Cognition Admission Test as Selection Method for Students’ Entry in a PBL Medical Curriculum [Letter]

Dear editor

We have read with great interest the paper by Kötter et al¹ The authors compared academic performance in the general practice of students in the fifth year of the medical course who were selected to entry the medical school based on pre-university grade point average (pu-GPA) with those who were selected based on pu-GPA and a 30-min panel interview. They observed that the former was considered more suitable for general practice than the latter by the supervising general practitioner. However, because of the small sample size, they correctly stated that the results should be received with caution. Furthermore, because the study was carried in a single center, they conceded that the generalizability of the study to other medical schools was limited.

The selection process for students to entry a medical course is still debatable. In other contexts, pu-GPA appears to be associated with outcome prediction² low rate of students dropout during medical course.³ Pu-GPA has been associated with a successful career following graduation as well.² Traditional interviews have been considered to have a poor outcome prediction in comparison with pu-GPA in the students’ selection process.² Therefore, the findings of Kotter et al¹ are somewhat surprising.

Cognitive admission tests (CAT) are another selection method for students to entry a medical school, which reliably predicts academic performance. At our university, we run a medical course based on a Problem-Based Learning (PBL) teaching approach.⁴ A preliminary study from our institution clearly shows that marks in the CAT are associated with academic performance at the end of medical course in terms of abilities and attitudes, as assessed by the Objective Structured Clinical Examination (OSCE). Furthermore, marks in the CAT are also associated with academic achievement in terms of cognition, as assessed by the Progress Testing (PT) and end-unit tests of the disciplines of the clerkship (Pediatrics, Internal Medicine, Surgery, Gynecology and Obstetrics, Family Medicine, and Urgency Medicine).⁵

Kötter et al¹ assessed students apparently formed in a Lecture-Based Learning (LBL) medical course using a Likert scale regarding suitability to work as a general practitioner by one general practitioner supervisors. This is simpler than to assess attitude, ability, and cognition by different methods of assessment as the OSCE and the PT, as recommended in the PBL approach. Therefore, we congratulate Kötter et colleagues¹ for their challenging work. Nevertheless, for medical courses running a PBL curriculum, we think that CAT is a preferable method for students’ selection to entry a medical course. Nonetheless, we concede that a randomized trial comparing different selection methods is needed in order that we can select students who might ultimately be a good doctor.