Domestic violence exposure and abuse in childhood: relationship to parent-child attachment and the impact of these risks on youth violence perpetration

Background and Purpose: Evidence documents co-occurance of children's exposure to domestic violence and physical abuse (Appel and Holden, 1998, Dong, 2004; Fantuzzo, 1997; Staus and Smith 1990), and a deleterious effect of these risks on parent-child attachment (Styron, 1997). Compromised attachment promotes risky behaviors in adolescence (Levendosky, 2002; Brook, 1993, Luster, 1994). More information is needed on the relationship between these three variables, as well as whether children dually exposed to abuse and domestic violence are worse off in this regard than those who experience one form of violence (Daro, 2004). Analyses use data from a prospective longitudinal study to examine the impact of childhood abuse and exposure to domestic violence on parent-child attachment, the effects of these risks on violence perpetration in adolescence, and the degree to which attachment acts as a moderator.

Methods:

Data are from the Lehigh Longitudinal Study, began in 1976. Data on physical child abuse were collected from official child welfare records, mothers' reports, and adolescents' retrospective reports at three points: preschool (children 18 months to six years); schoolage (ages 8 to 11); and adolescence (range 14-23 years; average 18). The measure of abuse and domestic violence exposure combined prospective reports from parents and retrospective reports from adolescents. Armsted and Greenberg's Inventory of Parent Attachment (1987) was administered to adolescents in the third assessment of the panel.

Children were divided into three groups: abuse only; exposed to domestic violence only; and abused and exposed to domestic violence. Analyses combined ANOVAs and logistic regression to examine group differences on levels of attachment and violence perpetration in youth, as well as group-by-attachment moderation effects.

Results: Results suggest a deleterious effect of combined abuse and domestic violence exposure on parent-child attachments, especially for boys; effects of abuse alone or domestic violence alone were less evident. Children who were abused and exposed to domestic violence were lower than those who had been abused only on alienation from parents and lower than the abused only and domestic violence only groups on parent trust.

Although the quality of parent-child attachments did not appear to moderate the effect of abuse and domestic violence exposure on later violence perpetration in youth, attachment by itself did appear to influence the likelihood of violence during adolescence.

Conclusions and Implications:

Findings indicate that exposure to abuse and domestic violence weaken attachments between children and parents. While the overall level of attachment of children and parents appears to predict violence in adolescence, attachment appears not to moderate the effect of abuse and earlier violence exposure on that outcome. The combination of experiencing and witnessing violence in the home may be more traumatizing and disruptive to relationships than either by itself (McClosky et al., 1995). Results suggest the need for assessment and treatment of dual exposure in children, with a goal of strengthening bonds of attachment to caring adults, including a non-abusive parent. Targeted interventions to enhance parent-child relationships may lessen the risk of violence perpetration in those impacted by family violence as children.