Patterns of Stroke Between University Hospitals and Nonuniversity Hospitals in Mainland China: Prospective Multicenter Hospital-Based Registry Study

Background and objective: In China, stroke has risen to the first commonest cause of death. Currently published data on stroke come mainly from university hospitals and less from community hospitals, especially lacking information on stroke focusing on the differences between university hospitals and nonuniversity hospitals. Therefore we aimed to investigate the patterns and differences of acute stroke between university hospitals and nonuniversity hospitals in China.

Methods: The survey was conducted in 281 hospitals in China: 62 in the west, 85 in the middle, and 134 in the eastern regions. The participating hospitals were sorted into university hospitals (n = 93) and nonuniversity hospitals (n = 118). We prospectively registered patients with acute stroke within 7 days of symptom onset between 1 April and 20 May 2006. The diagnosis of stroke was defined by World Health Organization criteria, and the pathologic types of stroke were determined by clinical and computed tomography/magnetic resonance imaging findings. The demographic data, pathologic types of stroke, and outcomes (death or disability) at discharge (or 30 days if not discharged) were collected. Disability was defined as modified Rankin Scale (mRs) score >2. Basic characteristics, pathologic types of stroke, and functional outcomes were compared between university hospitals and nonuniversity hospitals.

Results: We enrolled 5273 patients (3135 males; mean age, 65 ± 12 years), of which ischemic stroke accounted for 70.8% (3733), intracranial hemorrhage for 25.7% (1357), and subarachnoid hemorrhage for 3.5% (183). Most of the patients (3555, or 67.4%) were from nonuniversity hospitals, and 1718 patients (32.6%) came from university hospitals. There were no significant differences between university hospitals and nonuniversity hospitals in terms of age, sex, pathologic types of stroke, and history of stroke (all P > 0.05), except the less stroke severity (mRS) on admission (3.1 ± 1.4 vs. 3.2 ± 1.3; P = 0.005) in patients from nonuniversity hospitals. After adjustment for age, sex, and mRS on admission, only patients with intracranial hemorrhage from nonuniversity hospitals had significantly lower proportions of disability (40.8% vs. 47.1%; P = 0.005) and death/disability (49.9% vs. 57.1%; P = 0.008) at discharge than those from university hospitals.

Conclusion: This survey comparing stroke patients treated in university and nonuniversity hospitals in China showed that both cohorts were largely comparable including most demographic features, such as age and gender, history of stroke, and stroke subtypes. It may be concluded that patient cohorts treated on stroke units in university and nonuniversity hospitals are similar and stroke units in both types of institutions are also comparable in terms of workload and most patient outcomes.