Electronic Thesis and Dissertation Repository

Thesis Format



Master of Science




Mitchell, Marc S.


BACKGROUND: Mobile health (mHealth) apps may help promote physical activity and other health behaviours among office-based workers. Low app engagement, however, leading to little or no effect is typical. OBJECTIVE: To examine engagement with a rewards-based mHealth app and identify factors influencing engagement. METHODS: A one-year observational study was conducted with Canadian and U.S. users of the Sprout at Work app (N=2253; Female: 35.7%; Age: 39.3 years). Kaplan-Meier survival curves were used to examine engagement patterns from a ‘multiple-lives’ perspective (i.e., time to first disengagement, re-engagement, second disengagement). Regression models were used to identify factors influencing engagement. RESULTS: After one month of app use, 51.2% of participants disengaged. Nine out of ten did not re-engage. Risk of first disengagement was highest for 56-75 year-old participants (44%-106% higher), while rewards worth $10 per month lowered this risk (46% lower). CONCLUSION: Findings may help stakeholders address persistent low app engagement moving forward.

Summary for Lay Audience

Health promoting interventions delivered through mobile apps have increased in popularity as they are easily accessible and scalable. Although, in order to be effective, users must remain engaged with the intervention to achieve their desired health goals and adopt long-term behaviour change. This study examined engagement patterns over one year in 2253 users of the Sprout at Work app, a multicomponent app that encourages and rewards physical activity and other well-being behaviours. User activity with the app was examined to determine critical time points of user disengagement, re-engagement, and second disengagement. Furthermore, we explored if specific characteristics influenced users’ engagement with the well-being platform. User disengagement was highest during the first few weeks of app usage with only a small proportion of users re-engaging. Risk of disengagement was greatest for older adults and for those who were offered rewards at an inconsistent rate. On the contrary, risk of disengagement was lowered for users who were offered a financial reward of $30 per quarter. The only factor which influenced the likelihood of a user re-engaging was the duration of their initial engagement period with the app. The results of our study may be informative to future intervention developers looking to retain users and enhance their mobile heath platform. More research is needed to determine the optimal combination of app features that elicits the greatest engagement response.