Abstract
Background:
Increasing attention has been paid to concussions and especially sports-related concussions in youth. To prevent an inappropriate return to play while symptomatic, nearly all states have now passed legislation on youth sports-related concussions.
Purpose:
To determine (1) the incidence of sports-related concussions in high school athletes using a unique system to collect reports on concussions, (2) the proportion of athletes with concussions who play with concussive symptoms, and (3) the effect of the type and modality of coach education on the likelihood of athletes reporting symptoms to the coach or playing with concussive symptoms.
Study Design:
Cohort study; Level of evidence, 2.
Methods:
This study was conducted with high school football and girls’ soccer athletes playing in fall 2012 and their coaches and parents in 20 urban or rural high schools in Washington State. The main outcome was the incidence of concussions per 1000 athlete-exposures (AEs), the proportion of concussed athletes who played with concussive symptoms, and the association of coach concussion education with coach awareness of athletes with concussive symptoms.
Results:
Among the 778 athletes, the rate of concussions was 3.6 per 1000 AEs and was identical for the 2 sports studied. The cumulative concussion incidence over the course of the season was similar in girls’ soccer (11.1%) and football (10.4%). Sixty-nine percent of concussed athletes reported playing with symptoms, and 40% reported that their coach was not aware of their concussion. Most measures of coach concussion education were not associated with coach awareness of concussions in their athletes, although the modalities of a video and quiz were associated with a lower likelihood of coach awareness.
Conclusion:
More objective and accurate methods are needed to identify concussions. Changes in athlete attitudes on reporting concussive symptoms will likely not be accomplished through legislation alone.
In the past few years, there has been a burgeoning interest in concussions, including their incidence, pathophysiology, diagnosis, management, prognosis, and prevention. This interest has led to the realization that questions remain in each of these areas.9,21,22 Reports on the incidence of adolescent sports-related concussions have based their estimates on concussions identified by athletic trainers, coaches, or other adults.4,10,16 However, studies indicate that there are many barriers for high school athletes reporting potentially concussive symptoms to parents, athletic trainers, or coaches, including the difficulty recognizing that they are injured, the fear of being taken out of play, and the culture of the sport.3,20 As a result, many potential concussions go unreported, and some athletes continue to play while symptomatic. 4 Recent guidelines on the management of concussions recommend that concussed athletes be removed from play and have a gradual return to play once asymptomatic.9,21,22
To prevent an inappropriate return to play while symptomatic, nearly all US states have now passed legislation on youth sports-related concussions. 27 Washington’s Zackery Lystedt Law, 29 enacted in 2009, was the first such comprehensive effort in the country, mandating the education of coaches about concussions, signing of an information sheet on concussions by parents and athletes, removal of the athlete from practice or play at the time of a suspected concussion, and written clearance by a licensed health care provider trained in the evaluation and management of concussions before the athlete can return to practice or play.
This study was conducted to determine the incidence of sports-related concussions in high school athletes in Washington State using a unique system to collect reports on concussions, to examine the proportion of athletes with concussive symptoms who did not report their injury and continued to play while symptomatic, and to examine the effect of coach education on the likelihood of coach awareness of athletes’ concussive symptoms.
Materials and Methods
All study procedures were approved by our institutional review board.
Recruitment of Participants
A listing of all public school districts (N = 299) in Washington State was obtained and divided into urban (n = 87) and rural districts (n = 212), as classified by the Washington State Department of Health based on population size and proportion of the population commuting for work. 11 We then randomly selected districts in each of these 2 groups. Within districts, we randomly selected 1 high school that had a boys’ football team, a girls’ soccer team, and a boys’ soccer team (we piloted procedures on boys’ soccer in the prior spring before collecting the data on football and girls’ soccer reported herein). We contacted the athletic directors for the selected schools, explained the study, and asked them and the school coaches to participate. If a school’s athletic director did not agree to participate, we selected the next district. Schools were offered a US$100 incentive for participation. Our goal was to enroll 20 schools in the study.
Study staff attended each football and girls’ soccer team’s meeting with coaches, athletes, and parents before the start of the 2012 fall sports season to explain the study and recruit athletes and parents. We obtained consent from parents and athletes ≥18 years of age and assent from youth athletes.
Assessments
The coaches were asked to complete a questionnaire at the beginning of the athletic season, detailing their age, sex, level of education, years coaching, national certification, attendance of athletic trainers during games, and requirement for concussion education as well as the number and modalities of concussion education (written, video, PowerPoint, a 6-item quiz on concussion knowledge distributed by the Washington Interscholastic Activities Association, and in-person training). We also queried coaches on their awareness of the recommendation for graduated return to play after a concussion and of the Lystedt Law. Coaches were offered a US$20 incentive for completion of the questionnaire.
Athletes completed a baseline survey about prior athletic experiences and history of concussions. The preferred contact method (e-mail, text, or automated telephone call) was obtained from both athletes and parents. Athletes and parents were then contacted weekly during the 2012 fall sports season, reminding them each to call a toll-free number that activated an interactive voice response (IVR) system. Using the key pad of a telephone, athletes reported the number of practices and games in which they participated that week as well as whether they had suffered any hits to the head or body that resulted in any of the symptoms (see the Appendix, available in the online version of this article at http://ajsm.sagepub.com/supplemental) consistent with a concussion outlined in the Sport Concussion Assessment Tool version 2 (SCAT-2). 17 Parents answered the same questions, reporting on the number of practices and games that their athlete participated in during the week as well as any hits that resulted in concussive symptoms experienced by their athlete. Athletes were given a US$2 Amazon gift certificate for each weekly call, and parents were offered a US$10 incentive for completion of all calls.
If athletes or parents reported any concussive symptoms, the IVR system automatically notified the study team, and research assistants telephoned the athlete and parent separately to determine the date of the injury, occurrence during a practice or a game, severity rating of symptoms using a scale from 0 to 6 (higher scores indicating more severe symptoms) on the day of injury and in the 24 hours before the interview, whether they continued to play with symptoms, and if they reported the symptoms to their coach or if their coach was already aware of the concussion. For athletes who did not report a resolution of their concussive symptoms at the time of the initial call, both athletes and parents were called weekly until symptoms resolved. All interviews and follow-up assessments were reviewed by 1 of the authors (F.P.R.) to confirm the diagnosis of a concussion; however, the athletes were not examined in person as part of this study. Injuries were only labeled as concussions if athletes had ≥2 symptoms more severe than a score of 1 on the 6-point scale of the SCAT-2 or 1 symptom that was rated 5 to 6 in severity and that lasted more than a few hours. As has been done in prior studies of sports concussions5,8,19,20,25 and in the consensus statements on concussions in sports,21,22 a concussion was defined as an injury from a blow to the head or body that resulted in a temporary impairment of neurological function and resulted in a set of clinical symptoms.
Data Analysis
We evaluated the demographic and sports characteristics of athletes and coaches. We calculated the cumulative incidence of presumed concussions as the number of new concussions occurring during the football and girls’ soccer season reported by the athlete (or the parent if the athlete report was not available) divided by the total number of athletes enrolled in the study. We also calculated the incidence rates of concussions overall, by practice and game, and by sport using the number of athlete-exposures (AEs) reported by athletes with the IVR system. In this calculation, we only counted AEs reported to the IVR system and only counted possible concussions that occurred during the same week of play for which the AE was reported to the IVR system. If athletes missed a week of reporting, then neither numerator nor denominator data were counted. For athletes who dropped out during the season, only weeks that they played and reported to the IVR system were included in the count of AEs and possible concussions. We assessed the association between the requirement for concussion education, type of education, number of education modalities, coach awareness of graduated return to play and the Lystedt Law, and risk of the coach being aware of the athlete’s concussion using Poisson regression to estimate relative risks and 95% confidence intervals, accounting for clustering by school. We defined coach awareness as the athletes reporting their symptoms to the coach or the athletes stating that their coach was already aware of the hit resulting in symptoms.
Results
We contacted 68 eligible schools to obtain the 20 schools that agreed to participate. The 20 football teams had a beginning mean roster of 79 players (range, 47-114), and the 20 girls’ soccer teams had a beginning mean roster of 38 players (range, 22-67). We enrolled 778 dyads, consisting of 490 football players and their parents and 288 female soccer athletes and their parents, with a participation rate of 30.9% and 38.0%, respectively, based on the roster of the teams at the start of the season. During the course of the season, 217 study participants (155 football, 62 soccer) dropped out of the study either because they stopped playing for the remainder of the season because of an injury (n = 19) or for other reasons such as being cut from the team or not providing any data at all on AEs. A mean of 63.4% of athletes and 67.0% of parents called in weekly to report the number of practices and games in which they played that week as well as whether they had suffered a hit with associated concussive symptoms. There were a total of 23,212 AEs reported through the IVR system.
Characteristics of Study Participants
The athletes were distributed across all 4 high school grades, with football players more likely to be older, to be black, and to qualify for the free lunch program (Table 1). Most had been playing the sport for >2 years, with more girls playing it for ≥10 years. Two-thirds reported experiencing concussive symptoms after a hit to the head or body before the 2012 sports season, with the majority of these having >1 such episode and football players more likely to report ≥5 prior concussions. However, only one-third of athletes who had experienced symptoms consistent with concussions reported receiving a concussion diagnosis.
Demographic Characteristics of High School Athletes by Sport in Washington State, 2012 a
Values are expressed as n (%). Some columns do not add up to the totals due to missing data.
Missing data on 1 football player.
Statistically significant difference (P < .05).
Concussions in Athletes
During the football and girls’ soccer season, 122 athletes reported 147 concussions, of which 100 were incident (ie, new for that season) among the 778 players. Because we had AEs for only 83 of the 100 incident concussions, the cumulative incidence of players with a presumed concussion was 10.7% overall, with a similar cumulative incidence in girls’ soccer and football. There were 22 athletes with ≥2 presumed concussions during the season. The incidence of concussions at 3.6 per 1000 AEs was identical across the 2 sports (Table 2). For football, the rate of concussions in games was 6.3-fold greater than in practices, while girls’ soccer had a 12.4-fold greater rate of concussions in games compared with practices.
Incidence Rates of Concussions Among High School Athletes in Washington State, 2012 a
AE, athlete-exposure.
The most common symptom reported by athletes was headache (94%), followed by pressure in the head, not feeling right, and dizziness (Table 3). Symptoms with the highest reported severity were headache, memory loss, and fatigue, while the least severe were concentration problems, nervousness, and confusion. Loss of consciousness was only reported in 7% of concussions. We found 69% of athletes reported playing with symptoms.
Types and Severity of Concussive Symptoms Among High School Athletes in Washington State, 2012
Graded on a scale from 0 to 6, with higher scores indicating more severe symptoms.
Association of Law-Mandated Coach Concussion Education With Coach Awareness of Concussions
Coach age, sex, level of education, coaching experience and certification, and having an athletic trainer at the school did not significantly differ among players whose coach was or was not aware of their concussion (Table 4).
Characteristics of Coaches Who Were Aware or Not Aware of Their High School Athletes’ Concussions in Washington State, 2012 a
Values are expressed as n (%). Some columns do not add up to the totals due to missing data.
Among concussed athletes, 40% reported that their coach was not aware of their concussion, and this proportion did not differ by sport. Coaches who had used the video and quiz as concussion education modalities were 50% and 40% less likely to be aware of their athletes’ concussions, respectively (Table 5); there were no differences by sport. We found no differences in coach awareness by the number of educational modalities or awareness of the Lystedt Law.
Association Between Coach Education and Likelihood of the Coach Being Aware or Not Aware of Concussions a
Values are expressed as n (%) unless otherwise indicated.
Adjusted for coaching experience.
Statistically significant difference (P < .05).
Discussion
In this prospective study, we found that the cumulative incidence of sports-related concussions was more than 10% for both high school football and girls’ soccer players. The majority of athletes continued to practice or play while symptomatic, and 40% reported that their coaches were not aware of their concussive symptoms, despite having to sign a statement at the beginning of the season stating that concussive symptoms should be reported to the coach. Coach awareness of athletes’ concussions did not vary greatly by the different avenues for concussion education received by coaches, although some methods, namely the use of a video and accompanying quiz, may have been less effective than others.
There are a number of important strengths to this study. The sample consisted of schools from both rural and urban areas in the state and included both male and female athletes. We were able to calculate the incidence per 1000 AEs; we only counted the occurrence of symptoms that occurred during weeks that we also had data on the number of AEs. Our determination of concussions did not rely on the reporting of symptoms by athletes to coaches or athletic trainers or on recognition by the athletes or their parents that the youth had a concussion. Concussion determination only required recognition and reporting that athletes had symptoms associated with a hit during play and confirmation by an interview reviewed by physicians. We thus were able to capture potential concussive injuries that are missed by other injury surveillance systems.
There are also important limitations to consider when interpreting the results. The determination of concussions did not include a physical examination of the athlete or neuropsychological testing. However, the diagnosis of a concussion by history of a blow to the head or body followed by neurological impairment and reported symptoms is based on consensus guidelines9,21 and prior studies.18,19,25 A concussion remains a clinical diagnosis, based primarily on reported symptoms. Postural instability has a sensitivity of only 34% to 64% and is most useful only during the first 72 hours after the injury. 1 Most concussion diagnoses in clinical practice are made without the use of neuropsychological tests. 17 The type of education completed by coaches and athletes was based on self-report. This was not a longitudinal study, so we cannot compare the current educational program to that before the Lystedt Law was passed in Washington State. We did not have comparable sports between male and female athletes, limiting the ability to make comparisons by sex. Our study results may also have been affected by low participation rates for athletes and parents at our schools. Families that participated in the study may have been more likely to be interested in and report concussions, biasing our incidence estimates of concussions. Lastly, because ≥30% of athletes and parents did not report weekly to the IVR system, this likely resulted in missed concussions and AE time and some measurement error in our concussion rates. If athletes were more likely to report to the IVR system during weeks when they had symptoms compared with weeks that they played but had no symptoms, the resulting bias would be to overestimate the incidence of concussions.
The varying rates of concussions in different studies underscore the problem created by the lack of accurate and clinically useful biomarkers for a concussion. In the absence of such biomarkers, the current identification of concussions relies heavily on reported symptoms that are nonspecific. Some of the youth reporting concussive symptoms may not have had a brain injury. However, the reliance on athletic staff identification of injured athletes, in the presence of the current sports culture of not reporting symptoms, is also potentially flawed, albeit in the direction of much lower sensitivity for the identification of all injured athletes.
All studies, including ours, report higher rates of concussions in competitive games than in practices.1,16,23,24 The incidence of sports-related concussions that we found is about an order of magnitude higher than that which has been previously reported likely because of differences in injury identification. Marar and colleagues, 16 using a detailed athletic trainer determination of an injury, reported that the incidence of concussions in high school football was 0.6 per 1000 AEs, while we found a rate of 3.6 per 1000 AEs. For girls’ soccer, Marar et al 16 reported a rate of 0.3 per 1000 AEs in contrast to our findings of 3.6 per 1000 AEs. Other prior studies also found lower rates of concussions in football23,24 and in soccer 24 than those reported here. Our method of ascertaining concussions likely accounts for the higher incidence in our study and may be closer to the actual incidence than those in studies relying on reporting by trainers or coaches or based on the use of medical care. This is particularly true for soccer, as athletic trainers rarely are assigned to cover soccer, and thus, soccer concussions may not be well captured by data sets that rely on athletic trainer reports.
We found that the incidence of presumed concussions was the same in boys’ football and girls’ soccer, in contrast to other studies that report higher rates in the former.15,16 We are unable to determine if this is because of a higher rate of concussive symptoms in girls for a given hit to the head, a higher proportion of girls than boys reporting their symptoms with our IVR system, or a truly similar rate of concussive symptoms. Further study will require comparing girls and boys in the same sport, with the potential collection of force data in both sexes.
Many athletes in our study did not report their symptoms to their coach. The problem of the lack of recognition of concussions by high school athletes 1 and the reluctance to report2,20 have contributed to the well-known results of the underreporting of youth concussions.20,28 The Lystedt Law and other similar laws have tried to address this issue by mandating education to athletes and their families. The type and amount of education vary with each law, 12 and the reporting may be different in other states.
Management guidelines developed in recent years call for the removal of athletes from practice or play and a graduated return to sports that does not begin until symptoms have resolved.9,21 A central component of the Lystedt Law 11 and laws in other states is removal of the athlete from practice or play immediately and continued removal until evaluation by a health care provider and symptoms resolve. However, 69% of study athletes with concussions played with symptoms, likely at least in part because of many not telling their coaches of their injury. The amount and type of coach concussion education did not appear to be strongly associated with the coaches’ awareness of the concussion. While concussion laws and mandated education may improve the recognition and management of sports-related concussions, only 5 states explicitly require prospective evaluation of the efficacy of coach education or information requirements. 12 Some students also reported not telling their parents about their symptoms, which we learned when we had both parent and student reports of symptoms for the same event. We used athlete reports of their symptoms on the weekly IVR calls unless they did not call in; if so, we then used parent reports if available. This helped to maximize the number of athletes on whom we had weekly data.
There are a number of important implications of this study. Current estimates of the number of sports-related concussions from the Centers for Disease Control and Prevention based on the National Health Interview Survey 14 or emergency department surveillance 7 are probably unreliable. Even carefully conducted studies using athletic trainer reporting are likely underestimating the incidence of concussions by an order of magnitude or more. More accurate and objective methods of identifying concussions are necessary, and newer methods of reporting concussions, such as used here, are needed to better understand the epidemiology of this important problem.
In addition, attitudes of athletes regarding the reporting of concussive symptoms are a major barrier to the proper care of players with concussions, and a change in these attitudes will not be accomplished through legislation alone, although legislation can reinforce voluntary educational efforts.2,26 It will likely require a change in the culture of a team; the attitude of athletic staff, athletes, and parents; and educational interventions that are effective in underscoring the reasons for reporting. These interventions will require theory-based development and careful testing including randomized controlled trials before they are widely disseminated. A recent commentary promotes the idea that the use of health promotion and social marketing tools would improve the dissemination and implementation of concussion guidelines. 6 Further interventions to improve the recognition, reporting, and management of concussions are needed, and these should be tested with rigorous research designs. All these interventions may require time to ensure that the change in culture is fully embraced by the next, as well as current, generation of student athletes, their coaches, and their parents. 13
Footnotes
Acknowledgements
The authors thank the athletic staff of the participating schools, the athletes, and their parents for their involvement in the study. They also thank the research staff of Matthew Vaughn, Katherine Lepere, Adam Strizich, Asha Thomas, Charla Jones, Denise Hopkins, Sierra Rotakhina, Melissa Rudd, Adessa Churape, Rachel Silverstein, Reba Blissell, Jennifer Maeser, and Christopher Mack for their dedication and work.
One or more of the authors has declared the following potential conflict of interest or source of funding: The study was supported by the Public Health Law Research Program of the Robert Wood Johnson Foundation. The grant went to the University of Washington and not to any of the individual authors.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
