Comparing peer relationship ratings among adolescents with sport and non-sport related concussions

Abstract

Objective:

To evaluate if peer relationship ratings differ among adolescents who sustain a sport-related concussion compared to those who sustain a non-sport-related concussion.

Study Design:

We conducted a cross-sectional investigation of 123 adolescents who sustained a concussion within 21 days before their initial post-injury evaluation (mean = 9.2 ± 3.9 days post-concussion). Participants reported whether their injury occurred during organized sport participation (sport-related concussion group) or not (non-sport-related concussion group). Participants completed the PROMIS Global Pediatric 25 assessment, as well as the Post-Concussion Symptom Inventory (PCSI) to rate concussion symptom severity and a background/demographic intake form during the evaluation. We compared PROMIS peer relationship ratings between groups using independent t-tests and evaluated the association between peer relationship ratings and sport-related concussion using multiple linear regression after adjusting for covariates including PROMIS anxiety rating, age, sex, concussion and musculoskeletal injury history, loss of consciousness at time of concussion, physical activity participation prior to study evaluation, and concussion symptom severity.

Results:

Most (N = 96, 78%) participants reported sustaining a concussion during sports. The group who sustained sport-related concussion reported significantly better post-concussion peer relationship ratings compared to the group who sustained a non-sport-related concussion (52.7 ± 7.7 vs. 47.7 ± 10.4; p = 0.009, Cohen’s d = 0.60). After covariate adjustment, those who sustained a sport-related concussion had a significantly greater post-concussion peer relationship rating than those who sustained a non-sport-related concussion (β = 4.19, 95% CI = 0.39, 8.00; p = 0.03). Higher peer relationship ratings were also associated with lower anxiety ratings following concussion (β= −0.21, 95% CI= −0.38, −0.04; p = 0.03).

Conclusion:

Adolescents who sustained a concussion during sports reported significantly greater post-concussion peer relationship ratings than those in non-sport settings. Better peer relationship ratings were associated with less anxiety in the post-injury period. These findings suggest the potential effects of contextual factors (e.g. peer support) to facilitate improved outcomes following adolescent concussion.

Introduction

Concussion, a subset of traumatic brain injury, is common among adolescents [1]. One in five adolescents are estimated to have experienced a concussion in their lifetime [2]. Due to the frequency and complexity of concussions, identifying effective strategies to optimize recovery is a research and clinical priority [3]. However, because concussions have variable presentations, determining the best care pathway for each individual can be challenging [1]. Furthermore, concussions can impact the developing adolescent brain in biological, psychological, and social domains [4]. Specifically, concussions can lead to a variety of symptoms that may impact mental or physical health [5]. Concussions are associated with decreased quality of life and mental health challenges, possibly due to behavioral or social adaptions (e.g. removal from academic or social activities due to symptom exacerbation) [6]. Specifically, social challenges may occur when the individual is removed from their everyday, usual activities due to concussion-related restrictions in school and activity participation. As a result of this social activity change, feelings of isolation may occur, and lead to additional problems such as stress or loneliness [7]. While the prognosis of a concussion can be challenging based solely on the initial injury evaluation, the amount of support available may influence post-concussion clinical presentation and recovery, as social support may reduce anxiety, stress, or other negative impacts [8].

Adolescent athletes may have unique mental health challenges after concussion due to the way in which the injury can influence activities of daily life, such as missed academic time or time away from teammates during competitions [9–11]. Missing practices and games after a sport-related concussion can create feelings of isolation, being left out, or anxiety that is less commonly reported outside of sports [12]. Along with the external pressures (e.g. coaches, parents, teammates) placed on adolescent athletes to return to sport, adolescent athletes may also experience internal pressures, such as feeling they are letting their teammates or themselves down [12]. Past work investigating other sport-related injuries (e.g. ACL rupture) among adolescents suggests that many athletes have doubts in their ability to recover from injuries and social support may be a key factor in injury recovery [13]. In the context of sport-related concussion, specifically, teammates may provide both positive support (reassurance) or negative feelings (stress or peer pressure). While many athletes can rely on their teammates to succeed in their absence, anxiety around losing playing time, letting the team down, or other social pressures may be challenging. Peer support may vary between those who sustained concussions during sports and those who were not, and past research suggests that collegiate athletes with non-sport-related concussions had worse recovery outcomes (i.e. longer recovery duration, more severe symptoms) than those injured during sports [14]. Furthermore, academic and/or social activities may be altered after concussion [15]. Yet, access to guidance from healthcare professionals who can assist with academic or social guidance or adjustments may vary between those injured during sports relative to those injured outside of sports based on their availability to interact with qualified clinicians. Whether perceived peer support after adolescent concussion varies among those who were injured playing sports or not is not known. As such, comparing sport and non-sport related concussions may provide valuable insight into better ways to understand concussion recovery among individuals from varying backgrounds and life perspectives [16].

Concussion recovery is dependent on a myriad of factors, particularly for recovery of overall concussion symptoms or mental health difficulties [1,17–19]. Among adolescents, the setting where an injury occurs provides key context into the type of support they experience which may be associated with better or worse quality of life [20]. Many of the negative mental health symptoms associated with concussion (e.g. anxiety or depressive symptoms) [21] may be mitigated by the presence of peer support [22]. Therefore, the purpose of our investigation was to evaluate if adolescents who sustained a concussion during sports reported different post-injury peer relationship ratings than those who sustained non-sport-related concussions. We hypothesized that adolescents who sustained sport-related concussions would report higher peer relationship ratings compared to adolescents who sustained non-sport-related concussions.

Methods

Participants and study design

We conducted a secondary analysis of a larger prospective cohort study. For the purpose of this investigation, we performed a cross-sectional investigation of adolescents who sustained a concussion via sport-related and non-sport-related injury mechanisms. Participants were recruited and enrolled from sports medicine clinics associated with a tertiary care regional children’s hospital in the Denver Metro area. All participants were enrolled and completed assessments between April 2019 and May 2024. For all participants, inclusion criteria included: diagnosis with a concussion by a board-certified sports medicine physician using the most up to date consensus diagnostic criteria at the time of assessment [1,23], being between the ages of 13 to 18 years, completing the initial study evaluation within 3 weeks of concussion, and still being symptomatic at the time of enrollment (Post-Concussion Symptom Inventory [PCSI] rating ≥ 9 [10,24–27]. For all participants, exclusion criteria included: sustaining a concussion during a high-velocity mechanism (motor vehicle crash), neuroimaging abnormality documented if performed as a part of routine clinical care, or sustaining a previous concussion (excluding the index injury) in the 6 months prior to enrollment. Within each group, additional information was collected about the setting of injury (during organized sport participation or not), but no additional details regarding the mechanism of injury were collected. Participants were seen during a single research visit during which they completed all assessments described below. The study was approved prior to commencement by the Colorado Multiple Institutional Review Board. All participants and their parent/legal guardian provided informed consent and assent prior to any data collection procedures occurring.

Assessments and outcome variables

All participants completed a series of brief questionnaires, including the Patient-Reported Outcomes Measurement Information System (PROMIS) Pediatric Global 25 v.1.1 assessment [28] peer relationship [29,30] and anxiety [31] domains, injury characteristic questionnaire, and the PCSI [32]. For the PROMIS peer relationship domain (primary outcome variable) participants were asked to rate their experience in the previous week on four questions pertaining to the frequency of peer relationship elements (including: acceptance by other kids their age, being able to count on their friends, friends helping them out, and other kids wanting to be their friend). Each question was rated as a 0 (never) to 3 points (often). The summary of all four responses were summed and transformed into a T-score where 50 is the population average with a standard deviation of 10 (i.e. T = 45–55 is considered ‘normal,’ T < 45 is considered a poor peer relationship rating, and T > 55 is considered a good peer relationship rating) [29,30]. Therefore, a higher peer relationship T-score indicates a higher perception of peer relationships and support in the week prior to evaluation. Similarly, we assessed the anxiety domain through 4 questions rated on the same scale, answering questions about the past week on elements specific to anxiety (feeling like something awful might happen, feeling nervous, feeling worried, and worrying while at home). As described above, we also converted raw scores in the anxiety domain to T-scores, where a higher score indicated more severe (worse) anxiety symptoms.

Grouping variable

Participants reported whether the concussion occurred during sports participation or not, and this was used as our grouping variable. Consistent with past work [14], we classified concussions that happened as a result of competition, training, or practice as ‘sport-related’ and concussions that happened outside of this context as ‘non-sport-related,’ as self-reported by the participant. Therefore, ‘sport-related’ broadly encompassed any form of organized or casual physical activity when the concussion occurred while the ‘non-sport-related’ group were enrolled following any other mechanism of concussion, such as falls or collisions outside of sports.

Covariates

In addition to the primary grouping and outcome variables, we assessed several potential modifying factors during the post-concussion evaluation. This included the PCSI, a self-reported inventory for participants to report their symptom severity of 22 common concussion symptoms. In this rating, participants identified the severity of each symptom in the prior 24 hours on a scale from (0) ‘None’ to (6) ‘Severe.’ Total symptom severity was quantified as the sum of all 22 symptom ratings, with higher scores indicating more severe symptoms [32]. We also obtained other participant characteristics on a standardized intake form, including age, sex, time of injury, race, ethnicity, past history of concussion and musculoskeletal (MSK) injury, whether loss of consciousness (LOC) occurred during the concussion, sport played at the time of concussion (for the sport-related group), and whether they had begun physical activity prior to the evaluation. These measures were obtained via self-report and parents/guardians were available to assist with these elements if needed. While we collected data on sport- vs. non-sport-related concussion injuries, we did not collect data regarding overall athlete status (athlete vs. non-athlete). Therefore, the interpretation of our analysis is viewed from the perspective of the way in which the injury occurred rather than the athletic identity of participants.

Statistical analysis

Descriptive statistics are presented as mean and standard deviation for continuous variables and number of participants and corresponding percentage for categorical variables. We first compared background and demographic characteristics between the sport-related and non-sport related groups using independent t-tests and chi-square analyses. To address our primary purpose, we compared those who sustained their concussion during sport and non-sport settings on the PROMIS peer relationship T-score using an independent t-test and calculated Cohen’s d effect sizes to measure clinical meaningfulness. This was defined as d > 0.2 = small effect, d > 0.5 = medium effect, and d > 0.8 = large effect [33]. We then performed an adjusted multivariable linear regression analysis, evaluating the association between peer relationship ratings (outcome) and sport-related concussion (predictor, yes vs. no) while adjusting for covariates. We selected covariates to include in the multivariable model based on past work demonstrating an independent association between each factor with concussion-related outcomes [1,19,34–37]. These included anxiety rating (PROMIS T-score), age, sex, history of concussion, history of musculoskeletal injury, loss of consciousness, whether they had begun physical activity after the concussion, and PCSI rating. Collinearity between two variables was detected with a variance inflation factor (VIF) > 2.5 [38,39]. No collinearity was detected in the regression model for the current study (VIF <1.6). Statistical significance was defined a priori as p < 0.05, and all tests were 2-sided. Statistical analysis was performed using Stata Statistical Software: Version 18 (StataCorp, LLC, College Station, TX, USA).

Results

We enrolled and assessed 123 adolescents who sustained a concussion within the three weeks prior to evaluation (mean = 9 days after concussion, range = 2–21 days). Of those, the majority (N = 96, 78%) occurred during sports. The two groups (sport-related and non-sport-related) demonstrated similar background, demographic, medical history, and injury-related characteristics (Table 1). Univariable analysis indicated that participants who sustained a sport-related concussion reported significantly higher (better) peer relationship ratings compared to those who sustained a non-sport-related concussion (Figure 1; mean = 52.7, SD = 7.7 for sport-related group vs. mean = 47.7, SD = 10.3; p = 0.009; Cohen’s d = 0.60, medium effect). After covariate adjustment, those who sustained a sport-related concussion had a significantly greater peer relationship rating than those who sustained a non-sport-related concussion (Table 2). On average, those who sustained a concussion during sports reported an approximately 4-point higher peer relationship rating than those who did not. Additionally, our multivariable model indicated that higher peer relationship ratings were associated with lower anxiety ratings following concussion (Table 2). Additionally, there was a significant albeit weak correlation between higher peer relationship ratings and lower anxiety ratings (Figure 2; Pearson r= −0.24; p = 0.011).

Table 1.

Participant characteristic comparisons between groups. Data are presented as the mean value (standard deviation) for continuous variables, and the number of participants who reported ‘yes’ to each categorical variable (and corresponding percentage).

Variable		Sport-Related Injury (N = 96)	Non-Sport-Related Injury (N = 27)	P value	Cohen’s d
Age (years)		15.4 (1.6)	15.4 (2.0)	0.95	0.01
Sex (female)		53 (55%)	14 (52%)	0.83	–
Initial visit time (days since concussion)		9.3 (3.9)	8.9 (4.3)	0.65	0.10
Race	American Indian or Alaska Native	1 (1%)	0 (0%)	0.27	–
	Asian	1 (1%)	1 (4%)
	Black or African American	8 (8%)	1 (4%)
	Native Hawaiian or Pacific Islander	0 (0%)	1 (4%)
	White	75 (78%)	19 (70%)
	More than one race	10 (10%)	4 (15%)
	Unknown/Not reported	1 (1%)	1 (4%)
Ethnicity (Hispanic/Latino/a/x)		72 (75%)	18 (67%)	0.44
Concussion history		41 (43%)	14 (52%)	0.40
MSK injury history		63 (67%)	17 (63%)	0.81
LOC at time of injury		21 (22%)	8 (30%)	0.40
Began physical activity prior to initial assessment		31 (32%)	10 (37%)	0.64
Initial symptom severity (PCSI rating)		44.9 (28.2)	41.8 (24.7)	0.60	0.11
Anxiety symptom severity (PROMIS T-score)		46.9 (11.2)	50.6 (11.0)	0.15	0.33
Sport played during concussion event		Soccer: 37 (39%) Football: 23 (24%) Basketball: 7 (7%) Cheerleading: 7 (7%) Volleyball: 6 (6%) Lacrosse: 4 (4%) Wrestling: 3 (3%) Baseball: 2 (2%) Ice Hockey: 2 (2%) Karate: 2 (2%) Skiing: 2 (2%) Ultimate Frisbee: 1 (1%)	–	–

Figure 1.

Box and whisker plot describing the peer relationship ratings between the two groups. Higher peer relationship ratings (Y axis) indicate better perception of peer relationships. Findings indicate that the group who sustained sport-related concussion reported significantly better post-concussion peer relationship ratings compared to the group who sustained a non-sport-related concussion (52.7 ± 7.7 vs. 47.7 ± 10.4; p = 0.009, Cohen’s d = 0.60).

Table 2.

Multivariable linear regression model results evaluating the association between type of injury (sport-related vs. non-sport related) and peer relationship ratings (outcome) after concussion, adjusting for covariates. Unstandardized and standardized beta coefficients and corresponding 95% CIs are presented for each predictor variable.

Model predictors	Unstandardized Coefficients		Standardized Coefficients		P value
	β coefficient	95% confidence interval	β coefficient	95% confidence interval
Sport-related injury	4.19	0.39, 8.00	0.20	0.02, 0.38	0.03*
Anxiety rating	−0.21	−0.38, −0.04	−0.25	−0.45, −0.05	0.02*
Age	−0.47	−1.45, 0.51	−0.09	−0.27, 0.09	0.34
Sex	−1.07	−2.72, 0.59	−0.12	−0.31, 0.07	0.21
History of concussion	10.5	−2.35, 4.44	0.06	−0.13, 0.25	0.54
History of MSK injury	0.59	−2.91, 4.09	0.03	−0.16, 0.22	0.74
Loss of consciousness	−3.25	−7.12, 0.61	−0.16	−0.34, 0.03	0.10
Began physical activity before assessment	0.40	−2.99, 3.79	0.02	−0.16, 0.20	0.82
PCSI Symptom rating	0.04	−0.04, 0.11	0.12	−0.11, 0.36	0.31

^*p < 0.05.

Figure 2.

Scatterplot and line of best fit describing the univariable relationship between peer relationship ratings and anxiety ratings. Higher peer relationship ratings indicate better perception of peer relationships, and higher anxiety ratings indicate more severe anxiety at the time of assessment. There was a statistically significant albeit weak correlation between these two factors (Pearson r= −0.24; p = 0.011). Red dots indicate non-sport related injuries, and blue dots indicate sport-related injuries.

Discussion

The results of our investigation indicate that adolescents with sport-related concussions endorsed higher peer relationships ratings than those with concussions sustained outside of sport-related settings. We also found that higher peer relationship ratings were associated with lower anxiety ratings among all participants (regardless of sport-related injury status) following concussion. While the mean ratings for both groups were within the expected range for adolescents, the group difference observed indicates a potential benefit of environmental factors, such as teammates, coaches, or other supportive people who may be able to provide helpful during recovery following adolescent concussion.

Past studies have considered the role that mechanism of injury plays on injury-related outcomes [40], rather than the support system surrounding the patient. Specifically, while concussions sustained during sport-related activities remain the most common injury mechanism for adolescents, a substantial number of individuals may sustain a concussion outside of this setting [40]. As a result, it is important to consider potential differences in the recognition, diagnosis, referral, and overall support system available between those who sustain a concussion during organized sport participation, where medical personnel, coaches, and other supportive individuals may be present, compared to those outside of this setting. These different types of concussion occurrence settings may contribute to worsened outcomes. While participation in sports can lead to an increased risk of sustaining a concussion relative to other everyday activities [41], the support system for injured adolescents may be more readily available for those injured during sports than those not injured during sports. Given past findings that non-sport-related concussions in a sample of collegiate athletes are associated with longer recovery and more severe symptoms than sport-related concussions [42], our work provides insight into potential social mechanisms in which recovery is supported via better perception of peer relationships among those injured during sports. One recent study observed that peer support through a one-on-one buddy program may be helpful to navigate the complexity of concussion recovery [43]. In addition, sports can mitigate the impact of perceived stress within the context of academics [44], and this may translate into potential stress or fears related to concussion recovery often experienced by adolescent athletes [45]. The manifestation of various symptoms, the manner which adolescents interact with everyday life events, and the underlying physiology related to both short- and long-term outcomes has not yet been fully explored [46,47]. Future investigations are indicated to evaluate the interplay between concussion mechanism, peer relationships, and recovery.

A major aspect of quality of life is mental health, and mental health symptoms (e.g. symptoms of anxiety or depression) are commonly reported following adolescent concussion [48]. Among the various mental health symptoms investigated in past research, anxiety symptoms are common particularly for adolescents with concussion [34]. Our study results indicated that better peer relationship ratings were significantly associated with lower anxiety severity ratings for adolescents with concussion. However, our study was cross-sectional, and causality cannot be determined from this type of design. Assessing anxiety and peer and/or familial support during a multifaceted concussion evaluation may be an important clinical consideration. If anxiety symptoms, one of the main mental health concerns following concussion, can be alleviated through positive peer relationships or other supportive factors, then post-injury care may be improved. Peer relationships differed between sport-related and non-sport-related injury groups. Adolescents who were participating in sports during their concussion, on average, reported higher peer relationship ratings, which may contribute to lower anxiety ratings, although we did not observe a significant difference in anxiety ratings between sport-related and non-sport-related injury groups. However, our study highlights the potential importance of peer relationships after adolescent concussion, which may be especially relevant for those who sustain a non-sport-related injury.

Including the concepts and enhanced understanding of the role that peer relationships can play in concussion recovery may allow for the creation of more efficient treatment methods, particularly focused on the sociocultural elements that likely influence recovery outcomes [16]. Specifically, clinicians may provide better guidance to patients by seeking a thorough understanding of the circumstances that each individual patient has in their everyday life, including peer support, to help personalize treatment. As a result, our findings may be helpful in encouraging patients to engage with peer support as a method of facilitating recovery in adolescents. Additional considerations should be accounted for that were not evaluated in this study as well, such as the peer interaction differences within different school ages (e.g. middle school vs. high school) and the level of competition within sports (e.g. varsity vs. junior varsity teams). Future studies should consider how to support those who report lower peer relationships at initial clinical presentation in future research.

Limitations

Our study had several limitations, and our results should be interpreted with these in mind. Our study was cross-sectional in design, and as such, we cannot infer causality but rather only associations that were or were not present according to our analyses. In addition, our data was collected within a small geographic region in the greater Denver, Colorado metro area and our sample sizes were discrepant between our non-sport and sport-related concussion groups. As such, the unique characteristics of this region and participant demographics, such as the relatively homogenous characteristics of most participants, may have influenced the way in which participants reported peer relationship ratings. We also did not account for familial connections or relationships, which may also be associated with anxiety or other mental health outcomes. While we recorded the injury mechanism (sport- vs non-sport-related), we did not collect information regarding athletic participation or identity for the study participants, which may be another important factor to consider in evaluating peer relationships. Furthermore, we did not obtain data or any other information about the specific injury details, injury setting, or mechanism of injury beyond sport or non-sport contexts and the clinical characteristics described in Table 1. Inherent support structures for healthcare navigation following concussion may be discrepant between those who were injured during sports and those who were not (e.g. day of injury recognition, guidance, or referral). This discrepancy was not evaluated in this investigation but may have influenced the outcomes observed. In addition, based on our sample, we are not able to identify the potential impact of team vs individual sports participation on peer relationships for adolescents with concussion.

Conclusion

The data in our investigation indicate that adolescents who sustained a concussion during sports reported significantly higher peer relationship ratings than those who sustained a concussion in non-sport settings. In addition, better peer relationship ratings were associated with less severe anxiety symptoms after concussion. Collectively, these findings suggest the potential effects of contextual factors, such as peer support via teammates, coaches, or other supportive people, to facilitate improved outcomes following adolescent concussion. While pre-injury anxiety was not evaluated within this study, it is an important consideration for clinical guidance and identification of appropriate recovery strategies following adolescent concussion.