Introduction: Primary Stroke Center (PSC) certification was established to improve stroke care. The numbers of PSCs have significantly increased in the past decade. However, it remains unclear whether PSC certification has any impact on stroke mortality. We examined the short term mortality of hospitals that received initial PSC certification between 2009 and 2013 (new PSCs), compared to those received PSC certification before 2009 (existing PSCs) and those never received PSC certification (NSCs).
Method: The inclusion criteria was Medicare beneficiaries aged ≥65 years who were hospitalized between January 1, 2009 to December 31, 2013 with a primary discharge diagnosis of ischemic stroke. The patient information were obtained from the Medicare Provider Analysis and Review (MEDPAR) file. The list and characteristics of hospitals were obtained from the American Hospital Association Annual Survey Database. This study included only those general hospitals with emergency departments. All statistical analyses were performed using SAS Version 9.4 software.
Results: Among 1165,960 Medicare beneficiaries included in this study, 28.9% were treated at 2640 NSCs, 24.6% were treated at 634 new PSCs, and 46.6% were treated at 785 existing PSCs. Higher percentages of patients at new and existing PSCs had complicated hypertension, myocardial infarction, congestive heart failure, atrial fibrillation, prior history of cerebrovascular disease, any malignancy, metastatic cancer, peripheral artery disease and smoking (p<0.0001). New PSCs had the lowest unadjusted in-hospital all-cause mortality, followed by NSCs and existing PSCs (4.2%, 4.6% and 5% respectively). Both New and existing PSC groups had lower unadjusted 30 day compared to NSCs (12.5%, 13.2% and 13.7%). New PSCs had lower unadjusted and adjusted 30 day mortality than existing PSCs (Hazard Ratio 0.981, 95% Confidence Interval (0.968, 0.993)).
Conclusion: The PSCs that were newly certified between 2009 and 2013 had lower unadjusted in-hospital and 30 day mortality after stroke than existing PSCs and NSCs. It is important to further understand whether this difference results from change in patient population or quality of care.