It could be argued that the armamentarium of the chronic disease epidemiologist was forged in the heat of the controversy over the relationship between cigarette smoking and lung cancer. The methods used in the midcentury studies that fueled the debate, case-control and cohort studies, were soon to be used to study the causes of cardiovascular disease, and later a variety of noncommunicable diseases. They also had some noteworthy antecedents, including informal case-control studies of the association between syphilis and cancer1,2 and a formal multicenter case-control study of cancer in relation to a variety of risk factors.3 The latter study is worth resurrecting, for although it was too early to show an effect for cigarette smoking, as distinct from other tobacco use, it was the first analytic study to demonstrate the protective effect of eating vegetables.
See also p 2609.
But it was the reaction to the postwar studies