Mobile-based behavioral support by ex-smoking peers for smoking cessation : a pragmatic randomized controlled trial

Abstract

Background: Involving peer support for promoting smoking cessation was recommended. The widely used instant messaging apps (e.g., WhatsApp) were a promising modality for delivering instant smoking cessation support cost-effectively. Rarely study has rigorously assessed the effect of mHealth-based ex-smoking peers support on smoking cessation in community smokers. Methods: This thesis has two phases. In the first phase, a two-arm, pragmatic, randomized controlled trial proactively recruited biochemically validated daily adult smokers from the community. All the participants received brief smoking cessation advice (AWARD: ask, warn, advise, referral, do-it-again) at baseline. After individual randomization, each participant in the intervention group joined an instant messaging 3-person chat group, which consisted of an experienced smoking cessation advisor and a trained ex-smoking peer who had quit smoking for more than 1 year. Participants received 12-week tapering scheduled text messaging on smoking cessation, five positive psychological exercises, and real-time behavioral support from ex-smoking peers (i.e., quitting experiences and encouragement). Participants in the control group received 6 text messaging within 3 months on simple smoking cessation advice and reminders of follow-ups. All participants were followed at 1-, 2-, 3-, and 6-month. Blinding of participants and intervention deliverers was not possible, but statistic analysts were masked of the allocation. The primary outcome was biochemically validated abstinence at 6-month after intervention initiation. The main analysis followed the intention-to-treat principle. Poisson regression with robust variance was used to yield the risk ratio (RR) of intervention effect. Trial registry: ClinicalTrials.gov: NCT04909320. In the second phase, individual semi-structured interviews were conducted among a purposive sample of trial participants and ex-smoking peers to explore their experiences of participating in the trial. Data analyses followed the thematic analyses by Braun and Clarke. Results: In phase one, 1105 smokers were recruited and randomized into the intervention (n=553) and control (n=552) groups from June to October 2021. 78.4% of participants were male, and 54.2% intended to quit within 30 days. The intervention group showed a small but non-significant increase in the biochemically validated 7-day point prevalence abstinence at 6 months (8.9% vs. 6.7%; RR 1.32, 95% CI 0.88, 1.99; P=0.18) compared to control group. In phase two, 23 trial participants and 21 ex-smoking peers were interviewed. Both trial participants and ex-smoking peers appraised the benefits of participating in the study. The unmatching between the trial participants and ex-smoking peers might limit the supporting effect gained by trial participants. The trial participants’ low engagement in the intervention might cause supporting difficulty of ex-smoking peers, which might, in turn, lower the trial participants’ engagement. Several reasons for not engaging in the intervention and recommendations were identified for further intervention refinement. Conclusion: The trial has found that mHealth-based behavioral support by ex-smoking peers has increased smoking abstinence in community smokers but did not achieve statistical significance. Further studies are needed to explore the optimal engagement level and improve the intervention engagement. Studies on matching between the ex-smoking peers and participants, and ex-smoking peers’ supporting contents are further needed to enhance the ex-smoking peers’ supporting effect.