What factors are related to engagement with digital mental health interventions (DMHIs)? A meta-analysis of 117 trials.

Abstract

Identifying correlates of engagement with digital mental health interventions (DMHIs) can guide strategies to encourage therapy skill practice in targeted subgroups, supporting precision mental health. This three-level robust variance estimation meta-analysis examined correlates of DMHI engagement across 117 studies (1698 effect sizes; 279,791 participants), classifying engagement as initial uptake, study-specific usage, and intervention completion. Women engaged more than men in both unadjusted (Cohen's d = 0.34, 95% CI [0.06, 0.62]) and adjusted models (r_p = 0.05 [0.01, 0.09]). Other positive correlates were past mental health problems, guided versus self-guided modality delivery, therapeutic relationship, and positive expectancy. Meta-regressions indicated that completion was associated with older age, higher education, employment, and lower stress. Uptake was related to higher socioeconomic status, greater motivation, and lower symptom severity; usage was linked to human guidance and common mental disorders versus other conditions. Longer treatment duration amplified the effects of therapeutic relationship and guidance on engagement. Nearly half (48.7%) of studies had low risk of bias; minimal publication bias did not alter findings. Persuasive design strategies could enhance male engagement. Practitioners and policymakers should prioritize coaching that fosters productive relationships, pre-intervention materials to set realistic expectations, and methods to reduce the high risk of bias in future DMHI research.