Early Identification of Students at Risk for Academic Difficulty in a Doctor of Physical Therapy Program.

Purpose: Physical therapy education programs use preadmission factors to admit qualified applicants. These factors have limited ability to predict academic outcomes and 5% of enrolled students do not graduate. The purpose of this study was to investigate whether early assessment scores in a Human Gross Anatomy course can be used to identify students at greater risk of academic difficulty.
Methods: This is a retrospective analysis of data from 272 students enrolled in a Doctor of Physical Therapy program 2011-2013 and 2015-2019. Independent variables included scores on assessments in a Human Gross Anatomy course. Dependent variables included course score and first-year GPA. Receiver operating characteristic (ROC) curves were generated to evaluate each assessment's ability to differentiate between students who did and did not have academic difficulty and determine cutoff scores.
Results: 4% and 11% of students were found to have academic difficulty in the course and the program, respectively. Practical Exam #2 (AUC 0.95, 95% CI 0.89-1.00, p<0.001) best differentiated between students who did and did not have academic difficulty. The calculated cutoff score of 61.5% had the same sensitivity (90.91%) and greater specificity (91.95%) than the standard passing score of the program of (sensitivity 90.91%, specificity 72.41%). Scoring < 61.5% on Practical Exam #2 increased the odds of academic difficulty in the course and first year of the program.
Conclusions: This study demonstrated a method that could be used to identify students at greater risk of academic difficulty before submission of any course grades. Using this evidence-based approach can benefit students and programs.