Madagascar - Population and health sector review

Report No. 6446-MAG Madagascar Population and Health Sector Review July 7,1987 Country Department III Africa Regional Office FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. its contents may not otherwise beo'' -i without World Bank authorizatioh-..-  CU REMIVAS Currency Unit: Madagasy Francs (FMG) Exchange Rate at the Time of Field Mission and Used in Report (unless otherwise noted) 1US$ = FMG 650 F19CAL YEAR Fiscal Year is the Calendar Year STArMAS Weights and Measures are based or metric system  FOR OMCUL Use ONLY ABVTIACr Madagascar's population is estimated to have reached 10.3 million in 1985 with a current rate of growth of 3.0 percent. Assuming unabated growth. population would increase to over 2, million by 2015 creating extraordinary constraints on the country's development. Demographic and economic projections demonstrate that these constraints can over the longer term be eased by reducing the h4gh level of fertility. This, however, does not happen automatically. It requires sustained efforts on the part of the Government t' develop a population policy and to set up action programs with particular attention to strengthening and expatding the family planning structure in conjunction with maternal and child hbalth programs. Contrary to common belief. recent statistical information indicates that the pattern of morbidity and mortality, particularly infant aortality, is not better in Madagascar than in many sub-Saharan African countries. Molnutrition in children appears to have worsened with the deterioration of economic conditions. The rapid expansion of the primary health care network could not prevent further deterioration of the population's health status due to lack of resources and quality of health personnel primarily in the area of preventive care. In light of the health system's limited absorptive capacity and severe financial constraints, the Government faces difficult choices in reconciling the s"-stantial oversupply of medical and paramedical graduates with the needs for physical resources such as drugs and medical supplies and improvement in in-service training and in managerial skills of existing personnel. Cons3quently. the development of a health stratagy and long-term health plan on the basis of an improved health statistics system is of critical priority. Moreover. alternatives to the existing financing of health services in the public and Private sectors need to be studied. This report is based on the findings of a PHN mission which visited Madagascar in February/March 1986. The mission was composed of Richard ileaver (mission leader), Herbert Boehm (senior project economist). Judith Editrom (organization/management specialist), Dr. Jean Lecomte (public health specialist consultant), Alain Marcoux (population specialist consultant), and Jean-Bernard Rasera (health economist consultant). The report was written by Herbert Boebm. This document has a restricted distribution and may be used by recipients only In the performance of their offcial duties. Its contents may not otherwise be disclosed without World Bank authorizttion.  POPULATION AM HEALTH SECR REPORT Table of Content. Page No. Abbreviations .......i. . 1 Glosary ..... ii Executive Sumnry and Recommendations ........................... iv I. ECONOMIC AND SOCIAL SETNG I....................... ....' 1 II. POPULATION .........**......e**ooo**e***oeeaoe*ee*,eeee 3 A. Present Demograpbic Situation and Trends ...... .......... 3 B. Population Projections .............,.................... 6 C. Socio-economic Consequences of Population Growth ........ 10 D. Population and Family Planning Activties a 21 _III* HEALTH 2............................. 7 A. Health and Nutrition Status ............................. 27 B. Heeath Policies and Strategies ..* ...................... 31 C. Health Sector Organization and Management ............... 34 D. Distribution and Utilization of Health Fecilities ....... 37 E. Manpower and Training of Health Personnal ..... .......... 44 P. Health Sector Finances and Investment Planing *......... 51 TABLES IN TEXT Table 1: Population Projections for Madagasear .......*.......... 8 Table 2: Urban Population Projections for Madagascar ........... 9 Table 3: Total Fertility Rates and Implied Contraceptive Prevalence Rates . ...........0............ 10 Table 4: Total and Per Capital CDP Projections in Real Terms ... 11 Table 5: "Active Age" Population and Annual Growth of Nale and Female Labor Force ... .................... 12 Table 6: Projected Total Food Energy and Protein Requirements .. 14 Table 7: Projected Fuelvood Requirements ... .................... 16 Table 8: Projected Fuelvood Deficits and Resultant Tree Production Aresu Needed ...................e*o*oo, 17 Table 9: Projected Population Reaching Age 6 and Projected Total Compulsory School Enrollment ....................... 18 Table 10: Projected Size of MCM Target Population ................ 19 Table I1: Projected Health Ezpenditures in Real Terms ............ 20 Table 12: Increase of Medical and Paramedical Staff ...# .......... 44  Pas,e No. Table 13: Projectod Surplus of Graduated Physicians by 1995 ..... 49 Table 14: Total Eapenditures and Sources of Health Care Financing 1985 ...............***.......... .............. . 52 ANNEXES Annex I: Demographic Situation and Trends: Gurrent Estimates and Projections ........................... .... ... 58 Annex II: Health *eo*oo*eooe......... 79 Annex III: Organizational Charts .......... ... .... .. ............. 102 Annez IV: General Health Statistics ............................. 106 Annex V: Health Manpower Statistics ............................ 112 Annex VI: Financial Statistics ..................... ....... 117 MAP: ............... IBRD 20035  ADB African Development Bank ASFR Age-specific fertility rates AVS Association for Voluntary Sterilization BNI Banque nationale de l'Industrie (National Back for Industry) CBR Crude birth rate CDR Crude death rate CPR Contraceptive prevalence rate CRS Catholic Relief Service CSSP Centres de soine de sant& primaire (Prim&ry health care center) CUS Centres urbains secondaires (Secondary urban centers) IBN Enqufte Budget MAnage (Household Budget Survey) EDF European Development Fund BEC European Economic Community EPC Enqu8te post-censitaire (Post-Census Survey) RSD Enqu8te socio-dimographique (Socic-Demographic Survey) PAO Food and Agriculture Organization FARMAD Laboratoires pharmaceutiques malgaches FI4S. Fianakaviana Sambatra (Associatien for Family Well-Being and the Health of the Mother) FP Family planning PTK Association of Malgache Christian Union GCU Grands centres urbains (primarily urban centers) GDP Gross domestic product IBRD International Bank for Reconstruction end Development IDA International Development Association INC Information. education and communication IMF International Monetary Fund IMR Infant mortality rate INSRE Institut national des Statistiques et de la Recherche 6conaaique (National Institute for Statistics and Economic Research: Now called State Data Bank) IPPE International Planned Parenthood Federation IUD Intra-uterine device MCH Maternal and child health MOH Ministry of Health NGO Nongovenmental organization NRR Net reproduction rate OFAFA Pharmaceutical corpowation PHC Primary health care SSSD Service des statistiques sanitaires et d6mographiques (Service for sanitary and demographic statistics) STD Sexually transmitted disease TVR Total fertility rate TWE Tonnes of wood equivalent UNDP United Nations Development Program UNESCO United Nations Educational. Scientific and Cultural Organization UNICEF United Nations Children's Fund UNFPA United Nations Fund for Population Activities URGD Urban-rural growth differential USAID United States Agency for International Development WHO World Health Organization  - ii.- GIl crude Sixth Rate: The nutber of births per 1,000 population in a given year. Cmde Death Rate: The nuiber of deaths per 1,000 population.in a given year. Total Fetility Rate: The average muter of children that would be born alive to a wmn during her lifetime if she pass8ed throgh her chileing years conforming to the age-specififeilit rates of a given year. Gross erution Rate: The average n'mber of daughters that would be born to a wcan during her lifetime if during bhr c e years he were to bear children at each age in acordance with - prevailing age-specific fertility rates. Net P n Rate: The average mnber of daughters that would be born to a wmumn (or group of wcv1n) if during her lifetiz she were to confom to the age- specifi fertilty and mortality rates for a given year. Lif e cptancy at Blirth: The average mnter of years a newborn would live if age/fex-specific mortaity trends prealing at the time of birth wre Infant Mrtality Rate: The nuner of deaths of infants under one year of age in a given year per 1,000 live births in that year. Dependency Ratio: The ratio of the econQuicaliy dependent part of the population to the productive part, arbitrarily defined as the ratio of the young (under 15 years of age) and the elderly (those 65 years of age and older) to the working age population (those 15-64 years of age). Parity: The nurter of children previously born alive to a wcoan. Rate of Natural Increase. The rate at which a pcpulation is increasing (or decreasing) in a given year due to a surplus (or deficit) of births over deaths, exprexsed as a p ntage of the total poplation.  -.I{i; - Rate of P laon Groth: The rate of natural increase adjusted for (net) migration, and expressed as a percentage of the total population of a given year. Contraceptive Prevalence The perntage of married wa,en in Rate: tive ages who are using (or whose husbands are using) any fonm of contraception at any given point in tins. Family Planning: The conscious effort of couples to regulate the n1nber and timing of births. Morbidity Rate: The frequency of disease in a population in a given period. Mortality Rate: The frequency of death in a population in a given period. Nalnutrition: Deficiency or excess of nutrients that produce disorders within cells, tissues or the whole body, which are sufficient to interfere with a person's health, potential for growth, normal physiologic fnctions and ability to interact with other individuals and with his physical and social environment. When it is due to dietary inadequacies, the malnutrition i8 tenmed r When it is due to anorexia, nelabsorption or metabolic alterations, it is termed s Low Birth Weight: Infant weight at birth less than 2,500 grams. Weight-for-Age Malnutrition: A common method of assessing. child nutritional state, based on the deficit in weiqht-for-aqge ccrpared to a reference standard. stunting: A method of assessing a person's nutritional state, based on the deficit in height-for-acqe ccwpared to a reference chart. Wasting: A nethod of assessing a person's nutritional state, based on the deficit in weiqht-for- heiqht cmpared to a reference chart.  - iv - CTIm IrZ AM A. Population 1. Since its first census using modern demographic standards in 1975, very little data have been gathered on a national basis to track trends in Madagascar's population growth. Consequently, in order to project present population levels and trends, the sector mission had to derive estimates for the current fertility and mortality rates from various different statistical surveys. These estimates show that Madagascar's population increased from 7.6 million at the time of the 1975 census to 10.3 million by 1985 reflecting a very high growth rate of 3.0 percent per annum, against the Government estimate of 2.8 percent. 2. Based on this estimate, and various assumptions about future fertility, the sector mission prepared long-term population projections. The resulta of these projections (see following table) may have most serious implications for the country's future development. MADAGASCAR: PROJECTED POPULATION SIZE AND GROWTH 1985 2000 2015 Total Population (in millions) No Fertility Decline 10.3 16.7 28.1 Gradual Fertility Decline 10.3 15.6 22.2 Rapid Fertility Decline 10.3 15.1 19.8 Average Annual Growth Rate (percent) No Fertility Decline 3.01 3.31 3.54 Gradual Fertility Decline 3.01 2.65 2.18 Rapid Fertility Decline 3.01 2.34 1.46 Source: World Bank projections. If Madagascar's population continues to increase at the same rate as it has over the tlst ten years, there could be almost 17 million people in the year 2000 and over 28 million in 2015. Population growth would accelerate to a rate of 3.5 percent, allowing a further doubling to 56 million within the following 20 years. If, however, rapid fertility declines are achieved--and they can be achieved through effective population policy and programs-, the population would increase to 15 million in 2000 and close to 20 million in 2015 with a continuing fall in the growth rate down to 1.5 percent. These projactions show that the potential for growth in Madagascar's population is massive and needs to be of serious concern to the country's policy makers and health officials. It must be emphasized, however, that the issue is not whether Madagascar could ultimately support a population of this size, but  - v - the effects of the rate of population grovth on development in the medium term. Rapid population growth threatens steady, long-term social and economic development, because it virtually negates economic advances, thus perpetuating unsatisfactory living standards and having critical implications for the adequacy of agricultural and energy production and the financing of the social sectors. 3. The future growth of population in Madagascar has important implications for the standard of living of the individual families and for progress in development of the society in general. The World Bank estimates under certain assumptions that Madagascar's gross domestic product (GDP) in real terms would grow long term between 3.0 and 3.6 percent. Since real advances in per capita income and living standards are only possible if the annual population growth remains below national output gains, this scenario is only foreseeable if fertility rates decline in the future. Conversely. continued rapid population increases will have severe and unprecedented implications for job creation, food and energy consumption and improvements in social services. 4. If fertility continues unabated, it is estimated that the "active age group" (the population aged ten years and over) will grow from the present 6.9 million to 18.5 million in 2015. At present labor force participation, additional emnloyment opportunities for 200,000 job seekers per year over the period 1985-2000 and for 360.000 over the period 2000-2015 will have to be created. Rapid population growth would also create enormous pressures on the country, in terms of food energy and protein requirements. Without resorting to imports, food production may bave to triple over the next thirty years requiring improvements in agricultural yield and development of as yet uncultivated land which iF constrained by a number of factors including an intractable land tenure system. To improve yields, inputs, such as fertilizers and better seeds, as well as irrigation systems and distribution networks, would be needed. To expand the cultivated area utilizing the remaining, increasingly marginal land. would not only require more development investment in order to obtain the same yields, but land used for food crops would then not be available for industrial and export crops, the country's major foreign exchange earners. It also must be noted that this situation is made worse by the deteriorat4'on of international exchange terms due to the decline of the price of agricultural products in comparison to imported products. 5. About 80 percent of Madag,scar's total energy consumption is in the form of fuelwood and charcoal. While the long-term sustainable annual yield of forests is estimated at 6.5 million tonnes of wood equivalent (twe), and gross national demand in 1985 amounted to 5.6 million twe, large deficits occurred in the highland provinces creating serious deforestation. Over the long term, the projected fuelwood demand under any population growth assumption would by far outstrip its supply and may accelerate the deforestation process to catastrophic proportions. Without significantly slowing population growth, even measures such as increased plantation  - vi - development, -emand managemtnt and wood substitution would most likely not be sufficient to resolve the energy crisis. 6. Population growth may also affect the adequate provision of social services such as education and health services. The results of projections of school age children indicate that, between the rapid end no fertility decline scenarios, there would be a difference of 460,000 additional first- graders and three million additional compulsory school students in 2015. Not only would a less rapidly growing population permit annual savings in educational expenditures estimated at FMG 70 billion (about US$110 million). but it could also keep primary education expenditures in line with the recurrent budget, facilitating the maintenance of universal primary education and affording improvements in the quality of the education system. 7. A large part of the health services in Madagascar is devoted to maternal and child health services (MCH). If ftatility does not decline, the potential MCH clients would increase from the present 4.6 million to 12.6 million in 2015, a number larger than the current total population of Madagascar. Over the same period, deliveries at health facilities would increase from 450,000 to 1.2 million per year. In contrast, rapid fertility decline would translate into a drastic reduct.on in client population realizing substantial comparative savings (estimated for the year 2015 alone at FMG 40 billion or US$60 million) which could be used to improve the quality of health services. 8. Madagascar has neither an explicit policy on population nor a formal government family planning program. Nevertheless, Government awarenees about population issues is increasing. The main non-governmental organization involved in family planning, PISA, is gradually gaining recognition; but with only 84 service locations, primarily in urban areas, the coverage of services is still very limited. The contraceptive prevalence rate is estimated at 1Z, placing Madagascar very close to the bottom of the list of developing countries. Recommendations 9. Population growth presents Madagascar with its most serious threat to steady, long-term social and economic development. Government action toward fertility reduction is therefore justified and necessary to safeguarC the future welfare of its society. A second justification for public support to population-related programs is that with family planning services, the Government would increase freedom of choice as well as encourage responsible parenthood; or, inaction on the population issue by the Government could effectively curtail individuals' rights to choose the size of their families, could in the future foster economic stagnation and possibly lead to more drastic actions in the future. 10. It is therefore recommended that a high level government committee, supported by a strong technical secretariat consisting, inter alia, of the new population unit of DG Plan, be created to develop a multi-  - vii - stctoral population policy with explicit demographic objectives and a plan of action for the next five years. The developnent of a population policy in Madagascar would need to address the Lolloving: (a) The capability of the Government to undertake demographic analyses and projections including the collection of reliable data on population size, fertility and mortality; (b) The potential for coordination with UNFPA regarding the assistance to the Statistical Data Bank (Banque des Donn6es de l'Rtat) for the planned 1990-91 census; (c) Mobilization of sustained Government commitment for proposed population policies; and (d) Proposals for an institutional framework capable of carrying out the planning and implementation functions of these policies. 11. Madagascar still has a law in its books dating back to colonial times (May 30, 1933), which prohibits the distribution of contraceptives. Although it is no longer enforced, the Government may wish to repeal it as a first active step in the direction of an increasing commitment to a population policy. 12. On the basis of available information on fertility, field visits and meetings with Government officials, the sector mission gained the clear impression that unmet demand for family planning services does exist. Consequently, immediate efforts should be directed towards strengthening the existing family planning structure and rapidly expanding family planning activities within FICH services to all health outlets. This would as a first phase not only entail financial and administrative support to PISA, a private family planning organization which was set up with IPPF's support, but also a program of FP activities for a as large number of public health facilities as possible. To achieve the widest knowledge of family planning services, it is recommended that the Government develop IEC programs with the assistance of experienced specialists in this field. Purthermore, in- service training in family planning for existing health personne., provision of the required supplies and equipment, and strengthening curricula and practical training programs in family planning of medical and paramedical students are recommended. B. Health 13. Even before the 1978 International Conference on Primary Health in Alma--Ata, Madagascar had already placed great emphasis on the creation of a BEkry health care network, expanding it to a current 1,950 facilities. To accommodate this expansion, the number of health personnel was increased from 3,900 in 1975 to over 8,200 in 1984. As budget allocations to the Ministry of Health (MOh) were not raised to take account of this expansion.  - viii - non-personnel expenditures had to be severely cut resulting in serious shortages of medicine, technical equipment and supplies. The Government. as most recently reflected in its 1986-1990 Development Plan, correctly continues to give high priority to rehabilitating and completing the primary health care system. together with improvement of management and supervision. 14. The institutional framework of the health sector, with the MOH as the most important modern health service provider, is generally adequate: the degree of decentralization to the provinces and medical districts is reasonably effective: and the distribution of health facilities among the primary, secondary and tertiary tiers appears sound. Management weaknesses at lower levels, however, are major contributors to poor provision of health services. Long distances to the health facilities, difficult terrain and frequent severe shortages of transportation, communication and materials aggravate these weaknesses. Insufficient supervision, inadequate training in management, and the absence of tools for good management, such as job descriptions, performance targets and incentives, are the primary causes for managerial deficiencies. 15. The sector mission estimates that probably 65 percent of the total population has easy access to modern health care services in the form of basic curative care--a very favorable ratio by African standards. Nevertheless, the extensive primary health care network, plagued by lack of resources and skilled personnel, experiences extremely low utilization rates since it has been unable to properly carry out its expected responsibilities, particularly with respect to preventive care. Frequently, no more than one day per month is devoted to outreach activities for health and hygiene, education, house visits or community health involvement. Health education appears limited with scant use of mass media such as radio, posters, and video. Operations of the health facilities in Madagascar are severely affected by the chronic shortage of medicine, vaccines and medical equipment. Although quite a large number of private pharmacies and dispensaries exist in the country, prices of pharmaceuticals can be very high. As a result, the poorest segment of the population often remains without modern medication. 16. It is important to note, however, that with its solid foundation for a physical and manpower health infrastructure, the country's health system has the potential for considerable improvements in efficiency and productivity. As Madagascar's mortality (infant mortality has been estimated at 125 per 1,000 live births) and morbidity conditions are less favorable than previously thought, significant declines should be achievable through improvements in preventive interventions and modest increases in medical supplies and pharmaceuticals. 17. Pre-service training for health aides, nuzsing staff and physicians is severely hampered by the lack of textbooks, teaching materials, and often inadequate command of the French language. Enrollment of students in the Faculty of Medicine at the University of Madagascar totalled 5,340 with an estimated output of over 400 graduates in 1985/86.  - ix - Since the health sector has no systematic, long-term manpower plan, the supply-driven recruitment process to a large extent ignores the actual long- term need for more health personnel as well as the implications for the recurrent budget. 18. At the present time, no organizational structure, policy or specific budget allocation exists for in-service training or personnel development. MOR increasingly recognizes the need for more continued training. The proposed National Center for Health Development could play an important role in this context. 19. Total health care expenditures in 1985 are estimated at FMG 49.2 billion or US$76 million, representing 3.2 percent of GDP or US$7.40 per capita. While in nominal terms MOH total expenditures have iacreased by 7.2 percent per year from 1977 to 1985. in real terms they declired by an annual rate of 3.4 percent. Total funds allocated to MOH as pa-:t of the total Government budget have also declined steadily from about 9.2 percent in 1976 to an estimated 6.7 percent in 1986, a level which is still well above most sub-Saharan African countries. The material decliae of funding, particularly during the recent economic crisis years. coinciding with a period of rapid expansion of health infrastructure, led to a significant shift in resource allocation and almost certainly constrained improvements in morbidity and mortality conditions. Recommendations 20. Program Changes. As future spending on health services will most likely be restrained by continued budget stringency. the Government could take advantage of a number of cost-effective program changes in health care provision. This would include assigning higher priority to preventive interventions such as immunization programs. pre- and postnatal care. vector control, health education, and at the same time strengthening simple .curative care as well as establishing an essential drug program. In most of these areas, increased emphasis on outreach activities and the development of community-based programs can yield substantial returns over passive, clinic-based health care as is currently almost exclusively practiced in Madagascar. An added benefit of developing a strong outreach system is that it also can serve as a distribution basis for family planning which has proven to be very effective in many African countries to increase contraceptive prevalence. 21. Planning. Development of a health strategy and medium-term plan (five years) that goes beyond the annual budget preparation are basic prerequisites for the strengthening of Madagascar's health sector. A management/forward planning approach needs to be introduced at all management levels of MON. To this end. the newly created Department for Studies, Planning and Programming in MOR is being strengthened with the assistance of the African Development Bank (AD). The ADB assistance  - x - could be most effective if key studies that will have an impact on future strategies and expenditures were accelerated. 22. More specifically, a cohesive health strategy should include health sector objectives based on desireable and feasible morbidity and mortality improvements, and it should provide guidelines to allocate available resources responding to the epidemiological situation. The strategy should also define the roles and specify the extent of coordination between the main health care providers; i.e., central and provincial govertments. village communities. and NGOs. The mechanisms for coordinating and utilizing foreign assistance should also be defined as part of the health strategy. The medium-term plan should be developed in accordance with the health sector strategy, specifying health targets and geographic and functional distributions of recurrent and capital expenditures and establishing priorities on the basis of demographic, epidemiological and operational information as well as on the basis of cost-effectiveness considerations. The plan should be approved at top government levels, updated annually on a rolling basis. and it should service as basis for the annual budget preparation. 23. Each year. MOH's budget requests are substantially inflated and then cut back by the Ministry of Finance, thus making the preparation and approval of the budget a time contuming process. Availability of a strategy and plan as recommended in para. 3.19 would rationalize the process of allocating limited resources. In addition, as part of the improvement of MON's planning process, particular attention must be devoted to investment planning. Investment projects must be justified on technical and financial grounds, prepared so that they are consistent with established standards, and prioritized. It is recommended that appropriate technical expertise be contractid to develop the necessary procedures including analysis of recurrent cost implications;train the pertinent staff, and participate jointly with the trained staff in the actual'preparation of alternative investment proposals. 24. The new, integrated health statistics system will be an important complement to both planning and management. However, at the present time, the system is cumbersome, untimely and suffers severely from underreporting. The present system does not contain cost information or performance targets. Since the system could provide comprehensive and reliable information if properly amended, it is recommended that a review is made of the progress and results achieved so far and proposals are prepared as to how the deficiencies could be corrected. Specialist inputs will be needed for this task. It is further recommended that these specialists identify data collection needs which a health statistics system cannot handle adequately; e.g., non-facility related epidemiological and nutritional status data. 25. Nutrition. This report deals with nutrition only to the extent that nutrition affects Madagascar's health status. Despite weak statistical data, the sector mission's findings seem to indicate that malnutrition in the country is high and getting rapidly worse in the wake of severe economic  - xi - adjustment measures. Poor feeding practices of children and adequate access to food by the poor appear to be the two key areas which need to be addressed. In order to understand the underlying constraints, it is recommended that a thorough nutrition study be carried out as soon as possible. This study would review the causes for the present situation; determine on the basis of a food balance the past and future trends in the nutritional value of the diet; make proposals for upgrading nutrition status data; and develop a nutrition plan including the required institutional and financial arrangements. Moreover, the surprisingly high degree of infant malnutrition calls for. close monitoring and study. Such a study would. inter alia, design a weaning strategy to promote improved breast-feeding and weaning practices. A nutritional surveillance study began in July 1985 with the aid of UNICEF. This study covers two centers in every of the six provinces. It would be important to follow the results. Organization and Management 26. Supervision and in-service training deficiencies and problems in logistics reduce the effectiveness of the health sector. While the need to improve these critical areps; is already recognized ar4 supported by MOH's top management and some progress has been made, efforts need to be intensified and accelerated. It is recommendod that MOH, with initial expert assistance, establish management training and consulting capacities within the Ministry in order to analyze these problems, develop the necessary supervisory and logistic structures, and formulate and undertake training programs for high and mid-level supervisory staff. The proposed National Center for Health Development, which will be financed by ADB, should play an important role in this context. In the ADB project, it is planned to improve health management, to introduce new methods of work and equipment at the central level, and to assure the training of Malagasy cadres in matters of management. Manower 27. In order to more accurately match actual job requirements with job levels, encourage higher performance and improve staff motivation, MOR's personnel administration needs to be modernized. The preparation of job descriptions for the various kinds of positions requires priority attention. Improved personnel planning over the short and medium terms is also essential to better manage placement and recruitment needs. Assistance from ADB is foreseen in this area. 28. Especially the imbalance between supply and demand of medical and paramedical personnel-a surplus of 3,000 medical doctors, 330 dentists and 1>850 paramedical graduates has been estimated over the next ten years- needs to be addressed immediately and effectively. Because of the large number of students already enrolled and graduated, an immediate, serious surplus of medical and paramedical graduates cannot be avoided. To improve  - xii - the situation over a longer term, a reduction in output from health training institutions should be considered as a matter of high priority. As part of the proposed health plan, the future requirement for medical and paramedical personnel should be determined. Subsequently, the intake and output of new students from the Faculty of Medicine, the School of Dentistry and the paramedical training schools should be adjusted accordirgly. Expert assistance in developing alternative proposals is recommended. 29. For recruitment of medical and paramedical personnel. the Ministry should review its policy of recruiting graduates in the order in which they applied since, once a backlog arises, this means recruiting staff who have completed their training several years earlier, and may have outdated knowledge and skills. Staff should instead be selected from the best students graduating in recent years, and the others should be informed that there is little or no chance that they will be recruited into the public service, encouraging them to look elsewhere for employment. Financing Issues 30. The constraints on the recurrent budget have severely affected the delivery of health care in Madagascar resulting in serious shortages of pharmaceuticals, medical equipment and support infrastructure. While communities make a considerable contribution to the construction and maintenance of primary health infrastructure, foreign assistance has been a comparatively weak source of finance. Fortunately, the decline in provision of pharmaceuticals has been partially offset by the expansion of drug supplies from the private sector. But since prices at pharmacies and private dispensariea have been high, the poor segment of the population has been affected most. With few exceptions, health services in the public sector are free of charge. As the recurrent health budget is not expected to increase significantly in real terms from present levels, alternative possibilities for raising finance must be identified. Accordingly, studies are proposed to focus on (a) the existing financing of health services in the public and private sectors and the expected effects of alternative changes in prices on demand for service; (b) the current policy of free medical care and its alternatives; (c) the present system of pharmaceutical procurement, distribution and manufacture, at' its alternatives to current practices in these areas; and (d) increased development of *the private health care sector. The ADB project foresees an analysis of financial constraints in the delivery of medical services.  POPULATION AW DALT SBCTOR UVIW I. UCCNCIC AND SOCIAL SBTTTIN 1.01 Madagascar, the "Great Island," comprises the world's fourth largest island and several smaller island dependencies. About the size of France and the Netherlands combined (587.000 square kilometers) it is located 400 km from the southeastern African coastline in the Indian Ocean. The island is characterized by rugged topography and substantial regional variations in climate and ecology. The central plateau region enjoys temperate to subtropical climate with favorable soils. The south has to cope with an arid climate and infertile soils. The eastern region possesses subequatorial climate with rich agricultural resources. In the west. rainfall decreases from north to south. During summer. violent cyclones cause widespread damage to large parts of the country. 1.02 With a population estimated to have reached 10.3 million in 1985. Madagascar is one of the least developed countries in the world. Its record of economic development since independence from France in 1960 has been disappointing. Since the early 1970s when the Government adopted highly dirigistic economic policies. per capita income has declined almost consistently, dropping to US$235 in 1985. which in real terms amounted to only 80 percent of the 1980 level and 50 percent of the 1970 level. The present Government, which took power in 1975, had initially implemented policies aimed at nationalization and expansion of the public sector and the "Invest to the Hilt" program of 1978-80, resulting in a profound financial crisis. Poor project selection did not realize the expected benefits, and heavy commercial foreign borrowings could not be serviced frci project- generated funds. This exerted further pressures on budget expenditures already strained from falling agricultural output, heavy rice importation and high oil prices. Consequently, budgetary and balance of payment deficits rose rrecip4tously, and despite comprehensive price controls inflation accelerated. 1.03 With the support of International Monetary Fund (IMF) standby programs, the Government initiated in the early 1980. five severe demand management programs. In addition. several violent cyclones devastated large parts of the island and added to the already massive economic problems. Currently, a sixth program is in preparation and should run through the end of 1987. The stabilization efforts and economic reforms puraiued under the standby programs have also been supported by International Development Association (IDA) sector adjustment lending aimed at removing structural constraints from the economy, as well as by consultative group meetings through successive debt reachedulings and additional concessional assistance. Progress has been achieved in the tightening of credit ceilings, curbing of domestic and external demand pressures, reducing import volumes, controlling budgetary expansion, holding down inflation, d3creasing most consumer subsidies, and adjusting the country's exchange rate to more realistic levels. Moreover, the Government started to rationalize its investment program, encourage private sector interest and put emphasis on  -2- the rehabilitation of existing productive resources and infrastructure. While these considerable efforts achieved an economic turnaround and appear to have initiated a recovery phase. the country's gross domestic product (GDP) growtb rate remains below its population growth. per capita consumption continues to decline and productive activities remain severely restricted by foreign exchange scarcity and hampered by internal and external trade regulations. In order to ensure continued and sustained progress. the Government must implement a host of difficult policy measures to reform the economy. To this end, the Bank has expressed willingness to help the Government in its efforts to halt the economic decline and achieve real development. 1.04 Providing 35 percent of GDP and 85 percent of employmeza. generation, agriculture dominates Madagascar's economy. Agricultural products account for about 80 percent of the country's foreign exchange earnings. While the long-term devclopment potential of the agricultural sector is considered to be excellent, growth in agricultural production over the past decade has been sluggish with an average annual growth rate of less than 2 percent. Particularly disconcerting is the development of the sector's predominant commodities, rice and livestock, which constitute the population's primary food supply. Export crops (coffee, cloves, vanilla, etc.) and industrial crops (cotton, sugar. etc.) as well as the supply of raw materials to the predominantly agriculture-based industrial sector have experienced serious problems since the mid-1970s. As the economy began to falter, the ecountry even had to import rice as price subsidization of rice created significant shifts in dietary patterns and further fueled its demand. While the Government has, since 1982. as part of its economic recovery program, embarked on far-reaching agricultural reform programs, including the development of a short-term food security strategy, a key question over the longer term will be what pressures high rates of population growth will exert on the agricultural production system (see paras. 2.20 - 2.24). 1.05 Despite its size and significant potential energy sources (hydropower. oil, coal, uranium, etc.), Madagascar is facing considerable energy problems primarily as a result of a growing fuelwood shortage arising from overuse of its forest resources (which are the dominant energy supply), as well as its total dependence (except a small domestic coal production for its commercial energy needs) on oil imports. As the household sector accounts for over 83 percent of the country's net energy consumption almost entirely in the form of wood fuels, Madagascar's population growth would undoubtedly have serious repercussions on the country's future energy situation (see paras. 2.27 - 2.29). 1.06 The Malagasy population is descended solely from immigrants. No remnants of any prehistoric culture have been found. The first settlers, the mysterious Vazimba, arrived from Asia in the third century. From the fifth to the eight centuries, migrants originated mainly from Indonesia, followed by Arabs. and from 1500 onward, also by Portuguese. French, British and East African settlers. After a long succession of regional and national monarchies, the French assumed control over the country in 1884. The colonial period ended in 1958, and Madagascar gained full independence in  -3- 1960. The present Government adopted a Charter of Malag,asy Social.ist Revolution which initiated massive reform policies resulting in nationalizations. expansion of the public sector and decentralizing Government authority to the six provinces, the Faritany. and 11,500 village communities, the Fokontany. These reform policies included the promise to offer free medical services and education to all Malagasy. 1.07 With its insular location close to Africa and its Polynesian- emanated immigration, Madagascar possesses a unique blend of culture and customs rooted deeply in the long history of well functioning monarchies and missionary activities. In addition to the principal eighteen tribes. communities of Indians, Chinese and Arabs exist primarily along the northern and eastern coastline. Intermarriage betweon the groups occurs frequently. Despite its diversity of racial background, a single language, Malagasy (derived from the Malayo-Polynesian group of languages) with regional dialects, is spoken throughout the country. The 1975 census recorded 24 percent of the population as Catholic and 21 percent as Protestant. In the capital and in the coastal cities, Islam and other religions can also be found. 1.08 The family and its well-being form the primary focus of the Malagasy society. As shown later in this report with more specificity. the Malagasy family is large and close relationships with other families within the same village frequently exist. Deep respect for family ancestors (the Razana), fear of breaking taboos (Fady) and love for children are characteristic features of the Malagasy culture. Sterility is considered to be a curse. Children help their parents from a young age, planting. collectinjs fruit, harvesting. shepherding and selling agricultural products. II. POPUATION A. Present Demographic Situation and Trends Sources and Quality of Data 2.01 While population counts have been conducted in Madagascar since almost the beginning of this century. the recording of vital statistics over the last decade has deteriorated rapidly to the point where it no longer serves demographic needs. The first census using modern demographic standards was held in 1975. It was followed by i post-census survey. Since then, very little data have been gathered on a national basis to depict trends in fertility and mortality, and consequently estimates of the current size of the population and its growth rate rely on estimates derived from several other statistical surveys, such as the 1980 Household Budget Survey and the 1984 Socio-demographic Survey. The still unpublished Household  -4- Budget Survey, which was conducted, like the 1975 census and the post-census survey, by the National Statistical Institute (INSRE), covered the rural areas and smaller urban centers as follow-up to a 1977/78 survey of the seven major urban centers. The Socio-demographic Survey was conducted as part of a United Nations Educational. Scientific and Cultural Organization/United Nations Fund for Population Activities (UNESCO/UNFPA) population education project. With UNFPA assistance, the Ministry of Health (MON) established in 1983 a demograpaic unit to set up and maintain information on birth, death, and total population for a permanent health data collection system based on a sample of 1,439 communities (12.5 percent). The first returns for the year 1984, however, included only 340 communities (3.0 percent) and suffered from a considerable variation in statistical quality. 2.02 Thus, the regular, ongoing demographic data collection in Madagascar is not able to readily supply information such as the country's population size or its growth trends. Nevertheless, analyzing the various existing surveys, the mission was able to produce updated estimates for fertility and mortality rates and consequently project the present population levels and trends starting from actual 1975 census data. The methodology used for these projections is described in Annex I. Population Size and Growth 2.03 At the turn of the century, an administrative head count put the population of Madagascar at about 2.2 million people. Over the next fifty years, population grew on average at a modest one percent rate per year including several periods of actual decline caused by epidemics and hostilities. From 1950 onward, population growth accelerated with official records putting it at an average 2.3 percent per annum between 1950 and 1960, and at an average 3.5 percent per annum between 1960 and 1970. At the time of the 1975 census, Madagascar's population had increased to 7.6 million. Using this figure as a projection base without adjustment for possible undercounting, the population appears to have increased to 10.3 million by 1985 with a current growth rate of 3.0 percent per annum as against the Government estimate of 2.8 percent. At a continuing growth rate of 3.0 percent, the population would double in 24 years. Fertility and Mortality 2.04 The 1985 population estimate is based on a crude birth rate (CBR) of 45.6 per 1,000 population and a crude death rate (CDR) of 15.5 per 1,000 population. (See Annex I for source of data.) The total fertility rate (TFR) has increased slightly from 6.4 children per woman in 1975 to the present estimate of 6.6 children. The gross reproduction rate currently is about 3.3 daughters born alive per woman while the net reproduction rate is considerably lower; i.e., 2.4 due to the high mortality rates. 2.05 Life expectancy at birth has improved significantly over the last twenty years; in 1966, it was given at 38 years. The 1975 census reported  -5- an increase to 45.4 years. and while further advances were achieved up to 1980, a slight decline has reduced the hitherto achieved progress. The current estimate prepared by the sector mission-primarily dependent on the 1984 Socio-demographic Survey-shows life expectancy to be 51 years with a differential of about three years in favor of females. At considerable variance from past Government statistics, the present infant mortality rate (IMR) has been conservatively estimated by the sector mission to be about 125 per 1.000 live births as against the 68 per 1,000 figure estimated from women's report in the 1975 census and the 75 per 1.000 current figure implied in "Population de Madamascar." a bockle. published in January 1985 by the Ministry for Scientific Research and Technology for Development. As infant mortality is widely viewed as an important indicator for a population's health status and closely linked to economic and social welfare, the current high level of the IMR must give rise to serious concern. While some people in the Government suspect that the INR has been deteriorating since the past surveys, they are of the opinion that Madagascar's IMR would still be considerably better than those in the rest of sub-Saharan Africa. The mission's IMR estimate of 125 per 1.000, however, would point to infant mortality conditions which are not much different from the many sub-Saharan African countries (Ghana, 97; Kenya. 81; Mozambique, 109; Zaire. 106; and Ethiopia, 135). Age and Sox Structure of the Population 2.06 The mission has derived an age structure for mid-1985 on the basis of the above fertility and mortality estimates. No external migration was assumed. As can be expected in view of the high fertility and mortality rates, the age structure is youthful with 46 percent of the population under 15 years of age; and only 3 percent of the population over 65 years of age. The dependent population, that under 15 and over 65 years of age, is almost equal in size (96 percent) to the potentially econoeically active segment of the population aged between 15 and under 65; i.e.. a dependency ratio close to one. Age structure and dependency ratio are similar to those in Eastern African countries. 2.07 The sex distribution of the 1985 population estimates gives a sex ratio of 98.5 males per 100 females, which is in the normal range. Nevertheless, in the 25 to 69 years age group, Madagascar's women exceed the number of men. The average age at marriage is 19.6 years for women and 22.9 years for men. Female marriage is thus not particularly early by African or Asian standards. The number of married women without children (zero parity) is high. For example, seven to eight percent of married women in rural are"s between the age of 25 to 34 hve had no child. As this may either be linked to insufficient data or indicate a possible sterility problem, future investigation of this subject is warranted. Population Density and Spatial Distribution 2.08 Witb only 17.6 inhabitants per sq. km. Madagascar is still sparsely settled. However, only 20 years earlier in 1966, there were just  -6- 10.6 inhabitants per sq. km-, ten years later, 13.0. Results from the 1975 census exhibit a considerable variation in density of settlement between the six provinces with Antananarivo showing 37.2 inhabitants per sq. km. Toliary and Majahanga 6.4 and 5.5 respectively and the other three provinces close to the overall country average. Antananarivo province, where the capital is located, accounts for 28.5 percent of the population, but occupies less than one tenth of the country's territory. 2.09 Madagascar is still a predominantly rural country with 83.6 percent of its population living in rural areas according to the 1975 census. At that time, 47 urban centers with more than 4,000 people existed. The capital city. Antananarivo, alone accounted for about 430,000 inhabitants in 1975 or 34.8 percent of the country's total urban population. Although the 1975 census provided this valuable information, virtually no data are available on trends in the regional distribution of Madagascar's population. Only a modest flow of internal migrants to nearby urban centers has taken place. And even the capital city, where the inflow of the unemployed has been consistently discouraged. has seen only a relatively modest growth rate of less than five percent per annum, which is modest compared to overall growth rate of the country's total population. Therefore, net growth rates for urban and rural populations do not differ greatly. Consequently. the sector mission estimates the current rural proportion of the total population still to be as high as 80 percent. Similarly, external migration data are practically nonexistent. With the exception of some overseas students who remain abroad after their education. emigration seems to 1be extremely limited. Due to Madagascar's insular location, the same holds true for immigration. B. Population Projections Construction of Projections 2.10 The sector mission prepared a set of three population projections for the period 1985-2015 using the 1985 population estimate of 10.3 million as their base. For all three cases, it was assumed that future mortality rates would decline at a moderate rate resulting in higher life expectancy; i.e., an additional 1.5 years for females atd about 1.4 years for males for each five-year period. Different assumptions as tc expected fertility trends over the 30-year projection period were applied. The first trend alternative assumes constant age-specific fertility rates at the levels estimated for the period 1980-1985. As fertility appears to have been relatively stable over the last ten years, this no-change scenario becomes a distinct possibility if a national family planning program is not implemented and if factors conducive to fertility declines (e.g., improvements in education and living conditions) are offset by factors conducive to fertility increases (e.g., improvements in maternal health care resulting in better health for women and better nutrition). Based on the  -7- World Bank's "standard projection,"l as defined in the "World Development Report 1984." the second trend alternative assumes a moderate and continuous fertility decline beginning in 1985 reaching replacement fertility between 2030 and 2035. Hence, the TFR is assumed to fall from 6.6 in 1985 to 4.0 during the 2010-2015 period. The third trend alternative assumes rapid declines in fertility to achieve replacement fertility by 2015-2020; i.e., 15 years earlier than in the second trend alternative. The TFR would in this case fall to 2.8 during the period 2010-2015. This maximum decline was modeled on countries which have achieved a rapid fall in fertility over the recent past after their successful implementation of strong government policies and programs in family planning and substantial strides in areas such as adult literacy, health services and alleviation of poverty. Details of the projections are given in Annex 1. Results of Population Projections 2.11 Table 1 (on the following page) summarizes the projected size of the population of Madagascar ever the next 30 years. 2.12 Demographic projections should not be treated as forecasts; they only serve as illustrations of possible effects of selected assumptions on future population growth. The results in Table 1 indeed paint a dramatic picture of the potential for future population growth in Madagascar. If fertility remains at the past ten-year level, the country's population would grow by 62 percent by the year 2000, and more than two-and- one-half fold by 2015. At that time, 28.1 mi;llion people would occupy the island and the population growth rate would have accelerated at a rate of over 3.5 percent. At this rate, the country's population would double again over the next 20 years. Such rapid growth would probably make it increasingly difficult to continue the assumed mortality decline due to inordinate pressures on the country's resources just to keep up the same level of per capita expenditures for social services. The dependency ratio would remain very high-at around 97 percent. As shown later in this chapter, the prospects of uncontrolled fertility under this assumption are extremely serious. 2.13 Especially over the longer term. fertility declines could alter the picture significantly. A gradual fertility decline would yield a total population of 22.2 million at the end of the projection period# i.e., almost lFor the standard projection. the assumed year for replacement-level fertility in Madagascar was regressed on several predictors: the current total fertility rate for each country, the change in this rate over the previous ten years, the proportion of couples using contraceptives. and the current female life expectancy.  -8- Table 1: POPULATION PROJECTIONS FOR MADAGASCAR 1985 2000 2015 Total Population (in million) No Fertility Decline 10.3 16.7 28.1 Gradual Fertility Decline 10.3 15.6 22.2 Rapid Fertility Decline 10.3 15.1 19.8 Crude Birth Rate (per 1,000) No Fertility Decline 45.6 45.4 44.8 Gradual Fertility Decline 45.6 38.2 30.8 Rapid Fertility Decline 45.6 34.9 23.5 Crude Death Rate (per 1.000) Nc Fertility Decline 15.5 12.3 9.5 Gradual Fertility Decline 15.5 11.7 9.0 Rapid Fertility Decline 15.5 11.4 8.9 Average Annual Growth Rate (percent) No Fertility Decline 3.01 3.31 3.54 Gradual Fertility Decline 5.01 2.65 2.18 Rapid Fertility Decline 3.01 2.34 1.46 Source: World Bank projections. 6 million people fewer than without fertility change. The growth rate would have come down to 2.2 percent implying a further doubling of the population in 36 years. The dependency rate would show a significant reduction to about 68 percent, because the proportion of children in the population would be considerably lower. If rapid fertility declines can be achieved over the next 30 years. total fertility would drop from the current 6.6 live births per woman to 2.8 at the end of the projection period. This would result in a total population in 2015 of 19.8 million or 8.3 million less then without fertility change. The growth rate would then be 1.5 percent. lengthening the doubling time to 46 years. and the dependent ratio at the projection end would be only 55.4 percent as a result of a more moderate birth rate. Urban and Rural Population Growth Projections 2.14 For the three trend alternatives, the relative proportion of rural to urban population was projected, using the Government's definition of an urban center; i.e., a concentration of more than 4,000 inhabitants. However, the possible effects of alternative fertility assumptions on the urbanization process have not been taken into account. As indicated above, very little information is available on historic trends of urban versus rural populatiou growth in Madagascar and assumptions regarding urbanization are in general considersbly more subject to volatile factors than most other demographic events. The sector mission's projections are therefore merely an attempt to illustrate one likely scenario. in which the urban/rural growth differential was assumed to remain at a constant two percent over the  -9- next 30 years. Thus, the following projections are generated: Table 2: URBAN POPULATION PROJECTIONS FOR MADAGASCAR (in '000) 1985 2000 2015 No Fertility Decline 1,984 4,060 8,500 Gradual Fertility Decline 1,984 3,790 6,700 Rapid Fertility Decline 1,984 3,680 6,000 Proportion of urban to total 19.2 24.3 30.2 population (in percent) Source: World Bank projections. 2.15 Accordingly, without change in fertility. about 8.5 million people would live in urban centers by the year 2015-a more than fourfold increase. Over the same period, people in rural areas would increase from 8.3 million to 19.6 million and their share of the total population would fall from about 80 to 70 percent.2 While the urban/rural proportion would evolve the same. under declining fertility assumptions. the urban population would grow at slower rates, reaching 6.7 million under the moderate decline hypothesis and 6.0 million under the rapid decline hypothesis. Yet, even under the latter hypothesis, the urban population would triple over the next 30 years. Implied Contraceptive Prevalence 2.16 In order to provide an idea as to the magnitude of change in the overall reproductive behavior required to achieve the fertility decline assumed under the gradual and rapid fertility decline alternatives, the implied contraceptive prevalence rates were calculated. They are given for five-year intervals in Table 3. 2.17 These projections assume roughly the same relative composition of contraceptive methods in the future as presently used in Madagascar. The present contraceptive prevalence rate is estimated to be in the neighborhood of one percent. This would need to increase to 21-30 percent in 1998 and to 42-60 percent in 2013 under the gradual and rapid fertility decline assumptions. It must be noted that contraceptives are used at the present 2 The United Nations' project Madagascar's urban population to increase to 24.44 percent in 1990 and to 31.46 percent in 2000. See Patterns of Urban and Rural Population Growth. 1980.  -10- Table 3: TOTAL FERTILITY RATES (TFR) AND IMPLIED CONTRACEPTIVE PREVALENCE RATES (CPR) Gradual Fertility Decline Rapid Pertility Decline Year TFR CPR TFR CPR 1988 6.2 8% 6.0 11% 1993 5.7 152 5.4 212 1998 5.3 21% 4.7 30% 2003 4.9 28% 4.1 402 2008 4.4 35% 3.5 501 2013 4.0 42% 2.8 60% Source: World Bank projections. time in aub-Saharan Africa by less than 10 percent of married women and that the level of contraceptive use implied in the projections for 1998 would then be about the same as the current level in South Asia. The proportion implied for 2013 under the rapid fertility decline assumption would be close to the current usage in countries such as Thailand. Sri Lanka or Indonesia. C. Socio-economic Consequences of Population Growth 2.18 As shown in the previous chapter. even if fertility declines rapidly. Madagascar's population is likely to increase by almost 50 percent over the next 15 years and double over the next 30 years. By then, almost unavoidably, it would reach 20 million and may even be as high as 30 million. This chapter shows that this will not only have important implications for the individual families and for society in general, but will also represent an awesome challenge for planners and the country's policy makers. 2.19 Empirical evidence has shown that population growth above two percent "acts as a brake on development." 3 However, wide differences can be expected among countries as to what extent population growth affects their economies and their efforts to improve their citizens' welfare and at the same time reduce social inequalities. Even in countries with large unexploited natural resources, progress in development may be held back by limited availability of financial resources for investments and recurrent expenditures for projects and associated infrastructure as well as by the need for human skills to coustruct and operate these projects. This chapter investigates how the anticipated population growth could affect capital formation, labor force and food consumption in Madagascar. 3 See Chapter 5. World Development Report 1984.  -11- Implications for Economic Growth 2.20 From 1970 onward, Madagascar's economy has been growing at a slower pace than its population. The GDP for 1985 amounted in real terms to only 80 percent of its 1980 level. The Bank estimates4 that, under the assumption the Government would successfully implement a far-reaching set of accelerated structural adjustment policies, GDP could improve in real terms at an annual rate of 3.1 percent for the rest of the l?80s. The alternative population growth projections for 1985-1990 made by the sector mission are 3.1 percent for no fertility change, 2.9 percent for moderate decline, and 2.8 percent for rapid decline. The population growth would, consequently, to a large extent negate any advances in per capita income and living standards. For the subsequent five-year period, modest advances could be possible if the economy would grow at the projected rate of 3.6 percent. Beyond that time, the prospects for Madagascar's economy are still quite uncertain, but in view of the continued pressure of population increases, substantial gains in national output would be required to realize net advances in per capita income. However, if GDJP over the long term (1995- 2015) grows only at 3 percent per year, advances would only be achievable under declining fertility trends, as the following table illustrates. Table 4: TOTAL AND PER CAP;TA GDP PROJECTIONS IN REAL TERMS Rat. 1985 2000 2015 GDP (Billions of FMG) 1,399 2,245 3,495 GDP Per Capita ('000 FMG) No Fertility Decline 136 134 124 Gradual Fertility Decline 136 144 157 Rapid Fertility Decline 136 148 176 Source: World Bank projections. These figures, however, dc provide any insight as to how the projected economic and population gL. would influence income distribtition within Madagascar, particularly whether there would be a future decrease or increase in poverty. Increases in per capita income alone-projected for the next 30 years under declining fertility assumptions-may be too mfAest to lift a substantial proportion of the population out of poverty. Undoubtedly, if economic development increased substantially, the per capita income could improve. 4 Country Economic Memorandum for Madagascar, March 1985.  -12- Implications for Active Ase Group 2.21 If fertility continues unabated in Madagascar, the "active age group"--the population aged ten years and over--would grow from the present 6.9 million people to 18.5 million in 2015; i.e., at a rate equal to the total population increase over the next 30 years. Under declining fertility assumptions, the proportion of the active age group over the same period would inc-ease, thus reducing the dependency burden. The following table skows that even in the declining fertility scenarios the yearly need for additional jobs for both sexes would remain substantial. albeit less pronounced than without marked fertility changes. As these projections prasume the same labor force participation rates as prevalent in 1985, additional employment opportunities would need to be created if more people are looking for jobs. Table 5: "ACTIVE AGE" POPULATION AND ANNUAL GROWTH OF MALE AND FEMALE LABOR FORCE Percent Increase Active Age Population (in million) 1985 2000 2015 1985-2015 No Fertility Decline 6.9 11.1 18.5 169.6 Gradual Fertility Decline 6.9 10.9 16.5 140.0 Rapid Fertility Decline 6.9 10.8 15.6 127.3 Annual Labor Force Increase (in 000) 1985-2000 2000-2015 No Fertility Decline Males +112 +197 Females + 90 +159 Gradual Fertility Decline Males +108 +147 Females + 87 +119 Rapid Fertility Decline Males +107 +126 Females + 86 +101 Source: World Bank projections.  -13- Consequently, in order to just maintain productivity, the country's investments or stock of capital would have to be increased commensurate with the labor force. If this could not be achieved, productivity and thus income would fall. At the same time, wages would decline in proportion to profits and rents, and income inequalities would be exacerbated, particularly since the additional jub seekers may be expected to be relatively young, uneducated and unskilled. Unless the urbanization process accelerates beyond expectation, which in turn could create massive urban un- or underemployment in the urban centers, the labor force in agriculture would grow significantly. How to employ these job seekers productively in the future is of critical concern to Madagascar. Implications for Agricultural Production 2.22 While agriculture is the mainstay of Madagascar's economy, its food crop production for domestic consumption (rice, manioc, potatoes, maize, etc.) has been insufficient since the mid-1970s and had to be supplemented at times by heavy food imports. The importation of rice, the staple preferred by most people in Madagascar, peaked in 1982 with about 350,000 tonnes but declined sharply from about 185,000 tonnes in 1983 to 106,000 tonnes in 1985. While systematic data regarding the nutritional status on a national or regional basis are not routinely collected in Madagascar, several surveys, studies and clinical reports suggest that a serious food shortfall exists at the present time (for details see Annex II). 2.23 The following table portrays mission estimates of the growth in food demand in terms of food energy and protein requirements over the next 30 years under the three population growth alternatives, but not taking account of othier relevant determinants such as rising income, which could raise food demand further. Even with rapid fertility decline, food production would have to double over the next 30 years. Moreover, with no fertility decline and taking real income growth into consideration, food production may have to triple to meet requirements without resorting to food imports. These are, by any standard, formidable challenges. 2.24 Clearly, unchecked population growth would create a severe strain on the country's capacity to feed itself. In order to entirely eliminate shortfalls, domestic production would, without fertility decline, need to increase by 4.6 percent annually up to the year 2000 and 3.5 percent thereafter. It must be noted that only a few countries have ever been able to sustain agricultural growth of more than 3.5 percent per annum over more than two decades (e.g., Mexico, Thailand and the Philippines). Under the rapid fertility decline assumption. domestic production would need to be raised by 4.0 percent annually to the year 2000 and 1.8 percent thereafter. 2.25 For demonstration purposes, this would mean that if all additional food would come from rice production and if no productivity increases are assumed, the presently cultivated area for paddy production (paddy  -14- production grew at less than 0.5 percent a year over the past decade)5 would have to be almost doubled within thirty years in the rapid fertility decline scenario and expand almost threefold if fertility remains at a constant level. Depending on the different scenarios, the area of rice paddy production would thus have to increase between 1.2 and 2.2 million hectares over the next 30 years assuming a national average production of 1.8 tonnes per hectare. In the worst case, this would require 40 percent of the yet unused cultivable land, estimated at 5.5 million hectares. However, new lands for paddy production are increasingly marginal in the densely settled areas of Iae country, requiring more development investment and yielding less per hectare. Other factors wou.ld improve prospects for better rice productivity, but it is uncertain whether they could prevail in Madagascar in the future. For instance, overall yield improvements would reduce the amount of land that needs to be devoted to rice production. Advanced projects using water controls, fertilizers and pesticides obtain as much as four tonnes per hectare, while small holder paddy fields on average yield no more than 1.5 tonnes of paddy per hectare. Of course, other food sources; e.g., maize, cassava, beans, potatoes, could also be developed and may provide substantial economic advantages over rice production. To Table 6: PROJECTED TOTAL FOOD ENERGY AND PROTEIN REQUIREMENTS 1985 2000 2015 Eneruy Requirements (in billions of kcal/year) No Fertility Decline 7,256 11,745 19,753 Moderate Fertility Decline 7.256 11,138 16,114 Rapid Fertility Decline 7,256 10,937 14,728 Protein Requirements (in 000 tonnes of protein) No Fertility Decline 89 145 243 Moderate Fertility Decline 89 137 199 Rapid Fertility Decline 89 135 182 Source: World Bank projections based on latest Food Administration Organization (FAO)/World Health Organization (WHO) method for assessing nutritional requirements.a a Depending on sex/age, protein requirement expressed in egg protein varies between 23.7 and 25.2 grams per day and energy requirement between 1,926 and 2,036 kcal/day. 5 It is assumed that three units of paddy produce two units of rice and that 16 percent of the rice is used for non-human consumption.  -15- improve crop yields nationwide, inputs, such as fertilizers and better seeds. as well as irrigation systems and distribution networks, would need to be used extensively. Also. resettling farmers in relatively unpopulated areas of Madagascar entails high costs and is impeded, over the short- and medium-terms by socio-political factors, including intractable land tenure problems. 2.26 The agricultural statistics available may serve to indicate a general order of magnitude. At the present time, about five percent of the country is cultivated. Only an additional nine percent or 5.5 million hectares could still be potentially cultivated as the rest of the area comprises pastures, forests, water areas and uncultivable land. If 40 percent of the remaining area is used for food crop production and taking fallow into consideration, only about two million hectares would remain for growth in industrial and export crops, the country's major foreign exchange earners. Even if reasonable productivity increases can be achieved in food crop production in the future, additional land needed to grow sufficient food for a rapidly growing population would reduce the remaining area for industrial and export crops. Whether the presently uncultivated land can be used for food crop and whether sufficient water is available for the required irrigation, needs to be investigated further. In any case, the required land reclamation would substantially exceed the historic trend (e.g., paddy production grew at less than 0.5 percent a year over the past decade) and would be increasingly time-consuming and costly. Similarly. intensifying the use of existing land requires additional skills, investments and foreign exchange, all of which are in short sttpply in Madagascar. Implications for Fuelwood Requirements 2.27 With only a per capita use of 215 kg of oil equivalent. Madagascar's energy consumption is very low. Yet, although the country possesses an abundant hydroelectric potential and a considerable surplus in its current generating capacity, it faces serious energy problems. In addition to increasing import needs of petroleum for transportation and for diesel-fired electric power generation, over-exploitation of forest resources for fuelwood has assumed crisis proportions. 2.28 About 80 percent of Madagascar's total energy consumptiou is derived from fuelwood and charcoal. The contiguous forest cover, including plantations and mangroves. is estimated to amount to 13.1 million hectares with an manual long-term sustainable yield of 6.5 million tonnes of wood equivalent (twe)6. While gross national demand in 1983 amounted to 5.3 million twe creating an overall surplus on a national basis. large deficits occurred in the highland provinces of Antananarivo and Fianarantsoa, which have more than half of the population of the country but only about 12 6 Tonne of wood equivalent at 25 percent moisture content wet basis (mcwb) per unit weight. twe = 0.31 tonnes of oil equivalent.  -16- percent of the forest cover. Since wood surpluses in other areas cannot economically be transported to these provinces, except for a limited amount from the Toamasina province, the net fuelvood deficit from the highland provinces has been estimated at about 1.9 million twe per year. The consequences are that the existing forest stocks in the deficit areas, and to a serious extent particularly in the hinterland of Antananarivo, are increasingly being depleted. According to the FAO Tropical Forest Resourceg Assessment Project, the annual rate of deforestation is estimated at 150,000 hectares. At this rate, erosion problems and economic and social impact may become catastrophic over the long term. 2.29 Table 8 shows the projected fuelwood requirements on the basis of the 1983 per capita consumption of 680 kg wood equivalent for the highland population and 397 kg wood equivalent for the lowland population. Table 7: PROJECTED FUELWOOD REQUIREMENTS (in twe) 1985 2000 2015 Consumption No Fertility Decline 5,632 9,140 15,419 Moderate Fertility Decline 5,632 8,534 12,199 Rapid Fertility Decline 5,632 8,275 10,876 Surplus/(Deficit) Over Supply No Fertility Decline + 718 (2.790) (9.069) Moderate Fertility Decline + 718 (2,184) (5,849) Rapid Fertility Decline + 718 (1.925) (4,526) Source: World Bank projections. These projections assume constant supply. If shortfalls are, however, not satisfied by other means of energy, the deforestation process would reduce the forest stock and invalidate the constant supply hypothesis. Nevertheless, the table depicts the magnitude of the ensuing deficits under the three population growth assumptions. In the case of no fertility decline, the deficit in the year 2015 would reach 9.1 million twe; i.e., nearly 2 1/2 times the current annual sustainable nationwide fuelwood production. In the case of rapid fertility decline, the deficit in the year 2015 would be only half as large as in the former case but still exceed the annual sustainable fuelwood production by a substantial margin (about 71 percent). 2.30 One theoretical solution of course would be to increase the fuelwood supply particularly in the highest shortfall areas through plantation development and encouragement of rural tree plantations which, even if fast growing and productive eucalyptus trees are used. will have  -17- little impact on the supply situation before ten years hence. The next table demonstrates the acreage needed for additional tree production (eucalyptus) in the two highland provinces. Antananarivo and Pianarantsoa. Table 8: PROJECTED FUELWOCD DEFICITS AND RESMLTANT TREE PRODUCTION AREAS NEEDED Resultant Projected Deficit Acreage ('000 twe) ('000 hectares) Year 2000 No Fertility Decline 4,420 679 Moderate Fertility Decline 3,020 618 Rapid Fertility Decline 3,850 592 Year 2015 No Fertility Decline 8,640 1,328 Moderate Fertility Decline 6,500 1.000 Rapid Fertility Decline 5.620 864 Source: World Bank projections. The table includes only the limited wood fuels transport from Toamasina province to Antananarivo province since other transfers seem economically not feasible. Without fertility decline, the acreage which has to be readied for production in 2015 (1.3 million hectares) would require an 86.4 percent expansion over the present total forest cover of these two provinces. The rapid fertility decline scenario, however, would result in subs;antially less expansion, but still require a 56 percent increase in acreage. 2.31 Clearly, in addition to reduction of population growth, other strategies must also be pursued to overcome the already presently serious energy crisis. To this end, the World Bank has made detailed recommendations.7 But without slowing down significantly population growth, measures such as improved demand management and wood substitution would most likely not be sufficient to resolve this crisis. Implications for Social Services 2.32 As a country's population grows, increased investments and recurrent expenditures are needed for education, health and developing job skills just to maint.iin ezisting levels of development. The implications 7 Report on Madagascar: Issues and Options in the Energy Sector, November 1985.  -18- under the three population trend alternatives for the education and health sectors are illustrated in terms of projected demand for such services and estimated in public expenditures. Education 2.33 In 1975. the Government introduced a policy of free primary school education. The primary course was reduced to five years. and agriculture studies were introduced to make education more relevant to the rural population. As a result. primary school enrollment has increased from 88 percent in 1975 to almost 100 percent at the present time. School attendance is compulsory between 6 and 14 years of age. Table 10 shows the projected number of children entering the compulsory school system in 1985. 2000 and 2015 and the projected total primary school enrollment for children age 6 to 14 years for the same years. Table 9: PROJECTED POPULATION REACHING AGE 6 AND PROJECTED TOTAL COMPULSORY SCHOOL ENROLLMENT (in '000) 1985 2000 2015 Population Aged 6 Years No Fertility Decline 329 526 901 Moderate Fertility Decline 329 454 574 Rapid Fertility Decline 329 423 440 Population Aged 6 - 14 Years No Fertility Decline 2,507 4.113 6,969 Moderate Fertility Decline 2,507 3,708 4.829 Rapid Fertility Decline 2,507 3,534 3,931 Source: World Bank projections. Declines in fertility and infant and child mortality affect future enrollment of primary schools after only a few years. The implications of slower or aore rapid fertility decline are consequently readily evident. With no fertility decline. there would be about 460.000 more first-graders and about three million more compulsory school students in 2015 than if the rapid fertility decline were to take place. The estimated reduction in enrollment under the rapid fertility decline alternative would be greater than the present size of the primary school system. 2.34 A less rapidly growing perulation would produce considerable financial savings in education expenditures. Using IBRD/UNESCO estimates in  -19- constant prices for educational co0tB per student8 end under the assumption Table 10: PROJECTED SIZE OF MCH TARGET POPULATION ('000) 1985 2000 2015 Women AMed 12-49 Years Plus Children Under 5 Years No Fertility Decline 4,567 7,510 12,558 Moderate Fertility Decline 4,567 6,843 9,350 Rapid Fertility Decline 4,567 6,556 8,183 Annual Number of Deliveries No Fertility Decline 452 722 1,194 Moderate Fertility Decline 452 577 671 Rapid Fertility Decline 452 515 464 Delivery Related Days in Health Centers No Fertility Decline 1.017 2,272 4,297 Moderate Fertility Decline 1,017 1,818 2.416 Rapid Fertility Decline 1,017 1,624 1,669 Source: World Bank projections. that the private sector will continue to educate 13 percent of the primary student population, real recurrent educational expenditures without a fertility decline would double to FMG 80 billion by the year 2000 and quadruple to FMG 160 billion by 2015. In the moderate and rapid fertility decline scenarios, expenditures would grow to FMG 69-72 billion by the year 2000 and to FMG 90-110 billion by 2015. The potential annual financial savings between the highest and lowest hypotheses amount to almost FMG 70 billion (about US$110 million). Furthermore, assuming that the Government recurrent budget would increase in line with economic growth at 3 percent per annum. only the rapid fertility decline alternative would keep primary educational expenditures (2.7 percent per annum) in line with the recurrent budget. Rapid fertility decline can thus ease the pressures on the education system by reducing government spending on school construction, materials and supplies and teachers' training. This may facilitate the maintenance of universal primary education and permit improvements in the quality of the educational system. Health Services 2.35 A large part of the health services in Madagascar is devoted to 8 FMG 18,500 in 1985; FMG 22,400 in 2000; and FMG 26,300 in 2015.  -20- mothers and children under five years of age. Estimates of future deminds for maternal and child health (MCH) services provides a strong indication of differentials in volume, health manpower and costs of health services under the various fertility decline assumptions. Table 11 presents such estimates for Madagascar. 2.36 Without fertility decline, the target MCH group would increase from the present 4.6 million to 12.6 million in 2015-a number larger than the current total population of Madagascar. Under the rapid fertility decline assumption, the target MCH group would almost double. Although this would still represent a considerable increase in service requirements, it would be a more manageable increase from the present levels. Annual deliveries would increase without fertility change to 1.2 million in 2015. but with moderate fertility decline there would be only 670,000 in 2015. With rapid fertility decline, there would first be a modest increase in deliveries by 2000. but then a decrease in 2015 reading almost the current level. The latter scenario would realize substantial comparative savings that could be used to improve the quality of health services. The same holds true for the total number of days spent in hospitals and health centers due to delivery.9 2.37 The impact of the projected population growth on total health expenditures in constant terms is shown in the following table. Table 11: PROJECTED HEALTH EKPENDITURES IN REAL TERMS (in billion of PMG) 1985 2000 2015 Total Health Expenditures No Fertility Decline 49.2 79.6 133.9 Moderate Fertility Decline 49.2 74.4 106.0 Rapid Fertility Decline 49.2 72.1 94.5 Government-Financed Health Expenditures No Fertility Decline 19.5 31.6 53.1 Moderate Fertility Decline 19.5 29.5 42.0 Rapid Fertility Decline 19.5 28.6 37.4 Source: World Bank projections. 9 Projections assume hospital/health center deliveries to increase from the present 50 percent to 65 percent beginning in 1995 and to 80 percent beginning in 2010.  -21- These projections exclude inflationary increases and assume that total health expenditures per person remain at the-unsatisfactory-1985 level (US$7.50 per pe=son) and are financed about 40 percent from all public sources. In both fertility decline alternatives the annual growth rates of Government funding could remain below 3 percent (moderate decline, 2.6 percent; rapid decline, 2.2 percent). To maintain the present situation of health services without fertility decline, funding would need to increase by 3.4 percent or above the long-term economic growth projections. This would require that either the share of health expenditures in the national and local health budgets increases from 8.1 percent in 1985 to 21.9 percent in 2015 at the expense of other sectors or that the proportion of Government funding for health care be reduced. Depending on the size of Madagascar's future population, it would be increasingly difficult for the Government to even maintain the present insufficient levels of health services and it would be highly unlikely that health care could be extended and improved adequately if population growth persists at current levels. D. Population and Family Planning Activities General Attitudes 2.38 Except among an educated, urban minority, the predominant view of the people towards having children is that they are a gift of God, bliss to a family and fortune for the country. For an important proportion of the population. children represent social security for old age. For men, they are proof of continuing virility. The traditional wish made at Malagasy weddings is that the couple be blessed with seven sons and seven daughters. In a past survey, carried out in 1974, only seven percent of the wives interviewed in urban centers and in rural areas mentioned difficulties in educating, raising and feeding a large family. Among women with more than four children. 61 pexcent wanted to have more children and only 25 percent preferred no more children. More than half (54 percent) opposed family planning, while 28 percent favored it for health reasons. The husbands almost unanimously preferred large families. While no more recent surveys exist, it appears that the recent economic recession has hastened the slow change in traditional values associated with a large family. Anecdotal evidence for this trend is the increasing number of abortions and the recent rise in aba&doned children. 2.39 Official health statistics, most probably incomplete, give an induced abortion rate of 4.5 per 1,000 women of childbearing age (2.3 percent of live births). Since many abortions are conducted outside the official health system, the numbers may be much higher. According to one reportlO about 8 out of 1,000 women admitted for various illnesses in health 10 "Service de M&decine de Soins" rapport annuel, 1982.  -22- care facilities die as result of abortions. Another studyll carried out in the maternity hospital of Befelatanana reports that one out of six maternal deaths was the consequence of abortions. Probably the most useful study to date on abortion-related questions was carried out by Fianakaviana Sambatra (FISA. meaning "Happy Family")12 comprising 8.331 patients in the Antananarivo and Fianarantsoa proviaces: among them 1.375 (16.5 percent) admitted to having had at least one abortion. 2.40 These figures require cautious interpretation. Compared to sub- Saharan African countries, the problem in Madagascar still appears to be less pronounced. The International Planned Parenthood Federation (IPPF) estimates the clandestine abortion rate in Zaire. for example, to be between 5 and 20 percent of live births. In a large hospital in Zaire, some 40 percent of gynecological cases admitted were found to be the result of induced abortion and 75 percent of maternal deaths were caused by induced abortion. A study carried out at a major urban hospital in Ghana reported that one third of women with one previous pregnancy had terminated it by abortion. Population Policy and Programs 2.41 Madagascar has neither an explicit policy on population nor a formal government family program. Neverthelegs, Government awareness is increasing as evidenced by statements made by the Minister of Population. Social Affairs, Youth and Sports during the 1984 World Conference on Population in Mexico. "In Madagascar we are certainly aware of the difficulty to break the vicious circle of underdevelopment-high fertility, rapid demographic growth, underdevelopment-but it remains that the Malagasy perception regarding development and population problems is based on the conviction that economic and cultural emancipation must have priority over demographic solutions." By January 1985. the concern had increased as indicated in the booklet "Population de Madagascar" published by the Ministry of Scientific Research and Technology for Development: "...it is clear that in the present economic situation, a rapid or even moderate growth of the Malagasy population jeopardizes the country's food self- sufficiency in the year 2000. A population education project financed by UNFPA and executed by UNESCO is implemented by the Ministry of Education. The project is aimed at introducing population issues into the curricula of primary and secondary schools. Finally. it appears that the integration of the population variable amongst the explicit factors of economic and social development is a real necessity for Madagascar." 11 "A propos de la Planification familiale a Madagascar" by M.P. Ramakavelo, 1983. 12 See paragraph 2.45.  -23- 2.42 The following actions on population related matters undertaken by the Government indicate further evolution on this subject: (a) founding of a 12-member parliamentary group on population in December 1984 following the Mexico World Conference on population; (b) a government-approved and UNFPA-funded mission in 1985 to four foreign countries to observe their family health programs of representatives of the Planning Directorate, the Ministries of Health and of Population, Social Affairs, Youth and Sport, and of the National Assembly; (c) a request by the head of the Planning Directorate to United States Agency for International Development (USAID) to finance a Resources for the Awareness of Population Impacts on Development (RAPID) analysis demonstration (a computer-based demographic model) to be presented at a seminar on population and development. planned for late 1986/early 1987; (d) participat.on of senior officials of the Ministry of Health in a regional conference in Rwanda in August 1986 on family health organized by the International Planned Parenthood Federation (IPPP) and financed by the World Bank; (e) gradual recognition and limited support of FISA, the main private sector organization involved in family planning in Madagascar. and (f) the establishment of a Population and Development Unit in the Planning Directorate. with the assistance of UNFPA, to study implicatioxs of demographic growth for macro and sectoral planning and. if warranted, to prepare population policy recommendations. Recommendat.5ons Concerning Population Activities 2.43 If current fertility levels in Madagascar remain unchanged, the country's population will grow over 60 percent by the year 2000, reaching more than two-and-a-half times its current size by the year 2015. At that time, the present population of 10.3 million will have increased to over 28 million. By then, more than 8.5 million or 80 percent of the island's present population are likely to live in urban centers. The issue is not whether Madagascar could ultimately support a population of this size, but the effects of the rate of population growth on development in the medium term. Rapid population growth threatens steady, long-term social and economic development, because it virtually negates economic advances, thus perpetuating unsatisfactory living standards and having critical implications for the adequacy of agricultural and energy production and the financing of the social sectors. 2.44 Government action toward fertility reduction is justified to safeguard the future welfare of society. A second justification for public  -24- support to population-related programs is that with family planning servicee., the Government would increase freedom of choice as well as encourage responsible parenthood. By contrast. inaction on the population issue on the part of the Government would effectively curtail individuals' rights to choose the size of their families and would in the future foster economic stagnation. 2.45 It is therefore recommended that a senior level Government comittee be created to initiate and oversee the study of the population issue in order to propose explicit demographic objectives, develop specific progorms and recommend a clear plan of action for the next five years. The development of a population policy in Madagascar would need to address the following: (a) The capability of the Government to undertake demographic analyses and projections including the collection of reliable data on population size. fertility and mortality; (b) The potential for coordination with UNFPA regarding the assistance to the Statistical Data Bank (Banque des Donn6es de l'Etat) regarding the planned 1986/87 census; (c) Mobilization of sustained Government commitment for proposed population policies; and 'd) Proposals for an institutional framework capable of carrying out the planning and implementation functions of these policies. The United Nations Fund for Population Activities (UNFPA) has identified similar needs in the population sector and proposes an action program containing projects which are directed toward (i) development of health and family planning; (ii) improvement of demographic data collection and analysiso (iii) provision of information and research as basis for political population decisions; (iv) integration of education on population into the curricula; (v) creation of increased communication programs on population; and (vi) integration of population activities into development issues affecting women. 2.46 Madagascar still has a law in its books dating back to colonial times (May 30, 1933) which prohibits the distribution of contraceptives. Although it is no longer enforced, the Government may wish to repeal it as a first active step in the direction of an increasing commitment to a population policy. Family Planning Activities 2.47 The main organization involved in family planning in Madagascar is FISA, a non-governmental organization (NGO) affiliated with IPPF. PISA was founded in 1967 and over time gradually has gained recognition. With its staff of about 60, it has assisted the Ministry of National Education with the introduction of a family education program; the Armed Forces, with  -25- supplies of contraceptives and the Ministry of Finance with contraceptive imports. The Ministry of Population, Social Affairs, Youth and Sports oversees FISA's activities and provides some budgetary support. In 1984. PISA began to provide family planning services in 56 MOH health facilities. 2.48 Overall family planning services are offered in 84 locations, of which 12 clinics are operated by FISA alone; 69 by FISA in conjunction MOH. the Ministry of Health, the military and social centers; and 3 by para- statal companies. Thus, with only one facility for 27,500 women aged 15-49 years, the coverage of services is still very low. The 56 MOH facilities in which FISA provides family planning services represent only about 2 percent of the total MOH jervice outlets, but this represents an important first step by the Ministry in accepting family planning as a health intervention. The women of reproductive age (about 2.3 million) who were using modern contraception in 1985 are estimated at 40,000 (15,000 by FISA, 5,000 by public enterprises and 20.000 by pharmacies); i.e., a contraceptive prevalence rate of about 2 percent. This rate of contraceptive use undoubtedly puts Madagascar very close to the bottom of the list among developing countries, as against Kenya at 18 percent, Zimbabwe at 27 percent and Ghana at 10 percent. 2.49 The proportion of the various contraceptive methods practiced are as follows: oral contraceptives (pill) 51.4 percert; injectable contraceptives 43.4 percent; intra-uterine devices (IUD) 0.6 percent; and condom 4.6 percentl3. The pill and the injectable contraceptive are the most commonly used methods of birth control in Madagascar. This has changed little over the past seven years. Geographic distribution of contraceptive use closely follows the concentration of dispensing facilities in the different regions. Accordingly, with 26 of 81 FISA operated facilities in primarily urban areas of Antananarivo province, 53.3 percent of all FISA acceptors live in this province. Fianarantsoa and Toamasina provinces each share about 13 percent also mostly in urban areas. The rural population has practically no access to information or to services of modern contraceptive methods. The discontinuation rate of FISA clients is said to be low compared to usual experien.-e in Africa. 2.50 Due to the country's past predominantly pronatalist attitudes, FISA had to expand its operations cautiously imposing strict criteria on client eligibility. While these have been gradually relaxed, EISA still requires in most locations that clients present proof of marriage and spousal consent. To obtain services, membership at a nominal fee is required. Contraceptive services are provided against payment, but prices are on average about half of those charged by pharmacies. 2.51 FISA's activities in information, education and communication (IEC) are handled by 16 specialists with audiovisual support. Regular seminars on family planning education are conducted in almost all provinces. 13 A propos de la planification familiale a Madagascar by M.P. Ramevelu, 1985.  -26- PISA's budget in 1985 amounted to FMG 166 million (about US$255,000) of which FMG 90 million (54.2 percent) came from IPPF and the remainder from local sources including contributions in cash or kind from provincial governments. It received a US$21,000 contribution in 1985 from the Association for Voluntary Sterilization (AVS) for training of medical and paramedical personnel. UNFPA also provided assistance in 1985 in the amount of US$142,000 for the purchase of contraceptives, materials, equipment and transport. 2.52 This UNFPA assistance to FISA is part of a two-year (1986/87) project, whose total cost will be US$ 200,000. UNFPA will fund a US$680,000 project with MOH. This program will include services for pregnancy spacing. Under the overall objective of reinforcement of the MCH programs, the project also aims at enhancing the knowledge of medical personnel in MCH/Family Planning techniques and will provide family planning services in 48 additional health facilities. This project took five years to be agreed upon. Its finalization after such a long gestation period is probably a reflection of a changing climate towards family planning as a health intervention. 2.53 The Ministry of Population$ Social Affairs.-Youth and Sports has a multitude of responsibilities including a Department of Population and Social Affairs. The Ministry's organizational chart is included is Annex III (page 4). To date, the Department's resources have been devoted primarily to disaster relief. The demography unit is not staffed to undertake research and make policy recommendations. Institutional capacity would need to be developed to provide information and education to leaders, health workers and the public concerning the population situation and the availability of family planning services. The functions, staffing requirements and job descriptions would need to be reviewed and redefined as necessary so that the Department may play its role in assisting in the implementation of population and family planning activities. This ministry is responsible for the supervision of FISA. 2.54 The Catholic Church operates its own family life education service. the Association of Malgache Christian Union (FTK), whicn promotes natural family planning methods and informs participants (1,000 couples since 1978) about responsible parenthood. Recommendations Concerning Family Planning Activities 2.55 On the basis of the available information on fertility, field visits and meetings with Government officials, the sector mission gained the clear impression that unmet demand for family planning services does ezist. Consequently, immediate efforts should be directed towards strengthening the existing family planning structure and rapidly expanding family planning activities within maternal and child health services to all health outlets. This would. as a first phase, entail a program of PP activities for as large a number of public health facilities as possible. In order to achieve the widest knowledge of family planning services. the Government is recommended to develop IEC programs with the assistance of experienced specialists in  -27- this field. 2.56 With the ezistence of its large health infrastructure, Madagascar is in a favorable position to deliver family planning services immediately throughout its health network in conjunction with maternal and child health programs. However, it will be necessary to (a) develop a family planning program as part of MCH, working out the details of service modalities and supervision; (b) formulate an in-service training program for existing health aides. nurses and physicians; and (c) provide the required supplies and equipment. Furthermore, the respective curricula and practical training programs of medical and paramedical students should be amended to include skills and knowledge on contraceptive methods to reduce the need for in- service training in the future. This type of program is in the process of being implemented with assistance from UNFPA. It is essential to coordinate future projects with UNFPA. III. HEALTh A. Health and Nutrition Status 3.01 Mortality and morbidity data are collected at health facilities. This information gives only a partial picture of disease patterns in the population as a whole. In addition, the quality of data depends on accurate diagnosis as well as on timely and correct reporting by medical personnel in health facilities. The installation of an improved health information system began in 1979; but the system did not become fully operational until 1983, and therefore the first figures available are for the period January- June 1984. Hence, the most recent annual data are for 1982 using the old system. In addition, a few studies and surveys have been carried out, particularly in the area of communicable diseases and nutrition. The available iuformation. although limited. provides nevertheless a broad picture of the levels and trends of health conditions.  -28- Morbidity 3.02 Outpatient morbidity in 1981 were recorded under the following disease categories: respiratory diseases (33.2 percent); infections and parasitic diseases (28.3 percent); digestive diseases (7.2 percent); neurological diseases (5.7 percent); and dermatological diseases (3.6 percent). The residual group of other diseases amounted to 10.6 percent. Detail are given in Annex II, Table 1. The major causes of morbidity in the same year were the same year were acute disorders of the upper respiratory tract, intestinal infections, bronchitis, digestive diseases, flu and malaria (see Annex II. Tables 2 and 3). One third of the patients were less than five years old. Most of these children, weakened by nutritional deficiencies, were affected by infectious or parasitic diseases (72.8 percent). of which many could have been avoided by preventive care. 3.03 Significant variations in incidence and causes of morbidity exist in the six provinces (see Annex II. Table 4). Fianarantsoa appears to have the least morbidity incidence of all five of the leading causes. Respiratory diseases have the highest incidence in Antananarivo province while infectious and parasitic diseases are most prevalent in Toamasina province. These geographic variations are in part due to climatic differences-the colder highlands causing more respiratory diseases and the more humid conditions in the coastal provinces being more conducive to infectious and parasitic ailments. 3.04 Malaria, accounting for 24 percent of infectious and parasitic diseases, is a major public health problem in Madagascar primarily due to inadequate vector control measures and insufficiency of antimalarial drugs. The sector mission estimates the annual incidence of malaria in the range of 8 to 11 percent compared with 19 percent in Zambia and 8 percent in Ethiopia. About six percent of the population suffer from schistosomiasis with a high potential of further spread of the disease. With regard to sexually transmitted diseases, particularly syphilis is on the rise with Toliary province recording a disturbing 41 percent of all cases reported in 1984. Antananarivo province also reports consideraDle increases in sexually transmitted diseases. Tuberculosis. leprosy (there are 30 leprosariums and about 30,000 cases in the country) and plague (about 500 to 1,000 cases per year) are endemic in Madagascar and present a major public health concern to the authorities. Annex II provides details on these major communicable diseases in Madagascar with mission estimates on disease incidence and prevalence and also describes the lack of environmental sanitation which constitute a major health risk. Mortality 3.05 As described in Annex II of this report, the CDR has been estimated at 15.5 per 1,000 population, life expectancy at birth at about 51 years for both sexes, and infant mortality rate at 125 per 1,000 live  - 29 - births. The leading causes of mortality in 1982 were intestinal infections (12.2 percent), bronchitis (7.5 percent), malaria (5.1 percent). measles (4.9 percent) and bronchopneumonia (3.9 percent). 3.06 The health facilities' records show that in 1982 45 percent of the deceased were under five years of age with about half of the victims under one. Infants died primarily from infectious and parasitic diseases (31 percent). respiratory ailments (24 percent) and perinatal disorders (19 percent). Over the period 1978 to 1984, fetal mortality remained relatively constant at about 30 per 1.000 births. However, both neonatal mortality (from 5.8 to 8.6 per live births) and maternal mortality (from 140 to 520 per 100.000 live births) increased considerably. This development is particularly disappointing since over the same time span, the primary health care network (CSSPs) was substantially expanded with the specific aim of improving health conditions of mothers and children. However, the system was severely plagued by lack of resources and skilled personnel, and it was thus unable to properly carry out the expected responsibilities. particularly with respect to preventive care including MCH. Hence. vaccination coverage (see para. 3.44 and 3.45 for details) is still low (vaccinations against measles began only in 1985); vector control efforts are inadequate, and malaris prophylaxis is limited by chronic shortage of supplies; knowledge of and access to oral rehydration salts is not widespread family planning services are just beginning; and nutritional deficiencies are widely prevalent and lower the children's resistance to infectious and parasitic diseases. Nutrition 3.07 No comprehensive surveys on nutritional statue have been carried out to date on a nationwide scale in Madagascar. Several surveys as well as clinical observations in pediatrics provide indications that the nutritional status of Malagasy children is poor. Although malnutrition does not appear in the health statistics as a primary cause of morbidity in children under five years (18 per 1,000 against 56 per 1,000 for measles or 23 per 1,000 for whooping-cough). and according to a 1984 UNICEF analysis, it is directly responsible for less than five percent of child mortality. it is likely to be an underlying factor in deaths due to infectious and parasitic diseases. 3.08 The preliminary results of a nationwide survey in 1983. sponsored by WHO, of 1,855 children under two years of age. equally from rural and urban areas, showed "moderate wasting" (evidence of past malnutrition) in 31 percent of the children; i.e.. 31 percent weighed less than 80 percent of the normal weight for their height. The survey also revealed that 14.4 percent of the children had low birth weight (less than 2,500 grams). Other results from this survey, particularly on current malnutrition, are not yet available. A June 1984 survey of 1.000 children up to five years of age in the Antsirabe region, one of the areas most affected lby malnutrition, reveals a high degree of "stunting," reflecting past, chronic and long-term conditions of undernourisbment. Eighty-four percent of the boys and 50  - 30 - percent of the girls showed moderate stunting; i.e., were 90 percent of the normal height for that age. Moderate malnutrition, less or equal to 80 percent of normal weight-for-age, was found among 34.1 percent of children. Usually, malnutrition rises sharply after one year of age; i.e., during weanirg and from the age of two, seven out of ten children are below normal weight-for-height. 3.09 In the case of Madagascar. severe malnutrition occurs less frequently than in African countries. but stunting appears to be exceptionally high, pointing to chronic, long-term food scarcity. The prevalence of malnutrition is also higher in large families with more than six children (35 percent of the families of that size are affected, compared to 27 percent of the smaller families, with fewer than five children). Moreover, birth order appears to affect nutritional status of the infant; in 36 percent of the births of sixth or higher parity the babies showed nutritional deficiencies, compared to 29 percent of infants from first to fifth order of birth. 3.10 The monitoring of about 45,000 children aged between five months and five years by the Catholic Relief Service (CRS) over a six-month period (December 1984 to May 1985) shows similar results (see Annex II for details). The situation appears to have worsened with the deterioration of etonomic conditions, as indicated by hospital admission records. While, for example, approximately six percent of the children attending the pediatrics department of Befelatanana General Hospital showed some degree of malnutrition in 1978. rising to 9.5 percent in 1984, at the present time virtually all children under six show signs of malnutrition and 15 percent of them are hospitalized for severe malnutrition. Also, the Red Cross in Antananarivo reports a sharp increase in the last year in the number of children requiring supplementary feeding. 3.11 Fortunately, breast-feeding remains the primary method of feeding babies, practiced by 92.7 percent of mothers in urban centers and 98.7 percent in rural areas. On average, breast-feeding extends over 16 months with food supplements consisting primarily of rice given at around four to five months. Thus, the babies' food requirements are reasonably well met during the first year. However, as babies are weaned, protein-caloric consumption drops significantly and causes widespread primaryl4 malnutrition among children. Only in the cattle region of the South is cow milk fed to children. Elsewhere, it is too expensive and often unavailable. Consumption of most animal protein, including eggs and fish, is restricted by behavioral factors (Fady). Because of its value as a cash crop, fruit does not form an important part of children's food intake. 14 Due to dietary inadequacies.  - 31 - eacommendations Concerning Nutrition 3.12 This report deals with nutrition only to the extent that nutrition affects Madagascar's health status. Despite weak statistical data, the sector mission's findings seem to indicate that malnutrition in the country is high and getting rapidly worse in the wake of severe economic adjustment measures. Poor feeding practices of children and inadequate access to f3od by the poor appear to be the two key areas which need to be addressed. In order to understand the underlying constraints. it is recommended that a thorough nutrition study be carried out as soon as possible. This study would review the causes for the present situation; determine on the basis of a food balance the past and future trends in the nutritional value of the diet; make proposals for upgrading nutrition status data; and develop a nutrition plan including the required institutional and financial arrangements. Moreover, the surprisingly high degree of infant malnutrition calls for close monitoring and study. Such a study would. inter alia, design a weaning strategy to promote improved breast-feeding and weaning practices. Micronutrient Deficiencies 3.13 Few data are available on vitamin and mineral deficiencies in Madagascar. Iron deficiencies found during clinical examination of pregnant women have not been quantified. Food fortification programa are not carried out on a broad scale since little processed food is marketed. There is no evidence of Vitamin A deficiency and data on iodine deficiency are not available. In terms of malnutrition and vitamin and mineral deficiencies, the statistical information is fragmented and incomplete. A comprehensive study would be necessary to understand the extent and main determinants of malnutrition and to devise appropriate intervention strategies and programs. To this end, the Central Nutrition Laboratory Department, the Human and Dietetic Nutrition Division of the Ministry of Health, the Ministry of Agriculture and the Ministry of Scientific and Technological Research for Development have expressed interest in further research in these areas. Their efforts are to a limited extent supported by UNICEF but additional technical assistance and funding are required to carry out studies. B. Health Policies and Strategies 3.14 Even before the 1978 International Conference on Primary Health Care in Alwr-Ata, USSR, Madagascar had already placed great emphasis on the creation of a primary health care network. Since 1975, the Government's health policies have accorded priority (a) to the development of rural village-based health services in order to reduce the urban/rural health care imbalance in access to modern health care services; and (b) over the long term, to preventive over curative services. Accordingly, the primary health care network consisting of sanitary posts, nursing posts. maternity posts and primary health care centers (CSSP), has been built up to 1.950  - 32 - facilities. Also, as described later in this chapter, the number of health personnel was raised from about 3,900 to over 8,200 from 1975 to 1984, which substantiaily increased the relative share of personnel costs within the health budget (to 70 percent) and consequently resulted in a massive reduction in availability of non-personnel related resources (including pharmaceuticals) as overall health expenditures decreased in real terms. 3.15 As part of its 1986-1990 Development Plan, the Government restated its policy to further pursue its efforts in favor of both the rural population and preventive health activities. In order to complete the network of primary health care centers by 1990, 100 additional health aides (aides sanitaires) per year are scheduled to be trained. To improve management and supervision of these centers, further decentralization is envisaged by creating medical subdistricts. The objectives of the preventive program essentially remain unchanged focussing primarily on vaccination (DPT, Polio and BCG) to arrive at a "satisfactory" coverage; health education; maternal and child health care; combatting against epidemics (malaria, tuberculosis. leprosy, schistosomiasis and plague); and safe water supply and sanitation facilities. On the curative side, the Plan seeks to reinforce the secondary level (surgical hospitals, secoindary hospitals and medical centers) to achieve an improved referral system for the primary health care network. For the central health level (general and principal hospitals and specialized facilities), the Plan calls for achieving an adequate level of "software," i.e., rehabilitation or acquisition of specialized medical equipment. To achieve the objectives set out in this Plan, the Ministry of Health has developed a program focussing on: (a) Strengthening of human resources and management at various levels in the health sector by (i) adding specialized positions for human resource management (finance, logistics, pharmaceuticals, etc.) and for physical resource management (planning and programming, project implementation, supervision and evaluation); (ii) refresher training of all medical and para-medical personnel; and (iii) development of specialists in clinical practice, public health and research. (b) Reinforcement of health infrastructure and equipment by (i) construction of new health facilities at the tertiary level, primarily for the surgical hospitals; (ii) rehabilitation and reequipment of the provincial hospitals; and  - 33 - (iii) provision of technical equipment, where appropriate, at various levels. (c) Improvement of transport and logistics at all levels; and (d) Increase in supply of medicine by (i) increase of local pharmaceutical production; (ii) local manufacture of blood serum and vaccines; and (iii) establishment of an essential drug list. 3.16 While the Government correctly gives high priority to completing the primary health care network and emphasizes in its bealth programs the improvement of management and supervision and the rehabilitation and maintenance of existing facilities and equipment, the program does not outline targets, lacks adequate project justification and fails to set priorities among the proposed interventions. It also ignores, in certain respects, the health system's absorptive capacity and financial constraints. Recommendations Concerning Health Strategy and Medium-Term Plan 3.17 Development of a health strategy and medium-term plan (five years) that goes beyond the annual budget preparation are basic prerequisites for the strengthening of Madagascar's health sector. A management/forward planning approach needs to be introduced at all management levels of MOH. To this end, the newly created Department for Studies, Planning and Programing in MOH should be brought to full operational status as quickly as possible, and its personnal would need to be qualified in health sector and management planning. A staff development program on management techniques with special regard to planning, monitoring and evaluation would need to be designed and implemented. 3.18 More specifically, a cohesive health strategy should include health sector objectives based on desireable and feasible morbidity and mortality improvements and it should provide guidelines to allocate available resources responding to the epidemiological situation. The strategy should also define the roles and specify the extent of coordination between the main health care providers; i.e.. central and provincial governments, village communities, and NGOs. The mechanisms for coordinating and utilizing foreign assistance should also be defined as part of the health strategy. The medium-term plan should be developed in accordance with the health sector strategy, specifying health targets and geographic and functional distributions of recurrent and capital expenditures and establishing priorities on the basis of demographic. epidemiological and operational information as well as on the basis of cost-effectiveness considerations. The plan should be approved at top government levels.  - 34 - updated annually on a rolling basis, and it should serve as basis for the annual budget preparation. C. Health Sector Oraanization and Management Institutional Framework 3.19 The major providers of Madagascar's system of health services are the MOH, provincial and communal governments, missions, several foreign funded and operated health facilities. NGOs, private sector (primarily in the form of drug and pharmaceutical manufacturers) and traditional practitioners. By far the most important provider of modern health setvices is the MOH. It formulates and develops the country's health policy and planning, and provides the overwhelming share of health services at all levels throughout the country. The Minister is assisted by a Secretary General and a number of technical advisors. 3.20 There are five departments in MOH. each headed by a directot A department consists of between three and five services and divisions. The Department of Health and Medical Services is essentially responsible for curative medicine, and includes services for health care, communicable disease control, border health control, health and demographic statistics and personnel training. The operation of the two main hospitals (the so- called general hospitals) and several specialized health centers in Antananarivo is also supervised by this department. Preventive health activities fall under the responsibility of the Department of Community Health through its services of maternal and child health care, immunization, sanitation and hygiene, and a central nutrition laboratory. Procurement and distribution of drugs, pharmaceuticals and technical supplies for primarily the public health sector are under the Department of Pharmacies and Laboratories. This department also coordinates the laboratories of the Ministry. The Department of Studies, Planning and Programming prepares health plans and investment programs, and carries out special studies; in addition. it supervises programs and also supervises externally-financed investments. Finally, the Department of Administration and Financial Affairs handles the entirety of the Ministry's support activities including personnel, logistics and finance. Annex III (page 1) includes the current organization chart of the MOB. 3.21 In each of the six provinces, all MOH clinical services and their administrative support are under the direction of the Chief of Provincial Health Services who reports to MOH's Secretary General. These provincial chiefs have supervisory authority and responsibility over all staff. activities and public and private health facilities within their provinces. Their budgetary authority extends to all public sector health facilities with the exception of principal hospitals and surgical hospitals which enjoy budgetary autonomy. The provincial chief also serves as the director of the Provincial Paramedical School. Support staff at each provincial  - 35 - headquarters typi:cally number 70 to 80 and include about 3 to 5 professionals. Annex III (page 2) depicts a sample organization chart of a provincial health headquarters. 3.22 The provincial health service regions are in turn divided into 30 medical districts (circonscriptions m6dicales). The medical districts do not necessarily correspond to the official administrative districts (Fivondronampokontany). Depending on a number of factors such as population density, homogeneity of population and accessibility of health facilities within the medical districts, they may embrace one or several administrative districts. The medical inspector, always a physician who heads the medical districts, is responsible for the operation of typically about 50 to 70 health facilities which serve a population of between 200,000 to 500,000 support staff. Annex III (page 3) shaos a representative organization chart for a medical district. As already mentioned, due to the expanding responsibilities of the medical inspectors in overseeing the increasing number of health facilities in their medical districts as well as administering such services as mobile vaccination teams, anti-vectorial and communicable disease programs. ambulatory services. sanitation brigades. record keeping and health aides training, the Government is in the process of creating medical subdistricts. It is not contemplated at this time to give these subdistricts budgetary autonomy. ManaRement 3.23 MOHts Minister and some of his top managers at both the central and provincial levels have been in office for a long period of time, particularly in comparison to many other countries in the region. Within MOH and by colleagues outside the Ministry, they are respected for their high level of competence. The Ministry of Health has expressly stated the importance of good management as the key requirement for Madagascar's health system. 3.24 The organizational structure and lines of authority of the MOH are well defined and provide an adequate institutional framework for delivering health services to the population. The overall departmental alignments at the central level appear to be sound. Inter-departmental communications atc the central level are satisfactory as most directors and division chiefs have worked together for many years, channels of commuwication are well established and procedures are generally followed. Within the departments, however, there is a proliferation of small organizational units: departments are divided into services which are again divided into divisions. While the large number of hierarchical levels may be justified at the central level, the same breakdown is also maintained at the provincial and medical district levels. Especially at the district level, this can lead to particularly cumbersome organizational structures where there may be as many as 15 organizational units, each employing an average of two persons and most headed by a clerical staff.  - 36 - 3.25 The delegation of considerable maagement authority to the chiefs of provincial health services i8 reasonably effective. A potential weakness in the organization structure could be the dual line of responsibility of the personnel in the health facilities since they report to the provincial chief s on administrative matters, but on professional matters, they are supervised by the central staff of the Department of Health and Medical Services. However, this form of matriz management is not uncommon in the health sector and both supervisors and supervised staff appear comfortable with this system. 3.26 Management weaknesses at lower levels are major contributors to the poor provision of health services. Supervision of the primary health care facilities by medical district or provincial administrators is weak and often nonexistent. Long distances, difficult terrain and frequent severe shortageb of transportation, commuication and materials only explain in part this problem. lnsufficient quality and poor motivation of some provincial and district administrators, coupled with their inadequate training in management, are also contributing factors. These weaknesses in the field are exacerbated by the absence of tools for good management, such as job descriptions, performance targets, nd an adequate incentive system (para. 3.46-3.67). Recommendations on Proram Changeis 3.27 As future spending on health services will most likely be restrained by continued budget stringency. tho Government could take advantage of a number of cost-effective program changes in health care provision. This would include assigning higher priority to preventive interventions such as 4imunization programs, pre- and post-natal care, vector control, health education, and at the same time strengthening simple curative care as well as establishing an essential drug program. In most of these areas, increased emphasis on outreach activities and the development of community-based programs can yield substantial returns over passive. clinic-based health care as is currently almost exclusively practiced in Madagascar. An added benefit of developing a strong outreach system is that it also can serve as a distribution basis for family planning which has proven to be very effective in many African countries to increase contraceptive prevalence. Management Information/Health Statistics System 3.28 With UNFPA assistance, MOH introduced a new integrated health statistics system in 1983. following a test period in a number of medical districts and a major training effort of staff at all levels of the health system. Based on standardized registers differentiated by type of consultation (prenatal, postnatal, general) and health facility, each health center sends a monthly report to the district headquarters where the reports are manually consolidated into a district report. District reports are in  - 37 - turn manually consolidated iuto provincial reports which are submitted to the Health Statistics Service at the central level for publication of a national annual report. Financial or management information is not included in this reporting system. Due to a host of technical and administrative difficulties, no annual report has been produced since 1982. The publication of the 1984 report is scheduled for mid-1986. There continues to be massive underreporting: at least one-third of all health institutionsl5 ,ail to meet this reporting requirement. Thus, the benefits of the new health statistics system have yet to be assessed. Nevertheless, in the sector mission's opinion. the system when fully operational has the potential of contributing significantly to a management information system. In at least one province, the ceief of provincial health services bas already begun to publish epidemiological analyses of the data in a monthly newsletter to all staff in his region. 3.29 When the reporting and completion delays are reduced to reasonable levels and the data begin to provide a more accurate picture of the health situation in Madagascar, decisionmakers should be able to set viable quantitative objectives for performance of the health system. What is then required is the development and incorporation of financial and management indicators for monitoring and assessing perfortance against objectives at national, provincial and district health facility levels. Recommendations Concerningthe Health Statistics System 3.30 The new, integrated health statistics system will be an important complement to both planning and management. However, at the present time, the system is cumbersome. untimely and suffers severely from underreporting. The present system does not contain cost information or performance targets. Since the system could provide comprehensive and reliable information if properly amended, it is recommended that a review is made of the progress and results achieved so far and proposals are prepared as to how the deficiencies could be corrected. Specialist inputs will be needed for this task. It is further recommended that these specialists identify data collection needs which a health statistics system cannot handle adequately; e.g.. non-facility related epidemiological and nutritional status data. D. Distribution and Utilization of Health Facilities 3.31 The Government at the central, provincial and medical district level operated 2.082 medical facilities in 1985; i.e., 878 more than it did 15 Indicative reporting rates are: 58% of CSSP, 64% of secondary hospitals, 68% of surgical hospitals, and 75% of maternity hospitals.  - 38 - in 1978. The structure of this system can be depicted as a pyramid with the complexs.ty of medical interventions increasing toward the top (on the following p.ge). 3.32 The lowest level of facility-based care comprises 1.904 basic health care facilities of which 1.323 are "primary health centers" (centres de soins de sante primaire-CSSP). A primary health center is staffed by a health aide with five years of general education and 14 months of specialized training. The villages nominate a health aide from their midst and construct and maintain the center, a delivery room and a two-room house. The Government pays the salary and provides medical supplies and medicine. Since 1978. when 476 primary health centers were available. 120 additional such centers were added to the Vstem each year. In contrast. the number of the other types of facilities of the basic health care system-340 "health posts" (postes sanitaires), 144 nursing posts (Dostes d'infirmiers) General Hospital Provincial Third Tier Hospital Medical/Surgical Hospital Simple Secondary Hospital Second Tier Medical Centers Health Posts, Nursing Posts, Maternity Posts. Primary Health Centers First Tier The distribution, number and development of these facilities are given on page 1 of Annex IV. and 97 maternity posts (postes d'accouchements)-increased only marginally over the period 1978-85. These facilities are staffed by a nurse/ paramedical and/or midwife who must have a minimum of nine years of general education and three years of specialized training. There is little difference in the type of service offered by the four different types of basic outpatient and inpatient health care, namely simply curative treatment, vaccinations, and pre- and post-natal consultation and child  - 39 - deliveries. Frequently, no more than one day per month is devoted to outreach activities for health and hygiene, education, house visits or commnity health involvement. On average, the health aide of a primary health center handles about 150 consultations per month, the paramedics of the other facilities see about 400 patients monthly. Few patients are referred to facilities at a higher level (5-10 per annum). 3.33 While buildings are generally in good condition, medical equipment is often obsolete or in disrepair, technical manuals and educational aides missing and transport unavailable except for the odd bicycle provided by UNICEF. Operations of all facilities are, however, most affected by the chronic shortage of medicine and vaccines. They are supplied in insufficient quantities semiannually and often are depleted already after one to three months. In these cases, the patient receives a prescription which can be filled at the village drug dispensary or pharmacies. Although quite a large number of such private dispensaries (1.428) and pharmacies (74) exist in Madagascar. which usually are in close proximity to the health centers, prices for phamaceuticals are very high: officially 1.6 times the equivalent price in France, and often, contrary to regulations, inflated substantially more by the rural dispensaries. As a result, the poorest segment of the country's population must, when seeking modern health care, depend almost exclusively on medicine from the public health facilities. In addition to lack of resources, the sector mission found supervision inadequate to maintain and develop staff skills and to foster staff morale (visits often no more than once a year). Inspections are hampered by lack of transport, constraints on vehicles operating budgets, lack of roads (many centers are 10 to 20 km from the nearest road) and long distances to referral centers. Regular in-service training is not provided at any level. 3.34 The second tier of health service consists of 99 "medical centers" (centres m6dicaux) and 58 "simple secondary hospitals" (h8