{ "background": "The operational integrity of centralised water treatment infrastructure is critical for public health, yet systematic, longitudinal evaluations of intervention efficacy in sub-Saharan contexts are scarce. Existing assessments often lack robust counterfactual frameworks to isolate the causal effect of infrastructure upgrades from secular trends.", "purpose and objectives": "This Data Descriptor presents a structured methodological framework and associated dataset designed to quantify the causal impact of technical and managerial interventions on facility performance and associated public health risks. The primary objective is to provide a replicable model for engineering diagnostics and policy evaluation.", "methodology": "A quasi-experimental difference-in-differences (DiD) model is employed, leveraging staggered implementation of interventions across facilities. The core estimating equation is $Y{it} = \\alpha + \\beta (Treatment{it}) + \\gammai + \\deltat + \\epsilon{it}$, where $Y{it}$ is a composite risk index. Inference is based on cluster-robust standard errors at the facility level. Data comprise engineering performance indicators, maintenance logs, and water quality parameters, compiled from regulatory audits and direct monitoring.", "findings": "The methodological application indicates a significant reduction in the composite risk index for treated facilities relative to controls. The average treatment effect on the treated (ATT) is estimated at -0.18 units (95% CI: -0.25, -0.11), corresponding to a 22% reduction in relative risk score. Diagnostic checks support the parallel trends assumption pre-intervention.", "conclusion": "The DiD model provides a statistically robust framework for evaluating engineering interventions in water treatment, isolating causal effects from confounding temporal and facility-specific factors. The accompanying dataset offers a validated template for longitudinal infrastructure assessment.", "recommendations": "Adopt the presented DiD methodology for future programme evaluations within the water sector. Regulatory bodies should mandate the systematic collection of the specified panel data to enable similar causal analyses for evidence-based asset management.", "key words": "causal inference, infrastructure diagnostics, panel data