Domestic Homicide Reviews: evidence briefing

4. The Process of Conducting a Review

Implementing a DHR and establishing its process requires consideration of several aspects, including the setup of the panel, who is included, what methodology is used, whether family will be involved and how other death reviews might interact with the DHR. The following sections will address these questions. Table 3 in Annex 1 provides an overview of the DHR process for each jurisdiction.

4.1 Review panels

Key findings:

Most jurisdictions work with a multi-agency panel, with representation by police, ministerial departments, health providers, domestic abuse service providers, victim advocates and domestic abuse experts. The panel can be established at jurisdiction or local level and can be a permanent team or created ad hoc.

The literature suggests that membership of these multi-disciplinary panels should be inclusive and culturally sensitive. Moreover, clarity on roles and responsibilities and, especially where there are multiple local panels operating under the same model, of the DHR process is important. Lack of clear guidance might lead to differences in how DHRs are conducted, what type of information is gathered and what is reported upon.

Most jurisdictions' DHRs operate on a multi-agency basis, with the majority including representation by police, ministerial departments, health providers, domestic abuse service providers, victim advocates and domestic abuse experts (e.g. academics). Other approaches include South Australia, for example, where a research officer is responsible for gathering information on domestic homicides, which can lead to a coronial inquest, where different agencies will be called to 'give evidence'. In West Australia the Ombudsman is responsible for DHRs. The office of the ombudsman will work with all agencies involved and is supported by an advisory panel.

Multi-agency teams can be established at a jurisdiction-wide or at a local level, and can either be dedicated, permanent teams or established ad-hoc as required. The literature suggests that permanent panels can allow for a good team dynamic to develop and skill development, while ad hoc panels can allow for a greater number of professionals to gain experience with DHRs (Rowlands 2020a). In most cases where teams operate on a local level, there is an overall team or board that pulls together the results of the different reviews at jurisdiction level. The literature suggests that an overarching body to draw together all the learning of local teams and identify recommendations is important for understanding the overall impact of the DHRs. In the cases of Australia, Canada and the USA, where DHRs are regulated at state-level, there is a country-wide network to support cross-jurisdiction learning.

Either permanent or ad hoc, at local or jurisdiction level, the literature suggests that membership of these multi-disciplinary panels should be "inclusive rather than exclusive" (Websdale 2020) and should be culturally sensitive. This latter point is important, as often wider factors such as discrimination, stereotypes and structural oppression might play a role and it is important to consider these (Jones et al. 2022). Rowlands (2020a) and Websdale (2020) point out there is a critical role for domestic violence specialists in helping panel members to understand the complexity of domestic violence.

Haines-Delmont et al. (2022) showed that in England and Wales, some professionals felt there was a lack of diversity and input, particularly from third sector specialist domestic abuse agencies. Jones et al. (2022) point out that setting up a multi-agency team will require transparency and honesty about tensions that can arise between members of a review team, acknowledging different organisational priorities and practices. They point out the need for a shared language and standardised terminology (Jones et al. 2022).

The literature also highlights the need for clarity on the roles and processes of the DHRs. Boughton (2021) point out that in England, the Home Office guidance for example states that the DHR panel must meet "an appropriate number of times", but no further commentary is made what is considered appropriate. Boughton's study showed that the role of the panel member and the Individual Management Review author in the English model also remain unclear. The guidance states who should be considered for the roles, but there is very little reference to the responsibilities of the role (Boughton 2021). Rowlands (2020a) argues that when reviews are implemented by multiple local teams this lack of clear guidance might lead to differences in how DHRs are conducted, what type of information is gathered and what is reported upon. For example, in England and Wales the timeframe that a review covers (i.e. how far a review looks back) seem to vary (Jones et al. 2022).

In several jurisdictions guidance stipulates that members of the panel should be of an 'appropriate level of seniority'. Reviewing this guidance in England and Wales, Boughton (2021) suggest that this has however led to participants being concerned that senior members might not have the knowledge of 'on the ground' processes. Other reviews, such as the Welsh Child Practice Review (CPR), recognise this and take a different approach, with CPRs holding at least two Learning Events per review, one for management and one for practitioners directly involved in the case (Boughton 2021). A core philosophy of the DHR is that it should not blame and shame individuals. However, in practice it has been reported that professionals can be defensive and try to shift responsibility, as there can be "a tension for representatives from organisations between what their organisation will allow them to say, the shifting of blame and the Chair's final report" (Haines-Delmont et al. 2022).

In most of the jurisdictions, the chair of the review team/committee has a significant role to play. The chair's skills and experience in dealing with sensitive and contentious issues are therefore important enablers of the DHR process (Haines-Delmont et al. 2020). Haines-Delmont et al. (2020) point out that it can be advantageous to have an independent chair, although this might require significant resources, and the chair might lack some knowledge on the local context.

4.2 Review process and methodologies

Key findings:

Broadly speaking the different jurisdictions tend to follow the same general process and use similar methodologies. These includes constructing a timeline, identifying agencies and individuals that the victim (and perpetrator) had contact with, and evaluating the sharing of information and collaboration between agencies.

Addressing privacy and confidentiality is an important aspect to consider when establishing a DHR. The literature suggests that teams should follow the "do no harm" principle and accessing information should not undermine confidence of victims in advocacy services.

Other aspects that are considered important, particularly for sustainability of the DHR, are review teams' capacity and resources as well as effective oversight and leadership.

Although there is variation in the exact process followed by the DHR models in different jurisdictions, in general the process will observe the following steps:

• Team or those responsible for deciding whether a DHR takes place receives information on domestic homicides;
• A DHR is initiated (if a jurisdiction works with ad hoc teams, a DHR team is established and a terms of reference is produced);
• Information is gathered from records and requested from agencies who have had contact with the victim and/or perpetrator;
• Family and friends are asked to provide information about the victim's (and perpetrator's) life;
• A meeting is called, bringing together the review team;
• The team reviews the information, establishing a time-line of events leading up to the homicide;
• Recommendations are formulated and reported on.

Often jurisdiction-wide panels will work with a two-tiered mixed methods data collection process: collecting standardised quantitative data on all domestic homicide cases and conducting in-depth qualitative reviews on a selection of cases.

In New South Wales, Australia, the Domestic Violence Death Review Team (DVDRT) is a multi-agency committee convened by the State Coroner, bringing together representatives from key government agencies, non-government service providers and sector experts.

The DVDRT reviews individual closed cases and identifies systemic issues; it understands domestic and family violence as a complex, intergenerational and 'wicked' problem that requires complex responses that reach across government, non-government and community actors. It takes a two-tiered approach to analyse the cases:

- Tier 1: 'real time' domestic violence homicide dataset (for quantitative data analysis)

- Tier 2: Examination of in-depth case reviews (qualitative data analysis), following a comprehensive examination and analysis of all available case material. The secretariat prepares a case review report, which is examined by the team in a series of workshops. Recommendations are developed in consultation with agencies.

Broadly speaking, most DHR models use the same methodologies for in-depth reviews. In general a timeline is constructed, agencies and individuals identified that the victim (and perpetrator) had contact with, and information sharing and collaboration between agencies is evaluated (Dale et al. 2017; Rowlands 2020a; Websdale 2020). Teams "identify 'red flags' or risk markers in the relationship [between the victim and perpetrator] that may have indicated an elevated level of danger" (Dale et al. 2017 p. 230). Some reviews will consider a broader context such as situational factors, broader social structures and cultural values and beliefs (Rowlands 2020a, Websdale 2020).

To conduct a DHR the review panel is dependent on receiving information from agencies, which will include personal information on victim and perpetrator. There are different approaches taken to confidentiality and privacy and how individual case information is accessed and shared. The literature suggests that teams should follow the "do no harm" principle and accessing information should not undermine general confidence of victims in service providers. Dale et al. (2017) point out that absolute confidentiality is seen as important, during the review process. In a number of jurisdictions panel members will sign confidentiality agreements to ensure privacy is protected and an open discussion is guaranteed. Many reviews will also only report aggregated data and learning, to protect the privacy of those involved. This may create a tension, however, with giving the victim's voice a central role which could mean that "the uniqueness of each person's story is lost" (Rowlands 2020a p.32).

Review teams' capacity and resources are key issues for sustainability, as are effective oversight and leadership. DHRs are resource intensive and unstable funding has led in some jurisdictions to difficulties in producing reports consistently (Jones et al. 2022). Jones et al. (2022) note a tension between speed and thoroughness, and with little known about how proposed timeframes play out in practice, there is a need to better understand these (Rowlands 2020a). Boughton (2021) found that due to Community Safety Partnerships in England and Wales (who conduct the DHRs) having to continuously balance their finances, it can sometimes lead to the decision to not review a case, even though the case would fall within the selection criteria.

4.2.1 A gendered approach

Key findings:

Domestic abuse and domestic homicide is a gendered issue.The literature suggests the need for a focus on the gendered dynamics of domestic homicides.

Establishment of DHRs have often come about through involvement of feminist advocacy and in several jurisdictions the DHR is embedded in the government's strategy for tackling violence against women.

It is less clear if a gendered approach is taken when conducting a DHR. In the 17 jurisdictions included in this briefing data is presented disaggregated by sex. However, other studies pointed out that a feminist perspective or methodology seems almost entirely absent from the process. While some reports highlight gendered drivers and impacts of domestic abuse, most recommendations tend to focus on service provision or risk assessment rather than on structural inequalities.

Domestic abuse and domestic homicide is a gendered issue. While the majority of homicide victims are men, most victims of intimate partner homicide are women (Haines-Delmont et al. 2022). Moreover, a UN study showed that in 2021 women and girls were most likely to be killed by a member of their own family (UNODC and UN Women 2022)^[12]. With this in mind, Sheeny (2017) argues that a feminist lens of analysis is necessary which focuses on the gendered dynamics of domestic homicide and societal barriers to change.

Although an in-depth analysis of the DHR methodologies and data representation is out of the scope of this briefing, a quick scan of the 17 jurisdictions did show an acknowledgement of the gendered dimension of domestic abuse and homicides, especially when establishing a DHR model. The establishment of DHRs often came about through involvement of feminist advocacy, and in several jurisdictions it is embedded in government's strategy tackling violence against women.

When conducting a DHR, however, it is less clear whether a gendered approach is taken. In the 17 jurisdictions covered, data was presented disaggregated by sex, showing the gendered nature of the homicides. Moreover, in New Zealand part of the DHR methodology is to use a structural analysis, taking an intersectional approach to the data and consider domestic abuse as a complex societal problem and acknowledge gender inequality (Rowlands 2020a). However, in a review of (a limited number of) DHRs in Canada, Sheeny (2017) points out that analysis from a feminist perspective seems almost entirely absent from the process. The UK Government has received some criticism from those who consider that they have failed to recognize the gendered nature of domestic abuse in the DHR model (Rowlands 2021).

The aim of many DHRs is to share information and evidence relating to the identification of domestic and family violence risk indicators and/or case characteristics. This analysis and presentation of data trends and patterns opens the possibility to highlight gendered impacts. In New South Wales the 2017-2019 DHR report, for example, mentions men's behaviour, its gendered drivers and societal attitudes towards women, and acknowledges the impact of power-imbalances and gendered structures in society. The sixth report of New Zealand's review committee includes a discussion on the reduction of men's use of violence, providing recommendations on healthy masculine norms and reconnecting men with positive forms of social support. Most often, however, recommendations tend to focus on service provision, coordination between agencies or risk assessment rather than on structural gender inequalities.

Other inequalities might also play a role in understanding domestic homicides. An intersectional approach to domestic homicides could take into account both the gendered nature of the issue as well as potential other (structural) inequalities that may play a role. In Montana for example, there is a specific Native American Domestic Violence Fatality Review Team (NADVFRT), focused on domestic homicides in Montana's Indian Country. This team aims to recommend culturally sensitive, proactive changes to prevent domestic homicides in the future. In the Scottish context, a DHR policy will be subject to an Equality Impact Assessment (EQIA) which will allow to flag potential inequality impacts.

4.3 Family involvement

Key findings:

Some review teams will include testimonies from family and friends of the victim (and sometimes perpetrator) to inform the DHR, and the literature recommends to include family (and friends) in the review process.

It is suggested that a DHR can provide families with a sense of healing, and give them a critical role in the process. The literature also notes that DHR teams recognise that there are challenges, both practical as well as concerns around the family's experience of loss and the risk of further traumatisation.

There is still limited understanding on how family engagement is undertaken and whether intended outcomes are met.

To conduct a DHR, review teams will draw on administrative data (e.g. police and court records) and data received from agencies involved. Some teams will also include testimonies from family and friends of the victim (and sometimes perpetrator). Not all jurisdictions covered in this briefing include family and friends, and those that do include them can differ in the level of engagement. In New Zealand the review committee has acknowledged for example that the current approach is not reciprocal, and they would like to embed the family more fully in the review process. In the 2017-2019 report of the New South Wales review, the review team states that while they are currently using document reviews to analyse a case, opportunities to engage with surviving friends and family are being explored.

In England, guidance for DHR panel states that "the review panel should carefully consider the potential benefits gained by including such individuals from both the victim and perpetrator's networks in the review process. Members of these support networks should be given every opportunity to contribute unless there are exceptional circumstances, for example, where there are suspicions of 'honour'-based violence".

The guidance mentions that one of the benefits of this approach includes "assisting the family with the healing process which links in with the objectives of the new National Homicide Service - supporting victims for as long as they need after homicide."

The review panel is, however, made aware of sensitivities and the need for confidentiality.

The literature recommends to include family (and friends) in the review process to provide their perspective and experiences. Mullane (2017) argues that families should be integral to the review process, as "the review is a journey for the family (…) and should allow the family to contribute, receive and review information" (p. 262). Rowlands (2020b) notes that family and friends can help provide a more victim-centred story, and it is important to treat their accounts with the same status as agency information, as well as facilitate multiple opportunities for meaningful involvement.

Rowlands and Cook (2022) point out that DHR teams increasingly recognise the input of families as valuable. However, the teams also recognise challenges, with both practical concerns (e.g. on confidentiality) as well as "the emotional burdens of contributing" (Rowlands and Cook 2022, p. 560). This includes concerns around the family's experience of loss and the risk of further traumatisation. Mullane (2017) points out, for example, that sometimes sensitive information previously unknown to the families might be revealed in a review. He suggests that a specialist advocate, preferably peer support, can help families to understand the DHR process, assist with coordinating the family's contribution, managing expectations, as well as supporting the family with discussing the review outcomes and report with the chair/panel (Mullane 2017).

In an analysis of family involvement in the DHR process in England and Wales, Rowlands and Cook (2022) argue that family involvement is often framed as offering two outcomes: contributing to the diagnosis and identification of any system-failures (and formulating possible solutions); and enabling relational-repair for families through their participation in a DHR. However, there is still limited understanding on how family engagement is undertaken and whether these outcomes are reached (Rowlands and Cook 2022). Dale et al. (2017) point out that when and how a family is approached is important, as well as ensuring that, in cases where family and friends are interviewed, there are trained interviewers to do so. Follow-up for family and friends should also be addressed, both in offering care and information on the DHR.

Jones et al. (2022) states that there is limited involvement of children, "despite their active role in experiencing [Domestic Violence and Abuse], witnessing the homicide and calling for help" (Jones et al. 2022 p.4).

4.4 Other death reviews

Key findings:

Apart from a DHR a jurisdiction might have other (death) review processes in place. It appears that often there is no clear process in place to coordinate multiple review processes. This can lead to several reviews being conducted on the same case.

In a couple of jurisdictions there was specific mention of other review processes, such as child death reviews, and how the DHR should interact with those. In England, for example, the case selection criteria include a note that where the victim is between 16 and 18 years old, both a Serious Case Review and a DHR are required. These may be run in parallel, and some aspects can be commissioned jointly.

Often however, there does not appear to be a clear process in place to coordinate multiple review processes and this can lead to several reviews being conducted on one case simultaneously. In Wales, for example, one case was identified where seven different reviews took place (not including the police investigation and coroner's inquest). It shows the need to streamline these processes, as "the duplication of evidence gathering, where single incidents trigger numerous reviews (…) would seem to be unwieldy, unfair to family and not in the spirit of multi-professional, inter-agency working" (Robinson et al. 2018 p. 11). In Wales, this led to commissioning an evaluation of its review process (Robinson et al. 2018). The evaluation focused in particular on the overlap between DHRs and other death reviews, in order to improve coordination between these reviews. It led to the design of a new unified review process. The Single Unified Safeguarding Review model in Wales is currently being finalised and is expected to be rolled out across Wales in 2023.