Feasibility of an interactive voice response system for monitoring depressive symptoms in a lower-middle income Latin American country

Abstract

Background: Innovative, scalable solutions are needed to address the vast unmet need for mental health care in low- and middle-income countries (LMICs). Methods: We conducted a feasibility study of a 14-week automated telephonic interactive voice response (IVR) depression self-care service among Bolivian primary care patients with at least moderately severe depressive symptoms. We analyzed IVR call completion rates, the reliability and validity of IVR-collected data, and participant satisfaction. Results: Of the 32 participants, the majority were women (78% or 25/32) and non-indigenous (75% or 24/32). Participants had moderate depressive symptoms at baseline (PHQ-8 score mean 13.3, SD=3.5) and reported good or fair general health status (88% or 28/32). Fifty-four percent of weekly IVR calls (approximately 7 out of 13 active call-weeks) were completed. Neither PHQ-8 scores nor IVR call completion differed significantly by ethnicity, education, self-reported depression diagnosis, self-reported overall health, number of chronic conditions, or health literacy. The reliability for IVR-collected PHQ-8 scores was good (Cronbach's alpha=0.83). Virtually every participant (97%) was "mostly" or "very" satisfied with the program. Many described the program as beneficial for their mood and self-care, albeit limited by some technological difficulties and the lack of human interaction. Conclusion: Findings suggest that IVR could feasibly be used to provide monitoring and self-care education to depressed patients in Bolivia. An expanded stepped-care service offering contact with lay health workers for more depressed individuals and expanded mHealth content may foster greater patient engagement and enhance its therapeutic value while remaining cost-effective. Trial registration ISRCTN ISRCTN 18403214. Registered 14 September 2016. Retrospectively registered