Implementation Fidelity in Community Mental Health Services Programmes in Ghana: Bayesian Hierarchical Estimation of Access Inequalities

Implementation fidelity—the degree to which community mental health services are delivered as designed—is rarely measured in ways that reveal how access inequalities accumulate across districts. Because no primary fidelity data were available for this analysis, the framework is specified as a methodological proposal and its properties are examined through the published distribution of a related access indicator: the DHS Program's 'Discontinuation due to access/availability' measure, drawn from 74 country-survey observations across 28 African countries between 1992 and 2025 (mean 1.712, standard deviation 1.37). Fidelity should be modelled as a hierarchical moderator rather than a fixed programme attribute, and that access inequalities estimated without fidelity adjustment are biased toward underestimating district-level disparities. The proposed estimator is presented with its identification assumptions, prior sensitivity, and decision-theoretic implications for Ghana's district mental health teams. The contribution is a transferable estimation strategy that makes fidelity an explicit object of inference in community mental health services research.