Longitudinal Patterns of Child Maltreatment Experiences and Adolescent Substance Use

Background and Purpose: Adolescent substance use is a significant public health concern that can lead to subsequent development of substance use disorders over the life course. Child maltreatment is a widely recognized, salient risk factor for adolescent substance use. Despite robust research on maltreatment and adolescent substance use, prior work has often failed to consider the complex nature of maltreatment, such as the type, timing, and chronicity of maltreatment. The present study addressed two research questions: (1) Are there heterogeneity in longitudinal patterns (e.g., timing: infancy/toddlerhood, early childhood, school age, adolescence) of physical abuse, sexual abuse, and neglect among at-risk youth? (2) How are different patterns of child physical abuse, sexual abuse, and neglect from birth to age 17 associated with alcohol, tobacco, and marijuana use at age 18?

Methods: The sample consisted of 899 adolescents from the Longitudinal Studies of Child Abuse and Neglect (LONGSCAN), a multisite cohort study designed to investigate child maltreatment in a high-risk sample of youth across the United States. Of the 899 adolescents, 48% were male, 26% were White, 53% Black, 7% Hispanic, and 14% other races. Participants reported the frequency of substance use (tobacco, alcohol, marijuana) in the past 30 days. Child maltreatment (physical abuse, sexual abuse, neglect) was measured using child protective services records (birth to age 17). For each maltreatment type, we created four binary variables (0 = no, 1 = yes) that correspond to Erikson’s four stages of development: infancy/toddlerhood (ages 0-2); early childhood (ages 3-5); school age (ages 6-12); adolescence (ages 13-17). We conducted repeated measures latent class analysis (RMLCA) using Mplus v.8.0.

Results: RMLCA identified three physical abuse classes (Stable low physical abuse; moderate-high physical abuse, spike at school age; High physical abuse in infancy/toddlerhood, sharply declining), two sexual abuse classes (Stable no/low sexual abuse; High sexual abuse at school age), and three neglect classes (Moderate-high neglect in childhood; Moderate neglect in infancy/toddlerhood, declining; Consistent low neglect, but moderate neglect at school age). Adolescents in the moderate-high physical abuse, spike at school age class showed greater alcohol, cigarette, and marijuana use, compared to other classes. Similarly, adolescents in the high sexual abuse at school age class showed greater substance use than those in the stable no/low sexual abuse class. Lastly, adolescents in the moderate neglect in infancy/toddlerhood, declining class showed significantly less substance use than those in the other two classes.

Conclusions and Implications: Exposure to maltreatment at school age appears to be a salient risk factor for increased substance use in adolescence, pointing to the need for additional support and services for youth who have experienced maltreatment during the school age period. On a promising note, our findings suggest that even adolescents who experienced physical abuse or neglect in infancy/early childhood can build resilience against substance use and achieve positive behavioral health outcomes, when they are protected from further victimization at later developmental stages. Overall, the present study highlights the importance of early intervention and ongoing maltreatment prevention (e.g., home visitation, parenting education).