Protocol and registration
The protocol was developed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Extension for Scoping Reviews (PRISMA-ScR) [43] and published on Open Science Framework: https://osf.io/frh2e [44]. The PRISMA-ScR checklist is available in Appendix 3.
Eligibility criteria
We included any UK-based secondary or tertiary prevention support services and interventions aimed at adults who had experienced or perpetrated DSVA. Randomised controlled trials (RCTs), non-randomised comparative trials, pre-post studies and service evaluations that reported effectiveness outcomes at two or more time-points or made comparisons to another group were included, so that cause and effect could be inferred. Perpetrator programmes were considered for several reasons. Firstly, including them allows the measurement of DSVA without placing the burden on people who have experienced DSVA to change someone else’s behaviour. This ties in to the UK government’s new perpetrator strategy [45], which intends to place the onus on perpetrators changing their behaviour, alongside the recovery of those who have experienced DSVA. Finally, perpetrator programmes provide support for perpetrators to change their behaviour, and many offer associated support to (ex)partners or referral to appropriate support. Details of the full eligibility criteria can be found in Table 1.
Table 1
PICO Inclusion and Exclusion Criteria
|
Inclusion
|
Exclusion
|
Population
|
• Adults (16 years or older) who have experienced DSVA.
• Adults (16 years or older) who have perpetrated DSVA.
|
• Children
• Adults who have not experienced and/or perpetrated DSVA
|
Intervention (and service)
|
• Any secondary or tertiary prevention intervention and/or service for DSVA e.g., housing (e.g., refuges, housing workers, resettlements), Advocacy (e.g., Independent Domestic Violence Advisers/Advocates (IDVAs), Independent Sexual Violence Advisers/Advocates (ISVAs)), outreach, open access (e.g., helplines, drop-ins, online chats), psychological support (e.g., support groups, counselling, befriending), legal support, financial support, multi-agency risk assessment conferences (MARACs), and police-based services.
• Perpetrator programmes (see Appendix 1).
• Entry to the intervention had to be determined by the either the experience of DSVA (for victim-survivor support interventions) or perpetrating DSVA (for perpetrator programmes).
|
• Pharmacological
• Primary prevention
• Not primarily aimed at people with experience of DSVA
|
Comparison
|
• RCTs:
o Another type of included intervention or service.
o Usual care
o No intervention or service
• Pre-post designs:
o Before and after the intervention or service
|
• No comparison
|
Outcome
|
• Any outcomes used to measure effectiveness of DSVA support services and interventions.
|
• Those not focussed on the effectiveness of the intervention or service (e.g., process evaluation outcomes such as satisfaction, staff training, etc)
|
Setting
|
• UK based
|
• Not UK based
|
Study design
|
• RCT
• Nonrandomised comparative
• Pre-post
• Service evaluation
|
• Cross-sectional, case control, case study
• Qualitative
• Letters to the editor, think pieces, descriptive only
|
Dates
|
• 1982-present
|
• Pre-1982
|
Information sources and search strategy
MEDLINE, EMBASE, PsycINFO, Social Policy and Practice, ASSIA, IBSS, Sociological abstracts and SSCI electronic databases were searched from inception until 21st June 2022. Grey literature sources were also identified, including searching electronic grey literature databases (National Grey Literature Collection, EThOS, Social Care Online and the Violence Against Women Network) (Appendix 2), a call for evidence which was shared with research networks and UK-based DSVA services and organisations, and searches of relevant UK charity and organisation websites. Finally, backwards and forwards citation tracking of all included studies, as well as reference searching of identified systematic reviews was undertaken to identify further relevant studies. The search results were exported into EndNote and duplicates were removed.
Selection of sources of evidence
The de-duplicated citations were uploaded into Rayyan [46]. Titles and abstracts, followed by full-texts, were screened according to the inclusion and exclusion criteria. A 20% sample was independently screened by a second reviewer at each stage, and all disagreements were resolved by discussion, with a third reviewer where needed. Risk of bias assessments are not necessary for scoping reviews [41].
Data charting process and data items
Data charting took place using a piloted excel spreadsheet. Data items extracted included the study citation, study design, participant demographics, description of the service or intervention and comparison (if applicable), setting, and the outcome measures used. Data extraction and charting was carried out by one reviewer, and checked for errors by a second reviewer. Where data was missing, corresponding authors were contacted and asked to supply said data.
Synthesis of results
A narrative synthesis was conducted to summarise the included interventions and services and the effectiveness outcomes they measured. The unit of analysis was the outcome measure(s) reported. Interventions and services were grouped by the type of support offered, based on the intervention and service types outlined in the protocol. Where studies reported more than one form of support as part of the intervention or service, they were classified as “multi-service”. Where the same intervention or service produced multiple reports (e.g., from different years), these were combined into one entry during data charting so that all outcomes ever reported by that service were extracted, but were not double counted if multiple reports used the same outcome.
Effectiveness outcomes reported by studies were grouped thematically by type to create categories, domains and subdomains. The groupings and the names for the sub-domains, domains and categories were developed iteratively. All outcome measures were firstly listed in an excel spreadsheet. This included standardised and unstandardised questionnaires and single item measures. A column was added, describing what the outcome measured (e.g., the Beck Depression Inventory [47] measures depression). These sub-domains were then grouped into broader domains (e.g., depression is a mental health-related outcome). Where applicable, these were grouped into again even broader categories (e.g., mental health is a form of overall health).
A list of all relevant outcomes and domains, and how frequently they were reported was made. Comparisons were made between outcomes reported by studies in different sectors, over time, and by service user (i.e., perpetrator or victim-survivor) and support type (i.e., psychological support, housing, combinations), or violence type (i.e., domestic violence and abuse [DVA], sexual violence and abuse [SVA], childhood sexual abuse [CSA]).
Patient and public involvement
An advisory group was set up comprising representatives from six specialist DSVA organisations who are involved in the delivery, planning, funding or support of specialist DSVA support services in the UK. The group inputted to the design of the study protocol, interpretation of the data, and provided insight regarding the challenges in measuring the effectiveness of support services in the third sector. Their input resulted in several changes, including the broadening of the scope of this review to all effectiveness outcomes, rather than a narrower focus on only outcomes relating to reducing violence, to reflect the priorities of the sector.