Impact of Sex and Pubertal Development on Anxiety in Adolescents After Concussion

Abstract

Concussion often results in psychological symptoms, including anxiety. Post-concussion anxiety has been well documented, although much of this research has focused on collegiate athletes. The purpose of this study was to compare (1) anxiety symptoms in concussed and healthy controls over time and (2) to explore sex differences in post-concussion anxiety within the context of pubertal development. Participants (N = 126, mean age = 15.1 years old), including concussed (n = 86) and healthy adolescents (n = 40), completed the Pubertal Development Scale (PDS) and the Screen for Child Anxiety and Related Disorders (SCARED-C). The concussed groups completed SCARED-C at three visits (<10 days, 4 weeks, 3 months). Results of an analysis of covariance (ANCOVA) and multi-variate analysis of covariance (MANCOVA) found concussed adolescents reported higher SCARED-C total, generalized, and panic anxiety scores than healthy controls, after controlling for sex, age, and PDS score (PDSS). A three-way mixed ANCOVA examined the effects of sex, PDSS, time, and their interaction on SCARED-C total score in concussed adolescents while controlling for age. There was a significant three-way interaction between sex, age, and PDSS on SCARED-C total score while controlling for age. Overall, we observed increased anxiety in concussed adolescents, compared with controls, as well as greater post-concussion anxiety reported by females compared with males, including within PDSS groups. Concussion providers should be prepared to receive training to administer well-validated measures of psychopathology and should consider that female adolescents, compared with males, regardless of pubertal development, may be at greater risk for post-concussion anxiety.

Introduction

Concussion is a major public health concern that affects approximately 33 million children and adolescents each year worldwide.¹ These injuries are heterogenous and may involve a myriad of symptoms and impairments that reflect different subtypes, including anxiety/mood, cognitive, headache/migraine, ocular, and vestibular.^2,3 Adolescents may be at particular risk for protracted concussion recovery because of ongoing brain development, including incomplete myelination, as well as increased acceleration-deceleration forces from reduced neck strength.^4
–6 Consequently, they may take on average 30 days to recover compared with young adults who often recover in 7–10 days.^7
–9

Researchers have demonstrated that concussion often results in psychological symptoms including depression, anxiety, and irritability.^10,11 Researchers have reported higher rates of post-concussion anxiety compared with post-concussion depression.¹² A variety of explanations have been posited for how concussion has been associated with anxiety, including overaction of the behavioral inhibition system, the physiological mechanism responsible for individuals' responses to situations that may result in consequence or punishment.^13,14

Hypervigilance and rumination of symptoms^10,15,16 and a potential for a neurological predisposition to fear-conditioning after concussion may result in increased anxiety. In addition, post-concussion vestibular symptoms often overlap with symptoms of anxiety. Interestingly, past research has shown no differences in anxiety within one week after injury between patients with concussion and those with orthopedic injuries.¹³ After a couple of weeks to months, the literature is inconsistent, with some studies showing elevated anxiety and others showing no differences in anxiety from controls.^17,18 For example, a study of collegiate football players found that in comparison with healthy controls, concussed individuals reported higher anxiety scores at three days and one week post-injury, but not at one month.¹⁹

Sex may play a role in anxiety after concussion. Higher rates of anxiety disorders are diagnosed in females; they demonstrate higher levels of anxiety symptoms after a concussion compared with males.^16,20
–22 Females are also at greater risk for prolonged recovery after concussion, and this pattern is more pronounced in adolescents.^23,24 There are potential physiological and neuroanatomical explanations for these findings, including differences in neck strength,²⁴ as well as sex differences in brain structure and functioning.^25,26 Overall, females report more severe concussion symptoms in general than males,^27,28 which may also account for the higher anxiety symptoms.

Adolescence is a sensitive developmental period often associated with anxiety, especially for females. Females report higher levels of generalized anxiety, social anxiety, and panic disorder than males.^29,30 Overall, females are more likely than males to report elevated levels of anxiety from early childhood through adolescence.³¹ Research has identified that pubertal development predicts mood and anxiety symptoms, for females, over and above chronological age.^32
–34

A study identified neural correlates of post-concussion anxiety in adolescents, with adolescents with more advanced pubertal maturation displaying white matter abnormalities, as measured by lower neurite density index (NDI), in key white matter tracts for emotional regulation.³⁴ Although it is well established that puberty is associated with anxiety in some adolescents, researchers have yet to examine the interaction between sex and puberty on anxiety in the early stages of concussion, or later.

There have also been many weaknesses in previous studies examining post-concussion anxiety, including the lack of a control group. In a systematic review of mental health outcomes after concussion, predominantly examining depression and anxiety, researchers reported that only two of seven studies included a control group.³⁵ Further, studies have often relied on measures such as the Beck Anxiety Inventory, which provides a total score, but lacks nuance regarding specific anxiety subtypes.^19,36,37

Many studies have focused on collegiate athletes and adults,¹² rather than adolescents, and studies have often been conducted across various post-injury time intervals, making synthesis of the data challenging.^35,38 In the studies comparing demographics, higher symptoms of anxiety are reported among females and those of pre-adolescent age.^20,39

The primary purpose of this study was to compare anxiety symptoms between concussed and healthy adolescents. A secondary purpose was to examine the longitudinal nature of post-concussion anxiety in adolescents to determine if post-concussion anxiety symptoms are transient in nature or if they persist for months after injury. Within this secondary purpose, we also sought to examine sex differences in post-concussion anxiety symptoms in the context of pubertal development.

We expected that concussed adolescents would report higher anxiety symptoms than healthy controls and, specifically, that concussed adolescents would report higher levels of panic and school avoidance, as concussion symptom triggers (e.g., demanding environments such as school) often produce anxiety symptoms that overlap with and are qualitatively similar to certain concussion impairments (e.g., panic associated with vestibular symptomology). We also expected that females would report higher anxiety symptoms after concussion than males and anticipated that females with advanced pubertal maturation would report higher anxiety after concussion compared with their male counterparts.

Methods

Design

This study was approved by the University of Pittsburgh Institutional Review Board. An initial sample of 86 concussed adolescents and 40 healthy controls was recruited through a pre-existing longitudinal study, the Investigating Concussion in Adolescents at Risk for Emotional Dysregulation (iCare) study (R01MH11488101; PIs: Versace, Kontos).

Participants

There were 126 eligible participants enrolled in the study. Ten concussed participants had incomplete data at the first time point. Seven healthy controls were excluded because of missing data. The concussed group consisted of adolescents aged 12.8–17.4 years with a formally diagnosed concussion, as per current consensus guidelines,⁴⁰ within six days of injury. Concussed adolescents were age-and-gender matched to controls, with no history of concussion. Exclusion criteria for all participants included neurodevelopmental, neurological, or systematic disease, significant orthopedic injury within the past month, medical history of major psychiatric disorder (e.g., schizophrenia, major depressive disorder, delusional disorder), and ongoing illicit substance abuse or dependence (within the past three months).

Measures

SCARED

Participants completed the Screen for Child Anxiety and Related Disorders, Child Version (SCARED-C) to quantify anxiety symptoms. Concussed participants completed SCARED-C at three post-injury time points: visit 1 (≤10 days [≤10D]), visit 2 (Four ±1 weeks [4W]), and at visit 3 (Three ±1 months [3M]). Controls only completed SCARED-C at visit 1. The SCARED-C consists of 41 items assessing anxiety symptoms on a 3-point Likert scale (0 = Not True or Hardly Ever True, 1 = Somewhat True or Sometimes True, 2 = Very True or Often True). Item responses are summed to generate a total anxiety score, as well as five subscales: generalized anxiety disorder, separation anxiety disorder, social anxiety disorder, significant school avoidance, and panic disorder symptoms. The SCARED has demonstrated adequate psychometric properties, including strong internal consistency (α = 0.90), as well as good convergent validity with other well-validated self-report measures of anxiety.^41
–43

PDS

The Pubertal Development Scale was completed by controls and concussed participants at visit 1 (<10D post-injury for concussed participants) only.⁴⁴ Overall, the PDS has been found to be a reliable measure, with high correlations with adolescent pubertal stage and hormone level.^44
–46 The PDS was well correlated with a picture-based interview about puberty (0.59–0.70), as well as physical examination characteristics (0.63–0.68).⁴⁵ While PDS scores range from 1–4, we converted the scale for Tanner-comparable staging.⁴⁵ A total PDS score (PDSS) was generated, aggregating scores based on gonadarche and adrenarche-related traits. Converted scores range from 1–5, with higher scores indicating advanced pubertal development. The PDSS was included as a continuous variable in statistical analyses.

Procedures

Participants with a suspected concussion were evaluated by trained medical professionals (e.g., neuropsychologists, physicians) at the University of Pittsburgh Concussion Clinic. All participants with a confirmed diagnosis of concussion were offered to participate in the study. Age- and sex-matched controls were recruited from the community through a local recruitment website (https://pittplusme.org). All participants, including parents and guardians when indicated, provided informed assent and consent to participate in this study.

Statistical analyses

A post hoc power analysis was conducted for the analyses comparing concussions and controls, using G*Power software, version 3.1.⁴⁷ To achieve 80% power in a one-way analysis of covariance (ANCOVA) model at 0.05 significance, with an estimated large effect size of 0.40,⁴⁸ a total of 80 participants is needed. The large effect size used in the power analysis was based on existing evidence of differences in anxiety between concussions and controls.⁴⁹

Descriptive statistics (means/standard deviations or count/frequencies) were calculated for the concussed (n = 86) and non-concussed cohorts (n = 40). Age, sex, pre-injury anxiety history, and PDSS were compared between groups with independent samples t tests or chi-square tests.

A one-way ANCOVA was conducted to assess between-group differences in SCARED-C total score for non-concussed controls (n = 33) and concussed adolescents (n = 76) at visit 1, controlling for age, sex, and PDSS. The interactions of all variables included in the models were assessed, and when the assumption of homogeneity of regressions slopes was met (p > 0.05), the interactions were removed from the model. All other assumptions of ANCOVA were evaluated and met. Post hoc analysis of adjusted means was performed with a Bonferroni confidence interval adjustment.

A one-way multi-variate analysis of covariance (MANCOVA) was also conducted to examine between-group differences in the SCARED-C subscales (SCARED-C panic, SCARED-C generalized anxiety, SCARED-C separation anxiety, SCARED-C social anxiety, SCARED-C school avoidance) at visit 1, controlling for age, sex, and PDSS. The interactions of all variables included in the models were assessed, and when the assumption of homogeneity of regressions slopes was met (p > 0.05), the interactions were removed from the model. All other assumptions of MANCOVA were evaluated and met.

Follow-up univariate one-way ANCOVAs were performed to determine between-group differences for each SCARED-C subscale. A Bonferroni adjustment was made such that statistical significance was accepted when p < 0.01 (i.e., 0.05/5 = 0.01; 5 SCARED-C subscale dependent variables). Statistically significant one-way ANCOVAs were followed up with pairwise comparisons with a Bonferroni adjustment.

For concussed adolescents with data at all three time points (n = 56), a three-way mixed ANCOVA was performed to understand the effects of sex, PDSS, time, and the interaction between these independent variables on the SCARED-C total score while controlling for age. Between-subjects factors were sex (male, female) and PDSS. For the PDSS factor, participants were grouped into those who scored <4 and >4 on the PDSS to reflect adolescents who were still developing versus those who had completed development, respectively. The within-subjects factor was time at visit 1 (≤10D), visit 2 (4W), and visit 3 (3M).

For a significant three-way interaction, univariate general linear models were performed to calculate simple two-way interactions of sex and PDSS at each time point. A Bonferroni adjustment was made such that statistical significance was accepted when p < 0.0167 (i.e., 0.05/3 = 0.0167; two-way interaction at each time point). For main effects, follow-up pairwise comparisons were performed with the Bonferroni procedure to control family-wise type I errors. For all analyses, statistical significance was set at p < 0.05. Estimations of effect size for mean differences were reported as partial eta-squared (partial η²) values and interpreted as small (0.01–0.08), medium (0.09–0.24), and large (>0.25).⁵⁰ All statistical analyses were performed using SPSS software (version 28.0; SPSS, Inc, Chicago, IL).

Results

Descriptive data

Descriptive statistics for the concussed sample are provided in Table 1. Descriptive statistics for concussed males and females are provided in Table 2. There were no statistically significant differences between adolescents with a concussion and healthy controls in age (15.1 ± 2.3 years vs. 15.2 ± 1.5 years, p = 0.64), sex (55.8% male vs. 57.5% male, p = 0.96), pre-injury anxiety history (11.8% vs. 7.5%, p = 0.4) or PDSS score (3.9 ± 1.1 vs. 3.9 ± 1.2, p = 0.94).

Table 1.

Descriptive Statistics for Continuous and Categorical Variables for Concussed Adolescents

	Concussed (N = 76)
Mean Age ± SD	15.1 ± 2.3
Female Sex (%)	37 (43)
Mean PDSS ± SD	3.90 ± 1.0
Mechanism
Sport (%)	55 (77.5)
Non-sport (%)	16 (22.5)
Days to First Visit (<10D)	3.5 ± 1.7
Days to Second Visit (4W)	27.0 ± 5.4
Days to Third Visit (3M)	98.3 ± 13.3
Days to Recovery	43.4 ± 37.4
ADHD/LD History^* (%)	6 (7.9)
Migraine History^* (%)	21 (27.6)
Motion Sickness History^* (%)	15 (19.7)
Ocular History^* (%)	4 (5.3)
Anxiety History^* (%)	9 (11.8)
Depression History^* (%)	1 (1.4)
Other Psychiatric History^* (%)	9 (11.8)
Previous Concussions^*(%)	18 (23.7)

SD, standard deviation; PDSS, Pubertal Development Scale Score; ADHD, attention-deficit hyperactivity disorder; LD, learning disorder.

2 levels: yes/no

Table 2.

Descriptive Statistics for Concussed Male and Female Adolescents

	Males (n = 43)	Females (n = 33)	p
Mean Age ± SD	14.9 ± 2.8	15.1 ± 1.5	0.96
Mean PDSS ± SD	3.67 ± 1.1	4.2 ± 0.83	0.09
Days to Visit 1 (<10 days)	3.4 ± 2.0	3.4 ± 1.2	0.56
Days to Visit 2 (4 weeks)	27.0 ± 5.7	27 ± 5.1	0.88
Days to Visit 3 (3 months)	100.5 ± 16.2	95.5 ± 7.9	0.23
Days to Recovery	33.5 ± 28.3	56.3 ± 44.2	0.08

SD, standard deviation; PDSS, Pubertal Development Scale Score.

Significant at p < 0.05.

The concussed sample had predominantly sport-related concussions (77.5%) and was 43% female. The overall sample's PDSS had a negatively skewed distribution, indicating the sample had a higher number of adolescents with advanced pubertal development. At visit 1, full data were available for 88.4% of concussed (n = 76) and 82.5% of non-concussed controls (n = 33). Control data were only collected at visit 1, and those values were used for comparison against concussed. Seven non-concussed controls were lost to missing data, because this study was conducted during the COVID pandemic. At visit 2, full data were available for 79.1% of concussed (n = 68). At visit 3, full data were available for 66.3% of concussed (n = 57).

Anxiety differences between concussed adolescents and non-concussed controls

SCARED-C total scores for male and female participants in the healthy control group at visit 1 and male and female participants in the concussed adolescent group at each visit (<10D, 4W, 3M) are provided in Figure 1. Results of the one-way ANCOVA revealed there was a statistically significant difference in the SCARED-C total score between concussed adolescents and healthy controls, F(1, 104) = 6.636, p = 0.011, partial η² = 0.1. Concussed adolescents had significantly greater total anxiety (Adj. mean = 13.3 ± 1.2) compared with healthy controls (Adj. mean = 7.81 ± 1.8; p = 0.011) after adjusting for age, sex, and PDSS (Table 3).

FIG. 1.

SCARED-C total score for male/female healthy controls at visit 1 and male/female concussed adolescents at each visit (≤10D, 4W, and 3M). SCARED-C, Screen for Child Anxiety Related Emotional Disorders.

Table 3.

Results for Between-Subjects Comparisons of Concussed Adolescents and Healthy Controls at Visit 1 for Screen for Child Anxiety Related Emotional Disorders-C Outcomes When Adjusting for Age, Sex, and Pubertal Development Scale Score

	Concussed (n = 76)		Controls (n = 33)
ANCOVA	M (SD)	M_adj (SE)	M (SD)	M_adj (SE)	p
SCARED-C Total	13.26 (11.33)	13.29 (1.17)	7.88 (7.55)	7.8 (1.78)	0.011^a

MANCOVA	M (SD)	M_adj (SE)	M (SD)	M_adj (SE)	p
SCARED-C Panic	2.21 (2.4)	2.21 (0.24)	1.00 (1.15)	1.02 (0.37)	0.007^b
SCARED-C Generalized	4.41 (3.96)	4.42 (0.40)	2.42 (2.39)	2.39 (0.61)	0.007^b
SCARED-C Separation	1.76 (2.21)	1.77 (0.22)	0.85 (1.35)	0.84 (0.34)	0.025
SCARED-C Social	3.63 (3.75)	3.64 (0.41)	2.94 (3.18)	2.92 (0.63)	0.345
SCARED-C School Avoidance	1.25 (1.48)	1.26 (0.17)	0.67 (1.65)	0.65 (0.26)	0.054

ANCOVA, analysis of covariance; M, mean; M_adj, adjusted mean for age; SE, standard error; SD, standard deviation; MANCOVA, multi-variate analysis of covariance; SCARED-C, Screen for Child Anxiety Related Emotional Disorders.

Significant at p < 0.05.

Significant difference between adjusted means based on Bonferroni adjustment (p < 0.01).

Results of the one-way MANCOVA revealed a statistically significant difference between concussed adolescents and controls on the combined SCARED-C subscales after controlling for age, sex, and PDSS, F(5, 100) = 2.54, p = 0.033, Wilks' Λ = 0.887, partial η² = 0.113. There were statistically significant differences in adjusted means between groups for SCARED-C panic (F(1, 104) = 7.57, p < 0.007, partial η² = 0.1) and SCARED-C generalized anxiety (F(1, 104) = 7.59, p < 0.007, partial η² = 0.1), but not for the SCARED-C separation anxiety, SCARED-C social anxiety, or SCARED-C school avoidance subscales. Concussed adolescents had significantly greater panic symptoms (Adj. mean = 2.21 ± 0.24) than healthy controls (Adj. mean = 1.002 ± 0.37; p = 0.007) and significantly greater generalized anxiety symptoms (Adj. mean = 4.42 ± 0.403) than healthy controls (Adj. mean = 2.39 ± 0.61; p = 0.007) after adjusting for age, sex, and PDSS (Table 3).

Anxiety in concussed adolescents over time

Results for the three-way mixed effects ANCOVA revealed a statistically significant interaction between time, sex, and PDSS on SCARED-C total score while controlling for age, F(2, 102) = 4.29, p = 0.016, partial η² = 0.1 (Fig. 2). There was no statistically significant within-subjects main effect of time, F(2, 102) = 0.332, p = 0.718, partial η² = 0.01. There was a statistically significant between-subjects main effect of sex, F(1, 51) = 10.194, p = 0.002, partial η² = 0.17, but there was no significant between-subjects main effect of PDSS, F(1, 51) = 0.01, p = 0.93, or age, F(1, 51) = 0.01, p = 0.95. There were no statistically significant simple two-way interactions between sex and PDSS on SCARED-C total score at visit 1, F(1, 71) = 0.31, p = 0.579, visit 2, F(1, 63) = 3.815, p = 0.055, or visit 3, F(1, 52) = 0.525, p = 0.472, while controlling for age.

FIG. 2.

Mean ± standard error SCARED-C total scores across visit 1 (<10D), visit 2 (4W), and visit 3 (3M) based on sex and (A) PDSS <4 versus (B) >4 for concussed adolescents. Overall, a significant interaction between time, sex, and PDSS on SCARED-C total score while controlling for age, F(2, 102) = 4.29, p = 0.016, partial η² = 0.1, was found. SCARED-C, Screen for Child Anxiety Related Emotional Disorders; PDSS, Pubertal Development Scale Score.

Pairwise comparisons for SCARED-C total score based on sex, PDSS, and time, while controlling for age, are reported in Table 4. Females had higher SCARED-C total scores across all three time points compared with males (≤10D: p = 0.02, 4W: p = 0.005, 3M: p = 0.002), while PDSS groups had similar scores at all three time points (≤10D: p = 0.901, 4W: p = 0.710, 3M: p = 0.988).

Table 4.

Screen for Child Anxiety Related Emotional Disorders Total Scores of Concussed Adolescents (n = 56) Over Time Based on Sex and Pubertal Development Scale Score Group While Controlling for Age

Factor	Visit 1 (<10D)	Visit 2 (4W)	Visit 3 (3M)
Sex
Female	15.72 (2.16)^a	16.86 (2.01)^a	13.76 (1.88)^a
Male	8.70 (1.89)^a	9.01 (1.76)^a	5.42 (1.65)^a
PDSS
<4	11.96 (2.49)	13.61 (2.32)	9.62 (2.17)
<4	12.45 (2.3)	12.26 (2.14)	9.57 (2.00)
Sex x PDSS
Male and <4	9.47 (2.49)	7.99 (2.32)^c	6.37 (2.17)
Male and >4	7.93 (3.23)^b	10.03 (3.01)	4.47 (2.81)^b
Female and <4	14.46 (3.90)	19.22 (3.64)^c	12.86 (3.39)
Female and >4	16.98 (2.78)^b	14.49 (2.59)	14.66 (2.42)^b

PDSS, Pubertal Development Scale Score; SCARED-C, Screen for Child Anxiety Related Emotional Disorders.

Data are presented as adjusted mean (standard error).

Females had higher scores at all three visits compared with males (≤10D: p = 0.02, 4W: p = 0.01, 3M: p < 0.01).

Females in the >4 PDSS group had higher SCARED-C total scores compared with males in the >4 group at visit 1 (p = 0.02) and at visit 3 (p < 0.01).

Females in the <4 PDSS group had higher SCARED-C total scores compared with males in the <4 group at visit 2 (p < 0.01).

Females in both PDSS groups had higher SCARED-C total scores compared with males in those groups (<4: p = 0.038, >4: p = 0.019). Females in the <4 group, however, had similar scores compared with females in the >4 group (p = 0.974), and males in the <4 group had similar scores compared with males in the >4 group (p = 0.898). Females in the >4 PDSS group had higher SCARED-C total scores compared with males in the >4 group at visit 1 (p = 0.024) and at visit 3 (p = 0.004) but not visit 2 (p = 0.225). Females in the <4 PDSS grouphad higher SCARED-C total scores compared with males in the <4 group at visit 2 (p = 0.007) but not at visit 1 (p = 0.246) or visit 3 (p = 0.086).

Discussion

The current study is the first to identify sex differences in post-concussion anxiety symptoms across multiple post-concussion time points, within the context of pubertal development. Overall, the findings indicated differences in anxiety symptoms between concussed and controls, supporting our primary hypothesis. Total anxiety, generalized anxiety, and panic anxiety were all higher in adolescents after concussion than in healthy controls. Within the concussed sample, the findings also supported our hypothesis for sex differences in anxiety symptoms, with females reporting higher anxiety symptoms compared with males. When considering pubertal development, females who had completed development (PDSS >4) and females who were still developing (PDSS <4) reported higher total anxiety symptoms across all three study time points compared with males at the same pubertal development level.

As expected, higher levels of overall anxiety were endorsed among our concussed sample compared with controls. This is consistent with the empirical literature demonstrating transient self-reported post-concussion emotional changes.^51
–53 There may be underlying physiological mechanisms to explain the onset of emotional symptoms post-concussion. Specifically, the neurometabolic dysfunction evidenced with concussion often mimics the neurometabolic changes displayed on neuroimaging in individuals with psychopathology, including anxiety.⁴ This pattern of overlap in the pathophysiology between concussion and psychopathology was also demonstrated in reference to neurological alterations in limbic-frontal circuitry, in addition to fluctuations in neurotransmitter level, specifically in serotonin and dopamine.^53

-56

Interestingly, only the generalized and panic anxiety subdomains of the SCARED-C were higher in the concussed sample compared with controls. There were no group differences for separation, social anxiety, or school avoidance, compared with the non-concussed sample. Considering that vestibular symptoms of concussion often overlap with panic symptoms and social anxiety, and occur in school settings, these findings were partially surprising.^53,57 Provocation of overall concussion symptoms in the environments associated with these subdomains of anxiety is common.

Currently, individuals with concussion are often being treated with a more active, exposure-based approach, including completing social exposures and their normal-day-to-day activities.⁵⁸ This contemporary concussion treatment strategy may promote an adaptive cognitive framework, interpreting symptom exacerbation as non-threatening, and may ultimately explain why adolescents endorsed lower levels of separation, social anxiety, and school avoidance.

The breakdown of survey items may provide some insight into why panic and generalized anxiety scores were elevated within the concussed group. The SCARED items encompassing the generalized anxiety subdomain often examine adolescents' feelings of worry related to ability and interpersonal acceptance (e.g., “I worry about being as good as other kids”). These sources of positive reinforcement are often disrupted by concussion and consequent removal from sports, where adolescents may find themselves worrying about their athletic performance and place within their peer groups. During adolescence, a developmental time where peer relationships begin taking precedence over parent-child relationships,⁵⁹ adolescents' inability to be connected to peers through athletics and other extracurricular activities may result in increased feelings of closeness to ones' immediate family.

The SCARED items encompassing the panic subdomain often assess feelings of anxiety related to somatic symptoms (“When I get frightened, I feel dizzy”), providing a potential explanation for why panic was elevated in the concussed group. Dizziness is present in approximately 50–84% after concussion^39,60 and may persist over weeks to months.⁶¹ These findings help to identify the potential types of anxiety that might be implicated during concussion recovery, which can then be targeted in throughout the course of recovery.

Our findings also support the hypothesis that females would endorse higher anxiety symptoms than males post-concussion, and this relationship persisted across all time points. Specifically, when taking sex and pubertal development into consideration together, female adolescents, compared with males, reported higher total symptoms across all three visits. These findings are reflective of previous research identifying that post-concussion vestibular dysfunction is more common in females, and this vestibular dysfunction is often associated with anxiety symptoms.^24,53,62

These findings also align with research on healthy adolescents,⁶³ which found that adolescent females endorsed higher mean levels of generalized anxiety, social anxiety, and panic disorder compared with adolescent males. Taken together, this pattern may suggest one reason why females present with a higher post-concussion symptom burden, both acutely and in the chronic stage of recovery, compared with males.¹⁴

Although females had higher anxiety than males in both pubertal groups, one potential explanation for the sex differences pertains to sex-specific hormonal changes with pubertal development, including fluctuations in adrenal hormones, growth hormone, and luteinizing hormone (LH). These hormonal differences set the stage for elevated levels of anxiety, especially in concert with increased psychosocial stressors, for example, elevated academic demands and increased social pressure for inclusion and high social functioning.^64,65

Clinical implications

This study highlights the potential clinical utility of administering a well-validated anxiety inventory with subdomains of anxiety. Understanding which subdomain of anxiety is elevated can help the clinician refine the treatment plan for those with post-concussion anxiety. For example, panic symptoms may be best addressed through exposures,⁶⁶ while generalized anxiety symptoms may respond to cognitive restructuring techniques.^67,68

This study also reinforces the notion that adolescent females are at risk for acute and chronic anxiety post-concussion. Clinicians should be aware of these sex discrepancies in outcomes and should inform their individualized treatment plan accordingly. For example, more proactive treatment for anxiety post-concussion may be indicated for adolescent females. The study also highlights the importance of clinicians administering validated measures of anxiety through all stages of recovery, because this will allow clinicians to quantify the magnitude of patients' post-concussion anxiety and appropriately respond. Treatment for post-concussion anxiety may include psychoeducation concerning the typical course of recovery from concussion, guided exercise, and referral to a mental health specialist.

Exercise is particularly salient given it is a treatment approach that can address multiple symptoms and impairments associated with concussion, in addition to anxiety. Subthreshold exercise results in increased cerebral blood flow, improved regulatory processes, including sleep, as well as autonomic nervous system regulation in patients with concussion.^10,69 Exercise may alleviate anxiety through multiple proposed mechanisms including acting as a form of exposure therapy, improved self-efficacy, and neurotransmitter and hormonal changes including release of endorphins.^70
–72

Limitations

There are several limitations to this study. The demographics of the sample, including both athletes and non-athletes, makes study results difficult to generalize, because athletes may present with different pre-injury levels of fitness and may engage in various post-injury exertion regimens that may impact their anxiety levels. Further, with this sample being collected at a specialty concussion clinic, results may not generalize to other healthcare settings (e.g., emergency department, pediatrician's office). In addition, patients with non-sport mechanisms of injury—for example, motor vehicle accidents—may have trauma-related symptomology or secondary gain, which was not measured in this study.

Approximately 12% of the concussed sample had a pre-existing history of anxiety, which may result in higher rates of post-concussion anxiety. The percentage of the concussed adolescents with pre-injury anxiety, however, was not significantly different than that of the controls. Nonetheless, future studies should examine the role of pre-injury anxiety and other clinical risk factors.

A higher proportion of physically developed adolescents were recruited, possibly resulting in a higher level of pre-injury baseline anxiety than would be expected in a more evenly distributed adolescent sample. Incorporating symptom severity, as measured by a post-concussion symptom scale, would add valuable clinical information to help quantify to what extent these finding may be mediated by symptom burden. Future studies should also examine treatment efficacy for post-concussion anxiety among adolescents. Last, a larger sample size should be recruited in future research, because our study was underpowered to meet our secondary analyses.

Conclusion

This study was the first to examine sex and pubertal development differences in anxiety among adolescents after concussion. The key findings suggested that adolescents reported higher levels of anxiety after concussion compared with controls, and females reported higher anxiety after concussion than males. Females who completed pubertal development and who were still developing reported higher levels of anxiety after concussion than males at the same respective pubertal developmental stages. These findings highlight the importance of assessing anxiety after concussion in adolescents. Healthcare professionals should include validated measures of anxiety in their multi-domain concussion assessment. Clinicians should also be prepared to provide both psychoeducation and referrals for psychological treatment for adolescents with anxiety, when indicated.

The results also emphasize the need to consider the role of both sex and pubertal development on anxiety after concussion in adolescents. Early identification of adolescents at risk for later development of post-concussion anxiety (e.g., females) should be used to inform earlier behavioral interventions and referrals to mitigate anxiety after concussion in this at-risk population.

Transparency, Rigor, and Reproducibility Summary

This study was a conducted through of a pre-existing longitudinal study, Investigating Concussion in Adolescents at Risk for Emotional Dysregulation (iCare) study (R01MH11488101; PIs: Versace, Kontos). The data analysis plan was formulated prior to beginning collection. The analysis plan was not pre-registered, but the lead statistician (Dr. Eagle) confirms that the proposed statistical analysis plan was pre-specified. A sample size of 80 participants was planned based upon 80% power required for the proposed statistical analyses. 126 participants (40 healthy controls, 86 concussed adolescents) were initially included in the study, with 97 participants remaining enrolled by the final visit. Seven healthy controls were lost to missing data, and 66.3% of the concussed sample was available by the third visit. Data analyses were performed by investigators aware of relevant characteristics of the study participants. Data was initially collected in concert with patients' presentation for concussion treatment. Data was collected using demographic questionnaires and standardized self-report measures. Inclusion/exclusion criteria and clinical outcomes were assessed by investigators with long-standing records of empirical research. Statistical tests were based on the assumptions that dependent variables and covariates were coded as continuous variables. Bonferroni corrections were applied as appropriate. No external validation studies are ongoing. The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. There is no analytic code associated with this study. The contributing authors all agree to provide the full content of the manuscript, on request, by contacting the corresponding author.

Footnotes

Funding Information

This study was funded through the pre-existing longitudinal study, Investigating Concussion in Adolescents at Risk for Emotional Dysregulation (iCare) study (R01MH11488101; PIs: Versace, Kontos).

Author Disclosure Statement

Drs. Kontos and Collins receive book royalties from APA Books, and funding for their research through the University of Pittsburgh from the Centers for Disease Control and Prevention, Chuck Noll Foundation for Brain Injury Research, Department of Defense (CDMRP, USAMRAA, USUHS), National Football League, National Institutes of Health (NICHD, NIMH, NINDS), and private donors.

References

Davis

, Anderson

, Babl

, et al. What is the difference in concussion management in children as compared with adults? A systematic review. Br J Sports Med, 2017; 51(12):949–957; doi:10.1136/bjsports-2016-097415

Collins

, Kontos

, Reynolds

, et al. A comprehensive, targeted approach to the clinical care of athletes following sport-related concussion. Knee Surg Sports Traumatol Arthrosc, 2014; 22(2):235–246; doi:10.1007/s00167-013-2791-6

Kontos

, Sufrinko

, Sandel

, et al. Sport-related concussion clinical profiles: clinical characteristics, targeted treatments, and preliminary evidence. Curr Sports Med Rep, 2019; 18(3):82–92; doi:10.1249/jsr.0000000000000573

Giza

, Hovda

. The new neurometabolic cascade of concussion. Neurosurgery, 2014; 75 Suppl 4(0 4):S24–S33; doi:10.1227/NEU.0000000000000505

McKay

, Brooks

, Mrazik

, et al. Psychometric properties and reference values for the ImPACT neurocognitive test battery in a sample of elite youth ice hockey players. Arch Clin Neuropsychol, 2014; 29(2):141–151; doi:10.1093/arclin/act116

Thomas

, Apps

, Hoffmann

, et al. Benefits of strict rest after acute concussion: a randomized controlled trial. Pediatrics, 2015; 135(2):213–223

Collins

, Lovell

, Iverson

, et al. Examining concussion rates and return to play in high school football players wearing newer helmet technology: a three-year prospective cohort study. Neurosurgery, 2006; 58(2):275–286; doi:10.1227/01.Neu.0000200441.92742.46

McCrory

, Meeuwisse

, Dvorak

, et al. Consensus statement on concussion in sport-the 5(th) international conference on concussion in sport held in Berlin, October 2016. Br J Sports Med, 2017; 51(11):838–847,\; doi:10.1136/bjsports-2017-097699

Williams

, Puetz

, Giza

, et al. Concussion recovery time among high school and collegiate athletes: a systematic review and meta-analysis. Sports Med, 2015; 45(6):893–903; doi:10.1007/s40279-015-0325-8

10.

Gillie

, Fazio-Sumrok

, Eagle

, et al. Clinical predictors of post-injury anxiety in adolescent patients following concussion. Appl Neuropsychol Child, 2022; 11(3):253–259; doi:10.1080/21622965.2020.1799790

11.

Harmon

, Clugston

, Dec

, et al. American Medical Society for Sports Medicine position statement on concussion in sport. Br J Sports Med, 2019; 53(4):213–225; doi:10.1136/bjsports-2018-100338

12.

Yang

, Peek-Asa

, Covassin

, et al. Post-concussion symptoms of depression and anxiety in division I collegiate athletes. Dev Neuropsychol, 2015; 40(1):18–23; doi:10.1080/87565641.2014.973499

13.

Covassin

, Crutcher

, Bleecker

, et al. Postinjury anxiety and social support among collegiate athletes: a comparison between orthopaedic injuries and concussions. Journal of athletic training, 2014; 49(4):462–468; doi:10.4085/1062-6059-49.2.03

14.

Gray

, McNaughton

. The neuropsychology of anxiety: reprise. Nebr Symp Motiv, 1996; 43:61–134

15.

Iverson

, Schatz

. Advanced topics in neuropsychological assessment following sport-related concussion. Brain Inj, 2015; 29(2):263–275; doi:10.3109/02699052.2014.965214

16.

Iverson

, Silverberg

, Mannix

, et al. Factors associated with concussion-like symptom reporting in high school athletes. JAMA Pediatr, 2015; 169(12):1132–1140; doi:10.1001/jamapediatrics.2015.2374

17.

Carlson

, Kangas

, Susa

, et al. Sport-related concussion is associated with elevated anxiety, but not attentional bias to threat. Brain Inj, 2020; 34(3):363–368; doi:10.1080/02699052.2020.1723698

18.

Fish

, Vanni

, Mohammed

, et al. Comparison of anxiety and depression symptoms in concussed and nonconcussed adolescents. Sports Health, 2023; 15(2):185–191; doi:10.1177/19417381221113840

19.

Singh

, Savitz

, Teague

, et al. Mood symptoms correlate with kynurenine pathway metabolites following sports-related concussion. J Neurol Neurosurg Psychiatry, 2016; 87(6):670–675; doi:10.1136/jnnp-2015-311369

20.

Luis

, Mittenberg

. Mood and anxiety disorders following pediatric traumatic brain injury: a prospective study. J Clin Exp Neuropsychol, 2002; 24(3):270–279; doi:10.1076/jcen.24.3.270.982

21.

Dorte

MC.

Examining Sex and Gender Differences in Anxiety Disorders. In: A Fresh Look at Anxiety Disorders. ( Federico

ed.) IntechOpen: Rijeka; 2015; Ch. 2.

22.

Merikangas

, He

, Burstein

, et al. Lifetime prevalence of mental disorders in U.S. adolescents: results from the National Comorbidity Survey Replication–Adolescent Supplement (NCS-A). J Am Acad Child Adolesc Psychiatry, 2010; 49(10):980–989; doi:10.1016/j.jaac.2010.05.017

23.

Brent

, Max

. Psychiatric sequelae of concussions. Curr Psychiatry Rep, 2017; 19(12):108; doi:10.1007/s11920-017-0862-y

24.

Marar

, McIlvain

, Fields

, et al. Epidemiology of concussions among United States high school athletes in 20 sports. Am J Sports Med, 2012; 40(4):747–755; doi:10.1177/0363546511435626

25.

Dougan

, Horswill

, Geffen

. Athletes' age, sex, and years of education moderate the acute neuropsychological impact of sports-related concussion: a meta-analysis. J Int Neuropsychol Soc, 2014; 20(1):64–80; doi:10.1017/s1355617712001464

26.

Graham

, Rivara

, Ford

, Spicer

. Sports-Related Concussions in Youth. National Academies Press (US): Washington (DC); 2014

27.

Broshek

, Kaushik

, Freeman

, et al. Sex differences in outcome following sports-related concussion. J Neurosurg, 2005; 102(5):856–863; doi:10.3171/jns.2005.102.5.0856

28.

Colvin

, Mullen

, Lovell

, et al. The role of concussion history and gender in recovery from soccer-related concussion. Am J Sports Med, 2009; 37(9):1699–1704; doi:10.1177/0363546509332497

29.

Bender

, Reinholdt-Dunne

, Esbjørn

, et al. Emotion dysregulation and anxiety in children and adolescents: Gender differences. Personality and Individual Differences, 2012; 53(3):284–288; doi:https://doi.org/10.1016/j.paid.2012.03.027

30.

McLean

, Asnaani

, Litz

, et al. Gender differences in anxiety disorders: prevalence, course of illness, comorbidity and burden of illness. J Psychiatr Res, 2011; 45(8):1027–1035; doi:10.1016/j.jpsychires.2011.03.006

31.

Deardorff

, Hayward

, Wilson

, et al. Puberty and gender interact to predict social anxiety symptoms in early adolescence. J Adolesc Health, 2007; 41(1):102–104; doi:10.1016/j.jadohealth.2007.02.013

32.

Hayward

, Sanborn

. Puberty and the emergence of gender differences in psychopathology. J Adolesc Health, 2002; 30(4 Suppl):49–58; doi:10.1016/s1054-139x(02)00336-1

33.

Carter

, Silverman

, Jaccard

. Sex variations in youth anxiety symptoms: effects of pubertal development and gender role orientation. J Clin Child Adolesc Psychol, 2011; 40(5):730–741; doi:10.1080/15374416.2011.597082

34.

Lima Santos

, Kontos

, Holland

, et al. The role of puberty and sex on brain structure in adolescents with anxiety following concussion. Biol Psychiatry Cogn Neurosci Neuroimaging, 2022; doi:10.1016/j.bpsc.2022.09.013

35.

Meehan

, 3rd, Taylor

, Berkner

, et al. Division III collision sports are not associated with neurobehavioral quality of life. J Neurotrauma, 2016; 33(2):254–259; doi:10.1089/neu.2015.3930

36.

Beck

, Epstein

, Brown

, et al. Beck anxiety inventory. J Consult Clin Psychol, 1993

37.

Meier

, Bellgowan

PSF

, Mayer

. 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

38.

Meier

, Bellgowan

, Singh

, et al. Recovery of cerebral blood flow following sports-related concussion. JAMA Neurol, 2015; 72(5):530–538; doi:10.1001/jamaneurol.2014.4778

39.

Kontos

, Elbin

, Schatz

, et al. A revised factor structure for the post-concussion symptom scale: baseline and postconcussion factors. Am J Sports Med, 2012; 40(10):2375–2384; doi:10.1177/0363546512455400

40.

Patricios

, Schneider

, Dvorak

, et al. Consensus statement on concussion in sport: the 6th International Conference on Concussion in Sport–Amsterdam, October 2022. Br J Sports Med, 2023; 57(11):695–711; doi:10.1136/bjsports-2023-106898

41.

Monga

, Birmaher

, Chiappetta

, et al. Screen for Child Anxiety-Related Emotional Disorders (SCARED): convergent and divergent validity. Depression and anxiety, 2000; 12(2):85–91; doi:10.1002/1520-6394(2000)12:2<85::Aid-da4>3.0.Co;2-2

42.

Birmaher

, Brent

, Chiappetta

, et al. Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED): a replication study. J Am Acad Child Adolesc Psychiatry, 1999; 38(10):1230–1236; doi:10.1097/00004583-199910000-00011

43.

Birmaher

, Khetarpal

, Brent

, et al. The Screen for Child Anxiety Related Emotional Disorders (SCARED): scale construction and psychometric characteristics. J Am Acad Child Adolesc Psychiatry, 1997; 36(4):545–553; doi:10.1097/00004583-199704000-00018

44.

Petersen

, Crockett

, Richards

, et al. A self-report measure of pubertal status: reliability, validity, and initial norms. J Youth Adolesc, 1988; 17(2):117–133; doi:10.1007/bf01537962

45.

Shirtcliff

, Dahl

, Pollak

. Pubertal development: correspondence between hormonal and physical development. Child Dev, 2009; 80(2):327–337; doi:10.1111/j.1467-8624.2009.01263.x

46.

Schmitz

, Hovell

, Nichols

, et al. A validation study of early adolescents' pubertal self-assessments. J Early Adolesc, 2004; 24(4):357–384

47.

Faul

, Erdfelder

, Lang

, et al. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods, 2007; 39(2):175–191; doi:10.3758/bf03193146

48.

Selya

, Rose

, Dierker

, et al. A practical guide to calculating Cohen's f(2), a measure of local effect size, from PROC MIXED. Front Psychol, 2012; 3:111; doi:10.3389/fpsyg.2012.00111

49.

Covassin

, Zynda

, Loftin

, et al. Changes in state and trait anxiety throughout concussion recovery in high school and college-aged individuals. J Athl Train, 2023; doi:10.4085/1062-6050-0536.22

50.

Cohen

. Eta-Squared and Partial Eta-Squared in Fixed Factor Anova Designs. Educational and Psychological Measurement, 1973; 33(1):107–112; doi:10.1177/001316447303300111

51.

Broshek

, De Marco

, Freeman

. A review of post-concussion syndrome and psychological factors associated with concussion. Brain Inj, 2015; 29(2):228–237; doi:10.3109/02699052.2014.974674

52.

Macartney

, Woodfield

, Terekhov

, et al. Anxiety, depression, and symptom experience in concussed children and youth. J Spec Pediatr Nurs, 2021; 26(1):e12310; doi:10.1111/jspn.12310

53.

Sandel

, Reynolds

, Cohen

, et al. Anxiety and mood clinical profile following sport-related concussion: from risk factors to treatment. Sport Exerc Perform Psychol, 2017; 6(3):304–323; doi:10.1037/spy0000098

54.

Barkhoudarian

, Hovda

, Giza

. The molecular pathophysiology of concussive brain injury. Clin Sports Med, 2011; 30(1):33–48; vii-iii, doi:10.1016/j.csm.2010.09.001

55.

Chen

, Johnston

, Petrides

, et al. Neural substrates of symptoms of depression following concussion in male athletes with persisting postconcussion symptoms. Arch Gen Psychiatry, 2008; 65(1):81–89; doi:10.1001/archgenpsychiatry.2007.8

56.

Smyth

, Sandhu

, Crawford

, et al. The role of serotonin receptor alleles and environmental stressors in the development of post-concussive symptoms after pediatric mild traumatic brain injury. Dev Med Child Neurol, 2014; 56(1):73–77; doi:10.1111/dmcn.12263

57.

Diagnostic and Statistical Manual of Mental Disorders: DSM-5,™ 5th ed. American Psychiatric Publishing, Inc.: Arlington, VA; 2013

58.

Collins

, Kontos

, Okonkwo

, et al. Statements of agreement From the Targeted Evaluation and Active Management (TEAM) approaches to Treating Concussion Meeting Held in Pittsburgh, October 15-16, 2015. Neurosurgery, 2016; 79(6):912–929; doi:10.1227/neu.0000000000001447

59.

Erikson

EH.

Identity: youth and crisis. Norton & Co.: Oxford, England; 1968

60.

Lau

, Kontos

, Collins

, et al. Which on-field signs/symptoms predict protracted recovery from sport-related concussion among high school football players?. Am J Sports Med, 2011; 39(11):2311–2318; doi:10.1177/0363546511410655

61.

Slobounov

, Slobounov

, Sebastianelli

, et al. Differential rate of recovery in athletes after first and second concussion episodes. Neurosurgery, 2007; 61(2):338–344; doi:10.1227/01.Neu.0000280001.03578.Ff

62.

Sufrinko

, Mucha

, Covassin

, et al. Sex differences in vestibular/ocular and neurocognitive outcomes after sport-related concussion. Clin J Sport Med, 2017; 27(2):133–138; doi:10.1097/jsm.0000000000000324

63.

Lewinsohn

, Gotlib

, Lewinsohn

, et al. Gender differences in anxiety disorders and anxiety symptoms in adolescents. J Abnorm Psychol, 1998; 107(1):109–117; doi:10.1037/0021-843x.107.1.109

64.

Patterson

, Mann

, Grotzinger

, et al. Genetic and environmental influences on internalizing psychopathology across age and pubertal development. Dev Psychol, 2018; 54(10):1928–1939; doi:10.1037/dev0000578

65.

Reardon

, Leen-Feldner

, Hayward

. A critical review of the empirical literature on the relation between anxiety and puberty. Clin Psychol Rev, 2009; 29(1):1–23; doi:10.1016/j.cpr.2008.09.005

66.

Meuret

, Wolitzky-Taylor

, Twohig

, et al. Coping skills and exposure therapy in panic disorder and agoraphobia: latest advances and future directions. Behav Ther, 2012; 43(2):271–284; doi:10.1016/j.beth.2011.08.002

67.

Curtiss

, Levine

, Ander

, et al. Cognitive-Behavioral Treatments for Anxiety and Stress-Related Disorders. Focus (Am Psychiatr Publ), 2021; 19(2):184–189; doi:10.1176/appi.focus.20200045

68.

Kaczkurkin

, Foa

. Cognitive-behavioral therapy for anxiety disorders: an update on the empirical evidence. Dialogues Clin Neurosci, 2015; 17(3):337–346; doi:10.3188/7/DCNS.2015.17.3/akaczkurkin

69.

Clausen

, Pendergast

, Willer

, et al. Cerebral blood flow during treadmill exercise is a marker of physiological postconcussion syndrome in female athletes. J Head Trauma Rehabil, 2016; 31(3):215–224; doi:10.1097/htr.0000000000000145

70.

Broman-Fulks

, Berman

, Rabian

, et al. Effects of aerobic exercise on anxiety sensitivity. Behav Res Ther, 2004; 42(2):125–136; doi:10.1016/s0005-7967(03)00103-7

71.

Jayakody

, Gunadasa

, Hosker

. Exercise for anxiety disorders: systematic review. Br J Sports Med, 2014; 48(3):187–196; doi:10.1136/bjsports-2012-091287

72.

Salmon

. Effects of physical exercise on anxiety, depression, and sensitivity to stress: a unifying theory. Clin Psychol Rev, 2001; 21(1):33–61; doi:10.1016/s0272-7358(99)00032-x