Exploring digital health user engagement: General app usage patterns from a clinical trial with the mLab App

The number of digital health applications has rapidly grown over recent years, but many of them are limited by sustainability and scalability. Paradata, the detailed interactions of a user with a piece of software, is straightforward to collect without disrupting the user-experience, and can provide a nuanced understanding of in-app user behavior. In this work we analyze paradata from the mLab App arm of a three-arm, multi-site randomized clinical trial (NYC and Chicago), in a sub-study that evaluates longitudinal, session-level interaction logs with multiple sessions per participant, (registered with Clinicaltrials.gov as NCT03803683), a mobile health application to facilitate at-home HIV testing in at-risk populations. We investigated application-level usage statistics to identify feature usage, as well as common navigational paths within the application. Temporal patterns were observed for login events and test-taking patterns. Despite being under-collected and underreported, paradata reveal feature gaps, guide targeted revisions for subsequent research and implementation, and deepen understanding of user behavior-enabling digital-health applications with lasting impact.