Engagement With a Mobile Chat-Based Intervention for Smoking Cessation

Importance Determining how individuals engage with digital health interventions over time is crucial to understand and optimize intervention outcomes. Objective To identify the engagement trajectories with a mobile chat-based smoking cessation intervention and examine its association with biochemically validated abstinence. Design, Setting, and Participants A secondary analysis of a pragmatic, cluster randomized clinical trial conducted in Hong Kong with 6-month follow-up. From June 18 to September 30, 2017, 624 adult daily smokers were recruited from 34 community sites randomized to the intervention group. Data were analyzed from March 6 to October 30, 2023. Intervention Chat-based cessation support delivered by a live counselor via a mobile instant messaging app for 3 months from baseline. Main Outcomes and Measures Group-based trajectory modeling was used to identify engagement trajectories using the participants’ weekly responses to the messages from the counselor over the 3-month intervention period. The outcome measures were biochemically validated tobacco abstinence at 3-month (end of treatment) and 6-month follow-ups. Covariates included sex, age, educational level, nicotine dependence, past quit attempt, and intention to quit at baseline. Results Of 624 participants included in the analysis, 479 were male (76.8%), and the mean (SD) age was 42.1 (16.2) years. Four distinct engagement trajectories were identified: low engagement group (447 71.6%), where participants maintained very low engagement throughout; rapid-declining group (86 13.8%), where participants began with moderate engagement and rapidly decreased to a low level; gradual-declining group (58 9.3%), where participants had high initial engagement and gradually decreased to a moderate level; and high engagement group (58 5.3%), where participants maintained high engagement throughout. Compared with the low engagement group, the 6-month validated abstinence rates were significantly higher in the rapid-declining group (adjusted relative risk ARR, 3.30; 95% CI, 1.39-7.81), gradual-declining group (ARR, 5.17; 95% CI, 2.21-12.11), and high engagement group (ARR, 4.98; 95% CI, 1.82-13.60). The corresponding ARRs (95% CI) of 3-month validated abstinence were 4.03 (95% CI, 1.53-10.59), 5.25 (95% CI, 1.98-13.88), and 9.23 (95% CI, 3.29-25.86). Conclusions and Relevance The findings of this study suggest that higher levels of engagement with the chat-based smoking cessation intervention were associated with greater biochemically validated tobacco abstinence. Improving engagement with digital interventions may increase intervention benefits. Trial Registration ClinicalTrials.gov Identifier: NCT03182790