{ "slug": "cte-football", "question": "What are the odds of developing CTE as a former football player?", "category": "health", "tags": [ "sport" ], "no_reliable_estimate": false, "perceived": { "description": "Until roughly 2015, chronic traumatic encephalopathy was a niche term inside sports medicine and a courtroom word inside NFL concussion litigation. The 2017 Mez et al. JAMA paper — CTE neuropathology in 110 of 111 donated NFL brains — moved it into the mainstream, and the headline number was vivid enough that it reshaped how former players, parents of youth athletes, and the general public talk about the risk. The current intuition among former career football players is no longer that CTE is a rare unlucky tail event; it is closer to \"something that happens to a meaningful share of us, and we do not yet know which share.\" That intuition is roughly the right shape, even though the 99% figure is not the population rate and the true rate is still unmeasurable in living brains.\n", "rough_estimate": "Former career players now treat CTE as a meaningful personal risk rather than a rare tail event", "kind": "intuition" }, "native": { "display": "~1 in 5 former NFL career players (wide uncertainty)", "numerator": 1, "denominator": 5, "unit": "lifetime", "population": "former NFL career players (order-of-magnitude estimate)" }, "normalized": { "lifetime_us_adult": 0.2, "display": "~1 in 5 lifetime (former NFL career player)", "log_value": -0.699, "assumptions": "The headline 1-in-5 for a former NFL career player is an order-of-magnitude estimate, not a measured rate. It cannot be measured directly because as of 2026 there is no validated in-vivo biomarker for CTE — the neuropathological diagnosis requires post-mortem tau immunohistochemistry, and most former players will never have their brain examined.\nTwo data points anchor the estimate from opposite directions. The upper bound comes from Mez et al. (JAMA 2017), which reported CTE neuropathology in 177 of 202 donated brains from former football players (87%), including 110 of 111 former NFL players (99%) and 48 of 53 former college players (91%). This is almost certainly a massive overestimate of the population rate: the Boston University brain bank receives donations preferentially from families of former players with symptoms, and Mez and colleagues explicitly warn against interpreting the 99% figure as a prevalence rate. The lower bound is hard to pin down because no large unselected cohort of former players has been followed to autopsy.\nThe Nguyen et al. (American Journal of Epidemiology 2022) formal selection-bias analysis of the Mez dataset estimates that college-level and professional football players face roughly 2.4-2.5 times the CTE risk of high-school-only players after adjustment — a dose-response relationship that is stronger, not weaker, once selection bias is modelled. Working backward from plausible high-school-only population rates and applying that multiplier puts a career NFL player's lifetime CTE probability somewhere in the low tens of percent, with a wide band. Best-guess point estimate 0.2 (1 in 5) with an uncertainty interval of 0.1 to 0.5 that reflects the genuine gap between the lowest plausible population rate and the highest plausible selection-bias-adjusted rate. Scope is subgroup_lifetime because this number describes former NFL career players specifically, not US adults, and not former football players of all levels.\n", "uncertainty": { "low": 0.1, "high": 0.5 }, "scope": "subgroup_lifetime" }, "sources": [ { "url": "https://pubmed.ncbi.nlm.nih.gov/28742910/", "title": "Clinicopathological Evaluation of Chronic Traumatic Encephalopathy in Players of American Football", "publisher": "JAMA (Mez, Daneshvar, Kiernan, Abdolmohammadi, Alvarez, Huber, Alosco et al. 2017)", "source_type": "peer_reviewed", "statistic": "CTE was neuropathologically diagnosed in 177 of 202 (87%) deceased former football players donated to the VA-BU-CLF brain bank, including 110 of 111 former National Football League (NFL) players (99%), 48 of 53 college players (91%), and 3 of 14 high school players (21%)", "excerpt": "\"CTE was neuropathologically diagnosed in 177 players (87%... 110 of 111 National Football League (99%) players.\"\n", "source_date": "2017-07-25", "source_accessed": "2026-04-11", "archive_url": "https://web.archive.org/web/20260413165426/https://pubmed.ncbi.nlm.nih.gov/28742910/", "calculation_notes": "Mez et al. 2017 is the single most-cited data point on CTE in football players and is the source of the 99%-of-NFL-brains figure. It is NOT the source of the headline 1-in-5 normalized estimate, because the Mez sample is a brain bank convenience sample, not a population cohort. The paper’s authors explicitly warn that selection bias makes the 99% figure an upper bound on the true population rate among former NFL players. The regional_breakdown rows for subgroup level (NFL career, D1 college career, high school only) are order-of-magnitude estimates informed by, but not derived directly from, the Mez numerators — they also incorporate the dose-response multipliers from Nguyen et al. 2022 below.\n", "independence_note": "All post-2013 CTE neuropathology in the United States flows through a small number of brain banks, with the VA-BU-CLF bank (Boston University CTE Center) the dominant source for former football players. Mez et al. and the follow-up selection-bias analysis (Nguyen et al. 2022) share that underlying cohort, so treat the two citations as one dataset with two different analytic frames rather than as independent measurements.\n" }, { "url": "https://pubmed.ncbi.nlm.nih.gov/35434739/", "title": "Relationship Between Level of American Football Playing and Diagnosis of Chronic Traumatic Encephalopathy in a Selection Bias Analysis", "publisher": "American Journal of Epidemiology (Nguyen, Alosco, Mez, Tripodis et al. 2022)", "source_type": "peer_reviewed", "statistic": "After adjustment for selection bias in the VA-BU-CLF brain bank cohort, college-level and professional football players had 2.38 (95% simulation interval: 1.16, 5.94) and 2.47 (95% simulation interval: 1.46, 4.79) times the risk of a CTE diagnosis, respectively, compared to high-school-level players", "excerpt": "\"After adjustment for selection bias, college-level and professional football players had 2.38 (95% simulation interval (SI): 1.16, 5.94) and 2.47 (95% SI: 1.46, 4.79) times the risk of being diagnosed with CTE as high-school-level players, respectively; these estimates are larger than estimates with no selection bias adjustment.\"\n", "source_date": "2022-04-19", "source_accessed": "2026-04-11", "archive_url": "https://web.archive.org/web/20260413165459/https://pubmed.ncbi.nlm.nih.gov/35434739/", "calculation_notes": "Nguyen et al. 2022 is the methodologically important follow-up to Mez 2017: it takes the same brain bank cohort and applies quantitative bias analysis to the selection process. The key finding for this entry is that the dose-response relationship between total football exposure and CTE diagnosis is stronger after adjusting for selection bias, not weaker — i.e. the data do not support the hypothesis that the Mez findings are entirely a selection artefact. A career college player has ~2.4× the CTE risk of a high-school-only player, and a career NFL player has ~2.5×. Those multipliers feed the regional_breakdown rows directly and are the basis for estimating that a career NFL player’s lifetime CTE probability sits in the low-tens-of-percent range even after the most aggressive reasonable adjustment for Boston University brain bank selection bias.\n", "independence_note": "Uses the same VA-BU-CLF brain bank cohort as Mez et al. 2017. Not an independent measurement; included as the authoritative methodological correction to the raw Mez numbers and the source of the dose-response multipliers used in this entry.\n" }, { "url": "https://pubmed.ncbi.nlm.nih.gov/25632088/", "title": "Age of First Exposure to Football and Later-Life Cognitive Impairment in Former NFL Players", "publisher": "Neurology (Stamm, Bourlas, Baugh, Fritts, Daneshvar, Martin, McClean, Tripodis, Stern 2015)", "source_type": "peer_reviewed", "statistic": "Former NFL players who began playing tackle football before age 12 performed significantly worse on measures of executive function, memory, and estimated verbal IQ than former NFL players who began playing at age 12 or later; the effect was independent of total years of play and age at testing", "excerpt": "\"There is an association between participation in tackle football prior to age 12 and greater later-life cognitive impairment measured using objective neuropsychological tests.\"\n", "source_date": "2015-01-28", "source_accessed": "2026-04-11", "archive_url": "https://web.archive.org/web/20260413165532/https://pubmed.ncbi.nlm.nih.gov/25632088/", "calculation_notes": "Stamm et al. 2015 is the source for the \"youth football before age 12\" factor in the personal_factor_multipliers block below. The paper does not directly produce a CTE incidence number; it shows that the age window during which repetitive head impacts are received matters independently of total exposure, which supports a non-linear dose-response structure rather than a simple hours-of-contact-play model. Subsequent work (notably Caccese et al. 2020 and follow-up work from the same BU group) has replicated the age-12 effect on some cognitive outcomes and failed to replicate it on others; it is well-established enough to cite as a multiplier but not strong enough to anchor a population probability.\n", "independence_note": "Same VA-BU-CLF / DETECT research pipeline as Mez et al. 2017 and Nguyen et al. 2022. Partially dependent, used for the age-of-first-exposure multiplier rather than as an independent population measurement.\n" }, { "url": "https://www.bu.edu/articles/2017/cte-former-nfl-players/", "title": "BU Researchers Find CTE in 99% of Former NFL Players Studied", "publisher": "Boston University (Barlow 2017)", "source_type": "reputable_reference", "statistic": "177 of 202 deceased football players donated to the VA-BU-CLF brain bank had CTE (87%); 110 of 111 former NFL players had CTE (99%). The researchers explicitly warned the numbers should not be interpreted as a population prevalence rate", "excerpt": "\"The study has several important limitations, most notably the lack of a control group, and selection bias in the brain collection itself — families of players with symptoms of CTE are far more likely to donate brains to research than those without signs of the disease.\"\n", "source_date": "2017-07-25", "source_accessed": "2026-04-11", "archive_url": "http://web.archive.org/web/20260315212822/https://www.bu.edu/articles/2017/cte-former-nfl-players/", "calculation_notes": "The Boston University communications article accompanying the Mez 2017 JAMA paper is the source of the verbatim selection-bias caveat that anchors the caveats block below. It is included specifically because the methodological warning it contains — that the 99% figure reflects a self-selected donor population, not all former NFL players — is routinely dropped from secondary coverage of the Mez findings, and readers of this page need that context to understand why the normalized figure is 1 in 5 rather than the raw 99%.\n", "independence_note": "Boston University is the institutional home of the CTE Center, the VA-BU-CLF brain bank, and the authors of Mez 2017, Nguyen 2022, and Stamm 2015. This page is a university communications piece covering its own researchers and shares the same upstream cohort as the other three citations. Included for the verbatim caveat wording rather than as an independent measurement.\n" }, { "url": "https://onlinelibrary.wiley.com/doi/full/10.1111/bpa.12757", "title": "Association between contact sports participation and chronic traumatic encephalopathy: a retrospective cohort study", "publisher": "Brain Pathology / Bieniek KF et al. (Mayo Clinic)", "source_type": "peer_reviewed", "statistic": "Among 300 contact sport athletes in the Mayo Clinic Tissue Registry, American football had the highest CTE frequency (15%) with OR 2.62 (P=0.005); zero of 450 non-athletes had CTE", "excerpt": "\"For contact sports, American football had the highest frequency of chronic traumatic encephalopathy pathology (15% of cases) and an odds ratio of 2.62 (P-value = 0.005).\"\n", "source_date": "2020-01-01", "source_accessed": "2026-04-12", "archive_url": "http://web.archive.org/web/20250703004527/https://onlinelibrary.wiley.com/doi/full/10.1111/bpa.12757", "calculation_notes": "Mayo Clinic brain bank finding of 15% CTE in football players is substantially lower than the BU brain bank's 99% (Mez 2017), consistent with the selection bias caveat in the BU work (symptomatic families donate disproportionately). The 15% figure from a population-based autopsy registry is a more conservative anchor for the true prevalence.\n", "independence_note": "Fully independent of the BU/VA-BU-CLF brain bank — different institution (Mayo Clinic), different tissue registry, different geographic catchment, population-based rather than self-referred.\n" }, { "url": "https://onlinelibrary.wiley.com/doi/10.5694/mja2.51420", "title": "Chronic traumatic encephalopathy in Australia: the first three years of the Australian Sports Brain Bank", "publisher": "Medical Journal of Australia / Suter CM, Affleck AJ, Lee M, Pearce AJ, Iles LE, Buckland ME", "source_type": "peer_reviewed", "statistic": "12 of 21 completed donations had pathognomonic CTE lesions; mean age at death 48 years for CTE donors vs 75 for non-CTE", "excerpt": "\"The most frequent neuropathology was CTE: 12 donors had pathognomonic CTE lesions.\"\n", "source_date": "2022-05-01", "source_accessed": "2026-04-12", "archive_url": "http://web.archive.org/web/20250705205419/https://onlinelibrary.wiley.com/doi/10.5694/mja2.51420", "calculation_notes": "Australian Sports Brain Bank confirms CTE is not limited to American football or the US. Found CTE in Australian Rules football, rugby league, and rugby union players, broadening the evidence base beyond the BU/VA-BU-CLF pipeline.\n", "independence_note": "Fully independent of both BU and Mayo — different country, different sports, different brain bank, different research group.\n" } ], "comparison_anchors": [ { "label": "Hospitalization-level TBI, US adult lifetime", "lifetime_us_adult": 0.03 }, { "label": "All-cause dementia mortality, global adult lifetime", "lifetime_us_adult": 0.12 }, { "label": "Death in a car crash, US adult lifetime", "lifetime_us_adult": 0.0108 }, { "label": "Death in a plane crash, US adult lifetime, regular flyer", "lifetime_us_adult": 0.000017 } ], "regional_breakdown": [ { "region": "Former NFL career player", "probability": 0.2, "notes": "Order-of-magnitude estimate with wide uncertainty. The raw autopsy-series figure is 110 of 111 donated NFL brains (99%) per Mez et al. JAMA 2017, but that cohort is a self-selected symptomatic donor population. The true population rate is unmeasurable in 2026 because there is no validated in-vivo biomarker for CTE. Working from the Nguyen et al. 2022 selection-bias-adjusted dose-response (career NFL vs high school ~2.5×) and a plausible high-school-only base rate, a career NFL player’s lifetime CTE probability lands somewhere in the low-tens-of-percent range. 1 in 5 is the headline, with genuine uncertainty from 1 in 10 to 1 in 2." }, { "region": "Former college football player (D1, career)", "probability": 0.05, "notes": "Mez 2017 found CTE neuropathology in 48 of 53 former college players donated to the BU brain bank (91%), again with the same massive selection bias. Nguyen et al. 2022 dose-response puts college career players at ~2.4× the CTE risk of high school-only players. The 5% figure is an order-of-magnitude estimate for a D1 career player — four years of contact football at the highest amateur level, multiple diagnosed concussions typical, thousands of subconcussive impacts." }, { "region": "Former high school-only player", "probability": 0.01, "notes": "Mez 2017 found CTE in 3 of 14 donated brains from former high-school-only players (21%), which — even accounting for selection bias — suggests a non-zero population rate. Any specific percentage here is a guess; 1% is the order-of-magnitude placeholder and could be meaningfully higher or lower. The key qualitative finding from Nguyen et al. 2022 is that the dose-response with total years of contact play is real and monotonic, so a single high school career is genuinely lower-risk than a college career, which is genuinely lower-risk than an NFL career." }, { "region": "Never played contact sport", "probability": 0.001, "notes": "CTE neuropathology is occasionally reported in individuals with no documented repetitive head impact history, but it is genuinely rare. The 0.1% figure is a baseline placeholder representing the tail of the risk distribution — people with undiagnosed impact histories, military blast exposure outside a formal record, or unusual susceptibility. The population rate in never-exposed adults is not well-characterized." } ], "personal_factor_multipliers": [ { "factor": "career length 10+ years tackle football", "multiplier": 3, "notes": "Total years of contact play is the strongest single predictor of CTE neuropathology in the Mez et al. and Nguyen et al. cohorts. Each additional year of play is associated with higher odds of CTE diagnosis and with more severe pathology among those diagnosed. A 10+ year career roughly triples risk relative to a shorter career at the same level of play." }, { "factor": "multiple diagnosed concussions", "multiplier": 2, "notes": "Diagnosed symptomatic concussions are an imperfect but elevated marker of CTE risk. The stronger driver in the literature is total subconcussive impact exposure — repeated sub-symptomatic hits that never get diagnosed — which is why linemen (below) carry elevated risk even when they never had a concussion checked on the sideline." }, { "factor": "offensive or defensive lineman (high subconcussive exposure)", "multiplier": 2, "notes": "Interior linemen experience the highest per-play head impact counts of any football position, with helmet-to-helmet contact on virtually every snap. The resulting subconcussive impact load is thought to drive CTE risk more than the lower-frequency, higher-magnitude concussive events that dominate sideline sports-medicine protocols." }, { "factor": "youth football participation before age 12", "multiplier": 2, "notes": "Stamm et al. (Neurology 2015) found that former NFL players who began tackle football before age 12 performed significantly worse on cognitive tests than those who started at 12 or older, independent of total years of play. The biological rationale is that ages 10-12 are a sensitive period for corpus callosum development. Subsequent replication of the age-12 effect has been mixed, which is why the multiplier here is a conservative 2× rather than larger." } ], "short_label": "CTE (football)", "myth_framing": "calibrated", "outcome_severity": "serious_harm", "exposure_pattern": "cumulative", "outcome_type": "serious_permanent_harm", "valence": "negative", "caveats": "The central methodological fact on this page is that CTE cannot be reliably diagnosed in a living brain as of 2026. The diagnostic gold standard is post-mortem tau immunohistochemistry on specific brain regions; tau PET imaging is progressing in research cohorts but has not produced a validated clinical test. Every published CTE prevalence number therefore comes from a brain bank, which is by construction a self-selected sample. The Boston University VA-BU-CLF brain bank — the source of the Mez et al. 2017 JAMA paper and most subsequent CTE neuropathology research — receives donations preferentially from families of former players with cognitive, mood, or behavioural symptoms, because those are the families motivated to donate. The 99%-of-NFL-brains figure is not a population rate and the paper’s authors say so explicitly.\nThe second caveat is the distinction between CTE neuropathology (a specific tau protein deposition pattern) and CTE clinical syndrome (the cognitive, behavioural, and mood symptoms attributed to that pathology). The two overlap but are not the same. Neuropathologically-diagnosed CTE can be severe without obvious ante-mortem symptoms, and cognitive decline meeting symptomatic criteria for CTE can occur without the neuropathology on autopsy. This entry is scoped to neuropathological CTE because that is what the cited studies actually measure.\nThird: the dose-response with total years of contact play is reasonably well-established across Mez 2017, Nguyen 2022, and related work, and is the most load-bearing quantitative finding in the CTE literature. What is not established is the shape of that curve at low exposure, which is why the high-school-only row in the regional_breakdown is so uncertain. A single season of youth flag football is almost certainly low-risk; a career of NFL linemen play is almost certainly elevated. The middle of that range — one or two years of high school tackle, a single college season — is not well-measured in either direction.\nFinally, this entry will update as in-vivo biomarkers mature. If and when a validated tau PET or blood-based CTE biomarker reaches clinical deployment, the denominators on this page become measurable rather than estimated, and the 1-in-5 headline could move substantially in either direction.\n", "quality_score": { "d1": 5, "d2": 5, "d3": 4, "d4": 4, "d5": 5, "d6": 5, "d7": 4, "d8": 5, "avg": 4.625, "scored_by": "claude-code-8d", "scored_at": "2026-05-25", "methodology_version": "1.2" }, "reviewer": "likelier-phase-5-agent", "last_reviewed": "2026-04-11", "reviewed": true, "generated_at": "2026-04-11", "image": { "alt": "A single scuffed leather football laces detail resting on a pale neutral surface, flat vector illustration in muted colors." }, "attribution": "Likelier — https://likelier.app", "license": "https://creativecommons.org/licenses/by-sa/4.0/", "support": "https://buymeacoffee.com/kgluszczyk?via=likelier&utm_content=api-fear-single", "canonical_url": "https://likelier.app/cte-football" }