Submitted by Brent HARPER* (1), Alex SIYUFY (1), Julia CASTLEBERRY (1), Angela MICKLE (2), Kristen JAGGER (1), Andrew, WAFF (3), Kenneth COX (4)

(1) Department of Physical Therapy, Radford University, Radford, VA (USA)
(2) Department of Health and Human Performance, Radford University, Radford, VA (USA)
(3) Athletic Trainer, Radford High School, Radford, VA (USA)
(4) Department of Communication Sciences and Disorders, Radford University, Radford, VA (USA)

*Corresponding Author – Brent HARPER, Radford University, 101 Elm Av SE, Roanoke, VA 24013 (USA

ABSTRACT
Identifying kids playing American football who have suffered significant head impacts is vital to ensuring the safety of the athlete and to providing a safe environment within which they can play and be monitored. There are multiple technologies available, but they may be prohibitively expensive for the average non-professional recreational league or high-school athlete. This paper is a clinician’s perspective of an attempt to monitor concussive and sub-concussive head impacts using a commercially available head impact monitor device.

KEYWORDS:concussion, concussion monitoring technology, head impact, safety

INTRODUCTION
Concussions continue to be a major issue in youth sports. Concussion rates for youth sports such as football, soccer, lacrosse and hockey have been well documented (10,18,25). A subcategory of mild traumatic brain injury (mTBI) with an incidence of 300,000 to 3.8 million (12,15), concussion is a serious public health issue (21), especially for younger athletes (12) and one that must be addressed in health promotion and injury prevention. High school male football and female soccer athletes have the highest incidences (7,14,17,18,22,26). Equipment manufacturers have responded to the demand for safer helmets by improving primarily American football helmets for both youth and adults. In addition, companies have developed detection systems for monitoring the number and severity of impacts to the head (6,25,28,30).

One such device is the Reebok CHECKLIGHT™. The Reebok CHECKLIGHT™ is comprised of a skull cap with a sensory array worn underneath the helmet. It has a simple green, yellow or red light alerting system to warn of a possible concussive event, but it is not a diagnostic tool. According to the manufacturer, the product is designed to objectively measure impact force and acceleration to the head, with higher forces triggering a red light, and lesser forces triggering a yellow light. Triggering the light system does not mean that a concussion has occurred, but rather the alert indicates that the device has detected a certain amount of force to the head.

There is a need for objective, validated and rapid sideline tests guiding remove-from-play decisions (4-6,11,28). The purpose of this article is to discuss the viability of using of the Reebok CHECKLIGHT™ system in two levels of youth football: A Parks and Recreation youth league football team and a high school football team. The article will discuss lessons learned while using the system, pitfalls and benefits of the system, and the author’s recommendations for individuals and/or teams who are contemplating the use of this type of technological device.

Study Synopsis
Methods
This study was approved by a University Institutional Review Board and informed consent was obtained from each participant. Sixteen subjects were recruited from a Virginia High School junior varsity (JV) and varsity football team (15 male, 1 female). Age range for the JV team was 13-15 years and varsity was 15-18 years. Thirteen subjects were recruited from the Parks and Recreation youth league football team. Age range for these subjects was 11-13 years.

During this study, each subject was fitted, free of charge, with a Reebok CHECKLIGHT™ system, which was commercially available for $150 dollars at the time of this study. The system consists of two components: a soft skull cap made of lightweight fabric and sensor which inserts into the skull cap. Each sensor was numbered so that it was consistently used by the same individual. Skull caps were sized small, medium, or large and, while each athlete got their appropriate sized skull cap, they did not get the same skull cap each day. The devices were worn under their helmets during all contact practices and competitions and were monitored by a clinically licensed health care provider researcher or a certified athletic trainer.

Data for the JV and varsity participants was collected during 12 contact practices, 1 scrimmage, and 3 games. Data for the recreation league was collected during 16 contact practices, 1 scrimmage, and 7 games. Baseline metrics obtained pre-season included standardized assessment of concussion (SAC) (2,19,20), King-Devick (K-D) (8,9,13,16), and modified balance error scoring systems (BESS) (23,24,29). A protocol was developed and implemented by the researchers when a yellow or red light was triggered. If the athlete demonstrated standard and/or obvious signs of concussion they were immediately assessed by a health care provider. The protocol flow was as follows: The athlete was first asked three basic questions regarding orientation depending if it occurred during practice or game situation (i.e. What day is it? Who are we playing? What quarter is it?). Then the athlete performed the “H” test for visual tracking and was observed for aberrant eye movements. If the participant failed these initial tests, he or she performed the K-D test. If the participant failed the K-D test, he or she would be re-tested and re-assessed with the SAC and modified BESS tests unless displaying obvious signs and symptom of a concussive event.

Results
The Reebok CHECKLIGHT™ was activated 32 times during the JV/Varsity season, which was higher than predicted, considering concussion rates in previous research (χ2= 8450.730, p>0.05) (10). There were a total of 22 lights triggered for the Rec League team (20 yellow, 2 red). There were 16 contact practices, 1 scrimmage and 7 games monitored with no concussive events.

Discussion
American football, by nature, is a violent game with multiple sub-concussive impacts over the course of a competitive season. One method used to estimate concussion rates is the athletic exposure (AE) rate, which is defined as one participating athlete in one practice or competition event. Using this calculation, the average American football high school rate in 2010 was 2.3 to 2.5 per 10,000 AE. Analyzed in isolation, the rate of concussion in games ranged from 5.3 to 6.4 per 10,000 AE and in practice, 1.1 per 10,000 AEs (10,18). Since time played in practice and games varies individually, Guskiewicz et al. (11) evaluated concussion risk per athletic exposure hours (AEH). Their calculated concussion risk for male high school American football athletes was 10.3 per 10,000 AEH.

Not all impacts result in obvious concussion, so Young et al. (30) evaluated two years of impacts using Head Impact Telemetry (HIT™) data in 7 to 8 year old male American football players (n=19). The average head impact per game was 11 ± 11 and per practice was 9 ± 6. The authors indicated that Pop Warner American football athletes seem to experience a similar magnitude of head impacts as high school and collegiate athletes. Cobb et al. (6) also evaluated HIT™ data, this time in boys aged 9 – 12 years old playing American football. The average number of impacts was 240 per season, compared to 565 for high school players. The authors did not find statistical significance of impact when correlated to concussive events, however they still recommended minimizing head impact exposure. Based on these and other studies (6,25,28,30), the collective understanding for the most effective methods to mitigate concussive events and to limit head impact exposure in American football include teaching proper technique and limiting practice sessions.

Talavage et al. (27) followed eleven high school football players, age 15 to 19 years through a ten week competitive season. They discovered three categories of participants: those with no clinically diagnosed concussion, those with a clinically diagnosed concussion, and those who demonstrate no clinical concussion symptoms but with quantifiable cognitive deficits despite being asymptomatic. This study is important because it provides quantifiable data that sub-concussive impacts can result in altered deleterious brain function despite a lack of standard concussion symptoms. A research review written by Bailes et al. (1) discussed the possible harmful role that repetitive sub-concussive impacts may have on brain and neurological function leading to both short- and long-term deficits. Repetitive sub-concussive impacts place one at risk for chronic neurodegenerative syndromes. These findings and their potential implications support the implementation of limiting practice in American football in an attempt to decreased head impact exposure. This is especially important as we currently do not know enough about the negative long-term effects of repetitive mTBI.

Breedlove et al. (3) compared functional magnetic resonance imaging (fMRI) from two competitive seasons of thirty-nine American football high school males between the ages of 14 to 18 years. They found a significant relationship between neurophysiological changes in the brain and repetitive blows to the head in those who were not diagnosed with a concussion and who were asymptomatic. This study reinforces the hypotheses that repetitive and cumulative sub-concussive blows to the head result in pathological neurophysiological findings.

Challenges of Head Impact Indicator Usage
Introduction
One major challenge for clinicians and health care providers is to monitor hits that may have the potential for causing concussion. Manufacturers have begun to develop devices to assist with monitoring these activities, and the Reebok CHECKLIGHT™ was used for this study. There were several issues that presented themselves when using the lights with the team which can be broken down in to three categories: threshold issues with the lighting system; light usefulness; and set-up, application and maintenance.

Threshold Issues
Rebook would not share their light activation threshold data with our research team. However, they did indicate that field studies were done in their laboratories and with sports teams when setting those thresholds. The four parameters that are measured by their device are 1. Linear acceleration, 2. Rotational acceleration, 3. Duration of impact, and 4. Location of impact on the head. These four factors are then calculated to provide a triggered yellow light for a “moderate” hit, and a red light for a “severe” hit.

As noted above, 32 lights were triggered for the JV/Varsity group. All but one tested negative during the initial sideline concussion testing, which included the “H” test and three cognitive questions. That one positive, initially indicated by the helmet as a yellow light impact, also failed the cognitive questions, “H” test, and the King-Devick test. Because the concussion was overtly obvious, the researchers stopped the protocol as the trainer referred the athlete to the local emergency department that later confirmed the concussion with a medical diagnosis. None of the 22 lights triggered for the recreation league group resulted in a concussion and none yielded a failure of the sideline screening protocol.

The HIT™ data, which measures forces and counts the number of real-time head impacts utilizes accelerometers to measure linear and rotational acceleration forces and records the number of sub-concussive impacts (6,28,30). Previous studies have identified that athletes are at higher risk of suffering a concussion from linear acceleration forces great than 96 g’s and rotational acceleration forces greater than 5,500 rad/s2 (4,5). Young et al. (30) studied 7 – 8 year olds and had median linear accelerations of 16 ± 2 g’s and rotational acceleration of 686 ± 169 rad/s2, eleven of which had linear acceleration forces greater than 80 g’s. None were diagnosed with concussion. Cobb et al. (6) studied 9 – 12 year olds and identified the median linear accelerations as 18 ± 2 g and rotational accelerations as 856 ± 135 rad/s2. The 95th percentile impacts sustained had linear accelerations of 43 ± 7 g and rotational accelerations of 2034 ± 361 rad/s2. Only 36 head impacts (0.3%) were greater than 80 g’s meaning the average player sustained 0.7 ± 1.2 impacts greater than 80 g’s. Four athletes were diagnosed with concussion within the range of linear accelerations from 26 to 64 g’s and rotational accelerations from 1552 to 4548 rad/s2. Urban et al. (28) collected head impact information from 39 male athletes between the ages of 14 to 18 years during a single season of high school American football using HIT™ data. The median linear head acceleration value was 21.9 g’s and the median rotational head acceleration was 973 rad/s2. The linear acceleration of 21.9 g’s is higher than the collegiate level of 18 g’s, which suggests that higher severity of impacts occurs at the high school level. During the 14 games there were 16,502 impacts, 76 (0.46%) of which were above the linear acceleration value of 98 g’s, which has been associated with concussion.

Reebok advertises this device for use in players 10 years and older. It is possible that, in and effort to cover the average player aged 10 and up, Reebok chose a relatively low trigger threshold. Our observations suggest that the CHECKLIGHT™ threshold may be set considerably lower than those average linear and rotational forces experienced by high school players. The threshold settings are probably consistent for those thresholds experienced by the recreational level football player.

It became apparent almost immediately that there were issues with the light systems at the high school practices. Lights would be triggered (in yellow mode) for situations where it seemed clear they should not. One author (AM) noted lights being triggered when an athlete came off the field for a water break, took off his cap placed it in his helmet and set the helmet on the ground to get water.
Due to the inconsistent findings on the sidelines during the study, one of the researchers (BH) collaborated with the Director of a Gait Analysis Laboratory (KJ) in the Department of Physical Therapy to test a hypothesis that the Reebok CHECKLIGHT™ device was not consistently or accurately representing impacts on the field.

An AMTI© OR6-7 force plate was utilized to assess the forces associated with dropping a team helmet on its crown from 1 feet, 2 feet, 3 feet, and 4 feet onto the bare force plate and onto a stack of towels (up to 14 layers to represent softer earth or player’s bodies). Helmet conditions included the following: Condition A) the helmet with the CHECKLIGHT™ sensor and skull cap stretched over an 8-lb. sandbag that represented a head within the helmet, Condition B) the helmet with the CHECKLIGHT™ sensor and skull cap sitting loosely in the helmet (as if dropped into the helmet by the player while on the sidelines), and Condition C) the helmet with the CHECKLIGHT™ sensor taped externally to the posterior aspect of the helmet.

In spite of vertical ground reaction forces peaking at 6450N (1450 lbs.) during drops onto 14 layers of toweling from a 4-foot height under helmet Condition A, the CHECKLIGHT™ failed to trigger any yellow or red lights. When the sandbag was removed from the CHECKLIGHT™ skull cap and the 7-lb. helmet (Conditions B & C) was dropped from 3 ft. and 4 ft. heights onto a bare force plate, intermittent yellow and red lights began to ensue. Peak vertical forces during these conditions ranged from 3800N (854 lbs.) to 5450N (1225 lbs.), but there was no discernable pattern between peak forces and light color or activation. Further assessment of data patterns revealed that the CHECKLIGHT™ sensor was triggered most consistently when a rapid oscillation between positive and negative forces in the X and Y planes was recorded. These rapid oscillations occurred within milliseconds of contact when the helmet was rebounding off of the force plate. Again, the magnitude of force did not appear to play a role in the color or activation of the light; only rapid oscillations resulted in activation of the sensor light.

These findings are consistent with the findings on the sidelines as well as with the technology incorporated. The CHECKLIGHT™ was more likely to display an activated yellow or red light when the un-weighted helmet was dropped onto a bare force plate (simulating an athlete dropping a helmet while standing on the sidelines) than when large forces were recorded during a weighted simulation of a head making an impact with layers of toweling at the crown (simulating contact during play). Because the CHECKLIGHT™ uses an accelerometer, rapid changes in direction of motion are necessary to reach the threshold required to trigger the sensor activation. In our basic and admittedly unrefined laboratory testing, shear and torsional accelerations/decelerations far exceeded the vertical decelerations associated with impact. As such, the CHECKLIGHT™ sensor appears to be adept at signaling motions associated with rapid alternating changes in velocity (similar to whiplash events), while not being capable of identifying forceful impacts (blunt force between players).

Light Usefulness
There were several challenges for clinicians monitoring the CHECKLIGHT™ devices that should be taken into account when using them. The LED lights are extremely difficult to see when the player is on the field. The lights could not be visualized adequately in direct sunlight or under the lights of the playing field during night games. This meant that monitoring the players’ impacts could only be done when the players came off of the field (during competition). As a result, those players whose lights were triggered could not be screened at the time of the head impact, but rather when the player left the field of play for the sideline.

Another challenge experienced by the researchers was how the athlete handled the device if it was not on his/her head. If the player inadvertently removed the cap during water breaks, for example, and dropped the cap with the sensor, then the light was often triggered and had to be reset. Each time the player entered a practice session or competition, the CHECKLIGHT™ had to be confirmed green, indicating it was functional.

Once a light was triggered, the cap had to be removed from the player, the sensor had to be completely removed from the cap and then the sensor was reset. The reset method was supposed to take seconds, but often it took much longer. At times it took multiple attempts to reset the light, resulting in several minutes of inactivity for the player. This was extremely frustrating for the player and the coach, and is a major deterrent for use during a game.

Device Management
Another issue with the CHECKLIGHT™ system is that it has to be recharged using a plug similar to a cell phone recharger. To do this, each sensor had to be removed from the skull cap, the skull cap had to be turned off and then plugged in for a minimum of 4-5 hours for a full charge. When doing this for a team, this presented several challenges. First, special large sized surge protectors had to be purchased to accommodate all of the CHECKLIGHT™. These were expensive and required an entire counter of athletic training room space. In addition, the total time required to disassemble each skull cap, turn it off, and plug it in was approximately 2 minutes each. Skull caps were washed each day and could not be dried in the dryer. Thus after being washed they had to be hung out to dry.

Before each practice the sensor had to be unplugged and then turned on. It then had to be inserted into the proper sized skull cap to be picked up by the appropriate athlete. We found that the easiest way to the do this was by laying them out by jersey number on the bench. The time to assemble each skull cap and lay them out for the athletes was approximately 4 min each since we had to ensure that each sensor and skull cap size was appropriate.

While the total time required to set up and breakdown an individual CHECKLIGHT™ is not prohibitively long, the time added up when dealing with an entire team. One person was essentially tied up for 45 minutes to 1 hour, both right before and directly after practice. This is, traditionally, a very busy time in any athletic training facility.

CONCLUSIONS
The Reebok CHECKLIGHT™ system is an instrument designed to help sideline personnel detect forces that might predict concussion in athletes. Our experience is that the system has limitations for use with athletes at the high school level, a fact that may relate to the threshold level preset by the company. In addition, the light design does not allow review by sideline personnel until after the athlete exits the field, which limits its usefulness for preventing continued play following a potentially concussive force. This also limits the ability to monitor cumulative sub-concussive impacts. Furthermore, the light’s propensity to spontaneously turn off limits its usefulness in many situations.

APPLICATIONS IN SPORT
The time required to breakdown the light components for recharging and washing at the end of a practice would not allow the system to be easily incorporated into many teams’ practice and game routines. The equipment required to recharge the sensors is expensive and the space required is enough to be bothersome in all but the largest athletic training facility. Furthermore, personnel time required is significant during some very busy times.
Concussive events and cumulative sub-concussive impacts should be addressed in youth athletics in conjunction with various cognitive and system specific assessments in order to make participation in youth sports as safe as possible. One piece of equipment to measure head impacts is electronic devices that specifically monitor and measure these forces. Our feeling is that the CHECKLIGHT™ system, may be beneficial to individual athletes below the high school level, but has limited usefulness for entire teams and/or high school athletes.

ACKNOWLEDGEMENTS
None

REFERENCES
1. Bailes, J. E., Petraglia, A. L., Omalu, B. I., Nauman, E., & Talavage, T. (2013). Role of subconcussion in repetitive mild traumatic brain injury. Journal of Neurosurgery, 119(5), 1235–1245. http://doi.org/10.3171/2013.7.JNS121822

2. Barr, W. B., & McCrea, M. (2001). Sensitivity and specificity of standardized neurocognitive testing immediately following sports concussion. Journal of the International Neuropsychological Society: JINS, 7(6), 693–702.

3. Breedlove, E. L., Robinson, M., Talavage, T. M., Morigaki, K. E., Yoruk, U., O’Keefe, K., … Nauman, E. A. (2012). Biomechanical correlates of symptomatic and asymptomatic neurophysiological impairment in high school football. Journal of Biomechanics, 45(7), 1265–1272. http://doi.org/10.1016/j.jbiomech.2012.01.034

4. Broglio, S. P., Eckner, J. T., Surma, T., & Kutcher, J. S. (2011). Post-concussion cognitive declines and symptomatology are not related to concussion biomechanics in high school football players. Journal of Neurotrauma, 28(10), 2061–2068. http://doi.org/10.1089/neu.2011.1905

5. Broglio, S. P., Schnebel, B., Sosnoff, J. J., Shin, S., Fend, X., He, X., & Zimmerman, J. (2010). Biomechanical properties of concussions in high school football. Medicine and Science in Sports and Exercise, 42(11), 2064–2071. http://doi.org/10.1249/MSS.0b013e3181dd9156

6. Cobb, B. R., Urban, J. E., Davenport, E. M., Rowson, S., Duma, S. M., Maldjian, J. A., … Stitzel, J. D. (2013). Head impact exposure in youth football: elementary school ages 9-12 years and the effect of practice structure. Annals of Biomedical Engineering, 41(12), 2463–2473. http://doi.org/10.1007/s10439-013-0867-6

7. Daneshvar, D. H., Nowinski, C. J., McKee, A. C., & Cantu, R. C. (2011). The epidemiology of sport-related concussion. Clinics in Sports Medicine, 30(1), 1–17, vii. http://doi.org/10.1016/j.csm.2010.08.006

8. Duenas, M., Whyte, G., & Jandial, R. (2014). Sideline concussion testing in high school football on Guam. Surgical Neurology International, 5, 91. http://doi.org/10.4103/2152-7806.134364

9. Galetta, M. S., Galetta, K. M., McCrossin, J., Wilson, J. A., Moster, S., Galetta, S. L., … Master, C. L. (2013). Saccades and memory: baseline associations of the King-Devick and SCAT2 SAC tests in professional ice hockey players. Journal of the Neurological Sciences, 328(1-2), 28–31. http://doi.org/10.1016/j.jns.2013.02.008

10. Gessel, L. M., Fields, S. K., Collins, C. L., Dick, R. W., & Comstock, R. D. (2007). Concussions among United States high school and collegiate athletes. Journal of Athletic Training, 42(4), 495–503.

11. Guskiewicz, K. M., Weaver, N. L., Padua, D. A., & Garrett, W. E., Jr. (2000). Epidemiology of concussion in collegiate and high school football players. The American Journal of Sports Medicine, 28(5), 643–650.

12. Halstead, M. E., Walter, K. D., & Council on Sports Medicine and Fitness. (2010). American Academy of Pediatrics. Clinical report–sport-related concussion in children and adolescents. Pediatrics, 126(3), 597–615. http://doi.org/10.1542/peds.2010-2005

13. King, D., Brughelli, M., Hume, P., & Gissane, C. (2014). Assessment, management and knowledge of sport-related concussion: systematic review. Sports Medicine (Auckland, N.Z.), 44(4), 449–471. http://doi.org/10.1007/s40279-013-0134-x

14. Kirkwood, M. W., Yeates, K. O., & Wilson, P. E. (2006). Pediatric sport-related concussion: a review of the clinical management of an oft-neglected population. Pediatrics, 117(4), 1359–1371. http://doi.org/10.1542/peds.2005-0994

15. Leddy, J. J., Sandhu, H., Sodhi, V., Baker, J. G., & Willer, B. (2012). Rehabilitation of Concussion and Post-concussion Syndrome. Sports Health, 4(2), 147–154. http://doi.org/10.1177/1941738111433673

16. Leong, D. F., Balcer, L. J., Galetta, S. L., Liu, Z., & Master, C. L. (2014). The King-Devick test as a concussion screening tool administered by sports parents. The Journal of Sports Medicine and Physical Fitness, 54(1), 70–77.

17. Lincoln, A. E., Caswell, S. V., Almquist, J. L., Dunn, R. E., Norris, J. B., & Hinton, R. Y. (2011). Trends in concussion incidence in high school sports: a prospective 11-year study. The American Journal of Sports Medicine, 39(5), 958–963. http://doi.org/10.1177/0363546510392326

18. Marar, M., McIlvain, N. M., Fields, S. K., & Comstock, R. D. (2012). Epidemiology of concussions among United States high school athletes in 20 sports. The American Journal of Sports Medicine, 40(4), 747–755. http://doi.org/10.1177/0363546511435626

19. McCrea, M. (2001). Standardized Mental Status Testing on the Sideline After Sport-Related Concussion. Journal of Athletic Training, 36(3), 274–279.

20. McCrea, M., Kelly, J. P., Randolph, C., Kluge, J., Bartolic, E., Finn, G., & Baxter, B. (1998). Standardized assessment of concussion (SAC): on-site mental status evaluation of the athlete. The Journal of Head Trauma Rehabilitation, 13(2), 27–35.

21. Meehan, W. P., 3rd, & Bachur, R. G. (2009). Sport-related concussion. Pediatrics, 123(1), 114–123. http://doi.org/10.1542/peds.2008-0309

22. O’Kane, J. W., Spieker, A., Levy, M. R., Neradilek, M., Polissar, N. L., & Schiff, M. A. (2014). Concussion among female middle-school soccer players. JAMA Pediatrics, 168(3), 258–264. http://doi.org/10.1001/jamapediatrics.2013.4518

23. Rahn, C., Munkasy, B. A., Barry Joyner, A., & Buckley, T. A. (2015). Sideline Performance of the Balance Error Scoring System during a Live Sporting Event. Clinical Journal of Sport Medicine: Official Journal of the Canadian Academy of Sport Medicine, 25(3), 248–253. http://doi.org/10.1097/JSM.0000000000000141

24. Riemann, B. L., & Guskiewicz, K. M. (2000). Effects of mild head injury on postural stability as measured through clinical balance testing. Journal of Athletic Training, 35(1), 19–25.

25. Rowson, S., & Duma, S. M. (2013). Brain injury prediction: assessing the combined probability of concussion using linear and rotational head acceleration. Annals of Biomedical Engineering, 41(5), 873–882. http://doi.org/10.1007/s10439-012-0731-0

26. Schneider, K. J., Meeuwisse, W. H., Kang, J., Schneider, G. M., & Emery, C. A. (2013). Preseason reports of neck pain, dizziness, and headache as risk factors for concussion in male youth ice hockey players. Clinical Journal of Sport Medicine: Official Journal of the Canadian Academy of Sport Medicine, 23(4), 267–272. http://doi.org/10.1097/JSM.0b013e318281f09f

27. Talavage, T. M., Nauman, E. A., Breedlove, E. L., Yoruk, U., Dye, A. E., Morigaki, K. E., … Leverenz, L. J. (2014). Functionally-detected cognitive impairment in high school football players without clinically-diagnosed concussion. Journal of Neurotrauma, 31(4), 327–338. http://doi.org/10.1089/neu.2010.1512

28. Urban, J. E., Davenport, E. M., Golman, A. J., Maldjian, J. A., Whitlow, C. T., Powers, A. K., & Stitzel, J. D. (2013). Head impact exposure in youth football: high school ages 14 to 18 years and cumulative impact analysis. Annals of Biomedical Engineering, 41(12), 2474–2487. http://doi.org/10.1007/s10439-013-0861-z

29. Valovich, T. C., Perrin, D. H., & Gansneder, B. M. (2003). Repeat Administration Elicits a Practice Effect With the Balance Error Scoring System but Not With the Standardized Assessment of Concussion in High School Athletes. Journal of Athletic Training, 38(1), 51–56.

30. Young, T. J., Daniel, R. W., Rowson, S., & Duma, S. M. (2013). Head Impact Exposure in Youth Football: Elementary School Ages 7-8 Years and the Effect of Returning Players. Clinical Journal of Sport Medicine: Official Journal of the Canadian Academy of Sport Medicine. http://doi.org/10.1097/JSM.0000000000000055