Marostica, Chance W.

Background/Objectives: Machine learning (ML) models can predict hospital admission from emergency department (ED) triage data with areas under the receiver operating characteristic curve (AUC) exceeding 0.85. Whether incorporating the assigned provider’s identity—as a proxy for unmeasured practice variation—improves prediction has not been systematically studied. We aimed to compare 10 supervised ML classifiers for predicting hospital admission at ED triage, with and without provider identity, and to characterize model reasoning using SHapley Additive exPlanations (SHAP). Methods: We conducted a retrospective cohort study of 186,094 ED visits (2020–2023, training) and 58,151 visits (2024, temporal holdout test) at one academic tertiary-care ED. Ten classifiers spanning linear, distance-based, tree-based, ensemble, probabilistic, and neural network families were each trained in two conditions: baseline (23 triage features) and with provider identity appended. SHAP TreeExplainer was applied to the top-performing models (CatBoost and XGBoost). Results: The admission rate was 31.3% (training) and 31.7% (test). CatBoost achieved the highest baseline AUC of 0.8906 (0.8878–0.8933). Adding provider identity produced negligible AUC changes across all models (ΔAUC range: −0.0029 to +0.0015; all DeLong p > 0.05). SHAP analysis identified ESI level, respiratory rate, temperature, complaint category, and age as the dominant predictors, with clinically intuitive directionality. Conclusions: Provider identity does not meaningfully improve ML prediction of hospital admission beyond standard triage variables. The observed 28-percentage-point variation in provider admission rates is explained by patient case-mix differences than with independent practice pattern effects on prediction. SHAP provides transparent, clinically interpretable explanations suitable for bedside decision support.

Background: An understanding of financial trends is important to advance agreeable reimbursement models in plastic surgery. This study aimed to evaluate trends in Medicare reimbursement rates for the 20 most commonly billed reconstructive plastic surgery procedures from 2000 to 2019. Methods: The Centers for Medicare and Medicaid Services Physician and Other Supplier Public Use File was used to identify the 20 reconstructive procedures most commonly billed to Medicare by plastic surgeons in 2016. Reimbursement data were extracted from The Physician Fee Schedule Look-Up Tool from the Centers for Medicare and Medicaid Services for each CPT code. Monetary data were adjusted for inflation to 2019 U.S. dollars. Average annual and total percentage changes in reimbursement were calculated based on these adjusted trends. Results: The average adjusted reimbursement for all procedures decreased by 14.0 percent from 2000 to 2019. The greatest mean decrease was observed in complex wound repair of the scalp, arms, or legs (−33.2 percent). The only procedure with an increased adjusted reimbursement rate was layer-closure of the scalp, axillae, trunk, and/or extremities (6.5 percent). From 2000 to 2019, the adjusted reimbursement rate for all procedures decreased by an average of 0.8 percent annually. Conclusions: This is the first comprehensive study evaluating trends in Medicare reimbursement in plastic surgery. When adjusted for inflation, Medicare reimbursement for the included procedures has steadily decreased from 2000 to 2019. Increased consideration of these trends will be important for U.S. policymakers, hospitals, and surgeons to ensure continued access to meaningful reconstructive plastic surgery care.