a b s t r a c t

Objective: Concussions and chronic traumatic encephalopathy (CTE) related to professional football has received much attention within emergency care and sports medicine. Research suggests that some of this may be due to a greater likelihood of initial helmet contact (IHC), however this association has not been studied across all age groups. This study aims to investigate the association between player age and IHC in American football.

Methods: Retrospective review of championship games between 2016 and 2018 at 6 levels of amateur tackle foot- ball as well as the National Football League (NFL). Trained raters classified plays as IHC using pre-specified criteria. A priori power analysis established the requisite impacts needed to establish non-inferiority of the inci- dence rate of IHC across the levels of play.

Results: Thirty-seven games representing 2912 hits were rated. The overall incidence of IHC was 16% across all groups, ranging from 12.6% to 18.9%. All but 2 of the non-NFL divisions had a statistically reduced risk of IHC when compared with the NFL, with relative risk ratios ranging from 0.55-0.92. IHC initiated by defensive partic- ipants were twice as high as offensive participants (RR 2.04, p < 0.01) while 6% [95% CI 5.4-7.2] of all hits were helmet-on-helmet contact.

Conclusions: There is a high rate of IHC with a lower relative risk of IHC at most levels of play compared to the NFL. Further research is necessary to determine the impact of IHC; the high rates across all age groups suggests an im- portant role for education and prevention.

(C) 2021

1. Introduction

Research on the epidemiology of Traumatic brain injury in American Football has increased in recent years. Much of this research has focused on the prevention, diagnosis, and long-term sequelae of concussions. Recent research has demonstrated an association between subconcussive head trauma and the incidence of chronic traumatic en- cephalopathy (CTE), with the association strongest among former players of the National Football League (NFL) [1-3].

The risk of subconcussive impacts in amateur players is less well de- fined but it is thought that children and adolescents who play football are more susceptible to head injury, subsequent morbidity following head injuries, and pathological changes from participation in football [4,5]. Recent studies have drawn the link between an earlier starting age of football and worse performance on neurocognitive testing [6,7]. Behavior modifying initiatives may play a role in reducing the burden of dangerous helmet impact among youth football players, especially

* Corresponding author.

E-mail address: [email protected] (A.M. Napoli).

during the highest risk portion of the game – tackling the ball-carrier [8-12]. However, the frequency of helmet impacts across levels of play is not well defined. Our hypothesis was that Initial Helmet Impact (IHC), contact between two players at the point of tackle that is initiated with one or both players’ helmets, occurs less in all amateur leagues when compared to professional football.

1. Materials and methods

A group of eight investigators, were all trained in a pre-specified def- inition of “hits” and IHC (Appendix, SDC 1, Definition of terms and stan- dardized video reviewer hit criteria) and then randomly assigned in pairs to watch original TV broadcasts of football games in their entirety. Game reviewers knew the goal was to assess the frequency of IHC but were blinded to the specific hypothesis of the study. Game-films were obtained via open-source locations on the internet, the National Foot- ball League (NFL), Youtube Channel, ESPN, or a private subscriptions service for film review (NFHS Network, Indianapolis, IN). All games took place 2016-2018 and were “championship-level” games for each level of play. Because these games were singular championship games

https://doi.org/10.1016/j.ajem.2021.03.039

0735-6757/(C) 2021

in corresponding years it was not possible to randomize allocation of game review. NFL championship games consisted of the Super Bowl, National Football Conference (NFC) championship games, and the American Football Conference (AFC) championship games from 2016 to 2018. The collegiate level games consisted of the semifinal and final games of the National Collegiate Athletic Association (NCAA) Division I Football Bowl Series subdivision (FBS) College Football Playoff from 2017 to 2018. High School Championships were state championship games. All football championships below High School consisted of quar- terfinal, semifinal, and final games from the Pop Warner(TM) Super Bowl for five different age/weight groupings (definitions for the division in- vestigated for this study are shown in Table 1).

• 1. Study design and setting

This was a retrospective observational study. All video review was conducted on computers, using software to slow down, pause and re- wind pre-recorded video. In particular, the use of VLC Media Player Ver- sion 2.2.4 (VideoLAN, Paris, France) was incorporated for this purpose. Each reviewer (eight in total) was trained by a single master trainer who developed the hit classification paradigm used. The trainer reviewed the use of the video software and showed reviewers represen- tative video examples of each category of hit: Helmet to Helmet Direct, Offensive Helmet Only, Defensive Helmet Only, No Helmet Involvement (see “Classification of Player Hits” below). All trainees reviewed a game with the trainer as a group.

Each game in the study had two reviewers randomly assigned to watch and code hit types. For each hit, they were required to mark down the positions of the two players involved and categorize their in- terpretation of the level of helmet involvement for the players. They were instructed to record disagreements in their interpretations of any hit, which would be reviewed by a third, randomly selected viewer who gave a final classification of the hit.

The Institutional Review Board of the principal investigator’s institu-

tion approved this study with waiver of consent.

• 1. Classification of player hits

A hit was defined as contact between a ball carrier and a defensive player in the act of tackling that led to the end of a play or fumble. Plays in which a hit as previously defined did not occur were not in- cluded in the analysis. Each hit was classified based on whether the ini- tial point of impact of the offensive (the ball-carrying player at the time of the impact) and defensive players involved either player’s helmet. Hits could be categorized as “Helmet-to-Helmet direct,” “Offensive hel- met only,” “Defensive helmet only,” or “No helmet”; IHC was defined as a hit that qualified as “Helmet-to-Helmet Direct,” “Offensive Helmet Only,” and “Defensive Helmet Only.” Total hits were the total number of hits in all games that were classified at that level. The total hits per game equates to the average number of hits across all games at that level. The complete set of inclusion criteria of hits and a definition of each classification category is provided in Appendix 1.

Table 1

Demographics of Pop Warner(TM) football divisions.

Football divisions	Football divisions	Football divisions	Football divisions
Jr. Pee Weea	8-10	60-115 lbs	124 lbs
Pee Weea	9-11	75-130 lbs	139
Jr. Varsitya	10-12	90-155 lbs	164
Varsitya	12-14	105-180 lbs	189
Unlimited	11-14	105 lbs. minimum

^a These divisions allow a player 1 year older than their upper limit if the player is 20lbs under the normal upper limit for weight in that division.

• 1. Statistical analysis

Data were entered into Microsoft excel prior to analyses conducted in SAS (version 9.4, Cary N.C.). The IHC rate across the games for each age division was calculated as a proportion of the total number of hel- met hits. To address the study hypothesis examining significant differ- ences between younger (non- professional) and professional (NFL) games we conducted a series of relative risk ratios (with 95% confidence intervals [CIs]), comparing the risk of IHC for each player division to the NHL (exposed group). To examine the effect of increasing age division on the hypothesized increasing frequency of IHC hit rates we conducted a trend test using the Extended Mantel-Haenszel chi square statistic.

There was disagreement between the two initial raters in 63 (2.2%) of the 2912 rated hits. Cohen’s kappa (with 95% confidence intervals) was calculated for each pair of raters and for the overall agreement rate across raters. For the individual rater pair agreement range from kappa = 0.83 (0.73, 0.93) to 1. The overall kappa across the 2511 paired ratings = 0.95 (0.93, 0.97), indicating an excellent level of agreement among and across raters [13].

1. Results

Over the course of the video reviews 37 games were rated across the seven divisions (see Table 2), and in total 2912 total hits were reviewed. The overall incidence of IHC was 16% [95% CI 15-17] across all groups. The rate of IHC ranged from 12.6% to 19.1% with a rate of 18.9% [95% CI 15.8-22.31] among NFL players and the lowest rate at the Pee Wee level, 12.6% [95% CI 8.7-17.3] (Table 2). The number of hits per game was significantly higher at high school and above (p < 0.01). Direct hit rate, defensive hit rate, and offensive hit rate as percentages of all hits were highest at the Pee Wee (7.8%), Junior Pee Wee (8.4%), and NFL (5.2%) levels respectively (Table 3).

All but two of the non-NFL divisions had a statistically reduced risk of IHC with relative risk ratios ranging from 0.55-0.92 (Table 4), compared to the NFL. There was a non-statistically significant trend toward an increased rate of IHC at higher age division levels (?2 (1) = 2.81; p = 0.09). IHC initiated by defensive participants were twice as high as offensive participants (6.5% vs. 3.2%, RR 2.04, p < 0.01) while 6% [95% CI 5.4-7.2] of all hits were helmet-on-helmet contact. While the relative risk of IHC from defensive players vs. offensive players, or the rate of helmet-to-helmet hits did vary, the pattern was generally consis- tent across all divisions of play (Table 4, Fig. 1).

1. Discussion

In recent years there has been much research on the relationship be- tween concussive or sub-concussive impacts and CTE, particularly in tackle football. Advocacy and education have also changed the approach to the game with a focus on risk reduction and safer tackling techniques at the youth level. This study is the first to characterize and quantify the frequency of IHC, regardless of impact force or player outcome, at differ- ent age levels of play. We demonstrate an inverse association of IHC with level of play but a relatively high frequency across all age groups. Tackle football is known to be one of the highest risk sports for trau- matic brain injury, with almost every play ending with the potential for IHC [5,22]. In this study we found that nearly 1 in 6 attempted tackles, across all age levels, results in IHC. While the overall rate of IHC shows a trend toward an increase at increasing levels of play, the frequency of IHC at the lower age groups is concerning. High school, college, and professional football may have nearly a 40% increase in the number of plays per game but younger participants often play both sides of the ball, potentially as much as doubling the risk to an individual player. A notable exception in the relative rate of decrease of IHC with declining age was at the Junior Varsity level. This level of play generally corre- sponds with male pubescence and may be an indicator of a short-term

increase in at risk tackling behavior.

Table 2

IHC rate by age division.
Division (games)	Total hits	Hits per game	Total direct hits	Total defensive hits	Total offensive hits	Total IHC	Overall IHC rate (%)
NFL (n = 6)	583	97.2	39	43	28	110	18.87
College (n = 6)	628	104.7	40	45	20	105	16.72
High school (n = 6)	541	90.2	21	24	16	61	11.28
Varsity (n = 4)	237	59.3	17	13	6	36	15.2
Junior varsity (n = 5)	299	59.8	23	20	14	57	19.1
Pee Wee (n = 4)	255	63.8	20	12	0	32	12.55
Junior Pee Wee (n = 6)	369	61.5	19	31	8	58	15.72

Table 3

IHC rate as a percent of all hits.

Division	Non-helmet contact rate	Direct hit rate	Defensive hit rate	Offensive hit rate
NFL	79.6	7.2	8.0	5.2
College	83.3	6.4	7.2	3.2
High school	88.7	3.9	4.4	3.0
Varsity	84.8	7.2	5.5	2.5
Junior varsity	80.9	7.7	6.7	4.7
Pee Wee	87.5	7.8	4.7	0.0
Junior Pee Wee	84.3	5.1	8.4	2.2

Table 4 Relative risk ratios with 95% confidence intervals of dangerous play rates compared to the NFL.

Division	Relative risk ratio	95% CI	p value
College vs. NFL	0.81	0.65, 1.06	0.13
High school vs. NFL	0.55	0.41, 0.74	<0.01
Varsity vs. NFL	0.74	0.57, 1.05	0.09
Junior varsity vs. NFL	0.92	0.70, 1.24	0.63
Pee Wee vs. NFL	0.61	0.42, 0.88	0.01
Junior Pee Wee vs. NFL	0.77	0.57, 1.03	0.07

This study highlights multiple potential areas for future research and player/coach education. We suggest that the overall rate of IHC

of about 1 in every 6 hits during games is relevant to all levels of play and a fact which coaches should generally be aware. At lower ages of play, players who play both offense and defense and play po- sitions most likely to result in IHC (running back, linebacker, etc.) should be monitored closely. It is notable that these results were found approximately 5 years after highly visible campaigns to re- form tackling techniques across age groups, but particularly in youth players. This suggests there may be limited impact to date of this education campaign particularly as recent studies have shown reduced head impacts and injury rates during practices that have not demonstrably translated into games [8,11,12, 14, 15].

There are several limitations in this study that may affect the conclu- sions drawn from this study. First, the videos included in this study were chosen from a subset of championship or near championship games at their respective levels. While one might assume IHC would be more common in the most consequential games, it is also possible that the op- posite may be true; IHC may be more frequent as the quality of play de- clines and these results could therefore underestimate IHC. Selection of a more diverse and larger sample of games could potentially introduce further variability in the results as described. Second, the raters in this study only examined an incomplete subset of IHC for football players: in-game impacts between ball carrier and tackler. Although tackling is the activity that has the most significant influence on head impact expo- sure and is seemingly the best target for behavioral modification, hits between players away from the ball could certainly contribute to repet- itive sub-concussive impact. Third, IHC may be best categorized at higher play levels as video review often includes multiple camera angles

Fig. 1. Dangerous Hit Rate by Division of Play.

allowing for better detection. While it is likely that greater viewing an- gles would allow for better specificity, and therefore lower the fre- quency of detection, it is possible the opposite is the case. Nonetheless, such an outcome would artificially increase the frequency of IHC in adolescent football players as compared with professional and collegiate. Drawing direct associations between the frequency of IHC and brain injury is far more complicated; this study did not examine im- pact forces at different levels of play. Although there has been a wealth of research investigating the force and acceleration thresholds of head impacts that determine greater risk of concussion [27], there is no re- search investigating the force of impact that risks sequelae. Further re- search is also needed to elucidate the impact of repetitive head impact on different age groups.

1. Conclusions

Overall, we found that nearly 1 in 6 tackles across all levels of play results in IHC. There is a trend toward a higher relative risk and absolute number of IHCs at increasing age levels of play however that difference in risk may be lower if players are playing both offense and defense. A greater than 2:1 ratio of IHC among defensive players, as compared with offensive, emphasizes the need for education at all levels regarding proper tackling technique. Further research is needed to assess what, or if, the severity of impact combined with frequency of impact de- monstrated in this study has on the brain at different age levels of participants.

Conflicts of interest statement

The authors of this work deny any professional relationships with companies or manufacturers who will benefit from the results of the present study.

The results of the present study do not constitute endorsement by ACSM.

The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.

Disclosures

Authors have no disclosures.

Grant/funding support

None.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2021.03.039.

References

1. Tagge CA, Fisher AM, Minaeva OV, et al. Concussion, microVascular injury, and early tauopathy in Young athletes after impact head injury and an impact concussion mouse model. Brain. 2018;141(2):422-58.
2. Mez J, Daneshvar DH, Kiernan PT, et al. Clinicopathological evaluation of chronic traumatic encephalopathy in players of American football. JAMA. 2017;318(4): 360-70.
3. Montenigro PH, Alosco ML, Martin BM, et al. Cumulative head impact exposure pre- dicts later-life depression, apathy, executive dysfunction, and cognitive impairment in former high school and college football players. J Neurotrauma. 2017;34(2): 328-40.
4. Bahrami N, Sharma D, Rosenthal S, et al. Subconcussive head impact exposure and white matter tract changes over a single season of youth football. Radiology. 2016; 281(3):919-26.
5. Nonfatal Traumatic brain injuries related to sports and recreation activities among persons aged <=19 Years - United States, 2001-2009. [cited 2019 Sep 6] Available from: https://www.cdc.gov/mmwr/preview/mmwrhtml/mm6039a1.htm; 2011.
6. Stamm J, Bourlas A, Baugh C, et al. Age of first exposure to football and later-life cog- nitive impairment in former NFL players. Neurology. 2015;84(11):1114-20.
7. Alosco ML, Kasimis AB, Stamm JM, et al. Age of first exposure to American football and long-term neuropsychiatric and cognitive outcomes. Transl Psychiatry. 2017;7 (9):e1236.
8. Cobb B, Urban J, Davenport E, et al. Head impact exposure in youth football: elemen- tary school ages 9-12 years and the effect of practice structure. Ann Biomed Eng. 2013;41(12):2463-73.
9. Schussler E, Jagacinski RJ, White SE, Chaudhari AM, Buford JA, Onate JA. The effect of tackling training on head accelerations in youth American football. Int J Sports Phys Ther. 2018;13(2):229-37.
10. Kerr ZY, Dalton SL, Roos KG, Djoko A, Phelps J, Dompier TP. Comparison of Indiana high school football injury rates by inclusion of the USA football “Heads Up Football” player safety coach. Orthop J Sports Med. 2016;4(5) 2325967116648441.
11. Kerr ZY, Yeargin S, McLeod TCV, et al. Comprehensive coach education and practice contact restriction guidelines result in lower injury rates in youth American football. Orthop J Sports Med. 2015;3(7) 2325967115594578.
12. Kerr ZY, Yeargin SW, McLeod TCV, Mensch J, Hayden R, Dompier TP. Comprehensive coach education reduces head impact exposure in American youth football. Orthop J Sports Med. 2015;3(10) 2325967115610545.
13. Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas. 1960;20 (1):37-46.
14. Crisco JJ, Fiore R, Beckwith JG, et al. Frequency and location of head impact exposures in individual collegiate football players. J Athl Train. 2010;45(6):549-59.
15. Campolettano ET, Gellner RA, Rowson S. High-magnitude head impact exposure in youth football. J Neurosurg Pediatr. 2017;20(6):604-12.

[22] Harmon K, Proescholdbell S, Register-Mihalik J, Richardson D, Waller A, Marshall S. Characteristics of sports and recreation-related emergency department visits among school-age children and youth in North Carolina, 2010-2014. Inj Epidemiol. 2018;5(1):1-14.

[27] Brennan JH, Mitra B, Synnot A, et al. Accelerometers for the assessment of concus- sion in male athletes: a systematic review and meta-analysis. Sports Med. 2017;47 (3):469-78.