Methodological Evaluation and Yield Optimisation of Public Health Surveillance Systems in Tanzania: A Multilevel Regression Analysis

{ "background": "Public health surveillance systems in sub-Saharan Africa are critical for disease control, yet their methodological rigour and yield—the proportion of true cases detected—are often suboptimal. Systematic evaluations quantifying the impact of specific interventions on surveillance yield are lacking.", "purpose and objectives": "This study aimed to methodologically evaluate an enhanced surveillance intervention and quantify its effect on case detection yield within the Tanzanian Integrated Disease Surveillance and Response system. The primary objective was to measure yield improvement across sentinel facilities.", "methodology": "We conducted a quasi-experimental intervention study across four regions. The intervention comprised targeted training, streamlined reporting protocols, and integrated data feedback loops. Performance was assessed using a two-level hierarchical model: $\\logit(\\pi{ij}) = \\beta0 + \\beta1 X{ij} + uj + e{ij}$, where $\\pi{ij}$ is the yield for facility $i$ in district $j$, $X{ij}$ denotes intervention status, and $u_j$ are district-level random effects. Inference was based on robust standard errors.", "findings": "The intervention was associated with a significant increase in mean surveillance yield. Adjusted analysis showed a 17.2 percentage point improvement (95% CI: 12.5 to 21.9) in intervention facilities compared to controls. The multilevel model indicated significant residual variance attributable to district-level factors.", "conclusion": "Methodological enhancements focusing on training and data feedback can substantially improve the yield of public health surveillance. The success of the intervention was moderated by contextual district-level variables.", "recommendations": "National programmes should adopt structured, data-driven feedback mechanisms as a core component of surveillance strengthening. Future interventions must account for and mitigate higher-level administrative heterogeneities to ensure equitable improvements.", "key words": "surveillance evaluation, yield optimisation, multilevel modelling, health systems strengthening, sub-Saharan Africa", "contribution statement": "This study provides a novel methodological framework for