Do Concussion Assessments Predict Simulated Driving Performance Following Concussion?

Individuals with concussion display simulated driving patterns that are associated with motor vehicle crash risk. Since most clinicians do not have the capability of directly assessing driving performance, nor is this clinically feasible, identifying key clinical predictors of post-concussion driving performance would positively inform clinical recommendations. PURPOSE: To identify widely used concussion assessments that uniquely predict simulated driving performance among concussed individuals, relative to controls. METHODS: College students within 72 hours of concussion (n=27, 19.0 ± 0.2 years) and non-concussed controls (n=46, 19.9 ± 0.3 years) were recruited. We focused on the following outcomes: BESS total score, Standardized Assessment of Concussion (SAC) total score, symptom severity, and tandem gait completion time. The following simulated driving outcomes were included (all count variables): total collisions, speed exceedances, stop signs missed, centerline crossings, and road edge excursions. Separate generalized linear mixed regression models were conducted with group, assessment, and group x assessment interaction effects and fit using a Poisson distribution. This abstract focuses on interactions. RESULTS: There was an interaction between group and SAC for speed exceedances (p=0.045). Each additional correct response on SAC was associated with 10.00% (95% CI: 1.00-1.21) increase in the number of speed exceedances in concussion group relative to controls. There was an interaction effect between group and tandem gait for the centerline crossings (p=0.001). Each additional second taken to complete the tandem gait task was associated with 6.40% (95% CI: 1.03-1.11) increase in the number of centerline crossings in concussion group relative to controls. There were no other statistically significant interactions (p>0.05). CONCLUSIONS: Counter to our hypothesis, better SAC performance was associated with an increase in speed exceedances. This may indicate that concussed individuals with better mental status had more aggressive driving behaviors. Tandem gait may capture post-concussion driving impairments. Further study is needed to determine whether other concussion assessments (e.g. neurocognitive, vestibulo-ocular) can better represent driving impairments. The research reported in the abstract was supported by Andee's Army Foundation and the National Center for Advancing Translational Sciences of the NIH under Award Number UL1TR002378. The content is solely the responsibility of authors and does not necessarily represent the official views of the NIH.