Background: Intravenous Thrombolysis with Recombinant Tissue Plasminogen Activator (rt-PA) availability at Chinese hospitals varies and may affect care quality for acute ischemic stroke patients. Limited research has shown whether there were differences in quality of care at China National Stroke Registry (CNSR II) hospitals based on rt-PA capability.
Methods: For acute ischemic stroke patients admitted to CNSR II hospitals between 2012 and 2013, care quality at hospitals with or without Intravenous rt-PA capability was examined by evaluating conformity with performance and quality measures. The primary outcome was guideline-concordant care, defined as compliance with 10 predefined individual guideline-recommended performance metrics and composite score. A composite score was defined as the total number of interventions actually performed among eligible patients divided by the total number of recommended interventions among eligible patients. Propensity score matching was used to balance the baseline characteristics. We used cox model with shared frailty model and logistic regression with generalized estimating equation to compare the relationship between hospitals with rt-PA capability and hospitals without rt-PA on quality measures.
Results: This study included 19604 acute ischemic stroke patients admitted to 219 CNSR II hospitals. Before matching, there were 7928 patients admitted to 86 (40.4%) hospitals with rt-PA capability and 11676 patients admitted to 133 (59.6%) hospitals without rt-PA capability. After matching, 7606 pairs of patients in rt-PA-capable hospitals and rt-PA-incapable hospitals were analyzed. Before matching, the composite score of guideline-concordant process of care was higher at hospitals with rt-PA capability than hospitals without rt-PA capability (74% versus 73%, P=0.0126). Hospitals with rt-PA capability were more likely to perform deep vein thrombosis prophylaxis within 48 hours of admission, dysphagia screening, assessment or receiving of rehabilitation, discharge antithrombotic, anticoagulation for atrial fibrillation and medications for lowering low-density lipoprotein (LDL) ≥100mg/dL. But hospitals with rt-PA capability were less likely to perform antithrombotic medication within 48 hours of admission and hypoglycemic therapy at discharge for patients with diabetes. After matching, differences of stroke care quality between hospitals with rt-PA capability and without rt-PA capability still exist after adjusting covariates.
Conclusions: The CNSR II hospitals were associated with better performance in some of the hospitals but not all of them. The difference in conformity between rt-PA-capable hospitals and rt-PA-incapable hospitals was modest for performance measures of stroke care. However, more room for improvement still exists in key quality performance measures and further studies should be explored.