Mild Cognitive Impairment and Dementia Reported by Former Professional Football Players over 50 yr of Age: An NFL-LONG Study : Medicine & Science in Sports & Exercise

Journal Logo

Advanced Search
EPIDEMIOLOGY

Mild Cognitive Impairment and Dementia Reported by Former Professional Football Players over 50 yr of Age: An NFL-LONG Study

WALTON, SAMUEL R.; BRETT, BENJAMIN L.; CHANDRAN, AVINASH; DEFREESE, J. D.; MANNIX, REBEKAH; ECHEMENDIA, RUBEN J.; MEEHAN, WILLIAM P. III; MCCREA, MICHAEL; GUSKIEWICZ, KEVIN M.; KERR, ZACHARY Y.

Author Information

1Department of Exercise and Sport Science and Center for the Study of Retired Athletes, University of North Carolina at Chapel Hill, Chapel Hill, NC

2Department of Neurosurgery/Neurology, Medical College of Wisconsin, Milwaukee, WI

3Datalys Center for Sports Injury Research and Prevention, Indianapolis, IN

4Division of Emergency Medicine, Boston Children’s Hospital, Boston, MA

5Department of Pediatrics and Emergency Medicine, Harvard Medical School, Boston, MA

6Department of Psychology, University of Missouri–Kansas City, Kansas City, MO

7University Orthopedics Center Concussion Care Clinic, State College, PA

8Sports Medicine Division, Boston Children’s Hospital, Boston, MA

9The Micheli Center for Sports Injury Prevention, Boston, MA

10Department of Pediatrics and Orthopedics, Harvard Medical School, Boston, MA

Address for correspondence: Zachary Y. Kerr, Ph.D., University of North Carolina at Chapel Hill, 209 Fetzer Hall, CB #8700 Chapel Hill, NC 27599; E-mail: [email protected].

Submitted for publication June 2021.

Accepted for publication September 2021.

Medicine & Science in Sports & Exercise 54(3):p 424-431, March 2022. | DOI: 10.1249/MSS.0000000000002802
  • Free

Abstract

Purpose 

This study aimed to estimate prevalence of mild cognitive impairment (MCI) and dementia diagnoses in former National Football League (NFL) players ≥50 yr old and examine the relationships among these diagnoses and an array of predictors of long-term brain health.

Methods 

A cross section of former NFL players (n = 922; mean ± SD age, 64.8 ± 8.9 yr) completed a questionnaire. Prevalence of self-reported medical diagnoses of MCI and dementia was reported alongside U.S. population estimates across 5-yr age intervals (e.g., 60–64 yr). Prevalence ratios (PR) were calculated for multiple predictors of long-term brain health.

Results 

Overall, MCI prevalence and dementia prevalence were n = 219(23.8%) and n = 82(8.9%), respectively. Each diagnosis was more prevalent in former NFL players across age-groups than U.S. norms, with greater disparities at relatively younger ages (e.g., 65–69 yr) compared with older ages. Greater prevalence of MCI and dementia was associated with self-reported concussion history (10+ vs 0; PRadjusted [95% CI] = 1.66 [1.02–2.71] and 2.61 [1.01–6.71], respectively); recent pain intensity (PRadjusted [95% CI] = 1.13 [1.07–1.20] and 1.15 [1.03–1.28]); and diagnoses of depression (PRadjusted [95% CI] = 2.70 [1.92–3.81] and 3.22 [1.69–6.14]), anxiety (PRadjusted [95% CI] = 1.96 [1.26–3.07] and 3.14 [1.47–6.74]), or both (PRadjusted [95% CI] = 3.11 [2.38–4.08] and 4.43 [2.71–7.25]). Higher MCI prevalence was related to sleep apnea (PRadjusted [95% CI] = 1.30 [1.06–1.60]); higher dementia prevalence was associated with age (5-yr interval, PRadjusted [95% CI] = 1.42 [1.26–1.60]) and race (non-White vs White, PRadjusted [95% CI] = 1.64 [1.07–2.53]).

Conclusions 

Self-reported MCI prevalence and dementia prevalence were higher in former NFL players than national estimates and were associated with numerous personal factors, including mood-related disorders and a high number of self-reported concussions. Predictors of higher MCI and dementia prevalence may be modifiable and warrant consideration by clinicians and researchers as potential targets to mitigate the onset of these conditions.

Mild cognitive impairment (MCI) (1,2) affects a substantial proportion of Americans (3), and the cumulative risk of conversion from MCI to dementia-related conditions is estimated around 24%–32% (4). Risk factors for cognitive decline and dementia-related disorders are not fully understood; however, demographic (e.g., age and race), social, educational, health-related (e.g., osteoarthritis and cardiovascular disease), and mood-related (e.g., depression) factors are associated with cognitive deficits and/or dementia in the general population (3,5–13). Traumatic brain injury (TBI) may also increase risk for developing MCI and dementia-related disorders (14–18). With respect to sport-related TBI, the prevalence of MCI may be higher in former National Football League (NFL) players with three or more self-reported concussions compared with those with fewer, but not necessarily for Alzheimer’s disease (AD) (16). Still, the overall prevalence of AD in former NFL players appears higher than that in American males, with a trend for the greatest disparity among those under 70 yr old. Further research is needed to better understand the factors associated with the onset of MCI and AD in this population (16).

To address this need, we surveyed former NFL players’ reports of MCI and dementia-related diagnoses and investigated the relationships among these diagnoses and a broad array of predictors of long-term brain health. We hypothesized the following: 1) diagnoses of MCI and dementia disorders would be higher in this sample of former football players than general population norms, and 2) higher lifetime self-reported concussion history (SR-CHx), as well as a number of other previously reported predictors of interest (3,5–13), would be associated with higher prevalence of MCI and dementia.

METHODS

Institutional approval and participant consent

Former NFL players completed an online (QualtricsXM; SAP America Inc., Newton Square, PA) or paper hardcopy questionnaire. This study was approved by the Institutional Review Board at the University of North Carolina at Chapel Hill. All participants provided written informed consent before participating in the study.

Participants and recruitment

We initially contacted 15,025 former NFL players of all ages via hardcopy mailing and a subset of these same individuals (n = 11,645) via electronic mail (based on availability; Fig. 1) to engage as many participants as possible. Individuals were eligible to participate if they had played at least one full season in the NFL and were 50 yr or older at the time of questionnaire completion. Given that the objective of the current study was to investigate prevalence rates and predictors of MCI and dementia in older adults, we selected 50 yr old as the cutoff age because it represents the lower-bound age in which neurodegenerative diseases with typical earlier onset first occur (e.g., early onset/autosomal AD, behavioral variant of frontotemporal dementia) (19,20) and symptom manifestation before selected cutoff is not common. Those who had complete data for each self-reported diagnosis and explanatory variable of interest were included in these analyses.

F1
FIGURE 1:
CONSORT diagram. Participant recruitment and inclusion.

Data collection

The comprehensive questionnaire used for this study was expanded from a previously administered questionnaire and has been described elsewhere (16,21–23) as part of an ongoing study, Neurologic Function across the Lifespan: A Prospective, Longitudinal, and Translational Study for Former National Football League Players (NFL-LONG). This survey explored the general health of former NFL players and acquired information regarding: personal demographics; football playing history; medical history; concussion history; musculoskeletal injury history; self-reported psychological, physical, and cognitive functioning; health-related quality of life; and current substance use- and health-related behaviors.

Outcome measures

For self-reported medical diagnoses, participants were asked, “Have you ever been told by a physician or health professional that you had/have any of the following conditions?” and conditions included MCI, AD, frontal temporal dementia, Lewy body dementia, vascular dementia, and other dementia. Because of the low prevalence for each of the individual dementia diagnoses in the current sample and the general population (e.g., vascular dementia) (12), all dementia disorders were combined into an “any dementia” diagnosis outcome before analyses.

Self-reported concussion history

Participants were provided with a definition of concussion used in previous research (24): “A concussion typically occurs from a blow to the head and is followed by a variety of symptoms that may include any of the following: headache, dizziness, loss of balance, blurred vision, ‘seeing stars,’ feeling in a fog, or slowed down, memory problems, poor concentration, nausea, or throwing-up. Getting ‘knocked out’ or being unconscious does NOT always occur with a concussion.” Each participant reported the total number of concussions they believed they sustained while playing football at various levels of participation: before high school, during high school, during college, as a professional, and during “other” (e.g., military). Participants were reminded to include both concussions that were and were not medically diagnosed.

Statistical analyses

Prevalence estimates of self-reported MCI and dementia were calculated by dividing the number of participants reporting a diagnosis of each condition by the total number of participants. To compare these estimates to U.S. population estimates, separate prevalences were calculated using 5-yr age bins typically used in dementia literature (3,12,13). Any dementia population-based prevalence estimates were extracted from two epidemiologic studies on the prevalence of multiple types of dementia in the United States (12,13). U.S. normative MCI and dementia prevalence estimates were representative of both males and females as there is little evidence that MCI diagnoses differ by sex (3,25) and overall dementia diagnoses may not be related to sex (12,13).

Potential predictors of the prevalence of MCI and dementia were selected a priori based on previous research (3,5–13,16,17). Lifetime self-reported concussion history (SR-CHx), based on the sum across all levels, was collapsed into five categories: 0, 1–2, 3–5, 6–9, and 10+. These categories are in congruence with previous literature (26) to reduce measurement error associated with continuous variables potentially containing extremely high values. Additional predictors included age (continuous); body mass index (continuous); race/ethnicity (operationalized as identified vs did not identify as White/non-Hispanic); relationship status (in vs not in a relationship); highest level of education attained (less than a bachelor’s degree vs bachelor’s degree vs graduate/postbaccalaureate degree); total years of football participation; self-reported diagnosis of pertinent comorbidities such as cardiovascular morbidities [coronary heart disease, hypertension, and hypercholesterolemia], diabetes, sleep apnea, any cancer, osteoarthritis, depression, and anxiety; musculoskeletal surgery history (0, 1–2, 3–5, 6–9, 10+); and PROMIS®-29 Profile version 2.0 Pain Intensity (27) (scale measure ranging from 0 for “no pain” to 10 for “worst pain” within the previous 7 d).

To calculate crude prevalence ratios (PR) for each predictor, separate Poisson regression models with robust model variance estimations were fit for each outcome diagnosis (MCI, any dementia). PR values with 95% confidence intervals (CI) excluding 1.0 were considered statistically significant. Only predictors that were statistically significant in bivariate (crude PR) analyses with MCI or any dementia were included in the multivariable prediction model for that respective outcome diagnosis. Analyses were performed with SAS version 9.4 (Cary, NC).

RESULTS

Participants

Overall, 922 participants were included in analyses (Fig. 1). The overall response rate for our questionnaire, including participants of all ages, was approximately 12% (n = 1784/15,025); however, it was not possible to ascertain an estimate of the total number of living former NFL players over 50 yr of age. As a result, we are unable to report an accurate response rate. Most participants self-identified as White/non-Hispanic; those not identifying as White/non-Hispanic identified as Black or African American/non-Hispanic, as one or more race that was not “White” or “Black or African American,” and/or as Hispanic ethnicity (Table 1). Most participants were in a relationship with a significant other and had obtained a bachelor’s or higher degree. The most prevalent self-reported medical diagnoses were high blood pressure/hypertension, followed by high cholesterol/hypercholesterolemia, and sleep apnea (Table 1).

TABLE 1 - Sample characteristics (n = 922).
Age, mean (SD), yr 64.8 (8.9)
Body mass index, mean (SD), kg·m−2 30.5 (4.4)
Race/ethnicity, n (%)
 Non-White 275 (29.8)
  Black or African American/non-Hispanic 252 (27.3)
  Reported one or more racial/ethnic identity other than Black or White 23 (2.5)
   Hispanic a 9 (1.0)
 White/non-Hispanic 647 (70.2)
Relationship status, n (%)
 In a relationship 767 (75.5)
  Married 731 (79.3)
  Dating 42 (4.6)
 Not in a relationship 151 (16.4)
  Separated/divorced 99 (10.7)
  Widowed 33 (3.6)
  Single (never married) 19 (2.1)
Highest level of education attained, n (%)
 Less than a bachelor’s degree 139 (15.1)
  High school graduate (diploma or GED) 2 (0.2)
  Some college, but no degree 115 (12.5)
  Associate degree (e.g., AA, AS) 22 (2.4)
 Bachelor’s degree (e.g., BA, BS) 546 (59.2)
 Postgraduate degree (e.g., MA, PhD, JD, MD) 237 (25.7)
Total years of football played, mean (SD) 17.7 (4.7)
Time since leaving the NFL (yr), mean (SD) 35.6 (9.4)
Primary playing position in the NFL, b n (%)
 Lineman 255 (27.7)
 Non-lineman 366 (39.7)
 Missing 301 (32.6)
Typical pain (PROMIS-29), c mean (SD) 4.3 (2.5)
Cognitive function (PROMIS), d mean (SD) 44.8 (10.2)
Lifetime sport-related concussion history, n (%)
 0 108 (11.7)
 1–2 140 (15.2)
 3–5 229 (24.8)
 6–9 194 (21.0)
 10 or more 251 (27.2)
Lifetime musculoskeletal surgery history, n (%)
 0 256 (27.8)
 1–2 231 (25.0)
 3–5 210 (22.8)
 6–9 126 (13.7)
 10 or more 99 (10.7)
Self-reported lifetime medical diagnoses, n (%)
 MCI/memory impairment 219 (23.8)
 Dementia (any) 82 (8.9)
  AD/dementia 41 (4.5)
  Frontal temporal dementia 13 (1.4)
  Lewy body dementia 4 (0.4)
  Vascular dementia 2 (0.2)
  Other dementia 22 (2.4)
 Depression 181 (19.6)
 Anxiety 163 (17.7)
 Cardiovascular disease 568 (61.6)
  Coronary heart disease 131 (14.2)
  High blood pressure/hypertension 448 (48.6)
  Hypercholesterolemia/high cholesterol 312 (33.8)
 Diabetes—type I 15 (1.6)
 Diabetes—type II 75 (8.1)
 Sleep apnea 286 (31.0)
 Cancer 141 (15.3)
 Osteoarthritis/degenerative arthritis 235 (25.5)
aParticipants identifying as Hispanic also reported one or more racial identity and were operationally included within “non-White” category.
bLineman included offensive and defensive line positions as well as tight ends, whereas non-lineman included defensive backs, linebackers, quarterbacks, running backs, wide receivers, kickers, punters, and kick/punt return specialists.
cPain was self-reported on a Likert scale from 0 (“no pain”) to 10 (“worst pain imaginable”).
dMissing data for 19 participants.

Prevalence estimates

Self-reported MCI and dementia diagnoses were more frequent in older age-groups of former football players (12,13). Prevalence for both diagnoses appeared higher in the former football players than the general population estimates in each age-group (Fig. 2).

F2
FIGURE 2:
Prevalence of self-reported MCI and dementia diagnoses by age in former NFL players from the current study compared with national estimates. A, Former NFL player and national estimates by age for MCI. National estimates for MCI diagnoses were based on the latest “Practice Guideline Update Summary” for MCI from the American Academy of Neurology, which included males and females and were based on meta-analysis of data from 34 separate reports (3). Prevalence of MCI in that Summary statement was reported in the following age-groups: 60–64, 65–69, 70–74, 75–79, and 80–84 yr. B, Dementia diagnoses in the former NFL players included Alzheimer’s, frontal temporal, Lewy body, vascular, or “other” dementia diseases. National trends for dementia diagnoses are based on multiple types of dementia in the United States as reported by Plassman et al. (12) and Langa et al. (13) Data from Plassman et al. were reported for males only from the “all dementia” category in the following age-groups: 71–79, 80–89, and ≥90 yr. Data from Langa et al. were reported for males and females combined using data from 2012, as all dementia types, and in the following age-groups: 65–74, 75–84, and ≥85 yr. The age variable used as a predictor in regression analysis for the present study was operationalized into 5-yr age-groups: 50–54 yr (n = 128), 55–59 yr (n = 179), 60–64 yr (n = 166), 65–69 yr (n = 169), 70–74 yr (n = 122), 75–79 yr (n = 107), and ≥80 yr (n = 51). For the purposes of aligning our study sample with dementia diagnoses reported in the studies referenced, we have also included the age-groups ≥85 yr (n = 12) and ≥90 yr (n = 3) in part B of this figure. There were very few participants from our study in each of the ≥85- and ≥90-yr age-groups—necessitating caution when interpreting these prevalence estimates.

PR

Many predictors yielded significant crude PR, leading to their inclusion in the multivariable (adjusted PR) models (Table 2). Lifetime SR-CHx of 10+ (compared with none) was associated with higher prevalence of self-reported MCI and dementia (Fig. 3). Depression, anxiety, and pain were each associated with the prevalence of MCI and dementia (Table 2). Sleep apnea was associated with a higher prevalence of MCI, whereas musculoskeletal surgery history (1 to 2 surgeries vs no surgeries) was associated with a lower prevalence of MCI. Finally, older age and race/ethnicity (not identifying vs identifying as White/non-Hispanic) were each associated with higher prevalence of any dementia (Table 2).

TABLE 2 - PR of MCI and dementia by self-reported concussion history and other health-related predictors.
Variable Value MCI Diagnosis Any Dementia Diagnosis
n (%) Crude PR (95% CI) Adjusted PR (95% CI) n (%) Crude PR (95% CI) Adjusted PR (95% CI)
Self-reported concussion history 0 14/108 (13.0) 1.0 1.0 5/108 (4.6) 1.0 1.0
1 to 2 21/140 (15.0) 1.16 (0.62, 2.17) 1.29 (0.73, 2.28) 5/140 (3.6) 0.77 (0.23, 2.60) 0.86 (0.26, 2.82)
3 to 5 38/229 (16.6) 1.28 (0.73, 2.26) 1.16 (0.70, 1.92) 14/229 (6.1) 1.32 (0.49, 3.57) 1.24 (0.47, 3.25)
6 to 9 50/194 (25.8) 1.99 (1.15, 3.43) a 1.55 (0.95, 2.53) 18/194 (9.3) 2.00 (0.77, 5.25) 1.80 (0.70, 4.61)
10 or more 96/251 (38.2) 2.95 (1.77, 4.93) a 1.66 (1.02, 2.71) a 40/251 (15.9) 3.44 (1.40, 8.48) a 2.61 (1.01, 6.71) a
Age (yr) 5-point increase N/A 1.06 (0.99, 1.13) – N/A 1.23 (1.09, 1.37) a 1.42 (1.26, 1.60) a
Body Mass Index (kg·m−2) 5-point increase N/A 1.12 (0.99, 1.27) – N/A 1.05 (0.84, 1.32) –
Race/Ethnicity White/non-Hispanic 137/647 (21.2) 1.0 1.0 49/647 (7.6) 1.0 1.0
Non-White 82/275 (29.8) 1.41 (1.11, 1.78) a 1.02 (0.82, 1.28) 33/275 (12.0) 1.58 (1.04, 2.41) a 1.64 (1.07, 2.53) a
Relationship status In a relationship 183/771 (23.7) 1.0 – 70/771 (9.1) 1.0 –
Not in a relationship 36/151 (23.8) 1.00 (0.73, 1.36) – 12/151 (7.9) 1.14 (0.63, 2.05) –
Education Less than bachelor’s degree 48/139 (34.5) 1.0 1.0 20/139 (14.4) 1.0 1.0
Bachelor’s degree 126/545 (23.1) 0.67 (0.51, 0.88) a 0.91 (0.70, 1.19) 44/545 (8.1) 0.56 (0.34, 0.92) a 0.68 (0.42, 1.11)
More than bachelor’s degree 45/239 (19.0) 0.55 (0.38, 0.77) a 0.94 (0.68, 1.31) 18/239 (7.6) 0.42 (0.29, 0.96) a 0.85 (0.47, 1.55)
Total years of football played 5-yr increase N/A 1.07 (0.95, 1.20) – N/A 1.04 (0.87, 1.25) –
Musculoskeletal surgical history 0 61/256 (23.8) 1.0 1.0 26/256 (10.2) 1.0 –
1 to 2 40/231 (17.3) 0.73 (0.51, 1.04) 0.67 (0.49, 0.93) a 14/231 (6.1) 0.60 (0.32, 1.11) –
3 to 5 42/210 (20) 0.84 (0.59, 1.19) 0.84 (0.60, 1.16) 13/210 (6.2) 0.61 (0.32, 1.16) –
6 to 9 39/126 (31) 1.30 (0.92, 1.83) 1.02 (0.75, 1.38) 14/126 (11.1) 1.09 (0.59, 2.02) –
10 or more 37/99 (37.4) 1.57 (1.12, 2.20) a 0.88 (0.64, 1.22) 15/99 (15.2) 1.49 (0.83, 2.70) –
Any cardiovascular disease No 67/354 (18.9) 1.0 1.0 31/354 (8.8) 1.0 –
Yes 152/568 (26.8) 1.41 (1.10, 1.82) a 1.15 (0.91, 1.44) 51/568 (9.0) 1.03 (0.67, 1.57) –
Diabetes—Type I No 215/907 (23.7) 1.0 – 80/907 (8.8) 1.0 –
Yes 4/15 (26.7) 1.12 (0.48, 2.62) – 2/15 (13.3) 1.51 (0.41, 5.59) –
Diabetes—Type II No 199/847 (23.5) 1.0 – 72/847 (8.5) 1.0 –
Yes 20/75 (26.7) 1.13 (0.76, 1.68) – 10/75 (13.3) 1.57 (0.85, 2.91) –
Sleep apnea No 120/636 (18.9) 1.0 1.0 50/636 (7.9) 1.0 –
Yes 99/286 (34.6) 1.83 (1.46, 2.30) a 1.30 (1.06, 1.60) a 32/286 (11.2) 1.42 (0.93, 2.17) –
Cancer No 187/781 (23.9) 1.0 – 70/781 (9.0) 1.0 –
Yes 32/141 (22.7) 0.95 (0.68, 1.32) – 12/141 (8.5) 0.95 (0.53, 1.71) –
Osteoarthritis/degenerative arthritis No 135/687 (19.7) 1.0 1.0 48/687 (7.0) 1.0 1.0
Yes 84/235 (35.7) 1.82 (1.45, 2.29) a 1.14 (0.91, 1.42) 34/235 (14.5) 2.07 (1.37, 3.13) a 1.02 (0.68, 1.54)
Depression and anxiety Neither 92/697 (13.2) 1.0 1.0 28/697 (4.0) 1.0 1.0
Only depression 31/62 (50.0) 3.79 (2.77, 5.18) a 2.70 (1.92, 3.81) a 13/62 (21.0) 5.22 (2.85, 9.55) a 3.22 (1.69, 6.14) a
Only anxiety 15/44 (34.1) 2.58 (1.64, 4.06) a 1.96 (1.26, 3.07) a 6/44 (13.6) 3.39 (1.48, 7.76) a 3.14 (1.47, 6.74) a
Both 81/119 (68.1) 5.16 (4,11, 6.47) a 3.11 (2.38, 4.08) a 35/119 (29.4) 7.32 (4.63, 11.57) a 4.43 (2.71, 7.25) a
PROMIS pain intensity 1-point increase in scale N/A 1.28 (1.22, 1.34) a 1.13 (1.07, 1.20) a N/A 1.35 (1.23, 1.47) a 1.15 (1.03, 1.28) a
aStatistically significant PR (CI does not include 1.0).
−, predictor was not included in the multivariable model for this outcome diagnosis; N/A, not applicable.

F3
FIGURE 3:
Self-reported MCI and dementia diagnoses in relation to self-reported concussion history. Dementia diagnoses included Alzheimer’s, frontal temporal, Lewy body, vascular, or “other” dementia diseases. Proportions of self-reported MCI diagnoses for each concussion history group were as follows: 0 concussions (n = 14/108); 1–2 concussions (n = 21/140); 3–5 concussions (n = 38/229); 6–9 concussions (n = 50/194); and 10+ concussions (n = 96/251). Proportions of self-reported dementia diagnoses for each concussion history group were as follows: 0 concussions (n = 5/108); 1–2 concussions (n = 5/140); 3–5 concussions (n = 14/229); 6–9 concussions (n = 18/194); and 10+ concussions (n = 40/251).

DISCUSSION

In this study of former NFL players, the greatest disparities in the prevalence of self-reported MCI and dementia diagnoses between former players and general population occurred at relatively younger ages (e.g., 60–64 yr old). These trends could be associated with earlier onset of cognitive dysfunction in former football players (16), heightened awareness and recognition of brain health concerns by former players and healthcare providers (28), or some combination thereof. When considered alongside numerous predictors of MCI and dementia diagnoses, SR-CHx was significantly associated with both diagnoses—but only when the number of lifetime concussions was 10 or more. Other predictors of MCI and dementia were depression, anxiety, sleep apnea, and pain—each of which may be potentially modifiable with targeted treatments or therapies.

Prevalence estimates of MCI and dementia

Diagnosis of MCI was reported by almost one in four participants in this study. The difference in prevalence estimates between current study participants and national trends decreased as age increased. Similar patterns were observed for dementia diagnoses (12,13). The proportional prevalence of dementia diagnoses in the present study (Table 1) approximated expected values based on national lifetime prevalence (12). Like MCI, the greatest disparity between estimates of any dementia prevalence in the current sample compared with national trends was observed at the lowest age-group for which national trend data were available (65 to 69 yr old) (12,13). It should be noted that the present sample is a subset of the larger former NFL player population, and it may not fully represent all those individuals. Still, these data impress the need to develop comprehensive epidemiological and neurobiological studies of cognitive decline in former NFL players alongside factors that may influence such decline, when present.

In a previous study, former NFL players under 70 yr of age reported a higher prevalence of AD compared with U.S. norms, but this was not observed in participants older than 70 yr (16). Greater disparities in prevalence estimates between former football players and the U.S. population at younger ages may be related to repetitive head trauma from playing football, leading to earlier clinical manifestations of cognitive decline (e.g., multiple lifetime concussions possibly contributing to diminishing cerebral reserve), as has been suggested previously (29). Additionally, given recent interest by researchers and popular media on the potential long-term neurologic sequelae of playing football (15,30–33), former players may be more actively seeking care from medical providers at younger ages than their nonathlete counterparts. Studies of objectively measured brain health are needed to understand the potential pathophysiologic relationships between football exposure and the development of neurodegenerative conditions.

Predictors of self-reported MCI and dementia prevalence

Self-reported depression and/or anxiety had the strongest association (adjusted PR values farthest from the null) with MCI and dementia in this study. The association between cognitive difficulties and clinical psychiatric diagnoses has been documented in previous studies of former football players (34,35). In the general population, individuals with a lifetime depression diagnosis (made at any age) had 1.7 to 2.9 times the risk for developing dementia-related diagnoses compared with those without depression diagnosis (6). As age of depression and anxiety diagnoses are unknown in the current study, we cannot determine whether depression and anxiety preceded MCI or dementia (as a risk factor or prodrome), or if these conditions coexist (e.g., depression is a common symptom of various dementias) (36). Future research with former football players should examine the onset and chronicity of objectively measured psychological and cognitive conditions and the relationships among these disorders.

A self-reported lifetime history of 10 or more concussions (compared with no concussions) was significantly associated with higher prevalence of both MCI and any dementia diagnoses. This finding suggests that the potential for long-term behavioral or physiological responses resulting from repetitive head trauma in athletes does not appear to be linear based on the present data, and long-term issues associated with concussion history may occur after a greater number of self-reported concussions than has previously been reported (16,37–41). Importantly, multiple studies have also observed no meaningful effect of concussion history on behavioral outcomes using those same cutoffs in current and former athletes (37,38,42,43). Our study expands on many previous studies by using a more granular approach to lifetime concussion history categorization (e.g., multiple ordinal categories among those with lifetime concussion history greater than three [3–5; 6–9; 10 or more]), and future work should seek to understand how the operationalization of concussion history might affect the interpretation of study outcomes. SR-CHx has been reported to be moderately stable over time (Cohen’s weighted kappa = 0.48) and may be influenced by the presence of medical conditions (e.g., MCI), time, and media attention—an association that is worth considering given the current findings (44). At present, there is no widely accepted threshold for the number of sport-related concussions that one might sustain before observing the functional and biological changes associated with early cognitive decline or dementia, and some individuals may never experience long-term issues as the result of their previous concussive injuries. Therefore, patients should be assessed and treated according to their own history and needs.

Higher recent pain intensity was also associated with increased reporting of MCI and dementia, although the chronicity of this pain was not measured in this study. Whitlock et al. (10) have described greater declines in memory and slightly higher probabilities of dementia in those with chronic moderate or severe pain. The reason for the significant associations with pain in the current study is unclear given that neither musculoskeletal surgery history nor osteoarthritis was associated with increased prevalence of either condition in the multivariable models. One theory is that pain was associated with ongoing medical conditions that caused systemic inflammation, a purported risk factor of MCI and dementia—but this was not explicitly measured (45,46). Alternatively, ongoing pain may have been a source of psychological distress, which has been associated with memory complaints that lead to increased risks of MCI and dementia (47,48). Depressive symptoms and chronic pain have also been associated with lower hippocampal volumes, further adding to the enigmatic relationship among these factors and cognitive dysfunction (49). Regardless of the role of pain in the development of MCI and dementia, it appears to be an important element to measure, given that it is potentially modifiable and has been associated with cognitive dysfunction and decline.

Sleep apnea was significantly associated with MCI. Sleep apnea has been reported to advance cognitive decline in older individuals as has been related to earlier onsets of MCI and Alzheimer’s dementia, whereas treatment with continuous positive airway pressure was associated with delayed age at onset of MCI (9). Older age at the time of participation in the current study and the binary race/ethnicity status were each associated with dementia. Age and race have well-established relationships with dementia (5,8,12,13,50). Our dichotomized race/ethnicity predictor was selected because of the distribution of racial and ethnic identifiers in our current sample. Most non-White participants identified as Black or African American/non-Hispanic, although some identified as a separate race or as multiple races, and a few reported being of Hispanic ethnicity (Table 1). We observed that 12% of those identifying as non-White reported dementia diagnoses versus 7.6% of those identifying as White (Table 2), despite non-White participants being 5.5 yr younger than White participants (60.9 vs 66.4 yr of age). Research suggests slightly higher rates of AD and related dementias in African Americans compared with non-Hispanic Whites in the United States (8); however, we did not have sufficient sample size to test the interaction between age and racial identity in the present study. Owing to the self-reported and cross-sectional nature of each variable in this study, the relationships observed in our sample should be interpreted with caution, and future studies of the incidence and risk of MCI and dementia in former football players should further assess such potential associations.

Limitations

Data were limited to a relatively small subset of former professional football players (922 out of thousands of former players over 50 yr of age), and therefore, the results of this study may not be generalizable to nonrespondents, the entire former football player population, or other populations. This study also contains bias inherent to self-reported survey data, including recall bias and social desirability bias. Particularly, self-reported concussion history may be subject to recall bias or may have been influenced by the presence of impairments or medical diagnoses—real or perceived (44,51). Given its cross-sectional design, the date of onset and the chronicity of medical diagnoses and pain, as well as the receipt of treatment for these issues, were not included in the analyses. Additionally, some participants may have reported the presence of MCI or dementia despite not having been medically diagnosed. These aspects are required to better understand the time course and relationships among MCI, dementia, and medical comorbidities. Finally, as with any estimations of effect estimates, model selection and model fit may yield imprecise results and caution should be taken. In this study, log-binomial models were also considered for estimating PR, and Poisson models were ultimately used because of greater computational efficiency (e.g., cases of failed model convergence) (52).

CONCLUSIONS

This study found that self-reported MCI and dementia-related diagnoses were higher in former NFL players than general U.S. population estimates among individuals 50 yr and older. In this subset of former NFL players, history of 10 or more sport-related concussions, lifetime diagnoses of depression and/or anxiety, and greater recent pain intensity were each associated with higher prevalence of MCI and dementia. Sleep apnea was also associated with a greater prevalence of MCI. Older age and self-identifying as non-White were associated with a greater prevalence of dementia. Potentially modifiable (e.g., depressive symptoms) and nonmodifiable (e.g., age) factors warrant further investigation as risk factors for MCI and dementia through longitudinal studies of former football players. This study suggests that there may be preventative and therapeutic targets that might mitigate the onset of MCI or dementia-related disorders.

The authors are grateful to Candice Goerger, Hope Campbell, Caprice Hunt, and Greggory Kobelski for their invaluable efforts in the administration, coordination, and execution of the NFL-LONG study. Special thanks are owed to the members of the NFL Alumni Association and Legends community for their assistance in recruitment of study participants. The authors also thank each of the clinicians, researchers, educators, and former players who helped them adapt and develop the current general health survey. The parent study from which these data were drawn (NFL-LONG) was funded by the NFL, but the sponsor provided no input into the present study, data analyses, or preparation of this manuscript.

B. L. B. acknowledges support from the National Institute of Neurological Disorders and Stroke (award number 2L30 NS113158-02) and the National Institute on Aging (award number K23AG073528-01) under the National Institutes of Health. A. C. receives funding from the National Collegiate Athletics Association as Director of the NCAA Injury Surveillance Program. R. J. E. is a paid consultant for the NHL and cochair of the NHL/NHLPA Concussion Subcommittee. He is also a paid consultant for Major League Soccer, the U. S. Soccer Federation, and Princeton University Athletic Medicine, and he occasionally provides expert testimony in matters related to mTBI and sports concussion. M. M. reports a grant from NFL during the conduct of the study as well as grants from NIH, CDC, DoD, and NCAA outside the submitted work. W. P. M. receives royalties from 1) ABC-Clio publishing for the sale of his books, Kids, Sports, and Concussion: A guide for coaches and parents and Concussions; 2) Springer International for the book Head and Neck Injuries in Young Athlete; and 3) Wolters Kluwer for working as an author for UpToDate. His research is funded, in part, by philanthropic support from the National Hockey League Alumni Association through the Corey C. Griffin Pro-Am Tournament and a grant from the NFL. Z. Y. K. reports grants from CDC, DoD, and NIH.

The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of the present study do not constitute endorsement by the American College of Sports Medicine.

REFERENCES

1. Winblad B, Palmer K, Kivipelto M, et al. Mild cognitive impairment—beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med. 2004;256(3):240–6.
2. Portet F, Ousset PJ, Visser PJ, et al; MCI Working Group of the European Consortium on Alzheimer’s Disease (EADC). Mild cognitive impairment (MCI) in medical practice: a critical review of the concept and new diagnostic procedure. Report of the MCI Working Group of the European Consortium on Alzheimer’s disease. J Neurol Neurosurg Psychiatry. 2006;77(6):714–8.
3. Petersen RC, Lopez O, Armstrong MJ, et al. Practice guideline update summary: mild cognitive impairment: report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology. Neurology. 2018;90(3):126–35.
4. Mitchell AJ, Shiri-Feshki M. Rate of progression of mild cognitive impairment to dementia—meta-analysis of 41 robust inception cohort studies. Acta Psychiatr Scand. 2009;119(4):252–65.
5. Bachman DL, Wolf PA, Linn RT, et al. Incidence of dementia and probable Alzheimer’s disease in a general population: the Framingham Study. Neurology. 1993;43(3, Pt 1):515–5.
6. Bellou V, Belbasis L, Tzoulaki I, Middleton LT, Ioannidis JPA, Evangelou E. Systematic evaluation of the associations between environmental risk factors and dementia: an umbrella review of systematic reviews and meta-analyses. Alzheimers Dement. 2017;13(4):406–18.
7. Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer’s disease: an analysis of population-based data. Lancet Neurol. 2014;13(8):788–94.
8. Matthews KA, Xu W, Gaglioti AH, et al. Racial and ethnic estimates of Alzheimer’s disease and related dementias in the United States (2015–2060) in adults age ≥65 years. Alzheimers Dement. 2019;15(1):17–24.
9. Osorio RS, Gumb T, Pirraglia E, et al. Sleep-disordered breathing advances cognitive decline in the elderly. Neurology. 2015;84(19):1964–71.
10. Whitlock EL, Diaz-Ramirez LG, Glymour MM, Boscardin WJ, Covinsky KE, Smith AK. Association between persistent pain and memory decline and dementia in a longitudinal cohort of elders. JAMA Intern Med. 2017;177(8):1146–53.
11. Weber A, Mak SH, Berenbaum F, et al. Association between osteoarthritis and increased risk of dementia: a systemic review and meta-analysis. Medicine. 2019;98(10):e14355.
12. Plassman BL, Langa KM, Fisher GG, et al. Prevalence of dementia in the United States: the aging, demographics, and memory study. Neuroepidemiology. 2007;29(1–2):125–32.
13. Langa KM, Larson EB, Crimmins EM, et al. A comparison of the prevalence of dementia in the United States in 2000 and 2012. JAMA Intern Med. 2017;177(1):51–8.
14. Lehman EJ, Hein MJ, Baron SL, Gersic CM. Neurodegenerative causes of death among retired National Football League players. Neurology. 2012;79(19):1970–4.
15. Nguyen VT, Zafonte RD, Chen JT, et al. Mortality among professional American-style football players and professional American baseball players. JAMA Netw Open. 2019;2(5):e194223.
16. Guskiewicz KM, Marshall SW, Bailes J, et al. Association between recurrent concussion and late-life cognitive impairment in retired professional football players. Neurosurgery. 2005;57:719–26.
17. Plassman BL, Havlik RJ, Steffens DC, et al. Documented head injury in early adulthood and risk of Alzheimer’s disease and other dementias. Neurology. 2000;55(8):1158–66.
18. Perry DC, Sturm VE, Peterson MJ, et al. Association of traumatic brain injury with subsequent neurological and psychiatric disease: a meta-analysis. J Neurosurg. 2016;124(2):511–26.
19. Spina S, La Joie R, Petersen C, et al. Comorbid neuropathological diagnoses in early versus late-onset Alzheimer’s disease. Brain. 2021;144(7):2186–98.
20. Bateman RJ, Aisen PS, De Strooper B, et al. Autosomal-dominant Alzheimer’s disease: a review and proposal for the prevention of Alzheimer’s disease. Alzheimers Res Ther. 2011;3(1):1.
21. Guskiewicz KM, Marshall SW, Bailes J, et al. Recurrent concussion and risk of depression in retired professional football players. Med Sci Sports Exerc. 2007;39(6):903–9.
22. Walton SR, Kerr ZY, Brett BL, et al. Health-promoting behaviours and concussion history are associated with cognitive function, mood-related symptoms and emotional–behavioural dyscontrol in former NFL players: an NFL-LONG Study. Br J Sports Med. 2021;55(12):683–90.
23. Brett BL, Walton SR, Kerr ZY, et al. Distinct latent profiles based on neurobehavioural, physical and psychosocial functioning of former National Football League (NFL) players: an NFL-LONG Study. J Neurol Neurosurg Psychiatry. 2021;92(3):282–90.
24. McCrea M, Hammeke T, Olsen G, Leo P, Guskiewicz K. Unreported concussion in high school football players: implications for prevention. Clin J Sport Med. 2004;14(1):13–7.
25. Au B, Dale-McGrath S, Tierney MC. Sex differences in the prevalence and incidence of mild cognitive impairment: a meta-analysis. Ageing Res Rev. 2017;35:176–99.
26. Kerr ZY, Marshall SW, Harding HP Jr, Guskiewicz KM. Nine-year risk of depression diagnosis increases with increasing self-reported concussions in retired professional football players. Am J Sports Med. 2012;40(10):2206–12.
27. Hays RD, Spritzer KL, Schalet BD, Cella D. PROMIS®-29 v2.0 profile physical and mental health summary scores. Qual Life Res. 2018;27(7):1885–91.
28. Stewart W, Allinson K, Al-Sarraj S, et al. Primum non nocere: a call for balance when reporting on CTE. Lancet Neurol. 2019;18(3):231–3.
29. Randolph C, Karantzoulis S, Guskiewicz K. Prevalence and characterization of mild cognitive impairment in retired national football league players. J Int Neuropsychol Soc. 2013;19(8):873–80.
30. Wolfson DI, Kuhn AW, Kerr ZY, et al. Chronic traumatic encephalopathy research viewed in the public domain: what makes headlines? Brain Inj. 2020;34(4):528–34.
31. Kuhn AW, Yengo-Kahn AM, Kerr ZY, Zuckerman SL. Sports concussion research, chronic traumatic encephalopathy and the media: repairing the disconnect. Br J Sports Med. 2017;51(24):1732–3.
32. Carone DA. The Role and Ramifications of the Media’s Reporting on Concussions. Oxford University Press; 2014.
33. McKee AC, Stein TD, Nowinski CJ, et al. The spectrum of disease in chronic traumatic encephalopathy. Brain. 2013;136(1):43–64.
34. Plessow F, Pascual-Leone A, McCracken CM, et al. Self-reported cognitive function and mental health diagnoses among former professional American-style football players. J Neurotrauma. 2020;37(8):1021–8.
35. Baker JG, Leddy JJ, Hinds AL, et al. An exploratory study of mild cognitive impairment of retired professional contact sport athletes. J Head Trauma Rehabil. 2018;33(5):E16–E23.
36. Bennett S, Thomas AJ. Depression and dementia: cause, consequence or coincidence? Maturitas. 2014;79(2):184–90.
37. Clark MD, Varangis EML, Champagne AA, et al. Effects of career duration, concussion history, and playing position on white matter microstructure and functional neural recruitment in former college and professional football athletes. Radiology. 2018;286(3):967–77.
38. Ford JH, Giovanello KS, Guskiewicz KM. Episodic memory in former professional football players with a history of concussion: an event-related functional neuroimaging study. J Neurotrauma. 2013;30(20):1683–701.
39. Kerr ZY, Thomas LC, Simon JE, McCrea M, Guskiewicz KM. Association between history of multiple concussions and health outcomes among former college football players: 15-year follow-up from the NCAA Concussion Study (1999–2001). Am J Sports Med. 2018;46(7):1733–41.
40. Collins MW, Grindel SH, Lovell MR, et al. Relationship between concussion and neuropsychological performance in college football players. JAMA. 1999;282(10):964–70.
41. Iverson GL, Silverberg ND, Mannix R, et al. Factors associated with concussion-like symptom reporting in high school athletes. JAMA Pediatr. 2015;169(12):1132–40.
42. Teel EF, Marshall SW, Shankar V, McCrea M, Guskiewicz KM. Predicting recovery patterns after sport-related concussion. J Athl Train. 2017;52(3):288–98.
43. Rosenblum DJ, Walton SR, Erdman NK, Broshek DK, Hart JM, Resch J. If not now, when? An absence of neurocognitive and postural stability deficits in collegiate athletes with one or more concussions. J Neurotrauma. 2020;37(10):1211–20.
44. Kerr ZY, Marshall SW, Guskiewicz KM. Reliability of concussion history in former professional football players. Med Sci Sports Exerc. 2012;44(3):377–82.
45. Shen X-N, Niu L-D, Wang Y-J, et al. Inflammatory markers in Alzheimer’s disease and mild cognitive impairment: a meta-analysis and systematic review of 170 studies. J Neurol Neurosurg Psychiatry. 2019;90(5):590–8.
46. Shen X-N, Lu Y, Tan CTY, et al. Identification of inflammatory and vascular markers associated with mild cognitive impairment. Aging. 2019;11(8):2403–19.
47. Mitchell AJ, Beaumont H, Ferguson D, Yadegarfar M, Stubbs B. Risk of dementia and mild cognitive impairment in older people with subjective memory complaints: meta-analysis. Acta Psychiatr Scand. 2014;130(6):439–51.
48. Mol M, Carpay M, Ramakers I, Rozendaal N, Verhey F, Jolles J. The effect of perceived forgetfulness on quality of life in older adults; a qualitative review. Int J Geriatr Psychiatry. 2007;22(5):393–400.
49. Ezzati A, Zammit AR, Lipton ML, Lipton RB. The relationship between hippocampal volume, chronic pain, and depressive symptoms in older adults. Psychiatry Res Neuroimaging. 2019;289:10–2.
50. Yaffe K, Falvey C, Harris TB, et al. Effect of socioeconomic disparities on incidence of dementia among biracial older adults: prospective study. BMJ. 2013;347:f7051.
51. Scheier LM. The spurious chicken and the confounding egg: commentary on cumulative head impact exposure and youth football (re: doi: 10.1089/neu.2016.4413). J Neurotrauma. 2019;36(2):408–10.
52. Petersen MR, Deddens JA. A comparison of two methods for estimating prevalence ratios. BMC Med Res Methodol. 2008;8(1):9.
Keywords:

FORMER ATHLETES; COGNITIVE DECLINE; DISEASE PREVALENCE; CONCUSSION HISTORY; MOOD-RELATED DISORDERS; MODIFIABLE RISK FACTORS

Copyright © 2021 by the American College of Sports Medicine