Efeito da reflexão deliberada e do diagnóstico diferencial na calibragem diagnóstica das doenças anorretais entre os estudantes do último ano do curso de medicina da Unifenas - Alfenas

Abstract

Introduction: To avoid diagnostic errors, physicians must recognize when their diagnostic hypotheses are fragile. The confidence they place in a hypothesis must be well aligned with theaccuracy of this same diagnosis. This relationship between confidence and accuracy is called diagnostic calibration and it is relevant to good clinical practice. Objective: To evaluate the effect of guided deliberate reflection, with and without a diagnostic “tip”, and differential diagnosis on diagnostic calibration in sixth-year medical students at UNIFENAS- Alfenas, when diagnosing clinical cases about anorectal lesions. Methods: Students were invited to diagnose, in a single session, six clinical cases about anorectal lesions, being randomly allocated to this analysis through DR (experimental group 1), DR with the tip (experimental group 2) and through differential diagnosis (control group). For each clinical case, participants were asked to write down the most likely diagnosis for the case and then indicate their confidence in this diagnosis, using a confidence scale. Results: 71 students were included, 24 allocated to differential diagnosis, 24 to deliberate reflection and 23 to deliberate reflection with the tip. There were no significant differences between the groups regarding initial accuracy (p=0.342), initial confidence (p=0.666) and initial calibration (p=0.877). There was also no significant difference between the groups regarding final confidence (p=0.375). The final accuracy was different between the groups (p<0.001), with the participants in the deliberate reflection with the tip group showing higher accuracy than the others. The final calibration was different between the groups (p<0.001), with the deliberate reflection with the tip group calibrating better than the others. Conclusion: Deliberate reflection with the tip increased diagnostic accuracy, which in turn increased calibration. We expected the increase in calibration to come from both increased accuracy and reduced confidence. Participants showed, in general, slight overconfidence in their diagnostic hypotheses