Appendix 4: Literature Review
1. Development of advocacy as a response to gender-based violence in the UK
Advocacy, as an intervention with victims  of gender-based violence ( GBV), has developed over the past 30 years.
Until the late 1990s, specialist women's support services such as Women's Aid and Rape Crisis were the main providers of advocacy. Their emphasis was on non-directive support and empowerment for women victims of domestic abuse and/or sexual violence, including support to report to the police and/or attend court.
Kelly and Humphreys (2000), highlighted the growing recognition of the need for more integration in response to domestic abuse, operationally and strategically  . The projects reviewed included one in which a team of 'civilian' support workers based in a police station used advocacy to follow up domestic abuse incidents reported to the police  . This proactive approach, and accepting third party referrals, was markedly different from the approach taken by most specialist women's support services up to then. The study notes that these new projects were working differently with victims, taking a more proactive approach and recognising:
'…individuals coming from positions of fear and isolation will often require the skills of an advocate to negotiate housing, legal support and benefit entitlements. It is the emphasis on rights and entitlements which distinguishes advocacy from other more familiar concepts like support.' (Kelly and Humphreys, 2000).
A shift in emphasis towards a more risk-focused approach is evident from 2001, with the establishment of the Women's Safety Unit ( WSU) in Cardiff, which provided a 'one stop shop' for victims of domestic violence  and 'known-perpetrator rape'. Goals of the project included increasing the proportion of cases resulting in arrest, charges and convictions, and reducing the level of repeat victimisation (Robinson, 2003).
Towards the end of her report, Robinson notes that a new initiative  , Multi Agency Risk Assessment Conferences ( MARAC) provides further illustration of the value of the work done by the unit. In particular, the role of the seconded police officer was seen as 'a bridge between community and criminal justice agencies and their respective approaches to handling cases of domestic violence.' (Robinson, 2003, p.36). The WSU-based officer had access to confidential information, not usually shared with community agencies, and could track cases through the criminal justice process. The officer also had 'in-depth information gathered by victim-oriented trained professionals that is not normally within the purview of criminal justice agencies' (ibid).
When shared at the MARAC, this range of information 'was instrumental in creating safety plans for victims' (ibid). Robinson's subsequent evaluation of the Cardiff MARAC (Robinson, 2004) was positive about the impact of the MARAC approach, and noted that 'the information provided … and actions undertaken by the Women's Safety Unit appear particularly significant.'
Funding from the UK Home Office in 2005 resulted in more domestic abuse advocacy projects in England and Wales, along with additional funding to develop specialist 'domestic violence' courts. The Home Office also funded Caada (Co-ordinated Action Against Domestic Abuse, now SafeLives), to provide accredited professional training for independent domestic violence advisers ( IDVAs) and independent sexual violence advisers ( ISVAs). These terms are now common in the rest of the UK  .
Since then, domestic abuse advocacy services have predominantly developed in tandem with the emergence of MARACs and other variants on what is known as the coordinated community response ( CCR) to domestic abuse.
The first domestic abuse advocacy project in Scotland, ASSIST, was set up in 2004 to support the pilot domestic abuse court in Glasgow. The evaluation of this court acknowledged the value of the advocacy service, and a subsequent feasibility study recommended that the court, including the advocacy service, should develop across Glasgow (Reid-Howie, 2007; Scottish Executive, 2008).
MARACs in Scotland were piloted in Glasgow and North Lanarkshire from 2005. In the MARAC context, advocacy services have subsequently developed in other areas of Scotland. In the absence of agreed standards or service specifications, this has happened in an ad hoc way. In some areas, advocacy services are linked to specialist courts. In others, they are linked to MARACs. A few have children's advocacy workers.
Sexual violence advocacy emerged alongside the development of multi-agency approaches to sexual violence. ISVAs were introduced in several areas in England and Wales in 2006 following research into Sexual Assault Referral Centres ( SARCs), which identified that victims of sexual assault needed support, advocacy and information in the period immediately after a sexual assault (Lovett et al, 2004). Rape Crisis Scotland has rolled out a National Advocacy Project following positive feedback from survivors who participated in a pilot project in Glasgow which offered advocacy and support to women reporting a sexual assault to the police (Brooks and Burman, 2016).
2. Training and accreditation
Caada (now SafeLives) was set up explicitly to encourage the use of independent advocacy to increase victims' safety. This depended on trained advocates. Caada has provided this training since 2005.
Training focuses on assessing and reducing risk to victims; providing consistent professional support; and liaising with other agencies that contribute to victim safety. The training model was originally based on best practice developed by Standing Together Against Domestic Violence, ADVANCE and the Cardiff WSU (Kail et al, 2007).
In Scotland, the feasibility study for the Glasgow domestic abuse court recommended that the Equality Unit/Scottish Government should consider whether the Caada standards and accredited training might be adapted for use in Scotland (Scottish Executive, 2008).
Subsequently, ASSIST and Scottish Women's Aid worked with Caada/SafeLives  to develop a qualification for IDAAs accredited by the Scottish Qualifications Authority. Based on the Caada training but tailored to the Scottish context, the Scottish Government funded the three organisations to develop and provide the training from 2011 to 2016.
3. Defining advocacy
The literature describes the characteristics of the advocacy services being offered but does not define it.
4. Evaluating advocacy
This review has primarily considered multi-site rather than individual-service evaluations. Multi-site evaluations offer a significant degree of comparison and commentary on the value of advocacy services and the challenges.
Most of the sites evaluated provided services predominantly to female victims. Howarth et al (2009) identified a small number of male victims (44) in the dataset (2,567) but noted:
'While it is increasingly recognised that both homosexual and heterosexual males can suffer abuse and that abuse can be inflicted by another family member, less is known about both of these areas. These cases may be marked by a different pattern of risk and it is feasible that different intervention strategies are required to address these issues. For this reason, and in recognition that there is a marked asymmetry in the extent to which males and females experience severe levels of abuse, it was decided to exclude these cases from the study sample.' (Howarth et al, 2009).
An evaluation of four services in London included three services working with male victims. As with the Howarth et al (2009) study, numbers were relatively small - 14 male victims, and two transgender victims. 732 (97.9%) of the victims were female, and study focuses on their experiences (Coy and Kelly, 2011).
The preamble to the Safety in Numbers report (Howarth et al, 2009) notes that studies of advocacy in the UK up to then 'mostly represent in-depth and rigorous evaluation of individual services'. The authors observe that 'Single site evaluations will naturally be influenced by local operating conditions and by the individuals involved, which may potentially limit the extent to which the conclusions derived from these studies are applicable to IDVA services more widely.'
Evaluation generally has focused on domestic abuse advocacy; there is little evaluation of sexual violence advocacy. 
This review considers three multi-site evaluations of IDVA services and one multi-site evaluation of ISVA services.
Domestic abuse advocacy
Robinson (2009a) reviewed four IDVA services, working with MARACs and/or specialist courts and based in specialist domestic violence services. The work was at times located in other settings, for example police station and court. Co-location was considered beneficial and improved partnership working. However, study participants said this had to be balanced alongside the role of IDVAs in providing independent advice and support.
Practitioners and victims valued the role of IDVAs in providing a 'seamless response' by agencies engaged in a coordinated community response (Robinson 2009a).
Safety in Numbers reviewed seven IDVA services in England and Wales. The services were based in urban, suburban and rural locations and ranged in size 'from one full time IDVA as part of a wider community based domestic abuse service, up to 12 IDVAs.' The study included new and long-established services. Some worked in communities with high BME populations and others in areas where these groups were under-represented (Howarth et al, 2009).
The study confirmed that, in this sample, IDVAs were working with complex high-risk cases. Outcomes were impressive; more than half of victims during the period of the study experienced a cessation in the abuse and around three quarters reported 'improved feelings of safety'. Victims were 'much safer' when they received intensive support and when multiple services were offered. The study highlights the role of the IDVA in co-ordinating services and improving the multi-agency response (Howarth et al, 2009).
Coy and Kelly (2011) evaluated four IDVA services in London. Reporting on service user perceptions, they note that those who participated in the evaluation (10% of those who used the service during the evaluation period) reported feeling and being safer. Two thirds reported that no further violence had occurred since contact with the advocacy service. Service users particularly valued the proactive approach, being enabled to recognise and name the violence, listening, safety planning, being given information about rights and options and the liaison with other agencies.
The study notes that 'internationally acknowledged integrated CCRs tend to be in small cities, with shared agency boundaries, low staff turnover and key players in post for extended periods'. The size of London brings some additional challenges when developing a coordinated community response, and creates some challenges for IDVAs as a result (Coy and Kelly, 2011).
Sexual violence advocacy
Along with her review of IDVA services (Robinson, 2009a), Robinson conducted a parallel review of six ISVA services (Robinson, 2009b); both studies commissioned by the Home Office.
Robinson (2009b) notes that the ISVA role includes providing non-therapeutic support to victims at the point of crisis, practical help and advice, information and assistance throughout the criminal justice process, and liaising with partner agencies in a multi-agency context.
Risk assessment was not defined as part of the role, though it may be required at times, depending on the context of the assault and the relationship between the victim and the perpetrator.
Victims valued the services, and appreciated having one key worker as their point of contact, including the liaison with other agencies.
The Rape Crisis Scotland pilot 'Support to Report' project ran in Glasgow in 2013. The evaluation (Brooks et al, 2015) found that, despite a lower than expected take-up, there was strong support for the service, with survivors of sexual violence needing support at the reporting stage and also during the later stages of the criminal justice process. The service was highly valued by those who used it. The evaluation highlighted some difficulties, for example with the service configuration and the realities of partnership working, and made various recommendations which have informed the design of a new national project which has been offered since December 2015. The new project is currently being evaluated.
Advocacy for other forms of violence against women and girls
There is a lack of research into advocacy for other forms of gender-based violence. This may reflect a lack of such advocacy services.
In 2010 the Ministry of Justice funded 11 pilot sites to employ an IDVA to support people seeking/or subject to a Forced Marriage Protection Order ( FMPO). The main aim of the pilot was to find out whether the designation 'Relevant Third Party' should be extended to include IDVAs. Of 158 case summaries from the pilot sites, 151 clients engaged beyond simple information provision. Only five applications for FMPOs were made, and these were made by solicitors, with support from the IDVA. While the advocacy service provided by the IDVAs was seen as valuable in itself, the conclusion of the evaluation was that there was insufficient evidence to support an extension to the RTP provision (Ministry of Justice, 2010).
Advocacy for children and young people
The impact of domestic abuse on children and young people is now well documented and all of the evaluations reviewed identified the risks to children living with domestic abuse.
Howarth et al (2009) reported that 69% of the victims in their study had children, mostly primary school age or younger.
None of the services included in the evaluations provided specific advocacy for children and young people affected by GBV. The lack of specific services for children and young people is consistently identified as a problem for services and victims (Howarth et al, 2009; Robinson, 2009a; Coy and Kelly, 2011).
The recommendations from Howarth et al (2009) note: 'It is not the role of the IDVA to work directly with children…However, the impact of the work of the IDVA in helping end the abuse that victims are suffering has clear implications for the safety of children also. Work needs to happen without delay to examine how links can be made between those whose work it is to safeguard children and those who are working with this high risk group of victims.' (Howarth et al, 2009, p17).
Another study goes further. Commenting on the needs of children and young people affected by domestic abuse, the authors assert that 'domestic violence risk assessment needs to address both adult and child victims…risk assessment and management should be done with children and victims, not to them' (Stanley and Humphreys, 2014).
Advocacy for women from marginalised groups
The lack of specific services for black and minority ethnic victims is also consistently identified as a problem (Howarth et al, 2009; Robinson, 2009a; Coy and Kelly, 2011).
Howarth et al (2009) note that nearly a quarter (23%) of victims in their sample were from BME communities, compared with 11.5% as the proportion of BME people nationally and higher than the proportion of BME women in the areas studied (14.4%). Noting that BME victims face significant barriers when trying to access services, they suggest that 'it should be viewed as a positive finding that the proportion of B&ME victims accessing services was higher than expected, indicative of IDVA projects being accessible to local minority communities.'
A 2012 report identifies that few studies map the experiences or support needs of black, minority ethnic or refugee ( BMER) women across the spectrum of violence against women and girls. A short-life study involving ten organisations across the UK (including one specialist refuge provider in Scotland) found that BMER women were most likely to talk to friends (54%) and family members (45%) about abuse. Only 15% of the women had approached agencies such as the police, health visitors, teachers/children's school, children's centres and women's organisations (Thiara and Roy, 2012).
A later report on service responses to BME women experiencing sexual violence concluded that all organisations should 'scrutinise their assumptions and practice to counteract a 'one-size-fits-all' approach to service delivery […] This includes the homogenous understanding of BME women and girls' needs, frequently linked to narratives of poverty and immigration, and which can disguise complexity of need and experiences in other areas' (Thiara et al, 2015).
One of the services reviewed by Coy and Kelly had three specialist community IDVAs, working with women from BMER communities. They note that 'research demonstrates that BME support services ensure that women's additional and specific needs are addressed (Gill and Rehman, 2004), particularly 'intense advocacy' (Thiara et al, 2015, p.7)' (Coy and Kelly, 2011, p.19).
Specific services for women with disabilities are also scarce. Hague et al (2011) in the first national study of domestic abuse and disability in the UK found that service provision for disabled women was proportionately less than for non-disabled women.
A 2014 review of the role of social care agencies in MARACs raised questions about the issue of 'capacity' and the response of MARACs (and by extension advocacy workers) to victims with additional needs or vulnerabilities:
'Assessing capacity can be particularly challenging in domestic abuse situations, where the person is cared for by, or lives with a family member or intimate partner and is seen to be making decisions which place them in danger.' (Robbins et al, 2014).
Impact of advocacy
Evaluations of domestic abuse advocacy services consistently identify that advocacy is an early intervention offered in a proactive way, and that a key goal of advocacy is to encourage victim engagement with the criminal justice system. As a result, advocacy:
- Increases engagement with the criminal justice process (Reid-Howie, 2007; Coy and Kelly, 2011)
- Enhances the safety of victims/survivors, and contributes to reducing or ending abuse (Howarth et al, 2009; Coy and Kelly, 2011)
- Is integral to a coordinated community response to domestic abuse and sexual violence (Howarth et al, 2009; Robinson, 2009a; Robinson, 2009b; Coy and Kelly, 2011; Brooks et al, 2015)
- Improves victims' health, wellbeing and quality of life (Reid-Howie, 2007; Coy and Kelly, 2011; Safe Lives, 2016b)
5. Main components of advocacy services
From the evaluations, whether about domestic abuse or sexual violence advocacy, common themes and principles emerge. These may not hold true in all cases. Advocacy is a developing field and more work is needed in order to understand the needs of marginalised groups and ensure that advocacy services are accessible to all who need them.
A crucial component of an advocacy approach is independence - the 'I' in ' IDVA' (Robinson, 2009a). Advocates in the services reviewed were based in various locations, including local authority hubs, police stations, A&E departments and voluntary organisations. However, regardless of physical location, all the evaluations concur that advocates must be independent of 'the system' in order to represent the best interests of victims. Their independence is critical to the success of the advocacy role and the extent to which victims and practitioners can trust them trust (Howarth et al, 2009; Robinson, 2009a; Robinson, 2009b; Coy and Kelly, 2011).
Advocacy workers respond to third party referrals, offering the service to victims rather than waiting for the victim to self-refer. This includes offering the service repeatedly if it is declined or the victim does not respond at the first approach (Kelly and Humphreys, 2000; Howarth et al, 2009; Robinson, 2009a; Coy and Kelly, 2011).
Advocacy is a short-term crisis intervention, designed to address the immediate risk to victims; reduce the risk of further abuse; help victims get other services; and promote access to justice and rights. The trigger for an advocacy intervention is usually a specific abuse-related incident. The advocacy may last a few days, or a few weeks, but rarely longer than three to six months (Howarth et al, 2009; Robinson, 2009a; Coy and Kelly, 2011; SafeLives, 2016b).
The police or a health care provider often make an initial assessment of risk before referring a victim to an advocacy service. The advocate's role is to develop the risk assessment further; to find out more about level of risk; and to take action to reduce the risk to the victim and enhance her safety. Risk assessment is dynamic. The advocate continues to assess risk throughout contact with the victim.
Most advocacy services use a standard risk assessment checklist. Advocates must also use their own professional judgment in assessing risk. Robinson (2007) notes that risk assessment relies on:
'The good judgement and experience of trained advocates rather than a simple matrix that can be completed by anyone with access to victims of domestic abuse. The 'science' of risk assessment is still in its infancy, and complex lives and dangerous situations cannot simply be reduced to a tick box form. It is important that a sophisticated understanding of domestic abuse and knowledge of risk is combined with an environment (both physical and human) that is supportive of victims, and helps them to feel comfortable disclosing features of their personal lives, in order to produce a process of risk assessment and classification that can help to identify those victims who are most vulnerable and at risk of further harm.' (p.4)
A consistent approach to risk assessment across agencies is also required. A recent baseline report of MARACs in Scotland highlights complications in assessing risk because different agencies use different risk assessment methods (SafeLives, 2016a).
Safety planning is dynamic and constantly updated to take account of the changing circumstances of the victim and the perpetrator. It is practical and tailored to the circumstances of each individual. It is a process done 'with' not 'for' the victim (Campbell, 2004).
Advocates provide information to victims. This includes information about process - 'what's happening now/next?'; about rights and entitlements to criminal justice/housing/welfare; and about the dynamics of gender-based violence. By sharing information about how perpetrators of abuse tend to operate, advocates can help victims understand more about patterns of abuse, perpetrator behaviour and how abuse affects the victims and children involved. This can help increase victims' understanding of 'coercive control' in intimate partner relationships and know more about the risks from perpetrators (Coy and Kelly, 2011).
Speaking with and for victims
Advocates act on behalf of victims at a time when they may be unable to do so themselves. A critical role for advocates is to keep the victim central to the process, representing their views at multi-agency discussions at which the victim is not present (Howarth et al, 2009; Robinson, 2009a; Robinson, 2009b; Coy and Kelly, 2011).
Multi-agency partnership/coordinated community response ( CCR)
Advocates liaise with colleagues in agencies involved in the multi-agency response to victims of GBV. They are integral to the coordinated community response.
They become the point of contact for the victim and the 'one stop shop' for information and updates about what other agencies are doing. They are also the point of contact for the agencies involved with individual women (Howarth et al, 2009; Robinson, 2009a; Robinson, 2009b; Coy and Kelly, 2011).
Slow responses from other agencies create barriers to effective advocacy work. An advocacy worker cannot do their job if the agencies around them are not responding effectively. The role of the advocacy worker in encouraging an effective multi-agency response is critical (Howarth et al, 2009; Robinson, 2009a; Robinson, 2009b; Coy and Kelly, 2011).
ISVAs operate within a slightly different multi-agency framework, and a narrower range of agencies may be involved. The limited literature on ISVAs suggests that they often have to negotiate with one agency at a time to advocate for women (Robinson, 2009b).
The role of advocacy workers in a coordinated community response is predominantly operational but they also work strategically. As they negotiate the criminal justice/housing/social work/welfare systems they form a picture of what is and what is not working. This contributes to plugging the gaps, overcoming barriers and improving system responses and processes (Howarth et al, 2009; Robinson, 2009a; Coy and Kelly, 2011).
There is a body of evidence about the processes and the outcomes of advocacy services as a response to domestic abuse and sexual violence.
There is little consideration of advocacy for other forms of gender-based violence.
From the literature, some common components of an effective advocacy response emerge. In brief, these are:
- Advocates are integral to the coordinated community response to gender-based violence, but must be independent of 'the system' in order to represent the interests of victims.
- Proactive outreach engages more victims at an earlier stage, and increases the likelihood that they will stay engaged with the criminal justice process.
- Risk assessment and safety planning are dynamic processes which advocates conduct with victims. The information which advocates gather from other agencies, including those involved with the perpetrator and/or with children/young people involved, is vital to these processes.
- Advocates provide information to victims, including information about the criminal justice process/what is happening with a case, and the outcomes of the MARAC, where appropriate; about legal, housing and welfare rights/entitlements; and information about the dynamics of abuse.
- Advocates represent the interests of the victim in information-sharing and risk-management forums; speak on behalf of the victim when they are unable to speak for themselves; and ensure that the interests of the victim stay central to multi-agency discussions.
- Advocacy is a time-limited crisis intervention. Once safety is achieved, advocates support victims (and their children) into follow-on services if required.
- Children and young people need advocacy in their own right.
Brooks, O., Burman, M. and Kyle, D. (2015). Evaluation of support to report pilot advocacy service: summary report. Glasgow: Scottish Centre for Crime and Justice Research.
Brooks, O. and Burman, M. (2016). 'Reporting rape: victim perspectives on advocacy support in the criminal justice process.' Criminology and Criminal Justice. 1-17.
Campbell, J.C. (2004). 'Helping women understand their risk.' Journal of Interpersonal Violence, 19(12).
Coy, M. and Kelly, L. (2011). Islands in the stream: an evaluation of four London independent domestic violence advocacy schemes. London: Child and Woman Abuse Studies Unit, London Metropolitan University.
Hague, G., Thiara, R. and Mullender, A. (2011). 'Disabled women and domestic violence: making the links - a national UK study.' Psychiatry, Psychology and Law, 18(1): 117-36.
Home Office (2005). Domestic violence: a national report. London: Home Office.
Howarth, E., Stimpson, L., Barran, D., Robinson, A. (2009). Safety in numbers: a multi-site evaluation of IDVA services. London: The Henry Smith Charity.
Kail, A., Miller, I. and Jarviner, J. (2007). Caada: NPC analysis 2007. London: New Philanthropy Capital ( NPC).
Kelly, L. and Humphreys, C., (2000). Reducing domestic violence? Outreach and advocacy approaches. London: Home Office.
Lovett, J., Regan, L. and Kelly, L. (2004). Sexual Assault Referral Centres: developing good practice and maximising potentials, Home Office Research Study 285. London: Home Office.
Ministry of Justice. (2010). Findings of the forced marriage IDVA support pilot (September 2009 - February 2010). London: Ministry of Justice.
Reid-Howie Associates. (2007). Evaluation of the pilot domestic abuse court. Edinburgh: Scottish Executive.
Robbins, R., McLaughlin, H., Banks, C., Bellamy, C. and Thackray, D. (2014). 'Domestic violence and multi-agency risk assessment conferences ( MARACs): a scoping review.' The Journal of Adult Protection, 16(6): 398-389.
Robinson, A. (2003). The Cardiff Women's Safety Unit: a multi-agency approach to domestic violence. Cardiff: Cardiff University.
Robinson, A. (2004). Domestic violence MARACs (multi-agency risk assessment conferences) for very high-risk victims in Cardiff, Wales: a process and outcome evaluation. Cardiff: Cardiff University.
Robinson A. (2007). 'Risk assessment and the importance of victim intuition.' Safe: The Domestic Abuse Quarterly, a national journal for practitioners, 21: 18-21.
Robinson, A. (2009a). Independent Domestic Violence Advisers: a process evaluation. London: Home Office.
Robinson, A. (2009b). Independent Sexual Violence Advisers: a process evaluation. London: Home Office.
Robinson, A. and Hudson, K. (2011). 'Different yet complementary: two approaches to supporting victims of sexual violence in the UK.' Criminology and Criminal Justice, 11(5): 515-533.
SafeLives. (2016a). MARAC in Scotland baseline report. Bristol: SafeLives. Unpublished.
SafeLives. (2016b). A cry for health: why we must invest in domestic abuse services in hospitals. Bristol: SafeLives.
Scottish Executive. (2008). Glasgow Domestic Abuse Court Feasibility Study Group: report to the Scottish Executive. Edinburgh: Scottish Executive.
Stanley, N. and Humphreys, C. (2014). 'Multi-agency risk assessment and management for children and families experiencing domestic violence.' Children and Youth Services Review, 47: 78-85.
Thiara, R. and Roy, S. (2012). Vital statistics 2: key findings report on black, minority ethnic and refugee women's and children's experience of gender-based violence. London: Imkaan.
Thiara, R., Roy, S. and Ng, P. (2015). Between the lines: service responses to black and minority ethnic ( BME) women and girls experiencing sexual violence. London: Imkaan.