Causes, risks, and probabilities: probabilistic concepts of causation in chronic disease epidemiology

Abstract

Identifying and understanding causes of disease is arguably the central aim of the discipline of epidemiology. However, while the discipline has matured over the past sixty years, developing a battery of quantitative tools and methods for data analysis, the discipline of epidemiology lacks an explicit, shared theoretical account of causation. Moreover, some epidemiologists exhibit discomfort with the concept of causation itself, concerned that it creates more confusion than clarity. This paper describes how, during the post-war period, epidemiologists began to think about causation in new ways as they encountered novel challenges in studying chronic diseases. The epidemiologic evidence linking cigarette smoking and lung cancer in the 1950s provided a focus for debates over causation. While some epidemiologists embraced probabilistic concepts of cause and effect, others maintained that causal mechanisms must ultimately be deterministic. The tension between probabilistic risk factors and deterministic causal mechanisms continues to haunt epidemiology today.