The growing rift between epidemiologists and their data

Excerpt

In 1662, John Graunt, a curious London haberdasher, authored the first epidemiology book, a concise monograph with the ungainly title, “Natural and Political Observations Mentioned in a Following Index and Made Upon the Bills of Mortality” [1]. The book was remarkable for many firsts, including the first life table and the first report that, among human births, there are more males than females [2]. It also set a standard for presentation of data in epidemiologic studies that prevailed until recent times: numerous, exhaustive tables that displayed raw data in support of the dozens of conclusions that Graunt drew from the Bills of Mortality (Fig. 1). Two centuries later, John Snow published his treatise on cholera, which is renowned for the intricate maps that it contained to illustrate his arguments about transmission of cholera [3]. Snow’s paper, however, contained more data than just maps. It also contained detailed tables filled with data from which he drew his inferences (Fig. 2).

…