How does the level of care change after proximal humeral fracture in older patients? A propensity score matching analysis

PURPOSE: In the context of an aging population, the impact of surgical versus non-operative treatment for proximal humeral fractures (PHFs) on level of care (LoC) in older adults remains poorly understood. METHODS: This retrospective study analyzed claims data from the German BARMER health insurance fund, including patients aged ≥ 65 years with PHF (ICD S42.2) between 01/2017 and 09/2022. Patients were categorized into non-operative or surgical treatment groups, with surgery further divided into locked-plate fixation (LPF), reverse total shoulder arthroplasty (RTSA), and other. A 1:1 propensity score matching adjusted for age, sex, pre-fracture LoC, year of fracture, and risk profile was performed. Primary endpoints were overall survival, LoC deterioration, and nursing home care. Hazard ratios were estimated using Cox proportional hazards or Fine and Gray models to account for competing risks. RESULTS: Among 54,595 patients (median age 79, 84% female), 44.3% underwent surgery. After matching (31,314 patients), 21.3% of patients experienced LoC deterioration one year after PHF. RTSA was associated with increased risk of LoC worsening (HR 1.40; 95% CI 1.31-1.50), while LPF showed no difference compared to non-operative treatment. Risk for progression to severe care dependency (LoC ≥ 3) or nursing home care did not differ between treatment groups. Surgical treatment was associated with reduced overall mortality (HR 0.88; 95% CI 0.84-0.91). CONCLUSION: While RTSA was associated with a higher risk of LoC deterioration, differences in severe care dependency were not observed. Surgical treatment was linked to lower mortality, supporting individualized treatment decisions that consider both survival and long-term care outcomes. LEVEL OF EVIDENCE: Level III, retrospective comparative study.