JOURNAL OF NEUROTRAUMA 37:1021–1028 (April 15, 2020) Mary Ann Liebert, Inc. Original Articles DOI: 10.1089/neu.2019.6661 Self-Reported Cognitive Function and Mental Health Diagnoses among Former Professional American-Style Football Players Franziska Plessow,1,2 Alvaro Pascual-Leone,1,3,4 Caitlin M. McCracken,1,5 Jillian Baker,1 Supriya Krishnan,1 Aaron Baggish,1,6 Ann Connor,1,7 Theodore K. Courtney,1,8 Lee M. Nadler,1,9 Frank E. Speizer,1,8,10 Herman A. Taylor,1,11 Marc G. Weisskopf,1,8 Ross D. Zafonte,1,12 and William P. Meehan, III1,13 Abstract Clinical practice strongly relies on patients’ self-report. Former professional American-style football players are hesitant to seek help for mental health problems, but may be more willing to report cognitive symptoms. We sought to assess the association between cognitive symptoms and diagnosed mental health problems and quality of life among a cohort of former professional players. In a cross-sectional design, we assessed self-reported cognitive function using items from the Quality of Life in Neurological Disorders (Neuro-QOL) Item Bank. We then compared mental health diagnoses and quality of life, assessed by items from the Patient-Reported Outcome Measurement Information System (PROMIS), between former professional players reporting daily problems in cognitive function and former players not reporting daily cognitive problems. Of the 3758 former professional players included in the analysis, 40.0% reported daily problems due to cognitive dysfunction. Former players who reported daily cognitive problems were more likely to also report depression (18.0% vs. 3.3%, odds ratio [OR] = 6.42, 95% confidence interval [CI] [4.90–8.40]) and anxiety (19.1% vs. 4.3%, OR = 5.29, 95% CI [4.14–6.75]) than those without daily cognitive problems. Further, former players reporting daily cognitive problems were more likely to report memory loss and attention deficit(/hyperactivity) disorder and poorer general mental health, lower quality of life, less satisfaction with social activities and relationships, and more emotional problems. These findings highlight the potential of an assessment of cognitive symptoms for identifying former players with mental health, social, and emotional problems. Keywords: cognitive function; football; mental health; quality of life 1 Football Players’ Health Study at Harvard University, Harvard Medical School, Boston, Massachusetts. 2 Neuroendocrine Unit, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts. 3 Hinda and Arthur Marcus Institute for Aging Research and Center for Memory Health, Department of Neurology, Hebrew SeniorLife and Harvard Medical School, Boston, Massachusetts. 4 Guttmann Brain Health Institut, Institut Guttmann, Universitat Autònoma de Barcelona, Barcelona, Spain. 5 College of Pharmacy, Oregon State University/Oregon Health Science University, Portland, Oregon. 6 Cardiovascular Performance Program, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts. 7 Berenson-Allen Center for Noninvasive Brain Stimulation and Division for Cognitive Neurology, Department of Neurology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts. 8 Department of Environmental Health, Harvard T. H. Chan School of Public Health, Boston, Massachusetts. 9 Department of Medicine, Dana-Farber Cancer Institute, Boston, Massachusetts. 10 Channing Division of Network Medicine, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts. 11 Department of Medicine, Cardiovascular Research Institute, Morehouse Medical School, Atlanta, Georgia. 12 Department of Physical Medicine and Rehabilitation, Brigham and Women’s Hospital, Massachusetts General Hospital, Spaulding Rehabilitation Hospital, and Harvard Medical School, Boston, Massachusetts. 13Department of Pediatrics and Orthopedics, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts.  Franziska Plessow et al., 2019; Published by Mary Ann Liebert, Inc. This Open Access article is distributed under the terms of the Creative Commons License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. 1021 1022 PLESSOW ET AL. Introduction diagnosis of depression and/or anxiety by a health professional leading to prescribed pharmacological treatment, (2) a lower M ental health and quality of life among professional American-style football (ASF) players beyond their active playing years has become a key area of interest in sports medi- quality of life, and (3) unhealthy lifestyle habits (frequent alcohol consumption, cigarette smoking, and/or low levels of exercise). cine. Evidence suggests that careers in professional football may be associated with long-term neurocognitive sequelae and mental health Methods problems,1–7 and mental health problems may be more prevalent Study sample among active and/or former professional ASF players with a history In February 2015, the Football Players’ Health Study at Harvard of concussion.3,8–12 In a cross-sectional study of 2552 members of University identified 14,538 former active professional players the National Football League (NFL) Players Association – Retired who had signed a contract with the NFL between 1960 and 2013 Section, Guskiewicz and colleagues reported a lifetime prevalence and were no longer listed as active players and believed to be alive of depression of 11.1% and a positive association between the and eligible. This start year 1960 was chosen, as the transition to the number of recalled concussions and diagnosis of depression.8 In a helmet with a hard plastic shell, which was accompanied by major follow-up study of >1000 members of the original sample, Kerr and rule changes, was complete and well established by 1960. Of these coworkers reported a 9 year incidence of depression of 10.2%, with 14,538 players, we had potential contact information for 13,403 the incidence increasing as the number of recalled concussions former professional players, all of whom we attempted to contact. increased.9 In comparison, a nationwide survey determined the We believe that we reached 13,200 potential study participants. At percentage of men with current depression in the general United the time of this analysis, 3779 former professional ASF players (28.6%) had responded. Newly retired players continue to enroll, States population to be 5.5% for 20–39-year olds, 5.2% for 40– and we plan to follow them longitudinally. The four-page ques- 59-year olds, and 6.1% among ‡60 years of age.13 tionnaire assessing self-reported sociodemographic, play-related, The Football Players’ Health Study at Harvard University is health, quality of life, and lifestyle characteristics was sent in either designed to pursue research that can lead to improvements in health paper or electronic format. In our sample, 1308 former players and quality of life among former professional ASF players.14,15 A completed the questionnaire in paper format and 2471 completed it core element of our research approach is the close collaboration via Research Electronic Data Capture (REDCapTM). This study among scientists, former professional ASF players, and their was approved by the Institutional Review Board (IRB) of Beth families when identifying, prioritizing, and pursuing medical Israel Deaconess Medical Center, an affiliate of Harvard Medical problems of interest. Early in the course of the study, our focus School. The IRBs at the collaborating Harvard institutions Harvard groups revealed that, whereas former professional players are Medical School, Boston Children’s Hospital, and Massachusetts General Hospital ceded IRB review to the Beth Israel Deaconess willing to report cognitive difficulties, which are viewed by many Medical Center’s IRB’s review. Consent was obtained via an im- as related to repeated impacts to the head during play, they are plied consent model with the participant’s willingness to complete reluctant to report mental health problems, such as depression and the questionnaire serving as implied informed consent. For addi- anxiety. A recent qualitative study interviewing 25 current and tional details regarding the design and objectives of the Football former NFL players together with 27 family members revealed in Players’ Health Study, study sample, and determination of eligi- 15 interviews that concerns around managing mental health chal- bility, please refer to the study by Zafonte and coworkers.22 lenges created a barrier to seeking support for experienced mental distress.16 Cognitive function These observations are further supported by the literature. Even We assessed cognitive function using items from the Quality of though the culture of football may be growing more accepting of Life in Neurological Disorders (Neuro-QOL) Item Bank v1.0 for self-reported physical injuries during play,17 the self-report of Applied Cognition – General Concerns,23 an established reliable emotional problems does not seem to have evolved at the same rate. and valid tool for assessing perceived difficulties in memory, at- Men, in general, are less likely to report or seek treatment for tention, and decision making. Each item describes a cognitive mental health problems.18–20 Former professional ASF players may problem (e.g., ‘‘I had trouble thinking clearly’’) and asks respon- be even less inclined to report or seek help for non-physical in- dents to indicate how frequently they have experienced this prob- juries.21 This is especially problematic, as clinical practice strongly lem over the past 7 days. Answers are provided on a five-point relies on the self-reporting of mental health problems. Moreover, Likert-like scale including the options ‘‘never,’’ ‘‘rarely (once),’’ this bias is of immediate clinical relevance, as many former pro- ‘‘sometimes (two or three times),’’ ‘‘often (about once a day),’’ and fessional players may be experiencing symptoms of depression and ‘‘very often (several times a day).’’ The item bank provides a total of 18 questions loading on a single latent construct (applied cog- anxiety that are not reported to their treating clinicians, particularly nition) and allows for selection of items of highest relevance to a considering the high prevalence rates of mood disorders in middle- target population with the recommendation to use at least eight aged men.13 items in order to reliably estimate the ability reflected by the latent Because our focus groups reported a willingness to acknowledge construct.23 The 11 items used in this study (Table 1) were pilot cognitive symptoms, we sought to determine whether the reporting tested on former professional ASF players prior to inclusion in the of cognitive symptoms was associated with a diagnosis of de- initial intake questionnaire. pression and/or anxiety. If so, the predictive value of reported cognitive function as related to potentially treatable mental health Mental health problems might prove useful to clinicians caring for former pro- Respondents were defined as having depression, anxiety, fessional players as a means of screening for mental health prob- memory loss, or attention deficit(/hyperactivity) disorder (ADD/ lems. To this end, we analyzed data from an initial intake ADHD), if they reported currently taking medication prescribed by questionnaire of a cohort study of former professional ASF players a healthcare professional for the treatment of that disorder. We and tested the following hypotheses: Former professional players chose this conservative definition, as there are common miscon- who report daily cognitive problems are more likely to have (1) a ceptions regarding mental health diagnoses among the general MENTAL HEALTH AND PROFESSIONAL FOOTBALL 1023 Table 1. Frequency (%) of Self-Reported Concerns Regarding Cognitive Function in 3758 Former Professional American-Style Football Players Captured with 11 Questions from the Quality of Life in Neurological Disorders (Neuro-QOL) Item Bank v1.0 for Applied Cognition: General Concernsa Rarely Sometimes Often (about Very often (several Item Never (once) (2–3 times) once a day) times a day) In the past 7 days, 1. I had to read something several times 806 (21.5) 1092 (29.1) 1,057 (28.1) 405 (10.8) 397 (10.6) to understand it. 2. I had trouble keeping track of what I was doing 873 (23.2) 975 (26.0) 1,022 (27.2) 475 (12.6) 412 (11.0) if I was interrupted. 3. I had difficulty doing more than one thing at a time. 1036 (27.7) 996 (26.6) 891 (23.8) 432 (11.5) 392 (10.5) 4. I had trouble remembering new information, like 894 (23.8) 944 (25.1) 916 (24.4) 509 (13.6) 494 (13.2) phone numbers or simple instructions. 5. I had trouble thinking clearly. 1144 (30.5) 1054 (28.1) 830 (22.1) 412 (11.0) 316 (8.4) 6. My thinking was slow. 1142 (30.6) 1031 (27.6) 832 (22.3) 434 (11.6) 296 (7.9) 7. I had to work really hard to pay attention or 1197 (32.0) 999 (26.7) 782 (20.9) 423 (11.3) 346 (9.2) I would make a mistake. 8. I had trouble concentrating. 987 (26.4) 986 (26.4) 905 (24.2) 453 (12.1) 409 (10.9) 9. I had trouble remembering whether I did things 953 (25.4) 991 (26.4) 924 (24.7) 492 (13.1) 388 (10.4) I was supposed to do. 10. I had trouble making decisions. 1334 (35.7) 1046 (28.0) 762 (20.4) 328 (8.8) 266 (7.1) 11. I had trouble planning out steps of a task. 1443 (38.7) 1007 (27.0) 717 (19.2) 332 (8.9) 234 (6.3) a Twenty-one out of 3779 participants provided valid responses for <8 out of 11 Neuro-QOL items and were not included in the analysis. Total values of <3758 for a question indicate missing responses. population.24 In addition, we assessed the correlation between white, and other), ethnicity (Hispanic/Latino: yes/no), currently cognitive symptoms and self-reported current symptoms of de- employed (yes, in football, yes, outside of football, retired, and no), pression and/or anxiety using sum scores on the Neuro-QOL and domestic status (living with partner: yes/no), seasons of profes- the Patient Health Questionnaire for Depression and Anxiety sional football played, position played (positions mirroring each (PHQ-4). The four-item PHQ-4 represents a screening tool for other on the offense and defense were grouped together as appro- depression and anxiety in outpatient or home settings. Based on priate, taking into account the type and number of physical impacts self-reported symptom presence over the past 2 weeks on a four- and body composition of the player: offensive/defensive lineman, point Likert-like scale from 0 (‘‘not at all’’) to 3 (‘‘nearly every linebacker, defensive back/wide receiver, running back, tight end, day’’), the sum score distinguishes between no (0–2), mild (3–5), quarterback, kicker/punter), current cigarette use (yes/no), and di- moderate (6–8), and severe (9–12) depression and/or anxiety, agnosis of sleep apnea (yes/no) were self-reported. For race and providing a quantitative measure of current symptom severity in position played, participants could mark all options that applied. addition to the more conservative definition of medical manage- Exercise regimen was assessed by self-reported weekly engage- ment.25 Finally, we used the established cutoff of a PHQ-4 score of ment in the following types of exercise: walking for exercise, jog- ‡6 for a positive screening result for current depressive and/or ging, running, other aerobic (e.g., bicycling, stationary/elliptical anxiety symptoms (moderate to severe symptoms).26 machine/other), low-intensity exercise (e.g., yoga, Pilates, stretch- ing), and weight training (e.g., lifting free weights, using weight machines). Low level of exercise was defined conservatively as en- Quality of life gaging in <1 h of accumulated exercise activity weekly. Quality of life was assessed employing the widely used Global Health Scale 10-Item Bank v1.0/1.1 of the Patient-Reported Out- Data analysis come Measurement Information System (PROMIS).27 The items for general quality of life (Global02), mental health (Global04), Outlier values were corrected where possible using publicly satisfaction with social activities and relationships (Global05), and available data (e.g., for years played in the NFL) or were set to emotional problems (Global10) were answered on a five-point missing. For all analyses concerning cognitive function, partici- Likert-like scale with the categories ‘‘poor,’’ ‘‘fair,’’ ‘‘good,’’ pants were included only if they provided valid responses to at least ‘‘very good,’’ and ‘‘excellent.’’ 8 out of the 11 Neuro-QOL items, resulting in a final sample of 3758 former professional players. Other measures Based on their responses to the Neuro-QOL items, respondents were divided into two groups: (1) former professional players who Alcohol use was measured by the number of alcoholic beverages reported at least one daily problem with their applied cognitive (i.e., 12 oz. of beer, 5 oz. of wine, or one shot [2 cL] of alcohol, such abilities and (2) former professional players who did not report as gin, rum, or vodka) consumed weekly. The measure was created such daily concerns, and group differences were assessed using the based on two questions that asked about (1) the number of days a v2 test. For mental health outcomes, odds ratios adjusted (aOR) for participant drank alcoholic beverages during a typical week and (2) age and sleep apnea, variables known to affect cognitive func- the number of drinks consumed on a typical day that the person tion,29,30 were determined using logistical regression analyses. In drank. Frequent drinking was defined as consuming ‡15 alcoholic addition, to test whether more severe cognitive symptoms are as- beverages weekly, which is the Centers for Disease Control and sociated with more depressive and/or anxiety symptoms among Prevention’s classification for heavy drinking.28 those with a diagnosis of depression and/or anxiety, we conducted Age, height, weight, race (American Indian/Alaskan Native, a nonparametric correlation analysis of Neuro-QOL and PHQ-4 Asian, black/African American, Native Hawaiian/Pacific Islander, sum scores in this subgroup. Finally, we examined the relevance of 1024 PLESSOW ET AL. the presence of current depression and/or anxiety for the depres- Of the 3758 former professional ASF players included in the sion/anxiety and cognitive function link. Using v2 tests, we com- analysis, the majority reported having had at least one cognitive pared how many former professional players with a diagnosis problem in the preceding 7 days (Table 1). More than a third of depression and/or anxiety but no current moderate to severe (n = 1502; 40.0%) reported experiencing at least one cognitive symptoms of depression and/or anxiety (PHQ-4 < 6) reported daily problem on a daily basis, whereas 2256 former professional players cognitive concerns (%) with that number among (1) former pro- (60.0%) did not report any daily cognitive problems. Former pro- fessional players with a diagnosis of depression and/or anxiety and current depression and/or anxiety (PHQ-4 ‡ 6) and (2) former fessional players with daily problems in cognitive function were professional players without a diagnosis of depression and/or anx- more likely to have a diagnosis of depression, anxiety, memory iety and no current moderate or severe depressive and/or anxiety loss, or ADD/ADHD (Table 3). These effects remained significant symptoms (PHQ-4 < 6). after adjusting for age and sleep apnea: depression: aOR = 6.17, 95% confidence interval (CI) [4.67–8.15]; anxiety: aOR = 5.06, Results 95% CI [3.93–6.52]; memory loss: aOR = 21.13, 95% CI [11.57– 38.58]; and ADD/ADHD: aOR = 4.73, 95% CI [3.22–6.95]. Respondents ranged from 24 to 89 years of age (mean age: 52.77 Among former professional ASF players diagnosed with de- years); 1.2% were American Indian or Alaska Native, 0.3% were pression and/or anxiety who also completed the PHQ-4, we found Asian, 37.5% were black/African American, 1.1% were Native a linear relationship between Neuro-QOL and PHQ-4 sum scores, Hawaiian or other Pacific Islanders, 60.1% were white, and 1.7% indicating that in this subgroup, worse cognitive function was as- were ‘‘other.’’ A total of 1.4% identified as Hispanic, 97.8% as sociated with more pronounced depressive and/or anxiety symp- identified as non-Hispanic, and 0.8% chose not to respond. They toms, rs = 0.65, p < 0.0001. Also, among former professional played an average of seven seasons in the NFL. All positions were players with a diagnosis of depression and/or anxiety who did not represented (Table 2). report significant current depressive and/or anxiety symptoms, Table 2. Basic Characteristics (n = 3779 Former 52.2% reported daily cognitive concerns. This percentage was Professional American-Style Football Players)a smaller than the percentage of former professional ASF players with a diagnosis of depression and/or anxiety who also reported Characteristic Mean – SD or n (%) current depression and/or anxiety (90.3%; p < 0.001), but larger than the proportion of former professional players who were not Sociodemographics diagnosed with depression and/or anxiety and did not report current Age (years) 52.77 – 14.14 moderate to severe depressive and/or anxiety symptoms (27.3%; Raceb p < 0.001). In addition, former professional players with a diagnosis American Indian/Alaska Native 44 (1.2) and significant current depressive and/or anxiety symptoms differed Asian 10 (0.3) from those with neither the diagnosis of anxiety and/or depression Black/African American 1417 (37.5) Native Hawaiian/Other Pacific Islander 41 (1.1) nor current symptoms ( p < 0.001; Fig. 1). White 2272 (60.1) Compared with those without daily cognitive problems, former Other 63 (1.7) professional ASF players with problems in memory, attention, Ethnicity and/or decision making on a daily basis also reported a lower Hispanic/Latino 51 (1.4) general quality of life, poorer mental health, less satisfaction with Not Hispanic/Latino 3593 (97.8) social activities and relationships, and more emotional problems Chose not to respond 30 (0.8) (Table 3). Finally, former professional players experiencing daily Currently employed problems in their cognitive function were more likely to be smokers Yes, in football 355 (9.5) and to report exercising <1 h per week. No difference in the report Yes, outside of football 2172 (58.3) of alcohol consumption was found between groups (Table 3). Retired 700 (18.8) No 497 (13.4) Domestic status: living with partner Discussion Yes 3061 (81.8) No 683 (18.2) Former professional ASF players who experienced daily cog- Play-related characteristics nitive problems had higher odds of also carrying the diagnosis of Number of seasons played professionally 6.78 – 3.84 at least one of the following: depression, anxiety, memory loss, or Position playedb ADD/ADHD than those who did not report day-to-day cognitive Offensive/Defensive line 1369 (36.2) issues (Hypothesis 1). An additional analysis in the subgroup of Linebacker 631 (16.7) former professional players diagnosed with depression and/or Defensive back/Wide receiver 973 (25.7) anxiety showed a linear correlation between the extent of cognitive Running back 410 (10.9) Tight end 313 (8.3) problems and the severity of depressive and/or anxiety symptoms, Quarterback 182 (4.8) suggestive of an association between cognitive and emotional Kicker/Punter 208 (5.5) problems in this population. Further, we observed that among those Physical health former players with a diagnosis of depression and/or anxiety but no BMI 31.22 – 4.96 significant current depressive and/or anxiety symptoms, the per- Sleep apnea centage of those with daily cognitive concerns was almost twofold Yes 843 (22.8) higher than in former players without a diagnosis or current report No 2853 (77.2) of depression and/or anxiety (52.2% vs. 27.3%, respectively). In a Total values of <3,779 for a question indicate missing responses. former players with a diagnosis of depression and/or anxiety and b Participants had the option to select more than one category. current experience of moderate to severe symptoms, the percentage BMI, body mass index; SD, standard deviation. of those with daily cognitive problems was as high as 90.3%. This MENTAL HEALTH AND PROFESSIONAL FOOTBALL 1025 Table 3. Mental Health, Quality of Life, and Lifestyle Factors in Former Professional American-Style Football Players Reporting Daily Problems in Cognitive Functiona Compared with Those Who Do Not Report Daily Cognitive Problems Daily cognitive problems No daily cognitive problems Characteristic (n = 1502) (n = 2256) OR [95% CI] b Mental health Depression Yes 259 (18.0) 73 (3.3) 6.42 No 1183 (82.0) 2140 (96.7) [4.90–8.40] Anxiety Yes 276 (19.1) 94 (4.3) 5.29 No 1171 (80.9) 2109 (95.7) [4.14–6.75] Memory loss Yes 137 (9.5) 12 (0.5) 19.14 No 1300 (90.5) 2179 (99.5) [10.56–34.66] ADD/ADHD Yes 118 (8.1) 37 (1.7) 5.16 No 1334 (91.9) 2157 (98.3) [3.54–7.51] Quality of life General quality of life Poor/fair 553 (37.5) 147 (6.6) 8.48 Good to excellent 923 (62.5) 2080 (93.4) [6.96–10.33] Mental health Poor/fair 935 (62.7) 246 (11.0) 13.63 Good to excellent 557 (37.3) 1997 (89.0) [11.51–16.14] Satisfaction with social activities and relationships Poor/fair 725 (48.6) 244 (10.9) 7.73 Good to excellent 767 (51.4) 1995 (89.1) [6.54–9.14] Emotional problems Often/always 717 (47.9) 163 (7.3) 11.77 Never to sometimes 779 (52.1) 2085 (92.7) [9.75–14.22] Lifestyle Frequent drinkingc Yes 326 (22.0) 447 (20.0) 1.13 No 1154 (78.0) 1787 (80.0) [0.96–1.33] Current smoking Yes 66 (4.4) 52 (2.3) 1.96 No 1419 (95.6) 2189 (97.7) [1.35–2.83] Low level of exercised Yes 416 (27.7) 310 (13.8) 2.41 No 1084 (72.3) 1943 (86.2) [2.04–2.84] Data are presented as n (%). a Assessed with 11 questions from the Quality of Life in Neurological Disorders (Neuro-QOL) Item Bank v1.0 for Applied Cognition – General Concerns. Twenty-one out of 3779 participants provided valid responses for <8 out of 11 Neuro-QOL items and were not included in the analysis. Total values of <3758 for an outcome indicate missing responses. bBased on self-report of currently taking medication for the treatment of that disorder. c Defined as ‡15 alcoholic beverages (i.e., 12 oz. of beer, 5 oz. of wine, or one shot [2 cL] of alcohol, such as gin, rum, or vodka) per week based on the Centers for Disease Control and Prevention’s classification for heavy drinking.26 d Defined as <1 h/week of walking, jogging, running, other aerobic (e.g., bicycling, stationary/elliptical machine/other), low-intensity exercise (e.g., yoga, Pilates, stretching), and/or weight training. ADD, attention deficit disorder; ADHD, attention deficit/hyperactivity disorder; CI, confidence interval; OR, odds ratio. finding could be interpreted as an increased vulnerability for cur- fective treatments are available when diagnosed. Previous studies rent daily cognitive concerns in those with a diagnosis of depres- suggest that high-level football players are reluctant to seek help sion and/or anxiety, even in the absence of current symptoms (e.g., for mental health problems, and links have been drawn to hy- because of successful pharmacological management). In addition, permasculinity and long-time exposure to a value system reinfor- compared with past players without day-to-day cognitive problems, cing masculine concepts, such as competition and toughness.21,31 former professional ASF players reporting daily cognitive concerns Given the reliance on the self-reporting of symptoms in making the reported a lower quality of life (Hypothesis 2) and were more likely diagnosis of mental health problems, this potential under-reporting to smoke and engage in exercise at a lower level (Hypothesis 3). of symptoms increases the risk that former players with depression If former players are, in fact, reluctant to disclose mental health and/or anxiety will be un- or misdiagnosed. If, however, former symptoms but willing to report cognitive symptoms, then the as- players are more likely to report cognitive symptoms, the assess- sessment of cognitive difficulties might be used as a means of ment of cognitive function could be used to identify former players identifying a group of former players at risk for depression and at higher risk for mental health problems. Although further research anxiety, conditions known to reduce quality of life, for which ef- is needed, the combination of questions chosen for this study 1026 PLESSOW ET AL. FIG. 1. Percentage of former professional American-style football players reporting daily cognitive problems depending on (1) diagnosis of depression and/or anxiety (defined by currently prescribed medication for these conditions by a healthcare professional; yes/no) and (2) current experience of moderate to severe depressive and/or anxiety symptoms (defined by a Patient Health Questionnaire for Depression and Anxiety [PHQ-4] score ‡6; yes/no). ***p < 0.001. (Table 1) might be a valuable screening tool to engage in further daily cognitive problems. Therefore, we cannot confirm that such conversations about cognitive challenges and conditions that former self-reports accurately reflect true cognitive deficits. Previous lit- players (and individuals from similar professional groups) might be erature, however, suggests that self-reported cognitive problems reluctant to raise with their clinicians, specifically mental health indicate developing cognitive problems even before standard problems, even if these problems are not initially endorsed. neuropsychological testing captures them.41 In addition, our self- The close link between cognitive problems and depressive and report measure of cognitive function was defined through the use anxiety symptoms in former professional ASF players shown in this of a unique selection of questions from the Neuro-QOL Item study is in alignment with previous reports of a close relationship be- Bank,23 which is an established reliable and validated assessment. tween the cognitive and emotional domains in other populations. In this population, however, it is possible that their self-perception Cognitive impairments are frequently reported in patients with a di- of cognitive difficulties has been affected by widespread media agnosis of depression or anxiety.32,33 They have been shown to repre- reporting about later-life cognitive problems in former ASF sent a marker of disease severity with the extent of cognitive problems players. Second, we included in our definition of mental health being correlated with symptom severity and illness duration.33,34 outcomes self-reports of prescribed medication for the according The current data represent an association and do not allow for mental health condition or recommendations for such medications. inferences about causation. We cannot determine whether among Therefore, there were likely respondents who had these condition former professional ASF players, cognitive problems lead to mental but were offered only non-pharmacological therapies. Because the health challenges, such as depressive and anxiety symptoms, whether information on symptoms and diagnoses were collected simulta- mental health problems cause cognitive impairments, or whether neously, we cannot infer the directionality of these findings. Third, third factors, such as injuries, cause both cognitive problems and our sample represents 28.6% of eligible, potentially contactable mental health challenges. Previous studies, however, suggest that former players. Although this response rate is similar to previous cognitive impairments often precede and predict emotional symp- surveys of professional athlete cohorts,42 the possibility of a bias toms.35–37 Failure to suppress engagement with threatening stimuli among the cohort cannot be excluded. Finally, additional factors, (anxiety disorders), to disengage from mood-congruent thoughts or such as financial stability, socioeconomic status, self-perceived external stimuli (depression), and to apply emotion regulation social status and support, and loss of a meaningful group role strategies (both depression and anxiety disorders), all relying on might play a role in the development and maintenance of and adequate cognitive function, may represent key contributing factors support-seeking for mental health challenges and might be useful to developing and maintaining depressive and anxiety symptoms.38 to include in follow-up studies. Moreover, alterations in the structure and function of neural networks Our study reveals that daily cognitive problems are strongly as- subserving cognitive function represent a transdiagnostic core fea- sociated with depression and anxiety among former professional ture linked to the emotional psychopathology, providing a biological ASF players. Therefore, as a primary care provider or sports medi- mechanism underlying the behaviorally observed link and its gen- cine expert, it is important to remember that when confronted with eralizability across mental disorders.39,40 Although further studies complaints about cognitive performance, depression and/or anx- are required to illuminate the causes of the reported relationship iety might coexist. Validated and standardized tools are available between self-reported cognitive function and depressive and anxiety to screen for current depressive and anxiety symptoms, including symptoms in their complex relationship with other biological, psy- the items (including short forms for efficient assessment) of chological, and social factors and characteristics of the players’ ex- the Depression and Anxiety domains of the PROMIS (Mental perience, the clinical implication of gaining knowledge about mental Health/Emotional Distress; www.healthmeasures.net/explore- health by assessing self-reported cognitive function represents a measurement-systems/promis)27 or the PHQ-4 (used in this valuable pragmatic strategy to identify individuals in need of further study).25 For more formal testing, the items assessing the Fear and mental health counseling. Sadness constructs as part of the National Institutes of Mental Our findings must be considered in light of the study’s limita- Health Toolbox (Emotion Battery/Negative Affect; nihtoolbox tions. First, we relied on self-reporting for the determination of .org) should be considered. MENTAL HEALTH AND PROFESSIONAL FOOTBALL 1027 Acknowledgments manuscript; and decision to submit the manuscript for publica- tion. Other authors report that no potential or actual competing We acknowledge and thank the former ASF player community, financial interests exist. our participants and advisors, and the past and present staff of the Football Players’ Health Study at Harvard University for their dedication toward our joint goal of improving the health of former References ASF players. 1. Amen, D.G., Newberg, A., Thatcher, R., Jin, Y., Wu, J., Keator, D., and Willeumier, K. (2011). Impact of playing American professional football on long-term brain function. J. Neuropsychiatry Clin. Neu- Funding Information rosci. 23, 98–106. The Football Players’ Health Study is funded by a grant from 2. Casson, I.R., Viano, D.C., Haacke, E.M., Kou, Z., and LeStrange, D.G. (2014). Is there chronic brain damage in retired NFL players? the National Football League Players Association (NFLPA), Neuroradiology, neuropsychology, and neurology examinations of 45 Washington, DC. retired players. Sports Health 6, 384–395. 3. Didehbani, N., Munro Cullum, C., Mansinghani, S., Conover, H. and Hart, J., Jr. (2013). Depressive symptoms and concussions in aging Author Disclosure Statement retired NFL players. Arch. Clin. Neuropsychol. 28, 418–424. 4. Hart, J., Jr., Kraut, M.A., Womack, K.B., Strain, J., Didehbani, N., Dr. Alvaro Pascual-Leone was partly supported by the Sidney R. Bartz, E., Conover, H., Mansinghani, S., Lu, H., and Cullum, C.M. Baer Jr. Foundation, the National Institutes of Health (R01MH100186, (2013). Neuroimaging of cognitive dysfunction and depression in aging retired National Football League players: A cross-sectional R21AG051846, R01MH111875, R01MH115949, R01MH117063, study. JAMA Neurol. 70, 326–335. R24AG06142, and P01AG031720), the National Science Foundation, 5. Multani, N., Goswami, R., Khodadadi, M., Ebraheem, A., Davis, DARPA, the Football Players’ Health Study at Harvard University, K.D., Tator, C.H., Wennberg, R., Mikulis, D.J., Ezerins, L., and and Harvard Catalyst j The Harvard Clinical and Translational Science Tartaglia, M.C. (2016). The association between white-matter tract Center (NCRR and the NCATS NIH, UL1 RR025758). Dr. Pascual- abnormalities, and neuropsychiatric and cognitive symptoms in retired professional football players with multiple concussions. J. Neurol. Leone serves on the scientific advisory boards for Neosync, Neuronix, 263, 1332–1341. Starlab Neuroscience, Neuroelectrics, Magstim Inc., Constant 6. Strain, J., Didehbani, N., Cullum, C.M., Mansinghani, S., Conover, H., Therapy, and Cognito, and is listed as an inventor on several issued Kraut, M.A., Hart, J., Jr., and Womack, K.B. (2013). Depressive and pending patents on the real-time integration of transcranial symptoms and white matter dysfunction in retired NFL players with magnetic stimulation with electroencephalography and magnetic concussion history. Neurology 81, 25–32. 7. Strain, J.F., Didehbani, N., Spence, J., Conover, H., Bartz, E.K., resonance imaging. Dr. Aaron Baggish has received funding from Mansinghani, S., Jeroudi, M.K., Rao, N.K., Fields, L.M., Kraut, M.A., the National Institutes of Health/National Heart, Lung, and Blood Cullum, C.M., Hart, J., Jr., and Womack, K.B. (2017). White matter Institute, the NFLPA, the American Heart Association, the Amer- changes and confrontation naming in retired aging National Football ican Society of Echocardiography and receives compensation for League athletes. J. Neurotrauma 34, 372–379. 8. Guskiewicz, K.M., Marshall, S.W., Bailes, J., McCrea, M., Harding, his role as team cardiologist from United States Soccer, United H.P., Jr., Matthews, A., Mihalik, J.R., and Cantu, R.C. (2007). Re- States Rowing, the New England Patriots, the Boston Bruins, the New current concussion and risk of depression in retired professional England Revolution, and Harvard University. Dr. Ross D. Zafonte was football players. Med. Sci. Sports Exerc. 39, 903–909. partially supported by the National Institute on Disability, Independent 9. Kerr, Z.Y., Marshall, S.W., Harding, H.P., Jr., and Guskiewicz, K.M. Living, and Rehabilitation Research (90DP0039-03-00, 90SI5007-02- (2012). Nine-year risk of depression diagnosis increases with in- creasing self-reported concussions in retired professional football 04, 90DP0060), the National Institutes of Health (4U01NS086090-04, players. Am. J. Sports Med. 40, 2206–2212. 5R24HD082302-02, 5U01NS091951-03), and United States Army 10. Lehman, E.J., Hein, M.J., Baron, S.L., and Gersic, C.M. (2012). Medical Research and Materiel Command (W81XWH-112-0210). He Neurodegenerative causes of death among retired National Football also serves as Co-PI on a T-32 and receives funding from the Football League players. Neurology 79, 1970–1974. 11. Schwenk, T.L., Gorenflo, D.W., Dopp, R.R., and Hipple, E. (2007). Players’ Health Study at Harvard University, which is funded by the Depression and pain in retired professional football players. Med. Sci. NFLPA. Dr. Zafonte received royalties from Oakstone for an educa- Sports Exerc. 39, 599–605. tional CD- Physical Medicine and Rehabilitation: A Comprehensive 12. Solomon, G.S. and Haase, R.F. (2008). Biopsychosocial character- Review and Demos publishing for serving as co-editor of the text Brain istics and neurocognitive test performance in National Football League players: an initial assessment. Arch. Clin. Neuropsychol. 23, Injury Medicine. Dr. Zafonte serves on the Scientific Advisory Board 563–577. of Myomo, Oxeia Biopharma, Elminda, and Biodirection. He also 13. Brody, D.J., Pratt, L.A., and Hughes, J.P. (2018). Prevalence of de- evaluates patients in the Massachusetts General Hospital Brain and pression among adults aged 20 and over: United States, 2013–2016, Body-TRUST Program, which is funded by the NFLPA. Dr. William NCHS Data Brief, No. 303. National Center for Health Statistics, P. Meehan, III receives royalties from ABC-Clio publishing for the Hyattsville, MD. 14. Churchill, T.W., Krishnan, S., Weisskopf, M., Yates, B.A., Speizer, sale of his books, Kids, Sports, and Concussion: A Guide for Coaches F.E., Kim, J.H., Nadler, L.E., Pascual-Leone, A., Zafonte, R. and and Parents, and Concussions; Springer International for the book Baggish, A.L. (2018). Weight gain and health affliction among former Head and Neck Injuries in Young Athletes; and Wolters Kluwer for National Football League players. Am. J. Med. 131, 1491–1498. working as an author for UpToDate. His research is funded, in part, by 15. Meehan, W.P., Weisskopf, M.G., Krishnan, S., McCracken, C., Zafonte, R., Taylor, H.A., Baggish, A., Pascual-Leone, A., Nadler, L.M., and philanthropic support from the National Hockey League Alumni As- Speizer, F.E. (2018). Relation of anterior cruciate ligament rears to po- sociation through the Corey C. Griffin Pro-Am Tournament and by a tential chronic cardiovascular diseases. Am. J. Cardiol. 122, 1879–1884. grant from the NFL. All of the other authors named on this study are or 16. McGraw, S.A., Deubert, C.R., Lynch, H.F., Nozzolillo, A., Taylor, L., were either partially or fully supported by the Football Players’ Health and Cohen, I.G. (2018). Life on an emotional roller coaster: NFL Study at Harvard University, which is in turned sponsored by the players and their family members’ perspectives on player mental health. J. Clin. Sport Psychol. 12, 404–431. NFLPA. Please note that the NFLPA had no role in: the design and 17. Anderson, E., and Kian, E.M. (2012). Examining media contestation conduct of the study; collection, management, analysis, and in- of masculinity and head trauma in the National Football League. Men terpretation of the data; preparation, review, or approval of the Masc. 15, 152–173. 1028 PLESSOW ET AL. 18. Addis, M.E., and Cohane, G.H. (2005). Social scientific paradigms of evidence of episodic memory dysfunction. J. Psychiatr. Res. 39, masculinity and their implications for research and practice in men’s 207–214. mental health. J. Clin. Psychol. 61, 633–647. 33. Snyder, H.R. (2013). Major depressive disorder is associated with 19. Addis, M.E., and Mahalik, J.R. (2003). Men, masculinity, and the broad impairments on neuropsychological measures of executive contexts of help seeking. Am. Psychol. 58, 5–14. function: a meta-analysis and review. Psychol. Bull. 139, 81–132. 20. Brooks, G.R. (2001). Masculinity and men’s mental health. J. Am. 34. Hasselbalch, B.J., Knorr, U., Hasselbalch, S.G., Gade, A., and Kes- Coll. Health. 49, 285–297. sing, L.V. (2013). The cumulative load of depressive illness is asso- 21. Jones, T.V. (2016). Predictors of perceptions of mental illness and ciated with cognitive function in the remitted state of unipolar averseness to help: a survey of elite football players. J. Ment. Health, depressive disorder. Eur. Psychiatry 28, 349–355. 25, 422–427. 35. Airaksinen, E., Wahlin, A., Forsell, Y., and Larsson, M. (2007). Low 22. Zafonte, R., Pascual-Leone, A., Baggish, A., Weisskopf, M.G., Tay- episodic memory performance as a premorbid marker of depression: lor, H.A., Connor, A., Baker, J., Cohan, S., Valdivia, C., Courtney, evidence from a 3-year follow-up. Acta Psychiatr. Scand. 115, 458–465. T.K., Cohen, I.G., Speizer, F.E., and Nadler, L.M. (2019). The 36. Franz, C.E., Lyons, M.J., O’Brien, R., Panizzon, M.S., Kim, K., Bhat, Football Players’ Health Study at Harvard University: design and R., Grant, M.D., Toomey, R., Eisen, S., Xian, H., and Kremen, W.S. objectives. Am. J. Ind. Med. 62, 643–654. (2011). A 35-year longitudinal assessment of cognition and midlife 23. National Institute of Neurological Diseases and Stroke (2010). User depression symptoms: The Vietnam Era Twin Study of Aging. Am. J. Manual for the Quality of Life in Neurological Disorders (Neuro- Geriatr. Psychiatry 19, 559–570. QOL) Measures, version 1.0, September 2010. http://www.neuroqol 37. Simons, C.J., Jacobs, N., Derom, C., Thiery, E., Jolles, J., van Os, J., .org (last accessed November 22, 2019). and Krabbendam, L. (2009). Cognition as predictor of current and 24. Parker, R. (2000). Health literacy: a challenge for American patients follow-up depressive symptoms in the general population. Acta Psy- and their health care providers. Health Promot. Int. 15, 277–283. chiatr. Scand. 120, 45–52. 25. Kroenke, K., Spitzer, R.L., Williams, J.B., and Lowe, B. (2009). An 38. Goschke, T. (2014). Dysfunctions of decision-making and cognitive ultra-brief screening scale for anxiety and depression: the PHQ-4. control as transdiagnostic mechanisms of mental disorders: advances, Psychosomatics 50, 613–621. gaps, and needs in current research. Int. J. Methods Psychiatr. Res. 23, 26. Lowe, B., Wahl, I., Rose, M., Spitzer, C., Glaesmer, H., Wingenfeld, Suppl. 1, 41–57. K., Schneider, A., and Brahler, E. (2010). A 4-item measure of de- 39. Buckholtz, J.W. and Meyer-Lindenberg, A. (2012). Psychopathology pression and anxiety: validation and standardization of the Patient and the human connectome: toward a transdiagnostic model of risk for Health Questionnaire-4 (PHQ-4) in the general population. J. Affect. mental illness. Neuron 74, 990–1004. Disord. 122, 86–95. 40. McTeague, L.M., Goodkind, M.S., and Etkin, A. (2016). Transdiag- 27. Cella, D., Riley, W., Stone, A., Rothrock, N., Reeve, B., Yount, S., nostic impairment of cognitive control in mental illness. J. Psychiatr. Amtmann, D., Bode, R., Buysse, D., Choi, S., Cook, K., Devellis, R., Res. 83, 37–46. DeWalt, D., Fries, J.F., Gershon, R., Hahn, E.A., Lai, J.S., Pilkonis, P., 41. van der Flier, W.M., van Buchem, M.A., Weverling-Rijnsburger, Revicki, D., Rose, M., Weinfurt, K., Hays, R. and PROMIS Co- A.W., Mutsaers, E.R., Bollen, E.L., Admiraal-Behloul, F., Wes- operative Group. (2010). The Patient-Reported Outcomes Measure- tendorp, R.G., and Middelkoop, H.A. (2004). Memory complaints in ment Information System (PROMIS) developed and tested its first patients with normal cognition are associated with smaller hippo- wave of adult self-reported health outcome item banks: 2005–2008. J. campal volumes. J. Neurol. 251, 671–675. Clin. Epidemiol. 63, 1179–1194. 42. Gouttebarge, V., Hopley, P., Kerkhoffs, G., Verhagen, E., Viljoen, W., 28. Centers for Disease Control and Prevention (2016). Alcohol use and Wylleman, P., and Lambert, M.I. (2017). Symptoms of common your health. http://www.cdc.gov/alcohol (last accessed November 22, mental disorders in professional rugby: an international observational 2019). descriptive study. Int. J. Sports Med. 38, 864–870. 29. Bucks, R.S., Olaithe, M., and Eastwood, P. (2013). Neurocognitive function in obstructive sleep apnoea: a meta-review. Respirology 18, 61–70. Address correspondence to: 30. Samson, R.D., and Barnes, C.A. (2013). Impact of aging brain circuits Franziska Plessow, PhD on cognition. Eur. J. Neurosci. 37, 1903–1915. Harvard Medical School 31. Steinfeldt, J.A., and Steinfeldt, M.C. (2012). Profile of masculine 401 Park Drive, Suite 504W norms and help-seeking stigma in college football. Sport Exerc. Per- form. Psychol. 1, 58–71. Boston, MA 02115 32. Airaksinen, E., Larsson, M., and Forsell, Y. (2005). Neuropsycholo- gical functions in anxiety disorders in population-based samples: E-mail: fplessow@mgh.harvard.edu