Malawi - First and Second Health Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 14741 PERFORMANCE AUDIT REPORT MALAWI HEALTH PROJECT (CREDIT 1351-MAI) AND SECOND FAMILY HEALTH PROJECT (CREDIT 1768-MAI) JUNE 30, 1995 Operations Evaluation Department 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 without World Bank authorization. Currency Equivalents (February 26, 1987) Currency Unit = Malawi Kwacha (MK) SDRI.00 US$1.27 US$1.00 = MK2.00 (September 10, 1993) US$1.00 = MK3.96 Abbreviations and Acronyms CBD Community Based Distribution CHSU Community Health Sciences Unit CMS Central Medical Stores CS Child Spacing DHS Demographic and Health Survey EEC European Economic Community EP&D Department of Economic Planning and Development EPI Expanded Program of Immunization FP Family Planning GNP Gross National Product IDA International Development Association HIS Health Information System IEC Information Education Committee KFW Kreditanstalt fur Wiederaufbau MCH Maternal and Child Health MOCS Ministry of Community Services MOH Ministry of Health MOI Ministry of Information MOWS Ministry of Works and Supplies ODA Overseas Development Administration PCR Project Completion Report PHC Primary Health Care PHN Population Health and Nutrition PIU Project Implementation Unit PPF Project Preparation Facility WHO World Health Organization SAR Staff Appraisal Report TA Technical Assistance UNDP United Nations Development Programme UNFPA United Nations Fund for Population Activities USAID United States Agency for Industrial Development Fiscal Year Government April 1-March 31 FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation June 30, 1995 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Malawi Health Project (Credit 1351-MAI) Second Family Health Project (Credit 1768-MA1) Attached is the Performance Audit Report (PAR) on the Malawi Health project (Credit 1351-MAI, approved in FY83) and the Second Family Health project (Credit 1768- MAI, approved in FY87) prepared by the Operations Evaluation Department. This report presents an audit of the first two IDA credits provided to Malawi for its health and population programs. The first (Credit 1351) focused on strengthening the planning and administrative capacity of the sector, reducing costs of pharmaceutical procurement and distribution, and introducing a family planning program. While continuing some of these activities, the second project (Credit 1768) aimed at a substantial expansion of the primary health care system (also supported in the first project) and introduced population and health related activities (e.g., literacy, nutrition and women's programs) in non-health sectors. Implementation of the first project took six years rather than the planned two because of shortages of skilled personnel and materials, inadequate counterpart funding and poor administrative procedures. Although most planned activities were eventually completed, the primary goals-strengthening planning and administrative capacity and reducing pharmaceutical costs-were not satisfactorily achieved. Hence, the project outcome is rated as unsatisfactory. Institutional development is rated as negligible, and sustainability as uncertain. Implicitly, the Project Completion Report (PCR) agrees with the second two ratings, but suggests that the project outcome was satisfactory. Implementation of the second project was satisfactory and most physical goals were achieved. Since expansion of the primary health system was a major part of this project, the outcome is rated as satisfactory; but institutional development is rated as negligible and sustainability as uncertain. These ratings are consistent with the findings of the PCR. The audit attributes the weak institutional development performance of these projects to shortages of skilled manpower, largely the result of budgetary shortages and poor pay scales, plus the generally poor performance of credit financed technical assistance. The audit concludes that if the performance of Malawi's health sector is to improve, greater allocations from the recurrent budget are essential, accompanied by improvements in efficiency and further capital investment. Attachment 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 without World Bank authorization.  FOR OFFICIAL USE ONLY Contents Preface . ............................................................... 3 Basic Data Sheet .. ........................................................ 5 Evaluation Sum m ary ..................................................... 9 1. Background ......................................................... 15 2. Health Project (Health I) .............................................. 17 Design and Objectives ................................................. 17 Implementation....................... ........................ ..... 19 Borrower Compliance with Credit Agreement ............................... . 22 B ank Supervision .................................................. 22 3. Second Family Health Project (Health II) ................................... 23 Design and Objectives ............................................... 23 Implementation .... 24 Borrower Compliance with the Credit Agreement ............................ 25 Bank Supervision .................................................. 26 4. Project Results and Achievements ........................................ 27 Health Status ..................................................... 27 Extension of the Health Delivery System into the Countryside.................... 28 Family Planning ................................................... 29 Efforts to Strengthen Health and Health Manpower Planning .................... 30 Community Health and Services Unit (CHSU) .............................. 31 Pharmaceuticals ................................................... 32 IEC and Multisectoral Activities ........................................ 34 Medical School .................................................... 34 Bank Performance .................................................. 36 5. Conclusions and lessons ................................................ 37 R atings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 L essons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Annex 1 ............................................................... 41 This report was prepared by Ronald G. Ridker (Task Manager) who audited the project in November 1994. Pilar Barquero provided administrative support. 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 widLout World Bank authorization.  3 Preface This is a Performance Audit Report (PAR) of the Government of Malawi's first two health projects. They were supported by two Bank credits, Credit 1351-MAI for SDR 6.3 million (US$ 6.8 million equivalent) approved on April 21, 1983 and Credit 1768-MAI for SDR 8.7 million (US$ 11.0 million equivalent) approved on March 24, 1987. The first credit closed on December 31, 1988, four years behind schedule. The last disbursement was on July 31, 1989 and the sum of SDR 37,824 was cancelled. The second credit was closed as scheduled on June 30, 1993. The last disbursement was on November 19, 1993 and the sum of SDR 2,233 was cancelled. This PAR is based on the Staff Appraisal Reports, the Credit Agreements, supervision reports, project files and government documents. Additional information was obtained through discussions with Bank staff, bilateral and other donors, government officials and others associated with the projects, medical personnel and project beneficiaries, the latter groups being contacted during a visit to Malawi in November 1994. Sections 2 and 3 review the basic facts about these projects and, with minor exceptions confirm the findings of the PCRs. Section 4 takes up several issues not covered in the PCRs and goes into others in more depth. Similarities and differences in ratings and lessons are discussed in Section 5. Following standard OED procedures, copies of the draft PAR were sent to the Borrower for comments on April 14, 1995. Comments received are attached as Annex 1.  5 Basic Data Sheet HEALTH PROJECT (CREDIT 1351-MAI) Key Project Data (Amounts in US$ million) Appraisal Actual Actual as % of estimate appraisal estimate Project Costs 8.7 n.a. - Credit Amount 6.8 6.20 91 Cancellation 0.0 0.04 0.06 Disbursed 6.8 7.49 110 Institutional Performance Negligible Cumulative Estimated and Actual Disbursements (SDR million) FY84 FY85 FY86 FY87 FY88 FY89 FY90 Appraisal Estimate 1,650 4,800 6,300 - - - - Actual 119 1,914 2,446 2,780 3,810 5,916 6,262 Actual as % of Appraisal 7 40 39 44 60 94 99 Date of Final Disbursement July 31, 1989 a. As of July 31, 1989, SDR 37,823.59 was cancelled. Last application: No. 157 for K 521,892.4 dated July 31, 1989. Project Dates Item Planned Date Revised Date Actual Date Health Sector Mission 10/1980 Identification Mission 09/1981 (Initial project discussion with IDA) 05/17-22/82 Preparation Mission 09/21/81 09/21/81 09/21/81 Appraisal Mission 07/07/82 07/07/82 07/07/82 Credit Negotiations 03/15/83 03/15/83 03/15/83 Board Approval 04/26/83 04/26/83 04/26/83 Credit Signature 05/20/83 05/20/83 05/20/83 Credit Effectiveness 07/01/83 08/22/83 08/22/83 Project Completion 06/30/85 06/30/88 Credit Closing 12/31/85 12/31/86 12/31/88 12/31/87 6 Staff Inputs (staff weeks) FY80 FY81 FY82 FY83 FY84 FY85 FY86 FY87 FY88 FY89 FY90 FY94 TOTAL Preappraisal 49.6 55.0 104.7 Appraisal 49.3 49.3 Negotiation 7.7 7.7 Supervision .5 31.2 19.5 3.9 13.4 12.1 6.4 12.5 .1 99.6 Other .5 .8 2.6 5.4 9.3 Total .5 50.4 57.6 62.9 31.2 19.5 3.9 13.4 12.1 6.4 12.5 .1 270.6 Mission Data Mission Time Participantsa Staff weeks Performance statusb Health Sector Mission 10/80 Appraisal 07/82 F M DI OS0 Supervision 1 09/83 2(A,PH) 2.0 1 1 - 1 Supervision 2 12/83 2(A,A) 2.0 1 1 - 1 Supervision 3 02/84 4(E,M,P,PH) 7.0 1 2 1 2 Supervision 4 06/84 4(A,H,P,PH) 7.0 1 1 1 2 Supervision 5 10/84 4(HP,M,P,PH) 4.0c 1 1 1 2 Supervision 6 04/85 3(A,HP,PH) 3.0c I 1 1 2 Supervision 7 06/86 4(A,E,MS,PH) 2.0c 1 1 1 2 Supervision 8 09/86 3(A,E,MS) 4.0c 1 1 1 2 Supervision 9 04/87 2(A,E) 2.0c 1 1 1 2 Supervision 10 08/87 1(E) 1.oc I 1 1 2d Supervision 11 11/87 3(A,E,PH) 2.0 1 1 1 2 Supervision 12 03/88 3(A,E,PH) 2.0c 1 1 1 1 Supervision 13 10/88 3(A,E,PH) 1.5c 1 1 1 1 Supervision 14 03/89 3(A,E,PH) 1.5c 1 1 1 1 TOTAL 45.0 Proj. Completion 11/89 2(A<E) 4 - a. A = Architect; E = Economist; H = Health Advisor; HP = Health Planning Specialist; M = MCH/FP Specialist; MS = Management Specialist; P = Pharmaceutic Specialist; PH = Public Health Specialist. b. F = Available Funds; M = Project Management; DI = Development Impact; OS = Overall Status. c. The Health Project was supervised together with the preparation/supervision of the Second Family Health Project. d. No Form 590 on record. 7 Basic Data Sheet SECOND FAMILY HEALTH PROJECT (Credit 1768-MAI) Key Project Data (Amounts in US$ million) Appraisal Actual Actual as % of estimate appraisal estimate Project Costs 24.9 25.1 100.8 Credit Amount 11.0 11.6 105.4 Cancellation 0.0 0.0 Disbursed 11.0 11.6 105.4 Institutional Performance Negligible Cumulative Estimated and Actual Disbursements (SDR million) FY88 FY89 FY90 FY91 FY92 FY93 Cumulative Estimate at 0.66 2.97 6.05 8.69 10.45 11.02 Appraisal Estimate as % of the Loan 11 27 55 79 95 100 Cumulative Actual 1.528 5.053 7.577 10.315 11.104 11.595 Actual as % of the Loan 13 44 65 89 96 100 Date of Last Disbursement June 9, 1993 Project Dates Step Planned Actual Identification 06/22-29/85 06/22-29/85 Preparation 09/19-30/85 09/19-30/85 Appraisal 01/86 12/02-20/85 Post appraisal 06/09-17/86 06/09-17/86 Negotiations 09/86 02/02-06/87 Board Approval 11/86 03/24/87 Signing Date 05/07/87 05/07/87 Credit Effectiveness 06/15/87 06/15/87 Completion Date 12/31/93 12/31/93 Closing Date 06/30/93 06/30/93 8 Staff Inputs (staff weeks) FY83 FY84 FY85 FY86 FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 TOTAL Preappraisal 8.6 3.4 34.3 46.3 Appraisal 55.0 30.1 .5 85.6 Negotiation 8.2 8.2 Supervision .6 5.1 14.1 10.4 11.5 19.6 9.0 13.9 21.2 105.3 Other .2 .4 5.5 12.8 18.9 Total .2 8.6 3.8 95.4 56.3 14.1 10.4 12.0 19.6 9.0 13.9 21.2 264.3 Mission Data Mission Time Participants' Staff weeks Performance status' Identification 06/85 2 (Ph,E) 2.3 Preparation 08/85 6 (Ph, Pop,E, 10.3 IEC, Arc, Demo) Appraisal 12/85 6 (2 Ph, E, Arc. 15 IEC, Res) Post Appraisal 06/86 5 (Ph, Arc, 3.2 IEC,E, Res) Implementation 11/86 3 (IEC, Arc, E) 3 Mission Launch 04/87 2 (Arc, E) 1.6 G P M F Supervision 1 11/87 3 (E, Ph, Arc) 2.1 1 2 2 nr Supervision 2 03/88 3 (E, Ph, Arc) 1.8 2 2 1 nr Supervision 3 10/88 3 (E, Ph, Imp) 2.1 2 2 1 nr Supervision 4 03/89 3 (E, Ph, Arc) 2.8 1 1 1 1 Supervision 5 10/89 2 (Ph, Arc) 1.1 1 1 2 1 Supervision 6 03/90 2 (Ph, E) 2.0 1 1 1 1 Supervision 7 06/90 2 (Arc, E), 2.8 1 2 1 1 Supervision 8 10/90 2 (E, Arc) 2.0 1 1 1 1 Supervision 9 03/91 2 (E, Ph) 1.6 1 2 1 1 Supervision 10 06/91 3 (E,Ph, Imp) 3.2 1 1 1 1 Supervision 11 12/91 2 (E,Ph) 2.6 1 1 2 1 Supervision 12 06/92 2 (Ph, Pop) 2.8 2 3 2 2 Supervision 13 03/93 2 (Ph, Pop) 0.7 2 3 2 2 a. Arc = Architect; Demo = Demographer; h = Economist; lEC = Information Education and Communication specialist; Imp Implementation Specialist; Ph = Public Health Specialist; Pop = Population Specialist; Res = Research Assistant. b. G = General Status; P = Procurement; M = Management; F = Availability of Local Funds; nr = not rated. 9 Evaluation Summary Objectives 1. This is a Project Audit Report for the first two health projects that the Bank has supported in Malawi (Credits 135 1-MAI and 1768-MAI), the goal of which was to improve the capacity of the health system to reduce mortality, morbidity and fertility rates. 2. The first project, approved in 1983, was to cover only two years and be focused on strengthening the planning and administrative capacity of the Ministry of Health and the development of a national health plan. The rationale behind this orientation was that such a plan was necessary before deciding on major investments and that, because of the grave financial problems of that time, the Government could not afford any project that significantly increased its recurrent budget requirements. In addition, the project assisted in the establishment of a Community Health Sciences Unit (CHSU) to collect and analyze epidemiological data, assisted the Central Medical Stores (CMS) improve its procedures and cut its costs, extended the primary health care system which had been piloted in a few locations to the whole of three districts, and introduced a child spacing (CS) and maternal and child health (MCH) program. While some of these components would increase recurrent budget requirements, the CMS component was expected to result in sufficient savings to more than offset these increases. 3. The second project was derived from the national plan produced in conjunction with the first project plus a newly completed population sector report. The first of its two parts supported a more substantial expansion of the primary health care (PHC) system, technical assistance to strengthen the management, manpower and support systems of the Ministry of Health (MOH), additional support for the CS and MCH programs, and provision for the establishment of a Project Implementation Unit. The second part (approximately one eighth of total project costs) was to support population-and health-related activities in non-health sectors-so-called multisectoral activities-which included: functional literacy, nutrition and women's programs; a nation-wide Information, Education and Communications (IEC) program; and youth programs aimed at introducing integrated family health messages. Implementation Experience 4. The first project was extended three times and eventually took six years rather than the planned two years to complete. The primary factors explaining these delays were shortages of skilled personnel, materials and budgetary resources, a situation exacerbated by the Mozambican war, inflow of refugees, and deteriorating economic conditions during the implementation period, plus poor administrative procedures which to some extent were symptoms of the personnel shortages. In the end, however, most planned activities were completed and most covenants complied with. Shortfalls, while few in number were important and help explain the weak performance of this project: in particular, key personnel and counterparts to work with consultants were often not appointed or appointed only with substantial delays, a number of consultants failed to perform up to expectations, and several important studies and a few items of construction associated with the CMS were not completed (or had to be redone because of design errors). 10 5. While Bank supervision missions identified these problems as they emerged, supervision reports provided ratings that were overly optimistic; and after the first two years, attention was devoted to development of the second project. The result was less forceful supervision than was warranted given the shortfalls experienced by this project. 6. The second project fared better in the sense that nearly all planned activities were completed within the originally specified time period. Counterpart funds were adequate and timely and most covenants were complied with. Once again, however, there were important shortfalls in staff appointments; and several studies, including in this case a mid-term review, were not undertaken. In addition, the overall coordination committee did not meet, which may have accounted for lack of coordination of the multisectoral activities. 7. Bank supervision was supportive and timely and ratings were somewhat more realistic. But here again, the Bank failed to strongly voice concerns about shortfalls and seemed at times more concerned with development of the third project than with the effective implementation of this one. Project Results 8. Health Status. While Bank documents suggest that the health status has not improved over the last decade despite the substantial external resources invested in this sector by the Bank and other donors, the audit argues that the data are too weak to support this or any other contention. Moreover, even if this contention were correct, donor inputs and in particular, these two Bank supported projects, may have offset a deterioration that was likely to have occurred in their absence. 9. Extension of Health Delivery System. But whatever the impact of Bank and other donor inputs on the overall health status, these inputs must have had significant positive effects on the individuals directly affected. Interviews with users and staff of primary health centers plus other materials suggest that the extension of the health system provided care of a quality that recipients were satisfied with and had two major effects: first, they significantly reduced travel time for people living on the periphery of catchment areas, and second, they appear to have raised expectations and effective demand for health care which should lead to better practices-for example, a larger fraction of births taking place in clinics with the assistance of a trained nurse and more children being brought in for immunizations and checkups on a regular basis. 10. Family Planning. Considering the pro-natalist stance of the Government prior to 1993 and the fact that importation and sale of modem contraceptives were illegal until 1982, Malawi has come a long way during the course of these two projects. This is one issue that the Bank did forcefully press on, and with good effect. 11. Government policy is now openly supportive of family planning. Among other things, non-medical personnel are being allowed to distribute contraceptives subject to certain guidelines, services are available at least once per week in all hospitals and more than half of Government health centers, and non-governmental organizations are being encouraged to become involved in this field. Since the early 1980s, the contraceptive prevalence rate appears 11 to have risen from close to zero to seven percent in 1992, as sign that these measures are beginning to have at least some positive effects. 12. While all to the good, Family Planning (FP) experts interviewed by this mission point out that most changes in Government policy so far have only been permissive, not actively promotional. They see few signs of a shift towards a more active policy and fear that progress from this point forward will be much slower if such a policy shift does not occur. 13. Efforts to Strengthen Health Planning and Manpower Development Capacity. Technical assistance provided to both the Planning Unit and the Manpower Development Unit of the MOH had little lasting effect on the capacity of these units. Everything went wrong: a number of consultants performed poorly, counterparts to work with consultants were not appointed in a timely fashion, and staff turnover diluted the value of the training and fellowships provided by the projects. 14. The effort to develop a national health plan is a case in point. A plan of good quality was eventually produced but not by the expatriate team called in to produce it; that team's output was considered unsatisfactory and the Ministry eventually produced its own plan. But the capacity and sense of ownership created by this experience has not been sustained because of staff turnover. Today, the Planning Unit is in no better shape to produce such a plan itself than it was 12 years ago, and it is thinking of calling in another expatriate team of advisors to produce the next National Health Plan. The text also points out several organizational issues that have caused problems, for example, the assignment of statistical and manpower planning functions to organizations independent on the planning unit. 15. The CHSU was established but has never functioned well, in part because of shortages of budget and failure to fill a number of important staff positions, both signs of lack of significant support from the MOH, but also because of organizational problems. 16. The Pharmaceutical Component. The decentralization of medical stores planned by the project was eventually accomplished and has significantly decreased time required to deliver drugs to more remote areas. But there is no evidence that this and other planned actions have resulted in significant cost savings, an anticipated outcome of the project. 17. The multisectoral activities do not appear as yet to have accomplished much, although most planned activities have been executed. Efforts to introduce health and FP messages into literacy, home economics, and other women's programs have been limited by shortage of funds to support these programs. While nearly all the planned youth programs were initiated, they ceased when the Malawi Young Pioneers, a semi-political organization responsible for their implementation, was closed down in 1993. Many of the IEC activities got off to a slow start because of the need for training before they could be implemented. No evaluation of their impact is available. Findings and Lessons 18. While most of the activities planned under the two projects eventually were implemented, they have failed to achieve much lasting improvement in the capacity of the 12 health system. The principal exceptions to this are the extension of the health system into underserved areas and the introduction of a family planning program. Other achievements are less important or ephemeral. For example, while a national health plan was developed, the capacity to undertake health planning was not significantly improved. Accordingly, the outcome of the first project is rated as unsatisfactory although the PCR appears to rate it as satisfactory, while the outcome of the second project, which contained a much larger construction component and was not subject to delays in implementation, is rated as satisfactory in this report as well as in the PCR. However, the institutional development efforts had negligible impacts, an observation that is consistent with PCR findings. 19. Sustainability of the gains made is more difficult to judge. On the one side, the Government's recurrent budget remains under extreme pressure so that if it were not for donor inputs, the primary health care system might collapse. On the other, these inputs, plus the decentralization measures so far undertaken and the recent elections have generated expectations on the part of the rural population that are putting pressure on the Government to allocate more funds to the health sector and to the rural areas. In these circumstances, this audit rates the sustainability of both projects as uncertain, in agreement with the second PCR, but not the completion report for Health I. 20. In terms of lessons, these projects are excellent examples of the need for greater focus on Government ownership, implementation and supervision. Other issues pertain to personnel problems, recurrent cost problems, technical assistance and general project orientation. 21. Many of Malawi's problems stem from shortages and high turnover of personnel. The Bank's remedy for these personnel problems has been to provide training and fellowships and to insist on covenants that certain critical posts will be filled. But the MOH does not have adequate control over the situation to take fully effective action. Frequently, it does not have the budget or the authorization to fill a post or to pay sufficiently high wages to attract and keep good people. The Bank must deal with issues lying outside as well as inside the sector if it is to effectively help the MOH solve its manpower problems. 22. Lying behind these personnel problems is lack of budgetary resources. This is the fundamental problem with the health sector in Malawi-in the 1970s and early 1980s because of the low priority given to the sector and later because of the worsening economic and revenue situation. Improvements in efficiency, while sorely needed, will not be sufficient. In this situation, nor will the provision of investment funds be sufficient; help in meeting recurrent expenditures is also required. What is needed is a package of inputs, capital plus recurrent, that is adequate to the task; otherwise, the task should be scaled back. The Bank could have helped with this problem had it been, at that time, more flexible in its rules about recurrent cost financing. 23. A surprisingly large number of foreign advisors funded by these projects turned in disappointing performances. A common factor in these cases was that they worked more or less in isolation from those who ostensibly needed their inputs. Several recommendations for dealing with this situation are discussed in the text. 13 24. Finally, the report suggests that these projects erred too much on the side of specifying and attempting to achieve process objectives (training certain numbers of persons, establishing certain numbers of clinics, etc.) rather than health status objectives (immunization rates, contraceptive prevalence rates, and over longer periods, morbidity, mortality and fertility rates). This was in keeping with the thinking of that time which, fortunately, is now changing.  15 1. Background 1.1 Malawi is a densely populated, land-locked country with a modest and narrow resource base that makes it highly vulnerable to external shocks such as changes in world prices, weather conditions and disruptions in transport across its boarders. Despite these circumstances, for almost two decades after independence in 1964, it experienced reasonably good economic growth. Since the early 1980s, however, the country was faced with a series of economic problems that included deterioration in terms of trade, two oil crises, disruptions in transport routes through, and an influx of refugees from, Mozambique, a series of droughts and several reversals in economic policy. These problems, plus rapid population growth which averaged 3.2 percent per year during the 1980s, have resulted in a Gross National Product (GNP) per capita ($210 in 1992) that is lower today than it was in 1980. Throughout this period budgetary resources for recurrent expenditures of public services have been very limited and development budgets have been funded largely by donors. 1.2 The average life expectancy at birth is 44 years, a result among other things of very high infant, child and maternal mortality rates. While data are very poor, the incidence of such diseases as malaria, pneumonia, tuberculosis, cholera, gastroenteritis, measles and HIV/AIDs-related diseases appear to be very high by African standards. The severity of these diseases have been made worse by an environment conducive to the growth of parasitic disease vectors, poor nutrition, unsanitary sources of water, and a weak public health system. 1.3 The weakness of the public health system, in turn, results in large part from lack of trained personnel, inadequate budgets, weak financial control systems, internal inefficiencies and misallocations (excessively favoring curative over preventive and urban over rural, and overcrowding of essentially free public facilities compared to more costly private facilities).' In 1980 when discussion of the first health project began, half the technical posts of the Ministry of Health (MOH) were vacant, over half the districts lacked medical officers, there was one physician per 53,000 population, and the return rate for medical students sent abroad was roughly 20 percent (according to the Staff Appraisal Report for Health 1). Also at that time, only four percent of Malawi's very low GNP per capita was devoted to health. Slightly over half of these expenditures were by Government institutions. In recent years about US$5.50 per capita has been spent on health when, according to the 1993 World Development Report, a minimum package of basic health services appropriate for poor countries is estimated to cost about US$12 per capita. 1.4 Prior to 1980, the Bank's involvement in the health sector was limited to small components in eight agricultural projects, amounting in total to about US$4 million. In early 1980, after the Bank announced its willingness to lend for free-standing health projects, the Government of Malawi asked the Bank to consider providing assistance to its health sector. Since then the International Development Association (IDA) has provided assistance for three projects: US$6.2 million for the First Health Project (approved April 1983, completed June 1988). US$11.6 million for the Second Family Health Project (approved March 1987, 1. The SAR for Health I describes private facilities as underutilized. But one of the reasons people are willing to pay to go to these facilities is because they are less crowded. 16 completed December 1993) and US$55.5 million for the Population, Health and Nutrition Sector Credit (approved on May 31, 1991, estimated completion December 30, 1996). 17 2. Health Project (Health I) Design and Objectives 2.1 The Government's initial suggestion was for a sizeable construction project to expand training facilities, replace or upgrade rural health centers and provide staff living quarters. The Bank responded favorably but suggested that a health sector review be undertaken as a first step. A Bank mission for this purpose, which visited Malawi in the fall of 1980, concluded that, in addition to serious financial constraints facing the sector, there were four main clusters of problems: weaknesses in administrative support for health services, serious shortages of staff at all levels, failure of Government to make effective use of the private sector, especially mission-run facilities, and negligible planning. The mission identified several areas that might be suitable for Bank group assistance-improvement and expansion of the paramedical training program, provision of service facilities and housing in remote areas, assistance to improve performance of the Central Medical Stores and institute drug quality control procedures, and measures to improve transport and communications and strengthen health education activities. But it recommended that such support be provided only in the context of a plan for the overall development of the sector which at the time did not exist. 2.2 In October 1981, a Health Identification Mission visited the country to discuss the health sector review and a possible project. At this point, the MOH proposed Bank assistance for four areas: training at all levels, primary health care development, supply and distribution of drugs, and strengthening of the administration, with decentralization as the first step. With a few exceptions, the Government's proposals coincided closely with the conclusions of the health sector review. Accordingly, the mission concurred and recommended Project Preparation Facility (PPF) financing to help prepare a project along these lines. 2.3 Shortly thereafter, grave financial problems led the Ministry of Finance (MOF) to request all ministries to withhold project proposals, including requests for PPF. Hence, when a project preparation mission reached Malawi in February, 1982, it found that no preparatory work had been undertaken. The Government's financial problems, plus concerns about implementation capacity, led the mission to propose that the project be scaled down to involve mainly system improvement, the centerpiece of which would be the development of a national health plan, with substantial system expansion being deferred to a second project. These problems also led the mission to be especially concerned about the recurrent cost implications of its recommendations. In the end, it proposed a $5 million project to be implemented over two years and designed to have little or no net impact on the recurrent budget. The project was appraised in July 1982 and negotiated in March 1983. 2.4 In its final form, the project had five objectives with a component devoted to each. a) Improving MOH capacity to plan and administer its programs (ten percent of total costs at appraisal). This was to be accomplished through technical assistance and training to establish a permanent planning, evaluation and monitoring capability and to improve financial management, accounting and service statistics systems. A principal output of this component was to be a draft national health plan backed up by a series of studies of, among other 18 things, organization of the public and private health system, manpower planning, health information retrieval and analysis systems, budgeting and accounting procedures, and options for health financing. This component also included support for studies on infant and child mortality, on possible linkages between traditional and modern health sectors and on ways to improve the low level of utilization of private hospitals. This work was to be undertaken by the Planning Unit of the MOH which would receive 43 months of consultant services and 12 months of overseas fellowships for its staff. Assurances were obtained during negotiations that four consultants-a health planner, an epidemiologist, a financial analyst and a manpower planning expert-would be employed by the Government by September 30, 1983. b) Establishing a Community Health Sciences Unit (CHSU) (13 percent of total costs at appraisal), which would collect and analyze epidemiological data and include necessary laboratory services. This was considered essential because lack of such information was constraining the ability to plan, monitor and evaluate. This component was to renovate and extend an existing building to house the unit and to provide housing, equipment and vehicles, fellowships and consultant assistance (funded in part by the United Nations Development Programme (UNDP) and the World Health Organization (WHO)). A staff of six persons were to be recruited and trained for this Unit. Two years was expected to be needed to establish the facilities and train staff. From January 1985, four consultants would be provided by WHO with UNDP funding for two years to assist in establishing work programs and routines. c) Assisting the Central Medical Stores (CMS) improve efficiency in pharmaceutical procurement and distribution (33 percent of total costs at appraisal). This component involved, among other things, improvement of procedures, in-service training, establishment of two regional centers for manufacture and distribution and development of CMS into an independent cost center. Two pharmaceutical consultants were to be employed by the Government by September 30, 1983 and some assistance was to be provided by the core consulting team assigned to the planning unit. This component was expected to result in a substantial financial saving in operating costs which would more than offset the modest increases in recurrent costs resulting from other components of this project. d) Developing a Primary Health Care System for three districts (26 percent of total costs at appraisal). This component was meant to be the first phase of an effort to expand a primary health care (PHC) system throughout Malawi. This system, which had been developed and applied on a pilot basis by the MOH, was evaluated by a WHO/UNICEF mission in January 1982. The work plan developed by that mission became the basis of this component. In essence, it involved extending the system to the whole of three districts in 19 which the pilot programs were implemented and adding monitoring and evaluation activities. e) Introduction of Child Spacing (CS) services into Malawi's Maternal and Child Health (MCH) program (13 percent of total costs at appraisal). In November 1982, the Banda Government, which had been pronatalist, approved a paper prepared by the MOH outlining a plan to introduce child spacing activities in its MCH program. The Bank strongly encouraged this activity and attempted to ensure that the program would not flounder on legal or practical impediments. The component provided for the renovation and equipping of antenatal facilities at the Zomba General Hospital and 15 district hospitals, for the training of staff, and for a family formation study to investigate the causes of high fertility and infant mortality rates and help determine future priorities for the MCH program. 2.5 The MOH was given overall responsibility for project implementation, with components assigned to different offices. A coordinating committee chaired by the Principal Secretary was supposed to meet quarterly to ensure smooth implementation. 2.6 Total project cost was estimated to be US$8.7 million, US$6.8 million to come from the IDA, US$0.6 million from cofinanciers and the remainder from the Government. Actual project costs are estimated to be US$6.9 million less than planned because of devaluation and underestimation of the Government's contribution.' 2.7 In general, the focus and goals of this project were appropriate for Malawi at the time. But despite the fact that it was significantly scaled down from its initial conception, it still proved to be excessively optimistic about what could be accomplished in a brief period with limited inputs. More than two years should have been allowed for implementation; more time and care should have been taken developing the CHSU and pharmaceutical components before starting implementation; and more consulting time should have been allowed for the long list of tasks assigned to the Planning Unit given dearth of background information with which to work. Other design flaws are noted below in discussing individual components. Implementation 2.8 The PCR correctly summarizes the main variances between planned and actual implementation. The most serious variance was the lengthy delays experienced almost across the board: in civil works, in program development of the CHSU and the pharmaceutical components, and in preparation of the national health plan. In 1985 when the project was originally scheduled to end, only 40 percent of the credit had been disbursed. It took another three years to reach 60 percent and two years beyond that to reach 99 percent (the credit was closed and the remaining funds, SDR 37,824, were cancelled on July 31, 1989). In the end, after three extensions, the project took six years rather than the planned two years, to be completed. 2. This is only a rough estimate because the Government did not keep complete records of its own contributions to the project. 20 2.9 The factors accounting for these delays varied by component but in general resulted from shortages of skilled personnel and materials (made much worse by the advent of the Mozambique war and the sharp deterioration in economic conditions during the implementation period), failure of a number of Technical Assistance (TA) sub-components to perform as planned, and poor administrative procedures which to some extent were symptoms of the personnel shortages '. The following component-by-component review provides more details. a) Improving MOH capacity to plan and administer its programs. The contract for the core consulting team was signed in November 1983 and the team leader arrived shortly thereafter. In February 1984, the team proposed changes in the terms of reference, the term of the contract and substitution of personnel originally identified for specific tasks-changes with which both the Government and the Bank disagreed. In April, the consultants prepared a working document for the plan which the subsequent supervision mission found to be of extremely poor quality. In May, the contractor agreed with Government to replace several team members and to send someone at its own expense to clean up the work done so far. In October, after the team leader announced his intention to leave the country before all work was completed, the MOH suspended disbursements and in November reached agreement with the contractor to terminate the contract. A final draft of the National Plan was sent by the contractor to the MOH by the end of January 1985. At this point the MOH planning staff, with some consultant advice, began work on a substantial revision which was completed in December 1985. While this work proved to be arduous, in the end both the National Health Planning Committee and Bank supervision staff judged the final version of the plan to be of good quality, and it formed the basis for the development of the Second Health Project. Time was lost and fewer background papers of usable quality were prepared than planned; but the net result of this experience was a national plan of reasonable quality which was clearly owned by the Government. Unfortunately, this experience in learning-by-doing did not result in a sustained improvement in planning capacity because of subsequent staff turnover (see below).' b) Establishing a Community Health Sciences Unit (CHSU). While the civil works portion of this component progressed slowly for the reasons identified in paragraph 2.9, the training subcomponent made steady progress. However, difficulties in finding an acceptable head for the CHSU and its epidemiological 3. For example, the MOH failed at times to provide adequate architectural briefs to the MOWS and the MOWS was slow in preparing detailed design and tender documents. In both cases, shortages of architects and other technical staff was an important causal factor. Another example involves shifts in priorities which resulted in the neglect of other components: during the first two years key staff focused on resolving problems caused by failure of the core consulting team to produce an acceptable national health plan; thereafter, the staff focused on preparation of a second health project. Other examples are provided in the next section on results. 4. The PCR suggests that a permanent capacity was established. But it was written prior to the time that staff turn-over became endemic. 21 unit, in retaining staff returning from overseas training, and in obtaining technical assistance from the UNDP resulted in delays in startup activities even after civil works were completed. A baseline epidemiological survey which was to be completed during the course of this project was transferred to the second health project. At the time the PCR was written, in December, 1990, this agency had not yet begun to function properly, in large part because of failure to create posts for and recruit a senior epidemiologist, a microbiologist and laboratory technicians. c) Assisting the Central Medical Stores (CMS) improve efficiency in pharmaceutical procurement and distribution. In addition to the construction delays common to other components, this component was delayed because of MOH's dissatisfaction with the work of the initial pharmaceutical consultants, lack of availability of housing built for the CMS but occupied by non project personnel, protracted disputes between contractors and the Ministry of Works and Supplies (MOWS) over building design and failure to create staff posts in a timely fashion. At the time the PCR was written, outstanding issues included staff housing, the possibility that the drug manufacturing plant in Lilongwe would have to be redesigned and reconstructed, the need to develop appropriate computer, inventory and distribution systems, and the recruitment of qualified personnel. d) Expansion of the Primary Health Care System. This component came close to fully meeting its objective of extending the PHC program into all villages in three districts where the program had been previously piloted, and it was completed ahead of schedule. As of December 1990, however, sustainability was in question. The national coordinator of the National PHC Committee was the only person in MOH attending to PHC matters; he had no support staff, no funds for field visits, no vehicles and no funds to effectively stimulate community-based programs (PCR, para 18). In addition, at the time of project completion, a number of centers that had been physically completed could not be opened because of lack of staff or water. In retrospect, the three PHC training centers constructed under this component were unnecessary. Because of delays in construction, training was completed elsewhere and these buildings have never been used for training. In addition, the Government (correctly) decided that any additional training would take place in the communities rather than in classrooms. As a result, the training units have not been properly staffed, supplied and maintained. e) Introduction of Child Spacing (CS). Surprisingly, this component proved to be a bright spot. During project preparation, it was not even clear that this initiative would be approved by the pro-natalist Government of that time. Once it was approved and services began to come on stream, demand for services grew beyond expectations. Construction and training programs were completed ahead of schedule and the training program exceeded original targets. The principal bottleneck to more rapid progress was unanticipated 22 shortages of contraceptives, which donors were unable to overcome quickly. Each supervision report flagged this problem. The Bank offered to provide injectables for one year until they were forthcoming from the United Nations Fund for Population Activities (UNFPA). Other factors that constrained progress included shortages of nurses qualified in child spacing, lack of in- service training and infrequent supervision of clinical services-all, at least to some extent, problems of unanticipated growth in demand. Borrower Compliance with Credit Agreement 2.10 Compliance with the credit agreements was mixed. In general, appropriate procedures were followed. But key personnel and counterparts to work with consultants were often not appointed or appointed only with substantial delays, audit reports were delinquent for every fiscal year during the project period, and records of project progress and non IDA expenditures were incomplete or lacking. Bank Supervision 2.11 Supervision reports indicate a good understanding of the problems encountered by this project. But performance ratings did not reflect the seriousness of these problems. Considering the management difficulties, staff shortages and implementation delays experienced, it is surprising to find ratings of one for project management and development impact and one (at the beginning and end of the project) and two but "improving" during the remainder of the project, for overall status. Part of the problem may have been the fact that three task managers were involved in this project. In addition, after the first two years, attention during missions with supervision responsibility turned toward preparation of the second and then later the third health project. It is quite likely, as the PCR notes, that more forceful supervision could have contributed to earlier solutions to the problems of this project. 23 3. Second Family Health Project (Health II) Design and Objectives 3.1 Following completion of an acceptable draft of the National Health Plan and a Population Sector Review, the MOH developed a proposal to obtain financing for the first half of the 1986-1995 plan period. The plan called for a reorganization and strengthening of the health system to prepare for its decentralization, substantial manpower development at all levels including the creation of a medical school in Malawi, expansion of the family health program (previously called the primary health program), expansion of the child spacing activities within that program, and efforts to involve non health sectors in the promotion of population and health messages. To cope with the recurrent cost implications of this program, which would be substantial, the MOH envisioned the introduction of additional cost recovery measures and studies to find ways to improve efficiency of both public and private health facilities. 3.2 The Bank, while approving the general thrust of the MOH proposal, agreed to provide funding for only the first three years, the main reason being to set aside for more detailed study two activities that would have commenced after the third year: the development of a medical school and a regional hospital for the north. The Bank also requested, among other things, rethinking of the initial proposal for the reorganization of the MOH before negotiations. 3.3 As it finally emerged, the project consisted of two parts, Part A to be implemented by the MOH and Part B to be implemented by various agencies coordinated by the Department of Economic Planning and Development (EP&D). Part A (81 percent of appraisal base cost) included four components: a) Strengthening management, manpower and support systems through technical assistance, training, studies (including baseline and evaluation studies left over from the first project and a study on the potential of health insurance), construction of pharmaceutical depots in district hospitals and training facilities, and provision of computers for MOH and hospital budget management. b) Expanding the PHC system from 9 to 15 districts by providing training, equipment and supplies for village health committees, constructing and equipping 19 new health centers, upgrading six existing health sub-centers, and replacing one existing district hospital. c) Support for the MCH and CS programs through technical assistance, training, provision of vehicles, equipment and construction (surgical contraceptive units in eight district hospitals and several new or replaced urban health centers). d) Support for a Project Implementation Unit for Part A. 24 Part B (19 percent of appraisal base cost) consisted of e) Functional literacy, nutrition and women's programs under the Ministry of Community Services (MOCS) which included construction and equipping of a regional training center; f) Youth programs under the Department of Youth, to introduce family health messages; and g) Development of nationwide Information, Education and Communications (IEC) programs for population and health activities under the Information Department. 3.4 Thus, this second project was similar to the first except that it contained a larger construction element, greater specificity in some areas (eg., family planning), and support for multi-sectoral activities. It does not, however, include support for the CHSU and the CMS components which were left unfinished at the end of the first project. The PCR and the files give no hint as to why support for these institutions was not continued. Interviews suggest that the Bank became disillusioned with the CHSU and decided to cut its losses because it was not accomplishing what it was supposed to and had lost its support in the MOH when new management came in. No explanation could be found for the decision to cease support for the CMS, a substantial (one third of base costs) component of the first project. 3.5 Total project cost was estimated at appraisal to be US$ 24.9 million, US$ 11.0 million to come from IDA, US$ 1.9 million from the Government, and the remainder from five cofinanciers. Actual costs totalled US$25.1 million after some shifts in contributions by various donors. Implementation 3.6 Implementation went much more smoothly than in the first project. Disbursements more or less followed the planned schedule (except for currency fluctuations) and only SDR 2,233 was cancelled. The project was closed (6/30/93) and completed (12/31/93) on schedule. Once again, however, many of the problems that existed are directly or indirectly related to shortages of skilled manpower. a) Strengthening Management, Manpower and Support Systems. The Government's Complement and Grading Review Committee developed a detailed reorganization plan for the MOH which was eventually approved and implementation initiated. Among other things, Regional Health Teams were established, a plan to introduce cost reduction measures at MOH and in the hospitals was developed, and the Planning Unit was reorganized and expanded. Unfortunately, lack of personnel at the periphery and resistance to loss of control at headquarters slowed down implementationIf the decentralization plan, the new Planning Unit was always short of personnel (because of rapid turnover, study leaves and expanded work program), and the cost reduction 25 measures were not fully implemented. Overall, these efforts resulted in quite modest improvements. The Zomba School of Nursing was completed and a Manpower Development Unit was created; but the regional training centers were never built because of cost overruns, and the manpower planning analysis was not satisfactorily completed (see below). In addition, a number of key studies including some left over from the first project failed to be completed or were deferred to the Population Health and Nutrition (PHN) sector credit. b) Expanding PHC System. The physical expansion was completed as scheduled, though with some price overruns because of cost increases. However, 9 of the 19 health centers were inoperative at the time the PCR was written because of lack of staff or water. All centers are now operative but some still have staffing or water problems (See Chapter 4.6). c) MCH and CS. With some shifting around of funds from different projects, the planned physical expansion took place. However, a manual for MCH health workers was not produced and integration of CS into MCH was slow due to. shortage of staff to produce training materials and undertake training. Thanks to assistance from UNFPA and the United States Agency for Industrial Development (USAID), availability of contraceptives improved. But because parallel financing by KFW failed to materialize, the surgical contraceptive units in hospitals were not constructed, and the construction of the urban health units were shifted to the PHN Sector Credit because of shortages of funds. d) Support for the Project Implementation Unit (PIU). The PIU was successfully established and an effective project coordinator recruited. This undoubtedly contributed to the improved performance of this project compared to the first. Even here, however-and despite the fact that the project provided salary support for senior positions in this unit-there were staffing problems.' e) Multisectoral Activities. This set of sub-components suffered from weak coordination between the MOH and the other ministries. Nevertheless most of the planned activities were undertaken. Borrower Compliance with the Credit Agreement 3.7 Most covenants were complied with and counterpart funds were adequate and timely. However, the overall coordination committee never met, accounts and audit reports were not always timely and up to standard, agreed to staff appointments were often delayed, and some studies and a mid-term review failed to be undertaken. These shortcomings contributed to some of the problems experienced by this project. 5. As noted in the PCR. the procurement officer was never appointed, the accountant position was vacant for two years, and the architect's post was only filled for a total of 26 months prior to the closing of this project. 26 Bank Supervision 3.8 Overall, supervision was supportive and timely and supervision ratings were somewhat more realistic than in the first project. However, as was the case in that project, the Bank failed to strongly voice concerns about shortfalls from credit agreements and seemed at times to be more concerned with the development of the third project than with the effective implementation of this one. 27 4. Project Results and Achievements 4.1 What of a longer term nature have these projects achieved? This section reviews the little evidence that is available to answer this question, focusing on topics of special interest for future operations. Health Status 4.2 A common perception in the donor community at the present time is that there has been little if any improvement in the health status of the Malawian population over the last decade, despite the substantial external resources invested in the sector. This is frequently interpreted as meaning that the resources invested in the sector have yielded very few benefits, in large part because of weak management of the health system. Both the statement and the interpretation are open to question. 4.3 First, there are no time series of comparable data on which to judge what is happening to the health status of the population over time. There is one survey-the Demographic and Health Survey (DHS) of 1992-that includes estimates of health indicators believed to be of reasonable quality. But there is generally only one or at most two earlier observations on similar variables and they are of dubious quality and comparability. The figures in Table 4.1, reproduced from the July 1994 Midterm Review of the PHN Sector Credit, are typical of what is available. While the 1992 figures are from the DHS, the earlier figures are from sources that were impossible to track down and study. The only thing clear about them is that the estimates of the maternal mortality rate are far too low-certainly the first of the estimates but probably also the DHS estimate as well. We conclude that it is not possible to establish any trend at all from available data. 4.4 Since the density of health delivery points has increased and the Expanded Program of Immunization (EPI) has achieved high levels of coverage over the last decade, some improvements in health status should have occurred. But they may have been offset by the health impacts of a deterioration in economic and agricultural conditions plus a growing number of new diseases (drug resistant malaria, HIV/AIDs and other sexually transmitted diseases) over that period of time. Available data do not help to determine what the net effect has been. 4.5 Second, even if one could say with confidence that there has been no improvement in health status, low productivity of investments in the health sector is only one possible explanation. Another is that these investments are too small to make a significant difference in the overall situation.' Yet another is that they have offset what would have been a disastrous situation in their absence. The final possibility is that these investments and the institutional and policy changes accompanying them (for example, decentralization measures) are laying the groundwork for significant improvements that will only show up later. This review concludes that each of these interpretations has some merit depending on the subsector. 6. As noted above, even with donor inputs, only roughly US$5.50 per capita is spent on health in Malawi, far below the US$12 believed to be minimally necessary. Moreover, the importance of economic conditions and new disease vectors should not be underestimated. 28 Table 4.1: Selected Health Indicators Indicator Project start At mid term Infant Mortality Rate (per 1000) 150 134 Maternal Mortality Rate (per 100000) 170 620a Total Fertility Rate (per 1000) 7.8 6.7 Contraceptive Prevalence Rate (per 1000) 3 7 Crude Birth Rate (per 1000) 54 47 Crude Death rate (per 1000) 20 20 HIV Positive cases not available Approx. 10% a. It the project start figure is accurate, this worsening of the maternal mortality (DHS) could be a result of a combination of factors including the drought and HlV/AIDS. Sources: Staff Appraisal Report and Demographic and Health Survey 1992 Extension of the Health Delivery System into the Countryside. 4.6 Whatever the overall impact of Bank and other donor inputs, there is little doubt of their positive effects on the people served by individual facilities established with these funds. This was very evident from interviews of clients and staff at two health centers visited by the audit mission and two officials from the MOH. Both centers were in very remote areas, some 15-20 km. from the nearest paved road and at least ten km. from the nearest alternative facility. 4.7 One, an MCH center at Kafukula, which was added next to an existing dispensary operated by the local government, was completed in 1984 with funds from Health I. The other, the Choma Health Center, is a full-service center with ten beds (six for maternity cases); it was completed in 1990 with funds from Health II. Its buildings, which compare favorably with many urban health centers, looks a bit strange sitting nearly by themselves in the bush. 4.8 Because of staff shortages, Kafukula did not start operating for three years and Choma for two years after construction was completed. Neither have a full complement of staff even today. Neither has running water: although they were both outfitted with indoor plumbing, a holding tank, a borehole and a handpump, pipes to connect the borehole with the tank and the main building were never installed. Neither facility has electricity or transport equipment (pharmaceutical being kept cool in a kerosine refrigerator). If an ambulance is needed, a call is placed to the district hospital over an hour and a half away in good weather. Kafukula has a telephone operated by a solar panel, but the nearest phone to Choma is several kilometers away. Drugs are in short supply: Choma had been without many used on a daily basis-including Fansidar for malaria-for two weeks (a partial shipment arrived during the mission's visit). 4.9 These problems are quite typical. More than half the centers constructed with funds from the two IDA credits opened late because of shortages of staff and had-in many cases still have-similar water problems. Even the Mzimba District Hospital, a 200-bed facility 29 constructed with funds from Health II, has serious water problems. Until the nearby river dried up in the recent drought, staff carried water from there to supplement the intermittent supply piped from the town. Two years ago, the Hospital requested a bore hole; it was drilled in August 1994 but as of November the well was not yet completed and operational. 4.10 Despite these problems, clients and villagers interviewed at both sites were generally pleased with the quality of care they received, their main complaint (in the few cases that any was expressed) being periodic shortages of supplies and personnel. These centers appear to have had two major effects on the people interviewed. First, they significantly reduced travel time for those living near the periphery of the catchment area; this has led them to make somewhat greater use of medical facilities. And second, they appear to have raised expectations and effective demand for health care. Among other things, more women now expect to give birth in a clinic with the assistance of a trained nurse and to bring their children in for immunizations and to be weighted for several months after birth. These are small changes, but very significant in the lives of the people experiencing them. Family Planning 4.11 Considering how long it has taken in some African countries to change policies and get a significant family planning program going, Malawi has made substantial progress since 1982 when then-President Banda agreed to permit the development of a child spacing program. Since then, and particularly during the last three years, a number of positive events can be pointed to. First, there have been significant changes at the policy level: from permitting these activities to publicly speaking out in favor of them-and in the process all but dropping the term child spacing in favor of family planning-and from allowing only trained doctors to prescribe pills to permitting traditional birth attendants and workers in the community-based distribution (CBD) program to provide pills even before attending a training program provided they abide by a fairly liberal CBD checklist in doing so. Second, contraceptives are no longer in short supply (thanks to donors, mainly USAID and UNFPA) and services of trained personnel are available in all hospitals and more than half of government health centers at least one day per week. This was confirmed by the field visits undertaken by this mission. Third, a number of agencies have been established to promote family planning and expand service provision independently of the MOH; they provide a legitimate base of operations which did not exist before for individuals interested in playing an active role. Fourth, the 1992 DHS indicates widespread awareness of family planning and, coupled with earlier estimates, a respectable increase in contraceptive prevalence rates-from close to zero at the beginning of Health I to about three percent at the end of that project and seven percent in 1992. These are very promising first steps. 4.12 But family planning experts interviewed by this mission painted a much less optimistic picture. According to them, most changes at the policy level are permissive or mildly promotional. There has been little followup in the form of orders, significant increases in staff or changes in incentives to encourage promotion of family planning by village health workers who are inherently conservative and unlikely to take any initiative on their own. In addition, there are reasons to be concerned about the sustainability of the contraceptive prevalence rates (CPR). So far, these rates reflect new users. There is anecdotal evidence suggesting that the number of persons actually using the contraceptives is substantially less than that implied by 30 data on distribution of contraceptives. These issues need careful empirical investigation. If they prove to be important, something more than further expansion of the present distribution system will be required to continue the rate of progress experienced so far. Efforts to Strengthen Health and Health Manpower Planning 4.13 Since the beginning of Health 1, efforts to strengthen the Ministry's planning capacity have involved the establishment of a Planning Unit initially with two posts but later with six, the establishment of a Manpower Development Unit with three posts, and the provision of fellowships and training, equipment and technical assistance. Both the Bank and USAID have supported these efforts, the Bank primarily concerned with the Planning Unit and USAID with the Manpower Development Unit. To date these efforts have not been successful in establishing the kind of planning and analysis capacity originally envisioned. The only significant document produced has been the National Health Plan for 1986-1995; while in the end it was produced by MOH staff rather than expatriate advisors, because of staff turnover, the MOH is hardly better prepared to produce a new health plan today than it was ten years ago. The main problems have been difficulties in recruiting, training and keeping staff, inadequate technical assistance, and flawed organizational arrangements. 4.14 Staffing. The Planning Unit was initially staffed by two professional health planners already employed by the MOH. Of the four persons recruited for the new positions, only one has actually done any work in the Unit.7 In 1988, the two senior planners, were relieved of their posts for political reasons. To fill in, three staff members of the Ministry of Economic Planning were seconded to the MOH. One of the three left for graduate work overseas; the other two left for training and have only recently returned. After the change in government last May, one of the two original planners returned to his post as head of the Unit; he is trying to recruit additional people. The Manpower Development Unit, established in 1990, has only recently filled the three posts allotted to it. 4.15 Technical assistance has not helped much. As noted above the Ministry rejected the draft of the National Health Plan provided by the original team of four consultants assigned to the Planning Unit. In September 1985, a supervision mission reported that the Ministry rejected a consultant's recommendations for revision of its financial and accounting systems and requested a new consultant. The same experience was repeated with a consultant for the pharmaceutical component. In each case Bank staff agreed with the judgements and actions of the Ministry. A final example pertains to the report of a consultant (funded by USAID) to the Manpower Development Unit recruited for a two-year period to produce a manpower development plan. The report proved to be more of a situation analysis than a plan and is currently being redone by the head of the Unit. The MOH did not appoint anyone to work with the consultant and did nothing to draw the work into its operations. 4.16 While each of incidents had its unique problems, there is one common thread. In all cases, the individuals worked in isolation without frequent in-depth meetings with MOH officials. Also, in none of these cases did the consultants leave much behind other than their 7. All were sent for overseas training. Of the first three to return, two took up other positions. The last one sent has not yet completed his training. 31 reports: training was not considered a significant part of their tasks and counterparts were rarely appointed. 4.17 Organizational issues. As originally contemplated, the Planning Unit, in addition to its planning and analysis functions, was supposed to have responsibility for the Ministry's statistical work, manpower planning, monitoring of health sector resource allocations, and management of donor-funded projects. However, most of these functions were hived off to other units-important statistical functions to the CHSU, manpower planning to the Manpower Development Unit and management of the IDA projects to a Project Implementation Unit-each reporting to a different line manager. 4.18 Several things could be done to improve the situation. (1) Statistical functions and manpower planning could be brought back into the Planning Unit. This would consolidate staff who are spread too thinly just now into something approaching a critical mass and would facilitate coordination amongst these interrelated functions. (2) The Planning Unit could be raised a notch or two in the hierarchy and given some discretionary authority over its budget and work program.! This would improve its capacity to do its job, improve morale and make it easier to recruit and retain good personnel. (3) A charge (in local currency) could be made against the budget of each agency receiving technical assistance. This could induce agencies to think more carefully about whether they need such help, induce them to select consultants with more care and encourage more thought about how to make best use of such assistance.' Community Health and Services Unit (CHSU) 4.19 This institution has not functioned as planned since its inception. The health information system (HIS) is producing very little of value at the current time and is at least two years behind in entering data coming to it from the districts. Few of the studies and program planning activities envisioned at its inception have been undertaken. Indeed, a baseline study included in Health I was deferred to Health II and then to the third project, the PHN Sector Credit. Laboratory facilities are impressive compared to that of many other African countries, but little is going on there except what is sometimes requested by the various disease control programs. The CHSU was supposed to provide training in sample survey methods to regional and district level personnel but has not done so for lack of staff and budget. Most of the activity is going on in disease control units supported by foreign assistance, but they tend to be erratic given shifts in donor priorities; and there is little or no coordination between these programs. 4.20 Some of the organization's problems stem from the original design of the CHSU. Its organizational chart-even its name-is puzzling. Typically, an epidemiological unit includes a reference laboratory, a program office (for program planning, monitoring and evaluation), and a disease surveillance unit. In addition to these elements, this organization includes responsibility for the whole HIS plus disease control units with line responsibility for program 8. At the present time, it has no budget for field work. Yet without a solid factual base on which to build and argue the case, planning exercises will be little more than that-exercises, not to be taken too seriously. 9. For a more complete discussion of this point, see Chapter 4 in Ridker, 1994, especially the section starting on page 85. 32 implementation. While it is awkward and untidy to combine line and analytical responsibilities, that arrangement might be made to work. But the HIS, which consists of far more than disease monitoring, should be associated with-ideally, under the direction of-the unit responsible for health planning. Nothing in Bank files explains why the organization was established in this way and it was not possible to find out during the course of the mission. 4.21 In addition, the health surveillance system is far too centralized. One of the main reasons for being more than two years behind is that all data are entered by hand at headquarters, instead of being entered and subjected to analysis for local purposes at the district hospital level." 4.22 But the CHSU's most immediate problems are shortages of staff and operating budgets, and excessive centralization and control of its functions by the MOH. Few posts have actually been established and many are filled with persons occupying established posts elsewhere in Government. Budget decisions are made in MOH with little prior discussion with CHSU staff. The laboratory has no budget of its own; it has been operating solely on transfers from various disease control programs when they need help. Vehicles (provided by Health I) are in very short supply, having been taken by the MOH when they arrived before the CHSU became operational, and then never returned. 4.23 The Japanese Government has recently agreed to provide substantial assistance in the form of equipment, technical assistance and budgetary support. This will ease the organization's most pressing problems for a time, at least. For the longer term, there is a need for an in-depth assessment of this organization and its role in the health system. Pharmaceuticals 4.24 The goal of this component of Health I was to improve the supply and distribution of pharmaceuticals and medical supplies in the country. This was to be accomplished primarily by improving the operating efficiency of CMS which is responsible for the procurement and distribution of 80 percent of all medical supplies (close to 100 percent of what is available outside the major urban centers). The implicit assumption behind this approach was that the overall budget for medical supplies would be adequate if it were used efficiently. Indeed, early papers in Bank files indicate that substantial budgetary savings could be achieved so that on net Health I would not result in any increase in recurrent health expenditures (see para. 2.4). These assumptions proved to be excessively optimistic, an indication that this component, like the epidemiological component (CHSU), had not been adequately thought out prior to implementation. 4.25 Efficiency improvements were to be achieved by three broad sets of measures: decentralization of facilities so as to reduce transportation costs and inventory requirements, development of facilities for local manufacture and processing of simpler compounds, and improvements in a variety of software elements (inventory control, accounting and pricing procedures, etc.). 10. The CHSU has recently taken a step in the right direction by decentralizing data entry to the regional level and has started providing some feedback from the regional to the district level. 33 4.26 While substantial delays were involved, the decentralization planned in the project (adding two regional depots) has been achieved. There is no evidence that this has resulted in significant reductions in transport costs or inventory requirements, but there are signs that it has substantially reduced the time required to deliver medical supplies to more remote areas-in one example cited, from three days to eight hours. Even so, problems remain. The housing constructed under the project has still not been turned over to the CMS staff; and flaws in the original design of some facilities-known about at least as far back as 1988-have required reconstruction which has not yet been completed. 4.27 Substantial cost savings should be involved in undertaking certain processes domestically-producing intravenous fluids and compounding and packaging of medicines from bulk supplies, for example. While some equipment was procured under Health I, the decision was made to contract out such activities to the private sector-a wise decision, but one that has apparently caused substantial delays and problems in its own right. Only in the last few months has a private firm secured the financing from local banks to establish facilities for production of intravenous fluids. A contract to a foreign firm which agreed to establish some local facilities is being questioned because the terms may end up costing the Government more, not less, than international procurement. 4.28 Most of the efforts to improve software elements have not yet paid off. Many started later than anticipated and are still in process of being implemented; others have not developed as planned. One of the more important elements was the plan to establish the CMS as a distinct self-accounting unit with appropriate business practices and organizational status. Health I encouraged some partial moves in this direction in 1984 when the drug budget was shifted from the CMS to the hospitals (which are responsible for distribution to health centers), and which then used the budget to purchase from the CMS. The CMS was provided with a fund for its purchases, which was supposed to be replenished by sales to the hospitals. However, the prices CMS could charge hospitals was controlled by the Government, the CMS remained under Government civil service regulations, and it maintained its near monopoly position in the country. 4.29 Since then, tight Government budgets and unwillingness to adjust prices for inflation have put increasing pressure on the CMS. While this may have led to some efficiency improvements, that is not very evident; the primary result has been the accumulation of overdrafts, periodic shortages of critical supplies and bail-out operations by one or another donor. The devaluations of 1994 have created the most severe shortages experienced since independence. In April 1994, the kwacha budget for drugs was sufficient to purchase $10 million worth of supplies. By June, that budget could purchase less than half that amount. 4.30 The PH-N Sector Credit attempted to deal with this situation by making it a condition of the credit that the CMS would be developed into an independent profit center. This will not have the desired effect so long as prices are controlled and the CMS is not faced with any serious competition for its services. 34 IEC and Multisectoral Activities 4.31 The need for programs focused on information, education and promotion of good health, child care and family planning practices is particularly great in a country like Malawi given its low education levels. Health I appears to have supported some IEC activities but no explicit component was identified. Under Health II, in addition to continuing these activities in the MOH, [19 percent] of the budget was devoted to initiating programs in three other ministries. 4.32 The Ministry of Women, Children and Social Welfare (successor to the Ministry of Community Service) received funds to develop and introduce health and family planning messages into functional literacy, home economics, and other women's programs. It started in 1989 with an advisor who helped develop the messages and establish a training program. Currently there are 58 master trainers who are supposed to provide one month courses to local promoters and trainers. These efforts are not progressing well in large part because of shortage of funds to support the core programs." Their impact is also limited by the small size of this program and by the fact that they do not involve men. 4.33 The Ministry of Youth, Sports and Culture was provided with funds for similar purposes targeted at youth. While nearly all the planned activities were initiated, they ceased when the Malawi Pioneers, which was responsible for implementation, was closed down in early 1994. This is unfortunate since, with the spread of AIDS, the need is greater today than in the past. 4.34 Support was provided to both the MOH and the Ministry of Information (MOI) for IEC activities. The MOH program is focused on activities attached to the health centers; it has been inadequately staffed and not much has been produced. The MOI program attempts to reach a broader audience using mass communications media supplemented by extension workers who attempt to enlist the help of village leaders. It got off to a slow start because of the need for training and only launched its first mass media campaign in June 1993. No evaluation has been undertaken. Medical School 4.35 During preparation of the first credit, the MOH raised questions about funding for a medical school. This was a non starter at the time because of the Government's financial difficulties, but it was taken up more seriously during preparation of the second health project. The Bank initially took the position that it could not consider such a proposal until it was developed, costed out in more detail and compared with alternatives. After a number of studies and discussions, the Bank finally declined to support the project. At that point, the I1. The EEC, which finances this subcomponent, provides funds only for the development of the child spacing materials, not for the literacy or the home economics programs themselves. There is a dearth of reading materials in the literacy program (UNICEF used to help with the production of literacy materials but has now turned to other things). Because of unhappiness with their salaries (which are very low and sometimes not paid or paid late), literacy teachers and homecraft workers often do not show up for classes. The only way around this problem is for donors to support the core activities as well, or for the recipients who want these activities to continue to find some way to generate income to support them. The PHN Sector Credit is providing some funds to help develop such income generating activities. 35 Government decided to fund the project itself.12 In October 1986, the first class of 20 students was sent to England for their first three years of study. In 1989, this class (minus only two people) returned to complete their last year in Malawi. Starting in September 1994, all four years are being taught in Malawi. According to Malawian informants, the retention rate has been over 90 percent, quality of graduates is good and costs are relatively low. Was it a mistake on the Bank's part to have declined to assist in the funding of this enterprise? 4.36 While there were a variety of reasons for wanting such a school, the soundest arguments pertained to retention rates, the character of the training contemplated and costs. Malawi has one of the lowest rates of doctors per capita, one of the reasons being that only one student in four was returning from overseas training.3 Moreover, the training received was not appropriate for Malawian conditions. Typically there was too much emphasis on diseases of the elderly and use of sophisticated equipment and treatment procedures, and insufficient focus on infectious, nutritional and tropical diseases and on treatment regimes appropriate for Malawi's regional hospitals. Finally, unit costs were expected to be much lower. Given the low retention rate, training costs would have to be four times higher in Malawi before it would be cost effective to send students abroad. But the per student cost of such training would be substantially lower in Malawi because plans called for adding marginally to an existing hospital rather than starting from scratch, less sophisticated and expensive equipment and techniques were to be utilized, and teacher salaries, at least for Malawians, would be less.'4 4.37 While the PCR and Bank files are unclear about the reasons for the decision not to assist this enterprise, discussions with staff suggest two possible reasons. First, there was a presumption that the Government wanted to establish a training facility patterned after European and North American institutions. That this was not the case can be seen from the Tripartite Study Report referenced above which served as the basis for the project approved by the President. It appears that the Bank did not investigate the situation thoroughly enough to determine whether its presumption was correct; but also, the Government appears not to have done what would have been necessary to convince the Bank differently (for example, detailed, convincing cost estimates that would have made Government intentions much clearer do not appear to have been provided to the Bank). Second, the Bank believed that the Government could not meet the recurrent costs of this project. But it did not explore the possibility that donors who were then giving scholarships to Malawian students might be willing to assist with the recurrent costs in leu of these scholarships-which is what, in effect, happened; nor did it consider funding the recurrent costs itself, at least on a declining basis until other sources could be found. 12. The Government has funded all capital costs and much of the recurrent costs. Some funds for scholarships for students to undertake preclinical studies abroad and for teacher salary supplementation was received. 13. Most students went to Europe and North America. The return rate for students going to other African countries for training was higher, but few seats have been available because of preference given to nationals. 14. See the 1986 Tripartite Study Report, A Plan for Medical Education in Malawi, with membership from Germany, Britain and Malawi, which strongly recommended this community oriented approach. 36 Bank Performance 4.38 The Bank's principal contributions to Malawi's health sector has been its assistance in extending the health system into the underserved periphery of the country, encouraging the rapid development of a family planning program, and applying pressure to redirect resources in certain ways (to decentralize the system, to increase the total budget allocated to health, increase the share going to rural areas and to primary care, and work towards greater cost recovery and improved efficiency of operations). There is general agreement amongst donors that these are the proper directions in which to move and that the Bank has effectively used its influence for this purpose, making their job that much easier. The Bank is also given high marks by donors and the Government for its insistence on proper reporting, auditing, procurement and other procedural aspects of the aid relationship. 4.39 This praise is qualified, however, by criticisms about the Bank's style of operations: lack of a sector specialist in the field, periodic large-scale missions that tie up Government officials for weeks during which time other work-including the work involving other donors-suffers, inadequate dialogue and communications with other donors, and periodic insistence on achieving goals without adequate understanding of the difficulties involved or provision of help in overcoming these difficulties. These are strongly-held opinions. Indeed, the ODA has felt so strongly about the lack of a health specialist in the field that it has provided funds to the Bank for such a position." 4.40 The Bank's role in project preparation was satisfactory in the sense that a decent sector study was obtained prior to initiating the first project and the Bank continued to build up its understanding of the sector so that its knowledge base for preparing the second project was substantial. But it permitted several design errors to be incorporated into these projects which have caused difficulties ever since. As noted above, the first project provided insufficient time and TA resources to implement all its components and did not fully and appropriately design the CHSU and Pharmaceutical components, while the second project did not continue assistance to the CHSU and Pharmaceutical components despite their need for continued support. It was probably also an error that the Bank did not assist in the funding of the medical school or of certain recurrent costs in the health sector, although these are arguable propositions. 4.41 The Bank's supervision activities have already been discussed. 15. In response the Bank staff associated with the country noted that this view ignores the Bank's efforts to enlist donors' assistance in financing this and similar initiatives. 37 5. Conclusions and lessons Ratings 5.1 While most of the activities planned under two projects eventually were implemented, they have failed to achieve much lasting improvement in the capacity of the health system. The principal exceptions to this are the extension of the health system into underserved areas and the introduction of a family planning program. Other achievements are less important or ephemeral. For example, while a national health plan was developed, the capacity to undertake health planning was not significantly improved, while the pharmaceutical component was ultimately implemented more or less as planned, there is no evidence that costs have been reduced, and while the planned multisectoral activities took place, there is no evidence that they have affected attitudes or behavior. Accordingly, in contrast to the PCRs which imply satisfactory outcome ratings for both projects, performance of the first project is rated as unsatisfactory, while the performance of the second project, which contained a much larger construction component and was not subject to delays in implementation, is rated as satisfactory. However, the institutional development efforts had negligible impacts in both projects, a judgement that is consistent with the PCR findings. 5.2 Sustainability of the gains made is more difficult to judge. On the one side, the Government's recurrent budget remains under extreme pressure so that if it were not for donor inputs, the primary health system might collapse. On the other, these inputs, plus the decentralization measures so far undertaken and the recent elections have generated expectations on the part of the rural population that are putting pressure on the Government to allocate more funds to the health sector and to the rural areas. In these circumstances, sustainability of both project outputs is rated as uncertain, a judgement that agrees with the PCR for Health II but not for Health I. Lessons 5.3 These projects are excellent examples of the concerns addressed in the Wapenhans report, among others, the need for Government ownership, realism in assessing implementation capacity and progress, and more attention to supervision. The Audit agrees with lessons of this sort pointed out in the two PCRs. Other issues of relevance to the health sector in Malawi include treatment of personnel problems, recurrent cost problems, technical assistance, and general project orientation. 5.4 Many of Malawi's problems stem from shortages and high turnover of personnel, especially at senior levels, though these problems often show up in other guises. For example, plans for system improvements do not get implemented. Such implementation requires sustained, detailed attention by senior staff with the authority to make relevant decisions. But the few people in such positions are typically overwhelmed trying to maintain daily operations, let alone think about system improvements. High turnover rates exacerbate the problem by wiping out what little progress has been made. A third factor that limited sector management capacity to make and implement decisions was the danger of doing so given the totalitarian nature of the government in power during the first and a large part of the second project implementation periods. 38 5.5 The Bank's remedy in these two projects has been to provide training and fellowships and to insist in covenants that certain critical posts are filled. These measures have frequently proved to be insufficient because the MOH did not have adequate control of the situation. Frequently, even today, it does not have the budget or the authorization to fill a post or to pay sufficiently high wages to attract good people and keep them once they have received training. Many of these problems lie outside the sector. Trying to solve them, as the Bank has done by focusing on the MOH alone-for example, putting pressure on the MOH to fill posts-will not work. 5.6 Lying behind these personnel problems is lack of budgetary resources. This is the fundamental problem with the health sector in Malawi-in the 1970s and early 1980s because of the low priority given to the sector and later because of the worsening economic and revenue situation. It is true that the sector has made inefficient use of the resources it has; but it is unrealistic to believe that sufficient economies could be achieved to significantly reduce the need for more inputs. Indeed, in many cases additional inputs-for example, to fill senior staff vacancies-are needed before any system improvements can be implemented. 5.7 In the case of such extreme budgetary shortages, it is not enough to provide funds for construction, equipment, technical assistance and training. Resources to help meet recurrent expenditures are also required. Put differently, capital and recurrent resources are to some extent fungible. If one element is in short supply, ways will be found to use what is available to meet the most pressing needs: funds will be diverted from maintenance and field inspections, foreign advisors will be used for current operations, funds for training will be sought and used mainly as salary supplements. What is needed is a package of inputs, capital plus recurrent, that is adequate in total. This has not been the case for any sustained period. 5.8 A surprisingly large number of foreign advisors funded by these two projects turned in disappointing performances. A common feature in these cases is the fact that the consultants were left more or less on their own, often without active counterparts, to complete a report. There was little interaction with the persons for whom the report was intended and little effort made to transfer skills and knowledge. Part of the reason for this situation is, once again, shortage of staff. But even so, it implies that lower priority was given to the consultant's activities than was required for the task to succeed. One way to change this situation is to establish a rule that agencies receiving technical assistance must pay something for it from their budget (for example, a senior civil servant's salary). This would ensure that TA is not accepted without care because it is a free good. It would also help to provide TA only for advisory functions, that is, not to write a national health plan but to advise the person or team with responsibility for writing the plan. This might be reinforced by providing the advisor on a short-term mission basis, rather than locating the person in the field for an extended period.'" 5.9 Finally, these projects were excessively process- rather than results-oriented. That is, their objectives were more to provide certain services or improve certain capacities rather than 16. These suggestions and their rationale are discussed more fully in Ronald Ridker. 1994. "The World Bank's Role in Human Resource Development in Sub-Saharan Africa: Education, Training, and Technical Assistance." A World Bank Operations Evaluation Study. Washington, D.C. 39 to achieve specified changes in health status. This is in keeping with the general orientation of the medical and health community of the 1970s and 1980s, which focused on meeting perceived or effective demand rather than more basic needs. This allowed the development of the situation observed in Malawi, where donors have invested substantial amounts of resources in the health sector with little in terms of results to show for it. It may also help explain why efforts to monitor changes in health status and undertake related epidemiological and other studies were given such low priority. The projects reviewed in this report were strongly influenced by this orientation-an orientation which, fortunately, has been changing in recent years. 5.10 The PHN Sector Credit, particularly as it has been reoriented after the mid-term review, is attempting to take many of these points into account. It is encouraging the Government to increase health sector funding and introduce cost saving measures, it is providing more funds for recurrent costs, it is somewhat more results- than process-oriented, it has attempted to cut back on use of non-African TA and encourage use of local and African TA, and it has not supported any long term overseas training awards (because of a belief that the bottleneck is less a shortage of skills than a set of management practices that result in people not being able to use their skills effectively). Unfortunately these changes have made little difference so far because the Government has not appointed qualified Malawians to available TA posts, despite their being interviewed, and has not changed its personnel and posting policies.  41 ANNEX I TdkPhow: Ldongm v7 a PRIVATE BAG 330 Tera: COUSEV P~ m Telx: 44361 COMSEV MI A IA CI F=: 732 796 LIDNGWE 3 MALAWI SECRETARY FOR WOMEN AND CHILDREN AFFAIRS AND COMMUNITY SERVICES Ref. No. CD4/86A 2nd June, 1995 Mr Roger-Slade, Chief Agriculture Human Development Division Operations Evaluation Dept. The World Bank 1881 StreetNW Washington DC 20433 Dear Sir, MALAWI HEALTH PROJECT (CREDIT 1351-MAI) SECOND FAMILY HEALTH PROJECT (CREDIT 1768 MAI DRAFT PERFORMANCE AUDIT REPORT Belatedly I wish to acknowledge receipt of the above document on which you asked for my comments. I have the following comments to make on the documents:- 1. Very little has been said about the Child Spacing Messages Project component may be this is because it is a small component in the entire project. We appreciate positive remarks made about the performance of the family planning activities generally on pages 45 and 46 especially noting that awareness raising, which is one of the areas in which high achievements have been recorded. 2. It is true that the main reasons for the limited success achieved by CSM project include shortage of teaching and learning materials for the literacy classes. However, it should be noted that the govenrment is taking corrective measures including the possibility of purchasing of a bigger printing press under one of the projects and the employment of permanent literacy instructors as opposed to voluntary ones. 42 3. The issue of the size of the project as contributing to the limited success may not hold. One needs to look at the objectives and coverage of the project. It should be noted that this was a project and not a programme and as such its coverage was defined in the project document. The process of implementing the projects was such that it determined the reaching out, as training of staff at all levels and the process of integration took most of the time. 4. With respect to men involvement, the HOme Economics Programme's target audience is women. However, the literacy programme involves men as well and over the years due .to the special efforts that the Ministry has made to increased male involvement, their rate has increase from 11% in 1991 to 14% in 1993. This we feel is positive achievement which should be acknowledged, considering the reasons that make the men shy away from participating in literacy programme. In conclusion, I wish to bring to your attention that the project was expected to contribute towards behaviour change of the target group by providing appropriate Child Spacing Messages. Change of behaviour is a slow process especially when the backup services, such as contraceptives are not readily available. The Ministry's move to participate in Community Based Distribution of contraceptives, will hopefully enhance the impact of our I & C efforts. Yours faithfully, . m a ra  DNIfVW I