Parental incarceration and adolescent externalizing behaviors and substance use: Reconciling discrepant findings

Mass incarceration in the United States has led to millions of parents of minor children being incarcerated, many of whom are Black Americans. The consequences of parental incarceration have rippling effects across generations and result in poor social, economic, and health outcomes among children of incarcerated parents throughout the lifecourse. Research on the effects of parental incarceration on children has focused primarily on externalizing behaviors as a means to control crime and the intergenerational transmission of crime. However, such perspectives are rooted in racism and negates the role of structural racism in maintaining social and economic policies that keep Black American families and communities disenfranchised and under the control of the majority. Theoretical perspectives suggest that parental incarceration increases the risk of adolescent substance use and externalizing behaviors, while others suggest that parental incarceration decreases the risk of adolescent substance use and externalizing behaviors. However, when examining the literature, there are discrepant findings regarding the direction and magnitude of the association between parental incarceration and adolescent externalizing behaviors. This dissertation focuses on explaining reasons for and reconciling these discrepant findings, and exploring and expanding the literature on parental incarceration and adolescent substance use which has not been adequately synthesized to date. This dissertation consists of five chapters. The first chapter is an introduction to the dissertation. Chapter 2 is a systematic review of the literature on parental incarceration and adolescent cigarette, alcohol, and cannabis use. Chapter 3 is an empirical study that assessed the association between parental incarceration and past year adolescent alcohol and cannabis use, and externalizing behaviors using data from the Fragile Families and Child Well-Being Study (Fragile Families). Fragile Families is well-suited for assessing the association between parental incarceration and adolescent substance use and externalizing behaviors because it includes a diverse sample of adolescents and their parents, various measures of parental incarceration over time, and a wealth of information on adolescent outcomes. This diverse sample also makes it possible to assess effect modification by race/ethnicity, which other studies may have been underpowered to assess considering the lack of racially and ethnically diverse samples. Chapter 4 used data from Fragile Families and The National Longitudinal Study of Adolescent to Adult Health (Add Health) to empirically examine if discrepant findings existed across these datasets regarding the association between parental incarceration and adolescent cannabis use. Prior studies have shown discrepant findings regarding parental incarceration and offspring externalizing behaviors across these datasets; however, adolescent cannabis use has not been assessed in Fragile Families to identify potential discrepancies. This dissertation concludes with Chapter 5, which provides a summary of the key findings, and discusses the public health impact of this research and future directions. The systematic review provided evidence that there is a positive association between parental incarceration and adolescent cigarette and cannabis use. However, more research is needed to determine if these associations are purely correlational or potentially causal. Findings from the review did not suggest that parental incarceration was associated with adolescent alcohol use. Two out of three studies included in the review assessed binge drinking/episodes of heavy drinking, which may have impacted results. In contrast, Chapter 3 showed support for a positive association between parental incarceration and adolescent alcohol use in the past year (any vs. none; relative risk [RR]=1.54, 95% CI=1.17-2.03) using data from Fragile Families. Chapter 3 showed that both maternal and paternal incarceration were associated with adolescent alcohol and cannabis use (RR=1.25, 95% CI=1.01-1.55), and externalizing behaviors (delinquency scale: incidence rate ratio [IRR]=1.18, 95% CI=1.01-1.36 and CBCL: IRR=1.14, 95% CI=1.02-1.28). Findings also suggested that the associations between lifetime parental incarceration and adolescent alcohol and cannabis use, and externalizing behaviors were different by adolescent race; however, data were not sufficient to assess interaction effects. In Chapter 4, the association between parental incarceration and adolescent cannabis use was shown to be different across Fragile Families and Add Health (association between maternal incarceration during adolescence and adolescent past month cannabis use: RRFragile Families=2.09, 95% CI=1.36,3.21 vs. RRnon-post-stratification-weighted Add Health=1.27, 95% CI=0.77-2.10, magnitude of difference=39.2%), which was expected based on prior studies assessing parental incarceration and adolescent externalizing behaviors across these studies. Generally, findings were in the positive direction, but the magnitude of associations were inconsistent. Reweighting Add Health to have the sample distribution of adolescent gender, age, and race as Fragile Families was hypothesized to lead to congruent findings across these studies. Nonetheless, reweighting Add Health resulted in measures of association being further away from those in the non-poost-stratification-weighted Add Health sample; thereby further away from the Fragile Families sample (association between lifetime paternal incarceration and adolescent past month cannabis use: RRFragile Families=1.34 vs. RRnon-post-stratification-weighted Add Health=1.44 vs. RRpost-stratification-weighted Add Health=1.53). In conclusion, results from these studies indicate that there is a positive association between parental incarceration and adolescent cannabis use, and externalizing disorder. However, findings are inconsistent regarding the effect of parental incarceration on adolescent alcohol use, and regarding the association between parental incarceration and adolescent outcomes across Fragile Families and Add Health. These studies attempted to reconcile discrepant findings and provide a novel method for reconciling discrepant findings across different exposures and outcomes in future studies