Predicting Engagement Patterns With Connected Wearable Devices in a Health System: Survival Analysis

The rapid advancement and widespread adoption of wearable devices provide opportunities to collect longitudinal objective activity and health data and integrate the information directly into a patient's electronic health record (EHR). Patterns of engagement and factors associated with the use and nonuse of wearable devices are currently not well understood.

This study aimed to quantify the number of patients still engaged and using wearable devices at 1 year since each patient's first day of use across a cohort collected over 6 years. We then aimed to identify demographic and behavioral factors that statistically significantly predict the likelihood of staying engaged and using wearable devices within the same 1-year time span since first use.

We analyzed connected device data from a large, nonprofit academic medical center, which began to incorporate wearable device data into the EHR system in April 2015. We conducted a survival analysis to evaluate time to early disengagement among connected device users and identify factors associated with long-term (1 y) engagement in multivariable Cox proportional hazard regression models.

The analysis included 8616 patients (mean age 45, SD 14.36 y; median 21, IQR 34-55 y; men: n=4489, 52.1%; women: n=4126, 47.9%) with available connected device data (eg, step counts) from the EHR between 2015 and 2022. A total of 5870 (68.13%) patients were engaged with active connected devices in the EHR at 1 year. Multivariable Cox regression models indicated no statistically significant differences between gender groups and race categories. Younger age categories (18-34 y) and lower median daily step counts (<5000) were associated with statistically significant increased hazards for early disengagement at 1 year.

The ongoing development of new sensors and algorithms presents opportunities to expand the capabilities of wearable devices, making them even more integral to health care delivery. It is important to quantify and enhance engagement to maximize the benefits of this technology and inform future use of the technology to improve health outcomes.