Field studies have evaluated the accuracy of sensors to measure head impact exposure using video analysis, but few have studied false negatives. Therefore, the aim of the current study was to investigate the proportion of potential false negatives in high school soccer head impact data. High school athletes (23 females and 31 males) wore headband-mounted Smart Impact Monitor-G impact sensors during competitive soccer games. Video footage from 41 varsity games was analyzed by 2 independent reviewers to identify head contact events, which were defined as visually observed contact to the head. Of the 1991 video-identified head contact events for which sensors were functioning and worn by the players, 1094 (55%) were recorded by the sensors. For female players, 45% of video-identified head contact events were recorded by the sensor compared with 59% for male players. For both females and males, sensitivity varied by impact mechanism. By quantifying the proportion of potential false negatives, the sensitivity of a sensor can be characterized, which can inform the interpretation of previous studies and the design of future studies using head impact sensors. Owing to the difficulty in obtaining ground truth labels of head impacts, video review should be considered a complementary tool to head impact sensors.
Keywords: head injury; kinematics; sport.