[THE ARGUMENT AGAINST THE USE OF NON-ACADEMIC CHARACTERISTICS IN THE SELECTION OF MEDICAL STUDENTS].

Medical schools rely on tests for the applicants' academic and nonacademic attributes to select students. The former is based on evidence that students with top past academic achievements outperform other students during the clerkship rotations. Testing for nonacademic attributes responds to the expectation that those selected will perform better as physicians than those chosen by academic standards alone. However, even if nonacademic attributes do contribute to identifying applicants with improved later performance, I believe that their use should be re-examined for two reasons. First, applicants rejected because of poor past academic achievements would understand why they were not admitted. Not so those who "fail" nonacademic tests: their rejection may affect their self-esteem. I am concerned that my profession, guided by the principle of "do no harm," does not consider the harm of admission processes that purport to select virtuous applicants. Second, society needs not only clinicians but also physicians with other subspecialties that require different qualities: compassion is important for people-oriented professions, but less so for technology-oriented professions. Therefore, the selection of medical students that is solely based on their past academic performance would be not only less expensive, but also more responsive to societal needs, and more consistent with the ethical requirement to not harm.