Driving Performance of Teenage Patients With mTBI: a Longitudinal Assessment

3A. Teen driving Motor vehicle crashes (MVCs) are the leading cause of death for teens in the U.S.1 Young people aged 15 to 24 years account for approximately $26 billion of the total costs related to motor vehicle injuries each year.2 In the state of Ohio for 2013, this age group accounted for over 240 deaths with costs exceeding $200 million for medical expenses and lost work time.3,4 Teen drivers aged 16 to 19 years account for much of this morbidity and mortality. Teens are at the highest risk of an MVC compared to any other age group5-per mile driven, they are three times more likely to be in a fatal MVC than drivers 20 years and older.6 Driving an automobile requires a complex set of cognitive, perceptual and motor skills that must be utilized in concert for successful and safe driving performance. Specifically, it involves the simultaneous control of the lateral and longitudinal position of the vehicle in combination with a high amount of visual attention to the roadway, which is necessary to inform decisions and resultant driving actions to manage current and future driving situations.7-9 Driving also requires timely and efficient responses to a complex array of stimuli (e.g., traffic signs and signals), and potential challenges (e.g., pedestrians and other automobiles). Accordingly, the latency of a driver's perceptual-motor responses (i.e., reaction times) to these stimuli and challenges may be critically important for the avoidance of MVCs.

Unfortunately, teen drivers are known to drive faster and allow shorter headways than older, more experienced drivers10. Even more troubling is that teen drivers are more likely to use attention-distracting technologies (e.g., cell phones) while driving11, and are more prone to attentional lapses with two or more passengers in the automobile than experienced adults.10 Thus, given their inattentiveness, teen drivers likely are forced to rely more heavily on reaction times and shorter cognitive and perceptual-motor responses to avoid adverse driving events than experienced drivers.

3B. Driving related deficits in patients with mTBI Closed head injuries such as concussion are becoming more prevalent in teenage populations: Over a recent 10-year period there was a 200% increase in sports related emergency room visits for concussion among teens 14 to 19 years of age.12 Negative sequelae associated with concussion include reduced attention span, longer reaction times, impaired cognitive function and deficits in oculomotor control13-17-all critically important factors for safe, effective driving performance. Moreover, these symptoms tend to linger even after patients are cleared to return to normal activities, and in some cases may persist for several months following the initial injury.18 Very little is known about the severity and duration of perceptual-motor and cognitive deficits related to acute concussion (≤ 1 week from initial injury) in pediatric populations, although some conclusions can be drawn from research on young adult mTBI populations. For example, adults with mTBI exhibit slower reaction times at the time of initial injury and for up to 6-weeks post-injury compared to healthy, matched controls13. Moreover, adult patients exhibited longer performance times on simple reaction-based response and memory tasks even after their physical symptoms resolved and the patients were cleared to resume normal activities14. In fact, in that particular study no patient returned to baseline reaction time performance on the day clearance was given to return to contact sports (M = 4 days). Similarly, a series of studies on high school and collegiate football players demonstrated that both groups exhibited greater cognitive impairment, as indexed by the Standardized Assessment of Concussion (SAC) test for the evaluation of neurocognitive functioning, in the first several days after initial injury.15,19 Specifically, both age groups exhibited below pre-injury baseline performance in orientation, concentration and memory for up to 48 hours after the initial injury; however, cognitive processing was found to have resolved 7 days post-injury for collegiate athletes.15 Disruptions to the neurological networks associated with oculomotor control impair visual task performance in patients with mTBI as well20. For example, adults with mTBI exhibited impaired eye movements17,20 which were manifested as longer, more frequent self-paced saccades and slower reflexive saccades. These impairments gave rise to lower visuospatial accuracy as well as decreased smooth pursuit tracking performance.17,20 Other work has demonstrated that the variability of saccades during working memory tasks correlates with white matter integrity in the brains of patients with mTBI, and indicates that oculomotor behavior may be linked to the degree of diffuse axonal injury in these patients.21-23 Similar oculomotor deficits have been observed in children (6-16 years of age) with acquired brain injury.24 Those deficits were shown to constrain their ability to process high frequency sensory information, making it difficult for the children to utilize anticipation strategies and feedback. Taken in combination, this evolving oculomotor symptom profile may be emblematic of problems these patients have participating in activities of daily living, including driving, and the symptoms that arise (e.g., nausea, dizziness, fatigue) due to participation in such activities.

3C. The profile of driving performance for patients with mTBI is incomplete At present, there is no data on if or when it is safe to allow acutely concussed teenagers to drive. This is a critical gap in the literature. Acute concussion symptoms have the potential to greatly exacerbate the already high-risk driving behaviors exhibited by healthy teenagers. Further, in the absence of return-to-drive criteria, physicians are forced to decide whether a teen is fit to drive following a concussion with limited information. This has the potential to put the teen at higher risk for MVCs and the accompanied costs.

Recently, research has begun to examine the relation between neurocognitive deficits and negative characteristics of driving performance in adults with mTBI.25,26 Patients with mTBI were required to watch videos of genuine traffic scenes filmed from a driver's point of view and to respond with a mouse click to potential traffic hazards as early as possible. The results indicate that patients demonstrated slower responses to traffic hazards in the initial 24 hours after injury compared to matched controls.26 More directly, actual driving performance of adults with mTBI up to 3 days post-injury has been assessed on a driving simulator.27,28 In one such study patients exhibited greater deviation from the center of the driving lane in the absence of deficits in speed management.27 However, a separate study by the same research group found speed management to be more variable, while overall deviation from the center lane was not found to differ between patients and controls.26 While conflicting, the results from both studies indicated improvements in measures of driving performance that positively correlated with improvements in cognitive performance and reaction time (assessed with ImPACT testing), as well as reductions in clinical deficits from three days post-injury (session 1) compared to when patients were cleared to return to normal activities (session 2).27,28 Thus, changes in driving performance in this population are believed to be correlated with changes in cognition and reaction time performance in the first several days after injury. However, the link between these measures and driving performance is not clear.

There are several additional gaps in the current literature on concussion and driving performance. First, while reaction time deficits are known to manifest as longer response latencies to identify adverse risks, the underlying mechanisms that drive such behavior are not well understood. In order to understand how longer response times manifest in this context, it is necessary to examine several testable components of reaction time behavior: (1) adequate and efficient visual attention to the roadway (2) the oculomotor control that drives the visual search, (3) the efficiency of cognitive-perceptual processes to identify the situation as one of "adverse risk", (4) the decision making that dictates whether or not action(s) be taken to avoid the risk, and (5) the motor output that drives the specific action(s) for risk avoidance. Therefore, the identification of response latencies in this population is merely the first step in understanding the underlying mechanisms that lead to any potential changes in their driving performance.

Adequate and efficient visual attention to the roadway is one potential mechanism for slower response times by patients with mTBI. Healthy and experienced adult drivers utilize a strategy called timesharing to safely shift attention to and from the roadway. This strategy involves a set of saccadic eye movements toward and away from the secondary task (e.g., changing a radio station) until it is completed (Figure 1),29,30 and successful timesharing limits the duration of each glance away from the roadway to less than or equal to 2 s29,31(this translates to approximately 160 feet of distance if driving 55 mph). However, 45% of teen drivers exhibit glances greater than or equal to 2.5 s away from the roadway (a critical threshold for MVC risk), compared to only 10% of experienced drivers.32 The investigation of this question therefore requires an analysis of oculomotor performance, and specifically self-paced and reflexive saccade behaviors. This is because patients with mTBI may timeshare equally to their healthy peers, but simply exhibit slow reaction times transitioning out of the timeshare. Alternatively, a lack of control of visual saccades may potentially lead to greater inefficiency in timesharing for these patients, as precious

milliseconds may be spent correcting for inexact self-paced saccades. This behavior might take away from time attending to a given task, and may thus lead to protracted glances away from the roadway. For similar reasons, inefficient self-paced and reflexive saccades may lead to greater delays (slower reaction times) before recognition of an adverse event, decreasing driver response time. Further, inefficiencies in tracking moving objects (smooth-pursuit tracking deficits) may lead to miscalculations in purposeful steering deviations for risk avoidance once an object has been identified, or potentially, poor risk assessment altogether as visual information (e.g., time to contact) may be degraded. Such smooth-pursuit tracking deficits, combined with slower cognitive and processing times in these patients, may ultimately lead to reduced performance capabilities that may be difficult for the patient to overcome.

Importantly, to the investigators knowledge no research has examined the effect of oculomotor control deficits on driving performance for individuals with mTBI. This is surprising given that such oculomotor processes likely underlie perceptual-motor and cognitive behaviors that are critical to safe and effective driving. Also, while standardized assessments of cognition and simple reaction time are sensitive and specific measures of concussion33,34, evidence is lacking as to their relation to driving performance for teenage patients with mTBI. In addition, no research has examined such a profile of driving performance in teenage drivers over the time-course of recovery.