Bias-Adjusted Study Finds About 75% of Deceased Former NFL Players Had CTE

A new study in JAMA Network Open estimates that 74.5% of former NFL players who died from 2008 through 2023 had chronic traumatic encephalopathy, or CTE, at death after researchers adjusted for the well-known bias in brain-bank donations.

That methodological step is the point of the paper. For years, some of the most cited CTE figures have come from donated brains, many from families seeking answers after a player showed cognitive or behavioral problems. Those studies helped establish the disease’s link to repeated head impacts, but they could not cleanly answer how common CTE was across all deceased former NFL players. The new study attempts to do that by estimating prevalence in a defined population, not just among donors.

The paper, titled “Bias-Adjusted Prevalence of Chronic Traumatic Encephalopathy Among Deceased National Football League Players,” was published online Oct. 28. Lead authors included Julia Gallini, Michael L. Alosco and Bobak Abdolmohammadi, with senior and collaborating authors including Ann C. McKee, Robert A. Stern, Jesse Mez, Christopher Nowinski and Yorghos Tripodis. The researchers are affiliated with Boston University, Mass General Brigham and the Concussion & CTE Foundation.

The study examined all former NFL players who died between Jan. 1, 2008, and Dec. 31, 2023, a group of 2,362 people. Within that cohort, researchers identified 396 brain donors and 1,966 nondonors. Of the donors, 387 received a neuropathologic evaluation, and 362 had CTE — a raw donor prevalence of 93.5%.

But the researchers said that donor-only figure is inflated by selection bias. Using data from the UNITE brain bank and mortality data from the U.S. National Death Index, they estimated that players with clear cognitive symptoms were about five times more likely to have their brains donated than players without CTE. After adjusting for that imbalance, the estimated prevalence of CTE neuropathology at death across the full deceased cohort was 74.5%, with a 95% confidence interval of 69.6% to 78.3%.

The study also estimated that 27.2% of the deceased former players had a history of all-cause dementia during life, with a 95% confidence interval of 22.6% to 31.6%. In sensitivity analyses, the estimated CTE prevalence ranged from about 70.3% to 86.1%, depending on the assumptions used.

The authors stressed the limits of the findings. The estimate applies only to deceased former NFL players in the study period. It does not apply to living former players or active players, and it cannot be used to infer how many living people have CTE.

CTE is a progressive neurodegenerative disease associated with repeated head impacts and defined by abnormal tau protein buildup in the brain. It can be definitively diagnosed only after death through neuropathologic examination. Previous Boston University brain-bank studies found CTE in more than 90% of former NFL player brains studied, but those findings have long come with the caveat that donated brains are not representative of all players. A BMJ study published in August by some of the same researchers estimated that at least one in four former NFL players who died between 2016 and 2021 had CTE, using a different method; the new paper is more directly focused on correcting donation bias.

“Providing a precise estimate for the prevalence of CTE among deceased NFL players helps define the scale of the problem and informs priorities for the potential current and future medical needs of this community,” Gallini said in a Boston University press release.

She added: “With CTE consistently found in less than one percent of donors to community brain banks, an estimated 75 percent prevalence among NFL players who died over a 16-year period is concerning, and suggests we need to increase efforts to support families, diagnose CTE during life, and treat CTE-related symptoms.”

The NFL said in a statement that it “continuously strives to make the game of football safer, including by implementing strategies to reduce concussions and head impacts.”

The study’s broader takeaway was that selection bias inflates donor-based estimates, but does not come close to erasing the burden of disease. For researchers and families, the findings offer a clearer measure of the long-term neurological toll among former players who have died — and of the medical and support needs that follow.