in Benin. However, the country is one of the rarest that does not have one or more mechanisms in place to address this problem. No solution appeared to be realistic to the eyes of the working group. This reflection needs to be pursued quickly. To reach the MDGs, and more particularly the one concerning maternal health, an important effort must be made to enhance staff’s competence (i.e. technical care quality). Several actions are necessary here: 18. Developing training programs for medical specialties that are either nonexistent or with insufficient numbers of practitioners; 19. Transferring legal control over all health workers’ training schools to the Ministry of Health 20. Promoting the creation of a school for health managers. 21. Creating a school for auxiliary nursing; 22. Updating the curricula of pre-service training (LMD system) ; 23. Developing rigorous selection criteria for admission in to training institutions (psycho- test etc.) Last but not least, it is indispensable to reduce absenteeism of health workers, racketeering of patients and embezzlement of public funds. Though a salary increase seems to be justified, it cannot by itself alter these practices, which have been lasting for too long in the health sector. It is then recommended to: 24. Put in place a policy of deconcentration/decentralization of HRH management; 25. Draw up a career plan for each cadre; 26. Establish a Result-Based Financing system (RBF). This last point is crucial. It seems to be the only one likely to restore a true culture of accountability, impartiality and mostly of respect to patients. At first, it would have to do with the healthcare facilities, for they are the sole ones able to objectively measure their performance59. 59 In a clinical setting, performing better is to have a growing number of patients. Health workers’ revenues have to be linked (at least partially) to this indicator. 223 2.4. A more efficient and more equitable financing system While waiting for the implementation (in the pipeline) of a universal health insurance scheme and considering the disappointment of the health mutuals experience, it seems advisable to keep on focusing on supply-side funding (by the Government). At first, it is necessary to financially support the process of administrative deconcentration: 27. Adjust the MoH allocation to the RHA/DDS (regional health authorities) according to their current responsibilities ; 28. Increase the MoH allocation to health districts ; 29. Provide allocations budgeted to health districts on the basis of objective criteria (results, activity, poverty level of the population, size of health districts and of the covered population) ; The Health Equity Fund has to dramatically review its identification mechanism of the beneficiaries, which implies that it: 30. Supports the communes in the execution of an annual survey at the community level to set up a database of the poorest households (with a view to establish a biometric card for every poor) ; 31. Reorients the Fund’s allocations in the poorest districts. 2.5. More autonomous and more efficient health services The aim of the health system deconcentration has to result in a greater administrative and especially financial autonomy of healthcare facilities (mostly hospitals). In financial terms, there are two possible (and non-exclusive) options. To simplify, we can talk about: - A Rwandan-style option, where the additional resources of the healthcare facilities derive essentially from public sources (preferably result-based, as in the RBF mechanism); - Or a Senegalese-style option, where these additional resources come from cost-recovery (of course, such a solution requires a considerable reinforcement of the exemption mechanism – HEF-style – to avoid worsening inequity). The two options are suggested here: 32. Increasing the volume of the budgeted allocations and reduce the discretionary nature of their amount. The implementation (ongoing) of the new result-based financing contracts would strengthen this process. With or without that type of mechanism, the budgeted allocations should be set for use, at least partly, to create an incentive for health facilities and staff members. 33. Reviewing the user fees grid (by reducing the drugs’ share and increasing fees for clinical services); allow the creation of a significant incentive out of cost recovery revenues. 224 In both cases, it will be necessary to: 34. Develop management accounting for healthcare facilities (at least hospitals) Some measures would also permit to enhance the functioning of the health pyramid in Benin, namely: 35. Collecting and regularly analyzing data on referrals; 36. Completing the mix of services, by creating a tertiary-level hospital, privately managed and with a regional vocation; 37. Preparing and implementing regional health plans; 38. Implementing a National Program for Quality Insurance; 39. Putting in place a National Program for promotion and control of traditional medicine. 2.6. More available, less expensive and better prescribed drugs A major concern in the pharmaceutical sector is about drugs’ prices. The suggestions previously made (i.e. to increase healthcare facilities’ financing) must give way to the implementation of the following actions: 40. Make sure that health facilities are applying reasonable drug margins, while receiving additional financial support to compensate for losses 41. Fight with determination against the illicit market. Though it constitutes a less urgent challenge, the availability of drugs could be yet enhanced, by: 42. Computerizing drug management at all levels. 43. Relocating the central stores of the CAME 44. Developing local production of essential generic drugs. The quality of the prescribed drugs (both for modern and traditional drugs) remains insufficient, as seen with the increase of kidney diseases in the country. One important measure would be to: 45. Develop and promote pharmacovigilance. Lastly, given the already significant use of traditional therapies, an effort of development of this type of medicine would assume: 46. Promoting and developing production of enhanced traditional drugs. 47. Encouraging traditional healers in research and development of traditional drugs 48. Reinforcing the ISBA/FSS research laboratory in the pharmacognosia study 225 2.7. A vaccination coverage extended to all regions and social categories Despite the excellent results reached by Benin in terms of immunization coverage, the actions aiming at the insufficiently covered districts remain to be reinforced, namely by: 49. Organizing immunization days in the hard-to-reach districts 50. Advocate towards religious authorities and local communities about immunization 51. Paying lump-sum amounts to the PEV nurses to cover motorbike-renting expenses for the advanced-strategy immunization 52. Renew motorbikes 53. Implementing a multimedia communication plan for the PEV 54. Ensuring the implementation of the PEV cold chains’ equipment renewal plan. A second problem is that of the financing’s durability. Benin already has a very precise financial planning of its needs at its disposal. It implies an increase of Government funding. 2.8. Blood products accessible to everyone The blood transfusion sector is one of the Benin health system’s least developed components. The whole chain (donations, collection, treatment and distribution) has to make way for a drastic reinforcement. The current weakness of this sector explains a significant number of maternal and child deaths. 55. Implementing the strategic plan for blood transfusion sub-sector; 56. Drawing up a program of blood-donation promotion with institutionalized encouragement and in relation with the blood donors’ association 57. Motivating blood donors through the organization of special days and certificate remittance. 58. Linking all components of the blood collection and distribution system. 59. Providing the regions mobile teams with adequate resources to collect blood; 60. Endowing all the 77 blood transfusion posts and the donor units in motorbikes. 61. Streamlining procedures for purchase of reagents and consumables, so as to make them permanently available for blood and its derivates’ qualification. 62. Endowing districts’ laboratories with Elisa chains. 63. Reinforcing the technical diagnostic, exploration and research capacity of laboratories at all the levels of the health pyramid 64. Encouraging self-transfusion In conclusion, beyond reinforcing the different components of the health system, two fundamental principles have to drive the transformations of this system: 1. a principle of institutional governance : through the deconcentration of the health system, basic healthcare facilities have to see their needs better taken into account 226 (hence the requirement of a decentralized planning) and particularly become more autonomous both financially and administratively ; 2. a principle of individual governance : health workers have to be strongly encouraged to enhance their performance (competence, productivity and respect to the patients). Given the failure of actions to strengthen inspections and other « top- down » type controls, this incentive cannot but come from the customers, either directly (i.e. bonuses based on cost recovery), either preferably indirectly with a mechanism of result- based payment financed by the Government and eventually by the donors. 227 Bibliography - ABDOULAYE I et alii (2006a) « Enquête sur le marché illicite des médicaments a Cotonou en Mars 2003 », Médecine Tropicale 2006 ;66 :573-576 - ABDOULAYE I et alii (2006b) « Evaluation des campagnes de sensibilisation sur les pratiques d’achat de médicaments dans la ville de Cotonou », Médecine Tropicale 2006; 66 :615-618 - ADEYA, G., A. Bigirimana, K. Cavanaugh, and L. Miller Franco (2007) “Rapid Assessment of the Health System in Benin”, April 2006, USAID - AFAP (2005) « Programme rural. Rapport de l’étude d’impact (enquête CAP) du programme d’appui à la réduction de la pauvreté dans les communautés d’Ahita, Agada-Hounmè et Djohoun-Colle, commune de Sakété » réalisé du 15 au 21 novembre 2005. - AGUEH V. et alii (2007) « Prévalence de l’obésité et les facteurs associés chez les personnes de 25 à 60 ans dans la ville de Porto-Novo » Cotonou Septembre 2007 - BOLLIG M et alii (2005) « Population projections for the Oueme catchment » - BENAUDIT (2006) « Etude sur mécanismes tarification des actes soins et médicaments dans les structures publiques et privées au Benin », Mai 2006 - BEYER Brigitte (1998) “The politics of the health district reform in the republic of Benin”, International Journal of Health planning and management, 13, 230-243 (1998) - BHUTTA Zulfi (2008) “Maternal and Child Undernutrition 3. What works? Interventions for maternal and child undernutrition and survival”, The Lancet, Vol 371 February 2, 2008 - BLUNDO G et J-P OLIVIER DE SARDAN (2003) « La corruption au quotidien en Afrique de l'Ouest : Approche socio-anthropologique comparative: Bénin, Niger et Sénégal », Working Paper, Institut für Ethnologie und Afrikastudien, Johannes Gutenberg-Universität, Forum 6, D-55099 Mainz, Germany, 2003 - CAME (2008) « Rapport d’activité sur l’exercice 2007 » - CAMPBELL Oona et alli (2006) “Strategies for reducing maternal mortality: getting on with what works”, Maternal Survival 2, The Lancet, Vol 368 October 7, 2006 - CAPUTO Angelika et alii (2003) « Undernutrition in Benin – an analysis based on graphical models », Social Science & Medicine, 56 (2003) 1677-1691 - CEDA (2006) « Opportunité de création de mutuelle de santé – Littoral et Atlantique » - CENTRE INTERNATIONAL DE DEVELOPPEMENT ET DE RECHERCHE (CIDR) (2001) « Etude de cas des mutuelles d’assurance maladie du Sud Borgou et du Nord Zou au Bénin », janvier 2001. - CHEN L, Evans T, Anand S, Boufford JI, Brown H, Chowdhury M et al (2004) “Human resources for health: overcoming the crisis.” Lancet, 2004, 364:1984–1990 - CLEMENS Michael (2007) “Do visas kill? Health effects of African health professional emigration” Center for Global Development, Working Paper Number 114, March 2007 - CNAN (1995) « Plan d’Action National pour l’Alimentation et la Nutrition » DANA –Secrétariat Permanent du CNAN - CNAN (1999) « Rapport du Séminaire sur l’harmonisation des mécanismes, méthodes et outils de surveillance nutritionnelle au Bénin » (Bohicon les 25, 26 et 27 janvier 1999) DANA –Secrétariat Permanent du CNAN - CNAN (2004) : Plan d’Action National de Lutte contre les Carences en Micronutriments Direction de l’Alimentation et de la Nutrition Appliquée ; Secrétariat Permanent du Comité National pour l’Alimentation et la Nutrition - CNAN/FAO (1998) « Programme d’Appui pour la Mise en Œuvre de la Déclaration du Plan d’Action de la Conférence Internationale sur la Nutrition – Rapport d’Exécution » - DANA –Secrétariat Permanent du CNAN - CNLS (Comité National de Lutte contre le Sida) (2008) “Rapport national de situation à l’intention de l’UNGASS”, 2008 - CREDES (2000) « Poursuite des actions de la coopération française dans le secteur des mutuelles de santé au Bénin – Étude de faisabilité », République du Bénin, décembre 2000. - CREDESA (2006) « Etude sur les collectivités locales et la décentralisation des services de santé », Avril 2006, Projet d’appui au secteur santé 8eme FED, Benin - DANA/PILSA (1999) : Rapport de l’Enquête Alimentaire et Nutritionnelle dans les zones du PILSA. Volume 1 – Synthèse Direction de l’Alimentation et de la Nutrition Appliquée ; BP 295 ; Tél. (229) 20 21 26 70 Fax (229) 21 39 63 Porto-Novo, Avril 1999 République du Bénin 228 - DANA/UNICEF (2001) « Rapport de l’Enquête Nationale sur la Prévalence des Troubles Dus à la Carence en Iode au Bénin » DANA –Secrétariat Permanent du CNAN Porto-Novo, juin 2001 - DE BROUWERE Vincent and Wim Van Lerberghe, ed (2001) “Réduire les Risques de la Maternité : Stratégies et Evidence Scientifique » Studies in Health Services Organisation & Policy, 18, 2001 - DECAILLET François et John May (2000) « Le secteur médical privé à Cotonou, Bénin, en 1999 », rapport HNP 22084, Banque Mondiale, Mai 2000 - GABF (2003) « Programme d’Appui à la Sécurité Alimentaire et Nutritionnelle pour la Santé et le Développement » Pobè, Novembre 2003 - GBANGBADE Sourou et alii (2003a) “Safe Motherhood Studies—Results from Benin. Competency of Skilled Birth Attendants. The Enabling Environment for Skilled Attendance at Delivery. In-Hospital Delays in Obstetric Care (Documenting the Third Delay)”, Novembre 2003 - GBANGBADE Sourou et alii (2003b) « Evaluation des Besoins en Soins Obstétricaux d’Urgence au Bénin », Décembre 2003 - GBANGBADE Sourou, Hounsa, Assomption, and Franco, Lynne Miller (2006) ”Systemwide Effects of the Global Fund in Benin: Final Report” Bethesda, MD: Health Systems 20/20. AbtAssociates Inc. - GRANDVOINNET Helene (2006) “Civil service reform in Benin: stocktaking and opportunities”, Public Sector Reform and Capacity Building Unit (AFTPR) Africa Region, World Bank, Report No. 36034-BJ, June 21, 2006 - HEILBRUNN John (1999) “The political economy of service delivery in Benin”, World Bank mimeo, October 1999. - HELVETAS (2002) « Etude socio-comportementale en rapport avec l'hygiène de l'eau de boisson, des mains et du milieu » Juin 2002, 41 pages. - HESSOU D. J. (1995) « Appui à l’intensification et à la promotion des cultures maraîchères et fruitières : Etude des aspects nutritionnels. Rapport de consultation pour la FAO » Cotonou, novembre 1995 - HESSOU D. J. et al (2007) « Rapport de consultation pour la formulation du Programme National de Sécurité Alimentaire du Bénin » (PNSA Bénin) Cotonou 2007 - HESSOU D. J. et VAN APPELDOORN (2003) : Evaluation du programme d’amélioration de l’Accès à l’Alimentation des Ménages et de la Réduction de la Prévalence des Maladies Infectieuses dans la Commune de Sakété, Rapport de consultation pour ICCO/AFAP 2003 - HESSOU D. J. et VAN APPELDOORN (2003) « Rapport de mission d’évaluation du Programme d’amélioration alimentaire et nutritionnelle des populations des communes de Pobè et d’Adja-ouèrè » ICCO/GABF 2003 - HILL Kenneth et alii (2007) “Estimates of maternal mortality worldwide between 1990 and 2005: an assessment of available data”, The Lancet, Vol 370 October 13, 2007 - HOTZ Christine and BROWN Kenneth (2004) “Assessment of the Risk of Zinc Deficiency in Populations and Options for Its Control”, Food and Nutrition Bulletin, vol. 25, no. 1 (supplement 2) 2004 - HOUESSOU K., R (2004) « L'extension de la sécurité sociale aux populations non couvertes – Rapport du Bénin » Association internationale de la sécurité sociale, Colloque des directeurs d’institutions de sécurité sociale des pays francophones d’Afrique, Limbé, Cameroun, janvier 2004, p.8. - INSAE (2008) « Enquête Modulaire Intégrée sur les Conditions de Vie des ménages (EMICOV) – Rapport de synthèse », 2008 - KPATCHAVI Adolphe et alii (2005) « Etude sur la demande en santé de la population de la zone sanitaire de Tanguieta – Materi – Cobly », Unidea, Aout 2005. - LABIE M et alii (2006) « Analyser l’articulation entre micro finance et micro-assurance sante : réflexions a partir de trois cas béninois », UMH, working paper 2006/3, 2006 - LEMIERE Christophe, Christopher H. Herbst, Negda Jahanshahi, Ellen Smith, and Agnes Soucat (2008) “Reducing Geographical Imbalances of the Distribution of Health Workers in Sub-Saharan Africa - A Labor Market Angle on What Works, What Does Not, and Why”, World Bank working paper 2008 - LOUVAIN DEVELOPPEMENT; PROMUSAF ; en collaboration avec STEP, AMCES et BØRNEFONDEN (2002) « Développement des mutuelles de santé : Rôles et places des acteurs dans les relations avec les prestataires de soins ? » – Rapport Général (Atelier des 29,30 & 31 janvier 2002), Cotonou, République du Bénin, janvier 2002. - MATHAUER Inke and Ingo Imhoff (2006) “Health worker motivation in Africa: the role of non-financial incentives and human resource management tools”, Human Resources for Health 2006, 4:24 - MBENGUE C et alii (2000) « Health zones, co-management and decentralization in Benin », Technical report n52, PHR, Septembre 2000 - Ministry of Health (MOH) (2006) / Ministère de la Santé Benin, “Annuaire statistique de la sante 2005”, 2006 - MOH (2005) Benin, “Annuaire statistique de la sante 2004”, 2005 - MOH (2007) Benin, “Annuaire statistique de la sante 2006”, 2007 229 - MOH (2007) Benin, “Plan National de Développement des Ressources Humaines”, 2007 - MOH (2004) « Rapport de l’atelier Profiles de la situation nutritionnelle et le plaidoyer pour la nutrition au Bénin » Analyse Profiles au Bénin, Cotonou, 19 au 30 janvier 2004. - MOH (2005) « Etude sur la rationalisation de l’utilisation du Fonds des Indigents au Benin », DPP, 2004-2005 - MOH (2005) « Mécanisme d’utilisation du fonds sanitaire des indigents », DRFM, MS, Juillet 2005 - MOH (2005) Arrêté interministériel 2005 n743/MFE/MSP/CAB/SP sur la mise en place du mécanisme d’utilisation du fonds sanitaire des indigents, 2005 - MOH (2007) « Plan directeur du développement du sous secteur de la transfusion sanguine 2008-2012 », Novembre 2007 - MULDER-SIBANDA Menno (2004) “Halving Benin’s malnutrition by 2015: What are the chances ?” May 2004 - NIANGALY A, Kante N, Soumare N'diaye A (2001) « Qualité des soins dans les centres de santé de la région de Kolikoro (Mali) » Médecine d'Afrique Noire 2001-48 (7) ; 300-4 - NKUNZIMANA C. (1996) « Disponibilité des médicaments essentiels sous noms génériques dans les formations sanitaires publiques et privées à but non lucratif du département de l'Atlantique (Bénin) » Cotonou 1996 ; Mémoire n°108/IRSP ; 66 p - PISAF/USAID (2006) « Rapport de l’EQGSS (volet hygiène Assainissement) dans les Zou-Collines en 2006 » - PNHAB, Programme National de l’Hygiène et de l’Assainissement de Base (PNHAB) - PNUD et MAEP (2001) « Étude sur les conditions de vie des ménages ruraux (ECVR 2) » Edition 1999 – 2000, République du Bénin, Novembre 2001 - PNUD (2000) « Etude sur les Conditions de Vie des Ménages en Milieu Rural (ECVR) » 2000, 84 pages. - RIDDE V, Nitiema AP, Dadjoari M (2005) “Improve the accessibility of essential drugs for the populations of one medical region in Burkina Faso”, Santé. 2005 Jul-Sep; 15(3): 175-82 - ROKX Claudia (2000) “Who Should Implement Nutrition Interventions? The Application of Institutional Economics to Nutrition and the Significance of Various Constraints to the Implementation of Nutrition Interventions”, HNP Discussion paper, December 2000, World Bank - SABI GANI, M. (2004) « Analyse des mutualistes sur les relations non-contractualisées avec les services de santé » Alliance Santé, CIDR, Parakou, République du Bénin, 2004. - SAIZONOU Jacques et alii (2006) « Audit des décès maternels dans quatre maternités de référence du Bénin - qualité des soins d'urgence, causes et facteurs contributifs.” African Journal of Reproductive Health, Vol. 10 No.3 December, 2006 - SCEN AFRIK (2006a), « Etude descriptive de l’offre privée des soins au Bénin », rapport Février 2006 - SCEN AFRIK (2006b), « Evaluation des activités exécutées dans le cadre du développement du partenariat entre les secteurs public et privé », rapport Février 2006 - SCEN AFRIK (2004), “Evaluation de la politique pharmaceutique nationale 200-2004”, Juillet 2004 - SENELLE Mélanie (2005) « Les relations contractuelles entre les mutuelles de santé et les formations sanitaires : Eléments théoriques et exemples pratiques en République du Bénin », UNIVERSITE DE LIEGE, 2005, - SODJINOU R. et al (2006-2007) « Etat nutritionnel des personnes âgées de 20 à 60 ans de la ville de Cotonou » Cotonou 2006-2007 - SOMDA Christian (2006) « Evaluation du système logistique de gestion des médicaments essentiels génériques dans la commune de Kpomassé (Benin) », mémoire IRSP n413 - WORLD BANK (1994) « Rapport d’évaluation République du Bénin. Projet d’Intervention Locale pour la sécurité Alimentaire » 9 mars 1994. Division des Opérations Agricoles. Département de l’Afrique Occidentale et Centrale. Rapport N° 12253 BEN. - WORLD BANK (2007) « Rapport de l’Atelier de Consensus sur la Mise en œuvre de la Politique économique des Réformes dans le Secteur de la Nutrition au Bénin », Novembre 2007 - WORLD BANK (2008) « Rapport national sur l’état de l’assainissement, Cotonou, 2008. - WORLD BANK (2004), Pour une plus grande efficacité de la dépense publique: Examen de trois secteurs clés, Rapport No. 29656-BEN, 20 décembre 2004 - WORLD HEALTH ORGANIZATION (2003) “Reducing maternal deaths: The Challenge of the New Millennium in the African Region”, 2003 - WORLD HEALTH ORGANIZATION (2006) World Health report 2006 - XU Ke et alii (2003), “Household catastrophic health expenditure: a multi country analysis”, The Lancet Vol 362 • July 12, 2003 - ZEBRA (2005) « Etude sur les modalités de la déconcentration et la forme de décentralisation répresentée par les COGES dans les centres de santé de commune et d’arrondissement » 230 ANNEXES 231 Annex 4.1 : Decision space within health districts MS DDSP MCZS COGECS/COGEA Communes Ð Ð Ð Ð Ð Ð DDS et ZS MCZS CSC/CSA CSC/CSA CSC/CSA CSC/CSA Pouvoirs (pour contrôler l’échelon inferieur) Gestion du personnel Recrutement FORT TRES FAIBLE FAIBLE MOYEN FAIBLE NUL Le MS décide du recrutement du MCZS et du La nomination du MCZS est décide par le MS et non Le DDS peut seulement recruter des agents Pour les personnel d'Etat, le MCZS n'a aucun Le pouvoir des COGECS/COGEA est nul personnel d'Etat. Il n'a en revanche pas de le DDS contractuels non qualifiés (aide-soignants, contrôle. En revanche, les recrutements de concernant les personnels de l'Etat. Ils peuvent pouvoirs pour le personnel contractuel chauffeurs, etc…) personnel communautaire doivent être par contre recruter du personnel communautaire, communautaire approuves par lui avec l'accord du MCZS. Promotion FORT TRES FAIBLE ??? ??? NUL NUL Le MS décide Le DDS propose et le MS décide Congés ??? ??? ??? ??? ??? ??? et absences Mutations FORT TRES FAIBLE (pour les mutations en dehors du TRES FAIBLE (pour les mutations en dehors du NUL NUL NUL Le MS décide des mutations d'un département département) département) a l’autre Le DDS propose et le MS décide Le DDS propose et le MS décide FORT (pour les mutations au sein du département) Sanctions FORT TRES FAIBLE TRES FAIBLE ??? ??? NUL disciplinaires Le MS décide des sanction (pour le personnel Le DDS propose et le MS décide Le DDS propose et le MS décide de l'Etat) Licenciement FORT TRES FAIBLE TRES FAIBLE ??? FAIBLE NUL Le MS peut décider seul du licenciement d'un Le DDS propose et le MS décide Le DDS propose et le MS décide Un COGECS/COGEA peut toutefois licencier un personnel d'Etat agent communautaire Formation MOYEN FORT FORT FAIBLE NUL NUL Le DDS décide, sur proposition du MCZS Le DDS décide, sur proposition du MCZS Primes ??? FAIBLE ???? ??? NUL NUL Il n'existe pas de prime contrôlée par le DDS, sauf s'il en crée une (exemple de Oueme/Plateau) Approvisionnement ??? NUL ??? FORT NUL en consommables et Le DDS n'a aucune attribution spécifique en la matière Le COGECS/COGEA contrôle l'inventaire et fixe médicaments les prix Bâtiments Programmation FAIBLE NUL NUL NUL FORT (investissement) (sauf pour les hôpitaux) Le DDS n'a aucune attribution spécifique en la matière La commune est juridiquement responsable de la construction des bâtiments (sauf pour l'hôpital de zone) Acquisition (y FAIBLE NUL NUL NUL FORT compris choix du (sauf pour les hôpitaux) Le DDS n'a aucune attribution spécifique en la matière prestataire) Equipements Programmation FAIBLE NUL NUL FORT NUL (investissement) (sauf pour les hôpitaux) Le DDS n'a aucune attribution spécifique en la matière Le COGECS/COGEA décide du programme d'investissement Acquisition (y FAIBLE NUL FORT NUL NUL compris choix du (sauf pour les hôpitaux) Le DDS n'a aucune attribution spécifique en la matière Le MCZS attribue les marchés prestataire) Equipements et FAIBLE NUL FORT NUL NUL bâtiments Le DDS n'a aucune attribution spécifique en la matière (entretien) Gestion financière FAIBLE FAIBLE ???? MOYEN NUL Le DDS na a priori aucun contrôle sur les dépenses des ZS. Le COGECS/COGEA doit approuver le budget du CSC/CSA et contrôler son exécution Gestion stratégique FORT FAIBLE MOYEN FAIBLE NUL Le MS peut imposer des activités non prévues Le PTA de la ZS doit normalement être valide par le DDS, mais il n'a pas les moyens de contrôler cela. Le MCZS élabore le PTA de la zone dans le PTA de la zone Ressources (pour mettre en œuvre ces pouvoirs de contrôle) Ressources FORT FAIBLE MOYEN MOYEN MOYEN financières Depuis 2000, une large part des crédits délégués est directement allouée aux ZS. Les MCZS reçoivent des crédits délégués, 10% Les COGECS/COGEA ne contrôlent que les Les communes disposent d'une taxe Les DDS reçoivent néanmoins une part des recettes communautaires. des recettes communautaires et quelques recettes communautaires locale et de subventions de l'Etat subventions Source : Arrêté n6723 du 30 Juillet 2004 232 Annex 4.2 : Degree and nature of COGECS involvement in various health facilities’ activities Frequency ACTIVITIES (in %) Planning Contributing to the health center’s planning activities 68,9 Conduct awareness activities on the importance of health services ofr the population 71,4 Budget Examine, vote and approve budget proposals 95,6 Follow-up on the implementation of the approved budget 10,7 Analyze causes of possible deficits and search for solutions 28,6 Financial and resources management Examine and approve revenues and expenditures for the previous 2 quarters 31,8 Advise on all expenditures decisions 0,2 Support collection of revenues 13,7 Decide on fees exemptions for the pooreste, on the basis of invetsigations by social 3,2 workers or community leaders Make sure official fees are enforced and fight against illicit drug sales 76,9 Control cash situation at least once a month 12,9 Notify to the health district officer and regional health offcier all cases of 21,7 embezzlement and other misconduct Make sure that involved health workers are sanctioned 0,3 Send to court members of COGECs that are suspected of being involved in case of 0,9 embezzlement and misconduct Help on annual inventory of assets of the health facility 63,9 Mobilize additional revenues for the health facility 76,5 Drug management Getting involved in ordering and receiving drugs and other medical supplies 95,8 Help on quarterly inventory of drug of the health facility 86,7 Getting involved in receiving donated drugs 91,2 Source : Survey Zébra 2006 233 Annex 5.1 : Number of healthcare facilities across districts (2006) Département/Zone sanitaire Arrond. Centre de Arrond Taux de Nombre de Santé(Disp couvert en couverture Centres de +Mat) CS sanitaire en % Santé Privés Alibori 41 38 38 93 Banikoara 10 10 10 100 M alanville/Karimama 10 9 9 90 - Kandi/Gogonou/Ségbana 21 19 19 90 - Atacora 47 44 44 94 24 Natitingou/Boucoumbé/Toucountouna 19 17 17 89 10 Kouandé/Péhunco/Kérou 13 13 12 92 9 Tanguiéta/M atéri/Cobly 15 14 15 100 5 Atlantique 74 71 71 96 Abomey-Calavi/So-Ava 16 15 15 94 - Allada/Toffo/Zè 33 28 28 85 - Ouidah/Kpomassè/Tori/Bossito 25 28 28 112 - Borgou 43 44 42 98 Bembèrèkè /Sinendé 9 9 9 100 - Nikki/Kalalé/Pèrèrè 19 18 18 95 - Parakou/N’dali 8 10 8 100 - Tchaourou 7 7 7 100 - Collines 60 49 49 82 108 Savalou/Bantè 23 17 17 74 31 Dassa-Zoumè/Glazoué 20 17 17 85 39 Savè/Ouèssè 17 15 15 88 38 Couffo 50 46 42 84 - Aplahoué/Djakotomè/Dogbo 24 22 19 79 - Klouékanmè/Lalo/Toviklin 26 24 23 88 - Donga 26 26 25 96 32 Bassila 4 3 3 75 5 Djougou/Copargo/Ouaké 22 23 22 100 27 Littoral 13 14 12 92 - Cotonou 1/Cotonou 4 2 4 4 200 - Cotonou 2/ Cotonou 3 4 2 2 50 - Cotonou 5 4 3 3 75 - Cotonou 6 3 5 3 100 - Mono 35 32 34 97 Lokossa/Athiémè 10 10 10 100 Comè/Grand-Popo/Bopa/Houéyogbé 25 22 24 96 Ouémé 52 57 50 96 - Adjohoun/Dangbo/Bonou 20 19 19 95 - Akpro/missérété/Avrankou/Adjarra 18 17 17 94 - Porto-Novo/Sèmè-Kpodji/Aguégués 14 21 14 100 - Plateau 29 28 28 97 - Pobè/Kétou/Adja-Ouèrè 17 17 17 100 - Sakété/Ifangni 12 11 11 92 - Zou 76 59 54 71 141 Abomey/Agbangnizoun/Djidja 29 24 21 72 59 Covè/Ouinhi/Zagnanado 18 14 12 67 12 Bohicon/Za-Kpota/Zogbodomè 29 21 21 72 70 Bénin 546 508 489 90 Source : Statistics directory 2006 234 Annex 5.2 : Functional health care facilities per districts (2007) Number of functional Number of facilities Number with management… facilities providing basic services Total Health districts Communit Faith- Commerci number Entirely Partially Full Partial Curative Public y-based based al functional functional package package care only OUIDAH KPOMASSE 24 0 0 0 24 24 0 24 0 0 TORI BOSSITO DJIDJA ABOMEY 24 0 0 0 24 18 0 18 0 0 AGBANGNIZOUN LOKOSSA ATHIEME 10 0 0 0 10 0 0 0 0 0 BANIKOARA 13 0 2 0 15 15 0 13 2 0 TANGUIETA COBLY 10 5 1 0 16 15 0 15 0 0 MATERI BEMBEREKE 10 0 0 0 10 10 0 10 0 0 SINENDE ADJOHOUN BONOU 17 1 1 1 20 18 2 17 3 0 DANGBO COTONOU 1 4 1 2 45 52 6 0 6 0 0 COTONOU 4 KOUANDE OUASSA- 14 0 3 0 17 18 0 18 0 0 PEHOUNCO KEROU NATITINGOU BOUKOUMBE 17 0 4 0 21 0 0 3 0 0 TOUCOUNTOUNA PARAKOU N’DALI 11 1 2 0 14 14 0 14 0 0 POBE KETOU ADJA- 28 0 0 0 28 28 0 0 0 0 OUERE KLOUEKANME 30 0 3 2 35 29 0 24 1 4 TOVIKLIN LALO SAVALOU BANTE 17 1 2 0 20 0 0 0 0 0 AVRANKOU ADJARRA 20 0 0 55 75 0 0 0 0 0 AKPRO-MISSERETE ABOMEY-CALAVI SO- 12 0 0 0 12 12 0 12 0 0 AVA COTONOU 6 5 1 2 0 8 8 0 1 3 4 DJOUGOU COPARGO 9 0 7 0 16 0 0 0 0 0 OUAKE COTONOU 2 3 2 118 123 5 118 1 2 118 COTONOU 3 SAVE OUESSE 17 1 18 18 18 MALANVILLE 1 1 KARIMAMA TCHAOUROU 7 1 3 1 12 10 0 0 10 6 BOHICON ZAKPOTA 19 5 4 28 27 1 18 10 ZOGBODOMEY APLAHOUE DJAKOTOMEY 20 2 22 22 22 DOGBO 235 Number of functional Number of facilities Number with management… facilities providing basic services Total Health districts Communit Faith- Commerci number Entirely Partially Full Partial Curative Public y-based based al functional functional package package care only PORTO-NOVO AGUEGUES SEME- 29 2 6 4 41 41 41 PODJI NIKKI KALALE 33 0 0 0 33 23 0 23 0 0 PERERE ALLADA TOFFO ZE 25 0 0 0 25 43 28 15 COVE OUINHI 16 0 1 32 49 47 2 16 2 31 ZANGNANADO KANDI GOGOUNOU 20 0 1 0 21 20 1 SEGBANA DASSA ZOUME- 24 0 3 0 27 27 0 27 0 0 GLAZOUE COTONOU 5 3 0 3 2 8 7 1 3 2 3 COME BOPA HOUEYOGBEY 34 0 0 0 34 34 0 28 6 0 GRAND-POPO BASSILA 5 1 0 0 6 7 0 6 0 0 SAKETE IFANGNI 12 0 1 0 13 13 0 13 0 0 Total 523 14 55 264 856 537 125 397 56 166 % 61% 2% 6% 31% 100% 63% 15% 46% 7% 19% Source : DPP/MS 236 Annex 5.3: Allocation of delegated credits (2006) 237 Source : Statistics directory 2006 238 Annex 5.4 : Revenues and expenditures from cost recovery (2006) Source : Statistics directory 2006 239 Annex 5.5 : DEA method for estimating technical efficiency of hospitals 1. Why using the a Data Envelopment Analysis method ? To measure technical efficiency of the Beninese hospitals, we used a Data Envelopment Analysis method. This method is much more relevant and robust than a productivity ratio analysis, for at least 4 reasons: 1. Hospitals are multi-product entities. In other words, they produce not only inpatient cases, but also outpatient visits, surgical procedures, imaging and lab tests. A productivity ratio analysis would therefore imply to aggregate these outputs onto a single one, which assumes that weights for each type of output have to chosen. With the DEA approach, these (highly simplistic) assumptions on weights are no longer needed as the DEA algorithm will generate its own weights, on the basis of the selected sample. Basically, with DEA, the selected weights will be based on the most efficient hospitals. 2. In hospital efficiency analysis, only the DEA approach allows entangling the “pure technical efficiency” from the scale efficiency. 3. There are DEA benchmarks for hospitals in many countries, including in Africa. 4. A DEA does not require a large sample. The reason is that DEA is based on linear programming and not statistical analysis. 2. Which hospital inputs and outputs have been retained in our DEA analysis ? Before describing the inputs and outputs used in the DEA, it is noteworthy mentioning a rule, defined by Cooper (Cooper 2006). This rule states that the number of observations (i.e. hospitals, in this case) must be at least higher than: Max {# inputs * # outputs, # inputs + # outputs)*3}. In our case, DEA has been carried out with 32 observations (out of 39 hospitals). After several tests, we concluded that – in order to comply with the Cooper’s rule60, while having a relevant analysis, selected inputs and outputs would be the following: OUTPUTS INPUTS Parameters selected for DEA 1. Inpatient cases 1.Surgeons 2. Visits 2. Other doctors 3. Obstetrical and surgical 3. Paramedics procedures 4. Other staff 5. Beds As for OUTPUTS, we retained surgical procedures, so that hospitals with high surgical production would be disadvantaged. 60 When this rule is not applied (i.e. too many inputs and outputs and few observations), DEA algorithms have difficulties to discriminate entities. As a consequence, a large number of entities are considered as fully efficient. Ideally, in a DEA, only 2 entities should be considered as fully efficient (as only two points are needed to draw the efficiency frontier). 240 As for INPUTS, the number of beds is viewed as reflecting the degree of capital use (cf. Wagstaff 1992). 3. Which data and parameters have been used ? We have used the DEAP software (http://www.uq.edu.au/economics/cepa/deap.htm). The DEAP parameters were the following : 32 NUMBER OF FIRMS 1 NUMBER OF TIME PERIODS 3 NUMBER OF OUTPUTS 5 NUMBER OF INPUTS 1 0=INPUT AND 1=OUTPUT ORIENTATED 1 0=CRS AND 1=VRS 0 0=DEA(MULTI-STAGE), 1=COST-DEA, 2=MALMQUIST-DEA, 3=DEA(1-STAGE), 4=DEA(2- STAGE) The selected algorithm is the “output-orientated” one (and not the « input-orientated » one), because we assumed that hospitals have more control on their outputs (especially on their case- mix) than on their inputs. Indeed, it is difficult for hospital managers to reduce their staff number (i.e. the main input), given that most of them are civil servants. 241 The data table used in DEAP was the following (2006 data) : Hospitals OUTPUTS INPUTS Visits Inpatients cases Surgical and Surgeons Other Paramedics Other staff Number obstetrical doctors of beds procedures HZ Kouandé 5896 1859 106 0 3 13 28 52 HZ Natitingou 12894 2675 261 2 5 23 61 94 HZ Tanguiéta (Hôp. Saint Jean de DIEU de Tanguiéta) 5764 1707 263 3 8 42 133 40 HZ Bassila 8749 2498 160 1 2 17 8 73 HZ Djougou (Hôp Ord de Malte) 19480 4730 581 1 8 25 48 76 HOMEL 21794 13722 280 24 8 151 244 299 HZ Abomey-Calavi 8691 3038 2118 4 6 57 52 86 HZ Ouidah 4133 2296 450 4 4 27 64 85 HZ Cotonou 2 et 3 (Suru Léré) 2192 113 0 3 5 38 85 12 HZ Cotonou 5 (Centre de Santé de Ménontin) 64746 9266 2528 4 17 51 72 119 BETHESDA 16150 1134 402 2 14 26 74 56 Hôp Saint Luc 150215 8836 0 4 17 69 131 120 Hôpital la Croix de Zinvié 11684 5312 2158 5 13 27 32 125 Hôpital Saint Joseph de Sotchanhouè 12988 1541 26 1 3 10 36 40 HZ Banikoara 8627 3151 315 1 2 24 35 53 HZ Malanville 11285 382 258 1 1 25 37 13 HZ Kandi 10581 3977 0 1 3 35 74 82 HZ Bembérékè 37438 6198 1979 2 2 69 62 115 HZ Saint-Jean de Dieu Parakou (Boko) 12304 1707 600 0 4 31 32 80 HZ Nikki - Sounon Sero 15565 9652 1112 1 4 15 37 75 HZ Tchaourou (Papané) 6451 3727 612 0 3 21 46 97 CHD-Mono 6334 4209 1118 6 19 54 77 121 HZ Aplahoué 923 2063 372 1 2 19 65 80 HZ Klouékanmé 7281 2026 307 1 5 16 70 64 HZ Comé 4407 2914 898 3 4 28 74 126 CHD-Ouémé 36790 16569 2622 10 32 173 317 366 HZ Adjohoun 3315 98 22 1 1 16 26 21 HZ Sakété 7920 3091 348 2 1 17 36 90 242 Hospitals OUTPUTS INPUTS Visits Inpatients cases Surgical and Surgeons Other Paramedics Other staff Number obstetrical doctors of beds procedures HZ Pobè 10168 1387 410 1 1 10 13 23 Hopital Auberge de l'Amour Redempteur Dangbo 3099 305 0 0 1 8 19 38 Clinique Louis Pasteur 1342 998 46 4 7 9 28 21 Hôpital El FATEH 14668 1323 485 8 9 25 83 72 243 4. What were the results ? They are presented below : Hospitals Efficiency Efficiency Scale effect (CRS) VRS HZ Kouandé 1 1 1 - HZ Bassila 1 1 1 - HZ Abomey-Calavi 1 1 1 - HZ Cotonou 5 (Ménontin) 1 1 1 - Hôp Saint Luc 1 1 1 - Hôpital la Croix de Zinvié 1 1 1 - HZ Malanville 1 1 1 - HZ Bembérékè 1 1 1 - HZ Saint-Jean de Dieu Parakou (Boko) 1 1 1 - HZ Nikki - Sounon Sero 1 1 1 - HZ Tchaourou (Papané) 1 1 1 - HZ Sakété 1 1 1 - HZ Pobè 1 1 1 - Hopital AAR Dangbo 1 1 1 - HZ Djougou (Hôp Ord de Malte) 0.71 0.725 0.98 irs Hôpital Saint Joseph de Sotchanhouè 0.666 1 0.666 irs HZ Comé 0.636 0.658 1.033 drs HOMEL 0.625 1 1.375 drs HZ Banikoara 0.621 0.736 0.844 irs HZ Kandi 0.574 0.578 0.993 irs HZ Aplahoué 0.453 0.485 0.932 irs CHD-Mono 0.439 0.449 1.023 drs Hôpital El FATEH 0.422 0.432 0.978 irs BETHESDA 0.421 0.446 0.944 irs CHD-Ouémé 0.415 1 1.585 drs HZ Tanguiéta (Hôp. Saint Jean de DIEU) 0.394 0.403 0.977 irs HZ Natitingou 0.375 0.376 1.002 drs Clinique Louis Pasteur 0.369 1 0.369 irs HZ Klouékanmé 0.357 0.381 0.937 irs HZ Ouidah 0.328 0.334 1.019 drs HZ Adjohoun 0.28 0.312 0.897 irs HZ Cotonou 2 et 3 (Suru Léré) 0.146 1 0.146 irs NB : irs = « increasing return to scale » drs = « decreasing return to scale » crs = constant return to scale” vrs = “variable return to scale” 244 Overall, one can see that, with 14 hospitals (out of 32) considered as fully efficient, quality of this DEA is not really high. This disappointing outcome is less due to the size of the sample (quite large and anyway very difficult to increase) and more to the high number of inputs and outputs. In other words, this DEA most probably overestimates the efficiency of Beninese hospitals. As a cautionary move, we retained the CRS index (at 62%). 62% of technical efficiency means that 38% of inputs (staff and beds) are wasted and could be removed without any impact on production. 245 Detailed results are below : Results from DEAP Version 2.1 Instruction file = bh2ins.txt Data file = bh3dta.txt Output orientated DEA Scale assumption: VRS Slacks calculated using multi-stage method EFFICIENCY SUMMARY: firm crste vrste scale 1 1.000 1.000 1.000 - 2 0.375 0.376 0.998 drs 3 0.360 0.394 0.914 irs 4 1.000 1.000 1.000 - 5 0.683 0.686 0.995 drs 6 0.625 1.000 0.625 drs 7 0.291 0.312 0.934 drs 8 0.234 0.240 0.973 drs 9 0.146 1.000 0.146 irs 10 0.957 0.990 0.966 drs 11 0.292 0.328 0.890 irs 12 1.000 1.000 1.000 - 13 0.610 0.679 0.898 drs 14 0.666 1.000 0.666 irs 15 0.621 0.736 0.844 irs 16 1.000 1.000 1.000 - 17 0.574 0.578 0.993 irs 18 1.000 1.000 1.000 - 19 1.000 1.000 1.000 - 20 1.000 1.000 1.000 - 21 1.000 1.000 1.000 - 22 0.270 0.396 0.683 drs 23 0.401 0.469 0.856 irs 24 0.300 0.342 0.878 irs 25 0.286 0.302 0.948 drs 26 0.362 1.000 0.362 drs 27 0.280 0.312 0.897 irs 28 1.000 1.000 1.000 - 246 29 1.000 1.000 1.000 - 30 1.000 1.000 1.000 - 31 0.369 1.000 0.369 irs 32 0.286 0.313 0.914 irs mean 0.625 0.733 0.867 Note: crste = technical efficiency from CRS DEA vrste = technical efficiency from VRS DEA scale = scale efficiency = crste/vrste Note also that all subsequent tables refer to VRS results SUMMARY OF OUTPUT SLACKS: firm output: 1 2 1 0.000 0.000 2 0.000 0.000 3 0.000 0.000 4 0.000 0.000 5 0.000 0.000 6 0.000 0.000 7 0.000 0.000 8 0.000 0.000 9 0.000 0.000 10 0.000 0.000 11 0.000 9.022 12 0.000 0.000 13 0.000 0.000 14 0.000 0.000 15 2849.614 0.000 16 0.000 0.000 17 0.000 0.000 18 0.000 0.000 19 0.000 0.000 20 0.000 0.000 21 0.000 0.000 22 2585.106 0.000 23 9790.716 0.000 24 0.000 0.000 25 967.758 0.000 26 0.000 0.000 27 0.000 671.199 28 0.000 0.000 247 29 0.000 0.000 30 0.000 0.000 31 0.000 0.000 32 0.000 0.000 mean 506.037 21.257 SUMMARY OF INPUT SLACKS: firm input: 1 2 3 4 5 1 0.000 0.000 0.000 0.000 0.000 2 0.534 0.000 0.000 16.304 25.878 3 1.966 5.571 20.663 94.939 0.000 4 0.000 0.000 0.000 0.000 0.000 5 0.000 2.735 2.173 2.880 0.000 6 0.000 0.000 0.000 0.000 0.000 7 2.520 0.184 33.452 0.000 0.031 8 2.179 0.000 3.985 18.558 0.014 9 0.000 0.000 0.000 0.000 0.000 10 1.883 8.169 15.896 0.000 27.111 11 0.186 7.873 0.000 19.942 0.000 12 0.000 0.000 0.000 0.000 0.000 13 3.925 9.182 10.146 0.000 49.383 14 0.000 0.000 0.000 0.000 0.000 15 0.000 0.000 4.517 7.361 0.000 16 0.000 0.000 0.000 0.000 0.000 17 0.000 0.000 7.132 32.367 0.000 18 0.000 0.000 0.000 0.000 0.000 19 0.000 0.000 0.000 0.000 0.000 20 0.000 0.000 0.000 0.000 0.000 21 0.000 0.000 0.000 0.000 0.000 22 3.714 11.000 16.429 0.000 4.429 23 0.000 0.000 0.000 31.805 9.971 24 0.000 0.969 0.000 30.631 1.285 25 2.000 0.000 13.000 37.000 51.000 26 0.000 0.000 0.000 0.000 0.000 27 0.000 0.000 0.000 3.400 2.000 28 0.000 0.000 0.000 0.000 0.000 29 0.000 0.000 0.000 0.000 0.000 30 0.000 0.000 0.000 0.000 0.000 31 0.000 0.000 0.000 0.000 0.000 32 6.266 2.461 0.000 23.630 8.528 mean 0.787 1.505 3.981 9.963 5.613 248 Annex 9.1 How to assess the performance of a health financing system ? A health financing system can be analyzed in 2 steps : 1. Identifying in the country all mechanisms for financing the health sector A health financing mechanism is a combination of two components ; (i) a funding source (resource mobilization) and (ii) arrangements for risk pooling and purchasing of health services. One should start by identifying all sources of funding. All countries use several sources of funding, but in Africa they are usually few. Three are essential: (i) the Government (out of its taxes and tariffs), (ii) households (out of their revenues after tax) and (iii) donors. One can add companies as a 4th source of funding, but their contribution is usually small. Each of these sources is channeled in the health system through various mechanisms. However one or two is usually prevalent. For instance, a Government will often fund directly health care supply, through its budgetary allocations. Similarly, households will usually fund directly these services, with out-of-pocket expenses. Other types of risk pooling and purchasing mechanisms can be used. For instance, a growing number of Governments strives to fund health systems in subsidizing health care demand, with Health Equity Fund (as in Benin) or in subsidizing mutuals (i.e. Rwanda). Conversely, the same risk pooling and purchasing mechanism may be used by several funding sources. As an example, mutuals (i.e. a risk pooling and purchasing mechanism) can be funded through premiums paid by household and through subsidies from the Government or from donors. In Benin, 7 risk pooling and purchasing mechanisms have been identified (main ones are in bold) : Funding Risk-poling and purchasing mechanisms sources Government Funding of health care supply Funding of health care demand : Health equity Fund (HEF) Funding of health care demand : free c-section program Households Out-of-pocket expenses Mutuals Health insurance companies Donors Funding of health care supply 2. Analyzing the performance of each identified mechanism Once mechanisms have been identified, their performance has to be assessed. That implies asking several questions, most of them being related to equity. 1. Is the funding amount (i.e. mobilized through this mechanism) sufficient (according to international benchmarks) ? 249 2. Is the arrangement for obtaining these revenues equitable for households (i.e. progressive) ? 3. Does this mechanism help to fund the most cost-effective interventions ? 4. Are the funded interventions pro-poor ? (that include looking at allocative efficiency and geographical allocation) 5. Is the mechanism efficient ? In other words, is it fully disbursed and are leakages limited ? It is worth noting that equity has several dimensions. It depends on the progressivity of resources mobilization and of risk pooling, but also on the nature of services that are purchased. 250 Annex 9.2 Main health mutuals in Benin (2006) PROMUSAF Alliance santé MUSANT ADMAB AssEF MSS Year of creation 1998 1995 1998 1998 1995 2000 17 mutuals 25 mutuals 12 mutuals 2 mutuals (MSSC in Cotonou and MSSP in Parakou) Technical ONG "Solidarité CIDR Lion's club MSA and AssEF Benin support mondiale" (WSN) Association (microfinance government France- group) and and STEP and Benin STEP CTB Beneficiaries 12,000 persones 17,800 persons 22,000 12,000 2,272 persons 1,200 persons (end 2005), mostly Borgou and persons, persons (end 2004) for MSSC women within Collines mostly urban mostly active and about 100 village groups women in for MSSP Cotonou Membership 500 - 1,000 FCFA 500 - 1,000 1,000 FCFA 5,000 FCFA premium FCFA Monthly 200 - 250 FCFA per 115 FCFA per 850 FCFA in 1,650 400 FCFA par 2,000 FCFA premiums member member average (100 FCFA per member + 100 - 200 for (but - 5,000 - member dependant annualized) 15,000) (family) Covered risks Primary care Primary care All health Primary Primary care All services services + care services care services deliveries + c- services + secondary sections + (covered care emergency through + some hospital care microcredit specialty care and surgery and a 10- 15% fee discount) + Other services (covered by the mutual) Prudential - 6 month delay for - 3 month delay - 6 month delay measures new members for new for new - copayment of 25- members members 30% - copayment of - copayment of - reserve fund 30% 10-30% - reserve fund - automatic radiation after 3 months of no payment Rate of premium 58% 83% collection Rate of 33% NC utilization of covered services 251 PROMUSAF Alliance santé MUSANT ADMAB AssEF MSS Rate of 60% 79% 75% sinistrality* Financial - subsidy from - Swiss - Lion’s club - Subsidy from support received WSM and the Cooperation (?) the Benin Belgian Fund for Government Survival Comments - According to - MUSANT is ADMAB - According to - The two Senelle 2005, the officially offers two Senelle 2005, mutuals are amount for paying endorsed by different the mutual heavily helath care services the products: a does not need subsidized by are entirely covered Association health any subsidy - - the by premiums. of Physicians micro-credit Contracting is Government. Subsidies are only (primary only with - Many for overheads. care care) private facilities members are in and a (AMCES) fact civil mutual (for servants other health (unlawfully) care using the services) system Sources - Senelle 2005 - Senelle 2005 - Senelle - Labie 2006 - Senelle 2005 - Labie 2006 2005 - Ceda 2006 - Ceda 2006 - Ceda 2006 252 Annex 9.3 Allocation of HEF credits to recipient facilities (2005-2007) in FCFA. Regions Health facilities 2005 2006 2007 ATACORA CHD 30.000.000 30.000.000 15.000.000 ZS Kouandé-Kérou-Péhunco 20.000.000 20.000.000 10.000.000 ZS Tanguiéta-Matéri-Cobli 15.000.000 15.000.000 25.000.000 ZS Natitingou-Boukounbé- Toukouunt - - 10.000.000 DONGA ZS Bassila 55. 000.000 40.000.000 10.000.000 ZS Djougou-Ouaké-Copargo - - 15.000.000 LITTORAL CNHU 200.000.000 160.000.000 100.000.000 CNHP Cotonou Ex Jacquot 20.000.000 20.000.000 25.000.000 Homel 20.000000 20.000.000 50.000.000 CNHPP Cotonou Ex Lazaret 20.000.000 20.000.000 5.000.000 ZS Cotonou 1 & 4 - - 5.000.000 ZS Cotonou 2 & 3 - - 15.000.000 ZS Cotonou 5 - - 20.000.000 ZS Cotonou 6 - - 5.000.000 ATLANTIQUE ZS Abomey-Calavi Sô Ava 50.000.000 40.000.000 20.000.000 ZS Ouidah-Kpomasse-Tori 55.000.000 50.000.000 12.000.000 ZS Allada-Toffo-Zè 15.000.000 15.000.000 25.000.000 Hôpital de gérontologie de Ouidah 18.000.000 15.000.000 3.000.000 BORGOU CHD Borgou 25.000.000 25.000.000 50.000.000 ZS Nikki-Kalalé-Pèrèrè 75.000.000 60.000.000 30.000.000 ZS Bembérèkè-Sinende 15.000.000 15.000.000 8.000.000 Zs Parakou-Ndali - - 10.000.000 ZS Tchaourou - - 30.000.000 ALIBORI ZS Kandi-Gogounou-Ségbana 75.000.000 60.000.000 30.000.000 ZS Malanville-Karamama 20.000.000 20.000.000 10.000.000 ZS Banikoara 20.000.000 - 15.000.000 MONO CHD Mono 25.000.000 25.000.000 40.000.000 ZS Lokossa-Athiémé 55.000.000 50.000.000 50.000.000 ZS Comé-Bopa-Grand Popo- Houéyogbé 15.000.000 15.000.000 25.000.000 COUFFO ZS Aplahoué-Djakotomey-Dogbo 65.000.000 50.000.000 25.000.000 ZS Klouékamey-Toviclin-Lalo 10.000.000 10.000.000 10.000.000 OUEME CHD Ouémé 25.000.000 25.000.000 60.000.000 ZS Adjohoun-Dangbo-Bonou 60.000.000 50.000.000 60.000.000 ZSPorto-Novo-Aguégués-Sèmè-Kpodji - - 5.000.000 ZS Avrankou-Adjara-Akpro-Misséréré - - 5.000.000 CHPP Akron Porto Novo - - 5.000.000 PLATEAU ZS Pobè-Adja Ouèrè-Kétou 15.000.000 15.000.000 25.000.000 ZS Sakété-Ifangni - - 15.000.000 253 Regions Health facilities 2005 2006 2007 ZOU CHD Zou 20.000.000 20.000.000 50.000.000 ZS Covè-Ouinhi-Zagnanado 65.000.000 50.000.000 60.000.000 ZS Abomey-Djidja-Agbangnizoun - - 10.000.000 ZS Bohicon-Zogbodomè-Za Kpota - - 10.000.000 COLLINES ZS Savalou-Bantè 60.000.000 50.000.000 50.000.000 ZS Savè- Ouèssè 15.000.000 15.000.000 25.000.000 ZS Dassa-Glazoué - - 25.000.000 TOTAUX 1.123.000.000 1.000.000.000 1.108.000.000 Source : Arrêtés N° 6261/MSP/DC/SGM/DPP/DRFM/SC de juillet 2005 N°1574/MSP/DC/SGM/DPP/DRFEM/SC de février 2006 N° 4106/MSP/DC/SGM/DPP/DRFM/SC/SAG d’avril 2007 254 Annex 9.4 : Disbursements of the HEF in 2006 Départements: Health districts, CHD, CNHU ALLOCATION CONSUMPTION RATE (%) ATACORA-DONGA ZS Tanguiéta-Matéri- Cobli 15.000.000 11.725.000 78% ZS Bassila 40.000.000 0 0% ZS Kouandé-Kérou-Pehunco 20.000.000 369.675 2% CHD Atacora 30.000.000 470.390 2% Sous-total Atacora Donga 105.000.000 12.565.065 12% ATLANTIQUE-LITTORAL ZS Abomey-Calavi-Sô Ava 40.000.000 14.062.040 35% ZS Allada-Zè-Toffo 15.000.000 14.999.980 100% ZS Ouidah-Kpomassè-Tori 50.000.000 0 0% CNHPP Lazaret 20.000.000 0 0% CNHP Jacquot 20.000.000 19.999.285 100% Hôpital Gérontologie de Ouidah 15.000.000 0 0% Hommel 20.000.000 19.999.285 100% Sous-total Atlantique-Littoral 180.000.000 69.060.590 38% BORGOU-ALIBORI CHD Borgou 25.000.000 24.996.830 100% ZS Nikki-Kalalé-Pèrèrè 60.000.000 4.174.800 7% ZS Bembèrèkè-Sinendé 15.000.000 1.405.530 9% ZS Kandi-Gogounou-Ségbana 60.000.000 15.481.550 26% ZS Malanville-Karimama 25.000.000 0 0% Sous-total Borgou-Alibori 185.000.000 46.058.710 25% MONO-COUFFO CHD Mono 25.000.000 23.464.220 94% ZS Lokossa-Athiémé 50.000.000 37.742.970 75% ZS Comé-Grand Popo-Houéyogbé-Bopa 15.000.000 14.444.450 96% ZS Klouékamé-Toviclin-Lalo 10.000.000 5.038.875 50% ZS Aplahoué-Djakotomey-Dogbo 50.000.000 13.296.000 27% Sous-total Mono-Couffo 150.000.000 93.986.515 63% OUEME-PLATEAU ZS Adjouhoun-Bonou-Dangbo 50.000.000 50.000.000 100% ZS Pobè-Adja-Ouèrè-Kétou 15.000.000 14.996.365 100% CHD Ouémé 25.000.000 25.000.000 100% Sous-total Ouémé-Plateau 90.000.000 89.996.365 100% ZOU-COLLINES ZS Covè-Ouinhi-Zagnanado 50.000.000 49.051.920 98% ZS Savè-Ouèssè 15.000.000 7.984.870 53% ZS Savalou-Bantè 50.000.000 34.742.010 69% CHD Zou 20.000.000 19.999.740 100% Sous-total Zou-Collines 135.000.000 111.778.540 83% CHNU 255 CNHU 160.000.000 22.739.760 14% TOTAL 1.000.000.000 423.445.785 42% Source : Rapport sur la gestion du Fonds sanitaire des indigents 2006. MSP/ DRFM 256 Annex 9.5 Main programs with external funding Source of funding 2004 2005 Axe 1 Réorganisation de la base de la pyramide sanitaire et renforcement de la couverture sanitaire Programme d’appui à la politique sanitaire du Benin (9eme FED) (PAPSBFED) FED 200 737 Projet socio-sanitaire suisse (PSSS) Suisse 200 200 Projet d'appui au développement du système de santé (PADS) BAD 0 0 Projet sante IBADEA (Santei Dadea) BADEA 800 725 Programme d'appui à la zone sanitaire de Bassila (PAZS Bassila) Belgique 636 636 Projet d'appui a la zone sanitaire de Come Belgique Programme d'appui à la zone sanitaire de Klouekanmey Belgique 650 650 Programme national de construction, d'équipement et de fonctionnalité des formations sanitaires Suisse 521 500 Axe 2 Financement et amélioration de la gestion des ressources du secteur Programme de renforcement des ressources humaines et des structures de gestion et de coordination du secteur sante Axe 3 Prévention et lutte contre les principales maladies et amélioration de la qualité Programme de promotion de l'hygiène et de l'assainissement (PPHA) Belgique 90 90 Programme élargi de vaccination phase 4 (PEV) OMS 697 697 FED ARIVA 56 56 UNICEF 128 128 Programme national de lutte contre la lèpre et l'ulcère de Buruli AFRF 39 39 Programme de sécurité transfusionnelle (SETRA) Belgique 0 96 Programme national de la médecine traditionnelle et de sa pharmacopée OMS 6 6 Axe 4 Prévention et lutte contre les maladies prioritaires Programme national de lutte contre le Sida (PNLS) OMS 34 34 Fonds mondial 815 615 Projet régional commun de prévention et de prise en charge des IST/VIH/SIDA le long du corridor de migration Abidjan-Cotonou-Lagos (projet corridor) IDA 200 200 Programme national de lutte contre le paludisme (PNLP) OMS 255 255 Fonds mondial 500 480 Unicef 7 7 Programme national de lutte contre la tuberculose (PNLT) Fonds mondial 200 200 Autres pays 39 39 Axe 5 Promotion de la sante familiale Programme intégré de sante familiale (PISAF) USA 700 540 Programme d'appui aux activités de la sante de la reproduction FNUAP 138 138 GRAND TOTAL 6911 7068 257 258 Annex 10.1. Estimated impact of community-based interventions on child mortality Baseline and target NNMR PNNMR IMR CMR Impact Estimation Table for U5MR coverages reduction reduction reduction reduction Baseline National Residual Residual Residual Residual Intervention packages coverage targets impact impact impact impact 1. Family oriented community based services 1.1 Family preventive/WASH services Insecticide Treated Mosquito Nets 16.2% 21.1% 0.0% 0.0% 0.0% 0.0% Quality of drinking water 66.0% 69.7% 0.0% 0.1% 0.1% 0.1% Supply of safe drinking water 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Use of sanitary laterin 38.0% 69.7% 0.0% 0.4% 0.2% 0.4% Hand washing by mother 13.0% 69.7% 0.2% 2.2% 1.2% 2.2% 1.2 Family neonatal care Clean delivery and cord care 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Early breastfeeding and temperature management 54.0% 54.0% 0.0% 0.0% 0.0% 0.0% Universal extra community-based care of LBW infants 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 1.3 Infant and child feeding Breastfeeding for children 0-5 months 43.0% 79.7% 0.1% 3.8% 2.0% 3.8% Breastfeeding for children 6-11 months 93.3% 93.3% 0.0% 0.0% 0.0% 0.0% Complementary feeding 68.0% 79.7% 0.0% 0.4% 0.2% 0.4% Therapeutic Feeding 43.0% 79.7% 0.0% 1.9% 1.0% 1.9% 1.4 Community illness management Oral Rehydration Therapy 27.7% 79.5% 1.2% 12.0% 6.3% 12.0% Zinc for diarrhea management 25.0% 19.1% 0.0% -0.2% -0.1% -0.2% Vitamin A - Treatment for measles 16.0% 19.1% 0.0% 0.0% 0.0% 0.0% Chloroquine for malaria (P.vivax) 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Artemisinin-based Combination Therapy for children 6.6% 7.6% 0.0% 0.2% 0.1% 0.2% Antibiotics for U5 pneumonia 21.7% 19.1% 0.0% -0.1% -0.1% -0.1% Community based management of neonatal sepsis 17.0% 8.2% -1.2% 0.0% -0.6% 0.0% Adjusted % mortality reduction 0.2% 18.4% 9.7% 17.8% Source : UNICEF MBB 2008 259 Annex 10.2. Estimated impact of preventive interventions on child mortality Baseline and target NNMR PNNMR IMR CMR Impact Estimation Table for U5MR coverages reduction reduction reduction reduction Baseline National Residual Residual Residual Residual Intervention packages coverage targets impact impact impact impact 2. Population oriented schedulable services 2.1 Preventive care for adolescents & adults Family planning 6.1% 15.0% 0.0% 0.0% 0.0% 0.0% HPV vaccination 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 2.2 Preventive pregnancy care Antenatal Care 42.0% 90.2% 1.5% 0.0% 0.7% 0.0% Tetanus toxoid 50.1% 90.2% 2.4% 0.0% 1.1% 0.0% Detection and treatment of asymptomatic bacteriuria 42.0% 90.2% 1.3% 0.0% 0.6% 0.0% Treatment of syphilis in pregnancy 50.0% 90.2% 0.3% 0.0% 0.1% 0.0% Intermittent preventive treatment (IPTp) for malaria in pregnancy 75.0% 90.2% 0.6% 0.0% 0.3% 0.0% 2.3 HIV/AIDS prevention and care PMTCT 6.0% 87.0% 0.0% 1.1% 0.6% 1.1% Cotrimoxazole prophylaxis for children of HIV+ mothers 4.0% 87.0% 0.0% 0.7% 0.3% 0.7% 2.4 Preventive infant & child care Measles immunization 85.0% 93.8% 0.0% 1.6% 0.9% 1.6% BCG immunization 99.0% 99.0% 0.0% 0.0% 0.0% 0.0% OPV immunization 69.0% 93.8% 0.0% 0.0% 0.0% 0.0% DPT immunization 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Pentavalent (DPT-HiB-Hepatitis) immunization 84.0% 93.8% 0.0% 0.0% 0.0% 0.0% Hib immunization 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Hepatitis B immunization 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Yellow fever immunization 65.7% 93.8% 0.0% 0.0% 0.0% 0.0% Neonatal Vitamin A supplementation 69.0% 93.8% 0.0% 0.0% 0.0% 0.0% Vitamin A - supplementation 94.0% 94.0% 0.0% 0.0% 0.0% 0.0% Zinc preventive 65.7% 93.8% 0.0% 1.3% 0.7% 1.3% Intermittent Presumptive Treatment (IPT) for children* 65.7% 93.8% 0.0% 5.1% 2.7% 5.1% Adjusted % mortality reduction 6.2% 9.6% 8.0% 9.3% Source : UNICEF MBB 2008 260 Annex 10.3. Estimated impact of clinical interventions on child mortality Baseline and target NNMR PNNMR IMR CMR Impact Estimation Table for U5MR coverages reduction reduction reduction reduction Baseline National Residual Residual Residual Residual Intervention packages coverage targets impact impact impact impact 3. Individual oriented clinical services 3.1 Maternal and neonatal care at primary clinical level Normal delivery by skilled attendant 58.0% 21.7% -6.0% 0.0% -2.9% 0.0% Resuscitation of asphyctic newborns at birth 17.6% 10.2% -0.2% 0.0% -0.1% 0.0% Antenatal steroids for preterm labor 17.6% 21.7% 0.4% 0.0% 0.2% 0.0% Management of neonatal infections 26.4% 21.7% -0.9% 0.0% -0.4% 0.0% 3.2 Management of illnesses at primary clinical level Antibiotics for U5 pneumonia 33.2% 35.6% 0.0% 0.1% 0.1% 0.1% Antibiotics for diarrhea and enteric fevers 28.2% 35.6% 0.1% 1.3% 0.7% 1.3% Zinc for diarrhea management 4.7% 35.6% 0.6% 6.1% 3.2% 6.1% First-line ART for pregnant women with HIV/AIDS 6.4% 19.1% 2.2% 0.0% 1.1% 0.0% 3.3 Clinical first referral care Basic emergency obstetric care (B-EOC) 4.7% 21.6% 0.4% 0.0% 0.2% 0.0% Normal delivery by skilled attendant 4.7% 21.6% 2.4% 0.0% 1.2% 0.0% Antenatal steroids for preterm labor 4.7% 21.6% 1.7% 0.0% 0.8% 0.0% Management of neonatal infections 4.7% 21.6% 2.8% 0.0% 1.3% 0.0% Universal emergency neonatal care (asphyxia aftercare, management of serious infections, management of the VLBW infant) 4.5% 21.6% 3.4% 0.0% 1.6% 0.0% Antibiotics for U5 pneumonia 5.1% 40.3% 0.0% 1.6% 0.8% 1.6% Antibiotics for diarrhea and enteric fevers 5.1% 40.3% 0.5% 5.2% 2.7% 5.2% Zinc for diarrhea management 5.1% 40.3% 0.5% 5.7% 3.0% 5.7% Chloroquine for malaria (P.vivax) 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Artemisinin-based Combination Therapy for children 4.7% 21.6% 0.0% 4.3% 2.2% 4.3% 3.4 Clinical second referral care Comprehensive emergency obstetric care (C-EOC) 7.4% 46.4% 6.1% 0.0% 2.9% 0.0% Management of neonatal infections 7.4% 46.4% 6.2% 0.0% 3.0% 0.0% Artemisinin-based Combination Therapy for children 7.4% 46.4% 0.0% 10.0% 5.2% 10.0% Management of severely sick children (referral IMCI) 7.4% 46.4% 0.7% 18.2% 9.5% 18.2% First line ART for children with HIV/AIDS 7.4% 46.4% 0.0% 0.7% 0.4% 0.7% Adjusted % mortality reduction 36.7% 42.0% 39.5% 41.2% Source : UNICEF MBB 2008 261 Annex 10.4. Estimated impact of preventive interventions on maternal mortality Maternal Baseline and target mortality Priority interventions coverages Phase III Baseline National Residual coverage targets impact 2. Population oriented schedulable services 2.1 Preventive care for adolescents & adults Family planning 6.1% 15.0% 0.3% HPV vaccination 0.0% 0.0% 0.0% Preconceptual folate supplementation 0.0% 0.0% 0.0% 2.2 Preventive pregnancy care Antenatal Care 42.0% 90.2% 0.0% Calcium supplementation in pregnancy 0.0% 0.0% 0.0% Tetanus toxoid 50.1% 90.2% 0.7% Deworming in pregnancy 5.4% 90.2% 0.0% Detection and treatment of asymptomatic bacteriuria 42.0% 90.2% 0.9% Treatment of syphilis in pregnancy 50.0% 90.2% 0.0% Prevention and treatment of iron deficiency anemia in pregnancy 70.0% 90.2% 0.0% Intermittent preventive treatment (IPTp) for malaria in pregnancy 75.0% 90.2% 0.0% Balanced protein energy supplements for pregnant women 0.0% 0.0% 0.0% Supplementation in pregnancy with multi-micronutrients 0.0% 0.0% 0.0% 2.3 HIV/AIDS prevention and care PMTCT 6.0% 87.0% 0.0% Condom use 6.0% 87.0% 0.0% Cotrimoxazole prophylaxis for HIV+ mothers 4.0% 87.0% 5.3% Cotrimoxazole prophylaxis for HIV+ adults 6.0% 87.0% 3.2% Cotrimoxazole prophylaxis for children of HIV+ mothers 4.0% 87.0% 0.0% Total 10.4% Source : UNICEF MBB 2008 262 Annex 10.5. Estimated impact of clinical interventions on maternal mortality Maternal Baseline and target mortality Priority interventions coverages Phase III Baseline National Residual coverage targets impact 3. Individual oriented clinical services 3.1 Maternal and neonatal care at primary clinical level Normal delivery by skilled attendant 58.0% 21.7% -12.2% Active management of the third stage of labor 17.6% 6.4% -4.2% Basic emergency obstetric care (B-EOC) 17.6% 21.7% 0.5% Resuscitation of asphyctic newborns at birth 17.6% 10.2% 0.0% Antenatal steroids for preterm labor 17.6% 21.7% 0.0% Antibiotics for Preterm/Prelabour Rupture of Membrane (P/PROM) 5.3% 21.7% 0.0% Detection and management of (pre)ecclampsia (Mg Sulphate) 0.9% 21.7% 1.3% Management of neonatal infections 26.4% 21.7% 0.0% 3.2 Management of illnesses at primary clinical level 3.3 Clinical first referral care Basic emergency obstetric care (B-EOC) 4.7% 21.6% 7.1% Normal delivery by skilled attendant 4.7% 21.6% 3.5% Active management of the third stage of labor 4.7% 21.6% 0.0% Comprehensive emergency obstetric care (C-EOC) 4.7% 21.6% 0.0% Resuscitation of asphyctic newborns at birth 4.7% 2.7% 0.0% Antenatal steroids for preterm labor 4.7% 21.6% 0.0% Antibiotics for Preterm/Prelabour Rupture of Membrane (P/PROM) 4.7% 21.6% 0.0% Detection and management of (pre)ecclampsia (Mg Sulphate) 4.7% 21.6% 1.0% Management of neonatal infections 4.7% 21.6% 0.0% Clinical management of neonatal jaundice 8.2% 4.7% 0.0% Universal emergency neonatal care (asphyxia aftercare, management of serious infections, management of the VLBW infant) 4.5% 21.6% 0.0% Total Mortality Impact First Referral Level Care 12.1% 3.4 Clinical second referral care Comprehensive emergency obstetric care (C-EOC) 7.4% 46.4% 14.7% Basic emergency obstetric care (B-EOC) 7.4% 46.4% 14.4% Normal delivery by skilled attendant 7.4% 46.4% 5.2% Active management of the third stage of labor 7.4% 46.4% 0.2% Resuscitation of asphyctic newborns at birth 7.4% 46.4% 0.0% Antenatal steroids for preterm labor 7.4% 46.4% 0.0% Antibiotics for Preterm/Prelabour Rupture of Membrane (P/PROM) 7.4% 46.4% 0.0% Detection and management of (pre)ecclampsia (Mg Sulphate) 7.4% 46.4% 1.9% Management of neonatal infections 7.4% 46.4% 0.0% Clinical management of neonatal jaundice 7.4% 46.4% 0.0% Universal emergency neonatal care (asphyxia aftercare, management of serious infections, management of the VLBW infant) 7.4% 46.4% 0.0% Total Mortality Impact Second Referral Level Care 38.2% Unadjusted % mortality reduction 48.3% Source : UNICEF MBB 2008 263