Abstract
Participation in collision and contact sport in Australia—specifically rugby league—is popular. With recent attention to the possible long-term health consequences of head impact exposure during a contact or collision sport career, the importance of understanding the contribution of modifiable risk factors as they relate to cognitive function has been highlighted. Risk factors for cognitive decline in the general population include cardiovascular health, sleep disorders, chronic pain, depression, anxiety, smoking, physical impairment, and physical inactivity. This study examined the associations between these risk factors and self-reported cognitive function in 130 former elite male rugby league players in Australia. Respondents were recruited through a survey distributed through former player groups and via word of mouth. Self-reported cognitive function was assessed using the Quality of Life in Neurological Disorders—Applied Cognition General Concerns questionnaire. Risk factors for cognitive decline were self-reported, with questions collated from multiple validated sources, with each selected to explore specific categories of cognitive function. They included: questions from the Football Players Health Study at Harvard; The Australian Mental Health and Wellbeing Survey 2007; the Patient Reported Outcome Measurement Information System Item Banks for Pain Interference and Physical Function; and the Patient Health Questionnaire. Of the 130 participants, 43.1% (n = 56) reported perceived cognitive impairment. When adjusted for age and number of concussion-related signs and symptoms experienced during their career, predictors of perceived cognitive difficulties included less than 5 h of sleep on average, history of stroke, current clinical symptoms of anxiety, physical impairment, and number of risk factors. The number of concussion-related signs and symptoms experienced was not related to perceived cognitive impairment, although it was associated with specific risk factors. Early education and intervention by medical professionals to manage these risk factors may provide a pathway for improving perceived cognitive health and functioning in former elite male rugby league players in the future.
Introduction
Contact and collision sport participation has been popular worldwide for decades. In Australia, the estimated participation rate in rugby league, one of Australia’s most popular collision sports, was 0.8% of all Australians in 2016 and increased to 1.1% in 2022. 1 In 2022, there were an estimated 189,556 people playing rugby league across all ages in Australia. The highest proportion of people was under the age of 17 years, with boys making up 75% of registered participants. 1 Injuries, including head injuries and concussions, are common in rugby league. Over three playing seasons, 13–17% of professional male rugby league players reported sustaining a concussion. 2 The concussion incidence in rugby league players at all levels ranges from 8.0 to 17.1 per 1000 playing hours. 3
There is evidence from retrospective cohort studies that former professional contact and collision sport athletes may experience later-in-life cognitive problems and depression at higher rates than comparable groups who were not exposed to contact and collision sports. 4 A recently published systematic review concluded that some studies examining former professional American-style football players and professional soccer players showed an increased risk of neurological diseases and dementia. 5 In a cohort study from Finland, former athletes who participated in boxing, wrestling, and soccer had higher rates of dementia compared with the general population. 6 The systematic review conducted by Iverson and colleagues, 5 identified a pressing need for future research to better evaluate potentially confounding variables that may be associated with cognitive outcomes, including socioeconomic status and social determinants of health, alcohol use and substance disorders, obstructive sleep apnea, cardiovascular disease, and cognitive reserve. 5
A range of potentially modifiable risk factors for cognitive impairment has been identified via population-level studies. These include sleep apnea, 7 short sleep duration, 8 obesity, 9 hypertension, 10 cardiovascular disease, 9 diabetes, 11 chronic pain, 12,13 depression, 11,14 anxiety, 15,16 smoking, 9 physical impairment, 17,18 and physical inactivity. 19 These risk factors were also identified in the Lancet Commission review, 20 which reported ∼40% of known dementia risk factors are modifiable (e.g., traumatic brain injury [TBI], hypertension, high alcohol intake, obesity, smoking, depression, social isolation, physical inactivity, air pollution, and diabetes). 20
Many of these risk factors have also been studied in former athletes. For instance, the Football Player’s Health Study at Harvard examined the association between modifiable risk factors for cognitive difficulties in 3803 former professional American-style football players. 21 Chronic pain, mood problems, sleep problems, obesity, and lack of physical activity were associated with participants endorsing significant cognitive problems on a self-report questionnaire. The health of former athletes in later life who participated in soccer and football, 5,22 –25 has been examined in studies involving medical conditions, 26 –29 cognition, 30 –33 and mental health issues. 34,35
To date, few studies have investigated the long-term health of former Australian rugby league or rugby union players. 3,36 –43 Functional and structural brain magnetic resonance imaging (MRI) studies have reported decreased glutathione in gray matter, 44 similar cortical thickness, 42 differences in white matter microstructure, 43 and functional connectivity differences in the cerebellum 36 in former elite-level rugby league players when compared with age and education matched controls with no history of brain trauma or multiple head impacts. One study reported that concussion history was not related to symptoms of depression in former elite-level rugby league players and that depression scores were associated with levels of anxiety, stress, resilience, and life-interfering pain. 40 A previous investigation of former Australian rugby league players did not find associations between perceived cognitive impairment and objective cognitive test performance. 41 Given the small number of studies to date, and the importance of brain health in former contact and collision sport athletes, more studies are needed relating to factors associated with cognitive difficulties in former rugby league players. The current study aims to extend the existing knowledge by determining the extent to which known risk factors for cognitive impairment are related to perceived cognitive difficulty in former elite male Australian rugby league players.
Methods
Participants
Any former elite-level Australian rugby league players were eligible to participate in this study. Rugby league did not become a “professional” league until the 1990s. As such, eligibility was based on participating at the “elite” level. The operational definition of elite included playing at least one game of first-grade rugby league in the New South Wales Rugby League, Queensland Rugby League, Australian Rugby League, Super League, or National Rugby League competitions. Two recruitment strategies were used, which included former player groups (e.g., club old boys’ groups), and word of mouth. Sport organizations with direct links to former athletes have actively promoted the study, including the National Rugby League Players Association, and the Family of League Foundation. Participant advocates have also reached out to former athletes and passed on information about the research program via their networks. This study did not include women because the premiership-level competition for women started in Australia in 2018. Participants completed a questionnaire using a paper form, which was mailed with a return self-addressed envelope. In order for a participant’s data to be used in the analysis, all of the questions pertaining to demographics, Quality of Life in Neurological Disorders–Applied Cognition General Concerns (Neuro-QOL) items, and concussion signs and symptoms had to be answered. Recruitment of participants occurred between December 2019 and 2024. The research was approved by the Local Health District Human Ethics Committee (Project number: 16/11/16/5.09).
The former rugby player’s health questionnaire
The survey included 105 questions. These questions were derived from a number of published sources: The Football Players Health Study at Harvard University 21 ; The Australian Mental Health and Wellbeing Survey 2007 45 ; Patient Reported Outcome Measurement Information System (PROMIS) Item Banks for Pain Interference and Physical Function 46 ; and the Patient Health Questionnaire (PHQ-4). 47 The questions asked about the participants’ perceived physical and mental health, presence and treatment of various diagnosed medical conditions, physical activity levels, smoking status, and self-reported concussion history.
Self-reported difficulties
Current cognitive difficulties were assessed using the short form of the Neuro-QOL-Short Form v1.0 (i.e., Neuro-QoL Cognition). The responses from the eight questions were summed to create a continuous variable and then transformed into a standardized T-score based on normative reference values from the United States general population. 48 Some analyses use the continuous T-score, while other analyses dichotomized the T-score into two groups (i.e., “perceived cognitive impairment” or “no perceived cognitive impairment”), 21 such that “perceived cognitive impairment” was defined as scoring ≤1SD (T-score ≤40) below the U.S. population mean, 21 which is equivalent with the 16th percentile or lower on standard psychometric conversion tables (i.e., 16% of the general population would be at or below this score).
Self-reported history of concussion signs and symptoms
In prior studies, 21 participants were queried with 10 questions assessing prior concussion-like signs and symptoms they experienced while playing American-style professional football to generate a composite variable related to their past concussion history. Respondents reported how many times they experienced each sign/symptom after a blow to the head, neck, or upper body during their athletic career. The signs and symptoms included headaches, nausea, dizziness, loss of consciousness, memory problems, disorientation, confusion, seizure, visual problems, and “feeling unsteady on your feet.” Response options for each symptom were: no, once, 2–5, 6–10, or 11+ times (numerically coded 0, 1, 3.5, 8, and 13, respectively, for analyses). A total score was calculated by adding the coded items for the 10 signs and symptoms. This total score was used for the multivariate linear regression. For other analyses, participants were grouped into four quartiles based on their total prior concussion-like signs and symptoms score, from the least number of signs and symptoms (Quarter 1 [Q1]) to the greatest number of signs and symptoms of concussion during playing years (Q4).
Self-reported physical activity
Participants’ current level of self-reported physical activity was calculated by using the average number of hours spent each week walking, jogging, and running, in other aerobic activities (e.g., cycling), low intensity exercise (e.g., yoga), and weight training. Jogging, running, and other aerobic activity was “high intensity.” The response options for number of hours per week engaged in each activity were 0, <1, 1–5, 6–10, and >10. These were multiplied by the metabolic equivalents (METS) for each activity according to a previous study, 49 to determine total METS for the week. In our survey, jogging was defined as slower than 10 min/km, and running was defined as faster than 10 min/km. The speed of these activities was used for calculating the equivalent METS. Other aerobic exercise was coded using the METS value for cycling at 20 km/h, and low intensity exercise was coded as per yoga.
Self-reported vascular risk factors and other health issues
Participants were asked to self-report if a medical provider had recommended or prescribed medication for high blood pressure, heart failure, heart rhythm problems, hypercholesterolemia, or diabetes. Participants were also asked to self-report a lifetime diagnosis of stroke, heart attack, heart surgery, and sleep apnea. Participants self-reported current weight, height, smoking status, and alcohol intake. High alcohol intake was defined as ≥14 standard drinks/week, whereas low alcohol intake was defined as <14 standard drinks/week. 50 Participants were also asked about how many hours they sleep on an average weekday. Low sleep duration was defined as <5 h/night. 51
Other questionnaires
The PHQ-4 was used to assess symptoms of depression and anxiety over the past 2 weeks. 52 There are two items for depression (PHQ-2) and two items for anxiety (Generalized Anxiety Disorder-2 [GAD-2]). Each question was rated on a 0–3 scale. The depression and anxiety scales were summed separately and were dichotomized per scoring recommendations to indicate screening positively for symptoms consistent with depression or anxiety (i.e., a score of ≥3).
The PROMIS Item Bank v1.1 Pain Interference Short Form 6b, 46 was used to determine the degree to which pain interfered with the participant’s daily life. This is a 6-item questionnaire that has respondents rank the degree to which pain interfered in their life in the last week, in five outcome categories from “none at all” to “very much.” The PROMIS Item Bank v2.0 Physical Function Short Form 6b, 46 was used to assess current physical functioning. This 6-item scale asks the respondents to rank their ability to complete common household tasks, (e.g., vacuuming/yard work) across five response options (i.e., “without any difficulty” to “unable to do”). High levels of interference (T ≥60) and low physical function (T ≤40) were categorized as ≥1SD worse than the U.S. population mean.
Statistical analysis
Demographics for the entire sample, as well as stratified quartiles based on participants’ prior concussion signs and symptoms scores, were examined using means, frequencies, and percentages. The association between possible cognitive risk factors and self-reported cognitive function was examined using a series of multivariate linear regressions. The continuous Neuro-QoL Cognition T-score was used as the dependent variable in each separate linear regression model, and the specific risk factor was the independent variable. For each risk factor, two models were run: one that only adjusted for age and one that adjusted for age and the total concussion signs and symptoms score. Each risk factor was entered as a binary value, with 0 meaning no and 1 meaning yes. The number of METS was divided in quartiles, and the highest and lowest were compared. For body mass index (BMI), normal weight (BMI <25) and obese (BMI ≥30) were compared. Diagnosis of cardiovascular disease (i.e., “Any Cardiovascular Disease”) included any diagnosis or prescription medication for hypertension, heart failure, arrhythmia, heart attack, any history of heart surgery, stroke, or high cholesterol. The “Risk Factor Sum” was calculated by adding the following 12 risk factors: Current/History of Smoking; Heavy Alcohol Consumption (≥14 units/week); Low Physical Activity (Quartile 1); Sleep Apnea; Low Sleep (≤5 h average night); BMI (≥30 kg/m2); Any Cardiovascular Disease; Diabetes; Significant Physical Impairment (T ≤40); Significant Pain Interference (T ≥60); Depressive Symptoms (PHQ-2 ≥ 3); and Anxiety Symptoms (GAD-2 ≥ 3).
Independent samples t-tests were used to examine if there was a group difference in participants’ concussion signs and symptoms scores within those who reported the analyzed risk factors. For categorical variables, chi-squared tests were used to compare the highest and lowest quartiles of concussion signs and symptoms as the dependent variable, and each risk factor as a binary value. Linear regression, adjusted for age, was used to examine the relationship between self-reported amount of physical activity and degree of physical impairment. Data were analyzed using SPSS version 29.0. Statistical significance was defined as p < 0.05.
Results
The contact details provided to the research team from the various recruitment sources were available for 316 former elite-level rugby league players, who were all provided access to the questionnaire. In total, 154 (49%) of former elite male rugby league players completed the survey; however, only 130 (41%) of respondents provided answers to all of the questions pertaining to demographics, Neuro-QoL items, and concussion signs and symptoms to allow for analysis. The respondents represent a small percentage of the estimated cohort of all eligible (i.e., alive) former elite-level rugby league players (estimated ∼5500 eligible individuals).
Demographics
The average age of respondents was 54.2 years (SD = 13.4, range = 29–82). Approximately 93% (n = 121) of former players identified as “White” with 5.4% (n = 7) as Aboriginal or Torres Strait Islander and 1.5% (n = 2) identifying as Māori or Pasifika (Pacific Islander). The average number of reported concussion-related signs and symptoms during the respondent’s career was 37 (SD = 27.7, range = 0–117). When stratified by concussion-related signs and symptoms, those with the most signs and symptoms were younger than those with the least concussion signs and symptoms (e.g., age for quartile 1: M = 62.1, SD = 11.2, age for quartile 4: M = 47.8, SD = 10.6; see Table 1).
Cognitive Risk Factors Stratified by Concussion Signs and Symptoms Sustained Over Elite Rugby League Career (N = 130)
Body mass index (BMI) ≥30 kg/m2 considered obese; high-intensity exercise = running METS + jogging METS + aerobic METS; cardiovascular disease = any diagnosis or prescription medication for either hypertension, heart failure, arrhythmia, heart attack, any history of heart surgery, stroke, and high cholesterol.
ATSI, Aboriginal or Torres Strait Islander; h, hours; IQR, interquartile range; kg, kilograms; M, mean; m, meter; Md, median; METS, metabolic equivalents; N/n, sample size; Neuro-QoL, Quality of Life in Neurological Disorders–Applied Cognition General Concerns; PHQ, Patient Health Questionnaire; Q, quarter; QoL, quality of life; SD, standard deviation; T, T-score.
Predictors of perceived cognitive impairment
A total of 43.1% (n = 56) of respondents reported perceived cognitive impairment (i.e., a T-score ≤40). The average age was similar between those participants with and those participants without perceived cognitive impairment (54.3 ± 13.9 years vs. 54.0 ± 13.1 years, t = −0.137, p = 1.77). Using linear regression, there were statistically significant associations between worse perceived cognitive function (measured as a continuous variable) and the following modifiable risk factors, when adjusted for age: lack of sleep (on average) (p = 0.006, unstandardized B = −10.67), history of stroke (p = 0.03, unstandardized B = −10.73), current clinically significant symptoms of anxiety (p < 0.001, unstandardized B = −10.17) or depression (p = 0.04, unstandardized B = −5.76), physical impairment (p = 0.003, unstandardized B = −9.73), the total number of risk factors (p = <0.001, unstandardized B = −2.10), and ≥3 risk factors (p = 0.007, unstandardized B = −4.69) (Table 2). There was no statistically significant association between concussion-like signs and symptoms during playing years and perceived cognitive impairment.
Association of Potentially Modifiable or Treatable Risk Factors with Moderate or Severe Impairment in Cognition-Related Quality of Life (N = 130)
The dependent variable for each model was the continuous Neuro-QoL Cognition T-score. For binary predictor variables, having the risk factor was coded 1 and the absence of the risk factor was coded 0, regardless of the scale of the raw or the T-scores. The unstandardized B values represent how many T-score points the Neuro-QoL Cognition scale changes when that risk factor is present. Each risk factor was entered as the independent variable in separate models adjusted for age at the time of questionnaire completion. For variables with more than one level, including alcohol, body mass index, and concussion signs/symptoms, only the most severe is shown. Physical activity in metabolic equivalent (METS) hours/week was calculated from the reported frequency of five types of exercises, and then divided into quartiles. Low T-scores indicate worse physical impairment, while high T-scores indicate worse pain interference. Current/history of smoking variable = any history of smoking (i.e., current smoker or history of smoking). There were 12 risk factors used to calculate the Risk Factor Sum: current/history of smoking; heavy alcohol consumption (≥14 units/week); low physical activity (Quartile 1 vs. 4), sleep apnea; low sleep (≤5 h average night); body mass index (≥30 kg/m2); any cardiovascular disease; diabetes; significant physical impairment (T ≤40); significant pain interference (T ≥60); depressive symptoms (PHQ-2 ≥ 3); and anxiety symptoms (PHQ-2 ≥3).
p < 0.05.
p < 0.001.
B, unstandardized B coefficient; CI, confidence interval; kg, kilograms; m, meter; Neuro-QoL, Quality of Life in Neurological Disorders–Applied Cognition General Concerns; PHQ, Patient Health Questionnaire.
When a linear regression model was used to predict perceived cognitive functioning (measured as a continuous variable) after adjusting for age and for the number of concussion-related signs and symptoms during playing years, results were similar to the results when only adjusting for age. The statistically significant predictors were lack of sleep (p = 0.006, unstandardized B = −10.70), current clinically significant symptoms of anxiety (p < 0.001, unstandardized B = −10.23), physical impairment (p = 0.003, unstandardized B = −9.67), total number of risk factors (p = <0.001, unstandardized B = −2.15) and ≥3 risk factors (p = 0.006, unstandardized B = −4.75) (Table 2).
Physical activity and physical function
The average number of METS completed each week for participants was 39.9 (SD = 34.5, interquartile range [IQR] = 15.5–54.4). Of those who had perceived cognitive impairment (measured as a binary variable, n = 56), 10.8% were in the lowest quartile of METS completed (n = 14/56), compared with 13.8% of those who did not perceive themselves to have cognitive impairment (n = 18/74). A total of 38.5% (n = 50) of participants reported no weight training in an average week and 28.5% (n = 37) of all participants did not participate in any high-intensity physical activity. Significant pain interference in life was reported by 10 (7.7%) participants (T-score ≥60 in PROMIS Item Bank Pain Interference Short form 6b). Of those 10 participants who reported high pain interference in their lives, six reported perceived cognitive impairment.
Vascular risk factors
Nearly one-third (31.5%, n = 41) of respondents reported having a history of cardiovascular disease. Cardiovascular disease was not associated with greater perceived cognitive impairment (measured as a continuous variable) when controlling for age (p = 0.85) or age and concussion signs and symptoms (p = 0.76, Table 2). When adjusting for age and concussion signs and symptoms, current and/or past history of smoking was not associated with worse perceived cognitive function (p = 0.051, unstandardized B = −3.02, Table 2). There were 34% (n = 44) of respondents who were in the obese category (≥30 kg/m2), and obesity was not associated with worse perceived cognitive impairment when adjusted for age or age and concussion signs and symptoms (p = 0.44 and p = 0.47, respectively). Sleep apnea was also not associated with worse perceived cognitive function when adjusted for age (−3.6T-score points, p = 0.11) or both age and concussion signs and symptoms reported during their career (−3.4T-score points, p = 0.14).
Mental health risk factors
A minority of respondents reported that, in the past, a medical provider either provided a diagnosis and/or recommended medication for depression (n = 23, 18%) or anxiety (n = 17, 13%). Only 10 respondents (8%) reported clinically significant symptoms of depression at the time of the assessment, and nine respondents (7%) reported clinically significant symptoms of anxiety. Clinically significant symptoms of depression were associated with worse perceived cognitive function when adjusted for age (p = 0.04 unstandardized B = −5.76), but not age and concussion signs and symptoms (p = 0.051 unstandardized B = −5.62). Current clinical levels of anxiety are highly significant when adjusting for both (p = <0.001, unstandardized B = −10.17 and p = <0.001, unstandardized B = −10.23), respectively (Table 2).
Discussion
In this survey of former elite male rugby league players, ∼43% of respondents expressed perceived cognitive difficulties. This is two to three times the expected percentage when compared with a normative reference sample from the U.S. general population, whereby we set the cutoff for defining perceived cognitive difficulties at 16% of the U.S. general population. When adjusting for age, worse perceived cognitive functioning (measured as a continuous variable) was significantly associated with lack of sleep, history of stroke, current clinically significant symptoms of anxiety or depression, physical impairment, total number of risk factors, and ≥3 risk factors. When adjusting for age and for self-reported estimates of prior sport-related concussions, worse perceived cognitive functioning was significantly associated with lack of sleep, current clinically significant symptoms of anxiety, physical impairment, total number of risk factors, and ≥3 risk factors. The combination of the small sample size and the methods of data collection limits our ability to generalize to the wider population of former elite rugby league players. Nevertheless, if the patterns observed in the respondents were similar among all former elite rugby league players, these results would have important implications. There are a variety of modifiable factors that, if properly managed, may contribute to better perceived cognitive functioning in this population.
Our study, and the Roberts and colleagues’ study on former American National Football League players, 21 found that when stratified by past concussion-related signs and symptoms, those with the most signs/symptoms were younger than those with the least concussion signs/symptoms. Over the last 3–4 decades, the game of rugby league has become faster, players are larger in size, and the contact moments involve greater force. 53 The recent change in the game, respondent’s ability to recall events which, on average, occurred 20 years ago, concussion literacy, and recall biases might all contribute to why younger former players’ retrospectively self-reported greater concussion signs/symptoms scores during their playing years.
Meta-analyses of data from the general population have previously reported that being physically active has protective effects with respect to developing all-cause dementia. 54 The participants in the current study reported relatively high levels of exercise. For comparison, the American Heart Association 55 recommends adults complete at least 12.4 METS per week to gain the health benefits of exercise, which 78.5% of our population met or exceeded. The lowest quartile of activity in our study was <15.7 METS per week. This differs from Roberts and colleagues’ study with former NFL players, in which the lowest quartile of activity was <5 METS per week. 21 Part of the reason our analyses did not find differences in perceived cognitive functioning between the lowest and highest physical activity quartiles may be due to the overall sample being physically active. Further, the amount of physical activity completed in each of these two studies may be influenced by the proportion of people who reported high levels of pain interference in their lives, given that 34.2% had significant pain interference in Roberts and colleagues’ study, compared with 8% in the current study. 21
Our sample had lower percentages who reported physical impairment (6%) and pain interference (8%) than was reported in the Harvard Football Players Health Study. The Harvard Football Players Health Study had 34.2% of respondents reporting high levels of pain interference in their daily life and 24.7% reporting high physical impairment. 21 The difference in physical impairment and pain interference between the participants in our study and former professional American football players could be due, at least in part, to the much higher injury rates in the NFL (395.8 per 1000 athletes at risk) compared with the NRL (26.4 per 1000 athletes at risk). 56,57
Most of the evidence for associations between hypertension, physical activity, obesity, and worse cognitive functioning in the general population has been through studies with older adults, or individuals in their 50s followed up for >10 years. 58 –60 Our study did not show a significant association between the presence of cardiovascular disease and perceived cognitive impairment. It is possible that the sample in the current study (average age 54.2 years) is not old enough to have developed the perceived cognitive impairment that has been reported in older samples in the general population. It will be important to continue to monitor this sample to identify if these associations develop as they age.
The prevalence of current clinically significant anxiety symptoms in the previous 12 months, as defined by the GAD-2, in this sample appears to be lower than the Australian general population (6.9% vs. 17.2%), though differences in methodologies may at least partially account for this discrepancy. 61 Current clinically significant symptoms of anxiety had the strongest association with perceived cognitive impairment in our participants. The two strongest predictors in the Football Players Health Study were the number of concussion-related signs and symptoms and pain interference in daily life. Pain interference and previous history of concussion signs and symptoms were not significant in the current study. Current clinically significant symptoms of anxiety were the third strongest predictor of perceived cognitive impairment in the Football Players Health Study at Harvard (with former NFL players), with a risk ratio of 2.46 when adjusted for age and race. 21 Anxiety has been shown to have a bi-directional relationship with mild cognitive impairment in adults over the age of 65, 62 but also modifiable, indirectly through exercise and physical activity. 62 –64
Perceived cognitive difficulties are common in people from the general population who are experiencing symptoms of depression, and these cognitive symptoms are cardinal diagnostic features of major depressive disorder. Former NFL players who report cognitive difficulties also endorse greater symptoms of depression. 21,65 The percentage of our sample with current, clinically significant symptoms of depression (7.7% in the past 2 weeks) was slightly higher than that reported by the Australian general population in the 12 months prior to 2022 (4.9%). 61 The figures from both of these studies are lower than the Harvard Football Players study, which reported that 20.8% of former NFL players had clinically significant depressive symptoms. 21 Depression is important to consider when conceptualizing cognitive difficulties reported by former athletes, and when conceptualizing treatment designed to improve daily functioning and quality of life.
Limitations
This study has several important limitations. First, despite the reasonable response rate from those who were contacted, the sample size is small and only represents a small percentage of the presumed population of former players. Second, recruitment bias is probable, such that individuals with serious cognitive impairment may not be able to complete a questionnaire. As such, it is difficult to determine whether the responses from this sample are reasonably representative of the entire former elite-level rugby league player population. Third, 15% of individuals who mailed the questionnaire back were excluded because they did not complete enough questions to be included in the analyses. Fourth, the use of normative reference values for some questionnaires is from the U.S. population because no Australian-based normative values were available. Fifth, the generalisability of the findings to the current generation of active players is unknown. It is possible that the improvement in the identification and the management of concussion—and interventions to reduce exposures to repetitive head impacts and risk of concussion—may mean that these results in former players who did not have the same level of identification and medical care are not generalizable. Finally, the past health history (including the concussion history) was self-reported and not able to be verified. Self-report information is well known for being prone to recall bias. 66 It is also possible that the “demand effect” (responding in a way that they believe the researcher’s want them to respond) may have influenced the participants’ responses, and “social desirability bias” may also have been at play, where participants may have over-reported physical activity and underreported smoking and alcohol intake, to provide answers that they perceive to be socially acceptable, rather than the truth.
Conclusion
In former elite-level male rugby league players, the factor with the greatest association with perceived cognitive function, after statistically adjusting for age, was a history of stroke, followed by lack of sleep, current clinical anxiety symptoms, physical impairment, current clinical depression symptoms, greater than or equal to three risk factors, and the total sum of risk factors. If the patterns observed in our sample were found to be similar across the wider population of former players, then early education and intervention to manage these risk factors could provide a path to reducing perceived cognitive impairment in retired rugby league players in the future.
Transparency, Rigor, and Reproducibility
This study was not preregistered. We recruited and surveyed former elite rugby league players. Eligibility for participation in this study included any former player who had played at least one first-grade rugby league game in the New South Wales Rugby League, Queensland Rugby League, Australian Rugby League, Super League, or National Rugby League competitions. Eligible participants with accompanying contact details were the only participants who were able to be recruited for this study. There were 41% of contacted eligible participants who responded fully to all questions. The respondents represent only a small percentage of the estimated cohort of all former elite rugby league players. The results of the analyses are presented in Tables 1 and 2 and the figure. All of the data specific to this questionnaire study are presented in the article. The data from the larger research program’s questionnaires will be analyzed and published in future studies. The data from these former athletes are not currently available to the public or other researchers for privacy reasons. This article will be published under a Creative Commons Open Access license, and upon publication will be freely available at https://www.liebertpub.com/loi/neu.
Footnotes
Authors’ Contributions
K.G.: Conceptualized the study design, curated the relevant data from the database, assisted in the conduct of the statistical analysis, drafted and edited the article, and approved the final version for submission. O.J.S.: Managed the research database (data curation), conducted the statistical analysis, wrote portions of the article provided editorial comments on drafts of the article, and approved the final version for submission. D.P.T.: Conceptualized the statistical analysis and provided support for conducting the analysis, wrote portions of the article and provided editorial comments on drafts of the article, and approved the final version for submission. K.L.Q.: Provided editorial comment on drafts of the article and approved the final version for submission. G.L.I.: Developed the survey questions and conceptualized the study design and methodology, provided editorial comments for all drafts of the article, and approved the final version for submission. A.J.G.: Developed the survey questions and conceptualized the study design and methodology, completed the data collection, provided supervision to the first author, wrote portions of the article provided editorial comment for all drafts of the article, and approved the final version for submission.
Author Disclosure Statement
K.G. and O.J.S. have no competing interests to declare. D.P.T. serves as a consultant (Senior Director of Research) for the National Football League (NFL) and a scientific adviser for HitIQ. He previously consulted for REACT Neuro, Inc. He has a consulting practice in forensic neuropsychology, including expert testimony, involving individuals who have sustained mild TBIs (including former athletes). He received research funding from Amgen, Inc. and Football Research Inc. K.L.Q. has been employed by New Zealand Rugby since 2000 and currently occupies the role of Chief Scientist, New Zealand Rugby. He was a member of World Rugby’s Scientific Committee from 2013 to 2024 and has contributed to various World Rugby working groups focused on player welfare issues from 2011 to the time of publication. He has received funding for travel and accommodation to attend World Rugby’s medical meetings. G.L.I. has served or serves as a scientific adviser for NanoDX®, Sway Operations, LLC, and Highmark, Inc. He has a clinical and consulting practice in forensic neuropsychology, including expert testimony, involving individuals who have sustained mild TBIs and former athletes. He has received past research support or funding from several test publishing companies, including ImPACT Applications, Inc., CNS Vital Signs, and Psychological Assessment Resources (PAR, Inc.). He has received research funding as a principal investigator from the NFL, and subcontract grant funding as a collaborator from the Harvard Integrated Program to Protect and Improve the Health of NFL Players Association Members. A.J.G. has a clinical practice in neuropsychology involving individuals who have sustained sport-related concussions (including current and former athletes). He is a contracted concussion consultant to Rugby Australia. He is the global clinical lead for the World Rugby Brain Health Service. He is a member of the World Rugby Concussion Working Group, and a member of the Australian Football League Concussion Scientific Advisory Committee. He has received travel funding or been reimbursed by professional sporting bodies, and commercial organizations for discussing or presenting sport-related concussion research at meetings, scientific conferences, workshops, and symposiums. Previous grant funding includes the NSW Sporting Injuries Committee, the Brain Foundation (Australia), an Australian-American Fulbright Commission Postdoctoral Award, a Hunter New England Local Health District, Research, Innovation, and Partnerships Health Research and Translation Centre and Clinical Research Fellowship Scheme, and the Hunter Medical Research Institute (HMRI), supported by Jennie Thomas, and the HMRI, supported by Anne Greaves. He is supported by a National Health and Medical Research Council (NHMRC) Investigator Grant. He acknowledges unrestricted philanthropic support from the Nick Tooth Foundation, and the National Rugby League for research in former elite-level rugby league players. None of the above entities were involved in the study design, analysis, interpretation, the writing of this article, or the decision to submit it for publication.
Funding Information
This research was part of a larger program of research conducted on former elite-level athletes. The principal investigators of this research program (A.J.G. and G.L.I.) have received unrestricted philanthropic support from the National Rugby League (NRL). The NRL was not involved in the study design; the collection, analysis, and interpretation of data; the writing of the report; or the decision to submit the article for publication. No honorarium, grant, or other form of payment was given to anyone to produce the article. G.L.I. acknowledges philanthropic support from ImPACT Applications, Inc., the Mooney-Reed Charitable Foundation, and the Schoen Adams Research Institute at Spaulding Rehabilitation.