Athletic Training and Sports Health Care

Original Research 

Concussion Reporting Intentions Among Division I Football Players: Consideration of Individual-Level and Team-Level Predictors

Monica R. Lininger, PhD, LAT, ATC; Heidi A. Wayment, PhD; Ann Hergatt Huffman, PhD; Debbie I. Craig, PhD, LAT, ATC

Abstract

Purpose:

To investigate the individual- and team-level factors associated with concussion reporting intentions using a hierarchical linear model.

Methods:

In this multi-site cross-sectional study, 248 athletes from three different National Collegiate Athletic Association Division I football programs completed a post-season questionnaire that assessed individual-level predictors of concussion reporting intentions: self-efficacy, norms about concussion reporting, and negative reporting attitudes. A hierarchical linear model analysis was used to examine the clustered data.

Results:

Nineteen percent of the variability in concussion reporting intentions was attributable to football program membership, leaving 81% of the variability in concussion reporting intentions predicted by individual-level variables. When adjusting for age and controlling for team-level influences, reporting attitudes and self-efficacy of reporting were significant predictors of concussion reporting intentions.

Conclusions:

Controlling for team-level factors, concussion reporting intentions were most strongly and significantly predicted by negative reporting attitudes and self-efficacy, across all three football programs. Development of methods to improve athlete self-efficacy and attitudes about reporting concussion symptoms should continue.

[Athletic Training & Sports Health Care. 201X;X(X):XX–XX.]

Abstract

Purpose:

To investigate the individual- and team-level factors associated with concussion reporting intentions using a hierarchical linear model.

Methods:

In this multi-site cross-sectional study, 248 athletes from three different National Collegiate Athletic Association Division I football programs completed a post-season questionnaire that assessed individual-level predictors of concussion reporting intentions: self-efficacy, norms about concussion reporting, and negative reporting attitudes. A hierarchical linear model analysis was used to examine the clustered data.

Results:

Nineteen percent of the variability in concussion reporting intentions was attributable to football program membership, leaving 81% of the variability in concussion reporting intentions predicted by individual-level variables. When adjusting for age and controlling for team-level influences, reporting attitudes and self-efficacy of reporting were significant predictors of concussion reporting intentions.

Conclusions:

Controlling for team-level factors, concussion reporting intentions were most strongly and significantly predicted by negative reporting attitudes and self-efficacy, across all three football programs. Development of methods to improve athlete self-efficacy and attitudes about reporting concussion symptoms should continue.

[Athletic Training & Sports Health Care. 201X;X(X):XX–XX.]

Research on sport-related concussions has grown exponentially over the past several years in an effort to better understand the etiology,1,2 pathology,3,4 methods to improve diagnosis,5–7 and rehabilitation outcomes8–11 for these injured student-athletes. These investigations are needed due to the high prevalence of sport-related concussions in athletics.12,13 Yet, there continues to be less known about the exact factors, both interpersonal and intrapersonal, that encourage a student-athlete to disclose a possible sport-related concussion. It has been noted that at least half of sport-related concussions are not reported by student-athletes.2,14,15 Many theories have been presented as to why student-athletes may not report potential sport-related concussions. These include the student-athlete not thinking the injury was serious enough,16–18 not knowing that it was indeed an sport-related concussion,19–21 not wanting to be removed from participation,16,19–22 and being afraid of letting the team down.17–21 Due to the lack of symptom reporting, valiant efforts have been made to increase the knowledge of sport-related concussions.23–29

One of the most used models in predicting concussion reporting behavior is the Theory of Planned Behavior.30 The Theory of Planned Behavior suggests that behaviors can be predicted from attitudes regarding the behavior, subjective norms, and perceived behavioral control.31,32 Another important predictor of behavior is the intention to perform that specific behavior.33 Concussion reporting intentions have been shown to be related to actual in-season concussion reporting behavior.34 Research has consistently shown that the Theory of Planned Behavior provides some explanatory power related to concussion reporting intentions.30,35 Kroshus et al.30 found that the Theory of Planned Behavior model provided significant prediction abilities of concussion reporting intentions, but only 25% of the variance in behavior and intention was explained in this sample. Additionally, attitudes, subjective norms, and perceived behavioral control have been associated with intentions to report in high school student-athletes.35 Yet one limitation of concussion reporting intention research using the Theory of Planned Behavior model is that the key variables within the model are individual-level factors that do not take into consideration team-level factors. We argue that behavior is a function of both individual- and team-level factors. This means individual-level factors (eg, positive or negative reporting attitudes) might initially be present with specific members of a team, but these factors will spread to other team members, making these factors distinct to the team (team-level negative reporting attitudes). These team-specific positive or negative reporting attitudes will then affect other individuals within the team, influencing individuals' reporting attitudes within the team. Research on teams has consistently shown that groups have a unique influence on individual behavior.36–38 Although this team-level effect is not an objective artifact, nor observable, it acts as a contextual factor that can be measured by assessing how much team members share in agreement related to the variable.39

To date, no research has been published that examines predictors of concussion reporting intentions in a model that allows assessment of the impact that individual- and team-level variables may have simultaneously. Therefore, the purpose of this study was to assess the factors that encourage concussion reporting intention, at both the individual and team levels. Specifically, we assessed the ability of individual- and team-level factors to predict concussion reporting intentions. To test our model, we assessed the four key variables in the Theory of Planned Behavior model (individual-level factors: negative reporting attitudes, social norms, self-efficacy, and concussion reporting intentions) and three team-level factors (team identification, safety climate, and player position) in a sample of collegiate football players. We hypothesized that the set of individual- and team-level factors, together, would be significantly associated with concussion reporting intentions.

Methods

Participants

Two hundred forty-eight student-athletes (age: 19.8 ± 1.5 years) on National Collegiate Athletic Association (NCAA) Division I football rosters for the 2017 season were invited to participate in this study (Table 1). Nearly 46% of the players self-identified as white, 50% were freshmen or sophomores, and they were equally divided as offensive and defensive players. All research procedures and instruments were approved by the institutional review boards from all institutions.

Demographic Characteristics and Response Rates of Student–Athletes in Sample

Table 1:

Demographic Characteristics and Response Rates of Student–Athletes in Sample

Instrumentation

Student-athletes completed a paper/pencil questionnaire at the conclusion of the 2017 football season. The questionnaire included demographic questions such as age, race/ethnicity, year in college, and six previously validated constructs (Table 2). Four scales23,30 were used for concussion reporting measures and two scales40,41 were selected as team-level assessments, as will later be discussed in detail. We used similar questions in the Fall of 2016 for data collection with 100 football student-athletes from Division I institutions.42,43

Previously Validated Scales With Individual Items Including Measure of Internal Consistency Used in Questionnaire

Table 2:

Previously Validated Scales With Individual Items Including Measure of Internal Consistency Used in Questionnaire

Individual-Level Variables

For this study, we used the four most studied predictors of concussion reporting intentions: negative reporting attitudes, social norms, self-efficacy, and reporting intentions.30 Negative reporting attitudes assessed a student-athlete's perceptions and attitudes about the consequences of reporting a potential sport-related concussion. This scale comprised 10 items, eight (Table 2) of which were identical to those found in the Kroshus et al.30 study on concussion reporting attitudes and behaviors in a sample of late adolescent ice hockey players. For the current study, only four items (Table 2) were included: “If I report what I suspect might be concussion, my teammates will. . . ‘think less of me,’ ‘respect me,’ ‘think I made the right decision,’ and ‘not support me’.” The construct of social norms about concussion reporting behavior23 measures the extent to which an athlete believes a variety of referents think about whether or not the student-athlete should report potential concussions. This measure comprised four items (Table 2). The third construct, self-efficacy of concussion reporting,30 measured the extent to which a student-athlete felt confident in his ability to report symptoms of a concussion under different conditions. This measure comprised five items (Table 2). The final construct, concussion reporting intentions, came from work by Register-Mihalik et al.23 and consisted of three items (Table 2): “I intend to report,” “I plan to report,” and “I will make an effort to report.”

Team-Level Variables

To assess team-level variability, two constructs (team identification and safety climate) were used. Fransen et al.'s40 measure of team identification was used to assess the extent to which a student-athlete's personal identity is tied to his team. The scale comprised three items (Table 2): “I feel very connected with this team,” “Being a member of the team is very important to me,” and “I am very happy that I belong to this team.” Next, the safety climate scale was used to determine a student-athlete's attitudes and beliefs about his head coach's role in promoting player safety in football. This eight-item measure (Table 2) was adapted from Beus et al.'s41 measure of organizational safety climate strength that sought to assess these same attitudes and beliefs regarding a supervisor's responsibility in promoting employee safety. Our adaptation replaced instances of the word “supervisor” with “coach,” “workgroup” with “team,” and “employees” with “players.” Sample items included: “My head coach is committed to improving player safety” and “My head coach encourage players to become involved in player safety.” Beus et al.'s41 measure was conceptually adapted from Zohar and Luria's44 multi-level model and measures of safety climate.

Procedures

We recruited three NCAA Division I football programs (one Football Bowl Subdivision program and two Football Championship Subdivision programs) from three different NCAA conferences. Prior to data collection, a member from the research team explained the purpose of the study, answered all questions, and then obtained informed consent from each participant. Across the three institutions, 248 student-athletes completed the post-season hardcopy questionnaire (Program 1 [Football Championship Subdivision] = 59, Program 2 [Football Bowl Subdivision] = 102, and Program 3 [Football Championship Subdivision] = 87) that took, on average, 15 minutes to complete. Participants received a $10 gift card once the questionnaire was completed. This sample was part of a larger investigation examining the culture of concussion reporting behaviors (see Craig et al.43 for more detail about methods and procedures).

Statistical Analyses

Frequencies and percentages were calculated for categorical data, whereas means and standard deviations were obtained for continuous variables. Preliminary analyses were conducted in SPSS software (version 24.0; IBM SPSS, Inc., Chicago, IL) to evaluate the normalcy, missingness, and linearity of the data. If a violation in normality was seen, a natural log transformation was used. For both team identification and the safety climate scale, a median split was applied to create a high/low level of team identification or safety due to violations in normality. There was approximately 12% missingness for race and year in school, but all other missingness was minimal (2%) and the assumption of linearity was met. In addition to the validated scales,23,30,40,41 football position was constructed to help assess team-level influence. A binary variable of position (offensive or defensive player) was used in the statistical analysis.

Pairwise comparisons were conducted among the study variables. Pearson's correlation coefficients were used to describe the relationship between two continuous variables (ie, reporting attitudes with norms of concussion reporting), whereas point-biserial coefficients described the relationships between a continuous variable and a nominal variable (ie, negative reporting attitudes with bowl game appearance), and phi coefficients were used for two nominal variables (ie, offensive/defensive player).

We used the hierarchical linear model analysis to examine the clustered data (student-athletes nested within football programs) of concussion reporting intentions. The hierarchical linear model allows the data to be modelled to determine how concussion reporting intentions differed between student-athletes while accounting for the clustered nature of the data. That is, student-athletes' attitudes and perceptions (eg, level 1 study variables of concussion reporting intentions, negative reporting attitudes, norms of concussion reporting, and self-efficacy of reporting) are “nested” within the team (eg, level 2 study variables of safety climate, team identification, and position [offense = 1, defense = 0]). The first model (Model 1: Unconditional Model) estimates variation attributable to athletes, individually. The second model (Model 2: Conditional Model) estimates variation attributable to team-level differences. Furthermore, the final model (Model 3: Adjusted Conditional Model) was adjusted for age and parent's median income as a marker of socioeconomic status. Also, to improve interpretation of the findings, the continuous predictors were grand mean centered, therefore showing the average response across groups. Hierarchical Linear Model software (version 7; Scientific Software International, Inc., Skokie, IL)45 was used for the multi-level modeling with alpha set to 0.05.

Results

Skewness and kurtosis estimates were in the range of normal (±2) along with non-significant Kolmogorov–Smirnov findings (P > .05), except for norms of concussion reporting (skewness: −1.32, kurtosis: 3.95, Kolmogorov– Smirnov: P < .0001), team identification, and the safety climate scale as seen in Table 3. A natural log was applied, transforming the data to a more normal distribution (skewness: 0.20, kurtosis: −0.76, Kolmogorov–Smirnov: P > .05). There were 14 statistically significant relationships between variables, nine positive and five inverse relationships (Table 4). Safety climate, age, and self-efficacy of reporting had the most number of significant relationships with other variables.

Descriptive Statistics and Normality Estimates of Concussion Reporting Behavior and Football Program Variables

Table 3:

Descriptive Statistics and Normality Estimates of Concussion Reporting Behavior and Football Program Variables

Pearson's Pairwise Correlations of Individual- and Team-Level Variables Used in Statistical Analyses

Table 4:

Pearson's Pairwise Correlations of Individual- and Team-Level Variables Used in Statistical Analyses

Equation 1 shows the Model 3: Adjusted Conditional Model:

where CRIij is the concussion reporting intentions for athlete j, γ00 is the level 1 intercept, γ01 adjusting for age, γ02 adjusting for socioeconomic status, γ03 is the effect of being either a defensive or offensive player, γ04 effect of team identification, γ05 effect of safety culture, γ10 is self-efficacy for each athlete, γ20 is the effect of reporting attitudes for each athlete, γ30 is the effect of norms for concussion reporting for each athlete, and u0j and rij are the error terms associated for each athlete measured in the study.

Using a two-level hierarchical linear model with student-athlete (N = 248) nested within team, 19% of the variability (τ00 / (τ00 + σ2; 0.31 / (0.31 + 1.32) in concussion reporting intentions was attributable to team-level differences, leaving 81% of the variability in the study variables attributable to individual-level differences. These results from Model 1: Unconditional Model only included self-efficacy of reporting, reporting attitudes, and norms about concussion reporting as the predictors of concussion reporting intentions. Moving to Model 2: Conditional Model, which now includes predictors at the team level (eg, offensive/defensive player, team identification, and safety climate scale), 17% of the variability is between schools. In this model, self-efficacy, reporting attitudes, and norms for concussion reporting were all significant predictors of intentions as seen in Table 5. However, when adjusting for age in Model 3: Adjusted Conditional Model, norms about concussions was no longer significant. Self-efficacy and negative reporting attitudes remained significant predictors in the adjusted conditional model. Model 3: Adjusted Conditional Model can be referenced in Equation 1.

Results of Multilevel Modeling for Concussion Reporting Behaviors

Table 5:

Results of Multilevel Modeling for Concussion Reporting Behaviors

Discussion

This study investigated both the individual- and team-level variables that were associated with concussion reporting intentions. We hypothesized that the set of individual- and team-level factors, together, would be significantly associated with concussion reporting intentions. We used a hierarchical linear model approach, a statistical procedure that has not been used in the literature on predictors of concussion reporting intentions, to assess both individual- and team-level factors. Compared to the traditional methods, such as the general linear model that requires an assumption of independence, the hierarchical linear model accounts for the naturally occurring clusters (student-athletes within team) in which the observations are not independent. Student-athletes from the same team are more likely to be similar than those from other schools. Our findings showed that across three Division 1 football programs (N = 248 student-athletes), 19% of the systematic variation in players' concussion reporting intentions could be attributed to team-level variables. Furthermore, when controlling for these team-level variables, and the influence of age and socioeconomic status, 81% of the variability in measured variables was due to individual-level differences. In other words, more than three-quarters of the variability in concussion reporting intentions comes from within the person as opposed to factors related to the team as a whole. Our results strengthen prior research showing the use of the Theory of Planned Behavior model as predictors of concussion reporting intentions. Additionally, to our knowledge, this is the first concussion reporting intentions study to assess the role that group level constructs have in the Theory of Planned Behavior model.

Taken together, our results suggest that the decision to report a possible sport-related concussion may be more likely influenced by negative reporting attitudes and self-efficacy, supporting earlier studies that have also shown that these two variables are important for concussion prevention efforts.22,34 For example, players are told that they should report symptoms during a game or immediately after a game. However, we are learning that players are more likely to report other orthopedic injuries when compared to a sport-related concussion.46 It is also reported that the sooner players report symptoms, the sooner they may recover and return to play.14 Although the empirical evidence is growing, the athletes themselves are not readily aware of it. Perhaps if players have practical and focused training about how and when to report symptoms, it would be easier for them to report. Methods to increase self-efficacy are well known,47,48 and include the cultivation of mastery experiences, role models, verbal persuasion, and motivational arousal. It should be noted that in our sample, norms of concussion reporting were not significantly associated with concussion reporting intentions, when adjusting for age and socioeconomic status, which contradicts previous work in NCAA Division I ice hockey student-athletes, where the reporting norm variable was focused on safety (not general reporting norms).49

In applying the findings of our study, clinicians and administrators may want to address encouraging individual athletes, coaches, and others associated with the team to shift away from viewing concussion reporting as a negative act and toward this being supported by all as a positive act. This simple shift in framing takes into consideration the individual athlete's potentially negative reporting attitudes and lack of self-efficacy, which accounts for 81% of the variability in our study. Only when these individual-level changes occur can a similar shift in team-level variables occur. How to best influence these specific individual variables toward positive changes has yet to be empirically studied using the hierarchical linear model. Previous research investigating why individual athletes choose to hide or not disclose a possible concussion demonstrates that not wanting to let their coach or teammates down is a primary cause of non-disclosure.16 Thus, changing the perception that the act of reporting a potential concussion is a negative act could alter concussion reporting intentions across many individuals on the team. We suggest that this singular effort to change individual-level reporting attitudes may influence the desired positive change in concussion reporting intentions.

Limitations and Future Research

Although this work is novel, it has limitations. The major limitation is that regardless of our ability to find variables that are correlated with concussion reporting intentions, we have no indication of whether intentions are truly related to actual reporting behavior. Another limitation is the lack of a comparison school for the Football Bowl Subdivision institution. Our current analysis used players from two Football Championship Subdivision and one Football Bowl Subdivision institution. The hierarchical linear model analysis would be improved if we had additional players from another Football Bowl Subdivision school. The culture at Football Bowl Subdivision schools may differ from that at a Football Championship Subdivision school. Another limitation impacting generalizability is the use of only three football programs. The findings of this study can also not be generalized to other levels of athletics, other sports, or female student-athletes. Future research can improve upon the current work by including that comparison institution and extending the research in a longitudinal study to see if factors differ across seasons. Building on this study, researchers should investigate the efficacy of interventions designed to improve negative concussion-related attitudes and self-efficacy. If these predictors are directly associated with concussion reporting intentions, interventions aimed at modifying an athlete's perception of performing a behavior along with his or her confidence to perform the behavior are vital to changing concussion reporting behavior.

Implications for Clinical Practice

The results from this study suggest that by using a hierarchical linear model statistical analysis that controls for both individual- and team-level sources of variation, concussion reporting intentions are most significantly affected by individual student-athlete's negative reporting attitudes and self-efficacy. Because student-athletes' reporting attitudes are less negative, and self-efficacy increases, so does their concussion reporting intentions. Instead of focusing interventions on additional education, this study postulates that future educational and behavioral interventions should focus on how best to improve negative concussion reporting attitudes and perceptions of self-efficacy to be able to report in specific competitive situations. Our measure of negative attitudes toward reporting reflected athletes' fears that they would not be respected by their teammates for reporting.

Our results imply that intervention efforts must more clearly aim to influence the root of why negative attitudes toward reporting persist. Although concussion education has been mandated at various sport levels by sports oversight committees and in many state legislative efforts, what is included in the educational sessions is less prescribed.20,50 This study supports that efforts beyond simple concussion knowledge (ie, relaying signs and symptoms of sport-related concussions) that go further to address changing negative reporting attitudes and improving self-efficacy would have a greater impact on concussion reporting intentions. Current research evidence suggests addressing negative reporting attitudes through discussion of perceived social norms may benefit concussion reporting intentions.30,35,51 Finally, interventions that are not presented within traditional “concussion education sessions,” but rather as team or small group discussions where athletes have the ability to express their concerns, may have a greater impact on decreasing negative attitudes at both the individual and team levels. This approach has yet to be empirically studied to our knowledge.

References

  1. Meier T, Bellgowan P, Mayer. A Longitudinal assessment of local and global functional connectivity following sports-related concussion. Brain Imaging Behav. 2017;11(1):129–140. doi:10.1007/s11682-016-9520-y [CrossRef]
  2. Zuckerman SL, Kerr ZY, Yengo-Kahn A, Wasserman E, Covassin T, Solomon G. 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. doi:10.1177/0363546515599634 [CrossRef]26330572
  3. Churchill N, Hutchison M, Richards D, Leung G, Graham S, Schweizer TA. Brain structure and function associated with a history of sport concussion: a multi-modal magnetic resonance imaging study. J Neurotrauma. 2017;34(4):765–771. doi:10.1089/neu.2016.4531 [CrossRef]
  4. Cripps A, Livingston SC. Differentiating concussion from intracranial pathology in athletes. J Sport Rehabil. 2017;26(1):101–108. doi:10.1123/jsr.2015-0043 [CrossRef]
  5. McCrory P, Meeuwisse WH, Echemendia RJ, Iverson GL, Dvorák J, Kutcher JS. What is the lowest threshold to make a diagnosis of concussion?Br J Sports Med. 2013;47(5):268–271. doi:10.1136/bjsports-2013-092247 [CrossRef]23479483
  6. Meeuwisse WH, Schneider KJ, Dvorak J, et al. The Berlin 2016 process: a summary of methodology for the 5th International Consensus Conference on Concussion in Sport. Br J Sports Med. 2017;51:873–87628446449
  7. Broglio SP, Cantu RC, Gioia GA, et al. National Athletic Trainers' Association position statement: management of sport concussion. J Athl Train. 2014;49(2):245–265. doi:10.4085/1062-6050-49.1.07 [CrossRef]24601910
  8. McCrea M. Standardized mental status testing on the sideline after sport-related concussion. J Athl Train. 2001;36(3):274–279.
  9. McCrea M, Barr WB, Guskiewicz K, et al. Standard regression-based methods for measuring recovery after sport-related concussion. J Int Neuropsychol Soc. 2005;11(1):58–69. doi:10.1017/S1355617705050083 [CrossRef]15686609
  10. Lovell MR, Collins MW, Iverson GL, Johnston KM, Bradley JP. Grade 1 or “ding” concussions in high school athletes. Am J Sports Med. 2004;32(1):47–54. doi:10.1177/0363546503260723 [CrossRef]14754723
  11. Valovich McLeod TC, Lewis JH, Whelihan K, Welch Bacon CE. Rest and return to activity after sport-related concussion: A systematic review of the literature. J Athl Train. 2017;52(3):262–287. doi:10.4085/1052-6050-51.6.06 [CrossRef]
  12. Hootman JM, Dick R, Agel J. Epidemiology of collegiate injuries for 15 sports: summary and recommendations for injury prevention initiatives. J Athl Train. 2007;42(2):311–319.17710181
  13. Kilcoyne KG, Dickens JF, Svoboda SJ, et al. Reported concussion rates for three Division I football programs: an evaluation of the new NCAA concussion policy. Sports Health. 2014;6(5):402–405. doi:10.1177/1941738113491545 [CrossRef]25177415
  14. Asken BM, McCrea MA, Clugston JR, Snyder AR, Houck ZM, Bauer RM. “Playing through it”: delayed reporting and removal from athletic activity after concussion predicts prolonged recovery. J Athl Train. 2016;51(4):329–335. doi:10.4085/1062-6050-51.5.02 [CrossRef]27111584
  15. Llewellyn T, Burdette GT, Joyner AB, Buckley TA. Concussion reporting rates at the conclusion of an intercollegiate athletic career. Clin J Sport Med. 2014;24(1):76–79. doi:10.1097/01.jsm.0000432853.77520.3d [CrossRef]
  16. McCreaHammeke T, Olsen G, Leo P, Guskiewicz K. Unreported concussion in high school football players: implications for prevention. Clin J Sport Med. 2004;14(1):13–17. doi:10.1097/00042752-200401000-00003 [CrossRef]14712161
  17. Delaney JS, Lamfookon C, Bloom GA, Al-Kashmiri A, Correa JA. Why university athletes choose not to reveal their concussion symptoms during a practice or game. Clin J Sport Med. 2015;25(2):113–125. doi:10.1097/JSM.0000000000000112 [CrossRef]
  18. Lininger MR, Wayment HA, Huffman A, Craig DI, Irving LH. An exploratory study on concussion-reporting behaviors from collegiate student athletes' perspectives. Athl Train Sports Health Care. 2017;9(2):71–80. doi:10.3928/19425864-20161116-01 [CrossRef]
  19. Kerr ZY, Register-Mihalik JK, Kroshus E, Baugh CM, Marshall SW. Motivations associated with nondisclosure of self-reported concussions in former collegiate athletes. Am J Sports Med. 2016;44(1):220–225. doi:10.1177/0363546515612082 [CrossRef]
  20. LaRoche AA, Nelson LD, Connelly PK, Walter KD, McCrea MA. Sport-related concussion reporting and state legislative effects. Clin J Sport Med. 2016;26(1):33–39. doi:10.1097/JSM.0000000000000192 [CrossRef]
  21. Broglio SP, Vagnozzi R, Sabin M, Signoretti S, Tavazzi B, Lazzarino G. Concussion occurrence and knowledge in Italian football (soccer). J Sports Sci Med. 2010;9(3):418–430.24149636
  22. Chrisman SP, Quitiquit C, Rivara FP. Qualitative study of barriers to concussive symptom reporting in high school athletics. J Adolesc Health. 2013;52(3):330–335. doi:10.1016/j.jadohealth.2012.10.271 [CrossRef]23427783
  23. Register-Mihalik JK, Guskiewicz KM, McLeod TCV, Linnan LA, Mueller FO, Marshall SW. Knowledge, attitude, and concussion-reporting behaviors among high school athletes: a preliminary study. J Athl Train. 2013;48(5):645–653. doi:10.4085/1062-6050-48.3.20 [CrossRef]23848520
  24. Anderson BL, Gittelman MA, Mann JK, Cyriac RL, Pomerantz WJ. High school football players' knowledge and attitudes about concussions. Clin J Sport Med. 2016;26(3):206–209. doi:10.1097/JSM.0000000000000214 [CrossRef]
  25. Bramley H, Patrick K, Lehman E, Silvis M. High school soccer players with concussion education are more likely to notify their coach of a suspected concussion. Clin Pediatr (Phila). 2012;51(4):332–336. doi:10.1177/0009922811425233 [CrossRef]
  26. Kroshus E, Daneshvar DH, Baugh CM, Nowinski CJ, Cantu RC. NCAA concussion education in ice hockey: an ineffective mandate. Br J Sports Med. 2014;48(2):135–140 6p. doi:10.1136/bjsports-2013-092498 [CrossRef]
  27. Kroshus E, Garnett BR, Baugh CM, Calzo JP. Social norms theory and concussion education. Health Educ Res. 2015;30(6):1004–1013. doi:26471918
  28. Kurowski BG, Pomerantz WJ, Schaiper C, Ho M, Gittelman MA. Impact of preseason concussion education on knowledge, attitudes, and behaviors of high school athletes. J Trauma Acute Care Surg. 2015:S21–8. doi:10.1097/TA.0000000000000675 [CrossRef]26308118
  29. Miyashita T, Timpson W, Frye M, Gloeckner G. The impact of an educational intervention of college athletes' knowledge of concussions. Clin J Sport Med. 2013;23(5):349–353. doi:10.1097/JSM.0b013e318289c321 [CrossRef]23528843
  30. Kroshus E, Baugh CM, Daneshvar DH, Viswanath K. Understanding concussion reporting using a model based on the theory of planned behavior. J Adolesc Health. 2014;54(3):269–274.e2. doi:10.1016/j.jadohealth.2013.11.011 [CrossRef]24560034
  31. Ajzen I, Madden TJ. Prediction of goal-directed behavior: attitudes, intentions, and perceived behavioral control. J Exp Soc Psychol. 1986;22(5):453–474. doi:10.1016/0022-1031(86)90045-4 [CrossRef]
  32. Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Process. 1991;50(2):179–211. doi:10.1016/0749-5978(91)90020-T [CrossRef]
  33. Glanz K, Rimer B, Viswanath K. Health Behavior and Health Education: Theory, Research, and Practice. San Francisco: Jossey-Bass; 2008.
  34. Kroshus E, Baugh CM, Daneshvar DH, Nowinski CJ, Cantu RC. Concussion reporting intention: a valuable metric for predicting reporting behavior and evaluating concussion education. Clin J Sport Med. 2015;25:243–247. doi:10.1097/JSM.0000000000000137 [CrossRef]
  35. Register-Mihalik JK, Linnan LA, Marshall SW, McLeod TCV, Mueller FO, Guskiewicz KM. Using theory to understand high school aged athletes' intentions to report sport-related concussion: implications for concussion education initiatives. Brain Inj. 2013;27(7/8):878–886. doi:10.3109/02699052.2013.775508 [CrossRef]23789865
  36. Nixon HL II, . Sport and Social Organization. Indianapolis, IN: Bobbs-Merrill; 1976.
  37. Cuellar AE, Cheema J. As roughly 700,000 prisoners are released annually, about half will gain health coverage and care under federal laws. Health Aff (Millwood). 2012;31(5):931–938. doi:10.1377/hlthaff.2011.0501 [CrossRef]
  38. Munroe K, Estabrooks P, Dennis P, Carron A. A phenomenological analysis of group norms in sport teams. Sport Psychol. 1999;13(2):171. doi:10.1123/tsp.13.2.171 [CrossRef]
  39. West BJ, Patera JL, Carsten MK. Team level positivity: investigating positive psychological capacities and team level outcomes. J Organ Behav. 2009;30(2):249–267. doi:10.1002/job.593 [CrossRef]
  40. Fransen K, Haslam SA, Steffens NK, Vanbeselaere N, De Cuyper B, Boen F. Believing in “us”: exploring leaders' capacity to enhance team confidence and performance by building a sense of shared social identity. J Exp Psychol Appl. 2015;21(1):89–100. doi:10.1037/xap0000033 [CrossRef]
  41. Beus JM, Bergman ME, Payne SC. The influence of organizational tenure on safety climate strength: a first look. Accid Anal Prev. 2010;42(5):1431–1437. doi:10.1016/j.aap.2009.06.002 [CrossRef]20538098
  42. Wayment H, Huffman A, Lane T, Lininger M. Relationship of athletic and academic identity to concussion reporting intentions. Musculoskelet Sci Pract. 2019;42:186–192. doi:10.1016/j.msksp.2019.04.003 [CrossRef]31014920
  43. Craig DI, Lininger MR, Wayment HA, Huffman AH. Investigation of strategies to improve concussion reporting in American football [published online ahead of print March 12, 2019]. Res Sports Med. doi:10.1080/15438627.2019.1586706 [CrossRef]
  44. Zohar D, Luria G. A multilevel model of safety climate: cross-level relationships between organization and group-level climates. J Appl Psychol. 2005;90(4):616–628. doi:10.1037/0021-9010.90.4.616 [CrossRef]16060782
  45. Raudenbush S, Bryk A, Congdon R. HLM 7 for Windows. Scientific Software International, Inc.; 2011.
  46. Baugh CM, Meehan WP 3rd, Kroshus E, McGuire TG, Hatfield LA. College football players less likely to report concussions and other injuries with increased injury accumulation [published online ahead of print June 27, 2019]. J Neurotrauma. doi:10.1089/neu.2018.6161 [CrossRef]
  47. Sarmiento K, Mitchko J, Klein C, Wong S. Evaluation of the Centers for Disease Control and Prevention's concussion initiative for high school coaches: “Heads Up: Concussion in High School Sports”. J Sch Health. 2010;80(3):112–118. doi:10.1111/j.1746-1561.2010.00491.x [CrossRef]20236412
  48. Glang A, Koester MC, Beaver SV, Clay JE, McLaughlin KA. Online training in sports concussion for youth sports coaches. Int J Sports Sci Coach. 2010;5(1):1–12. doi:10.1260/1747-9541.5.1.1 [CrossRef]20640175
  49. Kroshus E, Kubzansky LD, Goldman RE, Austin SB. Norms, athletic identity, and concussion symptom under-reporting among male collegiate ice hockey players: a prospective cohort study. Ann Behav Med. 2015;49(1):95–103. doi:10.1007/s12160-014-9636-5 [CrossRef]
  50. NCAA Academic and Membership Affairs Staff. 2016–2017 NCAA Division I Manual. Indianapolis, IN: NCAA; 2016.
  51. Lininger MR, Wayment HA, Craig DI, Huffman AH, Lane TS. Improving concussion-reporting behavior in national collegiate athletic association Division I football players: evidence for the applicability of the socioecological model for athletic trainers. J Athl Train. 2019;54(1):21–29. doi:10.4085/1062-6050-47-18 [CrossRef]30721092

Demographic Characteristics and Response Rates of Student–Athletes in Sample

Variable Frequency Percentage Number of Eligible Members of the Team Response Rate
Institution (n = 248)
  1 59 23.8 110 53.6
  2 102 41.1 110 92.7
  3 87 35.1 100 87.0
Race (n = 218)
  White 100 45.9
  African American 80 36.7
  Hispanic/Latino 4 1.8
  Native American/Alaskan Native 0 0
  Asian 1 0.5
  Other 33 15.1
Year in school (n = 217)
  Freshman 56 25.8
  Sophomore 52 24.0
  Junior 48 22.1
  Senior 38 17.5
  Fifth year senior 23 10.6
Position (n = 240)
  Offense 122 50.8
  Defense 118 49.2

Previously Validated Scales With Individual Items Including Measure of Internal Consistency Used in Questionnaire

Scale Items Rating Scale Cronbach's Alpha Variable/Level
Concussion Reporting Intentions23 I intend to report. I plan to report. I will make an effort to report. 1 (not very likely) to 7 (very likely) 0.98 Outcome
Self-efficacy of Reporting30 I am confident in my ability to. . . report symptoms of a concussion, even when I really want to keep playing, report symptoms of a concussion, even when I think my teammates want me to play, report symptoms of a concussion, even if I do not think the symptoms are all that bad, report specific symptoms, even if I am not sure that it is actually a concussion. 1 (strongly disagree) to 5 (strongly agree) 0.95 Predictor; Individual-level
Reporting Attitudes30 If I report what I suspect might be a concussion, my teammates will… respect me, think I made the right decision, not support me, think less of me. 1 (strongly disagree) to 5 (strongly agree) 0.86 Predictor; Individual-level
Social Norms Scale23 People I know think I should report. People who are important to me think I should report. It is expected of me to report. People who are important would approve of my reporting. 1 (strongly disagree) to 5 (strongly agree) 0.91 Predictor; Individual-level
Safety Climate Scale41 My coach. . . is committed to improving player safety, places a strong emphasis on player safety, ensures players have adequate player safety training, encourages players to become involved in player safety, praises players' safe behavior, shows support for football players who report possible concussions, openly discusses player safety with his team. 1 (strongly disagree) to 5 (strongly agree) 0.96 Predictor; Team-level
Team Identification Scale40 I feel very connected with this team. Being a member of this team is very important to me. I am very happy that I belong to this team. 1 (strongly disagree) to 5 (strongly agree) 0.94 Predictor; Team-level

Descriptive Statistics and Normality Estimates of Concussion Reporting Behavior and Football Program Variables

Variable Mean Standard Deviation Skewness Kurtosis
SER 3.89 0.94 −0.75 0.27
NCR 4.35 0.64 −1.32 3.95
Log(NCR) 0.19 0.16 0.20 −0.76
RA 3.86 0.79 −0.45 0.45
CRI 5.39 1.57 −0.99 0.43
TI 4.48 0.87 −2.79 9.33
SCS 4.25 0.85 −1.83 5.53

Pearson's Pairwise Correlations of Individual- and Team-Level Variables Used in Statistical Analyses

Age SES Position Team ID SCS SER Log(NCR) RA CRI
Age (y) −0.001 −0.13 0.19b −0.20b −0.18b 0.08 −0.12 −0.15a
SES 0.10 −0.03 0.03 0.05 −0.04 −0.03 −0.05
Position −0.04 −0.05 0.08 0.10 −0.06 0.08
Team ID −0.05 −0.14a −0.22b −0.02 −0.07
SCS 0.35b −0.02 0.39b 0.17b
SER −0.02 0.28b 0.51b
Log(NCR) −0.07 −0.07
RA 0.24b
CRI

Results of Multilevel Modeling for Concussion Reporting Behaviors

Model 1: Unconditional Model

Fixed Effect Parameter Coefficient SE df t

Intercept B00 5.80 0.14 207 41.51b
Self-efficacy B10 0.73 0.12 205 5.89b
Reporting Attitudes B20 0.28 0.12 205 2.38a
Norms of Concussion Reporting B30 −1.05 0.62 205 −1.68

Random Effect Parameter Variance Component SD df χ2

Individual difference of initial value r0j 0.31 0.56 207 251.58a

Model 2: Conditional Model

Fixed Effect Parameter Coefficient SE df t

Intercept B00 5.94 0.24 204 24.36a
Position B01 0.24 0.17 204 1.42
Team Identification B02 −0.13 0.17 204 −0.75
Safety Climate B03 −0.31 0.19 204 −1.66
Self-efficacy B10 0.74 0.12 205 6.02b
Reporting Attitudes B20 0.36 0.12 205 2.88a
Norms of Concussion Reporting B30 −1.23 0.61 205 −2.00a

Random Effect Parameter Variance Component SD df χ2

Intercept r0j 0.31 0.56 204 247.87a
Level 1 effect eij 1.32 1.15

Model 3: Adjusted Conditional Model

Fixed Effect Parameter Coefficient SE df t

Intercept B00 5.93 0.25 202 24.14a
Age B01 −0.07 0.06 202 −1.25
SES B02 0.0005 0.0004 202 −1.36
Position B03 0.26 0.17 202 1.56
Team Identification B04 −0.10 0.17 202 −0.55
Safety Climate B05 −0.37 0.19 202 −1.91
Self-efficacy B10 0.74 0.12 205 6.15b
Reporting Attitudes B20 0.36 0.12 205 2.96a
Norms of Concussion Reporting B30 −1.21 0.62 205 −1.94

Random Effect Parameter Variance Component SD df χ2

Intercept r0j 0.31 0.56 202 245.39a
Level 1 effect eij 1.31 1.14
Authors

From the Department of Physical Therapy and Athletic Training (MRL, DIC), Psychological Sciences (HAW, AHH), and W. A. Franke College of Business (AHH), Northern Arizona University, Flagstaff, Arizona.

Supported by the Research Grand Challenge: Mind Matters sponsored by the National Collegiate Athletic Association and the U.S. Department of Defense.

The authors have no financial or proprietary interest in the materials presented herein.

The authors thank the National Collegiate Athletic Association and the U.S. Department of Defense for their support. They also thank graduate assistants Patrick Doyle, Keragan Cavolo, and Emilee Jung for their work in support of this research.

Correspondence: Monica R. Lininger, PhD, LAT, ATC, Department of Physical Therapy and Athletic Training, Northern Arizona University, P.O. Box 15094, Flagstaff, AZ 86011. E-mail: monica.lininger@nau.edu

Received: March 01, 2019
Accepted: July 15, 2019
Posted Online: October 30, 2019

10.3928/19425864-20190822-01

Sign up to receive

Journal E-contents