In Reply We agree with Drs Smith and Siddiqi that the estimated mortality hazard may be clinically meaningful, but because it is based on a small number of deaths, further follow-up is necessary. We would encourage this interpretation even if the P value had fallen below .05.
Both letters question our choice of comparison group, but this choice should be viewed in the context of the prior literature, which used general population controls. Professional football players have higher levels of physical fitness and tend to abstain from unhealthy behaviors compared with individuals in the general population.1,2 The skill and ability to succeed in professional football are likely negatively correlated with baseline mortality risk. We believed that this healthy worker bias3 was substantial,4 likely explaining why prior studies demonstrated substantially lower mortality risks of NFL players relative to general population controls.5 We focused on replacement players because we believed these athletes—by virtue of being on the margin of making an NFL roster—had baseline risk factors that were more similar to NFL players than general population controls. We could not test this assumption because this was a historical case study and collecting baseline health data was infeasible. Nevertheless, our findings were consistent with this contention. We found higher estimated mortality risk among career NFL players, opposite to the conclusion of prior studies.
Venkataramani AS, Gandhavadi MB, Jena AB. Professional Football Participation and Mortality—Reply. JAMA. 2018;320(1):92–93. doi:10.1001/jama.2018.6717
Browse and subscribe to JAMA Network podcasts!
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: