INTRODUCTION
Concussion during athletic competition, termed sport-related concussion (SRC), is a form of mild traumatic brain injury (mTBI) induced by biomechanical forces that triggers a pathophysiological process resulting in functional derangements, typically in the absence of abnormal neuroradiographic correlates (McCrory et al., Reference McCrory, Meeuwisse, Dvorak, Aubry, Bailes, Broglio and Vos2017). Compared to the general population, athletes participating in contact or collision sports are more likely to experience concussion (Coronado, McGuire, Faul, Sugerman, & Pearson, Reference Coronado, McGuire, Faul, Sugerman and Pearson2012; Gessel, Fields, Collins, Dick, & Comstock, Reference Gessel, Fields, Collins, Dick and Comstock2007; McCrory et al., Reference McCrory, Meeuwisse, Dvorak, Aubry, Bailes, Broglio and Vos2017; Tator et al., Reference Tator, Davis, Dufort, Tartaglia, Davis, Ebraheem and Hiploylee2016).
Following SRC, immediate, objectively measured signs and subsequent self-reported symptoms should be assessed by a physician or other licensed healthcare provider (Belanger & Vanderploeg, Reference Belanger and Vanderploeg2005; Karr, Areshenkoff, & Garcia-Barrera, Reference Karr, Areshenkoff and Garcia-Barrera2014; McCrory et al., Reference McCrory, Meeuwisse, Aubry, Cantu, Dvorák, Echemendia and Turner2013, Reference McCrory, Meeuwisse, Dvorak, Aubry, Bailes, Broglio and Vos2017). These on-field signs often include balance/postural instability (Guskiewicz, Ross, & Marshall, Reference Guskiewicz, Ross and Marshall2001; McCrea et al., Reference McCrea, Guskiewicz, Marshall, Barr, Randolph, Cantu and Kelly2003), memory (Cantu, Reference Cantu2001; Graham, Rivara, Ford, & Mason Spicer, Reference Graham, Rivara, Ford and Mason Spicer2014; McCrory et al., Reference McCrory, Meeuwisse, Dvorak, Aubry, Bailes, Broglio and Vos2017), concentration/working memory (McCrea et al., Reference McCrea, Guskiewicz, Marshall, Barr, Randolph, Cantu and Kelly2003), and visual tracking/pupil dysfunction (dilatation, non-reactive to light, unequal size, convergence insufficiency, etc.; McCrea, Kelly, Kluge, Ackley, & Randolph, Reference McCrea, Kelly, Kluge, Ackley and Randolph1997; Mucha et al., Reference Mucha, Collins, Elbin, Furman, Troutman-Enseki, DeWolf and Kontos2014; Wojtys et al., Reference Wojtys, Hovda, Landry, Boland, Lovell, McCrea and Minkoff1999).
During recovery, individual symptoms and their severity are monitored via self-reported indices such as the Symptom Scale from the Sport Concussion Assessment Tool (SCAT; McCrory et al., Reference McCrory, Meeuwisse, Dvorak, Aubry, Bailes, Broglio and Vos2017). SRC symptoms have been empirically analyzed to yield four primary symptom clusters: (1) cognitive-migraine-fatigue (CMF), (2) affective, (3) somatic, and (4) sleep (Table 1; Kontos et al., Reference Kontos, Elbin, Schatz, Covassin, Henry, Pardini and Collins2012).
On-field signs and symptom cluster
a Adapted from the SCAT3.
b Derived from Kontos et al., Reference Kontos, Elbin, Schatz, Covassin, Henry, Pardini and Collins2012.
c Postural instability consisted of three Modified Balance Error Scoring System (M-BESS; Hunt, Ferrara, Bornstein, & Baumgartner, Reference Hunt, Ferrara, Bornstein and Baumgartner2009) trials. Positions listed were recorded as a fail if a single major loss of balance occurred during the 20-s period.
d Working memory/concentration was based on whether athletes were able to successfully state the days of the week backward. Performance on days of the week backward was recorded as a fail if they were unable to state the entire sequence correctly. Working memory/ concentration also consisted of the portion of the SCAT3 that involves repeating strings of digits backward. Digits backward were coded as a fail if they inaccurately recited one of the two the strings (two and three numbers for consecutive strings).
e Visual tracking and pupil abnormality (abnormal constriction/dilation, non-reactive to light, unequal size, convergence insufficiency) was recorded as a fail if any qualitative abnormality was observed.
f Immediate memory was characterized by an athlete’s ability to repeat three words for encoding and later spontaneously recall the words. Should athletes not be able to repeat or recall all three words, their performance was recorded as a fail.
Preliminary research has shown provisional associations between on-field signs and symptoms in the acute phase post injury. One study of 78 concussed high school and collegiate athletes demonstrated that on-field amnesia was significantly associated with reduced neuropsychological scores and increased total symptom scores at 2 days post injury (Collins et al., Reference Collins, Iverson, Lovell, McKeag, Norwig and Maroon2003). Pearce and colleagues (Reference Pearce, Sufrinko, Lau, Henry, Collins and Kontos2015) showed that student-athletes who experienced oculomotor (convergence insufficiency) and cognitive impairment reported significantly greater total symptom scores than those who did not within the first 10 days post-injury (Pearce et al., Reference Pearce, Sufrinko, Lau, Henry, Collins and Kontos2015). Moreover, objectively measured signs have also been linked to specific self-reported symptom patterns. A study of 138 male high school football players suggested that those with greater cognitive deficits (i.e., memory and reaction time) were more likely to endorse posttraumatic migraines during the initial course of recovery between 1 and 7 days (Kontos et al., Reference Kontos, Elbin, Lau, Simensky, Freund, French and Collins2013).
Structural equation modeling (SEM) is a statistical method well suited to study associations between signs and symptoms after SRC due to its capacity to examine complex networks of multiple constructs. Using SEM, the aim of the current study was to investigate the relationship between immediate, objective on-field signs and self-reported symptoms during the first day post injury in a sample of high school athletes who sustained a SRC. Due to the limited opportunity for athletes to experience certain symptoms during the first day of recovery (i.e., sleep and affective clusters), the current study examined symptom endorsement in two of the four previously mentioned clusters: (1) CMF and (2) somatic.
Several previous studies formed the basis for our empirically-driven hypotheses. First, due to the previously demonstrated relationship between verbal memory scores and CMF cluster symptoms (Broglio, Sosnoff, & Ferrara, Reference Broglio, Sosnoff and Ferrara2009; Kontos et al., Reference Kontos, Elbin, Lau, Simensky, Freund, French and Collins2013), it was hypothesized that athletes’ on-field immediate memory would be associated with the CMF cluster. Additionally, based on the previously demonstrated relationship between memory functioning (i.e., posttraumatic amnesia) and total symptom scores (Collins et al., Reference Collins, Iverson, Lovell, McKeag, Norwig and Maroon2003), it was hypothesized that immediate memory would also be significantly associated with the somatic cluster. Second, due to an expected relationship with subjectively reported CMF symptoms and objectively measured cognitive performance, it was hypothesized that on-field working memory/concentration deficits would be associated with the self-reported symptoms in the CMF cluster, which is also supported by studies investigating dual task paradigms (Howell, Osternig, & Chou, Reference Howell, Osternig and Chou2015).
Third, a relationship between postural instability and the CMF cluster was hypothesized due to a previously demonstrated association between objectively measured postural stability and cognitive measures of working memory. Specifically, Guskiewicz et al. (Reference Guskiewicz, Ross and Marshall2001) showed that athletes exhibiting postural instability also demonstrated poorer working memory performance (Trail-Making Test B and Wechsler Digit Span Test Backward) at day 1 post injury. This relationship is further supported by prior studies documenting the relationship between objective and subjective measures of balance within 48 hr of concussion (Broglio et al., Reference Broglio, Sosnoff and Ferrara2009). Fourth, visual tracking/ pupillary dysfunction was thought to predict increased CMF cluster endorsement based on previously demonstrated associations between vestibulo-ocular dysfunction and dizziness and concentration difficulties (Ellis et al., Reference Ellis, Cordingley, Vis, Reimer, Leiter and Russell2017).
METHODS
Participants and Setting
The sample included 237 concussed, student-athletes participating in a variety of sports in the southeastern United States (Table 2). Inclusion criteria were: (a) SRC diagnosed by a certified athletic trainer (ATC) and confirmed by team physician based on the international Concussion In Sport Group guidelines (CISG4; McCrory et al., Reference McCrory, Meeuwisse, Aubry, Cantu, Dvorák, Echemendia and Turner2013), (b) sign and symptom documentation within the first 24 hr post injury, (c) high school athlete ages 13 to 18, and (d) English speaking. All athletes not meeting these criteria were excluded from analysis.
Demographics
Note. Number and percentages of recorded demographics.
Of the 237 who met criteria for inclusion in the study, 175 (73.8%) were male and 65 (27.4%) reported a prior history of concussion. Objective signs were evaluated at the time of injury or immediately after the participating game or practice (i.e., in the locker room). Symptoms were recorded by the ATC within 24 hr of the injury and entered into an institutional database. Injuries occurred either during a game (184; 77.6%), practice (45; 19.0%), or other (e.g., weight room, school, etc.; 8; 3.4%). Data included in this manuscript were obtained in compliance with the Helsinki Declaration and Institutional Review Board approval was obtained for the purpose of the study (108202017190522).
Materials
On-field signs, which were based on interpretations of test responses/ results, were derived from the SCAT3 assessment (McCrory et al., Reference McCrory, Meeuwisse, Aubry, Cantu, Dvorák, Echemendia and Turner2013) and adapted for ease of rapid on-field administration. The constructs assessed included; (1) immediate memory, (2) balance, and (3) concentration/ working memory (Table 1). Additionally, consistent with the National Athletic Trainers’ Association Position Statement on the management of SRC, (4) visual tracking and pupil abnormality (abnormal constriction/dilation, non-reactive to light, unequal size, convergence insufficiency, etc.) was included and assessed as an observed on-field sign (Broglio et al., Reference Broglio, Cantu, Gioia, Guskiewicz, Kutcher and Palm2014; Guskiewicz et al., Reference Guskiewicz, Bruce, Cantu, Ferrara, Kelly, McCrea and Valovich McLeod2004). Immediate memory was characterized by an athlete’s ability to repeat three words for encoding and later spontaneously recall the words. Athletes were administered one trial per task and should athletes not be able to repeat and recall all three words in a single trial, their performance was recorded as a fail.
Also derived and adapted from the SCAT3, working memory/ concentration was based on whether athletes were able to successfully state the days of the week backward. Performance on days of the week backward was recorded as a fail if they were unable to state the entire sequence correctly in one trial. Working memory/ concentration also consisted of adapted trials of the SCAT3 that involves repeating strings of digits backward. The digits backward component included two trials, the first with two numbers and the second with three numbers. Digits backward was coded as a fail if athletes were unable to successfully complete both trials. Postural stability consisted of three mBESS positions (i.e., single, double, and tandem stances) and was recorded as a fail if a single major loss of balance occurred during the 20-s period for each position.
To assess self-reported symptoms, the Symptom Scale from the SCAT3 was used (McCrory et al., Reference McCrory, Meeuwisse, Aubry, Cantu, Dvorák, Echemendia and Turner2013). The SCAT3 Symptom Scale is comprised of 22 symptoms commonly associated with concussion, each of which is rated on a 0–6 Likert scale, with 0=none and 6=severe. The 22 symptoms have previously been factor analyzed and classified into four primary postconcussion symptom categories (viz., clusters). These four clusters include: (1) CMF, (2) affective, (3) somatic, and (4) sleep (Kontos et al., Reference Kontos, Elbin, Schatz, Covassin, Henry, Pardini and Collins2012). As stated above, only the CMF and somatic symptom clusters were included as outcomes for the current study. Given that the presence or absence of a symptom (binary variable) was of primary interest, student-athlete responses were coded as “No” with a value of 0, and “Yes” for a value of 1 or greater. The number of endorsed symptoms within a particular cluster was included as the dependent variable.
Statistical Analysis
A two-step SEM analysis was performed (Anderson & Garbing, Reference Anderson and Garbing1988). First, the measurement model was estimated using the maximum likelihood method and evaluated based on a chi-square test and several other fit indices, which included comparative fit index (CFI), standardized root-mean-square residual (SRMR), root-mean-square error of approximation (RMSEA), and normed fit index (NFI). To examine the factor structures, confirmatory factor analysis (CFA) was used to test the measurement model by examining the relationship between individual measured signs (e.g., single leg stance) and the constructs/latent variables that represent them (e.g., postural instability).
Second, the full proposed structural model that examines the relationship between signs and self-reported symptoms was estimated using a similar process described above. For instances in which on-field signs were significantly predictive of symptom clusters, follow-up binary logistic regression analyses were performed to further examine the relationship between on-field signs and specific self-reported symptoms in significant clusters. All analyses were performed using the computer software LISREL Version 9 (Fleishman & Benson, Reference Fleishman and Benson1987) and Statistical Package for the Social Sciences (SPSS Statistics 24.0).
RESULTS
Gender, sport, concussion history, and injury setting (practice or game) of the 237 subjects are provided in Table 2. The most commonly observed on-field sign following SRC was the inability to recite strings of numbers backward (N=84; 35.4%) and the more frequently endorsed symptoms were from the CMF cluster (M=3.1; SD=1.97). The frequency and percentages of student-athletes exhibiting failed trials of each on-field sign is presented in Table 3. Gender differences were not observed in the endorsement of CMF, t=0.75, p=.94, or somatic symptom clusters, t=−0.52, p=.61. Gender differences were also not observed on pupil and postural instability signs, including pupil/ ocular dysfunction, χ2=0.49, p=.49, double leg stance, χ2=0.61, p=.43, single leg stance, χ2=0.07, p=.79, and tandem stance, χ2=0.07, p=.79. Furthermore, gender differences were not observed in working memory and immediate memory signs, including days backward, χ2=0.02, p=.90, numbers backward, χ2=0.0, p=.99, repeat words, χ2=0.37, p=.54, and remember words, χ2=0.87, p=.35.
Correlation matrix of signs and symptoms
Note. Correlation table demonstrating relationships (Pearson correlation) of all the included variables in the model.
a Number of symptoms endorsed within each cluster; presented as mean and standard deviation.
b Number and percentage of athletes exhibiting a failed performance of each on-field sign.
Measurement Model
Fit indices and parameter estimates from the CFA indicated a very good fit of the measurement model (Table 4). All pattern coefficients of observed variables (e.g., single leg stance) were significant, indicating that measured on-field indicators were significantly associated with their corresponding latent variables (e.g., postural instability). Coefficients for latent variables/ constructs fell above the acceptable cutoff (0.70; Hair, Anderson, Tatham, & Black, Reference Hair, Anderson, Tatham and Black1998), ranging from 0.74 to 0.81 (postural instability), 0.72 to 0.75 (immediate memory), and 0.79 to 0.84 (concentration/ working memory). A correlation matrix displaying relationships between the variables is provided in Table 3.
Fit indices for proposed measurement and structural models
Note. Values suggesting good fit for the model have previously been proposed as comparative fit index (CFI > 0.95), normed fit index (NFI > .90), root-mean-square error of approximation (RMSEA ≤ 0.06), and standardized root-mean-square residual (SRMR ≤ 0.08; Byrne, Reference Byrne1994; Hu & Bentler, Reference Hu and Bentler1999)
a 95% CI.
Goodness of fit of the measurement and structural model was based on acceptable values, including the CFI (> 0.95), NFI (> .90), RMSEA (≤ 0.06), and SRMR (≤ 0.08; Byrne, Reference Byrne1994; Hu & Bentler, Reference Hu and Bentler1999). Fit indices indicated an acceptable fit of the measurement model χ2(23,237)=38.82, p=.02; CFI=0.97; NFI=0.94; RMSEA=0.05 (95% CI [0.02, 0.08]); SRMR=0.01.
Structural Model
To test the proposed structural model, or relationship between signs and two self-reported symptom clusters, the two symptom clusters were added to the measurement model. Initial paths from latent on-field signs to designated symptom clusters were added based on hypotheses described above. The structural model demonstrated a good fit to the data, χ2(27,237)=48.35, p=.01; CFI=0.96; NFI=0.92; RMSEA=0.05 (95% CI [0.02, 0.10]); SRMR=0.02 (Table 4; Figure 1).
Final structural model of signs and symptoms.
Significant Predictors in Final Model
Two paths from on-field signs to self-reported symptom clusters during the first day post injury were significant. Immediate memory, β=0.20, p=.04 and postural instability, β=0.19, p<.01, significantly predicted a greater likelihood of student-athletes endorsing symptoms in the CMF cluster (Table 5). Follow-up bivariate regression analyses of the significant on-field signs and individual symptoms in the CMF cluster were also significant for particular symptoms (Table 6). Bonferroni corrected alpha levels were set at 0.0025 to control for inflated type-I error associated with multiple comparisons.
Standardized coefficients for the structural model
a Paths significant at the p > .05 level.
b Paths significant at the p > .01 level.
*Paths not analyzed within the model.
Individual symptoms of significant on-field signs and cluster
Note. Bivariate regression analyses of significant on-field signs and symptoms in cognitive-migraine-fatigue cluster.
a Chi-squared statistic of bivariate regressions.
b Standardized beta regression weights.
c Bonferroni corrected alpha level controlling for multiple comparisons set at 0.0025.
d Lower bound 95% CI of OR.
e Upper bound 95% CI of OR.
*Significant at the Bonferroni corrected alpha level of 0.0025.
Immediate memory was significantly associated with student-athletes’ reporting of trouble remembering, χ 2 =37.92, p<.001, odds ratio (OR)=3.89 (95% confidence interval (CI) [2.47, 6.13]), and concentration difficulties, χ 2 =10.84, p=.001, OR=2.13 (95% CI [1.37, 3.30]). Postural instability was significantly associated with student-athletes’ reporting of trouble remembering, χ 2 =12.08, p<.001, OR=1.76 (95% CI [1.29, 2.40]).
Non-significant Predictors in Final Model
Multiple paths in the proposed structural model did not achieve statistical significance and indicated that certain on-field signs did not effectively predict the likelihood of reporting particular symptoms within the first day post injury. For instance, the occurrence of visual tracking/pupillary dysfunction, β=0.00, p=.95, and working memory/concentration, β=0.14, p=.11 during on-field evaluation, as assessed in the current study, did not increase the likelihood of high school athletes endorsing CMF-related symptoms following SRC. Furthermore, immediate memory difficulties did not predict a greater likelihood of reporting symptoms in the somatic cluster, β=0.06, p=.41.
DISCUSSION
The current study examined whether on-field signs immediately following injury could accurately predict student-athletes’ symptom reporting within 1 day following SRC. In a sample of 237 concussed student-athletes, immediate memory was significantly associated with the subsequent self-reported symptoms of trouble remembering and concentration difficulties. Furthermore, postural instability at the time of the concussion was significantly associated with subsequent self-reported symptoms of trouble remembering.
Significant Associations
Immediate memory at the time of injury was associated with the CMF symptom cluster (e.g., trouble remembering and difficulty concentrating). The association of these specific signs and symptoms is similar to a previous study of 32 concussed collegiate athletes. At 48 hr following injury, a relationship between objectively measured memory (i.e., ImPACT verbal memory score) and symptom endorsement of difficulty remembering and/or concentrating was demonstrated (Broglio et al., Reference Broglio, Sosnoff and Ferrara2009).
While changes in cognition are likely associated with the metabolic demands following SRC (Giza & Hovda, Reference Giza and Hovda2014; Yoshino, Hovda, Kawamata, Katayama, & Becker, Reference Yoshino, Hovda, Kawamata, Katayama and Becker1991), the exact physiological mechanism is not fully understood. Prior functional magnetic resonance imaging studies suggest that neurotransmitter and synaptic alterations may be the mechanism driving changes in cognition during the acute phase of recovery. For example, reduced task-related blood oxygen level dependent signaling has previously been associated with poorer performance on a task of attention/ working memory in high school athletes acutely following SRC (Keightley et al., Reference Keightley, Saluja, Chen, Gagnon, Leonard, Petrides and Ptito2014). Given the relationship between memory performance and cognitive symptoms, these results suggest that altered neurotransmission may begin as quickly as minutes after injury and persist further during the acute phase of recovery (McCrea et al., Reference McCrea, Guskiewicz, Marshall, Barr, Randolph, Cantu and Kelly2003).
Measured postural instability also increased a student-athlete’s likelihood of reporting CMF-related symptoms. Specifically, postural instability was associated with an increased likelihood of self-reported trouble remembering during the first day post-injury. The relationship between postural instability and cognitive-related self-reported symptoms (i.e., trouble remembering) is highlighted through studies assessing SRC with a dual-task paradigm (Lee, Sullivan, & Schneiders, Reference Lee, Sullivan and Schneiders2013). In a study of 10 concussed individuals, the inclusion of a divided attention component in a dual-task gait balance test was a more sensitive measure of balance than a single gait balance task within 48 hr post-injury (Parker, Osternig, Lee, Donkelaar, & Chou, Reference Parker, Osternig, Lee, Donkelaar and Chou2005).
As highlighted above, the energy crisis and hypometabolism of glucose may be the mutual mechanism driving the association between postural instability at the time of injury and self-reported memory difficulties 1 day later. While not yet directly demonstrated in mTBI or concussion, previous research in the area of moderate and severe TBI has shown a disproportional acute hypometabolism in the frontal and temporal lobes (Wright et al., Reference Wright, McArthur, Alger, Van Horn, Irimia, Filippou and Vespa2013), two regions associated with postural instability (executive functions of attention and concentration on balance task; Reilly, van Donkelaar, Saavedra, & Woollacott, Reference Reilly, van Donkelaar, Saavedra and Woollacott2008) and memory (Squire & Zola-Morgan, Reference Squire and Zola-Morgan1991).
Within the same study, this acute hypometabolism was significantly associated with later measured atrophy in the same regions and more distal attention, executive function, and psychomotor deficits. While mTBI is unlikely to result in long-term atrophy and permanent cognitive impairments (Giza & Hovda, Reference Giza and Hovda2014), the higher levels of frontal-temporal hypometabolism seen in severe TBI may exist to a lesser extent in mTBI, with a return to physiologic and neurologic homeostasis and without irreversible cellular changes such as protease activation and altered cytoskeletal proteins.
Non-significant Associations
Results did not support the hypothesized relationship between immediate memory and the somatic symptom cluster. Other previous studies suggesting this potential relationship included athletes who suffered from a more severe memory condition (i.e., posttraumatic amnesia), which may account for the absence of the expected relationship observed in the current result (Collins et al., Reference Collins, Iverson, Lovell, McKeag, Norwig and Maroon2003). Furthermore, results failed to show an association between working memory/ concentration and self-reported CMF symptoms. The lack of relationship between the two in the current study was unexpected, considering the significant correlation between subjective cognitive-related symptom reporting and objective neurocognitive measures reported in the study of 32 concussed collegiate athletes by Broglio and colleagues (Reference Broglio, Sosnoff and Ferrara2009).
Furthermore, in contrast to Ellis et al. (Reference Ellis, Cordingley, Vis, Reimer, Leiter and Russell2017), the current study failed to demonstrate an association between visual tracking/ pupillary dysfunction and CMF symptoms. The positive association, as compared to the null findings in the current study, may be attributable to differences in study methodologies. Specifically, Ellis et al. (Reference Ellis, Cordingley, Vis, Reimer, Leiter and Russell2017) involved a neurosurgeon performing a comprehensive vestibulo-ocular exam in an office setting, whereas the current study used a general screening of pupil/ocular motor dysfunction by an ATC on the field. The discrepancy between studies may have been reduced with consistent use of a vestibular and ocular impairment measure, such as the Vestibular/Ocular Motor Screening assessment, which is becoming a standard instrument in the SRC research field.
Clinical Utility of the Model
Clarifying the relationship between measured on-field signs and symptom reporting has potential to improve SRC management (Graham et al., Reference Graham, Rivara, Ford and Mason Spicer2014). This is especially true when considering underreporting by student athletes (Guskiewicz et al., Reference Guskiewicz, Bruce, Cantu, Ferrara, Kelly, McCrea and Valovich McLeod2004; Kuhn et al., Reference Kuhn, Zuckerman, Yengo-Kahn, Kerr, Totten, Rubel and Solomon2017). A clinician’s confidence in making an initial diagnosis is bolstered when discrepancy exists between concerning signs without symptom endorsement, which an athlete may choose to conceal. Furthermore, using objective signs to predict immediate symptoms can be of use in the case of delayed symptom onset (Duhaime et al., Reference Duhaime, Beckwith, Maerlender, McAllister, Crisco, Duma and Greenwald2012; McCrory et al., Reference McCrory, Meeuwisse, Dvorak, Aubry, Bailes, Broglio and Vos2017). In one study of 450 collegiate athletes wearing Head Impact Telemetry technology, half of the 48 diagnosed concussions had delayed or unclear timing of symptom onset (Duhaime et al., Reference Duhaime, Beckwith, Maerlender, McAllister, Crisco, Duma and Greenwald2012). Observed on-field signs can preemptively alert parents and coaches to potential symptoms that are more likely to occur, should there be a delayed onset.
While treatment guidelines for the initial day of recovery are often limited and involve primarily rest (McCrory et al., Reference McCrory, Meeuwisse, Dvorak, Aubry, Bailes, Broglio and Vos2017), the associations between immediate signs and symptoms have implications for individualized, preventative education-based interventions. Education of coaches and parents, which has been recommended strongly through several position statements (Broglio et al., Reference Broglio, Cantu, Gioia, Guskiewicz, Kutcher and Palm2014; Echemendia et al., Reference Echemendia, Iverson, McCrea, Broshek, Gioia, Sautter and Barr2012; Halstead, Walter, & Council on Sports Medicine and Fitness, Reference Halstead and Walter2010; Harmon et al., Reference Harmon, Drezner, Gammons, Guskiewicz, Halstead, Herring and Roberts2013; McCrory et al., Reference McCrory, Meeuwisse, Aubry, Cantu, Dvorák, Echemendia and Turner2013), can be individualized based on the specific signs and symptoms observed and may potentially reduce symptoms, psychological distress, and disruption of social activities.
For example, in separate studies of adolescents and adults in the general population, individuals who received an informational booklet on common symptoms following head injury with suggested coping strategies reported fewer symptoms and experienced less psychological distress at 3 months follow-up (Ponsford et al., Reference Ponsford, Willmott, Rothwell, Cameron, Ayton, Nelms and Ng2001, Reference Ponsford, Willmott, Rothwell, Cameron, Kelly, Nelms and Curran2002). Furthermore, parental education by providers who may be able to predict future symptoms from documented signs at the time of injury can aid in easing potential family distress, as parents and caregivers may be able to anticipate and expect the occurrence of specific symptoms, as opposed to being surprised and alarmed by them. Given that mTBI in children in the general population has been shown to increase family distress (Ganesalingam et al., Reference Ganesalingam, Yeates, Ginn, Taylor, Dietrich, Nuss and Wright2008), more specific parental counseling may allay family concerns and foster a healing home environment.
These results may also have implications for individualized return-to-school interventions. In a recent study conducted by Ransom et al. (Reference Ransom, Vaughan, Pratson, Sady, McGill and Gioia2015), symptomatic students who returned to school following SRC reported higher levels of concern for injury-related adverse effects on learning, school performance, and increased school problems. Parents exhibited similarly heightened anxiety. Prophylactic individualized education-based reassurance can be provided to students and parents early in the recovery process to alleviate concerns related to learning and school performance. For instance, should the student exhibit immediate memory difficulties and postural instability at injury, parents and students can be provided education and reassured about subsequent memory and concentration difficulties that may likely ensue. With this knowledge, symptom-specific academic accommodations can be arranged in advance of the student’s return to the learning environment (Gioia, Reference Gioia2016).
Future Directions
Logical follow-up to this study would include extending the current findings to assess longitudinal recovery and outcomes beyond the 1-day timeframe. Additionally, although this study included a diverse set of on-field signs as variables of interest, there is a wide array of on-field signs that can be used in the detection and management of SRC (e.g., orientation, coordination, etc.). Future studies should seek to replicate the associations demonstrated in the current study, while integrating additional on-field signs that are commonly assessed. Our knowledge of associations between on-field signs and symptoms would improve with prospective studies examining the current findings throughout the duration of recovery. For example, future studies should examine if the relationship between on-field signs and symptoms remains significant at days 3, 7, and furthermore, until full recovery is reached.
Limitations
This study is not without limitations. In particular, we were not able to control for many modifying factors previously shown to influence recovery from SRC and symptom reporting, such as a diagnosis of learning disorder/attention deficit hyperactivity disorder (Bonfield, Lam, Lin, & Greene, Reference Bonfield, Lam, Lin and Greene2013; Iverson, Atkins, Zafonte, & Berkner, Reference Iverson, Atkins, Zafonte and Berkner2016; Zuckerman, Lee, Odom, Solomon, & Sills, Reference Zuckerman, Lee, Odom, Solomon and Sills2013), history of treatment of depression and/ or anxiety (Iverson et al., Reference Iverson, Silverberg, Mannix, Maxwell, Atkins, Zafonte and Berkner2015; Yengo-Kahn & Solomon, Reference Yengo-Kahn and Solomon2015), history of previous concussion (Guskiewicz et al., Reference Guskiewicz, Bruce, Cantu, Ferrara, Kelly, McCrea and Valovich McLeod2004; McCrea et al., Reference McCrea, Guskiewicz, Marshall, Barr, Randolph, Cantu and Kelly2003), etc. Future research should attempt to replicate the current findings, while also examining these modifying factors as potential mediators of signs and symptoms.
Furthermore, although signs were objectively measured and assessed, items such as the mBESS stances did involve some subjective interpretation from ATCs, requiring them to individually gauge a “single major loss of balance.” While this was not codified operationally, it was performed by the same trainers, maintaining consistency in the rating. Further work should look to advance the findings of the current study through the inclusion of baseline measures and comparative post injury performances. Additionally, the current cohort was drawn from a southeastern geographical region, which may not be generalizable to other areas of the country or to international groups.
Conclusions
The current study examined whether on-field signs assessed immediately following injury could accurately predict student-athlete symptom reporting at 1 day following SRC in a sample of 237 concussed student-athletes. Findings indicated that immediate memory difficulties were significantly associated with the subsequent increased symptom reporting of trouble remembering and concentration difficulties. Postural instability at the time of the concussion was significantly associated with increased symptom reporting of trouble remembering. Through increased awareness of these associations, prophylactic management of SRC consisting of multimodal education-based interventions can be implemented based on anticipated symptoms from on field-signs that are observed immediately following SRC.
ACKNOWLEDGMENTS
Mr. Brett, Mr. Kuhn, and Drs. Yengo-Kahn, Jeckell, and Zuckerman report no conflicts of interest. Dr. Solomon is a consultant for the Nashville Predators, Tennessee Titans, and the athletic departments of Tennessee Tech University and the University of Tennessee, fees paid to institution. He is also a consultant to the National Football League Department of Health and Safety. No contents of this work have been previously presented. No grants or outside funding supported this work.
