Prevalence and correlates of concussion in children: Data from the Adolescent Brain Cognitive Development study

Document Type

Journal Article

Publication Date









Concussion; mTBI; Neurocognition; Pediatrics


© 2020 Introduction: Concussions are one of the most common causes for emergency room use in the United States (US) among youth and adolescents; however, prevalence data on concussion in this population are inconsistent. A growing body of literature has explored associations of a range of variables with pediatric concussion, but they have not been explored simultaneously in a well-powered sample in the US. The present study aimed to present lifetime concussion prevalence, evaluate demographic, psychological, and cognitive correlates of concussion, and assess for differences across these variables based on age of first concussion in a large sample of US children. Methods: We analyzed the Adolescent Brain Cognitive Development (ABCD) sample, which monitors biopsychosocial development in 11,875 children at 21 sites across the US between ages 9 and 10. Along with presenting rates of concussion, we also evaluated the association of demographics, sleep disturbance, cognitive functioning, and externalizing and internalizing symptoms with concussion history using backwards binary logistic regression. We further conducted univariate comparisons of all variables between those who experienced their first concussion before and after age 5. Significance was based on α =.02, with Benjamini-Hochberg FDR adjustments for multiple comparisons. Results: We found approximately 4% of the sample had experienced a concussion, and significant correlates of experiencing a concussion were male sex, increased family income, and higher somatic symptoms after FDR correction. Symptoms of ADHD were also noted as nominally significant. No differences based on age of first concussion were found. Discussion: Our analyses provided updated prevalence estimates of pediatric concussion in the US that aligns with many hospital records-based studies. Our findings largely mirrored those in the literature with the exception of somatic symptoms. Limitations of findings and implications of individual findings are discussed.

This document is currently not available here.