The Forgotten Ones? Exploring Factors Related to Child Maltreatment Reports and Substantiation Among Adolescents

Background and Purpose: Adolescents account for nearly a quarter of the child maltreatment reports and 35% of the child welfare out-of-home population (U.S. DHHS, 2010). However, little is known about the risk factors leading to these reports, substantiation rates, and reports’ outcomes (Mersky, Berger, Reynolds, & Gromoske, 2008). Much of the research on child maltreatment has focused on younger children. Less attention has been paid to adolescents’ entry and involvement in the child welfare system (Simmel, 2012). This is unfortunate as adolescents represent a qualitatively different group due to dramatic differences in psychosocial development and the varying contextual circumstances in which adolescents are involved. Thus their involvement with child welfare likely differs from their younger counterparts. The goals of the study include identifying; 1) developmental differences among type of report and associated risk factors, and 2) factors associated with substantiation of a report.

Methods: This study uses the National Child Abuse and Neglect Data System (NCANDS) for years 2005-2010. NCANDS is administrative data which yields rich information regarding referrals made to state-level child protective service (CPS) agencies.  In our data, approximately 3.6 million referrals (or 17 percent of all referrals) involved a youth aged 14-18.  As not all states submit data on our variables of interest, our analytic samples are smaller and include fewer states. We use linear probability models to determine which factors predict substantiation. Our preferred model includes 26 states with 451,000 youth reports and an alternative model includes 42 states with 1.9 million youth reports.  In either model, youth continue to comprise approximately 17 percent of the sample, implying the sample reduction is not correlated with the child’s age. 

Results: Descriptives reveal pronounced differences among early childhood (0-5), middle childhood (6-13), and adolescents (14-18).  The proportion of neglect cases declines steadily with age, with a prevalence rate of 67% among young children but only 55% among adolescents. Conversely, physical, sexual, and emotional abuse increase significantly from early childhood to adolescence.  Among adolescents, the most common type of report involves a female sexual abuse victim, and this pattern intensifies among substantiated referrals.  According to a linear probability model, adolescent characteristics significantly associated with increased substantiation are adolescent drug or alcohol use (9 percentage points [pp]) and the presence of an adolescent disability (3 pp).  The predicted probability of substantiation increases the most for reports involving sexual abuse (18 pp), caretakers in violent relationships (20 pp), and inadequate housing (31 pp).

Conclusions and Implications: These findings provide important insight into the differences among reporting and substantiation of child maltreatment across development. Results suggest that adolescents have distinct risk factors for being reported to CPS and that there are identifiable predictors for a report’s substantiation. These findings have implications for maltreatment prevention among adolescents. Prevention programs and policies may be tailored to meet the unique needs of different age groups. Special attention should focus on the risk factors associated with adolescents. Future research is needed to better understand effective child maltreatment prevention and interventions for all ages of children.