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 5
1 Introduction The historically high population growth rates experienced in Africa were the result of sustained high fertility levels as infant and child mortality began to fall. Recent data from the Demographic and Health Surveys conducted in sub-Saharan Africa indicate that the total fertility rates of a number of countries, most notably Botswana, Kenya, and Zimbabwe, have begun to decrease. Although fertility is declining in these and possibly other parts of sub-Saharan Africa (see Cohen, 1993), it remains high in most of the region. The declines in Botswana, Kenya, and Zimbabwe have been more apparent and rapid than those of other countries in the region. Similarly, analysis of child mortality trends (i.e., deaths among children under age 5) also suggests that declines are continuing to occur in a number of countries. The rate at which the declines are occurring varies across the region. One of the factors contributing to the decline in infant and child mortality has been the provision of health services. A number of initiatives over the past 25 years have fostered child survival through promoting relatively simple, affordable, and proven technologies. In 1978, the World Health Assembly set the goal of ''health for all by the year 2000." This goal includes ensuring a life both long and free of a heavy burden of illness. In Africa, between 1985 and 1990, an estimated 4.1 million (Heligman et al., 1993) died annually before their fifth birthday and are deprived of the most basic requirements for a healthy life. Therefore, the first aim of many health programs in Africa is to provide children with a reasonable chance of living a long and healthy life.
OCR for page 6
This emphasis on lowering mortality is also evident in the U.S. Agency for International Development's (USAID) Child Survival Initiative, which started in Africa in 1985. The program's goal is to reduce infant mortality in USAID-supported countries from the 1985 average of 97 death per 1,000 live births to less than 75. The program builds on primary health care programs, and focuses on immunizations, oral rehydration therapy, improving nutrition, reducing numbers of high-risk births, and improving maternal health. Since 1985, USAID has invested U.S. $1.5 billion in the Child Survival Initiative. In September 1990 at the World Summit for Children, representatives of 159 countries agreed on a Plan of Action for Implementing the World Declaration on the Survival, Protection and Development of Children in the 1990s. The plan includes seven major goals, the first of which is targeted to be completed between 1990 and the year 2000: a reduction of infant and under-5 mortality rates by one-third, or to 50 and 70 per 1,000 live births, respectively, whichever is less (United Nations Children's Fund, 1991). Among the more specific supporting goals were the elimination of neonatal tetanus by 1995 and the reduction of measles deaths by 95 percent by the year 2000, as well as the reduction of deaths due to diarrhea in children less than 5 years of age by 50 percent, and the reduction of deaths due to acute respiratory infections in children under 5 by one-third (United Nations Children's Fund, 1991). The plan included many other goals for child health: for example, increasing birthweights and reducing the prevalence of malnutrition, and of iodine and vitamin A deficiencies. However, reduction of child mortality is the major element of the goals for child health. Although infant and child mortality rates are declining in most of sub-Saharan Africa, only two countries—Botswana and Zimbabwe—currently are estimated to have infant and child mortality rates as low as the major goal calls for by the year 2000. Most other countries will require declines of more than one-third to achieve the goals of infant mortality rates of 50 per 1,000 and child mortality rates of 70 per 1,000. In the past, mortality reductions probably have been the result of a number of different factors, including health programs (and control of epidemics), changes in diets and health behaviors, and general economic development visible in education and road building, among others. These factors will all continue to play a role in reducing mortality in the future. However, the most direct way in which governments can intervene to reduce mortality in the short run is through increased provision of health services. The ability of African governments to provide health services is limited by the small amount of money available for the health sector. Half of the population of Africa lives in countries where expenditures on health were in the range of $1.50 to $6.80 per capita in 1985. The average African lives in a country in which the government expenditure on health was only $5.32
OCR for page 7
per person (United Nations Children's Fund, 1991; Ogbu and Gallagher, 1992). Between 1975 and 1985, the average African experienced a decline of 11 percent in the government's real per capita expenditures on health. The severe shortage of funds for health programs has led to a debate about the best strategy for developing health services. Although there are many dimensions to this debate, the central argument has been between those advocating "selective" primary health care and those advocating "comprehensive" primary health care. (See, for example, Walsh and Warren, 1979; Habicht and Berman, 1980; Newell, 1988, for more discussion.) The advocates of a selective approach argue that the shortage of resources requires that efforts be focused on the most cost-effective interventions. The advocates of a comprehensive approach argue that relying on cost-effectiveness leads to a short-term strategy that will make it difficult reach the long-term goal. In particular, they argue that the most cost-effective programs are often individual programs aimed at selected health problems, for example, vaccination-preventable diseases and malaria control. However, if too much emphasis is placed on vertically organized intervention-specific, the goal of integrated programs may never be achieved, and only a limited number of services will be available to the general public. Whether programs are integrated or not, the severe financial constraints on health programs in Africa require choices, such as which drugs to purchase and which skills to emphasize in training health personnel, among others. Without massive increases in the funds available for health programs, African governments will have to set priorities (implicit or explicit) for the use of scarce resources. In addition to cost-effectiveness, there are many other criteria for setting these priorities, including costs, efficacy, concerns for equity, and the preferences of the population. Given the goals set by the World Summit for Children and the high mortality rates in Africa, it seems initially reasonable to stress the expected effect of programs on mortality. Comparing programs according to their expected effect on mortality has several advantages. First, mortality rates provide a common measure that facilitates comparison of programs. For example, mortality rates allow us to compare the effect of a measles vaccination program with the effect of a program for treating diarrhea. Second, reductions in mortality often reflect reductions in the duration or severity of illnesses. For example, a program that treats or prevents malaria may lower mortality by reducing the incidence, prevalence, or severity of malaria. It is difficult to produce a simple measure of morbidity that incorporates both prevalence and severity. However, if mortality due to infectious diseases can be reduced through a program, it is likely that morbidity also declines. It is not feasible to measure the mortality effect of every health program
OCR for page 8
in every country. However, it is important to measure the mortality effect of various types of interventions. When we know that a specific type of program can reduce mortality, we can reasonably assume that any similar program that reaches a substantial proportion of the population with quality services probably reduces mortality. This volume examines the scientific evidence of the effectiveness of a variety of health interventions in reducing child mortality in sub-Saharan Africa. We have not evaluated the expected effect of potential new technologies yet to be tried in Africa, such as vaccines to protect against malaria or AIDS. We have also limited the review to interventions that are generally the province of health ministries. For example, we have not evaluated the impact of water and sanitation programs, agricultural projects, or adult literacy programs, although these might be important components of government strategies for reducing child mortality. For the majority of these types of programs, there is little evidence of a mortality effect. In addition, we have limited ourselves to programs whose effect on mortality can be evaluated. For this reason we have not evaluated the evidence on AIDS programs. The delay between human immunodeficiency virus (HIV) infection and death is so long that it is not feasible to measure directly the effect of programs on mortality. We have limited the scope of this report to the demographic and epidemiologic evidence that programs have reduced mortality. This focus provides only half of the evidence needed to carry out a ranking of programs in terms of cost-effectiveness. However, by limiting our review to programs that are already common in Africa, we have focused on programs that are generally assumed to be cost-effective. This review examines whether the evidence of effectiveness justifies the common perception that these programs are effective. It would be useful to determine whether studies of the costs of these programs (including both public and private costs) justify the usual perceptions. However, a complete review of cost studies would extend beyond the resources and the technical expertise of the working group. In evaluating the mortality effect of health interventions, we go beyond the evidence that these technologies are "safe and effective." Health programs do not include medical technologies unless there is evidence that their medical benefits outweigh the medical risks for individuals. We are interested in the benefits to populations, and these can be very different from the benefits to individuals. In the terms of evaluation studies, we need to go beyond estimating intervention "efficacy," the biological effect as measured in carefully controlled clinical trials. Instead we need to measure the ''effectiveness" of programs, which can be reduced by improper procedures, low compliance rates, or factors that select individuals who receive treatment. For example, programs to prevent measles, malaria, or tuberculosis
OCR for page 9
can change the epidemiology of the disease by changing the risk of infection for the entire population, including those not receiving treatment. Decision makers should be aware of the differences in results based on biological models and those from field studies. Theoretical results based on models rarely serve as a substitute for field studies, which incorporate the difficulties and shortcomings arising in carrying out a program. Often, the field-based evidence that health interventions reduce mortality rates is weak. The justification for many programs is based on the argument that a given disease causes death in a certain percentage of cases, and that the intervention is effective at a certain level in preventing (or curing) the disease and has few or no side effects. These arguments rely on biological models of the disease process and scientific tests of various parts of the model (e.g., the efficacy of a vaccine or treatment). This approach is most convincing when the evidence suggests very few serious side effects and when the disease has serious implications for health (e.g., the use of measles vaccination). The evidence to support these biological models may not be adequate to produce reliable population-based estimates of program effectiveness. For example, many of the deaths of children in Africa result from the interaction of several health problems. It is difficult, if not impossible, to design models that describe these interactions in sufficient detail to produce reliable estimates for the mortality-reducing effect of different types of programs. Estimating the effect of a health intervention becomes more difficult when the program does not reach all children at risk for the disease. For example, a program that reaches children only in the upper socioeconomic groups will probably have less effect on mortality than one that targets malnourished children in impoverished families. As a result of the unequal distribution of risks, there can be large differences in the effectiveness of programs that employ the same medical technologies. For this reason, we must examine the effectiveness of programs in a number of different settings, not just the potential efficacy of technologies. Much of the evidence about the effectiveness of interventions on mortality comes from research projects in small populations. Thus, the evidence from these projects may not provide reasonable estimates of the effects these interventions will have when implemented in large-scale government programs. When programs are expanded beyond small populations, the quality of the services can change drastically as can the social groups they reach. Vaccine failures, incorrect diagnoses, and reduced effectiveness of supervision or patient education are all more likely in large-scale programs. This volume begins with a discussion of mortality trends and levels over the past three decades in the region. Mortality has decreased in all parts of Africa, but at varying rates, as illustrated by regional differentials.
OCR for page 10
The cause-of-death profile varies across areas, but a small number of diseases accounts for most of the infant and child mortality throughout the region. However, because data are generally of poor quality or nonexistent, it is difficult to rank these specific diseases. Most of this volume examines various types of health programs. Chapters 3, 4, and 5 review evidence on the effectiveness of interventions aimed at individual diseases. Chapter 3 examines the diseases that can be prevented or reduced by immunization: measles, pertussis, tuberculosis, leprosy, and tetanus. Chapter 4 focuses on interventions targeted at other communicable diseases: diarrheal diseases, malaria, and acute respiratory infections. Nutritional conditions, including protein-energy malnutrition, low birthweight, and vitamin A deficiency are discussed in Chapter 5. The discussions of these specific diseases and conditions examine the etiology, symptoms, and prevalence of each disease; its epidemiology and natural course; intervention studies conducted in Africa and other parts of the world; and program coverage in Africa. The organization of the bulk of the report by disease reflects the fact that almost all of the studies of the effectiveness of health programs examine single interventions. Although we would have preferred to organize the report along other lines, the available research does not support other approaches. In particular, there is very little research on the effectiveness of the most common type of programs, namely those programs that combine interventions against several diseases. What evidence there is on the effectiveness of integrated and general health programs is reviewed in Chapter 6. The final chapter provides conclusions and recommendations.
Representative terms from entire chapter: