PUB7291 A WORLD BANK COUNTRY STUDY Indonesia Health Planning and Budgeting The World Bank Washington, D.C. Copyright e 1991 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing January 1991 World Bank Country' Studies are among the many reports originally prepared for internal use as part of the continuing analysis by the Bank of the economic and related conditions of its developing member countries and of its dialogues with the governments. Some of the reports are published in this series with the least possible delay for the use of governments and the academic, business and financial, and development communities. The typescript of this paper therefore has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. Any maps that accompany the text have been prepared solely for the convenience of readers; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country', territory, city, or area or of tile clL&:horities thereof or concerning the delimitation of its boundaries or its national affiliation. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to Director, Publications Department, at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to photocopy portions for classroom use is not required, though notification of such use having been made will be appreciated. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list (with full ordering information) and indexes of subjects, authors, and countries and regions. The latest edition is available free of charge from the Publications Sales Unit, Department F, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'Iena, 75116 Paris, France. ISSN: 0253-2123 Library of Congress Cataloging-in-Publication Data Indonesia: health planning and budgeting. p. cm. "Report no. 7291-IND." "February 28. 1989." "Country Department V. Asia Region," ISBN 0-8213-1609-5 1. Health planning--Indonesia. 2. Health planning--Economic aspects--Indonesia. 3. Health planning--Indonesia--Finance. I. ~orld Bank. II. International Bank for Reconstruction and Development. RA395.I55I53 1990 362.1'09598--dc20 IN 90-12736 '1J ;,1"' !',FO~'i"bJmON CErHER I,,'!, ,.,,~, .... I.r !. YI CIP (PIC) I Ace NO : ................. ~Pr.,............... 0) b"f CLI\SS : / ................ . I . .. '""KED ' , " j ' . , u ··· _n~_~ ..········..···..·ND ' j . . T~ .. :::::::::::CJf ":cFi aimed at improving the quality of Class C and D hospitals which serve the majority of the population. At the same time, overall levels of recurrent expenditure have at best been held constant in real terms in the face of massive underfunding of the operations and maintenance requirements of exist.ing capacity. However, this appears to have been achieved at the cost: ~. an erosion in spending on non-personnel expenditures. These deficiencies i'e O&M funding are exacerbated by an inappropriate allocation of resources - xiv favoring tertiary hospitals at the expense of lower level facilities. This is exemplified in a pattern of overstaffing of Class A and B hospitals, particularly with doctors and specialists, coexisting with understaffing in the Class C and D facilities. Third, there is an acute shortage of necessary planning and budgeting information, reflected in the difficulty of identifying and consolidating data on budgetary expenditures on hospitals, and in the lack of clear estimates or norms for staffing and other recurrent expenditure requirements in different types of hospital. (viii) Given limited budgetary resources, the Government's main focus in REPELITA V should be on achieving more efficient use of available resources in the hospital sector to help raise the quality and utilization of existing physical infrastructure. The main elements should include the following measures. On the recurrent expenditure side: (a) overall budgetary resources for O&M should be substantially increased by continuing the policy of tight restraint in public investment on hospitals. Incremental O&M requirements are estimated at around Rp.208 billion per year, compared to recurrent expenditures in 1985/86 of about Rp.173 billion; (b) the allocation of existing staff (particularly doctors in the four basic specialty areas of surgery, obstetrics, pediatrics and internal medicine) should be redistributed from the overstaffed tertiary hospitals to the understaffed Class C and D facilities, including payment of supplementary incentives as appropriate; (c) complementary emphasis should be placed on supporting specialist doctors in lower level hospitals with improved staffing of the diagnostic services (radiography and laboratory technicians) and equipment repair and maintenance facilities, together with adequate resources to overcome the serious shortage of funds for spare parts and expendable items such as laboratory reagents. With regard to investment expenditure: (a) the level of investment in new expansion of hospital capacity should be limited by the affordability of incremental recurrent costs, taking into account the need to first meet the full O&M requirements of existing infrastructure which are estimated to total around Rp.349 billion per year; (b) to facilitate this restraint, rigorous project evaluation procedures for review of hospital investment proposals should be instituted, including a realistic projection of anticipated O&M requirements together with staffing needs; and (c) overriding priority should be given to investment spending which raises the quality and impact of the existing hospital stock, for example the rehabilitation and upgrading program for the Class C and D hospitals, and in referral support facilities including laboratories and equipment repair and maintenance workshops. (ix) This focus on improved efficiency of public resource management in the hospital subsector should be supported by strong efforts to improve the information base for planning and budgeting. These should include initiatives to: (a) as part of a general effort to improve the classification of budgetary disbursements in the government accounts, to refine the classification of hospital spending, particularly to disentangle hospitals from other expenditures such as health centers and general administration which are often lumped together in the district routine budgets; (b) to develop a proper hospital accounting system to generate consolidated accounts for individual hospitals, and to include financial data as part of the routine hospital reporting system; and (c) to make detailed estimates of O&M requirements, including staffing for hospitals, to provide appropriate planning and budgeting norms for different hospital facilities. - xv Community and Preventive Health Services (x) Considerable progress has been made in expanding coverage with community hl~alth infrastructure through the INPRES grant program, and more recent:ly in enhancing village level outreach through the POSYANDU delivery system for preventive health services. However, utilization rates remain low in Indonesia. This reflects the low density and unequal distribution of facilities, which still leaves many rural residents beyond the effective range of th.:. health center system. This implies a long-term need for investment in addit:i.onal capacity to improve access to services in poor remote areas. However, there are also indications of substantial underfunding of present O&M requirements, implying low quality of existing services. This has been exacerbated by recent fiscal constraints. resulting in roughly constant recur:rent expenditure levels on O&M in the face of a growing demand for addit:i.onal O&M support created by rapid expansion of the posyandu network. Tenta~::ive estimates suggest that the order of magnitude of annual O&M requirements for the community health system might be as much as Rp.350 billion. compared to recurrent expenditure levels in 1985/86 of some Rp.140 billion. A similar pattern of underfunding the communicable disease control (CDC) programs has emerged from the recent adjustment experience. with large reductions in the already low share of sectoral spending on CDC implying a reallocation of expenditure priorities away from preventive health to curat:i.ve health services. Although some programs have been protected others. such as malaria and tuberculosis control. have sustained devastating cutbacks. This ~::rend is not consistent with the strong case for public sector invol'.ement in financing preventive interventions on externality grounds and should be reversed. (xi) Given the tight outlook for budgetary resources the main focus of ~)enditure priorities in the near term should be on substantially increasing O&M funding for the community and preventive health programs. With regard to community health. these requirements include: (a) increasing funds for nonsalary expenditures on transport to enable health center staff to travel to villages to provide support to the POSYANDU; and (b) increasing funds for maintenance of buildings. equipment and vehicles to help maximize the returns from past investments. and minimize the need for costly replacements. Turning to communicable disease control programs. there is an urgen":: need for the Ministry of Health to reassess the full requirements of a core program of major CDC activities. and to assign the core program a priority claim on sectoral resources which would be protected in the future. Incremental O&M requirements for CDC programs probably total around Rp.40 billion annually. Over the medium term, as additional resources become available. it may be possible to expand the presently low provision of commw1ity health infrastructure. but this should be contingent on the affordability of the additional O&M requirements created by new investment. taking into account the need to provide adequate O&M funding for existing serviee capacity. In setting investment priorities. consideration should be given to expanding the provision and role of subcenters which may prove to be a more cost-effective instrument than health centers in extending access to basic curative care and providing technical backup for preventive services delivered through village-level POSYANDU. - xvi (xii) Planning and budgeting for efficient resource use in community and preventive health services needs to be strengthened with significant improvements in the information base. In particular steps should be taken to: (a) develop institutional capacity to monitor levels and trends in utilization rates by income class of the different components of the community health system, in order to improve assessment of its effectiveness and distributional impact; (b) undertake research into the determinants of utilization differentials, including focus group methods to gain insights into client perceptions of the quality of community health services; (c) strengthen the budgetary system by refining the classification of health center expenditures in order to provide a baseline for evaluating the adequacy of overall expenditure levels; (d) to develop an integrated financial accounting system for health centers, and to incorporate financial data into the routine health center reporting system to help evaluate adequacy of spending in individual centers; and (e) develop detailed estimates of O&M requirements, including staffing, for individual community health services as a basis for setting budgeting norms for adequate funding of service capacity in the future. Health Manpower (xiii) Excess supply of paramedical manpower. During REPELITA IV, public paramedical schools graduated several thousand more paramedics than the government was able to absorb given fiscal constraints on hiring rates. In addition, graduates from private paramedical schools are also likely to greatly outnumber available openings in the private sector. The feasibility of various policy responses needs to be explored, including: (a) diversion of some pekarya kesehatan posts to paramedics; (b) selective school closings via enforced accreditation standards, especially in provinces with the highest levels of excess supply (with some of the cost-savings devoted to retraining or placement of affected students); (c) reabsorption of existing staff or new graduates into supplementary training programs. Among these, the pekarya issue is of particular importance. The Ministsry of Health (MOH) has devoted a significant part of its scarce new paramedical posts to these auxiliary paramedics, who receive only four months of training but are paid at the same scale as paramedics with three years of education. In theory, pekarya will greatly improve rural health services, since they are recruited from the neighborhood of under staffed health centers, but this theory needs to be verified. (xiv) Allocation of new posts. Via the INPRES program, MOH has achieved a relatively equitable distribution of new posts among provinces. Routine posts are less equitably distributed, with Jakarta taking a disproportionately large share. However, scattered evidence suggests that the intraprovincial distribution of posts deviates from goals of equity and efficiency, resulting in substantial mismatches between facility staffing and utilization. To some extent this may result from a pragmatic recognition of the difficulties of assigning staff to rural areas. It may also reflect political pressures on local administrators combined with a lack of staffing standards and staffing data. The new utilization-based staffing standards developed under the ISN system are addressed to the latter problem. An ISN like system is necessary in an era where the unpredictability of annual budgets requires flexible, prioritized manpower plans. It is important that the ISN system adopt a feasible system of data collection and be institutionalized in time to aid in planning. - xvii (xv) Placement and distribution of manpower. Although some provinces produce more paramedics than they hire, and others suffer from a deficit of paramedics, there is relatively little intraprovincial placement of paramedical graduates. Recognizing this problem, the MOH has instituted a quota system for allocating public paramedical school slots by province. The success of this strategy has yet to be assessed. Inadequate reliable information exists on the placement of new graduates within provinces, or on the filling of vacancies. It is widely thought that difficulties in placing paramedics in rural areas have led to understaffed health centers and overstaffed hospitals and health offices. Provision of doctors to rural areas and remote health centers is accomplished by rotating recent medical school graduates on two-year mandatory duty tours. It is therefore extremely difficult to augment the doctor/population ratio, or to maintain health center staffing in these areas. In eleven provinces, more than one-quarter of all health. centers are reported to lack a doctor. It is possible that provision of generous and innovative incentives for rural residence might help the situation. Such incentives might include, for instance, subsidized education for the doctors' children at high-quality boarding schools. Such incentives might be financed, in part, by recovering the cost of medical education from doctors who practice in urban areas. This would also serve to increase the relative attractiveness of the rural po~~~. (xvi) Worker producti_v:ty. Scattered evidence suggests that average productivity is low, but highly varied, throughout the health system. To a large extent, low productivity may be rooted in a failure of the personnel allocation process to match staff with existing workload. The new ISN staff planning system is directly addressing this problem. Low productivity may also stem from structural barriers to utilization, e.g., lack of specialized equipment, staff, or drugs, or inconvenient opening hours. Another problem lies in the lack of an incentive structure. With few exceptions, the system has lacked any penalties for lack of effort, or rewards for exceptional effort. Current MOH plans to link promotions in salary rank to work effort are a laudable step in the right direction. Further mechanisms for improving employee management should be explored. (xvii) The quantity/quality tradeoff. The conventional view is that there is an urgent need for additional paramedical staff. For this reason, REPELITA IV chose to emphasize rapid expansion of staff quantity, making quality improvement a long-term goal. From this viewpoint the recent necessity to reduce paramedical output is a regrettable, almost paradoxical, short-term expedient. It is appropriate, however, periodically to reexamine the quantity-quality tradeoff in light of changing conditions and experience. As noted above, there is evidence of areas of low labor productivity throughout the health sector. To the extent that the existing labor force is inefficiently utilized, one may question the necessity for adding additional staff. It may in fact be more productive to devote more resources to improving the quality and efficiency of existing staff, while slowing the rate of growth of the labor force. The recent period of retrenchment may offer an opportunity to redirect manpower policy in this direction. (xviii) Staffing information. The manpower information system is inadequate for the management of the Ministry of Health's most important single resource, accounting for over Rp.243 million in annual recurrent costs. The system: (a) does not cover about one-third of all employees working in - xviii MOH-supervised facilities; (b) provides no information on the facility or function to which a staff member is assigned. These shortcomings hobble basic planning and analysis, especially regarding the allocation and distribution of employees. The MOH recognizes these problems and is implementing an improved personnel information system. So far, a low response rate to baseline data collection has prevented operationalization of the system. It is important that this problem be diagnosed, and that the solutions be enthusiastically supported. Overlaps and redundancies between competing information systems should be minimized. Cost Recovery and Insurance (xix) With the provision of adequate health services threatened by the scarcity of fiscal resources it is appropriate to review options for increasing cost recovery directly from the beneficiaries of publicly provided services. Although Indonesia has adopted a policy of charging fees for services, the revenue yield is low with only 10% of total recurrent expenditure recovered through user charges. While cost recovery ratios are higher for hospitals, averaging about 20%, even these are low compared to the performance of some developing countries. In Thailand cost recovery in hospitals averages 40%; in China it ~~~rages about 80%. These figures suggest that there may be substantial scope for mobi.lizing additional resources by raising user charges. However, caution should be exercised before any major increase in charges is adopted in order to ensure that equity and efficiency objectives are also adequately reflected in the fee structure. It is important to note that striking an appropriate balance between the revenue generation, equity and efficiency objectives of pricing policy in the health sector will still entail a substantial role for government expenditure, particularly in the poorer regions. There also remains substantial scope for increasing the sector share of central government expenditure from its presently low level (about 2.5%) relative to Indonesia's international comparators (about 5%). (xx) The main elements of an appropriate policy on cost recovery for the health sector in REPELITA V include the following: (a) fees should be consistent with ability to pay in order to maintain utilization of necessary health services. For equity reasons it is desirable in principle to charge higher prices to the better-off and low prices to the poor. Since administration of a direct means test to distinguish the poor from the nonpoor is not practical, this price discrimination can be achieved in three main ways: within facilities, by targeting price increases to achieve full cost recovery for the higher quality classes of hospital inpatient accommodation (VIP, Class I and Class I) which tend to be self-selected by the better-off; among individuals, by targeting price increases at those who are protected by health insurance coverage, in particular at ASKES beneficiaries who presently pay only about one-fifth of the cost of public sector services; and among geographic regions, by targeting price increases to facilities in higher income regions. Thus, instead of a uniform national tariff for health center outpatient visits, central government guidelines would need to specify differential fees by region. Regional differentiation of fees by income level would need to be linked explicitly with a policy of redistributing central subsidies from richer to poorer regions in order to compensate for the lower ability to pay and higher unit costs entailed by the need for incentives to induce skilled manpower to move to poor areas; (b) the system for granting - xix certificates of exemption to the poor who are absolutely unable to pay should be strengthened by formalizing eligibility criteria, eliminating abuses by local officials responsible for granting certificates, and establishing a public fund to reimburse facilities for exempted services; (c) fees should be structured to encourage efficient utilization of services. In particular, for curative care, higher fees need to be charged for hospital than for health center out patient visits, and also for non-referred use of higher level hospitals, in both cases in order to encourage efficient use of the health service referral system. At the same time fees should not be charged for preventive health services with public goods characteristics, including immuni2:ation, other communicable disease control activities (e.g. malaria and tuberculosis control), prenatal care and health education; and (d) revenue retention by facilities should be formally adopted both in order to provide incentives to mobilize additional revenues, and to help ensure that extra fees are use,d to improve the quality of health services provided in hospitals and health centers. (xxi) These elements illustrate the broad directions of a new cost recovery strategy responding to the changing fiscal environment in Indonesia. However, additional data collection and analysis will be needed to translate this strategy into specific actions. Priority areas for support include the following: (a) the effect of fees on utilization of services by households in differe,nt income groups needs to be studied carefully as a basis for targeting price discrimination for public services; (b) institutional responsibility and mechanisms need to be established to monitor levels and trends in unit costs, tariff structures and revenues for public services in order to provide a continuing basis for recommending periodic adjustments in official cost recovery policy; and (c) improvements in financial accounting and reporting systems. are needed for government hospitals and health centers to generate comprebensive and reliable data on unit costs and revenues. (xxii) Measures to strengthen cost recovery policy in REPELITA V need to be C:Lccompanied by improvements in the availability and efficiency of health insurance coverage in Indonesia. This requires the institutionalisation of a strong central policy framework governing the future development of health insurance. Since insurance coverage is a key criterion for price discrimination, the presently low base of enrollment presents a major obstacle to significant increases in cost recovery. Options for consideration include the following: (a) for ASKES, available evidence suggests that there is a strong case for substantially raising ASKES premiums in order to reimburse the full cost of services. Estimates show that ASKES reimbursements presently recover only about 20% of actual costs, resulting in a large and regressive public subsidy to these beneficiaries. In addition to raising reimbursements mediated through ASKES, those covered could also be required to pay an out-of pocket copayment, either in the form of a deductible or of coinsurance for services provided, instead of benefiting from first-rupiah insurance coverage. As well as serving the objectives of revenue generation and equity, this would also help to promote efficiency by reducing the strong incentive on the demand side for ASKES beneficiaries to overuse services in the face of presently zero net prices; (b) efforts to implement the DUKM proposal to extend health insurance coverage to private sector employees should be encouraged, but only on a st.rictly experimental basis in order to clarify an appropriate design. To satisfy the revenue objective, particular attention needs to be given to ensuring that payroll deduction rates and associated reimbursement levels are - xx set high enough to achieve full cost recovery. And for efficiency reasons, consideration needs to be given to the inclusion of private insurers within the DUKM framework, including both conventional insurance carriers and pre paid plans, to help ensure adequate competitive incentives for efficient provision of services; and (c) the priority given to extension of rural health insurance through the dana sehat should be reassessed, on the grounds both that past performance has demonstrated their inability to provide efficient insurance coverage, and also that poor rural villagers should be protected from price increases which would necessitate provision of health insurance coverage. I. SECTORAL PERFORMANCE AND EXPENDITURE TRENDS A. Introduction 1.01 Interventions to improve health are an important policy instrument in the Government's overall strategy to alleviate poverty and improve the welfare of the Indonesian population. Three main factors justify this policy concern with the health sector. First, relief from the burden of illness and premature death satisfies directly a basic consumption need which is an important social policy goal in itself. Second, improvements in health constitute an investment in human capital formation leading to future yields in increased productivity among the poor: better health promotes learning, reduces absenteeism and increases energy output. And third, reductions in infant and child mortality also contribute indirectly to reducing poverty by helpir..g to lower high fertility rates: lower mortality not only helps parents to achieve their desired family size with fewer births, it leads them to want smaller families as well. 1.02 Indonesia's record of health improvement over the past two decades has been solid and impressive. The infant mortality rate has been roughly halvec: from around 132 per 1,000 live births in the late 1960s to about 71 in the e2.rly 1980s. These improvements were associated with the substantial expansion in coverage with government-financed community and preventive health progr2lms which took place during the oil boom period of the 1970s and early 1980s. However, since 1982 the external environment faced by Indonesia has deteriorated considerably as a consequence of the sharp decline in real oil prices and the rising burden of external debt service. The resulting necessity for tight public expenditure restraint to facilitate macroeconomic adjustment has led to a dramatic reduction in central government spending on the hE!alth sector. Direct central spending on health (excluding intergovernmental transfers) fell by 45% in real terms between 1982/83 and 1987/a8. The severity of this fiscal adjustment was enough to cut real spendtng by central government back to the levels attained in the late 1970s, thus reversing the budgetary gains made by the health sector over the previous decadE;. This deterioration reflects the sector's heavy fiscal dependence on a low budgetary share of declining central government expenditure. Although,it is too early to assess what impact these adverse expenditure trends have had on health conditions in Indonesia, they have clearly threatened the prospects for sllstaining the rate of progress achieved during the previous decade. 1.03 In this changing economic environment the major challenge facing Indonesian health policymakers is to define a sectoral adjustment strategy which in the short term protects past gains and which in the longer term promo1:es a sustained recovery in the rate of health improvement. This strate.gy will have to pursue two complementary policy objectives: restr\lcturing government expenditure patterns to make more efficient use of scarce public resources, while at the same time mobilizing additional resources needed to finance higher levels of sectoral spending. With regard to the use of public resources, the main issues include the need to strike an appropriate balance between curative and preventive health services, between capit,ll investment and recurrent expenditure on operations and maintenance (O&M) , and between subsidies to rich and poor beneficiaries. With respect to - 2 resource mobilization, key issues are: the scope for increasing the low sectoral budget share of central government spending, the prospects for mobilizing additional resources from local governments--particularly those with greater fiscal capacity, and the potential for selective increases in cost recovery from users from public services. Preparation of the next five year development plan; REPELITA V, provides a timely opportunity to define and introduce this policy agenda. 1.04 The purpose of this Report is to contribute insights into the broad directions of a health sector strategy for REPELITA V responding to the changing economic environment in Indonesia. Chapter I provides a broad over view of recent trends in sectoral performance and in levels and patterns of government expenditure on health. Chapters II, III and IV then focus in detail on the r~cent adjustment experience and implications for planning and budgeting of the major subsectors: hospital services, community and preventive health programs, and health personnel respectively. Chapter V complements this review of expenditure issues with an assessment of cost recovery policies and related health insurance arrangements which might mobilize additional nonbudgetary resources for the health sector. B. Sectoral Performance Indicators Mortality Decline 1.05 Past trends. Indonesian statistics do not facilitate a comprehensive assessment of mortality trends. Because Indonesia does not have a usable vital registration system for continuous recording of births and deaths, the main sources of national mortality data are indirect estimates of infant and child mortality derived from periodic population censuses and intercensal surveys carried out between 1971 and 1985. However. reliance on these sources for analysis of mortality trends is unsatisfactory for several reasons: (a) estimates of infant and child mortality derived from census and survey data on children ever born and children surviving by age of mother are subject to a variety of sampling, reporting and methodological errors; (b) these sources do not yield reliable estimates of adult mortality and life expectancy; (c) they do not provide information on the structure of causes of death; and (d) the resulting mortality estimates are necessarily infrequent and do not permit up to-date monitoring of mortality trends. 1.06 The available data suggest that Indonesia's record of mortality reduction has been solid and impressive. A key indicator is infant mortality which accounts for around 30% of total deaths. Periodic census and intercensal survey data show that over the last two decades the infant mortality rate has been roughly halved from around 132 per 1,000 live births in the mid-1960s to a level of about 71 in the early 1980s (see Table 1.1). This reflects an accelerating pace of mortality decline, increasing from an average annual rate of 1.2% between the 1971 and 1980 censuses to 9.1% between the 1980 census and the SUPAS 1985 intercensal survey. These results suggest that the REPELITA IV target of reducing the infant mortality rate to 70 by 1989 had already been achieved in the early 1980s. However, these figures are only indicative of trends over time and may not provide reliable estimates of absolute levels. The World Bank projections based on United Nations estimates - 3 provide a more conservative assessment of mortality levels (see Table 1.2). These estimates show that, despite the significant mortality decline achieved over the past two decades, Indonesia's performance still lags far behind the levels achieved in comparator countries. Infant mortality averages only around 45 in the East and Southeast Asia region as a whole. Corresponding rates in the individual ASEAN countries are also much lower than in Indonesia: 54 in the Philippines, 49 in Thailand, 36 in Malaysia and 20 in Singapore. Table 1,1: INFANT MORTALITY RATES, 1971-85 (per 1,000 live births) Census/survey Reference date date Urban Rural Total 1971 Census 1968/69 104 137 132 1976 SUPAS 1972/73 95 113 110 1980 Census 1977/78 88 112 112 1985 SUPAS 1982/83 57 74 71 Source: Central Bureau of Statistics (1987), Proyeksi Penduduk Indonesia. 1985-2005, Jakarta. Table 1.2: COMPARATIVE MORTALITY PROJECTIONS, 1985-2000 Infant mortality (per Life expectancy 1.000 live births) at birth (years) 1985 2000 1985 2000 Indclnesia 89 57 56 63 Malaysia 36 25 68 71 Philippines 54 35 64 69 Tha:l.land 49 32 65 69 Singapore 20 15 73 75 East & Southeast Asia 45 30 68 71 Source: Zachariah, K.C. and Vu, M.T. (1988), Wgrld Population Projections; 1987/88 Editign, The World Bank. - 4 1.07 The poor have benefited less than the better-off from Indonesia's progress in mortality reduction. This disadvantage is reflected in the regional variations in infant mortality levels within Indonesia, ranging from a low of 29 in Yogyakarta to a high of 146 in West Nusa Tenggara. In fact, these interprovincial differences have widened over time because faster declines have taken place in provinces with initially lower mortality rates (see Statistical Annex Table 1.1). These variations are closely associated with regional differences in per capita expenditure and the incidence of poverty. There is a strong negative association between the provincial infant mortality rate and household expenditure per capita, with an estimated elasticity of -0.74. In other words, a 10% increase in provincial expenditure per capita is associated with a 7% lower infant mortality rate. This pattern is confirmed by household survey data showing large mortality differentials by income class as proxied by levels of education. Estimates from the 1987 National Indonesia Contraceptive Prevalence Survey (NICPS) for the ten-year period 1977-87 indicate that infant mortality rates averaged around 90 per 1,000 for mothers with none or only some primary schooling, compared to only around 30 per 1,000 for those with secondary or more schooling (see Table 1.3). Similar differentials stratified by education of the head of household are shown by the 1986 Household Health Survey (HHS). Tablel 1.3: INFANT MORTALITY RATES BY LEVEL OF EDUCATION Level of Education ~ HHS 1986 NICPS 1987 .& None 87.5 98.8 Some primary 81. 9 82.5 Primary completed 53.5 60.1 Secondary or more 45.6 33.9 ~ Level of education refers to the head of household in HHS 1986 and to the mother in NICPS 1987 . .& Average mortality rates computed for the period 1977-87. Source: Budiarso,' L.R. et. al. (1986) Survai Kesehatan Rumah Tangga, Ministry of Healh; and Central Bureau of Statistics (1987) Indonesia: National Contraceptive Prevalence Survey. 1.08 Clos,ing the large infant mortality gap between the poor and the nonpoor will require intensified action against the main causes of high infant mortality among the poor. Four major groups of diseases accounted for more than three-quarters of all infant deaths reported in the 1986 Household Health Survey (see Table 1.4), The highest proportion, 28%, was caused by the group of immunizable diseases, the most important of which was tetanus which alone caused 19% of infant mortality. The second largest proportion of infant deaths was contributed by perinatal causes, most often due to obstructed labor - 5 and un-hygienic delivery, often exacerbated by low birth-weight or prematurity. Diarrhea was the third largest cause, contributing 16% of infant deaths, followed by acute respiratory infections (ARI) which caused 14% of the total. Table 1.4: UNDERLYING CAUSES OF INFANT MORTALITY, 1986 (% of deaths) Neotanal Post-neonatal All Infants lnununl~ableJl11eases 40.9 19.9 28,0 Tetanus 39,5 6.6 19.4 Measles, diptheria & pertussis 1.4 13.3 8,6 Perinatal causes 42.3 2.9 18.4 Diarrhea 2,3 24.3 15,6 Acute respiratory infection 2.3 21. 7 14.4 Other 12.2 31.2 23.6 Total .l.QQ...Q 100.0 100.0 Source: Government of Indonesia - UNICEF (1989) Situation AnalYlis of Children and Women in Indonesia, Tables 3.6 and 3.7. 1.09 Future prospects. Substantial further progress in infant mortality reduction will be essential to enable Indonesia to achieve its poverty alleviation goals. Government population projections envisage a continuing but slower pace of infant mortality decline averaging 4% between 1985-90, 3% between 1990-95 and 2% between 1995-2005 leading to a rate of 45 in the year 2000. These implied targets are modest in relation to the magnitude of the task facing the health sector. Table 1.2 shows that achievement of these targets by the end of this century would only reduce infant mortality to the current average for the region, thus leaving Indonesia well behind the much lowel' rates likely to be achieved by its comparators in the future. More rapid. improvements will therefore be needed to enable Indonesia to reach the level of social development enjoyed by neighboring countries. This will depend primarily on further progress in prevention and cure of the major com municable diseases, 1.10 The information system needed to document mortality change is necessarily part of the development process. Thus it is not surprising that the mortality data currently available in Indonesia are inadequate to meet the needs of routine sectoral performance monitoring and planning. However these deficiencies will become more acute in the longer term as Indonesia undergoes the epidemiological transition from a mortality structure dominated by communicable disease mortality among children to noncommunicable disease mortality among adults. This transition will result from: (a) the gradual ageing of the population induced by continuing declines in infant and child mortality and fertility; and (b) the higher age-specific incidence of - 6 noncommunicable diseases resulting from lifestyle changes associated with economic development such as increasing prevalence of smoking, consumption of higher fat diets, and rising incidence of traffic accidents. Strengthening the information system needed to monitor, analyze and plan appropriate policy interventions in response to these mortality trends is a high priority. Particular consideration needs to be given to development of a reliable vital registration system and also the capacity to undertake epidemiological analyses of the causes and underlying determinants of adult mortality. Utilization of Health Services 1.11 The level and pattern of health service utilization are important complementary indicators of health sector performance because they are suggestive of morbidity improvements which are not fully reflected in mortality data. Available information shows that Indonesia's high mortality rates are associated with remarkably low utilization rates, especially among the poor. 1.12 Provider choice. Estimates from the 1987 National Socioeconomic Survey (SUSENAS) show that the sick poor use fewer modern curative health services and choose lower quality providers than the nonpoor (see Table 1.5). A higher proportion of the poor reporting illness did not receive any treatment. Furthermore, a higher proportion of the sick poor resorted to self-treatment only. Of those reporting illness, fewer of the ill poor used modern health care providers. Among modern providers, government health centers turn out to be by far the most important choice for all persons reporting illness, with only a slight decline in frequency of use between the poor and the nonpoor. However, for high-quality modern providers (doctors and hospitals) there is a strong pattern of higher use among the better-off: Table 1.5: PROVIDER CHOICE BY EXPENDITURE CLASS, 1987 (% of those reporting 111 L4) llman lhu;:al Tgtal Poor Nonpoor Total Poor Nonpoor Total Poor Nonpoor Total Not treated 4.8 2.1 2.4 7.3 3.9 5.0 7.1 3.4 4.4 Self-treated 37.7 22.4 23.8 31.8 26.8 28.5 32.2 25.6 27.4 Traditional 2.4 1.5 1.5 5.4 4.7 5.0 5.2 3.8 4.2 Modern 55.0 l..L.2 ll...Q ~ .§!!...2. ll...2. 55.4 ~ 64.0 Doctor 7.7 26.8 25.1 2.6 8.8 6.8 3.0 13.9 10.9 Hospital 7.9 11.9 11.6 2.6 4.8 4.1 3.0 6.8 5.8 Health center 27.2 25.4 25.6 30.9 30.3 30.5 30.6 28.9 29.4 Clinic 1.9 2.8 2.7 3.4 2.9 3.0 3.3 2.9 3.0 Paramedic 10.3 7.0 7.3 16.0 17.7 17.1 15.5 14.6 14.9 LA Based on three-month recall. Source: World Bank staff estimates from 1987 SUSENAS survey. - 7 fewer of the ill poor visited a hospital, compared to the ill nonpoor; and, fewer of the ill poor visited a doctor, compared to the nonpoor. Better access to high-quality providers in urban areas is reflected in higher proportions of the poor visiting doctors and hospitals than in rural areas. 1.13 Utilisation rates. The combination of low reported morbidity and low use of modern health providers results in low absolute rates of health service utilization by the poor (see Table 1.6). Outpatient visits per person per yea): among the poor were significantly lower than among the nonpoor. Most of this difference is due to the much higher utilisation rates of high-quality pro"iders (doctors and hospitals) among the better-off, with little difference in levels for the lower-quality sources of care. The differential in hospital admission rates is much wider, averaging three times as many for the nonpoor comllared to the poor. On average, these rates are somewhat lower than were rep(lrted by the 1986 Household Health Survey but the pattern of differentials by :~ncome class as proxied by levels of education are broadly consistent (see Tab;.e 1. 7) . Tahl~ 1,6: UTILISATION OF MODERN PROVIDERS BY EXPENDITURE CLASS, 1987 (Rates per year LA) UIhm Blu:al Totiil Poor Nonpoor Total Poor Nonpoor Total Poor Nonpoor Total Hospital admis sion Lh 6.55 22.44 21.72 4.90 12.72 10.72 5.00 15.79 13.63 Outpatient v:lsits Ls 0.23 0.31 0.31 0.24 0.35 0.33 0.24 0.34 0.32 Igta1 v;i.sits Ls 0,24 .Q..ll 0,33 0.26 Q...jj! 0.35 0.25 0.36 0.34 Doctor 0.03 0.12 0.11 0.01 0.05 0.04 0.01 0.07 0.06 HI)spital 0.03 0.05 0.05 0.01 0.03 0.02 0.01 0.04 0.03 Health center 0.12 0.12 0.12 0.14 0.18 0.17 0.14 0.16 0.16 Clinic 0.00 0.01 0.01 0.02 0.02 0.02 0.02 0.02 0.02 Pa.ramedic 0.05 0.03 0.03 0.07 0.10 0.10 0.07 0.08 0.08 LA Annualized rates based on three-month recall. Lh Per 1,000 persons. Ls Per person. Source: World Bank staff estimates from the 1987 SUSENAS survey. - 8 Table 1,7: UTIl.ISATION OF MODERN PROVIDERS BY LEVEL OF EDUCATION, 1986 (Rate ~ per year) Some Primary Secondary None primary completed or more Total Hos~ital Admission ~ 7.04 9.65 11.75 21.14 li...1l Government hospitals 6.10 8.23 9.26 16.96 16.13 Private hospitals 0,94 1.42 2.49 4.18 3.58 Out~atient Visits 1£ 0.380 .Q...ill Q.467 0.567 0.452 Government hospitals 0.023 0.023 0,042 0,071 0,037 Private hospitals 0.003 0,004 0,006 0,012 0.006 Health centers 0.224 0.252 0.241 0.234 0.239 Private clinics 0.001 0.003 0.005 0.005 0,003 Doctors 0,044 0.065 0,088 0.186 0,088 Paramedics 0,084 0.086 0.085 0.060 0.080 /.§. Based on one-month recall. LQ Per 1,000 persons. Is:. Per persons Source: 1936 Household Health Survey. 1.14 Internationa~arisons. Even more striking than the relatively low utilization rates among the poor in Indonesia, especially for hospital admissions, is the fact that these rates are extremely low even for the nonpoor compared to levels prevailing in other countries. Household survey data derived from the World Bank's Living Standards Measurement Study (LSMS) program show average hospital admission rates between 5 and 10 times higher than the Indonesia average of 13.6 per 1,000 per year (see Table 1.8). Even Table 1,8, COMPARATIVE HOSPITAL ADMISSION RATES BY CONSUMPTION QUINTILE (per 1,000) ConsumRtion Quinti1es I II III IV V Total Cote d'Ivo:f..re 45.4 6005 55.3 60.4 111.8 66.6 Ghana 87.0 126.4 156.6 143.3 214.1 145.4 Jamaica 68,8 85,9 85.6 17 .0 94.6 70.3 Peru 40.3 55.1 95.2 87.6 90.1 73.6 _-~"O~~_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ Source' ':1f);c1d Bank staff estimates derived from the Living Standards ~k,,1",$'l!1r';!"'~nt Surveys. - 9 in the poorest consumption quintile hospital admission rates in these four countries are 3 to 6 times higher than the average for Indonesia as a whole. Thus the poor in Indonesia face differential access to a health which itself delivers relatively few services even to the nonpoor. 1.15 Distance. Travel time is an important determinant of levels of utilisation and patterns of provider choice. The opportunity cost time is a bigger access barrier for the poor than the nonpoor. Unfortunately the SUSENAS 1987 only reports travel time as reflected in distance travelled by actual users, instead of for all respondents; as a result, it does not reflect differentials in access for the population as a whole. However, the results do suggest that the poor face higher access costs, and that they are sensitive to these costs (see Table 1.9). For example, the average distance travelled to a health center is higher for the poor than the nonpoor. Distances travl~lled to hospital treatment are much higher, and those poor who decide to use -them are willing to travel less far than the nonpoor. Table 1.9: DISTANCE TO MODERN TREATM.ENT BY EXPENDITURE ClASS, 1987 (Average distance per user in km Ls) Urban Rural Total Poor Nonpoor Total Poor Nonpoor Total Poor Nonpoor Total Doctor 3.5 3.1 3.1 10.2 9.9 9.9 9.0 6 .J '" 6.4 Hospital 1.8 5.6 5.5 10.9 15.2 14.6 9.5 10.6 10.5 Heal-::h center 2.4 1.4 1.4 3.4 3.4 3.4 3.4 3.0 3.0 Polyclinic 0.4 2.0 2.0 2.8 2.9 2.9 2..8 2.7 2.7 Paramedic 2.5 1.8 1.8 3.2 3.0 :-Lt 3.2 2.9 2.9 All providers 2.4 2.7 2.7 4.9 4.9 4.7 3.9 4.3 4.2 ~ Based on three-month recall. Source: World Bank staff estimates from the 1987 SUSENAS survey. 1.16 These data demonstrate that the poverty alleviation impact of available health services in Indonesia is modest. Access to modern facilities is ll:)w compared to other Asian countries, and utilization is much lcnqer among the poo!'. As with mortality data, substantial improvements in the availability and quality of utilization statistics are needed to facilitate sectoral monitoring and planning. At present, there is no institutional mechanism to generate regular and comprehensive utilization data based on national-level household surveys to permit analysis of trends in levels and patterns of use among different income classes. Development of this capacity should be a priority to ensure better assessment and targeting of the poverty alleviation impact of health sector in~erventions. - 10 C. Government Expenditure Trends Bud&etaty Sources and Structure 1.17 Before examining government expenditure trends, it is important to understand the complexity of the existing pattern of government finance and its implications for planning and budgeting in the health sector. Three major problems with the system of budgetary finance are: (a) the fragmentation of funding between different budget channels split between the central and regional governments; (b) the lack of a comprehensive data source which aggregates these mUltiple sources into a consolidated account of sectoral expenditure; and (c) the obscurity of accounting classifications used in different budgets. Together. these three features of the budgetary system mean that there is no readily available information on total government expenditure on health. or on its composition between recurrent and investment spending, or on its functional allocation between different sectoral programs, or on its distribution between regions. In other words, no comprehensive view of actual budgetary allocations is normally available to sectoral policymakers as a basis for assessing the need for changes in expenditure priorities. 1.18 The fra~entation of financing between multiple budget channels split between the central and regional levels of government is illustrated in Table 1.10. The actual financing picture is more complex than suggested by this list because it only includes services delivered under the auspices of the central Ministry of Health and the regional health offices; related health expenditures funded through the Armed Forces, the Ministry of Education, and other departments and state enterprises are excluded. There are at least ten different budgetary sources of funds for the main public health system. Figure 1.1 illustrates the flow of funds through the three main tiers of government in Indonesia. At central government level the most important sources take the form of central grants to the regions: the Subsidi Daerah Otonom (SDO) and the specific Instruksi Presiden grant for health (INPRES Kesehatan) channeled for routine and development expenditure respectively. An important feature is the rigidity associated with the earmarking of those grants for specific uses. The SDO covers salaries for all regional government staff whose appointments are centrally approved, including a large number of staff working in regional health facilities such as hospitals and health centers. The sectoral INPRES grant mainly finances health center construction and operating costs through a per capita subsidy for drugs. Next in importance are the national budget (APBN) allocations to the Ministry of Health for routine (Daftar Isian Kegiatan or DIK) and development (Daftar Isian Proyek or DIP) expenditure, both of which finance directly a large amount of spending incurred in the regions. Additional central sources include foreign grants and loans (Bantuan Luar Negeri or BLN) channeled through the development account, and the Subsidi Bantuan Biaya Operasional Rumah Sakit Umum Daerah (SBBO-RSUD) for routine expenditure. The SBBO-RSUD complements the SDO grant with the specific purpose of subsidizing nonsalary expenditure (excluding drugs) requirements in provincial and district hospitals. Corresponding regional sources include the development and routine budgets allocated to health offices attached to the provincial and district governments (level I and II regions respectively). - 11 · Table 1.10: MAJOR BUDGETARY SOURCES OF HEALTH SECTOR FINANCE Central Government APBN-DIP Central Ministry of Health development budget APBN-DIK Central Ministry of Health routine budget INPRES Kesehatan Health sector grant to regional governments BW Project aid (grants and loans) SDO Salary grant to regional governments SBBO-RSUD Nonsalary grant for regional hospitals Recional Governments APBDl-DIP Provincial health office development budget APBDl·DIK Provincial health office routine budget APBD2-DIP District health office development budget APBD2-DIK District health office routine budget 1.19 There is no comprehensive data source which aggregates these multiple budget channels into a consolidated account of budgeted or actual expenditure on health. As a result it is very difficult to identify and coordinate all relevant financial flows to the health sector or to particular services within the sector. This reflects the absence of any institutional responsibility for maintaining an up-to-date information system for budgetary statistics in the health sector. The resulting paucity of data is manifested in two main ways: (a) available information on the different central government funding sources is substantially incomplete because the Ministry of Home Affairs is unable to provide a breakdown of the sectoral distribution of the SDO grant which is transferred to the regional government routine budgets to pay for staff salaries (the health sector share can only be identified from the salary component of the regional government routine budgets); and (b) data on regional government spending on health are normally not available. As a result only about one-half of total government spending is readily identifiable without undertaking a major survey of the individual regional government budgets. Although regional government accounts are published annually by the Central Bureau of Statistics, these only provide an economic classification of expenditure. A sectoral breakdown can therefore be obtained only by referring directly to the individual provincial and district budgets, which are often not accessible at central level. 1.20 For budget data that are available, the obscure classification of budceta[y disbursements often fails to reveal the actual composition of expenditures: (a) the distinction between development and routine budgets does not adhere to the conventional classification of investment and recurrent expenditures because a large amount of recurrent spending is funded through the development budget. Thus a useful breakdown between investment and recurrent expenditure can only be obtained by examining the detailed economic classification of the development budgets. However this cannot be done for the sectoral INPRES grant or for the provincial and district development budgets which do not provide any economic classification of expenditures; and (b) the functional breakdowns provided by individual budget accounts are often too inconsistent and general to permit meaningful aggregation across budgets in order to identify overall patterns of resource allocation between - 12 Figure 1.1: BUDGETARY FLOWS FOR HEALTH, 1985/86 (Rp. billion) Centrol Government Provinclol Govem in pract1.Ce,cherl!> is obviously great uncertainty about the bc:(:'i:~eeil inputs .and outputs, whether these are measured in terms of thaiL ulttill~tc impact: in loweX'ing morb:idity and mortality or in terms of hrt~i·iiiJ.ed:Lt!.t:1i> su.:::h. as ii.'1nillbers of hosp:ital admissions or inpatient days ~ NevGycI-~\2:1,,,,s:::l, it shvuld be Fossible to establish that a given level of staffing anJ0>t:her: is necessary to achieve a minimally acceptable IB';YE~[!1/il:dj'.CS" 4.25 For paramedics. the intco:')b l:h,~lt th(~ 'm1>l:klD.:'I.d (,1: ,:" particular type of employee be constant a,cr,j,!.i:iS (w;:Llh:ies 0 Ic"muld p:<,_'oba'bly be inefficient if nurses at hospital x handl""d 50 !,l;)l::ie"t:"l a J2.!Y, while nut'ses at hospital y averaged 10; a transfer of nU:L:::';'"s '-:Jould df nongovernment hospital emplc.yment over that three year period, Assessment of the situation requires much better information on hiring trends and prospects in the private sector. Other factors in the supply/demand equation also are in need of quantification. There does not appear to be readily available data on the attrition rate of staff, so the extent to which job vacancies make it possible to absorb additional paramedics is not known. Resional Balance 4.35 Measures of aggregate excess supply obscure substantial regional variations in excess supply. There are number of provinces, and probably many scattered districts, apparently experiencing excess labor demand. Posts in these areas are unfilled despite the presence of excess supply elsewhere. This means that, for labor surplus areas, excess supply is more severe than the aggregate figures would indicate. 4.36 Although the MOH has allocated a favorable proportion of new paramedical posts to outlying provinces, it has often been difficult to fill these posts.1/ Of the 22,071 INPRES positions created over 1979/80 to 1985/86, about 83% are ever-filled (the qualification is necessary, because some of the posts may have since been vacated). There are substantial disparities between provinces in the success in filling these positions. More than 40% of the INPRES positions have never been filled in six provinces: Central, West and South Kalimantan; South Sumatra; Irian Jaya; and East Timor. In contrast, more than 97% of INPRES positions were filled in North Sumatra, Jakarta, West and East Java, Yogyakarta, and Bali. Routine posts, which tend to be in hospitals and urban areas, are much more easily filled. More than 96% of the 4,114 routine posts created during 1984/85 and 1985/86 were filled; West Java was the province with the lowest record, at 89% filled. As noted 11 See Statistical Annex, Table 5.7. - 78 earlier, the MOH has an incentive to fill routine positions first, or to create these positions in places which will be easy to fill, since routine positions are forfeit if not filled. 4.37 A substantial number of posts remain unfilled at the same time that aggregate manpower supply exceeds aggregate demand. This suggests that supply and demand do not equilibrate between provinces. The available data !I, although rough and not comprehensive, show excess supply in some provinces and excess demand in others. These data cover only supply and demand for SPK (high school) level nursing positions created under INPRES, which are primarily health center positions. Supply here includes only those graduates who have applied for a public position; those who seek or have found private sector positions are excluded. Graduates are classified by their province of education, not their province of origin. Of the 4,244 open posts, 1,364 remain unfilled; these are concentrated in the excess demand provinces of South Sumatra, Bengkulu, Central Java, all of Kalimantan, and Irian Jaya. At the same time, East Java has a surplus of 1,298 unplaced graduates, and Jakarta has an excess supply of 569. In sum, there are around 2,600 unplaced graduates. While some of these unplaced graduates may subsequently have obtained positions created under the routine budget, only 1,000 such positions were available nationwide. Thus: (~~ there is a mismatch between production and absorption of paramedical personnel by province; (b) this mismatch is not equilibrated by interprovincial placement or migration. 4.38 Once in place, a paramedic is very unlikely to move between provinces. In 1985/86, there were a total of 513 interprovincial moves among the hundred thousand or so employees of rank I or II, a class that includes most paramedics. There are a number of plausible explanations for the reluctance of paramedical students to apply for out-of-province posts. Unlike physicians, most paramedical staff do not receive free housing; hence there is a strong incentive for the student to stay within the sphere of her social network, where housing can be arranged with friends or relatives. Cultural barriers between different regions of Indonesia can be formidable, and single women may have a particularly hard time in unfamiliar settings. Marriages contracted while the student is in paramedical school may be another constraint on reassignment. However, all these explanations remain untested hypotheses. It would be worthwhile to undertake a student survey (especially in East Java) to determine the actual constraints on placement. It is possible that some streamlining of administrative procedures, or an increase in the transportation and resettlement allowance, might be useful. Student Recruitment Policy; Quotas 4.39 Recognizing the difficulty of equilibrating supply and demand through outplacement of graduates, MOH has adopted a policy of encouraging recruitment of students in excess demand areas; the goal is that each province (if not each district) should be able to supply its own SPK-level, and much of its academy-level, nursing personnel. This policy takes the shape of provincial quotas for admission to nursing schools; provinces with insufficient training capacity are assigned slots at schools in neighboring provinces. For instance, for academic year 1986/87, 87 school slots were reserved for 11 See Statistical Annex, Table 5.8. - 79 students from Central Kalimantan. Forty of these were assigned to the SPK at Palangkaraya, the provincial capital. The remaining 47 were assigned to a variety of types of schools outside the province, in Banjarmasin, Bandung, Surakarta, and Jakarta. Quotas also exist at the district level. The quota scheme applies only to the 96 vertical schools, funded and operated directly by MOH. These schools offer 6,200 of the total 14,480 slots open for entering students. The quota schedule is devised by the central admissions committee, chaired by the head of Pusdiknakes. There is no formal representation of the Planni~g Bureau or the Personnel Bureau on the Committee.2/ E. Doctor Supply and Demand A&®ate Balance 4.40 General doctors are trained at thirteen public universities operated by the Ministry of Education and Culture; and at ten private universities. Upon gr.aduation, doctors have four possible sources of employment: (a) as a MOH employee (though possibly seconded to a local government); (b) as a MOEC employee; (c) the armed forces; (d) certain private employment interpreted as meeting the national interest (there is however no formal list of private employer.s which satisfy this requirement). 4.41 There is some disagreement concerning the equilibrium between supply and dem,and. Although all doctors are required to register with MOH upon graduation, MOH keeps statistics only on its own hiring.1Q/ However, MOH staff maintain that most general doctors find employment after graduation; the excepti.on is a group of approximately 150 long-term unemployed doctors who seek a .Jakarta based job, usually because of a spouse employed there. Accordi~ng to MOH roughly 300 graduates per year obtain positions with MOEC or the armed forces, and about 100 obtain private positions. MOEC, however, maintains that there is a high degree of unemployment among recent graduates. There is agreement, however, that there is a substantial degree of unemplo:yment among dentists. This reflects the surplus of dental graduates over ne'N posts.!l/ In addition, there is some tendency for dentists (the majorit:y of whom are women) to defer to their spouses' job locations (as evidenced in the large queue of requests for Jakarta jobs).ll/ 2/ The quota scheme and student selection procedure is described in detail in the Pedoman Pelaksanaan. Seleksi Penerimaan siswalMahasiswa Baru. published by Pusdiknakes. lQ/ See Statistical Annex, Table 5.10. !l/ See Statistical Annex Table 5.10. l2J See Statistical Annex, Table 5.11. - 80 - Relional Balance 4.42 Two to five years of mandatory service is required of all medical school graduates, public and private. Completion of this service is a prerequisite for admission to specialist training. To promote better geographical distribution of doctors, the mandatory service period is shorter for the less attractive provinces. The classification of provinces by service period has varied over time. The most recent one is shown in Table 4.7. Table 4.7: MANDATORY SERVICE REQUIREMENTS FOR DOCTORS Five years Region I: Java. Three years Region II: Sumatra, Bali, West Nusa Tenggara. Sulawesi (except Southeast Sulawesi), South Kalimantan. Two years Region III: Southeast Sulawesi, Kalimantan (except South Kalimantan), East Nusa Tenggara, Maluku, Irian Jaya, and East Timor; plus designated remote areas within Regions I and II. 4.43 Placement procedure. Graduates applying for MOH positions specify their preferred provinces. These preferences are matched, on a first-come, first-serve basis, with available positions; INPRES positions are filled first. Applicants are not obliged to accept any position, but the consequence of refusal is likely to be at least a year of unemployment. The provincial supply/demand situation for doctors can be illustrated with data showing for each province the number of INPRES positions available (i.e., health center positions), and the number of applicants specifying that province as first choice.11/ The overwhelming popularity of Jakarta is clear. In 1986/87, 192 applicants listed Jakarta as their first choice, although no INPRES positions were available. MOH Bureau of Personnel staff estimate that approximately 150 of these applicants are long-term unemployed graduates who seek a Jakarta position because their spouses work there. Other popular areas include the rest of Java, Bali. South Sulawesi (Ujung Pandang) and North Sumatra (Medan). Turning to deficit provinces, where open positions exceed applicants, Kalimantan heads the list, with 23 applicants for 65 positions. Other deficit areas include South and West Kalimantan, Aceh, East Nusa Tenggara, Maluku, Irian Jaya, and East Timor. 4.44 The results of the assignment process over the past two years mirror the selection priorities discussed above.~ Among newly created INPRES positions over 1984/85-85/86, most new physician posts in South and Central Kalimantan were left unfilled; about a quarter of the new posts in Maluku and Southeast Sulawesi were also not filled. Dentists were reluctant to be posted to health centers in East Nusa Tenggara, Irian Jaya, Central Kalimantan, and 11/ Statistical Annex. Table 5.12. ~ Statistical Annex, Table 5.13. - 81 South Kalimantan. By contrast, virtually all routine posts over 1985/86 and 1986/87 were filled; the routine posts are more likely to be located in hospit~ils, and in provincial capitals, and are thus more attractive. 4.45 Transfers. attrition and understaffing. In essence, the current system provides doctors to remote areas by rotating new graduates through on two-year tours of duty. Although some doctors may decide to stay on at their posts, most will opt for specialist training, which offers the possibility of an urb~m posting with a lucrative side-practice. It is therefore hard to augmen1: the doctor/population ratio in outlying regions. 4.46 The rapid throughput of doctors implies that it is very difficult to maintain full staffing of facilities in outlying regions, explaining the sub stantial proportion of health centers without doctors. Eleven provinces have more than one-quarter of their health centers operating without a doctor in 1985. These are official statistics on placement; the informal reassignment process discussed previously means that these statistics probably under estimate the proportion of doctorless health centers. A very rough estimate of the rate of accumulation of medics (doctors and dentists)12I clearly suggests that the system acts to accumulate medics in Jakarta, and to a lesser extent, Central and West Java. By contrast, East Nusa Tenggara, South Kal ima:rltan , and Aceh seem to be suffering from a net loss of medics over time. On a pE!r capita basis, the accumulation of doctors in East Java lags far behind other Javanese provinces. Outside Java, Bali, and Sumatra, the annual net increment of medics is in the single-digit range. This places a binding constraint on the possible expansion rate of the health center network. 4.47 Another constraint on the distribution of doctors is the lack of a centralized mechanism for filling vacated posts (lowongan). In theory, new graduates are supposed to be assigned only to newly-created INPRES posts. Vacancies at old posts (left, for instance, by the departure of a doctor after his two-year service period) are not filled through a centralized process, but by application to the provincial health office. If this rule is strictly adhered to, it is hard to see how the vacancies could be filled in places such as Central Kalimantan where even the new positions are left vacant. 4.48 Provision of doctors' services to outlying areas depends on the throughput of doctors as they traverse a period of public service between basic and specialized medical education. In many provinces, that throughput is insufficient, despite direct cash incentives (which may, however, only offset the higher cost of living in these areas) and the incentive of a reduced service period. Even where the quantity of doctors is adequate, the rapidity of the throughput means that a health center doctor never has time to become truly familiar with local people and conditions. A possible al tert1.ative to this system is to provide large incentives for doctors to remain. in outlying areas. Such incentives might include a salary comparable to a Java-based doctor's outside income, paid leaves to Jakarta or province of 121 See Statistical Annex, Table 5.14. The estimate calculates the inflow from INPRES and routine appointments, outflow to specialist training, and in- and out-flows from transfers. Outflows from specialist training are not available, but are surely concentrated on Jakarta and other urban centers. - 82 birth, and subsidized high-quality education for the doctor's children. In assessing the expense of such as scheme, it must be remembered that the current system supports the large expense of subsidized medical education in exchange for a rather small output of public goods, namely two or three years of service by some medical graduates in areas unable to support a private physician. Over the long run, it might be desirable to shift away from the indiscriminate subsidy of medical school students toward highly targeted incentives for service in the public interest. F. Policy Issues 4.49 Personnel information. The current personnel information system is inadequate for the management of the Ministry of Health's most important single resource, accounting for over Rp.243 million in annual recurrent costs. The current system: (a) does not cover about one-third of all employees working in MOH-supervised facilities; (b) provides no information on the facility or function to which a staff member is assigned. These shortcomings hobble basic planning and analysis, especially regarding the allocation and distribution of employees. The MOH recognizes these problems and is currently implementing an improved personnel information system. So far, a low response rate to baseline data collection has prevented operationa1ization of the system. It is important that the low response rate problem be diagnosed, and that the solutions be enthusiastically supported. Overlaps and redundancies between competing information systems should be minimized. 4.50 If it is necessary to continue using the current information system as a stopgap, consideration should be given to negotiating with BAKN for direct, continuing access to civil service data. The current procedure is to update MOH information independently of the BAKN. Direct access would save time and effort, provide more timely information, and could provide up to date information on health facility employees formally belonging to other departments. 4.51 Excess supply of paramedical personnel. During REPELITA IV, public paramedical schools have graduated several thousand more paramedics than the government has been able to absorb given fiscal constraints on hiring rates. In addition, graduates from private paramedical schools are also likely to greatly outnumber available openings in the private sector. This represents a policy problem for two reasons. First, it is wasteful to spend Rp.0.5 million per student year in providing a skill that may never be utilized. Second, there is some danger of a cobweb cycle: the paramedical glut could deter new admissions, resulting in quality or quantity shortfalls three years hence. 4.52 There is a need for estimating the extent of, and determining if there are any barriers to, private-sector absorption of these graduates. Indeed, if MOH is to continue as central planner of all paramedical supply for the country, this information should be gathered on a routine and continuing basis. Further research is also needed to explore the feasibility of various policy responses, including: (a) diversion of some pekarya kesehatan posts to paramedics; (b) selective school closings via enforced accreditation standards, especially in provinces with the highest levels of excess supply (with some of the cost-savings devoted to retraining or placement of affected - 83 students); (c) reabsorption of existing staff or new graduates into supplementary training programs. Among these, the pekarya issue is of particular importance. MOH has devoted a significant part of its scarce new paramedical posts to these auxiliary paramedics, who receive only four months of training but are paid at the same scale as paramedics with three years of education. In theory, pekarya will greatly improve rural health services, since they are recruited from the neighborhood of understaffed health centers. The success of this theory in practice needs to be verified. 4.53 Allocation of new posts. Via the INPRES program, MOH has achieved a relatively equitable distribution of new posts among provinces. Routine posts are less equitably distributed, with Jakarta taking a disproportionately large share. However, scattered evidence suggests that the intraprovincial distribution of posts deviates from the goals of equity and efficiency, resulting in substantial mismatches between facility staffing and utilization. To some extent this may result from a pragmatic recognition of the difficulties of assigning staff to rural areas. It may also reflect political pressures on local administrators combined with a lack of staffing standards and staffing data. 4.54 The new utilization-based staffing standards developed under the ISN system are addressed to the latter problem. An ISN-like system is necessary in an era where the unpredictability of annual budgets requires flexible, prioritized staffing plans. It is important that the ISN system adopt a feasible system of data collection and be institutionalized to aid in planning. 4.55 Another planning problem results from the fragmentation of budgetary responsibility for health employees. Approximately one-quarter of all workers at MOH-supervised facilities are nonmedical staff appointed at the provincial or local level by the Ministry of Home Affairs. The resultant coordination problems are said by MOH to result in some under-provision of nonmedical staff to local-level hospitals. 4.56 Placement and distribution of personnel. Although some provinces produce more paramedics than they hire, and others suffer from a deficit of paramedics, there is relatively little intraprovincial placement of paramedical graduates. Recognizing this problem, the MOH has instituted a quota system for allocating public paramedical school slots by province. The success of this strategy has yet to be assessed. Inadequate reliable information exists on the placement of new graduates within provinces, or on the filling of vacancies. It is widely thought that difficulties in placing paramedics in rural areas have led to understaffed health centers and overstaffed hospitals and health offices. Verifying the extent to which this is true, and verifying the accepted hypotheses on why it is difficult to place paramedics, will require research. 4.57 Provision of doctors to rural areas and remote doctors is accomplished by rotating recent medical school graduates on two-year mandatory duty tours. It is therefore extremely difficult to augment the doctor/population ratio, or to maintain health center staffing in these areas. In eleven provinces, more than one-quarter of all health centers were reported to lack a doctor in 1985. It is possible that provision of generous and innovative incentives for rural residence might help the situation. Such - 84 incentives might include, for instance, subsidized education for the doctors' children at high-quality boarding schools. Such incentives might be financed, in part, by recovering the cost of medical education from doctors who practice in urban areas. This would also serve to increase the relative attractiveness of the rural posts. 4.58 MOH is currently in the process of establishing staffing coordinators at the provincial level. It is hoped that these officers will be able to improve both the provincial supply/demand situation and the intraprovincia1 allocation of manpower. 4.59 Worker productivity. Scattered evidence suggests that average productivity is low, but highly varied, throughout the health system. To a large extent, low productivity may be rooted in a failure of the personnel allocation process to match staff with existing workload. The new ISN manpower planning system is directly addressing this problem. Low productivity may also stem from structural barriers to utilization, e.g., lack of specialized equipment, staff, or drugs, or inconvenient opening hours. Another problem lies in the lack of an incentive structure. With few exceptions, the system has lacked any penalties for lack of effort, or rewards for exceptional effort. Current MOH plans to link promotions in salary rank to work effort are a laudable step in the right direction. Further mechanisms for improving employee management should be explored. 4.60 The QuantitY/Quality tradeoff. The conventional view is that there is an urgent need for additional paramedical staff. For this reason, REPELITA IV chose to emphasize rapid expansion of staff quantity, making quality improvement a long-term goal. From this viewpoint the current necessity to reduce paramedical output is a regrettable, almost paradoxical, short-term expedient. It is appropriate, however, periodically to reexamine the quantity-quality tradeoff in light of changing conditions and experience. As noted above, there is evidence of areas of low labor productivity throughout the health sector. To the extent that the existing labor force is inefficiently utilized, one may question the necessity for adding additional staff. It may in fact be more productive to devote more resources to improving the quality and efficiency of existing staff, while slowing the rate of growth of the labor force. The current period of retrenchment may offer an opportunity to redirect personnel policy in this direction. - 85 V. COST RECOVERY AND HEALTH INSURANCE A. Introduction 5.01 Previous chapters have shown that the provision of adequate health services is threatened by the scarcity of fiscal resources in Indonesia. This provides an appropriate context to review options for increasing cost recovery to offset the decline in budgetary revenues. Experience in some developing countries shows that prices can play an important role in mobilizing supplementary resources to help finance provision of public sector services. Section B reviews current practice and performance in Indonesia. However, raising cost recovery would also increase the need for risk-pooling health insurance systems in order to alleviate the potentially catastrophic financial consequences of illness, and to ensure that utilization of necessary services is not reduced by higher fees. Sections C, D and E review the status of insurance coverage for civil servants (ASKES), private sector workers (DUKM) and rural villagers (dana sehat) respectively. Section F identifies selected policy issues in health insurance development. Section G then outlines appropriate policy dir~ctions in Indonesia. B. Pricing and Cost Recovery Tariff Structure 5.02 Unlike many developing countries, Indonesia has adopted a system of user charges for public sector health services. For hospitals the pricing structure consists of three main tariffs: (a) a fixed fee per outpatient visit; (b) a fixed fee per inpatient day, differentiated by the class of accommodation; and (c) a schedule of fees for a wide range of special servic.es. including operations, diagnostic services (such as X-ray and laborsLtory examinations) and drug sales. Of these three tariffs, the largest source of hospital revenue generation is typically the fees for special services, which account for about half of total revenues; most of the remainder is generated by inpatient fees, with outpatient fees contributing only a small proportion. Responsibility for setting hospital tariffs rests with the level of government which owns the facility. For vertical hospitals owned by the central government, tariff levels are determined by the MOH Directorate General of Medical Care. New guidelines for setting tariffs were issued in 1987 and summarized in Table 5.1. Tariffs for provincial or distrtct hospitals are determined by the relevant government authority. These are expected to conform roughly to the guidelines for central government hospitals, but the central MOH plays no role in determining or monitoring regional government fee schedules. The level of fees corresponding to this tariff structure is illustrated for selected hospitals in Table 5.2. These tariffs are set well below the actual cost of delivering hospital services, for which recent estimates are summarized in Table 5.7. · 86 . Table 5.1: GUIDELINES FOR HOSPITAL TARIFFS Outpatient Visit Class A 50% (Drug Cost Index) Class B - 40% (Drug Cost Index) Class C - 30% (Drug Cost Index) Class D - 20% (Drug Cost Index) Inpatient Day Class III A - 1.5 (Food Cost Index) LA Class III B - Class III A/3 Class II - 2 to 5 (Class III A) plus 30% medical consultation Class I - 6 to 9 (Class III A) plus 30% medical consultation Class VIP - 10 to 13 (Class III A) plus 30% medical consultation LA Food Cost Index - Rp.l,200 in 1987. Source: Regulation 66/MENKES/SK/II/1987 (Pola Tarip Rumah Sakit Pemerintah). Table 5.2: SELECTED HOSPITAL TARIFFS (Rupiah) RSU Cipto RSU Mataram Class A, Jakarta Class B, NTB 1987 1984 Outpatient Visit 500 350 Inpatient Day LA Class III B 600 600 Class III A 1,800 1,500 Class II 12,000 2,500 Class I 21,060 6,000 Class VIP 30,420 LA Tariffs for Classes II, I and VIP include charges for medical consultation but exclude additional charges for drugs. Classes III A and III Bare exempt from charges for medical consultation and drugs. 5.03 For health centers the basic user charge comprises a fixed fee per outpatient visit which is supposed to cover medical consultation plus three days supply of drugs. Central government guidelines issued jointly by the - 87 Ministries of Health and Interior in 1977 1/' established a maximum health center fee of Rp.150 per outpatient visit, with exemptions for the certified indigent and those requiring immunization or treatment for communicable diseases. However, in practice many local governments set higher fees ranging from Rp.300 to Rp.l,OOO per visit. New guidelines issued in 1988 have raised the official outpatient fee to Rp.300 per visit. As with hospital tariffs, the health center fee is set far below the full cost of service provision. Recent estimates for health centers in four provinces show an average cost per outpatient visit of around Rp.1,370. Cost Recovery Ratios 5.04 Although the Indonesian health sector does have a system of user fees, it generates a low volume of revenues. Table 5.3 summarizes cost reco"\l'ery data for 1983/84-1985/86 by program. In 1985/86 revenues totalled only Rp.47 billion, equivalent to 10% of recurrent expenditure. The majority of he,alth sector revenues are generated by hospital charges, which accounted for Rp.35 billion or about three-quarters of sectoral revenues in 1985/86. The cost recovery ratio in the hospital subsector averages around 20%, or about: double the sectoral average. By contrast, only about Rp.2 billion was gene:rated by health centers, yielding a very low cost recovery ratio of 3%. However this figure may be biased by classification errors in the government accotlnts (see para. 3.11). Table 5.3: COST RECOVERY, 1983/84-1985/86 Total LA Revenue Revenue as % of CRp.billion) recurrent expenditure 1983/84 1984/85 1985/86 1983/84 1984/85 1985/86 HospUa1s 24.2 30.8 34.5 20.2 22.0 19.9 Health centers 1.8 1.0 2.1 3.8 1.6 3.0 CDC 0.0 0.0 0.0 0.0 0.0 0.0 Training 0.0 0.0 0.0 0.0 0.0 0.0 Othel:' 11.7 10.5 10.6 9.6 8,0 6.0 Tl...J. 42,4 li....l U ll....2 10.2 LA Includes central, provincial and district levels of government. Source: ANNEX I. 5.05 Cost recovery performance in Indonesia compares favorably with some developing countries. For example, cost recovery in the Philippines is estimated to have declined to about 5% of public sector spending on health in 1/ Regulation l79/KENKES/SK/VIII/77 (Pedoman Pelaksanaan Pemungutan Biaya Pelayanan Kesehatan), - 88 1985, However, Indonesian performance is modest compared to the polar case of China. A recent World Bank survey of Chinese hospitals showed cost recovery ratios averaging over ae% of rec.1,.1.rrent expenditure. These high ratios are achieved by restric;t;ing guve:n1m.ent: subsidies to labor costs, while recovering all other operation.al c(.'sts from service fees and for-profit drug sales. Introduction of a similar cost recovery policy for Indonesia would triple hospital revenlles from 8xound lip. 34 billion to Rp. 91 billion in order to cover current levels of nons~.1I:'.r.y recurrent expenditure on hospitals. Revenue Retention 5.06 Government regulat::i.ons require fee revenues to be remitted to the level of government which owns the facility. These revenues tend to treated as an earmarked charge rather than an addition to general revenues. As shown in Table 5 .l~, provincial goverrunents budget more (about three times as much) than they collect in revenues from the health sector. But district governments budget. less: overall the size of this negative local subsidy is small but in some distris II. 5.10 Affidavit of Indi&ency. Indigent person61 can their village chief or other authorized official to issue aE affidavit ,,fhich e}~empts them from paying fees for health services at all pub1.ichealth centers or hospi::als. The effectiveness of this affidavit is uncleal:. However, evidence that affidavits can be obtained ,those who need them badly is th'it a maj or reason given by some Dana Sehat leaCl.el'S! for nOI: to provide hospital coverage for participants in th6>8e insurance schemes is that the indigent can easily obtain letters of from payment from the village chief. - 90 C. ASKES Insurance for Government Employees 5.11 ASKES (formally BPDPK or Perum Husada Bhakti) is the compulsory health insurance system covering civil servants. active and retired. GOI has provided health benefits to civil servants since 1950. Initially, civil servants had the right to use any health care provider. and submit claims for payment to the Ministry of Health. As the number of civil servants expanded. this system became unsupportable. In 1968. a version of the current system was introduced. Premium payments were required and civil servants were allowed free choice of providers who submitted claims to the government for payment. The system underwent several reorganizations as a result of continued financial difficulties and eventually ASKES members were restricted mainly to government health facilities. In 1984 ASKES was reorganized as a parastatal company. Perum Husada Bhakti. which is responsible to both the MOH and the Ministry of Finance. which must jointly approve its budget. As a perum (public corporation), however. the organization has greater flexibility in management. especially personnel management, than a government agency. 5.12 In 1986. the ASKES rolls included 3,256.330 cardholders. including civil servants and pensioners. This represents an increase of about 35% over 1981/82 enrollment. For provinces providing a breakdown. 21% of cardholders are pensioners. It is interesting that a significant proportion of civil servants do not hold ASKES cards; in 1983/84. for example. civil servants holding ASKES cards number 11.7% fewer than the Civil Service Commission (BARN) count of 2.722.766. ASKES coverage also extends to the card holder's spouse and not more than three children. BARN reports an actual average of 2.22 dependents per active civil servant. This yields an enrollment estimate of about 10.5 million or about 6% of the total population. If pensioners have fewer dependents, this is an overestimate. ASKES statistics assume an average of chree dependencs per cardholder, yielding a total enrollment estimate of about 14 million or about 9% of the total population. 5.13 ASKES coverage entitles beneficiaries to free use of government health centers, including basic drugs. Many health centers offer special afternoon sessions exclusively for ASKES participants. These are supposed co be less crowded and more convenient than the morning sessions which are open to the general public. Parcicipants are also entitled to free use of govern ment hospitals. Until recently, participants were entitled only to Class III accommodations (the lowest category) without paying additional charges. Under new regulations, however. rank III civil servants (8.5% of the total) may receive Class II accommodations without copayment, and rank IV staff (0.6%) are entitled to Class I. Patients may upgrade their accommodations by paying the difference in rates, and many elect to do so. They can also choose to use private hospitals, but have to pay for any difference between the charges and the official ASKES reimbursement schedule. - 91 - RevenuE! and Expenditure 5.14 The system is principally funded by a payroll deduction of 2% of base salary for active employees and pension payments for retirees. These funds are collected directly by the central office of the Ministry of Finance. Premium revenue rose, in nominal terms, from Rp.18.14 billion in 1979/80 to a projected Rp.79 billion in 1986/87. ASKES also receives interest on its substantial time and savings deposits. In 1987 as in past years ASKES revenues will be more than sufficient to pay the costs of the system. However, ASKES reimbursement rates are based on subsidized tariffs which are set well below the full cost of provision of the services. For ASKES patients as for all other users there is, therefore, a government subsidy implicit in any USE of public health services at the official fee schedules. 5.15 The 1987 budget projects operational expenditure of Rp.84.6 billion, of which Rp.7l.4 billion is allocated to health care, and the remainder to administration, support, and depreciation. As shown in Table 5.5, ASKES support.s health care through three different types of payment: (a) Claims for medical service. ASKES reimburses claims for outpatient treatment, inpatient treatment and selected other medical services. These reimbursements account for a large proportion of total cost recovery in the health sector; (b) Pro,vision of dru&s. ASKES provides direct physical allocations of drugs to health centers and hospitals; and (c) Staff honoraria. ASKES also makes direct payments to health center and hospital staff to reward them for attention to ASKES patients, e.g., for offering special health center sessions in the afternoon. These payments are monthly lump sums, the size of the payment being tied to the type of facility regardless of the number of staff or number of patients served. Reimbursement Policy 5.16 Until recently. claims were made by health facilities to ASKES for reimbursement on a fee-for-service basis. This situation had several adverse consequ.ences for ASKES: (a) ASKES had little control over the level of fees set by local governments; (b) there was an incentive for local facilities to provide more billable services (e.g. radiological and laboratory exams) than optimal; (c) detailed itemization of charges led to costly and time-consuming claims-processing procedures. These problems are addressed by a set of new hospital reimbursement policies which became effective in 1987.21 Under these policies, ASKES will reimburse inpatient services on the basis of a ~ packet price per day. In order to contain costs, length of stay will be monitored and reimbursement subject to guidelines on maximum length of stay by diagnosis. The packet price is designed to cover three main components: (a) hospital services, including room, food, standard medicines, and use of equipment and operating room; (b) materials (e.g., X-ray film, laboratory supplies); and (c) medical services including surgery, medical consultations, laboratory and radiology tests and administration. Selected special services such as intensive care, heart operations, and hemodialysis will be billed separately. The packet reimbursement rates set for 1987 are shown in Table 5.6. Although these are no longer the same as official tariffs for non 21 Regulations 751/KENKES/SK/X/1986 on vertical hospitals and 68/KENKES/SKB/II/1987 on local hospitals. - 92 ASKES patients, it is clear that they are set far below the full cost of service provision. Table 5.7 shows average estimates of unit costs in a sample of government hospitals, which are around five times higher than ASKES reimbursement levels. As such they entail a large subsidy for ASKES users of hospital services. For example, ASKES reimbursement per inpatient day in a Class B hospital is Rp.4,500, compared to an estimated average cost of around Rp.25,OOO. Table 5.5: ASKES REVENUE AND EXPENDITURE, 1987 (Rp. billion) Revenue Premium contributions 79.0 Interest 8.0 Expenditure Reimbursement of Claims 31.4 Health center outpatient 4.8 Hospital outpatient 3.7 Hospital inpatient 15.2 Other i.J! 7.7 Drugs 33.8 Health centers 18.4 Hospitals 15.4 Staff Honoraria 6.2 Health centers 5.5 Hospitals 0.7 Administration i.J! Includes:Other medical services (e.g. hemodialysis), births, eyeglasses, prostheses and hearing aids. Source: ASKES · 93 Table 5.6: ASKES HOSPITAL REIMBURSEMENT RATES, 1987 Hospital Class Rp. per Inpatient Day Class A 7,500 Class B3 5,000 Class B2 4,500 Class Bl 4,000 Class C3 3,500 Class C2 3,000 Class Cl 2,500 Class Dl 2,000 Source: ASKES. Table 5.7: ESTIMATED lliilT COST OF HOSPITAL SERVICES (Rp. ) Per outpatient visit Per inpatient day Class B 9,862 24,469 Class C 3,691 13,052 Class D 3,948 12,554 Source: STATISTICAL ANNEX, Table 3.11. 5.17 The new regulations also introduced major changes in revenue retention policy. Vertical hospitals are formally allowed to retain all ASKES revenJ.e. ASKES is to make a monthly deposit against drug and material costs, rather than waiting to process claims. Local hospitals are required to pass to th,e local government the 30% of reimbursement allocated to the hospital services component. However, the remainder of the fee packet is to be retained at the facUity. 5.18 Utilization and subsidies. As shown in Table 5.8, ASKES utilization rates are high, especially for health centers. The health center utilization rate is about 2.5 visits per covered person per year, or about five times the national average. The hospitalization rate is about 38 per 1,000 persons enrolled per year, again about 5 times the national average of around 8 per 1,000 for Ministry of Health hospitals. However, it is not clear whether ASKES members use health services at a higher rate than private workers with comparable income levels. Most ASKES families are probably in the top - 94 quartile of the Indonesian income distribution and, indeed, ASKES members comprise a substantial proportion of this group. The combination of five times higher utilization, as might be expected from the zero net price facing ASKES users, and the large subsidy per unit of service, implies that ASKES members capture a disproportionately large share of public subsidies for health. As shown in Table 5.9, this per capita ASKES subsidy amounts to around Rp.6,500 compared to only Rp.l.200 per capita among the general non ASKES population. Table 5.8: ASKES MEDICAL SERVICES: QUANTITY AND COST IN 11 PROVINCES LA Utilization Total Unit rate (per 1,000 Monthly monthly cost per enrollees) per cases cost (Rp.) case (Rp.) year Lh Health center outpatient 918,141 142,753,689 155 2,511 Hospital outpatient 69,685 84,018,937 1,206 191 Minor surgery (outpatient) 9,162 9,796,500 1,069 25 Hospital care (kabupaten) 6,743 182,321,437 27,039 19 Hospital care (province/ central) 5,431 351,473,227 64,716 15 Maternity (hospital delivery) 1,542 26,911,499 17,452 4 LA Aceh, West Sumatra, Riau, Jambi, South Sumatra, West Java, Central Java, South Kalimantan, North Sulawesi, West Nusa Tenggara, East Nusa Tenggara. Number of ASKES cardholders in these 11 provinces - 1,362,336. Number of cardholders nationwide - 2,650,276. Lh Enrollees calculated using BAKN average family size of 3.22 per civil servant. Source: ASKES, "Laporan Penyelenggaraan Program Pemeliharaan Kesehatan Pegawai Negeri Penerima Pensiun and Keluarganya, Tahun 1985," Table 1. Unit costs recalculated from cases and total cost. - 95 Table 5. 9: All ESTIMATE or tIlE ASUS SI1BSIDY !umber Percent Shares ASJCES Ifon-ASJCES Total ASUS Ron-ASJCES Malbership 10,485,383 154,144,235 164,629,618 6 94 Utilisation rat, A~ Hospital Inpatients per 1000 14.86 1.29 2.16 C'D Hospital Inpatients per 1000 18.45 0.55 0.63 Total Hospital Inpatients per 1000 33.30 6.76 8.45 Hospital Outpatients per person 0.22 0.11 0.12 Health Cnter Outpatients per person 2.51 0.24 0.38 Total Utilisation A~ Hospital Inpatients 155,778 199,522 355,300 44 56 C.o Hospital Inpatients 193,410 841,750 1,035,160 19 81 Total Hospital Inpatients 349,188 1,041,272 1,390,460 25 75 Hospital Outpatients 2.261,573 17,291,899 19,553,472 12 88 Health Cn.ter Outpatients 26,335,090 36,566,585 62,901,675 42 58 12tal Costs ~!e. billionl 89.98 245.93 335.91 II II A'. Hospital Inpatients 33.78 43.26 77.04 44 56 C'D Hospital Inpatients 14.54 63.30 77.84 19 81 Total Hospital Inpatients 48.32 106.56 154.88 31 69 Hospital C~tpetients 13.03 99.62 112.65 12 88 Health CU.ter Outpatients 28.63 39.75 68.37 42 58 Iotal B!!.nues '!e. bllllonsl 22.12 62.19 !L..ll. 1§ l! A~ Hospital Inpatients iQ.Oi 12.91 22.99 44 56 C.o Hospital Inpatients 5.23 22.76 27.99 19 81 Total Hos~'ltal Inpatients 15.31 35.67 50.98 30 70 Hospital Outpatients 2.73 20.85 23.58 12 88 Health C~ter Outpatients 4.08 5.67 9.75 42 58 Ret Subsi~! 'Re. billLonsl 67.86 .!!!.:..ll ~ II II ... HospitaL Inpatients 23.7ii 30.35 54.05 44 56 C'D Hospital Inpatients 9.31 40.54 49.85 19 81 Total HosF.ltal Inpatients 33.01 70.89 103.90 32 68 HospLtal ClUtpatients 10.30 78.77 89.07 12 88 Health Cer,ter Outpatients 210.54 34.08 58.62 42 58 Bet SUbSl~! '!e. eer caeital 6,472 ~ 1.528 A'. HospUal Inpatients 2,260 197 328 C.o HospUal Inpatients 888 263 303 Total HosF"ltal Inpatients 3,148 460 631 Hospital (lutpatients 983 511 541 Health Center Outpatients 2,341 221 356 Source. ~orld Bank sta££ est~t' ·. D. PKTK Insurance for Private Employees 5.19 PKTK (Pemeliharaan Kesehatan Tenaga Kerja or 'health care for workers') refers to a government-sponsored pilot health insurance scheme for private employees. As such it is one element of the still nascent DUKK (Dana Upaya Kesehatan Masyarakat or 'funds for public health') policy framework which will encompass all health insurance activities in Indonesia. This framework will be based on the following principles: (a) mandatory, universal coverage through a variety of programs; (b) cross-subsidization of the poor by the better-off; (c) a mixture of public and private sector participation under - 96 government regulation; (d) an emphasis on programs offering capitated coverage of defined populations; (e) emphasis on a family doctor as entry point into the health system, with a well-defined referral system. 5.20 Implementation of the DUKM framework will ultimately involve drafting national legislation on health insurance and the establishment of a health insurance supervisory body. The extent to which the government will itself provide the envisioned insurance or health services is still under discussion. The voluntary PKTK insurance program is a private-sector analog of ASKES, providing health care for employees and their families, almost exclusively at government facilities at the regular subsidised tariffs, in exchange for employer-paid premiums. PKTK was initiated in Jakarta in April 1985, and has since been extended to 18 different regional schemes. PKTK: Jakarta Pilot Scheme 5.21 Management. PKTK is a collaborative effort between the Jakarta office of the Ministry of Health and ASTEK, a parastatal company under the Ministry of Labor which provides social security insurance for workers. ASTEK is responsible for marketing the program and for collecting the premiums. The Health Department, through an office called DUKM-Jakarta, is responsible for providing and paying for health services. Collected funds are divided as follows: 2% to the DUKM coordinating body; 8% to ASTEK for administration; Health for administration; 10% to the Ministry of Health for administration; 70% for health care. 5.22 PKTK offers comprehensive health insurance to participating firms' employees and their families. Table 5.10 shows the basis for calculating the premium. These are costs for medical care only. In Jakarta, an additional 25% is added for administration and contingencies. An average family size of 3.4 was assumed, and an average wage of Rp.llO,OOO/month. Based on these considerations, the Jakarta premium was set at 7% of wages, with an applicable salary range of Rp.50,OOO to Rp.300,OOO monthly. This covers the employee, spouse, and up to three children living at home. Additional family members can be enrolled at Rp.2500/month. Employers need not enroll all employees, and generally do not enroll management or other highly-paid staff. However, firms whose enrolled employees have an average salary under Rp.llO,OOO per month must pay an alternative minimum premium of Rp.2,500 per covered person per month. 5.23 Benefits. The initial health service point is one of 38 specially- designated health centers. These include some of the best in the city, including a newly built, Rp.500 million showcase center. These centers receive PKTK participants during special afternoon sessions, shared with ASKES participants. Like ASKES members, PKTK members receive basic medications without additional charge. In case of hospitalization, PKTK members are entitled without cost to either class II accommodations at government hospitals or class III accommodations at private hospitals. - 97 Table 5.10: EXPECTED COSTS OF PKTK EXPERIMENT (Rp. per enrollee per month) Jakarta Semarang Medical and paramedical services 500 462 (health center) Dental services 125 Medicine 500 Inpatient care 416 215 Specialist consultation 32 Diagnostics 26 25 Operations and special care 65 82 Matert1.ity 20 15 Eyeglasses and hearing aids 11 22.5 Other 84 25 Total 1,779 841.5 Note: Categorization of service may not be strictly comparable between Jakarta and Semarang. Health center drugs apparently not included for Semarang. 5.24 PKTK reimburses participating health center staff on a capitation basis. To allow for low utilization during the program's startup, capitation is on a sliding scale, with a lump sum of Rp.150,OOO/month for 250 enrollees or less, Rp.400 per enrollee per month for the next 250 enrollees, and Rp.200 per additional enrollee per month for up to a maximum registration of 3,000 enrollees (including family members) per health center. The supervising doctor receives 60 to 70% of these funds, with the remainder distributed among paramE!dical staff. Drugs are funded separately. PKTK reimburses participating hospitals, public and private, for inpatient care at a fixed rate (If Rp.27,000 per day, including drugs, food, and medical service. This rate ts substantially higher than the standard ASKES reimbursement rates established in 1987. However, there are indications that this reimbursement rate i.s not sufficient to achieve full cost recovery for the level of services expected. This signals a potential danger with projected expansion of PKTK coverage. Unless reimbursement rates are carefully set to recover actual costs the additional utilization of public facilities that will be induced by provision of free coverage for PKTK beneficiaries may increase the need for publi(~ subsidies instead of reducing it. 5.25 Participation has been far below original expectations. In February 1987, almost two years after the program's inception, there were 51 participating companies and a total enrollment of 11,000 individuals (including workers and their dependents). About 90% of companies approached by ASTEK declined to join PKTK; four companies joined but later dropped out (three as a result of bankruptcy). Participating companies are small; of the - 98 companies enrolled at mid-year 1986, 82% had fewer than 100 employees, and 27% had fewer than 20. Almost all are service-sector companies rather than manufacturing industries. 5.26 Utilization of health centers has been slightly below ASKES levels, at 0.168 visits per enrollee per month. Most of the visits have been concentrated at 11 of the 38 participating health centers. Given fixed costs of administration, and the sliding scale for capitation, including a guaranteed minimum payment, the current enrollment level is substantially below the break-even point. Compounding PKTK's fiscal difficulties is its lack of control of drug use. Health center drug usage is currently running at Rp.3,330/visit, perhaps partly reflecting diversion of PKTK drugs to usage by other health center clients. PKTK management hopes to be able to reduce costs to Rp.2,OOO/visit in the near future, and eventually to implement a capitated system of drug allocation to the health centers. Causes of Low Enrollment 5.27 Voluntary enrolment in PKTK programs relative to ASTEK membership is very low. By 1988 PKTK covered only around 90,000 private sector workers and their dependents, or less than 1% of the entire population. This compares to an actual ASTEK membership base of 3 million employees, and a potential base of 5.5 million. 5.28 The marketing plan for PKTK appears to have been based on three premises: (a) most formal sector employers offer health benefits to their employees; (b) employers have difficulty containing the cost of these benefits; (c) PKTK, through efficient management, can offer equivalent health care at a lower price. Premise (a) is strongly supported by a 1984 ASKES sponsored survey of private employers in Jakarta; 171 out of 173 were found to provide health benefits to their employees. Premise (b) is plausible; for instance, many companies were found to be unable to provide breakdowns of health expenditure. Companies which were able to provide breakdowns reported spending a large proportion on drugs. Premise (c), however, requires a leap of faith. In the 1984 survey, average health expenditure per employee was about Rp.132,OOO per employee. PKTK proposes to provide equal or better quality service for an average of Rp.92,400 per employee, of which 30% is deducted for administration and reserves. 5.29 The lack of response to PKTK's offer almost certainly indicates lack of acceptance of premise (c) by potential participants. Some detailed hypotheses about the nature of that failure include: (a) PKTK is expensive relative to self-insurance. A comparison with ASKES is useful here. PKTK and ASKES provide roughly comparable benefits to employees, but ASKES costs 2% of wages while PKTK costs 7%. Private employers can and do achieve comparable coverage simply by paying their employee's charges at government health centers and hospitals, and selected private hospitals. In doing so, the employer takes advantage of the subsidized fee levels for the general public. As long as PKTK remains voluntary and government health facility prices remain low, this will be a rational decision. - 99 (b) B~alth centers are unattractive to employees. From the employees' viewpoint, a well-maintained, reliably staffed clinic at the workplace is far more attractive than a public health center with limited hours. Although PKTK has taken steps to improve health center conditions and image, their general appearance remains inferior to that of most private clinics. Doctors sometimes do not show up for the PKTK sessions, or are present for less than the requisite three hours. The ratio of patients to staff is often higher at the exclusive afternoon session than at the morning sessions for the general public. PKTK is aware of these problems. One proposed solution is to offer PKTK services at some of the private 24-hour clinics now open in Jakarta. Another is to permit employers to use in-house clinics for primary health care under the umbrella of DUKK coverage. (c) Some employers prefer to offer higher-quality benefits. Pertamina is an obvious example of a firm preferring to offer better care than is available through PKTK; it has traditionally offered its employees health care considerably better than that available in public facilities, at a cost per employee several times the level of ASKES or PKTK. Many companies now offering high quality care simply will not desire to reduce the quality of care offered. In principle, PKTK could respond to the market by offering a variety of coverage plans at different premium rates, but it has not yet done so. (d) Employers are waiting for PKTK to demonstrate viability. Still in its pilot stages, PKTK has had various operational problems. The future of PKTK, and indeed of the Indonesian health insurance industry, is still in flux. Under these conditions, companies may be understandably reluctant to dismantle a working health care system in favor of one with an uncertain future. 5.30 These hypotheses about the causes of lack of success of the pilot DUKK project in Jakarta are mutually compatible, and it is likely that all are to Sl)me extent valid. To the extent that hypothesis (a), high cost relative to ~~at is available to all, is the main explanation, PKTK will become more successful if hospital and health center fees are raised so that costs to the non-insured increase. To the extent that the hypotheses of low quality and of uncertain viability are true, the problems of PKTK are probably more deep seated, and need to be resolved before the system can expand. These latter hypotheses, if true, would also suggest that it may be advisable not to make EKIK mandatory, at least under the present plan in which only public sector providers are allowed. PKTK is a potentially valuable effort, but should be regarded as experimental. Caution should be exercised before replacing the existing, well-functioning system of employer health benefits with an experiment still in process. Providing Insurance for Wage Earners 5.31 DUKK's general principles, including the goal of mandatory insurance for the wage-based sector are appropriate. This group, which may comprise more than 20 million individuals (including both wage-earners and their dependents), is a disproportionately heavy user of curative health services, - 100 and therefore a prime recipient of a large and regressive public subsidy for those services. These families can afford to pay the full cost of their health care provided that they can insure against the costs of catastrophic illness. Provision of insurance is therefore a crucial complement to a policy of raising health facility fees. (a) Desian and incentives. The design of a health insurance system has important incentive implications for both participants and providers. Recognizing this, DUKK places great faith in capitated funding of providers as a cost-containment mechanism. However, in any insurance system (especially one employing capitated payments) competition among insurance carriers and among health providers is essential. A monopoly insurance carrier paid by capitation does have an incentive to contain costs, but no incentive to pass these savings to the consumer. The incentive to contain costs may jeopardize quality; without competition, there is no automatic check to make sure that quality standards are upheld. Competition is equally important in the context of fee-for-service reimbursement. A monopoly insurance carrier has no incentive to minimize administrative costs, or to set premiums in accordance with expenses. For this reason, the Government should not impose a single, government-run insurance system on the wage-based sector, or allow a private provider to monopolize the market; there would be a tendency for this to result in an inefficient, unresponsive bureaucracy. It is especially important not to scrap the current, working system of employer self-insurance without a demonstrated alternative. (b) Role of Government. The government's role in the insurance market is an important one, however. At a minimum, the government must set up the legal and regulatory framework for the market, including the requirement that all wage-earners be insured. The government may want to act as a collector of premiums, allowing firms or individuals to apply those premiums to anyone of a competing number of plans. The advantage of this scheme is that it facilitates cross-subsidization of the poor by the better-off through differential premium rates, while maintaining the advantages of competition. The disadvantage is that it may be difficult to implement, judging by the example of ASTEK. Although all employers are theoretically obliged to belong to ASTEK, in fact only a small proportion are members. The majority of employers apparently view the benefits of ASTEK membership as smaller than the costs. 5.32 PKTK-type government insurance systems may be able to play their most important role if they are only one of a number of competitors. Current PKTK schemes require management development as well as further experimentation with premiums, benefits, and choice of service providers. · 101 · E. Rural Health Insurance: Dana Sehat Dana Sehat in Theory 5.33 Dana sehat (or "health funds") are village·level organizations intended to insure village members against the costs of primary health care. Small fixed contributions are levied on each family; the proceeds are primarily used to pay basic health center charges. In addition, the funds may be used to finance public health activities or household income-generation projects. 5.34 A handbook with suggested guidelines for rural dana sehat was published in 1986 by the MOH Directorate for Community Participation. In this conception, dana sehat are not standardized, and they are not officially linked with the government. Rather, they are to be designed and operated by the village, with the advice and technical assistance of the local health center. The handbook suggests piggybacking dana sehat on existing economic, social, or religious organizations. It presents a hypothetical illustration of the economics of a dana sehat, summarized in Table 5.11. This simple example does not allow for secondary level outpatient care, or for management costs. Nonetheless, it suggests that, in theory, dana sehat should be able to provide coverage at very low cost. The premium, equivalent to Rp.40 or 50 per person per month, is small even compared to the fourth percentile rural consumption level (1984) of Rp.5,OOO per person per month. It is important to note, however, that a working system of dana sehat would not increase cost recovery by the government unless prices were raised. The dana sehat would simply collect money for a fund to pay costs already being paid by the individuals who would be members of the dana sehat. Table 5.11: DANA SEHAT: AN ILLUSTRATION Initial contact with health system: health volunteers (kaders). Assumed morbidity rate - 14% per month. Sick people are supposed to consult first with their neighborhood health volunteers, who are supplied with very basic drugs (e.g. aspirin). Monthly cost: 280 patients x Rp.75/case - Rp.2l,OOO. First referral level: health center. Kaders refer 50% of their patients to the health center for treatment. The dana sehat pays the standard health fee of Rp.150 (this is the standard, subsidized price which an uninsured member of the general public would pay). Monthly cost: 140 cases x Rp.150/case - Rp.2l,000. Second referral level: district hospital. Health centers are assumed to refer 2% of visitors to the local hospital for inpatient treatment. Monthly cost: 3 cases x 8 days/case x Rp.1500/day - Rp 36,000. Required premium per family - Rp.(2l,OOO + 21,000 + 36,000)/400 Rp.195/month. Source: Directorate for Community Participation, Ministry of Health. ~ 102 ~ 5.35 Some policymakers hope to establish a goal of universal rural coverage by dana sehat. In this vision, the dana sehat would be tied into nationwide health insurance, under the umbrella of DUKK; presumably this mechanism would allow some cross-subsidization of rural health care. Detailed plans for achieving this goal have not yet been drawn up. Dana Sehat in Practice 5.36 Dana sehat first emerged in the 1970's, principally in Java and Bali. For the most part, these were spontaneous or NGO-sponsored efforts rather than government operations. A few villages with successful dana sehat attracted extraordinary attention. Unfortunately, the resulting popularity of the dana sehat concept may have fostered unrealistic hopes. Despite recent government encouragement, dana sehat have diffused very slowly. A recent estimate by the Ministry of Health placed the number of dana sehat at about 600 in nine reporting provinces; it is likely that some of these are inactive. In Bali, where social cohesion is very strong and the dana sehat concept might be expected to take root, the famous example of Pejaten village has inspired little emulation. Pejaten's success appears to be more a result of good economic fortune than of the dana sehat organization itself. As a result of a favorable market position in roof tiles, the village became wealthy and was able to set up an endowment for health. In general, the lack of diffusion of dana sehat indicates either low perceived benefits, significant organizational barriers, or simple inability of rural dwellers to pay useful amounts of money. 5.37 The operation of dana sehat can be illustrated by the experience of Karanganyar kabupaten (central Java), where 86 of the 177 villages have some kind of health fund. In Kerjo kecamatan (population 32,000), dana sehat coverage is nominally universal. The mechanics of the system are as follows. Villages or subvillages collect a fixed monthly contribution of Rp.50 or 100 from each family. Parallel funds have been set up among school children; some of these are financed by the sale of firewood or building stones gathered by the children. Sick people are supposed to consult first with their neighbor hood health volunteer. A volunteer covers approximately 30 households; in a month she might receive twelve patients, and write two or three letters of referral to the health center. These letters excuse the bearer from the Rp.150 health center fee; in turn, the health center can claim Rp.lOO from the patient's dana sehat. 5.38 Approximately half of the health center's 1,400 to 1,800 monthly visits are covered by dana sehat. The remainder includes about 180 visits by ASKES members, and self-referrals by members of the general public, who could not find or did not seek a health volunteer. About 1% of patients are referred to the district hospital. The costs of their treatment, which might total Rp.50,OOO for an episode of dengue fever, are not covered by dana sehat. However, hospital patients from Kerjo can generally obtain a certificate of poverty excusing them from payment. 5.39 This dana sehat arrangement imposes a tremendous administrative burden on the volunteer staff, who must not only gather and disburse funds but also keep detailed financial and patient records. However, the burden is clearly unsupportable in poorer villages, or where a strong, enthusiastic - 103 village chief is lacking. In their present form, dana sehat impose tremendous nonfinancial costs and in return yield only small financial benefits. The current schemes have the following disadvantages: (a) Large transaction costs. Very small sums of money must be gathered each month from a very large number of families. This money must be considered, accounted for, and disbursed against health center claims. At the health center, patients have to be recorded on special registers for each village, and claims must be drawn up. It is difficult to estimate the number of paid and unpaid person-hours involved in administration, but it is clearly substantial. Even valued at the subsistence wage of Rp.lOOjhour, unreimbursed time costs are probably large in relation to the size of the funds. (b) Minimal gains from insurance. Insurance is most beneficial when applied against catastrophes: low probability, high-cost mishaps. Dana sehat, however, insure against high probability (approximately 0.5 times per year), low cost (Rp.150 or the cost of three cigarettes) events. In other words, even very poor families can self-insure against health center costs. (c) No contribution to health center cost recovery. As generally set up, dana sehat substitute for, rather than augment, the standard fee paid by the patient to the health center. They may possibly increase health center utilization and revenues, since visits are free to insurers at the point of service. However, it is likely that the marginal revenue from these visits is insufficient to cover the marginal costs of treatment. 5.40 Providing rural health insurance. Implementation of rural health insurance is a much more difficult task than is provision for wage earners. It is important to note however that low income itself is not necessarily a binding constraint, given the modest contribution level required. Table 5.12 shows a rough estimate of the cost of catastrophic insurance for the rural population using actuarial data based on the experience of ASKES. The unit costs are standard health facility fees, which are less than full cost. However, even if true costs are four times greater, a premium of Rp.500 per person per month should be affordable by the majority of the population. Placed in perspective, rural families in the Rp.8,OOO-lO,OOO monthly per capita bracket of total household expenditure (about 9th to 23rd percentile) consumed an average Rp.4l4 worth of tobacco and betelnut per capita per month in 1984. Those between the 40th and 70th percentile consumed Rp.l,008 per capita per month. The principal obstacles facing rural insurance are ~, a lack of consumer demand for insurance against rare but catastrophic expenses; and second, a lack of an efficient administrative mechanism for collecting and managing funds. These obstacles are reflected in the operation of dana sehat. As a result of the lack of demand for hospitalization insurance, dana sehat tend to concentrate inefficiently on insurance against small, routine health center fees. At the same time, the dana sehat must expend a tremendous amount of organizational effort and volunteer time on frequent collection of very small premiums. Is!J:>le.~12: HYPOTHETICAL COST OF HOSPITALIZATION INSURANCE Frequency Expected cost per person Unit cost per person Service type per month (Rp. ) (Rp. ) Hospital outpatient vii>it: .0128 1,202 15.4 Inpatient car"! (kabt,p~_t.en level) .0012 26,835 32.2 Inpatient care (provi.n('.e/"'~ntr'Rl) 00010 64,202 64.2 Total (excluding drugs) 111.8 Total (includi.ng drugs) 138.6 Source: ASKES, Laporan P,:myelenggaraan Program Peme1iharaan Kesehatan Pegawai Negeri, P~.ne:rima Pens:hm ",rod Ke1uarganya, Tahun 1985; Table 1. Frequency -9.';S'.''1l''l:\, ~:::~,:n~:;\~!? L':'.!Uily size of 4.0. 5.41 There are tlH:.~e. maJ.Q policy problems associated with present health insurance arr~l.lgeme:o.'ts i~lJ_ ~~~l;J.'S)D,9s~_.a." Resource MobilizaugI1 5.42 Existing health tnS~JD3.nce arrangements are not an effective resource mobilization instn1.illent b~t:!alJ.se they all reimburse public services on the basis of ordina:ry t~xiffs '\ih:i.ch !;Ire heavily subsidized. On average this subsidy amounts to> around 75% of the cost of delivering public services. The effect of these insurance ~.r:r.angements is only to lower the net price to users by pooling financial r:lsks ,"!nta.Hed by existing public tariffs. It does not alter the prices :r.ecelverl pl.).blic facilities on the supply side. As a result this arra.ng-elll'Snt does not j.ncrease the volume of revenues generated by public facilities, ~xcept :i.nsofa,r as the zero net price charged to insured users increase~ the dem~nd for public services. For example, the utilization rates of ASKES beneficfa:des "I.re about five times higher than non-ASKES users. However. because of the s'll,os:i.dized reimbursement policy, any such increase in public revenues entails ,fl, pn)'por.:tionately larger requirement for additional public subsidies. Thl.~S the net fiscal impact of existing insurance arrangements tends to be sl.d)stlitD.tially negative. This reflects a failure to link pricin"&'...Ro1icy: ,,,i.th the provision of insurance coverage in pursuit of the resource mobHizaHon objective; this could be achieved by differential pricing with full co.st·,recovery tariffs for the insured. In addition, there may be resource mobilization lo~ses for insurance schemes which are not financially viable even r;,7ith snbsidized reimbursements to the public sector. This appears not to be the case for: ASKES but probably is the case for some PKTK and dana sahat schemes. 5,43 resource moM.lizatioH impact of any future extension of health ins\,m"i!9.1.C·" jJ:,s low base is necessarily limited by congtr~j.T\·t"" These reflect both institutional and 105 econe,mic factors. Institutional constraints include the low compliance of private sector employers with existing mandatory provisions for social security coverage, and the lack of an organized transactions base for colle.cting insurance premiums in the rural economy, Economic constraints reflect the trade-off between cost recovery and cleilliind for coverage. Unless insurance reimbursements are based on a level higher than presently subsidized pubU.c sector tariffs. expansion of insurance coverage would tend to have a negat:ive net fiscal effect instead of making a positive cOl1t:r'ibution to resource mobilization for the health sector Hc.rwever, if cost recovery from 0 insurance is increased the premium may become 'U"oaffot'dable, resulting in lower demand for voluntary coverage. Effie: iency 5.44 Existing health insurance arrangements are characte:rized by several efficiency problems. The ASKES and PKTK schellies are faced with a lack of effie!iency incentives on the demand side, providing f:f.rst~rupiah insurance coverage without any copayments to restrain utilization (deductibles or coinsurance). They also face a lack of efficiency incentives to restrain supplier-induced demand (such as limits on re:i..mhuL'seruent per diagnosis). The rural dana sehat are typically organized on too small a basis (village level) to exploit fully the efficiency benefits of risk'pooling, and are generally restricted to an inefficient emphasis on low-cost outpatient coverage (instead of p(ltentially catastrophic high-cost inpatient coverage) which does not expl(lit the welfare gains from insurance. 5.45 The tendency of existing insurance a::crang8li!ents to have a l1egative net fiscal impact means that they draw morepubli.::: JLasour.:::es to fhlance services for the better-off with insurance, thus the poor of the benefits which could be provided by those publie funrt;]. Subsidized health services therefore disproportionately benefit the bett.n·~off. Thus, the net subsidy per capita for ASKES users of public d6ctor hospitals aXld health centt~rs is about five times higher than for the l.'est of the population. Moreover, efforts to extend insurance coverage to t'hl 8,878 South Sumatra 72,088 Lampuns 105,080 West KalLmantan 8.389 Central Kalimantan 45,721 South KalLmantan 215,549 East KalLmantan 58,451 Morth Sulawesi 40,300 Central Sulawesl 10,448 South Sulawesi 28.412 Southeast Sulawesl 4,000 Maluku 566,861 Ball 0 West Musa Tenssara 101,474 East Musa Tenssara 22,191 Irlan Jaya 127,542 Benskuku 391,436 East Timor 0 - 123 ANNEX I Table 14 of 20 cmmtAL GOVElUIHBlIT IIPRES EXPDDI'l'UIB BY PIOI,;IlAM AID 1'IIOVIIrCB, 1984/85 (BUDGEt) (Ip. tbouaancla) SalarLea MaterLals Tranaport Other LaM Coaatzuct LOll EquLpment Total Proaram J! 40,344,!i02 J! J! .!l ~I,124,397 J! 98,441!. 999 Hospitals 0 0 0 0 0 0 0 0 Health Centezs 0 40,344,602 0 0 0 37,655,398 0 78,000,000 CDC 0 0 0 0 0 0 0 0 Tralnlna 0 0 0 0 0 0 0 0 Other 0 0 0 0 0 20,448,999 0 20,448,999 Provlnce J! 40.344,602 J! .!l .!l 58,124,397 J! 98,441,999 Central 0 0 0 0 0 10,080,065 0 10,080,065 DIU Jakarta 0 1,285,183 0 0 0 1,400,391 0 3,225,574 West Java 0 7,434,888 0 0 0 5,076,461 0 12,511,349 Central Java 0 6,669,350 0 0 0 5,798,055 0 12,467,405 YOllyakarta 0 710,998 0 0 0 1,146,213 0 1,857,211 East Java 0 7,634,798 0 0 0 5,855,670 0 13,490,468 01 Aceh 0 726,207 0 0 0 1,534,800 0 2,261,007 Horth Sumatra a 2,310,179 0 0 0 1,526,629 0 3,836,808 West Sumatra a 972,615 0 0 0 1,439,558 0 2,412,173 Rlau 0 607,150 0 0 0 1,307,784 0 1,914,934 Jamb 1 0 407,999 0 0 0 997,900 0 1,405,899 South Sumatra 0 1,278,206 0 0 0 1,304,199 0 2,582,405 Lampung 0 1,369,735 0 0 0 1,659,779 0 3,029,514 West Kalimantan 0 763,585 0 0 0 1,425,839 0 2,189,424 Central Kallalantan 0 305,169 0 0 0 914,076 0 1,219,245 South Kal1.mal1 tan 0 552,153 0 0 0 1,425,169 0 1,977,322 East Kalimantan 0 424,044 0 0 0 1,244,198 0 1,668,242 Horth Sulawed 0 567,140 0 0 0 1,093,849 0 1,660,989 Central Sula'Wesl 0 361,525 0 0 0 646,646 0 1,008,171 South Selavesi 0 1,607,995 0 0 0 2,016,955 0 3,624,950 Southeast Selaveal 0 261,709 0 0 0 665,306 a 927,015 Haluku 0 ...... ,909 0 0 0 1,275,394 0 1,720,303 Ball 0 6"9,865 0 0 0 937,704 0 1,587,569 West Husa Ten.llara 0 730,634 0 0 0 1,145,589 0 1,876,223 East Husa Ten.llara 0 834,393 0 0 0 1,598,013 0 2,432,406 Irlan Jaya a 375,773 0 0 0 2,198,399 0 2,574,172 Benakulu a 219,355 0 0 1,300,973 0 1,520,328 East Timor ° 299,045 ° 0 ° 0 1,088,783 0 1,387,828 - 124 ANNEX I Table 15 of 20 CENTRAL GOVERNMENT ROUTINE EXPEllDITURE AND REVEIIUB BY PROGRAM .AlII) PIIOYllfCE, 1983184 (ACTUAL) (Rp. tbouaaDCIa) Salaries Materials Mainte_e Traneport Othar Total Revenue Program 43,849,162 32,6~9.775 6,8~6,9~i 1 67 , 20 2 1,891.~14 16.1&4,596 1,714,814 Hospltal 20,811,375 23,757,573 4,130,866 60,138 1,867,471 50,627,423 7,978,465 Health Centers 755,618 475,940 183,081 12,890 0 1,427,529 141,400 CDC 41,537 5,716 11,879 526 0 59,658 Tralnlna Other 3,166,971 19,073,661 3,401,092 5,019,454 641,181 1,889,929 40,060 753,595 13,925 10,118 7,263,229 26,746,757 ° 32,821 562,128 Provlnce 43.849.162 32,659,775 6,856.936 867,209 1.191,514 16,124,596 8,714.814 Central 11,992,016 2,357,831 989,254 543,316 10,118 15,892,535 237,413 OIt! Jakarta 7,242,539 8,087,983 2,005,431 12,291 1,548,584 18,896,828 4,059,523 West Java 4,789,492 5,043,116 819,049 27,096 0 10,678,753 1,258,258 Central Java 4,941,116 4,923,050 801,376 28,028 282,284 10,975,854 1,662,477 Yos"akarta 1,442,953 1,286,613 336,889 13,195 0 3,079,650 282,619 East Java 3,401,592 1,698,614 221,199 31,084 13,925 5,366,414 139,725 01 .leeh 254,968 226,070 55,584 9,505 546,127 14,894 North Sumatra West Sumatra 607,746 1,352,429 581,222 1,547,881 101,325 261,632 13,096 14,528 ° ° 36,603 1,303,389 3,213,073 40,967 253,464 Rlau 223,522 116,486 35,381 9,201 0 384,590 5,600 Jambl 205,734 85,902 20,320 4,347 0 316,303 1,948 South Sumatra 1,593,899 1,974,021 382,673 11,879 0 3,962,472 301,244 Lampuna 298,048 168,652 610,894 4,456 0 516,050 2,526 West Kalimantan 328,716 313,932 59,229 7,146 709,023 18,465 Central Kalimantan South KalLmantan 111,801 349,340 81,641 279,291 21,299 51,692 6,243 8,080 ° 0 0 220,984 688,403 3,241 27,838 East Kalimantan 258,283 200,094 53,759 8,693 0 520,829 21,102 North Sulawesl 503,232 345,812 59,467 11,721 920,232 13,228 Central Sulawesl South Sulawesl 100,337 674,323 99,208 656,670 23,245 120,03. 8,205 16,772 ° ° 230,995 1,467,799 .,919 39,496 Southeast Sulawes1 Maluku 160,277 228,644 97,315 151,421 23,792 42,671 6,964 8,672 ° ° 0 288,3.8 431,408 5,1079 2,624 Ball 1,607,291 1,42.,382 169,112 9,680 3,210,465 286,059 West Nusa Tenasara East Nusa Tenssara 169,238 193,583 138,523 114,83 27,533 22,075 5,803 9,637 ° ° 0 341,102 340,129 13,747 4,495 lrlan Jaya 130,391 135,634 28,176 16,675 0 310,876 11,951 Benakulu 120,138 85,113 19,742 3,681 0 228,674 1,390 East Timor 567,514 438,459 60,103 17 ,215 0 1,083,291 122 - 125 ANNEX I Table 16 of 20 PROVINCIAL GOVElUIMENT ROUTINE EXPENDITURE AIm REVEIIUE BY PROGRAM AIm PROVINCE, 1983/84 (ACTUAL) (Rp. thousands) Salaries Materials Maintenance Transport Other Total Revenue Proa r !!!! 52,339,804 16,020,185 3,391,366 392,122 1,991,986 14,148,063 8,082,369 Hospitals 16,080,333 12,414,816 2,389,439 101,914 1,128,611 32,115,239 6,169,109 Health Centers 24,383 20,540 1,491 442 250 41,112 18,433 CDC 90,502 11.854 3.885 4,540 0 116,181 0 Training 0 0 0 0 0 0 0 Other 36,144,586 3,501,515 1,002,545 285,166 269,119 41,208,931 1,894,821 Province 52,339,804 16,020,185 :->,391,366 392.122 1,991,986 14,148,063 8,082,369 Central 0 0 0 0 0 0 DXI Jakarta 859,135 1,666,342 83,621 313 0 2,610,017 698,110 West Jav. 8,353,344 64,186 10,309 12,242 0 8,440,081 0 Central Java 10,406,158 115,960 1,114,401 23,243 131,013 12,451,435 1,332,125 Yogyakarta 1,844,625 82,015 3,984 0 2,611 1,933,295 10,258 East Java 7,724,323 4,619,600 415,978 17,711 1,207,610 13,985,288 1,141,846 DI Aceh 495,922 386,141 91,551 18,734 23,111 1,015,519. 114,505 North Sua,atra 3,462,439 2,323,134 373,135 59,934 43,140 6,261,782 996,438 West Sumatra 747,538 436,053 111,601 15,999 6,805 1,317,996 123,464 Riau 282,238 200,349 11,033 3,751 43,331 540,108 91,893 Jambi 335,240 41,069 1,725 18,262 37,621 433,971 39,264 South SUIIo&tra 830,068 44,403 7,545 11,802 3,699 897,517 7,088 Lampung 1,893,002 663,953 63,186 5,427 196,181 2,821,7119 546,115 West Xal1mantan 1,025,211 459,185 7,701 25,501 8,305 1,525,903 113,850 Central Kalimantan 276,131 110,392 12,191 15,851 39,618 453,849 63,141 South Xalimantan 1,530,195 374,889 26,136 12,149 10,348 1,954,311 161,812 East Xal1mantan 853,186 429,108 19,761 16,438 3,255 1,322,348 339,338 North Sulawesi 1,112,561 610,514 110,104 6,863 26,043 2,526,085 382,032 Central Sulawesi 939,412 149,181 100,010 4,021 0 1,192,624 166,815 South Sula_si 1,201,208 1,052,837 41,588 22,209 0 2,329,842 244,202 Southeast Sula_si 353,296 31,842 44,423 1,848 1,111 444,586 43,032 MaluJc.u 393,666 351,369 14,267 5,633 3,415 168,410 90,966 Bali 2,901,558 517 ,313 14,467 7,133 3,864 3,444,335 341,980 West Hus. Tenggara 500,921 208,243 22,309 0 61,903 793,382 147,109 East Hus. Tenggara 1,151,286 110,619 8,359 20,110 56,069 1,352,443 126,135 Irian Ja,a 1,083,816 839,579 710,789 51,192 31,496 2,080,872 127,416 Bengkulu 524,593 20,749 2,080 6,150 58,065 611,631 25,499 East Timer 584,926 45,100 4,500 3,000 0 638,126 0 - 126 ANNEX I Table 17 of 20 01 STRICT GOVERNMEBT llOUTIlfE EXPElfDITURE AlfD llEVElfUE BY PROGRAM AlfD PROVIBCE, 1983/84 (ACTUAL) (Rp. thousaDds) Salarles Materlals Malntenance Transport Other Total "venue Proar . . 61,669,415 9,904,006 4,~66,874 1,544,377 ~,§~5,941 !O,340,613 ~Q,808,~~Q Hospltals 20,367,270 5,402,800 1,804,389 456,005 1,646,726 29,677,191 9,945,339 Health Centers 5,841,476 821,491 188,811 61,817 111,286 7,024,881 1,651,573 CDC 85,323 4,170 0 2,709 1,018 93,220 0 Tralnlna 2,059 569 123 48 0 2,799 0 Other 35,373,288 3,674,977 2,573,551 1,023,797 896,910 43,542,523 9,211,338 P£!!l:lnce 61,669,415 9,904,006 4,566,874 1,544,377 2,655,947 80,340,613 20,808,250 Central 0 0 0 0 0 0 0 Ott! Jakarta 0 0 0 0 0 0 0 West Java 7,039,030 2,105,880 1,469,052 136,821 989,867 11,740,650 5,382,340 Central Java 8,604,127 1,975,148 901,755 67,298 255,302 11,803,631 5,832,228 Yoayakarta 499,926 92,699 29,637 4,145 354,851 981,259 367,473 East Java 9,775,637 1,305,568 211,001 140,134 417,304 11,849,643 3,149,792 01 Aceh 2,027,758 209,337 22,274 39,724 22,547 2,321,640 198,084 Borth Sumatra 5,828,021 472,733 78,304 43,173 84,906 6,507,136 522,360 West Sumatra 1,912,264 168,231 26,808 18,290 36,026 2,161,619 325,183 Rlau 2,751,206 182,763 343,933 511,001 96,887 3,885,790 221,410 Jamb 1 779,167 201,275 34,179 11,542 44,007 1,070,170 103,649 South Sumatra 2,470,313 227,635 149,201 40,618 58,360 2,946,127 313,884 Lampuna 896,483 71,979 20,706 1,435 0 990,603 330,627 West Kalimantan 734,824 217 ,625 2,895 12,090 8,766 976,200 219,527 Central KalLmantan 2,186,679 110,773 15,887 33,057 13,250 2,359,646 108,368 South Kalimantan 623,349 92,042 34,531 12,238 35,816 797,975 151,485 East Kalimantan 979,135 132,260 35,669 27,289 44,107 1,218,459 69,335 Borth Sulavesl 1,401,853 321,851 17 ,580 26,183 5,698 1,773,164 1,118,721 Central Sulavesl 862,171 125,159 25,447 18,318 25,278 1,056,372 424,001 South Sulavesl 3,875,184 368,984 36,652 24,014 20,544 4,325,378 539,594 Southeast Sulavesl 785,167 767,547 934,787 279,032 728 2,767,262 76,693 Malultu 1,224,912 150,983 14,789 23,302 6,423 1,420,409 124,311 Ball 0 0 0 0 0 0 0 West Husa Tenaaara 1,359,941 142,057 48,194 6,027 16,724 1,572,943 142,105 East Husa Tenaaara 1,308,987 249,893 92,401 38,681 61,828 1,751,790 560,140 Irlan Jaya 3,339,039 203,137 19,271 28,485 48,631 3,638,564 492,696 Benakulu 404,244 8,446 1,923 1,479 8,093 424,185 34,244 East TlIDor 0 0 0 0 0 0 0 - 127 ANNEX I Table 18 of 2Q CIJITR.AL 00'VER.HMEIn' DEVELOPHDl'1' EXPElDI1'URB BY PIlOGlWf AlID PROVIRCE, 1983/84 (BUDGET) (Rp. tbowlaM.) Salarie. Material. Tranaport Other LaM Construction Equi~nt Total ll:2&E!e 9.77~.a§1 U. 713.9!1 9,agl.!4J I.!IZ.§J& J.U6 .Ug J!,9Ia.~17 11 I 03.5 · ~18 94,378,448 Ho.pitals 1,069,014 3,055,554 1,090,757 1,505,301 939,910 28,740,700 8,891,848 45,293,084 Health Centara 2,937,393 688,272 1,100,862 2,039,184 0 1,164,122 738,910 8,668,743 CDC 2,574,113 6,280,501 3,227,034 1,104,570 40,530 776,465 559,979 14,563,192 Trainins 1,603,144 413,697 1,188,885 2,010,787 436,100 2,267,562 684,084 8,604,259 Other 1,591,597 3,275,025 2,601,106 2,837,789 1,749,590 4,033,668 1,160,397 17,249,171 Province 2,7Z~,a61 13.7U,048 9,208,643 9,497,6U 3,16!,130 3!,982,517 11,035,218 24 ,378,448 Central 1,706,913 1,647,698 1,977,936 1,899,432 696,000 877 ,341 733,705 9,539,025 DI.I Jakarta 217,894 526,600 69,356 215,534 19,875 2,518,061 2,405,607 5,972,927 Wa.t Java 851,187 1,175,024 516,252 672,838 80,000 4,401,849 1,536,590 9,233,739 Central Java 1,200,574 2,684,737 524,911 791,654 314,750 3,731,310 1,403,886 10,641,822 Yo.yakarta 216,602 353,006 161,726 209,503 10,000 1,021,283 382,192 2,345,311 E.at Java 1,069,409 1,898,709 563,766 935,301 124,700 3,286,820 871,437 8,750,142 DI Acah 271,948 210,933 291,230 181,318 8,000 1,009,455 214,917 2,187,801 Rorth Sumatra 351,181 691,000 434,314 342,908 67,500 805,075 352,379 3,044,357 We.t Sumatra 284,954 497,909 270,021 400,014 30,000 2,153,745 515,246 4,152,889 "lau 154,076 178,904 187,535 128,507 11,250 686,448 149,885 1,496,604 Jamb 1 180,807 144,067 187,077 178,951 3,750 671,946 102,369 1,468,966 South Sumatra 248,550 266,724 280,463 296,445 10,000 1,965,262 526,943 3,594,387 Lampuna 203,601 191,918 122,472 176,914 150,000 635,165 97,274 1,577,344 Wa.t KA1UDantan 166,263 260,610 240,384 149,082 5,000 825,743 87,680 1,761,764 Centr.l Kalimantan 177 ,162 131,812 269,585 175,430 2,000 618,985 80,686 1,455,658 South Kalimantan 199,437 229,165 243,031 322,823 28,250 949,350 353,087 2,325,143 E··t Kalimantan 155,246 227,359 233,583 147,473 14,855 771,520 161,571 1,711,606 Rorth Sulavesi 205,322 246,973 246,096 232,441 7,500 903,445 116,867 1,958,643 Cantr.l Sul.,....l 178,350 160,495 281,211 177,035 120,600 345,285 161,921 1,424,897 South Salaved 452,673 446,874 470,412 403,876 968,050 1,433,080 421,788 4,596,754 Southea.t Sal.va.l 125,382 113,072 163,281 126,497 120,600 517,285 143,681 1,309,798 Maluku 133,425 144,675 216,680 193,560 7,500 938,075 170,904 1,804,819 Ball 257,659 376,793 186,612 263,358 55,250 1,651,616 287,716 3,082,003 Wa.t Ru.. Tana.ara 196,260 197,616 170,213 186,013 179,000 813,263 141,717 1,884,081 Ea.t Ru.a Tana··r. 239,955 234,135 285,461 253,030 66,950 735,270 112,269 1,927,070 Ir1an Jay. 153,751 267,849 336,220 233,554 19,950 1,156,997 172,324 2,340,644 Janakulu 120,428 118,020 109,703 121,069 600 829,100 174,464 1,473,384 Ea.t Timor 56,252 90,372 166,114 93,072 44,200 711,746 155,120 1,316,875 - 128 ANNEX I Table 19 of 20 PROVINCIAL GOVERHHENT DEVELOPMENT EXPElIDITURE BY PROGRAM ARD PROVINCE, 1983/84 (ACTUAL) (Bp. thousands) Total Proar!!!! 25.049.363 Bospitals 5,223,496 Bealth Centers 1,815,070 CDC 296,278 Tra1nln« 500,348 Other 17,214,171 Provlnee 25.049.363 Central Ott! Jakarta West Java 2,489,880 748,985 ° Central Java 1,668,155 DI Yogyakarta 7,694,352 East Java o 01 Aeeh 558,529 North SUIII&tra 1,053,311 West Sumatra 459,639 Rlau 1,292,522 Jamb 1 651,991 South Sumatra 212,490 LIIIIlPUn« 419,228 West KalLmantan 295,498 Central KalLmantan 762,697 South KalLmantan 472,085 East KalLmantan 1,002,029 North Sulawesi 328,316 Central Sulawesl 285,639 South Sula_sl 397.743 Southeast Sula_sl 302,649 Haluku 254,007 Ba11 729,722 West Nusa Ten«gara 309,281 Bast Nusa Ten«gara 188,430 Irlan Jaya 2,276,311 Ben«kuku 180,399 Bast Timor 15,475 - 129 ANNEX I Table 20 of 20 DISTRICT GOVERNMENT DEVELOPMENT EXPEHDITtJItE BY PROGRAM AIm PROVINCE, 1983/84 (ACTUAL) (Rp. thouu.nds) Total Prosram 3,.590,841 Hospitals 1,137,192 Health Centers 1,2.5.5,740 CDC 219,046 TraLning 8,244 Other 970,619 Province 3 . .590 .841 Central 0 DKI Jakarta 0 West Java .574,021 Central Java 898,978 DI Yoayakarta .5.5,.531 East Java 834,118 DI Aceh 191,243 North Sumatra 2.56,200 West Sumatra .56,106 lUau 49,787 Jambi 8,440 South Sumatra .50,164 Lampung .5,6.50 West Kalimantan 3,.57.5 Central Kalimantan 13,97.5 South Kalimantan 101,692 East Kalimantan 164,.599 North Sulawesi 10,62.5 Central Sulawesi 8,543 South Sulawesi .59,381 Southeast Sulawesi 0 Maluku 102,630 Bali 0 West Nusa Tengaara 75,016 East Nusa Tenggara 37,118 Irian Jaya 19,742 Bengkuku 13,709 East Timor 0 - 131 ANNEX II HOSPITAL INVESTMENT PROGRAM IN REPELITA IV Table 1 Physical Hospital Investment Targets and Achievements in REPELITA IV Table 2 Financial Hospital Investment Targets and Achievement in REPELITA IV - 132 ANNEX II Table 1 of 2 PHYSICAL HOSPITAL INVESTMENT TARGETS AND ACHIEVEMENTS IN REPELITA IV Unit Cost N~ber Repel ita e' EbIslcal FacIlIties lIudaet ~DIPI la (Rp .mUlions.l1! IV Plan 1984/85 1985/86 1986/87 Total Local Government Upgrade hospitals. Class D to C 1.285 70 46 10 3 59 Upgrade hospitals. Class C to C 1.420 39 15 3 0 18 Upgrade hospitals. Class C to B 2,029 5 4 0 0 4 Hew satellite hospitals (C+) 5,001 7 0 1 0 1 Hew hospital rehabilitation units (C) 165 26 is 0 0 0 0 Rehabilitate hospitals, Class A 30.848 1 1 0 0 1 Rehabilitate hospitals. Class B 14 .157 6 4 2 0 6 Rehabilitate hospitals. Class 0 1.468 149 U. 41 5 3 49 Central Government New teaching hospitals. Class A 77 .122 · 2J..s. 2 0 0 2 New teaching hospitals, Class B 35,393 1 0 1 0 1 Improve hospital rehabilitation units. Class A 7,275 1 0 1 0 1 Improve hospital rehabilitation units, Class B 5,605 9 0 1 0 1 Hew hospital rehabilitation units, Class C 200 20 0 0 0 0 S2ecialit~ Hos2Itals Hew cancer hospitals 36,800 1 0 1 0 1 aew mental hospitals S,S33 3 2 0 0 2 Improve TB hospitals 3,480 6 1 1 0 2 Improve leprosy hospitals 2,498 3 1 2 0 3 Improve eye hospitals 5,742 1 1 0 0 2 Improve orthopedic hospitals 1,683 1 1 0 0 1 Rehabilitate mental hospitals 2,213 11 0 0 0 0 Strengthen mental hospitals NA 0 0 0 0 Referral SEe20rt New medical laboratory (central) 2,602 1 1 0 0 1 Strengthen medical laboratory (provincial) 979 27 L! 1 1 0 2 New environmental health laboratories 2.602 4 NA HA HA HA Hew equipment maintenance workshops 1,77S 4 NA NA NA NA New hospital family planning units 78S 27 26 1 0 27 Red Cross NA HA HA NA NA HA Private hospitals (local) NA NA 65 100 71 236 Private hospitals (central) NA NA NA HA NA NA Evacuation HA NA 6 12 NA 18 Drugs for local government hospitals NA NA 2S6 274 228 7S8 L! Data refer to indiVidual projects started in terms of numbers of facilities, and do not imply completion of the project . .l1! Unit costs are given in 1985/86 prices. is HOH notes give 36; 26 Is conslstent with unit costs and planned expenditure. U. HOH notes give 176; 149 is consistent with unit costs and planned expenditure. L! Alternative sources give elther SOO beds or 13S0 beds each. i.! Estimated number of units budgeted. NA Indicates not available. Source: Ministry of Health, Directorate of Hospitals - 133 ANNEX II Table 2 of 2 FINANCIAL HOSPITAL IJllVESTMENT TARGETS AND ACHIEVEMDTS IN REPELITA IV (Rp. ml111ona) Total Cost to Cost of Complete Repel ita Budseted (ArBN-PIP) Units Units IV Plan 1984/85 1985/86 1986/87 Total Started Started Local Goye~nt: Upgrade hospitals, Class D to C 89,950 1,906 2,445 1,6.54 6,00.5 7,.5815 66.292 Upgrade hospitals, Class C to C 55,380 1,597 1,714 668 3,979 2.5,.560 19,250 Upgrade hospitals, Class C to B 10,145 340 897 992 2,229 8,116 4,581 Nev satellite hospUals (C+) 35,007 o 2.53 o 253 .5,001 4,600 Nev hospital. rehabilitation units (C) 4,290 o o o o o o Rehabilitate hospitals, Class A 30,848 1,485 1,195 19.5 2,87.5 30,848 27.973 Rehabilitate hospitals, Class B 84,942 687 278 399 1,364 84,942 83,578 Rehabilitate hospitals, Class D 218,732 3,470 1,701 843 6,014 71,932 62,395 Central Government: Nev teaching hospitals, Class A 154,244 1,972 36.5 681 3,018 1.54,244 151,226 Nev teaching hospitals, Class B 35,393 o 986 700 1,686 3.5,393 33,707 Improve hospital rehabilitation units, Class A 7,275 o 36 o 36 7,27.5 7,239 Improve hospital rehabilitation units, Class B 50,445 20 99 o 119 5,60.5 5,486 Nev hospital rehabilitation units, Class C 4,000 o o o o o o Speciality Hospitals Nev cancer hospital La 36,800 o 96 37 133 36,800 36,667 Nev mental hospital 16,599 83 o o 83 11,066 10,983 Improve TB hospitals 20,880 109 66 o 175 6,960 6,785 Improve leprosy hospitals 7,494 1,263 1,542 109 2,914 7,494 4,.580 Improve eye hospitals 5,742 894 692 72 1,658 .5,742 10,084 Improve orthopediC hospitals 1,683 869 3.5.5 193 1,417 1,683 266 Rehabilitate mental hospitals 24,343 3,212 2,627 273 6,112 NA NA Strengthen mental hospitals 54,805 H.A HA H.A HA IfA NA Referral SUpp2£k Nev medical laboratory (central) 2,602 265 273 316 8.54 2,602 1,748 Strengthen medical laboratory (provineial) 26,433 574 1,189 51 1,814 H.A NA Hev environment health laboratories 10,408 IfA H.A H.A H.A NA Hev equipme~t maintenance vorkshops 7,100 IfA H.A NA IfA IfA NA Hev hospitals family planning units 21,195 394 412 1.5 821 21,19.5 20,374 Red Cross 2,500 220 200 47 467 NA IIA Private hospitals (local) 10,250 1,.550 2.58 78 1,886 NA HA Private hospital (central) 12,250 o 27.5 34 309 HA NA Evacuation 210,935 31 32.5 o 356 NA H.A Drugs for le,:al government hospitals 11,442 2,002 1,230 914 10,146 HA IfA Dental Heal t1:. Services 15,980 NA IIA IfA 1280,092 La Land vas purchased but there vas no construction by 1987 - ps ANNEX III HEALTH MANPOWER PATA SOURCES - 136 ANNEX III Page 1 of 5 A. Introduction 1. Existing health manpower data sources are severely limited in comprehensiveness and detail. A newly-instituted data system aims for complete and detailed coverage of public health manpower, but has not yet been fully implemented. Very little data exist on health manpower employed by quasipublic or private organizations. These data limitations are a major restriction on staff planning and evaluation. This annex briefly reviews the scope. coverage. advantages, and disadvantages of health manpower data sources. 2. Data coverage is usually defined on the basis of the employee's workplace or employer. It is therefore useful to present a concise taxonomy of health workers by their place of employment, as follows: (i) Private sector: Self-employed health workers; private clinics and hospitals; and industrial and commercial firms with in-house health services; (ii) Public enterprises (BUMN): e.g., Pertamina; (iii) ABRI (the armed forces); (iv) Depdikbud (Ministry of Education and Culture): employs the faculty of medical schools; (v) Health facilities under Ministry of Health responsibility: vertical facilities (centrally funded through MOH); regional facilities (funded by provincial or district governments.) Two things should be noted about this classification. First, it is possible for a health worker to hold two or more jobs simultaneously: most if not all doctors in private employment also hold a government position. Second, a government worker may work at one facility but be employed by another. For instance, centrally-funded workers may be assigned to work at provincial facilities: and Ministry of Education and Culture employees may work at Ministry of Health hospitals (e.g .· specialists). B. "Old System" (BAKN-Based) 3. Description. The principal existing source of manpower data is maintained by the MOH Data Center and is based on records from the Civil Service Bureau (BAKN). In 1983, the Data Center obtained BAKN data tapes describing all civil servants working in MOH-supervised facilities. Since obtaining the data. the Data Center has independently updated it using information forwarded by the MOH Personnel Bureau. - 137 ANNEX III Par:;e 2 of 5 4. Covera~e. The chief drawback of this system is that the Personnel Bureau only processes appointments, promotions, and transfers for employees of MOH (i.e., civil servants with NIP 14). However, many employees at MOH supervised facilities are technically classified as employees of other departments. In particular, many nonmedical employees of provincial and distI'ict facilities are formally employees of the Ministry of Home Affairs. Many specialist physicians are formally on the rolls of the Ministry of Education and Culture. An accurate count of these two important classes of employees exists only for 1983. 5. Detail. The BAKN records are concise, containing basic data for pay compl.;,tation. These data include: · Basic demographics: date and place of birth, sex, marital status, religion; · Educational level; · Ever-attendance at a training course; · Rank and seniority in the civil service; · Province and kabupaten of current workplace; · Whether or not a centrally-paid employee; · Whether or not on probationary (calon) status; and · Number of dependents. 6. The BAKN records omit some information which is important for planning and evaluation purposes. Data are not available on the following: · Characteristics of the workplace: it is not possible to distinguish between hospital, health center, and administrative personnel; · Precise location of workplace: for instance, it is not possible to determine whether an employee is station~d at the district head quarters or in the outlying rural areas; · Educational or functional specialization: among paramedics, for instance, it is not possible to distinguish nurses from technicians; and · Wages: however, base pay and allowances can be calculated from salary grade, seniority and number of dependents. 7. Accuracy and timeliness. Updating of the "Old System" records is a multt-step process, involving transfers of information from health facility to provtncial office to the Personnel Bureau and finally to the Data Center. At best the process takes several months; at worst, there may be some information loss. An indicator of the problem is the high proportion, approximately 40%, of eD[ployees listed as being on probationary status. Probationary status is offieially supposed to last only for the first year after recruitment, so the proportion should be no higher than the annual growth rate of employment, around 6%. To some extent the high reported proportion may reflect noncompliance with regulations on promotion. But it also indicates delays or lapsEls in updating. - 138 ANNEX III Page 3 of 5 C. SP2TK (Sistem Pencatatan Dan Pe1aporan Tenaga Kesehatan> 8. Description and scope. Recognizing the need for improved personnel manpower monitoring, the Data Center formally introduced a new system (SP2TK) at the end of December 1985. The system employs a set of well-documented forms which are filled out by employees and forwarded directly to the Data Center in Jakarta for input and processing. In principle, the system should cover all employees working at facilities under MOH responsibility. 9. Detail. In addition to the data gathered by the old system, the SP2TK gathers precise information on: · Type of education; and · Workplace. 10. Coverage and implementation issues. Assessing the degree of coverage is difficult because no accurate benchmark census of employees exists. The Data Center uses as a benchmark the total number of employees listed under the old system: the more-or-1ess current number of formal MOH (NIP 14) employees, plus the number of employees classified under other departments as of 1983. Since the number of non-NIP 14 employees has probably grown over the past four years, the benchmark total of 200,690 employees is probably an underestimate. 11. As it approaches its first anniversary, SP2TK is still in its startup phase. This involves the formidable task of collecting data on all of MOH's approximately 200,000 employees. Bottlenecks exist at both the reporting and data entry stages. As of 31 October 1986, forms had been received on 155,505 employees; of these, 60,073 had been entered and checked. The data entry bottleneck is temporary and simply reflects the large volume of start-up data as compared with a small data entry staff. The low reporting rate is more difficult to diagnose and may indicate that the system is not yet understood, or that it is encountering administrative or compliance problems. The provinces with the lowest reporting rates are shown in Table 1. Communications constraints may account for the low response rate from Ma1uku and Irian Jaya. In Jakarta, however, nonreporting by two large hospitals accounts for the comparatively low response rate. Table 1: SP2TK REPORTING RATES, 1986 Estimated , of employees reported Maluku 18 West Sumatra 23 Irian Jaya 31 West Nusa Tenggara 43 South Sulawesi 46 Jakarta 53 - 139 ANNEX III Pace 4 of 5 12. Another indication of implementation problems is the very low nationwide rate of employee file updating (promotions, transfers, etc). The Data Center sent teams to twenty provinces in December 1986 to diagnose bottlenecks and assist local administrators in institutionalizing the reporting system. Clearly the viability and usefulness of the system depends on this kind of institutionalization. D. Other Manpower Data Sources at Ministty of Health 13. YanKed hospital data. YanKed, the hospital services directorate, operatE!S an independent data center which collects personnel data from all Indonedan hospitals, public, semi-public, and private. At the hospital level, data are collected on number of employees by detailed skill classification by source of salary. The data therefore include personnel who work at MOH facilities but are formally classed as Home Affairs or DepDikbud emploYE!es. YanKed also collects employee data at the individual level, though here cClverage is less than universal. In principle, these data could be used to indi.cate the extent of multiple job holding by government employees. 14. Binkesmas health center data. Binkesmas, the health center directc1rate, compiles personnel data on all government health centers. The published compilation lists number of employees by type (doctor, dentist, nurse, other paramedic, nonmedic) by district. It has been suggested that inform~Ll reassignment of health center posts to hospitals and health offices undercuts the reliability of this data, however. 15. Personnel bureau data. The personnel bureau collects a great deal of data OIl recruitments, appointments, and some transfers; in general, the data are not: comprehensive and lacks detail on the employee's work unit. 16. Doctor license data. The Bureau of Personnel also processes applic~Ltions for, and renewals of, doctor's licenses (S.l.D.) for all doctors, regardless of employer. Because renewal is at five-year intervals, and becaUSE there may be some non-reporting, these data can only be used as an approximation; it remains, however, the best source of data on the national stock eof physicians. 17. Indicators of Staffin& Needs Data. The ISN project is an ambitious attempt to rationalize staff planning and allocation. It establishes, for each ty~e of MOH facility, detailed staffing norms based on facility utilizs.tion (e.g., number of nurses required at a hospital is a function of number of outpatients and number of inpatient-days.) Application of the ISN system requires data on both current utilization and current staff from all MOH fae ilities. E. Summary and ReCOmmendations 18. Labor costs constitute a major portion of public health expenditures, and this component is under direct government control. Efficient allocation - 140 - ANNEX III Page 5 of 5 of existing staff, and planning and budgeting for staff expansion. requires detailed, accurate and comprehensive information on the manpower situation. Such data currently do not exist. 19. The SP2TK labor reporting system is the designated solution to this problem. The system, which relies on timely centralized collection of individual-level data from all employees, faces severe compliance problems and may not be supportable. Evaluation of the system's prospects needs to be made; if necessary. a less ambitious system should be substituted. At the same time, efforts need to be made to eliminate redundant collection of staffing data. - 141 ANNEX IV WAGES AND PRODUCTIVITY - 142 - ANNEX IV Pale 1 of 5 A. Wales and Productivity Wales and Income 1. Essentially all employees of the Ministry of Health at the national or local level are civil servants (pelawai neleri). As such, they are subject to the civil service salary structure. Salary is a function of tenure (i.e., years in service) and rank (Iolonlan). Starting rank, in turn, is a function of education, as follows: Rank I: Starting salary Rp.33,200/month; less than high school education, e.g. , drivers, janitors; Rank II: starting salary Rp.55,500/month; high school or academy education, e.g. most paramedics; Rank III: starting salary Rp.8l,OOO/month; university degree, e. g. , general doctors; and Rank IV: starting salary Rp.93,200/month; advanced degree. 2. Within each rank, salary doubles after about twenty years of service and plateaus at 24. It is possible to advance from step to step within each rank, and sometimes to advance ranks. The maximum base salary achievable by, for example, a specialist doctor after 24 years of service, is Rp.265,600/month, or approximately US$2,OOO per year. 3. To this base salary a number of official supplements (tunjunlan) are added: (a) All civil servants are entitled to a rice allowance of 10 kg per family member per month, paid in kind. The retail price of rice is about Rp.400/kg, so this supplement could be relatively substantial for a low-ranking employee with a large family. The opportunity cost to the government of providing the rice may be less than the retail price; (b) Functional supplements are paid to employees who are in certain service-providing positions. These supplements are not incentives, because they are not contingent on whether the employee actually provides those services. Thus doctors at health centers or hospitals receive an additional Rp.50,OOO/month, and paramedics receive Rp.15,OOO; (c) Structural supplements are paid to the occupants of so-called structural posts, which are for the most part directorships of a bureau, section, or facility. For instance, a health center doctor receives Rp.25,OOO/month by virtue of being director of the health center. These supplements can range up to Rp.150,OOO/month for rank IV administrators; - 143 - ANNEX IV Page 2 of 5 (d) Cost of living supplements are paid by some outlying provinces to health center staff, especially doctors. Ranging up to Rp.lOO,OOO/month, these supplements probably serve as an incentive for recruitment and retention, in addition to compensating for higher prices; 4. In addition to these standard regular supplements, a system of informal incentives has arisen based on honoraria. travel per diems in excess of average actual expenses, and special project funds. These are used to reward staff who participate in special task forces or projects. In some cases, ,/1 DIP-funded project may be used to stimulate activities which strictly speakin,g would be classified as routine. While these payment mechanisms may be susc,eptible to abuse, they do have the noteworthy property of being incenti'ves to effort in a system which generally lacks performance-related rewards or sanctions. 5. Finally, almost all doctors, and many paramedics, have private practic·es or second jobs in addition to their government duties. For doctors. particularly specialists, this outside income can be very substantial, ranging up to a large mUltiple of their government salary. Data on private sector wages or earnings for health workers are not readily available. 6. It is useful to evaluate these income figures against the general income distribution in Indonesia. Table 1 provides information on the percentile ranking of individuals according to per-capita family expenditure: Table 1: RANKING BY PER CAPITA EXPENDITURE, SUSENAS 1984 Monthly per capita Percentile Rank expenditure (Rp.) Indonesia Jakarta 80,000 99.5 96.2 60,000 98.7 96.2 40,000 95.8 75.5 30,000 91.2 57.8 20,000 77 .5 25.6 Source: Biro Pusat Statistik, Penge1uaran Untuk Konsumsi Penduduk Indonesia per Province, 1984. 7. A consequence of the fragmentation of budgetary responsibility for health personnel is that no comprehensive data on the public-sector wage bill is available. MOH-appointed employees are paid, variously, from the central, provincial, or local routine budgets; some salaries are also paid through the development budget. In addition, about one-third of all employees working at MOH-supervised facilities are officially employed by the Home Affairs Ministry - 144 ANNEX IV Page 3 of 5 and therefore entirely beyond MOH budgetary purview. An incomplete estimate of the wage bill for 1985/86. derived from the central and regional health budget allocations is given in Table 2. Table 2: PUBLIC SECTOR WAGE BILL FOR HEALTH BY LEVEL OF GOVERNMENT. 1985/86 Budget Salaries Total expenditure , salaries Central routine 72.3 133.9 54.0 Provincial routine 78.4 112.5 69.7 District routine 81.8 104.1 78.6 Central development 10.6 112.5 9.4 243.1 463,0 52.5 8. These figures do not include Home Affairs employees. An alternative estimate of the bill can be made by applying the civil service salary scale to a breakdown of MOH personnel by rank and tenure. A rough estimate of the public-sector wage bill for health is given in Table 3. - 145 ANNEX IV Pase 4 of 5 Table 3: PUBLIC SECTOR WAGE BIU. FOR HEALTH BY TYPE OF EMPLOYEE, 1986 (Rp. thousands) Number of Monthly Monthly Annual employees Lf! wage Lb. supplements 1£ wage bill Ministry of Health 164,870.052 Doctor (specialist) 287 139 75 737,016 Doctor (general) 11,789 113 75 26,595,984 Dentist 2,630 113 75 5,933,280 Other graduate 1,971 113 25 3,263,976 Paramedic, academy ed 3,946 83 15 4,640,496 Paramedic, high school 57,656 79 15 65,035,968 Paramed.ic, Jr. high 22,759 58 10 18,571,344 Nonmedic, Rank III/IV 1,543 113 50 3,018,108 Nonmedic, Rank II 15,264 79 o 14,470,272 Nonmedic, Rank I 36,934 51 o 22,603,608 Ministry of Home Affairs Nonmedics 50,000 65 o 39,000,000 Ministry of Education and Culture Specialits doctors 3,300 139 75 8,474,400 Grand Total 212.344.452 Lf! Do,!s not include private, quasi-public or armed forces personnel. Lb. Base salary assumes average of six to seven years' tenure. 1£ Salary supplements estimates are informed guesses. Does not include in·· kind rice supplement of 10 kg per family member per month. - 146 ANNEX IV Page 5 of 5 B. Productivity Issues 9. Scattered data suggest that worker productivity is low on average, but highly variable between facilities. Berman and Suomi !J studied workload at health centers in 22 subdistrict of three rural districts in the neighborhood of Yogyakarta. Vorkers were asked to report the allocation of their time between curative, immunization, and MCH services; output per full-time equivalent day was then calculated for each of these activities. Curative output was found to average 11 contacts per day; however, the most productive subdistrict had 12 times the contacts/day of the least productive. Similarly, the average MCH productivity was 4.6, contacts per day, with variation over a factor of 19, and immunizations averaged 5.5 per day with variation over a factor of 6. 10. Likewise, hospital data reveals a very low workload at some facilities. An extreme case, for instance, is Sungguminasa hospital in South Sulawesi, where three doctors and 29 paramedics were reported to handle an average daily workload of about 20 outpatient visits and two occupied beds. 11. There are a number of possible sources of, and remedies for, low productivity. To a large extent, low productivity may be rooted in a failure of the manpower allocation process to match staff with existing workload. The new ISN manpower planning system is directly addressing this problem. Low productivity may also stem from structural barriers to utilization. For instance, some private health centers have achieved much greater utilization rates than public health centers through the simple expedient of offering clinic hours during evening and Sunday hours convenient to the client. Similarly, lack of drugs, specialized staff or equipment may deter use of some health centers and Class D hospitals. 12. Another problem lies in the lack of an incentive structure. With few exceptions, the system has lacked any penalties for lack of effort, or rewards for exceptional effort. Promotions, for instance, have been automatic at four-year intervals, and dismissals are extremely rare. However, MOH is preparing to implement a point system for medical and paramedical promotions. Nurses, for instance, will get points for each patient served; a threshold number of points will suffice for promotion between steps within a rank, and from the top step of a rank to the next rank. Ceilings on the rank achievable by a paramedic will be removed. 1/ Berman, Peter and Suomi Sakai. ftThe Productivity of Rural Health Manpower in Java ft . Draft, The Johns Hopkins University, Department of International Health, September 1986. - 147 ANNEX V PARAMEDICAL MANPOWER TRAINING 1 - 148 ~nv Page 1 of 14 A. Paramedical Training Schools 1. Training of paramedical staff in Indonesia is currently conducted at 365 institutions, which can be cross-classified in a number of different ways. 2. Specialization. There are approximately 22 different types of schools (see Table 1). However, more than half the schools (188) fall into a single category, the SPK, or high-school level general nursing school. Three other types of nursing schools (academy-level general nursing, dental nurses, and psychiatric nurses) comprise another 50 schools. The SGP category comprising four schools, trains nursing teachers. The remainder of paramedical manpower types are usually lumped together as non-nurse paramedics, a heterogenous category that includes sanitarians, nutritionists, assistant pharmacists and technicians. 3. Academic Level. There are three levels of institution. The basic level is a three-year course at the level of a vocational high school; SPKs are an example of this type. The academy level (abbreviation starting with A) is also a three year course; a high school degree (general or health-related) is a prerequisite for admission. AKPERs, the advanced or supervisory nursing schools, are the most familiar example. Finally, there are one-year continuing education schools, oriented to supplemental training of experienced paramedics; these include the SGP teacher-training schools, the lab technician schools and the psychiatric nursing schools. 4. OWnership and Funding. Slightly less than half of all paramedical schools are public schools, directly funded and operated by the central Ministry of Health. For convenience they can be called central or vertical schools. Another 38 public schools are operated by provincial or local governments. The Armed Forces operate 24 schools. The remaining 132 schools are privately operated; most of these are nursing schools, and many are attached to private hospitals. In theory, all schools are subject to the technical supervision of the Ministry of Health. 5. Distribution by Province. Table 2 shows the provincial distribution of paramedical schools. Jakarta, with 5% of the country's population, has 15% of its paramedical schools, including 13% of its SPKs. However, the Jakarta schools recruit some of their students from other provinces. Every province has at least one SPK. 6. Expansion of system. At the end of REPELITA III (December 1983) there were 282 paramedical schools. In the first two years of REPELITA IV, the number of private schools increased 31%, public schools 32%, and military schools 4%. By the beginning of 1987 there were 373 institutions; construction or refurbishing was underway for 30 institutions, including six multi-stream academies. The target for the end of REPELITA IV had been set at 413. Emphasis during REPELITAs V and VI is to be placed on the upgrading of basic (SPK-level) schools to the academy (diploma level). - 149 - ANNEX V Paie 2 of 14 Table 1: PARAMEDICAL SCHOOLS BY CATEGORY, ACADEMIC LEVEL, AND OWNERSHIP, 1987 OwnershiR Cateior~ Central Local Kilitary Private Total BASIC LEVEL iJ!. SPK General nursing 62 33 19 77 190 SPRG Dental nursing 15 2 1 0 18 SMAK Laboratory assistant 11 3 1 6 21 SMF Pharmacy assistant 6 2 2 29 39 a~~EMY LEVEL LQ AKPER Nursing 16 1 1 14 32 AKZI Nutrition 6 0 0 2 8 APKTS Sanitation technology 8 0 0 1 9 ATEM Electro-medical technician 1 0 0 0 1 APRO Radiology 2 0 0 0 2 AKNES Anesthesiology 4 0 0 0 4 AKFIS Physiotherapy 2 0 0 0 2 Other Optics, rehabilitation, etc. 1 0 0 4 5 CONTINljING i£. SGP Teacher training 4 0 0 0 4 SPPH Assistant sanitarian 19 1 1 0 21 SPKSJ Psychiatric nursing 4 0 0 0 4 SPAG Assistant nutritionist 9 1 0 1 11 SPTG Dental technician 1 0 0 0 1 STLKF Laboratory 1 0 0 0 1 TOTAL 172 42 25 134 373 MIDWIFE PROGRAM 12 0 0 15 27 iJ!. 3-year course, equivalent to high school. LQ 3-year course, post high school. i£. l-year course, continuing education for civil servants. Source:: Ministry of Health, Pusdiknakes. - 150 - ANNEX~ Pale 3 of 14 ~, PAlWllDICAL SCBDOLS IY CA~y a AIID PIOVIIICB. 1986 SPit SPUJ SGP IPItG SPM SMr S'fLU SIWt IPPB IPAG ADD. .APl.TS AUI A'l'IMA no AXIIES AUII APM AAM M1( AD RO TOTAL Jit.k.rt. 2.S 1 2 1 8 1 3 1 .5 1 1 1 1 1 1 1 .54 We.t J.va 2.5 1 1 2 1 2 2; .5 1 1 46 Centr.l J · ..,. 21 1 1 4 1 1 1 1 1 1 1 1 U Yo.,it.k.rt. .5 1 1 1 1 1 1 11 Ea.t J · ..,. 22 1 1 1 1 2 4 1 1 1 - 1 40 Aceh 18 1 4 2 1 2; 1 29 W··t S....tr. .5 1 2 1 1 1 12 Rl.u 2 1 J .... l 3 1 1 1 1 1 B South I ....tr. · 1 2 1 1 1 14 L.......... 1 1 1 We.t 4 1 1 6 ltallaantan Centr.l 1 1 ltal laant an louth 1 1 1 1 1 1 10 ltallaantan E·· t Sul._.1 " 3 1 North lul._.1 .5 1 1 1 1 1 10 Centr.l 2 1 3 Sul..... l South Sul._.l · 1 1 1 2 1 2 1 1 1 1 21 South···t 1 1 2 Sul.....1 Malulw 3 1 4 I.U 3 1 1 1 1 1 1 \I W··t ..... 2 1 1 Te.....re laat lu·· 3 1 1 1 6 rena··lta Irlan J.,. S 1 la...ku1u 1 1 B.at t~r 1 1 I2U.l 111 ! ! 11 1 a 1 a2 n 11 1.2 1 I 1 1 ! 1 1 1 1. 1 1. m a s·· t.bl. 1 for an .xplanatlon of aoron,.. d.notina o.t··ory of par...dl0.1 aohool. - 151 ~ny Page 4 of 14 7. Quality of Facilities and Instruction. The paramedical physical facilities schools have long been recognized as inadequate, and loans from the Asian Development Bank and the World Bank have been addressed to this problem. Pusdiknakes estimates that for 1986/87, only 48% of Ministry of Health schools possess 'complete' facilities, and only about 50% of private schools possess 'adequate' facilities. Cited as especially lacking are basic teaching aids, and library materials. Observers asked to generalize on quality differentials between private and public schools suggest that private schools are more variable, with some schools above the public school standard and others below. B. System Capacity. Enrollment and Output 8. The normative capacity of each school is 40 students per class: the three-year schools have a normative enrollment of 120 and the one-year schools accommodate 40 students. There is considerable variation around this norm. A couple of schools have only fifteen students, several have over 300, and one has over 1,100 students enrolled. To meet REPELITA IV's ambitious plans for expanding the work force, parallel classes (split shifts) were instituted for much of the cohort entering school in August 1984 (the graduating class of July 1987). The introduction of parallel classes effectively doubled the intake rate of participating schools. Parallel classes were continued in the entering cohort of 1985, but discontinued for students entering in 1986 as the emerging personnel glut became evident. 9. Parallel classes resulted in a bulge in student enrollments. Table 3 sho~'s student enrollment in academic 85/86, together with estimated enrollment of entering students in academic 86/87. The graduating class of 1986 numbered abo~,t 14,000, including about 2,000 graduates of one-year institutions. Centrally-run schools account for nearly half of enrollment, locally-run public schools another 10%. About one-third of enrollment is in private schc.o1s, and the remaining 7% in military-run schools. The class of 1988, a1rE,ady enrolled, number about 22,000, not counting one-year students. 10. The increase in enrollment is apparent in all four ownership categories. Projected graduates by year are shown in Table 4, which assumes a 5% Elnnua1 attrition rate (attrition data on drop-outs and failures have not beeTI compiled by Pusdiknakes, although the raw data exist). 11. In opting for parallel classes, a deliberate decision was made to trade-off quality of instruction of quantity of output. As a consequence, the parll11el class cohorts are experiencing overcrowded facilities, increased stu(lent/teacher ratios, and greatly decreased opportunities for patient contact via clinical or field experience. Despite the increase in intake, there is tremendous competition for admission to public paramedical schools. Puscliknakes estimates a 10:1 ratio of applicants to available places. A statldardized nationwide admissions test has just been put in place. However, there is not a nationwide passing grade; to encourage recruitment of students in outlying regions, each district is allocated a quota of students for vertical MOH schools. Test scores are used to rank students within each district. Non-vertical schools, public and private, are supposed to use the tes1: to rank students within their own applicant pool. - 152 ANNEX V Page 5 of 14 Table 3: PARAMEDICAL SCHOOL ENROLLMENT BY SCHOOL OWNERSHIP Graduating Class La 1986 1987 1988 1989 Public 8,512 11.469 12.122 NA Central 7,231 9,265 9,762 6,200 Local 7,231 2,204 2,364 NA Private 4,783 6,877 8,173 NA Armed Forces 919 1,378 1,559 NA Total 14.214 19.124 21.858 14,480 ~ Class of 1986 includes about 2000 graduates of one-year courses; classes of 87 to 89 do not. Source: Ministry of Health, Pusdiknakes. Table 4: PARAMEDICAL TOTAL SUPPLY AND GOVERNMENT DEMAND, 1979-1988 ~ New Excess of Excess of Total Lb. Government Government total Government Year Graduates Graduates LJ;,. Posts ,Lg graduates new graduates posts over new posts 79/80 5,941 2,789 5,651 290 -2,862 80/81 5,320 2,589 6,898 -1,578 -4,309 81/82 4,654 2,521 5,860 -1,206 -3,339 82/83 5,164 2,912 4,687 477 -1,775 83/84 5,601 3,033 5,220 381 -2,187 84/85 6,810 4,807 4,119 2,691 688 85/86 8,495 5,825 5,158 3,337 667 86/87 12,807 9,138 4,665 8,276 4,473 87/88 20,659 13,754 11,907 16,128 1,847 88/89 18,812 11,902 11,907 14,281 -5 ~ Data exclude Pekarya Kesehatan. Lb. All Indonesian paramedical schools. LJ;,. Central Government (MOH) and local government only; excludes armed forces. ~ Newly created formasi (routine plus INPRES). Numbers for 1988/89 are illustrative only. Source: Ministry of Health, Pusdiknakes and Personnel Bureau. Government: posts: MOH, Personnel Bureau. - 153 ANNEX V Page 6 of 14 C. Teachin& Staff 12. Quantity. In academic 1985/86 there were slightly more than 2,000 full-time teachers, of whom about half hold full-time appointments at public (non-military) institutions (see Table 5). The resultant teacher/student ratio of about 1:25 is augmented by heavy use of part-time teachers. A total of 8,250 part-time teaching appointments were reported nationwide. Because multiple-job holding is the rule rather than the exception, the actual number of individuals engaged in teaching is far less than 10,000. Most part-time teachers hold several such appointments, in addition to a full-time appointment as a doctor, nurse, administrator or (most often) teacher at another paramedical school. 13. Teacher/student Ratio. Table 6 examines the teacher/student ratio in more detail. The full-time teacher/student ratio falls far below the norm of 0.100 for basic and continuing schools, and 0.160 for academies. At centrally-run public schools, the ratios are 0.032 and 0.017 respectively. But by 1987 the ratio at all central schools combined had risen to 0.062. In compensation, the ratio of part-time teaching appointments to students is rather high. Academies, perhaps because they offer more specialized subject matter, place much greater reliance on part-time teachers. However, there is no standard conversion factor giving the full-time equivalency of a part-time teacher. 14. It is difficult to compare teacher/student ratios between government and private schools because of a lack of standardization of the definitions of full and part-time. Private institutions, on average, place more reliance on full-time teachers than public institutions. Private SPK-level schools have a full-time teacher/student ratio almost three times that of central public schcols; but the part-time teaching appointment ratio is twice as high at the public schools. Overall, the armed forces schools have the highest stuc.ent/teacher ratios by a slight margin. 15. Within ownership categories, there is substantial variation in teacher/student ratios. Several schools have no full-time faculty. One might expect substitutability between part-time and full-time teachers, so that schc-ols with a limited full-time staff compensate by hiring additional part time teachers. In fact, there is no significant statistical correlation between the full-time and part-time teacher/student ratios. That is, schools fortunate enough to have a large contingent of full-time staff also tend to have relatively high numbers of part-time staff. 16. Teacher Productivity and Job-sharin,. The low teacher/student ratio is ~xacerbated by inefficient utilization of full-time staff. Full-time public school paramedical teachers are estimated to devote an average of 0.8 hours daily to their principal job. This permits the typical teacher to hold sev~ral additional, part-time appointments, often at other government-run schclols. - 154 ANNEX V Pale 7 of 14 Table 5: PARAMEDICAL SCHOOLEKPLOYEES, ACADEMIC YEAR 1985/86 Fu11·time Part· time Other Ownership category Number LA teachers Lb. teaching employees appointments Private 129 797 2,721 1,020 Public central 154 803 4,362 2,960 Public local 32 248 664 285 Armed forces 23 178 503 235 TOTAL 338 .2....Q22 8.250 4.500 LA Number of schools reporting both enrollment and staff. Lb. Because many individuals hold multiple part-time appointments, often in addition to a full-time appointment, the total number of teachers is less than the total number of appointments. Source: Computed from data on employees and enrollment compiled by Ministry of Health. Pusdiknakes. Division IV. Table 6: TEACHER/STUDENT RATIOS. ACADEMIC YEAR 1985/86 Teachers J2er student School type Ownership Full· time Part-time LA Basic 3-year Central public 0,032 0.142 Local public 0.040 0.115 Private 0.084 0.067 Academy 3-year Central. public 0.017 0.233 Private 0.034 0.111 Continuing I-year Central 0.051 0.236 LA "Part·time" should be read as "part-time appointments per student". An individual may hold several part-time appointments at different schools. so it is not correct to interpret the ratio as part-time teachers per student. Source: Calculated from data on school enrollments and teachers. compiled by Ministry of Health, Pusdiknakes. Division IV. Averages exclude schools with missing data. - 155 ANNEX Y Pale 8 of 14 17. Anecdotal evidence suggests that the comparable teachers in non- health fields devote far more effort to their principal jobs. This may not be true of medical education, however; the University of Indonesia, for instance, employed 613 full-time teachers to instruct 843 students in 1984, but did not provide the tutorial instruction that ratio implies. To some extent, low utilization may reflect the need for specialized staff at the academy level i.e., there are economies of scale which are not realized by the average, 120 student school. To a greater extent, however, low utilization probably results from the temptations posed by high demand for, teachers in a period of rapid ~;chool expansion, combined with school managers' lack of effective sanctions against low effort. 18. The effect of this pattern may be that the government effectively employs paramedical teachers at well above the nominal civil service salary, fractionating their work among several institutions. The result is much less efficbnt than would be obtained if teachers were employed at a single institlltion, even at a compensation rate equal to their current total income. Fracti(mation of job assignments has resulted in a lack of identification with, ()r responsibility to, any single institution. 19. Development of multistream academies would be a partial solution to this problem. The multistream academy, by combining several schools on a single campus, potentially offers economies of scale in the use of both specialized facilities and specialized staff. It is hoped, therefore, that teachers would in effect consolidate their multiple job-holdings under one roof. There may also be economies of scale in management. 20. Teacher Traininl. The principal teacher training institution is the SGP, a one-year school which trains nursing teachers, primarily for SPK. Particlpants in this training, for the most part, have only an SPK (i.e., high school) education, and three years of practical experience. The four SGPs are all ceIltrally run public schools. Their output in 1986 was about 240, project:ed to increased to 340 in 1987. 21. Teachers for non-nurse paramedical schools are trained by Pusdiknakes in collaboration with the Ministry of Education and Culture under the Akta III and IV programs. These have as admission prerequisites a nursing academy degree, and a university degree, respectively. About 80 graduates of these programs were expected during academic 1986/87. Degree programs for nurses are under development. D. Costs and Finaocinl 22. Financing. Centrally-run public schools, in principle, have their operati.ng costs funded out of the national routine budget (DIK) and investment out of the development budget (DIP). In practice, as it is true throughout the health sector, a substantial portion of operating costs are supported through the DIP, and some investment costs through the DIK. A problem specifi.c to the training subsector, however, is that schools are not supposed to be funded through the routine budget until they are accredited by KENPAN and thE: Ministry of Education and Culture. In 1986 only 98 of the 171 - 156 ~"V Pa&e 9 of 14 vertical schools were fully accredited; 112 of the schools were listed in the routine budget. The schools also receive a small amount of funding from fees. Ber1ian et.al. 11 report average per-student receipts of Rp.50,OOO to Rp.280,OOO per year in a sample of 12 vertical schools: the higher figures include dormitory fees. In some cases, private contributions for school facilities can be substantial; the SPK attached to Cipto Hospital in Jakarta received almost one-quarter billion Rupiah over the period 1979-85. 23. Locally-run public schools are funded primarily out of the regional government budget, although they may receive development funds from the central budget. Private schools are funded principally through tuition fees, which range up to one million Rupiah annually. Other private contributions may be important, especially for church-run schools. 24. Unit Costs. The principal source of unit cost information is the study of 22 schools undertaken by Berlian et al, Strictly speaking, the study deals with unit revenue rather than unit costs. Unit revenues are computed based on a five year average of revenues and graduates. There appears to be no correction for inflation, or for variations over time in the enrollment to graduate ratio. In addition, the distinction between revenue and costs is blurred, and there may be some double-counting. Moreover, the sample is drawn entirely from Jakarta and West Java, and therefore may not be typical of the country as a whole. The results of the study are shown in Table 7. In this sample, central schools seem to be more costly than other types. The central AKPERs and SPKs are all in the range of Rp.8 to 9 million per graduate. In contrast, the private SPKs and AKPERs were in the two to four million range. The three one-year schools in the sample were correspondingly less expensive than the three-year schools, at about one to two million rupiah per graduate. It is important to maintain the distinction between revenues and costs. Although public schools are supposed to have a balanced budget, revenue data provides an imperfect guide to the actual structure of costs. The distinction is more important for private schools, some of which are suspected of having revenues substantially in excess of expenses. 25. An independent source of unit cost information can be obtained from data on enrollment and revenue at accredited central schools, since the national DIK provides most of the funding for these schools. Unit costs computed from these data will be underestimated to the extent that income from DIP, school fees, and private contributions is ignored. One interesting feature of these data is the tremendous variation between schools in average cost. Table 8 shows average cost data for 90 central schools for which data on FY85/86 DIK allocations were available. These statistics suggest that the sample taken by Berlian et.al. overrepresented above-average cost schools. 11 Ber1ian T.P. Siagian et,al. (1986) Studi Biaya Pendidikan Paramedis Serta Pendayagunaannya sampai tahun 2000. Ministry of Health, Pusdiknakes. - 157 ANNEX V Page 10 of 14 Table 7: UNIT COSTS PER GRADUATE OF SELECTED PARAMEDICAL SCHOOLS Type Ownership Name Cost per graduate (Rp. thousand) Academy Level AKZI Central Jakarta 3,398 APK Central Jakarta 6,757 AKPER Central Bandung 8,237 AKPER Central Jakarta 8,513 AKPER Private RS PGI Cikin! 2,748 AKPER Private Advent Bandung 4,644 Basic Level SPRG Central Bandung 4,608 SPRG Central Jakarta 6,246 SPK Central Bandung 8,727 SPK Central Cipto Mangunkusumo, Jakarta 9,128 SPK Local PEKDA Garut 2,105 SPK Private Immanuel, Bandung 2,048 SPK Private Yarsi 2,812 SPK Military RSAL Mintohardjo 3,088 Continuing (one-year) Schools SPAG Central Jakarta 1,361 SPPH Central Jakarta 2,354 SPAG Private LPIG Bandung 1,037 Source: Ber1ian Siagian etta1., (1986) Studi Biaya Pendidikan Paramedis Serta Pendayagunaanya sampai Tahun 2000, Ministry of Health, Pusdiknakest Table 8: AVERAGE COST PER STUDENT PER YEAR (Rupiah) Mean (over schools) 465,000 Standard deviation 350,000 Minimum 88,000 Maximum 2,130,000 - 158 ANNEX V PaKe 11 of 14 26. A regression analysis of these schools (excluding the three largest), including only basic (SPK-level) institutions, yielded the following statistically significant relation (t-statistic in parenthesis): TOTAL COSTS - 50,193,000 + 114,000 * ENROLLMENT (3.7) A quadratic term in ENROLLMENT was tested and rejected. This analysis indicates that the government tends to allocate a large fixed sum per school (about Rp.50 million), plus a marginal sum of Rp.114,OOO per student. Average cost thus tends to decrease as enrollment increases; at the nominal school capacity of 120, average costs would be about Rp.562,OOO per student per year, or 1,687,000 per graduate (assuming no attrition). It should be stressed that this analysis is descriptive, rather than prescriptive; there should be no presumption that this cost structure is optimal, even given funding constraints. 27. Total Costs. Comprehensive data on total costs do not exist for government-run schools, let alone private schools. The 1985/86 central budget for health training allocates Rp.9.30 b11lion from the routine budget, and Rp.lO.60 billion in recurrent costs from the development budget. This sum includes the budget of all vertical paramedical schools; Pusdiknakes and Pusdiklat headquarters; and an indeterminate amount of training activities. Data on provincial allocations for health training are not available. A very rough order-of-magnitude estimate can be obtained by applying the cost equation derived above to a system (under all ownerships) with 365 schools and 55,000 students, yielding annual costs of Rp.26.24 billion. E. Pekarya Kesehatan 28. Pekarya kesehatan, a type of auxiliary paramedic, constitute a major additional class of health worker, trained outside of the school structure discussed above. Recruited, trained for four months, and posted at the district level, pekarya are primarily assigned to health centers, though some are assigned to hospitals. The program was designed to meet two major staffing problems: first, the need to rapidly expand paramedical personnel to meet REPELITA IV goals; and second, the difficulties of assigning paramedical school graduates to remote rural health centers. 29. REPELITA IV set a five-year goal of training 20,000 pekatYa. Some 6,323 were actually recruited in the first two years of the plan, and posts for an additional 5,000 were authorized for 1986/87. Responsibility for training them was assigned not to Pusdiknakes, which supervises the paramedical schools, but to Pusdiklat, the KOH unit primarily devoted to in service training. Candidates are recruited ideally from the vicinity of the health center to which they are assigned. With the exception of those recruited in Irian Jaya and East Timor, all candidates are required to have a general high school education. Training takes place at the district level, in whatever facilities are available. The seventeen week training course includes 236 hours of classroom instruction and 436 hours of field practice. - 159 - ANNEX V Pale 12 of 14 Emphasis is placed on health center administrative tasks; assistance to paramedics and doctors (e.g., blood pressure readings, equipment sterilization, and first aid); and health education activities.1/ 30. Pusdiklat's proposed 1986/87 budget set aside direct costs of Rp.441 million for training 4,850 health center pekarya and Rp.70 million for training 1,000 hospital-based pekarya the implied unit costs are about Rp.90,OOO and 70,000 respectively. Costs vary considerably between provinces, however; the unit cost in Irian Jaya, for instance, is Rp.194,OOO. F. In-Service Trainin& 31. Mindful of the need to increase the quality of health staff, REPELITA IV proposed a significant amount of retraining activity, including technical training for 270,000 participants, management and administrative training for 67,000 participants, and the training of 4,400 trainers. Under the recent reorganization of personnel training, in-service training activities were split off from preservice activities and assigned to Pusdiklat and Pusdiknakes respectively. The reorganized Pusdiklat has not yet reached full capacity; most training activities are funded and conducted outside Pusdiklat. During 1985/86, only 517 of the 19,959 trainees were trained in Pusdiklat facilities (BLKKs). Pusdiklat was, however, projected to grow rapidly with 24 BKLKs by the end of REPELITA IV (see Table 9). Table 9: PUSDIKLAT BUDGET, 1986/87 - 1987/88 1986/87 1987/88 Operational budget (Rp. billion) 1.14 2.15 Employees 483 688 BLKK (training centers) 9 17 Trainees (in-house) 4,770 12,160 A review is being carried out to determine how best Pusdiklat should be structured and financed, to establish coordinating procedures for training and to develop a national training program. 1/ The curriculum is outlined in Pedoman Latihan Pekarya Kesebatan pusat Kesehatan Kasyarakat, Ministry of Health, Pusdiklat, 1986. - 11>1 ANNEX VI HOUSEHOlD EXP ENDlTURE PATTERNS - 162 - ANNEX VI Pa&e 1 of 11 A. Survey Estimates of Expenditure 1. Household expenditure encompasses most but not all private-sector health expenditure. The principal additional source of private expenditure is employer-funded health benefits for employees. Smaller additional sources are the quasipublic insurance funds such as Jasa Raharja (transportation accidents) and ASTEK (occupation injuries). There are three recent sources of data on household health expenditure and use of health providers: (a) Susenas Survey. 1984. The large (50,000 households) nationwide survey focused on household consumption patterns, particularly food consumption. A limited number of questions on health expenditures were included. Respondents were asked the value of expenditure during the past month on: doctors; perawatan ( "nursing" or "care" usually interpretable as inpatient hospital care); midwives; contraception; traditional healers; medicine with prescription; medicine without prescription; other direct costs; and health and accident insurance. A noteworthy feature of the survey is the availability of data on the household's economic status. The survey uses total household consumption (including value of nonmonetary home production) as a proxy for household income. All individuals are classified according to the per capita consumption level of their family. (b) MOH Household Health Survey 1985-86. This very large survey, fielded between August 1985 and May 1986, covered over 300,000 individuals in 58,000 households. It covered only seven provinces: Yogyakarta, Bali, North Sulawesi, Bangkulu, Yest Kalimantan, Maluku, and Yest Nusa Tenggara. The basic questionnaire includes demographic data on all household members, and basic household characteristics (not including, however, a monetary measure of total consumption, income or wealth). Supplemental questionnaires deal with morbidity, mortality, pregnancy, and births. Of particular interest is a series of questions in the morbidity questionnaire on health service providers and their unit costs. Unfortunately, the questionnaire does not seek to ascertain total spending on health, or total contacts with health providers. Instead, it asks for details (including cost) concerning the single most important contact made. It asks about drugs present in the household, but does not ask for expenditure on drugs. There are also questions on the cost of inpatient care, including whether the cost was borne by a third party. (c) FKH Rural Yest Java Survey. 1985. This survey, carried out during summer of 1985 in two Yest Java regencies, comprised 2,700 households. Unique features of the survey include: data on all contacts with health providers (including drug vendors); and differentiation between transportation costs and service costs. - 163 ANNEX VI Page 2 of 11 2. These sources of information unfortunately provide an inconsistent view of total household spending on health. SUSENAS estimates a nationwide total of Rp.327.02 billion in out-of-pocket household health spending for 1984; this is equivalent to Rp.2,088 per person per year (see Table 1). This sum may in fact be substantially augmented by employer spending. A sample of ASTEK-registered Jakarta firms spent an average Rp.132,OOO per employee during 1983; some large employers are known to spend much more. AS'suming a national average of Rp.25,OOO per employee per year in employer provided health benefits, employer-payments would total Rp.8,125 billion nationwide, implying an a&iitional Rp.525 per person for the nation. 3. The MOH Household Health Survey, as noted above, underestimates total expenditure by excluding multiple contacts with health providers and excluding most ,expenditure on drugs. It probably does however, include some third I party (mainly employer) expenditures. Average spending on inpatient and outpatient care over the entire sample (including those who did not report a sickness episode) was equivalent to Rp.3,542 per person per year. Blown up to the national level, this implies over Rp.584 billion in health spending. This figure does not, however, include most over-the-counter drug expenditures; maternity expenditures; or expenditures on patients who subsequently died. MoreolTer, the sample excludes Indonesia's five largest urban areas, which have high concentrations of hospitals, specialist doctors, and drug stores and can be expected to have higher than average levels of per capita health expenditures. Inclusion of these categories of expenditures would substantially boost the national total, probably to a level at least double the SUSENAS estimate. The discrepancy in the estimates based on the two different samples may be attributable to the greater precision and depth of the MOH Health Survey as well as the interposition of more than a year of inflation. 4. Preliminary findings from the FKM survey of West Java estimated annual per capita spending at Rp.3,124 per person; Rp.135 of this amount was for t['avel (cash cost of transportation only, not including imputed value of time costs). Of the three survey questionnaires, that of the FKM is certainly the most comprehensive in ascertaining household health costs. However, the FKM sample is not representative, because it excludes urban areas. Bold extrapolation from this West Java result to a national total, assUDling a 3: 1 urban/rural differential, gives a national spending level of about: Rp. 627 billion (exclusive of transportation costs), again not including spending by employers. 5. Neither of these household surveys provides results that appear to be conslstent with the estimate of drugs sales provided by POM (the Food and Drug Dire(!torate-General of the MOH). Based on a nonrepresentative but adjusted sample of pharmacies, POM estimates the nationwide wholesale value of drug sales at Rp.649 for 1985. Approximately 15% of this total is said to be purchased by the government, with the remainder marked up (the official markup is 4;1.4%) and sold to private retail purchasers. This implies estimated private purchases of over Rp.770 billion. POM is attempting to cross-check this estimate and reconcile it with the household surveys, which are nearly an order of magnitude lower. Table 1: HEALTH-RELATED SPENDING BY ECONOMIC STRATUM, 1984 Monthly Month1~ Rer caRita eXRenditure (ruRiah) /a household Number of Inpatient Mid- Contra- Megicine expenditure individuals Doctors care wives ception Dukun Pre- No pre- Other Total per capita (Bidan) scrip- scrip (Rp.thousand) tion tion Under 5 5,142,526 3 2 0 0 5 1 6 2 19 5 - 6 5,713,433 6 4 1 0 5 2 8 3 29 6 . 8 18,136,422 10 6 2 0 8 2 10 5 43 8 - 10 21,471,232 13 8 3 0 9 4 15 6 58 10 . 15 45,695,651 24 12 4 1 10 10 20 9 91 ..... 0 15 - 20 25,215,703 46 19 11 1 11 24 27 12 151 .po. 20 - 30 21,438,038 83 41 13 2 12 64 40 18 273 30 - 40 7,274,916 164 107 28 15 13 109 53 32 521 40 - 60 4,479,237 267 122 29 9 15 221 78 22 763 60 - 80 1,260,733 405 98 27 11 20 341 96 41 1,039 80 and over 791,029 941 659 34 11 21 1,428 136 151 3,381 All 156.§l§.920 21 II 1 Z 10 .n II 12. ill LA Health spending in the highest economic class may be misleading here. The economic classification is based on total spending during the survey month. Relatively low income households with high medical bills during the survey month will be classified in a high economic class because of that spending. Source: Central Bureau of Statistics, Penge1uaran Untuk Konsumsi Penduduk Indonesia per Provinsi 1984, 1-'3> Jakarta. Number of households by expenditure class: Seri I, pages 7-10. Monthly expenditure by ~~ ...... t:rl category: Seri 3, pages 133-134. t'Il X w< H o H> ..... ..... - 165 - ANNEX VI Page 4 of 11 6. In sum, it is difficult to develop a very precise estimate of private health spending with currently available data. The problem is compounded by lack of information on the impact of the October 1986 devaluation on the price of health care. However, current spending probably exceeds Rp.l.5 billion. B. Spending by Income Class 7. SUSENAS is the only source of data for spending by economic class. These data must be interpreted with caution. First, as noted, SUSENAS apparEmtly underestimates total health expenditure. Second, SUSENAS does not report actual household income; instead, it uses total household spending (on all goods and services) as a proxy for income. But the relation between total spending and health spending is necessarily more positive than the relation betwelm income and health spending. Consider, for instance, a household with a mont:hly income of Rp.50,000, and usual monthly expenditures of Rp.40,OOO. If, during the survey month, the household had to pay a hospital bill of Rp.100,000, it would be classified in the Rp.140,000 total expenditure class instead of the Rp.100,000 health expenditure class; the mean for this class would therefore be skewed upward. With this caution in mind, Table 1 shows monthly per capita health expenditure by income (actually total expenditure per f.unily member) class. Spending is highly income-elastic. In the lowest 7% of the population with income under Rp.6,000 per family member per month, average health spending was Rp.29l per person per year. In the Rp.10 l5,000/person/month income category, which spans the 32nd to 51st percentiles, health spending averaged Rp.12,468 per person. 8. Some perspective on the size of health expenditure is provided in Table 2 which compares spending on health with spending on tobacco and betelnut. Expenditure on tobacco and betelnut greatly dominates in all but the highest income categories. This is not to suggest that it would be feasible to divert tobacco expenditure to health, but merely to establish a crude benchmark for affordable levels of nonsubsistence expenditure. C, Spendin& by Service Cate&o[y 9. Again SUSENAS provides a nominally comprehensive picture, but one with potentially serious flaws. The SUSENAS breakdown is shown in Table 3. The largest single component of expenditure is for drugs, which comprise 36.2% of the total. As noted, however, this is drastically at variance with the MOH estimate. Doctors' fees absorb another 29.9%; another 15.5% is devoted to "care", presumably inpatient care for the most part. Note that these estimates implicitly include payments both to private and public providers. SUSENAS does not, however, ask an explicit question about payments to private practice paramedics and it is not clear whether respondents would include such payments in any of the listed categories. The household surveys show that this is an important item of expenditure. A lower bound for the ratio of spending on private practice paramedics to private doctors ranged from 0.07 (Maluku) to 1.02 (West Kalimantan) according to the MOH household survey. - 166 - ANNEX VI Pale 5 of 11 Table 2: SPENDING ON HEALTH AND TOBACCO BY ECONOMIC STRATUM, 1984 Monthly household expenditure Number of Monthly per capita spendin, (Bp,) per capita individuals La All health items Tobacco and Betel (Rp. thousand) Under 5 5,142,526 19 162 5 - 6 5,713,433 29 231 6 8 18,136,422 43 296 8 10 21,471,232 58 418 10 15 45,695,651 91 652 15 20 25,215,703 151 988 20 30 21,438,038 273 1,359 30 40 7,274,916 521 1,786 40 60 4,479,237 763 2,232 60 80 1,260,733 1,039 2,607 80 and over 791,029 3,381 2,857 156,618,920 ill 823 La Each individual is classified by his/her family's per person total consumption level (roughly equal to income per person). Source: Central Bureau of Statistics, Pen,e1uaran Untuk Konsumsi Penduduk Indonesia 1984, Seri 1, Table 3.1. Table 3: HOUSEHOLD HEALTH EXPENDITURE: SUSENAS 1984 BREAKDOWN Annual expenditure Percent (Rp. million) of total Doctors 97,730 29,9 Inpatient care 50,745 15,5 Midwives 15,035 4.6 Contraception 3,759 1.1 Oukuns 18,794 5.7 Drugs (prescription) 71,418 21.8 Drugs (other) 46,976 14.4 Other 22,553 6,9 Total 327.020 100.0 Source: Central Bureau of Statistics. - 167 - ANNEX VI Pase 6 of 11 10. Table 4 shows SUSENAS estimates of spending on health and accident insurance. Total reported spen9ing is about Rp.17 billion annually, equivalent to 5t of the SUSENAS-reported direct health payments. Insurance spending is highly concentrated among the higher-income classes, with the top 1.3t of the population in per capita total expenditures accounting for over 40t of health and accident premium payments. Amounts reported here may overlap somewhat with mandatory premiums paid to two quasi-public insurance firms, ASKES (civil servants' health insurance) and Jasa Raharja (health and accident insurance for airlines, trains and motor vehicles). ASKES collected Rp.41 billion in contributions during 1983/84; Jasa Raharja collected Rp.43 billion in payments in 1983. Table 4: EXPENDITURE ON HEALTH AND ACCIDENT INSURANCE, 1984 Per capita Number of Per capita Total expfmdi ture individuals insurance Expenditure class in class expenditures per class (Rp. thousand) (Rp) (Rp. thousand) Under 5 5,142,526 0 0 5 - 6 5,713,433 0 0 6 - 8 18,136,422 0 0 8 - 10 21,471,232 0 0 10 - 15 15 - 20 45,695,651 25.215,703 1 3 45,696 75.647 20 30 - - 30 40 21,438,038 7.274,916 10 30 214,380 218,247 40 - 60 4,479,237 56 250,837 60 - 80 1,260,733 172 216,846 80 and over 791,029 454 359,127 All. 156.618.920 2. 1.380,Z§1 Source: Central Bureau of Statistics, Penseluaran untuk Konsumsi Penduduk Indonesia 19§4, Seri 1, Table 6 (urban plus rural), item E5. D. User Costs for Health Care 11. The NOH household survey provides a wealth of information about user costs of health care (see Table 5). Respondents seeking outpatient health care (except self-treaters) were asked the cost of a single treatment at their most important source of care. Cost was defined to include the cost of prescribed the distribution of responses over all categories of providers. Just over 15t reported that care was "gratis". Some of these respondents may have had their expenses covered by ASKES or their employer (that is, the cost was "gratis" to the respondent). Just under 20t paid between Rp.100 and 399; - 168 ANNEX VI Pase 7 of 11 these are probably puskesmas clients, There are additional modal responses at Rp.500 (10.6%), Rp.l,OOO (18.1%) and Rp.2,000 (9.9%). The median payment is Rp.800, and 92% of patients pay less than Rp.6,OOO. There is a very small tail of high payments: 2% of reported payments are over Rp.20,000. Further investigation is needed to determine whether these are coding errors; but these are plausible figures for a specialist consultation plus medicine or a minor surgical procedure, In either case, these rare but high payments will boost computed average expenditure by a few hundred rupiah, and will result in very high standard deviations of expenditure. Table 5: OUTPATIENTS BY COST OF TREATMENT IN SEVEN PROVINCE, 1985/86 Cost class LA. (RuRiah) Percent in Cumulative From To class percentage 0 15.3 15.3 1 99 0.3 15.5 100 199 7.0 22.6 200 299 6.6 29.3 300 399 5.8 35.0 400 499 1.8 36.7 500 599 10.6 47.3 600 699 0.8 48.1 700 799 2.1 50.2 800 899 1.2 51.4 900 999 0.2 51.6 1,000 1,999 18.1 69.7 2,000 2,999 9.9 79.6 3,000 3,999 6.4 85.9 4,000 4,999 2.3 88.2 5,000 5,999 3.5 91. 7 6,000 6,999 2.9 94.5 10,000 19,999 3.5 98.0 20,000 29,999 0.9 98.9 30,000 and higher 1.1 100.0 Total 100.0 100.0 LA Respondents who reported being sick during the reference month were asked to designate the single most important source of treatment; this table includes all responses excluding self-treatment. Costs are for a single treatment (visit) including cost of drugs, but excluding transportation cost. Zero-cost responses may include patients insured by ASKES. Sourcs: Preliminary tabulations, Ministry of Health Household Survey. - 169 - ANNEX VI Page 8 of 11 12. Table 6 reports average costs by type of provider and province. There are substantial differences between provinces as well as substantial variation with provinces for each type of provider. PUSKESKAS visits are relatively inexpensive in the inner provinces of Yogyakarta and Bali (at Rp.259 and 463), but more expensive elsewhere, ranging from Rp.589 in West Nusa Tenggara to Rp.2,085 in West Kalimantan. In each case, the standard deviation is much greater than the mean, indicating that a small but extreme right: tail of high-spenders is pulling up the mean. These means are, of coune, higher than the official charges of Rp.150 to 300. It is impossible to dE!termine, however, whether the higher reported figures reflect payments for drugs not available at the PUSKESKAS, or whether they represent higher actu.nl service fees (official or unofficial). Patients who reported seeing the J'USKESKAS doctor also reported paying more than patients who only saw a nUrSE!; but this pattern is consistent with both hypotheses. 13. The next most expensive sources of care were dukuns (traditional healEirs) and private practice paramedics. Cost per dukun visit averaged Rp.l,lOO or below in four provinces, and from Rp.2,300 to 5,100 in the rematning three. For paramedics, provincial averages fell in the Rp.900 to 2,30() range. Doctors in private practice were significantly more expensive, with provincial averages ranging from Rp.3,600 to 9,500. Generally the most expensive source of outpatient care was at hospitals, both private and pu.blic, wherf~ provincial averages ranged from Rp. 5,200 to 14,700 (government hospitals) and from Rp.2,700 to 18,000 (private hospitals and clinics), In the c:ase of government hospitals, these charges are far above the standard offic:ial rates, and must indicate some combination of additional drug purchase, laboratory fees, and unofficial service charges. 14. Data on cost of inpatient care were based on a very small number of cases. Charges at government hospitals averaged Rp.26l,000 over 147 cases; the ]6 cases at private hospitals reported average charges of Rp.446 ,000. The standard deviations were 804,000 and 942,000 respectively, indicating a wide rang." of variation. IU1.L.i' PRIKCIPAL SOURCE AIID AYEltAGE COST 0,. PAID OUTPATIEIIT CARE IK SEVEIf PIIOVIBCES, 1985/86 a. PrOYlder YUYUarta Snare of. .1.lt. A·. coat par .1.1t SbAre ·of .1alt. lall "'._at par .1.lt lord.. Sllla_.l Sh.are of. .lalta A.. _.t hll&k»lu lha"e of A.. _at par .lalt .1.1ta per .1.1t __ ".,tof ..,. coat Share .1.1ta Iall"P5& ",Juku "',t .~. t'MltD Shan of A.. coat Share; ..,. coat per .1alt .1alta per .1alt .1.lt. par .1.1t Pwtb._. 62.6 259 :n.9 463 43.9 839 39.4 1,323 74.6 2,085 0\3.8 779 61.9 589 eo-_nt 6.5 5,168 5.9 6,421 5.9 8,207 8.0 9,505 1.4 14,704 13.9 6,247 4.4 6,762 ho.pltals Prh'ate ho.pltals Ji. 7.0 7,921 0.9 5,850 0.9 18,086 0.6 10,778 1.777 2,738 3.2 5,771 0.6 13,644 Par_dlc. a ,... "'-oJ Prl.ate practlc. 6.7 1,599 34.9 1,543 14.0 1,367 19.4 1,794 14.0 2,270 14.7 1,127 16.6 .19 0 Doctor. I Prl.at. practice 16.4 5,990 21.6 3,587 34.7 5,868 26.9 7,640 4.8 6,500 24.2 9,426 11.5 3,185 Dukun 0.8 938 4.8 927 0.7 1,100 5.7 2,353 3.6 5,142 0.6 4,056 5.0 796 laW. W!.:.2 1.a..1U 1Sl.!L.2 l...W. 1Sl.!L.2 L.W. W!.:.2 l..ll!l. W!.:.2 1..W. 1Sl.!L.2 1...U§. W!.:.2 LoW a. . .apoolld.enta I.nclud.e par.-. who _t tha fo11_111& crlteria. (a) alck dur1ll& r.ference _th, (b) .00000t tna_tl (c) dld 111. lI:ePOrt that t n a _ t _. w aratl. w, (d) d1d DOt report ~ aelf-tr.a~ aa tI:t.e prl.ncipal aourc. of trea_t. Bach re.ppndant va. a.bel to lI:ePOrt tI:t.e a1ll&1e . .at ~rtaot aouzce of trea_t, .0 th1. tabl. . .elude· ·ecoadacy contacta aDd therefore probably UD4er·· t~te. u.e of dukuAa, par....1c., aDd puab..... Trea~nt co.t la for a .1n&le treat.aD.t, includ1ll& prOYicler'a f ·· aDd any addlt10nal co.t for drua., ~ 1f obtained elaavbanl tranaportatlon DOt I.ncluded. Ji. wPri.ate ho.p1tala w lnclud·· both ho.p1tal. and clln1ca. a wPar_d1cw includ.e. both par_dica and wkacler Eaaehetanw. ~I~ ..... t:ri :>< II) ~I< H o HI ,... .... - 171 - ANNEX VI Pale 10 of 11 E. Choice of Provider 15. Table 6 also provides information on the market share of the different types of outpatient treatment providers. Again only the "principal" provider for each illness episode is included. 16. Patterns of provider choice vary tremendously between provinces, prevencing any simple generalization to the national level from this atypical sample. Dependence on puskesmas, for instance, varies from a low of 32% in Bali tl) a high of 75% in West Kalimantan. Of particular interest is the large market share achieved by private providers, who cover more than half the market in each of three provinces. For the most part, these providers are the same doctors and paramedics who provide services at public facilities at a fractil)n of their private fee. Their private clients pay a premium for some combinltion of more attention, more drugs (especially injections), shorter waiting time, and more conveniently scheduled office hours (or in the case of paramedics, house calls). Anecdotal evidence suggests that injections are a particularly powerful drawing card; it should be noted, however, that paramedics are not legally empowered to give unsupervised injections. TraditLona1 healers have a relatively small share, or else they are used mainly for initial consultations, rather than principal treatments. This result is consistent with other surveys (including the FKM survey), although anecdo·t:a1 evidence tends to suggest that dukuns play a more important role. F. Morbidity Rates and Utilization Rates 17. Table 7 shows morbidity rates and provider utilization rates for the seven provinces in the MOH household study. Self-reported illness rates are shown in the first row. The morbidity rate is in the range of 8 to 9% per month, with the outliers of Yogyakarta (4.7%) and West Kalimantan (10.4%). The se,:!ond row shows the proportion of the total population who treated their i11nes,9, including self-treatment. The third row, of principal interest here, is the proportion of the total population who sought treatment for an illness during the past month. With the exception of Yogyakarta, this rate is remark.ab1y constant between provinces, ranging between 4.2% and 4.7%. For comparison ASKES and DUKM populations use outpatient services at a rate of about 18% to 20% per month. 18. The final row of Table 7 shows that the hospitalization rate for the general population is on the order of 4 to 12 cases per 10,000 population per month. - 172 - ANNEX VI Page 11 of 11 Table 7: MORBIDITY AND TREATMENT RATES IN SEVEN PROVINCES, 1985-86 (per 1,000 population) North West West Nusa Yogyakarta Su1a- Beng- Kali Teng Bali wesi kulu mantan Maluku gara Illness during past month 47.4 78.8 82.0 83.4 106.9 91.8 88.3 Illness was treated 32.3 60. ,9 55.4 54.6 61.6 53.8 51.7 Treatment by doctor, para medic, or dukun 27.1 46.7 46.1 41.9 44.0 45.0 43.2 Hospitalization. 0.98 0.45 1.29 0.68 0.52 0.53 0.39 Total Respondents LA 41,842 38,065 41,832 45,561 43,883 43,771 45,591 LA Number of respondents includes all individuals, sick or well, regardless of age, covered in the survey. Source: Preliminary tabulations, Ministry of Health Household Survey. - 173 STATISTICAL ANNEX - 174 STATISTICAL ANNEX LIST OF TABLES Health Indicators 1.1 Infant Mortality Rates by Province, 1971-85. Public Finance 2.1 Central Government DIP Budget for Health by Program, 1979/80-1987/88 2.2 Central Government INPRES Budget for Health, 1979/80-1987/88 2.3 Central Government Routine Budget for Health, 1979/80-1987/88 2.4 Central Government SBBO-RSUD Grant, 1982/83-1986/87 Hospital Statistics 3.1 General Hospitals by Type and Ownership, 1978/79-1985/86 3.2 General Hospital Beds by Type and Ownership, 1978/79-1985/86 3.3 General Hospital Beds per 1,000 Population, 1978/79-1985/86 3.4 Summary Service Statistics by Class of Hospital, 1985 3.5 Service Statistics for all Hospitals by Province, 1985 3.6 Service Statistics for MOH Hospitals by Province, 1985 3.7 Service Statistics per capita ror all Hospitals by Province, 1985 3.8 Service Statistics per capita for MOH Hospitals by Province, 1985 3.9 Manpower at all Hospitals by Province, 1985 3.10 Manpower at MOH Hospitals by Province, 1985 3.11 Estimated Unit Costs for Hospital Services Health Center Statistics 4.1 Provincial Distribution of Community Health Facilities, 1986 4.2 Ratios of Puskesmas to Population by Province, 1986 4.3 Provincial Distribution of Posyandu, 1986 4.4 Health Center Staff by Province, 1985 4.5 Proportion of Puskesmas with Doctors by Province, 1985 Health Manpower Statistics 5.1 Ministry of Health Medical and Paramedical Staff by Province, 1986 5.2 Ministry of Health Medical and Paramedical Staff by Budgetary Category, 1986 5.3 Ratio of Ministry of Health Medical and Paramedical Staff to Population, by Province, 1986 5.4 New Ministry of Health Posts, 1979/80 to 1986/87 5.5 Paramedical Formasi Created 1984/85 and 1985/86 5.6 Doctors and Dentists Formasi, 1984/85 and 1985/86 5.7 Proportion of New Paramedical Posts Ever-filled by Province, 1979/80 1985/86 5.8 Supply and Demand for INPRES Nursing Positions, 1985/86 5.9 Provincial Distribution of Paramedical Staff, 1986 5.10 Medical Manpower Supply and Government Demand, 1979/80-1985/86 5.11 INPRES Formasi and Appointments for Doctors, 1986/87 5.12 INPRES Formasi and Appointments for Dentists, 1986/87 5.13 INPRES Formasi and Appointments for Doctors and Dentists, 1984/85 and 1985/86 Combined 5.14 Rough Estimates of Medics Outflow and Inflow by Province, 1985 - 175 - STATISTICAL ANNEX Table 1.1 INFANT MORTALITY RATES BY PROVINCE, 1971-85 Average Rate of Decline Per 1.000 liv~ ~irthl (:t 12 sA.l LA Province 1971 1980 1985 1971-80 1980 85 D. I. ACI~h 141 91 47 4.8 13.2 North Sumatra 120 89 64 3.2 6.6 West Sumatra 151 121 76 2.4 9.3 Riau 141 113 55 2.4 14.4 Jambi 155 118 60 2.9 13.5 South SlJDlatra 151 118 71 4.8 10.2 Bengkuh.1 166 106 62 5.0 10.7 Lampung 147 97 59 4.5 9.9 DKI Jakarta 126 80 36 4.9 16.0 West Java 165 129 89 2.7 7.4 Central Java 143 96 65 4.4 7.8 D.I. Yogyakarta 98 62 29 5.1 15.2 East Ja'\7a 119 99 74 2.0 5.8 Bali 127 88 58 4.0 8.3 West Nusa Tenggara 219 187 145 1.7 5.1 East Nusa Tenggara 151 124 74 2.2 10.3 East Timor 69 West Kalimantan 143 116 57 2.3 14.2 Central Kalimantan 128 100 58 2.7 10.9 South Kalimantan 165 121 83 3.3 7.5 East Kalimantan 151 99 40 0.7 18.1 North Sulawesi 114 94 50 2.1 12.6 Central Sulawesi 146 128 94 1.4 6.2 South Sulawesi 159 108 69 4.2 9.0 Southeast Sulawesi 191 114 73 5.6 8.9 Ka1uku 145 124 68 1.7 12.0 Irian Jaya 113 106 38 0.8 20.5 Indonesia 143 107 70 3.2 8.5 LA Reference periods are: 1968-69 for the 1971 census, 1977-78 for the 1980 cen,sus and 1982-83 for SUPAS 1985. Source: Central Bureau of Statistics. CENTRAL DEVELOPMENT EXPENDITURE (A!'BN-DIP) ON HEALTH BY PROGRAM, 1979/80-1987/88 (BUDGET) (Rp. billion,,) REP§LITA III REPELITA IV 1979/80 1980/81 1981/82 1982/83 1983/84 1984/8S 1985/86 1986/87 1987188 Youth mo_nt 0.2 0.3 0.4 0.4 0.4 0.4 0.4 0.2 0.0 Manpower training 2.1 6.0 6.6 9.7 10.0 13.0 IS.O 13.6 6.8 Health education 0.8 1.2 1.3 1.4 1.3 1.8 2.0 loS O.S Health aervice" 30.7 42.3 S6.S 66.8 68.4 62.3 S1.S 24.1 8.7 Communicable diaeaae control 10.7 16.S 19.5 22.5 23.8 2S.3 2S.9 13.9 4.3 Nutrition 1.0 3.0 3.3 4.0 3.S 4.0 S.2 4.0 1.6 I-' -.J Food and drus a~ini"tration 1.3 2.1 2.4 3.0 3.0 4.S 3.3 2.2 O.S 0'\ Women" movement 0.3 0.4 0.4 O.S 0.4 0.3 0.3 0.2 0.0 Water aupply 0.8 1.S 2.1 2.5 2.0 loS 1.7 O.S 0.3 Sanitation 0.2 O.S 0.6 0.8 0.6 0.6 0.6 0.3 0.1 Reaearch and development 0.8 0.9 1.0 1.1 0.9 1.4 1.4 0.7 0.2 Management 0.5 0.6 0.9 1.1 0.9 1.0 0.9 O.S 0.1 Infra"tructure 1.0 2.9 2.8 S.S 4.0 3.1 3.6 3.7 0.0 l2.W. ~ l.!:J. .!LJ. ll!..1 ll!..1 119.1 111.7 .§.l..! .u.:.2 HCI.l III H 0"> ..... H ro H CI.l NH · H 1-'0 ~ ~ ~ CENTRAL IIPRES EXPENDITURE OM HEALTH 1979/80-1987/88 (Ln Rp. bLllLons) RE~ELITA III REPELITA IV 1979/80 1980/81 1981/82 1982/83 1983/84 1984T85 1985/86 1986/87 1987/88 Health centers 21.1 34.4 55.9 74.0 72.6 78,0 84.9 84.9 74.6 Of whLch: druas (12.7) (21.6) (29.7) (38.0) (38.9) (40.3) (45.3) (55.2) (67.5) ...... '-l Water supply 7.8 14.5 20.6 21.4 21.9 18.8 28.0 28.0 1.7 '-l Sanitation 1.1 1.1 2.5 3.0 4.0 1.6 1.7 1.6 0.0 Toul l!L.9. ~ l2.:.!!. .!!:.l 98,5 ....2.l..! ~ ~ 1J...1 t-'lCll III t-'l 0"> 1-'t-'l CD H CIl Nt-'l · H NO ~ ~ f:J CENTRAL ROUTllIE EXPEIJDITUU (APBlf-DIlt) Olf HEALTH BY PROGRAM, 1919/80-1981/88 (BUDGET) (ltp. blll1on.) _____ mlLI TA 111 R.EPELlTA IV 1919/80 1980/81 1981/82 1982183 1983/84 1984/85 1985/86 1986/81 1981/88 Secretary .eneral 9.5 18.9 2.5.8 21.1 21.1 21.8 31 · .5 38.1 40.4 Education and trainina 1.9 9.2 12.2 12.0 In.pector .eneral Medical care 0.2 19.3 0.2 28.3 0.3 41.4 0.3 43.0 0.4 46.3 0.4 S3.2 0.4 63.4 0 · .5 72.6 0.5 73.6 .. ..... Community health 1.1 1.6 2.2 2.3 2 . .5 2.S 3.0 3.8 3.2 00 Communicable di.ea.e control 1.2 loS 2.2 2.3 2.S 2.6 3.2 3.1 3.7 Pood and drua adBdnl.tratlon 1.1 1.3 2.4 2.8 3.1 3 · .5 3.1 .5.1 4.1 Reaearch and development 1.1 1.9 1.6 2.1 I2W ~ a..l 74.4 lL1 82.4 !Ll ll.!.J. ll!.:.1 .u.!!...1 ~ tIl fIl~ 1-'1-3 fDH til 1-3 · IH ~ ~ trJ >< - 179 STATISTICAL ANNEX Table 2,4 CENTRAL SBBO-RSUD GRANT FOR HEALTH, 1982/83-1986/87 (Rp, millions) 8 Teaching 22 District 299 District Total Hospitals Hospitals Hospitals 1982/83 6,126 1,625 7,751 1983/84 6,126 2,017 ° 8,143 198~/85 6,390 1,810 ° 8,200 198~,/86 8,077 1,376 ° 9,453 1986/87 8,439 1,347 ° 7,805 17.592 "·'--1 F - 180 STATISTICAL ANNEX Table 3.1 NUMBER OF GENERAL HOSPITALS BY TYPE AND OWNERSHIP 1978/79-1985/86 Type 1978/79 1979/80 1980/81 1981/82 1982/83 1983/84 1984/85 1985/86 MOil CentIill 10 11 11 11 13 13 13 14 Class A 1 1 1 1 1 1 1 1 Class B 8 9 9 9 9 9 9 9 Class C 1 1 1 1 3 3 3 3 HOH LOCil1 265 ill 288 295 ill 295 302 300 Class A 1 1 1 1 1 1 1 1 Class B 5 5 5 5 5 6 6 6 Class C 41 42 42 42 41 76 76 76 Class D 218 230 240 247 248 212 219 217 HOH Total ill 289 299 306 308 308 315 313 Class A 2 2 2 2 2 2 2 2 Class B 13 14 14 14 14 15 15 16 Class C 42 43 43 43 44 79 79 79 Class D 218 230 240 247 248 212 219 217 Other ministry 129 129 130 130 115 115 115 115 NGO lJj 76 77 78 76 75 76 78 80 Private sector 132 134 135 149 157 167 169 175 Total public 394 407 418 425 410 410 417 415 Total private LQ 208 211 213 225 232 243 247 255 Grand total 602 618 631 650 642 653 664 670 lJj Non-governmental organisations. LQ NGOs and Private sector. - 181 - STATISTICAL ANNEX Table 3.2 NUMBER OF GENERAL HOSPITAL BEDS BY TYPE AND OWNERSHIP 1978/79-1985/86 Type 1978/79 1979/80 1980/81 1981/82 1982/83 1983/84 1984/85 1985/86 MOH Central 5.610 ~ 6.794 ~ 1......a2l 7,800 7.928 7,992 Class A 1,280 1,388 1,389 1,360 1,335 1,388 1,388 1,388 Class B 4,029 5,026 5,055 5,025 5,993 5,868 5,996 6,024 Class C 301 314 350 350 564 544 544 580 HOH Local 28,090 29,286 29,937 30,359 31,379 33,270 34,454 35,148 Class A 1,483 1,466 1,518 1,512 1,504 1,513 1,516 1,530 Class B 2,946 3,018 3,019 2,936 2,933 3,347 3,329 3,372 Class C 7,851 8,307 8,590 8643 8,780 14,176 14,303 14,667 C1as.s D 15,810 16,495 16,810 17,268 18,162 14,234 15,306 15,579 MOH Total 33,100 36,014 36.131 37.Q94 39,211 41.Q10 42,382 43,140 Class A 2,763 2,854 2,907 2,872 2,839 2,901 2,904 2,918 C1a~;s B 6,975 8,044 8,074 7,961 8,926 9,215 9,325 9,396 Class C 8,152 8,621 8,940 8,993 9,344 14,720 14,847 15,247 C1a!.s D 15,810 16,495 16,810 17,268 18,162 14,234 15,306 15,579 Other ministry 13,025 12,895 12,525 13,178 12,072 11,481 11 ,565 11,539 NGO LJ! 9,412 9,345 9,310 9,381 8,826 8,854 8,911 8,762 Priva~:e sector 15,515 16,430 17 ,200 18,270 19,069 19,704 20,257 20,947 Total public 41,115 42,181 42,462 43,537 43,451 44,751 46,019 46,687 Total private LQ24,927 25,775 26,510 27,651 27,895 28,558 29,168 29,709 Grand, total 66,Q42 67,956 68,91, 11,18a 21,346 13.J09 7:2.187 16,396 12 Non-governmental organisations. LQ NGOs and Private sector. - 182 - STATISTICAL ANNEX Table 3.3 NUMBER OF GENERAL HOSPITAL BEDS PER 1000 POPULATION 1978/79-1985/86 Type 1978/79 1979/80 1980/81 1981/82 1982/83 1983/84 1984/85 1985/86 tllnllttx 2' HIAlkb I..A 0.234 Q.244 ~ 0,241 .Q....lli 0,256 0,259 0.258 Class A 0.019 0.019 0.019 0.019 0.018 0.018 0.018 0.017 Class B 0.048 0.055 0.054 0.052 0.057 0.057 0.057 0.056 Class C 0.057 0.058 0.059 0.058 0.059 0.092 0.091 0.091 Class D 0.110 0.112 0.112 0.112 0.116 0.089 0.093 0.093 Other ministry 0.090 0.087 0.083 0.086 0.077 0.072 0.071 0.069 NGO .& 0.065 0.063 0.062 0.061 0.056 0.055 0.054 0.052 Private sector 0.108 0.111 0,114 0.119 0.121 0.123 0.124 0.125 Total public 0.324 0.332 0.327 0.327 0.327 0.328 0.329 0.327 Total private ~ 0.173 0.175 0.176 0.180 0.178 0.178 0.178 0.177 ~t.ng t2t.1 2...ill 2....22§. 2.....2.Q1 2...2Ql 0.504 0,506 Q.~OZ 0.504 LA Including central and local government administered hospitals . .& Non-governmental organisations. ~ NGOs and Private sector. - 183 STATISTICAL ANNEX Table 3.4 SUMMARY OF SERVICE STATISTICS BY CLASS OF HOSPITAL Hospital Number of Dis- Bed- Outpatient Category Hospitals Beds charges days Visits lUni ~trx of Health 313 43.140 1.390.46Q 9.211.216 12.553,472 Class A 2 2,918 85,340 801,400 2,017,356 Class B 15 9,396 269,960 2,347,044 4,229,048 Class C 79 15,247 528,216 3,323,696 5,780,396 Class D 217 15,579 506,944 2,745,076 7,526,672 OthE,r ministry 115 11,539 22,780 1,904,212 6,359,132 NGO!: i.J! 80 8,762 18,260 1,410,904 4,461,996 Private sector 175 20,947 583,312 4,215,148 5,799,792 Toted public 428 54,679 1413,240 11,121,428 25,912,604 Totd private .&. 255 29,709 601,572 5,626,052 10,261,788 Grand total 683 84,38~ 2.014,U2 16.141,480 36,174,392 i.J! Non-governmental organisations . .&. NGOs and private sector. - 184 - STATISTICAL ANNEX Table 3,5 SERVICE STATISTICS FOR ALL HOSPITALS BY PROVINCE, 1985 Number of Outpatient Lenath of Occupancy Hoapitals Beds Diacharaea Bed-days Vis ita Stay Rate (Day) 0.1. Aceh 18 1,382 28,980 179,068 701,676 6.18 0.355 North Sumatra 73 10,206 198,880 1,641,576 2,989,428 8.25 0.441 Weat Sumatra 19 2,228 56,236 467,228 1,126,824 8.31 0.575 Riau 20 1,260 30,052 193,392 643,552 6.44 0.421 Jambi 9 682 11,620 59,724 291,416 5.14 0.2100 South Sumatra 32 3,351 91,540 602,112 2,120,636 6.58 0.492 Benakulu 5 290 6,920 102,3810 93,608 6.12 0.400 Lampuna 9 1,087 35,264 194,660 331,696 5.52 0.491 D.lt.I. Jakarta 36 9,681 255,196 2,178,904 5,047,3010 8.54 0.617 West Java 60 8,092 280,032 1,830,0410 4,246,188 6.54 0.620 Central Java 71 11,515 414,376 2,584,020 3,945,996 6.24 0.615 0.1. Yoayakatta 11 2,397 61,480 473,816 814,532 7.71 0.5102 East Java 89 13,704 432,612 3,020,372 6,241,800 6.98 0.604 Ball 16 1,836 65,908 438,132 686,240 6.65 0.654 Weat Nusa Tenaaara 9 6710 23,344 150,720 324,908 6.46 0.613 East Nusa Tenaaara 22 1,432 100,092 251,332 727,888 6.27 0.481 Eaat Timor 5 439 9,288 94,080 303,048 10.13 0.587 Weat Kal imantan 14 1,278 30,912 230,020 278,940 7.104 0.493 Central Kal1manatan 11 465 9,532 55,404 194,676 5.81 0.326 South Kalimantan 20 1,208 23,516 166,308 487,556 7.07 0.377 East Kal imantan 17 1,481 410,804 261,956 556,008 5.85 0.485 North Sulawesi 19 2,307 58,940 4100,356 700,568 7.47 0.523 Central Sulawesi 12 828 20,948 127,556 300,624 6.09 0.422 South Sulawesl 38 3,275 76,888 553,660 1495,856 7.20 0.463 Southeast Sula_si 12 648 14,324 86,788 490,272 6.06 0.367 Maluku 16 1,340 29,544 221,608 463,884 7.50 0.453 Irian Jaya 20 1,302 32,9410 202,260 569,268 6.14 0.426 Total ~ 84,388 2,384.172 16,747.480 36.174.392 !.:.!! .!L..!.!!! iie.iiOI!;III: Coefficient of variation 0.16 0.21 - 185 - STATISTICAL ANNEX Tagle 3,6 SERVICE STATISTICS FOR MOB HOSPITALS BY PROVINCE, 1985 Number of Outpatl.ent Lenath of Occupancy Bospl.tals Beds Dl.scharges Bed-days Vl.sl.ts Stay bte (Day) D.I. A"eh 9 86" 16,476 116,896 301,904 7.09 0.371 North ;;umatra 19 2,471 53,412 381,968 704,712 7.15 0.424 West S·.-tra 13 1,620 41,092 352,156 787,840 8.57 0.596 lll.au 7 611 15,292 97,452 309,808 6.37 0.437 Jamb I. 6 423 7,244 37,600 203,172 5.19 0.2U North Sumatra 10 1,293 41,364 286,656 718,636 6.93 0.607 Benskulu 4 268 6,496 38,796 78,236 5.97 0.397 LampunS 5 896 31,024 169,940 258,976 5.48 0.520 D.K. I. J.alta.rta 6 3,134 85,068 665,136 2,480,400 7.82 0.581 West Java 27 4,428 176,456 1.107,880 2,279,324 6.28 0.685 Centrll Java 37 7,164 288,188 1,757,992 2,777,8411 6.10 0.672 D.I. Yogy.alta.rta 5 1,0117 25,588 217,492 380,456 8.50 0.569 East ~f.va 37 7,387 270,780 l,793,U8 3525,632 6.62 0.665 Ball. 9 1,521 58,9211 391,852 600,1156 6.65 0.706 West Nusa Tenagara 6 570 20,388 128,808 261,1104 6.32 0.619 East Rusa Tenagara 13 786 26,732 144,904 571,688 5.42 0.505 East Umor 3 275 6,140 67,3110 188.428 10.97 0.671 West Kal1mantan 10 1,030 23,752 178,092 13::l,104 7.50 0.4711 Centr,l.l Kal1mantan 10 440 9,352 54,216 177.756 5.80 0.338 South Kal1mantan 11 698 13,544 89,696 306,332 6.62 0.352 Eastltal1mantan 7 809 27,024 156,588 279,512 5.79 0.530 North SulawesI. 6 961 29,336 212,864 329,740 7.26 0.607 Central Sulawesi 8 640 17,044 103,028 252,580 6.04 0.441 South Sulawesi 24 1,891 48,7611 344,500 803,028 7.06 0.499 Southeast SulawesI. 5 350 8,936 52,484 229.6611 5.87 0.411 Halulm 7 640 17,164 122,616 212,000 7.U 0.525 Irl.IU1. Jaya 9 923 24,880 146,816 395,840 5.90 0.436 I2U.!. ill 43,140 1,390,460 9,217.216 19.553,472 !.:E 0.496 "erao .ll!!!!: Coefficient of variation 0.18 0.23 - 186 - STATISTICAL ANNEX Table 3.7 SERVICE STATISTICS PER CAPITA 'OR ALL BOSPITALS BY PROVINCE. 1985 (per 1.000 populaUon) D18 Out Hacllcs Nurse Para- Non char.es patlent _cllcs ~lcs vlslts D.1. Aceh 0.46 9.72 60.08 235.40 0.04 0.27 0.04 0.11 North S~tra 1.08 21.06 173.82 316.54 0.13 0.45 0.06 0.39 West SUIII&tra 0.61 15.34 127.44 307.35 0.05 0.36 0.06 0.29 Rlau 0.50 11.95 76.92 25.5.96 0.04 0.28 0.0.5 0.21 Jamb 1 0.39 6.73 34.57 168.67 0.03 0.15 0.01 0.12 South SUIIII&tra 0.62 16.92 111.28 391.91 0.07 0.37 0.04 0.41 Be118kulu 0.31 7.39 45.29 100.03 0.04 0.25 0.03 0.15 0.18 5.89 32.52 5.5.41 0.03 0.12 0.01 0.12 D.I:.I. Jakarta 1.24 32.59 278.30 644.67 0.35 1.09 0.19 1.31 West Java 0.26 9.11 59.55 138.16 0.04 0.17 0.02 0.19 Central Java 0.43 15.38 95.94 146.50 0.06 0.20 0.02 0.28 D. I. YOI,.akarta 0.81 20.72 159.72 274.57 0.15 0.51 0.07 0.58 Bast Java 0.44 13.94 97.31 201.10 0.05 0.21 0.03 0.29 Ball 0.70 24.98 166.09 260.15 0.09 0.54 0.06 0.27 West Musa Te1181ara 0.22 7.66 49.47 106.64 0.02 0.12 0.01 0.09 East Nusa Tenaaara 0.47 13.24 82.97 240.30 0.01 0.22 0.01 0.20 Bast Timor 0.70 14.88 150.69 485.40 0.04 0.23 0.03 0.50 West KalLmantan 0.45 10.98 81. 70 99.08 0.02 0.15 0.01 0.10 Central 1lC&1Lmantan 0.41 8.36 48.61 170.81 0.02 0.36 0.04 0.09 South KalLmantan 0.53 10.28 72.67 213.04 0.04 0.30 0.04 0.18 Bast KalLmantan 0.96 29.14 170.37 361.61 0.07 0.5.5 0.04 0.31 North KalLmantan 0.97 24.82 185.41 294.97 0.10 0.51 0.02 0.37 Central Sulavesl 0.54 13.61 82.86 195.29 0.03 0.32 0.02 0.11 North Sulavesl 0.50 11.65 80.89 226.65 0.05 0.27 0.03 0.11 Southeast Sulavesl 0.60 13.22 80.10 452.51 0.04 0.41 0.04 0.19 Haluku. 0.82 18.09 135.73 284.11 0.03 0.35 0.02 0.17 Irlan Ja,.a 0.96 24.28 149.04 419.47 0.04 0.50 0.04 0.20 Memo It_= Coefflclent of varlatlon 0.44 0.46 0.53 0.51 1.03 0.57 0.88 0.89 - 187 STATISTICAL ANNEX Table 3,8 SERVICE STATISTICS PER CAPITA lOR MOB HOSPITALS BY PROVIIICIt, 1985 (per 1,000 populatlon) Beds Out Medlca aurae Para- lion Bed-~a pat lent ..alea _dles vhlts 0.1. AcfOb 0.29 5.53 39.22 101.28 0.02 0.16 0.03 0.07 0.26 5.66 40.45 74.62 0.07 0.21 0.03 0.16 West SUillAtra 0.44 11.21 96.05 214.89 0.04 0.28 0.04 0.20 Rlau 0.24 6.08 38.76 123.22 0.02 0.10 0.02 0.10 Jamb 1 0.24 4.19 21.76 117.60 0.02 0.11 0.01 0.07 South Sumatra 0.24 7.64 52.98 132.81 0.03 0.15 0.02 0.17 0.29 6.94 41.46 83.60 0.04 0.23 0.03 0.13 Lampuna 0.15 5.18 28.39 43.26 0.01 0.08 0.01 0.06 D.It.I .· ru.rta 0.40 10.87 84.95 316.81 0.18 0.38 0.06 1.48 West Ja',a 0.14 5.74 36.05 74.16 0.03 0.08 0.01 0.09 Central Java 0.27 10.70 65.27 103.13 0.04 0.12 0.01 0.16 D.I. Yog7u.rta 0.35 8.63 73.31 128.25 0.12 0.22 0.04 0.23 0.24 8.72 57.78 113.59 0.04 0.12 0.02 0.14 BaU 0.58 22.34 148.55 227.63 0.08 0.46 0.05 0.23 West lIu,.a Tena.ara 0.19 6.69 42.28 85.80 0.01 0.10 0.01 0.07 0.26 8.83 47.84 188.73 0.01 0.14 0.01 0.10 East Tloaor 0.44 9.83 107.86 301.81 0.04 0.23 0.03 0.50 West Ita lialantan 0.37 8.44 63.26 49.05 0.02 0.11 0.01 0.05 Central Xalialantan 0.39 8.21 47.57 155.97 0.02 0.32 0.04 0.09 0.30 5.92 39.19 133.85 0.02 0.22 0.03 0.09 East Xatialantan 0.53 17.58 101.84 181. 78 0.05 0.34 0.03 0.11 0.40 12.35 89.62 138.83 0.06 0.30 0.01 0.13 Central Sulavesl 0.42 11.07 66.93 164.08 0.02 0.28 0.02 0.09 South S"laved 0.29 7.39 52.20 121.67 0.03 0.17 0.02 0.05 0.32 8.25 48.44 211.98 0.01 0.23 0.02 0.08 Maluku 0.39 10.51 75.10 129.84 0.02 0.21 0.01 0.03 Irian J.,.a 0.68 18.33 108.18 291.68 0.02 0.40 0.02 0.09 !2t!! Memo It8lll: Coeffictent of variatlon 0.37 0.45 0.46 0.48 0.90 0.48 0.59 0.81 - 188 STATISTICAL ANNEX Table 3,9 IWfPOWER AT ALL HOSPITALS BY PROVINCE, 1985 Staff rat12 eer bed Medics Burse Para- Bon Med1c Nurse Total med1cs med1cs D.I. Aceh 107 790 126 322 0.077 0.572 0.973 Borth Sumatra 1,189 4,245 579 3,644 0.117 0.416 0.946 West Sumatra 198 1,316 236 1,0615 0.089 0.591 1.264 R1au 100 699 127 539 0.079 0.555 1.1153 Jamb 1 46 2150 25 200 0.0157 0.381 0.779 South Sumatra 352 1,988 207 2,211 0.105 0.593 1.420 Benakulu 38 233 215 137 0.131 0.803 1.497 Lampung 164 713 71 1595 0.151 0.1556 1.511 D.K.I. Jakarta 2,718 8,571 1459 10,529 0.281 0.885 2.377 West Java 1,302 5,132 587 5,831 0.1151 0.1534 1.588 Central Java 1,706 5,378 1537 7,417 0.148 0.467 1.315 D. I. Yogyakarta 455 1,510 200 1,7115 0.190 0.1530 1.1519 East Java 1,599 15,600 873 8,852 0.117 0.482 1.308 Ball 234 1,414 153 705 0.127 0.770 1.365 West Busa Ienagara 56 3154 36 279 0.083 0.540 1.091 East Busa Tenagara 44 1581 24 1512 0.031 0.4715 0.950 East Timor 24 146 115 311 0.055 0.333 1.132 West Kalimantan 153 424 23 289 0.049 0.332 0.625 CentraL KaLimantan 27 415 41 103 0.058 0.892 1.260 South Kalimantan 87 6915 95 421 0.072 0.576 1. 075 East Kalimantan 111 841 155 471 0.075 0.5158 1.005 Borth Sulawes1 243 1,203 48 871 0.105 0.521 1.025 Central Sulawes1 41 495 30 1159 0.050 0.598 0.888 South Sulawes1 3151 1,7815 223 731 0.110 0.545 0.947 Southeast Sulawesi 41 442 44 204 0.0153 0.682 1.128 Maluku 53 579 215 275 0.040 0.432 0.1596 Irian Jaya 57 1581 ·52 2157 0.044 0.523 0.812 !2W 11.416 47,602 6,029 48,597 .!hill 0.564 bill KelllO Ite!!!: Coeff1c1ent of variat10n 0.25 0.30 - 189 - STATISIIQAL ANNEX Iab1e 3.10 MANPOWER. AT MOB HOSPITALS BY PROVIICE, 1985 litaff ration J:!er beS Medics lurses Para- Ron- Medics luraea Total _dies medics 0.1. Aeell 71 486 94 199 0.082 0.563 0.984 lorth SUlUtra 663 2,017 S12 1,512 0.268 0.816 1.823 West SUIa..tra 164 1,016 158 747 0.101 0.627 1.287 R.iau 50 249 53 244 0.082 0.408 0.975 J_bi 36 183 20 114 0.085 0.433 0.835 South Suaatra 182 792 87 909 0.141 0.613 1.524 Benakulu 36 219 24 122 0.134 0.817 1.496 Lampuna 78 474 52 367 0.087 0.529 1.084 D.lt. X. Jakarta 1,412 2,953 469 3,784 0.451 0.942 2.750 West Java 846 2,513 282 2,725 0.191 0.568 1.438 Central Java 1,210 3,266 367 4,340 0.169 0.456 1.282 0.1. YOlyuarta 363 647 123 681 0.347 0.618 1.733 East Jav"a 1,157 3,590 488 4,222 0.157 0.486 1.280 Ball 205 1,207 136 611 0.135 0.794 1.419 West lu.a Tena.ara 39 297 31 211 0.068 0.521 1.014 East lu.a Tena.ara 33 420 16 295 0.042 0.534 0.972 East TUllor 24 146 16 311 0.087 0.531 1.807 West ltaUmantan 49 303 20 147 0.048 0.294 0.504 Central 'Kalimantan 26 368 40 103 0.059 0.836 1.220 South ltnlimantan 55 494 67 204 0.079 0.708 1.175 East ltaJ.imantan 74 520 46 162 0.091 0.643 0.991 lorth Sulawesi 137 711 24 312 0.143 0.740 1.232 Central SulaweSi 38 433 30 132 0.059 0.677 0.989 South Sulaweal 215 1,113 157 328 0.114 0.589 0.959 Southea"t Sulawesi 14 254 24 85 0.040 0.726 1.077 Maluku 31 342 13 43 0.048 0.534 0.670 Irian J,I.,.a 28 538 33 120 0.030 0.583 0.779 Iotal l....U! 25.551 3.182 23,030 .!L1!!! ~ ~ lIemo I t em: Coefficient of variation 0.17 0.24 0.35 - 190 - STATISTICAL ANNEX Table 3,11 ESTIMATED IllIlT COSTS FOR HOSPITAL SEiVIClS (Iluplah) So.pltal Per O\ltpatl_t Per lnpatient catelory and code v1alt day tpI lan B C 04 14,748.91 35,227.16 37 4,974.14 13,711.38 MOB Clan C 3,238.76 13.052.22 95 1,714.80 20.201.93 96 2,898.83 17,276.09 09 4.523.33 23,710.46 10 4,560.56 12,990.31 11 3,302.66 10,893.44 14 5,295.76 7,233.45 15 2,422.43 6,233.45 22 1,883.32 8,824.11 33 2,694.70 7,176.55 38 ".691.22 15,823.73 MOB Clan p 3.947.58 12.554.90 07 685.54 4,929.34 08 1,645.18 10,469.49 12 7,449.27 7,460.37 13 3,128.90 9,698.98 16 2,146.83 13,661.65 17 4,720.83 11,898.01 19 643.28 5,176.68 20 1,766.31 7,839.21 23 3,533.38 14,860.16 24 3,331.92 4,142.69 25 2,111.42 28,367.65 26 5,291.24 15,647.23 28 4,341.48 11,964.65 29 15,796.18 18,953.96 30 8,703.27 39,940.19 31 4,716.59 8,581.22 34 2,978.15 10,992.01 35 2,752.51 19.232.91 39 1,584.07 6,333.41 40 1,724.43 10,020.47 Armtd Fon" 01 7,780.41 22,195.14 18 6,957.90 14,165.95 32 5,708.19 21,130.20 State-Owped CompfQle. 02 44,200.35 232,010.56 36 12,998.76 14,971. 45 Prlyate 93 8,776.56 73,942.27 21 4,058.99 8,291.67 27 9,332.14 15,122.33 Source: Mlnl.try of Sealth and Unlver.ity of Iadone.ia (1988) Eya1vatlop and ABalr.i. of Ho.pltal Colt.; 'ha.e II, Jakarta. - 191 - STATISTICAL A1!HiX T.ble 4.1 PROVIIICL\L DISTl.IIU'fIOR OF COMMUJIITY IllALTB FACILITIIS, 1986 b·YII!!!· PuIYII!!!. lsbantu Iuprtl. lIcm-lDpre. rot.l Gov lIon rot.l a~nt Gova~t D.I. Aceh 99 47 146 387 22 409 Rorth S~tE"' 186 101 287 1,009 100 1,109 Wa.t S~tra 90 .52 142 396 4.5 441 Rlau 70 24 94 326 22 348 Jtlllbl .5.5 20 7.5 2.57 8 26.5 South S~tr. 12.5 .50 17.5 4.59 36 49.5 LMpuna 84 44 128 366 17 383 Banakulu 64 12 76 239 16 25.5 D.lt.I. JtUrta 60 232 292 0 0 0 We.t Jav. 430 213 643 948 302 1,250 Central Jav. 385 257 642 833 87 920 D.I. Yosy";arta 61 39 100 228 8 236 la.t J.v. 477 343 820 1,004 126 1,130 Wa.t J.v. 109 34 143 374 31 405 We.t ltallDMantan 109 34 143 374 31 405 Cantral ltal.1mantan 78 14 92 319 29 348 South ltall..llllntan 101 26 127 391 17 408 B··t ltaU.lDlontan 76 34 110 253 43 296 lIorth Sul.,.. a1 68 51 119 .518 62 580 Cantr.l Sul._.l 46 26 72 348 22 370 South Su1.,...11 121 94 221 186 60 846 s;ul.__ 1 Southe··t 47 22 69 252 11 263 lall 63 20 83 330 25 355 ....e lIua. ranasara 73 17 90 280 6 286 B··t Ru·· renas.r · 111 23 134 381 58 439 Ma1u1w 61 39 100 167 17 184 Irlan Jaya 104 22 126 287 142 429 B.at rl..lllor 44 24 68 108 10 118 Indone.l. .L.W. L.!!.9. 5,174 .u....!!§. ~ ~ - 192 - ITATISTI~ Table 4.2 ANNEX RATIOS OF PUSDSMAS TO POPULATIOif BY PROVIlfCE, 1986 1985 lteported ltetio Implied Population Puskes_s PUSDSMAS number (thou. .nd) '/86 Population 1:30,000 0.1. Aceh 2,981 146 1:20,418 99 lforth Sumatra 9,444 285 1,33,137 315 West Sumatra 3,666 14'1 1,26,000 122 lUau 2,514 92 1,27,326 84 Jambi 1,728 76 1:22,737 58 South Sumatra 5,411 167 1:32,401 180 o .IL I. Jakarta 7,829 280 1,27,961 261 West Java '0,733 615 1:49,972 1,024 Central Java 266,934 606 1:44,446 898 0.1. Yosyakarta 2,967 101 1:29,376 99 East Java 31,039 817 1:37,991 1,035 Lampuns 5,987 126 1:47,515 200 lenakulu 936 79 1:11,848 '1 West Kalimantan 2,815 137 1:20,547 94 West Kalimantan 1,140 92 1:12,'91 38 South Kal~tan 2,289 124 1:18,460 76 East Kalimantan 1,538 109 1:14,110 51 lforth Sula_si 2,375 114 1:20,833 79 Central Sulavesl 1,539 72 1:21,375 51 South Sula_si 6,600 222 1:29,730 220 Southeast Su1a_sl 1,083 69 1:15,696 36 Bali 2,638 83 1:31,783 88 West Rusa Tenagara 3,047 88 1:34,625 102 East Jlusa Tensgara 3,029 133 1:22,774 101 Maluku 1,633 100 1:16,330 54 Irlan Jaya 1,357 125 1:10,856 45 East Timor 624 61 1:10,230 21 Total 163.876 5.060 5.462 - 193 - STAI~STlSAL Table 4.3 ANNEX PROVINCIAL DISTRIBUTION OF POSYANDU. 1986 Ifumber of Percent of Ifumber of POSYANDU PUSUSMAS POSYANDU Reported Report ina Pro-rated 0.1. Aceh 5104 76 716 Iforth Sumatra 3.679 86 4,278 West Sumatxa 1.861 87 2.139 Riau 1,096 97 1,130 Jambi 1,644 94 685 Iforth Sumatra 1,556 85 1,831 Lampuna 2,513 93 2,702 Benair;ulu 517 93 556 D.It. I. Jalr;.a.rta 2,008 81 2,479 West Java 16,545 90 18,383 Central Ja·"a 13,381 90 14,868 n.l. Yosyat:a East Nus.. 40 70.0 732 86.3 441 12.8 1,213 80.9 221 90.5 T.nggal:a Maluku 31 64.5 916 83.0 412 92.5 1,359 85.4 149 80.5 Irian JaJ,a 28 75.0 596 86.9 87 94.3 711 87.3 153 93.5 East Tlmo)r 19 57.9 290 65.2 43 32.6 352 60.8 94 70.2 Subtotal Seadquart.rs 572 4.2 2,710 2.5 401 2.0 3,683 2.7 1,296 7.6 GrlUld Tot.al La Data for January 1986. a DPB d.not.s dip.rbantukan, i .··· paid by provinc. or kabupat.n/kotamaclya. 1£ Jakarta staff .xclud.s headquarters staff. Sourc.: Ministry of H.alth, Pusdakes BAIB files. - 198 - STATISTICAL ANNEX Table 5.) RATIO or MINISTRY or HEALTB MEDICAL AID PARAMEDICAL STAPF TO POPULATION BY BUDGETARY CATEGORY, 1986 Ssaff per million populaSion Relative Staff Ratio Doctors Paramedic Doctors Paramedics General Academy Bip school Junior hip All General Ac.de., Bilh echool Junior hilh All D.I. Aceh 86 ·.5 10.7 398 . .5 127.8 .537.1 1.35 0.52 1.19 0.94 1.09 North S....tr. 80.7 11.2 .587.6 341.9 940.7 1.26 0.55 1. 75 2.50 1.91 West Sumatr. 89.7 23.7 611.6 230.0 865.2 1.40 1.15 1.82 1.68 1.76 Rl.u 94.7 21.1 347.3 133.3 501.6 1.48 1.02 1.04 0.97 1.02 Jamb 1 90.3 34.7 409.1 94.9 538.8 1.41 1.22 0.69 1.09 South S....tr. 72.8 41.8 362.2 59.3 463.3 1.14 2.03 1.08 0.43 0.94 Benakulu 149.6 73.7 6.58.1 94.0 82.5.9 2.34 3 · .58 1.96 0.69 1.68 Lampuna 42.8 11 · .5 12.5.3 44.8 181.6 0.67 0.56 0.37 0.33 0.37 Jakart. 14.5.0 70.1 692.7 74.6 837.4 2.26 3.41 2.07 0.55 1.70 West J.v. 41.0 12 . .5 190.3 94.0 296.7 0.64 0.61 0.57 0.69 0.60 67.4 31. 7 466 . .5 146.3 644.4 1.05 1.54 1.39 1.07 1.31 Centr.l J.v. 43.1 10.0 229.8 126.2 366.0 0.67 0.49 0.69 0.92 0.74 E.st Jav. 43.9 16.1 2.51.1 115.7 382.9 0.69 0.78 0.75 0.85 0.78 West ltalimantan 81.0 14.2 316 · .5 171.6 502.3 1.26 0.69 0.94 1.26 1.02 Centr.l Kalimantan 100.0 21.9 492.1 332 . .5 846.5 1.56 1.07 1.47 1.43 1. 72 South ltalimantan 92.6 20.1 416.8 15.5.1 637.0 1.45 0.98 1.38 1.13 1.29 E.st Kalimantan 160.6 19 . .5 711.3 110.5 841.4 2.51 0.95 2.12 0.81 1.71 North Sul.ved 110.7 2.5.3 .536.8 343.6 905.7 1.73 1.23 1.60 2 . .51 1.84 Central Sul.vesi 103.3 39.0 407.4 2.56.0 702.4 1.61 1.89 1.22 1.87 1.43 South Sul.wesi 66 · .5 4.5.6 493.6 177.9 707.1 1.04 2.22 1.47 1.30 1.46 Southe··t Sul.vesl 91.4 48.0 .597.4 321.3 966.8 1.43 2.33 1. 78 2.35 1.96 Ball 94.8 39.4 684.6 201.7 925.7 1.48 1.91 2.04 1.48 1.88 West Husa Tenalara 69.6 13.5 220.9 78.1 312.4 1.09 0.65 0.66 0.57 0.63 Ee··t Husa Tenal.r. 73.0 13.2 241. 7 145.6 400.5 1.14 0.64 0.72 1.07 0.81 Kaluku 91.2 19.0 560.9 252.3 832.2 1.43 0.92 1.67 1.85 1.69 Irlan .laTa 112.7 20.6 439.2 64.1 523.9 1. 76 1.00 1.31 0.47 1.06 Eallt TlIIIor 1.50.6 30.4 464.7 68.9 564.1 2.35 1.48 1.39 0.50 1.1.5 Aver·le 64.0 20.6 33.5.3 136.7 492.6 1.00 1.00 1.00 1.00 1.00 LI Jakart. data exclude. Mlnist£)' of Be.lth he.dquarters .t.ff. Source.: Manpower d.t. from Mlnlet£)' of Be.lth (lAIN . .sterfile) - 199 - STATISTICAL ANNEX Table 5.4 NEW MINISTRY OF HEALTH POSTS, 1979/80 TO 1987/88 79/80 80/81 81/82 82/83 83/84 84/85 85/86 86/87 87/88 Medics; 1,953 L.ill. l....lli. 921 ~ 1.554 1.285 .L.l.6..2 ~ Inpres 600 610 660 610 660 700 700 700 1,000 Routine 1,353 973 916 311 674 854 585 565 1,425 E§.rm!!Elgics 2...lll .2.....m 5,860 !!....ill 5.220 4.119 5,158 4.665 11,907 Inpres 3,950 3,925 3,200 3,200 3,275 2,790 2,301 3,250 6,000 Rout:ine 1,701 2,973 2,660 1,487 1,945 1,329 2,857 1,415 5,907 E~k§.D'i Ke!!!eh§.:t§.n Q Q 800 800 1.225 2,210 ~ 5,000 2,600 Inpres 0 0 800 800 1,225 2,210 3,449 5,000 2,600 Routine 0 0 0 0 0 0 400 0 0 NOmneclics 1.~56 ~ 2......2M 2....1il .2......2.Ql 2.004 835 ill Q Inpres 0 0 0 0 0 0 0 750 0 Rout:ine 1,856 3,564 5,906 5,367 2,907 2,004 835 3 0 Total Inpres 4,550 4,535 4,660 4,610 5,160 5,700 6,450 9,700 9,600 Total Routine 4,910 7,510 9,482 7,165 5,526 4,187 4,677 11,983 7,332 ~[aml IQtal ~ 12.045 14.142 11.715 10.~86 ~ 11,12Z 11,683 16,932 Source: Ministry of Health, Personnel Bureau (unpublished compilations, 15 August 1986). - 200 - STATISTICAL ANNEX Table 5,5 PARAMEDICAL La FORKASI CREATED BY PROVINCE 1984/85 AND 1985/86: Inprell 1984185 " 1985/86 Jloutine 1984185 " 1985186 Humber Percent IlSJl IS. Number Percent JlSR. J.:R. D. I. Aceh 55 1.20 0.66 36 2.60 1.43 North Sumatra 193 4.20 0.73 71 5,12 0.89 Wellt Sumatra 146 3.18 1.42 83 5.99 2.68 Jliau 73 1.59 1.04 24 1.73 1.13 Jambi 45 0.98 0.93 38 2.74 2.60 South Sumatra 224 4.87 1.48 55 3.97 1.20 Ben&kulu 57 1.24 2.17 26 1.88 3.28 LampUXl.& 35 0.76 0.21 24 1.73 0.47 DltI Jakarta 1,382 30.08 6.60 85 6.13 1.28 Wellt Java 488 10.62 0.57 130 9.38 0.50 Yosyakarta 387 8.42 4.65 25 1.80 1.00 Central Java 188 4.09 0.25 183 13.20 0.80 Eallt Java 235 5.11 0.27 143 10.32 0.54 Wellt Kalimantan 44 0.96 0.56 9 0.65 0.38 Central Kalimantan 191 4.16 5.98 24 1. 73 2.49 South Kalimantan 54 1.18 0.84 23 1.66 1.19 Eallt Kalimantan 47 1.02 1.09 17 1.23 1.31 North Sulawelli 110 2.39 1.65 51 3.68 2.54 Central Sulawesi 71 1.55 1.65 30 2.16 2.30 South Sulawelli 225 4.90 1.22 116 8.37 2.08 Southeast Sulawesi 33 0.72 1.09 21 1.52 2.29 Bali 91 1.98 1.23 37 2.67 1.66 West Nusa Ten&sara 38 0.83 0.44 24 1.73 0.93 East Husa Ten&sara 66 1. 44 0.78 63 4.55 2.46 Maluku 44 0.96 0.96 15 1.08 1.09 Irian Jaya 36 0.78 0.95 22 1.59 1.92 East Timor 37 0.81 2.11 11 0.79 2.08 I2!!!. 4,495 !m!.:...Q.g .L..QQ 1,386 !m!.:...Q.g .L..QQ La Data includes nurses and nonnurse paramedics but exclude pekarya kesehetan. IS. JlSJl denoted the relative staff ratio which measures the provincial llhere of formas! to the provincial shere of population. Source: Ministry of Health, Personnel Bureau (unpublished data). - 201 - STATISTICAL ANNEX Table 5.6 DOCTORS AND DENTISTS PORMASI, 1984/85 AND 1985/86 BY PROVINCE Doctors Dentists Inpres Routine Inpru Routine No. RSK a No. % RSR a No. X RSR a No. X RSR a D.l. Ac··h 35 2.9 1.60 8 0.8 0.46 4 2.0 1.10 6 1.3 0.73 North S'JIII&tra 58 4.8 0.84 66 6.8 1.19 11 5.5 0.95 21 4.7 0.81 West SUlaatra 41 3.4 1.53 24 2.5 1.11 8 4.0 1.79 12 2.7 1.19 Riau 33 2.8 1.79 23 2.4 1.55 5 2.5 1.63 Jambi 26 2.2 2.05 14 1.5 1.38 6 3.0 2.85 5 1.1 1.06 South SCUDatra 44 3.7 1.11 49 5.1 1.54 10 5.0 1.51 10 2.2 0.67 Benekulu 29 2.4 4.23 19 2.0 3.45 7 3.5 6.13 4 0.9 1.56 Lampuna 36 3.0 0.82 30 3.1 0.85 8 4.0 1.09 8 1.8 0.49 Jakarta 12 1.0 0.21 107 11.1 2.32 2 1.0 0.21 68 15.1 3.17 West Java 152 12.7 0.68 157 16.3 0.87 19 9.5 0.51 9 20.3 1.08 Yogyakarta 16 1.3 0.74 17 1.8 0.97 4 2.0 1.10 3 3.6 1.97 Central Java 167 13.9 0.85 80 8.3 0.50 19 9.5 0.58 East Java 83 6.9 0.37 115 11.9 0.63 20 10.0 0.53 64 14.3 0.75 West lClIUmantan 31 2.6 1.50 7 0.7 0.42 4 2.0 1.16 8 1.8 1. 04 Central Kalimantan 36 3.0 4.31 11 1.1 1.64 5 2.5 3.59 4 0.9 1.28 South I.alimantan 31 2.6 1.85 7 0.7 0.52 6 3.0 2.15 5 1.1 0.08 East ICa limantan 39 3.3 3.46 12 1.2 1.32 7 3.5 3.73 4 0.9 0.95 North ~,ulawes i 33 2.8 1.90 22 2.3 1.57 8 4.0 2.76 6 1.3 0.92 Central Sulawesi 24 2.0 2.13 10 1.0 1.10 3 1.5 1.60 3 0.7 0.71 South ~.ulawesi 54 4.5 1.12 36 3.7 0.93 7 3.5 0.87 13 2.9 0.72 Southe,.st Sulawesi 27 2.3 3.40 8 0.8 1.25 4 2.0 3.03 6 1.3 2.02 Ball 27 2.3 1.40 43 4.5 2.77 7 3.5 2.17 13 2.9 1.80 West Nu.sa Ten&gara 28 2.3 1.25 19 2.0 1.06 6 3.0 1.61 6 1.3 0.72 East Nusa 'Ienagara 34 2.8 1.53 13 1.3 0.73 4 2.0 1.08 6 1.3 0.72 Haluku 32 2.7 2.68 8 0.8 0.83 4 2.0 2.01 8 1.8 1.79 Ir ian .laya 42 3.5 4.23 4 0.4 0.50 7 3.5 4.23 5 1.1 1.34 East T:.mor 30 2.5 6.57 56 5.8 15.24 5 2.5 6.57 3 0.7 1.75 L! RSR denotes the relative staff ratio which measures the ratio of provincial share of formasi to pr.)vincial share of population Source: Ministry of Health, Personnel Bureau (unpublished compilation). - 202 STATISTICAL ANNEX Table 5,7 PROPORTION OF NEW PARAMEDICAL NEW POSTS EVER-FILLED, 1979/80-1985/86, BY PROVINCE In1![~1 R2UUn~ Formasi , Filled LA Formasi , Filled D,l, Aceh 655 85,9 43 93,02 North Sumatra 1,265 98,7 98 92.86 West Sumatra 633 93,2 133 96,99 Riau 512 82,4 60 98,33 Jambi 431 73.1 38 97.37 South Sumatra 927 57,7 188 96,81 Bengku1u 460 72,0 30 96.67 Lampung 660 87,4 46 100,00 DKI Jakarta 252 98,0 1,688 96.56 West Java 2,380 98.7 392 88,52 Yogyakarta 368 97.8 188 100,00 Central Java 2,553 85.7 324 99,07 East Java 2,686 97,0 176 98.30 West Kalimantan 755 55.4 43 100,00 Central Kalimantan 595 51.4 105 99,05 South Kalimantan 735 55.1 47 89.36 East Kalimantan 540 88.9 44 90.91 North Sulawesi 763 96.2 67 100,00 Central Sulawesi 546 66.3 59 100.00 South Sulawesi 978 91.5 60 98.33 Southeast Sulawesi 375 75,5 29 100,00 Bali 452 98.5 78 100,00 West Nusa Tenggara 525 67.2 32 100,00 East Nusa Tenggara 519 80,3 38 97.37 Ma1uku 475 87.4 44 97.73 Irian Jaya 679 52.0 32 96.88 East Timor 342 60.5 32 93.75 Total 22.011 ll....!t: !L..lli .2.2.J!Q LA Inpres positions filled as of September 1986. Excludes pekarya kesehatan. Source: Ministry of Health, Personnel Bureau. - 203 - STATISTICAL ANNEX Table 5.8 SUPPLY AJID DEMAND FOIl. IlIPRES lIURSIIIG POSITIOBS, 1985186 Demand. Supply .LA Excess Intr. Ret Ret (open (araduatu) demand. provici.l open unplaced posts I..!.) pl.c_nt posts ar.duates 1.3. D.I. Aeeh 185 198 -13 184 1 14 133 473 -340 133 52 o 288 West S1.lID&t:c. 121 190 -69 101 1 13 83 IU.u 169 51 118 63 30 16 -12 Jambi 147 42 105 51 13 83 -9 210 141 69 93 4 113 48 Benakulu 170 38 132 40 8 122 -2 Lampuna 161 93 68 124 32 -31 Jakart. 11 584 -573 11 4 o 569 West J.v. 223 389 -166 223 12 o 1510 58 163 -105 58 o 105 Centr.l J.',. 619 348 271 1049 3 167 -101 East J.v. 257 1,5510 -1,297 256 1 1,298 West ltall.milntan 162 61 101 61 4 97 o Centr.l ltaUmantan 183 25 158 35 148 -10 South ltal~Gantan 160 53 101 100 120 13 East ltall.m;antan 133 84 49 96 1 36 -12 Rorth Sul.'.... i 161 153 -98 o 98 Centr.l Sul.wesi 83 76 7 59 24 11 South Sul.'.... i 161 235 -74 161 12 o 62 Southe.st :;'ul.wesi 100 77 23 52 48 25 B.ll 42 134 -92 41 1 93 West Bus. '!'ena··r. 152 36 116 81 4 67 -105 E.st Rus. 'rena··r. 46 62 82 4 22 -36 Maluku 128 14 131 3 8 -3 Irian J.y. 81 149 79 151 2 E·· t Timor 106 23 35 34 11 I..!. Open p".ts (sb. formasi) · never-filled IlIPUS po.itions for SPr. tr.1n.ed nurse., plus midwive·· 1k Supply (lulu.an) · new Ir.duate. who heve .pplied for public poSitions, cl···ified by province of educ.tlon. l.9. Out .nd in pl.c_t refer to interprovinci.l pl.c_nt. of Iraduate·· 1.3. Ret unp.lced Iraduate· ·hould never, in theory, be nea.tive, the nea.tive entrie. indlc.te thet .ome Ir.dua~e. heve been cl···ified under the wrona provlnee, e.a., B.li-educ.ted students frOID RTB heve been ml.takenly cl··· lfled under thelr province of birth. Thu. e.timate. for indlvldual provlnce. .hould be taken wlth c.utlon. Column .um., however, .re not .ffected by ml.cl··· lflc.tlon. Source: "lnl.try of He.lth, Bure.u of Per.onnel "Aloka.l lulu.an tenas. p.ramedl. untuk IlIPUS d.ri p:coplnsl leblh ke proplnsl kurans tho 1985. Per.w.t le.ehetan W (31 Maret 1986). - 204 STATISTICAL ANNEX Table 5.9 PROVINCIAL DISTRIBUTION OF PARAMEDICAL STAFF, 1986 L! Acad!!!!!% Hll!ih School Junlor Hlab 611 Number X DPB ./J! Number X DP· ./J! Number X DPB ./J! Total X DPB ./J! Provlnce 3.374 ll.:.l 54.946 E:.! n....m 82.8· 80.718 70.7 D.I. Aeeh 32 62.5 1,188 84.6 381 97.6 1,601 87.3 North S. . .tra 106 59.4 5,549 84.8 3,229 95.4 8,884 88.3 West Sumatra 87 39.1 2,242 52.7 843 82.8 3,172 60.3 Riau .53 60.4 873 85.1 335 96.1 1,261 87.0 Jamb 1 60 50.0 707 84.0 164 85.4 931 82.1 South S. . .tra 226 25.2 1,960 45.1 321 72.0 2,507 46.7 Benakulu 69 62.3 616 89.4 88 98.9 773 88.1 Lampuna 69 52.2 750 83.5 268 93.7 1,087 84.0 Jakarta 549 15.7 5,423 24.7 584 24.0 6,556 23.9 West Java 384 38.3 5,848 60.3 2,888 75.8 9,120 64.3 Yoayakarta 94 33.0 1,384 38.1 434 50.7 1,912 40.7 Central Java 269 46.8 6,190 60.3 3,399 73.5 9,858 64.5 East Java 499 62.9 7,793 73.3 3,592 88.0 11,884 77.3 West Kal1mantan 40 70.0 891 84.1 483 79.1 1,414 82.0 Central Kalimantan 25 72.0 561 88.4 379 93.9 965 90.2 South Kalimantan 46 58.7 1,057 81.6 355 91.3 1,458 83.3 East Kalimantan 30 60.0 1,094 83.4 170 92.4 1,294 84.0 North Sulavesl 60 56.7 1,275 90.2 816 81.1 2,151 85.8 Central Sulavesl 60 55.0 627 85.0 394 96.4 1,081 87.5 South Sulavesl 301 73.4 3,258 90.6 1,174 9.5.7 4,733 90.8 Southeast Sulavesl 52 78.8 647 87.3 348 98.6 1,047 90.6 Ball 104 49.0 1,806 48.6 532 72.9 2,442 53.9 West Nusa Tenaaara 41 65.9 673 85.9 238 98.3 952 88.1 East Nusa Tenaaara 40 70.0 732 86.3 441 72.8 1,213 80.9 Maluku 31 64.5 916 83.0 412 92.5 1,359 85.4 Irian Jaya 28 75.0 596 86.9 87 94.3 711 87.3 East Timor 19 57.9 290 65.2 43 32.6 352 60.8 Central J..s:. ill U 2.710 2.5 ill ~ 3.683 U SekJtm 156 3.8 1,004 3.6 113 3.5 1,273 3.6 Inspjen 19 5.3 55 29.1 3 33.3 77 23.4 linkesmas 48 10.4 125 1.6 26 3.8 199 4.0 P3M 112 3.6 199 0.0 42 0.0 353 1.1 Yankes 148 2.7 764 0.7 151 0.7 1,063 0.9 POH 23 0.0 349 1.1 38 2.6 410 1.2 Lltbanakes 30 3.3 139 2.2 14 0.0 183 2.2 Pusdlklat 36 8.3 75 2.7 14 0.0 125 4.0 !!!Ul 3.946 iLl 57,656 tL.1 22,799 !L.l 84.401 ll:.1. L! Data for for January 1986 from Mlnlstry of Health, PusdakKas (BAXH fl1es) . ./J! DPI - dlperbantukan, 1.e., pald by provlnce or kabupaten/kotamadya. J..s:. Staff listed under the dlrectorate-aenerals are all Jakarta-based, but are not included ln the Jakarta provlnelal totals. - 205 - STATISTICAL ANNEX Table 5.10 MEDICAL MANPOWER SUPPLY AND GOVERNMENT DEMAND, 1979-1986 Graduates Ministry of Health Ib Minus Year Graduates fA New Posts Appointments Appointments Doctors 79/80 1,124 1,482 927 197 80/81 1,079 1,092 839 240 81/82 1,173 1,132 1,075 98 82/83 1,648 700 690 958 83/84 1,245 910 866 379 84/85 1,023 910 482 541 85/86 1,336 1,003 1,059 277 Denti~:n 79/80 302 358 317 -15 80/81 288 324 280 8 81/82 363 347 344 19 82/83 426 148 141 285 83/84 413 260 242 171 84/85 390 300 132 258 85/86 447 162 170 277 fA G::'aduates of all schools, public and private. L.Q Nl?,w posts and appointments include both central and provincial position (pusat dan diperbantukan). i.£:. G.omera1 doctors (dokter umum) only. Sourc,s~: Ministry of Health, Personnel Bureau, New Posts Section, - 206 - STATISTICAL ANNEX Table 5.11 INPRES FORMASI AND APPOINTMENTS FOR DOCTORS, 1986/87 ~ Availaile Posi;lon Unfilled Unplaced Backlog Nev Total Applications Appointments positions appllcants D. I. Aceh 2 28 30 20 20 10 North Sumatra 24 24 38 24 14 West Sumatra 20 20 29 20 9 RLau 1 15 16 16 16 Jambi 18 18 18 18 South Sumatra 24 24 35 24 11 Benakulu 7 7 10 7 3 Lampuna 33 33 35 33 2 DKI Jakarta 102 192 West Java 2 59 61 87 61 26 Central Java 48 48 72 48 24 D.!. Yogyakarta 5 10 15 15 15 East Java 1 56 57 118 57 61 BaLL 8 8 26 8 18 West Nusa Tenagara 21 21 21 21 East Nusa Tenagra 20 20 19 19 1 West KalLmantan 3 18 21 14 14 7 Central KalLmantan 43 22 65 23 23 42 South KalLmantan 20 23 43 16 16 27 East Kalimantan 1 16 17 17 17 North Sulavesl 13 13 18 13 5 Central SulawesL 15 15 15 15 South Sulavesl 33 33 33 33 Southeast Sulavesl 8 16 24 19 19 5 Maluku 8 14 22 17 17 5 Irian Jaya 1 16 17 14 14 3 East Tlmor 23 23 21 2 !2!!1 II ~ ill .!l.2! ill 102 365 ~ Data as of September 1986. Source: Mlnlstry of Health, Personnel Bureau. - 207 - SIATISIICAL ANNEX Iab1e 5,12 INPRES FORHASI AND APPOINTMENTS FOR DENTISTS, 1986/87 LA ~ailable Positions Unfilled Unplaced Backlog Hew Total Applications Appointments positions app Uat ions D,l, Aceh 2 2 6 2 4 Horth Sumatrlt 5 5 15 5 10 West Sumatra 5 5 15 5 10 Riau 2 2 10 2 8 Jambi 3 3 3 3 South Sumatr~. 2 2 10 2 8 Bengkulu 2 2 3 2 1 Lampung 5 5 9 5 4 DttI Jakarta 213 213 West Java 16 16 96 16 80 Central Java 16 16 36 16 20 D.l. Yogyaka:t'ta 6 6 12 6 6 East Java 17 17 71 17 54 BaU 2 2 5 2 3 West Husa Te:nagara 3 2 S S 5 East Husa Te'nagara S 5 S West ltalimantan 1 2 3 3 3 Central ltal~nantan S S S South ltallmantan 6 6 6 6 East ltal imantan 1 2 3 3 3 North Sulawe.. i 7 7 6 6 1 Central Sulawesi 2 2 2 2 South Sul_e..i 1 S 6 6 6 Southeast Sulawesi 7 7 6 6 1 Maluku 2 2 2 2 Irian Jaya S S 10 6 6 4 East Timor 2 2 2 2 !2!!l! 46 !.22 ill ill ill !! ill U. Data as of Septe~ber 1986 Source: Ministry of aealth, Personnel Bureau, I - 208 - STATISTICAL ANNEX Table 5.13 INPRES FORKASI AND APPOINTMENTS FOR DOCTORS AND DENTISTS , 1984/85 AND 1985/86 COHBINED Doctors Dentists Total Unfilled Percent Total Unfilled Percent Formasi Formasi Unfilled Formasi Formasi Unfilled D. I. Aceh 35 0 0.0 4 0 0.0 North Sumatra 58 0 0.0 11 0 0.0 West Sumatra 41 0 0.0 8 0 0.0 Riau 33 1 3.0 5 0 0.0 Jambi 26 0 0.0 6 3 50.0 South Sumatra 44 0 0.0 10 0 0.0 Bengkulu 29 0 0.0 1 0 0.0 Lampung 36 0 0.0 8 0 0.0 Jakarta 12 0 0.0 2 0 0.0 West Java 152 2 1.3 19 0 0.0 Yogyakarta 16 0 0.0 4 0 0.0 Central Java 161 0 0.0 19 0 0.0 East Java 83 0 0.0 20 0 0.0 West Kall.ma.ntan 31 3 9.1 4 0 0.0 Central Kall.ma.ntan 36 32 88.9 5 4 80.0 South Kall.ma.ntan 31 20 64.5 6 6 100.0 East Kall.ma.ntan 39 1 2.6 1 1 14.3 North Sulawesi 33 0 0.0 8 4 50.0 Central Sulawesi 24 0 0.0 3 2 66.1 South Sulawesi 54 0 0.0 1 1 14.3 Southeast Sulawesi 21 1 25.9 4 4 100.0 Bali 21 0 0.0 1 0 0.0 West Nusa Tengsara 28 0 0.0 6 1 16.1 East Nusa Tengsara 34 0 0.0 4 4 100.0 Kalultu 32 8 25.0 4 0 0.0 Irian Jaya 42 1 2.4 1 5 11.4 East Timor 30 0 0.0 5 0 0.0 Total 1.200 12 L.1 200 35 ~ Source: Personnel Bureau, Ministry of Health (unpublished compl1shed). - 209 STATISTICAL ANNEX Table 5.14 ROUGH ESTIMATES OF MEDICS OUl?L0W AND INFLOW BY PROVI~,;E, 1985/86 Inflow Outflow Net Appointments In Gross Out Advanced Gross Inflow Routine Ie Inpres Transfers 1£ Inflow transfers1£ Training Inflow 14 Aceh 8 16 o 24 20 19 39 -15 North Sumatra 11 25 8 44 2 13 15 29 West Sumatra 4 18 7 29 o 4 4 25 Riau 6 19 1 26 2 16 18 8 Jambi 3 15 o 18 1 4 5 13 South Sumatra 9 19 6 34 o 8 8 26 Bengkulu 4 20 1 25 1 7 8 17 Lampung 5 23 3 31 2 9 11 20 Jakarta 245 4 88 337 9 5 14 323 West Java 29 105 17 151 36 15 51 100 Yogyakarta 4 12 10 26 6 2 8 18 Central Java 15 118 9 142 19 18 37 105 East Java 23 45 12 80 11 9 20 60 West Kalimantan 2 15 o 17 5 8 13 4 Central Kalimant:an 6 4 o 10 5 4 9 1 South Kalimantan 2 5 3 10 3 11 14 -4 East Kalimantan 3 19 2 24 7 9 I 16 8 North Sulawesi 6 13 o 19 9 6 15 4 Central Sulawesi o 15 2 17 5 4 9 8 South Sulawesi o 24 7 31 3 20 23 8 Southeast Sulawesi 4 10 1 15 3 5 8 7 Bali 8 17 5 30 3 1 4 26 West Nusa Tenggara 4 13 4 21 5 8 13 8 East Nusa Tenggara 3 16 1 20 11 11 22 -2 Maluku 2 13 1 16 7 3 10 6 Irian Jaya 2 18 2 22 6 6 12 10 East Timor 9 18 o 27 11 15 26 1 417 814 Medics refe~s to general doctors plus dentists. This table should be read as a rough attempt at establishing orders of magnitude of interprovincial personnel flows, not as an authoritative statement. Estimated routine appointments in 1985/86 is the number of routine appointments for the two year period 1984/85, 1985/86, multiplied by the 85/86 share of routine formasi of those two years. Outtransfers and intransfers include all interprovincial transfers of Golongan (rank) III and IV employees; a majority of these are doctors. Advanced training refers to the number of doctors accepting for specialist study from each province during calendar year 1985; all other columns refer to FY85/86. Data on inflow from advanced training are not available. Source: MOB, Pe~sonnel Bureau. Distributors ofWorld Bank Publications AllGENTINA flNLAND MEXICO For rullst:ripl;o,. o",us: Carl.. HinIcII, SRL Akateeminen Kirja'k.auppa [NFOTEC In.lemauot1a1 Sub6lCt'i:ption Service Caieri.aGuet:nm P.O. Box 128 Apartado Poatal22-860 P.O. Box 41095 Florid_US, 4th AooroDlc. 453{465 SF-0010l I4060TIalpan. Mexico D.P. Craighall 1333 Buenos Aires He!si.nlti 10 Johannesburg 2024 MOllOCCO AUSTMLlA. PAP\.1A NEW GUINEA. FRANCE Soc:i&! d'EtudeaMarketinS Maroc..... SPAIN FIJI. SOLOMON ISLANDS, World "Sank Publications 12 rue Mozart.. Bd. d' Anla Mundi-r-... Ubrn&, SA VANUATU. AND WESTERN SAMOA 66, avenue d'lca Caaal>lanca c..tell037 DA. Boob '" Joum.1h; 7S1161'ari8 28001 Madrid 648 Whitehone Road NETHERLANDS Mitcham 3132 GERMANY, FEDllRALllEl'UBLIC OF lnOr~Publilc.atiesb.v. Ubrerfa In,,,,"",,,onal AEDOS Vidoria UNQ.Verlag P.O. Box 14 CONe1J. d. C""~ J9I Popp.lodcrl.. All.. 55 7240 8A Loch"", 08009 Barcelona AUSTRIA D-SJOO Bonn I Cerold and Co. NEW ZEALAND Slll LANKA AND THl! MALDIVES Graben 31 GREECE Hill. Liorary and Informati.on Service Lake Houae Bookallop A~1011 Wien KEME Private Bag P.O. Box 244 '24. lppodamou Street Ptatia Plasma NewMarket 100. Sir OUttampalam A. Car-diner BAHRAIN Alhe..11635 AuckLand Mawatha Bahrain Re&earch and Consultancy Colombo '2 AI5Od.atee Ud. GUATEMALA NIGERIA - P.O. Box 22100 Libreriafl Piedra Santa Univenity PrIl!:85 Limited SW£DEN M..,ama Town 317 50. Calle 7-55 Zonal ThreeCruWl\ll BuildingJerid\O Private Mail Bag 5095 Ftw slMgle 'a": Fri_ fackboksl""",,&" BANGLADESH Micro :ndUltries llevelopment Aooi.....'" Soddy (MIDAS) Cuatemala City HONG KONG, MACAO ... NORWAY R.&eringIIg- 12, Ba>< 1tiJS6 S-103l7 Stoddlolm Houoe S, Road 16 Aata 2!lOO lJd. N.uveeen. Infonnation Center for SltbsO'iptkm orlm: 'DI\anmondi R/ Area 6 Fl... 146 Prince Edward Book Dep_.... Wennergren-WiUianu: AB DIIa"" 1209 Road,W. P.O. Bo.6125 Eti-.d Ba>< Dll4 Kowloon N.{lW2 0010 6 5-104 2SStod PubUcacoes Tecnic.ulntemacionaw ISJN. Heted;.Marg EdilOrial Deoarrollo SA lJda. Ballard ElIIlate Apartado 3824 TANZANIA Rua Peixoto Comide. 209 Bombay - 400 008 Um. Oxford Univemty PrrM 01 4)9 Sao Paulo. SP P.o. Box S29\l 13{14 Asal Ali Road PHILIPPINES OarellSalaCt CANADA New Del1U - 110002 National BookStore IA DiJfuIIow: 701 Ri.t.alA...."" THAILAND C.P. &5, 1SO! B rue Amp~ 11 Otittaranjan Avenue P.O. Box 1934 Central Oeporment Store Boud",...illr, Qu Bangkok J.yod... Hoo'e! Buildl:1& u.......Uooal Book Center CHINA 5th Main Road Gandhlnagar Filth Floor, IIlipinaB Life Buildirlg TRINIDAD .. TOBACC, ANTIGUA Chin. fi.nandal& Ec:Dnomic Publiahing Bangalore - 560 009 Ayala Avenue, Mabt:i BARBUDA. .AllBADOS, Ho_ Metro Manila DOMINICA. GRENADA. GUYANA, iI, 0. Fo 51 DongJ'" 3-5-1129 K.achiguda CraM Road JAMAICA. MONTSEIUlAT, ST_ Beijing Hyd....bad - SOO 027 POLAND KITTS .. NEVIS, ST. LUCIA. ORPAN ST, VINCliNT .. GllENAD.NES COLOMBIA Ptarthma R ..... ?J:\d Floor Pala< KultwyiNauki SystematiB Studies Unit lnfoenL'k:e Ltda. Near Thakore Baug. Navrangpura 00..901 Wanuwa 19 Wat.. StNet Apartado Mreo 34210 Atunedabad -380009 Curope BagdaD.E. POllTUGAL Tril::Udad. West India Patiala HOUSe Uvraria Portugal COTE O'IVOIRE to-A Aahok Marg aWl Do Carmo 7lJ..74 TURKEy Cenln! d'Edi_ eI de Diffusion Luclmow - 226001 1200 Liobon H.... Kitapevi, A.s. Africaine:e (CEDA) Istildal C'.addl'!i. No. 469 GUP.541 INDONESIA SAUDI AMBIA. QATAR Bul JI. Sam llatul_ngi 37 P.O. Bo.31% CYPllUS P.O. Box 181 Riyadh 11471 UGANDA MEMRB Jnfonnation Servicr:!:IJ JakarlaP......t Ugondailoabhop P.D. Box: 2098 MEMRB Intonnatio. Setvl"" P.O. Box 7145 Nicosia ITALY liCQNI: CommilJe.ionaria Sat1llOlli SPA '_do offias' A1AlM_ Kampala DENMARK ViAl Benedetto Fotiirti. 12WIO AI Oohrul Cent.r UNITtiIJ AllAB EMIllATES Somfundfl..itteratur c-n. 1'oota1.552 Fint F100r MEMRB Gull Co. R,*"Ofll'lUl AU' 11 3)125 Florence P.O. Box 7188 P.O. Box 6097 DK-I970 ftedg Kha!ed SIr... Microinfo Ltd. Restauracion. e laabell.4 Cat6lic.a 3C9 T"yo P.O. Box 3969 P.O. Ba><3 Apartado POIIIlal2190 Damman Alton. llampthin: CU34 21'G s.nto DaJ:ningo KENYA J!n&land Africa Book S.... u.mpur P.O. Bo.I141 CapeTownm> IBRD 22382 ,OS' 125~ 130' 135~ 1«)" Banda Aceh 5' ~ /,-' PHILIPPINES 5' ./ ./ MALAYSIA BRUNEI (i\ ...... 11. r.r-.~ ........ ® Madan \ ../ '" f (' INDONESIA /' / (-' / MALAYSIA ! ® Province Heodtworters i --,--:::·.SINGAPORE \."r-o,,--,.~j Manado® Provlnce BOlJndor!l;l~ Pekanb.ru ..............--. .-. HALMAHERA Internotionol BolJndoties 0' ® 0" KALIMANTAN Ponlianak@ ., Samari~a ~81u @Jamb! BANGKA KE;PULAUAN SULA ® 081 Palangkarava Jayapura® K.1 JAKARTA SUMATERA SULAWESI Pa\embang @ IRIAN JAYA BEUTUNG gengkulu@ C£:.RAli Baniarmnin ® ®Kendari BtfRU " SUMATERASELATAN RIAU TarJlJl'Igli:a'a"9 Ujongpandanu® JAMBl ® JAKARTA (8 BARAT KALIMANT AN TENGAH aanoun!) MAOIJRA KALIMANTAN SF:LATAN ® Semaranjl KALIMANTAN TIMUR J A WA (!)Suraba.ya SULAWESI TENGAH ® SUMBAWA Yoglfakarta BALI UTARA FLORES MaUlrAm ® .. o.,npa'lar I.OMBOK 100 '00 ,o0 400 500 TIMOR 'D' MILES Kupan g SUMBA e 2.00 300 400 500 600 700 800 KILOMETERS 10:S~ 110~ 120Q 1250 !30e 135'