{ "background": "Public health surveillance systems are critical for disease control, yet their methodological evaluation, particularly regarding efficiency and predictive capacity, remains underdeveloped in many low-resource settings. Existing approaches often lack robust frameworks for quantifying uncertainty and integrating heterogeneous data streams.", "purpose and objectives": "This study aimed to develop and apply a novel Bayesian hierarchical model to methodologically evaluate the efficiency of national public health surveillance and to identify key leverage points for its optimisation.", "methodology": "We constructed a Bayesian hierarchical model $y{it} \\sim \\text{Poisson}(\\lambda{it})$, $\\log(\\lambda{it}) = \\alpha + \\beta X{it} + ui + vt$, where $ui$ and $vt$ are structured random effects for region and time. The model integrated longitudinal surveillance performance data, resource allocation metrics, and outcome indicators. Efficiency was measured via a stochastic frontier analysis embedded within the Bayesian framework. Model parameters were estimated using Hamiltonian Monte Carlo.", "findings": "The model identified that a 10% increase in the timeliness of case reporting was associated with a posterior probability of 0.92 for a 4.2% to 7.1% gain in overall system efficiency. Substantial regional heterogeneity was observed, with the random effects $u_i$ indicating that infrastructural factors accounted for approximately 30% of the variance in performance.", "conclusion": "The Bayesian hierarchical model provides a robust methodological tool for the quantitative evaluation of surveillance systems, demonstrating that efficiency gains are achievable through targeted improvements in data timeliness and by addressing regional disparities.", "recommendations": "Surveillance strengthening programmes should prioritise interventions that reduce reporting delays. Resource allocation should be informed by sub-national efficiency analyses to address inequities. The adopted modelling framework should be incorporated into routine system evaluations.", "key words": "Bayesian inference, health systems strengthening, stochastic frontier analysis, health metrics, predictive modelling", "contribution statement": "This paper provides