Epidemiology and policy: the pump handle meets the new millennium.

Abstract

The direct linkage of epidemiologic evidence to the formulation of policy intended to advance public health is widely acknowledged (1). Almost universally, epidemiologists tell the story of John Snow and the Broad Street pump to illustrate the immediacy of observational findings for solving public health problems. Based on observation of cholera cases clustered along Broad Street in London, Snow recommended the removal of the municipal pump handle, and his advice was heeded (2, 3). This example is particularly compelling, because Snow demonstrated waterborne transmission of cholera before there was knowledge of the existence of the Vibrio cholerae organism. Numerous other examples also considered triumphs of epidemiologic inquiry include the establishment of cigarette smoking as a cause of lung cancer and other diseases; the identification of powerful and remediable causes of cancer, such as asbestos exposure and diethylstilbestrol administration during pregnancy; and the characterization of risk factors for acquired immunodeficiency syndrome. In spite of these evident successes, the place of epidemiologic evidence in supporting policy-making, as well as the role of epidemiologists in the process of policy-making, remains controversial and in flux. As a core discipline of biomedical research, epidemiology is not unique in generating evidence relevant to policy: the ultimate goal of all biomedical research is to advance the health of people. Epidemiology as a scientific method brings evidence that bears directly on the health of the population, and it is this direct linkage that distinguishes epidemiology from other branches of biomedical research. As a consequence, epidemiologic findings generally have