By Leo Lim Class of 2025

Multivariate Analysis and Prediction of the Risk Factors for Concussion in High School Students

Editor’s Note: The Taft STEM Research Fellowship is a yearlong, advanced study program for students pursuing interdisciplinary STEM research beyond the classroom. Fellows work closely with faculty and expert mentors, collaborate with peers, and apply their research to real-world problems, culminating in a public presentation to faculty and field professionals. The course combines independent scholarship with structured support and offers opportunities to explore research that bridges multiple STEM disciplines.

Abstract

Concussion, also known as mild Traumatic Brain Injury, is prevalent among athletes, with short and long-term consequences impacting neurological health and athletic performance. Current diagnostic methods rely on subjective measurements, leading to high rates of undiagnosed concussions, and advanced methods that are highly inefficient and expensive. This study thus aims to develop a multivariate logistic regression model to enhance early detection and risk assessment of adolescent concussions. Using data from federal repositories, such as FITBIR and Normative Athlete Data, 19 risk factors were identified using a logistic regression model through the programming language R. Then, with the identified factors, a logistic regression model was designed, trained, and tested using a 7:3 train-test ratio. To validate the model, a calibration plot and a Receiver Operating Characteristic (ROC) curve were used. The Area Under the Curve (AUC) of 0.7045 indicates acceptable predictive performance. As the final product, the model was implemented into a ShinyApp-based website, allowing athletes to readily input data and receive concussion probability estimates.

Poster

TO LEARN MORE, CLICK HERE TO VIEW PDF POSTER.