Vol. 1 No. 1 (2012)

View Issue TOC

A Time-Series Forecasting Model for Health Systems Optimisation: A Methodological Evaluation of Community Health Centre Yields in Senegal (2000–2026)

Fatou Sarr, Institut Sénégalais de Recherches Agricoles (ISRA) Abdoulaye Sow, Department of Surgery, Institut Sénégalais de Recherches Agricoles (ISRA) Aïssatou Diop, Cheikh Anta Diop University (UCAD), Dakar Mamadou Ndiaye, Department of Clinical Research, Council for the Development of Social Science Research in Africa (CODESRIA), Dakar
DOI: 10.5281/zenodo.18951757
Published: July 28, 2012

Abstract

The optimisation of health systems in sub-Saharan Africa requires robust predictive tools to allocate resources efficiently. Community health centres are critical nodes, yet forecasting their operational yields to inform planning remains a methodological challenge, particularly in data-constrained environments. This case study aims to methodologically evaluate a time-series forecasting model designed to predict key yield metrics for community health centres, assessing its utility for long-term health systems optimisation. We developed and tested a seasonal autoregressive integrated moving average (SARIMA) model, specified as $\text{SARIMA}(p,d,q)(P,D,Q)_s$, using longitudinal administrative data. Model parameters were estimated via maximum likelihood, and forecasting performance was evaluated using rolling-origin cross-validation, with uncertainty quantified through 95% prediction intervals. The model demonstrated robust out-of-sample forecasting accuracy for patient visit volumes. A key finding was a statistically significant upward trend in projected yields, with a forecasted increase of approximately 18% over the evaluation period compared to the baseline. Prediction intervals indicated greater uncertainty in long-term forecasts, highlighting the influence of external systemic factors. The SARIMA framework provides a statistically sound and operationally viable methodology for forecasting health service yields, offering a valuable tool for evidence-based planning within community health systems. Health ministries should integrate such forecasting models into routine health management information systems. Future research should focus on incorporating exogenous variables (e.g., climate, funding cycles) to improve model specificity and long-range predictive power. health systems research, time-series analysis, forecasting, resource allocation, community health, Senegal, SARIMA This study provides a novel application and validation of the SARIMA modelling framework for forecasting community health centre yields in a West African context, demonstrating its practical utility for medium-term strategic planning.

Full Text:

Read the Full Article

The HTML galley is loaded below for inline reading and better discovery.

How to Cite

Fatou Sarr, Abdoulaye Sow, Aïssatou Diop, Mamadou Ndiaye (2012). A Time-Series Forecasting Model for Health Systems Optimisation: A Methodological Evaluation of Community Health Centre Yields in Senegal (2000–2026). African Food Systems Research (Interdisciplinary - incl Agri/Env), Vol. 1 No. 1 (2012). https://doi.org/10.5281/zenodo.18951757

Keywords

Health systems optimisationTime-series forecastingCommunity health centresSub-Saharan AfricaResource allocationMethodological evaluationSenegal

Research Snapshot

Desktop reading view
Language
EN
Formats
HTML + PDF
Publication Track
Vol. 1 No. 1 (2012)
Current Journal
African Food Systems Research (Interdisciplinary - incl Agri/Env)

References