Objectives: Using saliva cotinine as a quantitative marker, we examined the contribution of factors other than parental smoking to children's passive exposure to tobacco smoke.
Methods: Saliva specimens from a random sample of 734 7-year-old schoolchildren in Edinburgh, Scotland, were analyzed for cotinine. Their parents completed a questionnaire covering smoking habits and conditions in the home.
Results: A number of independent predictors of cotinine were identified in addition to the main one of smoking by household members. These predictors included home ownership, social class, day of the week, season of the year, number of parents present, crowding in the home, the number of children in the household, and sex. Cotinine was higher in children from less advantaged backgrounds, during winter, on Mondays, in girls, and when fewer other children were present. The effects were similar between children from nonsmoking and smoking homes.
Conclusions: Questionnaire measures of parental smoking are insufficient to fully characterize young children's exposure to passive smoking. Because socioeconomic variables contribute to measured exposure, passive-smoking studies that treat class as a confounder and control for it may be overcontrolling.