BACKGROUND: Interhospital transfer invasive management patterns and implications for older non-ST-segment elevation myocardial infarction (NSTEMI) patients initially presenting to non-revascularization-capable hospitals have not been explored. HYPOTHESIS: Patients admitted to hospitals with a higher transfer proportion have lower risk of long-term mortality. METHODS: We linked CRUSADE Registry data on 5678 patients age ≥65 years from 65 United States non-revascularization-capable hospitals (2003-2006) with inpatient Medicare longitudinal claims. Hospitals were categorized according to hospital-level patient transfer-out rates, low (≤40%) vs high (>40%). The associations between transfer-out rates and 30-day, 6-month, and 3-year mortality risk were evaluated using Cox proportional hazard models. RESULTS: Hospital-level transfer-out rates varied widely (median, 43%; interquartile range, 31%-54%). Compared with patients from low-transfer-out hospitals (n = 2715), patients from high-transfer-out hospitals (n = 2963) were more likely to be male, less likely to have renal insufficiency and prior heart failure, and had lower long-term CRUSADE mortality risk scores. These patients also more commonly received evidence-based acute medications before transfer and underwent subsequent revascularization after transfer. The adjusted risks of mortality at various time intervals were similar for those from high- vs low-transfer-out hospitals: 30 days (hazard ratio: 0.95, 95% confidence interval: 0.79-1.14), 6 months (0.97, 0.84-1.12), and 3 years (1.01, 0.91-1.11). CONCLUSIONS: Transfer rates for older NSTEMI patients vary widely among non-revascularization-capable hospitals. Despite lower predicted mortality risk and higher rates of post-transfer revascularization, patients from high-transfer-out hospitals had a similar risk for short- and long-term mortality compared with those from low-transfer-out hospitals.