Community Mental Health Services Task-Shifting to Traditional Birth Attendants and Perinatal Depression Detection in Rwandan Health Centres: A Cluster-Randomised Implementation Evaluation

Perinatal depression remains under-detected in Rwandan health centres, where midwife and nurse cadres are already saturated and traditional birth attendants (TBAs) retain trusted community access. The DHS extract comprised 187 country-survey observations from 36 African countries between 1990 and 2025, with a mean crude birth rate of 35.914 (SD 5.662), a minimum of 22.20 in South Africa and a maximum of 43.77 in Chad. The protocol specifies a two-arm parallel cluster design with health centres as the unit of randomisation, TBAs as the delivery agents, and the Edinburgh Postnatal Depression Scale administered at six weeks postpartum as the primary detection outcome. The binding constraint on TBA task-shifting is not TBA willingness but the absence of a defensible sampling and power architecture that accounts for the clustering of births within health-centre catchments and the heterogeneity of fertility across districts. The DHS crude birth rate distribution is used to specify catchment-level birth volumes, intra-cluster correlation assumptions and detectable effect sizes. The contribution is a transferable design template that couples a real, publicly documented fertility indicator to the power architecture of a cluster trial, so that implementation evaluations of community mental health task-shifting can be planned against observed demographic structure rather than assumed birth cohorts.