Predictors of undergraduate and postgraduate clinical performance: A longitudinal cohort study

Abstract

OBJECTIVE: Though many medical schools applied various admission criteria in the selection process, the evidence of using those criteria is unclear. This study examined the predictive validity of each admission criterion for student competency. DESIGN, SETTING, AND PARTICIPANTS: We conducted a prospective cohort study of all students who matriculated to Seoul National University School of Medicine from 2002 to 2008. Demographic characteristics, admission criteria scores, and clinical competencies based on grade point average (GPA), objective structured clinical examination score, and internship score were obtained for each student to analyze the predictive validity of admission criteria. RESULTS: Graduate GPA at the end of 4 years positively correlated with preadmission GPA (p < 0.0001) and written test score (p = 0.012) but negatively correlated with essay test (p = 0.049). Internship score significantly correlated with preadmission GPA and graduate GPA. Regression analysis revealed that the preadmission GPA of the affiliated college and young age at admission could predict GPA, and preadmission GPA and graduate GPA could predict the internship score, which indicates postgraduate clinical performance. CONCLUSIONS: These findings suggest that preadmission GPA is a reliable predictor of academic achievement during medical school and postgraduate clinical performance. For assessing nonacademic competencies, further research is needed.