Rates of attrition and engagement in randomized controlled trials of mindfulness apps: Systematic review and meta-analysis

Abstract

Randomized controlled trials (RCT) have established the efficacy of mindfulness apps for numerous mental health and behavioral outcomes. However, studies of this kind are characterized by significant problems with attrition and engagement, making it challenging to interpret these findings and realize the full potential of mindfulness app-based solutions. This systematic review and meta-analysis aimed to examine rates of attrition from, and engagement in, mindfulness apps in 70 RCTs located through four databases. The weighted meta-analytic attrition rate from 9258 participants allocated to a mindfulness app was 24.7%. This estimate was higher (38.7%) when restricting analyses to larger studies (n ≥ 100 per condition) and for specific target samples (e.g., general population of adults). Studies that offered participants monetary compensation and required in-person enrolment methods were associated with the lowest attrition rates. Attrition was higher in mindfulness app conditions relative to waitlists, but was not different to placebo or active apps. Although engagement data were inadequately reported, qualitative synthesis highlighted clear problems with sustained user engagement with mindfulness apps. Findings have implications for powering, designing and appraising research trials of mindfulness apps, as well as the development of potentially effective retention and engagement strategies worthy of evaluation in future work.

Keywords: Attrition; Digital health; Meta-analysis; Mindfulness; Randomized controlled trials; Systematic review.