Medical statistics, patronage and the state: The development of the MRC Statistical Unit, 1911–1948

Abstract

The development of medical statistics based on the concepts of probability and error-theory developed by Francis Galton, Karl Pearson, George Udny Yule, R A Fisher, and their colleagues, has had a profound impact on medical science in the present century.This has been associated especially with the activities of the Medical Research Council (MRC) Statistical Unit at the London School of Hygiene and Tropical Medicine (LSHTM).It was here in 1946 that the world's first statistically rigorous clinical trial was undertaken, and where in the 1950s Austin Bradford Hill and Richard Doll revealed a statistical relationship between smoking and lung- cancer.But how and why was this key institution founded, and why were its methods probabilistic?'One might assume that the sheer brilliance of the work produced by Pearson and his statistical followers was sufficient to win over the medical profession, and to attract funding for the MRC Unit.However, one might question the extent to which an association with Pearson, and with the mathematical technicalities of his statistics, immediately endeared the fledgling discipline of biostatistics to the medical com- munity.As J Rosser Matthews has shown recently, Pearsonian statistics were only haltingly accepted by medical scientists in Britain in the 1920s and 1930s, and were little understood by medical practitioners.2How then was such a university-based infrastructure for medical statistics established?Ultimately, of course, the Statistical

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 21 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.