Predictors of Engagement in Multiple Modalities of Digital Mental Health Treatments: Longitudinal Study

Technology-enhanced mental health platforms may serve as a pathway to accessible and scalable mental health care; specifically, those that leverage stepped care models have the potential to address many barriers to patient care, including low mental health literacy, mental health provider shortages, perceived acceptability of care, and equitable access to evidence-based treatment. Driving meaningful engagement in care through these platforms remains a challenge.

This study aimed to examine predictors of engagement in self-directed digital mental health services offered as part of an employer-based mental health benefit that uses a technology-enabled care platform.

Using a prospective, longitudinal design, we examined usage data from employees who had access to an employer-sponsored mental health care benefit. Participants had access to a digital library of mental health resources, which they could use at any time, including daily exercises, interactive programs, podcasts, and mindfulness exercises. Coaching and teletherapy were also available to. The outcome was engagement with the self-directed digital mental health resources, measured by the number of interactions. Poisson regression models included sociodemographic characteristics, patient activation, mental health literacy, well-being, PHQ-9 and GAD-7 scores at baseline, primary concern for engaging in treatment, and the use of coaching or teletherapy sessions.

In total 950 individuals enrolled in the study, with 38% using any self-directed digital mental health resources. Approximately 44% of the sample did not use the app during the study period. Those using both self-directed digital and 1:1 modalities made up about one-quarter of the sample (235/950, 24.7%). Those using only coaching or therapy (170/950, 17.9%) and those using only self-directed digital mental health resources (126/950, 13.3%) make up the rest. At baseline, these groups statistically significantly differed on age, PHQ-9, GAD-7, MHLS, and primary concern. Receipt of coaching and teletherapy was associated with the number of self-directed digital mental health resources interactions in adjusted Poisson regression modeling. Use of any coach visit was associated with 82% (rate ratio [RR] 1.82, 95% CI 1.63-2.03) more self-directed digital mental health resource interactions while use of any teletherapy session was associated with 80% (RR 1.80, 95% CI 1.55-2.07) more digital mental health resources interactions (both P<.001). Each additional year of age was associated with increased digital mental health resources interactions (RR 1.04, 95% CI (1.03-1.05), and women had 23% more self-directed digital resources interactions than men (RR 1.23, 95% CI 1.09-1.39).

Our key finding was that the use of coaching or teletherapy was associated with increased self-directed digital mental health resource use. Higher self-directed digital resource engagement among those receiving coaching or therapy may be a result of provider encouragement. On the other hand, when a participant engages with 1 modality in the platform, they may be more likely to begin engaging with others, becoming "super users" of all resources.