Are Outcomes of Ischemic Strokes Improved When a Neurologist is Attending?
BACKGROUND: Whether admission of stroke patients to neurologists is associated with improved outcomes is uncertain. Though prior studies suggested ischemic stroke patients under the care of neurologists had lower rates of in-hospital mortality than those treated by internists, these studies were uncontrolled for the possibility that patients with better prognosis were admitted to neurologists. METHODS: The University HealthSystem Consortium administrative database contains patient information from 84 large academic health centers and associates. Discharge abstracts for ischemic strokes admitted through emergency rooms 1997–1999 were obtained. Database variables were validated by comparison with a detailed chart review of 927 patients at 36 institutions. Attending physician specialty was evaluated as a predictor of in-hospital mortality using chi-square statistics and multivariable logistic regression. To determine whether hospital rates of stroke admission to neurologists were predictive of in-hospital mortality, generalized estimating equations (GEE) were used. This multivariable method accounts for clustering of observations at institutions, which broadens confidence intervals (CI). All multivariable analyses were adjusted for age, gender, race, admission status, and treatment volume. RESULTS: Of 28,571 ischemic strokes admitted through the emergency department, 58% were admitted to neurologists. Univariate analyses demonstrated a lower risk of in-hospital mortality in cases admitted to neurologist (4.7%) compared to non-neurologists (9.4%; p<0.001). Adjustment for case-mix did not alter the association (odds ratio, 0.48; 95% CI, 0.43–0.53; p<0.001). However, risk of death was not lower at hospitals admitting a larger portion of ischemic stroke cases to neurologists (p=0.54) as would be expected if admission to neurologists led to improved outcomes. CONCLUSIONS: In academic medical centers, ischemic stroke patients admitted to neurologists are less likely to die in the hospital compared to those admitted to other services. However, this may be due to selection of patients with better prognosis for admission to neurologists.