Infant mortality traditionally has been viewed as the measure of health, more strictly ill health, that is most sensitive to poverty. Let us look at one historical example.
Benjamin Seebohm Rowntree was the son of Joseph Rowntree, a chocolate manufacturer, Quaker, and philanthropist in York, England. B.S. Rowntree conducted a study to draw attention to the conditions of what he called the working-class population. He studied three typical areas of the town that housed the working class, grading them according to their degree of poverty.3
Infant mortality rates (the number of deaths in the first year of life compared with the number of live births) varied according to area. In the worst-off area the rate was 247 per 1,000 live births; in the middle working-class area, 184 per 1,000; and in the highest, 173. By contrast, among York’s “servant-keeping class” it was 94.
Rowntree was ready to attribute the high rate in the poorest area to overcrowding and poor-quality housing. He wondered why the rate in the highest working-class area was double that of the servant-keeping class. In the highest working-class area there was no overcrowding and no back-to-back houses but, rather, wide streets and houses with gardens. To Rowntree, this did not provide a ready explanation. He concluded that the cause must be ignorance—ignorance in the feeding and management of infants “rather than to other causes arising out of the poverty of the people.”
Ignorance versus poor conditions.
This view—that there is no relation between poverty and health and that it is all due to ignorance—was still being propounded in Britain in the 1980s by government ministers. To put it kindly, it was a limited view in the 1980s, as it was eighty years earlier. We now have a different view.
Even were it true that the high infant mortality of the higher working class was due to ignorance, how are we to account for an infant mortality rate of 94 per 1,000 among the wealthiest people of York around 1900? In England and Wales infant mortality in 2000 was 3.7 per 1,000 among infants born to fathers in the top social class and 8.1 among those born into the bottom class.4 Among single mothers, the rate was 7.6. The richer members of the community at the end of the nineteenth century had infant mortality rates that were much higher than the worst-off members of the community at the end of the twentieth century.
What are the implications of this comparison? The major determinants of high infant mortality are those associated with poverty of material conditions: lack of sanitation; malnutrition; low-quality housing and overcrowding; and lack of medical care including care before, during, and after childbirth. The threefold higher infant mortality rate of the poorest people of York was the result of worse conditions. It is less clear why the best-off people of 1900 should have so much higher rates than the worst-off in the country 100 years later. We can guess that although privileged economically, they were “deprived” of the conditions for low infant mortality: good sanitation, nutrition, and medical care.
Condition of the community.
This is a rather dramatic clue that factors other than individual income play a powerful role in the determinants of health conditions that we associate with poverty. This is a conclusion reached by Sam Preston.5 Two tentative conclusions from this example run through the discussion that follows. First, we should not view individual incomes in isolation from the community in which people are located. The rich people of York in 1900, in some relevant respects, lived under worse conditions than do poor people in the same city a century later. The community is richer now. Money and technical knowledge have allowed the community to invest in conditions that favor an alleviation of the conditions that lead to high infant mortality. If we are using individual income as a measure of standard of living, then it does only a partial job, because it misses out on the benefits to be derived from living in a richer community.
Black-white health differences.
The second conclusion relates to the first. In Preston’s terms, factors “exogenous” to income have been responsible for much of the health improvement in the twentieth century. Putting the infant mortality of social class V or single mothers for the year 2000, 7.9 per 1,000, beside that of the servant-keeping class of York for 1900 suggests that the problems of ill health due to material deprivation have, to a large extent, been solved in today’s industrialized countries. Why then should such countries continue to suffer from large inequalities in health?
One could argue that the “high” infant mortality rate of 15 per 1,000 among U.S. blacks in 2000, although a fraction of the servant-keeper rates of the past, was still due to residual problems of material deprivation: poor sanitation, inadequate nutrition, and poor housing.6 This is a possibility that must be considered.
But infant mortality is not the main reason for black-white differences in life expectancy in the United States. Arline Geronimus studied sixteen U.S. communities, black and white.7 In the United States as a whole, the probability of a fifteen-year-old man surviving to age sixty-five was about 77 percent. For a young black man in New York the probability of survival was 37 percent. The three major causes of death contributing to this tragic waste of life were HIV-related factors, homicide, and cardiovascular disease. We do not think that heart disease is related to poor sanitation, malnutrition, and overcrowded conditions in houses without gardens; not, in other words, to material deprivation of 100 years ago. Can material deprivation explain that? If we want to describe coronary heart disease as a disease of poverty, reversing the decades-long practice of describing it as a disease of affluence, we must take a hard look at what we mean by poverty.