Social Determinants of Health Disparities Increase Five-Year Revision Rates but Not Postoperative Complications Following Primary Hip Arthroscopy

Stephen M Gillinov,Zachary L LaPorte,Jonathan S Lee,Bilal S Siddiq,Kieran S Dowley,Nathan J Cherian,Christopher T Eberlin,Michael P Kucharik,Scott D Martin
DOI: https://doi.org/10.1016/j.arthro.2024.07.038
2024-08-19
Abstract:Purpose: To investigate the impact of social determinants of health (SDOH) disparities on 30-day emergency department (ED) visits, 90-day postoperative complications, and five-year secondary surgery rates after primary hip arthroscopy using a large national database. Methods: A national administrative claims database was used to identify patients who underwent primary hip arthroscopy with femoroplasty, acetabuloplasty, and/or labral repair between 2015 and 2022. Patients were queried for whether they experienced any SDOH disparities, including economic, educational, environmental, or social disparities; those experiencing SDOH within 1 year prior to primary hip arthroscopy were matched 1:1 by age, sex, Elixhauser Comorbidity Index score, diabetes, obesity, and tobacco use to patients not experiencing any lifetime SDOH disparities. Odds of 90-day complications and 30-day ED visits were compared using multivariable logistic regression. Rates of five-year revision hip arthroscopy and of any secondary surgery (revision hip arthroscopy or total hip arthroplasty) were compared by Kaplan-Meier analysis. Results: A total of 3,383 primary hip arthroscopy patients who experienced SDOH disparities were matched 1:1 to a control cohort of 3,383 patients who had not experienced SDOH (age, 41.0 years, 79.6% female, for both cohorts). Odds of adverse events following arthroscopy were low and not different between cohorts (SDOH: 1.51%, vs. No SDOH: 1.57%; P=0.09). Additionally, there was no difference in odds of 30-day ED visits (SDOH: 5.65%, vs. No SDOH: 4.79%; P=0.10). The rate of five-year revision hip arthroscopy was significantly greater among patients experiencing SDOH disparities (5.4% vs. 4.1%; P=0.02); however, there was no difference in rates of any secondary surgery between cohorts (11.8% vs. 10.4%; P=0.10). Conclusion: Patients experiencing SDOH disparities had similar odds of postoperative complications and ED visits after primary hip arthroscopy but greater rates of five-year revision hip arthroscopy compared to a matched-control cohort of patients not experiencing SDOH disparities. Level of evidence: 3; retrospective case-control study.
What problem does this paper attempt to address?