Schlesselman James J
Department of Biostatistics, University of Pittsburgh Graduate School of Public Health, and University of Pittsburgh Cancer Institute, Pittsburgh, PA 15213, USA.
Prev Med. 2006 Oct;43(4):251-5. doi: 10.1016/j.ypmed.2006.07.014. Epub 2006 Sep 25.
Two influential case-control studies that clearly implicated cigarette smoking as a cause of lung cancer are reviewed in terms of their respective strengths and weaknesses. The findings from a U.S. study reported in 1950 by Wynder and Graham were strikingly similar to those arising from a U.K. study reported later that year by Doll and Hill. The methodological rigor of these investigations effectively ruled out alternatives to smoking as plausible explanations for the increased risk, although additional investigations in animals and man were needed to buttress the original results. The exceptionally high relative risk of lung cancer that was found has had far-reaching medical, social, and economic consequences.
