Proficiency testing for identifying underperforming students before postgraduate education: a longitudinal study.

Background: Efficient selection of medical students in GP training plays an important role in improving healthcare quality. The aim of this study was to collect quantitative and qualitative validity evidence of a multicomponent proficiency-test for identifying underperforming students in cognitive and non-cognitive competencies, prior to entering postgraduate GP Training. From 2016 to 2018, 894 medical GP students in four Flemish universities in Belgium registered to take a multicomponent proficiency-test before admission to postgraduate GP Training. Data on students were obtained from the proficiency-test as a test-score and from traineeship mentors' narrative reports.

Results: In total, 849 students took the multicomponent proficiency-test during 2016-2018. Test scores were normally distributed. Five different descriptive labels were extracted from mentors' narrative reports based on thematic analysis, considering both cognitive and non-cognitive competences. Chi-square tests and odds ratio showed a significant association between students scoring low on the proficiency-test and having gaps in cognitive and non-cognitive competencies during GP traineeship.

Conclusion: A multicomponent proficiency-test could detect underperforming students prior to postgraduate GP Training. Students that ranked in the lowest score quartile had a higher likelihood of being labelled as underperforming than students in the highest score quartile. Therefore, a low score in the multicomponent proficiency-test could indicate the need for closer guidance and early remediating actions focusing on both cognitive and non-cognitive competencies.