Concussion Incidence and Trends in 20 High School Sports

Data originated from the National High School Sports-Related Injury Surveillance Study (High School Reporting Information Online [HS RIO]) during 5 academic years (2013–2014 to 2017–2018). This study was deemed exempt by the Colorado Multiple Institutional Review Board, Aurora, Colorado. The methodology of HS RIO has been previously described^17,18  but is summarized below.

HS RIO data originated from a sample of high schools with 1 or more athletic trainers (ATs) that had valid e-mail addresses. Two data collection panels during the study period were available. The first panel collected data from a random sample of ∼100 high schools that were recruited annually to report data for 9 original sports of interest (boys’ football, wrestling, soccer, basketball, and baseball and girls’ volleyball, soccer, basketball, and softball). These high schools were categorized into 8 strata based on cross-sections of school population (enrollment ≤1000 or >1000) and US Census geographic region (Midwest, Northeast, South, and West).¹⁹  If a school dropped out of the system, a replacement from the same stratum was selected. The second panel consisted of an additional convenience sample of high schools recruited annually that reported data for additional sports of interest (eg, boys’ cross country, ice hockey, lacrosse, swimming, and track; girls’ cross country, field hockey, lacrosse, swimming, and track; and cheerleading) and any of the original 9 sports of interest. ATs at schools from the first panel could also contribute data to the second panel by reporting data for sports outside of the original 9 sports of interest. Data considered for this study originated from both panels and featured 20 sports with participation varying by sport and year. The 20 sports included in this study were as follows: boys’ football, wrestling, soccer, basketball, baseball, cross country, ice hockey, lacrosse, swimming and diving, and track and field; girls’ volleyball, soccer, basketball, softball, cross country, field hockey, lacrosse, swimming and diving, and track and field; and coed cheerleading.

ATs from participating high schools reported injury incidence and athlete exposure (AE) information weekly throughout the academic year using a secure Web site. For each injury, the AT completed a detailed injury report (body part, diagnosis, mechanism, activity, position, recurrence, etc). A reportable AE was defined as 1 athlete participating in 1 school-sanctioned practice or competition. Our injury of focus, concussion, was defined as (1) occurring as a result of participation in an organized practice or competition, (2) requiring medical attention by an AT or physician, and (3) being diagnosed as a concussion. Although other injuries collected via HS RIO were only included if they resulted in participation restriction time ≥24 hours, all concussions regardless of participation restriction time were included. We did not provide a definition for concussion because we relied on the expertise of the ATs and any additional team medical staff. Nonetheless, concussion consensus statements were available to medical staff.^15,20  Internal validity checks conducted by HS RIO staff have consistently found sensitivity, specificity, positive predictive value, and negative predictive value >90%.²¹ 

Data were analyzed by using SAS (version 9.4; SAS Institute, Inc, Cary, NC). Concussion frequencies, distributions, and rates per 10 000 AEs with 95% confidence intervals (CIs) were calculated. Injury rate ratios (IRRs)²²  compared concussion rates by event type (ie, competition or practice) and by sex in sex-comparable sports (ie, soccer, basketball, baseball or softball, cross country, track, and swimming). Lacrosse was not considered a sex-comparable sport because rules regarding contact vary between the boys’ and girls’ sports. An example of an IRR comparing competition and practice rates is as follows:

Injury proportion ratios (IPRs)²²  compared distributions by injury mechanism and recurrence. An example of an IPR comparing the proportion of concussions that were due to contact with another person between boys and girls is as follows:

IRRs and IPRs with 95% CIs excluding 1.00 were considered statistically significant. Linear regression was used to analyze trends in concussion rates across the study period and estimate average annual changes (ie, mean differences). Mean differences with 95% CIs excluding 0.0 were considered statistically significant.

During the 2013–2014 to 2017–2018 academic years, 9542 concussions were reported from the study sample (Table 1). Of these, 63.7% occurred during competitions, and 36.3% occurred during practices. Most concussions were reported in the regular season (81.6%), followed by the preseason (14.2%) and the postseason (3.3%); 0.9% of concussions did not have time-in-season data.

The 9542 reported concussions occurred during 22 870 364 AEs for an overall rate of 4.17 per 10 000 AEs (95% CI: 4.09 to 4.26; Table 1). Boys’ football had the highest overall concussion rate (10.40 per 10 000 AEs), followed by girls’ soccer (8.19 per 10 000 AEs) and boys’ ice hockey (7.69 per 10 000 AEs). Findings were similar when examining concussion rates in competition. When examining concussion incidence in practice, the highest rates were observed in boys’ football (5.01 per 10 000 AEs), followed by cheerleading (3.60 per 10 000 AEs) and boys’ wrestling (3.12 per 10 000 AEs).

Overall, the concussion rate was higher in competition than practice (10.37 vs 2.04 per 10 000 AEs; IRR = 5.09; 95% CI: 4.89 to 5.31; Table 1). Sport-specific findings were generally consistent with overall findings in this regard. However, in cheerleading, the concussion rate in competition was lower than that in practice (2.22 vs 3.60 per 10 000 AEs; IRR = 0.62; 95% CI: 0.48 to 0.79).

When examining sex-comparable sports (ie, soccer, basketball, baseball or softball, cross country, swimming, and track and field), the overall concussion rate was higher in girls than in boys (3.35 vs 1.51 per 10 000 AEs; IRR = 2.22; 95% CI: 2.07 to 2.39; Table 2). This finding was retained when examining specific sex-comparable sport pairs and when restricted to competitions or practices only; however, some IRRs were no longer statistically significant.

Sufficient evidence of temporal trends across the study period was found only in boys’ football (Fig 1). Whereas the competition rate had increased over time in boys’ football (annual average change of 1.56 per 10 000 AEs; 95% CI: 1.13 to 2.00), the practice rate had decreased (annual average change of −0.28 per 10 000 AEs; 95% CI: −0.37 to −0.18).

Of the 3463 practice-related concussions reported, most occurred after the first hour of practice (66.9%; Table 3). Although competition-related concussion data were stratified by sport, larger proportions of concussions were typically reported in the latter half of competitions. Among concussions reported in sports with quarters or halves (ie, boys’ football, lacrosse, soccer, and basketball and girls’ lacrosse, field hockey, soccer, and basketball), 54.5% were reported to have occurred in the second half, or third or fourth quarters compared with 29.4% in the first half, or first or second quarters. Precompetition- and/or warm-up–related concussions accounted for relatively small proportions of all reported concussions in most sports; the 2 exceptions were girls’ volleyball (27.5%) and girls’ softball (26.3%).

Most concussions were due to contact with another person (62.3%), followed by contact with the surface (17.5%) and contact with equipment or an apparatus (15.8%; Table 4). In sex-comparable sports, the proportion of concussions due to contact with another person was higher in boys than in girls (54.5% vs 42.1%; IPR = 1.29; 95% CI: 1.20 to 1.39). In contrast, the proportion of concussions due to contact with equipment or an apparatus was higher in girls than in boys (33.6% vs 23.5%; IPR = 1.43; 95% CI: 1.27 to 1.62). Table 5 summarizes the most sport-specific common injury mechanism, injury activity clusters, and positions with concussions.

Overall, 8.3% of reported concussions were recurrent (Fig 2). The sports with the largest proportions of concussions that were recurrent were boys’ ice hockey (14.4%), boys’ lacrosse (12.1%), and girls’ field hockey (12.1%). Among sex-comparable sports, the proportion of concussions that were recurrent was higher in girls than in boys (9.3% vs 6.4%; IPR = 1.44; 95% CI: 1.11 to 1.88). However, the proportion of concussions that were recurrent was higher in baseball than in softball (11.0% vs 4.4%; IPR = 2.48; 95% CI: 1.15 to 5.37).

Linear trend tests found evidence of decreases in the recurrent concussion rate across the study period (Fig 1). This annual average change was the greatest in boys’ football (annual average change of −0.13 per 10 000 AEs; 95% CI: −0.19 to −0.07). Decreases were also observed in all other sports (annual average change of −0.03 per 10 000 AEs; 95% CI: −0.05 to −0.02).

Understanding concussion incidence and prevention is important given the growing population of high school athletes and rapidly changing field of concussion management.¹⁰  Our study updates previous literature on high school sports concussion incidence^2,13,14  and provides additional information about emerging trends in incidence. Our study is unable to directly examine the causes of increased concussion risk or the effects of past prevention efforts. Still, because the van Mechelen et al¹²  sequence of prevention framework emphasizes ongoing monitoring of injury incidence, we believe our findings will benefit clinicians by helping guide future prevention and research directions.

Concussion rates in the current study were generally higher than previous estimates from similar surveillance-based studies using data from other time periods.^2,13,14  Likewise, we observed decreasing rates of recurrent concussions. Data from previous studies were generally collected during a period in which mandatory concussion legislation governing high school sports did not exist (such legislation currently exists in all 50 states and the District of Columbia).¹⁶  Such legislation has been found to be associated with increased concussion rates (likely due to increased reporting) and decreased recurrent concussion rates (hypothesized to be due to better concussion management that legislation advocated, including mandatory removal from play and requirements associated with return to play).^23–25  Future research may inform concussion prevention in 2 manners. First, concussion legislation varies by state¹⁶ ; future research should identify specific legislative components most strongly associated with changes in concussion incidence to help improve and clarify existing requirements.²⁶  Second, because the reported concussion incidence may be affected by the potential increase in reporting,^14,27  continued longitudinal research is needed.

As previously seen in high school sports^2,12,13  and other settings,^28,29  football maintained the highest observed concussion rate. Although this was likely due to the inherent element of collision in the sport, it is important to consider why football consistently has a concussion rate higher than other sports in which intentional contact is allowed, such as wrestling, boys’ ice hockey, and boys’ lacrosse. Although we found evidence of decreasing practice-related concussion rates in football (possibly due to efforts to limit contact exposure in practices),^30,31  there was also evidence of increasing rates in competitions. Thus, concussion prevention efforts in football should target competitions alongside practices. The American Academy of Pediatrics advocates teaching proper tackling and enforcing rules that discourage improper technique.³²  Simultaneously, kickoff rule changes at the collegiate level have been found to be associated with reduced concussion rates.³³  It is also, unfortunately, possible that decreasing contact exposure may impact skill development and place players at increased injury risk during games. More in-depth examinations for such changes in concussion rates are warranted. Furthermore, policy and rule changes must be evaluated to ensure that they provide their intended outcomes, particularly when introduced from higher levels of play (eg, professional and collegiate) to lower levels.

Finally, it is important to consider that cheerleading was the only sport in which the concussion rate was higher in practice than in competition. Not all states recognize cheerleading as a sport, and cheer squads may practice in less-than-ideal places (eg, hallways and on asphalt)³⁴  and may have less access to medical oversight and coaching support compared with other high school sports. Pediatricians should ensure that youth athletes and their families are aware of the concussion risk associated with their sports of interest. Furthermore, pediatricians working with youth sports organizations should advocate the use of safety measures to help prevent concussions.

Concussion rates were higher among girls than boys in sex-comparable sports. This result was found within multiple sports, although statistical significance was likely not achieved in sports such as cross country and track because of low counts of concussions during the 5-year study period. Findings related to sex differences are prevalent throughout concussion literature,^2,13,28,29  and potential associated factors have included differences in disclosure, neck musculature, cerebral blood flow and/or cerebrovascular organization, hormonal regulation, etc.^35–40  However, few studies have attempted to directly assess these factors in relation to incidence. Also, our study found that the proportions of concussions among sex-comparable sports that were recurrent and due to equipment contact were higher in girls than in boys, which may also contribute to the differences in concussion incidence between the sexes. Research that isolates the specific etiological factors associated with sex differences is needed to assist clinicians in properly highlighting potential preventive factors to reduce concussion incidence.

Approximately 1 in 6 concussions overall were due to contact with equipment or an apparatus. In a recent report from The Aspen Institute,¹¹  large proportions of youth sport coaches were found to not be trained in 6 core competencies, including sport-specific skills and tactics. Although high school athletes may have had experience playing their sports at earlier ages and lower levels of play, they may still lack aspects of skill development and physical literacy. All high school coaches need to be properly educated in their roles to ensure age-appropriate training and gameplay.

In both competition and practice, more concussions were generally reported in the latter segments of each event type, which concurs with previous research.^41–43  In sports in which contact occurs intentionally or unintentionally, athletes who are more fatigued may be less able to defend themselves and/or be less aware of oncoming collisions. Also, athletes on the opposing team who are fatigued may put other players at risk due to playing with less technical craft.⁴⁴  Game-officiating and referee-education programs should be adapted to be more sensitive to instances in which fatigue is resulting in reckless play. Primary prevention strategies including monitoring and removing fatigued athletes from play when feasible are recommended.

Despite this finding, we also found that more than one-quarter of competition-related concussions in girls’ volleyball and girls’ softball occurred in precompetition or warm-ups. We were unable to pinpoint the factors associated with these findings. In girls’ volleyball, both teams may warm up on their respective court sides at the same time; thus, players may not pay attention to the actions of the opposing team, which could lead to unintended ball contact when balls were spiked over the net by the opposing team. In response, the National Federation of High School Associations released a rule change in 2018–2019 not allowing balls to be hit over the net during warm-ups.⁴⁵  In girls’ softball, warm-ups may include numerous balls in motion in various areas of the field at once, which may place athletes at risk for being hit by a ball. Additional exploration of such potential causes and appropriate prevention strategies to reduce risk are needed.

Only data from high schools with ATs were included. Therefore, the findings of this study may not be generalizable to the overall population of high schools, particularly those without ATs.⁴⁶  Underreporting may have occurred if players chose not to disclose concussion symptoms, thus compromising our ability to fully understand athlete injury histories. Our study also did not provide a working definition of concussion, nor could it account for underreporting and/or misdiagnosis. However, all injuries were assessed and documented by medical professionals who were trained to accurately detect and diagnose injuries. Furthermore, previous research has found ATs to have high agreement with physicians in diagnosing injuries⁴⁷  and to provide higher-quality data collection compared with nonmedically trained individuals, such as coaches or other players.⁴⁸  This study was unable to account for factors potentially associated with concussion incidence, such as implemented injury prevention programs and athlete-specific characteristics (eg, maturation status, previous injury history, functional capabilities, etc). Lastly, our study calculated injury rates using AE, which may not be as precise an at-risk exposure measure as minutes or hours, or the total number of game plays across a season.²⁷  HS RIO also does not acquire AE by position or starter status, thereby precluding our ability to calculate more specific injury rates. However, collection of such exposure data is more laborious and may be too burdensome for ATs contributing to HS RIO.

Using surveillance data from the 2013–2014 to 2017–2018 academic years, we found that football remained the high school sport with the highest concussion rate, with decreasing practice concussion rates observed; however, increasing competitions rates highlight the continued need for prevention strategies. Decreasing recurrent concussion rates may be associated with changes in concussion diagnosis and management in high school sport settings, such as those elicited by concussion-based legislation. Future research should also target risk and preventive factors in other sports that allow contact, the role of skill development to reduce equipment-related contact, and strategies to mitigate increased concussion risk in the latter halves of events. Pediatricians should provide such updated information to youth athletes to help them and their families understand the risk of concussion in their sports of interest.

We thank the many ATs who have volunteered their time and efforts to submit data to HS RIO. Their efforts are greatly appreciated and have had a tremendously positive effect on the safety of high school student athletes.

AE

athlete exposure

AT

athletic trainer

CI

confidence interval

HS RIO

High School Reporting Information Online

IPR

injury proportion ratio

IRR

injury rate ratio