Efficient screening of current smoking status in recruitment of smokers for population-based research

Population-based samples of smokers are necessary for tobacco behavior monitoring and surveillance and for evaluating tobacco control programs and policies. We evaluated the sensitivity and specificity of a simple, one-question screener as a tool to maximize efficiency of obtaining a population representative sample of current smokers. This analysis was based on 5,002 respondents from the Ontario Tobacco Survey (OTS), a regionally stratified longitudinal survey of adult smokers and cross-sectional survey of nonsmokers in Ontario, Canada. Overall, the question "Have you smoked one or more cigarettes in the past six months?" achieved at least 99.7% sensitivity and 87.1% specificity when compared with several standard definitions of current smoking status. The brief screening question minimized respondent burden and data collection costs, and may have had a positive influence on response rate. Having a more conservative measure of smoking status permitted atypical smokers to be included in the survey which will allow us to track their behavior change and evaluate the performance of accepted smoking status definitions. We recommend that studies, which specifically sample smokers, utilize any past 6-month smoking as a brief screener for smoking status.