Background and Purpose: Stroke or cerebrovascular accident is a leading cause of neurological impairment and long term disability in the older adult. Projections of the elderly indicate they are more vulnerable to stroke than in the general population. African Americans experience not only a higher stroke incidence than whites, but also higher stroke mortality and, among stroke survivors, worse residual physical impairments. Because of these impairments, a valid and reliable measure of stroke impairment is needed that can be utilized in an acute hospital setting to determine the most appropriate rehabilitation plan for African American patients. The National Institutes of Health Stroke Scale (NIHSS) has been determined to be a valid and reliable tool for patients with an acute stroke. However, few data exist about the appropriateness of this instrument for African American stroke survivors. Therefore, the purpose of this study was to determine the reliability of the National Institutes of Health Stroke Scale (NIHSS) for African Americans with an acute stroke. Methods: This study was a reliability/correlational study of African Americans who were admitted to the hospital with the diagnosis of a cerebrovascular accident. On admission to an acute, urban hospital, patients with the diagnosis with an acute CVA were referred to physical therapy by the primary physician, and within 48 hours of referral underwent a physical therapy examination. Part of the initial examination included the National Institutes of Health Stroke Scale (NIHSS) and patients were re-examined three months post discharge with the NIHSS. The main outcome measures were the comparison of the initial score and the score 3 months post discharge. Results: The study included a cohort of 35 older African Americans between 60 and 92 years of age. Sixty percent of the subjects were under 80 years of age; 77% women; 23% men; 57% had a left cerebral insult and 43% had a right; 83% had a thrombus. A Pearson product correlation, Spearman rho correlation and Chi Square were done to determine level of significance and test of independence. A significant relationship was found between the initial score and the repeat score and discharge destination at p<.01. There was a less significant relationship found between the intitial score and gender at the p<.05. There was no relationship found between the initial score and repeat score for side of lesion, patient age or type of CVA. Conclusions: The test-retest reliability coefficient for the initial raw score and the repeat raw score was strong. The NIHSS identified the severity level of neurological involvement in this cohort of African American subjects with an acute stroke. The NIHSS appears to be a useful tool for physical therapy clinicians in determining destination and rehabilitation planning for African American patients after an acute stroke, but more study is needed.