A Mixed-Methods Evaluation of Empower@Home: A Novel Digital Mental Health Intervention for Depression Among Older Adults

Background and Purpose: Digital mental health interventions can enhance access to treatment, yet few are designed specifically for depression in older adults. Empower@Home is an innovative internet-based cognitive behavioral therapy (iCBT) program tailored to reduce depression among this demographic, developed in collaboration with community agencies and stakeholders following user-centered design principles. It incorporates nine online lessons, a physical workbook, and optional support from lay coaches. This presentation consolidates findings from various study phases to evaluate the intervention's feasibility and efficacy.

Methods: Participants (aged 50+ with mild depressive symptoms, PHQ-9 ≥ 5) were recruited through community referrals and research registries, excluding those with probable dementia, elevated suicide risk, or current substance use disorders. Device ownership and internet access were provided free of charge as needed. The study as a whole included uncontrolled and randomized controlled trials with waitlist attention control. This analysis pooled data from all study phases and combined qualitative and quantitative data to present a comprehensive evaluation of the intervention. We analyzed data from 224 participants using intention-to-treat, assessing changes in the primary outcome of depressive symptoms (PHQ-9), and secondary outcomes of anxiety (GAD-7) and loneliness (PROMIS Social Isolation-8a) from pre-treatment to post-treatment. CBT skill acquisition, behavioral activation, and autonomy were explored as potential mediators. A random subset (N=145) participated in qualitative interviews. Paired-sample t-tests were conducted to examine within-group differences in primary and secondary outcomes. Thematic analysis was conducted to extract themes from the qualitative interviews.

Results: The average participant age was 67 (SD=8.2), with over half reporting an annual income under $50,000. Treatment completion was high (90%, N=202), with 93% likely to recommend the program. Significant reductions were observed in depressive symptoms (Cohen’s d = .90), anxiety (Cohen’s d = .67), and loneliness (Cohen’s d = .35) at post-test (all p<.001). Effects were most pronounced in those with moderate baseline depression (Cohen’s d = 1.40). Causal mediation analysis highlighted the role of CBT skills, behavioral activation, and improved autonomy as change mechanisms. Qualitative findings emphasized the program's impact on mental health literacy, empowerment, and sustained engagement. Themes related to engagement included: (1) a structured, skills-based, and self-paced approach, (2) supportive accountability from weekly coaching, (3) usability and software, (4) narrative engagement dynamics, (5) enhanced learning through multimodal instruction, and (6) tailoring challenges in iCBT.

Conclusions and Implications: Lay coach-supported iCBT is acceptable and can reduce depression among older adults. Important lessons learned and implications for designing mental health interventions for older adults will be discussed, as well as the potential integration into aging services to enhance public health impact.