Open Source: Reflections From Over Ten Years of Treating Depression at Scale | StrongMinds

Open Source: Reflections From Over Ten Years of Treating Depression at Scale

After operating for more than 10 years and treating over 500,000 people for depression, it is fitting with StrongMinds’ mission of scaling depression treatment and the spirit of transparency with which we conduct all our work that we take this opportunity to share some of the most important lessons we have learned over the years. 

StrongMinds encourages other mental health organizations and mental health practitioners to provide feedback and share their own experiences in the same spirit of transparency. We hope that by encouraging and demonstrating transparency we can ultimately learn from one another and provide the best care possible to those most in need.

Developing Criteria for Volunteers

A principal goal for StrongMinds has always been to take care often provided by professionals —who are expensive and in short supply in some locations—and train and empower community members to deliver that care. Though we had initial trials in which lay providers led therapy sessions, the bulk of interpersonal therapy sessions were led by staff when we began operations in 2013. To realize our initial vision of extending therapy to more beneficiaries, we needed to recruit and train more volunteer facilitators to lead therapy sessions. Naturally, these included Village Health Teams (VHT)/Community Health Workers (CHW)—community members with established relationships that provide health information and referral services—and peer facilitators working mostly voluntarily. Informed by outcome data and experience, we developed criteria for these volunteers to ensure fidelity and quality therapy. Criteria included: the ability to read and write; being depression-free at the time of the group sessions (Determined through PHQ-9 screenings); possessing previous community volunteer experience in a health or development capacity; completing a reference check; being of 18 years of age or older; and residing in the community they are serving. Along with other changes, these updated volunteer criteria contributed to better client outcomes: the average depressive symptom score reduced from 3.00 in 2022 to 1.98 in 2023, as measured by the PHQ-9.

Treating Depression Through Partnerships

To continue to grow, StrongMinds began working with partner organizations and governments. Treating depression through partners came with its own set of challenges and learnings. For example, to ensure the same results and quality treatment as we had been providing through our staff and staff-trained volunteers, we needed to directly supervise partner training sessions for volunteers. We also developed specific training manuals for partner training sessions. It was also necessary to strongly emphasize the importance of privacy to maintain high standards, as we found some partners photographed clients during treatment, which negatively affected their experience and outcomes.

Treating Adolescents

When StrongMinds began, our focus was on treating depression in women as they have the highest need and are least likely to have access to care. As adolescent girls also have high rates of depression, we expanded care into this demographic as well. 

Through a 2018 partnership with BRAC, we piloted the treatment of adolescents through BRAC’s ELA program. This marked the first time that StrongMinds treated adolescent populations, delivered therapy through an NGO partner, and relied on youth mentors to facilitate groups. We observed multiple areas for improvement regarding the adolescent program. Though we continue to provide treatment for girls who left school, one of our first learnings was that it was important to provide treatment to girls in school and girls out of school separately because of how different their life experiences were from one another. Grouping these two populations together also created scheduling challenges and complicated supervision.

After the BRAC partnership, StrongMinds hired a human-centered design firm, which studied the entire adolescent program from a user perspective. This led to multiple changes in the program, including: the implementation of emotion cards and other visual aids to assist different types of learners; the introduction of icebreakers to create comfortable atmospheres; and the use of journaling to help engage clients. We determined that IPT-G-trained teachers and Village Health Technicians (part of the VCT) were more effective in facilitating adolescent therapy groups than youth.

We also learned the importance of educating parents, teachers, and school administrators about mental health to help reinforce the healthy behaviors learned in therapy. These changes contributed to a 39% decrease in student absence from therapy, reaching 89% attendance in 2023.

Implementing Teletherapy

During the COVID-19 pandemic, StrongMinds implemented teletherapy to provide depression treatment during lockdowns. Despite achieving short-term success (an external evaluation in 2022 showed our teletherapy clients experienced the same average symptom reduction as in-person clients, 11 points on the PHQ-9 scale, though entered and left treatment with slightly higher depressive symptom burdens), we discovered that teletherapy was not financially efficient in its implementation. The average cost-per-patient for teletherapy was three times higher than other delivery channels due to the smaller group sizes needed for therapy, the necessity of numerous radio ads to raise awareness about the program, and the expenses linked to teleconferencing. As a result, StrongMinds paused teletherapy until more research could be conducted to improve cost-effectiveness.

Shortening the Length of Treatment Cycles

One of the most significant changes we made over time dealt with the length of treatment cycles. Initially, our treatment model took 16 weeks to complete. Our Mental Health Facilitators observed that clients became demotivated or unavailable to attend sessions after the fourth or fifth session, which resulted in poor therapy attendance. We concluded that it would be more efficient to use the time spent on the later poorly attended sessions to initiate additional treatment groups.

To make this transition while maintaining quality, we needed to shorten our treatment model through rigorous experimentation and testing. Before implementing any changes, we collected client feedback and analyzed data, which helped us learn that high attendance requirements can interfere with income-producing work and childcare responsibilities. Over ten years, we gradually reduced our treatment cycle and implemented related changes—including grouping clients with similar depression triggers—and saw a significant increase in attendance. 

Our most recent reduction in cycle length was from eight weeks to six weeks. In partnership with researchers from New York University Langone Health, we conducted a randomized controlled trial (RCT) comparing this six-week therapy model to our previous eight-week model. The study showed this model yields excellent results. In fact, clients in the six-week program had larger improvements in their depression scores, by two points on average, than the eight-week model (15.3 vs. 13.3, measured by PHQ-9). In addition to meeting client needs, this model is also highly cost-effective and scalable.

Future Learnings

Though StrongMinds has learned a tremendous amount in its first ten years of work, we are not done learning. We have expanded our work to treat anyone who needs care, which includes men. We are also treating incarcerated individuals, as their need for therapy is high. The way we deliver care is also changing as we are now working with governments to train facilitators and lead sessions. As we learn from this work, we will continue to share our data, as well as our challenges so that the global mental health community can close the mental health treatment gap.