| ||Literature Coverage Dates||Number of Studies||Number of Study Participants|
|Meta-Analysis 1||1991 - 2010||13||2141|
Rivas and colleagues (2016) conducted a meta-analysis to examine the effects of advocacy interventions for women who experience intimate partner violence (IPV). Between 2011 and 2015, the researchers searched electronic databases for international peer-reviewed and non-peer-reviewed studies. The researchers also conducted website searches, manually searched journals, tracked citations, and emailed authors to inquire about potential new studies. The search spanned literature available from 1948 to 2015. Studies were eligible for inclusion if participants were at least 15 years of age and were randomly or quasi-randomly assigned to receive either an advocacy intervention or no care/care as usual. Eligible studies recruited women from healthcare settings, criminal justice facilities, refuges/shelters, or domestic violence agencies. Interventions could be brief (less than 12 hours) or intensive (12 or more hours) and could be delivered in individual or group settings. Interventions incorporated safety planning with the women or facilitated access to and use of community resources, such as refuges or shelters, emergency housing, and psychological care. Interventions that provided these services, either with or without ongoing informal support or counselling, were included.
In a prior review (Ramsay et al. 2009), the researchers only included multicomponent intervention studies where advocacy was the sole difference between the treatment and comparison groups (i.e., both groups received intervention A, but only the treatment group received the additional advocacy intervention). For the present review, multicomponent intervention studies included advocacy in addition to some other form of intervention, and these were compared with no care or care as usual. The present review also incorporated a more stringent definition of “usual care,” defining this type of care as including a minimal element of advocacy, such as providing information cards or leaflets listing addresses and telephone numbers of local support agencies. If more substantive advocacy was offered, then the study could be included only if fewer than 20 percent of women in the control group had received such care.
The search yielded 13 eligible studies from 1991 to 2011. Eleven of the studies were randomized controlled trials, and two were quasi-experimental designs. Twelve of the studies were published in peer-reviewed journals, and one was a published doctoral dissertation. Eight of the studies recruited women in healthcare settings, three in domestic violence shelters, one primarily in a domestic violence shelter but also through social and family service agencies, and one from a community center in a large urban area. The severity of abuse faced by the women varied considerably across studies, ranging from minor to severe physical, sexual, or emotional abuse. A total of 2,141 women took part in the trials. Sixty-one percent were recruited because they were experiencing current (within the last 12 months) physical or sexual abuse. Seven studies extended this inclusion to women who were experiencing emotional or non-physical abuse. Most of the women were still living or intimately involved with the abuser at study entry. The women ranged in age from 15 to 65; however, most were between the ages of 24 and 45. Seven studies included white, black, and Latina women; four studies focused on single ethnic groups (one was with Latina only, one was with black women only, and two were conducted in Hong Kong with Chinese women); one study focused predominantly on women who identified as Mestizo; and one study was conducted in Australia where nearly half of the women were born in Vietnam or a country other than Australia.
Interventions varied in length. In general, interventions described in non-healthcare settings were longer, typically taking place over a period of 2 to 4 months. Most of the interventions recruiting women in healthcare settings were relatively brief; three were one-off sessions and lasted 20 to 30 minutes, 30 minutes, and 90 minutes. Four other trials offered advocacy sessions over a prolonged period but were still generally brief. Follow-up length also varied by studies. Four studies did not conduct a further follow up after assessing outcomes upon completion of the intervention, and the remaining nine studies conducted follow ups from 10 weeks to 3 years postintervention.
The 13 studies included 25 outcome measures. Eleven studies measured some form of abuse (using eight different scales), six assessed quality of life (three scales), six measured depression (three scales), and three measured anxiety or psychological distress (three scales). Eight studies measured physical abuse, six of which evaluated brief advocacy interventions, and two that evaluated intensive interventions. Effects sizes were calculated for five of the six studies that offered brief advocacy. Fixed-effect models were used in all analyses given that all analyses were based on subsets of studies that were deemed clinically heterogeneous. Effect sizes were restricted to the latest sample or latest follow up. The researchers did not indicate use of weighting.