Patient Engagement in a Virtual Cardiac Rehabilitation Platform: A secondary analysis of the ERA-Nudge trial (Preprint)

Background: Mobile health interventions have the potential to motivate healthy behaviors. Optimal strategies for engaging patients in cardiac rehabilitation (CR) in mobile health interventions remain unclear. Objective: We compare minutes in app by whether patients received text message and app notification nudges or app notification nudges alone. Methods: We analyzed data from the ERA-Nudge study, a pragmatic randomized trial conducted at an urban safety-net hospital testing nudges (short messages to elicit behavior change designed using behavioral economics) to enhance CR adherence and log healthy activity. Trial patients were given access to a mobile application (app). Nudges were initially sent as app notifications. Later in the study, patients received nudges as both app notifications and text messages. We compared time in the CR app among patients receiving nudges as app notifications only versus app notifications and text messages. The trial enrolled patients who could engage with messages in either English or Spanish at an urban safety net hospital with a qualifying CR diagnosis and smartphone from May 2019 to May 2023. We describe minutes in app by type of nudge delivery and compare the two approaches using unadjusted and multivariate negative binomial regression over the 12 weeks of CR. Results: In total, 223 patients received nudges: 74 app notification only, 149 app and text. Patients receiving only app nudges spent a mean of 167167 minutes in the app and patients receiving text messages and app nudges spent 245306 minutes in the app (Ratio 1.5, 95% CI 1.1 to 2.0; unadjusted p = 0.02). After adjusting for age category, sex, primary language, CR location (hospital vs. home), and highest education achieved, patients receiving nudges as text messages and app notifications spent approximately 1.8 times as many minutes in the app compared to those getting nudges as app notifications (95% CI: 1.2 to 2.5, p = 0.002). Conclusions: Among an urban safety net population, the addition of text message nudges was associated with greater time in app compared to nudges delivered as app notifications only. These findings suggest that text messages may result in greater patient engagement for mobile health interventions.