Impact Evaluation Study of the Bank Assistance in the Health and Population Sector in Bangladesh, Pakistan, Papua New Guinea, and Sri Lanka

8/3/2019 Impact Evaluation Study of the Bank Assistance in the Health and Population Sector in Bangladesh, Pakistan, Papu…

http://slidepdf.com/reader/full/impact-evaluation-study-of-the-bank-assistance-in-the-health-and-population 1/44

t_ B O A R D

O F

D I R E C T O R S

ASIAN DEVELOPMENT BANK

IN.2S0-9721 November 1997

IMPACT EVALUATION STUDY

The following Impact Evaluation Study prepared by the Post-Evaluation Office is

attached for information:

Bank Assistance in the Health and Population Sector in Bangladesh,Pakistan, Papua New Guinea, and Sri Lanka



ASIAN DEVELOPMENT BANK IES: REG 97025

IMPACT EVALUATION STUDY

OF BANK ASSISTANCE

IN THE HEALTH AND POPULATION SECTOR

IN

BANGLADESH, PAKISTAN,

PAPUA NEW GUINEA, AND SRI LANKA

November 1997



CURRENCY EQUIVALENTS

Bangladesh(December 1995)Taka (Tk)

Tk1.00 = $0.0245$1.00 = Tk40.75

Pakistan(May 1996)Pakistan Rupee (PRs)

Papua New Guinea(October 1996)Kina (K)

K1.00 = $0.76$1.00 = K1.31

Sri Lanka(October 1995)Sri Lankan Rupee (SLRs)

PRs1.00 = $0.029$1.00 = PRs34.74

SLRs1.00 = $0.0195$1.00 = SLRs51.25

ABBREVIATIONS

BMECBRCDRDHCDMCFP,GDPGHCHFA 2000H&PIMRMCHO&MPHCPHM

PNGRHS1RHS2SDHCTA

Benefit Monitoring and EvaluationCrude Birth RateCrude Death RateDivisional Health CenterDeveloping Member CountryFamily PlanningGross Domestic ProductGramodaya Health CenterHealth for All by the Year 2000Health and PopulationInfant Mortality RateMaternal and Child HealthOperation and MaintenancePrimary Health CarePublic Health Midwife

Papua New GuineaRural Health Services Project 1Rural Health Services Project 2Subdivisional Health CenterTechnical Assistance

NOTES

(i) The fiscal year (FY) of the Bangladesh and Pakistan governments ends on 30 June.FY before a calendar year denotes the year in which the fiscal year ends, e.g.,FY1997 ends on 30 June 1997. The fiscal year of the Papua New Guinea and SriLankan governments is the same as the calendar year.

(ii) In this Report, "$" refers to US dollars.

IE - 44



CONTENTS

Page

EXECUTIVE SUMMARY

I.

II.

III.

IV.

V.

VI.

APPENDIXES

ii

INTRODUCTION 1

BACKGROUND 1

A.B.C.

Study Objectives, Approach, and MethodologyInitial Conditions against Which Impact was AppraisedBank Operations in the Health and Population Sector

125

OBJECTIVES, SCOPE, AND IMPLEMENTATIONAL EXPERIENCEOF THE PROJECTS 6

IMPACT OF BANK OPERATIONS 7

A.B.C.D.E.F.

Impact on Health StatusImpact on Policy AdjustmentImpact on Government ExpenditureImpact on Institutional DevelopmentSocial ImpactEnvironmental Impact

71213131415

KEY ISSUES FOR THE FUTURE 16

A.B.C.D.E.

Sustainability of Impact of the Bank's AssistanceCost RecoveryPrivatization of Health ServicesProject Design and ImplementationBenefit Monitoring and Evaluation

1616171718

CONCLUSIONS AND RECOMMENDATIONS 18

21



EXECUTIVE SUMMARY

The main objective of the Study was to assess the impact of the Bank'sassistance in the health and population (H&P) sector. The Study covered completed projects,for which postevaluation has been done, in four developing member countries (DMCs);

namely, Bangladesh, Pakistan, Papua New Guinea (PNG), and Sri Lanka.

Macroeconomic conditions common to the four DMCs at the time of projectdesign and implementation include chronic budget and current account deficits as well aspolitical instability and threats to national security. Such conditions led to inadequate financingof social expenditures. Although social conditions in the DMCs varied, poverty incidenceremained high. Based on such indicators as incidence of poverty, literacy rate, and lifeexpectancy, Sri Lanka was the most socially developed of the four DMCs. Pakistan (3 percent)and PNG (2 percent) had the highest population growth rates. As for morbidity rate, only SriLanka's was similar to that of more advanced DMCs. The four DMCs adapted the Health for Allby Year 2000 campaign of the World Health Organization to their policies for providing primaryhealth care.

By September 1996, the Bank had approved 31 H&P loans and 70 technicalassistance projects in 14 DMCs totaling about $956 million. Although all Bank projects aimedat improving health in the DMCs, approaches differed in the four countries during the earlyyears of project implementation. The Bank's experience was characterized by implementationdelays with a few cases of partial loan cancellation.

The Bank's assistance in the H&P sector was assessed in terms of its impact onthe health status of the beneficiaries, policy adjustment, government budget, institutionaldevelopment, social dimensions, and environment of the DMCs.

The impact of Bank's assistance on the health status, as reflected bydemographic and health indicators, morbidity indicators, beneficiary perceptions of the facilitiesprovided under the projects, and service provision indicators, varied in intensity among theDMCs. On the whole, however, the impact was positive. Demographic and health indicatorsshowing the positive impact of the Bank's assistance were most pronounced in Sri Lanka andleast in PNG. Morbidity indicators were unclear in a number of project areas. In Pakistan andPNG, for instance, some data showed increasing morbidity rates. Nevertheless, perceptions ofbeneficiaries confirmed the general downtrend of disease patterns and improvement in health.Beneficiaries were generally satisfied with the services provided by projects under the Bank'sassistance.

Service provision indicators comprised the ease of access to health facilitiesand the extent of home visits by midwives, the degree of prenatal care and immunizationcoverage, the frequency of attendance at childbirth by health workers, and the pervasivenessof family planning services. Access to health care improved with the availability and proximityof health centers. Home visits were facilitated with the provision of residential quarters athealth centers, although these were constrained by deteriorating law and order situations insome DMCs. The health centers also improved the availability of family planning services,prenatal care, and immunization services.



iii

Except for Sri Lanka, where the success of the village health centers led theGovernment to focus more attention on providing health care at the most peripheral level, theBank's assistance did not appear to have any significant or long-lasting impact on policyadjustment. Cost recovery, a policy encouraged under the projects, had limited success andwas confined to relatively minor expenses incurred by the beneficiaries. The projects affected

government expenditure not only in regard to the provision of counterpart funding duringproject construction, but also of subsequent operation and maintenance (O&M) requirements,thus aggravating the tight fiscal positions prevailing in DMCs.

The provision of health infrastructural facilities and training improved thecapacity of health providers. Such improvement was not evenly spread, however, and in somecases, was nullified by the lack of maintenance in health facilities, inadequate supply of drugs,lack of staff for training fellowships, and other factors.

The social impact of the projects was positive, with women as the majorbeneficiaries of the health and training facilities under the projects. Impact on poverty reductionwas indirect as general improvement in health status led to better employment opportunitiesand lower work absenteeism from sicknesses. Impact on the environment was positive,particularly when projects provided for water sanitation and services by health inspectors. Thepossible exception was the case of the dichlorodiphenyltrichloroethane (DDT) factory upgradein Bangladesh under another Bank project, though the factory subsequently closed down.

Sustainability of the positive impact of the Bank's assistance remained a keyissue, especially in the face of inadequate funds for O&M. A related issue was cost recovery.Attempts by the governments to collect user fees ran counter to the perception of many peoplethat health care should be provided gratis. Although the privatization of health services wasfully supported by the DMCs, the fact that government doctors were allowed to have privatepractices led to instances of abuse and conflict of interest. An important lesson arising from the

study was that to sustain and reinforce the positive impact of past projects, new projects,rather than focusing on new areas, should be designed to build on the successes of the oldprojects. A last issue was that benefit monitoring and evaluation systems, though deemedbeneficial, were not vigorously pursued by the DMCs, especially when initial difficulties wereencountered.

In conclusion, while health improvement in the DMCs might not be fullyattributed to Bank projects, its assistance in the H&P sector has positively contributed tohealth, lower population growth, and capacity building in the DMCs. Impact on governmentexpenditure, policy adjustments, and the environment was less positive.

Recommendations for the Bank include providing adequate project preparation

with in-depth sector analysis and, in order to facilitate future impact evaluation, preparinglogical frameworks; follow-on assistance to reinforce positive impacts; and the assurance of fullcommitment by the executing agencies. Recommendations for the DMC governments includethe provision of specific budgetary allocations for O&M and qualified staff for training.Recommendations for both the Bank and the DMCs are to obtain the full participation of staffat all levels, as well as that of the community, in the design and implementation of projects,and ensuring adequate monitoring of project implementation based on appropriateperformance indicators.



I. INTRODUCTION

1. Despite its status as the world's most economically dynamic region, Asia ishome to a disproportionately large percentage of people living under poor health conditions. Inparticular, the low-income developing member countries (DMCs) of the Bank have a high

prevalence of contagious, parasitic, and immuno-preventable diseases, as well as high rates ofpopulation growth. Although attempts by DMC governments to address these issues havevaried from country to country, the need to devote more public resources to basic health andpopulation (H&P) services is generally recognized as a prerequisite to economic development.

2. Since 1978, the Bank has provided $932.55 million for 31 projects in the H&Psector. In addition, grants worth $24.02 million were provided for 70 technical assistance (TA)projects. Bank assistance in the early years was focused primarily on improving healthconditions in DMCs, but increasingly the Bank has recognized the need to become involved inpopulation activities and to integrate these with its health activities.'

3. No comprehensive assessment has been made to determine the impact of Bankoperations in health and family planning. The importance of an impact evaluation of Bankhealth projects is underscored by the Medium Term Strategic Framework (1995-1998), whichidentified the social sector as a priority area for Bank assistance. Accordingly, the Post-Evaluation Office initiated a Study of the Impact of Bank Assistance in the Health andPopulation Sector (the Study) under a regional technical assistance.! The impact evaluationrepresented an important step in the project cycle and was needed to improve futuredevelopment policies, strategies, and implementation arrangements in the Bank and theDMCs.

II. BACKGROUND

A. Study Objectives, Approach, and Methodology

4. The main purpose of the Study is to assess the impact of Bank's assistance inthe H&P sector and derive lessons concerning the improvement of the quality of its H&Pprojects. The Study would (i) provide an overview of the Bank's experience in the sector andhighlight key issues requiring attention; (ii) identify strengths and weaknesses in project designand implementation arrangements; (iii) extrapolate lessons from completed projects; and (iv)recommend measures for attaining project objectives and improving Bank operations.

5. The Study was based on seven completed health projects, six of which havebeen postevaluated." There are two projects each from Bangladesh, Pakistan, and Papua NewGuinea (PNG); and one from Sri Lanka. All seven projects were approved during the first half

of the 1980s, prior to the preparation of a regional study" which has since guided subsequentBank assistance in the H&P sector. Details of the selected projects are given in Appendix 1.

2Population Policy-Framework for Assistance in the Population Sector, Asian Development Bank, 1994.TA No. 5629:Regional Study of the Impact of Bank Assistance in the Health and Population Sector, for$300,000, approved on 27 April 1995.The remaining project (Loan No. 710-PAK:Second Health and Population Project), for which a projectcompletion report has just been completed, was added to provide a more complete picture, even though aproject performance audit report was not prepared. Pakistan had five Bank-assisted projects of which onlyone has been postevaluated.TA No. 5294: Regional Study of the Health and Population Sector, for $198,000, approved in June 1988.

3



2

These four DMCs were chosen because of the Bank's significant involvement in thesecountries' H&P sectors.

6. Beneficiary surveys were conducted by locally recruited consultants in each of

the four DMCs. International consultants, who supervised the local consultants, wereresponsible for the preparation of the country impact studies and consolidated impact report,except for PNG. Due to initial difficulties and delay in recruiting suitable local consultants, thebeneficiary survey in PNG did not have the benefit of supervision by international consultants.Instead, the survey in that country was undertaken by a highly qualified locally recruitedexpatriate consultant with assistance from Health Department staff. More details on the surveymethodology and its constraints are given in Appendix 2. The surveys were complemented bya desk study and an in-depth review of all project documents (including those of H&P projectsin other countries), field visits, and discussions with Bank staff and executing agencies. Theusefulness of secondary data from the DMCs is limited by the quality of data in thesecountries.

7. A caveat must be applied to the approach adopted in undertaking thesesurveys. The macroeconomic environment and overall economic management in these fourcountries differed substantially from DMCs in East and Southeast Asia when these projectswere implemented. Besides their greater capacity to provide the necessary resources becauseof a higher developmental stage, the DMCs in East and Southeast Asia were able to releasemore resources for the H&P sector by spending less on income transfers and by leaving themost productive investments to the private sector. Such releases of resources created a morefavorable environment for positive impact than any other kind of intervention in the sector. Itwould therefore be more accurate to view this Study of the four selected countries which had aless favorable environment as a subset of the Bank's overall assistance in the H&P sector.

B. Initial Conditions against Which Impact was Appraised

1. Macroeconomic Background

8. The economic performance of the four countries covered in the Study wasgenerally characterized by chronic budget deficits and negative current account balances.Bangladesh was one of the poorest countries in the world with a per capita income of around$250 in 1995. Economic growth had remained low and poverty was prevalent among both ruraland urban households. In Pakistan, fiscal and external macroeconomic imbalances andstructural problems persisted, even if the gross domestic product (GOP) exhibited sustainedgrowth rates averaging 6 percent over the past decade and a half. For Sri Lanka, economicperformance had been sluggish, but this improved substantially in the 1990s following the

revitalization of an economic liberalization program in 1989. The economy of PNG during thesame period, on the other hand, was unsteady, characterized by sluggishness during the firstpart of the 19805, followed by a burst of economic activity during the 1985-1988 period, then ageneral decline from 1989 after the closure of two of its most important mines, except for1991-1993 when the economy picked up.

9. Political instability and threats to national security were a common experience inall four countries. This resulted in fiscal management problems that involved not only budgetdeficits, but the inability of governments to finance other social costs. As a large portion of the



3

expenditures was devoted to national defense, such social concerns as health and educationwere inevitably affected. In Bangladesh, this was further aggravated by the periodicoccurrence of such natural calamities as floods and typhoons, further impinging on whateverresources remained for addressing these social problems.

2. Social Development

10. While the four countries covered under the TA operated within diversesocioeconomic and cultural environments, they had some similarities. Significant proportions ofthe population in each country fell below the poverty line. The highest incidence of poverty forthe period 1980-1991 was recorded for Bangladesh, with 78 percent below the poverty line,followed by PNG with 73 percent. In contrast, Pakistan's poverty incidence rate was 28 percentand Sri Lanka's was 39 percent. Varying rates of progress had been made in the reduction ofpoverty among the countries under study. In Sri Lanka, the poverty levels declined by about 18percent between 1985 and 1990.

11. Among the four countries, Sri Lanka was the most socially developed. Itsliteracy rate was about 90 percent, compared with PNG's 72 percent, Bangladesh's 38percent, and Pakistan's 37 percent. The average life expectancy was also longer in Sri Lankaat 72 years against 62 years in Pakistan, 56 years in Bangladesh, and 55 years in PNG.Women enjoyed better status in Sri Lanka as indicated by the much higher female literacy rateof 87 percent compared with 63 percent in PNG, 26 percent in Bangladesh, and 24 percent inPakistan (Table 1, Appendix 3).

3. Overview of the Health and Population Sector

12. Two of the countries were characterized by high population growth rates.

Pakistan's rate of 3.0 percent was the highest for the period 1990-1995, while PNG's was 2.0percent over the same period. In contrast, Sri Lanka had the lowest rate (1.2 percent) followedby Bangladesh (1.8 percent).

13. Pakistan's population grew from 82.6 million in 1980 to 129.8 million in 1995.Over the same period, PNG's population grew from 3.0 million to 4.1 million; Bangladesh, from88.5 million to 116.9 million; and Sri Lanka from 14.8 million to 18.0 million. While the declinein the population growth rate generally reflected the declining crude birth rate (CBR), theoutmigration of workers since 1981 also played a key role.

14. Mortality measured in terms of crude death rate (CDR) remained at around 11per thousand population in 1993 for Bangladesh and PNG, with Pakistan slightly better off at 9

per thousand population. But Sri Lanka's CDR at around 6 per thousand population wascomparable to the more advanced DMCs. Infant mortality rate (lMR) was highest inBangladesh at 106 per thousand live births, followed closely by Pakistan with 88 per thousand.Predictably, Sri Lanka's IMR was the lowest at 17 per thousand live births. PNG's IMR of 67per thousand, although it was lower than Bangladesh or Pakistan, was still considered serious.In general, health conditions in Sri Lanka were the most improved while those in Bangladeshand Pakistan were the least favorable (Table 2, Appendix 3).



4

15. Differences in the health and family status among the countries were partlyattributed to the degree of effectiveness achieved by the government's health and familyplanning (FP) programs. In Bangladesh and Pakistan, health conditions and family planningactivities did not improve Significantly over the Study years. In addition to social and religious

factors, low literacy rates made the implementation of health and family planning programsdifficult, particularly in the rural areas. On the other hand, Sri Lanka's family planning programswere very successful in reducing the country's population growth rate. This was attributed tosocioeconomic and cultural conditions, and high literacy rates, which made the people moreresponsive to modern health and family planning strategies.

4. Policy Environment

16. Because of the budgetary problems discussed earlier, public expenditures onhealth in the four DMCs had lagged over the years. Health expenditures had remained below 2percent of GOP, which partly accounts for the poor local health conditions in these countries.In recent years, however, the governments of these countries had become more cognizant ofthe need to improve the overall health status.

17. The Health for All by the Year 2000 (HFA 2000) campaiqri' was endorsed by theDMCs under study and provided the impetus for reviewing policies in the sector. InBangladesh, the campaign was launched by shifting the public health system's emphasis fromcurative to preventive care. Health policies have since focused on strengthening healthmanagement and paramedics, local production and supply of essential drugs, and integratinghealth services with the national family planning and nutrition program.

18. In Pakistan, the Government launched its Social Action Program in 1992. Thisprogram aimed at improving the provision of basic social services in four key subsectors, oneof which was primary health. Under the Social Action Program, the main priorities for the healthsector were (i) to improve the quality of care by strengthening managerial effectiveness andthe delivery of health care services; and (ii) to increase accessibility, particularly by women,through improving the vertical health programs, provision of drug supply, and mobilization ofresources. The main priority of health subsectors was the expansion of health facilities.Additionally, the Government encouraged greater participation of the private sector in theprovision of health care.

19. In PNG, much of the health infrastructure was already in place at the time of thecountry's independence. Since then, public health administrators have concentrated mostly onupgrading existing facilities. There have been four Five-Year National Health Plans sinceindependence. The main health priorities identified under the current Plan are to (i) increase

health services to the rural majority; (ii) expand health promotion and preventive services; (iii)reorganize and restructure the national health system; (iv) develop skills of health workers; and(v) upgrade and maintain the health infrastructure.

20. Sri Lanka's health services were primarily provided by the public sector andmedical treatment was generally provided free in all hospitals, clinics, and dispensaries. In

The goal of this World Health Organization campaign is to assure that all member countries achieve astate of health enabling all citizens to lead socially and economically productive lives by the year 2000.



5

1992, the National Health Policy was formulated, emphasizing health promotion and diseaseprevention, improvement in the quality and range of existing services, and decentralization ofhealth administration. A Presidential Policy Statement in 1995 reiterated the Government'scommitment to achieving higher health status for the people and identified such additional

thrusts in health services as preventive disease control and availability of primary-levelfacilities, targeting of poor income groups and rural regions, and eradication of childmalnutrition.

C. Bank Operations in the Health and Population Sector

21. Bank assistance to the H&P sector was based on the premise that improvementin the health status and moderate growth of populations had a beneficial influence on thesocioeconomic development of DMCs. The Bank's first assistance to the sector was extendedto the Sha Tin Hospital-Polyclinic Project in Hong Kong, amounting to $19.5 million, approvedon 14 September 1978.

22. By September 1996, the Bank had approved 31 H&P projects dispersed in 14DMCs and totaling $932.55 million (about 1.6 percent of total lending). Of this amount, 43.25percent (11 projects totaling $403.33 million) were funded from ordinary capital resources, and56.75 percent (21 projects totaling $529.22 million) from the Asian Development Fund.Additionally, 70 TA projects have been approved amounting to $24.03 million, representing2.34 percent of the Bank's total TAs.

23. Bank-financed H&P sector projects in the past concentrated mainly on theimprovement of service delivery through the upgrading/construction of middle-level referralhospitals, large urban-based hospitals, and primary health care (PHC) facilities. The RegionalStudy on the Health and Population Sector conducted in 1989 identified priority areas forfuture assistance. These areas were sector planning and policy development, healthadministration and management, PHC services and targeting of benefits to high-riskpopulations, family planning services, cooperation with nongovernment organizations, humanresource development, essential support services, and sector financing.

24. Long-term considerations for future H&P projects were also identified under theRegional Study. These included (i) increasing the geographical coverage of Bank assistance,(ii) providing a more balanced hardware-software financing mix, (iii) improving projectsupervision and implementation monitoring, (iv) enhancing institutional support though regionalTAs and advisory TAs, (v) financing recurrent costs, (vi) using the sector and program lendingapproach whenever appropriate, (vii) improving project justification techniques, and (viii)enhancing intersectoral coordination within the Bank.

25. As a result, the Bank's involvement in the H&P sector over the last two decadeshas successively moved from (i) supporting middle-sized hospitals to rural health infrastructure,to development of health personnel (including equipment and other softer aspects), and tosectorwide policy reforms; (ii) purely Ministry/Department of Health prescribed programs andactivities to nongovernment organizations and beneficiary defined programs; and (iii) purelystate-owned, funded, and delivered services to a mix comprising private sector, insurancemarket, and contracted services. Such trends would not have been fully captured by theprojects under study.



6

III. OBJECTIVES, SCOPE, AND IMPLEMENTATIONAL EXPERIENCEOF THE PROJECTS

26. While all of the projects in the Study were aimed at improving the health status

of the beneficiaries, the approaches for achieving this goal varied for each country. InBangladesh, the initial focus of health projects was on the provision of essential drugs and theimprovement of quality of health services. To ensure the adequate supply of drugs intendedfor PHC, a pharmaceutical company was established from an existing publicly ownedpharmaceutical production unit. Additionally, a DDT (dichlorodiphenyltrichloroethane) factorywas refurbished to boost the national campaign against malaria. Health services would beenhanced both through improved hospitals and better auxiliary services. These servicesincluded warehousing and repair and maintenance of medical facilities and equipment. Thesubsequent projects included the establishment of health centers.

27. In Pakistan, on the other hand, health projects initially concentrated on providingtraining to medical staff and supplying medical equipment. Nurses and paramedical staff weremajor beneficiaries of the training components. Family planning activities also assumedimportance in the projects as training for field staff were conducted and family planningservices in health outlets were provided. Attention to the provision of basic health centers wasgiven in later projects.

28. Improvement in the rural health status was the main objective of the twoprojects in PNG. Existing health facilities were provided with medical equipment and medicalstaff. The specific focuses of both projects centered on health infrastructure, water supply andsanitation, communications and transport for health service delivery, and health education.

29. The first Health and Population Project in Sri Lanka was intended to improve thedelivery of integrated PHC by providing both preventive and curative services to thecommunity, and concentrated on strengthening the health infrastructure by establishing healthcenters at various administrative levels, concomitantly providing them with medical equipment,furniture, and transport services. Additionally, in-service training on public health and familyplanning was provided (Appendix 4).

30. All seven projects incurred implementation delays ranging from one to fouryears (20-124 percent time overrun). Projects in Bangladesh had the longest delays, averagingabout four years. Projects in PNG had the lowest time overrun, averaging about one yeardelay. Common factors cited for the delays were institutional inefficiencies of the governmentsand executing agencies such as poor intra-agency and interagency coordination and red tape,and lack of familiarity with the Bank's implementation procedures. Other causes for the time

overrun included (i) lack of counterpart funds; (ii) problems with procurement; (iii) difficultieswith land acquisition, siting, and technical problems; and (iv) civil unrest and weatherdisturbances (Table 1, Appendix 5).

31. Except for the Second Rural Health Services Project (RHS2) in PNG, the rest ofthe projects, despite substantial delays, were completed at lower costs than the appraisalestimates. Cost underrun ranged from 3.4 percent in the Health and Family Planning ServicesProject in Bangladesh to 39.6 percent in the Health and Population Project in Pakistan. In thecase of PNG's Second Rural Health Services Project, with an overrun of 4.7 percent,



7

additional costs were incurred for the pharmaceutical component that was added duringimplementation, to which initial loan savings had been reallocated. Overall, the average costunderrun for the seven projects was about 24 percent. The main reasons cited for the costunderrun were the reduction or cancellation of some project components, and savings arising

from the devaluation of the local currencies. Another factor that contributed to lower actualproject costs was the high provision for contingencies set during appraisal (Table 2, Appendix5).

32. The implication of cancellation of project components during implementation isnoteworthy in terms of assessing the impact of Bank's assistance. Impact on capacity buildingcould be lessened with the cancellation of the component on training and related consultingservices, as in the case of the Second Health and Population Project in Pakistan. Similarly,cancellation of the financing of part of the DDT supplies in the Public Health Program inBangladesh reduced the effectiveness of the malaria campaign (incidences of malariasubsequently increased), though it could be argued that longer-term adverse impact on theenvironment and health of those exposed to DDT would have been lessened.

IV. IMPACT OF BANK OPERATIONS

33. In assessing the impact of the Bank's assistance in the H&P sector, one has tobear in mind the change in the Bank's orientation as regards project appraisal. Concerns oversector linkages, social dimensions, project quality, issues of ownership, and beneficiaryparticipation, for instance, were only emphasized in recent years. Extracting lessons learned inthe context of these present concerns, however, may be difficult because they were notapplicable in the past. Attempts to assess the impact of the Bank's assistance is furtherconstrained by the failure in most loan documents to indicate the criteria by which impact ismeasured. Nevertheless, an attempt is made here.

34. Because of the nature of the Bank's assistance in the H&P sector, thediscussion focuses on the impact of the seven completed projects across the four differentcountries. The Study assesses impact in the following six categories: health status, policyadjustment, government budget, institutional development, social dimensions, andenvironment. The geographical impact under each of these categories, however, differed bycountry because of the scope of the projects and the nature of the components. Somecovered the whole country (e.g., PNG) while others were confined to administrative units (e.q.,specific provinces in Pakistan and districts in Sri Lanka).

A. Impact on Health Status

35. The impact of the Bank's assistance on the health status of the DMCs wasreflected by demographic and health indicators, morbidity indicators, service provisionindicators, and beneficiary perceptions of the facilities provided. Importantly, however, theseindicators often reflected health improvement efforts as a whole rather than just the Bank'scontribution. While the indicators may have their limitations, they do represent generalmeasurements of impact on H&P as put forward by experts engaged under the Study.



8

1. Demographic and Health Indicators

36. Achievements presented by demographic and health indicators varied bycountry. The most significant achievements were in Sri Lanka. Vital H&P indicators such as

CBR, CDR, IMR, and maternal mortality rate in project districts between 1988 and 1994showed a gradual improvement in line with national trends. The average percentage reductionin CBR from 1989 to 1992, for instance, was 13.7, whereas the figure for Sri Lanka over thesame period was 7.8 (Table 1, Appendix 6). In Bangladesh, these vital health indicators alsoshowed a declining trend in some of the project districts. In Chandpur, the CBR dropped by 6.1percent between 1990 and 1992, while the IMR dropped by 6.4 percent in line with nationaltrends (Table 2, Appendix 6). The picture was less clear in Pakistan given the paucity of timeseries data for comparison in the project areas before and after project implementation. InPNG, on the other hand, many indicators showed an increasing trend. National IMR, forinstance, increased from 65 per thousand births in 1985 (during implementation of the first andsecond Rural Health Services Projects [RHS1 and RHS2]) to 67 per thousand in 1993 (afterimplementation). While improvements in health status could not be specifically attributed to the

Bank's projects, these figures generally did indicate a positive contribution of the projectinterventions in varying degrees toward the improvement of the health status of the people inthe project districts. Even in the case of PNG, health officials indicated that the projects helpedto arrest the rapid deterioration of the health environment brought about by the worseningsocial conditions.

37. The projects also appear to contribute to the reduction in population growth. Thelowest percentage population growth was recorded for Sri Lanka, though the migration ofworkers, especially to the Middle East, was a very significant factor. While the populationgrowth rates in the project districts show a slightly declining trend in Bangladesh and SriLanka, a more direct indicator of impact is CBR which showed a steadily declining trend,except for Ratnapura district in Sri Lanka (Tables 1 and 2, Appendix 6).

2. Morbidity Indicators

38. The pattern of major diseases varied from country to country. In Sri Lanka,upper respiratory tract infections, infectious and parasitic diseases, and complicationsassociated with pregnancy, childbirth, and puerperium, were the most common ailments.Elsewhere in Bangladesh, infectious and parasitic diseases (particularly tuberculosis), anddiarrhea were common. Pneumonia, malaria, and perinatal conditions (neonatal sepsis) werekiller diseases for all ages in PNG, with typhoid and AIDS (acquired immunodeficiencysyndrome) the main new and significant health problems. In Pakistan, the most commondiseases were acute respiratory infections, gastrointestinal disorders, and malaria in the

project districts. The impact of Bank projects in reducing the incidence of these diseasesdepended to a great extent on the nature of the projects. Where projects had importantcomponents in providing basic health units at the most peripheral levels (for example,Gramodaya Health Centers [GHCs] in Sri Lanka or Basic Health Units in Pakistan), the positiveimpact of the projects was more discernible, even though some results were mixed.

39. Although morbidity data based on government hospital records do not give acomplete picture of the extent of morbidity in the community (because patients also soughttreatment from facilities other than government hospitals), these data do indicate the morbidity



9

trend in the country. In a number of instances, the morbidity rates increased. In Pakistan, therate in the tehsil headquarters hospitals increased from 1.3 percent in 1981 to 2.5 percent in1990. But in project District Headquarter hospitals the morbidity rates came down slightly from2.8 percent in 1981 to 2.3 percent in 1990. In the case of Sri Lanka, morbidity trends by

districts were mixed, but for the country as a whole, the decreasing trend was evident duringthe period 1988-1994. Specifically, the incidence of the three most prevalent immunizablediseases decreased in most districts as well as in the nation as a whole, due to the verysuccessful Expanded Program of Immunization implemented through project facilities.Certainly in Sri Lanka at least, the Bank's assistance had positive impacts in reducingmorbidity in the project districts, especially because of the extensive immunization activitiescarried out at GHCs established by the projects.

40. The downward trend of disease patterns was confirmed by beneficiariessurveyed in various field studies-by as much as 84 percent of the beneficiaries in Sri Lanka.Reasons given for the decline were increased coverage of immunization, better healthpractices, increased awareness of health and diseases, and the seeking of early treatment forillness. These reasons are directly attributable to the services provided by the various healthcenters, especially those at the most peripheral levels, established through Bank assistance.

3. Service Provision Indicators

41. The nature of services provided by the health centers under Bank projectsvaried from country to country according to whether the emphasis was on curative orpreventive health services. In Sri Lanka, the GHCs were oriented towards preventive andpromotive health care, especially in the area of maternal and child health care and familyplanning. The opposite was true in Bangladesh where curative services were the dominantactivities in the village health centers. While there was an increasing emphasis towardspreventive and promotive health care, especially in those countries fully subscribing to HFA2000, curative health services were still much sought after. Access to the health facilities, bethey oriented more towards curative or preventive health care, are an important element in theevaluation of the impact of Bank's projects. Other than curative services, the health centersprovide prenatal, natal, and postnatal care; family planning; immunization; growth monitoring;nutrition supplementation; and health education.

a. Access to Health Facilities and Home visits

42. Access to health care by the beneficiaries improved with the availability andproximity of the health centers constructed under the Bank's projects. Access to health carecan be assessed either in terms of the ease with which beneficiaries visit the health centers, or

in terms of the ease and frequency of health workers visiting the beneficiaries in their homes.Generally, where the facilities were provided under the projects, there was a reduced need totravel long distances for health assistance. Between 63 and 72 percent of respondentsinterviewed in various districts of Bangladesh reported that the need to travel long distancesfor health care was definitely reduced. However, in many instances, access was marred by thepresence of other constraints. In PNG, for example, it was tempered by the lack of medicalsupplies at aidposts, irregular patrols by health workers, and poor maintenance of the facilitiescompounded by the poor security situation in the country.



10

43. Home visits by health workers formed an important link in the provision of healthservices. An important function of the public health midwife (PHM) in Sri Lanka, for instance,was to visit homes for the purpose of registering and providing care to children and pregnantwomen. Though the average number of home visits by a PHM had not changed significantly

over the years, the fact that the PHM now resided in her area of work at the GHC should resultin more efficient visits and time savings for other activities associated with the GHC, such asimmunization. However, in PNG, home visits were made difficult by the breakdown of law andorder. Irregular maternal and child health (MCH) patrols had been cited by as large as 47percent of the beneficiaries interviewed in the project areas of RHS1 as the reason for the lackof health improvement in the project areas.

b. Prenatal Care and Immunization Coverage

44. The Bank's assistance improved the availability of prenatal care andimmunization services, particularly in the rural areas. In Sri Lanka, the registration of pregnantwomen by a PHM before the end of the fourth month of pregnancy and the immunization of

pregnant women with tetanus toxoid showed progressive increase in health standards over theyears as a result of the existence of GHCs constructed under the first Health and PopulationProject (Table 1, Appendix 7). In the case of PNG, prenatal care coverage increased from 56percent of pregnant women receiving care (as against those attending clinics) to 70 percentduring the 1987-1991 period. Similarly, prenatal coverage by trained personnel in the projectareas in Bangladesh was comparatively higher. Prenatal care in Mymensingh district in 1993(73.3 percent) and in 1994 (72.2 percent) were comparatively much higher than thecorresponding figures for the Rajshashi district, a nonproject district (Table 2, Appendix 7). Inthe case of another project district, Jessore, which shows a lower figure, the trend is increasingas against the decreasing national trend.

45. Immunization programs were greatly facilitated by the existence of healthcenters at all levels provided under the Bank's assistance. In Bangladesh, the immunizationfigures in some project areas were higher compared to the national figures. The Mymensinghdistrict's registered immunization rate against BCG (bacillus Calrnette-Guerin, the tuberculosisvaccine) exceeded 100 percent compared to the national figure of 95 percent in 1994 (Table 2,Appendix 7). The picture for PNG was similar. The annual immunization coverage for childrenunder one year for the vaccines of third Triple Antigen, third Polio, measles, and BCG showedvery clear improvement in coverage from 1982 to 1990. However, over the same period, infantmortality in PNG rose. As immunization coverage increased for the vaccines mentioned, thechildren were not immunized against the major killer diseases of infancy and childhood(pneumonia, malaria, and diarrhea).

c. Attendance of Deliveries

46. Attendance at childbirth deliveries by a health worker was facilitated by thepresence of peripheral health centers in the rural areas. While over 85 percent of deliveries inSri Lanka took place in the maternity units of hospitals, PHMs, by virtue of their 24-houravailability as they resided at the GHCs, could be called upon to assist home deliveries inemergencies. In Bangladesh, the percentage of deliveries by trained health personnel in someof the project areas increased. In the district of Jessore, for instance, the percentage increasedfrom 40 percent in 1993 to 52 percent in 1994 (Table 2, Appendix: 7).



11

d. Family Planning

47. FP activities have generally increased in the project areas. In Bangladesh, from

1990 to 1994, they have increased considerably except in the case of sterilization. Thecontraceptive acceptance rate increased from 58.6 percent in 1993 to 62.0 percent in 1994,while intrauterine device (IUD) insertions increased from 5,529 in 1990 to 14,377 in 1994 in theKishoregonj district. In Sri Lanka, distribution of oral pills and condoms improved from 1988 to1994, pills increasing from an average 0.2 to 0.4 packets per eligible couple and condomsmoving up from an average of 0.3 to 1.2 per eligible couple. In PNG, FP activities wereintegrated with the MCH programs. Bank assistance has contributed to ensure widespreadprevalence of FP services down to the aidpost level (where condoms, pills, and ovulationmethods are predominant), the health center and subcenter levels (where condoms, pills,depo-provera, ovulation methods, and loops are available), and at the hospital levels (allmethods plus ligation and vasectomy).

4. Beneficiary Perceptions of Services

48. The types of beneficiaries of the H&P projects differed according to the projectcomponents. These varied from those who derived direct benefits, such as women andchildren attending basic health care units at the village levels, to those who derived indirectbenefits, such as hospital patients who benefited from speedier diagnoses of medicalequipment workshops funded by the projects. Additionally, health service providers benefitedthrough the upgrading of their expertise through training and the provision of medicalequipment. The beneficiary surveys normally covered two groups of beneficiaries: (i) the usersor primary beneficiaries of health facilities, consisting mostly of women and children in the caseof the most peripheral health units; and (ii) the health care providers, including seniorgovernment health officers. Beneficiary perceptions of the impact of the projects have beenevaluated in terms of whether health services have improved, whether they were satisfied withthe services provided, and whether they perceived that there had been an improvement inhealth. Beneficiary perceptions of the services provided by the health centers were generallypositive. However, wide variances in the proportion of respondents reporting the positiveperceptions occurred not only across countries, but even within the country among the projectareas.

49. An average of about 65 percent of the primary beneficiaries in Sri Lankareported marked improvement in the services provided by the Bank-assisted health centers.Regularity of services, improved quality of MCH and FP services, availability of PHMs 24 hoursa day, and easy access were given as reasons for the improvements. In PNG, 95 percent of

the primary beneficiaries in the project areas of RHS1 but only 55 percent in some of theproject areas of RHS2 reported that services of the health centers had improved. The majorreason for this perception was the daily availability of the services. In the case of Pakistan,over 70 percent of the primary beneficiaries from project District Headquarter hospitals andabout 80 percent from the Tehsil Headquarter hospitals agreed that the services provided bythe hospital had improved considerably.

50. Most beneficiaries in Sri Lanka were satisfied with the services provided by theProject health centers, a few being dissatisfied for reasons of lack of expansion of services, ill-



12

maintained buildings, irregularity in the provision of services, and long waiting time. In PNG,there were more satisfied beneficiaries from the first project compared with the second project:over 90 percent of the primary beneficiaries from RHS1 were satisfied while only an average ofabout 65 percent of the beneficiaries from RHS2 were recorded as being satisfied with the

services provided.

51. The perception of primary beneficiaries with respect to improvement in healtharound them was similar. About 80 percent of those in RHS1 areas indicated that family healthhad improved. This is in contrast to the RHS2 areas where only about 48 percent of thebeneficiaries mentioned that family health had improved.

52. Perceptions of service providers generally supported the view that the projectshad contributed to the improvement of health services and health levels of the areas served bythe projects. In Pakistan, almost all health care providers interviewed in the Project hospitalsagreed that there was considerable improvement in the provision of services as a result of thesupply of essential medicinal and surgical equipment provided through the projects. Their

diagnostic and treatment capabilities had also improved considerably. Even in PNG, where theoverall morbidity indicators were not encouraging, the health care providers were of theopinion that the projects assisted by the Bank helped to arrest the seriously declining trend inhealth brought about by the deteriorating social conditions of the country. About 55 percent ofthose interviewed thought that the health of the people had improved on the basis of lesssickness episodes and improved nutritional status of the family. Such improvement wasbelieved to have been facilitated by the close proximity of health services in the case of 75percent of responses and by the provision of MCH patrols in the case of 56 percent. InBangladesh, response from the health care providers confirmed the usefulness of theinfrastructure, equipment, and drug facilities provided by the projects.

B. Impact on Policy Adjustment

53. Except for perhaps Sri Lanka, it does not appear that the Bank's assistance hadany significant or long-lasting impact on policy adjustments in the DMCs concerned. In thecase of Sri Lanka, the success of the GHCs led the Ministry of Health to modify the PHCmodel by focusing more on the provision of preventive health care through GHCs. The changein operational policy led to the construction of more GHCs in other parts of the country fromthe Government's own funds and external assistance from other funding agencies.

54. Cost recovery as a policy encouraged under various projects did result in someattempts by the governments to impose user charges. But these were limited to payment forrelatively minor expenses like payment for contraceptives. While the governments recognizethe validity of the policy of cost recovery, they are faced by the perception and expectation ofthe people that health services should be provided gratis and are therefore disinclined toimplement this policy.

55. While governments welcomed private sector participation in the provisron ofhealth services, attempts to promote privatization as a matter of policy were limited, except inthe case of one of the projects in Bangladesh where a Government pharmaceutical productionunit was converted into the private Essential Drug Company Limited. Provision of healthservices by Government doctors on a private basis was either officially sanctioned as in the



13

case of Sri Lanka, or tolerated as long as the services were provided outside of office hoursand premises.

c. Impact on Government Expenditure

56. Expenditure of government counterpart funds during construction of healthfacilities and electro-medical workshops, the consequent recruitment of health staff, andsubsequent operation and maintenance (O&M) requirements, caused increases in thedevelopment and recurrent expenditure of the governments. The degree to which the strainwas felt by the respective governments depended on the fiscal position of the countries and, inall cases, there were competing demands for resources such as maintenance of security orlaw and order. Although O&M allocation for health facilities was often inadequate, however,health expenditure in general increased in nominal terms. In Sri Lanka, the provision formaintenance of GHCs in general was generally considered inadequate by the officialsinterviewed. Isolating the impact of Bank-assisted projects on government expenditure fromthat produced by other health projects is difficult. Nevertheless, the Bank's assistance hasdefinitely contributed to the increasing demand for greater budget allocation.

57. Public expenditure on health in Sri Lanka has increased fivefold in nominalterms from SLRs1.34 billion in 1980 to SLRs6.54 billion in 1992. As a proportion of totalGovernment spending, it increased from 4.7 percent in 1980 to nearly 5.5 percent in 1992.Over the same period, public development expenditure in Pakistan on health and nutritiontogether with population planning increased from PRs813 million to PRs3.1 billion. In PNG,Government expenditure on health increased from K55 million in 1980 to K114 million in 1992,while in Bangladesh it increased from Tk1.34 billion to Tk6.83 billion over the same period.

58. However, the impact is mixed when viewed in the context of the share of the

GOP devoted to health expenditure. In Bangladesh, health expenditure as a percentage of thenation's GOP increased, though marginally, from 0.7 percent in 1980 to 0.8 percent in 1992;while in PNG, the share decreased from 3.2 percent to 2.8 percent over the same period. Incontrast, public health expenditure in Sri Lanka as a share of the GOP increased from 1.3percent in 1985 to 1.5 percent in 1992. In Pakistan, the share as a percentage of the grossnational product had remained constant at about 0.7 percent over the same period, except for1988 and 1989 when the share rose to about 1.0 percent.

D. Impact on Institutional Development

59. Construction of new health facilities, upgrading of existing infrastructure, supplyof equipment, and provision of transport and repair facilities to the project centers, coupled

with greater supply of pharmaceutical products, has enhanced the capacity of theseinstitutions to better serve the community. In particular, the capacity of the health providers toprovide better health care, especially those who served at peripheral levels, has beenaugmented (e.g., the provision of PHMs with residential facilities at the village level in SriLanka). However, the impact on institutional capacity was not evenly spread and, in manycases, nullified by the lack of maintenance in the health centers. An example is the case ofPakistan, specifically Sindh Province, which had a policy of no allocation of maintenanceexpenditure for new buildings for the first two years. Significant constraints in PNG includedthe unavailability of drugs in some aidposts, the lack of security for health providers working in



14

the villages, and the reluctance of health providers to stay in the village health centersbecause of the absence of civil and social amenities. In a few cases, the full impact of theprojects in terms of institutional development, and ultimately the efficient provision of healthservices was impaired by the non-usage or inoperational condition of some of the equipment

(e.g., the unpacked X-ray machine in Sindh funded under the Second Health and PopulationProject') or facilities (e.g., a medical equipment workshop occupied by law and order personnelin Pakistan and wrongly sited village health units in Sri Lanka and Pakistan).

60. Capacity building components of the projects tended to be generally designed insupport of institutional development plans for the projects. Nevertheless, there were caseswhere the capacity measures were specifically designed to facilitate project implementationsuch as in the case of the Sri Lankan project which provided for training personnel insupervision of project implementation as well as monitoring and evaluation of project benefits.

61. The training components of the projects served to expand and upgrade thecapacity of the existing health providers (paramedical personnel, medical technologists,

laboratory technicians, nurses, and family planning personnel) as well as the capacity of theDMCs to provide health services in both the public and private sector. While it is difficult toquantitatively measure the impact of the training components of the projects due to their long-term and indirect effects, the impact is certainly positive. Training accorded to PHMs and publichealth inspectors in Sri Lanka, for instance, assisted in improving the technical and managerialskills of both. The in-service training enabled PHMs to improve the quality of their services and,as acknowledged by the beneficiaries, their knowledge of family planning has also increasedconsiderably. In colleges assisted by the Bank, the impact would have been even moreextensive if not for the limitations imposed by physical intake capacity. Of about 150 studentsapplying to the College of Technology under the Institute of Health in Pakistan for a medicaltechnologist course, only 20 students could be accepted in 1990. Greater impact of the trainingcomponents of the projects has been marred by the inability of the DMCs to provide adequatestaff to take advantage of the fellowships provided under the projects. Such inability in fact hasled to cancellation of some of the components by the Bank as in the case of the first Healthand Population Project in Pakistan.i

E. Social Impact

62. The social impact of the projects has been favorable with regard to women andpoverty reduction. However, it must be pointed out that at the time of project formulation, socialconsiderations were not given prominence. Performance indicators in poverty reduction, forinstance, were not specified, even though health programs have obvious implications forpoverty reduction.

1. Impact on Women

63. The major beneficiaries of the Bank's assistance were women and children.Given the role of women in all these DMCs with regard to health matters, this was to beexpected. In Sri Lanka, almost all the services provided at GHCs by PHMs and the other

2Loan No. 710-PAK(SF):Second Health and Population, for $16 million, approved on 29 November 1984.Loan No. 562-PAK(SF):Health and Population, for $15 million, approved on 15 December 1981.



15

health staff were targeted at women and children. Considering that the profiles of beneficiariesin the village health facilities in other DMCs were similar, the impact on the improvement ofhealth status of women has been considerable.

64. MCH and FP clinics conducted at the health centers, such as those in SriLanka, were also occasions for social gatherings of women. Discussions among women,particularly mothers, commonly centered on health issues. One significant activity at the clinicswas the health education talk on current health topics given by PHMs to mothers and children.The result was increased awareness of mothers about health and disease. Improved literacyamong females, increased women's employment, and women's general contributions tosocioeconomic development are at least partly attributable to the improved health status ofwomen as well as reduction in family size.

65. The impact of the projects on women should also be considered in terms oftraining. Many trainees were women, notably nurses, PHMs, and FP workers, besides those in

categories of health providers as medical technologists. For women already employed, thetraining served to upgrade their skills. For those not employed at the time of training, theprojects afforded them better opportunities for employment with higher qualifications in eitherthe public or private sectors.

2. Impacton Poverty

66. The incidence of poverty in rural areas was higher than that in urban areas. Theselection of socioeconomically deprived areas for Bank assistance was appropriate inconsideration of the positive impact improved health status made in enhancing the incomelevels of individuals and families. It is difficult to determine in quantifiable terms the economicbenefits to the family brought about by improved health. Easy access and availability of health

facilities reduce travel time and cost, reduced family size due to increased family planningacceptance, and overall improvement in health status (thus increasing employmentopportunities and reducing work absenteeism) have all had a contributory impact on incomegeneration and poverty reduction in the family.

F. EnvironmentalImpact

67. Except for the possible effect from the DDT factory upgraded under the PublicHealth Program Project in Bangladesh, no direct adverse impact on the environment wasexpected given the nature of the Bank's assistance. The possibility of adverse effects from thespraying of DDT to contain the malaria vector was resolved when the factory was closed in1991. In general, project interventions have had positive impacts, directly or indirectly, on theenvironment. Examples are the facilitation of available potable water supply in PNG andimproved sanitation in all the countries brought about by the increased awareness of people,especially mothers, of the sources of environmental pollution. Knowledge about properdisposal of household and garden refuse and the need to use sanitary toilets was generallyimparted in the health education talks given to mothers at the clinic centers. Further, theadverse effects of excessive population growth on the environment could be reduced byeffective FP methods. Environmental sanitation was the primary responsibility of the publichealth inspectors who were part of the PHC complex. Their knowledge and skills regarding thisaspect were improved by the in-service training provided through the projects.



16

V. KEY ISSUES FOR THE FUTURE

A. Sustainability of Impact of the Bank's Assistance

68. The initial impact generated by the Bank's assistance must be sustained.Sustainability of the benefits of the projects depend on the availability and operationalefficiency of the facilities established under the projects. The lack of adequate budgetprovisions for O&M and related delays in making project facilities operational have significantlyreduced the full potential impact of the Bank's assistance. In all the DMCs under study, theproblem of inadequate funding for O&M of the health centers (including hospitals), the trainingschools and institutes, the electro-medical workshops, and equipment was constant. In mostcases, this resulted from competing demands on an already tight budget. In some cases,however, it was the result of the manner in which the allocation for maintenance was made,normally allocated as a one-line vote at the headquarters level of the department or ministry.This situation invariably resulted in health centers distant from the headquarters obtainingmuch less allocation than requested. Given their low priority in the administrative hierarchy,

therefore, the most peripheral health units (the GHCs in Sri Lanka, the Basic Health Units inPakistan, or the aidposts in PNG) had little or no allocation for maintenance, even though itwas recognized that the clientele they served were in greater need of health care. Moreover, insome other cases the lack of maintenance was the result of policy measures. In SindhProvince in Pakistan, for example, possibly in response to budgetary constraints, no allocationfor maintenance was normally made for any new buildings for the first two years. The problemin most cases was that more money was required to rehabilitate the health facilities than wouldhave been required had they been maintained regularly.

B. Cost Recovery

69. Related to the above issue is the question of cost recovery. If user charges areimposed, the sums collected could contribute to minor maintenance expenditures. The policiesof the governments reflected their reluctance to impose fees on users of public health services,particularly the more vulnerable groups. Some shift in the thinking of these governments wasdiscernible perhaps faced with the strain on the budget. Current attempts at cost recoveryincluded nominal charges collected at GHCs for condoms and contraceptive pills in Sri Lanka,or charges for certain categories of wards in Pakistani hospitals. These attempts remaininadequate to rationalize the demand for health services or to provide for significantmaintenance expenditure. The thinking of some governments appeared to be that, whilepreventive and promotive health care should continue to be free of charge, there are a numberof service areas, particularly in secondary and tertiary curative care, where cost recovery couldbe made possible without affecting those unable to pay. In the case of Sri Lanka, theGovernment initiated studies to look into the mechanism for some form of cost recoverymeasures such as health insurance and prepayment schemes. However, given the welfareorientation of most governments in regard to social services and the traditional political climateon such a sensitive sector as health, any attempts at cost recovery require considerable effort.Yet the DMC governments need to consider alternative financing strategies because theincreasing demand for health services are not matched by increasing budget allocations. Toimprove cost recovery, unit costing of services must be developed. Such information wouldhelp the governments to decide on the methods of cost recovery.



17

C. Privatization of Health Services

70. An important approach for the successful development of the health sector is toinvolve the private sector. Private sector participation in the provision of health services was

encouraged by DMC governments. Although the projects themselves did not actively solicitprivate sector delivery of health care, some project components contributed indirectly to fullerprivate sector capability in the delivery of services such as the training of medical technologistsor nurses, many of whom will ultimately find employment in the private sector. In practice,Government support for private sector participation took the form, either tacitly or explicitly, ofallowing government doctors to engage in private practice outside of office hours and outsidethe health facilities in which they were working, with the hope that once established they wouldturn to private practice. The majority of the 3,948 Government doctors in Sri Lanka, forinstance, are presently engaged in part-time private practice.

71. Apart from the possible abuse of the privilege, there was the inherent conflict ofinterest as the same patients visited the doctors both during and outside office hours. Giventhat the provision of curative services is more lucrative than the provisi.on of preventiveservices, doctors invariably focused on the former. This situation is not conducive to theintegration of preventive and curative services, as doctors are more inclined to devote attentionto hospital work (curative service) at the expense of preventive work, which also involves fieldvisits. Hospital work, after all, serves to keep the doctors in touch with the types of servicesrequired in private practice.

D. Project Design and Implementation

72. A frequent observation of health providers in the DMCs is the apparent lack oflinkage between the Bank's projects. While the first project in Bangladesh, for example, was

focused on the provision of drugs production, the second was concerned with establishment ofhealth centers. Similarly, in Sri Lanka, the successful component of the first project was notpart of the design of the second project. Realizing the importance of sustainability, it has beensuggested that in future the Bank should design projects that build on the successfulcomponents of earlier ones. In effect this means that projects must be designed andimplemented keeping in mind not only lessons learned from earlier interventions, but the needto reinforce the positive impacts of early interventions. While there are constraints in designingsuch integrative projects, given the manner in which projects are proposed to the Bank forassistance, the resolution of this issue merits consideration if the impact of the Bank'sassistance is to be sustained.

73. Lessened impact from cancellation, non-use, and implementation delay of

project components could have been minimized with more thorough assessment of theconstraints facing the sector. Unopened X-ray equipment provided under the Second Healthand Population Project to Pakistan, for instance, resulted from the lack of technical staff to setup and operate the equipment. Only 20 percent of funds for in-service training of public healthmidwives and inspectors under the first Health and Population Project in Sri Lanka wereutilized because similar training was provided by other funding agencies. Other inadequaciesinclude the absence of impact indicators and relevant baseline data at project preparation,poor monitoring at implementation, faulty design due to inadequate preparatory work, andminimal involvement of health care providers and beneficiaries throughout the project cycle-



18

which in the case of Sri Lanka resulted in the non-use of some GHCs because of poorlocation. These constraints indicate the need for an in-depth analysis of the sectoral andinstitutional issues prior to project formulation.

E. Benefit Monitoring and Evaluation

74. The importance of setting up benefit monitoring and evaluation (BME) systemswas recognized by both the Bank and the DMC governments in all these projects. The issue iswhether the setting up of the system was vigorously pursued. BME components were often notimplemented for various reasons, such as when difficulties prevented the recruitment ofconsultants in the Sri Lankan project, or when appropriate counterpart staff were not assigned.Another problem was the lack of performance indicators by which to assess the effectivenessof the design or the impact of the projects.

75. The installation of an appropriate BME system would have improved theefficiency of health care delivery systems. In particular, it would have helped to monitor

benefits both during and after project implementation, and also to ensure that relevantbenchmark data by which to quantify the impact of project intervention would be compiled. Theexistence of viable BME systems would certainly have alleviated the data collection problemsoften encountered in postproject assessments caused by transfers and unavailability ofrelevant government officers intimately connected with the projects. Also, BME data wouldhave provided essential data for the proper design and implementation of follow-upinterventions by the Bank.

VI. CONCLUSIONS AND RECOMMENDATIONS

76. Although health improvement in the DMCs is fully attributable not only to Bankprojects but to factors such as general income growth and improvement in education, it wouldbe incorrect to deny that, overall, the Bank's assistance in the H&P sector has contributedpositively to the health of the population in the DMCs. The veracity of this statement is basedon the various health indicators as well as the perception of the beneficiaries. Similarly, withregard to the reduction in the population growth rate, Bank assistance has contributedpositively to increased family planning acceptance and dissemination of family planning adviceand information. Other significant contributions of the projects are the institutional developmentand capacity-building impact from the establishment of health-related facilities and theimplementation of training programs, though instances of poor maintenance and cancellationof fellowships somewhat detracted from potentially greater impact.

77. In terms of the impact of the Bank's assistance on policy adjustment and

Government expenditure, the result was somewhat nebulous. Only in Sri Lanka, where theproject helped the Government make the necessary policy changes to focus more on thedelivery of health services through the most peripheral health centers (GHCs), was the impactof the project on Government policy apparent. In the other three countries, while it wasexpected that health projects would add to the demand for greater expenditure in absoluteterms, relative allocation for health expenditure declined. The social impact of the projectswould be significant, except that no performance indicators can be formulated based on thedata available to measure the impact in quantitative terms. Environmental impact of theprojects was insignificant on the whole, though if viewed on a project basis, as in the case of



19

the PNG projects with water sanitation components, the impact could be viewed as distinctlypositive.

78. Based on the findings of the Study, the following recommendations are offeredfor consideration.

1. Recommendations for the Bank

(i) Especially in a social sector like H&P, in-depth sector, intersectoral, andmacroeconomic analysis is crucial when designing and formulating aproject for which specific measures including policy changes andinstitutional improvements would be necessary to ensure the long termsustainability of benefits.

(ii) Appropriate provisions for future impact evaluation, particularly with theformulation of performance indicators in a logical framework, should beconsidered from the start of project preparation. Clear definitions ofanticipated health and related impacts should be formulated, and therelevant impact indicators and benchmark data should be identified forfuture use.

(iii) The Bank should consider providing follow-up assistance, if needed, toprojects with successful components upon which it can build, rather thaninitiating new projects. It will help to reinforce the positive impact fromcompleted projects, particularly when full benefits are not immediatelyrealizable, because financial and other problems continue to affectproject sustainability.

(iv) The Bank should encourage full commitment of the executing agenciesin providing the necessary support, such as provision of required humanresources, training, and maintenance of facilities, so that project benefitsare maximized and sustained over the long term.

2. Recommendations for the DMC Governments

(i) The DMC governments should ensure that adequate specific budgetaryallocations are made for the repair and maintenance of buildings,equipment, and vehicles provided by projects so that all potentialbenefits are enjoyed by the beneficiaries. Regular in-service training of

personnel in project health centers should be conducted to develop theirtechnical and managerial skills.

(ii) DMC governments should ensure that appropriately trained and qualifiedpersonnel are appointed early to staff the project implementation units.Incentives could be given to the staff to implement projects smoothly andefficiently.



20

(iii) DMC governments should ensure that appropriately qualified staff aremade available for project components such as fellowships and otherforms of training; otherwise, they should consider partial loancancellation rather than incurring further debt.

3. Recommendations for the Bank and the Government

(i) Project formulation and design should always be done with the fullparncipation not only of key government officials at the Central level, butof regional health administrators and care providers. Wherever possible,targeted beneficiaries should also be involved. This would create asense of ownership among all concerned.

(ii) Project implementation based on monitorable performance indicators at

different levels of the project cycle should be conducted moreextensively and regularly so that problems can be identified early andcorrective measures taken immediately. This monitoring exercise shouldbe done not only in conjunction with the key officials in the ministries butwith regional health administrators, health care providers, and, whereverpossible, the beneficiaries. Project implementation units composed ofhealth and other officials should be established right at the beginning ofproject implementation.

(iii) All avenues of community involvement and participation should beexplored during project implementation. Construction of buildings at lowcost through community organizations, and donation of land and

maintenance of facilities by the beneficiaries, are examples of possiblecommunity participation. Active beneficiary involvement would give thecommunity a sense of self-reliance and responsibility toward thesustainability of the project.



21

APPENDIXES

Number Title Page Cited on(page, para.)

1 Bank-assisted Health and Population Projectsin Bangladesh, Pakistan, Papua New Guinea,and Sri Lanka 22 1 , 5

2 Beneficiary Survey Methodology 23 1 , 6

3 Social Development and Health Indicators 25 3 ,14

4 Major Components of Projects Includedin the Impact Evaluation Study 27 6 ,29

5 Reasons for Implementation Delaysand Cost Divergence 32 6 , 30

6 Health and Demographic Indicatorsin Selected Districts 35 8 ,3 6

7 Prenatal Care Indicators 37 10,44



22

Appendix 1

BANK-ASSISTED HEALTH AND POPULATION PROJECTSIN BANGLADESH, PAKISTAN, PAPUA NEW GUINEA, AND SRI LANKA a

Country/Loan No. Project Title Amount($ million) Approved

Bangladesh

Loan No. 504-BAN(SF): Public Health Program 15.6 Dec 1980Loan No. 672-BAN(SF): Health and Family Planning

Services 27.5 Dec 1983Loan No. 1074-BAN(SF): Second Health and Family

Planning Services 51.0 Jan 1991

Pakistan

Loan No. 562-PAK(SF): Health and Population 15.0 Dec 1981Loan No. 710-PAK(SF): Second Health and Population 16.0 Nov 1984Loan No. 850-PAK(SF): Third Health 30.4 Oct 1987Loan No. 1200-PAK(SF): Health Care Development 60.0 Dec 1992Loan No. 1277-PAK(SF): Population 25.0 Dec 1993

Papua New Guinea

Loan No. 586-PNG: Rural Health Services 12.0 Sep 1982Loan No. 747-PNG: Second Rural Health Services 5.4 Oct 1985Loan No. 746-PNG(SF): Second Rural Health Services 8.5 Oct 1985Loan No. 1054-PNG(SF): Special Interventions 10.5 Nov 1990Loan No. 1097-PNG(SF): Third Rural Health Services 21.0 Sep 1991Loan No. 1225-PNG(SF): Population and Family Planning 7.11 Apr 1993

Sri Lanka

Loan No. 516-SRI(SF): Health and Population 9.3 Jul1982Loan No. 1189-SRI(SF): Second Health and Population 26.1 Nov 1992

a Completed projects which form the basis for the Study are given in bold.



23 Appendix 2, page 1

BENEFICIARY SURVEY METHODOLOGY

A. Introduction

1. Beneficiary surveys in Bangladesh, Pakistan, Papua New Guinea (PNG), andSri Lanka were implemented under TA 5629: Regional Study of the Impact of Bank Assistanceto the Health and Population Sector. The surveys were designed to gauge the perceptions ofthe beneficiaries of the impact of Bank-assisted projects. Tabulated results of these surveysare shown in various consultants' reports available on file.

B. Constraints

2. Given the absence in most cases of an updated and reliable sampling frame,the limited financial resources, and the time constraint, it was impracticable in most cases toundertake a statistically sound survey based on probability sampling techniques. Additionally,the likelihood existed of introducing nonsampling errors due to the lack of time to build up a

cadre of well-trained enumerators.

C. Survey Design

3. A purposive survey approach along the lines of rapid appraisal was adoptedwith an anecdotal reporting style as in the case of PNG. The concern for a relatively higherdegree of accuracy is less stringent in this type of survey because the information required isless numerate in character. Conclusions derive mainly from impressions and value judgmentsof the respondents.

4. Based on the Inception Report prepared by the international consultants, andagreed to by the Bank, the basic frame of questionnaires, including data to be collected, wasprepared by the international consultants for the domestic consultants in three of the fourcountries. The exception was PNG where the survey was undertaken after the engagementperiod of the international consultants. The domestically recruited consultants, nevertheless,adapted the questionnaire prepared by the international consultants to suit PNG conditions.

D. Sample Size

5. In Bangladesh, a total sample of 700 beneficiaries from 26 selected locationswere interviewed. Of these, 692 valid questionnaires were used. They consist of 150respondents from six union health family welfare centers, 155 from six rural dispensaries, 203from six thana health complexes, 80 from two district hospitals, 20 from one specialized

hospital, and 72 from two thana hospitals.

6. In Pakistan, two beneficiary surveys were undertaken because the SecondHealth and Population Project was included in the Study at a later stage. For the first Healthand Population Project, 11 district headquarters hospitals, 11 tehsil headquarters hospitals,and 3 reproductive health services hospitals were selected. The number of respondents is notindicated in the consultants' report. For the Second Health and Population Project, a total of340 respondents were interviewed: 205 from 22 basic health units, 65 from 4 regional traininginstitutes, and 70 from 5 nursing schools. The sample of beneficiaries comprise people who




receive services from their present closest health outlets, trainers, service providers, andmanagers in various institutions that benefited from the improved facilities in the institutions.

7. In PNG, the beneficiary survey was conducted through health services

personnel. Because of the high number of non-respondees, the final sample was comprised ofonly 134 respondents from the community.

8. In Sri Lanka, the sample size consisted of 831 respondents: 103 from 4divisional health centers, 258 from 11 subdivisional health centers, and 470 from 30Gramodaya Health Centers.



25

Appendix 3, page 1

SOCIAL DEVELOPMENT AND HEALTH INDICATORS

Table 1: Selected Social Development Indicators inBangladesh, Pakistan, Papua New Guinea, and Sri Lanka

Bangladesh Pakistan Papua New Guinea Sri Lanka

Mid-year Population (million)1980 88.5 82.6 3.0 14.81985 97.5 96.2 3.3 15.81990 106.8 112.1 3.7 17.01995 116.9 129.8 4.1 18"0

Average Annual Population GrowthRate, 1990-1995 (%) 1.8 3.0 2.0 1.2

Rate of Natural Increase, 1990-1995 (%) 2.4 3.2 2.3 1.5

Population in Poverty,Average: 1980-1991 (%) 78 28 73 39

Average Life Expectancy, 1993 (yrs) 56 62 55 72Male 56 61 56 70Female 56 63 57 74

Adult Literacy Rate, 1995 (%) 38 37 72 90Male 49 50 81 93Female 26 24 63 87

Population with Access to SafeWater, Average: 1988-1993 (%)

Urban 82 85 94 100 a

Rural 85 50 20 64

Population with Access toSanitation, Average: 1988-1993 (%)

Urban 63 60 57 73 a

Rural 26 17 10 56

Average for years 1988-1991.Source: ADB Key Indicators, 1995 and 1996.



26

Appendix 3, page 2

Table 2: Selected Health Indicators inBangladesh, Pakistan, Papua New Guinea, and Sri Lanka

Bangladesh Pakistan Papua New Guinea Sri Lanka1985 1993 1985 1993 1985 1993 1985 1993

Crude Birth Rate (per 1,000) 40 35 44 40 37 33 25 20

Crude Death Rate (per 1,000) 15 11 15 9 13 11 6 6

Infant Mortality Rate (per 1,000) 128 106 120 88 65 67 30 17

No. of Persons per Physician 6,703 5,143 3,153 1,918 12,416 10,083 8,276 4,745

No. of Persons per Hospital Bed 3,638 3,265 1,695 1,548 208 297 358 360

No. of Births per Woman 5.7 4.3 6.1 6.1 5.4 5.0 3.2 2.4

Contraceptive Prevalence Rate 25 40 11 14 4 n.a. 62 62(% of Women, 15-49 years)

n.a. = not available.Source: ADB Key Indicators, 1996.



27

Appendix 4, page 1

MAJOR COMPONENTS OF PROJECTS INCLUDEDIN THE IMPACT EVALUATION STUDY

BANGLADESH

A. Public Health Program-Loan No. 504-BAN(SF)

Part I: Upgrading of the Pharmaceutical Production Unit

Conversion of the Government's Pharmaceutical Production Unit into anautonomous drug manufacturing corporation.

Part II: Malaria Control

1. Importation of raw materials (benzene) for production of DDT.2. Rehabilitation of the DDT factory.

B. Health and Family Planning Services Project-Loan No. 672-BAN(SF)

Part I: Health Centers/Hospitals (in four selected districts:Mymensingh, Comilla, and Jessore).

Dinajpur,

1. Establishment of 150 Union Health Family Welfare Centers.2. Upgrading of 150 rural dispensaries.3. Strengthening of 70 Upazila Health Centers.4. Strengthening of 10 district and subdivisional hospitals.

Part II: Medical Stores and Workshops

1. Upgrading of the Central Medical Store.2. Strengthening of the Central Stores of the Population Control and Family

Planning Wing.3. Strengthening of the National Electro-Medical Equipment Workshop.4. Establishing 19 district warehouses and workshops for repair and maintenance

of medical equipment.5. Strengthening of workshops for repair and maintenance of medical equipment in

seven medical colleges and two specialized institutes.

Part III: Strengthening of Project Management

1. Provision of equipment, vehicles, additional staff, and consultant services toassist the Construction Management Cell and the Building, Planning and DesignUnit of the Ministry of Health and Family Welfare in implementing civil worksconstruction and upgradation.

2. Provision of consulting services to assist the Ministry of Health and FamilyWelfare in the management of medical supplies.



Appendix 4, page 2

28

PAKISTAN

A. Health and Population Project-Loan No. 562-PAK(SF)

Part I: Paramedical Training Schools, Hospitals, and Workshops

1. Establishment and equipping of three new paramedical training schools atJhelum, Sialkot and Sahiwal.

2. Provision of essential equipment and supplies to 12 district and 10 tehsilheadquarters hospitals.

3. Establishment and equipping of one central and two subsidiary workshops formaintenance and repair of medical equipment.

Part II: National Institute of Health

Establishment of a two-year training course at the National Institute of Health formedical laboratory technologists to prepare them as trainers at the paramedicaltraining schools.

Part III: Reproductive Health Services

Provision of equipment, vehicles, and supplies for four reproductive healthservices centers in support of family planning activities.

B. Second Health and Population Project + Loan No. 710-PAK(SF)

Part I: Basic Health Units, Electro-Medical Workshops, and Training ofNurses and Paramedical Staff

1. Establishment and equipping of 70 basic health units in Sindh Province.2. Establishment and equipping of one medical equipment workshop at Jamshoro.3. Establishment of a training course for biomedical technicians and provision of

fellowships for instructors at Jamia Millia Polytechnic, Karachi.4. Construction/upgrading and equipping of 10 nurses training schools.5. Construction and equipping of one paramedical training institute at Jamshoro.

Part II: Population Welfare Program

1. Establishment and equipping of five regional training institutes for familyplanning and maternal and child health (MCH) workers.

2. Provision of training in population services for health workers under theDepartments of Health, Education, and Local Government in Sindh.



29

Appendix 4, page 3

Part III: Federal Nursing Program/Management of Health Services at Ministry

of Health

1. Establishment of a Nursing Curriculum and Planning Cell.2. Provision of equipment and training materials to colleges and schools of

nursing.3. Consultancy Services for establishment of Health Services Administration

Program.

PAPUA NEWGUINEA

A. Rural Health Services Project-Loan No. 586-PNG(Central, East Sepik, Gulf, Madang, West Sepik, and Western Highlands provinces)

I. Rural Health Services

1. Construction, renovation, and equipping of health centers, subcenters,and aidposts.

2. Construction of staff housing.3. Upgrading of rural health facilities through provision of medical, dental,

and laboratory equipment.4. Improvement of transport and communication facilities.5. Provision of education materials and equipment for management

improvement.

II. Rural Water Supply and Sanitation Facilities

1. Construction of potable water supply and waste disposal facilities atselected health centers, schools, and villages.

2. Provision of transport facilities.

III. National Capital District Health Clinics

Construction and renovation of health clinics in the National CapitalDistrict.

IV. Consultant Services and Fellowships

1. Provision of consultant services in the fields of management, healthtraining, water and sanitation, building design and maintenance, andproject implementation.

2. Provision of fellowships in management training, health education, andvillage midwifery.



30

Appendix 4, page 4

B. Second Rural Health Services Project-Loan Nos. 746-PNG(SF) and 747-PNG

Part I: Improvements to Rural Health Services in the Primary Project AreaComprising East New Britain, Eastern Highlands, Milne Bay, Morobe,Simbu, Southern Highlands, West New Britain, and Western (SouthFly District only) Provinces

1. Construction/renovation/upgrading of (i) health centers, subcenters, aidposts,and staff housing, and (ii) rural water supply and sanitation facilities.

2. Procurement and distribution of pharmaceuticals,

Part II: Improvements to Rural Health Services in the Secondary ProjectArea Ccmprislnq Enga, Manus, New Ireland, North Solomons, Oro,

and Western (South Fly district only) Provinces

1. Construction/renovation/upgrading of (i) health centers, subcenters, aidposts,and staff housing, and (ii) rural water supply and sanitation facilities.

2. Procurement and distribution of pharmaceuticals.

Part III: National Health Programs on Human Resource Planning, HealthWorker Education and Training, and a Study on Rural HealthServices Costs

Consultancy services, fellowships, and training.

SRI LANKA

A. Health and Population Project-Loan No. 576-SRI(SF)

Part I: Health Centers

1. Construction of 400 new Gramodaya Health Centers (GHCs).2. Upgrading of 75 existing Subdivisional Health Centers (SDHCs).3. Upgrading of 30 existing Divisional Health Centers (DHCs).4. Construction of suitable residential accommodation for medical and paramedical

staff of the GHCs, SDHCs, and DHCs.

5. Provision of medical equipment, basic office furniture, and vehicles at each ofthe GHCs, SDHCs, and DHCs.

Part II: In-Service Training

Provision of financial support, teaching aids, and materials for in-service trainingin general public health and family planning for:

(i) Assistant Medical Practitioners.



31

Appendix 4, page 5

(ii) divisional health officers, public health nurses, family health workers, and

supervisory family health workers and their trainers.

Part III: Consulting Services

Provision of consultant services to assist the Ministry of Health in supervisingProject implementation and to train and assist personnel in monitoring andevaluation Project benefits.



REASONS FOR IMPLEMENTATION DELAYS AND COST DIVERGENCETable 1: Reasons for Time Overrun/Underrun for Projects Under Health and Populat ion with PPAR

Project Number Project Title Approval/Actual DelayLoan Project Completion Years Percen- Major Causes of Delay

Effectiveness tage

F irst inSector

13053-00-BAN Public Health 24 Mar 81 31 Dec 83 3.27 117.8 (i) Delay of loan effectiveness because of theProgram 23 Jun 81 06 Apr 87 covenanted subsidiary loan agreement between

Bangladesh Chemical Indust ries Corporation andCCC could not be concluded on t ime( ii ) lengthy discuss ions within the Government onthe stipulated reorganization of PPU/EDCL(ii i) the t ime-consuming reorganiza tion process ofMinist ry of Health and Family Planning (formerlyMinistry of Health and Population Control),

particul arly r egarding the integration of malar iacontrol activities into general health services(iv) problems of counterpart funding because ofbudgetary constraints(v) coordinat ion problems among var ious par tiesinvolved in the overall program implementation(vi) pervasive procurement problems partlyexplained by lack of familiarity of counterpartauthorities with Bank procedures and guidelines onprocurement

14051-00-PAK Health andPopulation

23 Mar 8227 May 82

30 Jun 8630 Jun 90

4.00 93.7 Lengthy procurement procedures

15003-00-BAN Health andFamily PlanningServices Project

4.25 123.9 (i) Executing Agency's lack of experience inimplement ing Bank projec ts , associated with poormanagement and lengthy Governmentdecisionmaking procedures(ii) lack of t imely avai labi li ty of local counterpar t

funds( ii i) problems of land acquis it ion, pol itica l unres t,and floods in 1987 and 1988 (when major civilworks were in progress)(iv) cancellation of the technical assistancefinanced by the United Nations DevelopmentProgramme

25 Apr 8404 Jan 85

30 Sep 8731 Dec 91

15045-00-PNG 1.00 20.4 Implementati on probl ems during the initial phaseof the Project ar ising from the decentralization offinancial and administrative functions from theCentral Government to the provincialgovernments. Major areas contributing to the delaywere(i) problems concerning civil works andrecruitment of key long-term consultants;

Rural HealthServices Project

03 Feb 8314 Jan 83

31 Dec 8731 Dec 88

y

N

N

y



, '

Project Number Project Title Approval/ActualLoan Project Completion

Effectiveness

DelayYears Percen- Major Causes of Delay

tage

F irst inSector

6024-00-SRI

6157-00-PNG

Health andPopulationProject

Second RuralHealth ServicesProject

03 Nov 8220 Oct 82

30 Jun 8731 Mar 90

2.75

0.95

(ii) siting, commissioning, and other technicald ifficul ties experienced under the communicationscomponents;(iii) the institutional deficiency of the provincial

authorities in implementing the rural water supply andsanitation component.

59.1 (i) MOH's lack of awareness of Bank procedur es, thecomplexity of Project design(ii) inadequate analysis of Government administrativeprocedures during Project preparation(iii) inadequate implementation arrangements(iv) deteriorating security situation

21.1 The Project was completed without any formalextension of the loan closing date (in spite of anadditional component).

Y

N3 Jul 8620 Jun 86

31 Dec 9012 Dec 91

PPAR = Project Performance Audit Report. CCC = Chittagong Chemical Complex.PPU/EDCL = Pharmaceutical Production Unit/Essential Drugs Co. Ltd. MOH = Ministry of Health



Table 2: Reasons for Cost Divergence for Projects Under Health and Population with PPAR

ProjectNumber Project Title Appraisal

Pro jec t Cos t (US $ '000)

OverrunlUnderrun

(%)ctual Comments Explaining Project Cost Divergence

13053-00-BAN (27.0) The lower-than-expected Program expendituresmainly stemmed from the Bank's decision not tofinance certain subcomponents such ashandsprayers, microscopes, and part of the DDTsupplies originally included in the Program.

14051-00-PAK

15003-00-BAN

15045-00-PNG

16024-00-SRI

16157-00-PNG

Public Health Program

Health and Population

Health and Family PlanningServices Project

Rural Health Services Project

Health and Population Project

Second Rural Health ServicesProject

20,660

22,830

34,510

20,825

12,210

19,956

15,080

13,780

33,340

13,884

8,140

20,889

(39.6) The cost of the completed Project was considerablybelow the Appra isal es timate . The reductions in thecost were due to: (i) devaluation of the Pakistanirupee; ( ii ) nonpayment of land acquisi tion costs forthe paramedica l schools because the schools werebui lt on avai lable land; and ( ii i) non-ut il ization of theOnly a four percent underrun on total actual Projec tcosts was r ecorded. The actual Project cost wouldhave been more if several Proj ect components hadnot been dropped.

(3.4)

(33.3) In dollar terms, the Project incurred a total costunderrun pr imar ily resul ting f rom the deprec ia tionof the kina against the dollar; the lower-than-est imated cos ts for vehicles , equipment , and long-term consultants; and the high provision forcontingencies.

(33.3) The reduction in Project cost was due to thedepreciation of tM rUf)aa against tha QoUar;

cancellation of major portions of the in-servicetraining and consultant components; reduction inthe physical facilities program; and lower-than-estimated costs of civil works, equipment, andvehicles.

4.7 Originally, the Project had a loan savingssubsequently reall ocated to the pharmaceuticalcomponent. The savings were due to theweakening of the dollar against the SDR, andunderutilization of some funds allocated for .theProject components on human resources andt raining and on the secondary Projec t area. Large lybecause of the pharmaceutical component, theProject had a cos t overrun of about f ive percent.

PPAR - Project Performance Audit Report. DDT - dichlorodiphenyltrichl oroethane. SDR - Special Drawing Rights.

. '.




HEALTH AND DEMOGRAPHIC INDICATORS

Table 1. Sri Lanka - Health and Demographic Indicatorsin Selected Districts

Indicators

Districts YearCrude Birth Crude Death Infant Mortality Population

Rate Rate Rate Growth Rate

(per thousand population)(per thousand (Natural) (%)live births)

N u w a r a E l i y a 1988 n . a . a n . a . 27.4 n . a .

1989 27.8 9.1 n . a . 1.91990 n . a . n . a . 33.8 n . a .

1991 26.3 n . a . 27.7 1.91992 26.3 7.6 n . a . 1.8

R a t n a p u r a 1988 n . a . n . a . 22.6 n . a .

1989 24.9 5.5 n . a . 1.91990 n . a . n . a . 29.3 n . a .

1991 28.0 n . a . 22.7 1.81992 19.5 5.1 n . a . 1.4

K e g a l l e 1988 n . a . n . a . 14.1 n . a .

1989 16.4 6.1 n . a . 1.01990 n . a . n . a . 17.7 n . a .

1991 14.9 n . a . 14.6 1.01992 13.4 4.9 n . a . 0.9

N a t i o n a l 1988 n . a . n . a . 20.2 n . a .1989 21.8 6.2 n . a . 1.31990 n . a . n . a . 19.3 n . a .1991 21 n . a . 17.2 1.51992 20.1 5.7 n . a . 1.4

a n . a . - n o t a v a i l ab l e .

S o ur c e : P o pu l a t i on D iv i s i on , M in i s t ry o f H ea l th .




Table 2. Bangladesh - Health and lOemographic Indicatorsin Selected Districts

Indicators

Districts YearCrude Birth Crude Death Infant Mortality Maternal Population

Rate Rate Rate Mortality Rate Growth Rate(per thousand population) (per thou!and live births) (Natural) (%)

Mymensingh 1990 32.8 11.4 94 4.8 2.151991 31.6 11.2 92 4.7 2.061992 30.8 11.0 88 4.7 1.981993 28.4 9.2 84 4.5 1.82

Chandpur 1990 32.8 11.3 94 n.a. n.a.1991 31.6 11.2 92 n.a. 2.121992 30.8 11.0 88 n.a. n.a.1993 n.a. a n.a. n.a. n.a. n.a.

National 1990 32.8 11.4 94 4.78 2.141991 31.6 11.2 92 4.72 2.041992 30.8 11.0 88 4.68 1.981993 28.8 10.0 84 4.52 1.92

a n.a. = not available.Source: Bangladesh Bureau of Statistics, 1994.




PRENATAL CARE INDICATORS

Table 1 . Sri Lanka-Performance and Quality of Careat Gramodaya Health Centers (GHCS). 1986-1994

Indicator 1986 1988 1990 1991 1992 1993 1994

A. Performance

1. Population served by one GHCA. Below 1 year 57.0 65.0 58.0 66.0 54.0 53.0 52.0

B. 1-5 years 218.0 263.0 248.0 277.0 273.0 288.0 244.0

C. Total population 2655.0 2888.0 3059.0 3336.0 2922.0 2873.0 2870.0

D. Eligible families 374.0 402.0 413.0 476.0 441.0 422.0 429.0E. Registered mothers 68.0 76.0 67.0 79.0 68.0 59.0 57.0

2. Home visitsA. Average prenatalvisits per registeredmother 5.3 4.4 4.5 4.5 4.6 4.6 4.7

B. Average postnatalvisits per mother 2.6 2.8 2.2 2.2 2.3 2.1 2.1

C. Average visi ts per infant 10.5 8.3 9.5 9.0 12.1 10.0 10.6D. Average visi ts per child 4.9 2.4 4.8 2.9 3.4 3.0 3.9E. Average visits for

family planningper eligible couple 0.3 0.2 0.3 0.3 0.3 0.4 0.4

3. Clinical visitsA. Average visits per

registered mother 4.1 3.8 4.0 3.3 4.2 4.5 4.3

B. Average visits per infant 3.8 4.8 4.7 5.3 7.1 6.4 5.94. Other

Total no. of deliveriesattended by the publichealth midwife (PHM) 7.0 19.0 31.0 39.0 26.0 14.0 26.0

B. Quality of Care1. Maternity Care

A. Percentage of mothersregistered by the PHMbefore 4 months 67.3 69.3 68.8 69.2 70.4 75.1 77.7

B. Percentage of mothersreceiving tetanus toxoid 34.8 48.9 42.9 60.7 61.5 70.7 71.6

C. Percentage of deliveriesattended by trainedpersonnel 96.9 97.2 98.1 98.9 98.6 97.6 99.5

D. Number of deliveriesattended by nontrainedpersonnel 8.0 18.0 15.0 20.0 13.0 11.0 5.0

2. OtherA. No. of diarrheal cases

attended 175.0 110.0 247.0 182.0 207.0 187.0 227.0B. Percentage of diarrheal cases

referred b~ the PHM 45.7 36.4 42.9 37.4 42.0 55.6 46.7Source: Consultant's Country Report.




Table 2: Bangladesh-Service Provision Indicatorsin Selected Districts, 1993 and 1994

(Percentages)

DistrictsType of Service Year Rajshahi Je$sore Mymensingh National

1. Pregnant women receivingprenatal care by trained 1993 35.0 28.0 73.3 29.0personnel 1994 40.0 32.0 72.7 27.7

2. Deliveries attended by trained 1993 44.0 40.0 43.3 n.a.personnel 1994 27.0 52.0 38.6 n.a.

3. Coverage of infants fullyimmunized

(i) BCG 1993 98.0 84.0 101.0 96.01994 107.0 88.6 115.0 95.0

(ii) DPT3 1993 97.0 84.0 100.0 88.01994 112.0 87.7 108.0 74.0

(iii) Measles 1993 96.0 80.0 71.0 88.01994 112.0 91.6 96.0 86.0

BCG=bacillus Calmette-Guerin,DPT3 = diptheria pertusis tetanus 3.

Note: Percentages over 100 are due to more children getting immunized than the numbers registered (with health workers)

which are used to calculate percentages

Source: Consultant's Country Report.

Impact Evaluation Study of the Bank Assistance in the Health and Population Sector in Bangladesh, Pakistan, Papua New Guinea, and Sri Lanka

Documents