A Methodological Evaluation and Time-Series Forecasting Model for the Cost-Effectiveness of Public Health Surveillance Systems in Kenya

{ "background": "Public health surveillance is critical for food systems security and disease control, yet its economic evaluation remains methodologically underdeveloped, particularly in resource-limited settings. Existing frameworks often lack predictive capacity for strategic resource allocation.", "purpose and objectives": "This study aimed to develop and validate a novel time-series forecasting model to assess the cost-effectiveness of public health surveillance systems, using Kenya as a case study. The objective was to provide a tool for predicting future cost-per-event averted to inform budgetary planning.", "methodology": "We conducted a methodological evaluation of surveillance system components and cost structures. A hybrid forecasting model integrating an Autoregressive Integrated Moving Average (ARIMA) component with exogenous economic variables was developed: yt = \μ + \∑i=1ᵖ\ yt-i + \ + \∑i=1q\ \εt-i + \∑j=1ᵏ\ Xj,t. Model parameters were estimated using maximum likelihood, and robustness was assessed via heteroskedasticity-consistent standard errors.", "findings": "The model demonstrated strong predictive accuracy, with a mean absolute percentage error of 12.3% in out-of-sample forecasts. A key finding was a projected 18% increase in the cost-effectiveness ratio for integrated disease surveillance over a five-year forward forecast, with a 95% confidence interval of 14% to 22%.", "conclusion": "The proposed forecasting model provides a robust, forward-looking tool for economic evaluation of surveillance, moving beyond static cost-effectiveness analyses. It enables proactive financial planning for sustaining public health surveillance functions.", "recommendations": "We recommend the adoption of this forecasting methodology by health and agricultural ministries for medium-term expenditure framework planning. Further application and validation in other country contexts with similar surveillance architectures is advised.", "key words": "economic evaluation, forecasting, health economics, predictive modelling, resource allocation, sentinel surveillance", "cont