Written and maintained by CASRAI Editorial Board
Last updated
A realist review (also called a realist synthesis) is a theory-driven approach to evidence synthesis that asks not just whether a complex intervention works, but why it works, for whom, and in what circumstances. Developed by Ray Pawson and colleagues, it is built around context-mechanism-outcome (CMO) configurations: the reviewer develops and refines a program theory describing how an intervention’s mechanisms are triggered (or not) by different contexts to produce different outcomes, then tests that theory against the available evidence — which can include quantitative, qualitative, and grey literature, unlike a conventional systematic review’s typically narrower inclusion criteria. Realist review is most often used to evaluate complex social, organizational, or policy interventions where a single “does it work” answer would obscure the fact that the intervention works well in some settings and not others.
Last verified 2026-08-24 against the EQUATOR Network’s RAMESES reporting-guideline pages and the original RAMESES publications.
The RAMESES Publication Standards
Reporting standards for realist work were developed under the RAMESES project (Realist And Meta-narrative Evidence Syntheses: Evolving Standards), led by Geoff Wong, Trish Greenhalgh, Gill Westhorp, Jeanette Buckingham, and Ray Pawson, using a Delphi consensus process with an interdisciplinary panel. The project produced two related sets of standards, published simultaneously in BMC Medicine and the Journal of Advanced Nursing in 2013:
- RAMESES publication standards: realist syntheses — 19 standards covering how to report a realist review (a secondary-research synthesis of existing literature), addressing the study’s rationale, its program theory, the search and selection process, appraisal and use of documents as evidence, and how CMO configurations are developed and reported.
- RAMESES publication standards: meta-narrative reviews — 20 standards for meta-narrative review, a related but distinct approach (also associated with Greenhalgh) for synthesizing evidence across research traditions that use conflicting paradigms and vocabularies to study the same broad question.
A separate RAMESES II standard applies to realist evaluation — primary research that applies realist methodology to a specific intervention in a specific setting, rather than reviewing existing literature about it. The three are related but not interchangeable: use realist-synthesis standards for a literature review, meta-narrative standards for a review that must reconcile conflicting research traditions, and RAMESES II for a primary realist evaluation.
The full standards, with the numbered checklist items, are maintained on the EQUATOR Network’s realist syntheses standards page and RAMESES II realist evaluations page.
Realist Review vs. Systematic Review vs. Umbrella Review
A conventional systematic review asks a narrower question — typically whether a specific intervention produces a specific outcome — using pre-specified inclusion criteria and often a PRISMA-reported search and screening process. A realist review deliberately casts a wider evidentiary net (it may include process evaluations, theory papers, and grey literature that a systematic review would exclude) because its goal is to explain variation in outcomes across contexts, not to produce a single pooled effect estimate. It is therefore a poor fit for questions with a clean, well-defined intervention and outcome, where a systematic review or meta-analysis remains the right tool — see CASRAI’s rapid review vs. systematic review and scoping review guides for those alternatives, and the umbrella review entry for reviews that synthesize other reviews rather than primary studies. For a general map of which evidence-synthesis method and reporting guideline fits which question, see CASRAI’s EQUATOR Network guide.
Frequently Asked Questions
What is a realist review?
A theory-driven approach to evidence synthesis that explains why, for whom, and under what circumstances a complex intervention works, rather than simply whether it works on average. It develops and tests context-mechanism-outcome (CMO) configurations against a deliberately broad range of evidence.
What does RAMESES stand for?
Realist And Meta-narrative Evidence Syntheses: Evolving Standards — the project that produced the current reporting standards for both realist syntheses and meta-narrative reviews.
How many items are in the RAMESES realist-synthesis reporting standards?
19 publication standards, developed through a Delphi consensus process and published in 2013 in BMC Medicine and the Journal of Advanced Nursing.
What is the difference between a realist review and a realist evaluation?
A realist review (or realist synthesis) is secondary research — a literature review, reported against the RAMESES realist-synthesis standards. A realist evaluation is primary research that applies realist methodology to study a specific intervention in a specific setting, reported against the separate RAMESES II standards.
What is a meta-narrative review?
A related evidence-synthesis approach, also developed by Trish Greenhalgh’s group, for reconciling research that studies the same broad question from conflicting disciplinary traditions and vocabularies. It has its own 20-item RAMESES publication standard, distinct from the realist-synthesis standard.
Related reading: How to Conduct a Scoping Review — the right tool when the goal is mapping the extent of evidence on a topic rather than explaining why an intervention works differently across contexts.
See also: Umbrella Review — a synthesis of existing systematic reviews, a different evidence-synthesis approach from realist review’s synthesis of primary and grey literature.
