Letters Section Editor: Robert M. Golub, MD, Senior Editor.
To the Editor: Dr Godtfredsen and colleagues1 report the results of a well-performed, large, observational study of the association between smoking reduction and the risk of lung cancer. Because this was an observational study, I believe that the authors should have been more circumspect with their conclusions. The abstract and comment sections state that smoking reduction by 50% significantly reduces the risk of lung cancer, which indicates a causal relationship. Although the association demonstrated by this study is convincing, causality cannot be concluded from an observational investigation. Observational studies are vulnerable to bias resulting from nonrandom assignments of patients to therapy.2 A randomized controlled trial is required to appropriately infer cause and effect.
Berger J. Smoking Reduction and Lung Cancer Risk. JAMA. 2006;295(9):1001-1002. doi:10.1001/jama.295.9.1001-a