Children affected by HIV in their families and communities face multiple risks to their health, education and psychosocial wellbeing. Community interventions for children who have been orphaned or rendered vulnerable take many forms, including educational assistance, home-based care, legal protection and psychosocial Support. Despite a recent influx of funding for programme implementation, there exists little evidence to inform policymakers about whether their investments are improving the lives of vulnerable children and meeting key benchmarks including the Millennium Development Goals. This paper reviews the current evidence base on evaluations of community interventions for orphans and vulnerable children (OVC) in high HIV-prevalence African settings, focusing on studies' methodologies. Sources reviewed include published research studies and evidence from the unpublished programmatic "grey literature" located through database and internet searches. A total of 21 studies, varying in scope and generalisability, were identified. Interventions reviewed address children's wellbeing through various strategies within their communities. Evaluation methodologies reflect quantitative and qualitative approaches, including surveys (with and without baseline or comparison data), costing studies, focus groups, interviews, case studies, and participatory review techniques. Varied study methodologies reflect diverse research questions, various intervention types, and the challenges associated with evaluating complex interventions; highlighting the need to broaden the research paradigm in order to build the evidence base by including quasi-experimental and process evaluation approaches, and seeking further insights through participatory qualitative methodologies and costing studies. Although findings overall indicate the value of community interventions in effecting measurable improvements in child and family wellbeing, the quality and rigour of evidence is varied. A strategic research agenda is urgently needed to inform resource allocation and programme management decisions. Immediate imperatives include building local technical capacity to conduct quantitative and qualitative evaluation research, and strengthening monitoring and evaluation systems to collect process and outcome data (including costing) on key support models. Donors and implementers must support the collection of sound empirical evidence to inform the development and scale-up of OVC programmes.