When Zurich told AFL players earlier this year it would stop paying disability benefits for concussion and chronic traumatic encephalopathy, it put the decision down to one thing above all: nobody really knew how big the CTE problem was. A study out of Boston this week offers the closest thing yet to an answer, and Australian insurers will not like it.
Researchers at Boston University estimated that 74.5% of the 2,362 former NFL players who died between 2008 and 2023 had CTE, the degenerative brain disease linked to repeated head knocks. That is roughly 1,760 men. The paper, published on 28 October in JAMA Network Open, is the first to put a population-wide figure on the disease in a professional football code rather than a count drawn only from donated brains.
American football is not Australian rules or rugby league. They have helmets and padding. But the uncertainty the study chips away at is the same one underpinning exclusions, sublimits and nervous reserving across the local contact-sport market.
CTE can only be confirmed by examining brain tissue after death, and brains tend to be donated by families who already suspected something was wrong. That skews every brain-bank tally upwards.
The Boston team tackled the skew to try to get a clear view of the stats. Of 387 former NFL players examined through the UNITE brain bank, 362 (about 94%) had CTE, according to the university's summary of the study. The researchers then matched that against US National Death Index records for every player who died in the 16-year window, and found that players with cognitive decline were about five times more likely to have their brains donated. Correcting for that brought the estimate down to 74.5%. By comparison, fewer than 1% of brains in community brain banks show the disease.
The new figure is three times the floor the same research group published in August in The BMJ. That study found CTE in at least 24.5% of NFL players who died between 2016 and 2021, but only by treating every undonated brain as disease-free, and it put the theoretical ceiling at 97.7%. The new model lands much closer to the top of that range than the bottom.
"An estimated 75 percent prevalence among NFL players who died over a 16-year period is concerning," said lead author Julia Gallini, a PhD candidate in biostatistics at Boston University.
The authors stress the limits. The estimate covers players who died in one era, many of them long before modern concussion protocols, and it does not predict how many living or active players have the disease. Not every player with CTE pathology goes on to develop symptoms, particularly in early-stage cases, the university noted, although earlier research has found former NFL players develop dementia at four times the rate of the general population.
Read next: AFL players lose brain injury cover in insurance shift
The local market has been moving on CTE for some time, and mostly in one direction. From 1 May, Zurich removed total and permanent disability cover for concussion, CTE and other brain injuries from the group policy inside the AFL Players' Association superannuation fund. The insurer cited "widespread uncertainty" over the potential scale of CTE, and was reportedly the only carrier to bid in the trustee's tender.
The pull-back followed a jump in claims after Melbourne premiership player Angus Brayshaw was medically retired with concussion issues in 2023. Brayshaw remains in a public payout dispute with Zurich.
Meanwhile the class action led by former Geelong player Max Rooke, filed in the Supreme Court of Victoria in 2023, has grown to more than 100 former players. It names the AFL and a string of clubs, and covers players who allegedly suffered concussion injuries between 1985 and 2023.
Local pathology is pointing the same way as the US data. In September, data scientist Jennifer Cropley presented Australian Sports Brain Bank findings to the International Sports Concussion Conference in Melbourne showing CTE in 31 of 51 donated brains from rugby league players. Some were in their 20s, 30s and 40s, and they had played at every level from amateur to professional. Those figures carry exactly the donor bias the Boston model was built to correct, which is why actuaries will be watching to see whether anyone applies the same method here.
The pool of exposure is large, and it starts young. About 284,000 children aged 14 and under played organised Australian football outside school hours, along with 149,000 people aged 18 to 24, according to the latest AusPlay figures from the Australian Sports Commission. Most of that risk sits not with the AFL but on the community club, school and association policies that brokers place every season.
Law firm Barry Nilsson expects the claims landscape to develop rapidly, and to spread beyond elite sport to amateur athletes, local leagues and schools. The tools it suggests carriers consider include revised underwriting models, sublimits, tighter retroactive dates for CTE claims and specialist training for claims assessors.
A credible prevalence number can cut both ways. It can harden terms for clubs, schools and associations that keep the exposure. Or it can give carriers enough confidence in the size of the risk to keep writing it, which is precisely what was missing when Zurich walked away from the AFL players' TPD cover.
The pressure is not unique to Australia. In Canada, NFP's Sarah Hanson, broker to Rugby Canada, has said head-injury wordings vary widely and that carriers routinely carve out long-term neurological risk. "Specifically, neurodegenerative conditions can be excluded," she said.
For brokers placing sporting clubs and associations, the practical message is to get ahead of renewal. Expect underwriters to ask harder questions about concussion protocols, return-to-play rules, junior programs, waivers and head-injury claims history.
Read next: Concussion claims rising: Why brokers are the key to sports club survival
The US also offers a lesson in policy wording. On 1 October, a New York appeals court ruled in Discover Property & Casualty Co. v National Football League that each former player's concussion claim is at least one separate occurrence, because the generic grouping language in the league's old liability policies did not name repeated head impacts as the triggering event.
That left Fairfax unit TIG's primary layer exposed player by player, while a Zurich excess layer that only kicked in above US$51 million (about A$73 million) in underlying limits owed nothing.
Australian courts apply their own aggregation principles, so the ruling is not a direct precedent here. But the drafting point travels. A policy that wants CTE claims treated as one event, or as many, needs to say so plainly.
The bigger shift may come from the lab rather than the courtroom. Researchers want to diagnose CTE in living people, and Barry Nilsson has flagged that claims could rise significantly if that happens. For now, the disease can only be confirmed at autopsy. Once it can be confirmed in a clinic, a claim that today surfaces decades late could arrive while the claimant is still working.
"We recognize these findings may cause concern for the community of living players and their families," Tripodis said in the university's statement. For Australian insurers writing contact sport, from the AFL down to the under-12s, the question is no longer whether that exposure is real. It is how to price it.
Read next: Insurers "nervous" over long-tail concussion claims
