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

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

During the early 1950s the University of North Carolina at Chapel Hill expanded from a two-year to a four-year medical school (the first new school after World War II) and created a new school of public health. Bernard Greenberg, Frank Williams, Dan Martin, Bob Huntley, and I were among the first faculty members appointed. As a consequence of our attempts to bridge the emerging gulf between public health and medicine, those of us in the medical school arranged for several faculty members of the School of Public Health to teach in our classes. These included Bernie Greenberg, then chairman of the department of biostatistics and later dean of his school, and his colleagues, Sydney Kark and the late John Cassel, two pioneering social epidemiologists. Several of us, in turn, participated in seminars and as thesis advisors in the School of Public Health. From the very beginning our friendships flourished. About 1954, in addition to our teaching innovations, we established what we called a Medical Care Research Group-one of the first, if not the first, in the country. We undertook what we then thought was a relatively new line of enquiry concerned with patient referral patterns, the adequacy of communications between patients and physicians and, among physicians, the quality of care, and randomized clinical trials.

Most of this research was centered in what we called a "General Medicine Clinic," which William Fleming, chairman of preventive medicine, Frank Williams, and I ran. Undergraduates and residents, as well as all faculty members in the department of internal medicine, had extended rotations and assignments in the General Clinic coping with the problems of general medical patients. The goal was to expose all faculty, medical students, and residents in the department of medicine to the problems faced by "generalists." In addition, we were committed to preparing adequate numbers of "general physicians" for North Carolina as promised to its state legislature in return for the funding of UNC's new medical school.

Bernie Greenberg also helped frequently with problems in research design, tutored us patiently, and was largely responsible for enlarging my own interest and knowledge of health statistics as distinguished from bit-statistics. Several of us, guided by Bernie, also participated in one of the first double-blind randomized control studies conducted in this country-but that is another story (Dorsett, Woods, White, et al. 1958). Now on to the origins of the logo adopted for this Institute with which my name is now associated, and its initial publication in "The Ecology of Medical Care" in The New England Journal of Medicine in 1961 (White, Williams, and Greenberg 1961).

In the fall of 1960 I returned from a sabbatical year in London to Chapel Hill where I was an associate professor of internal medicine. Charles Burnett, the original chairman of that department, who had hired me, was absent because of a long-lasting illness. His deputy, the late Louis Welt, a highly skilled and widely regarded nephrologist, replaced him as chairman. Lou believed that medicine's future lay in the direction of ever-increasing sub-specialization. To his everlasting credit he later changed his views substantially, but at that period he saw the General Clinic and all its works as an outmoded and generally ill-conceived effort. He intended to see it replaced by an ever-increasing series of sub-specialty clinics.

Chuck Burnett had a broad view of medicine's mission and had participated enthusiastically in the creation of the General Clinic. The Clinic had a seven-year history of well-received teaching, research, and service supported by the Commonwealth Fund, the Rockefeller Foundation and, of course, the State of North Carolina. Those of us responsible for running the General Clinic were dismayed by Lou's behavior. I had numerous talks with him about the Clinic's future and the need for general physicians to provide what we later called, for the first time in the "Ecology" article: "primary medical care." All of these discussions were unsatisfactory from my viewpoint. Sometime early in 1961, Lou and I discussed a forthcoming site visit from the Commonwealth Fund's officers to assess our progress and the prospects for renewing our grant for the General Clinic. Lou was unimpressed with my insistence that we prepare a written report for the Fund accounting for our use of the grant funds and relating our accomplishments. Lou's response was, "Just tell them we've spent the money!" This reply and the preceding discussion infuriated me. I recall returning to my office and thinking (and later recounting to others), "Lou and his ilk just don't understand the problems of providing appropriate medical care to all who need it."

While licking my wounds following this heated altercation with Lou Welt, I determined that the only way to demonstrate the relative needs of patients and the different requirements for generalists and specialists was with facts. As Alice in Wonderland remarked, "I would sooner see it done on paper with numbers!" I recalled a 1954 article published by two general practitioner friends of mine in London, John and Elizabeth Horder (1954). A diagram based on data generated from their own practice and other sources consisted of three squares of decreasing size. In the course of three months, only one-third of the patients enrolled in their general practice had sought care. About a quarter of these had been referred to a hospital-based specialist or consultant as an outpatient or inpatient since general practitioners in Britain did not have hospital privileges. Would these relationships hold for general populations in the United States as well as those in Britain, and could the relationships be documented more precisely?