Profiles of Child-Welfare-Involved Caregivers Identified by Caseworkers as Having a Domestic Violence Problem: Then and Now

Abstract

Over the past 10 years, there has been a significant decline in the rate of domestic violence (DV) experienced among caregivers involved with the child protective services (CPS) system. It is unclear whether this shift is related to changes in caregiver characteristics. Furthermore, despite evidence that suggests CPS caseworkers poorly identify DV and fail to link families to DV services, limited research exists on whether the current CPS interventions that are known to improve caseworkers’ DV identification will also improve chances for DV service receipt. The present study uses data from the first and second cohorts of the National Survey of Child and Adolescent Well-Being (NSCAW) to compare differences in demographic characteristics and DV experiences between caregivers in NSCAW I (1999-2000; n = 2,758) and NSCAW II (2008-2009; n = 2,207). We also examine the effects of CPS interventions on NSCAW II caregivers’ receipt of DV services external to the CPS agency (i.e., external DV services). Caregivers with caseworker reports of active DV in NSCAW I and II were similar in their demographic characteristics and external DV service experiences. However, caregivers in NSCAW II generally reported lower rates of victimization for specific types of violence than NSCAW I caregivers. Finally, caregivers with active DV involved with an agency that used DV assessment tools were 7.03 times more likely to receive external DV services than those in agencies without DV tools (95% confidence interval [CI] = [2.33, 21.22]). Whereas caregivers in agencies that sometimes (odds ratio [OR] = 0.16, 95% CI = [0.03, 0.99]) or always (OR = 0.15, 95% CI = [0.02, 0.98]) had a DV specialist available were less likely to receive external DV services than those in an agency that never/rarely had a DV specialist available. We recommend CPS agencies use specialized assessment tools to identify DV-affected families and link them to services. Additional research is needed to understand what types of services DV specialists offer within CPS agencies and whether these services meet caregivers’ needs.

Keywords

domestic violence children exposed to domestic violence intervention/treatment

The most recent study to examine the prevalence of domestic violence (DV) among child protective services (CPS) cases used the two cohorts of the National Survey of Child and Adolescent Well-Being (NSCAW I and NSCAW II) to compare rates of self-reported DV (Casanueva, Smith, Ringeisen, Dolan, & Tueller, 2014). This study found a significant decline in mothers who reported DV between the time of the first (1999-2000) and second NSCAW (2008-2009). Although this study compared DV prevalence between the two cohorts, it failed to examine whether unique characteristics may exist among caregivers with identified DV in NSCAW II compared with those in NSCAW I. Knowing such information may be helpful in understanding potential reasons for the significant decrease in DV prevalence. Furthermore, although there is evidence that suggests under-identification of DV may lead to difficulty with CPS caseworkers linking families to DV services, less is known about the effects of CPS interventions that may be helpful in improving caseworkers’ DV identification on CPS-involved caregivers’ receipt of DV services. The current study aims to fill these gaps in knowledge.

National Rates of DV Prevalence Among CPS Cases

Results from the study conducted by Hazen, Connelly, Kelleher, Landsverk, and Barth (2004) using the NSCAW I dataset revealed 29.0% of all children involved with child welfare (who remained in the home following the maltreatment investigation) had a female permanent caregiver who reported experiencing physical violence perpetrated by a spouse or partner during the past year, and 44.8% reported this type of abuse in their lifetime. The most frequently reported types of violence were being pushed, grabbed, shoved, or hit, or attempting to be hit with an object by a partner.

Using the same dataset to examine DV prevalence among female permanent caregivers of children who remained in their home following the maltreatment, Kohl, Barth, Hazen, and Landsverk (2005) found a substantially lower rate of DV identified by CPS caseworkers. Caseworkers identified DV in only 12% of these CPS cases. Because caregivers in NSCAW I reported higher rates of DV than that identified by CPS caseworkers, the authors conclude CPS caseworkers may be under-identifying DV in their cases. This conclusion reinforces other research using county-level data (Coohey, 2007), as well as study findings from current research using the second cohort of NSCAW (Casanueva et al., 2014).

When comparing DV prevalence of mothers who reported DV in NSCAW I with those who reported DV in NSCAW II, Casanueva and her colleagues (2014) found that at NSCAW I baseline, 28.9% of mothers reported DV, whereas only 24.7% reported DV at NSCAW II baseline. This represents a 15% decline in DV between 1999-2000 and 2008-2009.

DV and CPS

When compared with CPS-involved families that do not experience DV, CPS-involved families that experience DV are at greater risk of poorer case outcomes. Researchers have found CPS-involved children in families that experience DV were more likely to be placed in out-of-home care (Ogbonnaya & Guo, 2013), re-enter foster care (Hess, Folaron, & Jefferson, 1992), and achieve lower rates of reunification with birth parents (Marsh, Ryan, Choi, & Testa, 2006) than children in families that did not experience DV. In addition, children in families affected by DV may have poorer psychosocial outcomes, such as internalizing (e.g., depression, anxiety, and somatic symptoms) and externalizing (e.g., aggression, delinquency, and conduct problems) behaviors, posttraumatic stress disorder, low self-esteem, low levels of perceived competence, and self-blame (Wolfe, Crooks, Lee, McIntyre-Smith, & Jaffe, 2003). Thus, it is important that these families are detected and receive the help that they need to address their DV situation and improve the well-being of their children. Recognizing the risks to children in families with DV, increasing attention is being given to CPS interventions that aim to improve the response of child welfare workers to the co-occurrence of DV and child maltreatment. Such efforts may be effective, as suggested by findings from the Casanueva et al. (2014) study showing an overall decline in the number of caregivers with self-reported DV.

CPS Interventions Used to Address DV

One approach to improving CPS’ response to DV is to implement strategies designed to aid in the identification of DV and improve the outcomes of families affected by DV. Risk assessment tools to help identify these families are one such strategy. DV screening instruments have proven successful in increasing CPS’ DV detection rates and intervention service linkage (Magen, Conroy, Hess, Panciera, & Simon, 2001; Rivers, Maze, Hannah, & Lederman, 2007). Magen et al. (2001) conducted a 6-month pilot study to assess the effects of a DV questionnaire used during child maltreatment investigations in New York City. Using the questionnaire, they found that 35 of 125 caregivers in current relationships, or in a relationship during the past 6 months, reported DV. Among these 35 caregivers, only 17 (48.6%) had previously been reported as experiencing DV. Thus, the use of a DV questionnaire helped to identify several more cases by eliciting new caregiver reports of DV. In addition, the study revealed a positive relationship between caseworkers’ DV detection and victimized caregivers’ DV service referral and use. Once identified, caseworkers were likely to refer DV-affected families to preventive services and/or provide them with assistance obtaining protection orders.

A second approach is the incorporation of DV specialists into CPS practice. Rather than solely relying on CPS caseworkers to screen for DV and link DV-affected families to services, Aron and Olson (1997) suggested that the most effective way to address DV issues within the CPS population is through hiring DV specialists to consult with CPS caseworkers and model best-practice approaches. This suggestion is in accordance with the Greenbook Initiative, an effort by the National Council of Juvenile and Family Court Judges to create DV-related policy and practice for child welfare agencies (Banks, Hazen, Coben, Wang, & Griffith, 2009). Collaborative efforts between child welfare agencies and DV specialists have proven successful in helping clients understand DV and its impact, helping with referrals to other services such as DV shelters, providing direct advocacy support to clients, and supporting clients during court hearings (e.g., Aron & Olson, 1997; Rosewater, 2008). For example, when Greenbook principles and recommendations were implemented in five CPS agencies, Banks, Landsverk, and Wang (2008) found a significant increase in service referrals from Time 1 to Time 3 for victims of DV (35%-65%, p < .000).

Goals and Overview of Current Study

This study is an extension of Hazen et al.’s (2004) and Kohl et al.’s (2005) studies, which used data from the NSCAW I to examine DV prevalence and linkage to DV services, respectively. The current study compares demographic characteristics and DV experiences (i.e., types of physical abuse and DV service linkage) among families with caseworker-reported DV that were recently involved with CPS to families with caseworker-reported DV involved with CPS over a decade ago. In addition, this study adds to the literature on CPS’ response to DV by using the NSCAW II dataset to investigate the effects of current CPS DV interventions on caregivers’ use of DV services external to the CPS agency. Analysis of the NSCAW II allows for the examination of practice behaviors following nearly a decade of attention given to CPS agencies’ response to the co-occurrence of DV and child maltreatment. This study further expands upon the 2005 study by assessing how DV service outcomes vary depending on the following interventions, not measured in NSCAW I: (a) DV assessment tools and (b) DV specialists. Specifically, the study answers the following questions:

How do demographic characteristics of caregivers with caseworker-identified DV in NSCAW I compare with demographic characteristics of caregivers with caseworker-identified DV in NSCAW II?

What is the estimated change in rates of victimization for specific types of physical DV among caregivers involved with CPS during the NSCAW I study compared with those involved during the NSCAW II study?

How do caregivers’ report of DV service need, referral, and receipt vary when comparing reports among caregivers with caseworker-identified DV in NSCAW I with reports among caregivers with caseworker-identified DV in NSCAW II?

What CPS interventions measured in NSCAW II are likely to increase chances for caregivers’ receipt of DV services external to the CPS agency?

Method

This study involved secondary data analysis of the NSCAW I and NSCAW II. The NSCAW is a nationally representative sample of children and their families investigated by CPS for potential child abuse or neglect. The NSCAW sample included children who received ongoing services as well as children who did not receive services because their cases were not substantiated or because the CPS investigation determined services were not required. When a family had more than one child involved in an investigation, one of the children was randomly selected to be the study participant. Finally, CPS cases in which the study child was not the focus of an abuse or neglect investigation (e.g., cases investigating other family members, or cases investigating allegations other than abuse/neglect of the study child) were excluded (Administration for Children & Families, 2005, 2011).

NSCAW I Sample Design

The NSCAW I consists of a national probability sample of 5,501 children from families that were referred to and investigated by CPS for potential child abuse or neglect during a 15-month period (October 1999 to December 2000). Children had to be no more than 14 years old (aged 0-14 years) to be eligible for the NSCAW I study. The children were selected from 100 primary sampling units (PSUs), representing CPS agencies located across counties or groups of counties within the United States. During the time of the NSCAW I, four states (representing eight PSUs) refused to participate in the study (referred to as “agency first contact” states).

The NSCAW I data were gathered using information from face-to-face interviews with children, their caregivers, CPS caseworkers, and directors. Data were collected across four waves of follow-up to the CPS investigation: baseline (Wave 1), 12-month follow-up (Wave 2), 18-month follow-up (Wave 3), and 36-month follow-up (Wave 4). Given the focus of this study is on DV, we used a sample of 2,758 female caregivers with (a) children remaining in the home following a maltreatment investigation and (b) a caseworker report of active of DV (yes or no). In addition, to compare results with NSCAW II findings, we only used data from Waves 1 and 3. All NSCAW I analyses are based on caregiver and caseworker reports. Our study excluded 2,218 cases with (a) male caregivers or (b) without a caseworker report of whether DV was a risk, and (c) 525 cases with missing information on key variables. Consequently, our study findings may be generalized only to female caregivers whose children remain in the home following a CPS child maltreatment investigation.

NSCAW II Sample Design

NSCAW II used the same two-stage stratified sample design as NSCAW I to identify CPS-involved children; however, the target population is slightly different. Specifically, the NSCAW II included (a) 5,872 children aged birth to 17.5 years (upper age range expanded from 14 to 17.5) who were subject to a CPS investigation from February 2008 through May 2009; and (b) four new agency first contact states (representing nine additional PSUs excluded). To date, three waves of data have been collected—baseline (Wave 1), 18-month follow-up (Wave 2), and 36-month follow-up (Wave 3).

We used NSCAW II data collected from CPS caseworkers, directors, and caregivers during Waves 1 and 2. Similar to our NSCAW I sample, the NSCAW II sample in the current study is focused on female caregivers whose children remained in home following the maltreatment investigation with caseworkers’ reports of active DV (yes or no). To compare results with those in NSCAW I, this sample was limited to children aged 0 to 14 and included 2,207 female caregivers with children remaining in the home following an investigation. Overall, we excluded 3,053 caregivers because they were male or had cases without a report of whether DV was a risk, and 612 cases with missing information on key variables or with children older than 14 years.

Measures

With the exception of DV service reports, all study variables were measured using Wave 1 data. Measures related to DV service in NSCAW I were assessed using both Waves 1 (baseline) and 3 (18-month follow-up) data, whereas NSCAW II DV service data were analyzed using Wave 1 (baseline) and Wave 2 (18-month follow-up) data. Unless otherwise stated, all variables in NSCAW I and II were measured using the exact same variables.

DV variables

Caregiver DV reports

Caregiver reports of DV were measured using the Conflict Tactics Scale 1 (CTS-1; Straus, 1979) physical assault subscale in NSCAW I and CTS-2 (Straus, 1990) physical assault subscale in NSCAW II. In both NSCAW studies, the CTS was administered via audio computer-assisted self-interviews. Because the NSCAW I and II used different versions of the CTS, we could only compare CTS items worded exactly the same. Therefore, we compared six CTS items in NSCAW I and II that measure physical abuse behaviors between intimate partners: having something thrown, slapping, choking, beating up, threatening with a knife or gun, and using a knife or gun. Caregivers responded yes or no to being a victim of the physical abuse behaviors during (a) the past year (i.e., recent DV) or (b) a time prior to the past year (i.e., lifetime DV).

Caseworker reports of DV

Caseworker reports of DV were measured using a risk assessment completed by caseworkers. The risk assessment consisted of several questions used to determine how safe children were in their current environments (i.e., how likely the child is to suffer maltreatment in the immediate future). As part of the risk assessment, caseworkers were asked the following questions regarding DV: (a) “At the time of the investigation, was there active DV?” (yes or no) and (b) “Was there a history of DV against the caregiver?” (yes or no). Risk assessments were conducted 2 to 6 months after the investigation completion date, allowing enough time to determine whether potential risks existed. All caseworker interviews were conducted in-person at the CPS agency to facilitate access to confidential case records.

DV services

Data were collected from caregivers to assess whether they were (a) referred to DV services external to the CPS agency and (b) received external DV services. Specifically, all caregivers were asked the following questions regarding DV service referral and receipt, respectively: “In the last 12 months, have you been referred to domestic violence services, like a battered women’s shelter or a program to help you deal with an abusive partner?” (yes or no); and “In the last 12 months, have you stayed in a battered women’s shelter or received any other DV services to help you deal with an abusive partner?” (yes or no). If caregivers responded they did not receive DV services, they were asked to rate their need for DV services (a lot, somewhat, a little, or not at all). Responses were dichotomized into DV service need or no DV service need. Because all study questions pertaining to DV services asked about services external to the CPS agency, from this point forward, we refer to DV services as external DV services.

As previously mentioned, all external DV service reports were measured at both baseline and 18-month follow-up. We used both waves of data because it was possible some caregivers were referred to external DV services but did not receive such services until after the baseline interview.

DV assessment tools

Information about agencies’ DV-related practices was gathered from a field representative designated by the agency and applied at the child-level. This designated person was typically the county or regional director. To assess whether the agency used a DV assessment tool, directors were asked, “Please tell me if your agency has used a standardized DV assessment instrument during the investigation” (yes or no). This variable was only measured in NSCAW II.

DV specialists

Directors were asked the following question about their use of DV specialists:

We would like to know about the availability of various professionals or groups to assist in your investigation process. Assistance may include a variety of activities, such as providing information, consultation, or participation in the investigation. How available are DV specialists for your investigations?

Response items for this question included always, sometimes, rarely, or never. Because relatively few directors in our sample responded that they rarely (n = 238) or never (n = 31) used a DV specialist, these two categories were combined. This measure only exists in NSCAW II.

Demographic variables

We used information gathered from caregivers and caseworkers on a range of demographics as statistical control variables in the study analyses. The selection of caregiver variables was based on empirical evidence that suggests these variables are important correlates of DV among CPS-involved families (e.g., Hazen et al., 2004; Kohl et al., 2005). These variables include physical health, mental health, active substance use, previous substantiated case with CPS, prior reports to CPS, child in home with child welfare services, race, age, marital status, education level, number of children in household, employment status, trouble meeting basic needs, and urbanicity. Caregivers self-reported on their race, age, marital status, education level, number of children, and employment status. Substance use, mental health, previous substantiated case with CPS, and prior reports to CPS were based on caseworker risk assessment reports. A caregiver was determined as having a mental health problem if her caseworker reported on the risk assessment that she had a serious mental health problem during the time of the investigation (yes or no). Substance use was examined using caseworker risk assessment reports of whether there was active alcohol abuse (yes or no) or drug abuse (yes or no) by the caregiver at the time of the investigation. Whether a caregiver had a child that received in-home services and/or lived in an urban area were general variables available in NSCAW.

Physical health

Caregiver physical health was measured using Short Form Health Survey (SF-12; Ware, Phillips, Yody, & Adamczyk, 1996). The scale consists of items measuring the following health characteristics during the past 4 weeks: general health; health limitations related to moderate activities, stair climbing, accomplishments in regular daily activities, and work activities; and pain interference with normal work. Higher scores indicate better health. Prior research has found the SF-12 to have acceptable test–retest reliability for physical health (.89) and has supported its validity (Ware et al., 1996).

Analytic plan

Descriptive statistics and bivariate analyses (i.e., F tests and t tests) were used to summarize caregivers’ characteristics and DV reports, comparing findings in NSCAW I with those in NSCAW II. In addition, bivariate analyses were conducted with NSCAW II data to explore the relationship between caregivers with reports of active DV and caregivers without reports of active DV. Finally, a logistic regression analysis was conducted to assess the influence of each DV intervention (DV assessment tools or DV specialists) measured in NSCAW II on caregivers’ receipt of external DV services.

All analyses were conducted using Stata 12 software (2011). Although unweighted sampling sizes are reported in the tables, analyses were conducted using weights. The weights used varied depending on the wave of data collection. Therefore, because 18-month follow-up data were included in measures involving external DV service, sample sizes for analyses involving external DV services slightly differed from analyses not involving external DV services. Both NSCAWs use a complex weighting strategy accounting for stratification, clustering, weighting, and oversampling of some subgroups so that PSUs are nationally representative. Applying the weights resulted in study findings that generalize to CPS-involved families with children in the home following the child maltreatment investigation.

Calibration weights

To compare NSCAW I and II populations and test for differences, a set of comparison weights (also known as calibration comparison weights) were developed for NSCAW II. Specifically, the NSCAW I data were used to develop a set of calibration adjustments that were used to adjust the NSCAW II data for the missing PSUs. These NSCAW II comparison weights were also post-stratified to the full set of states that participated in NSCAW I. This common target population includes children aged 14 years or younger at the time of the investigation in states that participated in NSCAW I.

Calibration weights were used in all analyses. The NSCAW datasets and the comparison weights are available from the National Data Archive on Child Abuse and Neglect (NDACAN) at Cornell University upon request (www.ndacan.cornell.edu).

Results

Sample Demographics: NSCAW I and NSCAW II

We found very few significant differences in demographic characteristics when comparing caregivers with caseworker reports of active DV in NSCAW I (n = 437) with those with active DV reports in NSCAW II (n = 406; see Table 1). Comparisons between the two groups’ characteristics revealed similarities in their age, open case, functioning, education, number of children, employment, prior reports of maltreatment, prior substantiated abuse/neglect, urbanicity, marital status, and trouble meeting basic needs. However, significant differences existed when comparing the caregivers on their race/ethnicity and education level. Specifically, there were more Hispanic female caregivers with caseworker reports of active DV in NSCAW II than NSCAW I, F(1, 86) = 11.84, p = .0009. In addition, significantly more female caregivers with active DV in NSCAW II had an educational background with a bachelor’s degree or other degree compared with their counterparts, F(1, 86) = 12.24, p = .0007, whereas female caregivers in NSCAW I tended to have more vocational/associate degrees, F(1, 86) = 7.86, p = .0063.

Table 1.

Sample Characteristics of Caregivers With Child Welfare Worker Identified Active DV.

	NSCAW I (n = 437)	NSCAW II (n = 406)	p Value
	(%)	(%)	p Value
Caregiver age (in years)	M (SE) 31.26 (0.79)	M (SE) 30.30 (0.81)	.422
In home with CWS	35.26	27.16	.1549
Caregiver race/ethnicity			.0067
White	57.42	48.50
Hispanic	10.89	31.47
African American	24.18	14.33
Other	7.51	5.70
Caregiver functioning
Caregiver physical health	M (SE)49.73(1.06)	M (SE)48.17(1.53)	.382
Caregiver mental health problem	25.28	26.00	.9154
Caregiver substance use problem	17.93	16.85	.8728
Caregiver education			.0004
No degree	29.50	23.17
High school/GED	42.82	44.76
Vocational/associate	24.01	11.51
Bachelor’s degree or other	3.67	20.57
Number of children in household			.7438
One	30.72	23.11
Two	25.65	29.55
Three	25.15	27.40
Four	7.24	9.73
Five or more	11.24	10.22
Caregiver employment status			.8229
Full-time	39.56	34.01
Part-time	15.11	16.00
Unemployed, looking for work	14.59	17.29
Do not work or other	30.74	32.69
History of involvement with child protective services
Any prior reports of maltreatment	49.25	60.39	.1197
Prior substantiated abuse/neglect	53.87	64.02	.4116
Urban county of residence	71.74	73.53	.8027
Marital status			.2950
Married	29.33	27.75
Separated	27.17	27.58
Divorced or widowed	11.80	5.79
Never married	31.70	38.88
Trouble meeting basic needs	37.67	39.97	.7812

Note. F tests were conducted on the categorical variables, and t tests were conducted on the variables caregiver age and caseworker physical health. Unweighted sample size reported, however, proportions and means are based on weighted data. DV = domestic violence; NSCAW = National Survey of Child and Adolescent Well-Being; CWS = child welfare services; GED = general educational development certificate, equivalent to a high school diploma.

Description of Caregivers’ DV Experiences: NSCAW I and NSCAW II

As indicated in Table 2, of the 4,965 female caregivers in NSCAW I and II with DV reports in the study, 11.33% (unweighted n = 437) of NSCAW I caregivers had a caseworker report of active DV and 12.59% (unweighted n = 406) of NSCAW II caregivers had reports of active DV. Overall, 26.08% (unweighted n = 910) of the caregivers in NSCAW I had a caseworker report with a history of DV as a risk factor compared with 29.07% (unweighted n = 732) of NSCAW II caregivers. Bivariate analyses revealed no significant difference in the rate of caregivers in NSCAW I and II with reports of active DV, F(1, 86) = 0.72, p = .3971, or DV history, F(1, 86) = 1.93, p = .1681.

Table 2.

Prevalence of Domestic Violence.

	NSCAW I (n = 2,758)			NSCAW II (n = 2,207)			p Value
	Never	Past Year	Lifetime	Never	Past Year	Lifetime
	(%)	(%)	(%)	(%)	(%)	(%)
Types of physical violence
Had something thrown	73.58	15.59	10.93	76.75	14.89	8.36	.3489
Slapped	74.66	13.05	12.29	83.77	9.81	6.42	.0005
Beat up	81.58	8.06	10.36	88.58	6.66	4.76	.0001
Choked	82.62	8.19	9.19	88.54	6.90	4.57	.0011
Threatened with knife or gun	91.79	3.81	4.40	93.16	4.26	2.57	.1486
Knife or gun used	97.75	0.89	1.36	98.04	0.83	1.13	.8839
	NSCAW I (n = 2,758)			NSCAW II (n = 2,207)
Child welfare worker DV report
Active DV	11.33			12.59			.3971
DV history	26.08			29.07			.1681

Note. F tests were conducted for each categorical variable. Unweighted sample size reported, however, proportions and means are based on weighted data. NSCAW = National Survey of Child and Adolescent Well-Being; DV = domestic violence.

However, caregivers in NSCAW I and II significantly differed in self-reported rates of victimization for specific types of violence. In general, caregivers in NSCAW II seemed to experience lower rates of recent and lifetime victimization than those in NSCAW I. When examining victimization for specific types of violence among caregivers in NSCAW I who experienced at least one form of lifetime DV, rates ranged from 1.36% of caregivers having a knife or gun used on them to 12.29% of the sample being slapped. In the NSCAW II sample with reports of a lifetime DV experience, rates ranged from 1.13% having a knife or gun used on them to 8.36% having something thrown.

With the exception of having a knife or gun used, caregivers in NSCAW I who experienced recent DV reported higher rates of each type of violence compared with NSCAW II caregivers who experienced the same types of violence during the past year. This was specifically true when comparing rates for the following acts of violence experienced during the past year: being slapped, F(1, 86) = 11.47, p = .0011, beaten up, F(1, 86) = 19.85, p = .0000, and chocked, F(1, 86) = 12.73, p = .0006.

External DV Service Referral and Receipt: NSCAW I and NSCAW II

There was no significant difference in the rate of referral to external DV services when comparing female caregivers with active reports of DV in NSCAW I and II: 42.45% (unweighted n = 180) of NSCAW I caregivers and 50.81% of NSCAW II caregivers reported that at either baseline or 18-month follow-up, they were referred for external DV services during a time period 1 year before their interview, F(1, 86) = 0.95, p = .3317. Caregivers with reports of active DV in NSCAW I and II were also similar in their external DV service use at baseline and 18-month follow-up: 18.67% (unweighted n = 95) of the caregivers in NSCAW I reported that they used external DV services within the past year compared with 18.88% (unweighted n = 94) of caregivers in NSCAW II, F(1, 86) = 0.002, p = .9670. Among caregivers with active DV reports who did not receive external services, 39.01% of caregivers in NSCAW I reported a need for external DV services, whereas 42.73% of caregivers in NSCAW II reported such a need, F(1, 86) = 0.17, p = .6836.

Active DV Versus No Active DV: NSCAW II

Comparisons between the characteristics of caregivers in NSCAW II with and without caseworker reports of active DV (see Table 3) revealed similarities in their previous substantiated case with CPS, prior reports to CPS, child in home with child welfare services, race/ethnicity, age, number of children in household, employment status, physical health, and urbanicity. On average, caregivers in the NSCAW II sample were 30.68 years old and had a physical health score of 47.92. Most of the sample was White (52.64%), did not work (32.18%) or had full-time employment (31.22%), had one (28.40%) or two (26.01%) children in the household, and lived in an urban area (71.90%). In addition, 24.23% of the sample had an open case after their maltreatment investigation, and 53.44% had prior reports to CPS. Of those with prior CPS reports, 50.30% had a previously substantiated case with CPS.

Table 3.

NSCAW II Sample Characteristics of Caregivers With Child Welfare Worker DV Report.

	Total (n = 2,207)	No Active DV (n = 1,801)	Active DV (n = 406)	p Value
	(%)	(%)	(%)	p Value
Caregiver age (in years)	M (SE)	M (SE)	M (SE)	.622
	30.68 (0.26)	30.73 (0.28)	30.30 (0.81)
In home with CWS	24.23	23.81	27.16	.3884
Caregiver race/ethnicity				.0690
White	52.64	53.24	48.50
Hispanic	21.26	19.79	31.47
African American	20.84	21.78	14.33
Other	5.26	5.19	5.70
Caregiver functioning
Caregiver physical health	M (SE)	M (SE)	M (SE)	.864
	47.92 (0.54)	47.88 (0.58)	48.17 (1.53)
Caregiver mental health problem	12.36	10.40	26.00	.0002
Caregiver substance use problem	9.95	8.96	16.85	.0366
Caregiver education				.0324
No degree	26.70	27.21	23.17
High school/GED	46.39	46.63	44.76
Vocational/associate	14.86	15.34	11.51
Bachelor’s degree or other	12.05	10.82	20.57
Number of children in household				.5069
One	28.40	29.16	23.11
Two	26.01	25.51	29.55
Three	22.77	22.10	27.40
Four	12.50	12.90	9.73
Five or more	10.31	10.33	10.22
Caregiver employment status				.8048
Full-time	31.22	30.82	34.01
Part-time	15.97	15.96	16.00
Unemployed, looking for work	20.63	21.11	17.29
Do not work or other	32.18	32.11	32.69
History of involvement with child protective services
Any prior reports of maltreatment	53.44	52.44	60.39	.2169
Prior substantiated abuse/neglect	50.30	48.15	64.02	.0620
Urban county of residence	71.90	71.66	73.53	.7076
Marital status				.0011
Married	29.17	29.37	27.75
Separated	15.67	13.96	27.58
Divorced or widowed	17.13	18.76	5.79
Never married	38.04	37.91	38.88
Trouble meeting basic needs	22.28	19.73	39.97	.0004

Note. F tests were conducted on the categorical variables, and t tests were conducted on the variables caregiver age and caseworker physical health. Unweighted sample size reported, however, proportions and means are based on weighted data. NSCAW = National Survey of Child and Adolescent Well-Being; DV = domestic violence; CWS = child welfare services; GED = general educational development certificate, equivalent to a high school diploma.

Although the two groups were similar in some ways, they were significantly different in respects to education level, marital status, substance use, mental health, and trouble meeting basic needs. When compared with caregivers who did not have a caseworker report of active DV, caregivers with a report of active DV were more likely to be separated from a spouse, F(1, 74) = 13.5248, p = .0004, have a bachelor’s degree, F(1, 74) = 6.58, p = .0124, substance use issues, F(1, 74) = 4.5293, p = .0366, mental health issues, F(1, 74) = 14.8976 p = .0002, and trouble meeting basic needs, F(1, 74) = 14.0166, p = .0004. Conversely, they were less likely to be divorced, F(1, 74) = 30.776, p = .000.

CPS DV Interventions: NSCAW II

Among caregivers in the NSCAW II sample with active reports of DV (unweighted n = 406), 27.36% were investigated by an agency that used a DV assessment tool, and many were investigated by an agency that sometimes (47.01%) or always (47.30%) used a DV specialist.

CPS interventions and caregivers’ receipt of external DV services

As indicated in Table 4, after taking into account all control variables, caregivers with active DV involved with a CPS agency that used a DV assessment tool were 7.03 times more likely to report receiving external DV services compared with those served by agencies without DV assessment tools (95% confidence interval [CI] = [2.33, 21.22]). On the contrary, caregivers involved with agencies that sometimes (odds ratio [OR] = 0.16, 95% CI = [0.03, 0.99]) or always (OR = 0.15, 95% CI = [0.02, 0.98]) had a DV specialist available were less likely to receive external DV services than those involved with an agency that never/rarely had a DV specialist available. Furthermore, caregivers who had an open case involving child welfare services (OR = 3.26, 95% CI = [1.17, 9.12]) or a high school diploma or general educational development certificate (GED; OR = 5.47, 95% CI = [1.41, 21.18]) were more likely to receive external DV services than caregivers who did not have an open case or had no degree, respectively.

Table 4.

Logistic Regression Predicting Caregivers’ Receipt of Domestic Violence Services (N = 335).

Variable	B	OR	95% CI
Logistic intercept	−4.42
DV specialist
Never/rarely	Ref
Sometimes	−1.82	0.16*	[0.03, 0.99]
Always	−1.92	0.15*	[0.02, 0.98]
DV tool	1.95	7.03**	[2.33, 21.22]
Caregiver age (in years)	0.004	1.00	[0.92, 1.10]
In home with CWS	1.18	3.26*	[1.17, 9.12]
Caregiver race/ethnicity			[0.08, 1.68]
White	Ref		[0.11, 1.64]
Hispanic	0.03	1.03	[0 .25, 4.23]
African American	−1.29	0.27	[0.04, 1.81]
Other	−0.64	0.53	[0.05, 5.45]
Caregiver functioning			[0 .70, 11.16]
Caregiver physical health	0.06	1.06	[0 .99, 1.14]
Caregiver mental health problem	−0.71	0.49	[0 .14, 1.76]
Caregiver substance use problem	0.43	1.53	[0 .55, 4.28]
Caregiver education			[1.12, 4.45]
No degree	Ref
High school/GED	1.70	5.47*	[1.41, 21.18]
Vocational/associate	0.10	1.10	[0.17, 7.12]
Bachelor’s degree or other	1.18	3.24	[0 .54, 19.39]
Number of children in household			[0 .10, 2.02]
One	Ref		[0 .85, 4.16]
Two	0.32	1.38	[0.35, 5.47]
Three	−1.20	0.30	[0.75, 1.21]
Four	1.58	4.85	[0.70, 33.73]
Five or more	1.32	3.74	[0.53, 26.41]
Caregiver employment status			[0 .31, 2.94]
Full-time	Ref		[0.75, 11.46]
Part-time	−0.89	0.41	[0.12, 1.40]
Unemployed, looking for work	0.59	1.80	[0.64, 5.07]
Do not work or other	−1.17	0.31	[0.07, 1.37]
Any prior reports of maltreatment	0.74	2.09	[0 .61, 7.17]
Urban county of residence	−0.98	0.38	[0.07, 2.11]
Marital status			[0.22, 2.02]
Married	Ref
Separated	−0.22	0.81	[0 .28, 2.28]
Divorced or widowed	−1.36	0.26	[0 .04, 1.52]
Never married	−0.39	0.68	[0.23, 2.00]
Trouble meeting basic needs	−1.92	2.06	[0.68, 6.21]

Note. OR = odds ratio; CI = confidence interval; DV = domestic violence; Ref = reference category; CWS = Child welfare services; GED = general educational development certificate, equivalent to a high school diploma. F(35, 38) = 4.80***.

p < .05. **p < .01. ***p < .001.

Discussion

This study examined the DV experiences of CPS-involved female caregivers participating in the first cohort of the NSCAW study (NSCAW I) compared with those involved in the second cohort (NSCAW II). In addition, we investigated whether current CPS interventions are helpful in increasing the likelihood of NSCAW II caregivers’ receipt of external DV services. We found that caseworkers identified DV at similar rates during both NSCAW studies, approximately 11% of NSCAW I caregivers had a caseworker report of active DV and 13% of caregivers in NSCAW II. However, when asked about the specific types of DV acts recently experienced, caregivers in the two cohorts differed in their reported rates: Caregivers in the NSCAW I self-reported higher rates of being slapped, beaten up, and chocked than caregivers in the NSCAW II. In addition, we found significant differences when comparing the racial and educational backgrounds of caregivers with reports of active DV in NSCAW I and II. Caregivers with active DV in NSCAW II were more likely to identify as Hispanic than caregivers with active DV in NSCAW I.¹ Caregivers with active DV in NSCAW II were also more likely to have attained a bachelor’s degree. Although findings suggest that caregivers in the NSCAW I experienced higher rates of some types of physical violence than caregivers in NSCAW II, because not all types of violence on the CTS were measured in this study, we could not conclude whether this was an overall trend. However, our study findings indicate that DV-affected caregivers who were involved with child welfare over a decade ago share similar demographic characteristics as those more recently involved with child welfare, and their DV was identified by caseworkers at similar rates.

When examining external DV service referral and receipt among caregivers with caseworker reports of active DV in NSCAW I and NSCAW II, we found that caregivers from both samples reported similar rates. Unfortunately, we also found a disparity existed between (a) external service referral and external service receipt and (b) external service receipt and external service need. Both groups of caregivers reported greater amounts of external service referral than they received, and many of the caregivers who did not receive external services reported a need. This finding is consistent with prior NSCAW studies (Casanueva et al., 2014; Kohl et al., 2005).

Although a disparity exists in external DV services, it is promising to learn that many CPS agencies nationwide are using interventions that may improve DV identification. For example, most CPS directors in this study reported that their agency sometimes or always had a DV specialist available during child maltreatment investigations. However, contradictory to prior research findings, we found that having a DV specialist available during investigations decreased chances that DV-affected caregivers reported receipt of external DV services. Although we are unable to test why this is so, we highly suspect caregivers receiving services from agencies with DV specialists may have their service needs met internally; hence, there is no need for services externally. As such, our findings are not an indication that DV specialists do not improve DV-affected caregivers’ chances of receiving services. Instead, they highlight the need for additional research to understand what types of services are provided by the DV specialists working with CPS agencies.

Interestingly, relatively fewer agencies reported using DV assessment tools during investigations. Only a little over a quarter of agencies reported using DV assessment tools. This is worrisome considering our finding that suggests, after controlling for caregivers’ characteristics, DV assessment tools improve chances for caregivers’ receipt of external DV services. It may be that DV assessment tools lead to greater levels of DV identification, which then lead to better chances for external DV service receipt. However, further research is necessary to determine whether DV identification mediates the relationship between DV assessment tools and external DV service receipt.

Like the Kohl et al. (2005) study, our findings also indicate caregivers who had an open case were more likely to receive external DV services than those without an open case. Given the similar findings, it seems CPS service plans for DV-affected caregivers with children remaining in the home include external DV services. In addition, our results suggest caregivers who had a high school diploma or GED were more likely to receive external DV services than those with no degree. While unable to determine the cause of lower education as a barrier to external service receipt in this secondary data analysis, we propose a couple of explanations. Low education is highly correlated with poverty, and limited financial resources may prevent female caregivers from accessing external DV services. It may also be that individuals with high school diplomas have had more exposure to information about DV risks, and this may increase their likelihood of using external DV services compared with individuals who are not as highly educated.

Limitations

Although the current study adds to the extant child welfare literature on DV and DV services, it is not without limitations. First, because different versions of the CTS were used, we could not fully compare DV experiences between NSCAW I and NSCAW II. In addition, because different self-reported DV measures were used in NSCAW I and II, we could not classify caregiver samples based on self-report. Instead, we used caseworker reports of DV. This is problematic because prior research suggests caseworkers under-identify DV. Thus, findings may not be generalizable to all CPS-involved caregivers affected by DV. We call for research that uses DV self-reports to compare DV trends among caregivers involved with CPS. Second, rather than including the entire NSCAW II sample (i.e., children aged 0 to 17), the NSCAW II sample in this study was limited to caregivers of children aged 0 to 14 years. Therefore, study findings are not generalizable to CPS-involved caregivers with older children. Third, the DV intervention data were limited to information available in the NSCAW II study; therefore, we were unable to examine specific mechanisms by which DV assessment tools and specialists may influence chances for caregivers’ receipt of external DV services. In a related vein, because the NSCAW II did not measure specific types of internal services offered by DV specialists working with CPS agencies, we were unable to determine whether DV-affected caregivers’ needs may have been internally met by CPS. Notwithstanding these limitations, this study highlights the need for CPS practice, policy, and research that improve external DV service linkage.

Practice Implications

Given the disparity in external service receipt among caregivers in NSCAW I and II with reports of active DV who (a) are referred to external DV services and (b) self-identify as needing external DV services, CPS agencies must work to improve their caseworkers’ ability to identify DV service needs. Once a need for external DV services is identified and a referral is made, CPS workers should follow-up with caregivers to ensure receipt of these services. This is especially important given evidence that shows that CPS-involved caregivers matched to DV services are more likely to have improved outcomes compared with those not matched to DV services (Smith & Marsh, 2002). As recommended by the Greenbook initiative (Banks et al., 2009), one way CPS agencies may improve chances that caregivers receive services after referral is to involve DV agencies in creating safety plans for adult and child victims of DV. By collaborating with CPS agencies, workers at DV agencies may feel more invested in CPS clients’ outcomes, and, therefore, more likely to follow-up with CPS caseworkers to confirm whether caregivers receive services.

Although we did not examine the role of poverty, substance abuse, and mental health in the receipt of external DV services, we did corroborate the findings of others (e.g., Benson, Fox, DeMaris, & Van Wyk, 2003; Edmond, Bowland, & Yu, 2013) that caregivers who have experienced DV are at high risk of experiencing both mental health and substance abuse problems and having trouble meeting basic needs. Therefore, in addition to linking families to DV crisis intervention services, these families may need integrated services that include financial assistance, substance abuse, and mental health treatment. It is important that CPS caseworkers are aware of such possible co-occurring risk(s) and are prepared to address multiple needs.

Policy Implications

Based on the study’s findings, we recommend child welfare agencies create policies that require caseworkers to use assessment tools specifically designed to help identify DV. Such an activity is low cost and may be an effective approach to ensuring DV service receipt. This recommendation aligns with recommendations in the Greenbook, emphasizing the importance of proper assessment for DV in child welfare cases, and child welfare policies and practices to increase safety for the adult–child dyad (Banks et al., 2009). However, we caution this recommendation by suggesting DV assessment tools only be used in conjunction with methods that are in place to ensure caregivers are, in fact, receiving services and these services are helpful. A systematic review of research studies examining DV screening by health professionals indicate that, while DV screening may increase referral, there is little evidence that suggests DV referral leads to improved outcomes (e.g., decreased exposure to violence; Ramsay, Richardson, Carter, Davidson, & Feder, 2002). However, given this systematic review by Ramsay and her colleagues was conducted over 10 years ago, we recommend more current research examining the impact of DV screening on DV survivors’ outcomes.

Research Implications

Because the current study solely focused on physical DV and the number of DV acts studied was limited, research that includes other forms of DV (i.e., psychological and sexual violence; economic control) is required to determine the overall prevalence of DV among CPS-involved families. In addition, research is needed to understand how DV specialists work to assist CPS caseworkers with external DV service linkage and how their services may be influencing this outcome. For example, researchers may want to investigate how external DV service linkage varies by the types of services offered by DV specialists. Researchers may also want to study how service outcomes vary depending on the type of DV assessment tool(s) used. We were unable to delve deeper into these areas given limitations of the NSCAW data.

Conclusion

Given the extensive number of CPS-involved families identified in this study that experienced DV, CPS agencies can play a vital role in providing services to these families. However, CPS agencies sometimes struggle with how to help. To the best of our knowledge, this is the first study to compare demographic characteristics and types of DV victimizations between CPS-involved caregivers in NSCAW I with those in NSCAW II, and examine the relationship between CPS interventions and DV service linkage using a nationally representative sample. Therefore, it is important that additional research is conducted to advance this area of knowledge. Further research is needed to understand (a) changes in characteristics of CPS-involved families affected by DV and (b) how CPS interventions assist DV-affected families. We uphold recommendations that CPS agencies use specialized interventions, such as assessment tools, to identify DV-affected families and link them to services. Although many CPS workers are working with DV specialists based in their agencies, we also suggest they collaborate with DV specialists outside of their agencies to help ensure clients referred to external DV services are, in fact, receiving these services.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Ijeoma Nwabuzor Ogbonnaya, MSW, PhD, is an assistant professor at the School of Social Work at San Diego State University. She received her PhD from the University of North Carolina at Chapel Hill School of Social Work in August 2012. Following her doctoral education, she completed a 2-year fellowship as a postdoctoral research associate at the University Southern California. Her primary research interests are domestic violence, mental health, and racial/ethnic disparities among child welfare populations. She is working to build a research portfolio that supports the need for specialized services for child-welfare-involved families affected by domestic violence. She hopes to use this portfolio to support her long-term goal, which is to design and implement child maltreatment prevention interventions for families affected by domestic violence.

Patricia L. Kohl, MSW, PhD, is currently an associate professor and associate Dean for Social Work. She received her PhD from the University of North Carolina at Chapel Hill in 2006, after which she joined the faculty of the Brown School. Her scholarship is focused on strengthening, implementing, and sustaining evidence-based interventions that improve the quality of care provided to vulnerable families. Through her research, which is informed by several years of clinical practice with children and families, she seeks to close the gap between research and community-based practice. Her intervention and implementation research is informed by analyses of large datasets to examine how characteristics of parents within the child welfare service system, such as substance abuse and mental illness, influence parenting behaviors, child emotional and behavioral health, and the safety of children. She is also currently serving as the interim associate dean for Social Work. She is affiliated with two research centers at the Brown School: Center for Mental Health Services Research and the Center for Injury and Violence Prevention. She is also affiliated with the National Institute of Mental Health’s Implementation Research Institute. Her work has received financial support from the National Institute of Mental Health and the Centers for Disease Control and Prevention.

References

Administration for Children & Families. (2005). National Survey of Child and Adolescent Wellbeing (NSCAW): CPS sample component: Wave 1 data analysis report. Washington, DC: U.S. Department of Health and Human Services. Retrieved from http://www.acf.hhs.gov/sites/default/files/opre/cps_report_revised_090105.pdf

Administration for Children & Families. (2011). NSCAW II Baseline Report: Introduction to NSCAW II. Washington, DC: U.S. Department of Health and Human Services, Administration for Children & Families, Administration on Children, Youth and Families, Children’s Bureau. Retrieved from http://www.acf.hhs.gov/sites/default/files/opre/nscaw2_intro.pdf

Aron

L. Y.

Olson

K. K.

(1997). Efforts by child welfare agencies to address domestic violence: The experiences of five communities. Washington, DC: Urban Institute. Retrieved from http://www.urban.org/publications/406798.html

Banks

Hazen

A. L.

Coben

J. H.

Wang

Griffith

J. D.

(2009). Collaboration between child welfare agencies and domestic violence service providers: Relationship with child welfare policies and practices for addressing domestic violence. Children and Youth Services Review, 31, 497-505. doi:10.1016/j.childyouth.2008.10.005

Banks

Landsverk

Wang

(2008). Changing policy and practice in the child welfare system through collaborative efforts to identify and respond effectively to family violence. Journal of Interpersonal Violence, 23, 903-932. doi:10.1177/0886260508314693

Benson

M. L.

Fox

G. L.

DeMaris

Van Wyk

(2003). Neighborhood disadvantage, individual economic distress and violence against women in intimate relationships. Journal of Quantitative Criminology, 19, 207-235. doi:10.1023/A:1024930208331

Casanueva

Smith

Ringeisen

Dolan

Tueller

(2014). Families in need of domestic violence services reported to the child welfare system: Changes in the National Survey of Child and Adolescent Well-Being between 1999-2000 and 2008-2009. Child Abuse & Neglect, 38, 1683-1693. doi:10.1016/j.chiabu.2014.05.013

Coohey

(2007). What criteria do child protective services investigators use to substantiate exposure to domestic violence? Child Welfare, 86, 93-122. Retrieved from https://www.cwla.org/articles/cwjabstracts.htm

Edmond

Bowland

(2013). Use of mental health services by survivors of intimate partner violence. Social Work in Mental Health, 11, 34-54. doi:10.1080/15332985.2012.734180

10.

Hazen

A. L.

Connelly

C. D.

Kelleher

Landsverk

Barth

(2004). Intimate partner violence among female caregivers of children reported for child maltreatment. Child Abuse & Neglect, 28, 301-319. doi:10.1016/j.chiabu.2003.09.016

11.

Hess

P. M.

Folaron

Jefferson

A. B.

(1992). Effectiveness of family reunification services: An innovative evaluative model. Social Work, 37, 304-311. doi:10.1093/sw/37.4.304

12.

Kohl

P. L.

Barth

R. P.

Hazen

A. L.

Landsverk

J. A.

(2005). Child welfare as a gateway to domestic violence services. Children and Youth Services Review, 27, 1203-1221. doi:10.1016/j.childyouth.2005.04.005

13.

Magen

R. H.

Conroy

Hess

P. M.

Panciera

Simon

B. L.

(2001). Identifying domestic violence in child abuse and neglect investigations. Journal of Interpersonal Violence, 16, 580-601. doi:10.1177/088626001016006006

14.

Marsh

J. C.

Ryan

J. P.

Choi

Testa

M. F.

(2006). Integrated services for families with multiple problems: Obstacles to family reunification. Children and Youth Services Review, 28, 1074-1087. doi:10.1016/j.childyouth.2005.10.012

15.

Ogbonnaya

I. N.

Guo

(2013). Effect of domestic violence on the risk of out-of-home placement: A propensity score analysis. Journal of the Society for Social Work & Research, 4, 198-213. doi:10.5243/jsswr.2013.14

16.

Ramsay

Richardson

Carter

Y. H.

Davidson

L. L.

Feder

(2002). Should health professionals screen women for domestic violence? Systematic review. British Medical Journal, 325, 1-13. doi:10.1136/bmj.325.7359.314

17.

Rivers

J. E.

Maze

C. L.

Hannah

S. A.

Lederman

C. S

. (2007). Domestic violence screening and service acceptance among adult victims in a dependency court setting. Child Welfare, 86, 123-144. Retrieved from http://www.cwla.org/child-welfare-journal/

18.

Rosewater

(2008). Building capacity in child welfare systems: Domestic violence specialized positions. Retrieved from http://www.thegreenbook.info/documents/BuildingCaps.pdf

19.

Smith

Marsh

J. C.

(2002). Client-service matching in substance abuse treatment for women with children. Journal of Substance Abuse Treatment, 22, 161-168. doi:10.1016/S0740-5472(02)00229-5

20.

StataCorp. (2011). Stata: Release 12 [Computer software]. College Station, TX: Author.

21.

Stepler

Brown

(2015). Statistical portrait of Hispanic in the United States, 1980–2013. Retrieved from http://www.pewhispanic.org/2015/05/12/statistical-portrait-of-hispanics-in-the-united-states-2013-key-charts/

22.

Straus

M. A

. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of Marriage and Family, 41, 75-88. doi:10.2307/351733

23.

Straus

M. A

. (1990). The Conflict Tactics Scale and its critics: An evaluation and new data on validity and reliability. In Straus

M. A.

Gelles

R. J.

(Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8,145 families (pp. 49-73). New Brunswick, NJ: Transaction.

24.

Ware

J. E.

Jr. Phillips

Yody

B. B.

Adamczyk

(1996). Assessment tools: Functional health status and patient satisfaction. American Journal of Medical Quality, 11, S50-S53.

25.

Wolfe

D. A.

Crooks

C. V.

Lee

McIntyre-Smith

Jaffe

P. G.

(2003). The effects of children’s exposure to domestic violence: A meta-analysis and critique. Clinical Child and Family Psychology Review, 6, 171-187. doi:10.1023/A:1024910416164