‘Their Mum Messed Up and Gran Can’t Afford to’: Violence towards Grandparent Kinship Carers and the Implications for Social Work

Child and adolescent violence towards grandparent kinship carers is a significant and yet under-researched phenomenon. This study draws on data from thirty-six in-depth interviews which include grandparent carers who are experiencing such violence, and professionals from a range of backgrounds whose work intersects with this problem. The study highlights how the kinship care context shapes the violence, its impacts and, in turn, carers’ help-seeking practices. The findings highlight that social workers must better understand the barriers that prevent grandparent kinship carers from asking for help, and improve their responses to such requests. Recommendations for social work practice include asking the right questions, engaging in effective risk assessment, taking a trauma-informed approach, avoiding the language of coercion and improving its response to grandparents’ own articulated support needs. Teaser text Child and adolescent violence towards grandparent kinship carers is a signiﬁcant and yet under-researched phenomenon. This study draws on data from thirty-six in-depth interviews which include grandparent carers who are experiencing such violence, and professionals from a range of backgrounds whose work intersects with this problem. The study highlights how the kinship care context shapes the violence, its impacts and, in turn, carers’ help-seeking practices. The ﬁndings highlight how social workers need to better understand the barriers that prevent grandparent kinship carers from asking for help and improve their responses to such requests. Key recommendations are made for social work practice.


Introduction
The problem of children and young people's use of violence in the home, particularly towards parents, has received significant research attention over the past decade. Violent and abusive behaviours reported in the research include verbal abuse (e.g. insulting, threatening the parent), economic abuse (e.g. stealing money, making financial demands), physical abuse (e.g. kicking, hitting and spitting) and emotional abuse (e.g. humiliating and intimidating the parent). Once invisible in the research landscape, an increasing number of predominantly criminological studies from Europe, North America and Australasia has identified it as a significant problem: community surveys with young people estimate prevalence rates of 5-15 per cent for physical violence and 45-62 per cent for verbal abuse (Holt, 2021). Such violence produces a range of immediate and long-term harms including physical harm (e.g. injury and death), psychological harm (e.g. anxiety and depression), financial harm (e.g. property damage and loss of income) and social harm (e.g. isolation from friends, family and communities). As with other forms of family violence, there is evidence that the problem has been particularly exposed during the Covid-19 global pandemic as children spend significantly more time at home and families face additional stresses and strains (Condry et al., 2020). That this represents a further example of gender-based violence is highlighted in cases reported to the police, which are significantly more likely to feature female victims and male perpetrators, with the peak age of perpetration at around 15 years (Condry and Miles, 2014).
However, most of the research on children and young people's use of violence towards parents has focused on 'parents' as a singular group. Little attention has been paid to the nature and arrangement of the caring relationship between parent/carer and child and how this shapes the violence that is experienced and the support options that are available. This research deficit is particularly pronounced in the context of kinship care, which refers to situations where children live with other family members (such as grandparents, aunts/uncles or older siblings) because their parents are unable to care for them. An Australian survey of kinship carers who had experienced family violence since the placement Teaser text Child and adolescent violence towards grandparent kinship carers is a significant and yet under-researched phenomenon. This study draws on data from thirty-six in-depth interviews which include grandparent carers who are experiencing such violence, and professionals from a range of backgrounds whose work intersects with this problem. The study highlights how the kinship care context shapes the violence, its impacts and, in turn, carers' helpseeking practices. The findings highlight how social workers need to better understand the barriers that prevent grandparent kinship carers from asking for help and improve their responses to such requests. Key recommendations are made for social work practice.
reported that 46 per cent of the children being cared for exhibited violent and aggressive behaviours towards other family members, the majority of which (89 per cent) were directed towards their kinship carer. The majority of these carers were grandparents and the aggressive behaviours included physical violence, property damage and verbal and emotional abuse. For 50 per cent of the carers, this violence was experienced daily or weekly (Breman and MacRae, 2017).
There is no existing data to estimate prevalence rates on violence towards kinship carers in the UK (nor, indeed, on violence towards parents more generally). However, in her opening speech at the launch of the report by the Parliamentary Taskforce on Kinship Care in September 2020, Vicky Ford MP, the Minister for Children and Families, identified the increasing problem of violence towards kinship carers as a cause for concern. Further, a recent UK survey into the impact of the Covid-19 global pandemic on kinship carers reported that children's behaviour, manifesting in violence and aggression, was the most commonly expressed concern amongst the kinship carers surveyed (Ashley et al., 2020). To address this research gap, we draw on data from the first UK study of violence towards grandparent kinship carers to analyse how the context of kinship care shapes the violence, its impacts and help-seeking practices, and discuss what this means for social work practice.

Background
Current estimates suggest that one in seventy-four children in the UK are placed in kinship care, with grandparents acting as primary carers in 51 per cent of cases (Wijedasa, 2015). Whilst there are many reasons why a child might be placed in kinship care, it is often due to a complex and inter-related set of problems including parental substance misuse and/or mental health problems, parental maltreatment, child abandonment, intimate partner violence, parental incarceration, parental illness or death and/or problems with the parents' housing Ashley and Braun, 2019).
In most cases, kinship care is arranged informally and privately between parent(s) and kinship carer. However, in cases where children's social care services are involved, then arrangements may be arranged through a formal order issued by the family court-for example, a Child Arrangements Order (previously a Residence Order) or a Special Guardianship Order (SGO). Each of these arrangements differ in terms of their degree of permanence, legal authority and parental responsibility awarded. The 1989 Children Act requires that kinship care is used wherever possible for the long-term placement of children who require out-of-home care. The reasons for this preference relate to assumptions about the importance of familiarity for the child, perceived better outcomes for the child and the relative low-cost and speed of kinship care placement compared with other options such as foster care and adoption (McCartan et al., 2018). However, the lack of status afforded to kinship carers and the poor levels of support offered means that kinship care is often referred to as the 'Cinderella' of the care system (Kiraly, 2015).
The challenges involved in kinship care, for both carers and for the young people being cared for, are well documented. Many of these represent structural challenges: for example, a high proportion of kinship care households live in poverty, a disproportionate number of children in kinship care are from ethnically minoritised backgrounds and children living in kinship care are twice as likely to have a long-term health problem or disability (Wijedasa, 2015). For many grandparents who become kinship carers, their new role may not have been anticipated and they may feel conflicted about the obligation of full-time care and the restrictions it imposes on their lives (Shakya et al., 2012). Further complications can arise from difficult relationships with the child's biological parent(s), which can be exacerbated by ongoing conflict over contact and may, in some cases, result in violence towards the kinship carer from the parent(s) (Breman, 2014). Whilst most children and young people are positive about kinship care , there can be confusion about why they are not living with their biological parent(s) and this can impact on their wellbeing and identity development (Staines and Selwyn, 2020). Compared with the general population, high levels of trauma have been found in children and young people who are placed in kinship care, with an associated risk of emotional and behavioural problems, stress-response functioning and challenges in executive functioning (Kemmis-Riggs et al., 2018).
The interaction between the care-giving context and the impact of children and young people's use of violence was highlighted in Selwyn and Meaking's (2016) research with adoptive parents. In thirty-eight out of the forty-five cases which featured adoption disruption, the child's violence towards their adoptive parents was a significant factor in the adoption breakdown. The majority of the adoptive parents described violent and controlling behaviours emerging early on when the child was still at primary school, though it was only during adolescence that parents became fearful of their child and 'normal' parental behaviour adapted to accommodate the violence. The adoption context made parents feel particularly vulnerable-for example, parents were fearful of their child's threats to make allegations of abuse against them, a tactic of control that was often used by the children. Furthermore, the adoptive parents' feelings of failure were exacerbated by poor responses from social workers, who often framed the problem not as one of 'family violence', but as 'challenging behaviour' as a consequence of 'poor parenting' which required intervention in the form of 'anger management' (Selwyn and Meakings, 2016). Such a framing is inherently blaming of the victim of the violence and fails to recognise the seriousness of the violence and its impacts. It also fails to appropriately address the needs of both the victim and the young person (see also Nixon, 2012;Holt and Retford, 2013).
There is little acknowledgement of this issue in current policy. In 2015, the UK Home Office produced 'Adolescent to Parent Violence and Abuse (APVA): Information Guide' (Home Office, 2015) for practitioners who may come across this problem in their work. This document framed the problem within the UK government's ongoing 'Violence against Women and Girls' agenda, and outlined how different local agencies, including children's social care, might better respond to it. However, this document does not mention kinship care at all. Even the section specifically written for children's social care workers (p. 15) only refers to cases that involve 'parents', with only one sentence referring to children in out-of-home care. The policy failure to recognise kinship carers is reflected in the research literature generally, and this knowledge deficit extends to social work practice where, to date, there has been little engagement of what this problem means for social work and, in particular, for the kinship care families that social workers engage with.

Methods
This study draws on data from thirty-six participants, which included twenty-seven grandparent kinship carers and nine professionals drawn from across England, Wales and Scotland. Rich qualitative interviews were conducted either face-to-face or remotely (e.g. through Skype) and lasted 60-90 minutes. The audio data were transcribed and subject to thematic analysis.
Grandparents were asked questions about their family relationships, the nature of the violence, its impacts and their responses. During analysis, key themes were identified whilst also attending to the gendered, generational, familial and life-course contexts of the participants' lives.
Information about the study, including requests to participate, was distributed through gatekeepers from a range of organisations, including family support services, local Youth Offending Services, kinship care support groups and child/adolescent family violence intervention programmes. The criteria was that the grandparent was a kinship carer and was experiencing violent and/or abusive behaviours from the grandchild(ren) they were caring for. We did not impose an upper or lower age limit on the children who were of concern because we were keen to understand the developmental context of the violence, and the imposition of age-based parameters would prevent this (see Holt and Shon, 2018).
In terms of the grandparent participants, their main characteristics were: Twenty-four grandmothers and three grandfathers (in two cases, we interviewed both the grandmother and grandfather from within the same kinship care household). Age range was 40-74 years (modal age ¼ 60 years). Two-thirds were married/co-habiting, the other third separated/divorced or widowed. Two-thirds were retired or unemployed, the other third in employment.
Of the twenty-five kinship care households: 19 ¼ maternal grandparents; 6 ¼ paternal grandparents. Eight had a Residence Order, eight had an SGO, two were foster kinship carers and seven had informal arrangements in place.
In terms of the grandchildren who were the focus of concern within the twenty-five households: Twelve grandsons and thirteen grand-daughters. Age range was 5-20 years (modal age ¼ 15 years). Approximately a third of the children had neurodevelopmental conditions (either diagnosed or suspected), including attention deficit hyperactivity disorder (ADHD), autistic spectrum condition (ASC) and foetal alcohol spectrum disorder (FASD). The children had been living with their grandparent(s) for a number of years, often from an early age; sometimes they had returned for a short period to live with their biological parent(s). In two kinship households, the grandchildren had been removed due to the violence: in one case permanently and in the other case temporarily.
Most of the participants were female and we only hear their stories. Yet even this tells a story: in cases where both a grandmother and a grandfather were living together, we always asked if we could interview both grandparents, but this offer was rarely taken up.
The nine professionals who were interviewed were selected because their work intersected with kinship care and children's use of violence in the home. Their professional backgrounds included children's social care; youth offending services; domestic abuse services; family law; the police; education and intervention work for adolescent family violence. These interviews enabled us to understand more about the challenges that different practitioners face in responding to violence towards grandparent kinship carers.
Prior to the start of the study, ethical approval was obtained under the procedures of the University Ethics Committee and the project was guided by an expert advisory board. All data are pseudonymised and any identifying features have been removed in the extracts.

Findings
In the analysis that follows, we explore how the kinship care context shapes (i) grandparents' understanding of the child's violence, (ii) its impact and (iii) help-seeking practices and engagement with social workers. This is followed by a discussion about what the findings mean for social work practice.

The kinship care context
The reasons why children live with kinship carers are common to many situations that involve out-of-home placements. However, unlike in adoptive or foster families, those reasons continue to play a significant role in the kinship care family's life. Ongoing difficulties that shape the parent's life, such as incarceration or drug dependency, need to be both emotionally processed and administratively managed by kinship care families. In many cases, this includes managing contact between child and parent.
In this study, grandparents described a range of circumstances which led to their role as kinship carer. Most cases involved a complex history of intimate partner violence, substance misuse, mental health issues and child abuse and/or neglect. In some cases, the incarceration or death of a parent or sibling also featured. As one grandmother explained: My daughter has a problem with drugs and her partner has a problem with alcohol. My daughter also has mental health issues and there is domestic violence between the two of them. My granddaughter also has a dad who is in and out of prison all the time, for domestic abuse. (Deborah; carer of granddaughter, 15) In cases of kinship care, these circumstances are more than 'background info' to understand the reasons for the placement or, indeed, the reasons for the violence. They also form part of the kinship carer's own personal and familial life story, and the difficult events described involved the carer's own child. In many of these cases, the trauma experienced by all family members was evident: grandparents narrated stories about visiting their daughter in hospital who was 'beaten black and blue' by a violent partner or about seeing their son 'ravaged' by the physical and emotional harms caused by drug dependency. Each of their stories featured some form of loss: My son and his partner had another baby and it died from cot death, and my grandson actually found his little sister and witnessed all the In a minority of cases, the grandparent's decision to take responsibility for their grandchild's care was a voluntary one, made together with the grandchild's parent(s). Sometimes, this was because the child's aggression was already a problem, and it was felt that placing the child with their grandparent(s) would resolve it. However, the majority of the grandparents did not feel that the decision to become a kinship carer was taken freely and many of the grandparents said that they felt coerced into taking on the role by being given an ultimatum by their social worker: I never actually met her until she turned up on my doorstep with a social worker when she was five. It was a case of 'if you don't have her, she'll go into the care system.' So that was that. She's been with us ever since. (Sandra; carer of granddaughter, 12) The ultimatums did not end there. Another consistent finding concerned the role of formal court orders, which were put in place in sixteen of the twenty-five kinship households following a period of informal kinship care. Many grandparents reported that they did not particularly want a formal order granted but, again, they felt bullied by their social worker to have one: We had a lot of pressure for an SGO. I mean, at our household review this year, they told us if we don't do it by next year, they'll remove her. (Rebecca; carer of granddaughter, 8) Many of the grandparents took on the care of their grandchildren in addition to other existing unpaid caring commitments. Some were also looking after their own elderly parents, other grandchildren and, in some cases, their adult child(ren). There was a gendered element to this: many of the grandmothers (but not grandfathers) in the kinship households had given up paid employment to care for their grandchild(ren): To summarise, the kinship care context was highly challenging for grandparents. It was often emotionally painful and fraught with ongoing challenges in terms of juggling caring and other responsibilities, managing ongoing difficulties with the child's biological parent, and managing their grandchild's own complex needs and emotional difficulties. Many grandparents reported feeling coerced by their social worker in terms of the kinship care arrangement, despite their own concerns about it. It is within this context that we need to consider the violence, its impact and grandparents' help-seeking practices.
Understanding the child's use of violence All of the grandchildren in this study displayed physical violence, often daily. Sometimes this was directed at property and objects around the house, but very often it was directed towards grandparents themselves: The minute that it's not her way, that will start violence which will be kicking, smacking in the face, pushing downstairs, you know, a good shoving, punching in the back, throwing things, and quite recently she's taken to biting. ( In many cases, the violence was severe and had caused injury. There appeared to be no link between the age or the gender of the child, and the severity of the violence. In one case, an eight-year-old girl had twice hospitalised her grandmother. The grandparents also described a range of verbally and emotionally abusive behaviours, some of which had a sexual undercurrent: She'll start, you know, 'Shut your f-ing mouth, you c.u.n.t' and it's just unbelievable the way she gets to me and she'll chant it, she'll come and she'll stand right next to me and she'll just chant the 'cunt' word over and over and over and over and over and over until my fingers are in my ears. (Bev;carer of granddaughter,15) In addition, many grandparents experienced financial abuse. Almost all of the grandparents were experiencing financial difficulties, so this was not only emotionally harmful, but had real financial impacts on the grandparents' lives: I never seem to have much money. We've hardly done anything really since we've been here . . . I was pretty certain that she'd taken money out of my wallet-thirty pounds. I looked in her room and I found thirty pounds in a box. So I took it back [. . .] and then she discovered it wasn't there, and all hell broke loose. Physical confrontation and all that, and she managed to get my wallet off me and take this bloody thirty pounds back. (David; carer of granddaughter, 14) Several of the grandparents described how their grandchildren targeted emotional and physical vulnerabilities. Treasured family photographs or Violence towards grandparent kinship carers Page 9 of 18 special possessions would be purposely damaged or destroyed, and physical vulnerabilities would be also targeted, as one grandmother explained I suffer from a bit of arthritis in my knees and she knows it and so she kicks into them and that's where she targets. (Wendy; carer of granddaughter, 16) The kinship care context profoundly shaped the grandparents' interpretations of the violence. It was indeed understood as 'violence' (rather than as 'challenging behaviour'). It was also understood in terms of the child's past loss and trauma, and grandparents often described specific early experiences in their grandchild's life that they felt was at the root of the violence but that was difficult to discuss with their grandchild: He's angry all the time. He doesn't know why he's angry, but I knowway back he was in a womb with all these drugs, and it's affected his whole life. But I don't want to tell him that, I cannot tell him that. How would you feel if somebody told you you were born addicted? I cannot say that to him. (Nancy; carer of grandson, 20) The context of early trauma meant that grandparents rarely blamed their grandchild for the violence, and instead sympathised with them. This presented challenges about how to resolve the violence. For example, as their carer, many grandparents explained that they were the recipient of the violence because they were the only 'safe', unconditionally loving person that their grandchild had in their lives. Other grandparents explained that their grandchild's violence was a way of communicating the pain 'that she [her grand-daughter] desperately wants you to understand'.
Other grandparents understood the violence in terms of the hurt felt by the child as a result of the kinship care situation and the sense of loss it provoked in the child: She blamed me for taking her away from her mum. She blamed me for taking her away from her sister. And at that point I was being physically assaulted every single day. (Diane; carer of granddaughter, 9) It can be common for victims of family violence to sympathise with those who instigate it. However, the kinship care context, and the unique awareness of the child's biography that this enables, makes those sympathies particularly acute, and this appears to shape both the impacts of the violence and the help that is sought.

The impact of the violence
The violence created significant health impacts. Many grandparents spoke of depression, anxiety and suicidal thoughts. The violence also caused physical injuries which often took a long time to heal, with some grandparents requiring regular medical treatment as a result.
However, there were also wider impacts that were shaped by the kinship care context. Living with daily violence profoundly affected the grandparents' relationships with their families, who often blamed the grandparents for the violence because they had 'taken on' the responsibility of caring for their grandchild. Family members, such as the grandparents' other adult children, were often (though not always) unsupportive and their advice for managing the violence was often to either reciprocate with violence or to 'return the child to social services'. Although the decision to care for their grandchild was rarely experienced by grandparents as voluntary (see earlier), it was framed by family members as though it was, and that therefore their victimisation was their choice.
In kinship care households with two grandparents, often it was the grandmother who was both the target of the violence, and the one responsible for de-escalating it. For several of the grandparents, the strain of caring for a grandchild who is violent impacted on their marriage/ partnership: there was often disagreement on how best to manage the violence, with grandmothers often reporting that their husbands/partners wanted to take a stricter approach to discipline (this was also reported by two of the three grandfathers). Many of the partnered grandmothers also reported that their husbands/partners often dealt with the situation by removing themselves, either physically (in three cases, the violence resulted in marital separation) or emotionally, which exacerbated grandmothers' feelings of isolation: I couldn't even talk to my husband because he couldn't deal with it [. . .] I know his response back then would have been, "Well let's just get rid, get social services to come and pick her up". (Diane; carer of granddaughter, 9) The child's biological parent(s) were often in and out of the child's life and the time around contact was often identified by grandparents as a time when the violence escalated. Many grandparents reported that it was difficult to navigate contact between their child and grandchild. This was particularly the case in situations when they felt that contact would not be in the interests of the child (e.g. when the parent 'turned up drunk') and they had to manage their child's consequent upset about not seeing their parent, for which the grandparent was often blamed.
Often the biological parent(s) were not particularly supportive of the grandparent. Many grandparents understood this to be due to resentment that their child was living with them: My daughter was a drug addict [. . .] I always said to her: 'If you can't, I will' and I think she's taken umbrage to that statement. But I don't Violence towards grandparent kinship carers Page 11 of 18 think it's a bad statement to make-I think it's an honest statement. And her children didn't go into care '. (Martha;carer of grandson,18) These strained relationships with their husbands/partners, children and other family members resulted in isolation for the grandparent who was doing the majority of the caring and some grandparents also became estranged from their wider families: My parents are alive, but they walked out of our life about three years ago, because they cannot understand my granddaughter. They cannot understand her behaviour. And as far as they are concerned, she needs a good hiding. So I've walked away from them. (Rebecca; carer of granddaughter, 8) This isolation was compounded by the reactions of neighbours and others in the grandparents' local communities, who were often wary of the child and, as a result, avoided the grandparents. This social isolation was often entrenched by financial difficulties caused by taking on such caring responsibilities. For example, many of those grandparents who were still in work were frequently called away to deal with problems at school, which often resulted in further loss of income. The impact of the violence was overwhelming-it affected all aspects of the grandparents' lives and the intersecting emotional, familial and economic difficulties related to the kinship care context only exacerbated its impact.
Uncertainty, help-seeking and the double-bind: engagement with social workers For many grandparents, their kinship care situations were very uncertain: many did not know how long their grandchild(ren) would be living with them, whether they would return to their biological parent(s) or whether they would eventually be placed elsewhere. Furthermore, whilst formal court orders (such as SGOs) are considered to be preferable in cases of long-term kinship care because they establish permanence to the kinship care role, this perception is not necessarily shared by grandparents themselves: We feel like we've just gone on this rollercoaster without ever having a chance to get off. Once the children came, you just thought: 'Well, once [my daughter] has sobered up, she will come and get them' but she didn't [. . .] They fast-tracked it to court to get the Interim Care Order [. . .] And then they decided, when it did go to a Care Order, that the children were returning [. . .] but then we found out that [my daughter] was using heroin and we decided that that was it, there was probably no escape. And that's when we decided to go for an SGO. So then it became more permanent. But even then, until [my daughter] actually died, we always felt that we were looking after them for her, and that there always would be a time when they would be back, and they would be 'mum and children ' again. (Wendy;carer of granddaughter,16) This context of uncertainty presents additional challenges for grandparent kinship carers seeking help for the violence because they often understand the situation (and therefore the violence) to only be temporary. The uncertainty also meant that the grandparents felt particularly vulnerable to their grandchild being removed from them if they did seek help for the violence: I was scared that they would take him off of me as well and put him into care because I was only his grandparent-I wasn't his guardian or anything like that. I had no rights, so I was scared that if he carried on the way he was, he would be put in care. (Connie; carer of grandson, 13) Nevertheless, many grandparents did seek help and, because social workers were often already a part of the grandparents' lives, they were almost always the first person that grandparents approached. All of the grandparents reported that they had continually asked social workers for all kinds of help to support them in managing the violence and its impacts: they asked for financial help, for emotional help, for therapeutic help and for physical help (in terms of respite from the violence). Grandparents reported that they did not receive a positive response and, out of all the services where they requested help (including schools, the police, CAMHS and youth offending services), they reported that they had the most difficulty in getting support from children's social care services.
As discussed, the difficulties of engaging with social workers began as soon as their grandchild was placed with them in a process that was experienced as coercive. As the kinship care placement continued, grandparents were faced with the continual challenge of requesting help and information whilst also reassuring the social work team that their child would be safe with them: violence and reassuring their social workers that they can manage. The need to reassure social workers was very strong because, in the context of a placement that was experienced as both coercive and uncertain, the removal of their grandchild felt like a very real risk. The dilemma never resolved itself.
As discussed in Introduction, social work framings of child and adolescent violence in the home means that it is rarely understood as a problem of family violence that involves a parent/carer victim that needs support. Indeed, one social worker we interviewed explained that responding to violence towards carers is difficult because '[social workers'] orientation is always towards: is the child being harmed?' This orientation seemed evident in the grandparents' reported experiences of help-seeking: They used to send people down to the house four times a week, the social worker team were trying to keep the family together. But basically, if everything suited the kids, they weren't really bothered about the adults. If there was ever any aggression showed, it was 'Just walk away, walk out the house', things like that. They basically just wanted you to go out and leave the kids alone to do what they wanted. (Ian; carer of granddaughter, 15) [The social worker] knew what was going on, how I was being assaulted and everything. And she basically told me that I had to just stand there and let the child assault me and I couldn't do anything about it. Because if I did anything back to her, the child would be removed from our care. (Erica; carer of granddaughter, 8) With social services, it was all softly softly, let's not upset the child. I can understand that to a certain extent, but also understand that we are the ones that are the victims and I actually got quite cross with our social services. [I was] saying to them that we were the victims, we weren't hurting the child. (Laura; carer of granddaughter, 12) From the grandparents' perspectives, this response from social workers appeared both insensitive and inappropriate, and did nothing to either prevent the violence or to support the grandparents. The threat that the child would be removed from their care if they could not manage the violence themselves left the grandparents in an impossible 'doublebind': either 'cope' and live with the violence, or 'don't cope' and lose a much-loved member of the family. As one grandmother explained, 'they didn't ask to be born, their mum messed up and Gran can't afford to'. The uncertainty that continued to frame their kinship care role, coupled with a history of feeling coerced by their social worker in terms of this role, and knowledge of the child's painful history, only intensified the double-bind.

Discussion: implications for social work practice
This study provides original and important insights into the lives of grandparent kinship carers who are experiencing violence from their grandchild and highlights how profoundly the context of kinship care shapes the violence, its impact and help-seeking practices with social workers. The findings point to a number of areas where social work practice must improve to better support kinship care families where there is child and adolescent violence in the home.
First, it is vital that social workers understand how the kinship care context presents additional challenges to help-seeking. Kinship care households are often fraught with emotional, familial, social and financial difficulties which can exacerbate the impact of living with child and adolescent violence in the home and complicate grandparents' understandings of its causes. The perceived uncertainty of the placement (even following a court order) and the knowledge that they are the child's 'last resort' (which is often reiterated to them by social workers) contributes to the grandparents' need to reassure social workers that they are 'fine' and that their grandchild is safe with them. Social workers must understand this context and recognise their own role in making it feel 'unsafe' for grandparents to reach out for help. Social workers must learn to ask the right questions in the right way and understand why it is so difficult for grandparents to share difficult answers. This needs to begin prior to placement and should involve assessing potential risks to the kinship carer as well as to the child. Furthermore, this risk assessment must continue throughout the placement, including after a court order is issued.
Second, as the findings illustrate, kinship carers are uniquely bound to the traumatic context of the child's placement. This may produce 'secondary trauma' for kinship carers as they continue to deal with their own family tragedies of domestic abuse, mental health problems, substance misuse, incarceration and death. Yet kinship carers, who are perhaps more vulnerable in many ways than others who provide out-of-home care (such as adoptive parents or foster carers), receive very little support to help them parent a highly traumatised child. Furthermore, they receive little to help them process their own intersecting traumas that relate to both the past (i.e. the circumstances that led to their grandchild being placed with them) and to the present (i.e. being the victim of family violence). A trauma-informed approach to social work engagement with kinship carers is essential and some of the important trauma-informed practice that has started to emerge in the field of child/adolescent family violence (e.g. Evans, 2016) would be a good place to start.
Third, social workers need to be aware of the language they are using when talking to grandparent kinship carers about the options available to them, and of the threats that they may be making, however unintentionally. In such discussions, social workers should consider how the 'grandparent' component of the 'kinship carer' may feel about any perceived threat to 'remove' a loved member of their family, particularly in the context of ongoing trauma and loss. Of course, as one social worker pointed out during an interview, the practice of delivering such ultimatums to kinship carers may be driven by court practices. Nevertheless, social workers might consider if, and how, they can challenge such wider institutional practices that serve to re-traumatise individual families.
Fourth, social workers must ensure that kinship carers are fully informed about the provision that currently exists for them. In cases where kinship care is formally arranged, carers are legally entitled to post-order support until the child reaches 18 years. Furthermore, the Adoption Support Fund (ASF) now provides funding for therapeutic services for children subject to an SGO. However, despite grandparents continually emphasising that their grandchild was in need of therapeutic intervention, only one of the grandparents we interviewed were aware of this fund (see also Mervyn-Smith (2018) who found that only 16 per cent of the kinship carers surveyed were aware of the ASF). Furthermore, no such funding is available in cases where kinship care is organised through private arrangements, and there is an urgent need for policy change to ensure that all kinship carers have access to the same levels of support.
Finally, the grandparents' reported need for respite has been identified elsewhere as a primary support need (Robson and Conqueror, 2011), and this is doubly so for victims of family violence. Yet it was rarely available for the grandparents. That kinship care is the 'Cinderella' of the care system is not only something discussed in academic and practice circles. The grandparents in this study were also painfully aware of the support disparities, financial and otherwise, offered to them compared with the support that foster carers and adoptive parents receive. Because they are the child's grandparents, they are erroneously assumed to not need such support. Furthermore, caring for a traumatised child, particularly one with complex needs, is not necessarily something that grandparents have experience in, and they may not necessarily be well-placed to take this on when also processing their own trauma. Again, the erroneous assumption is that, because 'its family', then it is not needed (despite some of the grandparents not even particularly knowing the child before becoming their full-time carer).
The findings from this research are disturbing and its implications go far beyond social work practice. An urgent change in response is needed from a whole range of agencies whose work intersects with kinship care families-including the police, health care (particularly CAMHS), schools, youth offending services and victim support organisations (see Holt and Birchall, 2020). However, social workers were already part of the grandparents' lives, had an understanding of their traumatic familial context and understood the nature of the kinship care placement. Social workers also play an important role in double-binding grandparents to a situation which is difficult for the grandparents to resolve. For those reasons, it is clear that a change in social work understanding and practice is fundamental to changing how 'family violence' in kinship care families is addressed.

Funding
This work was funded by the British Academy/Leverhulme Trust [Grant No. SRG\171528].