Background: Provision of non-pharmacological interventions is a common policy objective for people with dementia, and support groups are an increasingly common intervention. However, there have been few attempts to synthesize evidence on the effectiveness of support groups for people with dementia. This review investigated the outcomes of support groups for people with dementia, explored participant characteristics and reviewed group formats. Methods: A systematic review was undertaken and a narrative synthesis of data from 29 papers (reporting on 26 groups and a survey of a range of groups) was conducted. Results: Support groups seem acceptable to people with dementia. Qualitative studies report subjective benefits for participants but there is limited evidence of positive outcomes based on quantitative data. Samples have tended to be homogenous and this may limit the generalizability of findings. Conclusions: Although qualitative studies will remain important in this area, further mixed-methods randomized controlled trials (RCTs)or comparison group studies with longer follow-up periods are needed to strengthen the evidence base.