Physical Abuse Safety Concerns and Substance Use Among Child Welfare Adolescents

Background: Over the past decade, researchers have moved the field of adolescent substance use forward by identifying a longitudinal relationship between physical abuse and later substance use in child welfare samples. However, previous research has not consistently considered the potential roles of initial levels of substance use and/or later physical abuse in this relationship. Such designs would clarify whether there are in fact delayed or long-term effects of physical abuse on substance use, whether later substance use is being overestimated as a result of more recent physical abuse, or whether physical abuse has a relatively short-term effect on substance use but then substance use persists over time. The current study advances knowledge about the development of substance use by examining the longitudinal relationship between physical abuse and substance use frequency among youth who were investigated by Child Protective Services (CPS) for maltreatment and remained in their homes. Additionally, this study examined initial substance use and later physical abuse as potential paths connecting this relationship.

Methods: A sample of 654 adolescents aged 11–16 at baseline (investigation; T1) and 12-17 at wave 3 (18 months) and in in-home care was drawn from the National Survey of Child and Adolescent Well-Being (NSCAW-I), a nationally representative study of children investigated by the child welfare system. At T1 and T2 youths self-reported past-year physical abuse (any severe or very severe assault on the Parent-Child Conflicts Tactic Scale) and past 30-day substance use (i.e. overall frequency of alcohol, marijuana, hard drug, inhalant, and nonmedical prescription drug use). Control variables included child’s gender (girl, boy), age, ethnicity (Hispanic, Non-Hispanic), race (Black, White, Other), and caregiver substance dependence. Path analysis was conducted using MPLUS v. 7.11. Bias-corrected bootstrap confidence intervals were calculated for all parameter estimates and indirect effects.

Results: Physical abuse reported at T1 was significantly related to substance use at T2 (B = .439, 95% CI: .035, .834, p = .033). However, this relationship was attenuated by T1 substance use and T2 physical abuse. Specifically, past-year physical abuse reported at T1 predicted past 30-day substance use reported at T1, which in turn predicted past 30-day substance use at T2 (Standardized Indirect Effect = .147, 95% CI: .053, .241, p = .002). Past-year physical abuse at T1 predicted past-year physical abuse at T2, which in turn predicted past 30-day substance use at T2 (Standardized Indirect Effect = .067, 95% CI: .011, .123, p = .019). R2 indicated that the model explained 19.0% of the variance in T2 substance use frequency, 5.7% of the variance in T2 physical abuse, and 10.5% of the variance in T1 substance use frequency.

Conclusions and Implications: The findings highlight the roles of ongoing physical abuse and persistent substance use in the development of substance use behaviors after CPS investigation among adolescents remaining in their homes. Results also highlight the importance of substance use assessment and intervention at the point of CPS investigation to prevent recurrent substance use.