Journal Logo

EP-05 Epidemiology and Biostatistics

Comparison Of Mortality Rates Among Elite American Football Athletes From 1965 To 1986


Wilber, Stephanie; Smoliga, James M. FACSM; Wilber, Z. Taggart

Author Information
Medicine & Science in Sports & Exercise: August 2021 - Volume 53 - Issue 8S - p 218
doi: 10.1249/01.mss.0000761588.57670.82
  • Free

Over the past decades, evidence has suggested there are increased health risks associated with playing American football. These health risks may be variable by position, due to repeated head impact exposure.

PURPOSE: To analyze potential differences in life expectancy between National Football League (NFL) players named to All-Pro teams from 1965 to 1986.

METHODS: Football players were identified as “elite” if they were named to an All-Pro team between 1965 and 1986. was used to identify these athletes and collect preliminary information. Additional information was gathered using personal interviews, news articles, blogs and forums. These data included the year that the athlete was first named to an All-Pro team, his position, team, date of birth, date of death (if applicable), the length of his NFL career, and reported height and weight. Kaplan-Meier survival analysis was used to compute the median lifespan by position and Cox regression, using forward stepwise selection, was used to determine if any of the covariates collected, including position, influenced lifespan. Special teams’ positions were excluded due to low sample size (n = 23).

RESULTS: 746 elite male NFL players were identified and included. Of those, 171 were deceased as of May 1, 2020 with a median age of death of 67y. Survival analysis revealed the median life expectancy was 84y [SE = 1y; 95% confidence interval 83-86y], with offensive linemen having the shortest median lifespan (82y; 95%CI = 79-85). Cox regression revealed BMI [Exp(B) = 1.12, 95%CI = 1.06-1.19; p < 0.001] and career duration [Exp(B) = 0.943, 95%CI = 0.89-1.00; p = 0.033] significantly influenced mortality. Position was not a predictor of lifespan in the stepwise model, but was a significant predictor of lifespan if entered into the model itself (p < 0.001), with linemen having shorter lifespans than skill players.

CONCLUSIONS: These findings suggest that BMI during one’s football career may be predictive of lifespan among elite NFL players. Athletes with longer NFL careers may live longer despite potentially increased exposure to repeated head impact. This paradox may be due to long careers being facilitated by the absence of major injuries and resulting in greater socioeconomic status, which may be associated with superior health post-retirement.

Copyright © 2021 by the American College of Sports Medicine