Cost-effectiveness of psychological intervention within services for depression delivered by primary care workers in Nepal: economic evaluation of a randomized control trial

Background. Integrating services for depression into primary care is key to reducing the treatment gap in low- and middle-income countries. We examined the value of providing the Healthy Activity Programme (HAP), a behavioral activation psychological intervention, within services for depression delivered by primary care workers in Chitwan, Nepal using data from the Programme for Improving Mental Health Care. Methods. People diagnosed with depression were randomized to receive either standard treatment (ST), comprised of psychoeducation, antidepressant medication, and home-based follow up, or standard treatment plus psychological intervention (T + P). We estimated incremental costs and health effects of T + P compared to ST, with quality adjusted life years (QALYs) and depression symptom scores over 12 months as health effects. Nonparametric uncertainty analysis provided confidence intervals around each incremental effectiveness ratio (ICER); results are presented in 2020 international dollars. Results. Sixty participants received ST and 60 received T + P. Implementation costs (ST = $329, T + P = $617) were substantially higher than service delivery costs (ST = $18.7, T + P = $22.4) per participant. ST and T + P participants accrued 46.5 and 49.4 QALYs, respectively. The ICERs for T + P relative to ST were $4422 per QALY gained (95% confidence interval: $2484 to $9550) - slightly above the highly cost-effective threshold - and -$53.21 (95% confidence interval: -$105.8 to -$30.2) per unit change on the Patient Health Questionnaire. Conclusion. Providing HAP within integrated depression services in Chitwan was cost-effective, if not highly cost-effective. Efforts to scale up integrated services in Nepal and similar contexts should consider including evidence-based psychological interventions as a part of cost-effective mental healthcare for depression.