Implementation Fidelity in Medical Programmes in Ghana: Bayesian Hierarchical Estimation of Access Inequalities

Implementation fidelity is routinely invoked to explain why Ghanaian medical programmes underperform, yet the construct is rarely modelled as a hierarchical phenomenon in which fidelity is nested within facilities, districts and regions. The framework treats fidelity as a latent facility-level construct measured through multiple indicators, with partial pooling across administrative levels to separate genuine between-facility inequality from sampling noise. Using a variance-structure specification with between-cluster standard deviation of 6.0, residual standard deviation of 4.5 and intraclass correlation of 0.64, the design analysis characterises the operating characteristics of a before-after-control-impact estimator across 2000 replications per design cell. Power curves, minimum detectable effect curves, receiver operating characteristic behaviour and coefficient precision are reported for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The analysis indicates that 52 clusters are required to detect a 5 percentage point fidelity gap at 80 per cent power under the declared variance structure. The contribution is a transferable estimation strategy that reframes fidelity from a programme-level audit score into a distributional property of the access system, with consequences for how Ghanaian medical programmes are evaluated and how access inequalities are attributed.