Risk factors for youth brain injury and post-injury behavioral change in the ABCD Study cohort
Open Access
- Author:
- Mongilio, Jessica
- Graduate Program:
- Criminology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- May 08, 2024
- Committee Members:
- Pamela Wilcox, Professor in Charge/Director of Graduate Studies
Lisa Kopp, Outside Unit Member
Jennifer Maggs, Outside Field Member
Kyle Thomas, Special Member
Derek Kreager, Major Field Member
Jeremy Staff, Chair & Dissertation Advisor - Keywords:
- adolescence
delinquency
alcohol use
TBI
head injury - Abstract:
- In recent years, an increasing amount of both scholarly and public attention has focused on traumatic brain injury (TBI) and its impacts on behavior. Prior work has established evidence of associations between TBI and problem behaviors (e.g., delinquency, externalizing behavior, substance use) among adolescent populations, but recent reviews and meta-analyses have cited inconsistencies as undermining our understanding of the effects of TBI among youth. Several threats to validity limit prior research, such as selection bias through sampling, temporal ordering of covariates and outcomes, and insufficient comparison groups. Critically, prior studies of TBI have tended to neglect the importance of shared risk factors for both TBI and behavioral outcomes of interest, and it is speculated that these pre-injury differences between those who do and do not experience TBI might render the focal associations spurious. For instance, youth who are more impulsive, in general, might be more likely both to experience a TBI and to initiate alcohol use earlier in adolescence than their uninjured peers. Without accounting for threats to validity such as these, previously reported estimates of the effect of TBI may be upwardly biased. The present dissertation utilizes the Adolescent Brain Cognitive Development (ABCD) Study to identify shared risk factors for TBI and externalizing behavior and to estimate the effects of TBI at age 11 on early adolescent externalizing behavior and alcohol use onset at age 12, after accounting for these confounders, among both boys and girls. Group differences between those with and without TBI are first explored, and shared risk factors are identified through regressions of TBI and the externalizing behaviors of interest. Then, KHB decomposition is used to demonstrate how previous estimates of associations between TBI and externalizing behaviors can be explained by the inclusion of these shared risk factors in models. Finally, coarsened exact matching (CEM) is used to approximate a random block experimental design, such that youth with TBI are only compared to their uninjured, but similarly at-risk, peers. Effects of TBI on externalizing behavior and alcohol use onset are then estimated for both boys and girls. Broadly, the present findings are indicative of the need to account for the nonrandom distribution of TBI, as the inclusion of the identified shared risk factors—pre-injury externalizing behavior, impulsivity, peer substance use, and prior TBI—dramatically change estimates. For girls especially, the inclusion of these shared risk factors accounts for about half to three-quarters of unadjusted associations between TBI and externalizing behaviors. In matched models, even when compared to a strict control group that should be similar on levels of pre-injury behavior, boys who experience TBI at age 11 have increased externalizing behavior at age 12. However, TBI does not appear to increase the risk of alcohol use onset among boys. In matched models, TBI among girls does increase rule-breaking behavior, but decreases subsequent aggressive behavior. For girls, TBI also greatly increases the risk of alcohol use initiation in the year following injury. The present dissertation contributes to our understanding of youth TBI and its impacts on subsequent behaviors and is suggestive of TBI as a risk factor for earlier onset and potentially longer-lasting trajectories of externalizing behavior, substance use, and delinquency. Implications for criminological theory—namely, the integration of biological and social risk factors for crime—and for future studies of TBI are discussed.