Concussions and suicide

Speidel raises some skeptical points about our methods1 because we lack clinical details for the 235 110 patients included in our study.

We agree that more research is justified because concussions can vary in symptoms, severity, context, construal and recovery. We also agree that ascertainment of suicide is challenging, reflects an uncertain number of false-negative results and causes most studies to underestimate the risk of suicide.

Speidel invites us to consider the article by Baron and associates.2 We agree that this study is widely misquoted and appreciate the opportunity to correct several common misunderstandings. First, the study examined cardiovascular mortality and was not powered to investigate self-harm events. Second, the article does not discuss suicide and relegates intentional self-harm to a single line in one table. Third, the overall mortality from all other causes of violent death was distinctly low, as would be expected among wealthy, retired professional athletes. Fourth, the study ended follow-up in 2007 and excluded at least nine former NFL players who committed suicide in the past five years alone (Jovan Belcher, Kurt Crain, Mike Current, Dave Duerson, Ray Easterling, Paul Oliver, Lawrence Phillips, Adrian Robinson and Junior Seau).

The statement from Speidel has one more caveat, since almost half of those included by Baron and associates were African-American men, a group historically that had a low suicide rate during the study enrolment interval (1959–1988). That may explain why the authors never stated that NFL players “experienced concussions at a rate one-sixth that of the general US population.” Otherwise, the authors might imply excess suicide deaths in their sample.