Depression and Neurocognitive Performance After Concussion Among Male and Female High School and Collegiate Athletes

1751

ORIGINAL ARTICLE

Depression and Neurocognitive Performance After Concussion Among Male and Female High School and Collegiate Athletes Anthony P. Kontos, PhD, Tracey Covassin, PhD, R.J. Elbin, PhD, Tonya Parker, PhD ABSTRACT. Kontos AP, Covassin T, Elbin RJ, Parker T. Depression and neurocognitive performance after concussion among male and female high school and collegiate athletes. Arch Phys Med Rehabil 2012;93:1751-6. Objectives: To prospectively examine the relationship of sport-related concussion with depression and neurocognitive performance and symptoms among male and female high school and college athletes. A secondary objective was to explore age and sex differences. Design: Pretest, multiple posttest, repeated-measures design. Setting: Laboratory. Participants: High school and collegiate athletes (N⫽75) with a diagnosed concussion. Interventions: Not applicable. Main Outcome Measures: Beck Depression Inventory-II and computerized neurocognitive test battery (Immediate Post-concussion Assessment and Cognitive Test), which includes concussion symptoms (Post-concussion Symptom Scale) at baseline and at 2, 7, and 14 days postinjury. Results: Concussed athletes exhibited significantly higher levels of depression from baseline at 2 days (Pⱕ.001), 7 days (P⫽.006), and 14 days postconcussion (P⫽.04). Collegiate athletes demonstrated a significant increase in depression at 14 days postconcussion than did high school athletes (P⫽.03). There were no sex differences in depression levels. Neurocognitive decrements at 14 days were supported for reaction time (P⫽.001) and visual memory (P⫽.001). Somatic depression at 7 days postconcussion was related to slower reaction time at 7 days postconcussion. Somatic depression at 14 days postinjury was related to lower visual memory scores at 14 days postinjury. Conclusions: Although not clinically significant, athletes experienced increased depression scores up to 14 days after concussion that coincided with neurocognitive decrements in reaction time and visual memory. Somatic depression appears to be most salient with regard to lower neurocognitive performance. Mood assessments after concussion are warranted to help monitor and enhance recovery.

From the UPMC Sports Medicine Concussion Program, UPMC Center for Sports Medicine, Department of Orthopaedic Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA (Kontos, Elbin); Department of Kinesiology, Michigan State University, East Lansing, MI (Covassin); and Department of Movement Science, Grand Valley State University, Allendale, MI (Parker). Supported by the National Operating Committee on Standards for Athletic Equipment. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Anthony P. Kontos, PhD, UPMC Sports Medicine Concussion Program, UPMC Center for Sports Medicine, Dept of Orthopaedic Surgery, University of Pittsburgh School of Medicine, 3200 South Water St, Pittsburgh, PA 15203, e-mail: [email protected]. In-press corrected proof published online on May 11, 2012, at www.archives-pmr.org. 0003-9993/12/9310-00143$36.00/0 http://dx.doi.org/10.1016/j.apmr.2012.03.032

Key Words: Brain injuries; Concussion, mild; Depression; Rehabilitation; Symptoms. © 2012 by the American Congress of Rehabilitation Medicine ECENT CONSENSUS STATEMENTS have advocated R for research on the emotional sequelae (eg, depression and anxiety) that accompany sport-related concussion. However, 1

researchers know little about the prevalence, severity, or nature of depression after concussion. In spite of the well-documented relationships between age,2 sex,3,4 and concussion outcomes, researchers have yet to examine the effect of these factors on depression after concussion. Previously, researchers have commented on the similarity between the signs and symptoms (eg, confusion, change in mood, and fatigue) indicative of depression and those observed after concussion.5 This overlap may also increase the difficulty in managing concussion in athletes who present with comorbid depression. Thus, additional research to “disentangle” depression symptoms from other postconcussion symptoms has been advocated for by researchers.5 Clinical depression occurs in approximately 180,000 (ie, 6% of the 1.6 –3.0 million concussions per year in the United States) patients with mild traumatic brain injury or concussion.6 This statistic is likely an underestimation of the true incidence of clinical depression in patients with mild traumatic brain injury because of the underreported nature of concussion. The diagnostic symptoms of depression often overlap with those of concussion (eg, sadness, irritability, fatigue, and sleep problems).5 Depression can adversely affect academic performance,7 social functioning, and baseline neurocognitive test performance.8 Increased incidence of depression has been associated with a history of concussions among retired boxers9 and professional football players.10 While these findings suggest a link between depression and sport-related concussion, these studies focused on chronic effects of concussions on depression; did not control for comorbid explanations for depression in these populations; and relied on retrospective, selfreport methods. Neurobiological and psychosocial factors may account for the relationship between brain injury and depression. For example, individuals with clinical depression have been found to exhibit structural and morphologic changes of the brain’s mood centers involving the hippocampus,11 amygdala, and prefrontal brain regions,12 which may be affected after concussion. In addition, the uncertainty of prognosis (ie, lack of an active rehabilitation), removal from sport, isolation, and lack of social support in dealing with concussion may influence mood in athletes with concussion.13,14 List of Abbreviations ANCOVA BDI-II ImPACT

analysis of covariance Beck Depression Inventory-II Immediate Post-concussion Assessment Cognitive Test

Arch Phys Med Rehabil Vol 93, October 2012

1752

DEPRESSION AND CONCUSSION, Kontos

Elevated levels of depression and mood disturbances have been documented in athletes with concussion.15-17 Mainwaring et al15 used the Profile of Mood States to prospectively examine the emotional function in collegiate athletes with concussion, uninjured collegiate athletes, and nonathletes. Although similar at baseline, the concussion group demonstrated significant postconcussion increases in depression symptoms (ie, nonclinical depression), confusion, and total mood disturbance persisting for 3 weeks before returning to baseline. These findings were recently supported in a sample of athletes with concussion who reported increased levels of total mood disturbance and depression symptoms on the Profile of Mood States that were 3 times greater than their depression scores at baseline.16 Postconcussion changes in total mood disturbance and depression in the preceding sample resolved within 14 and 7 days postinjury, respectively.16 Interestingly, depression symptoms resolved prior to cognitive decrements, which lasted on average 25 days after injury. The researchers concluded that removal from play did not cause depression or emotional disturbances after concussion.16 However, these prospective findings should be interpreted with caution because they were derived from small sample sizes, used a nonclinical measure of depression, and although statistically significant, were of low magnitude. Age (ie, high school vs college) and sex differences in depression after concussion have not been examined, which is surprising given that these variables influence concussion outcomes and may influence depression in general. Females are at a higher risk for concussion18,19 and exhibit more prolonged neurocognitive impairment after concussive injury than do males.20 Females also report higher levels of clinical depression than do males and are at a higher risk for developing postconcussion syndrome, which often includes emotional symptoms such as depression.21 High school athletes demonstrate longer neurocognitive recovery than do collegiate athletes,2 and depression is more prevalent among first-year college athletes than in older college athletes.21 Given the lack of information regarding depression after concussion among male and female high school and college athletes and its potential impact on recovery outcomes, a prospective investigation of these variables is warranted. The current study prospectively examined, while controlling for nondepression concussion symptoms, the relationship of sport-related concussion with depression in male and female high school and college athletes. We expected that depression would increase immediately after concussion but would return to baseline levels within the first 7 to 14 days after injury. We also hypothesized that computerized neurocognitive test performance would be negatively correlated with depression scores. With regard to sex difference, we expected that after concussion females would report more depression than would males. Given the lack of information in the literature, we did not speculate a directional hypothesis with regard to age and depression after concussion. METHODS Research Design A 2-year prospective design was used for the current study. The independent variables were age (high school, collegiate), sex (male, female), and time postinjury (baseline, 2d, 7d, 14d). The dependent variables were total depression scores on the Beck Depression Inventory-II (BDI-II) and computerized neurocognitive test scores (ie, verbal memory, visual memory, reaction time, and processing speed) and symptom scores from Arch Phys Med Rehabil Vol 93, October 2012

the Immediate Post-concussion Assessment and Cognitive Test (ImPACT). Participants A total of 75 athletes from multiple research sites participated in the study. Athletes were recruited and enrolled in the present study if they attended 1 of the participating institutions in an ongoing concussion surveillance program and provided informed consent or parental consent/assent (for minors younger than 18 years). This convenience sample included only those athletes who completed the depression measures at baseline and 2 postinjury time intervals. Inclusion criteria included any athlete who sustained a sport-related concussion that was diagnosed by a sports-medicine professional. Exclusion criteria included a history of treatment for substance abuse, psychiatric disorder, special education, years repeated in school, speech problems, and current score of 20 or greater on the BDI-II at baseline in the current study (these athletes were referred to a mental health professional). Evaluation Measures and Instrumentation Operational definition of concussion. Concussion was operationally defined as “a complex pathophysiological process affecting the brain, induced by traumatic biomechanical forces”1(p186) and was required to meet the after criteria: (1) presence of on-field signs (eg, posttraumatic amnesia and loss of consciousness) and symptoms (eg, dizziness and headache) as determined by a sports-medicine professional trained to identify concussions, (2) decrease from baseline levels in at least 1 postconcussion neurocognitive score determined by reliable change estimates, and (3) increase from baseline levels in postconcussion symptoms determined by reliable change estimates. Beck Depression Inventory-II. The BDI-II is a valid and reliable measure of depression symptoms that consists of 21 self-report items that ask subjects to indicate on a scale of 0 to 3 (except for items 16 and 18) their current status for each item. The BDI-II items cover hopelessness, irritability, cognitions (eg, guilt), and physical symptoms (eg, fatigue). Item scores are totaled for an overall depression score ranging from 0 to 63. Total scores are typically subdivided into minimal (0 –13), mild (14 –19), moderate (20 –28), and severe (29 – 63) depression categories but can be adopted to fit the sample distribution. Given the large numbers of 0s reported in the current sample, we subdivided the minimal group into none (0) and minimal (1–13) groups. In addition, the BDI-II comprises an affective (eg, pessimism, suicidal thoughts, and worthlessness) factor comprising 8 items and a somatic (eg, loss of energy, tiredness/ fatigue, and agitation) factor comprising 13 items. Immediate Post-concussion Assessment and Cognitive Test. The ImPACT computerized neurocognitive test battery was used to assess neurocognitive function and symptoms at baseline and after concussion. The ImPACT comprises 3 general sections that include demographic information, the 22-item Post-concussion Symptom Scale, and 6 neurocognitive test modules. Verbal memory, visual memory, visual processing speed, and reaction time composite scores are provided by ImPACT. Test-retest reliability, validity, and specificity/sensitivity data for ImPACT are reported elsewhere.22-24 Procedures Baseline evaluation. The study was approved by each participating institution’s university institutional review board prior to data collection. Consenting athletes were administered the baseline ImPACT followed by the BDI-II in groups of 10 at their respective institution’s designated computer laboratory.

1753

DEPRESSION AND CONCUSSION, Kontos

Postconcussion evaluation. Each participating institution’s sports-medicine professional referred all athletes with concussion to the researchers for follow-up testing. Athletes with concussion completed the ImPACT and BDI-II at 2 days, 5 to 7 days, and 10 to 14 days or until they were cleared to return to play by a sports-medicine professional. Data Analysis A series of 4 (time: baseline, 2, 7, and 14d) ⫻ 2 (sex) ⫻ 2 (age) repeated-measures analyses of covariance (ANCOVAs) were used to evaluate postconcussion changes on the BDI-II. Ten symptoms (headache, nausea, vomiting, balance problems, dizziness, sensitivity to light, sensitivity to noise, numbness/ tingling, visual problems, and fogginess) listed on the 22-item Post-concussion Symptom Scale were identified as “nondepressive” symptoms on the basis of established diagnostic criteria and consensus among 6 sport concussion clinicians and researchers. These symptoms were totaled and used as a covariate in the depression analyses to account for the potential confounding effect of other concussion symptoms on depression. A separate series of repeated-measures ANCOVAs were also performed to evaluate change across time for the cognitive-affective and somatic subscales of the BDI-II. Nondepressive concussion symptoms were also used as a covariate similar to the preceding analysis. A series of 4 (time: baseline, 2, 7, and 14d) ⫻ 2 (sex) ⫻ 2 (age) repeated-measures ANCOVAs were also used to evaluate ImPACT composite scores and total concussion symptoms. A series of Pearson product-moment correlations were used to assess the relationships among the BDI-II total, cognitive, and somatic scores and between ImPACT composite and total concussion symptom scores. Data were analyzed using SPSS 18.0,a and statistical significance was set at P⬍.05. RESULTS Demographic Results A total of 75 athletes sustained a concussion and completed a baseline and 3 postconcussion BDI-II and computerized neurocognitive assessments. The sample comprised 54 high school (mean age ⫾ SD, 15.74⫾1.28y) and 21 college athletes (mean age ⫾ SD, 19.68⫾1.33y). There were 40 high school males (mean age ⫾ SD, 15.90⫾1.28y) and 14 high school females (mean age ⫾ SD, 15.29⫾SD⫽1.20y), and 11 collegiate males (mean age ⫾ SD, 19.75⫾1.05y) and 10 collegiate females (mean age ⫾ SD, 19.60⫾1.65y) in the study. The sample reported a mean ⫾ SD of 1.08⫾1.29 previous concussions. A series of independent t tests revealed no significant differences in concussion history between males (1.20⫾1.34) and females (.83⫾1.17) (P⫽.26) or between high school (.54⫾1.11) and collegiate (1.43⫾1.66) athletes (P⫽.15). At baseline, athletes were classified as reporting either no depression (BDI-II⫽0; 40 athletes) or minimal depression (BDI-II⫽

Table 2: Means and SDs for Total Depression Scores on the BDI-II (Nⴝ75) Variable

Age High school College Sex Male Female Total

Baseline

2d

1.86⫾2.91 4.73⫾4.18 1.68⫾2.11 4.53⫾4.46

7d

4.00⫾4.05 4.21⫾5.61

14d

2.27⫾3.67 5.21⫾7.00*†

1.85⫾2.68 4.02⫾3.88 3.04⫾3.86 2.30⫾3.76 1.73⫾2.78 6.05⫾4.67 6.18⫾5.07 4.73⫾6.63 1.81⫾2.70 4.68⫾4.23*‡ 4.06⫾4.50*‡ 3.09⫾4.96*‡

NOTE. Values are mean ⫾ SD. *P⬍.05. † Significantly higher than high school at 14 days. ‡ Significantly higher than baseline.

1–13; 35 athletes) (table 1). The average score for the BDI-II at baseline was 1.81⫾2.67, and severe depression was not reported at any time during the study. Athletes reporting moderate depression levels at any of the postinjury time points were referred to mental health professionals at their institutions. Depression Results Means and SDs for postconcussion depression scores are presented in table 2. The results of a 4 (time) ⫻ 2 (age) ⫻ 2 (sex) repeated-measures ANCOVA (covaried for nondepression concussion symptoms) supported a significant withinsubject effect for time (Wilk’s ␭⫽.85, F3,69⫽3.53, P⫽.001, ␩2⫽.15), with athletes exhibiting higher levels of depression symptoms from baseline at 2 days (Pⱕ.001), 7 days (P⫽.006), and 14 days postconcussion (P⫽.04) (fig 1). There were no significant between-subject effects for age (F1,60⫽.59, P⫽.45, ␩2⫽.01) and sex (F1,60⫽3.74, P⫽.06, ␩2⫽.06) on depression. However, there was a significant time by age interaction (Wilk’s ␭⫽.87, F3,69⫽2.98, P⫽.04, ␩2⫽.13) (fig 2). Post hoc univariate analyses revealed that collegiate athletes demonstrated a significant increase in depression at 14 days postconcussion than did high school athletes (P⫽.03). The results of one-way repeated-measures ANCOVAs revealed a significant within-subject main effect for time on the somatic subscale of the BDI-II (Wilk’s ␭⫽.85, F3,60⫽3.58, P⫽.02, ␩2⫽.15). Somatic depression scores on the BDI-II were significantly below baseline at 2 days (P⫽.001) and 7 days (P⫽.001) postconcussion. Somatic depression scores were not significantly different from those at baseline at 14 days postinjury (P⫽.10). There were no significant differences

Table 1: Number of Athletes Endorsing Depression Symptoms Across Data Collection Time Periods (Nⴝ75) Depression Category

None Minimal Mild Moderate Severe

2d 7d 14d Baseline Postconcussion Postconcussion Postconcussion

40 35 0 0 0

16 54 3 2 0

24 49 2 0 0

39 33 2 1 0

Fig 1. Depression scores on the BDI-II after sport-related concussion (Nⴝ75). *P<.05. †Significantly higher than baseline.

Arch Phys Med Rehabil Vol 93, October 2012

1754

DEPRESSION AND CONCUSSION, Kontos

somatic depression was related to lower visual memory scores (r⫽⫺.24) and higher symptom scores (r⫽.26). At 14 days postconcussion, total depression was related to lower visual memory scores (r⫽⫺.24) and somatic depression was related to slower reaction time (r⫽.29). Cognitive depression at 14 days was related to slower reaction time (r⫽.35) and higher symptom scores (r⫽.52) at 14 days. Total depression at 14 days (r⫽.57) was related to higher symptom scores at 14 days.

Fig 2. Age ⴛ time interaction for depression scores on the BDI-II (Nⴝ75). *P<.05. †Significantly higher than high school athletes at 14d.

across time for the cognitive-affective depression subscale of the BDI-II (Wilk’s ␭⫽.95, F3,61⫽1.00, P⫽.40, ␩2⫽.05). Means and SDs for these subscales for each postinjury time period can be found in table 3. Neurocognitive and Symptom Results Results from a series of 4 (time) ⫻ 2 (sex) ⫻ 2 (age) repeated-measures ANCOVAs supported significant withinsubject effects for time for verbal memory (Wilk’s ␭⫽.87, F3,67⫽3.26, P⫽.03, ␩2⫽.13), visual memory (Wilk’s ␭⫽.78, F3,67⫽6.20, P⬍.01, ␩2⫽.22), motor processing speed (Wilk’s ␭⫽.86, F3,67⫽3.73, P⫽.02, ␩2⫽.14), and reaction time (Wilk’s ␭⫽.85, F3,67⫽4.06, P⫽.01, ␩2⫽.15). Athletes demonstrated significantly worse verbal memory performance at 2 days postinjury compared with baseline (P⫽.03). Visual memory performance at 2 days (P⫽.01) and 14 days (P⫽.001) postconcussion was also significantly lower than baseline levels. There were no significant decreases in motor processing speed at 2 days (P⫽.14), 7 days (P⫽.99), or 14 days postinjury (P⫽.99) compared with baseline levels. Reaction time was significantly slower at 2 days postconcussion (P⫽.001) and 14 days postconcussion (P⫽.001) compared with baseline levels. The results supported only 1 between-subject effect for sex on visual memory (F1,69⫽7.44, P⫽.008, ␩2⫽.10), with males (72.28⫾15.57) scoring higher than females (63.87⫾15.21). The results did not support any interactions for sex and age on neurocognitive performance. The results of a 4 (time) ⫻ 2 (sex) ⫻ 2 (age) repeatedmeasures ANCOVA yielded a significant within-subject effect for time (Wilk’s ␭⫽.44, F3,67⫽28.95, P⫽.01, ␩2⫽.56) on total postconcussion symptom scores. Total reported concussion symptoms were significantly higher than baseline symptom reports at 2 days (P⫽.001) and 7 days (P⫽.04) postinjury. There were no significant between-subjects effects or interactions for sex or age for total concussion symptoms. Correlations Between Depression and Neurocognitive Performance and Concussion Symptoms After Concussion The results of a series of Pearson product-moment correlations supported several significant relationships between depression, neurocognitive performance, and concussion symptom scores (see table 3). Specifically at 2 days postconcussion, total depression (r⫽.24) and somatic depression (r⫽.25) were related to slower reaction time. At 7 days postconcussion, Arch Phys Med Rehabil Vol 93, October 2012

DISCUSSION To the best of our knowledge, the current study was the first to prospectively examine the relationship between sport-related concussion, depression, and computerized neurocognitive performance in a combined sample of male and female high school and collegiate athletes. Overall, depression was elevated from baseline levels at 2 days, 7 days, and 14 days postconcussion for all athletes. We hypothesized that levels of depression would increase in the acute time period (ie, 2d) after concussion but would return to baseline levels within 7 to 14 days after injury. This hypothesis was partially supported because statistically, though not clinically, significant increases in depression from baseline were evident at 2 days postinjury. However, the increase in depression levels in the current sample persisted at both 7 and 14 days postconcussion, even after controlling for “nondepressive” symptoms. These results are dissimilar to those of Mainwaring et al16 who reported increases in postconcussion depression only up to 7 days postconcussion. The difference between the results of the current study and the results of the Mainwaring16 study may be attributable to small sample sizes and the use of a nonclinical measure of depression (eg, Profile of Mood States). Moreover, in the current study, we statistically controlled for the potential confounding influence that “nondepression” symptoms may have had on depression levels after concussion. This approach was employed in an attempt to disentangle depression from other postconcussion symptoms, as advocated by Iverson.5 In addition, our results supported relation-

Table 3: Correlations Between Depression Scores and Neurocognitive Performance and Symptoms After Concussion (Nⴝ75) Depression at 2d Variable

Total

r2

Cognitive

Somatic

Verbal memory Visual memory Processing speed Reaction time Symptoms

⫺.17 ⫺.21 ⫺.15 .25* .20

.03 .04 .02 .06 .04

.05 ⫺.04 .05 ⫺.04 .08

⫺.21 ⫺.22 ⫺.17 .28* .20

Verbal memory Visual memory Processing speed Reaction time Symptoms

⫺.10 ⫺.23* ⫺.04 .17 .16

.02 .05 .002 .03 .03

Verbal memory Visual memory Processing speed Reaction time Symptoms

⫺.16 ⫺.18 ⫺.10 .32† .57†

.03 .03 .01 .10 .32

Depression at 7d

.09 ⫺.04 .17 ⫺.07 ⫺.06

⫺.11 ⫺.24* .08 .21 .17

Depression at 14d

*P⬍.05. † P⬍.01.

⫺.12 ⫺.06 ⫺.11 .35† .52†

⫺.16 ⫺.21 ⫺.09 .29* .56†

DEPRESSION AND CONCUSSION, Kontos

ships between depression and lower neurocognitive performance and higher symptoms at several time intervals after concussion. In particular, somatic depression was related to slower reaction time, lower visual memory, and higher symptom scores. In the present study, high school athletes followed an inverted-U pattern of depression with a return to near-baseline levels at 14 days postconcussion, whereas collegiate athletes reported increased depression levels at 14 days postconcussion. The increased depression levels at 14 days postconcussion for the collegiate group may be attributable to factors specific to the collegiate sport environment. College athletes may be more invested than high school athletes in their sport because of high competition levels, stress, and scholarships and thus became more sad/irritable the longer they were out of sport because of their concussion. Increased depression may also be related to frustration from uncertain return-to-play timelines and academic performance difficulties associated with the effects of concussion. In addition, collegiate athletes may experience isolation from significant sources of social support (eg, family and friends), which are purported to alleviate the emotional effects of concussion.25,26 Conversely, high school athletes may have strong social support networks that are readily available (eg, family members and friends) to help them cope with their injury. In the current study, male and female athletes reported similar depression levels after concussion. This finding was unexpected, because females are more likely than males to experience symptoms of depression after an athletic injury.21 This finding also contrasts with previously reported sex differences in neurocognitive performance because females have been found to demonstrate worse neurocognitive outcomes after concussion.3,4,20 Although researchers have not previously examined sex differences in depression after concussion, the current study suggests that sex does not influence postconcussion depression levels. Although not addressed in the current study, diffuse injury to dorsolateral prefrontal cortex and limbic structures in the brain that are responsible for mood regulation may account for increases in depression after concussion.27 A recent study by Chen et al28 reported increased activation in limbic-frontal (eg, dorsolateral prefrontal cortex and striatum) and attenuated deactivation in medial frontal and temporal regions of the brain in athletes with concussion reporting depression symptoms. However, it is important to note that although the increases in depression levels reported in the current and previous studies were statistically significant, they were not reflective of clinical levels of depression. In fact, only 3 athletes in the current sample reported clinical (ie, more than minimal) levels of depression after concussion. While it is plausible that these postconcussion levels of depression may not be different from the normal variation of depressive symptoms that normally occur in these age groups, symptom reports of increased depression after concussion may require a different approach to concussion management that may call attention to other psychosocial factors (eg, stress management/interventions) that could mask prolonged recovery outcomes. Nonetheless, future research should focus on the underlying mechanisms of increases in postconcussion depression to better understand the reason for the current and previous studies’ findings. Neurocognitive impairments in verbal memory and motor processing speed and symptoms were mostly resolved within the first 7 days of injury. However, reaction time and visual memory impairment lingered to 14 days postconcussion. Impairments in reaction time and visual memory were evident up to 14 days postconcussion, whereas verbal memory and motor processing speed decrements resolved at 7 days. Persistent postconcussion impairments in reaction time have been previously documented29 and may reflect greater injury severity

1755

because reaction time has been reported as a significant predictor of recovery time.30 Similarly, visual memory has also been reported to be a significant predictor of recovery from concussion.30 Hence, the relationship of depression, particularly somatic depression, with slower reaction time and lower visual memory scores in the current study suggests that depression after concussion may predict greater cognitive impairment and potentially more protracted recovery times. Prolonged decrements in visual memory may also reflect the increased complexity associated with recalling abstract designs used for visual memory tasks compared with the simpler and more familiar word lists used for verbal memory tasks. Also of note, total reported postconcussion symptoms resolved quicker (ie, 7d) than neurocognitive scores and depression levels (ie, 14d). These data demonstrate the variable nature of concussion outcomes and highlight the need for a multifaceted approach for concussion management.1 The results of the current study indicated that depression scores were related to lower neurocognitive performance on reaction time and visual memory scores and higher reported postconcussion symptoms. These findings suggest that depression scores, even as early as 2 days postconcussion, may be a good indicator of lingering deficits in reaction time and visual memory that are associated with prolonged recovery.30 As expected, depression scores were related to higher symptom scores at each time interval, with higher magnitude relationships at 14 days postconcussion. These relationships are not surprising, given the overlap in depression and postconcussion symptoms. Interestingly, cognitive depression was related to symptoms only at 14 days postconcussion. This finding suggests that cognitive depression may become relevant only when recovery time becomes prolonged and athletes become more frustrated and socially isolated. Study Limitations There were several limitations in the present study. The researchers assumed that athletes were honest and accurate in completing the BDI-II, ImPACT, and Post-concussion Symptom Scale. Some subjects may have been motivated to report lower levels on the BDI-II because of the stigma associated with reporting depression. Only a small subsample of female athletes was included in the analyses, which may have limited the findings regarding sex. The majority (72%) of the sample comprised high school athletes, which may have limited the findings regarding age. In addition, there was no control group used in the present study, which would have provided the researchers additional control over normal variations in depression symptoms in high school and college athletes. Without a control group, we cannot account for the influence that removal from play, isolation from teammates, and diagnosis threat may have had on depression scores in the current study. In addition, we were unable to control for other variables (eg, loss of playing time, diagnosis threat, or social isolation due to removal from play) that may have influenced the relationship between sport-related concussion and depression, because they were not assessed in the current study. Finally, we examined depression only up to 14 days postconcussion. CONCLUSIONS Although there was no evidence of clinical depression, the increased depression scores after concussion documented in the current study warrant consideration from sports-medicine professionals. Specifically, athletes with concussion reporting increased levels of depression, particularly somatic depression, may need additional attention and monitoring to ensure that Arch Phys Med Rehabil Vol 93, October 2012

1756

DEPRESSION AND CONCUSSION, Kontos

depression symptoms do not prolong recovery time. Ignoring the emotional changes that can accompany concussion may negatively affect performance on tools used to manage concussion (eg, symptom reports and neurocognitive testing) and account for poor concussion outcomes. Moreover, given the relationship between depression and slower reaction time and lower visual memory scores in the current study, the assessment of depression may help to identify neurocognitive deficits and inform management of these deficits. The uncertain timeline for recovery and return to play after concussion can cause emotional upheaval for the injured athlete. Sports-medicine professionals should be cognizant of symptoms of depression that may exist after concussion and assess the emotional well-being of the athlete. This assessment can be an informal observation or communication with the athlete with concussion or involve more formal measures of assessing depressed mood (ie, depression inventories). Researchers should further explore the role of depression in specific neurocognitive deficits after concussion to confirm the relationships reported in the current study. Future research should also assess mood beyond 14 days postinjury to extend the current study’s findings regarding depression trajectories after concussion. Finally, additional mood measures (eg, anxiety), neuroimaging (eg, functional magnetic resonance imaging and diffusion tensor imaging), and interventional (eg, cognitive-behavioral, social support, and coping skills) studies for depression after concussion are warranted. References 1. McCrory P, Meeuwisse W, Johnston K, et al. Consensus statement on Concussion in Sport 3rd International Conference on Concussion in Sport held in Zurich, November 2008. Clin J Sport Med 2009;19:185-200. 2. Field M, Collins MW, Lovell MR, Maroon J. Does age play a role in recovery from sports-related concussion? A comparison of high school and collegiate athletes. J Pediatr 2003;142:546-53. 3. Covassin T, Elbin RJ. The female athlete: the role of gender in the assessment and management of sport-related concussion. Clin Sports Med 2011;30:125-31. 4. Colvin AC, Mullen J, Lovell MR, West RV, Collins MW, Groh M. The role of concussion history and gender in recovery from soccer-related concussion. Am J Sports Med 2009;37:1699-704. 5. Iverson GL. Misdiagnosis of the persistent postconcussion syndrome in patients with depression. Arch Clin Neuropsychol 2006; 21:303-10. 6. Jorge R, Robinson RG. Mood disorders after traumatic brain injury. NeuroRehabilitation 2002;17:311-24. 7. Hysenbegasi A, Hass SL, Rowland CR. The impact of depression on the academic productivity of university students. J Ment Health Policy Econ 2005;8:145-51. 8. Covassin T, Elbin RJ, Larson E, Kontos AP. Sex and age differences in depression and baseline sport-related concussion neurocognitive performance and symptoms. Clin J Sport Med 2012;22: 98-104. 9. Erlanger DM, Kutner KC, Barth JT, Barnes R. Neuropsychology of sports-related head injury: dementia pugilistica to postconcussion syndrome. Clin Neuropsychol 1999;13:193-209. 10. Guskiewicz KM, Marshall SW, Bailes J, et al. Recurrent concussion and risk of depression in retired professional football players. Med Sci Sports Exerc 2007;39:903-9. 11. Sheline YI, Wang PW, Gado MH, Csernansky JG, Vannier MW. Hippocampal atrophy in recurrent major depression. Proc Natl Acad Sci U S A 1996;93:3908-13.

Arch Phys Med Rehabil Vol 93, October 2012

12. Chen J-K, Johnston KM, Petrides M, Ptito A. Neural substrates of symptoms of depression after concussion in male athletes with persisting postconcussion symptoms. Arch Gen Psychiatry 2008; 65:81-9. 13. Kontos AP, Collins MW, Russo S. An introduction to sports concussion for the sport psychology consultant. J Appl Sport Psychol 2004;16:220-35. 14. Bloom GA, Horton AS, McCrory P, Johnston KM. Sport psychology and concussion: new impacts to explore. Br J Sports Med 2004;38:519-21. 15. Mainwaring LM, Bisschop SM, Green REA, et al. Emotional reaction of varsity athletes to sport-related concussion. J Sport Exerc Psychol 2004;26:119-35. 16. Mainwaring LM, Hutchison M, Bisschop SM, Comper P, Richards DW. Emotional response to sport concussion compared to ACL injury. Brain Inj 2010;24:589-97. 17. Hutchison M, Mainwaring LM, Comper P, Richards DW, Bisschop SM. Differential emotional responses of varsity athletes to concussion and musculoskeletal injuries. Clin J Sport Med 2009; 19:13-9. 18. Gessel LM, Fields SK, Collins CL, Dick RW, Comstock RD. Concussions among United States high school and collegiate athletes. J Athl Train 2007;42:495-503. 19. Covassin T, Swanik CB, Sachs ML. Epidemiological considerations of concussions among intercollegiate athletes. Appl Neuropsychol 2003;10:12-22. 20. Broshek DK, Kaushik T, Freeman JR, Erlanger D, Webbe F, Barth JT. Sex differences in outcome after sports-related concussion. J Neurosurg 2005;102:856-63. 21. Yang J, Peek-Asa C, Corlette JD, Cheng G, Foster DT, Albright J. Prevalence of and risk factors associated with symptoms of depression in competitive collegiate student athletes. Clin J Sport Med 2007;17:481-7. 22. Schatz P. Long-term test-retest reliability of baseline cognitive assessments using ImPACT. Am J Sports Med 2010;38:47-53. 23. Schatz P, Pardini JE, Lovell MR, Collins MW, Podell K. Sensitivity and specificity of the ImPACT Test Battery for concussion in athletes. Arch Clin Neuropsychol 2006;21:91-9. 24. Elbin RJ, Schatz P, Covassin T. One-year test-retest reliability of the online version of ImPACT in high school athletes. Am J Sports Med 2011;39:2319-24. 25. Rees T, Mitchell I, Evans L, Hardy L. Stressors, social support and psychological responses to sport injury in high- and lowperformance standard participants. Psychol Sport Exerc 2010;11: 505-12. 26. Yang J, Peek-Asa C, Lowe JB, Heiden E, Foster DT. Social support patterns of collegiate athletes before and after injury. J Athl Train 2010;45:372-9. 27. Cullen KR, Gee DG, Klimes-Dougan B, et al. A preliminary study of functional connectivity in comorbid adolescent depression. Neurosci Lett 2009;460:227-31. 28. Chen JK, Johnston KM, Petrides M, Ptito A. Neural substrates of symptoms of depression after concussion in male athletes with persisting postconcussion symptoms. Arch Gen Psychiatry 2008; 65:81-9. 29. Covassin T, Elbin RJ, Nakayama Y. Tracking neurocognitive performance after concussion in high school athletes. Phys Sports Med 2010;38:87-93. 30. Lau B, Lovell MR, Collins MW, Pardini J. Neurocognitive and symptom predictors of recovery in high school athletes. Clin J Sport Med 2009;19:216-21. Supplier a. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.