The historical rise of “overdiagnosis”—an essay by Scott Podolsky

In April 1924 a Winnipeg physician, James Douglas Adamson, gave an address on Over-Diagnosis of Pulmonary Tuberculosis at a conference of medical superintendents in Ottawa.1 “Phthisis is a disease which when incipient is hard to diagnose and easy to cure but when far advanced is easy to diagnose and hard to cure,” he began, hearkening back to Hippocrates. In the 20th century, noted Adamson, advances in treatment such as the sanatorium “have all in turn increased the demand for early diagnosis until now it is recognized as a sine qua non of adequate treatment.” And yet, such demand had “an undesirable by-product which must be recognized and guarded against.”

That by-product was overdiagnosis. Adamson said that it was “bad for the patient physically, financially, and psychically; bad for the physician mentally and morally; bad for the country economically.”

Many of the themes that would characterise late 20th and early 21st century concerns about overdiagnosis are manifested in Adamson’s paper: the desire for early detection and hence treatment and cure; the fear driving such a movement; and, of course, the stated potential for adverse consequences of such diagnoses. Yet there’s a distinction between the broad, vernacular use of the term “overdiagnosis” over the past century and the more precise way it’s been used since the 1970s, by people focused on the heterogeneity within existing disease categories (especially particular forms of cancer), as well as those focused on the definitional fuzziness at the outer boundaries of certain other disease categories such as hypertension and diabetes.

This essay examines the invocation of “overdiagnosis” in the context of both the broader history of detecting and defining …