Designing a Voice-Based Treatment Module for Treating Perinatal Depression in Rural Kenya

Abstract

Depression is the largest contributor to the worldwide disease burden of mental, neurological and substance use disorders. Addressing perinatal depression is a particular challenge in low- and middle-income countries where screening and treatment is not readily available. A potential solution to this treatment gap is shifting treatment from more skilled to less skilled providers. A treatment model using lay counselors in Pakistan demonstrated large reductions in depression rates but is challenging to implement at scale. This study investigated how mobile phones could be used to deliver treatment sessions remotely through an automated voice-based service, thus potentially easing the burden of implementing the program at scale in rural and remote regions. Working in Bungoma County, Kenya, we followed a Human Centered Design approach to adapting and testing a session of an efficacious treatment program developed in Pakistan called the Thinking Healthy Program. We conducted two rounds of testing with community health workers. Participants were randomly assigned to complete the session in person with a single facilitator or as part of a group, or remotely via a mobile phone with or without live operator support. Comprehension of session content was high and not significantly different across in person and phone conditions. Phone implementation was feasible and acceptable to participants, though refinement is needed. Results suggest that automated phone administration could be a viable method to deliver session content. Further testing should assess therapeutic benefits of such as system compared to standard, in person delivery.