"background": "Evaluating the cost-effectiveness of community health centres is critical for resource allocation in low-resource settings, yet standard frequentist methods often struggle with sparse, hierarchical data and do not fully quantify uncertainty. ", "purpose and objectives": "This case study presents a novel Bayesian hierarchical model for cost-effectiveness analysis and demonstrates its application to assess the efficiency of a national network of primary care facilities. ", "methodology": "We developed a Bayesian cost-effectiveness model using a hierarchical structure to account for clustering at the regional and district levels. The core effectiveness measure was disability-adjusted life years averted. The model, (ij) = \ + \ ij + u{i + vij +, where ui \ N (0, \²region) and vij \ N (0, \²₃₈ₒₓₑ₈₂ₓ), explicitly modelled uncertainty in all parameters, with posterior distributions estimated via Markov chain Monte Carlo sampling. ", "findings": "The model successfully synthesised sparse data, providing robust probabilistic inferences. A key finding was a high degree of variability in cost-effectiveness between districts, with the posterior probability that the incremental cost-effectiveness ratio exceeded a willingness-to-pay threshold being 0. 78 for the central region. The 95% credible interval for the national average cost per DALY averted was wide, reflecting substantial heterogeneity. ", "conclusion": "The Bayesian hierarchical approach offers a superior methodological framework for economic evaluations in complex health systems, formally incorporating variability and uncertainty that are often overlooked. ", "recommendations": "Health economists and policy analysts should adopt Bayesian hierarchical modelling for cost-effectiveness analyses where data are hierarchically structured and sparse. Future research should focus on integrating prior evidence from similar settings to strengthen inferences. ", "key words": "Bayesian hierarchical model, cost-effectiveness analysis, health economics, primary health care, uncertainty quantification, sub-Saharan Africa", "
Tadesse et al. (Sun,) studied this question.