Do you hear what you see? Utilizing phonocardiography to enhance proficiency in cardiac auscultation

Abstract

Introduction: Cardiac auscultation skills have proven difficult to train and maintain. The authors investigated whether using phonocardiograms as visual adjuncts to audio cases improved first-year medical students’ cardiac auscultation performance. Methods: The authors randomized 135 first-year medical students using an email referral link in 2018 and 2019 to train using audio-only cases (audio group) or audio with phonocardiogram tracings (combined group). Training included 7 cases with normal and abnormal auscultation findings. The assessment included feature identification and diagnostic accuracy using 14 audio-only cases, 7 presented during training, and 7 alternate versions of the same diagnoses. The assessment—administered immediately after training and repeated 7 days later—prompted participants to identify the key features and diagnoses for 14 audio-only cases. Key feature scores and diagnostic accuracy were compared between groups using repeated measures ANOVA. Results: Mean key feature scores were statistically significantly higher in the combined group (70%, 95% CI 67–75%) compared to the audio group (61%, 95% CI 56–66%) (F(1,116) = 6.144, p = 0.015, ds = 0.45). Similarly, mean diagnostic accuracy in the combined group (68%, 95% CI 62–73%) was significantly higher than the audio group, although with small effect size (59%, 95% CI 54–65%) (F(1,116) = 4.548, p = 0.035, ds = 0.40). Time on task for the assessment and prior auscultation experience did not significantly impact performance on either measure. Discussion: The addition of phonocardiograms to supplement cardiac auscultation training improves diagnostic accuracy and heart sound feature identification amongst novice students compared to training with audio alone.