University of Minnesota

Princeton Conference on Methodology in Epidemiological Studies of Cardiovascular Diseases

April 24-26, 1959

The Princeton Conference was the final one of a series of critical methods conferences called jointly by the American Heart Association and NIH to deal with issues of measurement in cardiovascular disease epidemiology. The topics considered in that early tumultuous period of research on CVD prevention were basic: the criteria for diagnosis and clinical evaluation of coronary heart disease and hypertension, measurements of diet, physical activity and biochemical characteristics; assessment of cultural/societal/familial/psychological and genetic influences; and a general approach to problems of study design and data analysis.

The chairperson opened proceedings with this challenge: “… The co-sponsoring organizations felt that the time had come to devote specific attention to problems of measurement faced by individual investigations and to the potential value of being able to combine, or at least compare, results of several investigators using standard measurements.”

The object was to obtain a minimum collection of standard baseline measurements that could be included in every CVD epidemiological study, aside from optimal measurements that might be made in some studies able to employ more precise methods.

The conference was a culmination of the work of nine working groups that had prepared preliminary reports. It was chaired by Herbert Pollack of the American Heart Association Committee on Nutrition, with Herman Hilleboe, director of the New York State Department of Health in Albany as co-chair and Dean Krueger from the National Heart Institute as secretary.

Many pioneers in CVD epidemiology were represented at the conference: the section on design and analysis was chaired by Felix Moore, formerly of the National Heart Institute, by that time at the University of Michigan; Marjorie Bellows and Len Cook from the American Heart Association; Harold Dorn and Phil Enterline from the Public Health Service; Zdynek Fejfar from WHO; Cuyler Hammond from the Cancer Society, Ralph Paffenbarger from NIH, and David Rutstein from Harvard, Knut Westlund, representing the studies in Norway, and Jacob Yerushalmy, statistician from Berkeley.

Many of the specialists in hypertension were there, and that session was chaired by Joseph Doyle, George Meneely, and Roland Schneckloth. Fred Epstein and Ernst Simonson represented the Minnesota Code and electrocardiographic criteria, with Bob Grant and Charles Kossmann. Clinical evaluation was led by Oglesby Paul and Roy Dawber, and those dealing with autopsy findings included David Spain and Jack Strong.

Ancel Keys was there on the Physical Activity Committee, along with Henry Taylor and Bill Zukel. Biochemical measurements had the representation of Gerald Cooper as well as John Peters from the Heart Association and George Mann from Vanderbilt.

The segment on cultural, psycho-social and other risk elements was chaired by Harry Kruse and had as a member Hans Selye, the originator of the stress hypothesis, along with Jerry Stamler and Larry Hinkle.

Other high level agency representatives included George Wakerlin, Luther Terry, Larry Lamb, and Abraham Kagan.

A letter from cardiologist A.C. Corcoran to James Watt at NIH, apparently set in motion the processes leading to the conference. It emphasized to Dr. Watt “the urgent need for standardization of procedures in cardiovascular surveys…”

The chairman, Herb Pollack, waxed historical: “Those attending this conference recognize that in the history of medicine, no disease has ever been conquered by an attempt to treat every infected individual. It is only by studying the etiology through the epidemiological approach that man has been able to stamp out mass disease. It is hoped that this approach can be applied to the problem of cardiovascular disease.”

The conference dealt with general problems of definition, design, and measurement. It addressed surveys involving clinical and pathological examination of individuals, eliminating household interviews or death certificate studies.

Felix Moore, speaking of design and analysis, characterized epidemiological studies as those undertaken with “persons known to have the disease under consideration, but also persons who are free or are apparently free of the disease as well.” Thus, making possible the finding of factors that differentiate two groups. He went on to lecture the multitude (splendidly) on the characteristics of epidemiologic study.

The segment on psychosocial variables and genetics was a long dissertation without definitions or criteria or precision, as was so heartily pursued by the other panels. The preparers of the report remarked that in regard to the psychosocial, genetic, and other influences considered in the section on miscellany, “It did not seem feasible in the limited time available to attempt to arrive at the synthesis of ideas on methodology in an area of study in which there are so many unknowns.”

Ralph Paffenbarger summarized study methods used in 21 studies in cardiovascular disease epidemiology.

The conference was co-sponsored by the American Heart Association and the National Heart Institute and published in the American Journal of Public Health and the Nation’s Health, Supplement to vol 50, number 10, October, 1960, pages 1-124. It refers to earlier studies such as the first National Conference on Cardiovascular Diseases (held January 18-20, 1950 in Washington, and published by the American Heart Association, New York, NY) followed by the Conference on Epidemiology of Atherosclerosis and Hypertension at Arden House in 1956 and the Beaconsfield Conference in 1957.

The Princeton Conference and its predecessors on methods, along with its 1960 publication, provided much-needed guidance for the growing discipline of CVD epidemiology. CVD field methods were eventually codified internationally in the 1968 WHO publication, “Cardiovascular Survey Methods.” (Henry Blackburn)


Pollack, H. & Krueger, D.E. (eds.) 1960, Epidemiology of Cardiovascular Diseases Methodology: Hypertension and Arteriosclerosis, Princeton, American Heart Association, National Heart Institute: New York, vol. 50(suppl), no. 10, pp. 1-124.