Abstract
{ "background": "Public health surveillance systems are critical for early detection and response to disease outbreaks, yet their operational reliability in resource-limited settings is often inadequately quantified. In Kenya, despite system improvements, methodological gaps persist in objectively measuring performance fluctuations over time.", "purpose and objectives": "This study aimed to develop and validate a novel time-series forecasting model to quantitatively assess the reliability of syndromic surveillance reporting within the Kenyan public health system, focusing on consistency and predictability of data flow.", "methodology": "We utilised national-level, weekly syndromic surveillance data. A Seasonal Autoregressive Integrated Moving Average (SARIMA) model, specified as $\\text{SARIMA}(1,1,1)(1,1,1)_{52}$, was fitted to a multi-year training set to generate expected weekly report volumes. System reliability was measured as the mean absolute scaled error (MASE) between forecasts and observed reports from a subsequent evaluation period.", "findings": "The forecasting model revealed significant temporal instability in reporting reliability, with a MASE of 1.87 (95% CI: 1.72, 2.04), indicating the observed data were 87% less accurate than the in-sample forecast. Reliability deteriorated markedly during known rainy seasons, with error spikes exceeding 40% above forecasted values.", "conclusion": "The applied time-series model provides a robust, quantitative tool for measuring surveillance system reliability, demonstrating that reporting consistency in the studied system is suboptimal and exhibits predictable seasonal weaknesses.", "recommendations": "Routine integration of forecasting-based reliability metrics is recommended for surveillance system performance monitoring. Resources should be prioritised to bolster reporting resilience ahead of periods of predictable degradation, such as prior to seasonal weather changes.", "key words": "surveillance, reliability, forecasting, time-series analysis, public health, Kenya", "contribution statement": "This paper introduces a novel application of SARIMA forecasting as a quantitative benchmark for public health surveillance reliability, providing programme managers with an objective tool for proactive