Domestic violence and abuse: multi-agency working

The PDG agreed that domestic violence and abuse occurs in all communities.

The PDG was clear that both women and men can experience domestic violence in heterosexual and same sex relationships. The likelihood of ever experiencing a physical assault from a partner or adult family member is higher among heterosexual women than men. Moreover, heterosexual women experience more repeated physical violence, more severe violence, much more sexual violence, more coercive control, more injuries and more fear of their partner than heterosexual men.

Although domestic violence and abuse research and services mainly focus on intimate partners, this type of violence and abuse takes many forms. Examples include: forced marriage, violence connected to 'honour', violence against adults by their children, abuse of older people and other intra-familial abuse. However, evidence of effective interventions in these areas is lacking.

The PDG recognised the important role that the experiences, views and preferences of those who have experienced domestic violence should have in the development of policy and services. However, it did not hear evidence from them directly; this was largely outside the scope of the evidence reviews due to a focus on evaluation of interventions and the quality appraisal system used. However, the PDG did receive very helpful expert evidence and reports from people working in the specialist domestic violence sector and directly with service users. Such organisations were also represented by PDG members.

The PDG agreed that, rather than use the terms 'victim' or 'survivor', the Group would refer to 'people who have experienced domestic violence and abuse'.

The PDG thought it likely that domestic violence and abuse services could also benefit the extended family and friends of people who directly experience domestic violence and abuse. However, these effects have not been studied.

The PDG was aware that much of the expertise and support for people who experience domestic violence and abuse lies in the voluntary and community sector, where funding and capacity is generally limited.

The PDG was aware that domestic violence and abuse is often one of several problems that a couple or family may face. For example, it may be combined with poverty, drug and alcohol misuse or mental health problems. Most of the evidence relates to male violence against women and children in heterosexual relationships. However, the PDG noted that domestic violence and abuse affects: bisexual, gay, lesbian and trans relationships, and reconstituted (or step) and more complex families. In such cases, the Group noted that people may face particular barriers to accessing support and may have specific needs.

The PDG recognised the wide range of ill-effects that exposure to domestic violence and abuse can have on children and young people, including the effect on their social, emotional, psychological and educational wellbeing and development. It also recognised that providing effective interventions and support may reduce the likelihood of them being affected by, or perpetrating, domestic violence and abuse in adulthood.

The PDG noted the importance of working concurrently with both the non-abusive parent or carer and child, rather than just focusing on the parent. Research on the effectiveness of parent/carer-child interventions has focused exclusively on mothers. Given the profile of domestic violence and abuse, that is where the biggest need for services is likely to be, but provision is needed for all families. The PDG agreed that evaluation of programmes where the father is the non-abusive carer will be especially important in light of the current lack of evidence about effective interventions.

The PDG noted that domestic violence and abuse – and children's exposure to it – often continues beyond the end of the adults' relationship.

The evidence linking effectiveness to the length of interventions for children is unclear. But it appears that longer interventions are more effective. This may be particularly true in complex cases.

The PDG did not consider evidence on the timeliness of interventions for children, but the Group was aware of a developing body of literature in this area.

The PDG noted the importance of ensuring services are appropriate to the age, gender and developmental stage of the child or young person. For example, teenagers may not want to be seen at the same time as their non-abusive parent or carer.

The PDG was aware that there is an ongoing debate about the effectiveness and desirability of screening, routine and targeted enquiries to identify people who are experiencing domestic violence and abuse. Currently there is insufficient evidence to recommend screening or routine enquiry in healthcare settings. Nevertheless, the PDG recognised that asking patients routinely about abuse in some specialised health care settings was considered good practice by professionals in those fields. The PDG acknowledged that people experiencing domestic violence and abuse may choose not to disclose it when asked by a healthcare or other professional. Or, if they do disclose, they do not want to be pressurised to give more details of the abuse or take a specific course of action (Feder 2006).

The PDG noted that healthcare professionals not trained to identify domestic violence and abuse may mislabel and misdiagnose people's problems, leading to inappropriate plans or ineffective remedies. (For example, specialists may be ordering unnecessary and expensive investigations and GPs may be prescribing inappropriate anxiolytics and antidepressants.)

There is no universally accepted understanding of what 'advocacy' means in the context of domestic violence and abuse. The PDG kept the term because it has been applied to a range of interventions that have been evaluated in research studies. A definition of advocacy was agreed for the purposes of this guidance.

The PDG noted that skill-building approaches might be of particular use in refuge settings, although they are also an intrinsic part of the advocacy and support role.

There is a lack of consistent evidence on the effectiveness of programmes for people who perpetrate domestic violence and abuse. The PDG noted that some evaluations take account of the partner's health and wellbeing and include their perception of any changes in the perpetrator's behaviour. However, these tend to be small-scale, uncontrolled studies.

The cost effectiveness analyses concluded that interventions with people who have experienced domestic violence and abuse are likely to be cost effective. However, this conclusion could not be extrapolated to interventions with perpetrators and the PDG was split on whether interventions with perpetrators should be recommended. However, members agreed that such interventions are an important part of domestic violence and abuse services and, provided they are supported by robust evaluation to inform future commissioning decisions, should be recommended.

The PDG noted that national programmes dealing with behaviour-change among perpetrators are aimed at heterosexuals. Members were unclear whether or not these programmes would also be effective for other groups.

Members of the PDG (and stakeholders) were disappointed that the review did not find sufficient evidence to make recommendations on primary prevention programmes. This was partly because it looked only at health and social care – and currently most primary prevention interventions are delivered in education settings. However, the PDG agreed that prevention is an important area for future research (see Recommendations for research).

The PDG discussed the relationship between training to support people affected by domestic violence and abuse and child safeguarding training. Overall, members agreed that there were obvious links between them. However, they did not necessarily think they should be combined. Members recommended that this question should be addressed in future research (see Recommendations for research).

The economic modelling showed that effectiveness and cost-effectiveness in the medium to long term was less certain than in the shorter term. This was partly due to the short follow-up period applied to the studies used as the basis of the model. It was also due to the lack of longitudinal studies. However, even using conservative assumptions, it seems likely that the interventions will be cost-effective in the long term by stopping the violence and improving the mental health of all those involved.

The PDG was aware that lack of evidence about the medium- and long-term consequence of interventions meant the economic models would underestimate their cost effectiveness. For example, a reduction in the incidence of post-trauma-related stress disorder is likely to lead to additional benefits, such as being less depressed or having improved self-esteem. However, limited data on these benefits meant they could not be estimated in the model.

Although the systematic review of cost-effectiveness studies only found one analysis (Devine 2012) and the economic modelling focused on 2 interventions, the findings are also relevant for interventions with similar benefits and similar (or lower) costs. The PDG noted that the potential health and non-health benefits of these interventions would outweigh the costs when the positive impacts on people experiencing the violence and abuse, their families and wider society were considered.