Tag Archives: economic determinants of health

Economic development and mortality: Samuel Preston’s 1975 classic

Originally published at The Pump Handle

In the late 1940s and 1950s, it became increasingly evident that mortality rates were falling rapidly worldwide, including in the developing world. In a 1965 analysis, economics professor George J. Stolnitz surmised that survival in the “underdeveloped world” was on the rise in part due to a decline in “economic misery” in these regions. But in 1975, Samuel Preston published a paper that changed the course of thought on the relationship between mortality and economic development.

In the Population Studies article “The changing relation between mortality and level of economic development,” Preston re-examined the relationship between mortality and economic development, producing a scatter diagram of relations between life expectancy and national income per head for nations in the 1900s, 1930s, and 1960s that has become one of the most important illustrations in population sciences. The diagram shows that life expectancy rose substantially in these decades no matter what the income level. Preston concluded that if income had been the only determining factor in life expectancy, observed income increases would have produced a gain in life expectancy of 2.5 years between 1938 and 1963, rather than the actual gain of 12.2 years. Preston further concluded that “factors exogenous to a country’s current level of income probably account for 75-90% of the growth in life expectancy for the world as a whole between the 1930s and the 1960s” and that income growth accounts for only 10-25% of this gain in life expectancy.

Preston’s next main task was to contemplate what these “exogenous factors” might be. Preston proposed that a number of factors aside from a nation’s level of income had contributed to mortality trends in more developed as well as in less developed countries over the previous quarter century.  These factors were not necessarily developed in the country that enjoyed an increase in lifespan but rather were imported and therefore are, according to Preston, less dependent on endogenous wealth. The exact nature of these “exogenous” factors differed according to the level of development of the nation in question.  Preston identified vaccines, antibiotics, and sulphonamides in more developed areas and insect control, sanitation, health education, and maternal and child health services in less developed areas as the main factors that contributed to increased life expectancy.

Preston’s paper continues to provide guidance in development theory and economics today. But there was and continues to be considerable resistance to Preston’s theory, mostly from economists. Shortly after Preston’s article appeared, Thomas McKeown published two books that argued essentially the opposite: that mortality patterns have everything to do with economic growth and standards of living. Pritchett and Summers argued in 1996 that national income growth naturally feeds into better education and health services, which in turn contribute to higher life expectancy.

How well does Preston’s analysis hold up today? For one thing, Preston did not foresee the seemingly intimate connection between development and the recent rapid increased incidence prevalence of some chronic diseases. As developing nations urbanize and become more affluent, noncommunicable diseases, such as cancer and heart disease, many secondary to “lifestyle” issues like obesity and lack of physical exercise, are now on the rise, with the potential to lower life expectancy significantly. So is wealthier healthier, to use the words of Pritchett and Summers? Not necessarily, as we are seeing increasingly.

Why is it so important to try to work out the relationship between health and wealth? If we assume that improvements in healthcare systems grow naturally out of increased wealth, then developing countries should be focusing primarily on economic growth in order to improve their healthcare. This must be true to a certain extent, but, as Preston is quick to point out, there are other factors that affect the health of a nation, and it is not sufficient to assume that economic growth will automatically lead to improved life expectancy. Preston’s analysis tends to emphasize instead that health systems strengthening and biological innovation must always take place beside economic growth to insure better health. Whether or not we can completely agree with Preston’s assertion that wealthier is not necessarily healthier, it is certainly the case that his landmark article stimulated an essential conversation about the relationship between economic development and mortality that continues avidly to the present day.

Leave a comment

Filed under Publication

Preventive care in medicine: Dugald Baird’s 1952 obstetrics analysis

Originally published at The Pump Handle

How much of a patient’s social context should physicians take into account? Is an examination of social factors contributing to disease part of the physician’s job description, or is the practice of medicine more strictly confined to treatment rather than prevention? In what ways should the physician incorporate public health, specifically prevention, into the practice of medicine?

These are the questions at the heart of Dugald Baird’s 1952 paper in The New England Journal of Medicine, “Preventive Medicine in Obstetrics.” The paper originated in 1951 as a Cutter Lecture, so named after John Clarence Cutter, a 19th-century medical doctor and professor and physiology and anatomy. Cutter allocated half of the net income of his estate to the founding of an annual lecture on preventive medicine. Baird was the first obstetrician to deliver a Cutter Lecture. Baird’s paper draws much-needed attention to the role of socioeconomic factors in pregnancy outcomes.

Baird begins by describing the Registrar General’s reports in Britain, which divide the population into five social classes. Social Class I comprises highly paid professionals while Social Class V encompasses the “unskilled manual laborers.” In between are the “skilled craftsmen and lower-salaried professional and clerical groups”; the categorization recognizes that job prestige as well as income is important in social class. Baird proceeds to present data on maternal and child health and mortality according to social group as classified by the Registrar General’s system. He makes several essential observations: social class makes relatively little difference in the stillbirth rate, but mortality rates in the first year of life are lowest for the highest social class (Social Class I) and highest for the lowest social class (Social Class V). Social inequality is thus felt most keenly in cases of infant death from infection, which Baird calls “a very delicate index of environmental conditions.”

Baird goes on to analyze data on stillbirths and child mortality from the city of Aberdeen, Scotland, which he chose because the number of annual primigravida (first pregnancy) deliveries at the time was relatively small and therefore manageable from an analytic standpoint and because the population in the early 1950’s was relatively “uniform.”  When comparing births in a public hospital versus a private facility (called a “nursing home” in the paper, although not in the sense generally understood in the U.S. today), many more premature and underweight babies died in the public hospital than in the private nursing home, even though only the former had medical facilities for the care of sick newborns. The difference could not, therefore, be explained by the quality of medical care in the two facilities.

Baird concludes that this discrepancy must have something to do with the health of the mothers. Upon closer examination, Baird recognizes that the mothers in the private nursing home are not only healthier but also consistently taller than the mothers in the public facility. According to Baird, the difference in height must have to do with environmental conditions such as nutrition, a reasonable conclusion although Baird in fact did not have available data on ethnicity or other factors that might have also contributed. As the environment deteriorates, the percentage of short women increases. Baird notes that height affects the size and shape of the pelvis, and that caesarean section is more common in shorter women than taller women. Baird began classifying patients in the hospital in one of 5 physical and functional classes. Women with poorer “physical grades,” who also tended to be shorter, had higher fetal mortality rates. He also observes that most women under the age of 20 had low physical grades, stunted growth, and came from lower socioeconomic statuses. Baird spends some time examining the effects of age on childbearing, looking at women aged 15-19, 20-24, 25-29, 30-34, and over 35. Baird found that the most significant causes of fetal death in the youngest age group (15-19) were toxemia, fetal deformity, and prematurity. Fetal deaths in women aged 30-34 tended to be due more frequently to birth trauma and unexplained intrauterine death. The incidence of forceps delivery and caesarean section grew sharply with age, and labor lasting over 48 hours was much more common among the older age groups.

In a turn that was unusual at the time, Baird considers the emotional stress associated with difficult childbirth and quotes a letter from a woman who decided not to have any more children after the “terrible ordeal” of giving birth to her first child. This close consideration of the patient’s whole experience is a testament to Baird’s concern with the patient’s entire context, including socioeconomic status.

Baird concludes by making a series of recommendations for remedying social inequalities in birth outcomes, some of which make perfect sense and some of which now strike us as outrageously dated. An example of the latter is his suggestion that “the removal of barriers to early marriage” would improve birth outcomes among young women. In fact, we now know that early marriage can have a negative impact on women’s sexual health, sometimes increasing incidence of HIV/AIDS.

Despite the occasional “datedness” of Baird’s paper, his analysis is not only a public health classic in its attempt to bring social perspective back into the practice of medicine but it also contains lessons that are still crucial today. Baird’s paper reminds us that gender is often at the very center of health inequities, and that maternal and infant mortality constitute a major area in which socioeconomic inequalities directly and visibly affect health outcomes. While maternal and infant mortality rates are not high in the developed world, they still constitute serious health problems in developing countries. Infant mortality in particular can be used as a useful indicator of socioeconomic development. Most importantly, Baird’s paper, written in an age when the medical field began relying increasingly on biology and technology, reminds us that it has much to gain from paying attention to social factors that have a crucial impact on health.

Leave a comment

Filed under Publication

Inequality, stress, and health: The Whitehall Studies

Originally published at The Pump Handle

The current state of public health research is increasingly aware of the effects of various kinds of inequality on health. Especially in the U.S. and other developed countries, the burden of chronic, non-communicable diseases is especially high among low-income individuals. Public health campaigns have thus developed initiatives to bring better health education to low-income communities and to encourage healthier eating, more exercise, and decreased use of tobacco and drugs.

What’s not always recognized is that the rationale behind many of these initiatives stems from the findings of a series of landmark studies among civil servants in England: the Whitehall Studies. Whitehall I began in 1967, at a time when awareness of the relationship between social inequality, work stress, and poor health was not necessarily keen. The Whitehall Studies thus set out to investigate the complex relationships among income, work status, psychosocial support, health behaviors, and resulting morbidity and mortality.

Whitehall I, published in Journal of Epidemiology and Community Health in 1978, followed British civil servants over ten years from 1967-1977 and ultimately showed an inverse association between employment grade and mortality from a variety of illnesses. After ten years of follow-up, those in the highest employment grade had one-third the mortality rate of those in the lowest grade.

The results of Whitehall I demanded some attempt at explanation. Whitehall II, a study of 10,314 British civil servants aged 35-55 published in The Lancet in 1991, represents this attempt. Importantly, while Whitehall I studied only male civil servants, Whitehall II also included women. Particularly important to Whitehall II is the effect of psychosocial factors, such as stress and social support, on disease outcome.

Whitehall II, conducted between 1985 and 1988, collected data primarily via surveys and questionnaires. Questionnaires distributed to the 10,314 British civil servants from 20 civil servanice department included questions about: health status, including questions about past medical history; work characteristics such as job demands and decision-making capacity; social networks and support; health behaviors, including smoking and physical activity; and the Framingham type A score, measuring hostility and stressful life events. In addition to questionnaires, subjects’ blood pressure was measured and screening was provided for angina and ischemia.

Whitehall II confirmed some of the results of Whitehall I, especially the finding that the lower the job status, the higher the prevalence of ischemic heart disease. New, landmark findings revolved around gender differences and isolation as possible causes of this inverse relationship. In general, women had greater morbidity than men in all grades of employment. The higher the job status of the man, the more likely he was to be married or cohabitating, but the opposite was true for women. Obesity was more prevalent among those in lower status jobs, and the risk factor for increased morbidity and mortality that differed most among job categories was smoking.

Overall, individuals in lower grades of employment were more resigned about their health. Individuals in lower employment grades were more likely to believe that it was not possible to reduce the risk of a heart attack. They were also more likely to rate their health status as poor and to report two or more of eight stressful life events in the previous year. Self-perceived poor health status is known to be a predictor of mortality.

Perhaps most significantly, the Whitehall II study discovered a firm connection among psychosocial factors, perception about work status and environment, and poor health outcomes. People in lower status jobs reported less social support than those in higher status jobs. They also reported less control, less use of skills, and less variety at work, as well as a more demanding workload and more psychological stress at work. All of these factors are associated with risk of cardiovascular disease.

The Whitehall Studies helped public health researchers develop a nuanced perspective on the relationship between socioeconomic position and health. The findings of Whitehall II established that it is not only factors most obviously related to obtainment of healthcare, such as time to attend doctors’ appointments or money to pay for insurance, that seriously affect health outcomes. Rather, psychosocial and socioeconomic factors that may seem unrelated to the provision of healthcare may have a direct impact on risk of developing cardiovascular disease and other chronic illnesses. Whitehall II in particular calls for careful attention to all of the conditions of every socioeconomic category as possible predictors of health outcomes. In many ways, these crucial studies established the interconnectedness of socioeconomic position, psychosocial factors, and health, a concept that is now at the very center of public health.

Leave a comment

Filed under Publication