In-the-wild data collection with digital apps and wearable devices: Insights from a longitudinal study on burnout...

Long-term monitoring via wearable devices is vital for mental health research, yet success depends on high participant adherence. We analyzed adherence and retention in a 9-month longitudinal study focused on work-related burnout detection, identifying key socio-demographic predictors of participants' engagement.

We conducted an observational study with N = 239 office and production workers. Participants wore devices for physiological tracking and completed monthly Shirom -Melamed Burnout Measure (SMBM) assessments. Adherence and retention were evaluated using beta regression and time-to-event analysis against socio-demographic factors.

We observed a median data collection adherence of 61.9%. Higher adherence was positively associated with older age, not being in a stable union, and holding a mid-level job position, while working in a production site was negatively associated with adherence. Median number of days into the study was equal to 242. We found an association of increasing age and work level with higher retention in the study, while working solely in a production site with lower retention. In our cohort, SMBM scores remained stable around a mean value of 3, showing an intra-class coefficient (ICC) of 0.71, with high between-person variation and low within-person changes over time.

Our analysis has identified key factors to improve adherence in studies involving wearable devices. We emphasize the importance of a truly user-centered design to improve participants' adherence and engagement. We recommend implementing automatic adherence reminders, offering alternative or complementary wearable devices, optimizing data synchronization procedures, and extending the duration of data collection for studies involving burnout changes.