Accuracy of Identifying Acute Stroke Admissions in a Michigan Stroke Registry
Author Information
Author(s): Reeves Mathew J., Wehner Susan, Organek Natalie, Birbeck Gretchen L., Mullard Andrew J., Jacobs Bradley S., Kothari Rashmi, Hickenbottom Susan
Primary Institution: Michigan State University
Hypothesis
The study aims to determine the accuracy of acute stroke diagnoses in Michigan hospitals participating in a stroke registry.
Conclusion
The accuracy of hospitals participating in a stroke registry to identify acute stroke admissions was very good, with a tendency to underreport rather than overreport stroke admissions.
Supporting Evidence
- The physician panel reached consensus on 91.3% of admissions.
- The registry teams had a sensitivity of 88.6% and specificity of 100%.
- Positive predictive value of the registry teams' designations was 100%.
- Negative predictive value was 90.5%.
Takeaway
The study shows that hospitals can get better at spotting strokes if they get the right training and support.
Methodology
Registry teams from 15 Michigan hospitals identified suspect acute stroke admissions and classified them, with a physician panel later reviewing the cases for accuracy.
Potential Biases
Potential for bias in case selection and misapplication of case definition criteria.
Limitations
The study's findings may not represent long-term accuracy as the audit was conducted early in the registry's operation, and the small sample size at each hospital limited specific findings.
Participant Demographics
Participants included hospitals of varying sizes and capabilities across Michigan.
Statistical Information
Confidence Interval
95% CI, 81.1%-93.3%
Want to read the original?
Access the complete publication on the publisher's website