Predicting Costs Exceeding Bundled Payment Targets for Total Joint Arthroplasty

Area of Science:

• Orthopedic Surgery
• Health Economics
• Predictive Analytics

Background:

• Bundled reimbursement models for total joint arthroplasty (TJA) set target prices for procedures.
• Currently, no tools exist to preoperatively predict patients who may exceed these cost targets.
• Identifying high-cost patients preoperatively is essential for managing bundled payments.

Purpose of the Study:

• To develop and validate a predictive model for excess cost-of-care in TJA.
• To identify demographic and comorbidity factors associated with exceeding TJA cost targets.
• To inform preoperative optimization and modification of bundled payment models.

Main Methods:

• Retrospective analysis of 2,084 primary unilateral TJAs at a single tertiary center.
• Development of a predictive model using 30 comorbidities and 7 demographic factors.
• Creation of a multivariable model and nomogram for visualizing excess cost probability.

Main Results:

• 29% of patients exceeded the target cost price.
• The predictive model showed adequate discrimination (AUC: 0.747).
• Key predictors included age, gender, marital status, ASA score, BMI, race, and 7 specific comorbidities.

Conclusions:

• A novel model using comorbidities and demographics effectively predicts excess TJA costs.
• This model aids in identifying patients for preoperative optimization.
• Findings support adjusting bundled reimbursement models for nonmodifiable risk factors.