- Email: [email protected]
Factors associated with post-concussion syndrome in high school student-athletes
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Science and Medicine in Sport journal homepage: www.elsevier.com/locate/jsams
Original research
Factors associated with post-concussion syndrome in high school student-athletes Zachary Y. Kerr a,b,∗ , Scott L. Zuckerman c,d , Erin W. Wasserman b,e , Christina B. Vander Vegt b,f , Aaron Yengo-Kahn c,d , Thomas A. Buckley g , Gary S. Solomon c,d , Allen K. Sills c,d , Thomas P. Dompier h a
Department of Exercise and Sport Science, University of North Carolina at Chapel Hill, USA Matthew A. Gfeller Sport-related Traumatic Brain Injury Research Center, University of North Carolina at Chapel Hill, USA c Vanderbilt Sports Concussion Center, Vanderbilt University School of Medicine, Medical Center East, USA d Department of Neurological Surgery, Vanderbilt University School of Medicine, USA e Datalys Center for Sports Injury Research and Prevention, USA f Human Movement Science Curriculum, University of North Carolina at Chapel Hill, USA g Department of Kinesiology and Applied Physiology, University of Delaware, USA h Department of Athletic Training, Lebanon Valley College, USA b
a r t i c l e
i n f o
Article history: Received 4 April 2017 Received in revised form 28 July 2017 Accepted 31 August 2017 Available online xxx Keywords: Brain injuries Adolescent Sports Epidemiology
a b s t r a c t Objectives: To identify factors associated with post-concussion syndrome (PCS) among a national sample of high school student-athletes from the 2011/12–2013/14 academic years. Design: Ambispective cohort study from sports injury surveillance data. Methods: Sport-related concussion data originated from the National Athletic Treatment, Injury and Outcomes Network (NATION) surveillance program, consisting of 27 sports from a convenience sample of 196 high schools across 26 states. All SRCs were reported by certified athletic trainers. The PCS and non-PCS groups consisted of concussed individuals with symptoms resolution time of >4 weeks and ≤2 weeks, respectively. Logistic regression estimated the association of athlete and concussion characteristics on the odds of PCS, and calculated adjusted odds ratios (OR) and 95% confidence intervals (CI). Results: Overall, 1334 concussed high school athletes met inclusion criteria: 215 in the PCS group and 1119 in the non-PCS group. In the multivariable analysis, concussion symptoms associated with increased odds of PCS included: retrograde amnesia (OR = 3.01, 95%CI: 1.31–6.91), difficulty concentrating (OR = 2.72, 95%CI: 1.56–4.77), disorientation (OR = 1.86; 95%CI: 1.04–3.33), insomnia (OR = 2.79; 95%CI: 1.62–4.80), loss of balance (OR = 1.76; 95%CI: 1.00–3.10), sensitivity to noise (OR = 1.80; 95%CI: 1.02–3.17), and visual disturbance (OR = 2.21; 95%CI: 1.23–3.97). Sex and recurrent concussion were not associated with PCS. Conclusions: As in previous research, somatic and cognitive symptoms were associated with PCS. The identification of factors associated with PCS may assist clinicians in identifying concussed athletes at greater risk of having longer symptom resolution time. © 2017 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
1. Introduction An estimated 1.6–3.8 million sports-related traumatic brain injuries occur annually in the United States.1 While sport-related concussion (SRC) has been extensively studied in college athletics, fewer studies have included high school student-athletes even
Abbreviations: CI, Confidence interval; OR, Odds ratio; PCS, Post-concussion syndrome; SRC, Sport-related concussion. ∗ Corresponding author. E-mail address: [email protected] (Z.Y. Kerr).
though they outnumber collegiate student-athletes.2 From a public health vantage, nearly 8 million high school student-athletes are at risk for SRC annually, and have access to fewer medical resources compared to collegiate and professional athletes. Additionally, high school sports injury surveillance has reported increases in SRC rates across time.3 Further study of SRC within the high school setting is warranted, particularly as it relates to prognosis and outcomes. Whereas 85–90% of athletes with SRC are asymptomatic by 10–14 days post-injury,4 a subset experience prolonged symptoms, referred to as post-concussion syndrome (PCS).5 PCS has many definitions, as described by the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10)6 and
http://dx.doi.org/10.1016/j.jsams.2017.08.025 1440-2440/© 2017 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
2
the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV).7 However, the most clinically-utilized definition, evidenced from a survey of over 500 physicians, is endorsement of at least 1 symptom for at least 4 weeks.8 Previously-discussed factors associated with PCS in sport populations include initial symptom severity,4 loss of consciousness,9 amnesia,4 and female sex,10 though considerable debate surrounds each factor. Very few of these studies have targeted specific playing levels (i.e. high school or collegiate student-athletes only). If associated factors can be identified in a homogeneous population, within a similar school environment, concussed athletes can be more appropriately managed to minimize risk of recurrent concussion, symptom duration, and improve quality of life. Given the potentially debilitating effects of prolonged postconcussive symptoms in high school student-athletes, we sought to further investigate risk factors for PCS after SRC. Using data from the National Athletic Treatment, Injury and Outcomes Network (NATION) surveillance program, this study examined factors associated with PCS among high school student-athletes who sustained SRCs during the 2011/12–2013/14 academic years.
2. Methods NATION was designed and implemented by the Datalys Center for Sports Injury Research and Prevention in 2011 and used a convenience sample of high schools drawn from 26 states to compile data on 27 sports. Data were collected across the 2011/12–2013/14 academic years.11 Forty-seven high schools participated in year one, 68 schools in year two, and 147 schools in year three; 30 schools participated all three years. Most high schools were public (84.4%), co-educational (98.6%), set in non-urban areas (75.5%), and had enrollment under 1000 students (51.0%).11 Both full-time and parttime Athletic Trainers (ATs) from these schools collected injury and exposure data, and were internally hired or contracted from nearby clinics or university graduate programs. The NATION project was reviewed by the Western Institutional Review Board (Puyallup, WA) and deemed exempt from human subject protections review. The funding organizations had no role in data collection, analysis, and interpretation, and did not have the right to approve or disapprove any resulting publications. All reported injuries were evaluated and/or treated by ATs who attended practices and competitions during the preseason, regular season, and postseason. All athletic injuries and exposures were recorded. Using a common data element standard, data were gathered from multiple electronic medical record applications. ATs were able to document injuries as part of their clinical practice using their preferred software application.11 For each SRC, the AT completed a detailed report on the injury. Symptoms were selected from a 17-item yes/no checklist originating from the National Collegiate Athletic Association (NCAA) Injury Surveillance Program (ISP).12 After initially inputting injury data, ATs could return to update the injury report as needed. Thus, delayed symptoms were reported, which consequently made our symptomatology measure an aggregate of symptoms reported at any point during recovery. We flagged concussions as recurrent when concussions were not the first sustained by athletes during their team’s participation in NATION, or when their AT noted the concussion as being recurrent. Data were exported and sent to the Datalys Center to be verified and analyzed. Prior to export, all identifying information was removed and the remaining variables were encrypted. ATs could modify injury data up to 30 days postseason. NATION data quality control staff reviewed data across the academic year for accuracy and completeness, reducing risk of memory decay. When invalid
Fig. 1. Flow-chart of final cohort.
data were flagged, the quality control staff and ATs worked together to rectify the error before data entered the aggregate database. A reportable injury in NATION was defined as an injury that: (1) occurred as a result of participation in an organized high school practice/competition, and (2) required attention from an AT or physician. A pre-determined definition of concussion was not provided as we relied on the medical expertise of the AT managing and reporting the injury. However, in lieu of local laws or guidance, ATs were encouraged to follow the international definition of concussion provided by the Consensus Statement on Concussion in Sport.13 Previous research in the high school setting has highlighted that ATs are suitable data collectors for injury surveillance purposes.14 Cases were defined as concussed student-athletes experiencing one or more symptom related to the SRC for greater than 4 weeks, as has been characterized in previous reports of PCS.15–17 Controls were defined as concussed student-athletes with symptom resolution at ≤2 weeks post-injury. Any student-athlete with symptom resolution in the intermediate area of >2 weeks but ≤4 weeks was excluded from our analysis in order to sharply demarcate the control and PCS groups, similar to methodology utilized in prior research.18 In addition, cases and controls with missing data were excluded. The primary purpose of this study was to assess factors associated with prolonged SRC symptomatology. Therefore, unlike other research that utilized sports injury surveillance data,19 the unit of analysis was not concussions, but each concussed individual. Thus, for each concussed individual used in the data analysis, only the most recent concussion reported in NATION was examined. Overall, 1334 concussed high school athletes met criteria for inclusion, of which 215 were in the PCS group and 1119 were in the non-PCS (control) group (Fig. 1); 271 were excluded for having symptom resolution time of >2 weeks but ≤4 weeks, and 241 had missing data. Those included and excluded in the study did not differ by sex (p = 0.07) or sport (p = 0.38).
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
Whereas previous research evaluating PCS has included individual and family medical histories,18 the aggregate data collection of NATION focused on injury incidence and lacked in-depth individual information such as past medical history or number of concussions outside of sport, for example. Variables assessed were sex, sport, contact level, helmet status, recurrence, number of symptoms, and prevalence of each of the 17 symptoms (Supplementary Table A). Contact level was determined by a pre-existing classification system.20 Helmet status was determined via groupings from previous research that identified sports in which athletes typically participated while wearing helmets.21 Those sports in which athletes played part of the time with helmets (e.g., baseball, softball) were included in the helmeted group. Also, in field hockey, athletes identified as goaltenders were in the helmeted group, as they were required to wear helmets. Number of symptoms was treated as a discrete variable. Data were analyzed to describe student-athletes with PCS. Quantitative data were presented as means with standard deviations. Categorical data were presented as percentages of the total population. Logistic regression estimated odds ratios (OR) and 95% confidence intervals (CI). Simple univariable logistic regressions predicted the odds of PCS. All variables whose univariate analyses yielded p-values <0.10 were included in a multiple logistic regression model. For this multivariable analysis, all ORs with p-values <0.05 were considered statistically significant. Given previous research suggesting that sex was associated with PCS,10 we reanalyzed the multivariable model stratified by sex. Data were analyzed using SAS-Enterprise Guide software (version 5.1; SAS Institute Inc., Cary, NC).
3. Results Boys’ football comprised the largest proportion of individuals in the PCS (45.1%) and non-PCS (52.1%) groups (Table 1). The PCS and non-PCS groups did not differ in regards to helmet status (p = 0.44) and recurrent concussion (p = 0.19) status. However, distributions differed between the PCS and non-PCS groups in regards to sex (p = 0.007), and sport contact level (p = 0.001). Also, the PCS group had a higher average number of symptoms reported than the nonPCS group (p < 0.001). In univariable analyses, the odds of PCS were lower in males than females (OR = 0.65; 95%CI: 0.48–0.89; Supplementary Table B). Compared to low-/non-contact sports, the odds of PCS were also lower in collision sports (OR = 0.54; 95%CI: 0.37–0.78) and high-contact sports (OR = 0.49; 95%CI: 0.32–0.75). In terms of symptoms, the odds of PCS increased 20% for each additional symptom reported (OR = 1.20, 95%CI 1.14–1.26). In addition, the odds of PCS were higher among 14 of the 17 concussion symptoms; the symptoms not associated with PCS were: excess excitability, loss of consciousness, and tinnitus. A multivariable model was constructed utilizing all the variables whose ORs in univariable analyses yielded p-values < 0.10 (Table 2). Controlling for other variables, the odds of PCS were lower in high-contact sports than low-/non-contact sports (OR = 0.51; 95%CI: 0.30–0.85); a similarly reduced OR was found among collision sports but did not reach the a priori defined level of statistical significance (OR = 0.61; 95%CI: 0.33–1.16). Concussion symptoms associated with increased odds of PCS, when controlling for other variables, included: retrograde amnesia (OR = 3.01, 95%CI: 1.31–6.91), difficulty concentrating (OR = 2.72, 95%CI: 1.56–4.77), disorientation (OR = 1.86; 95%CI: 1.04–3.33), insomnia (OR = 2.79; 95%CI: 1.62–4.80), loss of balance (OR = 1.76; 95%CI: 1.00–3.10), sensitivity to noise (OR = 1.80; 95%CI: 1.02–3.17), and visual disturbance (OR = 2.21; 95%CI: 1.23–3.97). In the multivariable model considering males only, retrograde amnesia, difficulty concentrat-
3
ing, and insomnia were the only significant predictors. In the multivariable model considering females only, contact level and visual disturbance were the only significant predictors.
4. Discussion Our findings originate from surveillance data collected from concussed high school student-athletes during the 2011/12–2013/14 academic years. Various factors were associated with PCS, including sport contact level and various concussion on-field markers and symptoms (retrograde amnesia, difficulty concentrating, disorientation, loss of balance, visual disturbance, sensitivity to noise, and insomnia). These findings follow previous research, which found similar symptom clusters associated with PCS.22 Our findings, in conjunction with previous research, highlight that many factors may be associated with PCS and may help identify concussed athletes at high-risk for PCS. The findings also highlight the imperative need to continue examining risk factors associated with PCS across multiple levels of competition, sports, and settings with various study designs. The symptoms associated with PCS in this study fell into three categories: cognitive (retrograde amnesia, difficulty concentrating, disorientation), somatic (loss of balance, visual disturbance, and sensitivity to noise), and sleep (insomnia). In previous research, PCS was also associated with sensitivity to noise, fatigue, retrograde amnesia, difficulty with concentration, and insomnia.22 Interestingly, the total number of symptoms was not associated with PCS in our study, as it has been in others.23 Meehan et al.23 found that total score on symptom inventory was associated with symptoms lasting more than four weeks. In a comprehensive systematic review of general traumatic brain injury,24 total symptom severity score was associated with PCS in 50% of the studies discussed. Although various definitions of PCS specify a minimum number of symptoms required,6 our findings suggest that the type of symptoms — not quantity — may be more indicative of who is most at risk to develop PCS. These results highlight specific symptoms that may alert high school ATs and clinicians about the potential risk of longterm symptom resolution time. However, an important difference is that Meehan et al.23 used an aggregate scale that summed the severity of each symptom. In contrast, our study used a scale based upon symptom presence. The variations in symptom measurement cannot be overlooked and warrant further study. Previous research has established that the highest concussion rates originate from collision sports including football, wrestling, and ice hockey.19 In our study, most concussed individuals were from collision sports, although this may be due to varying distributions of sports participating within NATION. To our surprise, when examining PCS by contact level, the highest proportion of concussed individuals with PCS originated from the low-/no-contact group. Our study cannot pinpoint the reason(s) for the increased odds of PCS within this group. Possible reasons may be that given the low number of head impacts in low/non-contact sports, the ones that do occur may have increased severity. Whereas collision sport athletes may expect to absorb head impacts routinely through practice and game, low/non-contact sport athletes may not be adequately prepared for such impacts and/or how to deal with them safely.25 Also, collision sport athletes may be more accustomed to symptoms and thus less likely to report them, or may overstate their symptom resolution in order to return to play sooner.26 Concussion management may also vary by sport contact level. Nevertheless, the findings suggest that PCS is not unique to collision and high contact sports. Furthermore, proper enforcement of rules restricting contact is essential in low-/no-contact sports to ensure the health and safety of their athletes.
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model
ARTICLE IN PRESS
JSAMS-1618; No. of Pages 6
Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
4
Table 1 Characteristics of concussed student-athletes in high school sports.
Sex Male Female Sport Boys’ football Boys’ wrestling Girls’ basketball Boys’ lacrosse Girls’ soccer Boys’ basketball Boys’ soccer Girls’ volleyball Girls’ field hockey Girls’ lacrosse Girls’ softball Boys’ baseball Girls’ track & field Girls’ swimming & diving Boys’ track & field Girls’ gymnastics Boys’ swimming & diving Girls’ tennis Girls’ cross country Girls’ crew Contact level Collision High-contact Low-/non-contact Helmet status Helmeted Un-helmeted Recurrent concussion Yes No Number of symptoms (Mean, SD)
Total
Post-concussion syndrome (PCS)
Non-PCS
986 (73.9) 348 (26.1)
143 (66.5) 72 (33.5)
843 (75.3) 276 (24.7)
680 (51.0) 81 (6.1) 79 (5.9) 79 (5.9) 77 (5.8) 65 (4.9) 57 (4.3) 51 (3.8) 43 (3.2) 35 (2.6) 34 (2.5) 14 (1.0) 12 (0.9) 7 (0.5) 7 (0.5) 5 (0.4) 3 (0.2) 3 (0.2) 1 (0.1) 1 (0.1)
97 (45.1) 16 (7.4) 10 (4.7) 13 (6.0) 13 (6.0) 6 (2.8) 7 (3.3) 10 (4.7) 12 (5.6) 7 (3.3) 9 (4.2) 2 (0.9) 7 (3.3) 2 (0.9) 1 (0.5) 2 (0.9) 1 (0.5) 0 0 0
583 (52.1) 65 (5.8) 69 (6.2) 66 (5.9) 64 (5.7) 59 (5.3) 50 (4.5) 41 (3.7) 31 (2.8) 28 (2.5) 25 (2.2) 12 (1.1) 5 (0.4) 5 (0.4) 6 (0.5) 3 (0.3) 2 (0.2) 3 (0.3) 1 (0.1) 1 (0.1)
761 (57.0) 357 (26.8) 216 (16.2)
113 (52.6) 49 (22.8) 53 (24.7)
648 (57.9) 308 (27.5) 163 (14.6)
893 (66.9) 441 (33.1)
139 (64.7) 76 (35.3)
754 (67.4) 365 (32.6)
118 (8.8) 1216 (91.2) 5.3 (2.9)
14 (6.5) 201 (93.5) 6.7 (3.2)
104 (9.3) 1015 (90.7) 5.0 (2.8)
1334 (100.0)
215 (100.0)
1119 (100.0)
p-Valuea 0.007
n/a
0.001
0.44
0.19
Total
<0.001
a p-Value from tests that examines differences between the PCS and non-PCS groups. For sex, contact level, helmet status, and recurrent concussion, chi-square tests were used. For number of symptoms, an independent samples t-test were used. Test was not run for sport.
Table 2 Multivariable analyses predicting odds of post-concussion syndrome (PCS) in concussed student-athletes in high school sports. Variable
Values
Odds ratio (95%CI) Overall (n = 1334)
Males only (n = 986)
Females only (n = 348)
Sex Contact levela
Male vs. female Collision High-contact Low-/non-contact (1-unit increase) Yes vs. no
0.85 (0.49–1.47) 0.61 (0.33–1.16) 0.51 (0.30–0.85)* 1.00 0.62 (0.33–1.16)
– 0.57 (0.18–1.77) 0.46 (0.14–1.52) 1.00 0.74 (0.45–1.19)
– – 0.48 (0.27–0.87)* 1.00 0.79 (0.37–1.70)
1.38 (0.73–2.61) 3.01 (1.31–6.91)* 2.72 (1.56–4.77)* 1.86 (1.04–3.33)* 1.39 (0.77–2.49) 1.34 (0.77–2.34) 1.44 (0.79–2.65) 4.23 (0.90–19.91) 2.79 (1.62–4.80)* 1.76 (1.00–3.10)* 1.19 (0.69–2.07) 1.32 (0.74–2.35) 1.80 (1.02–3.17)* 2.21 (1.23–3.97)*
1.31 (0.62–2.77) 3.60 (1.40–9.24)* 2.79 (1.43–5.43)* 1.78 (0.89–3.56) 1.36 (0.68–2.75) 1.42 (0.74–2.72) 1.35 (0.64–2.84) 3.34 (0.67–16.57) 3.29 (1.72–6.31)* 1.71 (0.87–3.36) 0.87 (0.44–1.70) 1.17 (0.58–2.36) 2.06 (1.03–4.13)* 1.60 (0.78–3.25)
1.62 (0.46–5.67) 0.99 (0.14–6.90) 2.14 (0.75–6.12) 1.76 (0.61–5.09) 1.33 (0.46–3.85) 0.96 (0.33–2.82) 1.42 (0.48–4.22) – 1.51 (0.54–4.18) 1.74 (0.62–4.90) 1.81 (0.67–4.91) 1.29 (0.45–3.66) 1.42 (0.51–3.96) 3.39 (1.15–10.02)*
Number of symptoms Symptom prevalence Posttraumatic amnesia Retrograde amnesia Difficulty concentrating Disorientation Dizziness Excess drowsiness Excess irritability Headacheb Insomnia Loss of balance Nausea/vomiting Sensitivity to light Sensitivity to noise Visual disturbance
Note: Univariable analyses found no association between PCS and the following variables, which were consequently excluded from this final multivariable model: helmet status, recurrent concussion, and symptom prevalence of excess excitability, loss of consciousness, and tinnitus. a The contact level category of “collision” was not assessed in the model with females only because no females played in collision sports. b Headache was not included in the model with females only because all females reported headache. * Denotes statistical significance (i.e., 95%CI does not include 1.00).
It is important to note that sex-stratified models indicated that this higher prevalence of PCS in low-/non-contact athletes was
attributable to females. Previous research has found that females generally have worse outcomes post-concussion than males.10
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
Also, previous concussion has also been found to be strongly associated with the development of PCS.22 However, our study found neither sex nor recurrent concussion to be associated with PCS. Much discussion has examined both biological and socio-cultural factors associated with sex differences in SRC.19,27 Future research needs to continue refining our methods to pinpoint specific factors that may be potentially attributable to sex differences, while accounting for variations in contact level in the sports played. Our findings parallel a recent study examining high school athletes with multiple concussions within a season28 ; although return to play time was longer with subsequent concussions, symptom resolution time did not differ. Our target population is relatively younger than collegiate or professional athletes, and in comparison to higher-level collegiate and professional athletes, high school athletes have sufficiently less exposure, which implies decreased opportunity for sustaining concussions. Given the lack of on-site management at lower levels of competition (e.g., recreational leagues, middle school sports), is possible that concussion history prior to high school was unknown. Regardless, it is essential for clinicians to not only obtain previous concussion history, but also work with young athletes of both sexes to ensure they are aware of the implications of concussion and the importance of disclosing and seeking appropriate care. Finally, compared to previous research utilizing a similar methodology with the NCAA-ISP,22 we found that the proportion of included athletes with PCS among included samples in the current study was higher [16.1% (215/1334) vs. 7.4% (112/1507)]. A previous study also found that the odds of return to play time ≥30 days post-concussion was higher in high school than college students, but symptom resolution time was not examined.29 One potential reason for this longer symptom resolution time may due to differences in day-to-day activities, changes in biochemistry and ongoing brain development. Whereas collegiate student-athletes have relatively less (albeit more challenging) class time, high school athletes are in class for 6–7 h continuously with minimal break time. The constant mental focus required for several hours may be affected adversely by concussion, whereas a collegiate athlete may be able to recover more quickly with multiple breaks. Furthermore, high school student-athletes may lack the resources college studentathletes receive from their institutions, such as tutoring, academic accommodations, and note-taking services. Given the proportions of concussed student-athletes with PCS, it is essential to identify interventions that not only can assist with return-to-play, but also return-to-learn. Because our study did not examine components related to return to learning, it is important for future research to consider methods that simultaneously examine post-SRC protocols for both return-to-play and return-to-learn. Recommendations exist for graduated return-to-learn protocols30 and should be evaluated for feasibility and effectiveness. Our study has limitations. NATION utilizes a convenience sample of high school sports programs and may not be generalizable to non-participating sports programs or other levels of competition (e.g., club sports, college sports). Data were also collected from a previous time-period that may not represent the current climate of concussion reporting. Although ATs were encouraged ATs to follow the Consensus Statement on Concussion in Sport,13 we could not ensure that ATs utilized this definition for diagnostic purposes. Conflicting PCS definitions impede researchers’ ability to properly estimate the prevalence and incidence of PCS and associated risk factors. In a survey of over 500 sports medicine physicians, the most commonly enlisted PCS criteria was at least one symptom for at least four weeks.8 Although NATION utilizes a 17-symptom checklist, it does not measure the severity of each symptom nor could it account for time of onset or duration. Future research should continue to examine how the type, number, and severity of symptoms are associated with long-term outcomes associated
5
with concussion at the high school (and other) level(s). Such additional information on symptoms will be helpful in sub-classifying PCS to help guide management strategies. As an ambispective study, we were unable to re-contact concussed individuals or participating schools to obtain additional information. Such variables of interest include those from the athlete (e.g., player position), school (e.g., staffing and resource allocations), and state (e.g., state-specific, concussion-related policy and legislation). In particular, NATION did not acquire medical history; thus, we did not have information on biopsychosocial factors such as learning disability, depression/anxiety, psychiatric illness, and family history that have been found to be associated with concussion outcomes.18 Our data were also collected via injury surveillance and includes only the information that team medical staff received during their clinical practice; thus, information on some variables related to athlete injury history may not be comprehensive.
5. Conclusion Our study examined factors associated with PCS at the high school level. Factors in a multivariable model predicting the odds of PCS included contact level as well as cognitive-, somatic-, and sleeprelated symptoms. Identify factors associated with PCS may assist sports medicine practitioners in better identifying those concussed athletes that may be at greater risk of having longer symptom resolution time. Further study is needed to better predict PCS and to develop and evaluate treatment strategies for at-risk individuals.
Practical implications • The specific symptoms endorsed by high school athletes following concussion may be more important than total symptoms in determining risk of PCS; athletes who endorse symptoms of memory, concentration, sleep, balance, noise, and visual difficulty may have a higher chance of developing PCS than those who do not. • Contact level of sport was not relevant in predicting the odds of PCS; PCS is seen in all athletes, both high and low contact sports. • Our findings, coupled with previous research, may suggest that high school student-athlete populations may have a greater prevalence of PCS than college student-athlete populations; thus, special attention should be given to the rigorous academic schedule and resources available to high school student-athletes in their return to school after a concussion.
Acknowledgments Funding for this study was provided by the National Athletic Trainers’ Association Research and Education Foundation and BioCrossroads in partnership with the Central Indiana Corporate Partnership Foundation. The organizations that funded this study had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the study sponsors. We thank the many athletic trainers who have volunteered their time and efforts to submit data to NATION. Their efforts are greatly appreciated and have had a tremendously positive effect on the safety of high school student-athletes.
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
6
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jsams.2017.08. 025. References 1. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of traumatic brain injury: a brief overview. J Head Trauma Rehabil 2006; 21(5):375–378. 2. National Federation of High Schools. Participation Statistics, 2016. Accessed 19 December 2016 http://www.nfhs.org/ParticipationStatics/ParticipationStatics. aspx/. 3. Lincoln AE, Caswell SV, Almquist JL et al. Trends in concussion incidence in high school sports: a prospective 11-year study. Am J Sports Med 2011; 39(5):958–963. 4. McCrea M, Guskiewicz K, Randolph C et al. Incidence, clinical course, and predictors of prolonged recovery time following sport-related concussion in high school and college athletes. J Int Neuropsychol Soc 2013; 19(1):22–33. 5. Ruff RM, Camenzuli L, Mueller J. Miserable minority: emotional risk factors that influence the outcome of a mild traumatic brain injury. Brain Inj 1996; 10(8):551–565. 6. Centers for Medicare and Medicaid Services. ICD-9-CM Diagnosis and Procedure Codes: Abbreviated and Full Code Titles. https://www.cms.gov/medicare/coding/ ICD9providerdiagnosticcodes/codes.html. Accessed 16 January 2017. 7. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-4, Washington, DC, American Psychiatric Publishing, 2000. 8. Rose SC, Fischer AN, Heyer GL. How long is too long? The lack of consensus regarding the post-concussion syndrome diagnosis. Brain Inj 2015; 29(7–8):798–803. 9. Luoto TM, Iverson GL, Losoi H et al. Clinical correlates of retrograde amnesia in mild traumatic brain injury. Brain Inj 2015; 29(5):565–572. 10. McNally KA, Bangert B, Dietrich A et al. Injury versus noninjury factors as predictors of postconcussive symptoms following mild traumatic brain injury in children. Neuropsychology 2013; 27(1):1–12. 11. Dompier TP, Marshall SW, Kerr ZY et al. The national athletic treatment, injury and outcomes network (NATION): methods of the surveillance program, 2011–2012 through 2013–2014. J Athl Train 2015; 50(8):862–869. 12. Wasserman EB, Kerr ZY, Zuckerman SL et al. Epidemiology of sports-related concussions in National Collegiate Athletic Association athletes from 2009–2010 to 2013–2014: symptom prevalence, symptom resolution time, and return-toplay time. Am J Sports Med 2016; 44(1):226–233. 13. McCrory P, Meeuwisse WH, Aubry M et al. Consensus statement on concussion in sport: the 4th International Conference on Concussion in Sport held in Zurich, November 2012. Br J Sports Med 2013; 47(5):250–258. 14. Lombardi NJ, Tucker B, Freedman KB et al. Accuracy of athletic trainer and physician diagnoses in sports medicine. Orthopedics 2016; 39(5):e944–e949.
15. McCauley SR, Wilde EA, Miller ER et al. Preinjury resilience and mood as predictors of early outcome following mild traumatic brain injury. J Neurotrauma 2013; 30(8):642–652. 16. Meehan 3rd WP, Mannix RC, Stracciolini A et al. Symptom severity predicts prolonged recovery after sport-related concussion, but age and amnesia do not. J Pediatr 2013; 163(3):721–725. 17. Tator CH, Davis H. The postconcussion syndrome in sports and recreation: clinical features and demography in 138 athletes. Neurosurgery 2014; 75(Suppl. 4):S106–S112. 18. Morgan CD, Zuckerman SL, Lee YM et al. Predictors of postconcussion syndrome after sports-related concussion in young athletes: a matched case-control study. J Neurosurg Pediatr 2015; 15(6):589–598. 19. Zuckerman SL, Kerr ZY, Yengo-Kahn A et al. Epidemiology of sports-related concussion in NCAA athletes from 2009–2010 to 2013–2014: incidence, recurrence, and mechanisms. Am J Sports Med 2015; 43(11):2654–2662. 20. Rice SG, American Academy of Pediatrics Council on Sports M, Fitness. Medical conditions affecting sports participation. Pediatrics 2008; 121(4):841–848. 21. Zuckerman SL, Lee YM, Odom MJ et al. Sports-related concussion in helmeted vs. unhelmeted athletes: who fares worse? Int J Sports Med 2015; 36(5):419–425. 22. Zuckerman SL, Yengo-Kahn AM, Buckley TA et al. Predictors of postconcussion syndrome in collegiate student-athletes. Neurosurg Focus 2016; 40(4):E13. 23. Meehan 3rd WP, Mannix R, Monuteaux MC et al. Early symptom burden predicts recovery after sport-related concussion. Neurology 2014; 83(24):2204–2210. 24. Silverberg ND, Gardner AJ, Brubacher JR et al. Systematic review of multivariable prognostic models for mild traumatic brain injury. J Neurotrauma 2015; 32(8):517–526. 25. Zuckerman SL, Totten DJ, Rubel KE et al. Mechanisms of injury as a diagnostic predictor of sport-related concussion severity in football, basketball, and soccer: results from a regional concussion registry. Neurosurgery 2016; 63:102–112. 26. Sallis RE, Jones K. Prevalence of headaches in football players. Med Sci Sport Exerc 2000; 32(11):1820–1824. 27. Broshek DK, Kaushik T, Freeman JR et al. Sex differences in outcome following sports-related concussion. J Neurosurg 2005; 102(5):856–863. 28. Currie DW, Comstock RD, Fields SK et al. A paired comparison of initial and recurrent concussions sustained by US high school athletes within a single athletic season. J Head Trauma Rehabil 2016. http://dx.doi.org/10.1097/htr.0000000000000240. Published online April 26. 29. Kerr ZY, Zuckerman SL, Wasserman EB et al. Concussion symptoms and return to play time in youth, high school, and college american football athletes. JAMA Pediatr 2016; 170(7):647–653. 30. Collins MW, Kontos AP, Okonkwo DO et al. Statements of agreement from the targeted evaluation and active management (TEAM) approaches to treating concussion meeting held in Pittsburgh, October 15–16, 2015. Neurosurgery 2016; 79(6):912–929.
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
ARTICLE IN PRESS Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Science and Medicine in Sport journal homepage: www.elsevier.com/locate/jsams
Original research
Factors associated with post-concussion syndrome in high school student-athletes Zachary Y. Kerr a,b,∗ , Scott L. Zuckerman c,d , Erin W. Wasserman b,e , Christina B. Vander Vegt b,f , Aaron Yengo-Kahn c,d , Thomas A. Buckley g , Gary S. Solomon c,d , Allen K. Sills c,d , Thomas P. Dompier h a
Department of Exercise and Sport Science, University of North Carolina at Chapel Hill, USA Matthew A. Gfeller Sport-related Traumatic Brain Injury Research Center, University of North Carolina at Chapel Hill, USA c Vanderbilt Sports Concussion Center, Vanderbilt University School of Medicine, Medical Center East, USA d Department of Neurological Surgery, Vanderbilt University School of Medicine, USA e Datalys Center for Sports Injury Research and Prevention, USA f Human Movement Science Curriculum, University of North Carolina at Chapel Hill, USA g Department of Kinesiology and Applied Physiology, University of Delaware, USA h Department of Athletic Training, Lebanon Valley College, USA b
a r t i c l e
i n f o
Article history: Received 4 April 2017 Received in revised form 28 July 2017 Accepted 31 August 2017 Available online xxx Keywords: Brain injuries Adolescent Sports Epidemiology
a b s t r a c t Objectives: To identify factors associated with post-concussion syndrome (PCS) among a national sample of high school student-athletes from the 2011/12–2013/14 academic years. Design: Ambispective cohort study from sports injury surveillance data. Methods: Sport-related concussion data originated from the National Athletic Treatment, Injury and Outcomes Network (NATION) surveillance program, consisting of 27 sports from a convenience sample of 196 high schools across 26 states. All SRCs were reported by certified athletic trainers. The PCS and non-PCS groups consisted of concussed individuals with symptoms resolution time of >4 weeks and ≤2 weeks, respectively. Logistic regression estimated the association of athlete and concussion characteristics on the odds of PCS, and calculated adjusted odds ratios (OR) and 95% confidence intervals (CI). Results: Overall, 1334 concussed high school athletes met inclusion criteria: 215 in the PCS group and 1119 in the non-PCS group. In the multivariable analysis, concussion symptoms associated with increased odds of PCS included: retrograde amnesia (OR = 3.01, 95%CI: 1.31–6.91), difficulty concentrating (OR = 2.72, 95%CI: 1.56–4.77), disorientation (OR = 1.86; 95%CI: 1.04–3.33), insomnia (OR = 2.79; 95%CI: 1.62–4.80), loss of balance (OR = 1.76; 95%CI: 1.00–3.10), sensitivity to noise (OR = 1.80; 95%CI: 1.02–3.17), and visual disturbance (OR = 2.21; 95%CI: 1.23–3.97). Sex and recurrent concussion were not associated with PCS. Conclusions: As in previous research, somatic and cognitive symptoms were associated with PCS. The identification of factors associated with PCS may assist clinicians in identifying concussed athletes at greater risk of having longer symptom resolution time. © 2017 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
1. Introduction An estimated 1.6–3.8 million sports-related traumatic brain injuries occur annually in the United States.1 While sport-related concussion (SRC) has been extensively studied in college athletics, fewer studies have included high school student-athletes even
Abbreviations: CI, Confidence interval; OR, Odds ratio; PCS, Post-concussion syndrome; SRC, Sport-related concussion. ∗ Corresponding author. E-mail address: [email protected] (Z.Y. Kerr).
though they outnumber collegiate student-athletes.2 From a public health vantage, nearly 8 million high school student-athletes are at risk for SRC annually, and have access to fewer medical resources compared to collegiate and professional athletes. Additionally, high school sports injury surveillance has reported increases in SRC rates across time.3 Further study of SRC within the high school setting is warranted, particularly as it relates to prognosis and outcomes. Whereas 85–90% of athletes with SRC are asymptomatic by 10–14 days post-injury,4 a subset experience prolonged symptoms, referred to as post-concussion syndrome (PCS).5 PCS has many definitions, as described by the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10)6 and
http://dx.doi.org/10.1016/j.jsams.2017.08.025 1440-2440/© 2017 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
2
the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV).7 However, the most clinically-utilized definition, evidenced from a survey of over 500 physicians, is endorsement of at least 1 symptom for at least 4 weeks.8 Previously-discussed factors associated with PCS in sport populations include initial symptom severity,4 loss of consciousness,9 amnesia,4 and female sex,10 though considerable debate surrounds each factor. Very few of these studies have targeted specific playing levels (i.e. high school or collegiate student-athletes only). If associated factors can be identified in a homogeneous population, within a similar school environment, concussed athletes can be more appropriately managed to minimize risk of recurrent concussion, symptom duration, and improve quality of life. Given the potentially debilitating effects of prolonged postconcussive symptoms in high school student-athletes, we sought to further investigate risk factors for PCS after SRC. Using data from the National Athletic Treatment, Injury and Outcomes Network (NATION) surveillance program, this study examined factors associated with PCS among high school student-athletes who sustained SRCs during the 2011/12–2013/14 academic years.
2. Methods NATION was designed and implemented by the Datalys Center for Sports Injury Research and Prevention in 2011 and used a convenience sample of high schools drawn from 26 states to compile data on 27 sports. Data were collected across the 2011/12–2013/14 academic years.11 Forty-seven high schools participated in year one, 68 schools in year two, and 147 schools in year three; 30 schools participated all three years. Most high schools were public (84.4%), co-educational (98.6%), set in non-urban areas (75.5%), and had enrollment under 1000 students (51.0%).11 Both full-time and parttime Athletic Trainers (ATs) from these schools collected injury and exposure data, and were internally hired or contracted from nearby clinics or university graduate programs. The NATION project was reviewed by the Western Institutional Review Board (Puyallup, WA) and deemed exempt from human subject protections review. The funding organizations had no role in data collection, analysis, and interpretation, and did not have the right to approve or disapprove any resulting publications. All reported injuries were evaluated and/or treated by ATs who attended practices and competitions during the preseason, regular season, and postseason. All athletic injuries and exposures were recorded. Using a common data element standard, data were gathered from multiple electronic medical record applications. ATs were able to document injuries as part of their clinical practice using their preferred software application.11 For each SRC, the AT completed a detailed report on the injury. Symptoms were selected from a 17-item yes/no checklist originating from the National Collegiate Athletic Association (NCAA) Injury Surveillance Program (ISP).12 After initially inputting injury data, ATs could return to update the injury report as needed. Thus, delayed symptoms were reported, which consequently made our symptomatology measure an aggregate of symptoms reported at any point during recovery. We flagged concussions as recurrent when concussions were not the first sustained by athletes during their team’s participation in NATION, or when their AT noted the concussion as being recurrent. Data were exported and sent to the Datalys Center to be verified and analyzed. Prior to export, all identifying information was removed and the remaining variables were encrypted. ATs could modify injury data up to 30 days postseason. NATION data quality control staff reviewed data across the academic year for accuracy and completeness, reducing risk of memory decay. When invalid
Fig. 1. Flow-chart of final cohort.
data were flagged, the quality control staff and ATs worked together to rectify the error before data entered the aggregate database. A reportable injury in NATION was defined as an injury that: (1) occurred as a result of participation in an organized high school practice/competition, and (2) required attention from an AT or physician. A pre-determined definition of concussion was not provided as we relied on the medical expertise of the AT managing and reporting the injury. However, in lieu of local laws or guidance, ATs were encouraged to follow the international definition of concussion provided by the Consensus Statement on Concussion in Sport.13 Previous research in the high school setting has highlighted that ATs are suitable data collectors for injury surveillance purposes.14 Cases were defined as concussed student-athletes experiencing one or more symptom related to the SRC for greater than 4 weeks, as has been characterized in previous reports of PCS.15–17 Controls were defined as concussed student-athletes with symptom resolution at ≤2 weeks post-injury. Any student-athlete with symptom resolution in the intermediate area of >2 weeks but ≤4 weeks was excluded from our analysis in order to sharply demarcate the control and PCS groups, similar to methodology utilized in prior research.18 In addition, cases and controls with missing data were excluded. The primary purpose of this study was to assess factors associated with prolonged SRC symptomatology. Therefore, unlike other research that utilized sports injury surveillance data,19 the unit of analysis was not concussions, but each concussed individual. Thus, for each concussed individual used in the data analysis, only the most recent concussion reported in NATION was examined. Overall, 1334 concussed high school athletes met criteria for inclusion, of which 215 were in the PCS group and 1119 were in the non-PCS (control) group (Fig. 1); 271 were excluded for having symptom resolution time of >2 weeks but ≤4 weeks, and 241 had missing data. Those included and excluded in the study did not differ by sex (p = 0.07) or sport (p = 0.38).
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
Whereas previous research evaluating PCS has included individual and family medical histories,18 the aggregate data collection of NATION focused on injury incidence and lacked in-depth individual information such as past medical history or number of concussions outside of sport, for example. Variables assessed were sex, sport, contact level, helmet status, recurrence, number of symptoms, and prevalence of each of the 17 symptoms (Supplementary Table A). Contact level was determined by a pre-existing classification system.20 Helmet status was determined via groupings from previous research that identified sports in which athletes typically participated while wearing helmets.21 Those sports in which athletes played part of the time with helmets (e.g., baseball, softball) were included in the helmeted group. Also, in field hockey, athletes identified as goaltenders were in the helmeted group, as they were required to wear helmets. Number of symptoms was treated as a discrete variable. Data were analyzed to describe student-athletes with PCS. Quantitative data were presented as means with standard deviations. Categorical data were presented as percentages of the total population. Logistic regression estimated odds ratios (OR) and 95% confidence intervals (CI). Simple univariable logistic regressions predicted the odds of PCS. All variables whose univariate analyses yielded p-values <0.10 were included in a multiple logistic regression model. For this multivariable analysis, all ORs with p-values <0.05 were considered statistically significant. Given previous research suggesting that sex was associated with PCS,10 we reanalyzed the multivariable model stratified by sex. Data were analyzed using SAS-Enterprise Guide software (version 5.1; SAS Institute Inc., Cary, NC).
3. Results Boys’ football comprised the largest proportion of individuals in the PCS (45.1%) and non-PCS (52.1%) groups (Table 1). The PCS and non-PCS groups did not differ in regards to helmet status (p = 0.44) and recurrent concussion (p = 0.19) status. However, distributions differed between the PCS and non-PCS groups in regards to sex (p = 0.007), and sport contact level (p = 0.001). Also, the PCS group had a higher average number of symptoms reported than the nonPCS group (p < 0.001). In univariable analyses, the odds of PCS were lower in males than females (OR = 0.65; 95%CI: 0.48–0.89; Supplementary Table B). Compared to low-/non-contact sports, the odds of PCS were also lower in collision sports (OR = 0.54; 95%CI: 0.37–0.78) and high-contact sports (OR = 0.49; 95%CI: 0.32–0.75). In terms of symptoms, the odds of PCS increased 20% for each additional symptom reported (OR = 1.20, 95%CI 1.14–1.26). In addition, the odds of PCS were higher among 14 of the 17 concussion symptoms; the symptoms not associated with PCS were: excess excitability, loss of consciousness, and tinnitus. A multivariable model was constructed utilizing all the variables whose ORs in univariable analyses yielded p-values < 0.10 (Table 2). Controlling for other variables, the odds of PCS were lower in high-contact sports than low-/non-contact sports (OR = 0.51; 95%CI: 0.30–0.85); a similarly reduced OR was found among collision sports but did not reach the a priori defined level of statistical significance (OR = 0.61; 95%CI: 0.33–1.16). Concussion symptoms associated with increased odds of PCS, when controlling for other variables, included: retrograde amnesia (OR = 3.01, 95%CI: 1.31–6.91), difficulty concentrating (OR = 2.72, 95%CI: 1.56–4.77), disorientation (OR = 1.86; 95%CI: 1.04–3.33), insomnia (OR = 2.79; 95%CI: 1.62–4.80), loss of balance (OR = 1.76; 95%CI: 1.00–3.10), sensitivity to noise (OR = 1.80; 95%CI: 1.02–3.17), and visual disturbance (OR = 2.21; 95%CI: 1.23–3.97). In the multivariable model considering males only, retrograde amnesia, difficulty concentrat-
3
ing, and insomnia were the only significant predictors. In the multivariable model considering females only, contact level and visual disturbance were the only significant predictors.
4. Discussion Our findings originate from surveillance data collected from concussed high school student-athletes during the 2011/12–2013/14 academic years. Various factors were associated with PCS, including sport contact level and various concussion on-field markers and symptoms (retrograde amnesia, difficulty concentrating, disorientation, loss of balance, visual disturbance, sensitivity to noise, and insomnia). These findings follow previous research, which found similar symptom clusters associated with PCS.22 Our findings, in conjunction with previous research, highlight that many factors may be associated with PCS and may help identify concussed athletes at high-risk for PCS. The findings also highlight the imperative need to continue examining risk factors associated with PCS across multiple levels of competition, sports, and settings with various study designs. The symptoms associated with PCS in this study fell into three categories: cognitive (retrograde amnesia, difficulty concentrating, disorientation), somatic (loss of balance, visual disturbance, and sensitivity to noise), and sleep (insomnia). In previous research, PCS was also associated with sensitivity to noise, fatigue, retrograde amnesia, difficulty with concentration, and insomnia.22 Interestingly, the total number of symptoms was not associated with PCS in our study, as it has been in others.23 Meehan et al.23 found that total score on symptom inventory was associated with symptoms lasting more than four weeks. In a comprehensive systematic review of general traumatic brain injury,24 total symptom severity score was associated with PCS in 50% of the studies discussed. Although various definitions of PCS specify a minimum number of symptoms required,6 our findings suggest that the type of symptoms — not quantity — may be more indicative of who is most at risk to develop PCS. These results highlight specific symptoms that may alert high school ATs and clinicians about the potential risk of longterm symptom resolution time. However, an important difference is that Meehan et al.23 used an aggregate scale that summed the severity of each symptom. In contrast, our study used a scale based upon symptom presence. The variations in symptom measurement cannot be overlooked and warrant further study. Previous research has established that the highest concussion rates originate from collision sports including football, wrestling, and ice hockey.19 In our study, most concussed individuals were from collision sports, although this may be due to varying distributions of sports participating within NATION. To our surprise, when examining PCS by contact level, the highest proportion of concussed individuals with PCS originated from the low-/no-contact group. Our study cannot pinpoint the reason(s) for the increased odds of PCS within this group. Possible reasons may be that given the low number of head impacts in low/non-contact sports, the ones that do occur may have increased severity. Whereas collision sport athletes may expect to absorb head impacts routinely through practice and game, low/non-contact sport athletes may not be adequately prepared for such impacts and/or how to deal with them safely.25 Also, collision sport athletes may be more accustomed to symptoms and thus less likely to report them, or may overstate their symptom resolution in order to return to play sooner.26 Concussion management may also vary by sport contact level. Nevertheless, the findings suggest that PCS is not unique to collision and high contact sports. Furthermore, proper enforcement of rules restricting contact is essential in low-/no-contact sports to ensure the health and safety of their athletes.
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model
ARTICLE IN PRESS
JSAMS-1618; No. of Pages 6
Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
4
Table 1 Characteristics of concussed student-athletes in high school sports.
Sex Male Female Sport Boys’ football Boys’ wrestling Girls’ basketball Boys’ lacrosse Girls’ soccer Boys’ basketball Boys’ soccer Girls’ volleyball Girls’ field hockey Girls’ lacrosse Girls’ softball Boys’ baseball Girls’ track & field Girls’ swimming & diving Boys’ track & field Girls’ gymnastics Boys’ swimming & diving Girls’ tennis Girls’ cross country Girls’ crew Contact level Collision High-contact Low-/non-contact Helmet status Helmeted Un-helmeted Recurrent concussion Yes No Number of symptoms (Mean, SD)
Total
Post-concussion syndrome (PCS)
Non-PCS
986 (73.9) 348 (26.1)
143 (66.5) 72 (33.5)
843 (75.3) 276 (24.7)
680 (51.0) 81 (6.1) 79 (5.9) 79 (5.9) 77 (5.8) 65 (4.9) 57 (4.3) 51 (3.8) 43 (3.2) 35 (2.6) 34 (2.5) 14 (1.0) 12 (0.9) 7 (0.5) 7 (0.5) 5 (0.4) 3 (0.2) 3 (0.2) 1 (0.1) 1 (0.1)
97 (45.1) 16 (7.4) 10 (4.7) 13 (6.0) 13 (6.0) 6 (2.8) 7 (3.3) 10 (4.7) 12 (5.6) 7 (3.3) 9 (4.2) 2 (0.9) 7 (3.3) 2 (0.9) 1 (0.5) 2 (0.9) 1 (0.5) 0 0 0
583 (52.1) 65 (5.8) 69 (6.2) 66 (5.9) 64 (5.7) 59 (5.3) 50 (4.5) 41 (3.7) 31 (2.8) 28 (2.5) 25 (2.2) 12 (1.1) 5 (0.4) 5 (0.4) 6 (0.5) 3 (0.3) 2 (0.2) 3 (0.3) 1 (0.1) 1 (0.1)
761 (57.0) 357 (26.8) 216 (16.2)
113 (52.6) 49 (22.8) 53 (24.7)
648 (57.9) 308 (27.5) 163 (14.6)
893 (66.9) 441 (33.1)
139 (64.7) 76 (35.3)
754 (67.4) 365 (32.6)
118 (8.8) 1216 (91.2) 5.3 (2.9)
14 (6.5) 201 (93.5) 6.7 (3.2)
104 (9.3) 1015 (90.7) 5.0 (2.8)
1334 (100.0)
215 (100.0)
1119 (100.0)
p-Valuea 0.007
n/a
0.001
0.44
0.19
Total
<0.001
a p-Value from tests that examines differences between the PCS and non-PCS groups. For sex, contact level, helmet status, and recurrent concussion, chi-square tests were used. For number of symptoms, an independent samples t-test were used. Test was not run for sport.
Table 2 Multivariable analyses predicting odds of post-concussion syndrome (PCS) in concussed student-athletes in high school sports. Variable
Values
Odds ratio (95%CI) Overall (n = 1334)
Males only (n = 986)
Females only (n = 348)
Sex Contact levela
Male vs. female Collision High-contact Low-/non-contact (1-unit increase) Yes vs. no
0.85 (0.49–1.47) 0.61 (0.33–1.16) 0.51 (0.30–0.85)* 1.00 0.62 (0.33–1.16)
– 0.57 (0.18–1.77) 0.46 (0.14–1.52) 1.00 0.74 (0.45–1.19)
– – 0.48 (0.27–0.87)* 1.00 0.79 (0.37–1.70)
1.38 (0.73–2.61) 3.01 (1.31–6.91)* 2.72 (1.56–4.77)* 1.86 (1.04–3.33)* 1.39 (0.77–2.49) 1.34 (0.77–2.34) 1.44 (0.79–2.65) 4.23 (0.90–19.91) 2.79 (1.62–4.80)* 1.76 (1.00–3.10)* 1.19 (0.69–2.07) 1.32 (0.74–2.35) 1.80 (1.02–3.17)* 2.21 (1.23–3.97)*
1.31 (0.62–2.77) 3.60 (1.40–9.24)* 2.79 (1.43–5.43)* 1.78 (0.89–3.56) 1.36 (0.68–2.75) 1.42 (0.74–2.72) 1.35 (0.64–2.84) 3.34 (0.67–16.57) 3.29 (1.72–6.31)* 1.71 (0.87–3.36) 0.87 (0.44–1.70) 1.17 (0.58–2.36) 2.06 (1.03–4.13)* 1.60 (0.78–3.25)
1.62 (0.46–5.67) 0.99 (0.14–6.90) 2.14 (0.75–6.12) 1.76 (0.61–5.09) 1.33 (0.46–3.85) 0.96 (0.33–2.82) 1.42 (0.48–4.22) – 1.51 (0.54–4.18) 1.74 (0.62–4.90) 1.81 (0.67–4.91) 1.29 (0.45–3.66) 1.42 (0.51–3.96) 3.39 (1.15–10.02)*
Number of symptoms Symptom prevalence Posttraumatic amnesia Retrograde amnesia Difficulty concentrating Disorientation Dizziness Excess drowsiness Excess irritability Headacheb Insomnia Loss of balance Nausea/vomiting Sensitivity to light Sensitivity to noise Visual disturbance
Note: Univariable analyses found no association between PCS and the following variables, which were consequently excluded from this final multivariable model: helmet status, recurrent concussion, and symptom prevalence of excess excitability, loss of consciousness, and tinnitus. a The contact level category of “collision” was not assessed in the model with females only because no females played in collision sports. b Headache was not included in the model with females only because all females reported headache. * Denotes statistical significance (i.e., 95%CI does not include 1.00).
It is important to note that sex-stratified models indicated that this higher prevalence of PCS in low-/non-contact athletes was
attributable to females. Previous research has found that females generally have worse outcomes post-concussion than males.10
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
Also, previous concussion has also been found to be strongly associated with the development of PCS.22 However, our study found neither sex nor recurrent concussion to be associated with PCS. Much discussion has examined both biological and socio-cultural factors associated with sex differences in SRC.19,27 Future research needs to continue refining our methods to pinpoint specific factors that may be potentially attributable to sex differences, while accounting for variations in contact level in the sports played. Our findings parallel a recent study examining high school athletes with multiple concussions within a season28 ; although return to play time was longer with subsequent concussions, symptom resolution time did not differ. Our target population is relatively younger than collegiate or professional athletes, and in comparison to higher-level collegiate and professional athletes, high school athletes have sufficiently less exposure, which implies decreased opportunity for sustaining concussions. Given the lack of on-site management at lower levels of competition (e.g., recreational leagues, middle school sports), is possible that concussion history prior to high school was unknown. Regardless, it is essential for clinicians to not only obtain previous concussion history, but also work with young athletes of both sexes to ensure they are aware of the implications of concussion and the importance of disclosing and seeking appropriate care. Finally, compared to previous research utilizing a similar methodology with the NCAA-ISP,22 we found that the proportion of included athletes with PCS among included samples in the current study was higher [16.1% (215/1334) vs. 7.4% (112/1507)]. A previous study also found that the odds of return to play time ≥30 days post-concussion was higher in high school than college students, but symptom resolution time was not examined.29 One potential reason for this longer symptom resolution time may due to differences in day-to-day activities, changes in biochemistry and ongoing brain development. Whereas collegiate student-athletes have relatively less (albeit more challenging) class time, high school athletes are in class for 6–7 h continuously with minimal break time. The constant mental focus required for several hours may be affected adversely by concussion, whereas a collegiate athlete may be able to recover more quickly with multiple breaks. Furthermore, high school student-athletes may lack the resources college studentathletes receive from their institutions, such as tutoring, academic accommodations, and note-taking services. Given the proportions of concussed student-athletes with PCS, it is essential to identify interventions that not only can assist with return-to-play, but also return-to-learn. Because our study did not examine components related to return to learning, it is important for future research to consider methods that simultaneously examine post-SRC protocols for both return-to-play and return-to-learn. Recommendations exist for graduated return-to-learn protocols30 and should be evaluated for feasibility and effectiveness. Our study has limitations. NATION utilizes a convenience sample of high school sports programs and may not be generalizable to non-participating sports programs or other levels of competition (e.g., club sports, college sports). Data were also collected from a previous time-period that may not represent the current climate of concussion reporting. Although ATs were encouraged ATs to follow the Consensus Statement on Concussion in Sport,13 we could not ensure that ATs utilized this definition for diagnostic purposes. Conflicting PCS definitions impede researchers’ ability to properly estimate the prevalence and incidence of PCS and associated risk factors. In a survey of over 500 sports medicine physicians, the most commonly enlisted PCS criteria was at least one symptom for at least four weeks.8 Although NATION utilizes a 17-symptom checklist, it does not measure the severity of each symptom nor could it account for time of onset or duration. Future research should continue to examine how the type, number, and severity of symptoms are associated with long-term outcomes associated
5
with concussion at the high school (and other) level(s). Such additional information on symptoms will be helpful in sub-classifying PCS to help guide management strategies. As an ambispective study, we were unable to re-contact concussed individuals or participating schools to obtain additional information. Such variables of interest include those from the athlete (e.g., player position), school (e.g., staffing and resource allocations), and state (e.g., state-specific, concussion-related policy and legislation). In particular, NATION did not acquire medical history; thus, we did not have information on biopsychosocial factors such as learning disability, depression/anxiety, psychiatric illness, and family history that have been found to be associated with concussion outcomes.18 Our data were also collected via injury surveillance and includes only the information that team medical staff received during their clinical practice; thus, information on some variables related to athlete injury history may not be comprehensive.
5. Conclusion Our study examined factors associated with PCS at the high school level. Factors in a multivariable model predicting the odds of PCS included contact level as well as cognitive-, somatic-, and sleeprelated symptoms. Identify factors associated with PCS may assist sports medicine practitioners in better identifying those concussed athletes that may be at greater risk of having longer symptom resolution time. Further study is needed to better predict PCS and to develop and evaluate treatment strategies for at-risk individuals.
Practical implications • The specific symptoms endorsed by high school athletes following concussion may be more important than total symptoms in determining risk of PCS; athletes who endorse symptoms of memory, concentration, sleep, balance, noise, and visual difficulty may have a higher chance of developing PCS than those who do not. • Contact level of sport was not relevant in predicting the odds of PCS; PCS is seen in all athletes, both high and low contact sports. • Our findings, coupled with previous research, may suggest that high school student-athlete populations may have a greater prevalence of PCS than college student-athlete populations; thus, special attention should be given to the rigorous academic schedule and resources available to high school student-athletes in their return to school after a concussion.
Acknowledgments Funding for this study was provided by the National Athletic Trainers’ Association Research and Education Foundation and BioCrossroads in partnership with the Central Indiana Corporate Partnership Foundation. The organizations that funded this study had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the study sponsors. We thank the many athletic trainers who have volunteered their time and efforts to submit data to NATION. Their efforts are greatly appreciated and have had a tremendously positive effect on the safety of high school student-athletes.
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025
G Model JSAMS-1618; No. of Pages 6
ARTICLE IN PRESS Z.Y. Kerr et al. / Journal of Science and Medicine in Sport xxx (2017) xxx–xxx
6
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jsams.2017.08. 025. References 1. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of traumatic brain injury: a brief overview. J Head Trauma Rehabil 2006; 21(5):375–378. 2. National Federation of High Schools. Participation Statistics, 2016. Accessed 19 December 2016 http://www.nfhs.org/ParticipationStatics/ParticipationStatics. aspx/. 3. Lincoln AE, Caswell SV, Almquist JL et al. Trends in concussion incidence in high school sports: a prospective 11-year study. Am J Sports Med 2011; 39(5):958–963. 4. McCrea M, Guskiewicz K, Randolph C et al. Incidence, clinical course, and predictors of prolonged recovery time following sport-related concussion in high school and college athletes. J Int Neuropsychol Soc 2013; 19(1):22–33. 5. Ruff RM, Camenzuli L, Mueller J. Miserable minority: emotional risk factors that influence the outcome of a mild traumatic brain injury. Brain Inj 1996; 10(8):551–565. 6. Centers for Medicare and Medicaid Services. ICD-9-CM Diagnosis and Procedure Codes: Abbreviated and Full Code Titles. https://www.cms.gov/medicare/coding/ ICD9providerdiagnosticcodes/codes.html. Accessed 16 January 2017. 7. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-4, Washington, DC, American Psychiatric Publishing, 2000. 8. Rose SC, Fischer AN, Heyer GL. How long is too long? The lack of consensus regarding the post-concussion syndrome diagnosis. Brain Inj 2015; 29(7–8):798–803. 9. Luoto TM, Iverson GL, Losoi H et al. Clinical correlates of retrograde amnesia in mild traumatic brain injury. Brain Inj 2015; 29(5):565–572. 10. McNally KA, Bangert B, Dietrich A et al. Injury versus noninjury factors as predictors of postconcussive symptoms following mild traumatic brain injury in children. Neuropsychology 2013; 27(1):1–12. 11. Dompier TP, Marshall SW, Kerr ZY et al. The national athletic treatment, injury and outcomes network (NATION): methods of the surveillance program, 2011–2012 through 2013–2014. J Athl Train 2015; 50(8):862–869. 12. Wasserman EB, Kerr ZY, Zuckerman SL et al. Epidemiology of sports-related concussions in National Collegiate Athletic Association athletes from 2009–2010 to 2013–2014: symptom prevalence, symptom resolution time, and return-toplay time. Am J Sports Med 2016; 44(1):226–233. 13. McCrory P, Meeuwisse WH, Aubry M et al. Consensus statement on concussion in sport: the 4th International Conference on Concussion in Sport held in Zurich, November 2012. Br J Sports Med 2013; 47(5):250–258. 14. Lombardi NJ, Tucker B, Freedman KB et al. Accuracy of athletic trainer and physician diagnoses in sports medicine. Orthopedics 2016; 39(5):e944–e949.
15. McCauley SR, Wilde EA, Miller ER et al. Preinjury resilience and mood as predictors of early outcome following mild traumatic brain injury. J Neurotrauma 2013; 30(8):642–652. 16. Meehan 3rd WP, Mannix RC, Stracciolini A et al. Symptom severity predicts prolonged recovery after sport-related concussion, but age and amnesia do not. J Pediatr 2013; 163(3):721–725. 17. Tator CH, Davis H. The postconcussion syndrome in sports and recreation: clinical features and demography in 138 athletes. Neurosurgery 2014; 75(Suppl. 4):S106–S112. 18. Morgan CD, Zuckerman SL, Lee YM et al. Predictors of postconcussion syndrome after sports-related concussion in young athletes: a matched case-control study. J Neurosurg Pediatr 2015; 15(6):589–598. 19. Zuckerman SL, Kerr ZY, Yengo-Kahn A et al. Epidemiology of sports-related concussion in NCAA athletes from 2009–2010 to 2013–2014: incidence, recurrence, and mechanisms. Am J Sports Med 2015; 43(11):2654–2662. 20. Rice SG, American Academy of Pediatrics Council on Sports M, Fitness. Medical conditions affecting sports participation. Pediatrics 2008; 121(4):841–848. 21. Zuckerman SL, Lee YM, Odom MJ et al. Sports-related concussion in helmeted vs. unhelmeted athletes: who fares worse? Int J Sports Med 2015; 36(5):419–425. 22. Zuckerman SL, Yengo-Kahn AM, Buckley TA et al. Predictors of postconcussion syndrome in collegiate student-athletes. Neurosurg Focus 2016; 40(4):E13. 23. Meehan 3rd WP, Mannix R, Monuteaux MC et al. Early symptom burden predicts recovery after sport-related concussion. Neurology 2014; 83(24):2204–2210. 24. Silverberg ND, Gardner AJ, Brubacher JR et al. Systematic review of multivariable prognostic models for mild traumatic brain injury. J Neurotrauma 2015; 32(8):517–526. 25. Zuckerman SL, Totten DJ, Rubel KE et al. Mechanisms of injury as a diagnostic predictor of sport-related concussion severity in football, basketball, and soccer: results from a regional concussion registry. Neurosurgery 2016; 63:102–112. 26. Sallis RE, Jones K. Prevalence of headaches in football players. Med Sci Sport Exerc 2000; 32(11):1820–1824. 27. Broshek DK, Kaushik T, Freeman JR et al. Sex differences in outcome following sports-related concussion. J Neurosurg 2005; 102(5):856–863. 28. Currie DW, Comstock RD, Fields SK et al. A paired comparison of initial and recurrent concussions sustained by US high school athletes within a single athletic season. J Head Trauma Rehabil 2016. http://dx.doi.org/10.1097/htr.0000000000000240. Published online April 26. 29. Kerr ZY, Zuckerman SL, Wasserman EB et al. Concussion symptoms and return to play time in youth, high school, and college american football athletes. JAMA Pediatr 2016; 170(7):647–653. 30. Collins MW, Kontos AP, Okonkwo DO et al. Statements of agreement from the targeted evaluation and active management (TEAM) approaches to treating concussion meeting held in Pittsburgh, October 15–16, 2015. Neurosurgery 2016; 79(6):912–929.
Please cite this article in press as: Kerr ZY, et al. Factors associated with post-concussion syndrome in high school student-athletes. J Sci Med Sport (2017), http://dx.doi.org/10.1016/j.jsams.2017.08.025