by Simone Salemme
For October 2026 we have selected Daneshvar DH, Nowinski CJ, Abdolmohammadi B, et al. Prevalence of chronic traumatic encephalopathy at death in National Football League players: retrospective population-based cohort study, 2008-21. BMJ. 2026;394:e100418. Published 2026 Aug 25. doi:10.1136/bmj-2026-100418
Chronic traumatic encephalopathy (CTE) can currently be diagnosed definitively only at neuropathological examination after death, making its epidemiology difficult to characterise. Previous brain-bank studies have reported high proportions of CTE among former American football players, but these estimates are vulnerable to selection bias because families may be more likely to donate brains when symptoms were present. Daneshvar and colleagues addressed this challenge by studying a fully enumerated cohort of former National Football League (NFL) players who died between 2008 and 2021, estimating the possible range of CTE prevalence at death and examining the association between severe CTE pathology and dementia.
The study included 1,712 former NFL players who died during 2008–21, of whom 338 donated their brains for neuropathological evaluation. Neuropathologists, masked to clinical and playing histories, assessed CTE diagnosis and stage, while clinicians masked to neuropathological status adjudicated dementia based on medical records and informant-based clinical histories. The authors calculated a conservative minimum CTE prevalence by dividing the number of donors with CTE by all NFL deaths, and a possible maximum prevalence assuming that all non-donors could have had CTE. They also used inverse probability weighting to account for differential brain donation when analysing the association between stage IV CTE and dementia among donors.
Among the 338 brain donors, 315 had CTE, corresponding to 93.2% of studied brains. Across all 1,712 NFL players who died during the study period, the possible prevalence of CTE at death ranged from 18.5% to 98.7%. During 2016–21, when brain donation was more frequent, 235 of 878 decedents donated brains and 215 had CTE, yielding a possible prevalence range of 24.5% to 97.7%. Dementia was also common among donors: 202 of 338 donors had clinician-adjudicated dementia, and 104 had stage IV CTE. Stage IV CTE was associated with dementia after inverse probability weighting and multivariable adjustment (risk ratio 1.44, 95% CI 1.16–1.78). Dementia was present in 94 of 104 donors with stage IV CTE.
The study also highlights an important limitation of relying on death certificates to capture dementia burden. Among donors with clinician-adjudicated dementia, only 40.6% had neurodegenerative disease listed as either a primary or secondary cause of death, suggesting substantial under-ascertainment when using death certificate data alone.
Overall, this study provides one of the clearest estimates to date of the possible burden of CTE at death in former NFL players from the hard-shell helmet era. Its findings indicate that CTE neuropathology was present in at least 24.5% of former NFL players who died during 2016–21, and that severe CTE pathology was associated with dementia among brain donors. At the same time, the findings should not be interpreted as estimating CTE prevalence among living football players or as generalisable to all American football populations. The key contribution is more specific and clinically important: in a highly exposed, fully enumerated professional football cohort, CTE at death was not rare, dementia was common among donors, and more severe CTE pathology showed a clear association with clinical dementia.
