A Multivariate Model to Predict Student Physician Assistant National Certification Exam Performance.

Area of Science:

• Medical Education
• Health Professions Education
• Physician Assistant Studies

Background:

• The Physician Assistant National Certification Exam (PANCE) is critical for assessing graduate medical knowledge.
• Predicting PANCE performance is essential for identifying students needing additional support.
• Early identification allows targeted interventions to improve student success.

Purpose of the Study:

• To evaluate the predictive accuracy of a combination of 10 standardized PA Education Association examinations for first-time PANCE scores.
• To develop and validate a predictive model for PANCE performance.

Main Methods:

• A retrospective analysis of scores from 4 PA program cohorts (n=91) was conducted.
• A multiple regression model was employed to assess the combined predictive power of 10 standardized exams.
• A predictive equation was developed and tested on a subsequent cohort (n=31).

Main Results:

• The multiple regression model demonstrated statistical significance (ANOVA, P < 0.0005).
• A strong coefficient of multiple correlation (R = 0.86) indicated high predictive accuracy.
• The model effectively predicted first-time PANCE scores.

Conclusions:

• The developed multiple regression model reliably predicts first-time PANCE scores.
• This provides validity for using standardized PA Education Association examinations for content assessment.
• Implementing this model can identify at-risk students, contributing to a 100% first-time pass rate in the latest cohort.