Key result
Gradient boosting model predicts hospitalization costs with ~0.84 R², identifying surgical complexity and comorbidities as drivers.
Why the study?
Elderly patients undergoing laparoscopic surgery often have substantial comorbidities leading to higher surgical complexity and hospitalization costs, necessitating efficient cost management strategies.
Observational (n=1,010)
No
Absolute Event Rate: 0.838% vs 0.8135%
GBRT modeling effectively predicts hospitalization costs and identifies key cost drivers (surgical complexity, LOS, CCI, surgical site) in elderly patients undergoing laparoscopic surgery, providing a framework for resource optimization.
May inform resource allocation in elderly laparoscopic surgery; leaves open prospective validation before practice change.
The aging population presents significant challenges to healthcare systems, particularly in the context of laparoscopic surgery, which is increasingly utilized in elderly patients due to its minimally invasive nature and faster recovery. However, elderly patients often present with substantial comorbidities, leading to higher surgical complexity and hospitalization costs, necessitating efficient cost management strategies. This study aims to address this issue by identifying key cost drivers and establishing a theoretical foundation for refined cost management through the development of a patient grouping model using advanced artificial intelligence techniques. A retrospective analysis was conducted on medical records of 1010 elderly patients who underwent laparoscopic surgery. Key factors influencing hospitalization costs were systematically examined using descriptive statistics, univariate analysis, and linear mixed-effects models (LMM). The LMM categorized patients into distinct subgroups based on surgical complexity, while a gradient boosting regression tree (GBRT) model was employed to predict hospitalization costs. The stratified heterogeneity analysis revealed distinct high-cost patient subgroups, particularly those experiencing extended length of hospital stay (LOS) who underwent complex laparoscopic surgery while presenting with substantial comorbidity burden. Through the GBRT modeling approach, surgical complexity, LOS, Charlson Comorbidity Index (CCI), and surgical site emerged as statistically significant predictors of hospitalization costs. This study demonstrates the effectiveness of GBRT in predicting hospitalization costs and enabling patient grouping for elderly laparoscopic surgery patients. The findings provide healthcare institutions with a data-driven framework to optimize resource allocation, standardize care pathways, and enhance cost management strategies. Future research should focus on external validation, incorporation of economic variables, and finer-grained cost component analysis to further refine the model’s applicability and robustness.
No takes yet. Share an insight, caveat, or question.
Hu et al. (2025) conducted an observational in Laparoscopic surgery (n=1,010). Gradient Boosting Regression Tree (GBRT) model vs. Random Forest Regressor (RFR) was evaluated on Hospitalization costs prediction accuracy (R² Score). The gradient boosting regression tree model accurately predicted hospitalization costs with an R² of 0.8380, identifying surgical complexity, length of stay, and comorbidities as key cost drivers.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: