Aims Participatory learning and action approaches with women’s groups have been shown to have significant impact on maternal and neonatal mortality in low-income countries. It is not clear however whether this method can be used to address health issues in other settings, such as the UK. We carried out a systematic review to answer the question: ‘Can participatory action learning improve health and social outcomes in high-income countries (HIC)?’

Methods We included randomised controlled trials testing any group-based action learning intervention in HIC. Studies were included if they worked with community or patient groups and measured any health or social outcome. MEDLINE, EMBASE and the Cochrane database were searched.

Results 15 trials fitted our criteria, investigating a total of 1,841 participants. The theory and structure of the tested interventions were varied and none followed the cyclical approach shown to have positive effects in low-income settings. Common to all was an emphasis on participatory discussion and problem-solving rather than didactic teaching.

Most trials worked with disease specific groups and outcomes, notably diabetes. All but one targeted high-risk groups, defined for example by gender. The only trial to take a universalist approach had a narrowly focused outcome measure (handwashing).

Reid et al tested ‘Incredible Years’ parent discussion groups and showed that mothers in the intervention group showed higher rates of supportive parenting than both the control mothers and mothers whose children received a classroom intervention alone. Trials typically showed mixed results, however, with significant differences between arms for some outcomes but not for others.

Conclusion Few trials have tested group-based action learning interventions to improve health outcomes, and none has used a formal cyclical approach. Only 15 RCTs were identified and results were mixed, possibly due to short intervention and follow-up periods. Further research should assess the use of cyclical approaches for mixed groups and broad outcomes, based on strong evidence from low-income settings.