Brief Treatment and Crisis Intervention Advance Access originally published online on September 14, 2007
Brief Treatment and Crisis Intervention 2007 7(4):364-382; doi:10.1093/brief-treatment/mhm013
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Relationship Between Domestic Violence and Child Neglect
From the Kent School of Social Work, University of Louisville (Antle, Barbee, Sullivan, Yankeelov, Johnson) and Department of Communication, University of Louisville (Cunningham)
Contact author: Becky F. Antle, Kent School of Social Work, University of Louisville, Louisville, KY 40292. E-mail: becky.antle{at}louisville.edu.
This study examined the comorbidity of child neglect and domestic violence (DV) in a sample of public child welfare cases. All cases investigated for child neglect in 1999 (N = 2,350 families) from a single county in Kentucky were included in the analysis. Findings indicate that DV was comorbid in approximately 29% of the cases and, although cases were more likely to be opened when child neglect was substantiated, they were less likely to be opened when DV was also present. A secondary random subsample of 100 cases were reviewed to examine the impact of this comorbidity on the child welfare workers' assessment of risk and problems in children's functioning, the relationship between DV and risk factors contained on the risk assessment, and the workers' response to DV through legal actions and incorporation of DV into the case plan. The results showed that child welfare workers rate families at significantly greater risk, particularly when there was previously unreported DV, and identified more problems in children's interpersonal functioning when DV was present. DV was significantly correlated with more severe neglect and a limited social support network for the family. Although workers took a number of legal actions in response to DV, DV was addressed in only 35% of child welfare case plans.
KEY WORDS: child abuse, child neglect, domestic violence, comorbidity
The purpose of this research is to examine the relationship between domestic violence (DV) and child neglect in a community sample. Although much is known about the independent etiologies and detrimental effects of these issues, there is little empirical research on their comorbidity in at-risk families, as much of the previous research on the co-occurrence of DV and child maltreatment has focused primarily on physical child abuse (e.g., Hughes, 1997). Nevertheless, these phenomena share many risk factors and consequences for affected children. An understanding of the relationship between DV and child neglect may begin to explain the prevalence and significant negative consequences of these forms of child maltreatment. In turn, the existence of such a relationship will have important implications for the assessment and intervention processes in social service settings.
Prevalence and Impact of DV
It is estimated that approximately 5.3 million women are victims of DV annually in the United States (Tjaden & Thoennes, 2000). Approximately 2 million of these incidents of DV will result in an injury to the victim, and 1,300 will result in death (Centers for Disease Control and Prevention, 2003). The Centers for Disease Control and Prevention (2003) projects that the costs of DV in 1995 exceeded an estimated $5.8 billion; these costs included nearly $4.1 billion in the direct costs of medical and mental health care and nearly $1.8 billion in the indirect costs of lost productivity. The same research found that victims of DV lose nearly 8 million days of paid work—the equivalent of more than 32,000 full-time jobs—and almost 5.6 million days of household productivity each year.
There are between 3 and 10 million children who are exposed to this abuse (Tolman & Edelson, 1995). The effect of witnessing DV on child development has been of particular concern in recent years. Children in these violent families demonstrate both immediate and long-term patterns of maladjustment (Hughes, 1997). DV also has distinctive effects on child outcomes, including poor problem solving skills and an external locus of control (Hughes, 1997), gender-based replication of parental behaviors (Jaffe & Suderman, 1995), and posttraumatic stress disorder (Groves, 1999).
A variety of theories have been proposed that suggest both direct and indirect mechanisms of the effects of witnessing DV. The direct mechanisms by which DV affects children include the modeling of aggression by the perpetrator and the stressful environment of the home (Hughes, 1997). Children may demonstrate behavioral problems because they have learned aggressive scripts for interpersonal coping and are imitating the behavior of the authoritative parent. There is also an element of disinhibition toward violence that leads to both acceptance and utilization of aggression. Another direct pathway alternative suggests that declines in children's functioning are a natural response to the chronic stress created by DV (Chalovich, 2004). Finally, the theory of learned helplessness, whereby children who are maltreated or exposed to violence are more likely to become adult victims of violence, has also been used as a rationalization of battered women's coping responses to DV (Renner & Slack, 2006; Walker & Browne, 1985). In a recent study of the intra- and intergenerational associations between DV and child maltreatment, Renner and Slack (2006) found that childhood physical abuse, sexual abuse, and witnessing DV were significant predictors of adult DV victimization. According to these researchers, this finding offered some support for the theory of learned helplessness. Public child welfare workers are most often called upon to address these problems of and the consequent impairments in functioning caused by child maltreatment and DV without the necessary data to illuminate their assessments and interventions.
Prevalence and Impact of Child Neglect
In 2002, an estimated 3,193,000 children received investigations or assessments from Child Protective Services (CPS) agencies, and approximately 896,000 of these children were found to be victims of child maltreatment (U.S. Department of Health and Human Services, 2004). Additionally, the U.S. Department of Health and Human Services (2004) reports that 61% of child maltreatment victims nationwide were victims of neglect, and child neglect accounted for 38% of child fatalities in 2002.
Data from the same sources indicate that 75% of maltreated children in Kentucky were victims of neglect.
A conservative estimate of the direct costs of child maltreatment in the United States is $24 billion per year, whereas an estimate of the indirect costs is $94 billion per year (Fromm, 2001). These direct costs are based on services provided by health care, mental health, and child welfare systems as well as law enforcement agencies and the courts. Estimates of indirect costs are based on the costs of special education, mental health and health care, juvenile delinquency, lost productivity to society, and adult criminality, which are all possible consequences of child maltreatment.
There is general agreement in the literature that both child abuse and neglect have deleterious effects on various areas of functioning. Some of the common effects of these problems include inappropriate weight levels (Crouch & Milner, 1993; Jaffe & Suderman, 1995), low academic achievement (Hughes, 1997), and withdrawn or aggressive behaviors toward peers and clinical depression. There are also unique effects of neglect on children such as failure to thrive (Crouch & Milner, 1993), insecure attachment to caregivers and psychological distress specific to the neglect (Gauthier, Stollak, Messe, & Aronoff, 1996). Gauthier et al. (1996) explained the latter effect as the result of children's inability to rationalize the neglect that is in contrast to cases of abuse where children are able to connect their maltreatment with behaviors. Research has shown that the effects of these forms of maltreatment are moderated by protective characteristics of the child (i.e., age, IQ, temperament) and the severity and timing of the maltreatment (Hughes, 1997; National Research Council, 1993).
The effects do not stop with the victimized children themselves. Research has shown that children of violent and neglectful parents show extremely high rates of the same maltreatment of their own children (Gellert, 1997; Weston & Colloton, 1993). Men who were abused or neglected as children are also more likely to engage in violent crime (Dutton & Hart, 1992). In fact, it is estimated that 13% of all violence can be linked to previous child maltreatment (U.S. Department of Justice [U.S. DOJ], 1996). The productivity that is lost by adult survivors of severe childhood maltreatment has been projected to cost the United States an estimated $656 million per year (Fromm, 2001).
Linkages Between DV and Child Neglect
In order to provide effective interventions to mediate these negative outcomes for individuals and society, an understanding of the dynamics within and between DV and child neglect is needed. Smokowski and Wodarski (1996) argue that research such as this can positively change the public child welfare system by extending our knowledge of the etiology, effects, and prevention of these issues. With a better understanding of the relationship between DV and child neglect, public child welfare workers may be able to begin to alleviate the harms caused by these problems and stop the cycle of violence and neglect in these families for future generations.
The relationship between DV and child neglect is suggested by the similarity of risk factors, effects, and tangential issues for these phenomena. Common risk factors for DV and neglect include violence in the family of origin, social isolation, substance abuse, and economic stressors (Gellert, 1997; National Research Council, 1993; Smokowski & Wodarski, 1996). It has also been found that children exposed to both DV and child maltreatment are more likely to experience depressive, withdrawn, and/or anxious behaviors and more likely to have high levels of total behavioral problems (Kernic et al., 2003).
There are also risk factors associated only with DV. These factors are personality traits of the batterer (Gellert, 1997), verbal conflict (Stets, 1995), and traditional values (Williams, 1995). Other risk factors specific to child neglect are characteristics of the child (Herzerberger, 1996), parental attitudes and attributions of blame (National Research Council, 1993), lack of parenting skills (Gaudin, 1993), and maternal depression (Smokowski & Wodarski, 1996).
Comorbidity rates of spouse and child abuse have been estimated to be between 40 and 60% (Hughes, 1997). Gender differences have shown that approximately 50% of wife batterers also abuse their children, whereas only 24% of victims engage in abuse (Ross, 1996). The significance of this variable is seen in those children who experience both DV, and child abuse are at even greater risk for maladjustment (Hughes, 1997).
Pathways From DV to Child Outcomes
There are several indirect mechanisms of influence used to explain the effects of DV on children. Rossman and Rosenberg (1997) have suggested that the effects of DV on children are the result of a failure to meet the developmental needs of competence, autonomy, and relatedness. Other researchers posit that the hostility and aggression of the marital relationship may be carried over in the form of negative emotions or responses to children (Hughes, 1997).
Other investigators suggest that DV influences parental behavior, and that change in parental behavior is responsible for many of the negative outcomes observed in children. Cummings and Davies (1994) found that interparent conflict may result in inconsistent and negative discipline, as well as ineffective child monitoring. Similarly, Wolfe, Jaffe, Wilson, and Zak (1985) found that maternal stress in DV situations significantly impairs the mother's parental functioning. This impairment was found to predict child adjustment significantly more than simply witnessing DV. The indirect impact of DV on child behavior was also addressed by English, Marshall, and Stewart (2003). These researchers found that DV greatly harmed family functioning, the caregiver's health and well-being, and the quality of the caregiver's interaction with the child. These deficits in family functioning were subsequently related to problems in child functioning and health.
Studies have shown significant rates of depression among women victimized by DV (Gellert, 1997; Hughes, 1997). This depression may affect the parent–child relationship and place children at risk for psychopathology and other forms of maladjustment. In a review of the literature on the relationship between parental depression and child functioning, Downey and Coyne (1990) confirm that depression resulting from marital conflict has significant implications for children, suggesting that depressed mothers experience difficulty in being warm and consistent toward their children (from Weissman & Paykel, 1974). They have little meaningful interaction; as such effortful behavior exceeds the capacity of their reduced energy levels (Fisher, Kokes, Harder, & Jones, 1980). The depressed mothers from this study were also less vocal, spontaneous, and positive. All these behaviors are indicative of neglect on the part of depressed mothers. In fact, depression has been identified as one of the greatest risk factors for child neglect (Smokowski & Wodarski, 1996). Hence, the relationship between DV and child neglect may be the function of maternal depression.
Similar findings on the effects of general stress levels of abused mothers also suggest neglectful parenting. Wolfe et al. (1985) demonstrated the significant effects of chronic stress on maternal functioning in physical, psychological and parental domains. This maternal stress accounted for 19% of the variance in child maladjustment.
Comorbidity Rate of Child Maltreatment and DV
Social service agencies have also begun to recognize the coexistence of DV and child maltreatment. The presence of "multiproblem families," who experience various forms of abuse, neglect and other social stressors is well established in the field of child welfare (Samples, 1981). There have been several attempts to determine the extent of comorbidity of social problems in families involved in social services. Tolman and Edelson (1995) reported that 32% of child maltreatment cases in Massachusetts also involved DV. However, this study did not differentiate between child abuse and neglect. Stewart, Senger, Kallen, and Scheurer (1987) retrospectively examined the rates of various forms of abuse and neglect among middle-income college students. These researchers found that 6% of students had witnessed DV, of which 11% reported some form of neglect. There was a significant positive correlation between child neglect and DV in these families.
Another such study was undertaken by Weston and Colloton (1993) to assess the relationship between historical abuse of mothers and consequent child neglect in nonorganic failure to thrive children. The researchers found that both childhood and adult abuse were significantly related to the mothers' neglect of their children. However, childhood abuse was a better predictor of neglect, and only sexual abuse differentiated the mothers of the failure to thrive children from the normal sample.
McGuigan and Pratt (2001) carried out an exploratory study of the co-occurrence of DV and three types of child maltreatment (physical abuse, psychological abuse, or neglect of that child up to the age of 5) among 2,544 at-risk mothers with first-born children who voluntarily participated in a home-visiting child abuse prevention program. The results of this study indicated that families in which DV was present during the first 6 months of a child's life were more than twice as likely to have child neglect confirmed during the next 5 years. Another finding of interest here was that DV preceded child maltreatment in 78% of the cases of co-occurrence, giving credence to the likelihood that DV leads to child maltreatment.
Beeman, Hagemeister, and Edleson (2001) examined differences between families experiencing both child maltreatment and DV (referred to by the researchers as "dual violence") and families experiencing child maltreatment only. Those findings indicated that dual-violence families (66%) were more likely than child maltreatment–only families (51%) to have had at least one neglect allegation in a 2-year period. Furthermore, with regards to determined type of maltreatment, disregard for safety, a subtype of neglect most commonly associated with DV, was significantly more prevalent among those families. Finally, dual-violence families were assessed by child protection workers as being at higher risk than child maltreatment–only families and workers were more likely to open the former cases for child protection services. However, dual-violence families with open cases were less likely than their child maltreatment–only counterparts to receive services, with the exception of referral to the county attorney. This finding led researchers to recommend more careful consideration of risk and protective factors by CPS workers.
Hartley (2002) studied the co-occurrence of DV and child maltreatment and found the following patterns among families experiencing both DV and child neglect: more single-parent households, fewer married parents, fewer biological fathers, more mothers with histories of substance abuse and mental illness, and more mother-only perpetrators compared to the maltreatment-only neglect families. In a 2004 study, Hartley noted a higher percentage of neglect, specifically lack of supervision, among families experiencing severe DV than among families experiencing less severe DV and child maltreatment only. These findings, which are identified by the researcher as being inconsistent with previous literature that supports more child physical abuse than neglect occurring among families experiencing severe DV, indicate that the co-occurrence of DV and child neglect warrants further study.
Although much is known about the independent dynamics of DV and child neglect and a few studies have touched on the relationship between these problems, there is a substantial gap in the literature on these issues. Limitations of the aforementioned studies included methodological problems, small sample sizes, and an inability to generalize to larger populations. Many studies that have addressed these issues have also confounded child neglect with physical abuse (Tolman & Edelson, 1995) and have failed to determine the precise comorbidity rate of DV and child neglect. In addition, the fact remains that there is little known about potential pathways of influence between DV and neglect. Some research suggests that one of the reasons mothers neglect their children is because they are dealing with their own issues of depression and stress from abusive spouses (Jaffe & Crooks, 2005). If this is the case, then the detrimental effects of DV may be better understood in the context of child neglect. Yet, no research has attempted to account for the variance in children's functioning in this way. Finally, there is a significant need for evaluation of the interventions for families and children experiencing DV and child neglect. This is particularly important for multiproblem families, for whom treatment alternatives are now well understood.
The Present Research
The current study addresses several of the gaps in the literature on these issues by assessing the relationship between child neglect and DV. This relationship will be studied by identifying the rates of reports of DV in cases of child neglect. This will establish valid comorbidity rates to be used in future analysis of the pathways of influence between these variables. Such interactions may account for the etiology of child neglect, as well as the negative outcomes of DV on children.
This study will also assess the integration of these joint issues into the treatment plans of public child welfare workers. The importance of such incorporation is revealed by significant effects that each of these issues has on child functioning. Many of these effects may be missed in child welfare workers' assessments of these families because direct physical abuse is more readily identifiable and considered more serious. However, the prevalence of children being affected by neglect and DV as well as the significantly greater harms of these problems than any other form of maltreatment points to the fallacy in these policies. Hence, a better understanding of the relationship between treatment of child neglect and DV will make a significant contribution to both the profession of social work and society at large.
 |
Method |
|---|
 TOP Method ResultsDiscussionFundingReferences |
|---|
Sample
The primary sample used for the analysis came from Systems Administration at the Cabinet for Health and Family Services of the Commonwealth of Kentucky. Researchers requested Systems Administration to generate a list of all reported neglect cases from the Jefferson County, Kentucky CPS investigative unit for a single year. The report included the case number, information about the substantiation decision, presence or absence of DV in the home, and whether or not the case was opened. In this management report, DV was classified as present or absent (vs. level of substantiation like child neglect). The maltreatment type was limited to child neglect in order to establish an accurate comorbidity rate and to establish an adequate sampling frame for further study. This produced a primary sample of 2,350 cases. A secondary systematic random (every 10th case) subsample (n = 100) was drawn for more detailed analysis.
Variables
The secondary sample was coded for a variety of child neglect and DV variables. These variables were operationalized through the forms used by CPS workers for intake, investigation, assessment, and case planning. The instrument that was utilized to collect this information was a self-designed Kent Case Evaluation Form. Child neglect and DV were operationalized through level of substantiation. This level of substantiation included (a) substantiated, (b) found and substantiated, (c) some indication, and (d) unsubstantiated. Found and substantiated is a designation used in the state of Kentucky for cases in which specific types of maltreatment that are not part of the original report are discovered during the course of the case. Furthermore, there were specific types of behaviors that were coded from the descriptions in the case files as indicative of neglect. These behaviors included a deficiency in any of the following: supervision, hygiene, housing, clothing, intervention, food, transportation, medical care, education, emotional neglect, and home safety.
There was documentation of the level of risk for the seven factors on the CPS risk assessment form. Each factor was assessed on a four-point scale, ranging from no risk (1) to high risk (4). Insufficient information or areas that are not applicable were rated as 0. Workers were provided with a general description of each of the seven factors, as well as anchor descriptions for each point on the rating scale. The workers completed this tool after interviewing the family, child(ren), and significant others to gather necessary information to rate the factors. This consensus-based risk assessment tool was developed by a panel of child welfare agency administrators, policy analysts, and practice experts based upon previous research on factors that place children at risk of future abuse or neglect. Unlike actuarial-based risk assessment tools that incorporate client characteristics statistically predictive of future maltreatment (Rycus & Hughes, 2003), consensus-based instruments require child welfare professionals to consider risk factors and arrive at a professional judgment about the overall level of risk (Knoke & Trocme, 2005). No reliability or validity data are currently available on this consensus-based risk assessment tool.
Summary information contained on the risk assessment tool involved seven factors including child's physical and mental abilities, severity of abuse and location of injury, severity of neglect and environmental conditions of the home, previous history of abuse and/or neglect, family risk factors (i.e., caretaker's physical and intellectual abilities, community resources, stresses/crises, financial stability, transportation, substance abuse, family violence, criminal activity, attitude toward child, level of cooperation, and parenting skills or knowledge), perpetrator's access to the child, and the strength of the family's support system (Table 1). As these means indicate, the factors rated with the highest risk were the family risk factors and perpetrator access to the child. These findings are consistent with the literature on risk factors for child neglect, such as family violence, substance abuse, and economic stress (National Research Council, 1993; Smokowski & Wodarski, 1996). Also, the higher rating of risk for perpetrator access to the child reflects the identification of mothers as perpetrators of neglect (48%) and the primary caregiver of the child (50%) in this study (see Table 2 for demographics). The factor with the lowest mean rating of risk was severity of abuse/location of the injury that was not surprising given that this study used a sample of child neglect cases. This risk factor is used only for child physical abuse. Therefore, only cases that involved both child neglect and physical abuse could have received an elevated score on this risk factor.
|
|
Another set of variables extracted from the case record involved problems in children's functioning, categorized as cognitive, psychological, behavioral, interpersonal, and physical. These problems in functioning were classified as present or absent, and the total number of such issues was calculated.
DV was measured in several ways. First, as previously mentioned, DV was reported dichotomously (present or absent) through the state management report, Second, in cases where DV was coded as present, there was often a formal report to Adult Protective Services (APS) on file through which a level of substantiation for DV was assigned. These levels of substantiation corresponded to the levels of substantiation for child maltreatment (substantiated, found and substantiated, some indication, and unsubstantiated). Third, previously unreported DV was documented through a review of the caseworkers' running record or case notes recorded after each interview, meeting, and so forth. The case notes were reviewed, and if there was documentation of any previously unreported DV, this was recorded dichotomously (yes or no to previously unreported DV).
Basic demographic data were collected through the Kent Case Evaluation Form, including information about the children, parents, or legal guardians, and alleged perpetrator in the case. These demographics included date of birth, gender, and adult's relationship to the child. Historical information was collected on past reports of abuse or neglect to CPS and parental experiences of childhood abuse or neglect and included potential contributing factors such as substance abuse, family violence, poverty, family size, juvenile delinquency, psychological problems, medical problems, and major life events.
The last section of the Kent Case Evaluation Form detailed the case plan. First, specific objectives for each system were listed, including those for the parents, children, CPS, and others involved in the case. The priority given to the issues of physical abuse, neglect, and DV in the case plan was established in two ways. First, the rank order of issues was determined through the order of objectives in the plan. Then, the number of objectives assigned to each issue was recorded. The rationale for this approach was that the ranking and number of objectives by type of family violence would reflect the prioritization of these distinct issues. Child welfare workers are trained to address the primary problem first in the case plan and to develop multiple strategies to address this problem. This information was also used to determine if DV was also incorporated into case plans which focused on children. The review of the case plan also documented whether the child welfare worker differentiated between physical abuse and neglect through separate case plan objectives specific to each type of maltreatment. Although this study focused on a sample of child neglect cases, there were still a number of cases that involved physical abuse providing limited information about whether treatment plans differ with respect to maltreatment type.
Finally, any legal action that was taken or planned was recorded on the instrument. These legal actions included Emergency Protective Orders, Emergency Custody Orders, Domestic Violence Orders, custody hearings, petitions, and general court orders.
|
Results |
|---|
|
TOPMethodResultsDiscussionFundingReferences |
|---|
Sample Description
The primary sample from Systems Administration consisted of 2,350 cases from which 100 cases were randomly sampled for further analysis (Table 2). The subsample consisted of primarily Caucasian children with a fairly equal distribution between males and females. Mothers were the most common caregivers and most frequently identified as the perpetrators of child neglect.1
Comorbidity of Domestic Abuse and Child Neglect
The primary analysis of the total number of neglect cases for the county revealed that the comorbidity rate of child neglect and DV for this county to be 29%. In other words, of the 2,350 cases of child neglect, 679 of the families also experienced DV.
Chi-square analysis of the larger sample indicated patterns of case opening based upon the level of substantiation of the case. Substantiated neglect cases were much more likely to be opened than those with only some indication of neglect, 
2 (2, N = 2,350) = 172.59, p < .0001. The likelihood that comorbid DV would result in a case opening was also explored through a series of 2 x 2 chi-square statistics. This analysis showed a consistent pattern of cases being less likely to be opened when DV was present. See Table 3 for descriptive statistics.
|
Of the 2,350 cases, more than half (61%) were substantiated for child neglect and, among these, 27% also involved DV. Substantiated cases were opened for ongoing services (out-of-home care, in-home services, etc.) 57% of the time, but substantiated neglect cases with comorbid DV were opened somewhat less (45%) often (
2 [1, n = 1,434] = 31.58, p < .0001). Of the total number of cases, 253 or 11% were found and substantiated for neglect (i.e., the original allegation was for another form of maltreatment and neglect was discovered sometime later), and 33% of these cases had comorbid DV. Again, a slightly lower percentage of these cases (55%) were opened than cases involving only neglect (63%). A similar difference was found among cases with some indication of neglect (n = 663), where 31% of such cases involved DV yet only 18% were opened for ongoing services (as opposed to 23% where DV was not involved). Although the pattern was similar (DV cases less likely to be opened) for cases that were identified as found and substantiated or some indication, these associations (chi-squares) were not statistically significant. In summary, substantiated neglect cases were much more likely to be opened than those with only some indication of neglect. Also, despite a fairly equal distribution of comorbid DV across levels of substantiation, cases involving DV that were fully substantiated were less likely to be opened for ongoing services.
Secondary Analysis of the Subsample (n = 100)
Types of Maltreatment.
Of the 11 types of maltreatment examined, the most frequently reported were lack of safety (92%), emotional neglect (91%), lack of supervision (86%), physical punishment (47%), and refusal of treatment (35%). The relationship between these different types of maltreatment and comorbid neglect/DV was also examined. Chi-square analysis of these 11 factors indicated a significant association between cases with comorbid DV and physical punishment (2 = 22.14, p < .0001).
Risk Assessment.
The risk assessment scores assigned to children in these cases were also examined (Table 1). The average composite score of all seven factors across cases was 18.61. The relationship between these risk assessment scores and comorbid neglect/DV was assessed. Because DV is one of the parental risk factors included in Factor 5 of the risk assessment tool, a revised risk assessment score was calculated prior to the examination of associations between comorbid DV and risk assessment scores. The mean risk assessment score for cases involving DV was 16.18, whereas the mean score for cases without DV was 14.03. An independent samples t-test utilizing the mean composite risk assessment scores indicated a significant relationship between comorbidity and the revised risk assessment score, t(70) = 2.68, p < .01.
The relationship between level of substantiation of DV and level of risk for each of the risk factors in the risk assessment tool was also analyzed. This level of substantiation was derived from the APS report and investigation and corresponded to the levels of substantiation for child maltreatment (substantiated, found and substantiated, some indication, and unsubstantiated). For the purpose of this analysis, the level of substantiation of DV was recoded as substantiated (included substantiated, found and substantiated, and some indication) or not substantiated (unsubstantiated). Also, the level of risk for the risk factors from the assessment tool was recoded as high or low risk. Chi-square analysis showed a significant association between substantiation of DV and level of risk for Factor 3 (severity of neglect and environmental conditions of the home), 2 (1, n = 76), 7.11, p < .01. Families in which DV was present were more likely to be rated at high risk for severe neglect/environmental conditions of the home. There was also a significant association between substantiation of DV and level of risk for Factor 7 (strength of the family's support system), 2 (1, n = 76) = 3.95, p < .05. Families in which DV was present were more likely to be rated at high risk due to limited social support.
Historical Factors.
Several historical factors relevant to the cases were examined. One such factor was the role of unreported DV in the family. Forty-seven percent of cases that had ongoing DV had not been previously reported to Adult or CPS for this violence. This case represented the first official report of DV despite documentation of an ongoing history of this violence. This unreported DV had a significant effect on caseworker assessment of risk, with total risk assessment score for children with unreported DV 16.86 compared to the total for children without unreported DV 13.54 (t(58) = 3.9, p < 0.01).
Problems in Children's Functioning.
The relationship between comorbid DV and problems in children's functioning was analyzed. There was an association between comorbid DV and problems in children' interpersonal functioning, 2 (2, n = 67) = 5.88, p = .053. Approximately 51.4% of children in cases with comorbid DV have problems in cognitive functioning, whereas 23.3% of those in cases without comorbid DV have such problems. Similarly, 70.3% of children in cases with comorbid DV have problems with interpersonal functioning, whereas 56.7% of children in cases without comorbid DV have such problems. There was also a significant relationship between comorbid DV and problems in physical functioning, 2 (2, n = 67) = 7.47, p < .05. However, in these cases, 18.9% of children in cases with comorbid DV have problems in physical functioning, whereas 36.7% of those in cases without comorbid DV have such problems. See Table 4 for summary statistics on problems in children's functioning by presence of DV.
|
Case Plan.
Factors related to the case plan were examined. First, the incorporation of DV into the case plan was evaluated. In 35.5% of cases where DV was present, steps were taken to address such concerns by incorporated related objectives into the case plan. The mean number of objectives by type of maltreatment was as follows: physical abuse, 0.95 (SD = 1.40); neglect, 2.61 (SD = 1.66); and DV, 0.98 (SD = 1.33). There was adequate differentiation between physical abuse and neglect in 71.7% of cases.
The taking of legal action by the worker or family was also of interest. In general, 65% of cases took some form of legal action around the DV issue. The frequency by type of intervention was as follows: court orders (45%), petitions (22%), Emergency Protective Orders (21%), Emergency Custody Orders (18%), Domestic Violence Orders (13%), and custody hearings (24%).
|
Discussion |
|---|
|
TOPMethodResultsDiscussionFundingReferences |
|---|
The findings of this research provide important information related to theoretical and practice issues in this area of child welfare. First, this study has revealed a high comorbidity rate (29%) of child neglect and DV. Although the literature has previously established such a relationship between child physical abuse and DV, only a small number of other studies have been reported in the literature examining the unique phenomenon of neglect (Beeman et al., 2001; Hartley, 2002, 2004; McGuigan & Pratt, 2001; Smokowski & Wodarski, 1996; Stewart et al., 1987; Weston & Colloton, 1993). In addition to this general comorbidity rate established through the larger sample, this study found that there is a significant positive correlation between level of substantiation of DV and the severity of neglect/environmental conditions of the home as measured by the risk assessment tool. Given the many findings that neglect produces more detrimental effects on children than other forms of maltreatment, this is an important contribution to the literature. Furthermore, the growing interest in dynamics of multiproblem families in the field of child welfare also makes these findings pertinent to service delivery.
There were several other significant contributions of this study related to general issues of child maltreatment and DV. One important finding of this study is the type of neglect that most frequently occurs when there is DV among parents in the home. Lack of safety, lack of supervision, emotional neglect, and refusal of treatment (e.g., mental health services) were the most common types of neglect present in cases with comorbid DV. Hartley (2004) also found that lack of supervision was the most common type of neglect when there is DV in the family.
Another contribution of this study is the finding that there is a significant positive correlation between comorbid DV and inadequate social support for the family as measured by the risk assessment tool. Families experiencing multiple problems (child maltreatment and adult violence) have weaker social networks or support systems. This finding is consistent with previous research that identified social isolation as a common risk factor between child neglect and DV (National Research Council, 1993). This preliminary evidence on primary risk factors, common types of neglect, and potential consequences for children is an important contribution to the current social service system.
Issues of Current Delivery System
Despite the evident need for services targeting these multiproblem families, there were a number of issues identified in the current system as exemplified by one metropolitan county in Kentucky. One strength of the current system was the taking of legal actions, such as EPOs. Approximately 65% of cases involved some type of legal action. Research has shown that EPOs are related to lower levels of DV (McFarlane et al., 2004). Hence, child protection workers appear to be assisting families with the necessary legal actions to maintain adult and child safety.
Although child welfare workers tended to take legal action in cases that were opened, analysis of the larger sample showed that cases with comorbid DV were actually less likely to be opened, raising concerns that child welfare workers did not act upon the potentially greater risk posed by multiple forms of family violence/maltreatment. This result differs from that of Beeman et al. (2001), who found that child protection workers were more likely to open cases with comorbid child maltreatment and DV.
Nevertheless, similar to Beeman et al. (2001), children in families with comorbid DV and child neglect in this research were rated at significantly greater risk with the child welfare agency's risk assessment tool. These children were also significantly more likely to experience problems in interpersonal functioning, a finding consistent with previous research on the harmful effects of exposure to DV (Groves, 1999; Kernic et al., 2003). Although workers recognized the greater risk for children with these multiple problems, they were no more likely to open these cases. As suggested by Schlonsky and Wagner (2005), there is a need to connect the assessment of risk with a contextual assessment of child and family functioning in the development of case plans.
Another weakness of the child welfare system was the lack of incorporation of DV into case plans. Despite higher child risk assessment scores in cases of comorbid DV and neglect, case objectives for DV were only included in 36% of cases with comorbid DV. Instead, more objectives tended to focus on neglect of the child. Although these issues are also vital, workers must recognize the importance of DV for children because of the numerous detrimental consequences of witnessing such violence (Caudill & Stoffel, 1998). For example, children exposed to DV exhibit symptoms of posttraumatic stress disorder (Groves, 1999) and cognitive limitations in areas such as problem solving (Hughes, 1997).
A final weakness identified by this study was the high rate of unreported DV. Forty-seven percent of cases involved previously unreported DV. The current case represented the first official report of DV, but there was documentation in the child welfare record that DV had been an issue for this family in the past. Hence, a substantial number of families are not being serviced by this system. Statistical analysis revealed that children in such homes had more potential risk factors as identified by this agency's risk assessment tool. Thus, greater measures must be taken to promote earlier identification of homes where children are being exposed to DV.
Limitations of Current Study
There were several limitations for the current study. First, the sample was restricted to cases that involved child neglect in order to determine an accurate comorbidity rate of this specific type of maltreatment. However, in restricting the sample in this way, comparisons could not be made between cases with various types of child maltreatment (e.g., physical abuse only, neglect only). A related sampling limitation was the inability to confirm that the secondary sample was representative of the primary sample along key demographic variables. The state data report generated for the primary sample did not include these demographic variables. Nevertheless, the demographics of the secondary sample are comparable to those reported at the national level (U.S. Department of Health and Human Services, Administration on Children, Youth and Families, 2005).
Second, this study was cross-sectional in nature, rendering it impossible to evaluate the effectiveness of interventions or outcomes over time. This study simply provides a "snapshot" of multiproblem families and the public child welfare system's response. Third, this study relied upon child welfare system's determination of substantiation of child maltreatment and DV as key variables. Substantiation is simply a legal indication that the allegations of maltreatment have been found by the CPS worker to be true. Drake, Johnson-Reid, Way, and Chung (2003) found that unsubstantiated cases are at an often similarly high risk of recidivism to substantiated cases in child welfare when followed over several years.
Furthermore, the consensus-based risk assessment tool used by the child welfare agency in this study did not have reported reliability and validity data. Knoke and Trocme (2005) identified two major factors driving the need for risk assessment tools with sound psychometric properties: the inadequacy of resources to meet an increasing demand for services in child welfare and concerns regarding the reliability/validity of worker judgments of risk. In the review of psychometric data on a number of risk assessment tools provided by these authors, although most lacked strong reliability and validity, the actuarial-based tools tended to perform better than consensus-based tools. When Baird and Wagner (2000) compared an actuarial approach to two consensus-based tools, they found the former differentiated between levels of risk much better than the latter. The findings derived from the risk assessment data in the current study should be considered in light of these concerns regarding the risk assessment tool. Hence, future research should examine the reliability and validity of this state's consensus-based tool and perhaps consider the use of an actuarial model for more accurate prediction of risk of repeat maltreatment.
Practice and Policy Implications
Findings from this research bear numerous implications for child welfare policy and practice. The equally high comorbidity rate of neglect and DV as compared with other forms of maltreatment and DV suggests a need for dissemination of these findings to child welfare workers. These workers need to be made aware of the significant detrimental effects of both neglect and DV on children, as well as the overlap of these effects. Furthermore, child welfare workers need to be informed regarding the likelihood of co-occurrence of these types of abuse so that they will assess for DV in CPS investigations and take appropriate action based upon the laws and resources of their given practice settings. This research found that children are at greater risk of physical abuse, lack of safety, and refusal of treatment (which include mental health and medical services) when there is DV in the home. These results are consistent with Hartley (2002), who found the most common type of neglect among families with DV was lack of supervision. CPS workers should be trained to be vigilant for these particular issues if there is DV present in the home.
The issue of not opening cases with multiple forms of family violence is troubling. However, more research is needed on the workers' decision-making process in cases where both adult violence and child maltreatment are present. Perhaps the DV in unopened cases was seen as less severe than the violence in which cases were opened. Research by Johnson (2005) finds that there are at least two distinct forms of DV. The most severe form he calls intimate terrorism. The less severe, but more prevalent form he calls situational couple violence. Assessment tools need to be developed to facilitate research on the differential effects of these two distinct forms of DV on child functioning and outcomes as well as which treatment protocols are best for each type of couple and family. These tools should then be utilized by social service workers to determine the best course of action when DV is present in homes inhabited by children.
Another area of need is the early identification of DV in families. The finding in the current research that unreported DV has deleterious effects on children suggests that earlier intervention is needed for such families. This could occur through several means. First, adequate assessment of DV by child protection workers should occur in every case. There are a several strong standardized tools available for the assessment of DV (Dutton & Kropp, 2000). Such tools should become a standard part of child welfare practice. Although child welfare assessments often include a brief assessment of the presence of DV and risk of harm to the child (Petrucci & Mills, 2002), there needs to be a more thorough consideration of the extent of violence and impact on the child.
Second, laws surrounding reporting of DV should be examined to determine whether they consider the best interests of the children. For example, the state of Kentucky has a mandatory reporting law for DV among married couples. Qualitative research by this team found that victims view this law as beneficial to themselves and their children (Antle, Barbee, Bledsoe, Daniel, & Yankeelov, 2007). Third, there is a need for collaboration among community professionals such as the courts, law enforcement, schools, and others (Jaffe, Baker, & Cunningham, 2004). One such example would be cross-reporting of DV by law enforcement to CPS. The current practice in Kentucky is for law enforcement to document whether children are present during the alleged DV incident in their formal report to APS. APS then makes a determination whether or not to refer to CPS based upon this report. Hence, although law enforcement is required by the mandatory reporting law in Kentucky to report DV to APS and best practice dictates documentation of the presence of children, there is no formal law or policy requiring a CPS report in every case. If such cross-reporting by law enforcement does not occur, children may be left at risk of harm through exposure to DV and greater risk of child maltreatment. The early identification of families with DV may prevent some problems experienced by children when they witness DV, but further research is needed in this regard. Some have cautioned the field of the risks for families when CPS becomes involved (e.g., Waldfogel, 2000). The protective benefits of child welfare system involvement must outweigh these risks to merit such early intervention efforts.
In addition, CPS workers need to be trained to incorporate DV into their case plans. In the current study, very few objectives were designed to address the issue of DV. Workers need to be made aware of the importance of this issue and given specific training on ways to address family violence. Treatment of children in violent families has primarily occurred in the context of psychotherapy (Grusznski, Brink, & Edelson, 1988; Jaffe, Wolfe, & Wilson, 1990). However, child welfare workers are in a unique position to address these multiproblem families because of the systemic nature of their interventions. To address these issues of the interface between child abuse, neglect and DV, Kentucky now trains all CPS workers on this interface and how to handle such multiproblem cases at the investigative, assessment, and treatment phases. These workers are required to develop case plan objectives for the entire family, as well as specific to the individual adults and children in the family.
Finally, the comorbidity of these issues suggests a need for the development of specialized services for multiproblem families. After this study was conducted, the county tested the use of specialized teams for children involved in DV situations. When a report of DV was made to APS, if children were involved, the case was sent to a team that included both Adult and CPS workers. The same was true for cases where DV is detected in CPS investigations. Hence, attention could be given to both issues of marital violence as well as its effects on children. Similar programs or approaches have been developed throughout the country in order to more effectively serve these families (e.g., Mills et al., 2000). Mills and Yoshihama (2002) describe the positive impact of training in DV for child protection workers on their willingness and confidence to respond to DV issues in their child welfare cases. Petrucci and Mills (2002) also confirm that 90% of child welfare agencies have incorporated an assessment of the risk of DV to the child into their standard child abuse assessments.
There is also a need for specialized services within the community for families who are not served by the child protection system. A large percentage of cases (82%) with some indication of neglect and DV were not opened in the current study. These families demonstrated some risks but did not meet the legal definition of abuse or neglect to warrant a case opening. Kentucky has developed a category of classification for such families to reflect this situation called "families in need of services." This innovation has been called for by Waldfogel (2000) and others. However, research on this Kentucky practice is missing and needed on whether this classification is effective in preventing escalation of family problems and/or future maltreatment. These families need support from community services in order to prevent significant protection issues. Therefore, community practitioners need to be trained in the prevalence and impact of comorbid child maltreatment and DV, appropriate legal and service responses to DV, and interventions for multiproblem families in crisis.
In conclusion, this study makes several significant contributions to the understanding of and practice with multiproblem families. First, this research provides confirmation of a substantial comorbidity rate of the unique phenomenon of child neglect and DV. Children in homes where there is DV are at greater risk of specific types of neglect and problems in functioning. These multiproblem families also have significantly lower levels of social support, a potential point of intervention for service providers who want to minimize risks of future maltreatment. Other service implications of this research include the need to address DV within the context of child maltreatment case plans and the need for better assessment and early intervention strategies for multiproblem families.
|
Funding |
|---|
|
TOPMethodResultsDiscussionFundingReferences |
|---|
Eastern Kentucky University, as part of the Child Welfare Training Assessment Project.
|
Acknowledgments |
|---|
Conflict of Interest: None declared.
|
Footnotes |
|---|
1 Comparable demographics were not available for the primary sample (N = 2,350) due to restrictions in the state data report. The inability to confirm the equivalence of the secondary sample to the primary sample along these demographics represents a limitation of this study. However, the demographics of the secondary sample are comparable to those reported in national studies (U.S. Department of Health and Human Services, Administration on Children, Youth and Families, 2005).
|
References |
|---|
|
TOPMethodResultsDiscussionFundingReferences |
|---|
-
Antle BF, Barbee AP, Bledsoe L, Daniel AK, Yankeelov PA. A qualitative evaluation of the effects of mandatory reporting on victims and their children (2007) Unpublished manuscript.
Baird C, Wagner D. The relative validity of actuarial and consensus-based risk asessment systems. Children and Youth Services Review (2000) 22:839–871.[CrossRef][ISI][find it @ Stanford]
Beeman SK, Hagemeister AK, Edleson JL. Case assessment and service receipt in families experiencing both child maltreatment and woman battering. Journal of Interpersonal Violence (2001) 16:437–458.
Caudill TL, Stoffel E. Effects of domestic violence on children (1998) Cincinnati, OH: National Conference on Child Abuse and Neglect.
Centers for Disease Control and Prevention. Costs of intimate partner violence against women in the United States (2003) Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Retrieved September 15, 2005, from http://www.cdc.gov/ncipc/pub-res/ipv_cost/ipv.htm.
Chalovich PS. The effects of chronic stress on cognitive functioning and academic achievement of homeless children. Dissertation Abstracts International: Section B: The Sciences and Engineering (2004) 64(10-B):5209.[find it @ Stanford]
Crouch JL, Milner JS. Effects of child neglect on children. Criminal Justice and Behavior (1993) 20:49–65.[Abstract]
Cummings EM, Davies P. Children and marital conflict: The impact of family dispute on resolution (1994) New York: Guilford.
Downey G, Coyne JC. Children of depressed parents: An integrative review. Psychological Bulletin (1990) 108:50–76.[CrossRef][ISI][Medline][find it @ Stanford]
Drake B, Jonson-Reid M, Way I, Chung S. Substantiation and recidivism. Child Maltreatment (2003) 8:248–260.
Dutton DG, Hart SD. Evidence for long-term, specific effects of childhood abuse and neglect on criminal behavior in men. International Journal of Offender Therapy and Comparative Criminology (1992) 36:129–137.[Abstract]
Dutton DG, Kropp PR. A review of domestic violence risk instruments. Trauma, Violence and Abuse (2000) 1:171–181.
English DJ, Marshall DB, Stewart AJ. Effects of family violence on child behavior and health during early childhood. Journal of Family Violence (2003) 18:43–57.[CrossRef][ISI][find it @ Stanford]
Fisher L, Kokes RF, Harder D, Jones JE. Child competence and psychiatric risk: IV. Summary and integration of findings. Journal of Nervous and Mental Disease (1980) 168:353–355.[ISI][Medline][find it @ Stanford]
Fromm S. Total estimated cost of child abuse and neglect in the United States: Statistical evidence (2001) Chicago, IL: Prevent Child Abuse America. Retrieved December 9, 2006, from http://member.preventchildabuse.org/site/DocServer/cost_analysis.pdf?docID=144.
Gaudin JM. Effective intervention with neglectful families. Criminal Justice and Behavior (1993) 20:66–89.[Abstract]
Gauthier L, Stollak G, Messe L, Aronoff J. Recall of childhood neglect and physical abuse as differential predictors of current psychological functioning. Child Abuse and Neglect (1996) 20:549–559.[CrossRef][ISI][Medline][find it @ Stanford]
Gellert GA. Confronting violence (1997) Boulder, CO: Westview Press.
Groves BM. Mental health services for children who witness domestic violence. The Future of Children (1999) 9:122–132.[CrossRef][Medline][find it @ Stanford]
Grusznski RJ, Brink JC, Edelson JL. Support and education groups for children of battered women. Child Welfare (1988) 67:431–444.[ISI][Medline][find it @ Stanford]
Hartley CC. The co-occurrence of child maltreatment and domestic violence: Examining both neglect and child physical abuse. Child Maltreatment (2002) 7:349–358.
Hartley CC. Severe domestic violence and child maltreatment: Considering child physical abuse, neglect, and failure to protect. Children and Youth Services Review (2004) 26:373–392.[CrossRef][ISI][find it @ Stanford]
Herzerberger SD. Violence within the family (1996) Boulder, CO: Westview Press.
Hughes HM. Research concerning children of battered women: Clinical implications. Journal of Aggression, Maltreatment and Trauma (1997) 1:225–244.[CrossRef][find it @ Stanford]
Jaffe PG, Baker LL, Cunningham AJ. Protecting children from domestic violence: Strategies for community intervention (2004) New York: Guilford Press.
Jaffe PG, Crooks CV. Understanding women's experiences parenting in the context of domestic violence: Implications for community and court-related service providers (2005) Retrieved December 13, 2006, from http://www.vaw.umn.edu/documents/commissioned/parentingindv/parentingindv.html.
Jaffe PG, Suderman M. Child witnesses of woman abuse: Research and community response. In: Understanding partner violence—Stith SM, Straus MA, eds. (1995) Minneapolis, MN: National Council on Family Relations. 213–222.
Jaffe PG, Wolfe DA, Wilson SX. Children of battered women (1990) London: Sage.
Johnson MP. Apples and oranges in child custody disputes: Intimate terrorism vs. situational couple violence. Journal of Child Custody (2005) 2(4):43–52.[CrossRef][find it @ Stanford]
Kernic MA, Wolf ME, Holt VL, McKnight B, Huebner CE, Rivara FP. Behavioral problems among children whose mothers are abused by an intimate partner. Child Abuse and Neglect (2003) 27:1231–1246.[CrossRef][ISI][Medline][find it @ Stanford]
Knoke D, Trocme N. Reviewing the evidence on assessing risk for child abuse and neglect. Brief Treatment and Crisis Intervention (2005) 5:310–327.
McFarlane J, Malecha A, Gist J, Watson K, Batten E, Hall I, et al. Protection orders and intimate partner violence: An 18-month study of 150 black, Hispanic, and white women. American Journal of Public Health (2004) 94:613–618.
McGuigan WM, Pratt CC. The predictive impact of domestic violence on three types of child maltreatment. Child Abuse and Neglect (2001) 25:869–883.[CrossRef][ISI][Medline][find it @ Stanford]
Mills LG, Friend C, Fleck-Henderson A, Krug S, Magen RH, Thomas RL, et al. Child protection and domestic violence: Training, practice, and policy issues. Children and Youth Services Review (2000) 22:315–332.[CrossRef][ISI][find it @ Stanford]
Mills LG, Yoshiama M. Training, children's service workers in domestic violence assessment and intervention: Research findings and implications for practise. Children and Youth Services Review (2002) 24(8):561–581.[CrossRef][ISI][find it @ Stanford]
National Research Council. (Ed.). Understanding child abuse and neglect (1993) Washington, DC: National Academy Press.
Petrucci CJ, Mills LG. Domestic violence assessment: Current practices and new models for improved child welfare interventions. Brief Treatment and Crisis Intervention (2002) 2:153–172.[Abstract]
Renner LM, Slack KS. Intimate partner violence and child maltreatment: Understanding intra- and intergenerational connections. Child Abuse and Neglect (2006) 30:599–617.[CrossRef][ISI][Medline][find it @ Stanford]
Ross SM. Risk of physical abuse to children of spouse abusing parents. Child Abuse and Neglect (1996) 20:589–598.[CrossRef][ISI][Medline][find it @ Stanford]
Rossman BB, Rosenberg MS. Psychological maltreatment: A needs analysis and application for children and violent families. Journal of Aggression, Maltreatment and Trauma (1997) 1:245–262.[CrossRef][find it @ Stanford]
Rycus JS, Hughes RC. Issues in risk assessment in child protective services: A policy white paper (2003) Columbus, OH: North American Resource Center for Child Welfare Center for Child Welfare Policy.
Samples P. Total family care for abusive families. Daycare and Elderly Education (1981) 21:44–47.[find it @ Stanford]
Schlonsky A, Wagner D. The next step: Integrating actuarial risk assessment and clinical judgment into an evidence-based practice framework in CPS case management. Children and Youth Services Review (2005) 27:409–427.[CrossRef][ISI][find it @ Stanford]
Smokowski PR, Wodarski JS. The effectiveness of child welfare services for poor, neglected children: A review of the empirical evidence. Research on Social Work Practice (1996) 6:504–528.
Stets JE. Verbal and physical aggression in marriage. Journal of Marriage and the Family (1995) 52(2):501–514.[find it @ Stanford]
Stewart C, Senger MM, Kallen D, Scheurer S. Family violence in stable middle-class homes. Social Work (1987) 32(6):529–531.[ISI][find it @ Stanford]
Tjaden P, Thoennes N. Extent, nature, and consequences of intimate partner violence: Findings from the National Violence Against Women Survey (2000) Washington, DC: U.S. Department of Justice. Retrieved December 9, 2006, from http://www.ncjrs.gov/pdffiles1/nij/181867.pdf.
Tolman RM, Edelson JL. Intervention of men who batter: A review of research. In: Understanding partner violence—Stith SM, Straus MA, eds. (1995) Minneapolis, MN: National Council on Family Relations. 262–273.
U.S. Department of Health and Human Services, Administration on Children, Youth, and Families. Child maltreatment 2002 (2004) Washington, DC: U.S. Government Printing Office. Accessed July 2007.
U.S. Department of Health and Human Services, Administration on Children, Youth and Families. Child maltreatment 2003: Reports from the states to the National Child Abuse and Neglect Data System (2005) Washington, DC: U.S. Government Printing Office. Retrieved http://www.acf.dhhs.gov/programs/cb/pubs/cm03/summary.htm.
U.S. Department of Justice, Office of Justice Programs, National Institute of Justice. Victim costs and consequences: A new look (1996) Washington, DC: U.S. Government Printing Office. Retrieved December 10, 2006, from http://www.ncjrs.gov/pdffiles/victcost.pdf.
Waldfogel J. Reforming child protective services. Child Welfare (2000) 79:43–57.[ISI][Medline][find it @ Stanford]
Walker LE, Browne A. Gender and victimization by inmates. Journal of Personality (1985) 53:179–195.[CrossRef][ISI][Medline][find it @ Stanford]
Weissman M, Paykel ES. The depressed woman: A study of social relationships (1974) Chicago, IL: University of Chicago Press.
Weston JA, Colloton M. A legacy of violence in nonorganic failure to thrive. Child Abuse and Neglect (1993) 17:709–714.[ISI][Medline][find it @ Stanford]
Williams KR. Social sources of marital violence and deterrence: Testing an integrated theory of assaults between partners. Journal of Marriage and the Family (1992) 54(3):620–629.[CrossRef][ISI][find it @ Stanford]
Wolfe DA, Jaffe P, Wilson SK, Zak L. Children of battered women: The relation of child behavior to family violence and maternal stress. Journal of Consulting and Clinical Psychology (1985) 53:657–665.[CrossRef][ISI][Medline][find it @ Stanford]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||