Developmental Trajectories of Externalizing Behavior in Maltreated Children: Roles of Exposure to Intimate Partner Violence (IPV) and Sibling Structural Factors

Background and Purpose: Maltreated children are at greater risk of IPV exposure and may be at heightened risk of developing externalizing behavior problems (EXT) than non-exposed children. However, studies also underline great variability among children; some exhibit rates of EXT comparable to their non-exposed peers. One potential risk factor is the developmental stage of IPV exposure, as some developmental stages may be more sensitive, and one protective factor may be the support offered through sibling relationships. This study examined (1) the extent to which development of EXT varied in maltreated children, (2) how timing of IPV exposure explained the variation in EXT development over time, and (3) the extent to which sibling structural characteristics (sibship size, birth order) influence EXT development.

Method:  Using data from the National Survey of Child and Adolescent Well-Being (NSCAW), EXT latent class trajectories were estimated using cohort sequential latent growth mixture modeling (CS-GMM) for children (n=1,770) who were between 0-5 years at baseline. Four waves of data collection were used to estimate EXT developmental patterns between birth to 10 years. EXT was measured by caregivers-reported Child Behavior Checklist (CBCL). IPV exposure was caregiver-reported using the Conflict Tactics Scale (CTS) physical assault subscale during three time periods: infancy/toddlerhood (birth-2 years), preschool age (3-4 years), and school age (5-10 years). Sibling structure factors included baseline sibship size and birth order. Control variables included child-level (gender, race/ethnicity, substantiated child maltreatment, previous Child Protective Services (CPS) involvement) and caregiver/family-level (income below poverty level, caregiver education, age) characteristics. IPV exposure effects were estimated separately for the singleton (43%) and sibling (57%) samples, using multinomial logistic regression adjusted for controls. Significance of sibling structural factors were examined in the sibling sample.    

Results:  Two resilient (low-stable, 32%; high-decreasing, 17%) and 2 non-resilient (high-stable, 27%; low-increasing, 24%) groups were identified. For the singleton sample, preschool-age IPV exposure increased odds of being in the low-increasing group versus the low-stable group (aOR=2.60, p<.001). For the sibling sample, school-age IPV exposure decreased odds of being in the high-decreasing versus the high-stable group (aOR=0.38, p=.003). Children who were second-born (aOR=0.44, p=.003) and third-born (aOR=0.47, p=.026), when compared to first-borns, had lower odds of being in the low-increasing group versus the low-stable group. 

Conclusions: Some children exhibit resilient EXT over time suggesting that risk and protective factors may have the potential to change the course of development. IPV exposure was associated with poorer outcomes over time among maltreated children. Sibling relationships may be protective, may function as additional attachment relationships, and buffer children from negative sequelae associated with violence exposure. First-born children were more likely to exhibit EXT than their younger siblings. Corroborating previous research that underlined how eldest siblings may experience more responsibility to intervene to prevent violence exposure or protect younger siblings and the non-offending parent, this added sense of duty may be an additional stress for first-born children resulting in an increased vulnerability to the development of EXT.