The ecology of medical care: origins and implications for population-based healthcare research

Health Services Research, April, 1997 by Kerr L. White

An extensive literature search produced an eclectic array of population-based, practice-based, record and interview surveys of health status and morbidity in the United Kingdom and the United States. There was no shortage of so-called vital statistics on causes of death, but precious little was available on the health problems of the living. The surveys were of varying sizes, credibility, and utility. Our own extensive survey based on a probability sample of all community-based physicians in North Carolina and their referred patients was also an important source of data (Williams, White, Andrews, et al. 1960).

Bernie Greenberg was especially helpful in examining the credibility of the available data and our underlying assumptions as we standardized and adjusted the data to one-month periods for general populations of adults over 16 years of age. We used data from both Britain and the United States and had at least two sources to support most of our findings. At any rate, the detailed data and the calculations are all set forth in our original article. Bernie also contributed substantially in helping to design the diagram itself so that the squares were accurately sized to reflect the underlying data [ILLUSTRATION FOR FIGURE 1 OMITTED].

We submitted the paper to The New England Journal of Medicine. Joe Garland, the editor, accepted it almost immediately without any changes in the text, tables, or the diagram. However, he objected strenuously to the term "ecology" in the title. I insisted that the word had to remain and, after an extended long-distance telephone call I made from a pay phone in Atlantic City, he capitulated. Following publication we were widely and severely criticized by many of our colleagues in academic medicine - much of it coming from younger sub-specialists who felt that their future careers were being challenged. Our reply was simple: Specify the inaccuracies and come up with different figures if they exist. The study has been replicated in many quarters over the decades and the relationships persist. For a 1973 article in Scientific American I used national annual data from the National Center for Health Statistics, and the fundamental relationships still held [ILLUSTRATION FOR FIGURE 2 OMITTED] (White 1973).

Now for the "Ecology" model's implications for population-based healthcare research. There are three arenas in which the problems of health and disease may be studied: the laboratory, the one-to-one clinical setting, and the population. One is not good or bad, right or wrong, hard or soft. The site selected depends on the problem being addressed and the question being asked. The credibility and utility of the findings are what count.

During the seventeenth, eighteenth, and nineteenth centuries our fore-bears paid considerable attention to the population dimensions of disease; there were not many useful clinical interventions to offer. In the first half of this century, however, medicine's increasingly successful focus on the laboratory study of microorganisms and on their eradication was accompanied by a diminished study of health and disease in populations. Only in the last three or four decades has the population perspective resumed its traditional place among the essential sites for furthering the healthcare establishment's efforts to improve the public's health (White 1991).