WITHDRAWN: Impact of Household Incarceration in Childhood on Substance Use Behaviors Among Hispanic Emerging Adults

Introduction: A burgeoning body of evidence implicates familial involvement in the criminal justice system as an important correlate of numerous health compromising outcomes for youth. Hispanic children are 2.5 times more likely to have a family member incarcerated compared to their non-Hispanic peers. We sought to evaluate the long-term effects of household incarceration on substance use outcomes among Hispanic emerging adults in California, the state with the highest concentration of Hispanics. We hypothesized that emerging adults exposed to household incarceration during childhood (before age 18) would report higher rates of past 30-day a) cigarette; b) alcohol; c) binge drinking; and c) marijuana use; and d) negative substance use consequences compared to non-exposed participants.

Methods: Longitudinal analysis of data from Project RED, a longitudinal study of acculturation process and health among Hispanics in California, was performed using coarsened exact matching to balance the data of the treatment and control groups under analysis—those exposed and non-exposed to household incarceration, respectively. Adolescents were matched on gender, SES, US or foreign born, family composition, alcohol, cigarette and marijuana use in 10^th grade and ACE. Separate multivariable negative binomial regression models assessed the association between household incarceration (primary explanatory variable) and past 30-day cigarette, alcohol and marijuana use frequency; and past 30-day substance use consequences.

Results: Household incarceration prior to age 18 was defined as the primary explanatory variable (treatment), and the measure of imbalance between exposed and non-exposed groups was GI=.46 before matching on covariates and GI=6.43 x -10^-6 after matching, nearly perfect balance. There was no significant difference between the exposed and non-exposed groups in terms of past 30-day cigarette use (IRR=1.21, 95% CI=.97, 1.49) or alcohol use (IRR=1.07, 95% CI=.96-1.20). However, there were significant long-term effects of household incarceration on the frequency of past 30-day binge drinking (IRR 1.23, 95% CI=1.07-1.68), marijuana use (IRR=1.48, 95% CI=1.16-2.01), and the number of negative substance use consequences experienced (IRR=1.41; 95% CI=1.08-1.64).

Conclusions: The present study demonstrates that the incarceration of a household member during childhood may be an antecedent of substance use outcomes that needs further consideration, especially among Hispanic emerging adults. Future research should continue to investigate the mechanisms that promote or inhibit resilience for children affected by incarceration and what policies and services are needed to effectively buffer the negative consequences of incarceration for children and their families in order to promote a healthy and successful transition into adulthood.