Objectives To compare new versus recurrent concussions with respect to constellation of symptoms, symptom severity, symptom resolution; evaluate potential subset differences with respect to gender and sport; and to compare mechanisms and activities associated with new versus recurrent concussions.
Methods Sports-related injury and exposure data were collected for nine sports from 2005 to 2010 from 100 nationally representative US high schools.
Results Nationally, an estimated 732,805 concussions occurred. Of these reported concussions, 13.2% were recurrent. The rate of new concussions was 22.2 per 100,000 athletic exposures while the rate of recurrent concussions was 3.1 per 100,000 athletic exposures (RR 7.23, 95% CI 6.39 to 8.17, p<0.001). While 0.6% of new concussion symptoms took >1 month to resolve, 6.5% of recurrent concussion symptoms took >1 month to resolve (IPR 10.35; 95% CI 4.62 to 23.16; p<0.001). Loss of consciousness was reported more often with recurrent (7.7%) than new concussions (4.4%) (IPR 1.76; 95% CI 1.02 to 3.03; p=0.043). A greater proportion of athletes sustaining recurrent concussions returned to play in >3 weeks (7.5%) or were medically disqualified (16.2%) than athletes sustaining new concussions (3.8%; IPR 1.95; 95% CI 1.01 to 3.77; p=0.047 and 2.9%; IPR 5.58; 95% CI 3.50 to 8.88; p<0.001, respectively). The majority of new and recurrent concussions resulted from contact with another person (73.4% and 77.9%, respectively).
Conclusions Athletes sustaining recurrent concussions had longer symptom resolution times, were kept out of play longer and reported loss of consciousness more frequently than athletes sustaining new concussions. With the possibility of long-term impairment and other negative sequelae, proper management and prevention of concussions at the high school level is imperative.
Statistics from Altmetric.com
In the USA, over 7.6 million students participated in organised high school sports in 2009–2010.1 An estimated 300,000 sports-related concussions occur annually within USA.2 Recurrent concussions are of particular concern given the potential long-term or life-threatening effects.3,–,5 As sports' participation among high school athletes has increased, the concern about concussions has also increased. In order to develop more effective preventive measures and then most appropriate recommendation regarding management, we must understand the epidemiology of new and recurrent concussions.
Previous investigations have explored risk factors for concussion including gender differences,6,–,8 sport6 9,–,12 and age.13,–,15 Several examined concussions in high school and college athletes6 8 13 15 16 while fewer focused on high school athletes alone.11 17 18 Since high school athlete's brains differ from college athletes due to a still-developing central nervous system,19 20 management and treatment of concussions for high school and adult athletes may differ. Few studies have compared new versus recurrent concussions. Research focused on recurrent concussions has discussed presentation of symptoms,18 the rate of recovery,21 the effect on the brain4 22 or prolonged cognitive effects.23,–,27 Little is known about risk factors for recurrent concussion among young athletes, like age, sport and gender. One study that did focus on high school athletes found athletes who had previously sustained two or more concussions exhibited more concussion symptoms including cognitive, physical and sleep difficulties than those athletes who had sustained one or no prior concussions.28
This study's objective was to compare the epidemiology of new versus recurrent concussions sustained by a large national sample of high school athletes. The specific aims were to compare the epidemiology of new versus recurrent concussions with respect to constellation of symptoms, symptom severity, symptom resolution, etc; to evaluate potential subset differences (gender and sport); and to compare mechanisms and activities associated with new versus recurrent concussion.
Data from the National High School Sports-Related Injury Surveillance Study, which gathers data via the internet-based surveillance system, reporting information online (RIO), were used. During the 2005–2010 academic years, schools with a National Athletic Trainers Association (NATA) affiliated certified athletic trainer (AT) with a valid email address were invited to participate. Willing participants were assigned a sampling strata based on school size (<1000 and ≥1000) and geographical region (South, Midwest, West, Northeast)29 with 100 schools randomly selected to obtain a nationally representative sample. If a school withdrew from the study, a replacement school was chosen from the same strata.
Previous studies have described the methodology of this high school sports surveillance system.6 30 31 In brief, ATs logged onto RIO weekly to submit injury reports and athlete exposure (AE) for five boys' sports (football, soccer, basketball, baseball and wrestling) and four girls' sports (volleyball, soccer, basketball and softball). Injury reports collected data on the athlete (age, height, weight, etc), injury (site, diagnosis, etc) and event (activity, mechanism, etc). ATs could view, edit and update injury reports throughout the study. An AE was defined as one athlete participating in one practice or competition. During the 2005–2007 academic years, a reportable injury was defined as one which (1) occurred as a result of participation in an organised practice or competition, (2) required medical attention by an AT or a physician and (3) resulted in restriction of the athlete's participation for 1 or more days. For the 2007–2010 academic years, the definition of injury was expanded to capture all concussions, even if there was no time lost from sports participation.
Data were analysed using SPSS Statistics v.17.0 (SPSS, Chicago, IL) and Open Epi v.2.3 (CDC). Unweighted case counts were used to calculate rates and rate comparisons. All other data presented used national estimates calculated by assigning sample weights (based on the inverse of the probability the school was selected for participation) to each injury. Standard errors for comparison were adjusted for the sampling scheme using the SPSS Complex Samples module. Rate ratios (RR) and injury proportion ratios (IPR) with 95% CIs and p values were used for data analysis. CIs not containing 1.00 and p values <0.05 were considered statistically significant. RR or IPR >1.00 were considered a risk association, and RR or IPR <1.00 were considered a protective association. Sample calculations follow:
The Institutional Review Board at Nationwide Children's Hospital approved this study.
Incidence and rates
During 2005–2010, 2417 concussions were reported of which 2110 were new (87.3%) and 292 were recurrent (12.1%) (new or recurrent was unknown for 15 concussions (0.6%)). The 2402 concussions designated as new or recurrent occurred during 9,503,641AEs for an injury rate of 25.3 concussions per 100,000 AEs (table 1). The injury rate of new concussions was 22.2 per 100,000 AEs, while the rate of recurrent concussions was 3.1 per 100,000 AEs (RR: 7.23, 95% CI 6.39 to 8.17, p<0.001). The highest rates of new concussions occurred in football (47.6), girls' soccer (25.8), boys' soccer (17.4) and girls' basketball (16.7) (table 1). The highest rates of recurrent concussions occurred in football (6.3), girls' soccer (4.1), girls' basketball (2.9) and wrestling (2.7). Among gender-comparable sports (ie, soccer, basketball and baseball/softball), girls had higher rates of concussions overall (20.4) (RR: 1.83, 95% CI 1.59 to 2.11, p<0.001), new concussions (17.6) (RR: 1.75, 95% CI 1.51 to 2.03, p<0.001) and recurrent concussions (2.8) (RR: 2.54, 95% CI 1.67 to 3.88, p<0.001) than boys (11.2, 10.1 and 1.1, respectively) (table 1).
The rates of recurrent concussion were significantly higher in competition compared with practice in all sports except volleyball and softball (there were no recurrent concussions sustained in baseball during practice) (table 1). Boys' soccer had the highest risk of recurrent concussions in competition compared with practice (RR: 22.21; 95% CI 5.17 to 95.36; p<0.001) followed by girls' soccer (RR: 17.02; 95% CI 5.98 to 48.42; p<0.001).
Nationally, an estimated 732,805 concussions (636,053 new, 96,752 recurrent) occurred among US high school athletes participating in the nine sports studied. Concussions represented 10.5% of all injuries. Football accounted for nearly half of all reported concussions (48.5%) followed by girls' soccer (17.7%), boys' soccer (12.2%) and girls' basketball (6.3%).
Of all concussions reported, 13.2% were recurrent (figure 1). The proportion of recurrent concussion was >10.0% in all sports except boys' basketball (8.3%). Among gender-comparable sports, boys (13.5%) and girls (15.0%) had a similar proportion of recurrent concussions.
The most commonly reported symptom for all concussions was headache (85.8%) followed by dizziness (64.6%), concentration difficulty (47.8%) and confusion/disorientation (39.5%); loss of consciousness was rarely reported (4.8%) (table 2). Overall, a mean of 3.61 (SD 2.17) symptoms were reported per concussion. The mean number of symptoms for new (3.59; SD 2.11) and recurrent concussions was similar (3.72; SD 2.50). Light/noise sensitivity was more frequent with recurrent (37.0%) than with new (29.6%) concussions (IPR: 1.25; 95% CI 1.01 to 1.55; p=0.047) (table 2). Loss of consciousness was also reported more often with recurrent (7.7%) than with new (4.4%) concussions (IPR: 1.76; 95% CI 1.02 to 3.03; p=0.043). Boys reported amnesia more often than girls (31.1% and 13.1%, respectively) for new concussions (IPR: 2.37; 95% CI 1.62 to 3.48, p<0.001); however, the proportions were similar in recurrent concussions (16.3% and 13.6%, respectively). Girls sustained a greater proportion of recurrent concussions that resulted in concentration difficulty (54.5%) than boys (22.9%) (IPR: 2.37; 95% CI 1.10 to 5.13; p=0.013).
In all sports, the highest proportion of new concussion symptoms resolved in 1 day to 1 week (figure 2). For recurrent concussions, in seven of the nine sports, the highest proportion of symptoms resolved in 1 day to 1 week (81.7% of symptoms within baseball resolved in 1 week to 1 month, and 41.2% of symptoms within softball resolved in >1 month). A greater proportion of recurrent concussion symptoms took 1 week to 1 month to resolve (20.9%) compared with new concussion symptoms (13.8%) (IPR: 1.52; 95% CI 1.10 to 2.10; p=0.012). Similarly, while 0.6% of new concussion symptoms took >1 month to resolve, 6.5% of recurrent concussion symptoms took >1 month to resolve (IPR: 10.35; 95% CI 4.62 to 23.16; p<0.001).
Number of symptoms was associated with symptom resolution time. For example, among new concussions with ≥4 symptoms, 20.6% resolved in ≥1 week compared with 9.0% of new concussions with <4 symptoms (IPR: 2.29; 95% CI 1.68 to 3.12; p<0.001). Among recurrent concussions with ≥4 symptoms, 36.6% resolved in ≥1 week compared with 19.1% of recurrent concussions with <4 symptoms (IPR: 1.92; 95% CI 1.11 to 3.30; p=0.014).
Return to play
A greater proportion of athletes sustaining a recurrent concussion returned to play in >3 weeks (7.5%) or were medically disqualified (16.2%) than athletes sustaining a new concussion (3.8%; IPR: 1.95; 95% CI 1.01 to 3.77; p=0.047 and 2.9%; IPR: 5.58; 95% CI 3.50 to 8.88; p<0.001, respectively) (figure 3). Among concussions whose symptoms took >1 month to resolve, 15.2% of athletes sustaining a new concussion and 7.9% of the athletes sustaining a recurrent concussion returned to play in <3 weeks, indicating that athletes returned to play while still symptomatic (there was no significant difference between new and recurrent concussions; p=0.464). In the three academic years when non-time loss concussions were captured, 3.0% of all concussions were non-time loss, and the proportion of new and recurrent non-time concussions was similar (3.3% and 1.3%, respectively).
Number of symptoms was associated with greater time loss. For example, among new concussions with ≥4 symptoms, 5.6% resulted in time loss of >3 weeks compared with 2.3% of new concussions with <4 symptoms (IPR: 2.39; 95% CI 1.25 to 4.56; p=0.006). Among recurrent concussions with ≥4 symptoms, 14.9% resulted in a time loss of >3 weeks compared with 1.0% of recurrent concussions with <4 symptoms (IPR: 14.81; 95% CI 2.37 to 92.48; p<0.001).
The majority of new and recurrent concussions resulted from contact with another person (73.4% and 77.9%, respectively). In gender-comparable sports, overall, boys had a higher proportion of sustaining concussions from player-to-player contact (73.4%) than girls (55.6%) (IPR: 1.32; 95% CI 1.15 to 1.51; p<0.001). This gender difference held true for new concussions (p<0.001) but not for recurrent concussions (p=0.89). Conversely, girls had a higher proportion of concussions from contact with the playing surface (25.3%) than boys (16.4%) (IPR: 1.54; 95% CI 1.06 to 2.25; p=0.021). Again, this difference held true for new concussions (p=0.027) but not for recurrent concussions (p=0.438).
This research, the first to examine the epidemiology of new versus recurrent concussions among a large national sample of the US high school athletes, found that 13.2% of all concussions were recurrent. As participation in high school sports continues to increase, the potential for recurrent injuries, including recurrent concussions, may also increase. With the possibility of long-term impairment and other negative sequelae associated with recurrent concussions,25,–,27 proper management and prevention of concussions at the high school level is imperative.
This study found higher rates of concussion among girls compared with boys overall and for new and recurrent concussions among gender-comparable sports (ie, soccer, basketball and baseball/softball). Several theories have been offered to explain these gender disparities including biomechanical differences in head and neck acceleration,32 hormonal differences33 and sociocultural differences such as the perception that boys must be tough and play through injuries. Additionally, coaches and/or medical professionals may be more likely to diagnose concussions among women, either because women are more likely to report concussion symptoms or because coaches and medical professionals are more concerned when women report symptoms. Interestingly although, we found in gender-comparable sports, boys (13.5%) and girls (15.0%) had a similar proportion of recurrent concussions. This suggests that once an athlete has been diagnosed with an initial concussion, gender difference is minimised. Boys and girls may be equally likely to report symptoms of a recurrent concussion, medical personnel and/or coaches may be more likely to identify a concussion in an athlete who previously sustained a concussion regardless of gender, or athletes of both genders who have been previously diagnosed with a concussion may simply be more familiar with symptoms. Such gender patterns in initial and recurrent concussions require additional research.
Similar to previous research,13 this study found that overall, the most commonly reported symptoms for concussions were headache (85.8%), dizziness (64.6%) and concentration difficulty (47.8%). The mean number of symptoms was similar for new (3.59; SD 2.11) and recurrent (3.72; SD 2.50) concussions; however, there were differences in the type of symptoms reported for new versus recurrent concussions. Loss of consciousness and light/noise sensitivity were reported more often for recurrent concussions compared with new concussions. In addition, there were also differences in the type of symptoms reported by gender. Boys reported amnesia more often than girls (31.1% and 13.1%, respectively) for new concussions while girls sustained a greater proportion of recurrent concussions that resulted in concentration difficulty (54.5%) than boys (22.9%). Future research is needed to further examine such differences and to determine potential clinical implications of patterns of symptoms experienced after new versus recurrent concussion.
This study also found that compared with new concussions, symptoms of recurrent concussions took longer to resolve, which is consistent with the research conducted in a collegiate population.34 Subsequently, athletes were kept out of play longer for recurrent concussions than for new concussions. However, the proportion of athletes sustaining new and recurrent concussions whose symptoms took >1 month to resolve who returned to play in <3 weeks (15.2% and 7.9%, respectively) was concerning. This finding indicates that a substantial number of athletes with lingering concussions returned to play while still symptomatic. Previous research which also used data from the National High School Sports-Related Injury Surveillance Study found for all concussions, one in six high school athletes returned to play too soon under the Prague guidelines.31 Given the potentially catastrophic consequences of returning to play before being fully recovered following concussion, particularly recurrent concussions, continued education on appropriate return-to-play guidelines is needed and these guidelines must be consistently applied. Sports medicine professionals, parents, coaches and sports administrators must work together to ensure that return to play guidelines are followed.
Like all studies, ours have limitations. Participating schools were limited to those with a NATA-affiliated AT. The findings may not be generalisable to schools lacking an AT. However, the quality and consistency of data provided by ATs justify this limitation. We also measured our AEs based on units of participation rather than time. While time-based AEs may be more accurate, due to our large study population, it was not feasible for ATs to submit time-based reports for every athlete under their care. Participating ATs were only able to report concussions they were aware of; therefore, this study likely underestimates the actual number of concussions sustained by high school athletes. However, to ensure that high-quality data were reported by medically trained professionals, reporting was restricted to ATs. Unfortunately, the National High School Sports-Related Injury Surveillance Study, which captured the data used for this study, did not capture the number of prior concussions sustained by the athlete, only whether or not the concussion was new or recurrent. Finally, this study focuses on nine high school sports, excluding sports with relatively high concussion rates including lacrosse and ice hockey. However, we accept this limitation because the nine sports studied have high participation rates in the USA. Despite these limitations, this was the largest epidemiological analysis of new versus recurrent concussions among a large nationally representative sample of US high school athletes.
Athletes sustaining recurrent concussions had longer symptom resolution times, were kept out of play longer and reported loss of consciousness more frequently than athletes sustaining new concussions. Further research is needed to determine the risk factors for recurrent concussions and what preventive measures may effectively decrease the number of recurrent concussions sustained by high school athletes. ATs, physicians, coaches, etc should carefully evaluate their return to play guidelines to ensure that athletes never return to play before all symptoms have resolved. ATs, physicians, coaches, policy makers and parents must work together to ensure that high school athletes are able to compete competitively and safely throughout their academic careers.
The content of this report was funded in part by the Centers for Disease Control and Prevention (CDC) grants #R49/CE000674-01 and #R49/CE001172-01. The content of this report is solely the responsibility of the authors and does not necessarily represent the official views of the CDC. The authors acknowledge the generous research funding contributions of the National Federation of State High School Associations (NFHS), EyeBlack and DonJoy Orthotics.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.