NFL’s 99% CTE Figure Was Inflated, But The Corrected Rate Is Still 74.5%

For years, one number has shaped the debate over football and brain disease: 110 out of 111. In a 2017 study, that many brains from former NFL players showed signs of CTE, a degenerative brain disease linked to repeated hits to the head. That number came with a well-known catch. Families who watched a loved one struggle with memory, mood, or behavior may have been more likely to donate his brain for study, which would push the rate far above what exists across all players. A new analysis from Boston University set out to measure how big that distortion really is, and even after the correction, the rate stays high.

After adjusting for which families chose to donate, researchers estimate that 74.5% of former NFL players who died between 2008 and 2023 had CTE in their brains at death. That estimate covers all 2,362 former players who died in that window, not just the few hundred whose brains reached a lab. It sits well below the near-perfect rate from earlier brain-bank studies, yet it still works out to roughly three out of every four players.

Doctors can only confirm CTE by examining brain tissue after death, so most of what scientists know comes from donated brains. That reliance on donation has long been the weak spot in this research. According to the authors, the new study, published in JAMA Network Open, is the first to estimate how common CTE is across an entire population of deceased NFL players while accounting for who donates.

A 99% figure built on a skewed sample

Most of what is known about CTE in football players comes from donors to a Boston University brain bank. Families decide whether to donate, and that decision is not random. In the general public, brain donation has been linked to age, race, marital status, depression, and dementia. Among football players, the authors point out, those who showed cognitive, mood, or behavioral symptoms during life may be even more likely to end up in a brain bank, which would inflate any rate calculated from donors alone.

Earlier attempts to correct for this produced answers too broad to be useful. A 2019 analysis concluded that the true rate among professional players could plausibly fall anywhere from about 10% to 100%.

Correcting brain-donor bias in CTE research on NFL players

To narrow that range, the team needed information on players whose brains were never studied. They turned to the National Death Index, a national database of U.S. death records, which identified 2,362 former players who had appeared in at least one regular-season or postseason NFL game and died between 2008 and 2023. Of those, 396 had their brains donated and 1,966 did not.

Among donors whose brains could be fully evaluated, 362 of 387, or 93.5%, had CTE. Death records cannot reveal whether a nondonor had the disease, so researchers needed a stand-in they could measure in both groups. They chose clear cognitive symptoms, reasoning that families are most likely to donate after seeing a loved one’s memory and thinking decline. Death certificates listing a dementia-related cause of death captured part of that group. Because many people with memory problems die of something else, the team also added estimated cases of mild cognitive impairment, an earlier and milder stage of memory decline, based on rates reported in a separate survey of former NFL players over 50.

A statistical model then compared donors with nondonors while accounting for age at death, race, position, years in the league, Pro Bowl appearances, and Hall of Fame status. Players with clear cognitive symptoms were about five times more likely to have their brains donated. Each additional season in the NFL also raised the likelihood of donation, while dying at an older age lowered it. That five-to-one ratio became the correction factor applied to the brain bank’s results.

CTE in NFL players after the correction

Applied to all 2,362 deceased players, the correction produced the 74.5% estimate, with a likely range of 69.6% to 78.3%. Selection bias was real and pushed donor-based numbers upward, but it could not erase a high rate of disease.

To see how much the result depended on their assumptions, the researchers reran the analysis nine more ways. Some versions limited the sample to older ages at death, others changed how memory problems were estimated, and one counted death by suicide as a sign of symptoms. Eight of those versions landed between 70.3% and 86.1%. A ninth was designed as a worst case that would drag the estimate down, and it still came in at 63.7%. Researchers also mapped how the share of players with CTE would shift under even more lopsided donation patterns: “Even if individuals with cognitive symptoms were 10 times more likely to donate their brains, the estimated prevalence would still exceed 60%.”

A second estimate looked at how many former players had a history of dementia during life. Among donors, 64.9% did. After the same correction, the estimate across all deceased players fell to 27.2%, or a little more than one in four. That figure likely understates lifetime risk. Donors died at an average age of 67, compared with nearly 75 for nondonors, and because dementia becomes more common with age, players who lived longer had more time to develop it.

Football’s long-running CTE debate

A 74.5% estimate does not mean three in four former players had CTE symptoms. It counts every stage of the disease found in brain tissue, and the authors note that many cases may have been early-stage and possibly symptom-free at death. It also cannot be applied to players alive today. Whether CTE itself shortens lives remains unknown, so the players studied here may differ from those still living. And because nondonor brains were never examined, the entire correction rests on memory symptoms standing in for the disease, an imperfect match the researchers acknowledge.

For years, skeptics have pointed to brain-donor bias as a reason to discount CTE numbers from football. This study measured that bias directly and found it real, large enough to pull the estimate down by about 19 percentage points. Even the most cautious version of the math left close to two-thirds of deceased NFL players with the disease. Donor bias explains part of football’s alarming CTE numbers. It does not explain them away.

---

Disclaimer: This article is for general information and does not provide medical advice or a way to diagnose CTE.

---