"background": "District hospitals are the cornerstone of primary healthcare delivery, yet systematic, longitudinal evaluations of their clinical performance remain methodologically underdeveloped. This gap hinders evidence-based health systems management and resource allocation. ", "purpose and objectives": "This study aims to develop and apply a robust panel-data methodology to evaluate the clinical performance of district hospitals over an extended period, identifying systemic trends and institutional determinants of outcomes. ", "methodology": "We constructed a novel, national panel dataset from administrative health records. Clinical performance was measured using a composite index of mortality and avoidable adverse events. The relationship was estimated using a two-way fixed effects model: Y{it = \ + \ Xit + \ + \ +, where Yit is the outcome for hospital i in year t, Xit contains time-varying covariates, and \ and \ₜ are hospital and year fixed effects. Inference is based on cluster-robust standard errors. ", "findings": "A one-unit increase in the nurse-to-patient ratio was associated with a 0. 15 standard deviation improvement in the clinical performance index (95% CI: 0. 09, 0. 21). Performance trajectories exhibited significant convergence, with historically poorer-performing facilities showing the most marked improvement. ", "conclusion": "The panel-data approach provides a rigorous framework for isolating institutional performance from temporal shocks. Results demonstrate that sustained input investments, particularly in staffing, are critically linked to enhanced clinical outcomes at the district level. ", "recommendations": "Health policy should prioritise the stabilisation and growth of the professional nursing cohort. We recommend the institutionalisation of panel-data performance monitoring to enable targeted, equity-driven interventions and longitudinal accountability. ", "key words": "health systems evaluation, panel data, fixed effects, clinical outcomes, resource allocation, health equity", "contribution statement": "This paper provides a novel methodological framework and a new national
Ndlovu et al. (Mon,) studied this question.