University of Minnesota

Luc Berlivet on Medical Science Advanced by Chronic Disease Epidemiology

French historian of science, Luc Berlivet, recently expanded on the evolution and historical contribution of chronic disease epidemiology, noting that: “In the second half of the twentieth century, epidemiology came to shape public health discourses and practices to an unprecedented extent” (Berlivet 2005; 39).

Addressing the attitude of some bench and clinical scientists in the 1950s and 60s debate about interpretation of the statistical relationship of cigarette smoking and lung cancer, Berlivet chastises those experts and dissects their problem: “Obsessed by the naïve idea of finding a single cause for every effect, they shake their heads and say that the evidence is only statistical, as if that invalidated it” (ibid.).

Increasingly, Berlivet argues, public health issues are defined by epidemiology, and now: “to be taken seriously as a legitimate public problem, a concern has to be framed as a risk, through the mobilisation of the epidemiological way of reasoning, its techniques and tools” (ibid.).

Thus, the principle of using inferential statistics to evoke causes of disease was tested and established in the lung cancer-cigarette debates. We point out in our history that CVD epidemiology also played a major role in formulating the risk factor paradigm. (1)

The great lung cancer-cigarette debate, also “provided Austin Bradford Hill with the opportunity to set out his ‘pragmatics’ of risk factor epidemiology,” his classic guidelines for arriving at causal inference from statistical associations (ibid.39). These fundamental criteria have immeasurably advanced epidemiology’s contribution to causality and thus, to prevention. They have helped stabilize what Mervyn Susser termed “the grammar of pragmatic epidemiology,” (ibid. 42) and provided greater understanding of and confidence in the discipline.

On the other hand, Berlivet characterizes as simplistic and satirizes the epidemiology- protagonists’ portrayal of the battle over accepting epidemiological evidence as “might against right” and “survival of the fittest,” saying:

The intellectual and moral quality of the epidemiologists and those who supported them is seen as substantiated by their ability to overcome their misguided, if not malicious, opponents to impose the truthfulness of their view. In a somewhat circular way, the sound scientific quality of the new style epidemiology is presented as the sole decisive element which could explain the final victory of the ‘causal side’ over opposing interpretations” (ibid.44)

Berlivet intends to go beyond such polar attitudes and moralism in his own addressing of the historical controversy, attributing strengths to both sides in the debates. He also laments that there has been no analysis of the role of epidemiologists or epidemiology in stimulating the larger social forces that have led to prevention policy and program and, we might add, to mass change in health behaviors. He maintains, however, that the volume of attacks on the central knowledge about causes derived from epidemiology is acknowledgment of its historical success as a scientific discipline.

(Footnote 1) Berlivet’s view [that of science historian] about the two primal studies, Doll and Hill and Framingham: “Although both studies were initiated in 1947-9, the tobacco–cancer studies are depicted as the landmark in the rise of ‘Modern Epidemiology’ while the ‘Framingham’ study is usually regarded, more narrowly, as the start of cardiovascular epidemiology – that is, a sub-discipline within the main framework” (Footnote 14 to Berlivet 2005, 63).


Berlivet, L. 2005 Association and causation : the controversy over the scientific status of “risk factor” epidemiology c.1947-c.1965, in : Berridge. V. Ed. Networks of knowledge and Power : Science, Research and Policy since 1945, V. Berridge (Dir.), London-Amsterdam, Rodopi (Wellcome Series in the History of Medicine), 43-74.