INTRODUCTION
Assessing self-reported symptomatology is an important aspect of concussion management as most clinicians will not allow athletes to return to play if they are still symptomatic, even if they are deemed “neurocognitively recovered.” For example, Arnett and colleagues (Reference Arnett, Meyer, Merritt and Guty2016) laid out an empirically derived algorithm for determining neurocognitive recovery, yet progression to the next steps following “recovery” is dependent upon an absence/reduction of symptoms as determined by self-report. However, what should one do if athletes report elevated symptoms, even at baseline when they are not concussed? Is it still appropriate to hold athletes from play because they report symptoms following injury, even if that is their individual “norm”? These questions highlight the need for a more individualized approach and understanding of how each athlete presents at baseline – especially with regard to symptomatology. One common tool that is used to assess symptomatology is the Post-Concussion Symptom Scale (PCSS) from The Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT). The PCSS is a self-report measure that is often utilized at both baseline and post-concussion to capture an athlete’s symptom profile. Despite its depiction as a “post-concussion” scale, previous research has demonstrated the utility of using the PCSS at baseline and has shown the value added in making comparisons to baseline evaluations following concussion. For example, Custer and colleagues (Reference Custer, Sufrinko, Elbin, Covassin, Collins and Kontos2016) found that athletes with high PCSS scores at baseline experienced greater decline in verbal and visual memory following concussion compared to those who reported no symptoms on the PCSS. This indicates that those reporting higher symptomatology at baseline may be at greater risk for negative outcomes following concussion than those who deny symptomatology. However, this begs the question: what leads an athlete to report symptoms consistent with concussion despite not experiencing an injury (i.e., at baseline)?
One possible factor associated with baseline symptom reporting is whether the athlete presents with significant affective symptomatology. Previous research has found that symptoms of stress, depression, and/or anxiety were positively correlated with post-concussion-like symptoms in individuals with no concussion history (Edmed & Sullivan, Reference Edmed and Sullivan2012; Garden & Sullivan, Reference Garden and Sullivan2010). Similarly, Iverson (Reference Iverson2005) demonstrated that patients who were clinically diagnosed with a depressive disorder demonstrated elevated symptom reporting on the British Columbia Post-concussion Symptom Inventory such that approximately 90% of patients in this sample met “liberal self-report criteria” for post-concussion syndrome (PCS) while over 50% met conservative criteria for PCS despite not having sustained recent head injury. Another study found that individuals with depression but without recent mild traumatic brain injury (mTBI), reported significantly greater symptom scores on the British Columbia Post-concussion Symptom Inventory than those who had sustained mTBI but denied depression, though these groups both reported significantly fewer symptoms than those with both mTBI and depression (Lange et al., Reference Lange, Iverson and Rose2011). Relatedly, Kontos and colleagues (Reference Kontos, Covassin, Elbin and Parker2012) demonstrated that, at 14 days post-concussion, depression was related to higher symptom scores. Previous work has also focused on symptom reporting in individuals with a history of mood disturbance. Despite the focus on history of disturbance rather than current complaint, these studies showed similar findings in that athletes with a history of mood disorder reported more symptoms with greater severity (Wallace et al., Reference Wallace, Learman, Moran, Covassin, Deitrick, Delfin and Shina2020; Weber et al., Reference Weber, Dean, Hoffman, Broglio, McCrea, McAllister and Dykhuizen2018), and that athletes with a history of psychiatric treatment were more likely to report emotional symptoms (Asken et al., Reference Asken, Snyder, Clugston, Gaynor, Sullan and Bauer2017). Lastly, Snell and colleagues (Reference Snell, Martin, Macleod, Surgenor, Siegert, Jean and Anderson2018) found that participants with chronic pain without history of mTBI also reported elevated post-concussion-like symptoms scores. Taken together, these findings suggest that there are many factors other than concussion that may lead to increased scores on self-report symptom measures.
Additionally, previous work by Riegler and colleagues (Reference Riegler, Guty and Arnett2019a) demonstrated the utility of using the PCSS affective factor as a screener for depression, suggesting that the affective symptoms captured on the PCSS demonstrate sensitivity similar to that of established depression screeners. Further, based on current understanding of depression and anxiety, these syndromes can lead to or exacerbate other difficulties such as cognitive difficulties (Castaneda et al., Reference Castaneda, Tuulio-Henriksson, Marttunen, Suvisaari and Lönnqvist2008), somatic and physical symptoms (Simms et al., Reference Simms, Prisciandaro, Krueger and Goldberg2012; Suls & Howren, Reference Suls and Howren2012), and/or sleep disturbances (Alvaro et al., Reference Alvaro, Roberts and Harris2013; Nyer et al., Reference Nyer, Farabaugh, Fehling, Soskin, Holt, Papakostas and Mischoulon2013). This understanding of the phenomenology of depression and anxiety is important because it may interfere with how the PCSS is interpreted given the overlap with potential concussion symptomatology. Specifically, a large number of symptoms of depression and anxiety overlap with items included in standard sports-concussion symptom checklists (e.g. sleep disturbance, mood changes, cognitive symptoms, and somatic symptoms) (Solomon et al., Reference Solomon, Kuhn and Zuckerman2015). This overlap may present challenges in discerning whether symptoms are pathognomonic of concussion or related to affective disturbance, which suggests that further research is warranted in order to improve clinical recommendations. Some of our previous work has focused on the association of affective symptomatology with neurocognitive performance (Guty & Arnett, Reference Guty and Arnett2018; Riegler et al., Reference Riegler, Guty and Arnett2020; Thomas, Guty, Riegler, & Arnett, Reference Thomas, Guty, Riegler and Arnett2021); however, it is unclear how affective disturbances are associated with symptom reporting across the PCSS. One might expect athletes experiencing either depression or anxiety, or both together, to have an elevated affective cluster on the PCSS given the nature of such symptoms and overlap, but we are interested in understanding associations with other symptom clusters as well. Given that comorbidity has been associated with more severe symptoms and complaints (LeMoult, Castonguay, Joormann, & McAleavey, Reference LeMoult, Castonguay, Joormann, McAleavey, Castonguay and Oltmanns2013; Pratt et al., Reference Pratt, Druss, Manderscheid and Walker2016), it is important to understand whether this is global or more circumscribed to the affective cluster. Thus, while the decision to evaluate symptom cluster scores in addition to total scores is predicated upon the limitations described above, the examination of the potential impact of elevated affective symptomatology on PCSS symptom cluster scores is largely exploratory, though it is possible that this approach may prove to be fruitful in helping to understand psychological influences on symptom reporting. Further, given research demonstrating that athletes tend to under-report symptomatology, particularly male athletes underreporting affective symptomatology and female athletes reporting more symptoms than male athletes at both baseline and following concussion, it is important to further explore patterns of symptom reporting (Asken et al., Reference Asken, Snyder, Clugston, Gaynor, Sullan and Bauer2017; Covassin et al., Reference Covassin, Swanik, Sachs, Kendrick, Schatz, Zillmer and Kaminaris2006, Reference Covassin, Elbin, Harris, Parker and Kontos2012; McCrea et al., Reference McCrea, Hammeke, Olsen, Leo and Guskiewicz2004; Meier et al., Reference Meier, Brummel, Singh, Nerio, Polanski and Bellgowan2015; Moser et al., Reference Moser, Olek and Schatz2019).
Current Study
To summarize, the current study aims to fill gaps in the neuropsychology and sports-related concussion (SRC) literature regarding affective disturbance, particularly elevated concomitant anxiety/depressive symptoms, and its association with baseline symptom reporting in college athletes. While previous research has investigated various aspects of symptom reporting and affective disturbance, the current study is believed to be novel in that it is aimed at assessing symptom reporting, in total and by symptom cluster, in college athletes who report affective disturbance at the time of baseline testing. It is important to properly identify and document baseline symptomatology because it may allow for more appropriate comparisons following concussion. In other words, if an athlete reports elevated symptomatology at baseline, it may be less concerning if the same individual reports elevated symptomatology (within a comparable range) following concussion – especially as opposed to an athlete who denies symptomatology at baseline. Thus, it allows for providers to make decisions relative to the individual and may be used in conjunction with neuropsychological data. We hypothesize that: 1) athletes reporting affective disturbance (e.g., athletes in the depressive symptoms only, anxiety symptoms only, and concomitant groups) will report higher PCSS scores, indicating greater symptomatology, compared to healthy controls (HCs); and 2) athletes reporting elevated concomitant anxiety/depressive symptoms will report higher PCSS scores than athletes experiencing either depressive symptoms or anxiety symptoms alone.
METHODS
Participants
This study involved 833 (M = 622,F = 211) collegiate athletes who were involved in a concussion management program at an NCAA Division I University. The mean age was 18.51 years (SD = 1.05) with a range from 17 to 24. All participants completed a comprehensive neuropsychological test battery at baseline which included self-report measures of anxiety, depression, and the PCSS. Testing was completed as part of standard operating procedures within the athletic programs. Athletes from the following sports underwent baseline testing: football, men’s and women’s soccer, wrestling, men’s and women’s lacrosse, men’s and women’s ice hockey, men’s and women’s basketball, baseball, softball, crew, volleyball, and rugby. Athletes in the current study were selected from a larger group of athletes (N = 1050) receiving baseline testing between 2002 and 2019. Athletes were only included in the current study if they completed the PCSS, the Beck Depression Inventory-Fast Screen (BDI-FS) (Beck, Steer, & Brown, Reference Beck, Steer and Brown2000), and the NEO-Five Factor Inventory (NEO-FFI) (McCrae & Costa, Reference McCrae and Costa2004) at baseline. All data were collected prior to the COVID-19 pandemic.
Procedures
This was a follow-up to previous work from a cross-sectional, retrospective research study conducted as part of the Sports Concussion Program at our NCAA Division I University (Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021). Our Sports Concussion Program is based on the “Sports as Laboratory Assessment Model (SLAM)” (Bailey et al., Reference Bailey, Samples, Broshek, Freeman and Barth2010; Levin et al., Reference Levin, Eisenberg and Benton1989). All participants were referred for baseline testing by their athletic trainer or team physician and completed a 2.5-hour comprehensive neuropsychological test battery at baseline. The neuropsychological test battery was administered by undergraduate research assistants or graduate students who were supervised by a Ph.D.-level clinical neuropsychologist. Our study was conducted in compliance with our university’s Institutional Review Board requirements and American Psychological Association ethical guidelines.
Measures
The BDI-FS was used to measure depressive symptoms in this study. The BDI-FS consists of seven items rated from 0 to 3, with higher scores indicating more severe symptoms. The BDI-FS is well-validated for use in medical populations and demonstrates value in measuring depressive symptoms in concussed populations due to its ability to discriminate between depression symptoms and symptoms of a concussion (Riegler et al., Reference Riegler, Guty and Arnett2019a). The distribution of the BDI-FS scores was not normal and demonstrated positive skewness and kurtosis in this sample. Further, a large percentage of individuals (52.3%) scored 0 on the BDI-FS. Given these considerations, and in order to maintain consistency with the cutoffs for the NEO-FFI subscale (described below), the BDI-FS cutoffs were derived based on scores one standard deviation greater than the sample mean (M = 1.06,SD = 1.73). As such, BDI-FS scores were dichotomized into two groups based on the presence of elevated depressive symptomatology (≥3) or absence of elevated depressive symptomatology (<3). These cutoffs were consistent for both males and females.
A modified subscale of the NEO-FFI was used to measure anxiety. Possible scores ranged from 0 to 16 with higher scores indicating more severe symptoms. The rationale and support for the development of this subscale can be found elsewhere (Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021). For purposes of this paper, it is important to note that the NEO-FFI Anxiety Subscale has been shown to correlate with the Anxiety Sensitivity Index (ASI: Reiss, Peterson, Gursky, & McNally, Reference Reiss, Peterson, Gursky and McNally1986) at levels comparable to the ASI’s correlation with other trait anxiety measures (McNally & Lorenz, Reference McNally and Lorenz1987; Reiss et al., Reference Reiss, Peterson, Gursky and McNally1986; Sandin et al., Reference Sandin, Chorot and McNally2001; Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021). The ASI has also shown reliability in assessing anxiety in college students (Peterson & Heilbronner, Reference Peterson and Heilbronner1987). Thus, the ASI appears to be an appropriate comparison for validation, and the NEO-FFI Anxiety Subscale can be considered an acceptable measure of anxiety for the current study.
Given that there is not a validated cutoff for the NEO-FFI Anxiety subscale, the current study used a score of one standard deviation above the sample mean (M = 6.51,SD = 2.84) to indicate clinical significance. As such, NEO-FFI Anxiety subscale scores were dichotomized into two groups based on the presence of elevated anxiety symptomatology (≥10) or absence of elevated anxiety symptomatology (<10) which is comparable to cutoffs used in previous work (Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021). These cutoffs were consistent for both males and females. Of note, the NEO-FFI Anxiety subscale scores were normally distributed. Rationale for using one standard deviation above the mean cutoffs, in addition to dichotomizing presence or absence of depression and/or anxiety, is predicated upon our interest in exploring the role of elevated affective symptomatology rather than clinical diagnoses. Moreover, we are specifically interested in evaluating differences between groups, particularly the concomitant depression/anxiety group compared to the other groups. Further, using a cutoff of one standard deviation is a commonly used method to indicate impairment (Lezak et al., Reference Lezak, Howieson, Loring and Fischer2004), and maintains symmetry with the approach utilized in our previous work (Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021).
The PCSS is a self-report symptom measure that is comprised of 22 items describing common post-concussion symptoms on a scale of 0 to 6 with 0 being ‘None’ and 6 being ‘Severe’ (Lovell et al., Reference Lovell, Iverson, Collins, Podell, Johnston, Pardini, Pardini, Norwig and Maroon2006). While much of previous research has focused on the PCSS Total Score when evaluating symptomatology at baseline and/or post-concussion (Covassin et al., Reference Covassin, Elbin, Harris, Parker and Kontos2012; McClincy et al., Reference McClincy, Lovell, Pardini, Collins and Spore2006), this approach may not be sufficient as it does not account for the athlete’s specific symptom profile. Thus, this approach may not be an accurate representation of the athlete’s overall symptomatology. Therefore, we decided to evaluate both PCSS Total Scores, as well as specific symptom clusters. To examine the latter, PCSS scores were broken down into 4 symptom clusters based on previous factor analytic work demonstrating that the PCSS items load onto 4 primary factors (e.g., Cognitive, Physical, Affective, and Sleep) with headache as a separate 5th factor (Merritt et al., Reference Merritt, Meyer and Arnett2015). Headache was also included as a separate factor because previous work has shown that headaches are one of the most common and persistent symptoms associated with sports-related concussion (Heyer et al., Reference Heyer, Young, Rose, McNally and Fischer2016; Kontos et al., Reference Kontos, Elbin, Lau, Simensky, Freund, French and Collins2013; Womble et al., Reference Womble, Henley, Fedor and Collins2019) and may predict prolonged recovery following concussion (Iverson et al., Reference Iverson, Gardner, Terry, Ponsford, Sills, Broshek and Solomon2017). This inclusion of headache as a separate factor, in addition to clusters of Cognitive, Physical, Affective, and Sleep symptoms, is supported by previous research (Guty & Arnett, Reference Guty and Arnett2018; Merritt et al., Reference Merritt, Meyer and Arnett2015; Riegler et al., Reference Riegler, Guty, Thomas and Arnett2021). These symptom clusters and their associated PCSS items are described in Table 1. Of note, the symptom clusters differ in the associated number of items, thus leading to different possible maximum scores.
Table 1. Composition of groups defined by self-report symptoms of depression and anxiety
1 These groups are all independent, meaning that athletes captured in the Depression Only or Anxiety Only groups are not counted toward the Concomitant Group. In doing so, we attempted to make the groups as diagnostically pure as possible. For example, the n of the Anxiety Only group is smaller than the Concomitant group because many of the athletes experiencing symptoms of anxiety also experience symptoms of depression, thus falling into the Concomitant group and not the Anxiety Only group.
Quantitative Approach
Athletes were then separated into four groups (HC (n = 581), Depression Only (n = 136), Anxiety Only (n = 54), and Concomitant Depression/Anxiety (n = 62)) based on their anxiety and depression scores (see Table 2). Of note, the groups were unequal in size and some of the groups differed in the proportion of male to female athletes, an issue that will be discussed in subsequent sections. The groups did not significantly differ in terms of mean age or proportion of athletes who reported previous concussion(s).
Table 2. PCSS symptom clusters and associated items
Table adapted from Merritt, V.C., & Arnett, P.A. (2014). Premorbid predictors of post-concussion symptoms in collegiate athletes. Journal of Clinical & Experimental Neuropsychology, 36(10), 1098–1111; Table 5, p. 1105.
We first conducted linear regressions to first examine the effect of Group (i.e., affective symptomatology presentation) on Total PCSS Scores. We employed a dummy coding approach so we could make meaningful comparisons between HCs and each of the affective groups separately while maximizing statistical power. Next, in line with previous research aimed at exploring PCSS Symptom Cluster Scores instead of PCSS Total Score (Merritt et al., 2014), we conducted a series of six multivariate analyses of variance (MANOVAs) comparing each of the affective groups to the healthy control group, as well as to the other affective groups. A Bonferroni correction was calculated to control for multiple comparisons (i.e., the six multivariate tests). As such, a p-value of .008 was used to indicate statistical significance for the multivariate tests. The univariate tests were not subject to correction given that utilizing a MANOVA framework effectively controls for multiple comparisons when evaluating univariate effects.
RESULTS
PCSS Total Scores
Overall, results revealed a significant effect of Group on Total PCSS Score, F(3,829)= 45.38, p < .001, η p 2=.14.Footnote 1 We found that, compared to HCs, the concomitant group reported significantly greater Total PCSS Scores, t(829)=11.25, p < .001, d = 1.02,Footnote 2 as did the depressive symptoms only group, t(831)=4.58, p < .001, d = .45. The anxiety symptoms only group did not show a significant difference in Total PCSS Score compared to HCs, t(829)=.99, p=.32, d = .18. Further, we found that the concomitant group reported significantly greater Total PCSS Scores than both the depressive symptoms only group, t(829) = -6.96, p < .001, d = .68, and the anxiety symptoms only group, t(829) = -7.32, p < .001, d = .91. The depressive symptoms only group showed marginally elevated Total PCSS Score compared to the anxiety symptoms only group, t(829) = -1.83, p = .07, d = .30. Total PCSS Score results are depicted in Figure 1. To further understand these findings, we conducted additional analyses aimed at exploring potential differences in symptom cluster scores. PCSS Total and Symptom Cluster Score means, standard deviation, and score ranges are presented in Table 3.
Fig. 1. Baseline PCSS total scores by group.
Table 3. PCSS symptom cluster means, SDs and score ranges
PCSS Symptom Cluster Scores
In line with previous research by Merritt et al. (2014), we conducted additional analyses, which were planned a priori, to explore the presentation of symptomatology broken down into 5 symptom clusters: Cognitive, Physical, Affective, Sleep, and Headache. Results are depicted in Figure 2.
Fig. 2. Baseline PCSS symptom cluster scores by group.
Affective groups compared to healthy controls
MANOVA results revealed that, overall, all three affective groups reported significantly greater symptomatology than HCs, with the concomitant group showing the largest difference compared to HCs. These results are further explored in subsequent sections.
Concomitant group vs healthy controls
Overall, the concomitant group reported significantly greater symptomatology than the HC group, F(5,637) = 44.09, p < .001, η p 2 = .26. Further, while all symptom clusters were significantly elevated in the concomitant group, the factors with the largest effects relative to HCs were the affective cluster, F(1,641) = 211.03, p < .001, η p 2 = .25, cognitive cluster, F(1,641) = 81.32, p < .001, η p 2 = .11, and sleep cluster, F(1,641) = 42.45, p < .001, η p 2 = .06 (see Table 4).
Table 4. PCSS symptom clusters: affective groups compared to healthy controls
a partial eta squared effect sizes: small (.01 – .05); medium (.06 – .14); large (>.14).
** significant at the .05 level.
*** significant at the .001 level.
*marginally significant at the .10 level.
Depressive symptoms only group vs healthy controls
Overall, the depressive symptoms only group reported significantly greater symptomatology than the HC group, F(5,711) = 7.49, p < .001, η p 2 = .05. Further, the depressive symptoms only group reported significantly elevated symptomatology, compared to HCs, on every symptom cluster except headache. The factors with the largest effects were the affective cluster, F(1,715) = 26.10, p < .001, η p 2 = .04, cognitive cluster, F(1,715) = 20.42, p < .001, η p 2 = .03, and sleep cluster, F(1, 715) = 16.28, p < .001, η p 2 = .02 (see Table 4).
Anxiety symptoms only group vs healthy controls
Overall, the anxiety symptoms only group reported significantly greater symptomatology than the HC group, F(5,629) = 7.49, p = .003, η p 2 = .03. However, the anxiety symptoms only group reported significantly increased symptomatology, compared to HCs, on only the cognitive symptoms cluster, F(1,633) = 9.73, p = .002, η p 2 = .02. Interestingly, the anxiety symptoms only group did not differ from HCs in affective symptomatology, F(1,633) = 1.17, p = .28, η p 2 = .002 (see Table 4).
Affective groups compared to affective groups (depressive vs. anxiety vs. concomitant symptoms)
Depressive symptoms only group vs anxiety symptoms only group
Overall, the depressive symptoms only group did not differ significantly from the anxiety symptoms only group, though there was a small effect size, F(5,184) = 1.85, p = .11, η p 2 = .05. Of note, the depressive symptoms only group showed a marginal increase in sleep symptoms, F(1,188) = 3.81, p = .052, η p 2 = .02, and affective symptoms, F(1,188) = 2.82, p = .095, η p 2 = .02, compared to the anxiety symptoms only group. These groups did not differ on any of the other symptom clusters (see Table 5).
Table 5. PCSS symptom clusters: affective groups compared to affective groups
a partial eta squared effect sizes: small (.01 – .05); medium (.06 – .14); large (>.14).
* marginally significant at the .10 level.
** significant at the .05 level.
*** significant at the .001 level.
Concomitant group vs. depressive symptoms only group
Overall, the concomitant group reported significantly greater symptomatology than the depressive symptoms only group, F(5,192) = 7.55, p < .001, η p 2 = .16. Further, while all symptom clusters were significantly elevated in the concomitant group, the factors with the largest effects relative to the depressive symptoms only group were the affective cluster, F(1,196) = 36.32, p < .001, η p 2 = .16, and cognitive cluster, F(1,196) = 13.68, p < .001, η p 2 = .07 (see Table 5).
Concomitant group vs anxiety symptoms only group
Overall, the concomitant group reported significantly greater symptomatology than the anxiety symptoms only group, F(5,110) = 8.18, p < .001, η p 2 = .27. Further, while all symptom clusters were significantly elevated in the concomitant group, the factors with the largest effects relative to the anxiety symptoms only group were the affective cluster, F(1,114) = 32.98, p < .001, η p 2 = .22, sleep cluster, F(1,114) = 13.06, p < .001, η p 2 = .10, and physical cluster, F(1,114) = 8.11, p = .005, η p 2 = .07 (see Table 5).
DISCUSSION
The current paper examined the relationship between significant affective symptomatology and self-report symptomatology across 5 symptom clusters on the PCSS, as well as total number of symptoms. Some of our previous work has found that those experiencing significant affective symptomatology show greater neurocognitive impairments at baseline (Guty & Arnett, Reference Guty and Arnett2018; Riegler et al., Reference Riegler, Guty and Arnett2020; Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021), and previous research has also explored aspects of the relationship between affective disturbance (particularly depression) and symptom reporting (Asken et al., Reference Asken, Snyder, Clugston, Gaynor, Sullan and Bauer2017; Garden & Sullivan, Reference Garden and Sullivan2010; Iverson et al., Reference Iverson, Gardner, Terry, Ponsford, Sills, Broshek and Solomon2017; Wallace et al., Reference Wallace, Learman, Moran, Covassin, Deitrick, Delfin and Shina2020; Weber et al., Reference Weber, Dean, Hoffman, Broglio, McCrea, McAllister and Dykhuizen2018). However, additional research is warranted to better understand how an athlete’s affective state, particularly when reporting elevated concomitant depressive/anxiety symptoms, is associated with specific symptom clusters at baseline.
Our findings suggest that athletes experiencing elevated concomitant depressive/anxiety symptoms reported significantly greater levels of symptomatology across all 5 clusters of the PCSS compared to HCs. Further, our findings demonstrate that athletes experiencing elevated concomitant affective symptoms tend to report greater symptomatology than those with elevated affective symptoms in only one domain (e.g., anxiety or depressive symptoms alone). Thus, it appears that the presentation of two co-occurring affective disturbances negatively impacts athletes over and above the presentation of only one affective disturbance. While the current study examined athletes with elevated symptomatology rather than those with clinical disorders, these findings are inline with previous research showing that the burden of disease increases with each additional diagnosis, and that those with comorbid psychological disorders typically demonstrate poorer functioning and less responsiveness to treatment compared to those with only one psychological disturbance (Emmanuel et al., Reference Emmanuel, Simmonds and Tyrer1998; Gadermann et al., Reference Gadermann, Alonso, Vilagut, Zaslavsky and Kessler2012; Walczak et al., Reference Walczak, Ollendick, Ryan and Esbjørn2018). This finding is important as concomitant affective disturbance, and its effects on athletes, has remained relatively unstudied to this point yet clearly plays an important role in athletes’ functioning.
Unsurprisingly, the effect size for the affective cluster was largest for the concomitant and depressive symptoms only groups, but it is important to recognize the cognitive symptom cluster was next largest. This latter finding supports previous research findings that show that those experiencing depression or concomitant affective disturbance perform significantly worse than HCs on objective neuropsychological measures (Riegler et al., Reference Riegler, Guty and Arnett2019b; Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021). Considering the widespread use of objective neuropsychological testing in determining recovery from concussion, exploring how factors other than concussion, like affective symptomatology, may skew cognitive performance is critical. Moreover, it is also important to highlight that athletes in the concomitant group report significantly greater headache symptomatology – something that is deemed to be a persistent and “hallmark” symptom of concussion. Thus, it is important to take this into account when making post-concussion comparisons as this presentation may not necessarily be pathognomonic of concussion. Relatedly, it is critical to note that the sleep cluster has the highest score for the healthy control and anxiety groups, and the second highest score for the depression and concomitant depression/anxiety groups. These results support previous findings regarding the negative impact of sleep disturbance on college athletes (Riegler et al., Reference Riegler, Guty, Thomas and Arnett2021), and also consistent with prior work suggesting that sleep disturbance is a widespread problem for all athletes (Asken et al., Reference Asken, McCrea, Clugston, Snyder, Houck and Bauer2016, Reference Asken, Snyder, Clugston, Gaynor, Sullan and Bauer2017; Iverson et al., Reference Iverson, Silverberg, Mannix, Maxwell, Atkins, Zafonte and Berkner2015). Considering that sleep may be easily modifiable, it may be an important point of clinical intervention. It is also important to note that the physical symptoms cluster showed relatively low scores across groups, despite possessing the potential for the greatest score due to the cluster containing 7 items. In contrast, the cognitive, affective and sleep clusters contained only 4 items, and the headache cluster only contained one item. Thus, it appears that very few people reported physical symptoms at baseline, which suggests the physical symptoms cluster may be more specific to concussion itself rather than secondary factors like depression, anxiety, or comorbidity.
Interestingly, the anxiety symptoms only group showed similar elevations on the cognitive symptom cluster, but this group did not show a significant increase in affective symptoms compared to HCs. One possible explanation is predicated upon the types of questions that make up the affective cluster of the PCSS, as well as our use of the NEO-FFI Anxiety Subscale to define the anxiety symptoms only group. Given that anxiety and depression often overlap, we attempted to obtain a pure anxiety measure by creating an Anxiety Subscale from the Neuroticism Facet of the NEO-FFI. The development of this subscale is described elsewhere (Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021), but it is important to note that the PCSS affective symptom cluster is comprised of 4 items, 3 of which phenomenologically describe depression and only 1 of which appears related to anxiety. Considering the Anxiety Subscale is designed to be a purer measure of anxiety that does not include depressive symptoms, it is perhaps not surprising that there was not as much overlap as one might have expected. Further, we believe this finding actually gives credence to using the NEO-FFI Anxiety Subscale as an anxiety measure, as it did not correlate with BDI-FS total score in our sample population, r(847)Footnote 3 = .01, p = .75.
Despite this possible support for using the NEO-FFI Anxiety Subscale as a relatively pure anxiety measure, this could also be considered a limitation of the study in that we did not use a stand-alone anxiety measure. While this anxiety subscale has been shown to correlate with the ASI at a level comparable to other anxiety measures as previously described, the scale has not been subject to rigorous validation. Another possible limitation is that, because we created the affective groups based on cutoff scores derived from the sample mean and standard deviation, rather than clinical cutoffs, we are capturing those who are experiencing affective symptomatology rather than true clinical disorders. However, we believe these cutoffs are still of utility given that they are derived from base rates within our sample, and also because of some athletes’ tendency to underreport symptoms (McCrea et al., Reference McCrea, Hammeke, Olsen, Leo and Guskiewicz2004; Meier et al., Reference Meier, Brummel, Singh, Nerio, Polanski and Bellgowan2015). Further, these findings demonstrate that, even if athletes do not meet full criteria for a clinical disorder, they may still be prone to reporting greater symptomatology which is of clinical significance. Another possible limitation is that the groups differed in the proportion of male to female athletes, such that the Anxiety and Concomitant groups had a significantly larger proportion of females to males than the HC and Depression groups. Given that previous research has demonstrated that female athletes tend to report more symptoms than males, both at baseline and post-concussion (Asken et al., Reference Asken, Snyder, Clugston, Gaynor, Sullan and Bauer2017; Covassin et al., Reference Covassin, Elbin, Harris, Parker and Kontos2012), it is plausible that the results are influenced by potential sex differences. For example, if the groups were evenly distributed in terms of sex, it is possible that the HC and Depression groups would report higher PCSS scores, or the Anxiety and Concomitant groups would report lower PCSS scores, which could lead to less significant differences between groups. Alternatively, if our sample included more female athletes, it is possible that BDI-FS cutoff scores would differ between males and females, thus requiring a different set of cutoffs. Similar issues have been discussed, with proposed solutions, in Arnett et al. (Reference Arnett, Meyer, Merritt and Guty2016). Lastly, this study is also limited in that we are not able to fully account for potential secondary influences, such as history of migraine, vestibular disorders, or medication usage, which may also influence symptom reporting. Further, given that we do not require athletes to report all medical and/or psychiatric history at the time of testing, there may be factors other than depression and anxiety that contribute to these findings.
In sum, the findings of this study are important because they demonstrate the impact of concomitant depressive, and, to a lesser extent, anxiety symptoms on PCSS symptom reporting – both in total and by symptom cluster. These relationships are important to understand given the emphasis placed on symptom reporting in concussion management and when making return-to-activity decisions. For example, despite not currently experiencing a concussion, the concomitant group demonstrated significantly elevated symptomatology on the PCSS, which could skew post-concussion comparisons if the athlete’s affective problems persist. Alternatively, if an athlete experiences concomitant depression and anxiety at baseline, but not post-concussion, symptoms that may actually result from concussion may be masked by their elevated symptom profile at baseline. Thus, it is recommended that athletes are routinely screened for mood disturbances using brief screeners, like the NEO-FFI Anxiety Subscale, BDI-FS, and/or the Affective Scale from the PCSS (Riegler et al., Reference Riegler, Guty and Arnett2019a), so that athletes who might be at risk for affective symptomatology can then be referred for a more thorough assessment and provided resources if necessary. Further, by routinely screening for affective symptoms at baseline, this allows providers to be aware that these symptoms are present and thus able to account for the potential influence of these symptoms, as well as potential differences in symptom reporting, during subsequent testing. Future research may be focused on incorporating the influence of affective symptoms in RTL and/or RTP decision-making algorithms similar to the algorithm present by Arnett and colleagues (Reference Arnett, Meyer, Merritt and Guty2016). Taken with previous findings (Thomas et al., Reference Thomas, Guty, Riegler and Arnett2021), it is evident that people with concomitant affective disturbance have cognitive difficulties. However, without awareness of such difficulties at baseline, it could be assumed that cognitive difficulties associated with injury are related to concussion rather than underlying affective disturbance. Thus, it is vital that providers are aware of athlete’s baseline functioning when making clearance decisions.
Lastly, it is also important to understand potential discrepancies in symptom reporting because symptom reduction plays a large role in determining whether an athlete is allowed to return to play or return to learn following concussion. Considering the important role of academics in the lives of student-athletes, it is critical that pre-existing mood disorders not interfere with return to learning after concussion and that return to learning should always precede unrestricted return to play. Further, considering that classroom performance may be impacted more by some symptom clusters over others (e.g., cognitive symptoms), it is important to gain a more nuanced understanding of an athlete’s symptom profile – particularly in their day-to-day functioning prior to sustaining concussion in order to make more thoughtful clinical decisions after concussion.
FINANCIAL SUPPORT
The authors do not have a financial investment in this study or the outcome of the data to report.
CONFLICT OF INTEREST
The authors do not have conflicts of interest.
ETHICAL STANDARDS
Our study was conducted in compliance with our university’s Institutional Review Board requirements and American Psychological Association ethical guidelines.
Open access
