The Longitudinal Relationship Between Concussion History, Years of American Football Participation, and Alcohol Use Among Former National Football League Players: an NFL-LONG Study

Abstract

Objective

Investigate the relationships between concussion history and years of football participation (repetitive head impact proxy) with alcohol use across multiple decades in former professional football players.

Methods

Participants (n = 348; mean age = 49.0 ± 9.4) completed health questionnaires in 2001 and 2019, which included self-reported concussion history and years of participation. Alcohol use frequency and amount per occasion were reported for three timepoints: during professional career, 2001, and 2019. Ordinal logistic regression models were fit to test associations of concussion history and years of participation with alcohol use at each timepoint.

Results

There were no significant associations between either concussion history or years of football participation with alcohol use (frequency and amount per occasion) at any timepoint. Effect estimates for concussion history and years of football participation with alcohol use were generally comparable across timepoints.

Conclusions

Later life alcohol use by former American football players is not associated with concussion history or years of exposure to football.

Introduction

Traumatic brain injury (TBI) and alcohol use are known to be highly correlated and are bidirectionally linked (Weil et al., 2018). That is, TBI, particularly during adolescence, can moderate or serve as a risk factor for problematic alcohol use. In one study, for example, ~20% of individuals identifying as abstainers or “light” drinkers prior to a TBI exhibited high-volume alcohol use post-injury (Corrigan et al., 1995). Conversely, problematic alcohol use is also a predictor of TBI, as one study has shown that approximately 45% of patients presenting to a level I trauma center for head injury met cutoff scores for problematic drinking prior to admission (Dikmen et al., 1995).

TBI and alcohol use are also both conjointly and independently associated with worse neurobehavioral functioning (Conner et al., 2009; Koch et al., 2019; Koponen et al., 2002; Ponsford et al., 2013; Schneider et al., 2022). Among former contact sport athletes who are at greater risk than the general public for sustaining sport-related concussion (SRC) and exposure to repetitive head impacts (RHI), alcohol use has been hypothesized to confound the relationships between head injury history (SRC or RHI) and poor neurobehavioral outcomes (Manley et al., 2017). Multiple studies have reported no significant association between head injury and increased alcohol use patterns among former athletes (Hind et al., 2022; Russell et al., 2020). However, all examinations, to date, have been cross-sectional or retrospective in nature. Investigating these associations longitudinally is necessary to understand the complex relationship between head injury history as a risk factor for the development of problematic alcohol use, which involves the potential to increase the risk of adverse long-term functional outcomes in contact sport athletes.

As such, the aim of this study was to examine the cross-sectional and longitudinal relationships between concussion history and years of American football participation with alcohol use in a sample of former professional football players across multiple decades.

Materials and Methods

The current study was part of an ongoing project entitled, Neurologic Function across the Lifespan: A Prospective, LONGitudinal, and Translational Study for Former National Football League Players (NFL-LONG). The study was approved by an institutional review board, and all participants provided informed consent.

The NFL-LONG study procedures are described in greater detail elsewhere (Brett et al., 2022). Briefly, data were collected at three timepoints (2001, 2010, and 2019); only data collected in 2001 and 2019 were used for the current study due to selective sampling in 2010. At the 2001 timepoint, players were asked about health and functional status in 2001 as well as during the players’ professional career. As such, the three timepoints referenced in the current study include during professional career (i.e., while playing football professionally), 2001, and 2019.

In 2001, a general health survey (GHS) was sent to all living members of the NFL Retired Players Association (n = 3,729), with 2,536 (68.7%) completing the GHS at that time. In 2019, a revised version of the GHS was sent out to an estimated 15,025 former NFL players via hardcopy and electronic mail (based on availability of contact information) as part of the ongoing NFL-LONG study. Overall, 1,784 (11.9%) completed and returned the 2019 GHS at the time these analyses were performed (GHS involves open enrollment). Among those completing the 2019 GHS, n = 348 (19.5%) had also completed the 2001 GHS. Inclusion criteria for the study required at least 2 years of participation in the NFL and completing the GHS in 2001 and 2019.

Participants provided demographic information, concussion history, and total years of lifetime football participation (a commonly used proxy for RHIs), and they completed questionnaires about various aspects of functioning as part of the GHS. The standard definition of concussion presented included: “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’(McCrea et al., 2004). Concussion history was recorded in 2001 and included total concussions sustained during high school, collegiate, and professional football experiences. Concussion history was also recorded in 2019 and included injuries at all levels of play and non-sport injuries. Total concussion history was binned into five groups: 0 (n = 74; 21.3%), 1–2 (83; 23.9%), 3–5 (105; 30.2%), 6–9 (43 (12.4%), and 10+ (43; 12.4%; Brett et al., 2022). To reduce recall bias, concussion history recorded in 2001 was used; however, sensitivity analyses indicated that results did not meaningfully change regardless of which timepoint was used.

Alcohol use frequency and amount per occasion were reported at all timepoints, but the scale of measurement varied slightly from 2001 to 2019, preventing direct measurement of change in alcohol use across time. On the 2001 GHS, players reported alcohol use frequency and amount of alcohol use per occasion both currently and during their professional play via ordinal ratings of days/week and drinks/occasion, respectively (Table 1). For the 2019 GHS, alcohol use frequency and amount of alcohol use/occasion were also recorded. The C-cutting down, A-annoyance by criticism, G-guilty feeling, and E-eye-openers (CAGE) is a screening measure that contains items reflective of problematic alcohol use (Ewing, 1984). Players also completed the CAGE for the during professional play and 2001 timepoints; summed scores were recorded and characterized as an ordinal outcome for analyses (Table 1). The de-identified data that support the reported findings are available from the corresponding author, upon reasonable request.

Table 1.

Alcohol use characteristics across timepoints

	Timepoint 1 (during pro-career)	Timepoint 2 (2001)	Timepoint 3 (2019)
Alcohol use history	M ± SD/N (%)	M ± SD/N (%)	M ± SD/N (%)
Frequency of use
None	66 (19.0%)	91 (26.2%)
1–2 days per week	141 (40.5%)	131 (37.6%)
3–4 days per week	84 (24.1%)	77 (22.1%)
5–7 days per week	57 (16.4%)	49 (14.1%)
Never			64 (18.4%)
Monthly or less			51 (14.7%)
2–4 times per month			77 (22.1%)
2–3 times per week			71 (20.4%)
4 or more times per week			85 (24.4%)
Drinks consumed per occasion
None	67 (19.3%)	92 (26.4%)	117 (33.6%)
1–2 drinks per occasion	82 (23.6%)	146 (42.1%)
3–5 drinks per occasion	121 (34.7%)	84 (24.1%)
6–7 drinks per occasion	53 (15.2%)	22 (6.3%)
8+ drinks per occasion	25 (7.2%)	4 (1.1%)
1 or 2			184 (52.9%)
3 or 4			41 (11.8%)
5 or 6			5 (1.4%)
7 or 9			1 (0.3%)
≥10			0 (0%)
CAGE score
0	251 (72.2%)	217 (62.5%)
1	44 (12.6%)	54 (15.5%)
2	33 (9.5%)	52 (14.9%)
3	14 (4.0%)	21 (6.0%)
4	6 (1.7%)	4 (1.1%)

Statistical Analysis

The full sample was maintained through multiple imputation using the Markov chain Monte Carlo method (five additional iterations of complete datasets generated). The subsequent analyses and reported results were based on pooled data from the multiple imputation datasets. Ordinal logistic regression models were fit to test the independent associations of concussion history and total years of football participation with alcohol use (frequency and amount per occasion) at each of the three timepoints. Unstandardized beta estimates and 95% confidence intervals (CIs) were calculated for the tested associations. Due to the inconsistencies in methods of recording alcohol use at between 2001 and 2019, measurement of change in alcohol intake across time could not be analyzed directly. Statistical analyses were performed with SPSS (version 28.0). Statistical significance was evaluated at α = 0.05 level.

Results

A total of 348 participants (mean age = 49.00 ± 9.38) with 15.9 ± 4.1 (median = 18.0; range = 23) total years of football participation met the study criteria. Eighty-one (23.3%) former players identified as non-White and 265 (76.1%) identified as White/non-Hispanic. Race was not reported by two participants. There were no significant associations between concussion history and alcohol use frequency/amount per occasion (Fig. 1) at any recorded timepoints. Similarly, there were no significant associations between total years of football participation and alcohol use at any recorded timepoints. Concussion history and years of football participation were not associated with scores on a measure of problematic alcohol use (CAGE) for the during professional career (b = − 0.01, SE = 0.10, p = .901; b = .04, SE = 0.03, p = .186, respectively) and 2001 timepoints (b = − 0.08, SE = 0.09, p = .375; b = − 0.02, SE = 0.03, p = .480, respectively). In general, effect estimates for concussion history and years of football participation with alcohol use outcomes were either variable (higher and lower) or stable across the different timepoints of interest (Fig. 1; CIs overlap with point estimates across timepoints). Sensitivity analyses indicated that the earlier results did not differ when the history of concussion across all levels of play and non-sport concussions collected at the most recent timepoint (2019) was included in the statistical models. Due to lower frequency endorsement of the highest drinks consumed per occasion categories in 2001 and 2019, additional sensitivity analyses that collapsed and binarized consumption groups were performed for 2001 (0–5 and 6+) and 2019 (0–4 and 5+). Results of these analyses again showed no significant associations between concussion history or total years of football participation and alcohol consumption per occasion.

Fig. 1.

Concussion history, years of NFL participation, and alcohol use over timepoints of during professional career, 2001, and 2019; Effect estimates derived from ordinal logistic regression models. Prof. Career = duration of time playing in the National Football League; Unstd Beta = unstandardized beta estimate; 95% CI = 95% confidence interval.

Discussion

In this longitudinal sample of former NFL players, concussion history and years of American football participation were not significantly associated with alcohol use. The discrepant findings from prior research supporting the link between subsequent alcohol use following TBI may be due to multiple factors. For example, severity of TBI could act as a moderator in this relationship. In a study examining the longitudinal trajectory of alcohol use in the 5 years after TBI in active-duty service members and veterans (Steffen-Allen et al., 2022), among those with a moderate to severe TBI, there was a drop in alcohol use during initial recovery, but there were overall increases in alcohol use across time. However, in those with a mild TBI, an initial spike in acute alcohol use was noted before a return to pre-TBI use levels. The absence of association observed in the current study may reflect the moderating influence of injury severity, as the large majority of head injuries sustained during sport are mild in severity.

Increased alcohol use has been theorized to be a consequence of greater concussion history, and as such, thought to confound associations between concussion history and neurobehavioral functioning later in life (Manley et al., 2017). However, by definition, a confounding factor needs to be associated with both the exposure and outcome variables of interest (Szklo & Nieto, 2018), which was not observed in the current study (i.e., no association between concussion history and alcohol use). This is not to say that alcohol use is not bidirectionally associated with physical (e.g., pain) and mental (e.g., depression) health, or does not adversely affect cognition and risk of neurocognitive disorders. In fact, previously reported findings from a larger cross-sectional (2019 timepoint) analysis in the NFL-LONG sample showed a positive association between higher alcohol use frequency and greater emotional behavioral dyscontrol (Walton et al., 2021). Instead, the current findings suggest that any associations between head injury history and neurobehavioral functioning in former football players is less likely to be due to alcohol use patterns specifically. This lack of relationship is important for clinicians to be aware of as they seek to identify and distinguish etiologies of neurobehavioral presentations as part of differential diagnosis and feedback for clients whose histories include factors such as head injury or alcohol use. Given the potential independent influences of head injury history and alcohol use on clinical outcomes, this is important from a clinical perspective, as alcohol use represents a target for treatment/intervention, where head injury history would not be modifiable. Other factors, such as pain, have been observed to influence patterns of substance use (e.g., opioids) among former NFL players (Cottler et al., 2011; Mannes et al., 2020), and these may similarly influence alcohol use and aspects of neurobehavioral functioning.

This study had multiple strengths, including a longitudinal design, prospective recording of concussion history, and large sample size. Study weaknesses and limitations included self-reported nature of alcohol use and concussion history (not corroborated with medical records) as well as potential bias that could arise from incorporation of only concussion/mTBI, but not potential history of moderate-to-severe TBI. Changes to methods of recording alcohol use at 2019 prevented the direct measurement of change across time. Potential bias due to sample enrollment/retention is possible; however, a wide distribution of concussion history, total years of football participation, and alcohol use is represented within the sample, allowing for testing of associations across variable ranges. Although the analytic sample size was not based on a planned a priori power analysis (sample size was determined by longitudinal enrollment procedures), with 348 subjects, the study had 80% power to detect a partial Spearman’s correlation of ±0.15 at a significance criterion of .05 for a two-tailed test. Finally, this study did not include the direct measurement or analysis of cognitive or behavioral functioning as an outcome; exploration of other variables that may account for poor neurobehavioral outcomes in this population is warranted.

Conclusion

In this study of former NFL players, alcohol use was not related to concussion history or total years of American football participation. These findings do not support the use of alcohol consumption behaviors as explanatory variables in linking head injury history to adverse neurobehavioral functioning in former football players. Future studies are needed to identify factors that influence long-term neurobehavioral health in former contact sport athletes, either independently or in relation to alcohol consumption.

Authors’ contributions

Brittany Lang (Conceptualization, Formal analysis, Writing—original draft), Zachary Kerr (Conceptualization, Data curation, Investigation, Methodology, Writing—review & editing), Avinash Chandran (Conceptualization, Investigation, Methodology, Writing—review & editing), Samuel R. Walton (Conceptualization, Data curation, Methodology, Writing—review & editing), Rebekah Mannix (Conceptualization, Investigation, Methodology, Writing—review & editing), Landon B. Lempke (Conceptualization, Methodology, Writing—review & editing), J. D. DeFreese (Conceptualization, Methodology, Writing—review & editing), Ruben J. Echemendia (Conceptualization, Funding acquisition, Writing—review & editing), Kevin M. Guskiewicz (Conceptualization, Funding acquisition, Investigation, Writing—review & editing), William P. Meehan III (Conceptualization, Funding acquisition, Investigation, Writing—review & editing), Michael A. McCrea (Conceptualization, Funding acquisition, Investigation, Supervision, Writing—review & editing), and Benjamin L. Brett (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing—original draft, Writing—review & editing).

Acknowledgements

We are grateful for the participation of the athletes, without whom this research would not be possible. The authors would also like to thank Alexa Wild (Department of Neurosurgery at the Medical College of Wisconsin), Candice Goerger (Center for the Study of Retired Athletes at the University of North Carolina at Chapel Hill), Caprice Hunt (Center for the Study of Retired Athletes at the University of North Carolina at Chapel Hill), and Danielle Hunt (Department of Orthopedics at Boston Children's Hospital) for study coordination and management; We would also like to thank Daniel Huber for his assistance with data quality and management (Department of Neurosurgery at the Medical College of Wisconsin).

Competing Interests

Dr. Brett reports grants from the National Institute on Aging and National Institute of Neurological Disorders and Stroke, and honoraria for conference presentations. Dr. Chandran discloses research funding from the National Collegiate Athletic Association as the Director of the NCAA Injury Surveillance Program, and separately from the North Carolina Translational and Clinical Sciences Institute, the Atlantic Coast Conference Center for Research in Intercollegiate Athletics, as well as the Canadian Institutes of Health Research- Human Development, Child and Youth Health Institute, on unrelated projects; he also discloses receiving honoraria from the Sports Neuropsychological Society as well as the American Medical Society for Sports Medicine. Dr. Kerr reports grants from National Institutes of Health; grants from Centers for Disease Control and Prevention; and grants from National Football League. Dr. Lempke reports grants from the Eastern Athletic Trainers' Association and VALD. Dr. DeFreese reports grants from the Atlantic Coast Conference and the Association for Applied Sport Psychology. Dr. Echemendia is a paid consultant for the National Hockey League and co-chair of the National Hockey League/National Hockey League Players Association Concussion Subcommittee and Major League Soccer, and provides testimony in matters related to mTBI. Dr Guskiewicz reports compensation from National Collegiate Athletic Association for other services and grants from Boston Children's Hospital (sub-award from the National Football League). Dr. Meehan receives royalties from ABC-Clio publishing, Springer International, and Wolters Kluwer. His research is funded by philanthropic support from the National Hockey League Alumni Association through the Corey C. Griffin Pro-Am Tournament and a grant from the National Football League. Dr. McCrea acknowledges research funding from the National Institutes of Health, U.S. Department of Defense, Centers for Disease Control and Prevention, National Collegiate Association and National Football League (via subaward from Boston Children's Hospital). Dr Mannix reports grants from U.S. Department of Defense; grants from NFL Foundation; and grants from National Institute of Neurological Disorders and Stroke.

Funding

This study (NFL-LONG) was funded by the National Football League (NFL) and supported by the National Institutes of Health National Institute on Aging (BLB- K23 AG073528).