To the Editor.
—The ongoing multicenter study by Fontham et al1 of lung cancer risk in nonsmoking women exposed to environmental tobacco smoke (ETS), which was updated in the June 8, 1994, issue of JAMA, is an impressive piece of work. The trend is clear: passive exposure to cigarette smoke is a risk. But I fear the authors dilute that point by incorporating too many tenuous variables in their statistical analyses. They acknowledge the problem of such factors as childhood recall of exposure and make attempts to control for it, but seem unwilling to admit that these and other data (eg, social exposure) are essentially not quantifiable by present methods. Inclusion of all these variables therefore led to a few implausible outcomes, which could give a window of solace to those who would defend ETS.The most implausible conclusion was that "[t]he excess risk of lung cancer among women
Kirkland LR. Environmental Tobacco Smoke and Lung Cancer in Nonsmoking Women. JAMA. 1995;273(7):519-520. doi:10.1001/jama.1995.03520310011006