‘The needs of the child have been met’: preliminary assessments regarding domestic violence in Swedish Child Protections Services

ABSTRACT

The Swedish Child Protection Services (CPS) are responsible for providing support and protection to children experiencing domestic violence, but in a high proportion of cases where there is a suspicion that children are experiencing violence, no investigation is opened.

The aim of this article is to explore on what grounds decisions were made not to open an investigation in cases initiated due to concerns that a child might have been experiencing domestic violence.

The analysis is based on qualitative data consisting of 116 preliminary assessments that resulted in the cases being closed without further action. The analysis uses a theoretical model in which the CPS sorting process is understood as a way of describing a case in such a way that it fits into the organisation’s problem categories, rules and procedures.

The analysis shows that a set of institutional assumptions and presuppositions affect the decision to screen out cases involving this form of child abuse. The legal requirement for voluntary parental cooperation has a profound impact on the outcome of a case. Cases are also screened out when the case workers find that parental cooperation is limited. Nor are there procedures for dealing with (psychological) violence that is not aimed directly at the child, but occurs between the parents. The outcome in these cases is that the parents’ interpretation is given precedence, while the needs of the child are seen as of less importance and become deprioritised.

KEYWORDS:

• Children experiencing domestic violence
• child protection services
• preliminary assessment
• child protection services funnel
• needs

Introduction

Experiencing domestic violence can have severe consequences for a child. Children who live in families where domestic violence occurs are at a higher risk of psychological, cognitive and social (school-related) problems during childhood and as adults (Holt, Buckley, and Whelan 2008; Olofsson et al. 2011; Graham-Bermann et al. 2012). The consequences of childhood exposure to such psychological abuse have been known for a long time, but it was not until the late 1990s that it attracted attention in the Nordic countries (Eriksson 2007a). The understanding that experiencing domestic violence is a form of child abuse has had important consequences both in Sweden and in other countries (Kaufman Kantor and Little 2003; Överlien 2010). In addition to its symbolic significance, this has strengthened the rights of young people and their access to public institutions (Coohey 2007). The responsibility of the Swedish social welfare system to provide support and help to women and children affected by experiencing domestic violence was clarified in the Social Services Act of 1997 and has been strengthen several times after that (Ekström 2016). As a result children experiencing domestic violence has received greater attention. Domestic violence has received more recognition as a problem area and the operations of the Swedish social welfare system have been modified in a number of ways: for example, staff have been trained in appropriate skills and dedicated social service units have been established that provide help to women, children and men who have been experiencing domestic violence (Mattsson 2013; Swedish Association of Local Authorities and Regions 2016). Thus, there is greater readiness and ability to manage the issue of children experiencing domestic violence. However, current studies show that the Child Protection Services (CPS) still have difficulty handling domestic violence (Eriksson 2012; Münger 2015; Mattsson 2017a). The pattern is the same from an international perspective (Coohey 2007; Stanley et al. 2011; Överlien 2010).

Figure 1. The child protection services funnel (Östberg 2010, 17–19).

A particular area of concern within the work of CPS is that of the preliminary assessments made in the initial stage of a case, which determine whether or not a child protection investigation is to be opened (Figure 1). The governmental supervisory agency, the National Board of Health and Welfare, has previously (2012) observed that, across the board, the number of notifications of child protection concerns is increasing, while the likelihood of an investigation being opened is decreasing. Thus, a high proportion of cases are screened out through preliminary assessments, though the extent varies between different municipalities. Between 30 and 60 per cent of all notifications of concern are closed without further investigation (Östberg 2010, 2014). Moreover, far too many preliminary assessments, regardless of the problem, are not conducted in accordance with regulations. For instance, in cases of male violence against women, risk assessments are not carried out in relation to the children, but only in relation to the woman (Health and Social Care Inspectorate 2014a). However, there is only limited knowledge about preliminary assessments, and there is inadequate understanding of the reasons why cases of child experiencing domestic violence do not trigger a child protection investigation (Health and Social Care Inspectorate 2014b, 2017).

This article focuses on preliminary assessments that are closed with no further action. The overall aim is to explore the grounds for the decision not to open a child protection investigation although there is concern that the child has been experiencing domestic violence. Key questions are:

• What is the understanding and evaluation of the concern about violence that triggered the assessment?

• How do the child’s and parents’ accounts affect the decision?

• What can be discerned in terms of the case workers’ difficulties and limitations in assessing whether an investigation should be opened?

The study was conducted at a CPS unit in one Swedish municipality. The data consists of preliminary assessments that were closed with no further action. All were initiated due to concern that children might have been experiencing domestic violence or a determination that domestic violence had in fact occurred. The article begins with a description of the Swedish context, with focus on CPS’ responsibility for and work with child protection investigations, after which the theoretical and empirical premises of the article are outlined. The analysis is organised under four thematic headings: Parents who speak for the child, Contact with CPS is declined, Violence but no concern and Violence eluding definition, derived on the level of theory from how a problem complex such as children experiencing domestic violence is dealt with and managed in the case management process of the child protection services.

The Swedish context

Based on the latest figures from the Swedish National Council for Crime Prevention (2014), an estimated 150,000 children in Sweden have been experiencing to psychological abuse through witnessing violence between their parents. The number of unreported cases of adult victims of intimate partner violence is high. Accordingly, a high proportion of the children in these families are also unseen. In Sweden, 6.8 per cent of the adult population report that they have been victims of violence perpetrated by a relative, friend or partner. Few have had any contact with public institutions such as the social welfare system, or the healthcare system. According to Melton (2005), the absence of statistics is evidence of a lack of tools to identify the problem. Not only does the violence remain a family secret, the lack of reliable data means that society is unprepared to deal with the problem.

Child Protection Services have the ultimate responsibility for intervention when parents fail to meet the needs of their children. The right and duty of CPS to intervene, if a child is maltreated or is a witness to domestic violence, is regulated by and clearly expressed in the Social Services Act (SFS 2001:453). Such psychological violence is considered equally as serious as when the child has been a direct victim of violence, and such children are also recognised as victims of crime under the Social Services Act. Grounds for a potential intervention are established through the child protection investigation. The preliminary assessments are conducted by reception units and are aimed at determining whether the child needs help and protection. If there is reason for concern that the child’s needs are not met, a child protection investigation must be opened (National Board of Health and Welfare 2015). The Swedish Social Services Act is based on the principle of voluntary participation, meaning that the parents must give their consent to CPS interventions, unless there are grounds for compulsory measures in order of the Care of Young Persons Act (SFS 1990:52). Although Östberg (2010, 2014) has shown that what she has called the ‘voluntary participation ideology’ plays a crucial role for when and whether CPS can intervene (Östberg 2010, 198).

This article is based on a case study carried out in 2012, and since then the regulations determining how the social welfare system must act when made aware of children experiencing domestic violence have been tightened. As of 2014, an investigation must be opened without delay if there is concern that the child has been the direct or indirect victim of domestic violence. Thus, the social welfare system is now required to initiate investigations in all cases of concern that a child has been experiencing domestic violence (National Board of Health and Welfare 2014). A lot of resources has been used on action plans and developing methods in order to implement the new regulations. However, governmental inspections (National Board of Health and Welfare 2016; Swedish Association of Local Authorities and Regions 2016; County Administration Board of Östergötland 2017: Health and Social Care Inspectorate 2017) and Swedish Government Official Report (SOU 2015:55) shows that problems remain and the regulations have not yet been implemented in the management and quality assurance systems of the CPS. Investigations are not always initiated and preliminary assessments are closed without further action so that children experiencing domestic violence are still at risk of being invisible within the CPS. Also, knowledge concerning the primary assessment process is limited which makes this study relevant in order to gain more knowledge in general on how and why cases are screened out and closed.

The work of CPS with children experiencing domestic violence

Studies from Sweden (Eriksson 2012; Münger 2015) and abroad (Devaney 2008; Humphreys and Absler 2011) show serious shortcomings in the ability of CPS to identify, confront and deal with children experiencing domestic violence in a satisfactory manner. What CPS do in connection with notifications of concern and child protection investigations is particularly limited in respect of this type of child abuse. The few studies that have been published illustrate the serious difficulties involved in identifying and managing this form of violence (Coohey 2007; Stanley et al. 2011; Stanley and Humphreys 2015 and, for Sweden, Sundhall 2015). In particular, there is a lack of studies on how violence is addressed in preliminary assessments (Irwin and Waugh 2007; Münger 2015). The violence is not by itself considered a risk factor: there must be other social problems for the social welfare system to intervene (Black et al. 2008). The violence becomes a subordinate factor in child protection investigations and the violent incidents are consequently downplayed or recast as mere arguments or disagreements (Coohey 2007; Överlien 2010; Stanley et al. 2011; Münger 2015; Mattsson 2017a). As a consequence, it has been difficult for children´s experiencing domestic violence to gain legitimacy as a specific social problem and it is not assessed based on the same criteria as other forms of child abuse. One explanation is that the work of CPS is informed by a discourse in which the woman is the victim of violence, the man the perpetrator and the child fades into the background (Fairtlough 2006; Eriksson 2007b; Hester 2011; Münger 2015). Another explanation might be that case workers lack necessary assessment tools and knowledge about the dynamics of domestic violence in order to assess the need of further investigation (Stanley and Humphreys 2014; Alaggia et al. 2015; Mattsson 2017a, 2017b). Also, the Swedish CPS are sometimes referred to as a family support system; that is, interventions should primarily be carried out with the parents’ consent and there is limited scope for compulsory measures this plays a crucial role for when and whether CPS can intervene (Östberg 2010, 2014). According to this parents are afforded a great deal of latitude even when the matter involves domestic violence (see e.g. Eriksson 2012). The focus on the child and the child’s perspective tends to be subordinated to the adult world and it is the parents’ perceptions of the violence and their willingness to cooperate with CPS that determine whether the social services are able to provide support and help for the children (Fairtlough 2006; Eriksson 2007b; Hester 2011; Münger 2015; Sundhall 2015).

Theoretical foundation

The premise of the article is that the preliminary assessment is understood as constructed in and for the contexts in which it serves a function. This reflects a public authority perspective in which the social worker’s identification of problems and needs is shaped by the institutional system’s logic, culture and rules (Järvinen and Mik-Meyer 2004). Preliminary assessments can thus provide a picture of how cases are categorised so as to fit in with or to be excluded from the CPS case management system. Östberg (2010) has noted that it is within the CPS institutional framework that a child’s needs for a social response must be identified, evaluated and addressed. A good illustration of how this response can express itself is provided by the now internationally recognised ‘child protection services funnel’:

The funnel is a metaphor for the CPS filtering process, where the topmost, larger opening illustrates the great number of reports received in relation to the few cases that are assigned some type of action. The size of the funnel’s bottom part is thus determined by the number of cases that are assigned an action. The dimensions of the funnel are determined by institutional conditions such as organisation, finances and regulations (Östberg 2010). Evaluations by social workers act as the filter that determines whether a case is to be closed without further intervention. This process is controlled by the mandate and institutional culture of CPS. It is accomplished through information gathering by the social worker, which takes place largely without the child’s participation. The information obtained is compiled based on the problem categories recognised within the organisation. Each case is therefore assessed within existing frameworks of understanding that determine how problems and needs should be interpreted. The process is complex, because a case must be identified as a problem and constructed in a way that fits the organisation’s categories, rules and procedures. In other words, for a social worker to intervene in a client’s life, the description of the problem, but also the solution to the problem, needs to fit into the organisation (Järvinen and Mik-Meyer 2004). Equally important, the client must agree with the organisation’s understanding of the problem, since interventions are based on the principle of voluntary participation. If the problems that have been identified are not aligned with the organisation there is a risk that they will be screened out and ignored by the system (Östberg 2010).

Method

The article is a qualitative document analysis (Mason 2002). The data was generated in a 2012 study which aimed at investigating how the social welfare system (CPS) identifies and handles children in families with problems involving violence. The study was conducted at a CPS unit in a large Swedish municipality, whose population structure and rates of crime, social problems, unemployment and poverty are highly representative of a Swedish municipality. During 2012 a total of 491 (375 + 116) referrals were made concerning children experiencing domestic violence. In 375 of them (16% of all child protection investigations that year) a child protection investigation was opened. This study is focused on the remaining 116 assessments which did not lead to a full investigation.

At the unit, all preliminary assessments closed with no further action were filed chronologically, based on the date the notification was received, in the so-called ‘trash binder’. The ‘binder’ for 2012 actually consisted of 45 binders, containing all notifications of concern (a total of 3,660 preliminary assessment), submitted to the social welfare system involving both children and adults. In ‘the trash binder’, half of the assessments were about children and young but none were related to children exposed to physical/sexual abuse or parental drug/alcohol misuse The assessments focused largely on neglect, for example submitted reports of bad clothes and bad eating habits from schools or dentists reports of missed appointments.

The quality of the assessments varied, some were just simple notes while others were drawn up as complete case files. All were read through in order to sort out assessments concerning children experiencing domestic violence. Of the 3,660 preliminary assessments in the binders, 116 were by the CPS coded by ‘violence’, ‘conflict’ or ‘honour-related’. These assessments were in general comprehensive and rich in information. They comprised between 12–15 pages and resembled ordinary case files. However, there were variations in how the situation that lead to the referral was described. In some of the assessments the violence was well described with comments and analysis from, for example, the police intervention. In others, the information was scarce and could consist of short notes that the parents could not be reached. In all 116 assessments, the actual referral was documented in some way, for example as an appendix. The referrals were most often done by the police (107) but also by preschool and school (3), other units within the social service (3) or private persons (3).

Processing and analysis

The analysis was carried out by hand and in three stages (Silverman 2006). The first stage, described above, focused on sorting out the cases concerning domestic violence. The 116 assessments was sorted out on the basis of that the CPS had coded as risk of a child experiencing domestic violence. The rest of the analysis was focused on these 116 assessments. In the second stage, the ideas and reasons found in the preliminary assessments were thematised and assessed against the theoretical premise that cases are categorised through logic and praxis to fit into administrative systems. This interpretation was further developed at the third stage of processing, where focus was on identifying the reasons discernible behind the decision not to open an investigation. The material showed wide variations in how parental violence and conflicts were described and how these descriptions contributed to the decision not to open a child protection investigation. However, through the analysis, four general themes could be derived from the various preliminary assessments: Parents who speak for the child, Contact with CPS is declined, Violence but no concern and Violence eluding definition. These themes have been used for the presentation of the material in the article.

Ethical considerations

The study was approved by the Regional Ethical Review Board, and was conducted in compliance with existing guidelines for research in the social sciences. In the first stage of the analysing process the data included all information of the clients. In the second stage the data was de-identified and personal details in the preliminary assessments were replaced by statements of gender and age. Incidents and situations that could be linked to individuals or the municipality were deleted or changed (Swedish Research Council 2011)

Findings

The needs of the child have been met

All notifications of concern are subject to preliminary assessment in the municipality in question. The assessment aims to gain an overall view of the child’s need of protection and support. The procedure in this municipality is for information to be obtained from the parents, but the caseworkers differ in the ways in which they collect and process the information. The common practice is that children are not interviewed during the preliminary assessment, although this would formally and legally be possible due to the Social Services Act (SFS 2001:453). The preliminary assessments that have been analysed routinely end with the standard phrase ‘the needs of the child have been met by other means’. The following analysis lays bare the reasoning that precedes this conclusion and the grounds for a decision to close the preliminary analysis with no further action.

Parents who speak for the child

A recurrent feature when caseworkers decide to close preliminary assessments with no further action is that it is the parents, and not the child, who have an opportunity to describe the situation in the family. Moreover, the parents’ accounts may in various ways tone down the extent of the child’s exposure to violence or the severity of the act of violence itself. Because it is the parents who provide the account of how the child perceives the situation at home, the opportunity for the caseworkers to inform themselves about child’s wellbeing is reduced. In one case, a notification of concern was submitted about a boy in his late teens by the head of the municipal agency where the boy’s mother works. According to the notification the boy had contacted his mother’s manager because his mother was being battered by her partner. The brief preliminary assessment notes that CPS had proposed several different times for a meeting with the mother. Finally, the mother rang up and the following was reported in the case record:

The mother says she is fine and that things are peaceful now. The caseworker asks how her son is doing and she says that he is fine too. She does not think it is necessary for us to meet at the suggested time./ – /. I wished her and her son well and pointed out that help is available and that no one should have to be beaten up.

The caseworker did not ask the mother any questions about the assault itself. Her statement that both she and her son are fine determined the decision not to investigate further, on the grounds that ‘there was no reason to open an investigation against the child’s or the mother’s wishes.’ The conclusion that they neither needed nor wanted contact was based on the mother’s statement. The assessment clearly shows that there were no conversations with the boy. The principle of voluntary participation makes it difficult for caseworkers to act when the mother declines discussion and help. Because they did not interview the son, even though he seemed to perceive the situation differently, the assessment was based solely on the mother’s statement.

The caseworkers may also be constrained when the parents provide contradictory accounts. This occurred, for example, in a case involving a twelve-year-old girl. In this case, the child’s fear of her father was reported via the mother:

She is still deeply affected by the parents’ separation and the conflicts between them. It emerges in the conversation with the mother that both she herself and the daughter are afraid of the father, as they perceive him as angry and sometimes threatening.

The father’s account was also reproduced, and his view was that the daughter’s fear was exaggerated. In the case record, the caseworker described his view of the relationship between him and his daughter:

He does not feel that his daughter is afraid of him/ – /. He feels that the mother is transmitting this fear to the daughter./ – /The father says that he asked his daughter to forgive him. She then texted back, “It will be great to see you, love you”

The case record stated that the father had asked for forgiveness, but that the daughter’s positive response had not been verified. There was no suggestion in the case record that the daughter should be contacted for her side of the story. The decision to close the case was justified by the caseworker’s opinion that the parents ‘seem eager for the daughter to have a secure childhood.’ Instead of opening an investigation that might make the situation in the family clearer, many caseworkers tend to rely on indications that the violence was inconsequential and that the parents have good intentions. On that basis the assessment is then closed with no further action.

Even when parents admit that acts of violence have occurred, cases are sometimes closed because the parents insist that the child was not harmed by what happened. Parents offer various reasons for why the violence did not affect the children, such as that the children did not notice or understand it, were not present or were sleeping. Such rationalisations related to the child’s wellbeing seem to weigh heavily with CPS. For example, one case was closed for the following reason:

The notification states that children were present, but did not in these cases notice the parents’ arguments. Nor does the mother admit that the children have any awareness of the parents’ arguments or that the husband has threatened or beaten her.

CPS came to this conclusion even though they seemed to regard the case as serious. It was based on a notification following a police intervention in a situation in which the father had hit the mother. In the preliminary assessment it is noted that the caseworker had engaged in the following discussion with both parents:

I pointed out the father’s serious responsibility to refrain from expressing his aggression with threats and violence against family members. The caseworker also makes clear that if the father is unable to accept his responsibility, it is the duty of the mother to protect her children. She insists that the children have never seen or heard the parents fight: on the night in question, the children were playing video games and did not hear what was said. The caseworker points out that conversations with children in families where arguments and assaults happen show that they do hear, even though the parents do not, or do not want to, believe it.

This discussion suggests that CPS had concluded that the father was inflicting violence on the family. Nevertheless, based on the mother’s statement about what the children had (not) seen or heard, their assessment were that the children had not been aware of the violence between the parents.

It is difficult to determine whether the decision to close a case is affected by the municipality’s standard practice of not interviewing the children during preliminary assessments. What does become clear, however, is that the parents are given preferential rights of interpretation and that this has a profound impact on the caseworkers’ determination of how the case should be categorised. When the caseworkers decide to close a case with no further action, this means that they will not open a child welfare investigation, which might have provided more information about the violence and the child’s wellbeing, for instance through conversations with the child. Research has shown that children often are very aware of domestic violence, even when the parents are convinced that they are not (Överlien 2010). Although the caseworkers seem to be aware of this, nevertheless the parents’ accounts are allowed to form the grounds for a decision that the matter is not a case for CPS.

Contact with CPS is declined

Another recurring reason for the decision to not open an investigation is that the parents decline contact with CPS. In a just above half of of the 116 cases, the decision was based on the parents’ unwillingness to have any contact with the CPS. Several such cases demonstrate that this can easily lead to the basis for the decision being inadequate. In one case where the mother declined contact, it was reported that no new information has emerged since the notification was received:

No information has emerged that could lead to concern that the boy might be maltreated in his home to such an extent that further investigation would be required. The mother also clearly states that the family does not need support.

Whether it was mainly the lack of new information or the mother’s unequivocal rejection of CPS that determined the matter is not clear from the case record. The combination of an absence of information and the parents’ refusal of further contact with the CPS is, however, common in the explanations of such decisions.

In just over a third of the analysed cases, CPS have unambiguous information about domestic violence right from the start of the case management process. This may, for example, be in the form of police reports where the documentation includes a pre-printed sheet, where the police can tick boxes indicating ‘child(ren) in danger’, ‘child(ren) experiencing domestic violence’ and ‘child(ren) in an unsuitable environment’. In several of these cases, it was nevertheless the parents’ refusal to have contact with the social welfare system that determined the matter. In the following case, the parents’ inadequate response to attempts by CPS to make contact was interpreted as a sign that no contact was wanted, and this is the reason given for closing the preliminary assessment with no further action:

The notification makes clear that there have been arguments between the parents and that when the police came to the home, they also met the girl. They [the police] do not express immediate concern for the child and perceive her to be curious and in a good mood. The parents have been called to two appointments, one of which by letter written in their native language. They have not attended at the suggested times or otherwise made contact. It is my assessment that the couple are not interested in contact with the social services office.

This case was preceded by police intervention in the family because the father had been violent towards the mother, for which he was remanded in custody. Even though the police had ticked the boxes on the pre-printed notification sheet indicating that the case involved both a ‘child experiencing domestic violence’ and a ‘child in danger’, the CPS decided not to open an investigation. From the remark in the police report that the ‘child was in a good mood and curious’ the caseworker concluded that there was no immediate reason for concern. It also appears that the difficulty contacting the family was crucial, as the parents’ silence was assumed to mean that they were not interested in help. The assessment did not consider that the parents may have had difficulty understanding the information, even though the appointment letter was written in their native language. In another case, the police report stated that the child had been experiencing domestic violence, and it included the following statement: ‘I [the police officer] can see that the small child is afraid. I tell both parents to stop shouting and that they must consider their children.’ The social worker later wrote in the preliminary assessment:

The mother states that violence has not previously occurred in the family, although the parents argue. The mother further states that the children were not present when everything happened/ – /The caseworker emphasises the seriousness of children being present when shouting, arguments and violence occur.

In the decision to close the preliminary assessment with no further action, the caseworker stated that, despite some unanswered questions, there was no reason to open an investigation:

I (the case worker) find it very odd that the mother reasons as she does. She appears to make light of the argument described in the notification and how this might have affected the children. The notification does not refer to any recurring violence/ – /. Considering this and the fact that the family does not believe they need support, it was decided that the preliminary assessment could be closed with no further action. The mother has been informed about the programme to end violence against women.

The decision in this preliminary assessment did not explicitly take account of the information documented in the police report. The assessment was entirely based on conversations with the mother, although these are very much at odds with the observations by the police. The caseworker notes in the case record that the mother makes light of the occurrence of violence and that the children may have been affected. On the contrary, CPS seems to have been persuaded by the mother that the violence should be regarded as ‘an isolated incident’ and that the family do not need help. Paradoxically, however, the mother was informed about the municipal programme to end violence against women.

There is thus a clear pattern in the analysed preliminary assessments of how parents’ reluctance to have contact with CPS reduces the case workers’ opportunities to obtain information about the family’s situation. The lack of information and the absence of apparent grounds under the Care of Young Persons Act (CFS become reasons for not opening an investigation. This can be understood as that what Östberg (2010, 2014) has called ‘the voluntary participation factor’ is of critical importance for whether or not child protection investigations are opened. One possible interpretation is that the caseworkers’ and the parents’ views of the family’s needs and potential problems diverge in these cases; in several, the caseworkers believe there is a problem, but when the parents themselves do not see the problem, the investigation goes no further. One interpretation of the reasoning of the caseworkers is that when they perceive the parents as unwilling to cooperate with CPS, they also regard the chances of conducting a more comprehensive investigation as limited. In such cases, the assessment is not only a matter of whether there are grounds for opening an investigation, but also, indirectly, of whether it would be fruitful to do so. At the same time, caseworkers clearly want to be able to support people in need. By supplying information about the municipal programme to end violence against women the caseworkers indicate that they have both identified and understood the challenges that exist within the family.

Violence but no concern

In several cases, CPS closed the preliminary assessment with no further action, even though they had been given clear indications that domestic violence was occurring. A consistent feature of these preliminary assessments is that the caseworkers base their decision on a judgement that there is nothing to be concerned about in the child’s situation. This is explained in various ways, such as the caseworkers’ focus on that the violence has stopped. Once again, focus is on the parents’ statements. In one case, for example, the mother of a boy, a year or two old, reported that she and her son had left their sheltered accommodation and returned to the home. She said that her husband ‘had hit her in the past, but does not do it anymore’. CPS acceped her statement that the violence had stopped and close the preliminary assessment with no further action. In another case, the relevant act of violence was briefly described: ‘Neither parent can explain why a knife was involved in the argument.’ The case record describes how the parents perceive their life situation and how the police intervention has affected them:

The parents cannot really explain why their relationship has become like it is./ – /. Both parents see the incident as a wake-up call. They believe they have changed their relationship.

The justification for the decision stated that the caseworker ‘finds the incident very serious’, but the violence was given less importance because the parents had reported that their relationship was changed. The caseworkers relied on what can be understood as a ‘change factor’, something that recurs in several cases. This kind of belief that the violence has stopped is based on varying descriptions in different cases, such as that the father has become kinder after starting medical treatment, that the substance misuse has ended or that the situation has improved since the couple divorced.

Marked indications of violence are thus sometimes disregarded in the preliminary assessments. Once again, the parents’ statements about the family situation are of central importance in the caseworkers’ assessment, but so are the caseworkers’ working methods, where the problems exhibited in the case get coded and classified at an early stage of the process. If domestic violence is not flagged at this point, there is considerable risk that the preliminary assessment will be limited to the other issues brought up at the initial stage.

Violence eluding definition

In several of the cases initiated due to apparent domestic violence, the problems and the violence prove difficult to assess, which in turn seems to affect the caseworkers’ decision to close the case with no further action. One example is a case that arose from a police report and which was categorised by CPS as ‘honour-related problems’. The case record stated that a three-year-old boy had been experiencing domestic violence and was living in an unsuitable environment. The case record stated that the mother and child had been living at a women’s shelter due to long-term physical and psychological abuse by the father and the extended family. When the prosecutor concluded that no crime could be proven and the police assessed the situation as low-risk, CPS stated the following:

It is very difficult to determine which of the parents is telling the truth about what happened in their relationship/ – /. Without dismissing the mother’s statements, what happened is, nevertheless, unclear.

CPS (and the legal system) maintained that it cannot be established beyond doubt that violence had occurred because the parents’ statements were contradictory. At the same time, the caseworker allowed for another interpretation and noted that:

If there has been violence and threats between his parents, it is in no way accepted that this is something a child should be exposed to. However, it is likely that such a destructive and dangerous relationship will end, because the parents will not be living together in the future.

The caseworker seems to vacillate as to whether violence has or has not occurred. The case was initially coded as honour-related problems, but this is not how the case record referred to the violence; instead it was defined as violence between the parents. These discrepancies do not trigger a more thorough investigation and the preliminary assessment was closed with no further action.

The absence of signs of physical violence can also make the violence harder to define and lead to the preliminary assessment being closed with no further action. One example is a case involving a ten-year-old girl in which the CPS assessment relied heavily on the following:

…nothing has emerged in school that indicates that the girl has been the victim of physical violence or that the girl was exposed to physical violence against her mother…

At the same time the case record includes several statements suggesting the presence of domestic violence, such as that the mother has said that she finds it hard to talk about her intense fear of her husband, and that she has also said that the child is afraid of him. The child has also, according to the mother, seen her father pushing and shoving her mother. Moreover, the father has already served a prison sentence for assault on the mother. Despite this, the statement that no indications of physical violence had been observed was significant for the decision to close the matter with no further action. In the absence of clear signs of physical violence it seems that it is difficult to take violence seriously (Mattsson 2017a).

Violence that seems hard to define and to put in concrete terms can thus become a reason for closing preliminary assessments with no further action – rather than a reason to initiate a child protection investigation, which would create opportunities for clarification. Because the violence is difficult to define the case then fails to be categorised as containing a concrete problem that needs to be investigated. The information becomes too vague for the caseworkers to decide to open an investigation. Instead of opening an investigation that could provide more information, the caseworkers choose to close the case with no further action.

Discussion

The child protection services funnel shows how child protection cases are handled within the institutional framework of CPS in such a way that it becomes possible to categorise them as ‘non-cases’ – matters that do not require investigation (cf. Irwin and Waugh 2007; Östberg 2010, 2014). This sorting process ensures that the case gets categorised in terms of the organisation’s established problem definitions, rules and procedures (Järvinen and Mik-Meyer 2004). The 116 cases analysed were screened out by the caseworkers, on the basis that there was not sufficient cause to open a child protection investigation, in the preliminary assessment entering the child protection services funnel. As current study is limited to one municipality, applications of the findings need to be sensitive to context. However, the more general character of the ways in which social workers do (not) identify and respond to this form of child abuse provide important knowledge to CPS. The implication of decisions to close cases is that the social welfare system has judged that the child’s home situation falls within the confines of what is normally accepted, meaning that there is no reason for society to intervene, even though most of the referrals came from the police and clearly identified risks of the child experiencing violence (cf. Stanley et al. 2011). The analysis shows that the decision to screen out a case is affected by a set of institutional assumptions, of which some can be interpreted as specific to matters related to children experiencing violence, while others are likely to generally affect the assessment process regardless of what initiated the assessment process. The first of these is the legal requirement for voluntary participation by the parents, which has profound impact on the outcome of the case (cf. Östberg 2010, 2014). The second involves what the caseworkers might do when, for example, they are unable to establish contact with the parents. When caseworkers feel that parents are unwilling to cooperate, they seem to see limited opportunities for a child protection investigation and thus choose to close the preliminary assessment. The third is that exposure to domestic violence as a specific form of child abuse has not yet been institutionalised or systematised within public institutions such as CPS. (Eriksson 2007b; Hester 2011; Münger 2015; Mattsson 2017a, 2017b). CPS largely lack procedures for dealing with non-physical violence that is not aimed directly at the child, but occurs between the parents. The child’s status as a crime victim is difficult to establish even when the caseworkers have a theoretical understanding of this form of psychological abuse and the child’s vulnerability in such cases (cf. Irwin and Waugh 2007).

The children in these cases are deprioritised. The wishes and opinions of the parents take precedence in the interpretation of (1) what happened, (2) how the children experienced the domestic violence and (3) the effect of the domestic violence on the children. Due to the absence of the child’s own voice, the child may be deprived of the opportunity to have some influence and define their own needs (cf. Eriksson 2012).

Under the new regulations, (National Board of Health and Welfare 2014) all matters involving children’s experience of domestic violence must be investigated. However the regulations have not yet been implemented in the management systems of the CPS (National Board of Health and Welfare 2016; County Administration Board of Östergötland 2017; Health and Social Care Inspectorate 2017). According to our analysis, based on preliminary assessments from 2012, one explanation might be in the CPSs filtering process, there is an insufficient tradition of defining experiencing domestic violence as a social problem – and this has not been changed. This form of child abuse has proved to be a problem that is difficult for the CPS to define and it is therefore at risk of being obscured.

Conclusion

This recurring pattern in the preliminary assessments highlights the importance of the child’s voice throughout the investigation process. The voices of parents are important, but preferential rights of interpretation must be shared among children, parents and caseworkers. It is obvious from this article that attitudes around children experience of domestic violence must fundamentally change. Experience of domestic violence must be regarded as child abuse in politics and practice (cf. Kaufman Kantor and Little 2003; Irwin and Waugh 2007). The existing legislation have to be implemented better within the community organisations. It is important to develop administrative processes and routines based on child perspective. Training of social workers is also an important part of the work to ensure that this form of abuse receives attention and is acted upon. Responsibility for this rests on the municipal council.

Additional information

Funding

This work was supported by the The Swedish Crime Victim Compensation support and Authority Grant number 3526/09.