Optimizing mHealth Instant Messaging-Based Smoking Cessation Support: A Sequential, Multiple Assignment, Randomized Trial (SMART)

AbstractMobile health (mHealth) messaging has been used to enhance quitting. Optimal sequencing of more personalized and intensive interventions may increase abstinence for mHealth non-responders (continuing smokers). We aim to test the effectiveness and cost of an adaptive design intervention based on chat-based personalized instant messaging (PIM) versus that based on regular instant messaging (RIM; non-personalized) on validated abstinence. Sequential, multiple assignment randomized trial proactively recruited adult daily cigarette smokers in Hong Kong. At baseline, participants received brief cessation advice plus referral assistance to cessation services and were randomized to receive PIM (PIM group, n = 422) or RIM (RIM group, n = 422). At 1 month, PIM non-responders were further randomized (ratio 3:1) to receive either combined cessation interventions (CCI, including multi-media messages, nicotine replacement therapy sampling, incentive for service referral, phone counselling, family/peer group chat) or maintained PIM for 2 months. RIM non-responders were further randomized (ratio 1:3) to receive PIM or maintained RIM. Responders (quitters) in either group continued to receive the respective initial intervention. Bio-validated abstinence at 6 months by intention-to-treat. The 844 participants were mostly male (82.3%). At 1 month, 370 (87.7%) and 373 (88.4%) non-responded to PIM and RIM, respectively. Of non-responders, 273 (73.8%) received CCI and 91 (24.4%) received PIM. At 6 months, PIM group had non-significantly higher validated abstinence than RIM group (10.2% vs. 8.3%, risk ratio [RR] 1.23, 95%CI 0.80 to 1.88) at doubled cost (US$33,228.8 vs. 15,985.5). In non-responders, receiving CCI (vs. maintained PIM: 4.8% vs. 6.2%, RR 0.77, 95%CI 0.30 to 1.97) or PIM (vs. maintained RIM: 3.3% vs 5.7%, RR 0.58, 95%CI 0.17 to 1.95) did not increase validated abstinence. The PIM-based adaptive intervention did not significantly increase validated abstinence than that of non-personalized IM. Non-responders to PIM or RIM did not benefit from more intensive interventions. ClinicalTrials.gov Identifier: NCT03992742