OBJECTIVES: To describe the characteristics of a sample of smokers recruited proactively into a smoking cessation trial, and to compare these characteristics with the wider population using data from the General Household Survey (GHS) and National Statistics Omnibus Survey. STUDY DESIGN: Sample recruited for a randomized controlled trial. METHODS: Between August 2007 and October 2008, 123 general practices mailed questionnaires to smokers in the U.K. identified from computer records. Smokers willing to participate in a trial of personalized computer-tailored feedback returned the questionnaires to the research team. The characteristics of the sample were compared with the wider population using data from the GHS and National Statistics Omnibus Survey, and Index of Material Deprivation scores. RESULTS: A response rate of 11.4% (n = 6697) was achieved. The sample was demographically similar to the population sample, with an even distribution of participants from areas of both high and low deprivation. The sample was more dependent than the GHS sample, but less dependent than clinic samples. Distribution by motivation and readiness to quit was similar to population estimates. CONCLUSIONS: Public health strategies targeting the entire population of smokers are needed to counter the low recruitment rates resulting from the traditional reactive methods of recruitment to smoking cessation studies. Using computerized records to identify and contact patients who are smokers is a simple method of recruiting a larger, more representative sample of smokers.