Abstract 269: Increasing Accuracy of Non-ST Elevation Myocardial Infarction Diagnosis Through Use of an Educational App Based on the Universal Definition of Myocardial Infarction

Background: There is a misconception among medicine trainees that an elevated troponin alone indicates a myocardial infarction (MI) even in the absence of ECG changes, imaging abnormalities, and/or typical symptoms. Misdiagnosis of non-ST elevation MI (NSTEMI) can lead to potentially harmful treatment, unneeded testing, and poor quality metrics. We tested the hypothesis that an educational app provided to trainees would increase accuracy of NSTEMI diagnoses.

Methods: The app was designed using criteria for NSTEMI from the Third UDMI as well as revisions anticipated in the Fourth UDMI. An algorithm was designed to improve diagnostic accuracy of AMI by UDMI. The app was tested from 12/2017 to 08/2018. Accuracy of NSTEMI diagnosis was evaluated 4 months before and then 4 months after app implementation. 431 patient charts were reviewed. MI type was assigned based on the UDMI and assessed for accuracy. Unclear cases were adjudicated by an expert cardiologist.

Results: 207 patients were in the intervention group and 224 were in the control group. Comparison revealed increased specificity of 33.6% (80.9% vs 47.3%) and increased positive predictive value of 19.7% (33.3% vs 13.6%) for NSTEMI diagnosis. Fisher’s exact test revealed a statistically significant difference for each change (p<0.0001).

Conclusion: Our app improved diagnostic specificity of NSTEMI by a statistically and clinically significant 33.6%. This was largely due to a substantial decrease in false positive diagnoses of patients with elevated biomarkers but no objective evidence of ischemia. This increased specificity saved patients and the hospital many unnecessary tests and procedures which likely led to improved quality of care and reduced cost. This app is a valuable tool that should be expanded to all hospital staff.