What is the impact of peer support groups on refugees’ mental health?
Evidence Database | Advanced Evidence Search | Evidence Summaries
Evidence Summary
July 2020
Peer support groups can be a valuable component of comprehensive refugee resettlement services.
- There is moderate evidence for positive impacts of peer support groups among some refugee communities. Such groups appear to provide a unique benefit to both participants and leaders in enhancing social connections, knowledge of community resources, improvements in mental health, acculturation, and other outcomes.
- Many peer support groups are inherently culturally appropriate and strengths-based.They recognize and build refugees' strengths, including by offering opportunities to support one another through community-based activities. They can also promote refugees’ meaningful engagement as agents of change within their communities.
Member characteristics should be considered in composing groups.
- Group work appears most effective with small groups of people who have had similar experiences; thus, groups are ideally composed of ethnically homogenous members. For members of cultures with certain defined gender roles, gender-segregated groups are preferable. Special steps should also be taken to promote engagement and retention of youth and families in multi-family groups.
It is possible that online peer support groups may be a feasible and effective option for refugees.
- Research indicates that online support groups are effective in the general population, even among those with low digital literacy. Resettlement providers should explore refugees’ interest in participating in online support groups, along with the considerations needed to make such groups accessible and appropriate, such as digital literacy and access to digital technology among participants.
More impact evaluation studies are needed on the effectiveness of peer support groups.
- Rigorous evidence about the effectiveness of peer support groups is limited. Randomized controlled trials of this modality have been shown to be feasible and acceptable by refugee participants, and more such studies, especially regarding online groups, should be conducted.
Studies included in the database focused on high-income or upper middle-income countries, including but not limited to the United States. Studies included must have been published since 2000. To identify evidence related to peer support groups among refugees, we searched the following websites and databases using the following population, methodology, and target intervention terms:
Websites and Databases | Population Terms | Methodology Terms | Target Intervention Terms |
Campbell Collaboration Cochrane Collaboration Mathematica Policy Research Evidence Aid Urban Institute Migration Policy Institute HHS OPRE Medline ASSIA Social Services Abstracts Social Work Abstracts ReliefWeb ALNAP |
refugeeOR “unaccompanied minor” OR asylee OR “temporary protected status” OR “victims of traffick*” OR “traffick* victims” ORT -Visa OR U-Visa OR Cuban OR Haitian OR Amerasian |
evaluation OR impact OR program OR intervention OR policy OR project OR therapy OR treatment OR counseling OR workshop OR review OR meta-analysis OR synthesis |
“support group” |
For databases or websites that permitted only basic searches, free-text terms and limited term combinations were selected out of the lists above, and all resultant studies were reviewed for relevance. Conversely, for databases or websites with advanced search capability, we made use of relevant filters available. All search terms were searched in the title and abstract fields only in order to exclude studies that made only passing mention of the topic under consideration. After initial screening, Switchboard evidence mapping is prioritized as follows: First priority is given to meta-analyses and systematic reviews, followed by individual impact evaluations when no meta-analyses or systematic reviews are available. Evaluations that are rated as impact evidence are considered before those rated as suggestive, with the latter only being included for outcomes where no evidence is available from the former.