mActive-Smoke: A Prospective Observational Study Using Mobile Health Tools to Assess the Association of Physical Activity With Smoking Urges

BackgroundEvidence that physical activity can curb smoking urges is limited in scope to acute effects and largely reliant on retrospective self-reported measures. Mobile health technologies offer novel mechanisms for capturing real-time data of behaviors in the natural environment. ObjectiveThis study aimed to explore this in a real-world longitudinal setting by leveraging mobile health tools to assess the association between objectively measured physical activity and concurrent smoking urges in a 12-week prospective observational study. MethodsWe enrolled 60 active smokers (≥3 cigarettes per day) and recorded baseline demographics, physical activity, and smoking behaviors using a Web-based questionnaire. Step counts were measured continuously using the Fitbit Charge HR. Participants reported instantaneous smoking urges via text message using a Likert scale ranging from 1 to 9. On study completion, participants reported follow-up smoking behaviors in an online exit survey. ResultsA total of 53 participants (aged 40 [SD 12] years, 57% [30/53] women, 49% [26/53] nonwhite) recorded at least 6 weeks of data and were thus included in the analysis. We recorded 15,365 urge messages throughout the study, with a mean of 290 (SD 62) messages per participant. Mean urge over the course of the study was positively associated with daily cigarette consumption at follow-up (Pearson r=.33; P=.02). No association existed between daily steps and mean daily urge (beta=−6.95×10−3 per 1000 steps; P=.30). Regression models of acute effects, however, did reveal modest inverse associations between steps within 30-, 60-, and 120-min time windows of a reported urge (beta=−.0191 per 100 steps, P