Understanding the Study’s Focus
This research analyzes cesarean section (CS) data from the “Federico II” University Hospital. It examines information from 9,900 women who underwent CS between 2014 and 2021. The study aims to identify factors affecting the length of stay (LOS) in the hospital after surgery. By using a multiple linear regression (MLR) model, the study estimates LOS based on various independent variables, including age, preoperative conditions, and complications during surgery. The findings aim to support healthcare decision-making and improve resource management.
Key Findings and Methodologies
- An MLR model achieved an impressive R² value of 0.815, indicating strong predictive capability for LOS.
- The random forest algorithm also performed well, with an R² of 0.813, but MLR was preferred for data processing.
- Decision tree (DT) and support vector machine (SVM) models showed the highest accuracy in predicting LOS classes, reaching 75%.
- Significant variables influencing LOS included abnormal fetus, cardiovascular issues, and multiple births, with p-values below 0.05.
Significance of the Research
This study highlights the importance of using data-driven approaches in healthcare to enhance clinical decision-making. The insights gained can help hospitals manage resources better and optimize patient care. By identifying critical factors that extend LOS, healthcare providers can tailor interventions for high-risk patients. Furthermore, the research underscores the need for multi-center studies to improve the generalizability of predictive models across different healthcare settings.
