"background": "Emergency care systems in South Africa face significant strain, yet a comprehensive methodological synthesis of clinical outcome forecasting models is lacking. This gap impedes the development of robust, evidence-based planning tools for healthcare delivery. ", "purpose and objectives": "This meta-analysis aims to critically evaluate methodological approaches and to develop an integrated time-series forecasting model for key clinical outcomes within the nation's emergency units. ", "methodology": "A systematic search identified relevant studies. Methodological quality was appraised using a modified Cochrane tool. Quantitative data were synthesised via random-effects meta-analysis. The core forecasting model is an autoregressive integrated moving average with exogenous variables (ARIMAX), specified as \ᵈ yt = c + =1^{p\ \ᵈ yt-i + =1^q\ -j + =1^m\ Xk, t + \, where Xk represents covariates including nurse-to-patient ratios. Uncertainty was quantified using 95% prediction intervals. ", "findings": "Methodological evaluation revealed that 68% of included studies had a high risk of bias due to inadequate handling of missing data. The synthesised ARIMAX model forecasts a significant negative association between higher patient crowding and positive clinical outcomes (β = -0. 23, 95% CI: -0. 31 to -0. 15). ", "conclusion": "Current forecasting methodologies exhibit substantial limitations, but the integrated model provides a more rigorous tool for predicting clinical outcomes, directly linking operational factors to patient care trajectories. ", "recommendations": "Future research must adopt more rigorous missing data protocols. Health policymakers should utilise advanced forecasting models incorporating real-time operational data for resource allocation and capacity planning. ", "key words": "emergency care, forecasting, meta-analysis, clinical outcomes, health systems, South Africa, time-series", "contribution statement": "This study provides
Merwe et al. (Thu,) studied this question.