Maltreated children who have also been exposed to domestic violence (DV) are at a greater risk of recurrent maltreatment. They are more likely to have additional allegations of maltreatment following child welfare involvement. DV exposure is detrimental to the child's safety, but most research in this area has not yet addressed whether the risk for re-report is lower when the DV has stopped or been reduced. The purpose of this study is to identify characteristics, including subsequent DV, which may contribute to the high levels of re-report among children exposed to domestic violence.

The subset of NSCAW used for this study is battered caregivers whose child remained in home following the initial child maltreatment investigation. Analyses examined the rate of report for new allegations of maltreatment over 36 months. Logistic regression analysis was used to determine child, caregiver and environmental characteristics that influenced the rate of re-report among families with caregiver reported DV.

In NSCAW, new reports for maltreatment were significantly more likely for families with reported DV (34%) compared to 25% for families without DV. Findings from the logistic regression analysis indicated that, when controlling for other factors, the level of DV over the study period did not significantly predict re-report among families with DV. Other caregiver characteristics appear to be more important. The children of battered caregivers meeting the diagnostic criteria for major depression were twice as likely to be re-reported compared to children of non-depressed, battered caregivers. Furthermore, the caregiver's own childhood victimization was a strong predictor of re-report. The location of the agency was also important and re-reports were less than half as likely when the child welfare agency was located in a non-urban community.

Findings from this study have important implications for child welfare policy and practice. Battered caregivers often have a complexity of risks in addition to the domestic violence and these findings suggest that these other factors may contribute more to the safety of the child. Interventions among battered caregivers must extend beyond reducing the domestic violence and also target other needs. These caregivers may need simultaneous mental health and safety services. Characteristics of the child welfare agencies are also an important consideration. Agencies in urban and non-urban areas often have differing levels of resources; therefore, characteristics of the agencies themselves may also be an important contributor to the long term safety of children. Multi-level modeling will be conducted to further examine how specific characteristics of the agency (e.g., fiscal resources, training specific to DV, and linkages with DV organizations) influence rates of re-report over the 36 months and results will be discussed.