Variation in the value of total shoulder arthroplasty.

Abstract

There is growing interest in maximizing value for patients undergoing discretionary orthopedic surgery but little data to guide improvement efforts. Integrating patient-reported outcomes with time-driven activity-based costing, we explored patient-level variation in the value of total shoulder arthroplasty (TSA) and characterized factors that contribute to this variation. Using our institutional registry, we identified 239 patients undergoing elective primary TSA (anatomic or reverse) between 2016-2017 with minimum 2-year follow-up. We calculated value as 2-year postoperative American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form (ASES) scores divided by hospitalization time-driven activity-based costs. This number was multiplied by a constant to set the minimum value of TSA to 100. Multivariable linear regression modeling was performed to characterize factors underlying variation in value. The value of shoulder arthroplasty ranged from 100 to 680, resulting in a variation of 580%. Reverse shoulder arthroplasty was associated with decreased value (79-point decrease vs. anatomic arthroplasty; P < .001; partial R² = 0.089), as were prior ipsilateral shoulder surgery (38-point decrease; P = .002; partial R² = 0.031), more self-reported allergies (4-point decrease per 1-unit increase; P = .029; partial R² = 0.015), diabetes (33-point decrease; P = .045; partial R² = 0.013), and lower preoperative ASES score (0.7-point increase per 1-unit increase; P = .045; partial R² = 0.012). We observed wide variation in the value of shoulder arthroplasty that was most strongly associated with procedure type and certain preoperative characteristics (eg, prior shoulder surgery, number of self-reported allergies, diabetes, ASES score). Awareness of these associations is important for implementation of targeted strategies to effectively reduce variation and redirect resources toward higher-value, cost-conscious care.