Document of The World Bank USE ONLY

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Document of The World Bank FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT REPUBLIC OF BURUNDI POPULATION AND HEALTH PROJECT (CREDIT 1862-BU) APRIL 16, 1996 Population and Human Resources Division Central and Indian Ocean Department Africa Region Report No This document has a restricted distribution and may be used by recipients only in the performance of their official duties. [ts contents may not otherwise be disclosed without World Bank authorization.

2 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF BURUNDI POPULATION AND HEALTH PROJECT (CREDIT 1862-BU) CURRENCY EQUIVALENTS Currency unit: Burundi Franc (FBu) (Period average, US $ 1 equivalent) 1987 = FBu = FBu = FBu = FBu = FBu = FBu = FBu = FBu = FBu 257 WEIGHTS AND MEASURES Metric system FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS AfDB AIDS BIP CPR EPI EPISTAT FP GPA HC HIS HIV IDA EEC MCH MOI MOPH NGO PDP PHC PIP SAR UNFPA UNICEF USAID WHO African Development Bank Acquired Immunodeficiency Syndrome Bureau de l'inspection et de la Planmfication Contraceptive Prevalence Rate Expanded Program of Immunization Health Information Service Family Planning Global Program for AIDS Health Center Health Information System Human Immunodeficiency Virus International Development Association Information, Education and Communication Maternal and Child Health Ministry of Interior Ministry of Public Health Non-Governmental Organization Programme de Depenses Publiques Primary Health Care Programme d'investissements Publics Staff Appraisal Report United Nations Fund for Population Activities United Nations Children's Fund United States Agency for International Development World Health Organization

3 FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT (ICR) REPUBLIC OF BURUNDI POPULATION AND HEALTH PROJECT (CREDIT 1862-BU) CONTENTS PREFACE... i EVALUATION SUMMARY.ii 1. PROJECT IMPLEMENTATION ASSESSMENT.I A. Statement and Evaluation of Project Objectives.1 B. Achievement of Project Objectives.2 C. Implementation Record and Major Factors Affecting the Project.5 D. Project Sustainability 7 E. IDA Performance.7 F. Borrower Performance.9 G. Assessment of Outcome.10 H. Future Operation.1.1 I. Key Lessons Learned STATISTICAL TABLES.13 TABLE 2-1: SUMMARY OF ASSESSMENTS.13 TABLE 2-2: RELATED IDA CREDITS TABLE 2-3: PROJECT TIMETABLE TABLE 2-4: CREDIT DISBURSEMENTS: CUMULATIV ESTIMATED AND ACTUAL TABLE 2-5: KEY INDICATORS FOR PROJECT IMPLEMENTATION TABLE 2-6: KEY INDICATORS FOR PROJECT OPERATION TABLE 2-7: STUDIES INCLUDED IN PROJECT TABLE 2-8: PROJECT COSTS TABLE 2-9: PROJECT FINANCING TABLE 2-10: ECONOMIC COSTS AND BENEFITS.20 TABLE 2-11: STATUS OF LEGAL COVENANTS IN CREDIT AGREEMENT.21 TABLE 2-12: COMPLIANCE WITH OPERATIONAL MANUAL STATEMENTS.24 TABLE 2-13: BANK RESOURCES: STAFF INPUTS.24 TABLE 2-14: BANK RESOURCES: MISSIONS.25 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed wiihout World Bank authorization.

4 3. APPENDICES A. Mission's Aide-Memoire B. Borrower Contribution to the ICR... 32

5 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF BURUNDI POPULATION AND HEALTH PROJECT (CREDIT 1862-BU) PREFACE This is the Implementation Completion Report (ICR) for the Population and Health Project in Burundi, for which Credit 1862-BU in the amount of SDR 11.0 million was approved on December 15, 1987, signed on February 9, 1988, and made effective on July28, The Credit was closed on September 30, 1995, about two years later than the original closing date. Final disbursementook place on November 21, 1995, when the remaining balance of SDR 0.6 million was canceled. Apart from the government's counterpart contribution of US$2 million (compared with the appraisal estimate of US$2.92 million), cofinancing for the project was provided by UNICEF for US$0.8 million equivalent (compared with the appraisal estimate of US$1.83 million). In addition, the Global Program for AIDS of WHO contributed during the project period US$4.76 million to the component "Support to the National AIDS Control Program", and the savings made were used to financc a greater number of health centers and additional offices for MOPH. The ICR was prepared by Ms. Michele Lioy of the Population and Human Resources Division, Central Africa and Indian Ocean Department of the Africa Region. It was reviewed by Mr. David Berk, Chief of the Population and Human Resources Division, and Mr. P. Hari Prasad, Operations Adviser of the Central Africa and Indian Ocean Department. Preparation of this ICR began during the Bank's final supervision/ completion mission in September It also uses extensively material in the project file and the results of the beneficiary assessment study that was carried out in late The Borrower contributed to the preparation of the ICR by providing views reflected in the mission's aide-memoire (Appendix A), preparing its own evaluation of the project's execution (Appendix B), and commenting on the draft ICR. The assessment of the project's success, and therefore the preparation of this ICR, were made particularly difficult by the disastrous consequences of the October 1993 military coup attempt and the political crisis that followed. As discussed in this ICR, the project brought about a number of improvements and in the quality of the health services in coverage by the health care system, and some of the objectives had been partially achieved before the crisis. Nevertheless, since many of these gains were reduced because of the crisis, the achievement of project objectives by the time of the closing of the IDA credit can only be assessed as negligible (for the financial objective) or partial, except for one objective that was achieved.

6

7 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF BURUNDI POPULATION AND HEALTH PROJECT CREDIT 1862-BU EVALUATION SUMMARY Introduction. 1. The first IDA operation in the sector, the Population and Health Project was an integral part of Burundi's Fifth Development Plan ( ) and complemented other donor assistance in population, health and nutrition. The project built on the Health and Population Sector Report of July 1983 and the subsequent dialogue on population issues which contributed to the development of a new awareness of demographic constraints. It also built on the experience of other donors (UNICEF, UNFPA, USAID, WHO) who were already active in the sector, and focused on interventions which were too weak or neglected by all donors, e.g., child nutrition, maternal health, and family planning (FP). Project Objectives 2. According to the development credit agreement, the objectives of the project were: (i) to contribute to the reduction of maternal and child morbidity and mortality; (ii) to promote birth spacing and strengthen the National Maternal and Child Health and Family Planning Program; (iii) to assist in controlling the AIDS epidemic; (iv) to improve the overall effectiveness and efficiency of the health system; and (v) to improve the demographic data base needed for socio-economic planning. The Staff Appraisal Report (SAR) included some quantitative objectives for reducing maternal and infant mortality and for reducing fertility. The objectives were to be achieved through the implementation of five components: (i) the strengthening of MCH/FP services; (ii) the development of an lec program; (iii) support to the National AIDS Control Program; (iv) the institutional strengthening of MOPH; and (v) population data development. Although the objectives were never formally changed during implementation, the component of support to the National AIDS Control Program was for all practical purposes dropped from the project, following the decision of other donors, particularly WHO, to finance it and the reallocation to other components of the proceeds of the IDA credit earmarked for that component (paragraph 1.1). 3. The development credit agreement included special covenants to promote the achievement of project objectives. These related to: project performance reviews and implementation plans; three-year rolling programs of investments and recurrent expenditures for the health sector; the acquisition of sites and the recruitment of additional nurses for health centers; the appointment of a manager and two advisors for the Health Education Division; the employment of two technical assistants (in health planning and architecture) to support the BIP (Bureau de l'inspection et de la Planification) of the MOPH and of a Certified Public Accountant to assist the budget division of MOPH; the

8 Evaluation Summary iii preparation of a five-year National AIDS Control Program; the appointment of multidisciplinary health supervisors in the provinces; the preparation of a policy paper on health financing; the preparation of a census master plan and of a plan of action for the improvement of national demographic and vital statistics; and the appointment of a full time deputy project coordinator to assist the inspector general of health (Table 2.11). 4. The project's objectives were important for Burundi and its health sector in view of the country's high maternal and child mortality. MOPH strategy to develop a strong family planning component in the MCH program was a sound one considering the high population density and fertility rate in Burundi. Since the Government had decided to give priority to the development of a comprehensive population program designed to reduce fertility and population growth, the project was responsive to changes in Burundi's circumstances and priorities. The objectives were also important for IDA's country assistance strategy which aimed inter alia at filling major gaps in the overall provision of external assistance. The project was relatively complex and very demanding for MOPH. The quantitative targets for reduction in mortality and fertility rates were too ambitious and difficult to achieve in the time frame of a five-year project, even if the political situation had remained stable (paragraphs 1.2 to 1.4). The design of the project was generally appropriate for achieving the objectives with respect to MCH services and institutional strengthening, but much less so for AIDS control (paragraph 1.9). Implementation Experience and Results 5. In October 1993, the Project (although somewhat delayed) was about to achieve some of its objectives, at least partially. Important gains had been made because of the Project, gains which have had a lasting impact (paragraph 1.17). The MOPH, which has been capable during the crisis to coordinate, with the help of WHO, emergency aid, is still functioning satisfactorily despite the difficult country situation; health sector personnel have been trained; systems have been put into place, and 41 health centers have been rehabilitated or built. All of those aspects of capacity building, which are key in a first operation, constitute elements of sustainability. The objective to improve the demographic data base needed for socio-economic planning was achieved. However, mainly because of the crisis, the success of the project as a whole and of most of its components in achieving their objectives is.either negligible (financial objective) or partial at the time of the closing of the IDA credit. The project activities were indeed very seriously affected by the disastrous consequences of the October 1993 military coup attempt and the political crisis that followed. Starting in October 1993, Burundi's health sector was almost paralyzed for nearly a year because of the insecurity and of the political uncertainty prevailing throughout the country. Large numbers of personnel fled to safer zones or were killed, and many Rwandese health personnel returned to Rwanda after fighting ended there, leaving health facilities inadequately staffed. Equipment, supplies and vehicles were often vandalized or stolen and national distribution channels deteriorated. The Government remains committed to improving the health status of the population and the national health policy is still promoting sectoral reforms, including decentralization of the health system and the introduction of new financing mechanisms of the health sector that would increase the participation of communities. In the second

9 iv ICR - Burundi: Population and Health Project semester of 1994, the MOPH started many of its activities again, and the management of the Ministry began to adapt the sector reforms which were planned in the context of this first project to the new situation. This led to a reappraisal of the Second Health and Population Project (Cr BU) which was approved in May This second project builds on the achievements of the first one, and continues and expands activities included in it. Because of the risks associated with the second project, due to the socio-econornic crisis, the sustainability of the first Population and Health Project achievements can only be assessed as "uncertain"(paragraph 1.17). 6. The project took about two years longer to complete than expected at appraisal. The reasons for these implementation delays include changes in project scope, uneven quality of management and delays in selecting staff, procurement and disbursement delays, and acute security problems (paragraphs 1.13 and 1.14). The actual project cost, not including some training activities financed by UNICEF which could not be costed, was $17.62 million, as compared with $18.8 million appraisal estimate. The IDA credit was almost fully disbursed (95% of SDR1 1 million)', and the contribution of the Government of Burundi and UNICEF were US$2 (instead of 2.9) million and US$0.8 2 (instead of 1.8) million respectively. 7. A number of factors affected the achievement of project objectives. Among those factors not generally subject to govermment control, the effect of the military coup attempt of October 1993 and the crisis that followed was substantial. In particular, the crisis has had an impact on the availability of personnel and of Govermment counterpart funds for the health sector. Among the factors generally subject to govermnent control, macroeconomic and sector policies, were significant: the decentralization policy was never implemented. A number of factors generally subject to implementing agency (MOPH) control also affected the achievement of the project objectives (paragraphs 1.14 to 1.16). On the positive side, the project benefited from its reliance on technical assistance which. was successful. Negative factors that affected the project included the unequal quality of project management over time and the high tum over of project coordinators, staffing difficulties, and procurement and disbursement delays. 8. IDA performance on this project was highly satisfactory at the identification stage, and satisfactory in preparation assistance, appraisal, and supervision (paragraphs 1.18 to 1.21). The Borrower's performance can be rated as highly satisfactory for preparation, satisfactory for implementation, and "partially" deficient for covenant compliance (paragraphs 1.22 to 1.25). 9. Mostly because of the crisis, the project's outcome deserves a rating in the quality of health services of "unsatisfactory". Before the crisis, however, there had been a number of improvements in the quality of health services and in coverage by the health care system, one objective had been achieved, and most of the others partially achieved. Apart 2 For disbursement in US dollars, see Table 2.4. This amount represents only the cost of technical assistance provided by UNICEF, and it does not include project activities such as health personal training.

10 Evaluation Summary v from contributing to the last census, the project was also successful in strengthening the institutional capacity of MOPH (including IEC and planning and management capabilities), and in developing a well integrated in-service training program complemented by a systematic decentralized integrated supervision system. In the final analysis, the Project had an impact, albeit limited, on institutional development. Studies for the reform of health financing had been done and changes in legislation were under preparation in October However, efforts to strengthen family planning services and design health financing strategies suffered from an insufficient Government commitment and weak institutional capacity. IEC activities for the National AIDS Control Program financed by the project contributed to increased knowledge and awareness about the means of transmission of HIV and the problems created by AIDS, but the seroprevalence and the number of AIDS cases has continued to increase, although less rapidly than in some neighboring countries (paragraph 1.26). Summary of Findings, Future Operations, and Key Lessons Learned 10. Although the project did not reach its objectives, which were too ambitious on balance, the project was implemented satisfactorily until the crisis; it brought about improvements in the quality of health services and in their coverage which were unfortunately reduced, and even nullified in some cases, by the disastrous consequences of the crisis. However, the Project has had a lasting impact in terms of capacity building at the central, intermediate and peripheral levels (particularly for planning and management and in IEC) in MOPH, increased sector knowledge, and reinforced peripheral health infrastructures. In 1995, IDA approved a new credit for a Second Health and Population Project. The objectives of that second project are to improve the health status of Burundi's population and contribute to sustainable per capita growth by slowing population growth, thus reducing pressure on resources. The project aims to do this by increasing the availability, quality and efficiency of health services and making them more responsive to the public's expectations. The design of the new project has incorporated the lessons learned from the experience of this Population and Health Project (see paragraph 11 below) and builds on its achievements (for example, infrastructures and systems put into place, and the development of technical and administrative skills). Given the country's current political situation, the new project has been designed in a flexible manner to permit the implementation of certain activities even in a context of political crisis and insecurity in certain parts of the country. In case of a continuous crisis, however, the new project would have to be restructured substantially (paragraphs 1.27 and 1.28). 11. The following lessons (paragraph 1.29), which are relevant for future project design both in health/population and in Burundi, can be learned from the experience of this Population and Health Project: * No matter how well prepared and implemented a project is, its development impact may be jeopardized in a situation of continuous political instability and civil disturbances: however, investments in human resources are more likely than other

11 vi ICR - Burundi: Population and Health Project investments to have a lasting impact and contribute to the sustainability of the project. * A project is implemented better when the Government is committed, when key staff of the implementing agency are involved in project design and preparation, and when beneficiary opinions are taken into account during project preparation. * Managerial and institutional capacities to develop project activities and key functions (such as project coordination, procurement and accounting) need to be carefully assessed during project preparation, and adequate strengthening must be provided throughout project implementation. * Strengthening health centers is not enough to ensure an improvement in overall health service delivery; reference health facilities must also be upgraded. * Rehabilitation activities should emphasize equally physical and qualitative inputs and outputs (increased service capacity and quality). * Long term technical assistance can contribute to capacity building and to project implementation if the adviser has a counterpart who participates in all project activities and who is given increasing responsibilities. - A lack of continuity in project management (due for instance to key personnel turnover) is likely to cause delays in project implementation. Mechanisms should be put in place which would help prevent too high a turnover of key staff. - Performance indicators should be designed and used to monitor the progress of the project on an annual basis in order to facilitate the assessment of the achievements of the project.

12

13 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF BURUNDI POPULATION AND HEALTH PROJECT CREDIT 1862-BU 1. PROJECT IMPLEMENTATION ASSESSMENT A. STATEMENT AND EVALUATION OF PROJECT OBJECTIVES 1.1 According to the development credit agreement, the objectives of the project were: (i) to contribute to the reduction of maternal and child morbidity and mortality; (ii) to promote birth spacing and strengthen the National Maternal and Child Health and Family Planning Program; (iii) to assist in controlling the AIDS epidemic; (iv) to improve the overall effectiveness and efficiency of the health system; and (v) to improve the demographic data base needed for socio-economic planning. The Staff Appraisal Report (SAR) included some quantitative objectives for reducing maternal and infant mortality and for reducing fertility (Table 2.5 of this report). The objectives were to be achieved through the implementation of five components: (i) the strengthening of MCH/FP services; (ii) the development of an EEC program; (iii) support to the National AIDS Control Program; (iv) the institutional strengthening of MOPH; and (v) population data development. Although the objectives were never formally changed during implementation, the component of support to the National AIDS Control Program was for all practical purposes dropped from the project following the decision of other donors, particularly WHO, to finance it. The project provided the National AIDS Control Program with equipment and furniture and some AIDS IEC activities were carried out in the context of the EEC component. 1.2 The objectives were clear and made very explicit by the inclusion in the SAR of not only process indicators for the major components (Table 2.5), but also of outcome and impact indicators related to services provided by the health system and to basic health data (Table 2.5). The objectives were fully consistent with the government's social objectives ("to improve living standards of the most vulnerable groups of the population") and economic adjustment objectives ("to improve the financial performance of technical ministries"). The objectives were very important for Burundi and its health sector, in view of the country's high maternal and child mortality. The MOPH strategy to develop a strong family planning (FP) component in the MCH program was a sound one considering the very high fertility rate in Burundi and population density close to 200 per sq.km., the second highest in Sub-Saharan Africa. Since the Government had decided to give priority to the development of a comprehensive population program designed to reduce fertility and population growth, the project was responsive to changes in Burundi's circumstances and priorities. The objectives were also in agreement with IDA's country assistance strategy. Following the Health and Population Sector Report of July 1983 and the

14 2 subsequent dialogue on health and population issues, IDA's strategy was to fill two major gaps in the overall provision of extemal assistance to the Population, Health and Nutrition (PHN) sectors: the first gap was the limited support to a large scale FP program, including IEC activities, and the second the lack of institution building for the health system. In addition, the National AIDS Control Program required a major foreign support. It was felt that, by participating in this program, which had implications for health expenditure and the overall health system performance, IDA would be able to influence its design and execution. 1.3 It was appropriate for Burundi to aim for a reduction in maternal and child mortality and for the development of birth spacing and family planning among couples, as well as for an improvement in the overall effectiveness and efficiency of the health system, all of which would be achieved progressively. However, the quantitative targets for reduction in mortality and fertility rates were too ambitious and difficult to achieve in the time frame of a five-year project, even if the political situation had remained stable. This difficulty was partly recognized in the Staff Appraisal Report (SAR), which stated that the objectives "will require a strong commitment from the Government and major efforts from the donor community". 1.4 The project was an integral part of Burundi's Fifth Development Plan ( ). Because the project dealt with many aspects of the functioning of the health system both at the central level and in the provinces, it was relatively complex. In addition, it was very demanding, not so much for the Ministry of Interior (MOI), whose Population Department senior staff had already carried out successfully the 1979 census, but mostly for the Ministry of Public Health (MOPH) which had limited qualified personnel in the fields of health policy and management. In fact, the SAR identified the relative MOPH inexperience in implementing a project of this size as the project's main risk. That is why the project design included the strengthening of various MOPH units by additional staff and a substantial input of extemal long- and short-term technical assistance. B. ACHIEVEMENT OF PROJECT OBJECTIVES 1.5 The assessment of the project's success in achieving its major objectives should take into account the significant differences between the various components, in terms of both content and relative importance. One has to distinguish between review of the project as a whole (reflected in Table 2.1 of Part II), and more qualified comments that can be made on individual components. An additional difficulty in assessing the project's success is the disastrous consequences of the October 1993 military coup attempt and the political crisis that followed. Starting in October 1993, Burundi's health sector was almost paralyzed for nearly a year because of the insecurity and the political uncertainty prevailing throughout the country. Large numbers of personnel fled to safer zones or were killed, and many Rwandese health personnel who had been working for years in Burundi retumed to Rwanda after fighting ended there, leaving health facilities inadequately staffed. Equipment, supplies and vehicles were often vandalized or stolen and national distribution channels deteriorated. Major health problems on a national scale were prevented only by

15 3 the substantial flow of humanitarian relief and intemational emergency aid. Mainly because of this political crisis, the success of the project as a whole and of most of its components in achieving their major objectives is either negligible (financial objective) or partial at the time of the closing of the IDA credit (see Table 2.1). Only the objective to improve the demographic data base needed for socio-economic planning was achieved, and remained so, despite the crisis. The discussion below tries to present a more balanced picture of the achievements of the project's components before the crisis and of what is now left, on which the second project can build. The quantitative criteria for judging the achievement of objectives were relatively few in the Staff Appraisal Report (SAR), but were much more detailed in the working papers produced for each project component (with the exception of the AIDS control program) during the preparation of the project. 1.6 The achievement of impact objectives is difficult to assess because of lack of reliable data. Since the beginning of the crisis, health information has not been kept consistently. The most recent report by the Health Statistics Division (EPISTAT) was issued in 1993 and presents the data for There is no information on the recent trend in the maternal mortality rate, but it seems that there has been some improvement in the infant mortality rate (from 125 per 1,000 in 1987 to 110 per 1,000 in 1992). 1 Prenatal coverage has improved: in 1986, 60% of women had at least one prenatal consultation, it increased to 92% in 1992 (percentage higher than the targeted 85%), but declined to 68% with the crisis. Percentage of deliveries medically assisted also improved (from 17% to 20%), although less than expected (27%). Child health monitoring (at least one visit a year for children 0-23 months) reached the SAR estimate for 1992 (80%) but declined after the crisis to less than it was in 1986 (46% versus 40% in 1995). The contraceptive prevalence rate improved slightly from 1,6% in 1987 to about 3% for the whole country in 1992 (and almost 10% in some provinces), but has since decreased and remains negligible. The fertility rate has continued to increase (to 6.8), instead of decreasing, and very little progress has been made in controlling the AIDS epidemic. Finally, the population data development has been successful with the development of a demographic data base. In conclusion, many objectives had been reached, at least partially, before the crisis, but most of the gains either were reduced or annihilated. 1.7 The project's physical objectives have been reached or even exceeded in some cases (for instance 41 health centers have been built or rehabilitated instead of 30). The project supplied to the various components all the tools (vehicles, computers, audiovisual equipment) that were envisaged in the SAR. Training was also carried out successfully: 2036 health staff were retrained instead of the 1000, and 24 central staff (instead of 20) and 50 provincial staff (instead of 80) were trained in management. All fellowships were awarded as planned in the SAR and all except two of the grantees, are now working for the MOPH, (one is still studying and the other works for the private sector). The 1987 estimate was based on data that were available when the SAR was prepared and that of 1992, on EPISTAT report for that year.

16 4 1.8 Regarding sector policies and institutional development, the achievement of objectives has been only partial. The institutional strengthening has had some very positive effects during the project period in terms of the management and planning capabilities of MOPH, the strengthening of MCH/FP services, and development of the integrated supervision system and of an IEC program. The EEC activities carried out have exceeded the SAR estimations: 60 programs were produced (instead of 30) and one of these received an award in a German film festival, 24 health promotion campaigns were carried out (instead of 18) and 18 training seminars held (instead of 16). The Health Information System was established. Although it was not computerized at the provincial level, for a lack of funding 2, the system collected and published health data until Since the crisis, data collection has been incomplete and EPISTAT has not published any report. Although some of those achievements have been partly offset by the consequences of the political crisis, it is worth noting that MOPH has retained most of its improved management and planning capabilities and its EEC capacity: it is one of the very few ministries which still functions satisfactorily and which has managed to cooperate and work well with donors to coordinate emergency aid. The project also contributed to developing a health map (carte sanitaire) which is used by all the health sector partners and to establishing norms for the health centers (two standard plans). 1.9 Even before the political crisis, the achievement of financial objectives was negligible, and it remains so after the crisis. Several studies were carried out, albeit somewhat belatedly, which laid the basis for looking at financing schemes. Before the crisis, a few hospitals had been made autonomous, a timetable had been set for implementing the reform of sector financing, and discussions had started with the Ministry of the Interior to change the municipal law so that health centers could keep and manage the income generated by cost recovery. However, no new financing mechanisms had been introduced before the crisis and little progress had been made regarding the feasibility of the financial and management autonomy of the health centers. In its last year, the project supported the preparation of the sector reforms which will be implemented under the Second Health and Population Project. In addition, there is some evidence of improvement in resource allocation in the health sector in terms of cost-effectiveness and efficiency. Salaries represented in 1995 less than 50% of MOPH's recurrent budget (versus 56% in 1986) and the hospital sector absorbs a smaller proportion of recurrent expenditures than in The execution of MOPH's budget improved significantly, thanks to the support of the PEP/PIP unit, with expenditures basically equal to the budgeted amounts until the crisis, and in 1994 and 1995, non-salary expenditures equal to 70% of budgeted amounts (whereas previously non-salary recurrent expenditures represented only about one-third of the budgeted amounts) The design of the project was generally appropriate for achieving the objectives with respect to MCH services and institutional strengthening, but much less so for AIDS control. Actually, the AIDS control component was not prepared in detail like the other 2 This computerization was to be financed by a USAID projecthat was postponed indefinetely because of the crisis.

17 5 project components, because the AIDS program was in its infancy and was supposed to be developed during the lifetime of the project; moreover, this component was added to the project after appraisal. C. IMPLEMENTATION RECORD AND MAJOR FACTORS AFFECTING THE PROJECT 1.11 The mission's aide memoire in Appendix A and the borrower's report in Appendix B discuss the implementation record for the project and the problems encountered. They provide a good account of what happened during the project implementation period. In spite of some problems, the implementation record is generally satisfactory The project faced some problems right from the beginning. Mostly because of the departure of the Inspector General of the MOPH (who had been on the job for about ten years and acted as Project Coordinator during project preparation) immediately after negotiations, the government wanted to renegotiate three aspects of the project: the technical assistance to be financed by the IDA credit, the proposed restructuring of the MOPH, and the financing with an IDA credit (rather than with a grant) of the National AIDS Control Program. IDA had to impress upon the government that a renegotiation of the IDA credit would not be opportune, but it seems that all government officials were not convinced and the same questions lingered on frequently during the first years of the project. The effectiveness of the IDA credit was delayed by a few months because of delays in the signing of the technical assistance contracts. As mentioned earlier, the National AIDS Control Program, although never formally canceled, was finally financed by WHO until 1992 and by others donors afterwards. During implementation, there were no major procurement problems, but there were procurement delays (because of the number of reviews and the lack of coordination) and, particularly at the beginning, delays with the replenishment of the special account, with a need for frequent reminders that proper procedures had to be followed. The project took two years longer to complete than expected at appraisal, with two postponements of the closing date of the IDA credit (one year at a time), from June 30, 1993 to June 30, In addition, the closing date was further postponed to September 30, 1995 for one contract of US$148,000 for the purchase of equipment for the maintenance workshops. The reasons for these implementation delays include changes in project scope, delays in implementing the sectorial reforms, uneven quality of management and delays in selecting staff, procurement and disbursement delays, and, above all, acute security problems. The actual project cost, excluding some training activities financed by UNICEF (which could not be costed) was $17.62 million, compared to the $18.8 million appraisal estimate. This amount covered activities of all project components with the exception of the support to the National AIDS Control Program that was financed by other donors 3. The project contributed only $715,000 to the program for equipment and furniture and financed AIDS IEC activities. The cost of these activities and goods was included in the IEC component. IDA funds budgeted for the AIDS component were reallocated to finance additional health centers 3 WHO contributed $4.76 million to this program for the period

18 6 and a building for the Project Unit's office, the Health Education Unit's office and the audio-visual production studio A number of factors affected the achievement of project objectives. Among those factors not generally subject to government control, the effect of the military coup attempt of October 1993 and the crisis that followed was substantial. To a lesser degree, cofinanciers were also a factor. UNICEF's collaboration with the project was very good at the beginning but did not last throughout the project period, and UNICEF ended up financing only two instead of three technical assistants (one for MCH/FP and one instead of two for IEC). WHO's decision to finance AIDS control activities affected both the project scope and its implementation (par. 1.12) Among the factors generally subject to government control, macroeconomics and sector policies were significant in slowing down project implementation in general and in delaying financial reforms in particular. The decentralization policy has never been implemented. No procedures were put into place so that the provinces could manage their own budgets. On the financial side, revenues collected by health center personnel (for payment by patients of the health insurance card and other services) are still channeled to the communal authorities, as required by the existing communal law, and only about 10 percent of the income is redistributed in kind to the health centers; such an arrangement tends to demotivate health center personnel since the bulk of the revenues generated cannot be used for buying drugs or financing other operational requirements. The project also suffered from a lack of counterpart funds from the beginning of the crisis until the end of 1994 when IDA agreed to increase its disbursement percentage to 100 percent (excluding taxes) for all disbursement categories A number of factors generally subject to implementing agency (MOPH) control affected the achievement of the project objectives. On the positive side, the project benefited from its reliance on technical assistance which was important during the preparation phase and early project implementation. There are many reasons why this technical assistance was successful. Despite earlier misgivings on the part of some government officials, MOPH commitment to use technical assistance efficiently was high. The assistance was specific, i.e. concentrated on specific topics such as management information systems, IEC, health organization, and maternal and child health. National professionals were involved from the very beginning and were organized in working groups focusing on specific topics; their work contributed to the efficiency of the technical assistance team. The leader of the technical assistance team worked in close collaboration with his counterparts, the Director and the Assistant director of the project, and assisted MOPH during the first 18 months of project implementation. The external technical assistance was phased out in parallel with: (i) long term training (fellowships) for senior MOPH officials; (ii) involvement of nationals in operations research; (iii) the organization of seminars/workshops; and (iv) the training of local staff in managerial skills (accounting, computer skills). Negative factors that affected the project included the lack of continuity in project management (the project had six project coordinators during the project

19 7 period) 4 and its unequal quality over time, staffing difficulties (particularly the absence of a qualified accountant for a whole year), and the procurement and disbursement delays mentioned in paragraph In summary, in spite of some problems, implementation was generally satisfactory. Respect of Bank procedures increased with time to become very satisfactory during the period preceding the crisis. Most activities planned (with the exception of the establishment of a health financing system) were carried out and, in some instance, targets were exceeded (rehabilitation, retraining, IEC activities). D. PROJECT SUSTAINABILITY 1.17 As noted earlier, because of the disastrous effects of the crisis, the achievement of project objectives has been either negligible (financial objective) or partial. However, important gains have been made as a result of the project. The MOPH is still operating satisfactorily in spite of the difficult country situation, Ministry personnel at all levels have been trained, systems for integrated supervision and health statistics were put in place, 41 health centers have been either rehabilitated (39) or built (2) and the health map still provides a framework to plan health activities. As shown by the beneficiary assessment, these project health centers offer a work environment more functional for health personnel and more hospitable for the population itself than before the rehabilitation. These infrastructures have not been damaged by the crisis (except for two of them which are being repaired) and remain an important contribution to the Burundi health system. All of those aspects of capacity building which are key in a first operation constitute elements of institutional sustainability. The Government remains committed to improving the health status of the population and the policy environment is not adverse to reforms, including a greater participation of communities. The Second Health and Population Project (Cr BU) continues and expands activities included in the first project, as discussed in the section below on "Future Operation". However, despite the gains which have been made under the first Population and Health Project, because of the risks, associated with the second project, due to the socio-political crisis, its sustainability can only be assessed as "uncertain". E. IDA PERFORMANCE 1.18 IDA performance on this project was highly satisfactory at the identification stage, and satisfactory in preparation assistance, appraisal, and supervision. As mentioned earlier, the project was very important for Burundi and for IDA's country assistance strategy, and project identification was on target. IDA involvement also encouraged other 4 Three of them left because they were awarded IDA-financed long-term fellowships. As they were civil servants (with no other compensation that their civil servant salary), it was difficult to keep them from competing for the fellowships which were awarded on a competitive basis. Among the others, one was replaced at IDA's request for his non-performance, one was dismissed for political reasons and the last one was the project coordinator at the end of the project.

20 8 donors to take into consideration certain aspects of the sector which had been generally neglected up to that time, such as institutional development and IEC activities. Thanks to IDA, MOPH has a strategy for the sector as a whole, rather than a project approach Regarding project preparation, IDA provided useful assistance to MOPH efforts to develop an overall plan and specific approaches. The design of the project was satisfactory, with the exception of the support to the National AIDS Control Program, which was included in the project after appraisal and which was basically a program in its infancy. These two reasons explain why this component had not been prepared and was limited to a few project ideas that remained to be defined more precisely and to a few lines in the Staff Appraisal Report. The satisfactory performance of IDA in assisting MOPH in project preparation was due to its substantial involvement (about 100 staff-weeks for the identification and preparation of the project), the right quality and quantity of staff and consultants (including skill mix and continuity), and prior IDA sector work Also because of the staff and consultants involved (about 60 staff-weeks), the performance of IDA in project appraisal was satisfactory. The commitment and capacity of the Government and MOPH were properly assessed, with the judgment that since capacity was limited there was a need for a significant amount of technical assistance. However, not all the problems that materialized, such as the lack of effective decentralization and the crisis, were anticipated or recognized as risks at the time of appraisal. Also, experience showed that for some components the implementation plans were not quite adequate, and process and performance indicators were not designed and therefore not used to permit the monitoring of the project on an annual basis. Finally, at the time of project preparation not much attention was paid to community participation and the availability of incentives for participants to sustain the project. Actually, as explained in paragraph 1.15, the financing system used to allocate the financial resources obtained from the sale of the health insurance card and other cost recovery mechanisms contributes to demotivating health personnel. Admittedly, at the time of project preparation, participation was not stressed as much as it is now IDA performance in supervising the project can also be rated as satisfactory. Sufficient attention was paid to the likely development impact, and by and large implementation problems (such as the demotivation of health personnel) were identified and assessed. Studies had been carried out a few months before the crisis, including a beneficiary assessment, in order to design new mechanisms for the financing of the health sector. However, the reporting of project implementation progress was not always as comprehensive as would have been desirable, particularly regarding compliance with covenants and the implementation of some sub-components, such as the health financing policy development (which became part of the public expenditures review exercise). As a result, some of the ratings on implementation status and development objectives given in Supervision Form 590 may have been too generous. Because most planned activities were taking place, enough attention may not have been paid to some of the existing problems. For instance, IDA should probably have insisted more on finding mechanisms to ensure more stability in the management of the project. Through regular contacts, IDA staff provided useful advice to MOPH. IDA showed flexibility in approving needed

21 9 modifications, especially in reallocating funds from the National AIDS Control Program to additional health centers and office buildings. Also, in view of the Government inability to provide counterpart funds after the crisis, IDA agreed in November 1994 to an increase of its disbursement percentages to 100 percent (excluding taxes) for all disbursement categories. Generally, IDA, the Borrower, MOPH, MOI and cofinanciers worked together successfully for most of the project period. With the obvious exception of the period of civil disturbances, the timing and length of time of supervision missions was adequate; on average 11 staff-weeks per year were spent on supervision, which is about normal for most of the region. There was no requirement for a mid-term review, and country implementation reviews did not have much impact; on the other hand, the Resident Mission in Bujumbura made a contribution since for the last year one of its staff was task manager for the project. F. BORROWER PERFORMANCE 1.22 The Borrower's performance can be rated as highly satisfactory for preparation, satisfactory for implementation, and "partially" deficient for covenant compliance With financing provided by a first Project Preparation Facility (PPF) advance, the MOPH set up a multi-disciplinary preparation team headed by the Inspector General of Health (the Head of the Bureau of Inspection and Planning - BIP), who played the role of project coordinator. This team began to prepare the project on the basis of terms of reference jointly prepared by MOPH and IDA, and was later reinforced by the recruitment of a foreign technical assistant to assist in project preparation coordination. Several working groups (involving about 60 local staff and students for research work) were established to prepare specific components. Other donors (UNICEF, UNFPA, USAID, AfDB, and WHO) were involved at an early stage of preparation and some of their technical advisers became part of the MCH/FP Committee in charge of developing the national program. In addition, six short-term international experts (health financing, MCH/FP, IEC, health planning, health statistics, and architecture) were able to build upon the preliminary work completed by local consultants and health staff. After appraisal, a second Project Preparation Facility (PPF) advance financed inter alia the preparation of final working drawings and final lists of equipment, and furniture and vehicles. This thorough and comprehensive preparatory work (except for the support to the National AIDS Control Program, which was added to the project after appraisal) was reflected in the good quality of the Government's project documents which included indicators, and was designed to ensure an effective start of the project in early It also resulted in a high stakeholder commitment The borrower's performance has varied during the project implementation period. It was highly satisfactory from the beginning of the project until the departure of the technical assistance in August 1990, and was satisfactory until the beginning of the civil disturbances in October 1993; from that time till the closing date of the IDA credit, it has been rather deficient. With an adequate and effective technical assistance, a generally good performance of consultants and contractors, the availability of counterpart funding,

22 10 and no government interference in operational decisions, MOPH was able to make a good start in project implementation despite the initial problems mentioned in paragraph After the departure of the technical assistance, the management of the project became less efficient, and the situation deteriorated further with the crisis when the project coordinator could not fulfill his duties on a full time basis because he was most of the time in hiding. It should be noted that, throughout the project period, project management and implementation suffered from a high turnover of staff, particularly at the top (there were six coordinators during the project period (see footnote No. 4), who had not always the right qualifications. Since the beginning of the crisis, the project was affected by an insufficiency of counterpart funds and very high absenteeism at all levels The extent to which the Government and MOPH have complied with major credit covenants/commitments has also varied over time for much of the same reasons. However, even before the crisis, some of the covenants were complied with partially or with delays. The failure of Government/MOPH to employ sufficient additional health personnel and staff health centers as planned and agreed upon, and to recruit all the needed advisers in order to ensure the most productive use of the new investments and systems, was particularly serious. On balance, compliance with covenants was less than satisfactory. G. ASSESSMENT OF OUTCOME 1.26 Mostly because of the crisis, the project failed to achieve most of its major objectives. Therefore, the project's outcome deserves a rating of "unsatisfactory". However, it should be noted that the project had yielded before the crisis worthwhile development results, some of which remained after the crisis. There had been a number of improvements in the quality of health services and in coverage by the health care system, and most of the objectives had been partially achieved. The number of health centers built or rehabilitated exceeded the appraisal estimate (41 instead of 30), and service indicators (such as the number of patients in health centers) improved in those health centers. As shown by the beneficiary assessment, 82% of beneficiaries who participated to the study, felt that health services had improved in the rehabilitated centers and, 77% and 62% indicated respectively that health personnel capacities and attitudes had improved. However the beneficiaries stressed the need for drugs and personnel in addition to infrastructures to improve services. Apart from contributing to the last census, the project was also successful in strengthening IEC (and to a lesser extent HIS) capabilities within MOPH, and in developing a well integrated in-service training program (based on high quality manuals developed in the context of the project and which is still in use) complemented by a systematic decentralized integrated supervision system. Generally, the quality of health services improved during the project period, implying that the project contributed to this improvement. Some dramatic improvements in certain areas were observed; for example, vaccination coverage had increased to 85%, and 80% of pregnant women had at least one prenatal consultation. However, efforts to strengthen family planning services and design health financing strategies suffered from an insufficient Government commitment and weak institutional capacity. The contraceptive prevalence

23 11 rate increased only slightly to about 3% (but up to 10% in certain provinces such as Ngozi) before falling back to an extremely low level of about one percent. The AIDS program has contributed to increasing knowledge and awareness about AIDS and the means of transmission of HIV and to promote a better acceptance of persons living with HIV, but the seroprevalence and the number of AIDS cases have continued to increase (although not as dramatically as in most neighboring countries). Finally, the most important impact of the project is in terms of capacity building and institutional strengthening, in particular increased sector knowledge on the part of MOPH, establishment of a sector strategy, and reinforced peripheral health infrastructures; those gains proved invaluable for the Second Health and Population Project which continues and expands activities initiated under the first project. H. FUTURE OPERATION 1.27 In 1995, IDA approved a new credit for a Second Health and Population Project. The objectives of that second project are to improve the health status of Burundi's population and contribute to sustainable per capita growth by slowing population growth, thus reducing pressure on resources. The project aims to do this by increasing the availability, quality and efficiency of health services and making them more responsive to the public's expectations. These objectives are in fact the continuation of the development objectives of the first project. In particular, the project plans to increase health personnel and make drugs available. In addition, the second project will implement the sector reforms (decentralization and financing) which were under preparation before the crisis in the context of the first project, after adapting them to the current situation. Specific objectives of the second project are to: (i) strengthen MOPH's capacity to design and implement key sector reforms; (ii) improve the availability and quality of health care delivery in rural areas; (iii) contribute to reducing population growth and maternal and child mortality rates; and (iv) promote beneficial health and family planning behaviors. Project components corresponding to these objectives consist of: (i) strengthening MOPH's capacity to effect key sector reforms in human resource management, hospital management autonomy, health financing, quality of health services, and maintenance; (ii) strengthening service delivery in five out of 15 provinces, by supporting basic health services and rehabilitating nine hospitals; (iii) implementing the maternal and child and family planning (MCH/FP) policy; and (iv) improving IEC activities. To monitor implementation and progress towards project development objectives, the new project includes key project performance indicators, which would be monitored, in some cases annually, by MOPH in close collaboration with IDA The new project not only continues the activities of the Population and Health Project (see next section), but its design has incorporated the lessons learned from this first experience. Given the country's current political situation, the new project has been designed in a flexible manner to permit the implementation of certain activities even in a context of a renewed political crisis and insecurity in certain parts of the country. Rehabilitation of project hospitals would be done in three phases that would be modified according to the prevailing state of security, starting with hospitals located in the safest

24 12 provinces. Likewise, management autonomy and health service quality would first be strengthened in these hospitals. Implementation of the MCH/FP policy and IEC activities at the provincial and community levels also would first be carried out in the safest areas of the country. However, activities to support design and implementation of sector reforms would be the most affected by a long lasting crisis. In case of a continuous crisis, the project would have to be restructured substantially. I. KEY LESSONS LEARNED 1.29 The following lessons, which are relevant for future projects both in health/population and in Burundi, can be learned from the Population and Health Project: * No matter how well prepared and implemented a project is, its development impact may be jeopardized in a situation of continuous political instability and civil disturbances: however, investments in human resources are more likely than other investments to have a lasting impact and contribute to the sustainability of the project. D A project is implemented better when the Government is committed, when key staff of the implementing agency are involved in project design and preparation, and when beneficiary opinions are taken into account during project preparation. * Managerial and institutional capacities to develop project activities and key functions (such as project coordination, procurement and accounting) need to be carefully assessed during project preparation, and adequate strengthening must be provided throughout project implementation (in spite of political crisis). * Strengthening health centers is not enough to ensure an improvement in overall health service delivery; reference health facilities must also be upgraded. * Rehabilitation activities should emphasize equally physical and qualitative inputs and outputs (increased service capacity and quality). * Long term technical assistance can contribute to capacity building and to project implementation if the adviser has a counterpart who participates in all project activities and who is given increasing responsibilities. * A lack of continuity in project management (due for instance to key personnel turnover) is likely to cause delays in project implementation. Mechanisms should be put in place which would help prevent too high a turnover of key staff * Performance indicators should be designed and used to monitor the progress of the project on an annual basis in order to facilitate the assessment of the achievements of the project.

25 13 2. STATISTICAL TABLES Table 2-1: Summary of Assessments A. Achievements of Obiectives Obiectives Substantial Partial Neglipible Not Applicable Macro policies O O O 0 Sector policies O Financial objectives O Institutional development a Physical objectives E Poverty reduction O Gender issues 0 O 0 0 Other social objectives a O O 0 Environmental objectives a O O 0 Public sector management Private sector development Other (specify) B. Project sustainability Likely Unlikely Uncertain a a 0 C. Bank performance Highly satisfactor Satisfactory Deficient Identification 0a0 a Preparation Assistance 0O 0 Appraisal Supervision D. Borrower Derformance Highly satisfactory Satisfactory Deficient Preparation 0l 0 Implementation Covenant Compliance Operation (not applicable in 1986) E. Assessment of outcome Highlv iighiy satisfactory Satisfactory Unsatisfactory unsatisfactor

26 Table 2-2: Related IDA Credits 14 Credit Title Purpose Year of Status Approval Preceding Operations No preceding Operation Following Operations Second Health and To improve the health status of 1995 The credit has Population Project Burundi's population and contribute been made (Cr BU) to sustainable per capita income effective, and the growth by slowing population project is being growth, thus reducing population implemented pressures on resources. satisfactorily. Effective 02/09/96.

27 15 Table 2-3: Project Timetable Steps in project cycle Date planned Date actual Identification (Executive Project Summary) N.A. Feb/Mar 1985 Preparation N.A. Oct 1985 to Nov 1986 Appraisal N.A. Apr 1987 Negotiations N.A. Oct 1987 Board Presentation N.A. Dec 15, 1987 Signing N.A. Feb 09, 1988 Effectiveness Jun 09,1988 Jul 28, 1988 Midterm Review N/A. N/A. Project Completion Dec 31, 1992 Jun 30, 1995 Loan/Credit Closing Jun 30, 1993 Sep 30, 1995 Table 2-4: Credit Disbursements: Cumulative Estimated and Actual (US$ million) FY88 FY89 FY90 FY91 FY92 FY93 FY94 FY95 FY96 Appraisal Estimate uu Actual Actual as % ofestimate l Date of final disbursement: November 21, 1995

28 16 Table 2-5: Key Indicators for Project Implementation Process Indicators SAR Actual Actual Estimate by end of by end of for end of MCH/FP Component Number of HCs rehabilitated Staffing of HCs (additional) Health staff retrained 1, % of health facilities delivering FP services IEC Component Programs produced Campaigns completed Seminars completed Institutional strengthening Staff trained in management - Central Provincial Development of HIS Completed Completed Completed Development of Health Financing Policy Completed Ongoing Ongoing Outcome Indicators Contraceptive prevalence (%) Prenatal coverage (I Consultation) (%) Deliveries attended medically (%) Child health monitoring (0-23 months, %) Impact Indicators Maternal mortality rate -20% -40%* N.A. Infant mortality rate (per 1,000) N.A. Total fertility rate *Estimations based on EPISTAT report for 1992: 1,000 per 100,000 live births in 1986, and 600 per 100,000 by end of 1992.

29 17 Table 2-6: Key Indicators for Project Operation Process Indicators Actual Actual by end of by end of Outcomes Indicators Revenue from sale of Health (%) insurance card (% MOPH recurrent budget) Health facilities delivering FP services Hospital bed occupancy (%) Availability of contraceptives (% of time) Population aware of population problem to be 80 (%) determined in 1996 Couverture vaccinale (%) Women of reproducible age to name one modern method contraceptive (%) Contraceptive prevalence (%) 1 10 Ante-natal coverage (1 consultation) % Delivery attended medically (%) Child Health Monitoring (0-23 months) Referral of high risk pregnancies Impact Indicators Infant Mortality rate 110* 95 Total Fertility rate Maternal Mortality rate (Per 100,000) *Estimations based on EPISTAT report for 1992: 1,000 per 100,000 live births in 1986, and 600 per 100,000 by end of 1992.

30 18 Table 2-7: Studies Included in Project Study Purpose as defined at appraisal Status Impact of Study Financing of health Improvement of health sector financing done Method improved sector j 19'Y Definr IEC services done Services greatly improved Institution building Define institution building done Better approach to institution in 1987 buildin Health information Improvement of the health information done Better data collection system system Study on launch of Clarifying financing and management of done Launch was effective. management of hospitals autonomous hospitals, 1993 Strategy of human Improve development of human resources done Human resources management resources, management greatly improved -August 1993 Study on IEC Improvement of IEC services done Services improved component, management February 1993 Project document: Define components and objectives of done Project is effective Health and project l opuiation II Project, June 1993 Socio-economic Identify the different income categories done Results could not be used survey and study, attitudes and behavior of because of the crisis and now beneficiary population towards (i) payment for health they are no longer valid. assessment 1993 services ; and (ii) various financial "insurance schemes" in order to improve access to health care. FlBeneficiarv Evaluation of results of the project done Results were used for the final assessment, 1995 components (SMI/lEC) on health centers' evaluation. services.

31 19 Table 2-8: Project Costs (US $ million) Project Components Appraisal Actual/Latest Estimate Estimate Strengthening MCH/FP Services * IEC Program Development * AIDS Control Program Institutional Strengthening Population Data Development Project Preparation Facility Total Project Costs * This amount corresponds only to the cost of technical assistance and exclude project activities such as health personnel training which could not be costed. ** This amount includes $ spent for equipment and furniture and IEC activities for the National AIDS Control Program. Table 2-9: Project Financing (US $ million Source Appraisal Actual/Latest Estimate Estimate IDA Credit Government of Burundi UNICEF * Unaccounted for 0.62 Total * This amount corresponds only to the cost of technical assistance and exclude project activities such as health personnel training which could not be costed.

32 20 Table 2-10: Economic Costs and Benefits List of benefits identiried in the SAR (a) Approximately one million people living in underserved areas will have access to a full range of essential PHC services. Actual (a) About 820,000 persons benefited from PHC services (b) Nearly one million women of (b) Before the crisis, in October 1993, 90% childbearing age will have access to effective of pregnant women. maternal care and family planning methods. During crisis, 70%. (c) About 800,000 under-five children will benefit from effective nutrition services as part of a comprehensive MCH package. (c) Information not available (d) The entire nation will benefit from the (d) The contraceptive prevalence rate had increasing contraceptive prevalence which increased only slightly to about 3%, but after will ease the demographic pressure. the crisis it has fallen back to about 1%. (e) The spread of AIDS would be reduced through the support to the National AIDS Control Program. (f) MOPH will benefit from institutional strengthening. (g) New sources of health financing will be developed. (e) The AIDS control program has contributed to increased knowledge, but it is difficult to assess whether it has slowed the spread of AIDS. (f) MOPH now has a team of competent cadres, and has some capacities for having a dialogue on policies and setting policies. (g) Although studies were done, no new sources of health financing were developed. (h) The demographic data base would be improved and used for population policy development and socio-economic planning. (h) Done. However, new surveys will need to be done because there have been some changes following the crisis.

33 21 Table 2-11: Status of Legal Covenants in Credit Agreement Section Covenant Present Fulfillment date Description of Covenant T_ype Status Original Revised Comments 2.02(b) 1, 3, 5 C Continuous Borrower to open and maintain two Special The two accounts have been opened and Accounts in dollars: Special Account A for the maintained, but especially at the beginning of Bureau de l'inspection et de la Planification of the project, there were problems with their MOPH, and Special Account B for the replenishments (including delays in having the Department of Population of MOI. accounts credited), and Borrower had to be reminded of the need to follow proper procedures. 3.01(a) 3, 4, 5 CP Continuous Borrowefs comnmitment to project objectives; Components other than the reconstruction / Borrower to carry out project with due diligence rehabilitation of health centers have not always and efficiency and in accordance with appropriate been implemented with due diligence and practices, and to provide, promptly as needed, the efficiency, contributing to the two years funds, facilities, services and other resources postponement of the closing date. Starting from required for project. the October 1993 crisis, Borrower had problems providing the agreed upon counterpart fund which were often insufficient , 9 C Before October 31 Borrower to exchange views with Association on Annual reviews held. The PIPIPDP unit is (b)&(c) of each year progress achieved in carrying out the project, and functioning although it still needs some to submit a project performance review, a project assistance. implementation plan and a revised three-year rolling program of Investments and Public Expenditures for the health sector , 3, 4, 5 CP Condition of Borrower to open and maintain a bank account in Starting from the October 1993 crisis, account (d)&(e) effectiveness, anid Fbu, deposit into it an initial amount equivalent to has not been replenished. thereafter $30,000, and thereafter deposit quarterly amounts continuous required for project CP Before 6/30/1988 In order to carry out the component for the All sites were acquired. Not all nurses were for health center Development of MCHIFP Services, Borrower to recruited: 2 additional nurses per health center rehabilitation, and acquire by 6/30/88 all sites not already o\ned for had been recruited by the end of continuous for HC to be rehabilitated, and to recruit at least four additional nurses. additional nurses for each of those centers CP Before 8/31/1988 In order to carry out the EC Program Only one adviser was appointed. Development component, Borrower to appoint by 8/31/88 a manager of the Health Education division _DA_ and two advisers, satisfactory to IDA. Key to abbreviations is on page 23

34 22 Section Co tant Present Fu-Hllir. nt date Description of Covenant 3 e Status Original _ Revised Comments 3.04 CP Before 3/31/b8, In order to carry out the Support to the National Borrower submitted a five-year AIDS control and thereafter AIDS Control Program component, Borrower to program, but there were no annual updates and continuous submit to IDA by 3/31/88 a five-year National exchanges of views as, for all practical AIDS Control Program, and annual updates on the purposes, this component was dropped from the occasion of the annual exchanges of views. project C By 12/31/1988 hi order to carry out the MOPH Institutional 24 supervisors nominated by December 31, by 12/31/1989 Strengthening component, Borrower to appoint by 12/31/1990 eleven multidisciplinary health supervisors for the Supervisors became "chefs de secteur", en provinces (in addition to the four already posted): 3 by 12/31/88, 3 by 12/31/89 and 5 by 12/31/ NC By 12/31/1988 Borrower to submit to IDA by 12/31/88 a policy Studies have been carried out, but no policy and paper on health financing policy development, plan of action have been developed. including recommendations and a plan of action , 9 C By 3/31/1988 and In order to carry out the Population Data General census completed in by 1/31/1990 Development component, Borrower to submit to Vital statistics greatly improved. IDA by 3/31/88 a census master plan, to submit progress reports on the 1989 census project, and to submit by 1/31/90 a plan of action for the improvement of national demographic and vital statistics CD By 3/31/1988 Borrower to appoint by 3/31/88 a full time deputy Project manager recruited in September project coordinator to assist the Inspector General of Health C continuous Procedures for procurement of goods, works and Project team followed procedures. consultants' services financed out of the proceeds of the Credit CD By June 30 Borrower to maintain adequate project records and In 1989, the audit report was submitted with of each year, for accounts, to have the accounts audited by delay. In one instance, auditors could not the preceding year. independent auditors acceptable to IDA, and to express an opinion because of the inadequacy of furnish to IDA, not later than six months after the the systems of accounts and of internal checks end of each fiscal year, the report of such audit by and controls; this was later corrected. said auditors. 5.01(a) 1, 5 C conditions of Borrower to employ (i) a technical assistant in effectiveness health planning for at least the first two years and a technical assistant in health architecture for at least the first three years, to support BIP, and (ii) a CPA to assist the budget division of MOPH. Key to abbreviations is on page 23.

35 23 Covenant types: I = Accounts/audits. 8 = Indigenous people. Present status: 2 = Financial performance/revenue 9 = Monitoring, review and reporting. generation from beneficiaries. 10 = Project implementation not covered by C = Covenant complied with. 3 = Flow and utilization of project funds. categories 1-9. CD = Complied with after delay. 4 = Counterpart funding. II = Sectoral or cross-sectoral budgetary or CP = Complied with partially. 5 = Management aspects of the project or other resource allocation.. NC = Not complied with. executing agency. 12 = Sectoral or cross-sectoml 6 = Environmental covenants. policy/regulatory/ institutional action. 7 = Involuntary resettlement. 13 = Other.

36 24 Table 2-12: Compliance with Operational Manual Statements All applicable Bank OMS were complied with. Table 2-13: Bank Resources: Staff Inputs Stage of Project Cycle Planned Revised Actual Weeks US$ Weeks US$ Weeks US$ Preparation to Appraisal N/A N/A N/A N/A N/A Appraisal N/A N/A N/A N/A 57.9 N/A Negotiations through N/A N/A N/A N/A 6.1 N/A Board Approval I I Supervision N/A N/A N/A N/A 77.0 N/A Completion N/A N/A N/A N/A 9.0 N/A TOTAL N/A N/A N/A N/A ,600

37 25 Table 2-14: Bank Resources: Missions Stage of Month/ Number Days Specialized staff Perfonmance Rating Types of Problenm Project Cycle Year of in sldlls represented Persons Field Implemen Develop- -tatlon ment Status Objectives Through apraisal Identification Feb PH, LO May PH _ Preparation Dec PH,IEC,FP,EC May PH Jul PH,IEC,FP,MA Preappraisal Nov PH,IEC,FP,RA,AR Appraisal through Board Approval Appraisal Mar/Apr PH,ARIEC,FP,EC, N.A. N.A.. RA Pre-eflectiveness Jun PH,AR N.A. N.A. Supervsbon Supervision I Oct PH,AR 1 I Supervision 2 Mar/Apr PH Supervision 3 Feb PH,IEC,EC 1 Supervision 4 Nov/Dec PH,AR,ESS,EC 1 2 PMP,SP Supervision 5 Feb PH,IEC,AR 1 2 PMP,SP Supervision 6 Sep PH, SS 2 2 PMP,SP Supervision 7 Feb PH,IEC,EC 2 2 PMP,TAP,SP Supervision 8 Oct-92 I 7 AR 2 2 PMP,TAP,SP Supervision 9 Mar AR,IEC 2 2 PMP,TAP,SP Supervision 10 Oct IEC 2 2 PMP, TAP, SP Supervision 11 Mar-94 - I n.a. AR 2 2 PMP,TP,TAP,SP Supervision 12 Sep EC N.A T N.A. Supervision 13 Dec IEC, SS 2 2 PMP, TP, TAP, SP Supervision 14 Apr-95 I n.a. SS S S PMP,PP,TP,TAP,SP Completion Sep IEC,SS S S PMP Key to specialized staff skills AR = Architect/Inplementation Specialist FP = Family Planning PH = Public Health EC = Economist IEC = Information, Education, Communication RA = Research Assistant ESS = EpidemiologistHealth Statistics LA = Loan Officer SS = Social Sector FH = Family Health MA = Management Key to perfornance ratina I = Problem Free 2 = Moderate Problems 3 = Major Problems 4 = Major Problems - Conective action to be takenl HS = Highly Satisfactory S = Satisfactory U.= Unsatisfactory HU = Highly Unsatisfactory Key to vroblems AF = Availability of Funds PMP = Project Management Performance SP = Studies Progress CLC = Compliance with Legal Covenants PP = Procurement Progress TP = Training Progress FP = Financial Performance TAP= Technical Assistance Progress.

38 26 3. APPENDICES A. MISSION'S AIDE-MEMOIRE Mission d'achevement du Projet Sante et Population I (PSP I) (Cr BU) septembre Mme Michele Lioy, Chargee de Projets au siege de la Banque Mondiale, a effectue, en collaboration avec M. Pamphile Kantabaze, Charge des Operations a la Mission residente du Burundi, une mission d'achevement du Projet Sante et Population I (PSP I), (Cr BR), au Burundi du 13 au 25 septembre Les objectifs de cette mission etaient de: (i) evaluer a posteriori les realisations du projet; (ii) reevaluer l'engagement de l'emprunteur envers les objectifs du projet; (iii) arriver a un accord avec l'emprunteur en ce qui concerne ses futures activites dans le secteur sante; (iv) confirmer les indicateurs de performance qui seront utilises pour le suivi des operations; et 1'evaluation de l'impact; (v) analyser les risques principaux au succes des operations et (vi) collecter les donnees necessaires pour les annexes statistiques du rapport. Ces objectifs ont.t atteints. La mission a travaille avec les membres de l'unite de Gestion du Projet (UGP) et les cadres du Ministere de la Sante publique et a consulte les bailleurs concernes par le secteur sante. Elle a effectue une descente sur le terrain au cours de laquelle elle a visite deux centres de sante (Bisoro et Mahwa) et l'h6pital de Bururi. 2. La Mission residente a et tenue informee du deroulement de la mission d'achevement. La mission remercie tous ses interlocuteurs, en particulier les membres de l'ugp et les cadres superieurs du Ministere de la Sante publique, pour leur excellente disponibilite. Cet Aide-Memoire a et discute avec son Excellence Monsieur le Ministre de la Sante publique et ses collaborateurs le 25 septembre L'UGP a prepare un rapport d'evaluation interne qu'elle a presente a la mission d'achevement. Elle preparera une synthese de ce rapport qui sera mise en annexe du Rapport d'achevement du Projet. En outre, une etude d'evaluation aupres des beneficiaires est en cours. L'objectif de cette etude est d'evaluer la disponibilite, la qualite et l'accessibilite des services de sante dans les centres de sante rehabilites par le projet. Les resultats de l'etude, qui seront disponibles a la mi-novembre 1995, seront utilises pour completer le Rapport d'achevement. Realisations du projet 4. Le PSP I 6tait constitue de cinq composantes: (i) Developpement des services de sante maternelle et infantile; (ii) Elaboration d'un programme d'information, Education et Communication; (iii) Appui au Programme National de Lutte contre le SIDA; (iv) Renforcement institutionnel du Ministere de la Sante publique (MSP); et (v) Etablissement d'une base de donnees en matiere de population. Les paragraphes 5 a 12

39 27 ci-dessous decrivent brievement les realisations de chacune des composantes. II faut noter que les &evnements qui ont eu lieu au Burundi ont, dans certains cas, ralenti les progres du projet (rehabilitation des centres de sante), et dans d'autres cas, ont annule ou diminue certains des acquis qui avaient ete observes dans le secteur sante (par exemple, la couverture vaccinale ou la prevalence contraceptive). En consequence, lors de l'appreciation des realisations du projet, il est necessaire de considerer, non seulement les realisations a la fin du projet, mais aussi les realisations, par rapport aux indicateurs definis lors de la preparation du projet, avant la crise. 5. Developpement des services de sante maternelle et infantile. Certains des objectifs specifiques de cette composante avaient ete en partie realises avant la crise 5. En outre, certains indicateurs de performance avaient atteint (centres de sante rehabilites) ou depasse les niveaux prevus (la couverture vaccinale). A la fin du projet, le nombre de centre de sante rehabilites etait superieur au nombre prevu au moment de levaluation. Des fonds ayant ete r&alloues, 41 centres de sante, au lieu des trente prevus, ont et rehabilites/construits et la plupart de ces centres ont ete meubles et 6quipes6. La troisieme phase ayant et perturbee par la crise, la rehabilitation de deux centres reste inachevee et les trois unites de maintenance qui ont ete construites n'ont pas pu etre equipees avant la fin du projet 7. Les activites de planification familiale n'ont pas et, comme prevu, financees par le projet parce que l'emprunteur a beneficie de dons de I1USAIS et du FAUNE pour leur realisation. Cependant, la planification familiale a re,u un appui indirect du projet: (i) dans les centres de sante renoves, la salle reservee a la MI/PAF oui les femmes pouvaient 8tre conseillees en prive a permis d'ameliorer la qualite des services de planification familiale; et (ii) la Sous-Direction d'education pour la Sante (EPS) a appuye la realisation d'activites d'iec dans ce domaine. En ce qui conceme les services de SMI, les outils de gestion et les fiches techniques et de supervision mis en place avec l'appui du projet, les formations, le developpement d'un manuel SMI/PF, prepare avec le concours de l'unicef, et la mise a disposition dans les centres r6noves d'une salle de SMI/PF ou les femmes peuvent 8tre examinees et conseill6es en prive a contribue a ameliorer la couverture vaccinale et le nombre des consultations prenatales (CPN). En ce qui conceme la qualite des CPN, elle ne peut 8tre evalu6e car les donnees necessaires n'etaient pas collectees. En 1992, derniere annee pour laquelle des donnees sont disponibles avant la crise, beaucoup des indicateurs de performance concemant la SMI avaient et atteints ou presque atteints. Cependant, ce n'est plus le cas, car le systeme de sante publique a ete tres touche par la crise. Beaucoup de centres de sante et d'h6pitaux 5 De 1986 a 1992, les progres suivants avaient ete accomplis: la couverture prenatale (au moins une une visite prenatale) est passee de 60 % de la population cible a 92% en 1992; le taux d'accouchements assistes par du personnel mddical est passe de 17 % a 20%; le percentage de nourrissons de 0-1 Imois qui etaient suivis au moins cinq fois par an a augmanete de 46 a 80%. La prdvalence contraceptive, cependant, n'avait augmente que de 1,6 a 2,6. 6 Certains des centres qui ont ete construits pendant la troisieme phase n'ont pas encore recu les meubles etlou l'equipement qui leur est destine, mais qui sont stockes au MSP. 7 Une etude de faisabilite est prevue dans le cade du deuxieme projet qui permettra de determiner comment seront gdr&es ces unites de maintenance.

40 28 n'ont plus le personnel necessaire pour fonctionner, I'approvisionnement en vaccins et en medicaments est tres irregulier et, dans certains cas, 1'equipement a disparu. 6. Elaboration d'un programme di'nformation, Education et Communication. Le projet a contribue a etablir une Sous-Direction d'education pour la Sante (EPS) au Ministere de la Sante publique qui a pour mission de planifier, coordonner et contr6ler la qualite de activites IEC du Ministere. En outre, I'EPS doit appuyer les activites IEC des programmes du MSP. Le projet a finance: (i) des formations pour le personnel de la Sous-Direction et les antennes IEC des programmes du MSP; (ii) de l'equipement audiovisuel 8 pour la production de materiel IEC; et (iii) de I'assistance technique ponctuelle 9. En outre, le projet a aide l'eps a apporter un appui technique (conception des campagnes et du materiel, formation, recherche) et financier aux programmes de sante du MSP. Entre autres, les programmes suivants ont beneficie de I'appui de I'EPS: le Programme National de Lutte contre le SIDA (PNLS), le Programme de Lutte contre les Maladies diarrheiques (LMD), le Programme de Lutte contre les Maladies transmissibles et carentielles (LMTC), et le Programme national de planification familiale (CPPF). Bien que des etudes sur la couverture vaccinale et l'etude CAP sur le V'IHSIDA indiquent que les activites LEC ont eu un impact positif sur les populations, il est difficile d'evaluer l'impact des activites EEC car I'EPS ne faisait pas d'evaluation systematique de toutes les activites IEC. II a aussi ete note une faiblesse dans les capacites de gestion et de coordination de I'EPS lorsque les demandes des differents programmes sont devenues nombreuses. 7. Une autre faiblesse etait le manque de capacites IEC aux niveaux provincial et peripherique. Pour remedier a cette carence, quelques mois avant la crise, des animateurs d'iec avaient ete nommes au niveau des provinces pour planifier et coordonner les activites de communication et former les techniciens d'assainissement comrnunaux. Cette decentralisation n'est jamais devenue operationnelle parce que la crise a empeche la formation des agents IEC provinciaux. De plus, il est a noter que depuis la crise, 1'efficacite de l'eps a diminue car il est plus difficile pour ses agents de sortir sur le terrain et beaucoup de son personnel est parti. 8. Appui au Programme National de Lutte contre le SIDA. Les fonds alloues a la composante SIDA ont ete re-alloues a la construction de: i) du batiment ou sont loges le siege du projet PSP I et les bureaux et studios de la Sous-Direction d'education pour la Sante; et 2) des centres de sante de la phase trois parce que le MSP a beneficie d'aides bilaterales et multilaterales pour la realisation des activites de prevention du SIDA. Le projet a cependant fourni un appui technique et financier aux activites IEC du PNLS (voir para. 6 ci-dessus). 8 L'UNICEF a fourmi un premier lot de materiel au debut du projet. Le projet a complete cet equipement et a remplace certains des appareils originels. 9 L'UNICEF a finance un assistant technique a long terme, specialiste en audiovisuel, qui a aide a demarrer l'eps.

41 29 9. Renforcement institutionnel du Ministere de la Sante publique (MSP). Dans le cadre de cette composante, le projet a financ& (i) une assistance technique a long terme (2 ans) en planification; et (ii) des etudes sur l'autonomie de gestion des h6pitaux et le financement du systeme de la sante en general et sur la decentralisation et le developpement des ressources humaines. Pour renforcer les capacites du MSP au niveau central, du personnel qualifie a ete recrute localement, notamment pour le Bureau d'inspection et de planification (BIP) et pour les sous-directions de l'epidemiologie et Statistiques Sanitaires (EPISTAT) et de l'eps. En outre, un appui logistique (vehicules et materiel informatique) a et fourni aux principaux services centraux. Le projet a assure aux cadres et personnels de secretariat la formation en traitement de texte et gestion de tableurs. Avec l'appui financier et technique de l'usaid, le service EPISTAT a ete informatise, a developpe les outils de collecte des donnees et forme le personnel de sante des niveaux intermediaire et local. Cependant, le manque de financement risque de compromettre les acquis du PSP I en ce domaine. Neuf bourses d'etudes de longue duree (maitrise) ont ete financees par le projet: 2 en administration et gestion des services de sante; 3 en IEC, 2 en sante communautaire et 2 en epidemiologie. Mis a part les deux epidemiologistes attendus cet automne, tous les boursiers sont rentres et ont ete affectes de maniere efficiente. Les services logistiques du MSP ont beneficie du soutien technique de la cellule architecturale de l'ugp tant dans l'elaboration que dans le suivi de l'execution des programmes d'investissement publics finances par le gouvemement et les autres partenaires. Le niveau d'execution du budget extraordinaire et d'investissement (BEI) est ainsi passe de 35% a 99% durant le cinquieme plan quinquennal Malgre les acquis mentionnes ci-dessus, les ressources humaines restent la principale contrainte relevee, tant au MSP qu'au niveau du projet. En effet, dans beaucoup de cas, le personnel forme dans le cadre du projet est engage par le secteur prive qui leur offre des emplois qui paient mieux. Le MSP accusait deja des deficits importants en 1993, mais la crise a fortement aggrave la situation. Plusieurs centres de sante (CS) fonctionnent avec un seul, voire sans personnel qualifie et certains hopitaux n'ont plus de medecins et fonctionnent temporairement avec le personnel d'ong d'urgence. Un plan de redeploiement et d'incitation du personnel en postes eloignes sera elabore et mis en oeuvre par le PSP II. Avec le concours technique et financier des autres partenaires, le PSP H aidera le MSP A developper une politique et des strategies de developpement des ressources humaines a long terme. 11. Les efforts entrepris avant 1993 avaient donne l'espoir au MSP de: (i) mettre en place une politique sectorielle et un sixieme plan quinquennal ; (ii) consolider les processus de programmation-budgetisation a travers la cellule PIP-PDP- PCT; (iii) demarrer le processus de decentralisation base sur les bureaux provinciaux de sante (BPS) et la mise en autonomie progressive des h6pitaux et CS; (iv) demarrer une strat6gie de financement basee sur I'assurance maladie pour le secteur non structure. Le PSP II appuiera le gouvernement pour re-analyser la situation generee par la crise et adopter un plan readapte pour la rehabilitation et la relance des reformes sectorielles concernant ces questions. Ce plan inclura notamment l'61aboration de normes et standards pour l'allocation et la gestion des ressources, le fonctionnement des h6pitaux et des CS et les prestations des services de sante qui devait etre faite pendant le premier projet.

42 Etablissement d'une base de donnee en matiere de population. Cette composante a realise ses objectifs. Le projet a contribue au financement de l'operation censitaire decennal de 1989, execute par le Ministere de l'interieur avec l'appui technique et financier du FNUAP. Depuis 1990, les resultats du recensement constituent la reference pour divers usagers publics et prives et sont disponibles au departement de la population. A l'occasion d'une recente mission d'evaluation effectuee par le FNUAP, la mission residente de la Banque et les evaluateurs se sont convenus de recommander au Gouvernement de preparer un plan d'action pour I'actualisation et l'amelioration des statistiques demographiques nationales Gestion du Projet 13. Le projet a ete gere par une succession de huit fonctionnaires du MSP nomme par le Ministre. Le Directeur de projet gerait les ressources de toutes les composantes, c'est a dire les responsables de composantes planifiaient et realisaient les activites de leur composante, mais ne geraient pas leurs ressources. Cette succession de directeurs et cette centralisation a cause des retards dans le demarrage des activites, exceptees celles de la cellule architecturale. Le systeme comptable n'a pas ete informatise ce qui rendait plus difficile la gestion des fonds. Bien qu'il n'y ait pas eu d'audits qualifies, certains problemes comptables ont ete observes en L'engagement d'un comptable qualifie en decembre 1994 a permis de remedier a la situation et, de maniere generale, d'ameliorer la gestion financiere du projet. Depuis la crise, la gestion du charroi est devenue un probleme a cause des nombreuses requisitions qui non seulement privent le projet de moyens de transport pour executer ses activites, mais qui entrainent des frais supplementaires, car les vehicules requisitionnes sont souvent endommages. Prestations des services de la Banque Mondiale 14. Le Gouvernement est generalement satisfait du service foumi par la Banque. II a constate qu'il y a eu une continuite dans les discussions, les negociations et le suivi, ce qui a ete important pour l'execution du projet dans un contexte ouj il n'y avait pas eu d'operations a l'echelle sectorielle au prealable. Le Gouvernement a aussi note qu'il etait particulierement satisfait par: (i) le fait que la banque avait une approche sectorielle et s'interessait a des domaines tels que l'appui institutionnel et l'iec qui allaient au dela de programmes specifiques et que les autres partenaires ne financaient pas; (ii) I'approche sectorielle qui a permis d'identifier les activites prioritaires dans le cadre de la politique sectorielle et ainsi de "canaliser" et de mieux coordonner les activites financees par les divers bailleurs de fonds concernes par le secteur de la sante; (iii) la flexibilite de la Banque qui a permis de re-allouer des fonds du projet quand des aides bilaterales et multilaterales ont ete mises a la disposition du Gouvemement pour financer des activites prevues pour le projet; (iv) 1'etablissement de normes et standards qui sont maintenant utilises par la plupart des partenaires; (v) le choix des membres des missions de preparation et de supervision qui avaient, en general, des specialites/competences pertinentes aux questions qui devaient 8tre traitees; et (vi) le fait que le projet a stimule le developpement du secteur prive (petites et moyennes entreprises) en donnant la preference

43 31 aux entreprises nationales et locales, en particulier dans le domaine de la construction. 11 aurait toutefois souhaite que les missions de supervision soient plus frequentes. Lecons-cles et future operation 15. Comme il l'a ete indique ci-dessus, le projet a etb realise d'une maniere satisfaisante jusqu'a la crise, moment ou il avait en partie realise ses objectifs. Cependant, des lecons peuvent Etre tirees de cette experience, en particulier: (i) le changement frequent des directeurs de projet a cause des retards et son mode de selection n'assurait pas qu'il avait toutes les competences requises pour le poste; (ii) le systeme de gestion financiere et administrative n'etant pas informatise rendait la gestion du projet moins efficace et moins transparente; (iii) la gestion tres centralisee du projet a cause des retards qui auraient pu 8tre evites si les responsables de composantes avaient ete plus impliques; (iv) les bureaux provinciaux ont besoin de ressources et d'inforrnation suffisantes pour planifier, gerer et superviser les activites de sante au niveau de la province; (v) les capacites de gestion et de planification de l'eps ont besoin d'8tre renforcees pour permettre a l'eps de repondre aux demandes de tous les programmes et assurer la sensibilisation de la population quand les reformes sectorielles seront mises en place; (vi) la mission de l'eps n'est pas claire pour tous ses partenaires et doit 8tre clarifiee; (vii) les capacites IEC ont besoin d'etre decentralisees pour que les activites de communication interpersonnelle au niveau peripherique puissent etre efficaces; et (viii) il est difficile de suivre la realisation des activites et d'evaluer la qualite de certaines activites (par exemple, des consultations prenatales) parce que le systeme d'information ne collecte pas toutes les informations necessaires, en consequence ces systemes ont besoin d'etre revises pour assurer que tous les indicateurs de performance et d'impact qui ont ete identifies lors de la preparation du projet puissent etre suivis regulierement. 16. Lors de la conception du Deuxieme Projet Sante et Population (PSP II) qui a ete n6gocie en mars 1995 et signe en juin 1995, ces lecons ont ete prises en consideration et une solution a ete proposee pour chacune de ces contraintes. Les objectifs de cette deuxieme operation dans le secteur de la sante sont de: (i) renforcer la capacite du MSP a concevoir et mettre en oeuvre des reformes sectorielles majeures; (ii) ameliorer la disponibilite et la qualite des soins de sante en milieu rural; (iii) contribuer a reduire le taux de croissance de la population et les taux de mortalite matemelle et infantile; et (iv) promouvoir les comportements benefiques vis A vis de la sante et de la planification familiale. Ces objectifs demontrent l'engagement continu du Gouvemement envers les objectifs du premier projet. les principaux risques au succes de cette nouvelle operation restent les m8mes que ceux qui sont presentes dans le Rapport d'evaluation du Projet (para. 6.2, page 40). Washington, le 17 octobre 1995