This study investigates the relationship between patient demographics, hospitalization factors, and length of stay (LOS) in hospitals. Statistical methods, including Kruskal Wallis tests and post-hoc analysis with generalized linear models with Gamma family and log link function, Random Forest Regressor, and, XGBoost are used to identify significant factors associated with LOS. The findings inform healthcare providers in identifying high LOS risk patients and optimizing hospital resource allocation. This project advances LOS prediction accuracy in hospitals through novel statistical methods.
Null Hypothesis: There is no significant association between patient demographics, including age, gender, and race, and hospitalization-related factors, suchas admission type, insurance, and marital status, with respect to the length of hospital stay at the time of admission. Alternate Hypothesis: There is a significant association between patient demographics, including age, gender, and race, and hospitalization-related factors, such as admission type, insurance, and marital status, with respect to the length of hospital stay at the time of admission.