SEARCH

SEARCH BY CITATION

Keywords:

  • child protection (policy and practice);
  • child welfare;
  • domestic violence;
  • violence (against women and children)

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Differential Response (DR) Models
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. References

Large numbers of domestic violence (DV) cases on child protection caseloads have necessitated the development of practices that address both DV and child safety. The first step in this process is to gain an understanding of the differences between DV-involved cases and other forms of maltreatment. The implementation of a differential response service model in Ontario offered an opportunity to compare risk assessment ratings, service outcomes and recurrence and to identify pathways of DV cases through child protection services (CPS). A sample (n = 785) of child protection investigations over a 4-month period was examined. Of these investigations, 26% cases were DV referred; 87% of the DV victims were mothers; perpetrating partners were mostly absent from investigations; non-white families were more often investigated for DV than white families; and DV cases were more likely to remain open for ongoing CPS. Only one-third of DV-exposed children were assessed as having been harmed and most community referrals were made for the victim parent. Mothers were the primary target of investigation, remaining in CPS for extended service provision although recurrence rates were lower than found in other investigations. Results are discussed to inform investigative procedures, assessment and service response to more adequately respond to children and families when DV is present.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Differential Response (DR) Models
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. References

Recent focus on child exposure to domestic violence (EDV) in child protection services (CPS) has contributed to enormous increases in identified cases of domestic violence (DV)-involved cases across North America (Trocmé et al. 2005a,b; Edleson et al. 2006). Canadian agencies have experienced notable increases in DV-related child protection investigations with the estimated number of substantiated DV cases going up dramatically between 1998 and 2003 and then stabilizing between 2003 and 2008 (Trocmé et al. 2005a,b, 2008) (see Note 1). In a recent analysis of the Canadian Incidence Study of Child Abuse and Neglect (CIS), EDV was the most frequently substantiated form of child maltreatment and was verified at the same rate as neglect (34%) (Trocmé et al. 2008). These reporting trends have created challenges in investigative practices, presenting workers and systems with service provision challenges (Bourassa et al. 2008).

As the significant rise in DV-related cases is a recent phenomenon, little scientific evaluation has been carried out on child protection practices in these cases specifically. Concerns continue about mother blaming formulations underpinning CPS responses wherein mothers, most often the victims of DV, are viewed as failing to protect their children from exposure (Devoe & Smith 2003; Davies & Krane 2006; Alaggia et al. 2007, 2012; Strega et al. 2008; Humphreys & Absler 2011).

Mothers are often expected to leave their abusive partners when CPS become involved; otherwise, they risk longer and more intensive intervention and are frequently scrutinized for their parenting capacities even when initial reporting is around their own victimization (Kantor & Little 2003; Hughes et al. 2011; Humphreys & Absler 2011). Yet, the literature is full of findings that point to the very real and serious barriers that prevent abused women from leaving abusive relationships. Structural and intra-/interpersonal barriers include financial and emotional dependence, cultural and religious prohibitions regarding separation and divorce, problematic custody and access orders, precarious immigration status and fear of deportation, safety threats and increased violence post-separation (McDonald 1999; Shaffer & Bala 2004; Shirwadkar 2004; Alaggia et al. 2009; Humphreys & Absler 2011; Stanley et al. 2011). Despite these often insurmountable obstacles, mothers are held responsible for continued child EDV if they stay with their partners.

One significant threat is the perceived role of child protection and the spectre of losing their children (Sev'er 2002; Alaggia et al. 2007; Nixon et al. 2007). Findings about children removed from their families for EDV are mixed at best. Some studies cite rates of apprehension as high as 80% when DV is considered to be moderate or high risk and when other forms of maltreatment are co-occurring (Humphreys 2000; English et al. 2005). Other research suggests that DV-involved cases are no more likely to result in removal than other forms of maltreatment (Trocmé et al. 2005a,b). More recent analyses of CIS data on service outcomes of CPS investigations found that response depended largely on whether there was co-occurrence of DV exposure with other forms of abuse (Black et al. 2008). In situations of co-occurring maltreatment, children were four times more likely to be apprehended than when DV was identified on its own. When child EDV was identified and substantiated on its own, workers tended towards shorter intervention, terminating involvement and closing 64% of the cases after initial investigation (Black et al. 2008). However, it is unclear under what circumstances these cases remain open for ongoing CPS and what happens to these families during extended involvement.

Differential Response (DR) Models

  1. Top of page
  2. Abstract
  3. Introduction
  4. Differential Response (DR) Models
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. References

Many jurisdictions across North America have adopted DR approaches for investigating child maltreatment, a model that has been strongly recommended in DV cases to ensure more appropriate response and services. DR models rely on valid and reliable risk assessment tools to assess actual risk or risk of harm to children. The DR approach has been developed to identify lower-risk families from high-risk families, whereby families to be deemed low risk are diverted out of traditional CPS and referred to community-based services (Waldfogel 1998; Conley 2007). Lower-risk families are viewed to better benefit from voluntary community-based services. Only children and families assessed as high risk are treated as requiring traditional CPS and are serviced within child protection systems (Waldfogel 1998; Edleson 2004; Conley 2007). Thus, DR approaches are perceived as keeping the best interest of the children as a central focus, reducing intrusive interventions and building on family strengths.

North American research on child protection involvement and outcomes based on DR is in its infancy and presents a perplexing picture in cases of EDV. American-based studies have found that DV factors are still under-identified in CPS, child protection intervention is described to be inadequate and a high number of these cases result in re-referrals within a relatively short period of time after closing (English et al. 2005; Kohl et al. 2005; Kohl & Macy 2008). English et al. (2005) concluded in their study, ‘the high level of re-referral to CPS for the moderate to high risk DV indicated cases and the similarly high level of re-referral for the no or low risk DV indicated cases is troubling’ (p. 1200).

In a longitudinal Canadian study comparing a family development response (based on DR) with a regular investigative approach in CPS, there was no difference found in recidivism between the two interventions. However, fewer children were removed from their homes in the family development response group (Marshall et al. 2010). It should be noted that this investigation focused on all forms of child maltreatment and did not distinguish outcomes with DV-involved cases specifically. Most evaluations of DR focus on all cases referred to CPS and do not separate out DV-involved cases for deeper analysis.

Given this state of research, the current investigation sought to examine CPS cases in a Canadian jurisdiction that had adopted a DR approach and compare case characteristics and outcomes. The research questions centred on the following: (i) establishing the incidence of DV in child protection cases in the study jurisdiction; (ii) describing families referred with DV as the presenting problem; (iii) determining characteristics that distinguish DV-involved families from families investigated for other forms of child maltreatment in terms of risk, case outcomes and recurrence; (iv) understanding the role and impact of DR with respect to child welfare outcomes for families with DV as compared with other forms of child maltreatment; and (v) describing the pathways of DV-involved families through the child protection system.

Method

  1. Top of page
  2. Abstract
  3. Introduction
  4. Differential Response (DR) Models
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. References

A mixed-method design was used to obtain information about the characteristics of children and families investigated for EDV and to ascertain common pathways through the child protection system under study. Information on child maltreatment referrals was collected over a 4-month period from five child welfare agencies in Southern and Eastern Ontario, resulting in a sample of 785 cases. Quantitative data gathered on this cohort included demographic and case characteristics of children and families, DV details, household information, assessment of risk to the children, referrals and service outcomes, and recurrence at 18 months. Qualitative data were gathered on 75 case records based on a coding framework to determine patterns in case pathways. Thematic analysis of case records was used to triangulate the quantitative data. The study received ethics approval from the University of Toronto.

Data collection proceeded after provincial training for DR was completed in July 2007. A period of 6 months was given for workers to be able to integrate and consolidate new practices based on this training. In three of the agencies, workers were trained to collect data manually on all investigated cases during the 4-month period based on a modified version of the survey used for the Canadian Incidence Study of Suspected Child Abuse and Neglect (Trocmé et al. 2005a) that was augmented to include DV-related information. Information for two agencies where electronic databases existed was retrieved by trained researchers using their form management information systems (including case notes) and supervised on-site by the lead researcher. A random sample (n = 357) of all investigations occurring during the 4-month period was used for these two agencies. The Ontario Child Abuse and Neglect Data System – a relational database (see Note 2) containing the child protection histories of children and families investigated for child maltreatment in Ontario – was used to follow children and families over time for recurrence rates.

Case record reviews were conducted on a subsample of 75 files for in-depth content analysis to understand DV case pathways, service needs and outcomes. Purposive sampling was used to identify files that were investigated for DV allegations across the five agencies: cases that had one opening or several openings, cases that were closed after the 30-day investigative period and cases that went to ongoing CPS.

Comparisons between DV-involved cases and other forms of maltreatment on child, family and household characteristics were done using chi-square tests for nominal level data and t-tests for interval level data. For significance of differences between groups, DV cases and non-DV cases, bivariate analysis was conducted comparing outcome indicators such as closure, ongoing service, risk assessment, assessment of harm to child and referrals to community services. Fisher's exact test was used to determine recurrence at 18 months by risk rating.

Case records were subjected to thematic analysis using a coding framework developed to label data. Each file was examined with specific codes in mind related to prior and subsequent openings, previous DV identification, other forms of child maltreatment, service provision and outcomes, actions taken with the perpetrating parent and recurrences. Open and selective coding was done to generate themes (Corbin & Strauss 1998). For pre-test purposes, 12 files were coded independently by two coders to ensure consistency in the use of the coding framework. The lead researcher checked the coding, discussed and resolved discrepancies. These codes were later collapsed into larger categories that represented frequently recurring patterns within the interview data. Next, interpretations were generated with the lead researcher and research assistants trained in qualitative analysis. These were then compared across interviews for major theme identification through a distillation process.

To ensure trustworthiness of the data (Drisko 1997), member checking meetings were held at the participating agencies with CPS workers from January 2010 to October 2011. Findings were shared to ensure authenticity and confirmability (Drisko 1997). Presentations and discussions were facilitated to fully understand the results.

Findings

  1. Top of page
  2. Abstract
  3. Introduction
  4. Differential Response (DR) Models
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. References

Sample characteristics

Twenty-six per cent of all investigated cases during the study period involved DV, and the identified victim was the mother in 87% of the cases. DV cases differed from other forms of child maltreatment in a number of ways. DV-involved cases were more often referred by police (Table 1), the families were younger (Table 2 and Fig. 1), as were the child victims (Table 3), and significantly more non-white families were investigated for DV than white families ( Fig. 2). Only about one-third (37.2%) of the perpetrating parents (predominantly fathers) were able to be contacted or interviewed for the investigation, while the victim parents (predominantly mothers) were investigated in over 90% of the cases.

figure

Figure 1. Caregivers' age by type of maltreatment.

Download figure to PowerPoint

figure

Figure 2. Caregivers' ethno-racial group by type of maltreatment.

Download figure to PowerPoint

Table 1. Source of referrala by type of maltreatment (n = 785)
 Domestic violenceOther forms of child maltreatment
FrequencyPercentageFrequencyPercentage
  1. a

    Percentages do not add to 100% because categories are mutually exclusive.

Family or friends3216%12221%
School2211%23440%
Police10050%7513%
Mental-health worker2714%7713%
Health-care worker137%5810%
Table 2. Primary and secondary caregiver age by type of maltreatmenta (n = 785)
 Domestic violenceOther forms of child maltreatment
FrequencyPercentageFrequencyPercentage
  1. a

    Percentages do not add to 100% because categories are mutually exclusive.

Primary caregiver age    
<21 years147%275%
22–30 years6533%11521%
31–40 years7136%22541%
41+ years4824%18333%
Secondary caregiver age    
<21 years43%113%
22–30 years4829%6017%
31–40 years6439%12535%
41+ years4829%16045%
Table 3. Average age of referred child and average number of children in the family by type of maltreatmenta (n = 785)
 Domestic violenceOther forms of child maltreatment
MeanStandard deviationMeanStandard deviation
  1. a

    Percentages do not add to 100% because categories are mutually exclusive.

Average age of referred child7.634.8779.664.639
Number of children in the family1.920.9601.891.000

Case outcomes

Figure 3 illustrates comparative case outcomes. DV-involved cases were substantiated more often (72.1%) in this sample as compared with 40.9% of all other forms of child maltreatment (P < 0.001). DV-involved cases remained open for ongoing CPS more often than for other forms of maltreatment (P < 0.01). Over thirty-five per cent (35.4%) of DV-involved cases remained open after investigation as compared with 24.4% for all other forms of child maltreatment. In the DV-involved cases, only one-third (33.3%) of the children were evaluated by workers as having been emotionally harmed from EDV. However, this was significantly higher than all other forms of child maltreatment of which 22.8% of the children investigated were assessed to have been harmed as a result of the alleged maltreatment (P < 0.01).

figure

Figure 3. Case outcomes by type of maltreatment.

Download figure to PowerPoint

More DV cases were referred to community-based services (59%) as compared with 49.1% for all other forms of child maltreatment (P < 0.01). Over half of the DV cases were referred to services mostly for parent/family counselling and victim services for the victim parent.

Risk assessment scores and recurrence

Final risk assessment scores on the Ontario Family Risk Assessment Tool (OFRAT) were obtained on 785 cases sampled in the study period. The OFRAT is an actuarial risk assessment instrument based on the Children's Research Center's Structured Decision-Making system, California version, which has been validated in California (Baird et al. 1999; Baird & Wagner 2000) as well as Ontario (Shlonsky et al. 2011). The OFRAT is administered during investigation and measures risk using a four-level classification scheme (low, moderate, high and very high risk). Comparative analysis showed that DV cases were less likely to be rated low risk – 5% vs. 18% for other forms of maltreatment (P < 0.001) – and more likely to be rated high risk – 38% for DV cases vs. 25% for other forms of maltreatment. The bulk of the risk assessment ratings for DV-involved cases were in the moderate- and high-risk range (88.5%) (Fig. 4).

figure

Figure 4. Final risk rating by type of maltreatment.

Download figure to PowerPoint

In terms of recurrence (cases re-investigated at 18 months), there were no significant differences between DV cases and other forms of maltreatment. The overall rate of recurrence was about 28% for all cases, with DV cases recurring about 31% of the time and non-DV cases recurring about 27% of the time with rates of substantiation under 20% (Fig. 5).

figure

Figure 5. Re-investigation and substantiation by maltreatment type (18 months).

Download figure to PowerPoint

Case record reviews

Thematic analysis of the 75 case record reviews revealed some notable features. The pathway of DV cases indicated that first openings for DV referred cases tended to result in longer and more intensive involvement supported by referrals to community services. Subsequent child welfare involvement in reopened files tended to be shorter. Community referrals were often aimed at the victim parent with most referrals being made to victim services (i.e. shelters) and family counselling (i.e. parenting groups) with fewer than expected referrals for the children such as children's mental-health services. Core themes that emerged are as follow:

  • Tendency towards more intensive child protection involvement in DV cases in first investigations of DV
  • Child welfare involvement was longer in cases where there was a concern that mothers might stay with or return to the abusive partner
  • Very little focus on perpetrating fathers and very little information on fathers in the workers' written narratives beyond the assault event
  • Referrals beyond the child protection system focused mostly on services for the victim parent–victims and parent counselling, and significantly less for the exposed children specifically

The following examples are case vignettes that illustrate these core and sub-themes.

Case Vignette 1: Nandi was visited by child protection workers after a police referral. Her partner Ben had physically assaulted her while her children, ages 9 and 10, slept in their bedroom. Both Nandi and Ben were drinking at the time, although the police report stated that he was clearly intoxicated and she was not. He was arrested and released on bail with a no contact condition with Nandi. She told the CPS workers she was ‘done with him’ and would cease all contact. The case had been opened twice previously with CPS. The first time was when her husband Jim (biological father of the children) assaulted her and was removed from the home. The case remained active for 10 months while Nandi re-established herself as a single parent. Jim was ordered through the courts to attend a perpetrator's treatment group. Her worker met with her regularly and referred her for community-based services for victims of violence. Nandi and her children attended a group program for children exposed to DV. The case was closed with CPS upon completion of that group. It was reopened later when her ex-husband assaulted her on an access visit where he verbally assaulted her and threw a hat at her. The police were involved and he was warned but not arrested because no physical assault occurred. CPS became involved again for 3 months in supporting her to secure a restraining order and arranging visitation through an access centre for child exchange purposes. It is not clear what intervention occurred with the father. The recent opening resulted in an investigation but Nandi did not want ongoing involvement assuring the worker that she would have no contact with Ben. CPS spoke with the family doctor and children's school and no concerns were noted. Her worker reminded her of the potential impact of DV on children and of services she could access for herself and children – shelters and parent counselling. The case was closed after 3 months.

In this first case example, the pathway is typified by the following:

  • Numerous CPS openings for DV
  • More than one perpetrating partner over time
  • Initial child welfare interventions are more intensive to support the mother
  • CPS is aimed at child and maternal safety and for securing services
  • Subsequent CPS involvement is often short term
  • Child protection involvement and intensity diminishes over time

Case Vignette 2: Amber is a 24-year-old single mother with a 3-year-old daughter. Her boyfriend Kyle, father of the child who periodically lived with them, assaulted Amber after she asked him to leave her apartment for good because he was dealing drugs out of the apartment. Kyle was unemployed and unmotivated to help out financially. Amber had returned to school to finish her degree, secure a job and start a better life for herself and daughter. The assault was very serious and she was hospitalized. Kyle was arrested but was out on bail within 24 hours. Later, he was court mandated to treatment and completed the program. Extended family cared for her daughter during Amber's recuperation period. CPS workers visited her at the hospital and started an investigation. Amber went to a shelter from the hospital as recommended by CPS and because of her fear of retaliation. The shelter staff helped Amber relocate and she started up her studies again. CPS closed the case after 4 months. Amber re-contacted them 1 year later because she felt her ex-boyfriend had located her or at least knew the school her daughter was attending. CPS did not get re-involved but instead referred her to the police and advised her to secure a restraining order. Amber was disappointed. She felt that the restraining order would be ineffective as she had heard from other women that the police do not follow-up on violations. She had no faith in the criminal justice system because she had to flee to a shelter in the first place when the perpetrator had been let out on bail.

The second case pathway is typified by the following:

  • Case moves from investigation to ongoing CPS because of the severity of the DV
  • Client takes up on services immediately
  • CPS is assured that mother is not going to return to the perpetrator
  • No contact is made with the perpetrator
  • CPS closes the case in a shorter period of time

Trends identified in these case vignettes indicated that different responses are applied based on specific aspects of each case. This is a customized approach believed to be more supportive especially because mothers consistently report fears of CPS involvement and possible apprehension of their children. However, mothers continue to be the primary focus of investigations with her actions determining the level of CPS involvement. In reading and rereading the case files closely, the concerns centred on the mother's ability or willingness to put distance between her and the abusive partner even when the partner was having visitation access to the children. Case notes on the abusive partners were scant.

With mothers as the primary focus of investigation, their capacity to parent was part of the investigation. This is evidenced by the frequency of referrals for the mother to parenting services. Additionally, unless the abusive partner was court mandated to treatment, there was almost no narrative evidence to indicate he was receiving any treatment for his abusive behaviour. Charges were laid in only one-third of the cases, so it is highly likely that at least two-thirds of the perpetrators were not mandated to treatment of any sort. In contrast, 59% of the victim parents were referred for services that included victim services, parent and family counselling, and about one-fifth (20.3%) of the children were referred for some form of child-focused counselling (Fig. 3).

Another trend to emerge in the case records was DV that occurred after the dissolution of the relationship, for instance, at access visits during the exchange of children. Despite the mother's actions to terminate the relationship with her abusive partner, she was often left to manage these visits on her own. Incidences of this type are not surprising given that many of the perpetrating parents would presumably have not received treatment for their offending behaviour and for understanding the negative impact on their children. Failing to involve the perpetrating father in the initial investigation holds ramifications far beyond the involvement of CPS outcome because children continue to have contact with perpetrating fathers and the fathers go on to partner with other women, involving other children, while having had no intervention.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Differential Response (DR) Models
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. References

Expected findings were that the rate of DV in child welfare systems in the current study sample is about one-quarter with mothers most frequently identified as the adult victim, which DV families are younger and more often detected through police intervention – all of which is consistent with other North American studies (Jones 2007; English et al. 2005; Kohl et al. 2005; Trocmé et al. 2008). Substantiation for DV-involved cases was significantly higher than for other forms of maltreatment, yet emotional harm to the child from EDV was only verified in about one-third of the DV cases, which was less than expected given the literature on negative effects of child EDV (Kitzmann et al. 2003; Wolfe et al. 2003; Holt et al. 2008). Of interest, the analysis showed that substantiation was often based on whether DV occurred, which is more easily verified because of supporting police reports. This finding indicates that substantiation refers more often to establishing that adult DV occurred and is less relevant in assessing harm to children. Substantiation of child emotional may be difficult to ascertain for a host of reasons. Workers may not have the necessary assessment tools; children may be too young to be assessed accurately, or children are not manifesting symptoms at the time of investigation.

Non-white families were over-represented in investigations for DV in this sample. One group that was significantly over-represented were Aboriginal children. This finding is consistent with what is cited in the literature regarding this persistent trend in child welfare systems (Blackstock et al. 2004; Blackstock 2007; Lavergne et al. 2008). In the current study, Aboriginal families were twice as likely to be investigated for DV and to be transferred to ongoing CPS.

The over-representation of non-white families in this study is difficult to interpret because structural and systemic variables could not be factored into the analysis – factors that may start to explain over-representation such as socio-economic variables – as a result of worker reluctance to fill out income and housing information. For example poverty has been associated with higher investigations of DV, financial stress can manifest in relationship conflict and newcomers (often non-white) to Canada are known to struggle financially in the first 5 years of their settlement (Alaggia et al. 2009; Maiter et al. 2009).

Another notable finding is that while fewer children than expected were assessed to be emotionally harmed, a significantly higher number of DV-involved cases remained open for CPS than cases for other forms of child maltreatment. This is in contrast to studies that show quite different case outcomes with workers closing DV-involved cases more often after investigation (Black et al. 2008) and inconsistently referring them to community services (Hughes et al. 2011). The higher rate of ongoing CPS involvement in this study is probably the result of the preponderance of DV cases assessed as moderate, high and very high risk. That is workers may have assessed DV-involved cases as riskier, more complex and needing greater levels of support, therefore commanding longer involvement. This resulted in the mothers' extended involvement with CPS and experiencing greater surveillance – a pervasive trend as evidenced by decades of research documenting the focus of CPS intervention to be predominantly with mothers (Humphreys & Absler 2011). As DR hinges on accurate risk assessment, in order to determine level of service involvement, workers may be erring on the side of caution in assessing risk as shown through case record analyses, putting weight on the actions of the parent who has been abused – primarily whether she is taking steps to leave the perpetrating partner.

This longer involvement may, however, be resulting in lower recurrence rates – rates lower than those reported in the USA (English et al. 2005). The study time frame allowed for analysis of recurrence at 18 months. Longer involvement may help workers to secure appropriate services and for families to connect with the community-based services beyond their involvement in the child protection system – perhaps more effectively addressing DV issues thus preventing re-referral to CPS.

A further explanation may be that CPS workers may be transferring cases to ongoing status because they are providing services that are not readily available and have wait times in community agencies. In other words, adequate funding has not been provided for preventive services and intervention programs at sufficient levels to address the increased volume of DV case referrals. Simply implementing a DR is insufficient. Increasing community capacity to service these families outside of CPS is critical; otherwise, the CPS system steps in with these families to fill service gaps.

One troubling finding that continues to persist in this and other studies is the under-involvement of perpetrating fathers in investigation or for ongoing CPS services (Strega et al. 2008; Humphreys & Absler 2011; Stanley et al. 2011). The main thrust of the work to be done logically lies with perpetrating parents. Yet their lack of involvement in the child protection system is profound. This study showed that only about one-third of the perpetrating partners were successfully contacted or investigated in DV cases. Case notes often indicated the inability to locate perpetrating partners, citing a number of factors. When perpetrating parents were on restraining orders, they were no longer in the household; therefore, they had no addresses to be reached at. Often, there were no responses to worker phone calls, messages were not returned or their phone services had been terminated. For those charged, the protracted nature of their cases going through the courts (sometimes up to 2 years) enabled them to still have access to their children during this time because they had not yet been proven guilty. This study finding echoes several others that have identified worker difficulty in locating perpetrators as well as the lack of worker training, concerns for worker safety, shortage of time to locate them and follow-up, and insufficient knowledge about services for perpetrating parents (Strega et al. 2008; Stanley et al. 2011).

Despite the advent of the DR in this jurisdiction, it appears that mothers continue to be the primary targets of investigation as their files remain open in the CPS system for extended service provision – significantly more often than for other forms of maltreatment. They are also the most frequently referred member of the family with only one-fifth of the children in this study referred for child counselling. These results beg the question, how much has changed over time? Gordon (1988) in her analysis of child protection files and more recently Humphreys & Absler (2011) in their content analysis of studies observe that mothers historically have been viewed ‘… as more amenable, introspective and self-critical: in other words, they were better subjects for professional casework’ (p. 467).

The one silver lining is the apparent lower CPS recurrence of DV cases in this jurisdiction. The longer involvement, as portentous as it might be for the mothers, may result in long-term diversion from further CPS intervention. This finding stands apart from other investigations that highlight the concerning trend of high recidivism of DV as compared with other forms of maltreatment (English et al. 2005; Marshall et al. 2010).

Limitations

This study examined child welfare outcomes for a short period of time – 18 months. Therefore, examination of the long-term impact of DR was not possible. As well, data were not available prior to the implementation of DR to conduct a comparative analysis of practice impact pre- and post-implementation. The sample was selected by the index referral complaint (DV or not DV), precluding accurate tracking of other forms of maltreatment that might have later emerged. It may also be the case that the training workers received had not translated into tangible practice, and a longer period of time is required to assess changes in practice and service patterns.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Differential Response (DR) Models
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. References

The implementation of DR in this study jurisdiction and within this sample may have resulted in DV cases, remaining in traditional CPS longer than cases referred for other forms of child maltreatment. In the current study, DV-involved families were rarely assessed to be low risk with half assessed as high or very high risk, therefore requiring more intensive CPS involvement. By and large mothers received the bulk of services, sometimes conjointly with their children. Of note, a significant proportion of these clients in CPS for DV issues were non-white women who were younger, with young children and possibly poorer.

Results such as these continue to shine a light on the pervasiveness of mother-blame trends running as undercurrents within child protection systems. Although the last few decades have ushered in models of CPS practice to redress this trend, the change focus has remained steadfastly fixed on mothers, even when they are as much the victims of DV as are their children. Accountability for exposing their children to violence is placed squarely on their shoulders. In the minority of cases where the perpetrator was charged and convicted, he was often presumed to have been dealt with through the criminal justice system and not in need of CPS intervention, especially if he had received treatment.

It is important to also keep in mind that the focus not be redirected to blaming workers. Instead, critical issues are raised in these findings. What is the true nature of the authority that CPS workers possess with perpetrators of DV from a parenting perspective? Currently, unless the perpetrator is dealt with in the criminal justice system, CPS workers appear to have few options to track and interview the perpetrator. In terms of assessment of emotional harm to the child, according to the case notes, workers turned to collateral reports where these existed to assess at arm's length child functioning, e.g. accessing medical records or school records – however, a number of the children investigated were not of school age. Is this the best way to assess child emotional harm and who is in the best position to evaluate this?

A murkier picture emerged around the longer period of CPS involvement with DV-involved families. Is the longer involvement resulting in effective long-term diversion? Are workers providing CPS in place of more suitable services in the community that may be otherwise lacking by ‘holding’ families during wait-list periods? Both anecdotal reports and a recent report published by the Canadian Institute of Health Research (CIHR) note the paucity of accurate numbers for service wait times but conclude that wait lists for children's mental-health services across Canada are at unacceptable levels – presumably the same services children suffering from negative effects of DV would utilize (CIHR – Institute of Human Development, Child and Youth Health 2010).

A number of systemic changes and actions seem necessary to take to improve DR to DV-involved families. In order to explain and address population or caseload issues, such as the over-representation of non-white families, CPS agencies should include a systematic approach to consistent collection of demographic data, such as income and housing, to support a longitudinal approach in tracking and to aid better understanding of case outcomes and recurrence.

Child protection policies may need to be reviewed and possibly reformed to enable workers to have the right type of authority to work with DV perpetrators from investigation through to follow-up services, including having a say in access visits. Worker training for perpetrator-specific engagement and management skills, within the context of parenting issues and impact on the parent–child relationship, is also indicated. As highlighted by Jaffe et al. (2003), policies also need to be regularly evaluated and evolve so as not to perpetuate unintended consequences for children and their families. Finally, adequate funding for community services and close partnerships between CPS and children's mental-health services would enable more timely access for those children deemed as potentially negatively impacted by EDV, so that they may be evaluated by professionals specifically trained in child assessment and treatment.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Differential Response (DR) Models
  5. Method
  6. Findings
  7. Discussion
  8. Conclusion
  9. References
  • Alaggia, R., Jenney, A., Mazzuca, J. & Redmond, M. (2007) In whose best interest? A Canadian case study of the impact of child welfare policies in cases of domestic violence. Journal of Brief Treatment and Crisis Intervention, 7, 275290.
  • Alaggia, R., Regehr, C. & Rischynski, G. (2009) Intimate partner violence and immigration laws in Canada: how far have we come? International Journal of Law and Psychiatry, 32 (6), 335341.
  • Alaggia, R., Regehr, C. & Jenney, A. (2012) Risky business: an ecological analysis of intimate partner violence disclosure. Research on Social Work Practice, 22 (3), 301312.
  • Baird, C. & Wagner, D. (2000) The relative validity of actuarial and consensus-based risk assessment systems. Children and Youth Services Review, 22 (11/12), 839871.
  • Baird, C., Wagner, D., Healy, T. & Johnson, K. (1999) Risk assessment in child protective services: consensus and actuarial model reliability. Child Welfare, 78 (6), 723748.
  • Black, T., Trocmé, N., Fallon, B. & MacLaurin, B. (2008) The Canadian child welfare system response to exposure to domestic violence investigations. Child Abuse and Neglect, 32, 393404.
  • Blackstock, C. (2007) Residential schools: did they really close or just morph into child welfare? Indigenous Law Journal, 6 (1), 7178.
  • Blackstock, C., Trocme, N. & Bennett, M. (2004) Child maltreatment investigations among Aboriginal and non-Aboriginal families in Canada. Violence Against Women, 10 (8), 901916.
  • Bourassa, C., Lavergne, C., Damant, D., Lessard, G. & Turcotte, P. (2008) Child welfare workers' practice in cases involving domestic violence. Child Abuse Review, 17 (3), 174190.
  • CIHR – Institute of Human Development, Child and Youth Health (2010) Access & Wait Times in Child and Youth Mental Health: A Background Paper. 15 October 2010. Available at: http://www.cihr-irsc.gc.ca/e/43055.html (accessed August 2012).
  • Conley, A. (2007) Differential response: a critical examination of a secondary prevention model. Children and Youth Services Review, 29, 14541468.
  • Corbin, J. & Strauss, A.L. (1998) Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Sage Publications, Los Angeles, CA.
  • Davies, L. & Krane, J. (2006) Collaborate with caution: protecting children, helping mothers. Critical Social Policy, 26 (2), 412425.
  • Devoe, E.R. & Smith, E.L. (2003) Don't take my kids: barriers to service delivery for battered mothers and their young children. Journal of Emotional Abuse, 3, 277294.
  • Drisko, J.W. (1997) Strengthening qualitative studies and reports: standards to promote academic integrity. Journal of Social Work Education, 33, 185197.
  • Edleson, J.L. (2004) Should childhood exposure to adult domestic violence be defined as child maltreatment under the law? In: Protecting Children from Domestic Violence (eds P.G. Jaffe, L.L. Baker & A.J. Cunningham), pp. 117. Guilford, New York.
  • Edleson, J.L., Gassman-Pines, J. & Hill, M.B. (2006) Defining child exposure to domestic violence as neglect: Minnesota's difficult experience. Social Work, 51 (2), 167174.
  • English, D.J., Edleson, J.L. & Herrick, M.E. (2005) Domestic violence in one state's child protective case load: a study of differential case dispositions and outcomes. Children and Youth Services Review, 27, 11831201.
  • Gordon, L. (1988) Heroes of Their Own Lives: The Politics and History of Family Violence: Boston, 1880–1960. Viking, New York.
  • Holt, S., Buckley, H. & Whelan, S. (2008) The impact of exposure to domestic violence on children and young people: a review of the literature. Child Abuse & Neglect, 32, 797810.
  • Hughes, J., Chau, S. & Poff, D. (2011) They're not my favourite people: what mothers who have experienced intimate partner violence say about involvement in the child protection system. Child and Youth Services Review, 33, 10841089.
  • Humphreys, C. (2000) Social Work, Domestic Violence and Child Protection: Challenging Practice. The Policy Press, Bristol.
  • Humphreys, C. & Absler, D. (2011) History repeating itself: child protection responses to domestic violence. Child & Family Social Work, 16(4), 464473.
  • Jaffe, P.G., Crooks, C.V. & Wolfe, D.A. (2003) Legal and policy response to children exposed to domestic violence: the need to evaluate intended and unintended consequences. Clinical Child and Family Psychology Review, 6, 205213.
  • Jones, L. (2007) Child welfare practice with families experiencing intimate partner violence. Journal of Aggression, Maltreatment & Trauma, 14 (4), 118.
  • Kantor, K.G. & Little, L. (2003) Defining the boundaries of child neglect: when does domestic violence equate with parental failure to protect? Journal of Interpersonal Violence, 18 (4), 338355.
  • Kitzmann, K.M., Gaylord, N.K., Holt, A.R. & Kenny, E.D. (2003) Child witnesses to domestic violence: a meta-analytic review. Journal of Consulting and Clinical Psychology, 71, 339352.
  • Kohl, P.L. & Macy, R.J. (2008) Profiles of victimized women among the child welfare population: implications for targeted child welfare policy and practices. Journal of Family Violence, 23 (1), 5768.
  • 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, 12031221.
  • Lavergne, C., Dufour, S., Trocmé, N. & Larrivee, M. (2008) Visible minority, Aboriginal, and Caucasian children investigated by Canadian protective services. Child Protection, 87 (2), 5976.
  • Maiter, S., Stalker, C. & Alaggia, R. (2009) The experiences of minority immigrant families receiving child welfare services: understanding how to reduce risk and increase protective factors. Families in Society, 90 (1), 2836.
  • Marshall, S.K., Charles, G., Kendrick, K. & Pakalniskiene, V. (2010) Comparing differential responses within child protective services: a longitudinal examination. Child Welfare, 89, 5777.
  • McDonald, S. (1999) Not in the numbers: immigrant women and domestic abuse. Canadian Woman Studies, 19 (3), 163179.
  • Nixon, K.L., Tutty, L.M., Weaver-Dunlop, G. & Walsh, C.A. (2007) Do good intentions beget good policy? A review of child protection policies to address intimate partner violence. Children and Youth Services Review, 29, 14691486.
  • Sev'er, A. (2002) A feminist analysis of flight of abused women, plight of Canadian shelters: another road to homelessness. Journal of Social Distress and Homelessness, 11, 307324.
  • Shaffer, M. & Bala, N. (2004) The role of family courts in domestic violence: the Canadian experience. In: Protecting Children from Domestic Violence: Strategies for Community Intervention (eds P. Jaffe, L. Baker & A. Cunningham), pp. 171187. Guilford, New York.
  • Shirwadkar, S. (2004) Canadian domestic violence policy and Indian immigrant women. Violence Against Women, 10, 860879.
  • Shlonsky, A., Noonan, E., Littell, J. & Montgomery, P. (2011) The role of systematic reviews and the Campbell Collaboration in the realization of evidence-informed practice. Clinical Social Work Journal, 39, 362368.
  • Stanley, N., Miller, P., Richardson Foster, H. & Thompson, G. (2011) A stop-start response: social services' interventions with children and families notified following domestic violence incidents. British Journal of Social Work, 41, 296313.
  • Strega, S., Fleet, C., Brown, L., Dominelli, L., Callahan, M. & Walmsley, C. (2008) Connecting father absence and mother blame in child welfare policies and practice. Children and Youth Services Review, 30 (7), 705716.
  • Trocmé, N., Fallon, B., MacLaurin, B., Daciuk, J., Felstiner, C., Black, T. et al. (2005a) Canadian Incidence Study of Reported Child Abuse and Neglect – 2003: Major Findings. Minister of Public Works and Government Services Canada, Ottawa.
  • Trocmé, N., Fallon, B., MacLaurin, B. & Neves, T. (2005b) What is driving increasing child welfare caseloads in Ontario? Analysis of the 1993 and 1998 Ontario Incidence Studies. Child Welfare, 84 (3), 341362.
  • Trocmé, N., Fallon, B., MacLaurin, B., Sinha, V., Black, T., Fast, E. et al. (2008) Characteristics of substantiated maltreatment. Canadian Incidence Study of Reported Child Abuse and Neglect: Major Findings (Chapter 4). Available at: http://www.phac-aspc.gc.ca/cm-vee/public-eng.php (accessed January 2013).
  • Waldfogel, J. (1998) The Future of Child Protection. How to Break the Cycle of Abuse and Neglect. Harvard University Press, Cambridge, MA.
  • Wolfe, D.A., Crooks, C., Lee, V., McIntyre-Smith, A. & Jaffe, P. (2003) The effects of exposure to domestic violence: a meta-analysis and critique. Clinical Child and Family Psychology Review, 6, 171187.
Notes
  1. 1

    The increase in DV cases is likely a result of legislative changes rather than an actual increase in the number of children exposed to DV.

  2. 2

    A relational database is a series of linked tables used to store multiple repeating events while preserving the unique features of observations and the relationships between individuals.