Predicting COMLEX-USA Level 2-CE using medical school performance and use for student advising

Area of Science:

• Medical Education
• Licensing Examinations
• Student Assessment

Background:

• COMLEX-USA Level 1 transitioned to Pass/Fail, increasing reliance on Level 2-CE scores for residency program selection.
• Developing predictive models for Level 2-CE performance is essential for medical student success.

Purpose of the Study:

• To evaluate the predictive value of various performance measures on COMLEX-USA Level 2-CE outcomes.
• To establish a predictive model and identify optimal cutoff scores for advising students on Level 2-CE preparation.

Main Methods:

• Retrospective analysis of 948 medical students' performance data.
• Correlational and multiple regression analyses to build a predictive model.
• Utilized pre-admission (MCAT, GPA), pre-clerkship, national exam (COMAT, COMSAE Phase 2, COMLEX Level 1 P/F), and clinical evaluation scores.

Main Results:

• A predictive model incorporating average COMAT scores, COMSAE Phase 2, pre-clerkship exam average, and COMLEX Level 1 Pass/Fail status was developed.
• This model explained 73.9% of the variance in Level 2-CE performance.
• Optimal cutoff scores were identified for passing (COMSAE=447, COMAT=94.4) and high performance (COMSAE=650, COMAT=700) on Level 2-CE.

Conclusions:

• The study validates previous findings and highlights the predictive utility of COMSAE Phase 2 for Level 2-CE outcomes.
• A robust predictive model combining multiple performance measures offers guidance for student advising.
• Established optimal cutoff scores aid institutions in supporting students toward achieving COMLEX Level 2-CE success and residency goals.