Michael J. Sharpston
Health has an obvious and fundamental influence on human welfare; and if the health of the bulk of the population in a developing country can be improved, this could well have a major impact on overall welfare distribution. This was the basis of the World Bank’s Health Sector policy paper released in 1975.
In recent years, two trends have tended to increase the Bank’s interest in health in developing countries. First, there has been a growing realization in the development community as a whole that gross national product (GNP) per capita alone is not an adequate measure of development and of its effect on human welfare. The result—both outside and inside the Bank—has been increased attention to income distribution and to the social sectors.
The second recent trend has been a growing awareness of the problems of pollution and of the ecological effects of man-made changes in the environment. In 1970 the World Bank Group established the Office of Environmental Affairs “… to review and evaluate every investment project from the standpoint of its potential effects on the environment.” Environmental concerns in Bank projects cover a wider field than human health, but health is often a major issue. For example, water resource projects, such as irrigation or hydroelectric schemes, can have an important effect upon the incidence of certain diseases, especially in a previously arid area, where new; breeding places for disease-carrying insects can be created. Thus both because of a desire to avoid health problems and because of an active interest in improving welfare, the Bank became increasingly involved in health considerations.
What affects health levels?
Before formulating a sensible health policy either for the World Bank or for the developing countries, perhaps the most fundamental question to be answered is: “What really determines the overall health level of a population?” One obvious and possible answer is health services and the number of physicians or hospital beds; yet there is clear evidence that at best this is only a very partial answer. For example, in the “west”—Northwestern Europe and the United States—life expectancy at birth rose from about 35–40 years in the eighteenth century, to 50–55 by 1900 (in Sweden it rose from below 35 in 1755–76 to over 55 by the first decade of the 1900s). Yet, very little indeed of this progress can be attributed to medical science. By 1900 major discoveries in medicine and surgery had already been made, but only smallpox vaccination would have been in sufficiently general use to have had a significant effect on mortality in the population as a whole. One must, therefore, look to other factors—better nutrition, a slow improvement in hygiene habits, and (from about 1870) the effects of a series of public health measures.
Similarly, in 1970, a country-wide health survey in Thailand found that, on the average, sickness occurred twice a year per person, but that only 17 per cent of the population utilized public sector health facilities during a year. Private clinics were important in Bangkok, but in the rural areas 61 per cent of the people applied self-treatment, sometimes with the aid of a pharmacy. Yet, despite this low coverage of the population by health services, health in Thailand and many other countries of the developing world has improved very substantially since World War II. Evidence of this is the population “explosion,” most of which can be attributed to decreased mortality rather than increased fertility. Average life expectancy at birth in the developing world has increased from about 32 years before World War II, to about 49 years at the end of the 1960s. Though this is still far short of the 70 years or more now typical of a developed country, it represents a very dramatic fall in mortality, over a rather short period. Some of the fall is attributable to major health campaigns (a notable example is malaria control in the South Asian subcontinent), but a large residual is left, which can only be explained by other factors. However, before one can understand fully how these other factors operate, it is first necessary to examine rather carefully what the real health problems of a typical developing country are and what causes them.
The ecology of poverty and disease
Man, the organisms which give him diseases, and the vectors (such as flies) which help transmit disease are all part of an ecological system. It is the interaction of man and his environment which determines the incidence of disease. Curative health care services very rarely affect the incidence of disease, though they may mitigate its effects. Rather, the disease pattern of a society intimately reflects its standard of living, and indeed its whole way of life.
One very important factor affecting the pattern of disease is the high level of fertility—typical of most developing countries and typical also of the poverty syndrome. This has an obvious effect on the age structure of the population. In any country, it is the old and the very young who are most susceptible to disease. Developing countries have proportionately very few old people, and many young children. As a result, diseases such as cancer and cardiovascular conditions are a far smaller part of the disease pattern than in developed countries. On the other hand, up to a half of all deaths may be those of children under five.
In addition, the high level of fertility common in the developing world, and the short interval between births, also have a direct impact on health, largely because less nutrition and care are available for each successive child. Morbidity (sickness) and mortality in a large family are higher, and later children in the family are at a particular disadvantage. Furthermore, from the third child onward, the risks to the general health of the mother increase. High fertility can cause health problems at a community level as well. Population pressure on the land may lead to overcropping and hence to poor general nutrition. Excessively rapid urbanization and overcrowding create major health risks. Population growth makes it harder to provide safe or sufficient water supply, garbage disposal, and sanitation for the community or other social infrastructure. More generally, population growth may slow the improvement of educational levels and per capita income; and, under a market system, tend to increase the inequality of income distribution. All these factors tend to have a strong detrimental effect on overall health conditions.
One of the most significant effects of the existence of large families living in poverty is malnutrition, which is of considerable importance as a direct cause of death. Still more importantly, nutritional deficiency is frequently an associated cause of death among young children, and is also a major factor in the widespread occurrence of immaturity (low birth weight and other similar characteristics, whatever the length of the pregnancy), often in its turn an associated cause of death in babies. In a major study of Latin America, a comparatively well-fed part of the developing world, “nutritional deficiency” or “immaturity” were the primary causes of death for only 6 per cent of deaths of children under five, but one or the other was an associated cause in 57 per cent of all deaths. Mere exposure to a disease agent need not produce clinical disease and very frequently does not do so. Malnutrition is of such significance essentially because it hampers the body’s resistance. Malnutrition acts “synergistically” with disease agents to increase the incidence of clinical disease and aggravate its severity. Equally, disease can bring on malnutrition by increasing food requirements at a time when effective food absorption is often diminished.
Diseases from human wastes
The most important single group of diseases in the developing world are those whose transmission cycle depends upon contamination by human wastes of food, drink, water, or soil. For most diseases generally, and for nearly all diseases in this group, man is the reservoir of infection; without contamination of the environment by the wastes of infected human beings, the disease cycles would come to an end. Among the diseases in this group are typhoid, dysentery, cholera, polio, and hepatitis. Diarrheal disease, also in this group, is probably the biggest single cause of death in children under five, and is a major cause of illness among adults. In Latin America, diarrheal disease was responsible for 26 per cent of deaths under five years of age in São Paulo, Brazil, and no less than 43 per cent in rural EI Salvador.
Many worm diseases also belong to this group, including tapeworms, hookworms, and bilharzia (also called schistosomiasis). These diseases only rarely cause death in adults; rather, they are diseases of life, a continuing burden on the sufferer. Case studies in Bangladesh, Sri Lanka, and Venezuela found infestation rates at age six to be 95 per cent, 97 per cent, and 93 per cent, respectively. A recent World Bank case study of the labor force engaged on civil construction at three sites in West Java, Indonesia, found 85 per cent infested.
Both water supply and sanitation bring about major changes in the man-disease ecological system. Essentially, all the various techniques for sewage disposal, or for the provision of a safe water supply, are intended to break the tight transmission cycle upon which a high prevalence of these diseases depends. With good water supply and sanitation, isolated cases may still arise—there may even be epidemics, or sporadic outbreaks of a disease problem—but disease is much less likely to be an endemic phenomenon, a part of the ecological “scenery.”
Airborne diseases are the second most important group. They are transmitted by the breathing-in of the disease agent, and include tuberculosis, pneumonia, diphtheria, bronchitis, whooping cough, meningitis, influenza, and measles. Crowded, ill-ventilated housing fosters the spread of these diseases, especially if windows are blocked against the cold. One study (Ruth R. Puffer, Carlos V. Serrano, and Ann Dillon, The Inter-American Investigation of Mortality in Childhood [Pan American Health Organization/UNICEF, 1971]) showed that in rural Viacha, Bolivia, on the high Altiplano, almost two thirds of all deaths of children under five years were caused by airborne diseases (and in Viacha over one child in ten dies even before the age of one).
Malnutrition, diseases caused by the contamination of the environment by human wastes, and airborne diseases form the core of the disease pattern of the developing world. This disease pattern is intimately related to large families with children born at short birth intervals, inadequate housing, water supply, sanitation, nutrition, and general hygiene. All these factors work together, and reinforce each other to emerge as the basic pattern of poverty and disease.
There are, of course, other important diseases, but they are generally more limited to particular areas or certain ways of life. The mundane diseases of the core disease pattern really take the biggest toll. In a major study of 22 locations in eight Latin American and Caribbean countries, diseases from contamination by human wastes, airborne diseases, and nutritional deficiency were responsible in all except two cases (both in Jamaica) for over 70 per cent of the deaths of children under five that were not due to congenital anomalies and birth-related causes.
Africa probably does have particularly difficult problems because its climate makes it ecologically very suitable for certain diseases. Nevertheless, in many ways the disease pattern of much of the present developing world is not so different from that which existed in many of today’s developed countries at about the turn of the century. It is, essentially, a disease pattern of poverty; and, in fact, the higher socioeconomic (largely urban) groups in a developing country do have markedly different health problems from the bulk of the population. They have much more housing than the rest of the population and often have smaller families. Their disease pattern will correspond more to that of a developed country than to that of a developing country, with influential members of this group requiring health care for the chronic and degenerative diseases typical of later life. Evidently, the health problems of the higher socioeconomic groups are minor compared to those of the poor, rural, or urban. However, these groups are of considerable significance in determining the effective health priorities of a developing country, both because of their economic power in terms of effective demand, and because of their sociopolitical influence.
Present health services
We can now examine the type of health and health-related services available in a developing country. As regards water supply and sanitation, in the poorest developing countries most of the rural population defecate in the fields, and indeed this continues to be the case until people reach quite high income levels. In few developing countries does more than a small portion of the rural population have access to functioning modern water supply systems. In the urban areas, water supply and sanitation services are somewhat better; but usually they are very inadequate, particularly in shanty-town areas, and such sanitary facilities as exist are often not well utilized.
Whereas in the United States expenditures on health services now average well over $400 per capita a year, in many developing countries expenditure is probably less than 1 per cent of this amount. Costs, however, are not nearly so different; a hospital can easily cost $8,000 per hospital bed, and $2,000 a year in recurrent costs (depreciation and interest on capital excluded). A doctor’s salary in some parts of Africa may easily be $10,000 a year—less than in the United States, to be sure, but still about one quarter of U.S. levels, whereas financial resources are only one hundredth. Any rational discussion of health policy must bear such factors in mind. High-level medical manpower is also very scarce in most African countries, although less so in Latin America and Asia. To take an extreme case, in 1970 Upper Volta had little more than one physician per 100,000 persons, as against one per 700–800 persons in the United States. Most developing countries also have much less physical infrastructure for health services than the developed world. For example, Indonesia recently had rather under 2,000 persons per hospital bed, as against about 120 per bed in the United States.
The allocation of resources
These statistics are traditional measures of health services, and although they do serve to give some idea of the limited availability of resources at a national level, they also show that there is gross distortion in the allocation of these resources. In nearly all developing countries, the bulk of government health expenditure is allocated to curative services. A large part of government funds is spent on hospitals, particularly on in-patient services. Hospitals, in turn, are concentrated in the urban areas, and most of the patients also come from the same urban area; organized referral is usually of insignificant importance. Some rural inhabitants go to an urban center, but even so, most of the medical care serves a few urban people.
Thus, in early 1969, Greater Accra contained about 8 per cent of the total population of Ghana, but over 45 per cent of all doctors in the country. Again, 55 per cent of Ministry of Health doctors worked at the three largest hospitals in the country, and the situation for nurses was similar. By contrast, the Upper Region of Ghana, with a population of nearly 1 million, had seven government doctors and five mission doctors, 2 per cent of all doctors for 10 per cent of the population; furthermore, all were non-Ghanaians, even the Regional Medical Officer. Within regions also, there was a heavy concentration of health services in the regional capitals. Thus, in the populous and prosperous Ashanti Region, 34 out of the 36 Ministry of Health doctors worked at Okomfo Anokye Hospital; one worked in the Regional Office; and only one was stationed outside the regional capital. Even after allowance for patient migration into the city for treatment, there were over six times as many Ministry of Health doctors available to care for the population of the regional capital as there were for the rest of the region.
The reasons why doctors (government or private), congregate in the principal towns and the government doctors stay in the large hospitals are not hard to find. The economic opportunities are in the large towns—and government doctors often do have private practices, whether or not it is officially permitted. Even if a doctor comes from the rural areas originally (and, in fact, most come from better-off households in the cities), by the time he has completed his training he usually has adopted the outlook of an urban professional man; his friends will live in the cities, and that is where the bright lights are. Furthermore, he is likely to associate high professional status with the sophisticated treatment of “interesting” cases; this will usually only be possible at a few major hospitals. Politically, private sector doctors will usually be able to determine where they practice, and so also will public sector doctors—doctors are a tight elite, with an arcane expertise that touches the frightening mysteries of life and death; few laymen will tackle them on their own terrain. In any case, the lay elite—politicians, civil servants, even trade union leaders—are also nearly all urban, and will want “proper” (that is, Western) levels of treatment for themselves and their families.
The result is that in most countries the doctors stay almost exclusively in the main towns, perhaps after a token year or so of exile in the hinterland, soon after becoming qualified. Additional output of doctors by the medical schools only leads to unemployment or underemployment of doctors in the capital cities, probably coupled with emigration; this, for example, is the case in the Philippines. The rural areas remain without doctors. On the other hand, the mere fact that doctors remain unemployed will not necessarily prevent the building of new medical schools; the output of doctors is frequently determined by the political power and social aspirations of middle-class parents.
‘there is a limit to what conventional health service can achieve in an unchanged physical and social environment”
This impressive array of factors tends to concentrate the resources of the health services in a few urban areas. Bearing this in mind, we can examine the three main reasons why health coverage in developing countries tends to be low. The first reason is geographical: it is known that the great bulk of patients at a health facility come from the immediate vicinity, say, within five miles. For example, a study in India showed that attendance at a dispensary halved for every half mile distance from it. Yet much of the population of most developing countries is dispersed and outside the immediate vicinity of a health facility. The second reason is administrative: it is very difficult to get referral systems to work. The transport from outlying areas to urban hospitals is not available, and few developing countries have an administrative control system strong and ruthless enough to prevent those in the immediate vicinity of a hospital from pre-empting the available beds. The third reason is cultural: particularly in Latin America, a peasant may feel that a doctor looks down on him, and prefer to go to a “curandero” or traditional healer. In Asia, people may have more faith in traditional healers, or go to western doctors only for certain conditions. Thus even in the cities many people do not make use of official health services. An example of the effects of “cultural distance” is Cali, Colombia, where one study found that despite a doctor-to-population ratio of 1 to 910, 17 per cent of children who die are not seen by a physician, and another 19 per cent have no medical attention during the 48 hours preceding death. The combined effect of geographical, administrative, and cultural distance is that for many developing countries the official health services may be a small and rather unimportant part of the total health care picture. Often other factors are of far more significance—such as self-treatment, use of traditional healers, advice from pharmacists or “injection men,” and among better-off groups in the cities, private practice.
The effect that the concentration of official health services in urban hospital facilities has upon the available health statistics is also significant. Even in a developed country, a health service “discovers” the illnesses for which it is looking; in New York City, reported cases of lead poisoning were one in 1950, but 1,925 in 1971, after lead poisoning attracted professional interest. In the same way, because official health services devote most of their resources to large urban hospitals with few pediatric beds, and in particular tend to collect their statistics from such sophisticated facilities, the picture the statistics tend to portray is that the important health problems of the country are the degenerative diseases of the more affluent and elderly city dwellers. There is thus a self-justifying circle; if a developing country adds a large kidney unit to its new teaching hospital, kidney problems will come to figure large in the health statistics—and in this way the emphasis on sophisticated and expensive kidney treatment facilities will be “justified.”
Why has health improved?
To some extent, major health campaigns have been a factor in the improvement of health in developing countries since World War II. Official health services for individuals, however, have probably had too low a coverage in many developing countries to have had much impact on overall national health levels. In any case, there is a limit to what conventional health services can achieve in an unchanged physical and social environment. For example, the medical school in Cali, Colombia had a program for hospital care of premature infants, with survival rates comparable to those in North America; but 70 per cent of the infants discharged from the premature nursery were dead within three months.
The existence of some attempt at modern water supply and sanitation in the towns of the developing world could well be one reason why health today in developing countries is better in the towns than in the countryside, whereas in nineteenth century Europe the reverse was the case. However, health in developing countries has also been improving in the rural areas—as the population explosion shows—and since water supply and sanitation usually remain very inadequate in such areas, there must be some other factor.
The most likely factors leading to health improvement, then, are a rise in the levels of nutrition and the slow spread of modern ideas of personal hygiene. Across the developing world, per capita incomes are rising, and transport systems are improving; the result is more food, better quality food, fewer localized food shortages, and a more varied diet. In other words, the principal factor behind the improvement in health and hence the population explosion in developing countries is probably not any form of health measure, but economic development itself.
Better health can increase production, increase the benefits of consumption, and improve the efficiency of investment for either purpose. As one example, a fall in morbidity can improve the efficiency of education or work time, and increase the benefit to the individual of his leisure time. A rather different type of possibility is that control of disease problems may open up new fertile land, permit exploitation of mineral resources, or increase trade and tourism. For example, the Panama Canal could only be built after disease problems had been brought under control. Again, in the past few years efforts at malaria control have permitted new settlements on fertile land in Ethiopia, Nepal, and Sri Lanka. Settlement of fertile land is one of the main anticipated benefits of the major project against “river blindness” now getting under way in West Africa.
To take another rather different situation, certain human diseases also affect animals, so that there is scope for programs, which improve human and animal health simultaneously, and facilitate exploitation of animal resources for meat or as stronger draft animals. Equally, provision of a good supply of water in an arid area may benefit livestock development or permit extra crop production, as well as benefiting human health. Or, again, everyone knows it is a waste if rodents eat crops while they are in storage; but it is just as much a waste when ingested food is eaten away by parasites within the body, when fever increases the metabolic demand for calories, or when enteric disease impairs intestinal absorption. Furthermore, if people are ill, there is a good chance that they will spend resources on attempted remedies (effective or not); it is, therefore, quite legitimate to count avoided costs of treatment as a benefit from better health.
Changes in health do, of course, affect population growth, and this must also be considered. In general, better health and nutrition—whatever their cause—tend to increase the ability to reproduce healthy babies with a good chance of survival; what one then has to consider is how the will to reproduce is affected.
Historical experience is varied; but on the whole, it seems likely that a decline in the birth rate is significantly related to a fall in the infant mortality rate, but that the birth rate falls less. It should be a deliberate aim of health policy to reinforce the link between a decline in mortality and a decline in family size, by having health workers recommend the benefits of smaller families, for the health benefit to the individual family, as well as on economic grounds.
Probably the most appropriate framework in which to analyze health policy is to view better health as part of a balanced socioeconomic development in the human environment. Better health is one aspect of man achieving more positive control of his environment and as such it helps to justify and facilitate planning of his future—social or economic, public or family affairs. Within this framework there are a large number of linkages. For example, family planning improves nutrition; nutrition improves health; health can improve attitudes to family planning and to development as well as directly increasing production. Emphasis on health is probably not essential for economic development (except for control of specific diseases in certain regions). However, better health in its own right should surely be one of the aims of development; and improving the health of the poorer members of a developing country could be a very important way of increasing their welfare.