Concussion Risk from Helmeted Sports; A Re-examination of Data and Methods

Prior authors have attempted to quantify concussion risk in the general population based on analysis of head impact data gathered from studies of helmeted football players. In the present study potential sources of bias and methodological error in prior concussion risk studies is discussed, including the use of multiple impacts to a single player as independent events, and the failure to describe statistical scatter among the data. A re-analysis of the data is presented in which the highest single impact sustained by each player was matched to concussion presence or absence, resulting in a series of concussion risk curves based on peak acceleration, head injury criterion (HIC), and angular acceleration, with associated 95% confidence intervals. These data have limited utility for forensic analysis of concussion risk in the general population.