An ML Model Predicts Engagement With A mHealth-Based RPM Program Among PWDs


Individuals who are enrolled in remote patient monitoring (RPM) programs commonly disengage over time. We developed a machine learning model that predicts risk of declining engagement (dropout) with an mHealth app (Glooko) that supports diabetes self-management.We trained a gradient boosting algorithm to predict risk of declining engagement in the next 28 days.

We selected 7134 people with diabetes (PWD) using meter and/or insulin pump devices who used the Glooko diabetes management app in a remote (outside of the clinic).

We used self-monitored blood glucose (SMBG), demographic, and app engagement data from the prior 28 days.Data was limited to 1/1/2019-7/1/2019 (40,349 total predictions).

Download to learn more.