The purpose of this study is to explore the reasons for treatment differences between African Americans and whites with cerebrovascular disease. The authors assessed race differences in the use of invasive procedures, the receipt of less invasive tests, as well as the use of anticoagulant
therapy. Data for this study were drawn from Medicare inpatient hospital records, and a 20%
national sample of patients admitted between January 1 and November 30, 1991 with a diagnosis of transient ischemic attacks (TIA) was selected. For the analysis of anticoagulant therapy, the sample was limited to those TIA patients with a secondary diagnosis of atrial fibrillation with no clinical contraindications.

Utilization rates of African Americans were significantly lower than those of whites for all of the services shown: noninvasive testing (40% for African Americans versus 48% for whites), cerebral angiography (4% versus 8%), carotid endarterectomy (<1% versus 2%), and anticoagulant therapy (21% versus 35%). Even after limiting the sample to those patients who received angiography, African American patients were still significantly less likely to undergo carotid endarterectomy.

There were a number of socioeconomic and clinical differences between African Americans and whites. African Americans were more likely to be dual Medicaid-eligible, live in a poverty area, have had a prior admission, have a secondary diagnosis of hypertension, and have been admitted to a teaching hospital, and they were less likely to have a neurologist as an attending physician. After adjustment for potential confounders, African Americans were less likely to have received non-invasive testing (OR=0.83), to have cerebral angiography among those with non-invasive testing (OR=0.54), to have carotid endarterectomy among patients who underwent angiography (OR=0.27), to have anticoagulant therapy (OR=0.62, not statistically significant), and to have a neurologist as an attending (OR=0.79).

The authors conclude that "Medicare African American patients with TIA may have less access to services for cerebrovascular disease, including both diagnostic tests and medical and surgical therapies aimed at preventing strokes. These suboptimal rates of evaluation and therapy have been noted on other studies and suggest an ongoing need for medical education and system-level practice improvement." Although interpretation of the racial patterns in the use of invasive procedures is limited by the lack of detailed clinical data, differences in severity should not affect the patterns of non-invasive tests. The authors suggest that physician bias (regarding racial patterns in cerebrovascular disease and ability to pay) or high rate of refusals among African Americans might also help explain these patterns.