Effect of recruitment method and setting on the composition of samples consisting of adult smokers

ObjectiveThis study aims to determine differences in samples of adult smokers recruited by passive or active recruitment in community pharmacy (PP and AP) or passive recruitment in general practice setting (PG), then comparing these samples to an unrecruited cohort of Dutch smokers.MethodsThe three recruited samples were compared on demographics, smoking behaviour, motivational determinants, stage of change and intention to use specific action plans when quitting using multinomial logistic regression analyses; this method was also used to assess whether recruitment affected the acquired samples concerning demographic variables compared with the cohort. PP, AP and PG response rates were compared with Tukey post hoc tests.ResultsSignificant differences were found for both AP and PG smokers, compared with PP smokers. Most important results include a higher rate of pre-contemplators in the AP sample and a lower educational level in the PG sample. All recruited samples appeared to be significantly different from the cohort sample.ConclusionRecruitment method and primary care setting does influence the sample recruited, this should be taken into consideration when recruiting a specific sub-sample.Practice implicationsThese results are valuable for researchers contemplating recruitment in a primary care setting.