Evidence-based policy is a dominant theme in contemporary public services but the practical realities and challenges involved in using evidence in policy-making are formidable. Part of the problem is one of complexity. In health services and other public services, we are dealing with complex social interventions which act on complex social systems--things like league tables, performance measures, regulation and inspection, or funding reforms. These are not 'magic bullets' which will always hit their target, but programmes whose effects are crucially dependent on context and implementation. Traditional methods of review focus on measuring and reporting on programme effectiveness, often find that the evidence is mixed or conflicting, and provide little or no clue as to why the intervention worked or did not work when applied in different contexts or circumstances, deployed by different stakeholders, or used for different purposes. This paper offers a model of research synthesis which is designed to work with complex social interventions or programmes, and which is based on the emerging 'realist' approach to evaluation. It provides an explanatory analysis aimed at discerning what works for whom, in what circumstances, in what respects and how. The first step is to make explicit the programme theory (or theories)--the underlying assumptions about how an intervention is meant to work and what impacts it is expected to have. We then look for empirical evidence to populate this theoretical framework, supporting, contradicting or modifying the programme theories as it goes. The results of the review combine theoretical understanding and empirical evidence, and focus on explaining the relationship between the context in which the intervention is applied, the mechanisms by which it works and the outcomes which are produced. The aim is to enable decision-makers to reach a deeper understanding of the intervention and how it can be made to work most effectively. Realist review does not provide simple answers to complex questions. It will not tell policy-makers or managers whether something works or not, but will provide the policy and practice community with the kind of rich, detailed and highly practical understanding of complex social interventions which is likely to be of much more use to them when planning and implementing programmes at a national, regional or local level.