7

Costs, Financing, and Setting Priorities

Average spending on health care in low-income countries is estimated at only (U.S.) $14 per person, of which less than one-half is from public funds; the corresponding figure in middle-income countries is $62 per person, of which just over one-half is from public funds (World Bank, 1993). Clearly, financial resources are tightly constrained. Our recommendations for improving reproductive health have to be considered in the light of limited financial, as well as managerial and administrative, resources. Funds for new or expanded programs would have to come from a combination of reallocations within the health sector, new resources from public or private sources, and greater efficiency in the use of resources in all types of programs.

In this chapter we set out the parameters within which these decisions will have to be made and discuss the principles that should guide those decisions. In Appendix C we present results from an illustrative model to show how conclusions about the costs of interventions depend on the setting, scale, and design of particular programs. Because of this variability and the lack of research on the cost and effectiveness of program types, we do not attempt to produce a global cost estimate for reproductive health programs. Enough is now known, however, to develop guidelines for setting priorities.

Experimentation and research should proceed in an interactive manner, so that lessons learned from research can be absorbed in new initiatives, which in turn spawn new areas for research. This chapter focuses on interventions to improve reproductive health within the health sector,



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions 7 Costs, Financing, and Setting Priorities Average spending on health care in low-income countries is estimated at only (U.S.) $14 per person, of which less than one-half is from public funds; the corresponding figure in middle-income countries is $62 per person, of which just over one-half is from public funds (World Bank, 1993). Clearly, financial resources are tightly constrained. Our recommendations for improving reproductive health have to be considered in the light of limited financial, as well as managerial and administrative, resources. Funds for new or expanded programs would have to come from a combination of reallocations within the health sector, new resources from public or private sources, and greater efficiency in the use of resources in all types of programs. In this chapter we set out the parameters within which these decisions will have to be made and discuss the principles that should guide those decisions. In Appendix C we present results from an illustrative model to show how conclusions about the costs of interventions depend on the setting, scale, and design of particular programs. Because of this variability and the lack of research on the cost and effectiveness of program types, we do not attempt to produce a global cost estimate for reproductive health programs. Enough is now known, however, to develop guidelines for setting priorities. Experimentation and research should proceed in an interactive manner, so that lessons learned from research can be absorbed in new initiatives, which in turn spawn new areas for research. This chapter focuses on interventions to improve reproductive health within the health sector,

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions while recognizing that interventions outside the health sector, such as investments in the quality and coverage of education for girls, also generate important health benefits. EXPENDITURES ON REPRODUCTIVE HEALTH Data on health expenditures, particularly on private out-of-pocket expenditures and on local and provincial public expenditures on health (in contrast to central-level spending) are quite weak in many developing countries. We draw here on World Bank (1993) estimates of the level of spending, activity, and source of finance based on information from government budgets and national accounts, household surveys, and estimates made by international organizations, supplemented by estimates predicted by an imputation model for countries for which direct data are unavailable. In the world as a whole, an estimated $1.7 trillion was spent on health care from all sources in 1990: about 10 percent, or $170 billion, was spent in developing countries, although about 78 percent of the world's population live in those countries. Two regularities characterize health expenditures across countries. First, as a country's average income level increases, so does the percentage of income spent on health: developing countries spend on average 4.7 percent of their gross national product (GNP) on health; in established market economies, the figure is 9.2 percent. Second, as incomes rise, the share of spending that is public also tends to increase: in developing countries about 50 percent of all spending is public; in established market economies, this share is 60 percent (World Bank, 1993). Health spending by region varies from as low as $11 per person in China to more than $1,800 per person in established market economies; see Table 7-1. There is also large variation within regions. Tanzania and Ethiopia, for example, spend only about $4 per capita on health, about one-half of which is public ($2 per capita), while spending in South Africa is almost $160 per capita. Such variation among countries in levels of expenditure means that there is no single set of near-term recommendations for reproductive health programs that is realistic in all settings. No available estimates cover reproductive health as a whole, but some estimates exist for public-sector spending on maternal and child health and family planning; see Table 7-2. These expenditures accounted for about 6 percent of public-sector health expenditures in 1990. The expenditures on family planning varied among the countries shown in Table 7-2 more than ten-fold, from less than (U.S.) $0.10 per person per year in Ghana, Lesotho, and Mauritania to more than $1.00 in El Salvador. No comparable data exist on private spending or on the other types of reproductive

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions TABLE 7-1 Domestic Health Expenditures by Region, 1990 Region Health Expenditures as a Share of GNP (percent) Per Capita Spending on Health (Public and Private) (U.S.$) Per Capita Public Spending (U.S.$) Public Spending as a Share of the Total (percent) Sub-Saharan Africa 4.5 24 13 55 India 6.0 21 5 22 China 3.5 11 6 59 Other Asia and Islands 4.5 61 24 39 Latin America and the Caribbean 4.0 105 63 60 Middle East and North Africa 4.1 77 45 58 Formerly Socialist Economies of Europe 3.6 142 101 71 Established Market Economies 9.2 1,860 1,116 60   SOURCE: World Bank (1993:Table A.9). health programs. Private expenditures are likely to account for a considerably higher proportion of total spending in the middle-income countries than in the lower-income countries, so the differences in total expenditures among countries are likely to be much greater than the public-sector estimates shown in Table 7-2. These estimates are not fully comparable across countries, however, because expenditures are easier to distinguish and assign to family planning in countries where it is centrally funded and operated as a vertical program with its own distinct headings in public budgets. Total expenditures on family planning in developing countries were between $4 and $5 billion in 1990 (Bulatao, 1993), about two-thirds of total health sector expenditures as estimated by the World Bank (1993). Total external assistance to the health sectors in developing countries was about $4.8 billion in 1990, accounting for less than 3 percent of 1990 spending on health (Michaud and Murray, 1994). Although their share is small, external resources can have a disproportionate effect on new investments and policy.1 1   In some countries the share of external aid is considerably larger: for example, aid accounts for more than one-half of all health expenditures in Burkina Faso, Chad, Guinea-Bissau, Mozambique, and Tanzania.

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions TABLE 7-2 Per Capita Public Sector Expenditures on Family Planning in Selected Countries, circa 1990 Countries Family Planning Expenditures, per Capita (U.S.$) Low-Income Economies   Bangladesh 0.51 Central African Republic 0.34 Ghana 0.05 India 0.35 Kenya 0.14 Lesotho 0.02 Mauritania 0.07 Nepal 0.20 Sri Lanka 0.18     Middle-Income Economiesa   El Salvador 1.22 Malaysia 0.53 Panama 0.13 Turkey 0.49 a Middle-income economies are defined as those with GNP per capita GNP between (U.S.) $600 and $2,500. SOURCES: Data from Ross, Mauldin, and Miller (1993); World Bank (1992, 1993). External assistance is provided though three main channels: official development assistance, multilateral loans, and through grants to nongovernmental organizations. Over 30 percent of external assistance programs are either directly or closely related to reproductive health (Michaud and Murray, 1994). Several important conditions are comparatively underfinanced in aid programs relative to the disease burden they impose, including pregnancy-related conditions (Michaud and Murray, 1994). Family planning has traditionally received most of the external assistance for reproductive health. In 1990 the category of population and family planning accounted for about 20 percent of all external assistance to developing countries, and external assistance represented about one-quarter of the total expenditures on family planning in developing countries (Bulatao, 1993). Maternal and child health accounted for 8 percent, and prevention and control of sexually transmitted diseases (STDs) and HIV accounted for 4 percent of external assistance for health in 1990. There are major problems in the quality of the data on external assistance, and the problems are greater when one tries to estimate flows to

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions specific health sector activities, such as reproductive health programs. 2 For our purposes, an important drawback is that they underestimate expenditures on reproductive health programs: they do not allocate to reproductive health any expenditures on general health services and hospital projects, even though these expenditures partially support reproductive health. ROLE OF THE PUBLIC SECTOR Governments have several approaches to influence reproductive health (Musgrove, 1996): inform people and health providers about health risks; directly provide health care; finance health care; mandate that certain activities be carried out; and regulate how health activities are carried out. All of these approaches can be used, and there is no single optimal mix for all situations. Economic theory and historical experience show that there are some important health-related activities that governments must finance if they are to be provided at all or at socially optimal levels. One set of such activities involve "public goods," activities that the private sector will not undertake, or will undertake at suboptimal levels, because users cannot be charged for them. Many public health interventions—such as spraying for malaria control and health information campaigns—are usually considered public goods. The testing and regulation of contraceptives is an example of a public good in the area of reproductive health. The private sector will not carry out testing and regulation to a socially optimal extent because individuals cannot be effectively charged for the services provided. Another set of activities that the public sector must help finance involves goods with large positive "externalities," benefits to others than those directly receiving the services. Treatment of STDs is an example of an intervention with large positive externalities. The individual who is 2   The two major data bases on external assistance are the Creditor Reporting System, maintained by the Development Cooperation Directorate, and the Development Assistance Committee of the Organization for Economic Cooperation and Development (OECD) statistics. There are major discrepancies in the numbers reported to these two systems; see Michaud and Murray (1994) for a detailed description of the reporting problems.

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions treated successfully derives benefits from that treatment, but so does the community at large as the risks of transmission to others are reduced. Therefore, an individual's willingness to pay for these services does not reflect the benefits they generate for society, and public support will be necessary to achieve the optimal level of STD control. Another type of "market failure" relevant to the health sector occurs when individuals cannot judge the value of a service, at least not without incurring large costs; in these cases, private markets for the service may be impossible to establish and so government financing is needed. Examples might be emergency care of obstetric complications or postabortion care, which occur so rarely in the experience of one person or family that there is no basis to judge whether adequate care is being provided. Governments have a necessary role in accrediting providers or facilities and fostering quality assurance. Another important rationale for public intervention in health financing relates to poverty. Poor people have the least capacity to pay for health services, and they suffer the largest burden of disease. This is especially true for reproductive health problems, such as maternal mortality. A strong equity rationale therefore supports public subsidies for health services for the poor. This rationale does not lessen the need to provide such services in a cost-effective way. Health spending patterns around the world indicate that governments typically choose to finance a much larger share of health services than would be narrowly justified by the economic criteria discussed here. At the same time, however, public spending often underfinances particular highly cost-effective public goods and health services with large positive externalities, many of which are reproductive health interventions. Public spending is inequitable in many countries. Developing countries often attempt to provide free, comprehensive services. But offering free care to all typically leads to some form of rationing, in which better situated populations often have an advantage. Resources may be excessively concentrated in urban facilities serving the middle and upper classes. The poor, particularly the rural poor, especially when referral systems are weak, are left with low-quality public services that are comprehensive in name only. Rather than achieving little in a vain attempt to provide everything for everyone, developing countries could provide significant health benefits to a large number of people by concentrating public spending on cost-effective public goods and services with large externalities and by subsidizing cost-effective clinical services for the poor. Much of this focus would include public health and clinical services for reproductive health programs.

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions Public Financing, Private Provision Even if services are to be financed by government, they need not be provided by government. Rather, combinations of public and private modes of finance and service delivery are possible; see Table 7-3. Governments can finance services and provide them directly, or they can finance care through private providers. Private sources, similarly, can be used to finance public providers or private providers. Most health systems rely on mixed financing and delivery modes. Many countries, especially OECD countries, use a large share of public finance to pay for private provision of services. The choice of who provides services ideally involves both costs and quality. Is it cheaper for public or private providers to deliver services of equal quality? Which subpopulations are best served by which type of provider? The choice also hinges on government's abilities to monitor and regulate contracted services. And, in some developing country settings, the private sector may be so poorly developed that contracting out is not a viable option. This is often the case in remote, rural areas. In general, however, most countries are increasingly accepting the view that competition among providers, on the basis of cost and quality, is preferable to the creation of a public monopoly in service provision. Nongovernmental organizations, financed publicly or privately, play an important role in the delivery of reproductive health services in developing TABLE 7-3 Public-Private Modes of Financing and Delivering Reproductive Health Services   Financing Mode   Delivery Mode Public: from general government revenues or publicly mandated insurance Private: direct, out-of-pocket expenditures and voluntary insurance Public Public-sector providers directly supply services. Governments charge for services they provide, particularly for the wealthy, in order to target public spending on the poor. Private, for Profit and Not for Profit Governments subsidize or contract with nongovernmental organizations or private for-profit providers to provide services. Individuals pay directly for services, which are often delivered through organized networks of private providers.

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions countries, and their role could be strengthened in many countries. Among 22 countries where Demographic and Health Surveys (DHS) were conducted during 1990-1993, for example, the proportion of users of modern contraceptives who obtained clinical services from nongovernment sources ranged from virtually none in a few African countries to well over one-half in several Latin American and Middle Eastern countries (Curtis and Neitzel, 1996:Table 7.2). For methods of modern contraception, at least 10 percent of users relied on sources other than government and pharmacies in one-half of the countries. In addition to provision of family planning services and supplies, many other tasks in reproductive health are suitable for nongovernmental organizations. For example, providing services to rape victims, providing high-quality abortion services and counseling, and providing information about warning signs of labor complications and what to do about them could all be handled effectively by such organizations. Sometimes, nongovernmental organizations are more efficient and provide services with greater consumer satisfaction than public providers. They are often innovative, testing out new services and delivery modes that are later adopted by other providers. For example, the Bangladesh Women's Health Coalition (BWHC) successfully developed menstrual regulation techniques in Bangladesh. BWHC clinics have also shown that integrated women's reproductive health services and improved care can increase the effective use of services at low cost in comparison with government family planning clinics (Kay, Germain, and Bangser, 1991). Nongovernmental organizations may reach poor women in underserved locations, and they are more accepted there than are government providers. The Aga Khan Development Network, for example, supports safe motherhood activities and other services in remote mountainous areas in rural Pakistan where government services are scarce. Governments can subsidize nongovernmental organizations or traditional medical practitioners to deliver reproductive health services as appropriate to poor families. For example, about 30 percent of the municipal governments in Brazil provide funding to the Sociedade Civil Bem Estar Familar No Brasil (BEMFAM), a nongovernmental organization that provides services in public-sector health posts and training to public-sector health personnel, as well as operating its own clinics providing a broad array of reproductive health services in state capitals (Gomes, 1994). BEMFAM services are concentrated in the poorest region of the country, the Northeast, where 90 percent of municipal governments have such funding arrangements. Governments can also facilitate such service provision by legalizing nongovernmental organizations, simplifying registration procedures, and offering training, office space, tax relief (import duty exemption), and supplies. In Botswana, government donations of

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions free vaccines and contraceptives to nongovernmental health providers have become a common way to target public subsidies to specific health intervention programs. Subsidies can be provided on a per case, per capita, or a block grant basis. Public policies can also encourage for-profit providers to deliver high-priority reproductive health services. Incentives range from directly financing for-profit providers to subsidizing in-service training for reproductive health services, such as in STD risk assessment and treatment or the use of new contraceptives. Governments can subsidize inputs for other critical prevention approaches to be bundled with reproductive health services, such as vaccines. However, the development of a private sector for some services, such as family planning, may be delayed by subsidization of public services, especially if subsidies are not targeted to the poor and are continued long after the idea of family planning has become familiar. Role of Mandates, Regulation, and Information Provision Public policies to encourage a diversified system of health service delivery, including nongovernmental organizations and for-profit private providers, need to be accompanied by efforts to strengthen the government's capacity to regulate health providers and, through mandates or regulation, to ensure minimum service delivery standards. Governments can mandate the content of educational curriculum to ensure that all graduating physicians have a minimum set of skills in reproductive health or that schools incorporate certain material on reproductive health in their curriculum. Governments can mandate that all insurers include certain high-priority reproductive health services in any insurance package. Governments can accredit hospitals and physicians. They can regulate what medical equipment and drugs and supplies are imported and what types of contraceptives are approved for sale. Such regulatory mechanisms tend to be weak in most developing countries. Providing information—for example, about how to minimize the risks of contracting STDs—is another extremely important role for governments in reproductive health. USER FEES Communities, families, and individuals pay some of the cost of publicly provided services in many countries through cash payments, in-kind contributions, and community-based prepayment schemes, as well as illicit fees (for example, tips in order to see a physician). Cost-sharing arrangements have grown in popularity in recent years in many developing

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions countries. A 1995 survey of 37 African countries found that 34 governments impose fees of some kind for government-provided health services (cited in Shaw and Griffin, 1995). This section briefly reviews the main objectives of user fees, the experience from implementation, and implications for women's reproductive health services. User fees tend to be used in the public sector for three reasons: to mobilize resources, to target public spending to the poor, and to improve efficiency by giving consumers appropriate price signals. For governments in developing countries, revenue generation is usually the main motivation for introducing user fees. Experience indicates that user fees usually contribute a modest amount to government costs—an estimated 4 to 20 percent in sub-Saharan Africa (Shaw and Griffin, 1995)—although in some cases, this is offset by fairly high collection costs. If user fees are largely retained and reinvested at the local facility level, they can improve the quality of services (Foreit and Levine, 1993). Some studies have shown that user fees can actually increase the use of health services by the poor if they are successfully used to improve the quality of services (Litvack and Bodart, 1993). A by-product of user fees is that they may make publicly run facilities more accountable to clients: clients tend to demand more responsive services when they are paying directly for them, especially if local communities are involved in the design and application of user fees. Some social services in developing countries attempt to target subsidies, either by charging fees on a sliding scale (so that the poorest clients pay less than others), by charging higher fees in locations used mainly by those who can afford to pay, or by charging fees for service at certain hours when waiting times are low (Grosh, 1994). Sliding-scale arrangements can be difficult to administer and are subject to abuse, but if the poorest clients can be readily identified, clinics can charge fees from those better able to afford them without discouraging use of services by the poor. Targeting subsidies by charging fees at some clinics and not others is feasible insofar as the poor live apart from the less poor or use different facilities. User fees can be used to improve efficiency in a variety of ways. User fees can be used to support a referral system by charging patients who go directly to tertiary facilities for care that should be provided at lower level facilities. User fees at tertiary facilities provide an incentive for patients to seek care at lower levels. User fees can be applied to influence the demand for services: services with high positive externalities (such as STD prevention) can be provided free or almost free of charge so that demand is not curtailed, while services with largely private benefits can carry higher user fees. Some have argued that user fees should not be imposed for prenatal and delivery care in order to help ensure that babies enter the maternal-child

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions health system and thereby have ready access to immunizations and other preventive services. Children who have been delivered at home tend to be more difficult for the health system to reach. After introduction of user fees in Zambia in 1989, pregnant women appeared to continue to use prenatal services, with modest user fees, but avoided hospital deliveries because of high fees (Booth et al., 1995; see also Prevention of Maternal Mortality Network, 1995). Women expecting complicated deliveries tried to delay hospital admittance until the last possible moment for fear of paying more, and this delay apparently led to higher numbers of infants who did not receive needed services. This in turn probably contributed to the decline in immunization coverage in Zambia from 80 to 50 percent in the early 1990s. Nongovernmental organizations in some developing countries rely on user fees for much of their revenue. The strategy sometimes has been to charge user fees for services that people are willing to pay for, such as abortions and pregnancy testing, and to use the revenue generated to cross-subsidize other services, such as the supply of contraception (Haaga and Tsui, 1995). This practice may not be appropriate if the services for which high fees are charged should be subsidized, at least for the poor. This might be the case, for example, with provision of safe abortions or prenatal and delivery care. Some of the reproductive health services for which nongovernmental organizations charge high fees—notably pregnancy testing and safe abortions—would be as appropriate or more appropriate to subsidize on public health grounds as the family planning services they are currently subsidizing. Experience in the 1980s and 1990s has shown that, while user fees for public services have served to mobilize resources, in many cases neither the poor nor certain types of critical services have been adequately protected from the demand-reducing effects of user fees. For reproductive health services, care must be taken to ensure that adequate exemptions are in place so that poor women have access to high-priority services, such as attended deliveries and emergency obstetric services. More data are needed on the practical concerns of how user fees are implemented, whether and how sliding scales for fees operate, and how efficiently fees are collected and whether they augment quality and the total resources available for reproductive health. SETTING PRIORITIES Recommendations to redirect public resources to reproductive health activities need to be considered in the context of total resources available for health and comparison with funding requirements for other cost-effective health interventions. A recommendation to increase public spending

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions on reproductive health to $10 per capita when the country's entire public-sector budget for health is only $8 per capita would likely be infeasible, for example, without some plan to mobilize additional resources for the health sector. Arguments to increase spending on reproductive health on the grounds of cost-effectiveness require an examination of the other highly cost-effective interventions that might receive less funding. Some rationing of health services occurs in all health systems in the world, either implicitly or explicitly. In highly market-oriented health systems, health services are implicitly rationed according to willingness to pay. This approach has both equity and efficiency problems. Equity problems arise because poor people have less access to health services, despite their disproportionately large share of disease burden, because of their lower incomes. Efficiency problems also occur if public goods are left to the market, because they will be undersupplied. Virtually all governments intervene actively in the health sector. Most governments finance a significant share of health spending (see Table 7-1, above) and in doing so must set priorities and ration the available funds. Developing country governments, with their limited health budgets, face difficult decisions about how to distribute their resources among the enormous needs of their citizens. One common approach is to avoid setting priorities explicitly, but to assume that comprehensive health care can be provided for all with the limited resources available. This strategy has resulted in practice in inefficient spending: tertiary care public hospitals—those that provide the most specialized and sophisticated services and where most clinical research, education, and training takes place—alone may consume 30 to 50 percent of the health budget. Even so, these hospitals often face chronic budget shortages. Health centers (secondary health facilities) typically are short staffed, particularly in rural areas, and face shortages of drugs and supplies. In planning investments in human resources and infrastructure, undue optimism about the future availability of funding has large costs: facilities may be built that are too expensive to run or are run at the cost of reallocating funds from higher priority programs. One of the most useful instruments for setting priorities for health expenditures is information on the effect of such spending, or value for money, which can be quantified in terms of cost-effectiveness. Cost-effectiveness measures the net gain in health (compared with a benchmark situation) in relation to the incremental cost of an intervention. If costs can be quantified and expressed in a common currency, and if health effects can be summarized in a single metric or index, different interventions can be compared in terms of a single ratio. As for other health fields, cost-effectiveness estimates for existing reproductive health interventions are imprecise. But even allowing for a

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions wide margin for error, these estimates suggest that several women's reproductive health interventions are among the most cost-effective in the health sector in developing countries (World Bank, 1993). In these estimates, the measured benefits include reductions in mortality, disease, and disability. Under the wider definition of reproductive health of the International Conference on Population and Development (ICPD), which incorporates the benefits of effective fertility control, the benefits of reproductive health programs would be higher. Continuing work is needed to estimate better the costs and effectiveness of health interventions at the country level—both to help set priorities for funding, and to monitor and improve efficiency. The World Bank has estimated the cost per case or per participant and cost per disability-adjusted life year (DALY), a measure of cost-effectiveness, for several reproductive health interventions.3 In this exercise costs were measured in terms of dollars and health outcomes were measured in terms of disability-adjusted life years saved; see Table 7-4. For all of these reproductive health interventions, saving a disability-adjusted life year could be achieved with expenditures less than about $110. The cost-effectiveness estimates are presented with a range to emphasize the fact that these are rough estimates. Estimates of cost per case or participant and cost-effectiveness are presented for two settings: low-income countries and middle-income countries. The estimates differ across these two settings because of differences in epidemiologic and demographic characteristics, levels of infrastructure available, and differences in the costs of specific inputs, such as labor. In an analysis of 47 child and adult health interventions in developing countries, cost-effectiveness ranged from as little as $1 per disability-life year saved to over $1,000 per disability-life year saved (Jamison et al., 1993). The reproductive health interventions shown in Table 7-4 are thus among the more cost-effective of those studied by Jamison and his colleagues. Cost-effectiveness analysis has been criticized on both technical and political grounds, but in our judgment, its advantages in making assumptions explicit outweigh the inevitable limitations of the methods. The criticisms of cost-effectiveness concern data limitations, the assumptions 3   The disability-adjusted life year is a measure that combines healthy life years lost because of premature mortality with those lost from disability. For disability, the number of life years lost was obtained by multiplying the expected duration of the condition (to remission or death) by a severity weight that measured the severity of the disability in comparison with loss of life. After combining death and disability losses, a discount rate is applied. Age weights are also applied so that years of life lost at different ages can be given different relative values. See Murray (1996).

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions TABLE 7-4 Costs and Cost-Effectiveness of Selected Reproductive Health Interventions in Low-Income and Middle-Income Countries (in U.S.$)   Annual Cost, Per Case or Participant Cost-Effectiveness (Cost per Disability Adjusted Life Years) Reproductive Health Intervention Low-Income Countries Middle-Income Countries Low-Income Countries Middle-Income Countries Public Health         EPI Plusa 15 29 12-17 25-30 Public health information (family planning, nutrition information) 2.4 5 N.A. N.A. AIDS prevention program 112 132 3-5 13-18           Clinical Interventions         Prenatal and delivery care 90 255 30-50 60-110 Family planning 12 20 20-30 100-150 Treatment of STDs 11 18 1-3 10-15 a EPI (Expanded Programme on Immigration) Plus includes micronutrient supplementation that is directed at women. SOURCE: Data from World Bank (1993:106, 117). used, average versus marginal measurements, political sensitiveness, and exclusion of consumer demand. Cost-effectiveness analysis requires detailed cost and health impact data. Some countries are investing in these data. In lieu of country-specific estimates, some analysts use data from countries with similar characteristics. Often, differences in cost-effectiveness between one intervention and another are much larger than the variation that can occur from one setting to another or from the range of error in the estimates. When this is the case, even rough estimates may be helpful in setting priorities across interventions. The disability-adjusted life year (DALY) as a measure of health outcomes incorporates a number of assumptions about valuing the future relative to today, weighing a year of healthy life lost at different ages, and rating the severity of disability (Murray, 1996). These assumptions are necessary to combine both healthy years lost of disability as well as premature

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions mortality into one metric. There are no ''correct" assumptions—they could vary by country setting, and they can be estimated through structured interviews with citizens. Sensitivity analysis can be used to illustrate how sensitive conclusions are to the assumptions used. Also, the measure does not incorporate nonhealth effects of interventions; to do so would require either a more inclusive measure of benefits or more complicated models. The DALY, like other single measures of outcomes such as lives saved, does not include the value of nonhealth outcomes. These can be significant, especially for interventions like family planning, which enhances control over reproduction and frees women's time for activities other than childrearing. These benefits would be valued in a full social cost-benefit analysis, but would not be easy to express in terms commensurate with the reduction of mortality and disability. Cost-effectiveness changes as interventions are extended throughout the population. Typically, one observes rising marginal costs and decreasing marginal effectiveness as interventions become more universal—but this is difficult to measure or to incorporate in simple models. Sensitivity analysis again can be used to indicate how cost-effectiveness changes as the intervention is extended throughout the population. Cost-effectiveness is only one element for priority setting in health. Clearly consumer demand must also enter into consideration. For potentially cost-effective services where demand is low, services must be accompanied by demand generation efforts. For less cost-effective services where demand is high, policy makers may wish to try to shift demand through education, user fees, or other measures, or supply the services for reasons of consumer satisfaction. COST ESTIMATION Both the potential value and the practical difficulties of estimating costs of reproductive health programs are illustrated by the results of cost models for the Mother-Baby Package of interventions defined by the World Health Organization and other international organizations (see Appendix C). The Mother-Baby Package is a program design for facility-based services, including basic health posts, better-equipped and -staffed clinics, and hospitals equipped for essential care of obstetric complications (see Chapter 5), and it includes clinic-based family planning services. It is not a complete cost model for the programs discussed above, however, because public information campaigns, community-based distribution of contraceptives, and efforts outside the health sector would all have to be accounted for separately. It does include the clinic-based services discussed in Chapters 3, 4, and 5, though (along with more child

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions health interventions than we have discussed), so qualitative results from the model should be relevant to planning reproductive health programs. The illustrative results show that estimates of the cost of a package of interventions are most sensitive to: the salaries of health care providers; the methods used to allocate shared costs of multipurpose workers, facilities, and equipment among different functions of health centers; and the assumptions about whether fixed facilities are operated at full capacity. Although we have not performed a similar exercise for programs outside the health sector intended to improve reproductive health, such estimates would likely depend to an even greater extent on how shared costs are allocated, since facilities and staff in education mass media and other sectors are primarily serving other purposes. Cost models like those used for the Mother-Baby Package are valuable mainly in specific applications, for planning and budgeting, and indeed have been adapted for use in several countries, for example, in World Bank sector analyses. Several qualitative inferences emerge from typical applications. One is that the cost-effectiveness of systems with large fixed-costs of facilities and specialized personnel can be increased when referral systems work, so that the clients with the most serious needs are referred and transported to specialized facilities. Screening and referral are particularly difficult tasks when primary care staff are poorly supervised and motivated, but as we note in Chapter 5, promising examples should be more widely replicated. The total cost of the package illustrated in Appendix C is greatly influenced by the number of pregnancies in the hypothetical population. This provides some grounds for optimism in countries that have recently seen sharp declines in fertility rates. The task of increasing coverage of attended births and improving the quality of care received should be less daunting. Urban-rural cost comparisons suggest that the increasing urbanization of the populations to be served will significantly reduce costs associated with vehicles and transportation now associated with rural health posts, but whether urbanization will result in any overall cost savings depends on whether salary costs and capital costs for facilities will be greater for expanded services in urban areas. Like our hypothetical ruralurban comparisons, cost estimates prepared using other models for India's Family Welfare Program were highly sensitive to the alternate assumptions about transport needs for rural health centers (Measham and Heaver, 1996:Ch. 6).

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions Training costs account for much of the total costs of new interventions and for much of the variation between high- and low-cost settings. For example, many of the interventions discussed in previous chapters call for health care and family planning providers to impart knowledge and counsel clients on many subjects about which the providers themselves may not currently know much. The services included in essential care for obstetric complications (Chapter 5) would require a major investment in training in most countries. Moreover, training should not be considered a one-time-only cost incurred in the introduction of reproductive health approaches. Evaluations of health care systems typically recommend proportionally more investment in refresher and in-service training. Training is essential if the reproductive health approach is to be effectively implemented. Planners should not assume that current programs with staff have no incremental costs. In allocating salary costs of multipurpose workers among functions, analysts typically assume that the workers are in their posts, working the statutory day at their assigned tasks. In countries where workers are poorly paid, poorly motivated, and poorly supervised, the actual output per worker is typically far below what is assumed (Janowitz and Bratt, 1994:46-48; Janowitz et al., 1996:Table III.B.1). The incremental costs of adding services to the job description of currently underemployed workers may appear deceptively low. The true costs of implementing new services would have to include organizational changes required to motivate the increased effort from workers. CONCLUSIONS Within the health sector, both the public and private sectors have important roles to play in delivering reproductive health services. Frequently, the focus of government actions is on service delivery. But governments can wield a variety of instruments to influence health—information provision, setting mandates, regulation—in addition to the financing and provision of services. Moreover, even when governments finance services, they do not necessarily have to provide them. In some cases, it may be more efficient to establish contracts with private providers, such as nongovernmental organizations or for-profit institutions, for service delivery. Users themselves currently bear a large share of the costs of the services they receive in developing countries. Private, out-of-pocket expenditures account for almost one-half of all health expenditures in developing countries. In the past several years, user fees for publicly financed services have become more common. What impact has this trend had on reproductive health? The evidence is mixed. While user fees, properly

OCR for page 178
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions implemented, can improve resource mobilization, equity, and efficiency, these benefits are often not achieved in practice because of indiscriminate application of user fees. The social benefits of user fees are best achieved if: (1) services with large positive externalities (such as prevention and treatment of sexually transmitted diseases) are fully exempted from fees so as not to deter demand, (2) fees are used to improve quality of care at the point of service, and (3) the poor are either exempted or face only modest fees for high-priority services. Many investments in women's reproductive health can yield substantial improvements in health in relation to costs. Current global cost-effectiveness estimates, such as those produced for the World Development Report (World Bank, 1993), alternatives prepared for the ICPD Programme of Action, or those we have explored for the Mother-Baby Package, would benefit from a better foundation of empirical studies. But even with wide bands of uncertainty surrounding point estimates, these analyses consistently support the qualitative conclusion that reproductive health programs compare favorably to alternative health-sector investments in both poor and middle-income countries. More external assistance to reproductive health programs, particularly to delivery and neonatal care, would remedy a past imbalance among programs considered for their potential contributions to reducing the burden of disease in developing countries. Low levels of current spending on the health sector as a whole in the low-income countries affect our recommendations for new expenditures on reproductive health. Our recommendations are designed to help in the planning for more efficient spending of resources that already are devoted to programs like family planning and mother-child health programs. But it is difficult to envision serious reforms and improvement coming with no increment in resources for the sector. Estimates prepared for India's reproductive and child health approach, for example, called for increases in public-sector recurrent spending on the order of 50-60 percent, for a relatively modest package of services (Measham and Heaver, 1996:Ch. 6). The arguments from welfare economics for public funding of health care are persuasive when applied to reproductive health, but even so, private sector funding will also have to be mobilized to achieve these goals in most countries. Lastly, the rationale for public funding of services is not to be confused with an argument for public provision of services, still less for provision by large centralized public-sector bureaucracies. The most effective organizational forms for delivering services will vary considerably, depending on the existing institutional infrastructure. A good deal of empirical evidence will be needed to test alternative ways to deliver on the promises of the reproductive health approach.