Do patient outcomes and follow-up completion rates after shoulder arthroplasty differ based on insurance payor?

Abstract

Background: Disparities associated with socioeconomic status (SES) and insurance coverage have been shown to affect outcomes in different medical conditions and surgical procedures. We hypothesized that patients insured by Medicaid will be associated with lower follow-up rates and inferior outcomes relative to those with Medicare or private insurance.

Methods: Patients undergoing shoulder arthroplasty, including anatomic total shoulder arthroplasty, reverse arthroplasty, and hemiarthroplasty, were enrolled preoperatively in an institutional database. Preoperative demographics, payor (Medicaid, Medicare, or private insurance), and baseline American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form (ASES) scores were recorded. Postoperatively, patients completed ASES scores at multiple time points. Follow-up completion rate was calculated as the number of follow-up visits completed relative to possible visits. Continuous variables were compared between groups with 1-way analyses of variance, and chi-squared tests were used for categorical variables. Significance was defined as P < .05.

Results: There were 491 shoulder replacements performed for 438 patients from 2012-2017. The mean follow-up completed percentage was significantly lower (P < .001) for Medicaid patients (62.6% ± 33.7%) relative to Medicare patients (80.2% ± 26.7%; P < .001) and private insurance patients (77.8% ± 22.1%; P = .001). The ASES Composite score increased significantly for all patients from baseline to final follow-up. At each time point, including before surgery and each postoperative time point, patients with Medicaid insurance had significantly lower ASES Composite scores. The final ASES Composite score was significantly lower in the Medicaid patients (66.1 ± 28.7) relative to private insurance patients (78.3 ± 20.8; P = .023). Medicaid patients had significantly lower preoperative (P < .001) and postoperative (P = .018) ASES Pain subscores. In multivariate regression analysis, Medicaid insurance was associated with both inferior preoperative and postoperative ASES scores relative to patients with Medicare or private insurance.

Conclusions: We observed that all patients, regardless of insurance payor, improved by similar magnitudes after shoulder arthroplasty, though patients with Medicaid insurance had significantly lower preoperative and postoperative ASES scores, primarily because of the ASES Pain subscore. Patients with Medicaid insurance also have lower follow-up rates than other payors.

Keywords: Medicaid; Shoulder arthroplasty outcomes; health care disparities; shoulder arthritis.