BACKGROUND: Electrocardiogram (ECG) interpretation is widely performed by emergency physicians. We aimed to determine the accuracy of interpretation of potential ST-segment elevation myocardial infarction (STEMI) ECGs by emergency physicians. METHODS: Thirty-six ECGs resulted in putative STEMI diagnoses were selected. Participants were asked to focus on whether or not the ECG in question met the diagnostic criteria for an acutely blocked coronary artery causing a STEMI. Based on the coronary angiogram, a binary outcome of accurate versus inaccurate ECG interpretation was defined. We computed the overall sensitivity, specificity, accuracy and 95% confidence intervals (95%CIs) for ECG interpretation. Data on participant training level, working experience and place were collected. RESULTS: 135 participants interpreted 4603 ECGs. Overall sensitivity to identify 'true' STEMI ECGs was 64.5% (95%CI: 62.8-66.3); specificity in determining 'false' ECGs was 78% (95%CI: 76-80.1). Overall accuracy was modest (69.1, 95%CI: 67.8-70.4). Higher accuracy in ECG interpretation was observed for attending physicians, participants working in tertiary care hospitals and those more experienced. CONCLUSION: The accuracy of interpretation of potential STEMI ECGs was modest among emergency physicians. The study supports the notion that ECG interpretation for establishing a STEMI diagnosis lacks the necessary sensitivity and specificity to be considered a reliable 'stand-alone' diagnostic test.