Abstract
{ "background": "Public health surveillance systems are critical for disease control, yet their operational reliability in low-resource settings is often assumed rather than rigorously measured. Methodological evaluations of these systems, particularly using experimental designs, are scarce.", "purpose and objectives": "This study aimed to methodologically evaluate the reliability of Rwanda's integrated disease surveillance system by quantifying its sensitivity and positive predictive value (PPV) for detecting priority reportable conditions through a randomised field trial.", "methodology": "We conducted a parallel-group, cluster-randomised field trial across 60 health facilities. Standardised, scripted case scenarios representing febrile illness, acute watery diarrhoea, and suspected measles were introduced into the surveillance workflow. The primary outcome was system sensitivity, analysed using a generalised linear mixed model: $logit(P(Y{ij}=1)) = \\beta0 + \\beta1 X{ij} + uj$, where $Y{ij}$ is detection, $X{ij}$ represents scenario type, and $uj$ is a random intercept for facility $j$. Robust standard errors were used for inference.", "findings": "Overall system sensitivity for detecting introduced cases was 78% (95% CI: 72 to 84). Sensitivity varied significantly by disease scenario, being highest for suspected measles (92%) and lowest for acute watery diarrhoea (65%). The PPV of signals generated by the system was 86% (95% CI: 81 to 91).", "conclusion": "The surveillance system demonstrated moderate to high reliability, though performance was condition-specific, revealing a critical vulnerability in gastrointestinal disease reporting. The randomised scenario method proved feasible for rigorous, operational evaluation.", "recommendations": "Programme managers should implement targeted refresher training for frontline health workers on case definitions for diarrhoeal diseases. The scenario-based evaluation methodology should be integrated into routine supervisory activities to provide continuous reliability metrics.", "key words": "surveillance evaluation, health information systems, health systems research, field