Interpretation of electrocardiograms by first-year residents: the need for change

Background: Prior studies have shown that misinterpretation of the electrocardiogram (ECG) can lead to inappropriate diagnoses and clinical decisions. This may be particularly true during the first month of postgraduate training. This study was designed to assess proficiency in ECG interpretation among residents at the start of their internal medicine (IM) residency. Methods: Ten ECGs were selected from IM department teaching files. All were representative of conditions that a starting IM resident should be able to identify. The ECGs had 1 correct primary diagnosis and a short list of secondary findings as determined by 2 cardiologists who reviewed them independently. Fifty-two first-year IM residents were given copies and asked to record their interpretations and an assessment of their certainty in each interpretation. Certainty was scored on a scale of 0 to 4 (0 representing a guess and 4 representing 100% certainty). Two blinded, independent graders scored each interpretation on a scale of 0 to 2 (0 = incorrect, 1 = partially correct, 2 = correct). Results: Overall, only half of all ECGs were read correctly. For the most critical diagnoses, the mean scores were as follows: 1.73/2.0 for acute myocardial infarction, 1.5/2.0 for atrial flutter, 1.11/2.0 for ventricular tachycardia, and 0.23/2.0 for complete heart block. The average level of certainty recorded by all participants was low at 18.5 of a maximum of 40. Conclusions: Internal medicine residents at the beginning of their residency training demonstrated low overall proficiency in interpreting ECGs and self-perceived confidence. Nearly all residents felt that their training was insufficient. These findings emphasize the need for improved and more effective training in ECG interpretation for physicians starting residency.