Medical school grades may predict future clinical competence. Issue 9 (23rd September 2022)

• Record Type: Journal Article
• Authors: Wu, Jr-Wei; Cheng, Hao-Min; Huang, Shiau-Shian; Liang, Jen-Feng; Huang, Chia-Chang; Shulruf, Boaz; Yang, Ying-Ying; Chen, Chen-Huan; Hou, Ming-Chih; Huey-Herng Sheu, Wayne
• Is Part Of: Journal of the Chinese Medical Association. Volume 85:Issue 9(2022)
• Page Start: 909
• Page End: 914
• Publication Date: 2022-09-23
• Subjects: OSCEs -- Prediction model -- School grades -- Structuralized interview; Medicine -- Periodicals; 610.5
• DOI: 10.1097/JCMA.0000000000000782
• Languages: English
• ISSNs: 1726-4901

Abstract

Abstract : Background: In real-world medical education, there is a lack of reliable predictors of future clinical competencies. Hence, we aim to identify the factors associated with clinical competencies and construct a prediction model to identify "improvement required" trainees. Methods: We analyzed data from medical students who graduated from National Yang-Ming University with clerkship training and participated in the postgraduate year (PGY) interview at Taipei Veterans General Hospital. Clinical competencies were evaluated using grades of national objective structured clinical examination (OSCEs). This study used data from medical students who graduated in July 2018 as the derivation cohort (N = 50) and those who graduated in July 2020 (n = 56) for validation. Results: Medical school grades were associated with the performance of national OSCEs (Pearson r = 0.34, p = 0.017), but the grades of the structured PGY interviews were marginally associated with the national OSCE (Pearson r = 0.268, p = 0.06). A prediction model was constructed to identify "improvement required" trainees, defined: trainees with the lowest 25% of scores in the national OSCEs. According to this model, trainees with the lowest 25% medical school grades predicted a higher risk of the "improvement required" clinical performance (Q1–Q3 vs Q4 = 15% vs 60%, odds ratio = 8.5 [95% confidence interval = 1.8-39.4], p = 0.029). In the validation cohort, our prediction model could accurately classify 76.7% "improvement required" and "nonimprovement required" students. Conclusion: Our study suggests that interventions for students with unsatisfactory medical school grades are warranted to improve their clinical competencies.