CVD Epidemiology and Prevention Council, American Heart Association
2004 San Francisco

The North Carolina Collaborative Stroke Registry: Design and First Year Experience

Wayne D. Rosamond, Larry B. Goldstein, Andrew W. Asimos, Robert Yapundich, Deborah Albritton, Charles H. Tegeler, Ana Felix, Kelly R. Evenson

Background
Established in 2002, the North Carolina Collaborative Stroke Registry is a pilot for the Paul Coverdell National Acute Stroke Registry. The North Carolina registry is evaluating prospective patient identification and a real-time data collection and analysis system.

Methods
Study Population

• Eleven registry sites in North Carolina participated

Enrollment and Data Collection Procedures

• Cases were identified prospectively based on presentation to the Emergency Department or through direct admission to Neurology or other appropriate services in the hospital
• Enrollment criteria
• Age 18 years or older
• Presents with signs or symptoms consistent with stroke (ischemic, intracranial hemorrhage, or subarachnoid hemorrhage) or transient ischemic attack
• In-hospital strokes or intracranial hemorrhage resulting from trauma excluded
• Eligible patients or their appropriate informants were interviewed in the Emergency Department and a Stroke Registry Card was filled out by hospital staff
• Abstraction of medical record was completed onsite after patient discharged
• Interview and abstraction data were transmitted to a central registry database via a secure website

Preliminary Results

• 2,620 cases meeting eligibility criteria were enrolled
• To date, abstraction of the medical record has been completed for 50 % cases enrolled (n=1,323).
• 68% of the medical records reviewed to date have documentation of symptom onset time (either specific time or estimate).
• 83% of the prospective interviews completed have recorded information on symptom onset time (either specific time or estimate).
• A specific date and time of symptom onset was found in only 34% of medical records compared to 54% of patient interviews.
• Thrombolytic therapy was used in 5.8% of all cases enrolled. Percent treated increased to 15.3% among patients discharged as ischemic stroke who had less than 2 hours pre-hospital delay.
• Of cases enrolled that were current smokers, only 14% received smoking cessation counseling.

Conclusion

• Preliminary experience suggests that a prospective, Emergency Department based registry of acute stroke patients is feasible in a variety of hospital settings.
• Our prospective methodology, including an interview using a stroke registry enrollment card, results in a high capture of stroke symptom onset time, which has been a limitation of retrospective registry models.
• Low rate of thrombolytic therapy use agrees with results from other studies
• The rapid data feedback available to hospitals through the registry website may facilitate quality improvement efforts for acute stroke.