Non-responders in a quitline evaluation are more likely to be smokers – a drop-out and long-term follow-up study of the Swedish National Tobacco Quitline

Abstract

Background: A previous randomized controlled trial (RCT) of the Swedish National Tobacco Quitline detected no significant differences in smoking cessation outcomes between proactive and reactive services at 12-month follow-up. However, the response rate was only 59 % and non-responders were over-represented in the proactive service. We performed a drop-out analysis to assess the smoking status of initial responders and non-responders. Methods: At 29–48 months after the first call, a postal questionnaire with six questions was sent to 150 random clients from the RCT database, with equal numbers from the proactive and reactive services as well as responders and non-responders at 12-month follow-up. Clients who did not return the questionnaire were contacted by telephone. The outcome measures were point prevalence (PP) and 6-month continuous abstinence (CA), and their associations with response status at 12 months were assessed by logistic regression. Results: The response rate was 74 % (111/150). Abstinence was significantly higher among initial responders than non-responders (PP 54 % vs. 32 %, p = .023 and CA 49 % vs. 21 %, p = .003). The odds ratios for initial responders vs. initial non-responders were, for PP = 2.5 (95 % CI 1.1–5.6, p = .024), and for CA = 3.7 (95 % CI 1.5–8.9, p = .004), after adjusting for proactive/reactive service. Conclusions: Non-responders to a 12-month follow-up smoking cessation questionnaire in a quitline setting were more likely to be smokers 1.5–3 years later. We propose a conservative correction factor of 0.8 for self-reported abstinence in telephone-based cessation studies if the response rate is approximately 55–65 %. Keywords: Effectiveness, Intention-to-treat, Non-response, Per protocol, Proactive, Reactive, Responder-only analysis, Smoking, Telephone, Questionnaire