Share this article on:

Elevated Mortality Associated with Neurodegenerative Disease Reported Among NFL Football Players — What the Study Says, What it Does Not

Samson, Kurt

doi: 10.1097/01.NT.0000421883.59885.46
Back to Top | Article Outline


THE NEUROLOGY TODAY archive of articles on concussion, football and neurological disease is available here: http://bit

THE NEUROLOGY TODAY archive of articles on concussion, football and neurological disease is available here: http://bit

Death related to all neurological causes among NFL football players was three times higher than the general US population, and was four times higher for two of the major neurodegenerative subcategories: Alzheimer's disease and amyotrophic lateral sclerosis.

Mortality associated with neurological causes among professional football players who played in the NFL between 1959 and 1988 appears to be significantly elevated over that of the general population, according to a Sept. 5 online report in Neurology.

Death related to all neurological causes among the football players was three times higher than the general US population, and was four times higher for two of the major neurodegenerative subcategories — Alzheimer's disease (AD) and amyotrophic lateral sclerosis (ALS), according to the study.

Moreover, among those in the football playing cohort who died from neurological causes, mortality was higher among those who played in “speed” positions — positions that required players to accelerate prior to being hit or tackled — than those who played in “non-speed” positions, such as lineman, that are more stationary.

However, both the study authors and experts who reviewed the report note that for at least two reasons findings from the study should not be regarded as conclusive: the total number of deaths from neurological causes in the football playing cohort was small, rendering statistical estimates for mortality ratios imprecise; and the data used for the study was lacking in information about concussions, injuries, or other non-football-related risk factors that may have contributed to neurological death at the individual level.

Still, the results are striking and are consistent with some previous research that has shown an increased risk of neurodegenerative disease among football players, and among individuals who experience head trauma or repeated concussions.

Lead author Everett J. Lehman, an epidemiologist with the Cincinnati office of the National Institute for Occupational Safety and Health, emphasized the findings related to “all neurodegenerative causes” because the aggregate statistic may point to chronic traumatic encephalopathy (CTE) — recently identified as a pathologically distinct neurodegenerative condition affecting individuals who have experienced multiple concussions — as an underlying cause of neurological death among the players.

“We would suggest that the analytical results for ‘all neurodegenerative causes of death’ is a more significant finding than the results for the individual causes (ALS, Parkinson's disease, AD) because other studies have shown that there may be a common etiology among one or more of the neurodegenerative diseases though they may take different developmental paths due to unique environmental exposures or genetic susceptibility factors,” Lehman told Neurology Today. “The overall result for all neurodegenerative causes was slightly less than for the individual causes, but was still more than three times the expected.”

The study authors drew data from the records of a 1994 report for Occupational Safety and Health on mortality among NFL players. Vital status was ascertained on a cohort of 3,439 NFL players identified by a pension fund database of vested players with at least five credited playing seasons between 1959 and 1988.



Players were matched to the National Death Index (NDI) beginning in 1979 (when the NDI began) with follow-up through 2007. The NDI provided underlying and contributing causes of death, coded to the ICD revision in effect at the time of death. (Death certificates were obtained from state vital statistics offices for deaths occurring prior to 1979 and for any information related to deaths before or after that date that was not provided by the NDI.)

For comparison, researchers used US male mortality rates (1960–2007) for 119 causes of death.

To examine possible neurologic mortality differences from high acceleration impact, Lehman and colleagues stratified the players into two categories based on position: “speed” positions (quarterback, running back, halfback, fullback, wide receiver, tight end, defensive back, safety, and linebacker) and “non-speed” positions (all defensive and offensive linemen).

Out of a total of 334 deaths among the football playing cohort, 17 were recorded with neurological disorders as either the underlying or a contributing cause: seven of these were Alzheimer's, and seven were ALS for standard mortality ratios of 3.86 and 4.31 respectively. Combining the two causes meant that football players had a four times greater likelihood of having one of those disorders as a contributing cause of death than did the general population.

Of the 17 deaths, 14 were among the “speed position” players. They had a standard mortality ratio of 4.74.

Back to Top | Article Outline


Commenting on the study, neurologist Walter Rocca, MD, MPH, professor of neurology and epidemiology at the Mayo Clinic in Rochester, MN, said: “Sports that involve head impact or concussion are a preventable factor and should be more strictly regulated. In addition, more studies are needed to understand the many concurrent factors that may contribute to the development of neurodegenerative disorders in subjects exposed to head trauma.”



“The strengths of the current study are the cohort design, the use of historical data free from recall biases, and the ability to compare players in speed positions with players in non-speed positions,” Dr. Rocca told Neurology Today. “The weaknesses are the small number of subjects who died with a neurodegenerative disease listed on the death certificate, the use of mortality as a surrogate measure for incidence, the inability to analyze specific positions among the speed players, and the lack of information about injuries or concussions or other risk factors at the individual level.”

Neurologist Carmel Armon, MD, chief of the division of neurology at Baystate Medical Center and professor of neurology at Tufts University School of Medicine, emphasized the low number of deaths in the cohort and the wide confidence intervals (reflecting a relative lack of precision about risk estimates). He cautioned that the estimated standardized mortality ratios (SMRs) for neurological deaths among the players may be artificially high.

One reason, he suggests, is that neurological deaths will stand out as apparently higher than normal as other possible causes of death diminish among a cohort of athletes who are less likely to smoke, be overweight or sedentary, and who are likely to have lived relatively healthy lifestyles.

“Standardized mortality ratios were calculated adjusting for race, age, and calendar year, but this adjustment does not take into consideration the increase in age-specific incidence of neurodegenerative disease that is expected to result from loss of competing causes of death,” Dr. Armon told Neurology Today. “The overall SMR for these players was 0.53 [for all causes of death]. This means that overall they experienced about half the mortality of US males. This was due primarily to reduction in mortality from the major causes of death — cardiovascular diseases and cancers, even as deaths from other causes were also reduced.”

And he drew attention to the fact that the “speed players” in the cohort who had a higher frequency of neurological death had a lower overall mortality than the non-speed players (8 percent versus 13 percent respectively). “This is consistent with the possibility that a loss of competing causes of mortality may explain the greater presence of neurodegenerative diseases,” Dr. Armon said.

A second reason why the SMRs for neurological death in football players may be artificially high is simply that any specific cause of death is more likely to be identified in a defined cohort whose records are monitored than in the general population. “It is difficult to adjust for the likelihood that a diagnosis of neurodegenerative disease is more likely to be made, and find its way to the death certificate, in a motivated and tracked cohort of individuals, than in the general population that serves as a source for the expected reference numbers,” he said.



Finally, he referenced a population-based study from Rochester, MN, published in the April 2012 edition of Mayo Clinic Proceedings that found no increased risk of neurodegenerative diseases, including ALS, in a cohort of high school footballers who played between 1946 and 1956.

“This would suggest that even if an increased risk of one or other neurodegenerative diseases is found in the NFL cohort, it may be appropriate to consider risk factors in addition to head trauma than those related to the game itself,” he said.

Back to Top | Article Outline




  • Overall player mortality compared with that of the US population was reduced — standardized mortality ratio (SMR) 0.53, 95% confidence interval [CI] 0.48–0.59.
  • Neurodegenerative mortality was increased using both underlying cause of death rate files (SMR 2.83, 95% CI 1.36–5.21) and multiple cause of death (MCOD) rate files (SMR 3.26, 95% CI 1.90–5.22).
  • Of the neurodegenerative causes, results were elevated (using MCOD rates) for both ALS (SMR 4.31, 95% CI 1.73–8.87) and AD (SMR 3.86, 95% CI 1.55–7.95).
  • In internal analysis (using MCOD rates), higher neurodegenerative mortality was observed among players in speed positions compared with players in nonspeed positions (SRR 3.29, 95% CI 0.92–11.7).
Back to Top | Article Outline


• Lehman EJ, Hein MJ, Gersic CM, et al. Neurodegenerative causes of death among retired National Football League players. Neurology 2012; E-pub 2012 Sept. 5.
    • Savica R, Parisi JE, Ahlskog JE, et al. High school football and risk of neurodegeneration: A community-based study. Mayo Clin Proc 2012;87(4):335–40.
      See Neurology Today's archive of articles about chronic traumatic encephalopathy:
        ©2012 American Academy of Neurology