Feature-level analysis of a smoking cessation smartphone app that uses a positive psychology approach (Preprint)

BACKGROUND Smoking cessation smartphone apps have emerged as highly accessible tools recommended by clinical guidelines to support smoking cessation efforts. It is unknown how specific app features contribute to user engagement over time and relate to smoking outcomes. OBJECTIVE To provide a feature-level analysis of the “Smiling instead of Smoking” app (Version 2; SiS2) and to link feature use to subsequent smoking cessation. METHODS Nondaily smokers (n=100) were asked to use the SiS2 app for a period of 49 days (1 week pre-quit, 6 weeks post-quit). Participants self-reported 30-day point-prevalence abstinence (PPA) at the end of this period and at 6-month follow-up. The SiS2 app engaged app users in both positive psychology content designed to maintain their positive affect, and traditional behavioral smoking cessation content to guide their quit attempt. The app sent push notifications to prompt participants to complete prescribed content (i.e., Happiness Exercise every day; Behavioral Challenge to use the app’s smoking cessation tools on 15 out of 49 days). Actions that participants took within the app were timestamped by the app and recorded. RESULTS Participants used the app on 24.7±13.8 days out of the 49 prescribed days, interacting with the happiness content on more days than the smoking content (23.8±13.8 vs. 17.8±10.3 days, t(99)=9.28, p<.0001). Prescribed content was completed frequently (45% of Happiness Exercises; 57% of Behavioral Challenges); ad libitum tools were used on ≤7 days. Most participants used each ad libitum smoking cessation tool within the app at least once, with higher use of personalized content (i.e., ≥92% used Strategies, Cigarette Log, Smoke Alarms, and Personal Reasons) than purely didactic content (i.e., 79% viewed Benefits of Quitting Smoking). The number of days participants used the app significantly predicted 30-day PPA at end of treatment (1.05 [1.02,1.09], p=0.002) and 6-month follow-up (1.04 [1.008,1.07], p=0.01). When looking at the specific content used, analyses showed that the number of days participants engaged with the happiness content significantly predicted smoking abstinence at both end of treatment (1.05 [1.02,1.08], p=0.002) and 6-month follow-up (1.04 [1.007,1.07], p=0.02). This effect was not significant for the number of days participants engaged with the smoking cessation content of the SiS2 app, either at end of treatment (1.04 [0.996,1.08], p=0.08) or 6-month follow-up (1.02 [0.98,1.06], p=0.29). CONCLUSIONS Greater SiS2 app usage predicted greater odds of self-reported 30-day PPA at both treatment end and long-term, suggesting therapeutic benefit of the SiS2 app. Positive psychology content and prescriptive clarity may promote sustained app engagement over time. CLINICALTRIAL ClinicalTrials.gov Identifier: NCT03951766