Abstract
The increasing evidence of the ill-health effects of environmental tobacco smoke (ETS) has prompted the search for accurate measures of exposure to ETS. The present study examined whether it was possible to enhance the ability of questionnaire-derived assessments of ETS exposure, to predict salivary cotinine. Salivary samples were obtained from 258 nonsmoking bank employees, who simultaneously answered questions detailing their exposure to second-hand smoke within the last three days. Exposure models were created, to take into account the number of smokers nearby, length of time in their presence, half-life of cotinine in bodily fluids, level of aversion to cigarette smoke and time of year. All models, including the consideration of intensity and duration of exposure combined, explained an equal amount of variance of log cotinine levels (approximately 16%). The weak relationship between questionnaire estimates of ETS exposure and cotinine, found in the present study, suggests that further investigation is needed to improve the assessment of recent ETS exposure.
