National study on emergency department visits for transient ischemic attack, 1992-2001.

OBJECTIVES: To describe the epidemiology of U.S. emergency department (ED) visits for transient ischemic attack (TIA) and to measure rates of antiplatelet medication use, neuroimaging, and hospitalization during a ten-year time period.
METHODS: The authors obtained data from the 1992-2001 National Hospital Ambulatory Medical Care Survey. TIA cases were identified by having ICD-9 code 435.
RESULTS: From 1992 to 2001, there were 769 cases, representing 2,969,000 ED visits for TIA. The population rate of 1.1 ED visits per 1,000 U.S. population (95% CI = 0.92 to 1.30) was stable over time. TIA was diagnosed in 0.3% of all ED visits. Physicians administered aspirin and other antiplatelet agents to a small percentage of patients, and 42% of TIA patients (95% CI = 29% to 55%) received no medications at all in the ED. Too few data points existed to measure a statistically valid trend over time. Physicians performed computed tomography scanning in 56% (95% CI = 45% to 66%) of cases and performed magnetic resonance imaging (MRI) in < 5% of cases, and there was a trend toward increased imaging over time. Admission rates did not increase during the ten-year period, with 54% (95% CI = 42% to 67%) admitted. Regional differences were noted, however, with the highest admission rate found in the Northeast (68%).
CONCLUSIONS: Between 1992 and 2001, the population rate of ED visits for TIA was stable, as were admission rates (54%). Antiplatelet medications appear to be underutilized and to be discordant with published guidelines. Neuroimaging increased significantly. These findings may reflect the limited evidence base for the guidelines, educational deficits, or other barriers to guideline implementation.