Predicting total healthcare demand using machine learning: separate and combined analysis of predisposing, enabling, and need factors

BMC Health Services Research, Journal Year: 2025, Volume and Issue: 25(1)

Published: March 12, 2025

Predicting healthcare demand is essential for effective resource allocation and planning. This study applies Andersen's Behavioral Model of Health Services Use, focusing on predisposing, enabling, need factors, using data from the 2022 Turkey Survey by TUIK. Machine learning methods provide a powerful approach to analyze these factors their combined impact utilization, offering valuable insights health policy. Seven different machine models—Decision Tree, Random Forest, Support Vector (SVM), K-Nearest Neighbors (KNN), Logistic Regression, XGBoost, Gradient Boosting—were utilized. Feature selection was conducted identify most significant influencing demand. The models were evaluated accuracy generalization ability performance metrics such as recall, precision, F1 score, ROC AUC. identified key features affecting For predisposing gender, educational level, age group significant. Enabling included treatment costs, community interest, payment difficulties. Need influenced smoking status, chronic diseases, overall status. demonstrated high recall (approximately 0.90) strong scores (ranging 0.87 0.88), indicating balanced between precision recall. Among models, Boosting, Regression consistently outperformed others, achieving highest predictive accuracy. Forest SVM also performed well, showing robust classification capability. findings highlight effectiveness in predicting demand, providing policy allocation. emerged reliable demonstrating superior performance. Understanding separate effects can contribute more efficient data-driven planning, facilitating strategic decision-making service delivery.

Language: Английский

---

Problematic internet use: A growing concern for adolescent health and well-being in a digital era

Journal of Global Health, Journal Year: 2024, Volume and Issue: 14

Published: Aug. 30, 2024

Language: Английский

---

Psychometric properties of the Turkish version of the universal mental health literacy scale for adolescents

Journal of Pediatric Nursing, Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 1, 2024

Language: Английский

---