A Mobile Health Intervention for Patients With Depressive Symptoms: Protocol for an Economic Evaluation Alongside Two Randomized Trials in Brazil and Peru

BACKGROUND: Mobile health interventions provide significant strategies for improving access to health services, offering a potential solution to reduce the mental health treatment gap. Economic evaluation of this intervention is needed to help inform local mental health policy and program development. OBJECTIVE: This paper presents the protocol for an economic evaluation conducted alongside 2 randomized controlled trials (RCTs) to evaluate the cost-effectiveness of a psychological intervention delivered through a technological platform (CONEMO) to treat depressive symptoms in people with diabetes, hypertension, or both. METHODS: The economic evaluation uses a within-trial analysis to evaluate the incremental costs and health outcomes of CONEMO plus enhanced usual care in comparison with enhanced usual care from public health care system and societal perspectives. Participants are patients of the public health care services for hypertension, diabetes, or both conditions in São Paulo, Brazil (n=880) and Lima, Peru (n=432). Clinical effectiveness will be measured by reduction in depressive symptoms and gains in health-related quality of life. We will conduct cost-effectiveness and cost-utility analyses, providing estimates of the cost per at least 50% reduction in 9-item Patient Health Questionnaire scores, and cost per quality-adjusted life year gained. The measurement of clinical effectiveness and resource use will take place over baseline, 3-month follow-up, and 6-month follow-up in the intervention and control groups. We will use a mixed costing methodology (ie, a combination of top-down and bottom-up approaches) considering 4 cost categories: intervention (CONEMO related) costs, health care costs, patient and family costs, and productivity costs. We will collect unit costs from the RCTs and national administrative databases. The multinational economic evaluations will be fully split analyses with a multicountry costing approach. We will calculate incremental cost-effectiveness ratios and present 95% CIs from nonparametric bootstrapping (1000 replicates). We will perform deterministic and probabilistic sensitivity analyses. Finally, we will present cost-effectiveness acceptability curves to compare a range of possible cost-effectiveness thresholds. RESULTS: The economic evaluation project had its project charter in June 2018 and is expected to be completed in September 2021. The final results will be available in the second half of 2021. CONCLUSIONS: We e