[Skip to Navigation]
Comment & Response
February 25, 2019

Variation in National Survey Estimates and Youth Traumatic Brain Injury—Where Does the Truth Lie?

Author Affiliations
  • 1School of Nursing, University of Michigan, Ann Arbor, Michigan
JAMA Pediatr. 2019;173(4):399. doi:10.1001/jamapediatrics.2019.0001

To the Editor Haarbauer-Krupa et al1 present data from the National Survey of Children’s Health (2011-2012) to estimate the lifetime prevalence of parent-reported traumatic brain injury in a national sample of US children. The NSCH is a cross-sectional telephone survey (response rate of 23%) of adults. These adult respondents were asked the following question regarding their selected child between ages 0 and 17 years: “Has a doctor or other health care provider ever told you that [your child] had…a brain injury or concussion?”1 Based on this question, Haarbauer-Krupa et al1 found that parent-reported traumatic brain injury among children aged 0 through 17 years was 2.5% in the United States (5.9% among children aged 15 to 17 years). While the authors present needed epidemiologic information on traumatic brain injury/concussion rates across a critical developmental period, it must be highlighted that these estimates in large-scale surveys are sensitive to several biases that include how the question was worded (eg, lifetime vs past year), who responded to the survey (eg, adult vs child), and how the survey was delivered (eg, telephone vs pen and paper). To illustrate the sensitivity of these estimates to survey bias, 3 national surveys assessing head injuries among children and adolescents are presented: the National Health Interview Survey (2016 child sample), the Monitoring the Future Survey (MTF; 2016), and the Youth Risk Behavior Survey (2017). First, the National Health Interview Survey is a national household survey that uses computer-assisted personal interviewing.2 Parents answer the questions for their children. The survey question was asked in the following manner: “IN {fill: his/her} LIFETIME, has {fill1: S.C. name} EVER had a significant head injury or concussion?” Among children aged 3 through 17 years (n = 9247), the prevalence of lifetime concussion was 7.0%; among children aged 14 through 17 years (n = 2771), the prevalence of lifetime concussion was 11.1%. Second, the MTF is a school-based national survey of 8th, 10th, and 12th graders (ages 13 to 18 years; typical response rate between 80% and 90%) in the United States and uses self-administered paper-pencil questionnaires in classrooms.3,4 The question in the 2016 MTF survey was asked in the following manner: “Have you ever had a head injury that was diagnosed as a concussion?” Based on MTF data (n = 13 088), 19.5% of adolescents indicated at least 1 diagnosed concussion during their lifetime. Third, the Youth Risk Behavior Survey is a national survey of high school students (ages 14-18 years; typical response rate between 60% and 70%) that uses similar methods as the MTF.5 The question in the Youth Risk Behavior Survey 2017 was asked in the following manner: “During the past 12 months, how many times did you have a concussion from playing a sport or being physically active?” Among the sample of adolescent respondents (n = 14 765), 15.1% indicated at least 1 concussion during the past year (note that lifetime prevalence of head injuries could not be obtained because the measure assessed past 12 months). As seen here, 4 national studies provide very different estimates, and the substantial variation in the estimates is likely owing to how the question is asked, who is responding to the questions, and the mode of survey delivery. Clinicians and researchers need to be aware of this sensitivity and be cautious of these measurement issues in large-scale epidemiologic surveys.