CURRENT RESEARCH
Proximity and Risk in Childrenās Witnessing of Intimate Partner Violence Incidents
Abigail H. Gewirtz
Amanuel Medhanie
The past two decades have seen an upsurge of interest in documenting and understanding childrenās witnessing of intimate partner violence (IPV; e.g., Cicchetti & Toth, 1995; Edleson, 1999; Geffner, Jaffe, & Sudermann, 2000; Gewirtz & Edleson, 2007; Margolin & Gordis, 2000). In particular, studies examining the prevalence, incidence, and impact of exposure to violence on development have resulted in efforts to develop and document early interventions for child witnesses (Berkowitz, 2003; Cohen, 2003; Gewirtz, Harris, & Avendano, 2006). A key challenge for the field is to gather information on event parameters, childrenās experiences, and reactions during and immediately after a traumatic event. Such information is more readily accessible for events occurring in the community than for violence occurring within homes, which is often underre-ported (Straus & Gelles, 1990). Hence, not surprisingly, there is a dearth of data regarding contextual or incident details and childrenās responses during and soon after exposure to IPV (see Kracke & Hahn, 2008, this issue). Much of the research on childrenās exposure to IPV focuses on the relationship between exposure and childrenās functioning months to years after the violent event or events (McCloskey, Figueredo, & Koss, 1995; McCloskey & Walker, 2000; OāBrien, John, Margolin, & Erel, 1994; Zuckerman, Augustyn, Groves, & Parker, 1995), with little focus on the details of single events or shorter time periods (an exception is Jouriles et al., 1998). Although such data offer ābig pictureā information, there are few studies that provide basic descriptive information about children in the context of single violent incidents: their involvement, proximity, acute adjustment, and other descriptive information (Ernst, Weiss, & Enright-Smith, 2006; Fantuzzo, Boruch, Beriama, Atkins, & Marcus, 1997). This article contributes to the current literature on childrenās exposure to IPV and acute traumatic stressors by providing key descriptive details about the nature, proximity, and parameters of childrenās involvement as witnesses to IPV, and their variation by age, gender, and family relationships. In addition, this is the first article that we know of that provides concurrent data on acute risk and functioning in the immediate aftermath of an incident of IPV.
The lack of information on childrenās experiences in IPV incidents is not surprising (see Kracke & Cohen, 2008, this issue). First, the underreporting of childrenās involvement in IPV incidents has been amply documented (e.g., Bachman & Coker, 1995; Edleson, Mbilinyi, Beeman, & Hagemeister, 2003; Mihalic & Elliott, 1997). Several authors have documented parentsā underestimation of the extent of childrenās witnessing of violent incidents even as their children describe details of events they were not supposed to have seen (e.g., Edleson, 1999; Jaffe, Wolfe, & Wilson, 1990). This underreporting may be due, in part, to policies in some jurisdictions that mandate the reporting to child protective agencies of incidents where children are actively involved in IPV.1 Second, the acute nature of violent events precludes the use of standard research designs to elicit information about childrenās experiences and involvement during violent events (Edleson et al., 2003). Third, despite increasing interest in the investigation and prosecution of IPV incidents, particularly those involving children (Gewirtz, Miller, Weidner, & Zehm, 2006), most police departments do not routinely mandate documenting the presence of children in an officerās police reports. Hence, many child witnesses to IPV are, literally, āinvisibleā to official documentation (Osofsky, 1997; Rosenbaum & OāLeary, 1981). Finally, the lack of a formal classification system or national surveillance system for childrenās exposure to IPV precludes gathering national epidemiological data on the extent and range of childrenās experiences with regard to IPV compared with, for example, child abuse and neglect experiences (see concerns regarding definitions in Kracke & Hahn, 2008, this issue).
Despite these drawbacks, multi-informant data on IPV incidents, when available, can be extremely useful for understanding childrenās involvement. Sources of information may include parent report, child self-report, other witness report and/or police report (when such incidents are reported to police, and when police document child witness details). When police are called, police reports can be key sources of information about event parameters, physical evidence, witness descriptions, and childrenās involvement at the scene of an IPV incident. For example, in their description of childrenās witnessing of IPV through the Spousal Abuse Replication Study (SARP), Fantuzzo et al. (1997) used police reports and victim interviews to describe childrenās involvement in misdemeanor IPV incidents in five cities. Findings indicated that children under 5 were disproportionately involved in IPV incidents, and that there was a range of child involvement, including child calls to 911. However, the study did not track details of childrenās exposure to violence or developmental differences in childrenās exposure.
Knowledge of childrenās event-related experiences and functioning during and immediately after IPV incidents is critical in the following ways: (a) to provide details that can further delineate the similarities and differences in childrenās reactions to varied traumatic stressors; (b) because event/exposure characteristics have consistently been identified as significant predictors of recovery and adjustment in both the short and longer term aftermath of traumatic events (e.g., La Greca, Silverman, Vernberg, & Prinstein, 1996; Pine & Cohen, 2002), and hence are important for the development of early preventive interventions for such children; and (c) because context- or incident-specific details of violent events can also help to further our understanding of specificity in the relationship between witnessing violence and adjustment outcomes (McMahon, Grant, Compas, Thurm, & Ey, 2003).
RISK AND CHILD ADJUSTMENT FOLLOWING VIOLENT INCIDENTS
Several factors have been identified as important in understanding recovery and adjustment among traumatized children (Green, Korol, & Grace, 1991; Chamberlain, 2008, this issue; La Greca et al., 1996; Pine & Cohen, 2002). These include event parameters (details of the event and the childās involvement, particularly, degree of exposure to the traumatic event), individual child risk and protective factors (e.g., history of trauma, psychopathology, and socio-demographics), and environmental variables (e.g., parental reactions, social and parental support of the child, etc.).
For all types of trauma, the level of exposure to the event has consistently predicted the later level of psychopathology (Pine & Cohen, 2002). This has been documented more clearly in studies of disaster and community violence (see Kracke & Hahn, 2008, this issue), such as in a study of a school shooting that indicated that children who were more proximal to the sniper or who visually witnessed the event were more likely to manifest subsequent traumatic stress symptoms (Nader, Pynoos, Fairbanks, & Frederick, 1990; Pynoos & Nader, 1989).
Developmental psychopathology approaches to the study of stress and adversity in development have emphasized the vulnerability to subsequent adjustment associated with childrenās experience of prior trauma events (Cicchetti & Toth, 1995). For example, McCloskey and Walker (2000), in a sample of 237 school-age children, found that prior trauma experiences (i.e., child abuse, domestic violence, and death or illness of a close family member) were predisposing factors for post-traumatic stress disorder. We could find no studies, however, that investigated the relationship between prior or event-related risk and childrenās concurrent adjustment in the near aftermath of a violent incident. Such data would be important for understanding the determinants of recovery among traumatized children, and in particular, the development of a comprehensive model of post-trauma functioning that takes into account the relationships between prior risks, event-related risks, functioning shortly after a traumatic event, and subsequent disorder in the months following.
The goals of the current study were (a) to gather basic descriptive details of childrenās participation in violent incidents and (b) to understand predictors of child functioning soon after the incident. Data were analyzed from an existing data set that provided a combination of police, clinician, and self-report data about child witnesses to IPV incidents resulting in 911 calls.
Research Questions
1. What is the nature and extent of childrenās involvement in IPV incidents? Are there developmental and/or gender differences in involvement in violent incidents?
2. Do a combination of acute (i.e., proximity) and chronic risks (i.e., prior trauma exposure) predict child functioning within the days following a violent incident?
METHOD
Sample
Data for this study were gathered from the records of the Minneapolis Child Development Policing Program (CDPP). The CDPP is a multi-disciplinary community-university partnership aimed at providing acute clinical and advocacy services to children and families following their exposure to violence. The purpose of the program is to increase access to care for children exposed to violence and ultimately to ameliorate the impact of violence on childrenās development and adjustment. The program, which is voluntary for families, is a partnership of police officers and mental health professionals who respond jointly, as soon as possible after a v...