Sports-related traumatic brain injuries (TBIs) are an important public health concern. Annually, approximately 1.6 to 3.8 million sports-related TBIs occur in the United States alone.1–5 The indirect and direct costs of these TBIs are piling up to a total of $76.5 billion every year.6
American football is the most popular sports in the United States.7 Football is known for its aggressive, fast-paced character which is prone to causing injuries, including head trauma. In fact, recent studies have shown that football accounts for the highest proportion of concussion in all sports.4,8 The incidence in professional athletes playing in the National Football League (NFL) is estimated at 4.56 concussions per 1000 athletic exposures (AEs). By contrast, the prevalence in nonprofessional football ranged from 0.81 to 1.03 per 1000 AEs.8 In the past years, attention is increasingly raising on the long-term effects of TBIs, especially mild traumatic brain injury (MTBI), also called concussion. Interestingly, in 2013 over 4500 former NFL players suffering from long-term effects of concussion filed a lawsuit against the NFL, accusing them of denying the link between football and brain injury. They claimed that they were not informed about the health consequences of TBI.9 The NFL reached a settlement of $765 million with all the retired players with medical evidence of brain injury.9 Also the recent deaths of several former American football athletes, of which some were achieved by suicide, have raised public and scientific attention.6
Several reviews have appeared on the long-term consequences of concussion in sports in general.1,8,10–12 They repeatedly reported psychological, cognitive, and somatic changes as a result of repetitive concussions. However, few have solely evaluated American football. To be more specific, they have not yet evaluated sports-related concussion from an occupational point of view. Because TBI occurs more often in professional football players and is caused by the profession itself, it can be called as an occupational disease. The purpose of this review is to inventory aspects of functioning during and after the career of professional American football players who are affected by TBIs. We will evaluate the physical, cognitive, psychological, and sports-related functional effects of concussion. The latter include return to play and on-field performance.
The answer to limiting the extent of this public health concern is prevention of TBI. One of the first steps in prevention is proper informing about the consequences of TBI to all those involved in professional American football, such as players, coaches, team physicians, and clinicians. To be able to provide clear and up-to-date information, research is needed. Furthermore, if concrete aspects of functional impairments because of TBI are identified, early detection of functional problems in retired players with a concussion history would be conceivable, as well as early treatment of symptoms.
The databases PubMed and SPORTDiscus were systematically searched for articles published between January 1990 and January 2015. We used a set of MeSH terms and keyword terms, divided into 3 concepts, each representing a part of the research question. The terms in the concepts, respectively, referred to professional American football players, all forms of TBI and 4 functional outcome variables. Table 1 provides an overview of the complete search strategy.
Selection of studies was performed using a set of criteria for inclusion and exclusion, which are listed in Table 2. To be as inclusive as possible, very little restrictions were made on study design. Only reviews were excluded, but references were checked. Selection of studies was performed by first removing duplicate studies whereupon title and abstract were screened. Then, the remaining articles were screened for full-text, applying criteria for inclusion and exclusion. When the first reviewer (B.C.V.) was uncertain about including a study, a second reviewer (K.N.) assessed and included or excluded it.
To assess the methodological quality of the included studies, a 5-item critical appraisal checklist was designed, based on existing checklists on observational studies from the STROBE-statement (STrengthening the Reporting of OBservational studies in Epidemiology), checklists for evidence-based guideline development available at the Dutch Cochrane Center and examples from former observational studies.13 We designed a new checklist because of the lack of checklists for evaluating both longitudinal and cross-sectional designs. We wanted to use a checklist that was applicable for all included study designs. All studies were assessed and were assigned either a “+,” a “−”, or a “±” for each critical appraisal item, resulting in a 5-point scale. Studies were considered to be low quality for scores of ≤2.5, moderate quality for scores between 3 and 4, and high quality if they scored ≥4. In the critical appraisal, the studies were checked for selection bias, information bias, and correct reporting of the study characteristics and the severity of the exposure. An overview of the critical appraisal checklist can be found in Table 3.
After critical appraisal, data were extracted and summarized. These results can be reviewed as Supplemental Digital Content 1 (see Appendix 1, http://links.lww.com/JSM/A129).
To synthesize the results of the studies, a data analysis was conducted on the findings. Because there is great heterogeneity between the studies, they were considered not suitable for meta-analysis. The extracted data were divided into themes, which correlated with the different functioning outcomes. This thematic data analysis is based on an example from Gärtner et al, 2010.13
Five levels of evidence were distinguished, based on study design, number of tested relations per theme, and percentage of significant relations per theme. A longitudinal study was considered as a higher form of evidence, compared with cross-sectional studies. Narrative studies (case studies) were included in the evidence synthesis to perceive a complete overview but did not contribute to the level of evidence because of low methodological quality. The 5 different levels of evidence are based on Gärtner et al, 201013 and are described below:
- One or more significant relations in the same direction are found in (1) longitudinal study/studies. More than 50% of the tested relations are significant.
- Three or more significant relations in the same direction are found in cross-sectional studies. More than 50% of the tested relations are significant.
Two significant relations in the same direction are found in cross-sectional studies. More than 50% of the tested relations are significant.
One significant relation is found in a cross-sectional study. More than 50% of the tested relations are significant.
Less than 50% of the tested relations within a theme are significant.
Within a theme, significant relations in different directions are found.
The literature search yielded a total of 247 studies, of which 21 remaining studies were eligible for inclusion. A flow diagram of the whole selection process is depicted in Figure 1.
Characteristics of Included Studies
Detailed information about the reference, study population, methods, and results of the studies can be found in the Supplemental Digital Content 1 (see Appendix 1, http://links.lww.com/JSM/A129). Included are 14 cross-sectional studies, 2 case reports, and 5 longitudinal studies. Sample size ranged from 1 (21, 22) to 2552 (29) participants. All participants were men. Eight studies examined active players,3,14–20 and these active players were aged between a mean of 20.16 and 27.6 years. The age of retired players ranged between a mean of 44 and a mean of 71 years. Educational level was mainly the same across the studies, ranging from a mean of 15.6 to 16.5 years of education. All studies examined MTBI and the method for measuring MTBI was usually based on self-report, diagnosis by a team physician, or use of the Sport Concussion Assessment Tool 2. Concussion history differed greatly between studies. Across the studies, the mean number of concussions ranged from 0.65 to 8.24. However, 1 study6 reported a mean of 350.8 concussions, with a maximum of 20 000 concussions experienced by a single player. The methods for measuring the outcome varied between studies, but the most used methods were the Beck Depression Inventory II (BDI-II) for measuring depression and the Immediate Post-concussion Assessment and Cognitive Testing (ImPACT) tool for measuring cognitive functioning.
The results of the critical appraisal are shown in Table 4. Thirteen studies were considered of high quality. Six studies were of moderate quality. The 2 case reports21 had a low methodological quality and were therefore not included in the evidence synthesis.
Aspects of Functioning
The levels of evidence per individual theme are shown in Table 5.
There were 9 studies examining the psychological function.2,21–28 In all studies, psychological functioning was called as depression. These studies were performed in a relatively old sample of players, ranging from a mean age of 44 to 71 years. All studies examined a sample of retired players, except for Amen et al (2011),22 who studied a group including younger active players. Most of the studies used BDI-II as a method for measuring depression. Of the 9 studies, 5 showed a significant relation between exposure to concussion and depression.2,21–24 Guskiewicz et al (2007)23 revealed that retirees with 1 or 2 concussion had an RR of 1.5 to have been diagnosed with depression, and retirees with 3 or more concussions had an RR of 3 to have been diagnosed with depression, relative to retired players with no concussion history (P < 0.005). Kerr et al (2012)24 examined the 9-year risk of a depression diagnosis and found a significant linear dose–response relationship between the number of concussions and risk of depression diagnosis (P < 0.001). Overall, there is strong evidence that there is a relationship between MTBI and depression later in life. In addition, these results are highly suggestive that there is a positive dose–response relationship between the number of concussions endured by a player and the risk of depression diagnosis.
Three studies examined the relation between concussion and sports-related functioning.3,14,16 This functional outcome was translated to “on-field performance” and “return to play.” All studies had a longitudinal design and studied active players. None of the studies found a significant relationship between concussion and a sports-related outcome measure. Kumar et al (2014)3 found no significant difference between preinjury and postinjury on-field performance. Pellman et al (2006)14 showed that players recover quickly from acute concussion and that they can return to play within a week. Overall, there is inconclusive evidence that there is an association between concussions and impaired sports-related functioning.
The relation between concussions and physical functioning was subject of 3 studies.17,18,27 Two of these studies showed significant relations between concussion and physical functioning.17,18 These studies showed that concussed players had significantly more deficits in balance control and had lower voluntary muscle activation and a declined muscle force, compared with their healthy counterparts. We can conclude that there is strong evidence that concussion is related to a declined physical function.
For cognitive functioning, there was such a great variety in outcome variables that we decided to split this category up into 5 subthemes of cognitive functioning: memory impairment, mild cognitive impairment (MCI), executive dysfunction, reaction time, and visual-motor speed. In total, 13 articles focused on cognitive functioning. With all subthemes combined, the overall evidence for the effect of concussion on cognitive functioning is inconclusive.
Mild Cognitive Impairment
Mild cognitive impairment is a diagnostic classification to describe cognitive decline in usually older individuals. It is frequently described as a transitional state between the cognitive impairments that normally come with age, and dementia. The cognitive decline found in MCI is usually in the domain of memory, but it can also be a generally lower score on neurocognitive testing.29 Three cross-sectional studies examined MCI.22,26,29 Two of the studies found a significant correlation between concussion exposure and MCI.22,29 Guskiewicz et al (2005)29 found that retired players with a concussion history of ≥3 concussions were twice as likely to have MCI, compared with players with a history of 1 to 2 concussions, and 5 times more likely to have MCI compared with players with no concussion history whatsoever (P = 0.02). Hart et al (2013)26 compared cognitively impaired players with nonimpaired players, and found no differences between concussion history in these groups. In conclusion, we found moderate evidence that a history of concussion is associated with MCI.
Nine studies examined the relation between concussion and memory impairment.4,6,14,15,19–21,27,29 Several methods for measuring memory impairment were used. These are described in Supplemental Digital Content 1 (see Appendix 1, http://links.lww.com/JSM/A129). Four studies found a significant relation between concussion and memory impairment.6,21,27,29 Some interesting significant findings were that retired players with ≥3 previous concussions had an RR of 3 to report significant memory problems, compared with retired players without a history of concussion.29 Seichepine et al (2013)6 found significantly impaired working memory in retired players, compared with normal healthy adults. Because of the great proportion of nonsignificant results, we conclude that there is inconclusive evidence for the relationship between concussion and memory impairment.
Four studies examined reaction time,14,19,20,25 which is a component of the ImPACT. Only Casson et al (2014)25 found a significant relationship between concussion history and a slower reaction time. They studied a sample of retired NFL players. All nonsignificant studies had active NFL players as their subjects. There is inconclusive evidence for a relationship between concussion and a slower reaction time.
Five studies focused on the association between visual-motor speed,14,15,19,20,25 for example, the processing speed of visual input into motor output, which is also a composite score of the ImPACT. Again, only Casson et al (2014),25 the only study examining retired players, showed a significant relation between concussion and an impaired visual-motor speed, and the evidence for this relation is therefore inconclusive.
The association between concussion and executive dysfunction, or the inability to perform complex tasks, was subject of 3 studies.5,6,21 Two studies6 found a significant positive relation between concussion and executive dysfunction. Hampshire et al (2013)5 used functional MRI whereas the subjects were performing “the Spatial Planning Task” (SP-task), a test commonly used to asses executive functioning. Compared with controls, the NFL alumni did not perform significantly worse on the SP-task. Seichepine et al (2013)6 compared NFL retirees with normal healthy adults, evaluating 9 areas of executive functioning. National Football League retirees scored significantly worse on 7 of the 9 areas. Because of the low number of significant findings, it can be concluded that there is limited evidence for the association between TBI and executive dysfunction.
This systematic review found strong evidence for the association between repetitive brain trauma in professional American football players and depression later in life. There is also strong evidence for the correlation with physical dysfunctions. However, these dysfunctions were subclinical of nature and the results for the association between concussion and sports-related functioning showed that it did not affect players' time to return to play or their on-field performance. The evidence for the effect of concussion on several aspects of cognitive functioning remains inconclusive, although there was moderate evidence for the relationship between concussions and MCI.
Interestingly, when we looked at players' ages, the most significant associations were found in the older age groups, involving retirees. The overall effect for cognitive dysfunctions was found to be inconclusive, but when the studies involving active players are removed from the evidence synthesis, there would have been strong evidence for the relation between TBI and cognitive dysfunctions. Moreover, the strongest relation we have discovered, between concussion and depression, all occurred in retired players older than 45 years. Depression is a common consequence of brain injury. It has even been reported as a consequence of a single mild traumatic injury,30 but the biological or psychological pathway of that association has not yet been clarified. One possible explanation of the relationship between repeated concussions and depression can be found in the occurrence of a disease called Chronic traumatic encephalopathy (CTE). Chronic traumatic encephalopathy is a neurodegenerative disease, characterized by atrophy of various regions of the brain and deposition of tau protein, also called tauopathy.27 It was first identified in boxers as a long-term effect of repetitive TBI, and later also in other contact sports such as hockey, soccer, martial arts, and American football.31 Chronic traumatic encephalopathy causes various behavioral symptoms, cognitive decline, and mood disorders, such as depression, suicidality, irritability, disinhibition, aggression, and substance abuse.27 Remarkably is that these symptoms typically present in the fourth to fifth decade of life, after a latency period of decades after the exposure to multiple TBIs. The disease can eventually develop into dementia.6,27 This supports our findings of a higher chance of depression later in life and a higher risk for developing MCI in older individuals. However, CTE may be the explanation for the symptoms we observe and can be suspected clinically, CTE remains a pathological diagnosis, and it can only be diagnosed at autopsy in deceased individuals.21 Therefore, early detection of this disease is not yet possible which may impede early clinical diagnosis and associated early interventions.
Because we considered TBI in football as an occupational disease, the question arises whether concussion really does affect optimal functioning in work. As we revealed, performance in current work of players is not affected. However, depression can affect an optimal performance in work after retirement from the NFL. Depression can negatively influence one's ability to maintain relationships, to be productive and concentrated at work, and to maintain an acceptable level of self-care.24 Furthermore, it is expected that MCI in retired players is also not conducive to optimal work functioning, as it is accompanied with memory loss and cognitive decline.
Limitations and Strengths
There are a number of limitations to this systematic review. The main limitation is the great heterogeneity between the studies. The included studies differ in sample size, concussion history, outcome measures, and measurement methods. This heterogeneity makes good comparison difficult. Another limitation is that more than half of the studies had a relatively small population size of <100 participants. It is not known whether larger population sizes would have shown the same results. However, the 3 largest studies23,24,29 with over 1000 participants, all showed a significant dose–response relationship between concussion and a functional outcome. There were also limitations to the individual included studies. First, almost all information about concussion history was based on self-report measures. It can be questioned whether the participants recalled the number of concussions they endured correctly, especially when there might be memory impairments in some players. One study participant even reported 20 000 concussions retrospectively.6 The recall of symptoms and functional problems may be inaccurate as well. This might underestimate or overestimate the results. Second, many of the methods for measuring functional outcomes were based on questionnaires and tests performed by the player itself. Participants were rarely assessed by a blinded physician or researcher. Participants were obviously aware of their concussion history, which could have affected their responses on questionnaires and tests. Third, selection bias may have played a role in an overestimation of the results. Many studies recruited their participants from the NFL players' association, which was almost always through self-referral, local advertising, and word of mouth. Because (retired) athletes were aware of the study purpose, athletes who were already experiencing functional impairments may have been more likely to apply to a study.
The main strength of this systematic review is the broad search strategy in 2 databases, with almost no limitations to study design and no less than 4 outcome measures. This is, to our knowledge, the first review to assess and compare such a broad range of functional outcome measures. With 21 included studies, we can give a complete overview of the literature on this subject so far. Another major strength is that levels of evidence were used to synthesize the results, which enabled us to create more transparency and draw more objectivized conclusions.
Implications for Research
In our included studies, the age group from 30 to 45 years was under-represented, and we recommend more studies in this age group. The typical career of an American football player ends at an average age of 28 years.32 Therefore, the 30 to 45 years age group is the postathletic period in which the retirees need to start pursuing some other career. As for now, it remains unclear whether the consequences of TBI affect functions in this age group and whether this would make it harder for these retired players to start a new career. Furthermore, more longitudinal studies on long-term effects of concussion on cognitive and physical functioning are needed. Only 1 study,24 which assessed depression, had a long follow-up period of 9 years. Longer follow-up periods will enhance the evidence for certain associations. Finally, the use of more heterogeneous populations and standardized outcome measures in future research would enable meta-analyses to be performed.
Implications for Practice
This review may be of practical relevance for American football players, coaches, team physicians, and future employers of NFL retirees. It is clear that the consequences of TBI need to be limited. The first step into tackling this problem is prevention of TBIs. This can be achieved by properly informing players and their coaches. Raising awareness may cause players to be more careful during practice or games and to make more informed choices about their health. Furthermore, this review provides input for the NFL to develop more stringent game rules preventing high-impact collisions, to increase medical surveillance and to design better protective equipment. These results can also be used to achieve early detection and intervention of consequences in retired athletes. Future employers can take into account that employees with a history of concussions may experience early cognitive or psychological decline and apply accurate interventions. Finally, the findings of this study may be applicable for other high-contact sports.
There is strong evidence for a relationship between a history of concussion in American football and depression diagnosis later in life, and there are strong hints that this is a dose–response relationship. Also cognitive dysfunctions such as MCI are seen in older American football players with a history of TBI. These dysfunctions may be related to occurrence of Chronic Traumatic Encephalopathy in these players. On the short term, concussion mainly causes subclinical physical dysfunctions like balance control deficits and decreased muscle force. These findings provide insights on the consequences of TBI in other high-contact sports, as well as input for actions to prevent TBI and their consequences in American football players.
1. Jordan BD. The clinical spectrum of sport-related traumatic brain injury. Nat Rev Neurol. 2013;9:222–230.
2. Didehbani N, Munro Cullum C, Mansinghani S, et al. Depressive symptoms and concussions in aging retired NFL players. Arch Clin Neuropsychol. 2013;28:418–424.
3. Kumar NS, Chin M, O'Neill C, et al. On-field performance of national football
league players after return from concussion. Am J Sports Med. 2014;42:2050–2055.
4. Ford JH, Giovanello KS, Guskiewicz KM. Episodic memory in former professional football
players with a history of concussion: an event-related functional neuroimaging study. J Neurotrauma. 2013;30:1683–1701.
5. Hampshire A, MacDonald A, Owen AM. Hypoconnectivity and hyperfrontality in retired American football
players. Sci Rep. 2013;3:2972.
6. Seichepine DR, Stamm JM, Daneshvar DH, et al. Profile of self-reported problems with executive functioning in college and professional football
players. J Neurotrauma. 2013;30:1299–1304.
7. Luker R. Fan base complexity sets up intense competition for attention. Sports Business J. 2011. Available at: http://www.sportsbusinessdaily.com/Journal/Issues/2011/06/27/Research-and-Ratings/Up-Next.aspx
. Accessed June 25, 2015.
8. Clay MB, Glover KL, Lowe DT. Epidemiology of concussion in sport: a literature review. J Chiropr Med. 2013;12:230–251.
9. Martin JP. NFL, ex-players reach $765 million deal in concussion case. 2013. Available at: http://articles.philly.com/2013-08-31/news/41622052_1_retired-players-former-players-ray-easterling
. Accessed June 25, 2015.
10. Elbin RJ, Covassin T, Henry L, et al. Sport-related concussion: “how many is too many?” Transl Stroke Res. 2013;4:425–431.
11. McKee AC, Daneshvar DH, Alvarez VE, et al. The neuropathology of sport. Acta Neuropathol. 2014;127:29–51.
12. Daneshvar DH, Nowinski CJ, McKee AC, et al. The epidemiology of sport-related concussion. Clin Sports Med. 2011;30:1–17, vii.
13. Gartner FR, Nieuwenhuijsen K, van Dijk FJ, et al. The impact of common mental disorders on the work functioning of nurses and allied health professionals: a systematic review. Int J Nurs Stud. 2010;47:1047–1061.
14. Pellman EJ, Lovell MR, Viano DC, et al. Concussion in professional football
: recovery of NFL and high school athletes assessed by computerized neuropsychological testing–part 12. Neurosurgery. 2006;58:263–274; discussion 74.
15. Pellman EJ, Lovell MR, Viano DC, et al. Concussion in professional football
: neuropsychological testing–part 6. Neurosurgery. 2004;55:1290–1303.
16. Pellman EJ, Viano DC, Casson IR, et al. Concussion in professional football
: injuries involving 7 or more days out–part 5. Neurosurgery. 2004;55:1100–1119.
17. Powers KC, Cinelli ME, Kalmar JM. Cortical hypoexcitability persists beyond the symptomatic phase of a concussion. Brain Inj. 2014;28:465–471.
18. Powers KC, Kalmar JM, Cinelli ME. Recovery of static stability following a concussion. Gait Posture. 2014;39:611–614.
19. Solomon GS, Haase RF. Biopsychosocial characteristics and neurocognitive test performance in National Football
League players: an initial assessment. Arch Clin Neuropsychol. 2008;23:563–577.
20. Solomon GS, Kuhn A. Relationship between concussion history and neurocognitive test performance in National Football
League draft picks. Am J Sports Med. 2014;42:934–939.
21. Omalu BI, Hamilton RL, Kamboh MI, et al. Chronic traumatic encephalopathy (CTE) in a National Football
League Player: case report and emerging medicolegal practice questions. J Forensic Nurs. 2010;6:40–46.
22. Amen DG, Newberg A, Thatcher R, et al. Impact of playing American professional football
on long-term brain function. J Neuropsychiatry Clin Neurosci. 2011;23:98–106.
23. Guskiewicz KM, Marshall SW, Bailes J, et al. Recurrent concussion and risk of depression in retired professional football
players. Med Sci Sports Exerc. 2007;39:903–909.
24. Kerr ZY, Marshall SW, Harding HP Jr, et al. Nine-year risk of depression diagnosis increases with increasing self-reported concussions in retired professional football
players. Am J Sports Med. 2012;40:2206–2212.
25. Casson IR, Viano DC, Haacke EM, et al. Is there chronic brain damage in retired NFL Players? Neuroradiology, neuropsychology, and neurology examinations of 45 retired players. Sports Health. 2014;6:384–395.
26. Hart J Jr, Kraut MA, Womack KB, et al. Neuroimaging of cognitive dysfunction and depression in aging retired National Football
League players: a cross-sectional study. JAMA Neurol. 2013;70:326–335.
27. Mitsis EM, Riggio S, Kostakoglu L, et al. Tauopathy PET and amyloid PET in the diagnosis of chronic traumatic encephalopathies: studies of a retired NFL player and of a man with FTD and a severe head injury. Transl Psychiatry. 2014;4:e441.
28. Strain J, Didehbani N, Cullum CM, et al. Depressive symptoms and white matter dysfunction in retired NFL players with concussion history. Neurology. 2013;81:25–32.
29. Guskiewicz KM, Marshall SW, Bailes J, et al. Association between recurrent concussion and late-life cognitive impairment in retired professional football
players. Neurosurgery. 2005;57:719–726; discussion 26.
30. Levin HS, McCauley SR, Josic CP, et al. Predicting depression following mild traumatic brain injury. Arch Gen Psychiatry. 2005;62:523–528.
31. Rabadi MH, Jordan BD. The cumulative effect of repetitive concussion in sports. Clin J Sport Med. 2001;11:194–198.
32. Hadavi T. Professional athletes prepare for life after sports. 2011. Available at: http://www.voanews.com/content/professional-athletes-prepare-for-life-after-sports-118377659/163130.html
. Accessed June 23, 2015.