Article Text
PDF
Abstract
Objective Concern exists about possible problems with later-in-life brain health, such as cognitive impairment, mental health problems and neurological diseases, in former athletes. We examined the future risk for adverse health effects associated with sport-related concussion, or exposure to repetitive head impacts, in former athletes.
Design Systematic review.
Data sources Search of MEDLINE, Embase, Cochrane, CINAHL Plus and SPORTDiscus in October 2019 and updated in March 2022.
Eligibility criteria Studies measuring future risk (cohort studies) or approximating that risk (case-control studies).
Results Ten studies of former amateur athletes and 18 studies of former professional athletes were included. No postmortem neuropathology studies or neuroimaging studies met criteria for inclusion. Depression was examined in five studies in former amateur athletes, none identifying an increased risk. Nine studies examined suicidality or suicide as a manner of death, and none found an association with increased risk. Some studies comparing professional athletes with the general population reported associations between sports participation and dementia or amyotrophic lateral sclerosis (ALS) as a cause of death. Most did not control for potential confounding factors (eg, genetic, demographic, health-related or environmental), were ecological in design and had high risk of bias.
Conclusion Evidence does not support an increased risk of mental health or neurological diseases in former amateur athletes with exposure to repetitive head impacts. Some studies in former professional athletes suggest an increased risk of neurological disorders such as ALS and dementia; these findings need to be confirmed in higher quality studies with better control of confounding factors.
PROSPERO registration number CRD42022159486.
- brain concussion
- athletes
- depression
- suicide
- neurology
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Survey studies reveal that some former contact and collision sport athletes report difficulties with cognitive functioning and mental health.
Some cross-sectional studies of former contact and collision sport athletes, compared with control subjects, have identified changes in brain structure, physiology or biochemistry.
Some death certificate studies of former professional soccer and American-style football players have reported associations between amyotrophic lateral sclerosis (ALS) and dementia as a cause of death and participation in sports.
WHAT THIS STUDY ADDS
The case-control and cohort studies included in this review suggest that (i) former amateur athletes are not at increased risk for depression or suicidality, (ii) former professional soccer players are not at increased risk for psychiatric hospitalisation and (iii) former professional American-style football and soccer players are not at increased risk for mortality from psychiatric disorders or suicide.
The studies included suggest that men who participated in amateur sports are not at increased risk for cognitive impairment, neurological disorders or neurodegenerative diseases compared with men from the general population.
In contrast, studies of men who participated in professional sports reported an association with neurological diseases (eg, ALS) and dementia in former professional American-style football players and professional soccer players.
Well-designed case-control and cohort studies with better control of confounding factors, including genetic, demographic and environmental factors, are needed to determine if these associations are causal.
We did not identify any published case-control or cohort studies that (i) examined age of first exposure to sports-related head impacts and later-in-life health risks, (ii) used neuroimaging as an outcome, (iii) examined later-in-life risks for women or (iv) were considered postmortem cohort studies of chronic traumatic encephalopathy neuropathological change.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
There are many methodological factors to consider when studying whether there is an association between concussions or participation in sports during adolescence and early adulthood and the development of neurological conditions or neurodegenerative diseases much later in life.
These include genetic, demographic, health and environmental factors, socioeconomic status and social determinants of health, alcohol use and substance abuse disorders, obstructive sleep apnoea, cardiovascular disease, peripheral vascular disease and cognitive reserve.
Future well-designed case-control and cohort studies that include more careful control of confounding factors are urgently needed.
Introduction
There is considerable interest, and concern, about possible problems with later-in-life brain health, such as cognitive impairment, mental health problems and neurological diseases, in former athletes who have sustained concussions and repetitive head impacts during participation in youth and adult sports. Later-in-life health, in this context, refers to many years or decades after athletes’ participation in sports. Several systematic reviews on this topic have been published over the past 5 years.1–13 The two primary exposure variables of interest in this steadily growing literature have been sport-related concussions (SRC) and exposure to repetitive head impacts.14 Of note, most studies estimate exposure to repetitive head impacts by the number of years of participation in contact or collision sports, although some studies take additional factors such as position played into account when estimating exposure to repetitive head impacts.
Problems with later-in-life brain health usually refer to the development of psychiatric conditions, such as a depressive disorder or suicidality, and neurological or neurodegenerative disorders or diseases such as Alzheimer’s disease or other causes of dementia, Parkinson’s disease and amyotrophic lateral sclerosis (ALS).13 It can also refer to mental health problems or cognitive decline15 but not to the point of meeting clinical criteria for a disorder or disease per se. Other topics of interest include changes in brain structure, physiology or biochemistry that are quantifiable using neuroimaging,12 16 17 the accumulation of pathological protein surrogates during life using positron emission tomography18–22 and pathological protein aggregates after death using immunohistochemistry.23–26
Chronic traumatic encephalopathy
There are a number of case series and cross-sectional studies related to postmortem neuropathological changes in former contact, collision and combat sport athletes in the past 17 years,23–30 and many recent reviews of the literature relating to chronic traumatic encephalopathy (CTE).4 17 29 31–57 Historically, chronic traumatic brain injury in boxers was described as a clinical condition using terms like punch drunk,58 dementia pugilistica,59 traumatic encephalopathy of pugilists60 and CTE.61 62 In recent years, CTE has been described as a neuropathological entity,26 63 with the defining feature (ie, the ‘pathognomonic’ lesion), based on the updated 2021 consensus criteria, described as ‘p-tau aggregates in neurons, with or without thorn-shaped astrocytes, at the depth of a sulcus around a small blood vessel, deep in the parenchyma and not restricted to the subpial or superficial region of the sulcus’ (p. 217).64 We refer to the postmortem neuropathology as CTE neuropathological change (CTE-NC).
CTE-NC is a postmortem diagnosis by means of neuropathological examination. The clinical phenotype (ie, clinical presentation) associated with CTE-NC during life has not been determined. In 2021, the first consensus criteria for traumatic encephalopathy syndrome (TES) were published.65 This consensus effort was sponsored by the National Institute of Neurological Disorders and Stroke (NINDS). A personal history of substantial exposure to repetitive head impacts is required before considering a clinical diagnosis. The consensus group emphasised that the diagnosis of TES required cognitive impairment, neurobehavioural dysregulation or both—and the condition must have a progressive course. These new criteria need to be validated through diagnostic studies.
Neuroimaging, surveys and neuropsychological studies
Cross-sectional neuroimaging studies have examined differences between former professional athletes and control subjects in neurochemistry, as measured by magnetic resonance spectroscopy66 67; neurophysiology, as measured by functional MRI68 69; microstructure, as measured by diffusion weighted imaging70–72 and macrostructure.73–80 Differences between former athletes and control subjects have been reported in some studies using each of these imaging modalities. The imaging findings are heterogeneous, and systematic reviews of these studies have encouraged more refined work to better understand the associations between the imaging findings and clinical outcomes.2 12
Survey studies, with and without control groups, have examined self-reported symptoms of depression,81–87 cognitive impairment15 84 88 89 and symptoms of TES.90 These studies reveal that some former athletes report significant difficulties with cognitive functioning15 89 91 92 and mental health.15 93–95 In a recent survey, more than one-third of former National Football League (NFL) players reported being ‘extremely concerned’ about memory problems, their thinking skills and/or developing CTE.96 Another survey of former NFL players found that modifiable factors, such as depression, anxiety, pain, physical impairment, sleep apnoea and lack of exercise were associated with impairment in cognition-related quality of life.97
There are also cross-sectional studies involving objective neuropsychological testing, with and without control groups.6 7 98 99 Systematic reviews relating to the literature on neuropsychological outcomes have concluded that former contact, collision and combat sport athletes perform, on average, more poorly on cognitive testing than control participants.5–7 These reviews have also concluded that the general quality of the evidence was low, and high-quality case-control and cohort studies are needed.
Research designs needed to examine risk for future health problems
There is a large and steadily growing body of cross-sectional studies on problems with brain health, broadly defined, in former amateur and professional athletes, and these studies have been the focus of multiple past narrative and systematic reviews.1–13 29 31 34 37–41 47 49 50 53 55 100 101 Case series are often the first type of study to appear when investigating new entities that may not have an established clinical diagnostic phenotype in living patients and may only have a pathological definition, such as is currently the case with CTE-NC.63 64 Survey studies are important because they illustrate how common it is for a group to have certain health problems, such as depression, and they can describe factors associated with those health problems. These types of studies are important because they raise hypotheses to be investigated in future cohort and case-control studies. However, because they are a snapshot in time and include only prevalent cases, they should not be used to infer causation.102 Case series and cross-sectional studies are not designed to examine causal relationships, but they help us direct efforts to examine these issues in more rigorous study designs.103 104 The Concussion in Sport Group consensus paper, published in 2017, emphasised the importance of future well-designed case-control and cohort studies for testing causal hypotheses.105 Moreover, the authors of recently published narrative and systematic reviews have emphasised the need for better designed epidemiological studies that can assess risk of disease and problems with later-in-life brain health, such as cohort and case-control studies, to further understand possible risks associated with SRC and exposure to repetitive head impacts.
Purpose of this review
We designed this systematic review to examine future risk for possible adverse health effects associated with single and multiple SRC, or exposure to repetitive head impacts, during participation in sports. Our a priori focus was on studies that can measure future risk (cohort studies) or approximate that risk (case-control studies). Relying on those types of studies, our two primary questions for this review were: (i) what are the possible long-term effects of single and multiple SRC and (ii) what are the possible long-term effects of exposure to contact sports and/or repetitive head impacts? Subquestions relating to these primary questions were as follows: (i) is the age at commencement of contact sport associated with long-term outcomes; (ii) what is the risk of developing either neurological or neurodegenerative diseases in athletes after SRC and (iii) what is the risk of developing either neurological or neurodegenerative diseases in athletes after exposure to contact sports and/or repetitive head impacts?
Methods
The following databases were searched: MEDLINE(R) and Epub Ahead of Print, In-Process & Other Non-Indexed Citations and Daily (via Ovid), Embase (via Ovid), Cochrane Central Register of Controlled Trials (via Ovid), Cochrane Database of Systematic Reviews (via Ovid), CINAHL Plus with Full Text (via EBSCO) and SPORTDiscus with Full Text (via EBSCO). The initial search was conducted in October 2019 and a top-up search was run in March 2022 (PROSPERO registration: CRD42022159486). The search strategies for all databases are available in the online supplemental file. The Medline search is annotated to provide a search narrative, outlining the different components of the systematic search.106 See the online supplemental material for more information relating to the search strategy and search terms.
Supplemental material
Study eligibility
Studies were eligible for inclusion if they were: (i) original research in humans (cohort and case-control studies); (ii) included former athletes and examined repetitive head impacts or SRC; (iii) evaluated incidence/prevalence, risk factors or causation related to brain health, such as neurodegenerative disease and (iv) evaluated potential long-term sequelae (defined as ≥5 years after sports-related injury or participation in sports). Studies were excluded if they were: (i) review articles, commentaries, editorials, case studies, case series or cross-sectional studies; (ii) not related to former athletes and/or SRC or (iii) conducted <5 years following concussion or participation in sports.
The authors designed this review to search the literature for research designs that could measure future risk to former athletes (ie, cohort studies) or approximate that risk (ie, case-control studies). In cohort studies, an exposed and an unexposed group from a representative sample or population of ‘persons at risk’ are followed through time to the outcome. In a prospective cohort study, data are collected into the future. In a retrospective cohort study, the data already exist for other reasons (eg, health registries and health administrative data). The incidence of the disease is then compared between the exposed and unexposed groups, producing a relative risk. In a case-control study, cases of the disease of interest are collected first, and their past exposure status is compared with a control group of similar subjects. Case-control studies are retrospective by design and the relative risk is approximated by the odds of exposure in the cases divided by the odds of exposure in the controls. The resulting OR approximates the relative risk in a well-designed case-control study.
Study selection
A rapid screen was completed by one author (GMS) to remove citations that were clearly irrelevant to the review (ie, non-human research, opinion papers, conference proceedings, etc). For all screening, data extractions and risk of bias assessments, reviewers independently completed each phase and to capture both methodological quality and content expertise, one designated methods author and one content expert coauthor were paired. For the title and abstract screen, inter-rater agreement was assessed by authors completing 50 randomly selected citations with an acceptable level of 80% agreement. The title and abstract screenings were then completed by pairs of two authors independently (RJC, JDC, KJS, GMS). Conflicts were settled by a third author. Combinations of two authors independently reviewed the articles for the full-text screen (RJC, JDC, GMS). Any disagreements resulted in engagement of a third author who also completed the full-text screen and a meeting between the three authors was held to come to a final decision.
Risk of bias
Risk of bias was investigated using a modified version of the SIGN criteria for cohort and case-control studies.107 These criteria prompt the reviewer to examine the methodology of the study in detail, including the purpose, selection of subjects, attrition, measurement bias, observer bias, control of confounding factors and statistical methods. Teams of two reviewers (GMS, GLI, JDC, RJC, RJE, KJS), one methods author and one content expert coauthor completed these forms on each paper and came to consensus on risk of bias before presenting to the larger group of authors for consideration and final rating of each study.
Equity, diversity and inclusion statement
All eligible studies were included in this systematic review, and no exclusions were applied relating to sex, gender, race, ethnicity, socioeconomic status, social determinants of health or representation from marginalised groups. The author team included 12 men and 3 women; all mid-career or senior in their careers; from North America, Europe and Africa; with education, training and expertise in diverse medical and scientific disciplines.
Results
The search identified 14 813 records (figure 1). After duplicate removal, 7512 underwent rapid title screening and 1318 underwent regular title and abstract screening. Of those, 75 were retrieved for full-text review. Of those 21 met criteria for inclusion. An additional seven articles that were published within the timeline of this systematic review were identified by authors, or by reviewing reference lists from recently published systematic reviews on this topic. No postmortem studies relating to CTE-NC or neuroimaging studies met criteria for inclusion as case-control or cohort studies. No studies relating to age of first exposure to contact or collision sports met criteria for inclusion.
PRISMA. APA, American Psychological Association; CCRCT, Cochrane Central Register of Controlled Trials; CDSR, Cochrane Database of Systematic Reviews; WofS, Web of Science.
There were 28 studies that met inclusion criteria, and they were published between 2003 and March 2022.94 108–134 Four studies included women,111 121 123 131 but no specific outcomes were reported for women. Some of the studies had some modest control for confounding variables,110 111 114 120 122 124 with one study controlling for many confounding variables134 (table 1). There was only one study that was rated as having low-medium (ie, ‘acceptable’) risk of bias.134 All other studies were rated as having high risk of bias, primarily because of lack of control for confounding.
Risk of bias assessment
Seventeen studies were conducted in the USA,94 108–119 125–127 134 three in Scotland,120 122 124 seven in Italy128–133 and one each in Sweden121 and Australia.123 Ten studies involved former amateur athletes108–113 121 125 131 134 (table 2). Four of the studies with amateur athletes used the same longitudinal cohort.108–111
Summary of findings from studies of former amateur athletes
Eighteen studies involved former professional athletes.94 114–120 122–124 126–130 132 133 One study with current athletes was a study of concussion with professional jockeys.123 Two studies with former professional football players119 126 and three studies with former professional Scottish soccer players120 122 124 used the same subjects.
Sixteen of the 28 studies included athletes who played American football.94 108–119 126 127 134 Seven studies involved men who played football during high school108–113 134 and nine studies involved men who played professional football94 114–119 126 127 (table 3). Seven studies examined the association between playing football and later-in-life neurological disorders (eg, dementia) and neurodegenerative diseases (eg, Alzheimer’s disease, Parkinson’s disease and ALS). Two were with former high school football players, using a medical record linkage methodology,112 113 and five were with former professional football players and relied on data recorded on death certificates.115 117 119 126 127 Two studies with former high school football players112 113 and two studies with former professional players126 127 did not find an association, but three studies with former professional players did find an association.115 117 119 The positive findings, and the null findings, all reflect the presence or absence of a minimally adjusted association between playing football and later-in-life neurological diseases. A minimally adjusted association means that only age and sex were considered in the analyses, and thus the analyses were unadjusted for many factors that could confound associations with later-in-life neurological diseases.
Summary of findings from studies of former professional American football players
Ten of the 28 studies did not have American football as the primary focus.120–125 128–131 Eight of those studies examined data relating to former professional Scottish and Italian soccer players120 122 124 128–130 132 133 (table 4).
Summary of findings from studies of former elite and professional athletes from Europe and Australia
Discussion
This systematic review focused on studies that could inform our understanding of risk for future problems with brain health and functioning associated with single or multiple SRC or exposure to repetitive head impacts during participation in sports. No postmortem neuropathology studies or neuroimaging studies met criteria for inclusion. No studies included in this review examined possible long-term effects in women, and thus we can draw no conclusions relating to women. There were also no studies of former university or college-level athletes that met criteria for inclusion in this review. One of our review questions pertained to age of first exposure to contact or collision sports and whether earlier age was associated with worse later-in-life brain health.81 100 135–137 There were no studies on this topic included in this review, and we cannot draw any conclusions.
All the studies that involved former amateur athletes reported no association between participation in amateur sports and later-in-life brain health problems.108–113 121 125 131 134 In contrast, most studies examining neurological disorders or neurodegenerative diseases with former professional athletes reported mostly weak associations between playing professional soccer and football and problems with future brain health.115 117 119 120 122 124 128–130 132 133 Some of these studies reported associations without consideration of potential confounding factors, and in others, control for confounding was considered rudimentary and insufficient. The same applies to some of the studies with former amateur athletes, such as the studies examining risk for future neurological diseases112 113; only minimally adjusted associations were reported without considering other potential confounding factors.
Minimally adjusted associations mean that the studies did not consider factors known to be associated with the exposure and later-in-life mental health, neurological diseases or cognitive functioning. The studies did not examine genetic factors,138 such as the apolipoprotein E genotype,139–143 and usually did not consider or control for known and potential confounding factors related to brain injury and brain health in the general population, such as socioeconomic status, educational attainment, cognitive reserve, smoking, hypertension and cardiovascular disease, diabetes, sleep apnoea, substance abuse, white matter hyperintensities, social isolation, diet, physical activity or exercise.144–154 The one study that did examine genetic factors in a subanalysis reported that it did not affect the lack of association between exposure to football and worse mental health or cognitive functioning in older age.134 To establish a clear causal association between sports participation, cognitive impairment and dementia, or to quantify that association, it is important to consider and control for factors that could confound these associations.
Mental health
Depression, anxiety or both were included in five studies of former amateur athletes (mostly former American football players), and none of them identified an increased risk for these mental health problems in the former contact or collision sport athletes.108–111 134 Two studies of former professional football players found an association between a greater number of past concussions and an increased risk for depression.94 114 One of those studies reported that former players endorsed symptoms of depression similar to people from the general population and there was an association between symptoms of depression and worse physical functioning in the former players.114 Death certificate studies revealed that former professional football players are at significantly lower risk for death relating to psychiatric disorders,126 127 and a medical records study illustrated that hospitalisation for psychiatric problems was less common in former professional soccer players than men from the general population.122 In aggregate, the studies suggest that former amateur and professional athletes are not at increased risk for mental health problems compared with men from the general population.
Suicidality
Nine studies examined suicidal ideation, suicidal behaviours or suicide as a manner of death,108–111 118 122 126–128 four of which were with former amateur athletes108–111 and five with former professional athletes.118 122 126–128 None of the nine studies found an association between playing contact or collision sports and later-in-life risk for suicidality or suicide. Playing high school football was not associated with future risk for suicidal ideation in four studies.108–111 Specifically, there was not an association at an average age of 22 years,109 29 years109–111 or 38 years108 in men participating in the National Longitudinal Study of Adolescent to Adult Health.155 All studies that examined suicide as a manner of death in former professional football players or former professional soccer players reported either significantly lower risk118 126 127 or no difference compared with men in the general population.122 128 In aggregate, the studies suggest that former amateur athletes are not at increased risk for suicidality and former professional football players and soccer players are not at increased risk for suicide, compared with men from the general population.
Neurological disorders and neurodegenerative diseases
Two studies reported that playing high school football was not associated with increased risk for dementia, Parkinson’s disease or ALS.112 113 A case-control study of people with ALS found no association with sports participation in general, or with playing amateur soccer, and having ALS.131 Men who played high school football, compared with those who did not play football, did not have worse cognitive functioning at the age of 65 years in one longitudinal study.134 In a Swedish twin study, there was not an association between sports involving repetitive head hits and the experience of cognitive impairment during older adulthood.121 In aggregate, the studies included in this review suggest that most former amateur athletes are not at increased risk for cognitive impairment, neurological disorders or neurodegenerative diseases compared with men from the general population.
Neurological conditions (eg, dementia) and neurodegenerative diseases (eg, Alzheimer’s disease and ALS) were outcomes in 12 studies involving former professional athletes,115 117 119 120 124 126–130 132 133 and many of these studies reported worse outcomes for the former professional athletes. Most of the studies examined minimally adjusted associations without controlling for confounding factors and some were ecological analyses with no individual-level analyses. Greater risk for neurological conditions and neurodegenerative diseases was reported for both former professional football players119 and former professional soccer players.124 The study with former soccer players was unique in that they included a variable relating to socioeconomic status at the time of death, and they also conducted sensitivity analyses that considered competing risks for death associated with ischaemic heart disease and cancer.124 Regarding absolute numbers, the study examining causes of death of 1180 former professional soccer players reported that 222 (18.8%) had a neurological or neurodegenerative disease listed on their death certificate, compared with 6.0% (228/3807) of a matched control sample of men who had died.124 Another study reported that 7.5% of former professional football players (39/517) compared with 3.7% of former professional baseball players (16/431) had a neurological or neurodegenerative disease listed on their death certificate.117
Eight studies found an association between ALS and participation in professional sports.115 119 124 128–130 132 133 ALS, a rare disease with a possible genetic cause in a substantial number of men who develop the disease before age 50 years,156 involves a highly selective population of neurons, about half of which are in the spinal cord, which makes identifying a specific trauma-related mechanism challenging. Researchers have been examining possible environmental and lifestyle risk factors for ALS for many years.157 158 The total number of ALS cases reported in the studies included in this systematic review tended to be small, with five studies identifying fewer than 10 cases,119 128 129 132 133 which limits the statistical power. One study of former professional football players identified ALS on 7 of 334 death certificates (2.1%),119 and another study of former professional football players identified ALS in 38 of a cohort of 19 423 former players (0.2%).115 No study reported pathological confirmation. More research relating to possible mechanisms underlying the association is needed.
In aggregate, the studies show an association between playing professional football and professional soccer and some neurological disorders and diseases. Higher quality studies that control for potential confounding factors are needed to confirm or refute these findings.
Methodological considerations and limitations with the studies included in this review
By design, this systematic review focused on concussions and exposure to repetitive head impacts in former athletes. We did not attempt to review the literature relating to associations between mild traumatic brain injury and depression, suicidality, neurological disorders or neurodegenerative diseases in people from the general population or military service members or veterans.
The literature on possible long-term effects of concussions and repetitive head impacts experienced during participation in youth, amateur and professional sports, employing a case-control or cohort study design, is limited, but evolving. The studies involving amateur athletes yielded entirely negative results. Negative results in cohort studies are important in this regard because cohort studies provide the most rigorous study design available in this literature, but these studies must be interpreted within the context of a variety of methodological limitations that are common in this literature. For example, these studies, as a rule, did not quantify or examine the amount of exposure to concussions or repetitive head impacts. Although the studies with former amateur athletes all yielded negative findings, that does not mean that there are no possible later-in-life adverse health effects associated with participation in amateur sports. Future research, using well-designed case-control and cohort studies, is needed to better measure potential health risks.
Some of the studies in this review had some modest control for confounding variables,110 111 114 120 122 124 with one study controlling for many confounding variables and being considered to have low-to-medium risk of bias134 (table 1). All other studies were considered to have high risk of bias, primarily because of a lack of control for confounding. As a rule, the studies did not consider many genetic, lifestyle or medical factors that could potentially confound associations between later-in-life adverse health outcomes and earlier-in-life participation in sports. A 2020 Lancet Commission report, for example, presented a life-course model for potentially modifiable risk factors for dementia in the general population that included factors in early life (ie, age <45 years) such as less education, midlife (ie, age 45–65 years) such as hearing loss, traumatic brain injury, hypertension, excessive alcohol consumption and obesity, and later life (age >65 years) such as smoking, depression, social isolation, physical inactivity, diabetes and air pollution.159
Another methodological consideration is whether the ‘general population’ is the preferred comparison group for studies involving the brain health of former contact and collision sport athletes. It is unlikely that control subjects selected from the general population are similar in all risk factors as professional athletes. Comparing with the general population might underestimate or overestimate certain risks. For example, former athletes are more likely to experience the health benefits from a life of physical exercise. However, they may also experience more chronic pain and be prescribed more medications than controls from the general population.
Several studies included in this review relied on data from death certificates. Some of these studies have relatively small sample sizes (eg, fewer than 500 deaths).119 126 127 In these studies, there is no explicit exposure other than belonging to a group—the exposure is playing a professional sport, such as football. Thus, these are ecological studies with no control of confounding at the individual level. Cardiovascular disease, for example, is associated with risk for cognitive impairment and dementia, and one study included in this review reported that 96.3% of former professional football players and 52.2% of former professional baseball players had cardiovascular disease listed on their death certificates.117 These ecological studies are hypothesis generating and cannot be used to draw specific causal conclusions. Moreover, errors on death certificates are common, and include the coding of causes of death, comorbidities and manner of death.160 161 Additionally, there is evidence that dementia as a cause of death is under-reported on death certificates.162 Methodological limitations associated with the 28 studies included in this review are discussed in greater detail in the online supplemental material.
Chronic traumatic encephalopathy neuropathological change and traumatic encephalopathy syndrome
Case series and cross-sectional studies relating to CTE-NC in former athletes have been published over the past 17 years,23–30 as well as many published reviews.4 17 29 31–57 A recent narrative review paper57 examined the issue of association versus causation relating to repetitive head impacts and CTE-NC, through the lens of nine viewpoints proposed by Sir Austin Bradford Hill in 1965.163 The authors of that review concluded that the evidence supporting repetitive head impacts causing CTE-NC was compelling. (As part of that review, the authors identified six studies that they referred to as case-control studies and calculated ORs for an association between sports exposure and CTE-NC. In the present systematic review, we did not consider those six studies to be case-control or cohort studies that were designed to examine risk for CTE-NC or risk for later-in-life clinical conditions or disorders. In addition, the original authors of most of the studies did not conceptualise them as case-control studies designed to measure risk.)57 The postmortem studies of CTE-NC were not included in this systematic review because they did not meet the a priori inclusion criteria for research design. However, given considerable interest in this topic, a review of the neuropathology studies, a summary of gaps in knowledge and directions for future research is provided below and in the online supplemental material (which should be read in conjunction with this paper).
The brain donation programme at Boston University has specialised, for years, in procuring donations from families of former athletes, especially football players. Their research has identified CTE-NC in virtually all former professional American football players in their sample (ie, 99%), 91% of former college players and 21% of former high school players.24 Those researchers have reported that the prevalence of CTE-NC in all former football players and athletes from other sports is unknown, and ascertainment bias, associated with the decision to donate a person’s brain, would inflate the proportions of the samples identified as having the pathology. Other researchers have identified CTE-NC in former elite and professional athletes in smaller case studies and case series of former high exposure boxers (7/14 from the seminal 1973 Corsellis study),164 165 former soccer players (5/7),23 rugby union players (3/4),23 rugby league players (2/2),166 Canadian professional football players (11/24)25 and Canadian ice hockey players (6/11).25 The Canadian studies are from a brain bank established in 2010 at the Canadian Concussion Centre to study the neuropathology of brain degeneration in professional athletes.30
In 2020, Bieniek et al published a postmortem study of former amateur athletes, compared with those who did not participate in sports, and identified CTE-NC in 2.8% of the total sample (21/750), 5.0% (15/300) of those who played sports, 1.3% who did not play sports (6/450), 0% of women (of 273), 4.4% (21/477) of men, 7.9% of men who played football (11/140) and 2.4% (6/245) of men who did not play sports.167 Those men who had CTE-NC were, on average, 10 years older at the time of death than those who were negative for CTE-NC, and they were much more likely to have Alzheimer’s disease neuropathological change (ie, 47.6% vs 20.2%). The authors concluded that participation in football, specifically beyond the high school level, was associated with a greater OR for having CTE-NC. None of the aforementioned studies can provide an accurate estimate of the incidence or prevalence of CTE-NC in these sports, which are essential to calculating risk—but they raise important hypotheses concerning potential risks.
Recently published postmortem studies from a brain bank in Australia (n=636)168 and a community sample in the USA (n=532)169 have revealed very few cases of CTE-NC (<1%), and a large European study (n=310) reported no cases of CTE-NC,170 using ‘strict’ criteria for case ascertainment. In a study of 180 consecutive autopsies in Australia, four cases of mild CTE-NC were identified (2.2%).171 In a recent study of 225 former military personnel, CTE-NC was identified in 4.4%, and it usually involved minimal microscopic changes on p-tau immunohistochemistry—with half of the identified cases having only a single pathognomonic lesion—and the authors concluded that it is unclear whether the minimal histopathological findings were of clinical significance.172
There are important gaps in knowledge that provide directions for future research. At present, the incidence and prevalence of CTE-NC (a neuropathological entity), or TES (a proposed clinical diagnosis), in former athletes, military veterans and people from the general population is not known. Moreover, it is not known whether CTE-NC causes specific neurological or psychiatric problems, the extent to which larger amounts of CTE-NC can be clearly identified within the presence of Alzheimer’s disease neuropathological change, or whether the neuropathology is inexorably progressive. The NINDS consensus group made numerous important recommendations to address gaps in knowledge relating to CTE-NC and TEC.65 They encouraged researchers to address topics such as determining whether patchy areas of perivascular p-tau deposition at the depths of cerebral sulci and more limited amounts of CTE-NC have direct clinical correlates, and whether more abundant CTE-NC is associated with specific cognitive and neuropsychiatric symptoms. They also encouraged studies relating to the inter-rater reliability and predictive validity of the new proposed criteria for TEC. They emphasised that the new diagnostic criteria are designed primarily for research use, and it is possible that a number of factors unrelated to sporting history might be associated with the clinical features underlying this condition, including ‘medical and psychosocial variables, such as sleep disorders, vascular risk factors, chronic pain and racial and associated inequities in social determinants of health’ (p. 859).65
Conclusions and directions for future research
In recent years, there have been many published narrative and systematic reviews relating to possible long-term effects of SRC and participation in contact, collision and combat sports.1–13 29 31 34 37–41 47 49 50 53 55 100 101 The topics of those reviews included neurological diseases, CTE, psychiatric disorders, suicide, neuroimaging and neuropsychological outcomes. The authors of past reviews have often concluded that the general quality of the evidence was low, and that high-quality cohort studies were needed. The present systematic review differed from prior reviews in that it focused only on case-control and cohort studies with the goal of drawing conclusions about future risk to former amateur and professional athletes. An important finding from this review, and gap in the literature, was documenting the lack of case-control or cohort studies involving age of first exposure to sports, neuroimaging as an outcome, women and postmortem studies (ie, CTE-NC). The postmortem studies are a particularly important direction for future research, especially for antemortem biomarker development.
The case-control and cohort studies included in this review suggest that (i) former amateur athletes are not at increased risk for depression or suicidality,108–111 (ii) former professional soccer players are not at increased risk for psychiatric hospitalisation122 and (iii) former professional football and soccer players are not at increased risk for mortality from psychiatric disorders126 127 or suicide.118 122 126–128 However, some former athletes, like men in the general population, might experience serious mental health problems and suicidality as they age; therefore, it is important to study these athletes to understand why, and to provide treatment.
Four cohort studies and one case-control study suggest that former amateur athletes are not at increased risk for cognitive impairment, neurological disorders or neurodegenerative diseases compared with men from the general population.112 113 121 131 134 In contrast, some studies of former professional athletes examining causes of death reported an association with neurological diseases and dementia in former professional football players115 117 119 and professional soccer players.124 Moreover, there were six publications that reported an association between participation in professional soccer and ALS.124 128–130 132 133 Most of the studies with professional athletes are ecological in nature and the associations are not at the individual level. Well-designed case-control and cohort studies are needed to determine if these associations are causal.
There are many potential factors to consider when studying whether there is an association between concussions or participation in sports during adolescence and early adulthood and the development of neurological conditions or neurodegenerative diseases later in life, such as genetic, demographic, health and environmental factors, socioeconomic status and social determinants of health, alcohol use and substance abuse disorders, obstructive sleep apnoea, cardiovascular disease, peripheral vascular disease and cognitive reserve. Future well-designed case-control and cohort studies that include more careful control of confounding factors are urgently needed.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
Acknowledgments
A preliminary version of this systematic review was presented at the 6th International Conference on Concussion in Sport in Amsterdam, the Netherlands, on 28 October 2022. The authors acknowledge the work of librarians Drs Zahra Premji (search strategy), Diane Lorenzetti (PRESS review) and Shauna Rutherford (reference retrieval and organisation in Covidence) for their contributions.
References
Footnotes
Twitter @Kat_Schneider7, @jonpatricios, @ProfJiriDvorak
Correction notice This article has been corrected since it published. Typographical errors have been made to the online version only and not in print.
Contributors Original conceptualisation of the systematic review: IKK, JD, JDC, JB, AH, KJS, GLI, GMS, GM, RCC, RJC and RJE. The title and abstract screenings were completed by RJC, JDC, KJS and GMS. Reviewed the articles for the full-text screen: RJC, JDC and GMS. Completed all risk of bias assessments: GMS, GLI, JDC, RJC, RJE and KJS. Data extractions from the articles were completed primarily by GLI, RJC and RJE. Separate from the data extractions and risk of bias reviews, JDC and GLI reviewed all articles for clinical and methodological content. GLI was primarily responsible for designing the structure of the manuscript and drafting the manuscript. JDC provided leadership in methodology. A separate review of chronic traumatic encephalopathy-neuropathological change and traumatic encephalopathy syndrome was conducted primarily by GLI, RCC and RJC, with one author primarily responsible for writing the content for the article and the online supplement (GLI) and two authors carefully reviewing this content (RCC, RJC). All authors reviewed and edited drafts of the online supplement and the manuscript, and approved the documents for submission for publication. GLI as the guarantor accepts full responsibility for the finished work and/or the conduct of the study, had access to the data and controlled the decision to publish.
Funding Funding for some of the administrative aspects of the numerous systematic reviews prepared for the 6th International Consensus Conference on Concussion in Sport was provided through an educational grant from the Concussion in Sport International Consensus Conference Organising Committee through Public Creations for partial administrative and operational costs associated with the writing of the systematic reviews. Partial support for travel to the Consensus Conference, for some of the authors, also was provided by the International Olympic Committee.
Competing interests GLI serves as a scientific advisor for NanoDX, Sway Operations and Highmark. He has a clinical and consulting practice in forensic neuropsychology, including expert testimony, involving individuals who have sustained mild TBIs (including former athletes), and on the topic of suicide. He has received past research support or funding from several test publishing companies, including ImPACT Applications, CNS Vital Signs and Psychological Assessment Resources. He receives royalties from the sales of one neuropsychological test (WCST-64). He has received travel support and honorariums for presentations at conferences and meetings. He has received research funding as a principal investigator from the National Football League (NFL), and subcontract grant funding as a collaborator from the Harvard Integrated Programme to Protect and Improve the Health of NFL Players Association Members. He has received research funding from the Wounded Warrior Project. He acknowledges unrestricted philanthropic support from ImPACT Applications, the Mooney-Reed Charitable Foundation, the National Rugby League, Boston Bolts and the Schoen Adams Research Institute at Spaulding Rehabilitation. RJC is a collaborator on a grant funded by the NFL to study the spectrum of concussion, including possible long-term effects. He has a consulting practice in forensic neuropathology, including expert testimony, which has involved former athletes at amateur and professional levels. JDC has provided expert epidemiological testimony in court cases on the long-term effects of concussions in sports. GMS is an owner of a multidisciplinary practice (managing patients with MSK pain disorders). He is a board member of Hockey Calgary (Calgary, Alberta, Canada) and Chair of the Alberta Association of Physiotherapy. He received funding for the administrative aspects of the writing of two of the systematic reviews that informed the consensus process. KJS has received grant funding from the CIHR, NFL Scientific Advisory Board, International Olympic Committee Medical and Scientific Research Fund, World Rugby, Mitacs Accelerate, University of Calgary, with funds paid to her institution and not to her personally. She is an Associate Editor of BJSM (unpaid), Independent consultant to World Rugby and has received travel and accommodation support for meetings where she has presented. She coordinated the writing of the systematic reviews that informed Amsterdam International Consensus on Concussion in Sport, for which she has received an educational grant to assist with the administrative costs associated with the writing of the reviews (with funds paid to her institution). She is a member of the AFL Concussion Scientific Committee (unpaid position), Brain Canada (unpaid positions) and Board member of the Concussion in Sport Group (CISG)(unpaid). She works as a physiotherapy consultant and treats athletes of all levels of sport from grass roots to professional. RJE is a paid consultant for the NHL and cochair of the NHL/NHLPA Concussion Subcommittee. He is also a paid consultant and chair of the Major League Soccer concussion committee, and a consultant to the US Soccer Federation. He previously served as a neuropsychology consultant to Princeton University Athletic Medicine and EyeGuide. He is currently a co-Principal Investigator for a grant funded by the NFL (NFL-Long) through Boston Children’s Hospital. He occasionally provides expert testimony in matters related to MTBI and sports concussion, and occasionally receives honoraria and travel support/reimbursement for professional meetings. JB serves as the Chairman of the Medical Advisory Committee of Pop Warrner Football and is a member of the NFL Head, Neck, and Spine Committee, NFLPA Mackey-White Health and Safety Committee and the NCAA Concussion Task Force, all without remuneration. He is a consultant for BrainLab and has provided expert testimony on TBI medicolegal cases. IKK is a professor at Ludwig-Maximilians-Universität Munich (paid position). She serves as European Editor at Journal of Neurotrauma (unpaid position) and as Vice President of the European Neurotrauma Organisation (unpaid position). She receives research grant funding from the National Institutes of Health, the European Research Council, the German Ministry for Research and Education. She receives funding for a research study on sport-related concussion from Abbott. The university hospital received donations for her research from the Schatt Foundation and Mary Ann Liebert. She receives royalties for book chapters published by Thieme Publishers. Her spouse is employee at Siemens and she thus holds stock options at Siemens and Siemens Healthineers. In-kind contribution: PhD students working under her supervision receive scholarships from the Villigst Foundation, the China Scholarship Council collaboration with Ludwig-Maximilians-University Munich and Fulbright. GM discloses grants from the United States Department of Defense—TBI End points Development Initiative (grant #W81XWH-14-2-0176), TRACK-TBI Precision Medicine (grant # W81XWH-18-2-0042) and TRACK-TBI NETWORK (grant # W81XWH-15-9-0001); NIH-NINDS–TRACK-TBI (grant #U01NS086090) and the NFL Scientific Advisory Board—TRACK-TBI LONGITUDINAL. The United States Department of Energy supports GM for a precision medicine collaboration. One Mind has provided funding for TRACK-TBI patients stipends and support to clinical sites. He has received an unrestricted gift from the NFL to the UCSF Foundation to support research efforts of the TRACK-TBI NETWORK. GM has also received funding from NeuroTruama Sciences to support TRACK-TBI data curation efforts. Additionally, Abbott Laboratories has provided funding for add-in TRACK-TBI clinical studies. MMcN has the following disclosures: (i) Chair, Ethics Expert Group, WADA (2021–23) (paid); (ii) Member, International Boxing Association, Ethics and Integrity Committee, (2021–22; resigned October 2022) (paid); (iii) Chair, Therapeutic Use Exemption Fairness Committee (2020–Present) (paid); (iv) Member, Steering Group, Sex Segregation in Sport, IAAF/World Athletics (2019–20) (unpaid); (v) Member, International Ice Hockey Federation, Ethics and Integrity Committee (2019–21) (paid); (vi) Member, IOC Consensus Statement Expert Group on Injuries in Children and Adolescents (2017) (unpaid); (vii) Member, Ethics Expert Group, WADA (2016–21) (unpaid) and (viii) Member, IOC Consensus Statement Expert Group on Pain Management (2016) (unpaid). JP is a sports and exercise medicine physician at Waterfall Sports Orthopaedic Surgery, Johannesburg, South Africa. He is Director of Sports Concussion South Africa and a Visiting Professor, Wits Sport and Health (WiSH), School of Clinical Medicine, Faculty of Health Sciences, University of the Witwatersrand. As an Editor of BJSM, he receives an honorarium and serves as a consultant (unremunerated) to World Rugby Concussion Advisory Group, South African Rugby, EyeGuide Scientific Advisory Board and the Union of European Football Associations. He is co-chair of the Scientific Committee, 6th International Conference on Concussion in Sport (unpaid) and a Board member of the Concussion in Sport Group (unpaid). RCC serves as a scientific advisor for the NFL's Head Neck and Spine Committee, VP and Chair of the Scientific Advisory Committee for the National Operating Committee on Standards for Athletic Equipment (NOCSAE), and as co-founder and Medical Director of the Concussion Legacy Foundation. He currently receives research support from the NINDS UNITE and Diagnose CTE grants. He has received travel support and honorariums for presentations at conferences and meetings. He receives royalties from Houghton Mifflin Harcourt publishing. He has a clinical and consulting practice in forensic neurology and neurosurgery, including expert testimony, especially individuals with traumatic brain and spinal cord injuries. He is a member of the National Collegiate Athletic Association Student-Athlete Concussion Injury Litigation Medical Science Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.