Improvement and Maintenance of Clinical Outcomes in a Digital Mental Health Platform: Findings From a Longitudinal Observational Real-World Study (Preprint)

Abstract

BACKGROUND Digital mental health services are a growing employer health benefit that can improve access and remove barriers to mental health care. Stratified stepped care models, in particular, offer personalized care recommendations that can simultaneously offer clinically effective interventions while conserving resources. Nonetheless, clinical evaluation is needed to understand their benefit for mental health and use in a real-world setting. OBJECTIVE This study examined changes in clinical outcomes (i.e., depressive and anxiety symptoms, well-being) and evaluated the use of stratified stepped care among members of an employer-sponsored digital mental health benefit. METHODS In a large prospective study, we examined changes in depressive symptoms, anxiety symptoms, and well-being over 3 months in 509 participants (Mean age = 33.9 ± 8.7; 61.3% women, 34.4% men, 4.3% non-binary; 40.2% BIPOC) who were newly enrolled and engaged in care with an employer-sponsored digital mental health platform (Modern Health Inc., San Francisco, CA). We also investigated whether participants followed, underutilized, or overutilized services (i.e., therapy, coaching, or digital self-guided content) relative to the recommendations provided to them through a stratified, stepped care model. RESULTS Participants with elevated baseline symptoms exhibited significant improvement in depressive and anxiety symptoms and well-being across the study period (ps < .001), with the greatest average improvements observed in well-being (90% score increase), followed by depressive (37% score reduction) and anxiety symptoms (29% score reduction). Further, over half exhibited clinical improvement or recovery for depressive symptoms (65.8%), anxiety symptoms (59.2%), and low well-being (65.9%). Among participants with lower baseline symptoms, we found high rates of maintenance for low depressive (92.3%) and anxiety (86.2%) symptoms, and high well-being (90.2%). Two-thirds of participants (67.4%) utilized their recommended care, 16.9% stepped up their care beyond their initial recommendation, and 15.7% of participants underutilized care by not engaging with the highest level of care recommended to them. CONCLUSIONS Participants with elevated baseline depressive and/or anxiety symptoms improved their mental health significantly from baseline to follow-up, and those with no or mild symptoms successfully maintained their mental health at high rates. In addition, engagement patterns indicate that the stratified stepped care model was efficient in matching individuals with the most effective and least costly care, while also allowing them to self-determine their care and use combinations of services that best fit their needs. Overall, the results of this study support the clinical effectiveness of the platform for improving and preserving mental health, and support the utility and effectiveness of stratified stepped care models to improve access and utilization of digitally-delivered mental health services.