The Timing of Maltreatment and Delinquency: Do Substitute Care Placements Account for Some of the Developmental Effects?

Purpose: Victims of maltreatment are at an increased risk of delinquency. Research indicates that the risk of delinquency is approximately 47% higher for youths associated with at least one substantiated report of maltreatment. In recent years there has been increased focus on the developmental effects of maltreatment. However, such studies have failed to account for differences in response from the child welfare system. That is, do child welfare systems respond similarly to cases of childhood and adolescent maltreatment? If not, can these differential responses account for some of the developmental effects? The current study addresses this gap the knowledge.

Methods and Results: The current study utilizes a sample of two birth cohorts (1983, 1984) with at least one substantiated allegation (n=18,676). We use official child welfare and juvenile justice records from birth to 18 years of age. We create five developmental group to capture and estimate the timing of delinquency: maltreatment in early childhood only (Age 0 to 5, 45%); maltreatment in later childhood only (Age 6 to 11, 31%); maltreatment in both early and later childhood (7%); maltreatment in adolescence only (Age 13 to 18, 13%); and persistently maltreated from childhood to adolescents (4%). Covariates include demographic characteristics, type of maltreatment, and substitute care placement.

Approximately 9% of the sample is associated with a delinquency petition. Regarding youth characteristics, male (OR=1.89) and African Americans (OR= 1.49) were at greater risk of delinquency. Regarding maltreatment type, physically abused (OR=1.33) or neglected (OR=1.23) youth were at an increased risk of delinquency. The coefficients associated with the timing of maltreatment changed as mediators were entered into the model. When timing variables were the only predictors in model, three groups, later childhood-only, childhood persistent, and persistent from childhood to adolescent groups were are all significantly more likely (βs >0; ORs >1; p<0.05) to be associated with a delinquency petition as compared to early childhood only group, while adolescent only group (β = -0.14; OR = 0.87; p<0.05) appeared to be less likely. However, once substitute care placement was entered into the model, all but one developmental effect vanished. Later childhood-only (5 to 12 years of age) remained significant (β =0.12; OR = 1.13; p<0.05). Finally, children that did not enter a substitute care setting were significantly less likely to be associated with a delinquency petition (β = -0.26; OR = 0.77; p<0.001).

Conclusions: There is evidence to suggest developmental differences in the modeling of maltreatment and delinquency. Yet to date there exist no studies that explored these relationships with official records, or that accounted for differences in placement rates. The findings from the current study indicate that developmental differences do exist, but that substitute care placements moderate and in some cases completely diminish the relative importance of timing. These findings raise important questions about why the risk of delinquency remains high for maltreated youth between 5 and 12 years of age.