Abstract
{ "background": "Community health centres are a cornerstone of primary healthcare delivery in Kenya, yet long-term financial sustainability remains a critical challenge. Robust forecasting of their cost-effectiveness is essential for strategic resource allocation and policy planning, but the methodological rigour of existing approaches is unclear.", "purpose and objectives": "This systematic review aims to critically evaluate the methodological approaches used for cost-effectiveness forecasting in Kenyan community health centre systems, identifying prevalent models, their technical specifications, and key methodological gaps.", "methodology": "A systematic search of peer-reviewed literature and grey literature was conducted following PRISMA guidelines. Studies employing quantitative forecasting models for cost-effectiveness or economic evaluation were included. Data were extracted on model structure, variables, validation techniques, and uncertainty handling. Quality assessment used a bespoke tool for forecasting studies.", "findings": "Of the 27 included studies, a dominant theme was the reliance on deterministic Markov cohort models (63%), with few incorporating stochastic elements or advanced time-series techniques. A key concrete result is that only 22% of models reported using robust standard errors or bootstrapping to account for parameter uncertainty in their forecasts. The autoregressive integrated moving average (ARIMA) model, when used, was typically specified as $\\Delta^d yt = c + \\phi1 \\Delta^d y{t-1} + ... + \\phip \\Delta^d y{t-p} + \\theta1 \\epsilon{t-1} + ... + \\thetaq \\epsilon{t-q} + \\epsilont$, but often without adequate diagnostic testing.", "conclusion": "Methodological approaches for cost-effectiveness forecasting in this context are heterogeneous and frequently lack sophistication in handling uncertainty and temporal dynamics, potentially compromising the reliability of long-term projections.", "recommendations": "Future research should prioritise the adoption and validation of stochastic models, integrate high-frequency routine health system data, and adhere to formal forecasting evaluation protocols to improve predictive performance and policy utility.", "key words": "economic evaluation, health