{ "background": "Emergency care systems in low-resource settings face significant challenges in measuring clinical outcomes due to fragmented data and heterogeneous patient populations. Robust methodological frameworks for outcome assessment are lacking, hindering system evaluation and improvement.", "purpose and objectives": "This study aimed to develop and validate a novel Bayesian hierarchical model to measure clinical outcomes across diverse emergency care units, providing a methodological tool for system-level evaluation.", "methodology": "We analysed a multi-centre, retrospective cohort of patient encounters. The core model is a Bayesian hierarchical logistic regression: $\\text{logit}(P(y{ij}=1)) = \\alphaj + X{ij}\\beta$, where $y{ij}$ is the binary outcome for patient $i$ in unit $j$, $\\alphaj \\sim N(\\mu{\\alpha}, \\sigma^2{\\alpha})$ are unit-specific intercepts, and $X{ij}$ are patient-level covariates. Posterior distributions were estimated using Hamiltonian Monte Carlo.", "findings": "The model successfully quantified variation in risk-adjusted mortality across units, with unit-level intercepts ($\\alpha_j$) showing a posterior credible interval of [-2.1, 0.8] on the log-odds scale. A key finding was that over 30% of the variance in patient outcomes was attributable to unit-level systemic factors rather than patient characteristics alone.", "conclusion": "The Bayesian hierarchical model provides a robust, statistically coherent framework for comparing clinical outcomes in heterogeneous emergency care settings, accounting for both patient risk and systemic performance differences.", "recommendations": "Health system managers should adopt similar hierarchical modelling approaches for fairer unit comparisons. Future research should integrate this methodology with process metrics to identify specific drivers of outcome variation.", "key words": "Bayesian inference, health systems research, clinical outcomes, emergency medicine, hierarchical modelling, Uganda", "contribution statement": "This paper introduces a novel application of Bayesian hierarchical modelling for health system evaluation in low-resource emergency care, providing a