Individual advocacy and assistance with service navigation for women in or just leaving shelter is very promising and can improve women’s physical and psycho-social well-being by increasing their use of safety behaviours, and reducing their symptoms of psychological distress and depression (Feder, et al., 2009, based on Sullivan, 1991; Sullivan & Davidson, 1991; Sullivan, et al. 1992; Sullivan & Bybee, 1999; Bybee & Sullivan, 2002 and 2005).
More frequent and sustained contact with survivors may increase the impact of advocacy on the safety of women and their children, especially where interventions facilitate linkages to coordinated services (Howarth et al., 2009).
Research from developed countries assessing specific criminal justice system interventions responding to domestic violence shows:
- Women’s ability to effectively interface with the criminal justice system depends on a number of individual, relational and system-level factors (e.g. a survivor’s state of mental health, especially post-traumatic stress disorder; the availability of social support; and previous experiences with police), which also determine whether women seek orders of protection or prosecution of the abuser (Jordan, 2004).
- Women may be slightly more willing to report incidents to police when they are provided a second or follow-up effort, such as a police and/or domestic violence advocate team intervening with a family after an initial domestic violence call, although it does not affect the likelihood of new incidents of violence (Davis et al., 2008).
- Protection orders have demonstrated mixed results, although permanent protection orders may be more effective in reducing future violence than temporary orders, and that even when orders are violated, women with orders of protection may experience fewer incidents of violence. The specific impact varies according to setting and context (Benitz et al., 2010; Holt et al., 2002; Logan & Walker, 2010; and Logan et. al., 2005).
- Emerging interventions showing promise include legal advocacy for abused women, and specialist domestic violence courts (Cook et al., 2004; Bell and Goodman, 2001; Cattaneo et al., 2009).
See the full module dedicated to the Justice Sector.
Economic and housing support significantly affects the ability of women to end the abuse in their lives. Income variables have been shown to be the strongest predictor of women’s decisions to stay in or leave an abusive relationship, which can be further influenced by societal factors and policy decisions that either support or marginalize survivors. The ability to obtain affordable housing is often the most pressing need of women exiting a shelter, though policies are not normally responsive to the specific circumstances of survivors. Secure and affordable housing assistance can contribute to women’s well-being in similar ways as experienced by women who receive shelter services (Anderson & Saunders, 2003, Ham-Rowbottom et al., 2005; Howarth et al., 2009; Baker et al., 2009; Baker et al., 2010, Eby, 2004; Niolon et al., 2009;Staggs et al., 2007; Morrow et al., 2004).
Shelters have played a historic role in establishing and maintaining coordinated responses for survivors, although there is limited evidence on the specific impact of shelters (distinct from domestic violence advocates) within such mechanisms. For example, shelter advocates are involved in coordinated response models, such as the United Kingdom’s Multi-Agency Risk Assessment Conference (MARAC), which assesses and responds to the needs of high risk domestic violence survivors, and has shown positive contributions to rates of re-victimisation, as well as participant satisfaction, case efficiency and other process outcomes. While shelters should be engaged in coordinated response systems, the effectiveness of coordination models may vary due to the diversity of structures and operating mechanisms across models. See the full Case Study on the MARAC (Klevens et al., 2008; Robinson & Tregidga, 2007; Robinson, 2006; Robinson, 2004). See more detailed guidance on Coordinated Responses.