“…If sensitivity to the acute effects of exposure were linked to lung cancer susceptibility and/or cigarette smoking, selective attrition of susceptible workers could lead to a healthy worker survivor effect, with the appearance of an inverse dose-response relationship. Such a mechanism has been proposed for cotton industry workers as well [Hodgson and Jones, 1990]. Chance variations in smoking patterns are unlikely given the large overall size of the study, the use of an internal analysis, and the use of a matched analysis to control for year of birth sex, race, and plant [Siemiatycki et al, 1988].…”