Uganda - Population health and nutrition sector review

Report No. 5262-UG Uganda Population, Health and Nutrition Sector Review June 17,1985 Population, Health and Nutrition Department FOR OFFICIAL USE ONLY ,,~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ . - . , -- .-- ,- 1.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. Documert of the World Barnk This document has a restricted distnbution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorizatiorL .-~-  FOR OMCUIL USE ONLY Abstract With unabated growth, Uganda's current population of 14.3 million - would grow to 40 million by 2015. The costs of rapid growth, in terms of added government services and diverted resources, are a constraint to development. Projections demonstrate that these costs could be substantially reduced through a fertility decline. The most critical population issue is the development of sufficient public support to mount an effective population program. A start to increase family planning services can be made through the FPAU and NGOs, but rehabilitation of the health delivery infrastructure will greatly enhance program development. Deterioration in government services has contributed to declining health status, but the data base is insufficient to establish the morbidity and mortality levels accurately. Revamping of the health delivery system is a first priority. Effort should concentrate on the restoration of facilities, establishment of a reliable pharmaceutical distribution system, improved planning and management, increased support of recurrent costs and rationalization of staff establishment and remuneration. The PHC plan is too complex and extensive and should be revised in scope. Cost recovery is especially important as a revenue measure for government services. The viability of cost recovery is demonstrated by the current financing of NGO activities through the use of fees and charges for drugs. Protein calorie malnutrition is primarily an episodic problem arising from local agricultural and social conditions. Improved marketing and distribution of foodstuffs is needed. Additionally, effective agrometeriological surveillance should be reinstituted, particularly in high risk areas, to provide early warning of crop failures. Contingency planning for food failures should be carried out under the Food and Nutrition Council. Ihb documnt has a resicod distbution aod nay be used by re:ipien only in the perfonnce of tbe oflirld duti. Its contens may not othewi be disdosed witbout Wold Dank au1thortion.  UGANDAL POPULATION, REALTH AND NUTRITION SECTOR REPORT TABLE OF CONTENTS Page No. BASIC DATA. ..........................................-i ABBREVIATIONS ............................********************* ii GLOSSARY ......................................***............. iii-iv SUMMARY .........................*...... v-ix I. ECONOMIC AND SOCIAL ....... ........... I II. POPULATION 2.................................... ..... 2 A. Present Demographic Situation and Trends .... ...... 2 B. Population Projections .... o.......... ...0...... 4 C. Development Implications of Population Growth . .... 5 D. Population and Family Planning Policies ........... 8 E. Family Planning Services ..ooo ...o............... 9 III. HFALTH ..................................*................. 11 A. Health Status .............................. * ...... 11 B. Health Policies ...........*.. ..................... 12 C. Organization and Management ............. .......... 14 D. Del-ivery of Health Services ............... *........ 15 E. Health Personnel and Training ........... .......... 21 F Health Support Systems . ........................... 26 G. Health Expenditures and Finance ................... 28 IVo NUTRITION 3.......eo 34 A. Current Nutritional Status ........................ 34 B. Causes of Malnutrition .................. .......... 34 C. Nutrition Activities and Policies ................ 36 V. ISSUES AND RECOMMENDATIONS ....... .38 A. Population ...... 38 B. Health 40 ANNEXES ......................................................... 1. Population ........50..............0.................50 2. Health Statistics ........2.......52 3., Health Financing ... 59 4. Nutrition ....69 This report is based on the findings of a PHN mission that visited Uganda in October 1983. The mission was composed of Graham C,larkson (public health specialist, mission leader); Howard Barnum (economist); and John Kevany (nutritionist). The team was assisted in Washington by Ms. Altbea Hill (denographer). The report was written by Howard Barnum with the assistance of Althea till for the population section.  UGMADh BASIC DATA a/ Population: 14.3 million (1984, estlmted) Annual Rate of Population Growth: 3.2S (1984, estimated) Population Projected for Year 2000: 25 million Dependency Ratio: 101.2 GNP per capita (1981): US$220 b, Population Density (per KR2): 64 Crude Death Rate: 19 Crude Birth late: 50 Infant Mortality Rate: 120 Life Expectancy at Birth: 50 years Total Fertility Rate: 7.0 a/ All data estimated for 1984 unless otheivise indicated. Source of population data is the World Bank, World Development Report Projection. b/ PIn 1981 US$-  - ix - ABBREVIATIONS AIRS Administrator of Medical Services BCG Tuberculosis Inoculation CEM Country Economic Mnemorandum (World Bank) CMS Central medical stores (Government) CPS Contraceptive Prevalence Survey DANIDA Danish Agency for International Development DM0 District Medical Officer DPT Diptheria, Pertussis and Tetanus Im=mization EN Enrolled nurse EPI Expanded Program on ITnunization FP Family Planning FPAU Family Planning Association of Uganda GDP Gross Domestic Product GOU Governmnt of Uganda IEC Information, Education and Commmnication DME Infant Mortality Rate INTRAR International Training for Health IPPF International Planned Parenthood Federation IUD Intrauterine Device JmS Joint medical stores (NGOs) MA Medical assistant MCH Maternal and Child Health MD Medical doctor NOR Ministry of Health HOLG Ministry of Local Government NGO Nongovernmental Organization ORS Oral Rehydration Salts ORT Oral Rehydration Therapy PCM Protein-calorie malnutrition (PCH) PHC Primary Health Care PHCW Primary health care worker RN Registered nurse REP Revised Recovery Program TBA Traditional Birth Attendants TFR Total Fertility Rate UNDP United Nations Development Program UNESCO United Nations Educational, Scientific and Cultural Organization UNFPA United Nations Fund for Population Activities UNICEF United Nations Childrens Emergency Fund UPL Uganda Pharmaceuticals Ltd. USAID United States Agency for International Development WDR World Development Report (World Bank) WHO World Health Organization  - iii - GLOSSARY Crude Birth Rate: The number of births per 1,000 population in a given year. Crude Death Rate: The number of deaths per 1,000 population in a given year. Classifications of Malnutrition: Two major classification systems are currently used to describe the severity and the type of malnutrition among children. The Gomez classification system identifies 3 increasingly severe degrees of malnutrition by comparing weight-for-age to a reference standard. Those children whose weight-for-age falls between 75%-90Z of the reference median are mildly malnourished (10); those who fall between 60-75Z of the median are moderately malnourished (20); and those whose weight-for-age fall below 60Z of the median are severely mlnourished (30). The Waterlow classification differentiates between chronic and acute malnutrition, based on measures of weight-for-height and height-for-age relative to reference standards. Children with relatively low weight-for-height, but adequate height-for-age are wasted, the result of acute, or short-term malnutrition; adequate weight-for-height, but low height-for-age indicates stunting or chronic, long-term malnutrition; children with low weight-for-height and low height-for-age are concurrently wasted and stunted. Dependency Ratio: The ratio of the economically deperndent 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).  - iv - IWfant Mortality Rate: The nuuber of deaths of infants under one year of age in a given year per 1,,OOO live births in that year. Life Zxpectancy at Birth: The average number of years an infant will live if the current age/sex-specific nortality trends prevailing at the time of birtb were to continue. Morbidity: The incidence of disease in a population. Mortality: The prevalence of death in a population. Protein-calorie Malnutrition: A clinical spectrum of suboptimum nutritional states ranging from nild undernutrition and growth retardation to the most severe states of kwasbiorkor and uarasuns. Rate of Natural Increase: The rate at which a population is increasing (or decreasing) in a given year due to a surplus (or deficit) of births over deaths, expressed as a percentage of the total population. Rate of Population Growth: The rate of natural increase adjusted for (net) migration, and expressed as a percentage of the total population of a given year. Total Fertility Rate: The average number of children that would be born alive to a woman during er -lifetime if she passed through her childbearing years conforming to the age-specific fertility rates of a given year.  UGANDA SUMMARY 1/ A. Situation and Issues Population 1. Based on an analysis of the 1969 census and the estimated current age structure and fertility rate the current population growth rate of Uganda is nearly 3.2% per annum. With unabated growth, the estimated 1984 population of 14.3 million would grow to 40 million by the year 2015. Given the good agriculture potential, Uganda could acco modate this population growth if consumption were to be maintained at the current subsistence levels. However, the rapid rate of population growth will impede attempts to move beyond low agricultural productivity by diverting investment funds needed for agriculture, job creation, and long-term economic recovery to consumption and social infrastructure needed to support a larger population. The rate of current population growth rather than the ultimate population size, is a constraint to development; population growth needs to be slowed if the standard of living of the Ugandan people is to recover and then increase in the future. 2. Although family planning (PP) was introduced In Uganda in the early 1950s and the Family Planning Association of Uganda (PPAU) was formed in 1957, FP programs were neither recognized nor given assistance by the Government or political leaders until the early 1970s. Program support has never been strong due primarily to lack of interest among key political leaders until recently. In 1981, the Government of Uganda adopted sn explicit population oontrol and family planning policy and outlined its intentions at the UN conference on the less developed countries. The objective of its plan was to reduce the population growth rate, currently estimated to be 3.2% per annum, to 2.6Z per annum over a 15-year period (1981-1995). A strategy was outlined to achieve this objective. The main components of the strategy include integrating FP services in all government And missionary health facilities, including rural units, integrating population education Into all levels of formal and non-formal education, raising the legal age of marriage, and ftstering women's development with an emphasis on formal education. Starting In fiscal year 1985, this strategy will begin to be implemented. 1/ During the last fifteen mxnths there has been an increase in government and donor activities in the population and health sectors. The findings in the report do not reflect these recent changes.  -vi- 3. The most critical population issue is whether sufficient public support can be cultivated to implement an effective population program. The current contraceptive prevalence rate of about 1% is far from the level necessary to achieve the government goal. To reduce the growth rate to the government target of 2.6Z in 1995 set by the Government would mean that the prevalence rate would have to rise te more than 14Z. Constraints in implementing a government program at this time, however, are substantial. The constraints include not only a lack of public interest in population policies, but also a lack of sufficient supporting health infrastructure, lack of trained manpower and poor logistical support. Health 4. There has been a partial collapse of health services in Uganda as a consequence of the events of recent years. Because of the disruption of health services caused by the civil disturbances, country-wide vital statistics are unavailable and information on disease patterns and deaths is limited. However, morbidity and mortality have both clearly risen, although the extent is difficult to quantify. Shortages of drugs and other supplies have substantially reduced the effectiveness of health services. As government facilities have deteriorated there has come to be an increasing reliance on non-government organizations, now operating well beyond design capacity, to provide essential services. 5. With the deterioration in health services, the breakdown in the itinization program, the deterioration of safe water supplies and sanitation, and the increase in social instability, total and infant mortality rates are estimated to have risen in recent years. lnfant mortality is estimated to be substantially in excess of 100 per 1,000 live births. The crude death rate reported at approximately 15/1000 total population in the mid-1970s might well be close to 20/1000 now. Impressions from health workers and the limited data from hospitals indicate measles, gastro-enteritis, malaria, pneumonia and neonatal tetanus to be the most common causes of deaths in infants and young children. 6. The Revised Recovery Programme for 1982-84 and the National Plan of Action on Primary Health Care outline selected health programs but neither of these documents fills the need for a coherent strategy for long term development within the health sector. The Revised Recovery Programme, while containing a logical series of actions that need to be supported, addresses only the most pressing short term needs. The PHC plan, while providing a long term perspective, may be unworkable and does not address the requirements of the entire health sector. 7. The Revised Recovery Program includes projects for minimum rehabilitation of facilities, development of primary health care, accelerated immunization, and strengthening of health training and NOR  -vii- planniug capabilities. The projects are all modest in scope due to the stringent economic conditions and the required investment in all sectors. It is important to note, however, that during the first two years of the recovery program, the health sector projects were not fully iiplemented and were neglected relative to other sectors. The primary constraints were insufficient donor funds and a lack of managerial capacity in the sector. 8. The National Plan of Action on Primary Health Care is under revision and a final version is expected to be available In 1985, but the basic elements of the original plan, adopted in September 1983, are expected to remain unchanged. The plan embraces a wide range of curative, preventive and rehabilitative activities that include malaria control, immunization of mothers and children, treatment of diarrhea with oral rehydration, a school health program, family planning, environmental upgrading including water, sanitation and vector control as well as promotion of food production and nutrition, basic education and initiation of community development programs. The basic service delivery personnel are to be volunteer primary health care (PHC) workers with three months training, and birth attendants with one month training. When fully implemented there would be a total of 26,400 trained workers at the primary level. 9. Several constraints facing the health sector would prevent the effective implementation of the PEC program as it is currently envisaged. The major constraints are insufficient funding of recurrent health sector costs, inefficient and centralized management and administrative procedures, inadequate logistics and transportation support, and an insufficient supply of essential drugs. Added to these problems is an inadequate referral system for backup services. Experience with PEC programs, as it is beginning to accuimlate globally, demonstrates clearly that removal of these deficiencies is critical to the success of a PEC program. Nutrition 10. Despite the deterioration in services and the economy, general nutrition conditions in Uganda are better than for many countries In East Africa, with a comparatively low prevalence of malnutrition. Nutritional problems are thus of less priority than the pressing population and health issues identified in the report. Bowever, evidence Indicates that the level of endemic malnutrition is rising. Additionally, during 1979/80, drought and famine in the north eastern area, the breakdown of civil order in the central area and the uneven distribution of food have led to acute episodes of malnutrition in certain regions of the country. Protein-energy malnutrition affects up to 30Z of children under five years of age in rural areas in the center and north of the country. 11 Limited nutrition activities being carried out by the hinitries of Agriculture, Animal Resources and Fisheries, Cooperatives and Marketing, Culture and Commnity Development and Education. These activities are not  -viii- coordinated through a central body. The absence of a coherent food and nutrition policy in Uganda is an obstacle to effective sectoral program planning and inter-sectoral coordination. Until such a policy is developed and implemented, it will not be possible to integrate program activities in each sector to produce an overall improvement in nutrition conditions. 12. Nutrition emergencies arising from natural causes and social disturbance, have been frequent in Uganda over the past five years. To a great extent these emergencies have resulted, not from a lack of food at the national level, but from inadequate distribution and marketing. Solutions to these problems lie outside of the health sector and can only be developed through careful agricultural nalysis of the economic and social costs of alternative marketing and storage policies. Marketing, storage and transport of agricultural goods are dependent on the agricultural and transport projects outlined in the Revised Recovery Program. Contingency planning for food emergencies should be carried out. Effective agro-meteorological surveillance should be reinstituted as soon as possible, particularly in high risk areas, to provide early warning of crop and livestock production failures. B. Recommendations 13. Because of the deterioration of health services and the slow progress towards rehabilitation, the highest priority is the revamping of the health infrastructure that provides the primary means of delivering population, health, and nutrition programs. But a number of actions are needed before coherent progress can be made towards rehabilitation. The solutions to the major nutrition issues lie primarily outside of the health sector. Thus, the recommendations below concentrate on population and health. The recommendations distinguish between short term actions that are feasible within the next two years and medium term activities that are expected to take longer to design and execute. Population 14. Information and education (IEC) activites can be carried out in the short run to develop the commitment of government agencies and to increase public awareness of family planning. Analysis of the economic _ consequences of population growth by the MOP and seminars for government officials conducted by the MOP and MOB can be used to promote government commitment. Public awareness of the consequences of high fertility for health and economic welfare at the micro level should be the focus of an IEC program developed and implemented jointly by the FPAU amd MOR. 15. In the medium term, an interim population program can be carried out emphasizing the distribution of family planning services through the FPAU and other non governmental organizations (NGOs). The interim program would provide the basis for expansion in the future by increased emphasis on population and FP in the curricula of the medical school and the medical  -ix- assistant and nurse training schools. To promote these activities, the population unit in the HOP should be strengthened, an MCH/FP coordinator should be appointed, the IEC program should continue to be extended and FPAU nmnagement should be strengthened. Health 16. To provide for coherent rehabilitation activities and the long term development of the health sector, the planning capacity of the MOH should be improved through training and technical assistance to specify efficient planning procedures. A health strategy is required to coordinate resource allocation with health needs and long term recurrent cost capacity. Medium term planning should be eseablished consistent with the health strategy and the needs of the recovery program. The medium term plan should be updated on an annual basis. Bpcause a lag can be expected in the development of planning capacity, a start should be made now if the strategy is to direct the flow of resources in the medium term and beyond. 17. Restoration of health facilities and the repair of equipment must be addressed with highest priority. A complete inventory of the state of repair of facilities and equipment at all levels of care should be carried out to guide the restoration effort. Restoration of basic water, electricity, and sanitation should commence immediately upon completion of the inventory. The restoration should be accompanied by a program to train staff in maintenance. Increased quantity and improved management of supplies and pharmaceuticals are also needed to bring the use of government facilities to efficient levels. Short term technical assistance is needed to suggest immediate actions to improve procurement, inventory control, security, storage and distribution. 18. Before further steps are taken towards its implementation, the FEC plan should be modified in scope and simplified in concept to provide for the development of PEC services in a phased manner over a limited geographical area. The major target of the program should be reduced to dovetail with the current UNICEF and WHO focus on imuinization and oral rehydration. Phased development and limited scope would allow an assessment of needs and resources for the planning of subsequent expansion at a later stage. Phasing would also allow the program to be coordinated with the restoration of the referral and logistics systems. 19. Immediate steps are needed to rationalize staff establishment and remuneration. A 1983 Bank study recommended a reduction in staff and the number of geographical administrative areas, and the concomitant revision of working hours and remneration, to increase personnel efficiency and incentives. Consideration should be given to carrying out these suggestions within MDE facilities. 20. When recurrent funding needs are compared to a projection of available funds over the period from 1985 to 1989, the result is a total shortfall conservatively estimated to be US$100 millions. Greater support  -K- of recurrent. costs is required. The urge.ncy of finding addit-Ionl sources of recurrent funds requres that the Governmt reexauLne its policies on cost recovery for governmnt health services, A study is needed to suggest a fee structure and collection mchaisnmTax to accompany imroved governmnt services as rehabilitation proceeds. 21. A managemet study is required to provide guidance for -Lmeganization and improved ranagement within the MDR over the medium term. Ioproved capacity to execute and coordinate program and to handle crises within an organized framework is essential. The study should coosider the unification of all health services under one ministry rather than the separation of higber and lower level services between the MDH and 1=. 22. An Improved data base is an essential complement to both planning and innageuat. Baseline sample surveys should be carried out to establish bencbmrks and to measure progress. Regular reporting of service statistics and disease prevalence should be reestablished as a short term priority activity.  I. ECONOMIC AND SOCIAL SETTING 1.01 Uganda is a landlocked country in central East Africa, covering approximately 241,000 sq. km. of which 197,000 is land area (165,000 km2 i9 considered cultivable) and 39,000 is water. It is bordered on the east by Kenya, on the west by Zaire, on the north by Sudan, and on the south by Rwanda and Tanzania. The terrain varies from the densely settled hill areas bordering Rwanda, and the thinly settled but fertile central plains area, to the dry and sparsely settled region in the northeast bordering the Sudan and Kenya. Agriculture contributes over fifty percent of GDP and accounts for 99% of exports. Coffee is the major export crop. The 1984 population is estimated to be 14.3 million. The population is primarily rural with only twelve towns of over 10,000. Of the total labor force, 83% are engaged in agriculture, 6% in industry and lIZ in services. About 50% of adults are literate. The ethnic, linguistic, and religious heterogeneity of the population is an important factor that must be considered in the design and implementation of social programs. Sixty percent of the population are christian (primarily catholic), 30% adhere to tribal religions and 10Z are moslem. 1.02 Uganda's per capita income was US$230 in 1982, placing it in the lower 25% of African countries, but the low per capita income belies the true potential productivity of the country. When Uganda obtained independence from Britain in 1962, it had one of the most promising economies in Sub-Saharan Africa. The Government's fiscal situation was basically sound and a strong transportation and education infrastructure existed. Progress during the first decade after Independence was substantial with the real GDP of the country growing by 5% annually. Export earnings covered import requirements and the counLry enjoyed a current account surplus. A high rate of savings gave the Government the capacity to generate adequate internal investment capacity to allow further development of infrastructure including investment in health services. 1.03 After a military coup in 1971, however, the situation quickly deteriorated. Private firms were confiscated and added to the pa:astatal (quasi-public) sector or given, as patronage, to inexperienced or incompetent new owners. In both government and the parastatal sector, the administrative system was greatly reduced in effectiveness through fear, favoritism and misuse of funds. The impact of the domestic political situation combined with the external shock of risLng oil prices in 1973 to severely depress the economy. Many skilled personnel left the country and industrial facilities and government infrastructure deteriorated through poor management and a lack of maintenance, replacement equipment and supplies. The savings rate fell and, with an eroded balance of payment position, government investment in infrastructure and new development projects all but ceased. 1.04 A civil war, supported by Tanzanian troops, ended the regime of the 1970s. Physical damage from the war and subsequent looting and social instability added to the burden born by the already weak economy. A recovery plan was drawn up and initiated in 1981. Some progress has been made under the plan towards rehabilitation of the economy, but the effects of the war, legacy of the military regime of the 1970s and the continuing  -2- sociaL unrest remain the major factors that must pervade any examination of the current economic and social. status of Uganda, including the population, health and nutrition sectors. II. POPULATION A. Present Demographic Situation and Trends Introduction 2.01 Because of the civil war of 1979 and continuing political disturbances in the subsequent years, the reliability of the 1980 census is in doubt. Much of the Information from the 1980 census was lost in looting, making detailed analysis impossible. Nevertheless, sufficient information for an enumeration was obtained, and an analysis of broad population characteristics is possible through comparison of the 1980 enumeration with the 1969 census and corraboration from other sources. Attention to the 1969 patterns of population distribution and pre-1969 growth rates is warranted because these may give some indication of the patterns that will develop as social and political stability return. Population Size and Growth 2.02 The 1959-69 growth rate was 3.8% per annum. Compared to 1970-1980, higher net immigration and natural increase accounted for most of the higher growth rate, though the 1959-69 rate was probably inflated by an increase In the comDleteness of the census in 1969 compared to that of 1959. For the 196Os, the annual rate of natural increase has been estimated at 2.9%, with a crude birth rate of 48 and crude death rate of 19; the net migration rate was about 0.6%, with a net gain of over 400,000 migrants occurring during the decade. 2.03 The population of Uganda was enumerated at 12.6 million at the time of the 1980 census. This represents a 34% Increase and implies an anntal growth rate of 2.8Z since the 1969 census when the population was 9.46 million. However, the growth rate In the 1970s was probably lower than it otherwise would have been due to the bigher mortality resulting from violence, emigration and lower fertility during that time. Based on an analysis of the 1969 census and the estimated current age structure and fertility rate, the mission estimates the current population growth rate to be 3.2% per annum. Population Distribution 2.04 Relative to mast of Africa, Uganda is densely settled. The population density in 1980 was 64 persons per km2 (higher than a1l other African countries except Mauritius, Nigeria, Rwanda and Burundi) with 14 out of 33 districts having a density between 50 and 100. There is considerable variation in density of settlement, with the population concentrated heavily along the shores of Lake Victoria, in the southwest higblands of Kigezi (the districts of Kabale, Rukungiri and Bushenyi) bordering on Rwanda and Kivu Province of Zaire, and along the banks of the Nile in the northwestern corner of the country which borders Zaire.  -3- 2.05 The pattern of population growth between the 1969 and 1980 censuses varied throughout Uganda. The low annual growth rate of 1.8% in the Kigezi area is due to out-migration to the surrounding districts which are the most rapidly growing in Uganda with growth rates averaging 4.3% per annum. In Kigezi, Kabale, Rukungiri and Bushenyi are districts of high fertility (based on 1969 census) and high density with little reserves of arable land. High population fertility and limited opportunities for increasing agricultural production will probably continue to generate population pressure and out-migration from the Kigezi area in the future. Low rates of growth in the areas near Sudan and Kenya were primarily attributable to out-migration associated with political disturbances. The high fertility rates recorded in the north and northwest in 1969 probably have continued to the present time. Urbanization and Migration 2.06 The level of urbanization is low. Defining an urban center as 2,000 or more people, 7% of the population were classified as urban in 1980 compared to 6.4Z in 1969. Most of the small towns and trading centers are not distinguished by an urban lifestyle. Only the two largest towns, Kampala (the capital and also a commercial and industrial center) and Jinja (an industrial center) could be considered as having an urban environment. 2.07 Between 1950 and 1970, Uganda had been a focus of migration from the surrounding African countries. The attractions included the relatively strong economy; developed plantations, estates, and industries; a stable and relatively well-developed infrastructure and social service system; and reserves of relatively empty land for subsistence settlement. The largest net flows of economic migrants came from Zaire, Rwanda and Burundi, but there was also a sizeable influx of people from Kenya, Tanzania and southern Sudan. The population of non-Ugandan origin rose steadily from 414,000 in 1948 to 682,000 in 1959 to probably over 1 million in 1969; inmigrant communities thus made up 8.4Z, 10.6Z and 11% of the total population, respectively. 2.08 Trends since 1969 have been very different. Inflows ceased with the destruction of the economy. There was a mass expulsion of Asians during that period. There has also been an exodus of Ugandan refugees, both during the seventies (when the educated elite fled in large numbers) and after the 1979 civil war. At present, there are nearly 250,000 Ugandans registered as refugees in Zaire and in southern Sudan. The likelihood of a return of Ugandan refugees and a renewal of the migration from surrounding countries to Uganda is not great in the immediate future but, if civil stability returns with an upturn in economic development, Uganda could once again become a magnet for international migration. Mortality 2.09 M4ortality has declined throughout much of this century. Recent estimates indicate that expectation of life at birth rose from about 34 in the mid-1940s to 40 in the mid-1950s and then to 46 by the mid-1960s. While demographic data is not available to confirm recent trends, it is probable that mortality fell further until the early 1970s and then levelled off or even increased. The infant mortality rate (IMR) is now  -4- probably near 120 per thousand poptlation. The level of mortality up to 1969 approximates that of other African countries, fitting well with the general gradient of mortality across Sub-Saharan Africa, which is highest in the West and North and lowest in the East and South. The level of mortality in Uganda is lower than estimated levels in much of West and Central Africa but higher than in Kenya, Tanzania, Zimbabwe, Zambia, Zaire and Botswana. Fertility 2.10 The national total fertility rate (TFR) in the late 1960s was about 7.0. This is very high by world standards, but not remarkable in eastern and southern Africa where TFRs of 7.5-8.5 are common (for example in Kenya, Rwanda, Zimbabwe, Malawi and parts of Tanzania and Zambia). This national average, however, conceals regional and ethnic differences. The total fertility varies from 7.7 to 8.5 in the southwestern and northwestern parts of the country to 5.6 to 6.7 in the Buganda Region. The prime determinant of differentials in fertility is the prevalence of sterility or subfecundity; the causes of the subfecundity are not known. In 1969, the proportions of women aged 45-49 who had never had a live birth, hence were almost certainly sterile, were only 4X-7% in the fertile Southwest and Northwest, roughly 10%-15% in the intermediate west and east, and as high as 201-25% in Teso, Bunyoro, and in the most affected parts of Buganda. B. Population Projections 2.11 A set of four population projections for the period 1980-2015 was constructed using the 1980 enumeration of total population as the base. The age distribution for 1980 was derived by projecting the 1969 census results on the assumption of constant fertility at the late 1960s level and mortality that declined during the early 1970s, then rose again to approximately the level of the late 1960s. In three of the projections, mortality is assumed to decline according to a standard path used in Bank World Development Report (WDR) projectionsl/, with life expectancy rising from 47 in the period 1980-85 to 61 in the period 2010-15. Three separate trends in fertility are combined with this declining mortality path. The first assumes a constant TFR of 7.0 throughout the projection period, and represents a scenario of maximum population growth. The second follows the standard path projected for Sub-Saharan Africa in the Bank WDR projections, with the TFR remaining constant at 7.0 until 1995 and then falling to 4.1 by 2010-15; this belated and gradual decline represents the expected spontaneous evolution of fertility given the normal processes of development in the absence of large-scale government family planning programs. The third trend is an early and rapid fertility decline beginning after 1985 and resulting in a TFR of only 2.3 by 2010-15; this decline is modeled on the experience of several countries in the recent / Net migration is assumnd to be negligible throughout the projections. The fertility and mortality assumptions are set out in Annex 1, Table 1.1. Details of the underlying IWDR projection model are discussed in Elwan, Vu and K. C. Zachariah, Short-Term Population Projections. 1980-2000 and Long-Term Projections, 2000 to Stationary Stage for All Countries of the World. PENPR, World Bank, August 1983.  -5- past, and represents what could be achieved if a successful natiodal fimily planning program were immediately initiated. Finally, a fourth projection was made on the assumption of constant fertility and constant mortality at the levels estimated for 1980-85; this projection represents what might occur if the country fails to regain political stability and the levels of economic prosperity which existed in the early 1970s. Table 1: Projected End of Year Population (in millions) Projection 1980 1990 2000 2015 Assumptions 1. Declining Mortality 12.6 17.5 25.2 45.5 Constant Fertility 2. Declining Mortality 12.6 17.5 24.8 37.8 Gradual Fertility decline 3. Declining Mortality 12.6 17.0 21.0 25.9 Rapid Fertility decline 4. Constant Mortality, 12.6 17.4 24.1 39.4 Constant Fertility See Annex I, Tables 1.1-1.5, for details of projection. 2 .12 The tremendous potential for population growth is immediately obvious from the results reproduced in Table 1 and detailed in Annex 1, Table 1.2 through 1.5. In the absence of any rapid fertility decline, the 1980 population of 12.6 million will grow nearly 100% over the next 17 years, reaching 24 to 25 million by the year 2000. It will triple, or more than triple, within the next 32 years, attaining between 38 and 46 million in 2015. Even if a rapid fertility decline occurs, the population would still increase by two-thirds by the year 2000, and approximately double to 26 million by 2015. 2.13 Further potential for growth will remain at the end of the projection period. Without a rapid fertility decline, growth rates will be in the range of 2.5-4.1%, which would imply a further doubling of population to 76-91 million within the following 17 to 28 years. Even with a rapid fertility decline, resulting in a net reproduction rate of about 1 by 2010-2015, the _ate of growth would still be 1.4Z in 2010-2015, giving a further doubling to 52 million within the following 50 years. C. Development Implications of Population Growth 2.14 The rapidity of population growth as well as the ultimate size will play a critical role in determining the outcome of Government efforts to improve the standard of living and the welfare of the people. The trend of fertility in the next 10 years will be of prime importance in determining the rate of increase. Large-scale net emigration does not  -6-e appear a feasible outlet for Ugandan population growth. It is also clear from the projections that mortality decline Is only of secondary importance in determining the rate of increase; the pessimistic assumption of no fall in mortality over the next 35 years only cuts the projected population by 4Z in 2000 and 13X In 2015. By contrast, even the belated and gradual fertility decline, starting only in 1995, would cut the 2015 population by 17%, while the rapid, early fall in fertility would reduce the total population by 17% in 2000 and by 43% in 2015. Indeed, in the absence of fertility decline, the rate of natural increase will remain above 3Z throughout the next 35 years even if mortality stays constant; by contrast, even if mortality falls, fertility declines would cut growth to 2.5% (if gradual) and 1.4% (if rapid). In fact, according to these projections, the government's target of reducing population growth to 2.3% by 1995 could be achieved only if an early and rapid fertility decline takes place (when a rate of 2.3% would be reached by the early 1990s); the belated decline would not reduce growth to this level even by 2015. 2.15 Economic growth is projected by the Bank (Uganda CEM, October 1983) to remain above 3.5% for the rest of the decade. However, a decade or more of economic growth at a rate of at least 7Z per annum would be required simply to recover the level of prosperity achieved by 1970 before any real socio-economic improvements can be undertaken. A slowdown in population growth is required to allow substantial improvements in per capita gross domestic product (GDP). The Uganda CEM of December 1983 estimated that by 1990, per capita GDP will have recovered to only 75Z of its peak 1971 level based on a population growth rate of 2.8%. But subsequent analysis reveals that the actual population growth rate is above 3% and that recovery at the rate assumed in the CEM could only be achieved with rapid fertility decline; if the fertility decline is delayed, the projected population growth rate of 3.2 to 3.4% will only permit a recovery in GDP per capita to 70-72X of the 1970 level by 1990. 2.16 In the longer term, to the year 2000 and beyond, the need for a start now with an effective population policy is even clearer. Important long term effects of population growth arise from changes in age structure, unknown possibilities for increasing agricultural production, and required expenditure for social infrastructure. Implications for Age Structure 2.17 A fall in fertility would produce dramatic declines (Annex 1, Table 1.3) in the dependency ratio. By contrast, constant fertility would result in a constant dependency ratio if mortality does not change, and in an actual rise in the dependency ratio (to 108 in 2000 and 111 by 2015) if mortality falls. (This is the result of faster declines in childhood mortality than in adult mortality, and hence increasing proportions of children in the population.) Domestic investment may be reduced as the economic burden of dependents grows and national resources are diverted to consumption. The ratio of 102 in 1980 could be cut to 73 by 2000 and 47 by 2015 if fertility declines fast and early and even the belated decline would produce a fall in the ratio to 79 by 2015. This decline in the dependency rates could provide a valuable opportunity to improve rates of capital accumulation over the next 30 years.  -7- Implications for Agriculture 2.18 At the national level, Uganda's land reserves and potential for improvements in agricultural productivity could provide a 3 to 4Z per annum increase in agricultural output that would be adequate for all but the constant fertility-declining mortality scenario of projected population growth. However, this is not true of all parts of the country. Population pressure on the land is already acute in Kigezi, where substantial emigration is occurring in response, and is beginning to be felt in parts of the North and East where soil and climatic conditions are less favourable than in the South. In these areas, which are zones of high human fertility, part of the solution lies in continued resettlement of the population in other parts of Uganda. However, a reduction in the rate of natural increase is also essential to contain ultimate population densities to a level consonant with the potential of the land. Implications for Social Infrastructure 2.19 The population projections demonstrate (Annex 1, Table 1.4) that massive growth of the school age population is inevitable unless a rapid fertility decline occurs. Using the declining mortality projections, the number of school age children will double by 2000 and more than triple by 2015 if fertility remains constant. In contrast, with a rapid fertility decline there would be only a 50% increase in the school age population by 2000 and the numbers would actually decline to a 25% increase by 2015. This would reduce the strain on funding resources, allow a quicker recovery to former educational standards, and speed the achievement of universal primary schooling. 2.20 The implications of the population projections for educational expenditures can be set out in monetary terms (see Table 2). Using an estimated cost per primary pupil of US$6.00 per year (UNESCO estimate) and assuming no fertility decline, the annual recurrent expenditure on primary schooling would have to increase by 3.8% per year between 1982 and 2015 to a level of US$33 million per annum just to maintain the current enrollment ratio, and the rate of increase necessary to achieve the government goal of universal primary education (UPE) would be 5.5% per annum to a level of US$57 million in 2015. In contrast, with a rapid fertility decline the rate of increase necessary to maintain the current enrollment ratio would be less than 1%, and UPE could be achieved with an increase in educational expenditure of 2% per year to a level of US$19 million in 2015. While these estimates are obviously approximate because of changing student teacher enrollment ratios and unpredictable changes in other educational cost prior to 2015, the estimates do give a valid picture of relative costs under alternative fertility trends and demonstrate that an effective population policy would contribute dramatically to reducing government social expenditures and to the achievement of the Government's plans for expanded primary education.  -8- Table 2: Required Growth in Educational Expenditures (US$ 000,000) At Constant ifrollment Universal Primary Education Ratiol/ by 2000 Average Average Annual Annual Growth Growth 1982 2000 2015 Rate% 2/ 1982 2000 2015 Rate% 2/ No fertility decline 9.4 18.1 33.3 3.8 9.4 31.3 57.4 5.5 Gradual fertility decline 9.4 18.1 25.0 3.0 9.4 31.3 43.1 4.6 Rapid fertility decline 9.4 13.4 10.9 0.4 9.4 23.1 18.8 2.1 1 / Assuming: US$6.00 per primary student 0.58 enrollment ratio _, 1982 to 2015 2.21 A similar picture emerges from projections of the population for maternal and child health (KCH) services (Annex 1, Table 1.". Again, the number of potential MCH clients would nearly double by 2000 unless an early and rapid fertiIity decline occurs, and would increase to between 2.8 and 3.7 times their present numbers within the next 35 years. Yet, at present the health services are in disarray and need almost complete reconstruction to regain former levels of quality and coverage. This process will require time and additional expenditure, and certainly will be severely handicapped by high rates of growth in the client population. A rapid fertility decline again would greatly facilitate the process of reconstruction since it would reduce growth in the client population to only 57Z more than present numbers by 2000 and to about double that number by 2015. D. Population and Family Planning Policies 2.22 Although family planning (EP) was introduced in Uganda in the early 1950s and the Family Planning Association of Uganda (FPAU) was formed in 1957, FP prograus were neither recognized nor given assistance by the government or political leaders until the early 1970s. Program support has never been strong and some political leaders remain reluctant to advocate FP because it runs contrary to strongly pronatalist social mDres. Furthermore, a 'arge proportion of the Christian majority are Catholics and their religious beliefs place restrictions on FP activities.  -9- 2.23 In the Third Five Year Plan (1972-76) proposals, the Government recognized FPAU as the agency for implementing its policy of providing advice to women on FP and child spacing matters. The central objective was to inform married people of the problems created by rapid population growth and of the FP methods available to limit fertility. In 1976, the Government designated the MOH to take over as the implementing agency for FP services, while FPAU was to provide information and education services. Directives were issued to make FP services available at all hospitals and health centers and to provide staff training in FP, but the services and training failed to materialize. 2.24 In 1981, the Government of Uganda adopted an explicit population and FP policy. The objective of its plan is to reduce the population growth rate to 2.6% per annum, over a 15-year period (1981-1995). The main program strategies include integrating FP services in all government and missionary health facilities, including rural units, integrating population education into all levels of formal and non-formal education, raising the legal age of marriage, and fostering womens' development with an emphasis on formal education. These are positive policies towards an effectivc population program. Unfortunately, these policies are only now starting to be translated into an operational plan. E. family Planning Services 2.25 The MOH is currently providing only limited family planning services. MOB services are intended to be expanded over the next three years, but MOH capacity to offer services is constrained by a lack of trained manpower, poor logistical support, and a limited infrastructure for the delivery of MCH services in rural areas. In 1983, only 3 nursing tutors, out of approximately 45 on the MOB staff, were trained in FP. None of the health training programs provide practical training in FF. Lack of a continuous source of contraceptive commodities and a distribution, storage, and inventory control system is an additional problem. Deficiencies in MOB service capacity are sufficiently great to impede effective delivery of FP services on a wide scale through an integrated MCH PP program in the next 3 to 5 years. Rehabilitation of the health service infrastructure and building up the training base are logical prerequisites to the introduction of an integrated FP program. In the interim, existing government and non-governmental channels must be fully exploited. 2.26 Some preliminary steps to build the foundation for a family planning program are being planned by MOB for USAID funding. These steps would include the development of teaching capacity in schools for nurses, medical assistants and health visitors. If plans proceed on schedule, 22 tutors would be trained in 1984. An additional step will be the development of a national contraceptive logistics management and reporting system and training of personnel in supply management. The USAID funding would also provide baseline information through a contraceptive prevalence survey (CPS) to gather information on contraceptive usage, attitudes and knowledge. 2.27 Until the MOH program can be expanded the PPAU will remain the principal provider of FP services. FPAU operates a total of 60 clinics.  -10- Twenty-three of the clinics are housed in FPAU's own facilities and 37 are located in M)E hospitals and health units. All of the clinics are located in cities and large totwns and service a predominantly urban clientele. Charges are levied for contraceptives at the 23 free-standing clinics (20 Ush. for one cycle of pills or one injection of Depoprovera, 5 U Sh. for one condom, IUDs are free). All clients regardless of clinic site must pay an annual FPAU membership fee of 100 U Sh. (40 US cents). The clinics are operated and supervised by FPAU staff. In 1983 there were 80 staff involved directly in service delivery, the remaining 36 (of a total of 116) staff members provided project support, logistics and central administration. Staff were t'rained, primarily in Uganda, with the assistance of the International Planned Parenthood Federation (IPPY) and International Training for Health (INTRAH). Contraceptives are provided to the program by IPPF. Ninety percent of FPAU's funding comes from donors and it is expected that further donor support will be forthcoming for training, educational materials and contraceptives. 2.28 Clinical service statistics are kept by FPAU units and reported to the central administration. Although reporting is uneven it is noteworthy that the FPAU records are collected and summarized on an annual basis. FPAU's records show a total of 35,400 new or continuing acceptors in 1983, compared to 33,700 in 1982 and 30,900 in 1981. In the absence of information on continuation rates it is difficult to enumerate users but it is estiEated that nationally about 12 of women of reproductive age are contraceptive users. In urban areas the'percentage may reach 4 to 52. Breakdown by method is: oral contraceptives, 712; injectables, 15t; - condoms, 8Z; and IUDs, 3x. Survey information is neeted to ascertain the reason for the method mix, but the low percentage of IUD acceptors appears attributable to a lack of trained personnel for insertion and follow-up. 2.29 FPAU has had management problems and been short staffed which has hindered effective promotion, esecution and surveillance of programs in the past. More recently, an IPPF evaluation in 1982 noted that the FPAU management has shown marked improvement since 1979. However, the present five person FPAU management and program development team is too small to carry out expanded program activities. Other major constraints to program expansion are shortages of equipment, transportation for supplies and staff, trained staff, and inadequate information, education and comnimication (IEC) activities. Examination of the FPAU three year plan for 1984-86, to be funded principally by IPPF, indicates that actions are planned to remedy some of these deficiencies through in-service training, improved record keeping, management training, leadership education seminars for politicians and senior public servants, and intensified IEC activities including schools and youth organizations. The forecast budget (US$269,000 for 1984, US$287,000 for 1985, US$309,000 for 1986) is sufficient for only a modest scale of activities. Increased funding is needed to allow the FPAU to build its activities and to support the expansion of services delivered through rural units in selected areas of the country. The FPAU constitutes a highly positive asset in the context of the generally limited  -11- III. HEALTH A. Health Status 3.01 The breakdown of the epidemiological reporting system occurring during the 1970s and continuing to the present time precludes a statistical analysis of the current health status of Ugandans. Assessment of mDrbidity and mortality is largely based on the observations and experience of health workers and limited inforuation from NGO facilities. Bowever, the limited evidence that does exist points to a deterioration of the health status over the last 12 years. Mortality and Morbidity 3.02 Data collected In the 1969 census revealed a crude death rate of 19/1000 and an infant mortality rate of 120/1000, both of which represented a steady decline in mDrtality rates over the last few decades. Death rates for 1980 have been estimated at 18 and 96 respectively in the World Development Report for 1983. However, due to the loss of the 1980 census data and the absence of an effective vital registration system or health data reporting system, these rates were based on projections of the positive trends in health status up to 1969 and do not reflect the events of the 1970s. It is aore plausible to assume, in accordance with the opinion of health workers in Uganda, that mortality rates may have levelled off with a deterioration in health services or even risen to the levels of the early 1960s. 3.03 A WHO 1980/81 survey of 6460 deaths for all ages listed the causes as follows: Cause of Death Percentage of Deaths Measles 24 Respiratory Tract Infections 17 Gastroenteritis 11 Anemia 6 Malaria 5 Trauma 5 Tetanus 4 Malnutrition 3 Tuberculosis 2 Whooping Cough 2 Other 21 Total 100 The MOH's National Plan of Action on Primary Health Care, 1982 lists, but does not rank, the major diseases for all age groups as: malaria, diarrhea and vomiting, respiratory tract infections, malnutrition, childhood diseases (especially measles, whooping cough, polio), intestinal worms, accidents and tuberculosis. Information on major causes of disease and death for children was obtained from PEN mission field trips to hospitals  -12- and health centers in the East (towards Mbale), West (towards Mbarara and Kabale), and in the Kampala area. The information was consistent with the WHO findings. For children, the major causes of morbidity in order of frequency were measles, gastroenteritis, malaria, and respiratory tract infections. Measles ranked as the most common cause of death, followed by gastroenteritis, malaria and pneumonia. In the neonatal period (the first 28 days of life) premature birth, low birth weight and neonatal tetanus were the leading causes of mortality. 3.04 Health workers observe that the prevalence of malaria and of tuberculosis appears to have risen substantially. At present, malaria prevalence rates are unknown, due to the lack of epidemiological surveys and the absence of equipment and supplies that are needed to diagnose the presence of malaria in the blood. Hospitals in Kampala with facilities for diagnosis report that falciparum malaria, the most severe type, is common. Pregnant women and young children are at particular risk. As a result of this, an increase has been observed in premature labor, severe anemias during pregnancy, and infant deaths from malaria. 3.05 Tuberculosis is also reported to be on the rise, and more patients are being admitted to hospitals with advanced forms of the disease. No epidemiological data exist that can be used to establish the prevalence of tuberculosis, and services to detect and treat the disease are extremely limited. Many non-governmental organisation (NGO) hospitals require that the patient or relatives pay in advance for the cost of hospitalization for tuberculosis. This was introduced to reduce the number of patients who discharged themselves prematurely, but it has been observed that some patients are unable to obtain treatment because of the cost. B. Health Policies 3.06 The MWH does not have a comprehensive strategy for development of the health sector. To the extent that policies have been outlined that reflect the administration's plans for development within the sector they are contained in the Recovery Programme for 1982-84 and the National Plan of Action on Primary Health Care. Neither of these documents fills the need for a coherent strategy for development within the health sector. The Recovery Programme (discussed in paragraph 3.08), while containing a logical series of actions that need to be supported, addresses only the most pressing short term needs. The PEC plan (discussed in r3ragraphs 3.25 - 3.29), while providing a long term perspective, may be unworkable over the near future and does not address the requirements of the entire health sector. Recovery Program 3.07 The recovery program for 1982-1984, as it is unfolding through actual project implementation, reveals a strategy that emphasizes assistance for agriculture, industry and transportation and places a lower priority on rehabilitation of the social infrastructure including health. The original recovery program, formulated in 1981, was revised in late 1983. Five criteria were used in the selection of projects for the Revised Recovery Program (RRP): (i) enhancement of Uganda's foreign exchange position, (ii) rapid stimulus to industrial and agriculture development,  -13- (ill) minimal negative effect on the GOU recurrent budget, (iv) importance to sustain recovery and growth, (v) projects of an urgent humanitarian and social character. Health projects qualify primarily and directly under the fifth criteria and secondarily and indirectly under the fourth criteria. The ordering of the criteria approximately reflects the priorities plaL.d on the recovery projects. Because of the partial collapse of the health services, health projects were selected to provide the minimal restoration required to reverse the worsening health status of the population. The projects are summarized briefly: (a) rehabilitation of Mulago Hospital. The cost of complete rehabilitation would be enormous and would carry the project beyond meeting immediate priority needs. The planned project provides for building repairs, limited equipment, rehabilitation of operating theatres, technical assistance in administration and engineering, and increased supply of drugs; (b) rehabilitation of Government district and Rural Hospitals. The planned rehabilitation of 45 hospitals include water and sewage, electrical and mechanical services, essential repairs of buildings, equipment and drugs. No provision Is included for administrative or maintenance training; (c) rehabilitation of rural facilities. This project provides for rehabilitation of 445 small rural health units ranging from health centres to subdispensaries. The project would be in addition to a UNICEF project to rehabilitate 100 rural health units; (d) blood transfusion services. This project includes a building to house the transfusion service, and purchase of blood transfusion equipment and essential supplies; (e) strengthening of primary health care (PHC) Services. The title of this project is somewhat misleading as it primarily provides for rehabilitation of basic health services but does not provide sufficient funds for training of the extensive cadres needed for the planned PHC program (see para. 3.45). UNICEF is assisting this project which includes strengthening the planning unit within the MOB, rehabilitation of 100 health centres, nutrition services, strengthening transport maintenance and refresher training; (f) health training and planning. This project, being supported by CIDA, would provide retraining of rural health workers, and technical assistance and training for health planning. The project includes renovation of schools, equipment and materials and would provide for refresher training and seminars for 1910 rural health workers and 300 members of district and regional health teams; (g) accelerated immunization services. The project would provide vaccines, transport, cold chain and supplies, training and IEC materials with the objective of achieving 60 percent coverage of expectant mothers and children aged 0-2 by 1986. UNICEF is providing assistance for the immunization project.  -14- 3.08 Within the scope of the recovery program there is considerable flexibility in the identifitation and design of specific projects. It is important that as funding of the recovery effort occurs the projects are shaped in accordance with a health strategy that details how resource allocation relates to major health problems, where resources would be distributed and what priorities underlie the strategy. An important element of planning is the coordination of capital expenditures with long term recurrent cost capacity. Routine planning should also provide a process for relating annual plans and district budgeting to the health strategy. C. Organization and Management Institutional Framework 3.09 The organization and operation of government services is shared between the NOR and the Ministry of Local Government (MOLG). The NOR is responsible for the formulation and development of health policy, planning, monitoring and evaluation, and for the provision of technical and professional advisory services. In addition, the MOR is responsible for the provision and operation of hospital services. These services include the Mulago Referral and Teaching Hospital, situated alongside the medical school in Kampala, the Central Mental Hospital at Butabika, four regional hospitals :ad 41 district hospitals. Finally, the MOR operates the nursing and paramedical training schools, the Tutors College and the Central Public Health Laboratory which has only recently been re-opened in facilities provided by the medical school. 3.10 The MOLG provides local health services from non-hospital facilities, including 88 health centers, 54 dispensaries, 62 maternity/dispensary units and 80 aid-posts (Table 2.1). Operation of these units is the reponsibility of the MOLG district administrators. Professional and technical supervision is provided by District Medical Officers (DMOs) and technical staff deputed to the MOLG from the MOH. The nursing and paramedical staff who run the facilities are also deputed from the MOE. Staff salaries, medical supplies and drugs supplied by the MOB are paid out of the MDLG budget. There are complaints that the MOLG is occasionally in arrears or default on reimbursement of MOR staff. On the other side, there are complaints of inadequate supplies forthcoming from the MOH. These complaints are partially attributable to the insufficiency of available revenues and drugs at the national level, but poor coordination between the central level and local programs also plays a part. 3.11 Nongovernuent organizations provide approximately 50Z of health services in Uganda through 32 hospitals and 61 sub-hospital facilities. Health services provided by NGOs are approved by the BOB and conform to government health policies. The principal providers in order of volume are the Catholic Church, Protestant Denominations, and the Islamic Council (Anne 2, Table 2.3). Although individual NGO facilities are independently nunaged, the NGOs have voluntarily established central medical bureaus to coordinate their services. In addition, the Catholic and Protestant bureaus have instituted an effective joint service to facilitate the purchase, storage and distribution of drugs to member units, including the distribution of donated drug supplies.  -15- 3.12 During the last ten years the MOH planning, organization and managemnt of hospital services, and the provision of professional and technical services to hospital and non-hospital facilities have come under increasing strain due to deterioration of facilities, lack of supplies, and loss of key managerial staff. At the central level there appears to be confusion over priorities and courses of action and MOH staff are largely occupied resolving crises concerning logistics, supplies, and maintenance- As a result, middle level staff are unclear about their responsibilities. -Central level knowledge of activities at the district level is hindered by a lack of field visits and a failure to collect managerial information on a regular basis attributable to transportation difficulties and a breakdown in reporting systems. Because of these circumstances, the MOH has lagged behind in developing detailed work programs based on the priorities set out in the revised recovery program. 3.13 In 1982 a joint GOU and WHO team assigned to assess health sector needs for rehabilitation recommended that the MOH be reorganized and the planning unit strengthened. The MOD has yet to make changes in its organizational structure to respond to the priorities set by GOU. Thus, the present structure of the MDH, as illustrated in Figure 1, reflects the different cadres that make up the health services and the large hospital units rather than program function. While complete reorganization may be too costly and disruptive in the light of more pressing needs for immediate recovery of health services, limited reorganization to allow for effective planning and implementation of projects is needed. Coordination 3.14 The National Advisory Council on Health, established by the Government in the 1960s to advise on the development of health services, has been In abeyance for some years as have the various technical advisory comnittees to the MDE. The MOLG has no formal mechanism to coordinate the health services under its Jurisdiction or to liaise with the MOH at the central level. At the district level, the District Medical Officer (DMO) and his technical support staff supervise all health services except for the MOB district hospital, which is supervised by the assistant director of medical services in charge of hospitals. NGO activities are loosely coordinated through the protestant council and catholic bureau. Staff from the three NMO hospitals in Kampala (two Catholic and one Protestant) and the Mulago Hospital (MH) meet monthly to promote the coordination of hospital services. In general, although NGO organizations are benefiting from limited cooperation in logistics, there is a serious lack of coordination between the MOH, the Y1OLG and NGOs. The needed coordination concerns distribution and planning of services, manpower, logistics and administration.- D. Delivery of Health Services Volume and Quality of Services 3.15 Health services are provided through hospitals, health centers, dispensaries, maternity units, sub-dispensaries and aid posts. When fully staffed, a health center includes inpatient and outpatient care provided by  -16- a medical assistant, a qualified midwife and nursing assistants. A dispensary unit provides elementary curative care through a medical assistant and, in the case of maternity units, a trained midwife. Sub-dispensaries are manned by paramedical staff having only rudimentary training. Aid posts provide outpatient care through weekly or twice weekly visits by a paramedic. Tables 2.1 through 2.4 detail the facilities operated by the MDR, WILG and NGO organizations. In 1983 the MOH and MOLG ran 43 hospitals (with a total of 9000 beds), 88 health centers (with 2400 beds) and 485 sub health center units (62 maternity/dispensary units, 54 dispensaries, 289 sub-dispensaries, 80 aid posts). NGO facilities included 32 hospitals (5000 beds) and 61 non hospital facilities (2 health centers, 17 maternity/dispensaries, 25 dispensaries, 3 subdispensaries, 14 maternity units). 3.16 If fully functioning, the wide distribution of facilities would allow the provision of a good balance of urban and rural services with only some inequity geographically resulting from the higher density of facilities in the southern and western regions (this is detailed further in paragraph 3.63). However, the data on numbers of facilities masks the low quality and scarcity of services, especially MOR services, and must be interpreted cautiously as an indication of the potential infrastructure for future services but not as an accurate picture of existing services. Both NGO and Government health facilities have deteriorated since the early 1970s and have continued to deteriorate throughout the war of liberation and post-war period through loss of medical manpower, lack of maintenance of plant and equipment, and in some instances war damage and looting. However, for a number of reasons to be discussed below, government facilities and service delivery have been more severely affected. Very little data is available to describe service delivery quantitatively; any assessment of the current situation by necessity relies on a subjective evaluation derived from field trips and discussions with health administrators. 3.17 In 1972/73, at the peak of government service delivery, the number of outpatient visits to NOR and MOLG facilities was 28 million (2.6 per capita) and the number of inpatient admissions was 950,000. By 1976/77, the last year for which data are available, the number of outpatient visits had fallen to 18 million (1.5 per capita) and the number of inpatient admissions to 450,000. Since 1976/77 the volume of services has not grown and, in fact, probably fell still further during the civil war and the immediate rehabilitation period, but this cannot be verified by data. 3.18 The decrease in services is attributable to declines in the numbers of physicians, technical staff, medical supplies and equipment. Nearly half of the more than 700 physicians in the country at the beginning of the 1970s left when the order expelling all Asians from Uganda was enacted. Other key personnel continued to leave throughout the remainder of the decade. The fall in manpower was coupled with a drastic reduction in the fiscal and physical resources available to the NOR. The MDH budget for the last few years has been less than 10% of its real value in 1969 (see para. 3.57).  -17- 3.19 Staff are, accordingly, often underutilized, which adds to difficulties in maintaining morale and reasonable standards of care. Even the facilities that are busy, such as Mulago Hospital (the central teaching hospital), operate at a slow tempo because delays in diagnostic work and treatment increase the length of stay. Because the majority of facilities have deficiencies in the quality or quantity of water and sewage disposal as well as a shortage of soap and disinfectants, the risk of hospital infections presents a substantial hazard. Auditing of patient admissions, length of stay and standards of care are almost entirely absent. Quality control of diagnostic tests has clearly suffered as have the standards of radiology. Patients are subjected to the unnecessary hazard of repeat X-rays which further Increases the demnd for scarce films. Some patients, who would foruerly have been treated in Uganda, are now evacuated to Nairobi or Europe because of the lack of key specialists and supplies and the deterioration of equipment. 3.20 The NGOs, although also heavily hit by the declining economy, have been able to carry out limited construction and renovations through assistance from overseas donors and NG0 facilities have expanded their services to compensate slightly for the gradual breakdown in government services by operating their facilities well beyond their designed capacity. Unusual commitment of NGO staff coupled with the independent management authority of each of their hospitals has allowed a response to changing circumstances through local initiative. NGO facilities have also been able to partly offset operating cost and replenish drugs and supplies by increasing their charges to patients (see para 3.66 below). Reports from seven catholic hospitals showed an increase of inpatient admissions of 27% from 1981 to 1982. Outpatients also increased by 20%. A review of cases admitted to one large catholic hospital indicated that growth in admissions was made possible by an increase in the bed complement (beds set up in catholic hospitals increased by 12% from 1979 to 1983) and improved efficiency to permit earlier discharge. The pattern of inpatient admissions in protestant hospitals also showed an increase, growing 27% from 1979 to 1982. The beds in protestant hospitals set up increased by 20% in 1983 compared to 1979. These achievements are notable, but have resulted in considerable strain on the overworked staff and on aging facilities and equipment. Table 2.3 lists the number of MOE and NGO hospitals and hospital beds by region. Despite the fact that NGOs have only 432 of the total bed complement, it is estimated that well over 50% of inpatient hospital care is now provided by NGOs. 3.21 Retrenchment on the part of both the government and NGOs has occurred at the expense of the more peripheral units. MOLG facilities would provide a total of 83% of all beds below the hospital level if looted and war damaged facilities were re-equipped. While data is not available for comparison it is probable that, as is the case with hospitals, NGOs provide a higher proportion of total out-patient services than the complement of facilities would indicate. This probability is supported by UNICEF's 1982 inventory of 80 MOLG health centers and 43 dispensary/maternity units that has provisionally found that 78% of MOLG health centers and 86% of maternity units were less than 50% occupied. 3.22 The UNICEF study also found that fewer than half the government facilities had assistant health visitors on staff (the assistant health  -18- visitor provides outreach services in all peripheral units from bealth centers to sub dispensaries). Less than a quarter of MOLG facilities had adequate equiaent to carry out their designated roles in delivering basic curative and MCH services. Only 52Z of facilities had adequate stocks of chloroquine (for the treatment of malaria), aspirin, antiheluenthics and ergometrine (for obstetrical emergencies). Over 30% of health center facilities provided no inpatient care and approximately 25% of health centers and maternity units provided no inpatient obstetrical services. Less than 252 of MOLG facilities provided imnization services (only 25% of the facilities had a refrigerator that worked). Only 50X of facilities provided well baby clinics for children 0-5 years, but nearly all provided antenatal services and outpatient services. These findings, corroborated by limited epidemiological data showing a rising prevalence of commmnicable diseases, demonstrate the severe contraction in basic curative and preventive services. Health Facilities, Equipment and Logistics 3.23 The operating efficiency of health facilities started to decline in the early seventies and still continues to decline because of insufficient funds to maintain the existing facilities and equipment, build new facilities and purchase equipment and essential supplies. The lack of transportation due to shortages of vehicles, spare parts and fuel, and the poor condition of the roads has compounded the problem of supplies and has also hampered patient access to hospitals and the transfer of patients between hospitals. In general, the more peripheral the facility, the greater the problems. Theft, particularly in governmnnt facilities, has become an added difficulty. Only a few facilities suffered severe war damage, but looting was widespread, particularly in the northern area. Another important factor in the deterioration of physical plant and equipment was the attrition of skilled technical staff who are required to mintain and operate the health facilities. 3.24 A comprehensive countrywide survey, in cooperation with NGOs, is needed to assess deficiencies in the physical plant and equipment of all health facilities. UNICEF's 1982 survey of over fifty percent of government health centers and dispensary/maternity units, hawever, gives an indication of the physical status of government facilities. Less than one-third of facilities had water on the premises, but nearly two thirds reported a close source of water available if repairs could be carried out. In 25% of responses, water was stated not to be of acceptable quality. In many instances, not enough water could be obtained to meet the needs of the facility. Twelve percent of facilities reported no patient toilets or latrines. One-third reported the condition of the physical plant as -good', 60% "moderate'. Over half reported that the last time repairs had been carried out was prior to 1974. Approximately one-third of the facilities had electricity and only one-quarter reported a refrigerator in working order. Less than one-quarter of all units had the necessary minimal equipment to offer basic curative or maternity services. Only 11% had basic equipment to offer MCH services.  -19- Primary Health Care 3.25 In Septeimber 1983 the Government adopted, in principle, the National Plan of Action on Primary Health Care and allocated US$15 million under the Recovery Program (1982-84) for the rehabilitation of primary health care facilit:ies. The plan is under revision and a final version is expected to be available in 1985 but the basic elements of the original plan are expected to remain unchanged. The plan embraces a wide range of curative, preventive, and rehabilitative activities that include malaria control, immunization of mothers and children, treatment of diarrhea with oral rehydration, a school health program, family planning, environmental upgrading including water, sanitation and vector control as well as promotion of food production and nutrition, basic education, and initiation of community development programs. Outside of the PHC plan, experience in primary health care (PUC) delivery has been obtained from 22 projects at the local level, but no common concept or structure has emerged from these projects and they provide a variety of service packages ranging from the purely curative to the solely preventive in character. 3.26 According to the present plan, training of PHC workers and establishment of full services would be extended to four administrative districts every two years. At this rate, it would take 16 years to cover the entire country (33 districts). Because each district has an average of twenty sub-counties (each of which will have an average of twenty PHC units), approximately 13,200 PHC units would be required by the year 2000. Each unit will consist of a PHC worker with three months' training and a trained TBA with four weeks' training. At the sub-county level a PHC supervisor, usually an assistant health visitor, will be prepared to recruit, train, organize, supervise and support 20 PHC teams in her jurisdiction. In turn, at the district level, a senior PHC supervisor/coordinator will carry out similar responsibilities with respect to the 20 sub-county supervisors under the jurisdiction. The district supervisor will be a member of the district health team and will be under the administrative authority of the District Commissioner and the District Medical Officer but will also receive technical guidance and supervision directly from the Central PHC Unit under the direction of the AEMS/PRC. Figure II details the structure of the planned PHC system. 3.27 Some fundamental questions arise regarding the organization and administration of the PHC program. The plan envisages an autonomous National Health Council on PHC that will execute its decisions through the specialized PHC Unit, which in turn will exercise technical direction over PHC staff and activities at the district and local levels. This proposed structure presents a strong likelihood of developing a parallel administration for primary health care that will be, to a great extent, independent of the normal line management of health services at the regional, district and local levels. In place of the proposed structure it is recommended that PEC be coordinated through a single coordinating committee or council at the central level, with ADMS/PHC as its secretary and reporting to the DMS or DDMS. This would integrate the administration of PEC into normal service channels and avoid any risk of duplication of effort or parallel management. 3.28 The National Plan for PHC (1982) is an ambitious proposal designed to develop community-based health care and extend curative and preventive services to 85% of the country by the end of the century.  -20- However, it seems likely that the large scale upon which PHC is projected and the broad range of activities under the program will be beyond the country's ability to implement effectively. Based on experiences in other countries a number of preconditions have emerged as essential to the successful conduct of a PHC program. These include a strong commitment to health services from the highest levels of Government, an adequate support and referral system, and a regular and dependable supply of essential drugs and supplies. 3.29 None of these preconditions are currently present in Uganda. The expenditure section of the report details the insufficiency of financial resources for recurrent costs (para. 3.57 - 3.58); the need for rehabilitation of health infrastructure was discussed above (para. 3.06-3.24); the inadequacy of drugs and supplies are discussed in the pharmaceuticals section (para. 3.48 - 3.51). In addition, the manpower development requirements, especially for the volunteer health workers, would place an unreasonably large burden on the MOH (para. 3.45). For these reasons, the final section of the report recommends that the PHC program be reduced in scope and focus. Immunization and Oral Rehydration 3.30 After reaching a level of approximately 250,000 DPT and polio vaccinations per year in the early 1970s, the imimnization program declined and virtually ceased in 1979 and 1980. With substantial technical and financial support, principally from UNICEF, the Uganda Expanded Program of Immunization was revived in 1981. The current program target is to fully immm_ize children in tiheir first year of life against measles, diptheria pertussis, tetanus, polio, and tuberculosis. By 1982 the annual number of immunizations reported was 255,000 measles, 710,000 DPT and polio, 390,000 BCG (tuberculosis) and 50,000 antenatal tetanus. Reliable data on coverage is absent, but WHO estimated that the following coverage had been reached by 1983: measles 22%, DPT (3 doses) 9Z, polio (3 doses) 8%, BCG 18%, antenatal tetanus 5%. These figures are an encouraging start for the new program, but the coverage remains low compared to other countries in the East Africa Region. A number of problems, including data collection and logistics, are hindering further program development. Incomplete and inadequate reporting is a major obstacle to program management. In 1982, immunization returns were reported for fewer than five months in 21 out of 33 districts. The majority of returns do not breakdown immunizations by age or number of doses received. To encourage better data collection, reporting forms should conform to WHO standards and data collection should become the clear responsibility of the DMO. The program also suffers from high dropout rates with over 60 percent of children receiving a first shot failing to complete the DPT and polio series. Insufficient motivation of mothers and unreliable logistics support have contributed to the low coverage and high dropout rates. These problems have been recognized by UNICEF and MDR officials. The EPI program has a semi-vertical status, through technical support from UNICEF, that should allow some independent progress in removing program constraints, but achievement of continuous coverage ultimately remains intertwined with other aspects of the improvement of health services. In the short term a change in administrative responsibility could reduce the efficiency of the program and the importance of program stability dictates that the structure remain  -21- unchanged over the next few years, but in the longer term the full integration of EPI into basic health services should enhance the effectiveness of both. 3.31 Given the high diarrhea morbidity rates and poor state of water and sanitation infrastructure, the benefit from a full Implementation of oral rehydration therapy should be substantial. UNICEF oral rehydration salts (ORS) are distributed to MOLG facilities, but the salts are not used as part of a well planned program of clinic based or home therapy following WHO or UNICEF guidelines. Some clinics, notwithstanding ample supplies of ORS, continue to treat diarrhea routinely with locally made intravenous fluid. The ORT program requires improved surveillance, retraining of paramedical staff and an extensive client education and information effort. A model program in the Mbarara area, using visiting district health teams, provides a useful prototype for extension to other regions of the country. UNICEF is supporting the further development of ORT as an addition to the immunization program. In the future, the ORI program will provide an essential complement to the EPI as an element of basic health services. E. Health Personnel and Training Current Health Personnel 3.32 The main categories of health personnel, including both Government and NGO services, are summarized in Table 3. About 10% Table 3: Principal Categories of Health Personnel for 1982 Category Number a/ Duration Current Population of Output per Health Health Per Year Worker Trainilng (years) Doctors 550 6b/ 70 25,000 Medical assistants 700 3 120 19,000 Nurses 6,000 2 1/2 cl 665 d/ 2,250 a/ Mission estimates based on reconciliation of WHO, Country Health Profile, 1983; GOU, National Plan of Action on PHC, 1983; and unpublished MOH data. bl Including one year of compulsory internship. C Registered nurses receive 3 years training, current output is about 45 per year. d/ 365 nurses, 300 midwives. of the doctors are expatriates, most of whom work in mission facilities. There are nine dentists. Disparities in the distribution of trained health  -22- personnel among districts are even greater than the uneveness in distribution of facilities would indicate. Over 40% of the doctors and 35% of registered nurses are located in the southern region with less than 30% of the population. Similarly, the distributUon of enrolled nursing staff disproportionately favors the southern and eastern parts of the country. The regional distribution of medical assistants, who are generally male and have a level of training roughly equivalent to a nurse, is more equitable; over 90% of MOLG health centers and 80% of MOLG maternity/dispensary units have one or more medical assistants. 3.33 In spite of the graduation of approximately 70 physicians per year, the number of physicians has fallen from 780 in 1969 to 550 in 1982 and the physician population ratio has fallen to 1/25000 compared to 1/12500 in 1969. The falling ratio results from the loss of large numbers of physicians during the 1970s and the low level of salaries and staff morale continue to make it difficult to rebuild the physician cadre. 3.34 The program to train and introduce the medical assistant into Uganda started in 1948. The three year training includes medicine, surgery, pediatrics, public health and limited obstetrics. In hospitals the medical assistant screens outpatients and, if necessary, refers them to a doctor. In health centers, maternity/dispensary units and dispensaries, he is the officer-in-charge. He diagnoses and treats most of the cases he sees, referring those he is unable to diagnose or treat to the district hospital. The number of medical assistants increased only slowly from about 400 in 1969 to 700 in 1982, with the ratio of medical assistants to population improving slightly from 1/23000 in 1969 to 1/19000 in 1982. The increase in output of medical assistants as a result of the new school at Gulu, will be one-third more starting in 1986. Over the years, this will improve the referral and back-up support for dispensaries, sub-dispensaries and aid posts provided that conditions, such as poor housing and working environment, causing high attrition of medical assistants can be improved. 3.35 In contrast to the physician and medical assistant personnel, the nursing cadre has more than kept pace with population growth. The total number of nurses, midwives, and nurse/midwives in 1982 was 6000, giving a nurse population ratio of 1/2250 compared to 1/3500 in 1969. The training of registered nurses and midwives is longer and includes more training in complex nursing and midwifery procedures than enrolled nurses or midwives. Health visitors (public health nurses) and assistant health visitors are recruited from the ranks of registered nurses and midwives and earolled nurses and midwives respectively. Each group undergoes one year additional training. The assistant health visitors run the MCH clinics in non-hospital facilities. The health visitor supervises the assistant health visitor. There are only 42 health visitors and 88 assistant health visitors. Less than 50% of the MOLG health centers and maternity/dispensary units have an assistant health visitor on staff. The importance of this work and the lack of staff trained in MCH further suggests the need to review the basic curricula for nurse education. A breakdown of nursing by category is not available but approximately 40% of the 6000 nurses are trained in midwifery. While there are acute shortages of nursing cadre trained to perform MCH and FP, the overall supply of nurses in the government service is much more adequate than the supply of  -23- doctors and medical assistants. In fact, because of limited equipment and supplies many nurses are not fully utilized. Accordingly, class sizes should be reduced over the next few years. This would facilitate teacher upgrading and the revision of the curricula to incorporate NCR and FP training in a substantial way for all cadres of nurses. A detailed review of all health manpower requirements is necessary to establish nurse training requirements in the future. Projected Manpower Needs and Training Requirements 3.36 Given the incomplete data available on numbers of personnel and attrition rates it is not possible to project manpower needs and training requirements accurately. A health manpower survey and assessment is urgently needed to provide a basis for planning. Applying tentative assumptions regarding target staff/population ratios, baseline numbers of health workers, attrition rates and training capacity, a simulation model was used to project the actual and required manpower for 1990 through 2000. As improved data becomes available, especially with regard to attrition rates and baseline numbers of workers, the projections (reported in paragraphs 3.37 through 3.45) should be revised and made on the basis of specific institutional and program needs rather than solely through population ratios. 3.37 Physicians. If the current training and work environment persists the number of physicians in 1990 will be about 750 compared to a requirement of 1400 in order to return to the 1969 physician population ratio, giving a shortage of 650; by 1995 the difference between required and actual would be 850. To close the gap between required and actual numbers of physicians by 2000 the attrition rate would have to decrease to .03 by 1990 and the annual output to increase to 165 by 1993. Given the difficulty of increasing the output of the medical school it is doubtful that the targetted physician population ratio can be achieved and rebuilding of health progam will need to be structured around a broadened role for medical assistants. 3.38 At the end of the 1960s the medical school at Makerere University was recognized for its outstanding faculty and facilities, but the expulsion of Asians from Uganda in 1973, followed by an exodus of expatriates, removed many seasoned teachers from the medical school and the size of the faculty fell from 97 in 1972 to 59 in 1983. The effective full-time faculty strength has been reduced by another 50% as faculty divert time to augment their low salaries through private practice or o3ther income generating activities. A school of dentistry was established in 1982 and 10 students were admitted to the course which has reduced further the resources available for physician training. Because of an acute shortage of medical school faculty and other resources the number of medical students admitted per year was reduced from 100 to 70 in 1982. 3.39 Economies of scale, as well as Uganda-s major shortage of physicians, require that admissions not fall much below 60 students per year, provided basic standards can be maintained. The standards of medical education have deteriorated substantially over the last decade. Standards can only be restored through improving the quality and number of the faculty and the quality of care in Mulago Hospital. Recruitment of faculty is one of the primary steps needed to rebuild the medical school. At the  -24- present time approximately 40 Ugandan physicians hold academic posts in medical faculties in Nairobi and Lusaka. While there have been a number of appeals to Ugandan physicians to return home, most have remained abroad because of the low salaries in Uganda, the social instability, the lack of housing, inadequacy of teaching materials, drugs, supplies and equipment, and the deteriorated state of Nulago teaching hospital. Many of these constraints to rebuilding the faculty and the medical school can only be met in the next 5 to 10 years through general rehabilitation of health services and other social infrastructure. 3.40 Nurses. The key nursing cadres are registered nurses (RN), enrolled nurses (EN), and nurse midwives. Nurse training is provided in seven MDH schools and nine NGO schools. Table 2.6 lists the nursing schools currently operating and the courses offered in each school. Table 2.7 lists the minimum entrance requirements and length of training by course. The majority of nursing staff in NGO hospitals are graduates of NGO schools and the output is essentially in balance with NGO requirements. NDH output is also projected to be adequate for forecast needs, in fact at present training rates a slight surplus is expected over the next decade, but the total projections hide maldistribution of the enrolled nursing category. Nationally, it is estimated that at the present time only 70% of health centers and less than 60% of MDLG dispensary/maternity units have at least one enrolled nurse. In contrast, hospitals are generally overstaffed with enrolled nurses. The curricula for both registered and enrolled nurses emphasizes curative care rather than preventive services and hence a large number of nurses will need retraining co support greater coverage with basic health services. Given the management responsibilities that are often taken by registered nurses, selected retraining of RNs in administrative skills is also needed. 3.41 Schools of nursing have not experienced the extensive loss of faculty that has severely affected the medical school, but the quality of training has been jeopardized by lack of qualified tutors, shortages of supplies, refresher training for faculty, deteriorated facilities and inadequate housing and food for students. In some instances hostel accommodations of nursing students have been encroached upon for the use of service staff. This is an especially acute problem in Kampala where there is a need to provide night shift workers with accommodation on the hospital grounds because it is unsafe for them to commute to work after nightfall. Attrition among nurses is not as great as for other health personnel with the exception of the health visitor cadre. Of the approximately 80 who have graduated since 1976 as health visitors only half are still working in this role. 3.42 Medical assistants. In concept, medical assistants comprise the primary diagnostic and medical care treatment cadre, particularly in rural health areas, but shortage in their numbers has affected service delivery in some rural units. Deficiencies in their training also restrict their effectiveness in the rural setting. Selected retraining is needed in MCH and FP preventive care, family planning and obstetrics. 3.43 Projections of medical assistants reveal that at present training rates, and an assumed attrition rate of 9% per year, there will be a shortage of 700 by 1990 and 950 by 1995. If the attrition rate falls to  -25- .05 and annual output increases steadily to 220 by 1989, the medical assistant population ratio should have improved to approximately 1/10000 by the late nineties. There are three training schools for medical assistants - one in the West (Fort Portal), one in the East (Mbale), and a third school In the North (Gulu) that enrolled its first students in 1983. The training schools in the west and east have been particularly affected by shortages of tutors, housing and supplies. In 1986 output for the three scbools should reach 180, or approximately 80Z of the annual output needed to achieve a balance between the required and actual number of medical assistants by 1995. 3.44 These shortages, covering the main categories of health personnel, give a quantitative measure of the effort needed to provide adequate numbers of personnel to extend basic health service coverage to the entire population. Improvement in the numbers of personnel can be brought about through rehabilitation and selected enlargement of training programs, but this in itself will provide only marginal benefit unless the high attrition rates are substantially reduced. Unfortunately, reductions in the loss of personnel through migration or change of employment are difficult to achieve in the face of current financial constraints, and involve improvements in wages, working environment, housing and transportation. In addition the effective number of hours worked from any given number of personnel is reduced by the current need to take work time off to earn additional income to supplement leve'ls of pay that are in many instances no more than subsistence. 3.45 Manpower Requirements for the PHC Program. Training, including the training of trainers, supervisors and primary health care outreach workers, is the basic component of the planned PHC program. Full implementation of the program would require: (a) reorientation courses at the central level for administrators and supervisors in the MOH, HOLG, NGOs and other affected agencies; (b) reorientation courses for selected district level MOH and MOLG staff; (c) training of trainers and supervisors (health visitors and assistant health visitors) for the PHC workers; (d) training of PHC workers; (e) training of traditional birth attendants; (f) incorporation of a PHC component in all health related training programs, such as medicine, nursing and medical assistants. The cost and recruitment burden of this training program will be considerable. The plan calls for the addition of 660 additional health visitors, 13,200 traditional birth attendants (3 weeks training) and 13,200 PHC workers (3 months training). The PHC project document assumes that the largest part of the cost of PHCW and TBA training will be born by the commtmity, but experience in other countries demonstrates that this assumption is unrealistic and large central government financial inputs will be needed to cover the trainer costs, the incentive payments and the supply expenses of PHCW and TBA training, and to offset the high attrition rates that can be expected among PHCW (who are community volunteers), TBA and health visitors. The training program in existence for health visitors will have to be more than tripled in capacity and the curricula modified  -26- extensively to include logistics management and teaching components if health visitors are to have the capacity to act as supervisors for the PHC workers. In addition the training programs for other cadres will have to be developed. Given the severe resource constraints faced in the health sector, the PHC program appears too ambitious at the present time. F. Health Support Systems Health Information 3.46 Health information is not sufficient to provide a foundation for management or planning. Records prior to 1972 demonstrate that a basic monitoring system had been in existence but that reporting procedures have broken down at both the central and local levels over the last ten years. Bank field trips encountered health centers, dispensaries and hospitals that had ceased to keep records and where visits from central monitoring staff seldom took place. At the local level one constraint is a lack of materials (paper forms and record books), but a more fundamental constraidt is that, through disuse, the procedures and record keeping skills of staff have deteriorated. Similarly, at the central level monitoring is constrained by a lack of vehicles, fuel and supplies, but an even greater constraint is the deterioration of procedures for gathering and analysing data. 3.47 Uganda is fortunate in having a tradition of record keeping in the past and a high calibre of district level personnel to build upon. Rebuilding of the information system should capitalize on this by starting with basic information regarding births, deaths, inpatient and outpatient usage, and environmental health statistics and building towards a more extensive management information system in the future. District level staff should be able to provide guidance and motivation to local staff and consolidate and forward data to the MDH. In turn, the MDH should provide feedback to district officers in the form of an annual report and periodic bulletins on notifiable diseases, and in-house news letters summarizing performance in selected program areas. Pharmaceuticals 3.48 Drugs are imported and distributed through three main sources: the Government Central Medical Stores (CMS), the NGO Joint Medical Stores (JMS) and the commercial importing agency, Uganda Pharmaceuticals Ltd. (UPL). Total imports are only US$1.00 per capita and are far from sufficient to meet the needs of the health system. Lack of drugs often results in patients receiving inadequate treatment even in some of the major facilities. The major constraint to increased drug availability is foreign exchange. The total value of drugs imported in 1982 was US$13 million, of which US$4 million was purchased by the Government, US$1.5 million was donated to the Government, US$3.5 million was donated to or purchased by NGOs, and US$4 million was purchased commercially. Thus, less than 31% of im,Lrts are obtained through Government purchases and there is little likelihood of any substantial increase given overall demands on Government foreign exchange.  -27- 3.49 The inadequacy of total supplies is compounded by problems that limit the effective distribution of pharmaceuticals. Poor central and regional storage facilities and inadequate management procedures coupled with the acute shortages have given rise to a black market and It is estimated that over 40% of CHS drugs are lost through spoilage or theft. The theft of drugs demonstrates the importance of the private health sector that provides a market for drugs. The motivation for theft may be related to the need to earn additional income in view of the extremely low levels of pay. Distribution of supplies is also hindered by a shortage of vehicles and fuel. Because of a lack of transportation at the CMS, individual hospitals and health facilities have had to make private arrangements for the delivery of drugs. 3.50 The Catholic and Protestant medical bureaus have recently established a joint medical store to facilitate the distribution of drugs donated to their member hospitals and to purchase drugs on a bulk basis. They also purchase a small amount (US$0.4 million) of drugs through the Government CMS. Catholic and Protestant hospitals are experiencing problems in arranging transportation to distribute drug supplies, but they have been able to overcome some distributional difficulties by pooling transportation resources. 3.51 Limited drug production was carried out in Uganda in the early 1970s in factories in Jinja and Kampala, and in some hospitals. In 1974 domestic production accounted for 8% of drugs distributed, but in 1982 output from factories was negligible and less than 1Z of drugs used in hospitals were produced locally. Consideration is currently being given by the Goverment to developing domestic production of both wet drugs (such as ointments, solutions, syrups) and dry drugs. After weighing expected difficulties in obtaining equipment, raw materials and packaging materials, against advantages, including savings in foreign exchange, transport and storage cost, a team of consultants evaluating the pharmaceutical sector for the Chrictian Medical Commission recommended that central production of wet products be developed but that dry production would not be ef" cient during the next five years. This evaluation appears apt and instead of developing dry production the MDH should concentrate on increased efficiency of drug management including improved procurement, prescription practices, storage and distribLution, and the development o' an inventory management Information system. Private Medical Care 3.52 Private practice of medlcine bas increased rapidly in the past few years because high Inflation and low Government salaries have encouraged many physicians into part time practice and because the deterioration of Government services has forced the population to look to the private sector for services. In theory, regulation of private medical services is a responsibility of the MDR, but in actuality the MDH has very little information on private practice and does not regularly carry out its surveillh.ce function. Data concerning the number of physicians in private practice is not available, but private clinics are seen frequently throughout Kampala and in major towns. A 1978 survey of households in five districts estimated that about 10% of households use private care for the treatment of childhood diseases. Information on use of private care by  -28- adults was not gathered In the survey, but studies in other countries demonstrate that a disproportionately high percentage of private care is used by adults. In many cases private practice is carried out without access to laboratory or radiological back up services. It is reported that some physicians, who run multiple clinics, illegally employ government trained medical assistants as proxies. Midwives have also established private maternity clinics, often in association with physicians. Growth of private care has filled a part of the health service gap that developed over the 1970s and should not be discouraged by unwarranted regulation. Some surveillance is necessary, however, to insure that minimum quality standards are maintained and government licensing and surveillance by district health inspection teams should be carried out. Traditional Medicine 3.53 A large proportion of the rural population use traditional practioners and traditional birth attendants (TBAs) commonly provide maternity services. This is partially because of the limited accessibility of modern health services, but is also due to the strength of cultural traditions and the perceived effectiveness of traditional medicine. Even in urban areas traditional medicine is often used as a supplement to modern services. Estimates of the volume of traditional care are not available, but the five district studies mentioned in the preceding paragraph found that approximately two thirds of households hold traditional medicine to be very important. The MDH should consider the potential for capitalizing on the strength of traditional medicine, possibly through retraining and IEC activities to develop TBAs as a vehicle for selected aspects of basic FP and MCH care. C. Health Expenditures and Finance Total Expenditures and Sources of Funds for Health Care 3.54 Estimates of total expenditures on health care in 1982 by type of service provider and by source of funds are presented in Table 4. Total health care expenditures (public plus private and recurrent plus capital) are estimated to have been U Shl2,084 (US$128) million or 1.8% of gross domestic product (GDP). Non-government capital expenditures are not available but are thought to have been small. Although data on household expenditures are also unavailable, in kind or monetary recurrent private expenditures on traditional and modern private care have been estimated at U Sh6310 million and U Sh2370 million respectively, or approximately 1.25% of estimated total private consumption expenditures. The estimated proportion for Uganda is based on the proportion of total private consumption expenditures spent on health services in several other African countries (for example, 1.4% in Lesotho, 0.8-1.5% in Rwanda, 1-2% in Zambia and 1.6% in Zimbabwe). This may underestimate the private health expenditures in Uganda however, given the partial collapse of government services and the increased reliance on the private sector from 1972 to 1982. Including public and private expenditures, the total per capita expenditure of U Sh 900 (US$9.5) is lower than for most other countries in the region vhere estimates of total private and public expenditures have been made (in 1982 prices: Zambia, US$38; Lesotho, US$12; Zimbabwe, US$38; Malawi, US$10; Rwanda, US$7). This intercountry comparison uses the official exchange rate of 94.5 U Sh/US$ and may thereby distort thne Ugandan expenditure.  - 29 - Table 4: TOM a3 AM SO19 OF EEATr4 CUME FMES 1982 (O a00o U Sh) service Providers _ _ _ _ FAI-1Direct HE Iocal Qvr et D/ Sjub PrIvate Don= Sorce of Funds Urban DistrIcts Total )fLBss SBctor ExpenditreE Total G,veat: Central 1504 50 117 1671 3 1674 Local 127 254 38L 381 Subtotal 1504 177 371 2052 2055 :mIrtay 87 87 Ini krd 10 10 Foreigia Dors C/ 282 282 190 472 Private Sector d/ ,m 42D 2V3D 3150 Traditimal 6310 6310 Total 1786 177 371 2334 520 9040 190 JZW. a/ i ude i capital e1edLuesb the Qoienumt and fozelpi donors. Tbere ls m ai for estimatinig cpital expeittmes by missis or the priate sector. Capital eqxpdures for urban ald distdct bealth projects tobl 15 i1llXi U.sb.; Ugand, Aproisd Estimtes of De1opi.nt Expwditure 1981/82 and 1982/83. Other srces are ited in tie tabs to 3. b/ nistry of 1ocdl orenuwuI records far a suwe of 23 districts mm anlyed awd extro1ato on a p capita bsi to the total fxw 33 distrIs. Urbn wespese mt 22 tm council and dt fow largst cities. Ctral g tribztIA to local . gPr, are anteited froa the avage cenal V,uemint No&g grant a optio of total evremies for the salae of 23 districts. C/ Fozelgi donwr fmding Is estiuted fa VW records. Brw reortin to U?W is ImWete md as donr ffEi m be CUitted. d M reiords of A_iof asstae are alo 1noxe Direct dowr assistance is eatited to be 4(X of totg al. d There are no bsold surveys ca to bse ma eatimte of private ezxlituxes. The estimate in the table is be fiad In other East AfrIc countries (Botsuan, Zaidsa Zimbabwe, Raada.). Exenditures audem p 1titmrs are eastttel to be 0.51, and eqi Stus ca traditional practi s are estinted to be 0.75% of total private cocxw  -30- 3.55 Of the total sources of funds used in the health sector in 1982, the Government provided 17%, missionb ,-rovided 1% and foreign donors 4%. The private sector was estimated to provide the remaining 78X. Of the total expenditures of funds, government services accounted for 19%, missions 4%, foreign donors 2% and the private sector 75%. 3.56 A serious problem in health services financing in Uganda is the underfunding of goverament recurrent expenditures. Insufficient funds for supplies, including drugs, and for maintenance have substantially reduced the use of government facilities, though this has not affected expenditures for staff. Government services are provided free. In contrast, patients using NGO facilities are charged fees covering approximately 80% of recurrent costs; yet NGO facilities are being used in excess of their designed operating capacity. The difference in demand between government and NGO facilities is attributable to the reliability and perceived higher quality of care offered by NGOs whose recurrent funding base, including patient charges, enables them to provide basic services only infrequently interrupted by a lack of essential supplies. Trends in Ministry of Health Expenditures 3.57 In nominal terms (Annex 3, Table 3.1) annual Ministry of Health (MOR) total expenditures have increased from U Sh 128 million in 1972 to U Sh 1,786 million in 1982, but the nominal increase has hidden a large decline in the real value of health expenditures. Deflated by a price index for gross domestic product and measured in 1982 shillings, total NOR expenditures in real terms (Annex 3, Table 3.2) have fallen by 80%, from U Sh 8,861 (US$93.8) million in 1972 to U Sh 1,786 (US$18.9) million in 1982. The decline results from a fall in the real value of both capital and recurrent expenditures. The extent of the decline is emphasized when the effect of population growth is included (Annex 3, Table 3.3); from 1972 to 1982 real expenditures per capita fell by 85%, from U Sh 861 (US$9.1) to U Sh 133 (US$1.4). 3.58 The financial constraints facing the Ministry of Health reflect a general reduction in government revenues that started during the 19708, was exacerbated by the civil disturbances of 1979, and has continued during the recovery era from 1980 to the present. Total government expenditures as a percentage of GDP (Annex 3, Table 3.4) fell from 22% in 1972 to a low of 32 in 1981 and then recovered slightly to 6Z in 1982. MOH expenditures as a percentage of total government expenditures have declined mDre mDdestly, from an average of 6% between 1972 and 1978 to 4X for 1981 and 1982. The combined effect of the reduced total government budget and the declining proportion of the government budget going to health has been to reduce MOR expenditures as a percentage of GDP from 0.9% in 1972 to 0.2% in 1982. Thus, central government support of health in Uganda is about one-half the 0.4% of GDP found on average in other low income countries included in the World Development Report. The current low level of the government health effort is a cause for concern in light of the fact that, prior to 1972, Government support of the health sector had been among the highest in Africa.  -31- Recovery Program Expenditures and Donor Financing 3.59 The original recovery program Included projects for the rehabilitation of Mulago Hospital Complex, rehabilitation of district hospitals, development of primary health care facilities, blood transfusion services and purchase of drugs. These five health projects totalled US$30.2 million (with US$5 million of this being for recurrent drug expenditures) or 28% of the social infrastructure funds and 4% of the total rehabilitation program, and were ranked number five through nine in a list of twenty soc_al infrastructure projects in order of priority. Planned capital expenditure for the health projects for 1982/83, the first year of the plan, was US$11.6 million but actual expenditures totalled only US$4.6 million or 40% of planned. The primary reason for the funding shortfall has been a lack of donor support In the sector. 3.60 The revised recovery program places a greater eKDhasis on the rehabilitation of social services with the allocated funds now comprising 20% of the plan as compared with 152 In the original program. However, allocation to the health sector has not increased proportionately. Within the social infrastructure segment of the program, health projects now total US$37.5 million (if recurrent cost for drugs are included) or 15% of the total expenditures for the social sectors. Annex 3, Table 3.6 gives the planned expenditures for bh-lth projects according to the revised recovery program, but as experience with the first program demonstrates, the intentions set forth in the recovery program are not meaningful unless supported by donor funding (see para. 5.23 for an estimate of the funding gap). 3.61 Donor support in the health sector revolves around recovery activities. UNICEF, with expenditures of US$2 milion in FY82, is the largest donor in the bealth sector and provides approximately a third of donor funds (Annex 3, Table 3.5). UNICEF funds and technical assistance have played a major role in the provision of drugs and support for basic maternal and child health services, particularly in oral rehydration therapy and immunization. Bilateral support in health came from DANIDA (US$1.2 million), Italy (US$0.7 million) and the Netherlands (US$0.7 million); missions contributed US$1 million. Distribution of Health Services and Expenditures 3.62 Information on the geographical distribution of services throughout the country is limited due to the collapse of the usage reporting system and the absence of monitoring and surveillance information. A measure of the uneven distribution of government expenditures by district was obtained by summning local expenditures, district-specific MOH expenditures and an allocation of central government expenditures in proportion to district facilities. NGO expenditures were added to Government expenditures to get total district expenditures. The results, summarized in Annex 3, Table 3.8, suggest that distribution of expenditures is inequitable. In Rumi, which benefits from large regional and district hospitals as well as mission hospitals, the total per capita expenditure is U Sh271 (US$2.87). At the other extreme, in Iganga district, the annual expenditure per capita is U SHIOI (US$1.07). Excluding Kampala district and mission expenditures, government  -32- expenditures vary from U Sh216 (US$2.29) in Kotido to U Shl6 (US$0.17) in Rasese. The uneven distribLtion of services is further evidenced by the variation in the bed/population ratio from 3.1 per 1000 population in Luwero to 0.3 in Kasese. 3.63 Maldistribution o! services would be further underlined if it were possible to identify institutions that are largely inoperable due to the sustained lack of maintenance over the 70s and to looting or var damage. Unfortunately, such damage is not fully reflected in recurrent expenditures because damaged units continue to receive an allocation of funds. Tentative adjustments were made to a 1980 MOH inventory of facilities to estimate bed availability for all 33 districts grouped on a regional basis (Annex 2, Table 2.6). The Southern region is well serviced with beds while the Western region is particularly deficient. In addition, some beds such as those in NGO facilities may be providing services to full capacity while other beds represent services that are available only intermittently and at less than full capacity. Without further physical assessment it is difficult to come to firm conclusions regarding the uni,ormity of quality and distribution of health services. 3.64 Distributed across functions, the allocation of health expenditures is not undesirable in the sense that it is less biased towards central urban expenditures than in other countries. The MOH employs a partial program budgeting system, including the use of program categories for ex post expenditures, that gives more information on the functional distribution of expenditures than is available in many countries but the information is not routinely used in management or planning and a detailed disaggregation of expenditures is not readily available. To obtain some insights into functional distribution, expenditures on medical supplies included under the highly aggregated category of 'medical stores- were arbitrarily disaggregated across service categories in proportion to base expenditures to arrive at the estimates of functional expenditure given in Annex 3, Table 3.9. Central hospital expenditures receive 19% of allocated funds. The percentage of funds going to the central hospitals is small compared with many countries where the percentage often exceeds 30-40%. Regional expenditures, obtained by totalling district hospital and regional health expenditures, are over 50% of the budget. Cost Recovery 3.65 The usefulness of recovering a part of operating costs directly from users is strongly demonstrated in Uganda by a comparison of the level and quality of services offered by NGO facilities where patient fees are levied, with government facilities, where services are offered free (see paragraphs 3.17 to 3.20 above). In many NCO facilities, fees constitute approximately 80% of revenues and allow continued replenishment of drugs and other supplies. Representative fees are U Sh5O per inpatient day, U Sh5O per outpatient visit, U Sh500-800 per delivery, and U Sh200-500 per minor surgical procedure. Full drug costs plus 25% (at joint medical stores subsidized prices) are charged. NGO experience does demonstrate that service demand decreases with increased prices, and issues of fairness and access would be raised by the use of government charges, but these issues are sumountable through a well designed system of charges.  -33- Cost Effectiveness of Health Services 3.66 I3hmost countries that have not undergone the civil trauma that has affected Uganda, a ranking of alternative programs In order of cost effectiveniess would strongly favor primary health care and rural health delivery systems. From the point of view of long-term planning, Uganda is not different in this respect, but the present condition of Uganda's physical facilities and associated recurrent cost probleim imply that over the next four to five years the cost effectiveness of rehabilitation expenditure and recurrent cost support for facilities including district hospitals and health centers is likely to be as high as many primary health care activities. The unusually high return expected from increased expenditures on facilities stems from two sources: (1) the current underutilization of capital and fixed personnel expenditures from want of conplementary operating funds; and (2) the inadequate maintenance of existing facilities which threatens the loss of capital in the future. 3.67 Mission field trips revealed extensive underuse of governmaent facilities at all levels - hospitals, health centers and health post-dispensaries. The primary reason for the low level of utilization is a lack of drugs and other supplies. Although data are not available to actually measure unit costs, the cost implications of suboptimum utilization in Uganda are so extreme that they can be clearly discussed que'litatively. Because a fixed expenditure on plant, equipment and pe ionnel establishment is divided by a small number of patients in underutilized government facilities, many operating at less than 25% of bed capacity, the cost per bed or patient treated is much higher than for facilities operating at designed capacity. The unit cost for NGO facilities with a high utilization rate, often in excess of OOZ of bed capacity, is much lower than that of government facilities because the flxed expenditure is divided by a larger number of patients. The expenditure per patient for NGO facilities is, however, slightly higher than would be the case if the facility were used at the designed capacity. This is due to the increased variable costs, especially tbose involving maintenance, associated with use above capacity. As a result, resources would be used more effectively (unit cost would be reduced as the level of use of government facilities increased and NGO facilities decreased) if adequate recurrent funds were available for government facilities. 3.68 Underutilization of government facilities is also partially related to poor maintenance and failure to provide minimum rehabilitation to war damaged equipment and buildings. Besides adding to the high cost through encouraging underutilization, the poor physical state of facilities if continued will lead to even higher capital expenditures for rebuilding in the future. For example, blocked sewage ln buildings without water my 'ave caused permanent deterioration of plumbing lines and fixtures. Unused steam equipment in laundries and kitchens has been dismantled (for the sale of parts) and is now being allowed to deteriorate further. Leaking roofs or physically damaged walls encourage further deterioration. An expenditure on maintenance supplies and training for maintenance is required to stop further, more costly, depreciation in the future.  -34- IV. NUrRITION A. Current Nutritional Status 4.01 In common with other countries of the region, women of reproductive age and young children are most affected by nutrition problems. Adult malnutrition and undernutrition undoubtedly exist, particularly in areas of social conflict; however no attempt has been made to quantify this aspect or to determine its effects on health, wellbeing and productivity. Protein-energy malnutrition is the most prevalent clinical condition, predominantly affecting the post weaning child with its high protein requirenents. Nutritional anemia, principally of the iron deficiency type, is a widespread problem affecting reproductive women but also occurring in young children. Other micro-nutrient deficiencies are infrequent and do not constitute major public health problems. Protein Calorie Malnutrition 4.02 Available data indicates that severe protein-calorie malnutrition (PCM), affects approximately 1% of children under 5 years of age, moderate malnutrition affects less than 5%, and mild malnutrition affects less than 30%. Considerable geographic variation in the frequency of protein-calorie malnutrition (PCM) arises from local agricultural and social conditions. In a number of small surveys carried out in 1981, moderate malnutrition was found to affect 15% of children in Karamoja (northeast), 4X of children in Ibale (southeast) and 2% in Kampala. There is also considerable temporal variation in frequency arising from seasonal weather changes and cropping patterns. In the south towards the height of the rainy season in April/May and in the north from May to October, the incidence of common infectious diseases such as measles, malaria and gastroenteritis reaches its peak. Concurrently, protein-rich foods such as beans ard groundnuts are in short supply before new crops have begun to yield. It is during this period that the reported incidence of PCM reaches its highest level. Records of hospital admissions show that both the magnitude and the severity of child PCM are greatest from April to July. Anemia 4.03 Severe iron deficiency anemia is reported to be increasing both in mothers and in children. No data on frequency is available though it was ranked as the sixth cause of death among pediatric admissions to 24 government hospitals in 1982. No age or sex breakdown of the data is available. Maternal anemia is observed "frequently" in central and district hospitals. The causes of anemia, in both mothers and children, cited in order of Importance are malaria, hookworm, diet and sickle cell trait. Short birth intervals also contribute to depletion of maternal iron stores with inadequate fetal storage during the last trimester of pregnancy. This leads to low hemoglobin levels in the post weaning period. B. Causes of Malnutrition Infection 4.04 The collApse of the immunization program in 1979/80 led to an increase in common infectious diseases of childhood, of which measles was the most prevalent. Measles now accounts for 28.5Z of all hospital deaths, compared to 5% in 1970; the deterioration in environmental sanitation  -35- measures has led to an increase in gastro-intestinal disease and a breakdown in control measures has resulted in an increase in malaria prevalence. These three conditions are among the top four causes of death in hospital admissions for children and have undoubtedly contributed to a deterioration in nutritional status. Social Disruption 4.05 Widespread social chaos has taken a serious toll on the health and nutritional status of the family, with women and young children particularly affected. Broken families arising from recent social disturbances have also contributed to the increasing prevalence of malnutrition. In recent years, children from broken homes have represented 372 of admissions to the Mwanamugimu Nutrition Clinic at Mulago Hospital. In areas of political dissidence, entire comuiunities have been prevented from planting or harvesting essential food crops resulting in wiidespread chronic malnutrition. Survivors are rounded up by the army or volbntarily seek admission to displaced person camps in the central area. Survey of these camps shows very high levels of malnutrition (35Z+). A considerable number of malnourished children are taken to Mulago Hospital by relief workers for admission to special wards. During a six week period in July-August 1983, a total of 411 children, 50% of whom were between 5 and 9 years of age, were admitted with severe clinical malnutrition: 54% kwashiorkor (protein malnutrition), 14% marasmus (calorie malnutrition), 20% mixed types and 2% severe anemia. Most of the marasuus was due to breast feeding failure resulting from maternal malnutrition. Cultural Practices 4.06 During weaning, children are particularly vulnerable to nutritional hazards. Weaning foods, such as plantain, cassava or potatoes, that are introduced between 3-6 months, are usually of low energy density and have a low protein content. Insufficient protein intake (less than 1.5 g/kg per day) compounded by low caloric consunption results in protein catabolism in order to meet energy requirements. Typically, breastfeeding is abruptly terminated when the mother becomes pregnaut again and toddlers are sent away to relatives. This is due to the belief that the presence of the embryo may advJirsely affect the breastfeeding child through the mother's milk. The consequences more often than not, are emotional and physical disturbances in the newly weaned child, resulting in a deterioration in their general health and, ultimately, in malnutrition. Agricultujral and Economic Factors 4.07 The production of food crops in Uganda increased steadily during the 1970's until the civil war in 1979. Available statistics indicate that during the period 1970/80 there was a sharp drop in national food production (Annex 4, Table 1.1). Compared vith the mid 1970's, production levels of cereal production dropped by 32%, plantain production by 36%, pulses by 52Z and oil seeds by 60x. Root crops declined much less (13S), and in some areas increased. Such crops (cassava, sweet potato) are important food emergency resources as they require minimal agricultural inputs and are able to survive unfavorable environmental conditions. With the return of political normalcy and the subsequent decontrol of food  -36- prices, food production has increased steadily since 1981 and in some cases (millet, sorghum, rice, sweet potatoes, cassava) has met or exceeded 1978 levels. 4.08 There is wide consensus that total food availability did not decrease on a national scale during the economic disruption of the last few years. This reflects reversion to subsistence-type agriculture directed at producing sufficient food to meet family needs with the existing surplus marketed for profit. Almost everyone in Uganda has direct contact with a farm family. This capacity to revert to or increase small scale food production has enabled the population to survive, in nutritional terms, what would otherwise have been the disastrous consequences of social upheaval. However, the reported production of many cash crops dropped to levels representing only a fraction of their former value as the population switched to food production for survival. FAO food balance sheet estimates for 1982 indicate that the available food supply was sufficient to meet 83% of the energy requirement of 2070 kcal per capita per day and 91% of the protein requirement of 48g per capita per day. There is general agreement that the current food supply, taking into account the recovery in production since 1981, is now (in 1983) sufficient to meet energy and protein requirements on a national basis. 4.09 Although overall food production may be adequate, uneven distribution of food results in surpluses in some areas and acute shortages in others. Price data is of poor quality but the extreme variation in average rural prices for staple food crops is an indication of maldistribution. Prices for specific crops vary as much as five-fold between different areas of the country (Annex 4, Table 4.2). The nutritional value of the available food supply also varies greatly from area to area. The degree to which local production of major crops meets the nutritional needs of the area population varies from 66Z to 277% of per capita energy requirements and from 57% to 339Z of protein requirements (Annex 4, Table 4.3). Agricultural sector analysis is required to identify problems in market performance, but a possible explanation lies in transportation and storage problems. C. Nutrition Activities and Policies 4.10 The major nutrition activity in the health sector is the operation of surveillance and rehabilitation services at the district level. These services are based on nutrition units located in the pediatric sections of regional and district hospitals and linked with community outreach activities through the use of local volunteers. Five of the fourteen units established are currently operative and receive donor support. UNICEF is the pri2cipal donor agency for nutrition and currently plans to restore 110 health centers as bases for surveillance and rehabilitation activities. World Food Programme provides food assistance for emergency and development projects in the Southwest, Central and West Nile regions. Other nutrition-related activities directed at food supply and education are carried out by the Ministries of Agriculture, Animal Resources and Fisheries, Cooperatives and Marketing, Culture and Community Development, and Education. Details of nutrition activities are provided in Annex 4.  -37- Food and Nutrition Policy 4.11 Existing nutritional surveillance and rehabilitation services are limited and are based in static facilities at the regional and district level. The absence of a community-based health care system on a widespread scale prevents the delivery of nutrition services in a systematic manner at the local level. The lack of a food and nutrition policy at the national level results in minimal coordination between sectors, particularly those of health, education and agriculture, and the failure to develop a coherent strategy to resolve nutrition problems in the longer term. 4.12 The appropriate body to formulate nutrition policy is the National Food and Nutrition Council. It is suggested that the Council should report to the Office of the President to ensure effective implementation of sectoral recommendations and to facilitate its role ae an inter-sectoral coordinating body. The present chairman of the Council is the permanent secretary of Agriculture; given the crucial importance of the agricultural sector in policy making, it is recommended that the permanent secretary of Agriculture be made ex-officio chairman for a substantial period (five years) to assure continuity at the outset. The present composition of the Council should be reviewed and revision in membership undertaken to reflect the current range of agencies and institutions involved in nutrition in Uganda, including the NGOs. 4.13 Nutrition Emergencies. Nutrition emergencies arising from natural causes and social disturbance have been frequent in Uganda over the past five years. To a great extent these emergencies have resulted, not from a lack of food at the national level, but from an inadequate distribution and marketing system combined with insufficient warning of impending local scarcities. To correct this situation, inter-regional food marketing should be stimulated and local and national food stocks developed. Marketing, storage and transport of agricultural goods are dependent on the agricultural and transport projects outlined in the Revised Recovery Program. Effective agrometeorological surveillance should be reinstituted as soon as possible, particularly in high risk areas, to provide early warning of crop and livestock production failures. Contingency planning for food emergencies should be carried out under the direction of the Food and Nutrition Council. 4.14 Endemic Malnutrition. A program of nutrition education, directed particularly to women of reproductive age, is fundamental to the effective elimination of endemic malnutrition. Such education would have to be provided through a range of services including the formal education system, extension services in agriculture and home economics and the primary health care system. The revision of the health sciences curriculum for primary and secondary schools currently being undertaken by the Ministry of Education should place special emphasis on this topic. The trial first edition of the booklet -Child Nutrition Guidelines" prepared by the Health Education Section of the Ministry of Health in collaboration with UNICEF should be carefully scrutinized at all levels of application and the revised edition widely circulated to all relevant sections of government and non-government organizations. The development of a health education manual and ac'ompanying teaching aids for health workers, primary school  -38- teachers and community leaders that will be undertaken as part of the 1983-86 program of UNICEF should also emphasize this problem topic. 4.15 The reduction of endemic malnutrition will require continued short term measures based on surveillance, referral and rehabilitation. It will be essential that the network of nutrition rehabilitation services, linked to community surveillance and screening activities, is restored and maintained. The Ministry of Health, in collaboration with UNICEF, is already planning to strengthen the capacity of the fifteen existing rehabilitation centers by providing food (with the World Food Program) and basic equipment for preparation, feeding and assessment of recovery. To complement these programs the National Food and Nutrition Council, together with the MOH, should consider the incorporation of nutrition education into the existing framework of health personnel education with emphasis on appropriate treatment of diarrhea (including the preparation and application of ORT) and the preparation of weaning foods, and the development of an IEC strategy to support these OL]; and weaning food activities. V. ISSUES AND RECOMMENDATIONS 5.01 Because of the extent of deterioration of health services and management systems and the slow progress towards rehabilitation, the development of priorities and sequencing of recommended actiors is especially critical in Uganda. The priorities are based on the relative gains in well being and health status but also, crucially, on the appropriate order of steps in program execution dictated by a consideration of program interlinkages. Thus, the overriding priority is the rehabilitation of the health infrastructure that forms the basis for the delivery of health, population and nutritional intervention programs. But a number of steps, including an inventory of health services, development of planning capacity, development of information and monitoring systems, a manpower planning study, and design of a system of cost recovery, are needed before substantial and durable progress can be made towards rehabilitation of services. Additionally, interim measures, such as strengthened MOH, FPAU and NGO population activities, continued eupport o' the MOH/UNICEF emergency program of immunization and oral rehydration, restoration of water and electricity to health centers and district hospitals, and emergency training for maintenance are needed. Nutrition recommendations are detailed in Chapter IV (paras. 4.12 - 4.15). The critical steps to be taken in the short term (from the present up to two years) and medium term (two to six years) in the population and health sectors are outlined below. A. Population 5.02 At the current population growth rate of nearly 3.2Z per annum, the population of Uganda will double by the year 2005 and will nearly quadruple by the year 2025. Given the existing agricultural potential, this population growth could be accommodated if consumption were to be maintained at the current subsistence levels. Nevertheless, the rapid rate of population growth will impede Uganda's economic development and attempts to move beyond low agricultural productivity by diverting investment funds  -39- needed for agriculture, job creation and long-term economic recovery to the consumption and social infrastructure needed to support a larger population. The current rate of population growth, rather than the ultimate population size, is a constraint to development; population growth needs to be slowed if the standard of living of the Uganda people is to recover and then increase in the future. 5.03 A decrease in fertility is required if population growth is to be reduced. Without government action, all of the three componenents of population change-mortality, migration and fertility-can be expected to contribute to higner rates of growth. With increased effectiveness of health programs and a return to civil stability, a fall in mortality can be expected over the future. Civil stability may also result in the recommencement of the immigration of the 1960s. Fertility is unlikely to fall significantly through response to socio-econonic change alone in the next 10-15 years. Fertility reduction can only be achieved with government understanding of the problem and the development of a strong program of family planning services. Population Policy and Strengthening of Family Planning Services 5.04 The Government's adoption of an explicit population policy in 1981 recognizes that population growth is a major factor in determining the long term economic prospects for Uganda but there have not been significant efforts to achieve the policy objectives. The most critical population issuo is whether sufficient public support can be cultivated to implement an effective population program. The current contraceptive prevalence rate of about 1Z is far from the level necessary to achieve the government goal of a 2.6Z growth rate by 1995. To reduce the growth rate to the target set by the Government would mean that the prevalence rate would have to increase to more than 14Z. Constraints in implementing an expanded government program at this time, however, are substantial. The constraints include not only a lack of public interest in population policies but a lack of sufficient supporting health infrastructure. 5.05. Although implementation of a comprehensive program through government services requires revamping of the health delivery system as a first priority, a number of activities can be started Immediately to increase public demand and provide services. Greater goverpment commitment can be stimulated through seminars for high and middle level officers conducted by the MOP and MOR and through economic analysis by the MOP of the consequences of population growth. Public support and service demand can be stimulated through incorporation of population material into educational curricula, and through the further development of an IEC program to be Implemented jointly by the FPAU and MOH. Services should grow concomitantly with the IEC program. 5.06 An interim population program should be adopted gradually over the course of the next two to six years. As part cc the interim program a range of government and non-government channels of distribution for family planning, including the FPAU and NGOs should be encouraged to make  -40- contraceptive services increasingy available on a national basis, especially as a meaure of maternal bealth. The interim progrm would provide the basis for futcre program expaion by formal training in population and PP for mbers of the madical sebool as mell as nurse and medical assistant schools. The curriculum of these scbools should increasingly liclude PP subjects. As a subsequent activity, FP training progrin should be m ted for health workers and 1P training ald services should be included as a component in pri_mry health care. VP services in NMO and NGO facilities could then be extended as staff become trained. To promote these activities, the population unit In the MOP should be strengthened, an PKICIP trainiug coordinator should be appointed, an IEC strategy and program should be developed and the FPAU management capacity should be strengtbened. 5.07 The establiabsent of baseline population data is a prerequisite to better planning of all government services and development programs. A baseline fertility survey is particularly needed to guide the formulation of family plansing and MCR programs. USAID is expected to fund a fertility and contraceptive prevalence survey in 1Y 86. Also, to fill part of the void left by the disappearance of the 1980 detailed population census form, an interi sample survey should be conducted about 1985. Because of the high cost of a full census and current priority needs for the resources that would be involved it is not advisable to move the 1990 census forward, but a comprebensive national sample survey, such as that being planned for UMPFA funding 1n FP 86-87, is required to provide planning information about age structure, fertility, migration, education, employment and income. B. Health 5.08 Development of a health strategy that goes beyond the current plans for recovery is critical for the future of the health sector. Development of such a strategy and fiproved planning capacity in the NOR is a logical prerequisite to the priority projects for health sector rehabilitation. The recovery program has correctly identified the crucial areas for support to bring about the rehabilitation of the health sector, but within the activities outlined by the plan there is considerable flexibility to shape specific projects. Additional details are needed to ensure that project desigs are consistent with long term goals for the health sector. A health strategy would provide sector objectlves and establish guidelines to coordinate resource allocation with the epidemiological situation, mnpower availability, and long term recurrent cost capacity. The strategy would also provide a basis for coordination of Government and NGW activities. Because a lag can be expected in the development of planning capacity, a stsrt is needed now if a strategy is to be specificed in time to shape the recovery effort and direct the flow of resources in the last half of the decade and beyond. 5.09 Consistent with the health strategy, a medium term (3-5 years) plan should be formslated. The plan should detail the geographic and functional distribution of expenditures and give a careful consideration to  -41- the recurrent cost implications of capital expenditures. Priorities should be established on the basis of demographic, epidemlological and operational information irncluding a specific consideration of cost effectiveness and distributional objectives. In the next few years the detailed medium term plan would relate closely to the recovery programme. The medium-term plan should be updated annually on a rolling basis. 5.10 An important complement to both planning and management would be development of a data base to assess morbidity and mortality and evaluate the location and mix of health services. Baseline sample surveys should be carried out to establish disease incidence as a benchmark to allow strategies to be modified and to measure progress. Regular reporting of service statistics and disease prevalence needs to be re-established to form a basis for management. As the capacity of the MOH to andertake planning, monitoring and evaluation develops over the next five years, gradual steps should be taken to move towards the eventual introduction of a management information system integrating service usage and budget data. Such a system would take advantage of the system of budgeting already in place, and the high quality of district personnel (who would be responsible for the flow of information needed for monitoring and short term planning), to allow the application of performance measures to determine the best allocation of resources to achieve planning objectives. Hospitals and Other Health Care Facilities 5.11 The restoration of health facilities and the repair and replacement of equipment under the Government's recovery program is of highest priority and mtst be addressed in the short to medium term on a coherent, planned basis. The work of assesslng the state of repair of health centers and other non-hospital facilities and some of the training facilities, being supported by UNICEF and CIDA respectively, needs to be expanded and should include hospital facilities. Preferably this would be carried out in conjunction with the NGOB and include a mapping exercise to assess health manpower supply and the distribution of health manpower and facilities. This will allow priorities to be developed, taking Into account possible savings that might accrue in instances where unnecessary duplication exists. It is particularly important for the NGOs to be involved in these and other planning activities because of their major contribution to the provision of health services in the country. The program of plant and equipment restoration should be accompanied by a program to train and deploy staff in plant and equipment maiuntenance so that investments will have maximum long range beneflts. 5.12 Efficient levels of usage of facilities will not be achieved unless an adequate and reliable flow of supplies reaches the lower levels of the health system. Increased quantities of supplies are needed, but this will not be sufficient of itself. Donors are reluctant _o finance drugs and supplies unless a secure and reliable logistics system is developed. Improved procurement, inventory control and management, security, storage and distribution of drugs and supplies are required with high priority. Short term technical assistance is needed to carry out a study of logistics and make recommendations for Immediate action.  -42-- Primary Health Care 5.13 Several constraints facing the health sector would prevent the effective implementation of the PHC program as it is currently envisaged. The major constraints are insufficient funding of recurrent costs of the health sector, inefficient and centralized management and administrative procedures, inadequate logistics, and transportation support, and an insufficient supply of essential drugs. Added to these problems is an inadequate referral system for backup services. Experience with PHC programs, as it is beginning to accumulate globally, demonstrates clearly that removal of these deficiencies is critical to the success of a PHC program. 5.14 The current PHC plan raises some fundamental questions regarding the organization and administration of the PHC program. It envisages an autonomous National Realth Advisory Council that will execute its decisions through the central PHC Unit, which in turn will exercise technical direction over PHC staff and activities at the district and local level. This proposed structure presents a strong likelihood of developing a parallel administration for primary health care that will be to a great extent independent of the management of the rest of health services at the regional, district and local level. Instead of the proposed structure, the adiinistration of PHC should be made an integral part of health services with a central PHC coordinating committee reporting to the DMS to be established to ensure that the program is developed effectively. This would integrate the administration of PHC into normal service channels and avoid any risk of duplication of effort or parallel management. 5.15 . The practicability of the plan would be enhanced by shifting the emphasis from the training of a large number of unpaid volunteers to a smaller number of paid paramedics and assistants. Given the high turnover experienced elsewhere with unpaid vclunteers and the large number to be trained in the Uganda program, the training costs would be beyond the capacity of the MfH to maintain an operational program. 5.16 The training curriculum and proposed routine activities of the PHC worker understandably place major emphasis on their preventive and advisory role. It is nevertheless evidenced both from the field visits of the Bank team and the evaluation of the Ugandan Red Cross project, that rural commmnities are clear as to their priorities: they want prompt and effective treatment of existing ailments with modern drugs and procedures; furthermore, they will circumvent nearby health facilities and pay substantial cash amounts in order to obtain these. The primary health care worker and the back-up systems that are provided must be able to meet this demand effectively if promotional and preventive activities are going to be accepted and applied.  -43- 5.17 Before proceeding further with implementation, the plan of action should be modified in scope and simplified in concept to provide for the development of PHC services in a phased manner over a limited geographical area consistent with the ability to train supervisory staff and field workers effectively and the availability on an assured basis of equipment and supplies. The major target of the program should dovetail with the current UNICEF and WHO focus on immunization and oral rehydration. Phased development and limdted scope would permit an evaluation of the feasibility and effectiveness of the proposed service package and allow a more accurate assessment of needs and resources for the planning of subsequent expansion at a later stage. Phasing would also allow the development of the program to be coordinated with rehabilitation of the referral and logistics systems. Organization and management 5.18 Decisions within the NOR are highly centralized. However, central level resources are largely occupied with response to iainagement crises in logistics, supplies, training and maintenance. As a result, there is insufficient coordination of lower level activities and little response to service needs at the field level. Improved capacity to execute and coordinate programs and to handle crises within an organized framework is essential. 5.19 In response to tiiese problems a management study of the MOR is needed to provide guidance for reorganization and improved managerial procedures over the medium tern. The study should consider the feasibility and design of a decentralized system of budgeting, planning and management. Improved management should build upon the degree of independence of district units that has developed during the last few years of decreased central surveillance. The responsibilities of central level department and senior level staff should be clearly defined. Administrative efficiency and monitoring would be enhanced if the job functions of management units were defined on an operational basis. Ultimately, functional rationalization of ROR organization and decentralized mnagement would allow linkage of planning and budgeting as a regular part of MOR institutional procedures and routine reconciliation of actual expenditures and health service usage. The study should also consider the unification of all health services under one ministry rather than the separation of higher and lower level services between the MRE and MOLG as is presently done. As health services and supporting systems are strengthened in the future, unification of services will add to the efficiency of management. Health Personnel 5.20 The main issues related to health personnel are: (i) personnel planning; (ii) staffing patterns for various types of facilities; (iii) distribution of personnel; (iv) insufficient numbers of personnel of various types; (v) curriculum content, and (vi) staff remneration. The most pressing need, linking all of these issues, is for a comprehensive health personnel arnalysis that would take into account appropriate staffing patterns for various types of facilities and the needs of a revised PHC program. Such an analysis should be followed by the developmnt of a long term personnel plan. The recent manpower study by an MOH task force does not provide a comprehensive analysis and particularly needs to be  -44- supplemented by a detailed inventory of personnel and projections that can be used as a basis for development of training programs. A single unit of the MOR should be designatad to coordinate personnel development activities and to implement the personnel plan. 5.21 Existing patterns of staffing appear to be inequitable on a geographic basis and may be inappropriate for the level and type of treatment feasible at lower level facilities. More information is needed to assess the adequacy of facility staffing on a functional basis, but geographic distribution of personnel is clearly inequitable. The shortage of certain categories of health manpower, especially nurses and medical assistants, in the rural districts will need to be addressed through greater availability of staff housing and other incentives. 5.22 In lieu of a comprehensive analysis, Bank mission projections (summarized above in paragraphs 3.36 to 3.45) demonstrate the need to expand training and reduce attrition of MDs and MAs. Given the constraints to expanded training of MDs and the long lag between increased MD training capacity and output, the number of medical school graduates is expected to be insufficient to meet requirements in the near future. However, it would be unwise to increase the enrollment of medical students at Makarere without raising the quality of medical education. This not only requires additions to the faculty but the upgrading of the standard of care at the teaching hosk ital. While the motives for establishing a dental school at Makerere cannot be questioned, it will diminish the resources available for medical training and is inappropriate in the light of existing priorities. 5.23 There should be retraining of selected MDs, registered nurses, and medical assistants in management, and training course syllabi for all levels of staff should include MCH activities and emphasize preventative care. Retraining of medical assistants and administrative nursing personnel is needed to introduce an effective surveillance system. In-service training in FP (see also para. 5.06) and in preventive and simple curative measures for maternal and child care, is needed for all lower level service personnel. 5.24 The rationalization of staff establishments and remuneration are essential ingredients if work productivity and dependability is to be improved. Immediate steps are needed to bring about a more rational remueration policy if rehabilitation is to be effective. A large increase In the total wage costs is not possible, but within that constraint rationalization could result in smaller numbers of staff with increased average salaries. A 1983 Bank study of public administration recommended a reduction in staff and the number of geographical administrative areas and the reform of working hours to increase personnel efficiency and incentives 4J. These recommendations have not been acted on in health. Many hospital workers are unable to put in a full day's work at their primury job because their low wages compel them to earn additional income. Exceptions have already been made in one or two government units to increase remuneration to overcome this constraint. Similar exceptions need to be made for key health workers. 4/ R. Liebenthal, Public Administration Study Back to Office Report, World Bank, March 22, 1983. Also O.A. Meesook, Public Sector Pay in Uganda, World Bank, mimeo. May 16, 1983.  -45- Open recognition of present practices will enable the rationalization of the system and allow adequate staff rosters to be developed that ensure at least minimum coverage of facilities round the clock. 5.25 The MOH and Makerere University are presently drafting regulations to bring needed control to part-time practice. Such practice constitutes a conflict of interest-it is not permitted in NGO hospitals- and, in the interest of health services, provision should be made for substantial non-private practice allowance to faculty members and medical officers who devote their full-time professional activities to their university or government work. Financing Issues 5.26 Resource allocation in Uganda will continue to be determined over the next few years by the requirements of rehabiltation and restructuring of the economy. Given the limited financial resources, the rehabilitation activities ase not only mutually competitive for capital within the recovery program, but also compete for recurrent funding of ordinary government operations. The shortage of operating revenues can be expected to continue over the next two to five years. Within the health sector in particular, the shortage of revenues calls into question the government policy of providing health services without charge. The first section below discusses the allocation of funds within the recovery program. The second section gives the result of projections of NOR costs and available funds and draws implications for future recurrent costs requirements. The final section considers the possibilities for cost recovery from the client population. 5.27 Recovery Program Financing. The choices of health projects and priorities in the Recovery Program are fundamentally correct. Given the striagent economic conditions and the need to get Uganda moving forward once again, the program is consistent with a dynamic view of welfare maximization. But the minimum attention to the health sector outlined in the recovery program is essential, and it should be underlined that the experience to date under the recovery program has been that, due to lack of funds and sufficient managerial capacity, the social sector activities are not being implemented and have been neglected relative to other sectors. It is essential that, as an immediate priority, funds be allocated to restore minimum government health care services as outlined in the recovery program. 5.28 The included projects (summarized above in paragraph 3.08) are consistent with the recommendation in the preceding section and, when fully defined, the projects could include the activities suggested here for minimum rehabilitation of facilities, development of planning capacity, management and maintenance training and selective primary health care. Briefly, the projects in the RRP encompass rehabilitation of the Mulago, and district hospitals (including water and sewage, electricity, equipment, repair of facilities, and medical supplies); primary health care (including accelerated immun1zation services); health training; and health planning. The total cost of these projects is estimated to be US$27.8 million (see Annex 3, Table 3.7), but of this amount only US$ 14.6 million has been  -46- secured (including US$13.6 million from foreign donors and US$1.0 million from GOU) leaving a funding gap of US$13.2 million or 50% of the total. 5.29 Recurrent Cost Funding. Insufficient funding of recurrent costs is one of the major problems facing the health sector and is a basic constraint to carrying out the recommendations outlined above. Minimum estimates of the supplementary current expenditures needed for drugs and maintenance over the period 1985 to 1989 have been added to a projection of baseline expenditures and recurrent cost for the recovery program to obtain a conservative estimate of required recurrent funds. The projections are conservative because of the low level of per capita drug expenditures and maintenance expenditure per hospital bed and facility used in their formulation (see Annex 3, Table 3.10 for details). When recurrent funding needs are compared to a projection of available funds over the period from 1985 to 1989, the result is a total shortfall of U Sh28,600 (US$102) million over the five year period. Potential sources are the Government fiscal authority, donors, reliance on the private sector, and cost recovery from service users. A larger government tax effort is needed. When the present level of government funding is compared with previous years, it is apparent that larger government revenues are feasible, but greater social stability and progress with the economic recovery program are required to provide the practical base to carry out tax reforms. In the interim period it will be necessary to continue to rely on donors to carry some of the burden of recurrent costs, especially those such as drugs involving foreign exchange. But donors, especially UNICEF, are already providing support of selected recurrent costs in health and this source is unlikely to grow substantially in the near future. Over the last ten years the public has grown increasingly reliant on the private sector, including NGOs, to provide health services. The government should continue to support the use of private sector facilities, and plans for development in the health sector, and donor efforts, should recognize the dependence on private services. Finally, the urgency of finding additional sources of recurrent funds requires that the government reexamine its current policies on cost recovery for government health services. 5.30 Cost Recovery. The use of charges for selected health services can be justified (1) as a means of raising additional revenue to cover operating costs; (2) as a means of improving the efficiency of resource allocation; and (3) as a means of improving the equity of health service use. The major justification for fees in Uganda is as a revenue measure but efficiency and equity are also considerations. In general, efficiency is served if the price paid by the user reflects the additional benefits to society from consumption of the service and the additional cost to society of producing the service. Services provided without charge may be consumed beyond the point of economic optimality, i.e. the marginal cost to society exceeds the marginal benefit of the excess consumption. Equity is served if higher income consumers are required to pay the cost of services they receive while basic services received by lower income households are free or subsidized. A well designed system of fees coordinated with a properly functioning referral system can allow cross subsidization to promote a more equitable distribution of services and should be developed over the medium term. 5.31 These simple criteria can be used to describe a possible core list of services to be provided free in the context of Uganda. Because  -47- preventive services have substantial benefits that accrue not only to the individual receiving the service but to society at large, user charges paid by individuals would result in a level of consumption that is less than socially desirable. Services of this type should be provided without user charges and include (1) most disease control programs such as malaria, communicable childhood diseases, sexually transmitted diseases, leprosy and tuberculosis; (2) antenatal care; (3) well baby services (growth monitoring, immunizations, instruction in oral rehydration therapy); and (4) family planning services. 5.32 Beyond the core group of basic services, there are considerable possibilities for cost recovery that could be implemented over the next few years. A schedule of charges for drugs and supplies and charges for non-referred use of physicians and higher level care could be drawn up. In addition to providing revenue, receipts and accounting required for cost recovery from the sale of drugs would provide a means of inventory control to reduce theft. Pharmaceuticals for preventive care could continue to be provided free of charge but other pharmaceuticals would be provided at full cost or subsidized cost according to government schedules. Nominal fees could be levied for first visits to outpatient clinics to discourage unjustified use of services. Also, referral to physicians and higher level care could be allowed without charge but patients who desired to enter the health system at levels above primary health care and who are not emergency cases would be required to pay fees. This would help prevent misuse of higher level facilities and would shi:t a part of the cost of higher level care to upper income groups that are more apt to try to circumvent entry into the health system at the primary health care level. 5.33 In order to be effective as a revenue measure and to ensure regeneration of operating supplies, the fees should accrue as close as possible to the levels of the collecting agent, preferably at the district level or below. One efficient possibility would be for health posts or health centers to use collected fees to directly replenish a district level revolving fund for drugs and supplies. Auditing and monitoring would be carried out at higher administrative levels but the actual administration of the fund would be carried out at the facility level. The effectiveness of a system of cost recovery and the efficiency of collection would be seriously jeopardized if the fees accrued at the central level to the Ministry of Finance.  I_1 DDIIAIl4 w t0mWL wiIm.u ..~ ~ ~ ( ,W" .W WA1H.AW mmdiv ihith Dl TA.Hlt k b,' So bith lkta U ih lt aua WI MUc ae Dnvlc Ssrvm U ihkth Mm 4 b~bt W Mari kopectom aitsdm &nt?m QM*Daeiada Tum ustrit Usict MD to Md tual gt 10 t to NW hith sri"  -49- Figure II: STRUCTURE OF PRIMARY HEALTH CARE Minister of Health Deputy Minister of Health j Inter-Ministerial/ ! ku~~~~~flti-sectoral Coordination Committee/National Health I _ ~~~~~~~~~Council Permanent Secretary M i o Director of Medical Services D.D.M.S. ADKS (PHC)| Fte-ering Comoittee| t~~~~~~~~~~~~~~~~~~~~~~~~~~ I . ~ .^ District Medical Officer District Planning [IZCo]ttee [District Health Nurse Health Centre H Country Health Staff Sub-County Coordination I | Committee Aset. Health Visitor --T- ~~I Co. Health Worker/ Community TBA Development CommitteeJ  -50- Annex 1: POPULATION ANNEX 1 Page 1 of 2 Table 1.1: UGANDA: Projection Assumptions 1980-2015 1980 1985 1990 1995- 2000 2005 2010 Projectionl/ -85 -90 -95 2000 - 05 -10 -15 1. Constant Mortality e' 46.7 46.7 46.7 46.7 46.7 46.7 46.7 Constant Fertility TFR 7.0 7.0 7.0 7.0 7.0 7.0 7.0 2. Declining Mortality e