SI Vault
 
Dead Wrong
David Epstein
May 21, 2012
Two studies refute reports in the media about former NFL players' life expectancy
Decrease font Decrease font
Enlarge font Enlarge font
May 21, 2012

Dead Wrong

Two studies refute reports in the media about former NFL players' life expectancy

View CoverRead All Articles

With Junior Seau's death putting the spotlight back on the health of retired NFL players, journalists and radio hosts last week recycled a scary statistic. As ABC News reported it: "The average life expectancy of a retired football player is 58, according to the NFL Players Association ... a stark contrast to the average American man's life expectancy of 75."

The trouble is that this oft-repeated number (sometimes 58, other times 55) is false. That bit of gridiron legend has been around since at least 1990, when the NFLPA asked the National Institute for Occupational Safety and Health to investigate the statistic's accuracy. NIOSH tracked nearly 7,000 players and issued a report, in '94, concluding that NFL retirees were dying at about half the rate of their American male peers. In other words NFL players, in general, live longer.

In January, a second NIOSH report again found that NFL players were outliving their non-NFL cohort, partly due to lower cancer rates that could be the result of their lower rate of smoking. NIOSH also found that former players were at a decreased risk of dying from heart disease (with the exception of players who had a high body mass index). And the rate of suicide among NFL vets was 59% lower than in the general population. Overall, for the 3,439 ex--NFL players in the more recent study, NIOSH projected 625 deaths, using nonathlete mortality rates, but observed only 334 (chart).

The NFLPA tells SI that it did not provide the number used in the ABC News report, so it is difficult to know why the age 58 figure is still being disseminated. Certainly the current public discussion about player health is an important one to have. But for the good of those same players, it should remain grounded in data.

1