Efeito do Treinamento de Habilidades Clínicas usando a Dermatologia Comparativa na acurácia diagnóstica de alunos de Medicina

Abstract

Introduction: The clinical reasoning in dermatology is mostly based on the analysis of the visual characteristics of cutaneous dermatoses. The diagnostic process is predominantly used by the expert called non-analytical reasoning, therefore it is based on pattern recognition. The use of analogies could help medical students with little clinical experience and not well developed reasoning structured patterns to improve diagnostic accuracy. Objective: To evaluate the contribution of Comparative Dermatology as a strategy to facilitate the acquisition of the diagnostic accuracy in dermatological lesions at graduation level. Methods: Controlled experiments with students from the 8th term of UNIFENAS-BH medical school. The students were randomly divided into two groups (Comparative Dermatology (CDG) and Control (CG)). All students answered a pre-test where they needed to identify and diagnose lesions. Students were then placed in two training groups: the CDG should diagnose lesions after comparison with similar objects while the CG should diagnose dermatoses by the description of the elementary lesions. Both groups were evaluated in a posttest at 30 and 90 days after the training. The accuracy of the diagnostic in the three periods of evaluation was compared between the groups and within the same group using the statistical method ANOVA. The ability to make analogy was also analyzed. Results: A total of 60 students were included, of whom 31 were allocated in the CDG and 29 in the CG. Both groups showed a statistically significant improvement in diagnostic accuracy 30 and 90 days post-training. There was no difference regarding performance between groups (p = 0.378). Stratified analyzes by new (p=0,808) or old diagnoses (p = 0,518), and diagnostic difficulty: easy (p = 0,899) and difficult (p = 0,642), also the difference between the groups was not significative. The CDG was able to memorize the objects given in the training phase and performed better when making analogies compared to the CG (p < 0.001). There was no statistically significant correlation between the scores obtained by the CDG and the score for comparative object identification (p = 0,437). Conclusion: The diagnostic accuracy was similar between CDG and CG. It was observed that the group that used the comparison objects in the training performed better in making analogies, memorizing correct objects previously trained, however they appear to have failed to make associations between these objects and the correct diagnosis of dermatological lesions.