Purpose
The purpose of this research was to demonstrate the accuracy of a statistical model using binary logistic regression to predict the results (Pass or Fail) of medical students on the USMLE Step 1 through the 2-digit results of the Comprehensive Basic Science Examination (CBSE) developed by the National Board of Medical Examiners (NBME).

Method
A quantitative methodology based on applied statistics was implemented with binary logistic regression in a prediction correlation design. This research has gathered data from 198 students who have taken USMLE Step 1 with Pass or Fail results and the CBSE with 2-digit results since the changes to both exams began. In addition, it was analyzed in which month (January, May, or June) the medical students take the CBSE there is a better prediction of the USMLE Step 1 result. To implement these methods of applied statistics, version 29.0 of IBM SPSS Statistics was used.

Results
The predicted results using the binary logistic regression reflected that the institution's medical students with a score of 56 or more in the CBSE have a probability of passing (Pass) USMLE Step 1 of 90%. However, the model's prediction in 2023 with a May CBSE score of 51 reflected an accuracy of 88% passing the STEP in the assessed students. Another finding achieved with this research is that the January and May scores on the CBSE predict USMLE Step 1, while those from June do not.

Conclusion
By analyzing the influence of CBSE scores in USMLE Step 1 pass/fail rates, this research aims to contribute to the existing body of literature on medical education assessment practices and assist educators in making informed decisions regarding students' exam preparation. Furthermore, the findings from this research could influence curriculum development and student support strategies to enhance academic success in medical education.