The vast majority of existing research on outcomes following concussion is androcentric with women notably understudied in all settings where concussion commonly occurs, including sport, military, and civilian domains. Clinical care for concussed women is, by extension, far less evidenced-based as it is largely extrapolated from male-dominated studies. We therefore sought to center concussion outcomes of women in big data sources in this investigation to capitalize on both sample size and breadth of setting in which concussion occurs.

We identified all studies with publicly released data as of 4/7/21 that included both male and female adults, enough information to determine severity of injury consistent with concussion/mild traumatic brain injury (TBI), a measure of postconcussive symptoms (PCS), and objective measures of neurocognitive functioning from the Federal Interagency Traumatic Brain Injury Research (FITBIR) Informatics System. FITBIR is a collaborative effort of the National Institutes of Health (NIH) and the Department of Defense (DoD), developed to share data across the entire TBI research community. This resulted in inclusion of six studies with a total of 9370 participants, 32% female. PCS data was drawn from the Neurobehavioral Symptom Inventory (NSI), the Rivermead Post-Concussion Symptoms Questionnaire (RPSQ), and the Sport Concussion Assessment Tool - 3rd Edition (SCAT 3), specifically the 22-item participant symptom evaluation rated on a 7-point scale. The questionnaires were harmonized and the following symptom domains emerged: Somatic, Cognitive, and Affective. Data were analyzed using linear mixed-effects models.

We found a small to medium sized significant effect of sex, with women reporting higher symptoms overall. Mean symptom endorsement scores were higher overall on the NSI in comparison to the RPSQ and SCAT (ps < .001). Follow-up analyses revealed sex differences were largest for cognitive symptoms, followed by somatic symptoms, then affective symptoms. We also found significant main effects of population (military>sport/civilian, p = .003) and sex (women>males, p < .001) on the overall composite, as well as a significant population-by-sex interaction such that female service members/veterans endorsed the highest rates of symptoms (p < .001). Similar patterns of significance and effect sizes were observed for the somatic and affective composites. Cognitive symptom composites showed a similar pattern, but with smaller effect sizes overall. Racial and ethnic diversity was also limited in the sample.

In one of the largest samples of women to date, we found a small to medium effect of sex on symptom reporting such that women reported higher levels of postconcussive symptoms than males. Notably, however, the women in military/Veteran samples endorsed the highest levels of symptoms. Despite using a large publicly available dataset to maximize the representation of women, the current sample was still predominantly male and racial and ethnic diversity among the sample was not consistent with expected broader population demographics. Dramatically more concerted efforts need to be made to engage women in all spheres of concussion research (military, civilian, and sport). Strategies to be more inclusive in concussion research will be highlighted.