File(s) under embargo
Does hospital setting influence shoulder replacement revision rate? A national comparison of outcomes between private and public settings
journal contributionposted on 2023-09-14, 06:33 authored by Nicholas G Paltoglou, Stephen GillStephen Gill, Michelle Lorimer, Sophia Corfield, Richard PageRichard Page
AbstractBackgroundAustralian healthcare relies on both private and public sectors to meet the demand for surgical care. Rapid growth of shoulder replacement surgery highlights a disparity in service provision, with two‐thirds occurring privately. This study aimed to assess the influence of hospital setting on shoulder replacement revision rate at a national level.MethodsAll primary shoulder replacements recorded by the Australian Orthopaedic Association National Joint Replacement Registry from April 2004 to December 2020 were included. Private and public settings were compared for stemmed total shoulder replacement (sTSR) for osteoarthritis (OA), reverse total shoulder replacement (rTSR) for OA/cuff arthropathy (CA), and rTSR for fracture. The primary outcome was cumulative percent revision (CPR), with Kaplan–Meier estimates of survivorship to determine differences between private and public hospitals, recorded as hazard ratios (HR). Secondary analyses investigated differences between hospital settings, targeting hospital outliers for revision and prosthesis selection.ResultsPrimary sTSR (OA) demonstrated a higher revision rate in private hospitals compared to public hospitals (HR = 1.27; P = 0.001), as did rTSR (OA/CA) after 3 months (HR = 1.33; P = 0.003). However, no significant difference was observed for primary rTSR (fracture) (HR = 1.10; P = 0.484). Restricting rTSR (OA/CA) to the best performing prosthesis combinations eliminated the difference between private and public outcomes (HR 1.10; P = 0.415). No other secondary analysis altered the primary result.ConclusionsDifferences exist between private and public hospitals for revision rate following primary shoulder replacement. Prosthesis selection accounts for some of the variation. Further analysis into patient specific characteristics is necessary to better understand these disparities.