Date of Graduation


Document Type


Degree Name

Master of Athletic Training (MAT)

Degree Level



Health, Human Performance and Recreation


Jeffrey A. Bonacci

Committee Member

Brendon P. McDermott

Second Committee Member

Stephen W. Dittmore

Third Committee Member

Gary B. Wilkerson


Health and environmental sciences; Core stability; Injury prediction; Injury prevention; Injury risk; Predictive modeling; Reaction time


Context: Various intrinsic factors such as high exposure, poor endurance of core muscles, previous injury, strength deficits, suboptimal neurocognitive function, and orthopedic abnormalities have been found as predictors for sprains and strains among collegiate football players. Objective: Assess the applicability of pre-participation assessments as predictors of core or lower extremity injury. Design: Cohort Study. Setting: National Collegiate Athletic Association Division I football program. Patients or Other Participants: Athletes who underwent mandatory pre-participation examinations before preseason football training over two consecutive seasons (n=225). Main Outcome Measure(s): Associations between preseason protocols and injury incidence for core and lower extremity injuries were established for 225 players using three different injury definitions; all injuries reported (ALL), limited participation (LP), and removed (OUT). Receiver operating characteristic analysis was used to establish cut-points that classified cases as high-risk or low-risk for injury incidence. Logistic regression and Cox regression analyses were used to identify a multivariable prediction model for injury. Results: A 4-factor model (FM) for ALL identified ≥2 Positive Factors for differentiating between injured and uninjured athletes (P<.001, OR=3.21; 90% CI 1.98, 5.20 , Sens=77.3%, Spec=48.5%). A 3-FM for LP identified ≥1 Positive Factors to be the criteria (P<.004, OR=2.41; 90% CI 1.41, 4.10, Sens=82.8%, Spec=33.3%). A 3-FM identified ≥2 Positive Factors for OUT to be the criteria (P<.012, OR=2.25; 90% CI 1.27, 4.00, Sens=75.4%, Spec=42.3%). A 4-FM identified =4 Factors to be the standard for injury in the previous season (P<.001, OR=8.61; 90% CI 4.00, 18.53, Sens=58.5%, Spec= 85.9%). A 4-FM identified ≥3 Factors for subsequent injuries during both years (P<.011, OR=8.40; 90% CI 2.00, 35.70, Sens=44.4%, Spec=91.3%). Conclusions: Injury definition appears to be important for identifying risk factors for football injuries. Additionally, there are modifiable risk factors that can be determined from previous season injury and for athletes who are injured in consecutive years.

Key Words: injury prediction, injury prevention, injury risk, Core Stability, Reaction Time, Predictive Modeling