Table 3.
Summary of findings
| Study | Concussion profile; n (%) reporting concussion, mean concussions | Mental health outcomes assessed (assessment tool) | Timing of mental health assessment relative to concussion | Main mental health findings |
|---|---|---|---|---|
| Banks et al. [15] | n = NRa; mean concussions for boxing = 0.4, range 0–11; mean concussions for mixed martial arts = 0.6, range 0–10 | Impulsivity (BIS-15) | Self-reported lifetime concussions, sample did not experience recent concussion exposure | Athletes, compared to healthy male controls, had significantly lower total impulsiveness scores (60.5 vs 62.8; p = 0.012), lower self-control scores (11.5 vs 12.4; p = 0.032), and lower cognitive instability scores (5.3 vs 6.4; p < 0.001) and higher cognitive complexity scores (12.3 vs 11.3; p < 0.001). In MMA fighters, each increase of 10 years of fighting decreased attention scores by 1.5, p = 0.02 |
| Casson et al. [16] | n = NRa; mean concussions = 6.9 | Depression (BDI; PHQ-9) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | A total of 15 retired athletes (33%) reported any severity of depression; this was greater than the prevalence in the general population (i.e. 15–20%). Those with moderate-severe depression (n = 6; 13.3%) were consistent with general population rates. Similarly, 9 participants (20%) met PHQ criteria for depression, consistent with general population rates |
| Covassin et al. [17] | n = 63 (50%); mean concussions = NR | Anxiety (STAI) | Within 1 week of sustaining injury, either concussion or orthopaedic | Athletes in the concussion and orthopaedic injury groups reported equivalent state and trait anxiety, p = 0.193. The two groups were consistent in sources of social support sought during injury |
| Decq et al. [18] | n = 217 (68.5%)a; mean concussions for RPs = 3.1; mean concussions for OS = 0.68 | Depression (PHQ-9) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | Retired athletes RPs reported more lifetime concussions than OS, p < 0.001, and a higher incidence of mild depression (PHQ-9 > 9) (9.55% vs 6.72%), p = 0.04. PHQ-9 score increased with the number of repeated concussions, regardless of the type of sport, p = 0.026. Daily alcohol consumption was reported more frequently by OS than RPs (31 vs 20%), but more RPs reported heavy drinking (5 glasses a day) than OS (88 vs 71%) |
| Didehbani et al. [19] | n = 29 (50%)a; mean concussions for players = 3.97 | Depression (BDI-II) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | Retired athletes had higher total BDI-II scores (mean = 8.80, SD = 8.33) than control group (mean = 2.83, SD = 3.95), p < 0.001. In retired athletes lifetime concussions and total scores on the BDI-II were significantly correlated (r = 0.43, p = 0.02) and remained significant when cardiovascular risk factors, headaches, and arthritis were entered as covariates |
| Guskiewicz et al. [20] | n = 1513 (60.7%)a; mean concussions = NR | Depression (self-reported history of diagnosed depression) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | Retired athletes reported a retrospective self-reported lifetime depression diagnosis rate of 11.1% (of the 2434 respondents). There was an association between recurrent concussion and diagnosis of depression, p < 0.005, with a significant test for linear trend suggesting that the prevalence increases in a linear fashion with increasing concussion history. Relative to retired athletes with no concussion history, retired athletes reporting a history of ≥ 3 previous concussions were three times more likely (prevalence ratio of 3.06) to be diagnosed with depression, and those with a history of 1 or 2 previous concussions were 1.5 times more likely (prevalence ratio of 1.48; 95% CI 1.08–2.02) to have been diagnosed with depression |
| Hart et al. [21] | n = 34 (94.1%)a; mean concussions = 4.0 | Depression (clinician diagnosis, BDI-II) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | A quarter (n = 8, 24%) of the retired athletes were diagnosed with major depression, 6 of whom had not been previously diagnosed or treated. The prevalence of depression amongst the retired players (24%) was higher than expected for this age group (approximately 15%) |
| Hutchison et al. [22] | n = 20 (37.7%); mean concussions = 0.90 | Tension, depression, anger, vigour, fatigue, confusion, and self-esteem (POMS) | Within 4 days, 1 and 2 weeks post-concussion, included a baseline assessment | Athletes experienced short-term emotional reactions after concussion that were different from that of musculoskeletal injury. Factorial ANOVAs indicated that concussion produced an emotional profile characterised by significantly elevated fatigue and decreased vigour, and short-term mood disturbance. In contrast, musculoskeletal injury was associated with anger |
| Kerr et al. [23] | n = 307 (38.8%)a; mean concussions = NR | Mental health composite (VR-12) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | There was a non-significant trend level association (p = 0.06) reported between number of concussions reported by retired athletes and their mental health composite scores. Those without a concussion history reported marginally higher (i.e. better) mental health scores (mean = 53.4, SD = 5.5), relative to those with 1–2 concussions (mean = 50.7, SD = 10.1), or ≥ 3 concussions (mean = 50.4, SD = 10.5) |
| Kerr et al. [24]b | n = 307 (38.8%)a; mean concussions = NR | Depression (PHQ-9), impulsivity (BIS-15), aggression (BPAQ-SF) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | In retired athletes reporting ≥ 3 concussions, the prevalence of moderate-to-severe depression was 2.4 times that of former collegiate athletes reporting zero concussions (95% CI 1.0–5.7; controlling for alcohol dependence, family history of depression). Those reporting ≥ 2 or ≥ 3 concussions had significantly higher mean impulsivity scores compared to those reporting no concussions. Similarly, those reporting ≥ 3 concussions had significantly higher mean score for aggression, compared to those reporting no concussions |
| Kerr et al. [25] | n = 679 (65.0%)a; mean concussions = NR | Depression (self-reported history of diagnosed depression), wellbeing (SF-36) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes), mental health assessed at baseline and follow-up (9 years post-baseline) | Retired athletes self-reporting concussions were at greater risk of experiencing depressive episodes during the 9-year follow-up relative to retired athletes self-reporting no concussions. A total of n = 106 (10.2%) reported receiving a diagnosis of depression between the baseline and follow-up period. The 9-year risk of a depression diagnosis increased with number of self-reported concussions, ranging from 3.0% in the “no concussions” group to 26.8% in the “10+” group (p < 0.001). A strong dose-response relationship was observed even after controlling for confounders |
| Kontos et al. [26] | n = 75(100%); mean concussions = 1.08 (prior to current exposure) | Depression (BDI-II) | Within 2, 7, 14 days post-concussion, included a baseline assessment | Athletes post-concussion exhibited significantly higher levels of depression as assessed on the BDI between baseline (mean = 1.68, SD = 2.11), and at 2 days (mean = 4.52, SD = 4.46), 7 days (mean = 4.21, SD = 5.61), and 14 days (mean = 5.21, SD = 7.00), p < 0.05 |
| Mainwaring et al. [27] | n = 16 (31.4%); mean concussions = 1.4 | Tension, depression, anger, vigour, fatigue, confusion, and self-esteem (POMS) | Within 4, 7, 14 days post-concussion, included a baseline assessment | Pre-injury POMS performance in athletes was not a risk factor for concussion. There was a significant acute spike for depression, confusion, and total mood disturbance at 4 days post-concussion. These increases were transient and appeared to resolve by 14 days |
| Mainwaring et al. [28]c | n = 16 (5.1%); mean concussions = NR | Tension, depression, anger, vigour, fatigue, confusion, and self-esteem (POMS) | Within 4, 7, 14 days post-concussion, included a baseline assessment | Athletes with concussion and ACL injury reported significant increases in depression scores post-injury compared with uninjured controls. Athletes with ACL injury reported higher levels of depression for a longer duration than athletes with concussion |
| Meehan et al. [29] | n = 836 (22.6%)a; mean concussions = NR | Anxiety, depression, emotional and behavioural dyscontrol, positive affect, sleep disturbance (Neuro-QoL), alcohol use (PROMIS) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | Respondents with a history of concussion self-reported worse health on several measures, including positive affect, measures of anxiety, depression, negative consequences of alcohol use, sleep disturbance, emotional and behavioural dyscontrol, and fatigue (p values < 0.001) |
| Meier et al. [30] | n = 17(38.6%); mean concussions = 1.0 (prior to current exposure) | Anxiety (HAM-A), depression (HAM-D) | Within 3 days (T1: n = 17), 13 days (T2: n = 15), 44 days (T3: n = 13) post-concussion, included a baseline assessment | Compared to healthy controls, athletes with concussion showed partial recovery of clinically assessed mood symptoms by T2 and T3 relative to T1 post-concussion; however, mood symptoms remained elevated throughout subacute assessment phase. Anxiety was significantly higher for the post-concussion group relative to healthy athletes at T1 and T2 (p values < 0.001) and trending higher at T3 (p < 0.10). Depression was significantly higher for the post-concussion group relative to healthy athletes at T1, T2 (p values < 0.001) and T3 (p = 0.01). There was no significant main effect or interaction of prior concussions on recovery of anxiety or depression (p values > 0.10) |
| Meier et al. [31] | n = 40 (100%); mean concussions = 0.93 (prior to current exposure) | Anxiety (HAM-A), depression (HAM-D) | Within 7 days (1.92 days, SD 1.04) post-injury | Concussed athletes significantly underreported post-concussive symptoms to their athletic trainers according to the ImPACT Post-Concussion Scale. Standardised measures of depression and anxiety (p values < 0.001) were higher when reported in a confidential setting |
| Meier et al. [32] | n = 44 (45.7%); mean concussions = 0.75 (prior to current exposure) | Anxiety (HAM-A), depression (HAM-D) | Within 1 day (T1: 1.74, SD 0.93; n = 34); 7 days (T2: 8.44, SD 2.15; n = 34); 30 days (T3: 32.47, SD 4.68; n = 30) post-concussion | Concussed athletes showed improvement in mood symptoms at each time point, but had significantly higher mood scores than healthy athletes at every time point; concussed athletes had higher depression scores relative to healthy athletes at T1 and T2 (p values < 0.001) and T3 (p = 0.003). Concussed athletes had higher anxiety scores relative to healthy athletes at T1, T2 (p values < 0.001) and T3 (p = 0.033) |
| Montenigro et al. [33] | n = NRa; mean concussions = NR; median concussions = 20 | Depression (CES-D), apathy (AES) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | The cumulative head impact index predicted later-life clinical outcomes, outperforming other individual metrics such as concussion history, age at first exposure to American Football, and total duration of play. A dose-response relationship between estimated cumulative head impact exposure all and later-life risk for neurobehavioral impairment was observed. Risk of developing behavioural dysregulation, depression, and apathy nearly doubled with 2800 additional impacts above the threshold |
| Poltavski and Biberdorf [34] | n = 17 (40%)a; mean concussions = NR | ADHD (ASRS) | Self-reported lifetime concussions, currently competing athletes, pre-season assessment | Athletes with a self-reported history of concussion scored significantly higher for ADHD checklist (mean = 10.24, SD = 2.77) relative to athletes without a history of concussion (mean= 7.88, SD = 4.01) (p = 0.04) |
| Pryor et al. [35] | n = 27 (100%)a; mean concussions = NR | Depression (CES-D) | Self-reported lifetime concussions, currently competing athletes, pre-season assessment | Individuals with a CES-D score of ≥ 16 sustained a significantly greater number of lifetime concussions (3.8 vs 1.6) (p < 0.001). Significantly higher CES-D scores were observed in players who had sustained ≥ 3 concussions (24.0 vs 15.6) than those with ≤ 2 (p = 0.03) |
| Putukian et al. [36] | n = 32 (12.2%); mean concussions = NR | Anxiety (GAD-7), depression (PHQ-9) | Baseline (T1) and post-concussion (T2; mean = 283.3, SD = 259.6 days) | There was no significant interaction (p values < 0.1) between time of assessment (T1, T2) and group (concussed, control) for either anxiety or depression. Main effects were not reported |
| Roiger et al. [37] | n = 7 (33.3%); mean concussions = NR | Depression (CES-D) | 1 week, 1 and 3 months post-concussion, included a baseline assessment | Concussed athletes had higher depression symptoms 1 week post-concussion (mean = 11.0, SD = 5.3) compared to baseline (mean = 6.7, SD = 3.9), p = 0.02. There were no significant differences between baseline depression and depression at 1 or 3 months. There were no differences in depression between the concussed and injured groups at any time point |
| Singh et al. [38] | n = 18 (50%); mean concussions = 1.11 (prior to current exposure) | Anxiety (HAM-A), depression (HAM-D) | 3 days, 1 week, 1 month post-concussion | Concussed players reported significantly higher depression and anxiety at 3 days (p values < 0.001), and at 1 week (p values < 0.001), post-concussion relative to healthy athletes. At 1-month, depression remained higher in the post-concussion group (p = 0.041), though there was no difference between the groups at this time for anxiety |
| Strain et al. [39] | n = NRa; mean concussions = 3.43 (asymptomatic group); mean concussions = 5.6 (symptomatic group) | Depression (BDI-II) | Self-reported lifetime concussions, sample did not experience recent concussion exposure (retired athletes) | Of the 26 retired athlete participants, 5 (19.2%) were identified as currently symptomatic for depression (scoring > 18 on the BDI-II). Those in the symptomatic group reported on average 5.6 (SD = 3.29) lifetime concussions, relative to 3.43 (SD = 2.87) for those asymptomatic, although this difference was not statistically significant |
| Vargas et al. [40] | n = 84 (65.6%); mean concussions = 0.92 (prior to current exposure) | Depression (BDI-FS) | 2 days (n = 36), 5 days (n = 60), 1–7 weeks (n = 19) post-concussion, included a baseline assessment, controls re-assessed mean = 6.8 weeks post-baseline | For those in the post-concussion group, a total of 9 of 84 athletes (11%) at baseline and 19 of 84 athletes (23%) post-concussion scored above the cut-off (BDI-FS > 3) (p = 0.02). The difference between T1 and T2 for the controls was not significant (n = 3; 7% vs n = 4; 10%), p > 0.9. Significant baseline covariates of post-concussion depression were baseline depression (p = 0.03) and age first played sport (p = 0.005) |
| Yang et al. [41] | n = 67 (100%); mean concussions = NR | Anxiety (STAI), depression (CES-D) | Within 1 week, and depending on injury duration also 1, 3, 6, 9, 12 months post-concussion, included a baseline assessment | Concussed athletes who had symptoms of depression at baseline (pre-injury) were 4.59 times more likely (95% CI 1.25–16.89) to experience depression symptoms and 3.40 times more likely (95% CI 1.11–10.49) to experience state anxiety following the concussion, compared to concussed athletes who had no symptoms of depression at baseline. Concussed athletes with baseline (pre-injury) trait anxiety did not have increased post-concussion depression and state anxiety symptoms. Post-concussion symptoms of depression significantly co-occurred with post-concussion state anxiety (OR = 8.35, 95% CI 2.09–33.34) |
ACL anterior cruciate ligament, ADHD attention deficit hyperactivity disorder, AES Apathy Evaluation Scale, ANOVA analysis of variance, ASRS Adult ADHD Self-Report Scale, BDI Beck Depression Inventory, BDI-FS Beck Depression Inventory—Fast Screen, BDI-II Beck Depression Inventory 2, BIS-15 Barrett Impulsiveness Scale, BPAQ-SF Buss-Perry Aggression Questionnaire—Short Form, CES-D Center for Epidemiologic Studies Depression Scale, CI confidence interval, GAD-7 Generalized Anxiety Disorder-7 Item Scale, HAM-A Hamilton Anxiety Rating Scale, HAM-D Hamilton Depression Rating Scale, ImPACT Immediate Post-Concussion Assessment and Cognitive Test, MMA mixed martial arts, Neuro-QoL Quality of Life in Neurological Disorders, NR not reported, OR odds ratio, OS other sports, PHQ-9 Patient Health Questionnaire, POMS Profile of Mood States, PROMIS Patient-Reported Outcomes Measurement Information System, RP rugby player, SD standard deviation, SF-36 Short Form 36 Measurement Model for Functional Assessment of Health and Well-Being, STAI State-Trait Anxiety Inventory, T1 time 1, T2 time 2, T3 time 3, VR-12 Veterans RAND 12-Item Health Survey
aRetrospective report of concussion (i.e. > 4 weeks elapsed between concussion exposure and mental health assessment) or unable to determine time elapsed between exposure and assessment
bSame sample as Kerr et al. [23]
cSame sample as Mainwaring et al. [27]