KONSİLWhat is a systematic review?
A systematic review answers a pre-specified question by searching for all studies that meet eligibility criteria, appraising them and synthesizing the findings with explicit, reproducible methods that aim to minimize bias. A narrative review has no formal rules of its own and offers a summary with interpretation and critique.
The Cochrane Handbook describes a systematic review as an effort to gather every piece of empirical evidence that meets eligibility criteria set in advance, to answer one specific question with methods designed to minimize bias. It says the search should be as extensive as possible, that searching two or more databases lowers the risk of missing eligible studies, and that publishing the protocol and registering it on PROSPERO reduces duplication of effort and research waste. Study selection and data extraction are done by at least two people independently. PRISMA 2020 asks reports to say how risk of bias was assessed.
Greenhalgh and colleagues describe a narrative review as a scholarly summary with interpretation and critique; it may lack a focused question or a methods section, yet they call it a different, potentially complementary form of scholarship rather than a lesser one. Erol notes that such reviews have no formal rules of their own but stresses that every review, whatever its type, needs a method fixed in advance and described in the text.
Meta-analysis, the statistical combination of study results, can be a step within a systematic review (Cochrane). PRISMA 2020 covers systematic reviews with and without synthesis, for example when only one eligible study is found, so not every systematic review includes a meta-analysis. PRISMA 2020 (BMJ 2021;372:n71) is a reporting guideline, not a guide to conducting reviews.
Example
In a hypothetical systematic review, the team registers a protocol, searches several databases with a pre-specified strategy, has at least two reviewers independently screen titles, abstracts and full texts, assesses the risk of bias of each eligible study, and then synthesizes the results.
A point to watch
According to the Cochrane Handbook, a systematic review fixes its question, eligibility criteria and methods in advance, as far as possible independently of the findings of the included studies, and keeps the search as extensive as possible. On the reading of these statements, skimming a handful of papers would not be a systematic review, and a single search in one database would fall short of the Handbook’s standard for Cochrane intervention reviews, which requires CENTRAL, MEDLINE and (where available) Embase to be searched.
Related terms
Guides and tools
Sources
- Cochrane Handbook for Systematic Reviews of Interventions, Chapter 1: Starting a review — Cochrane, 2024
- Cochrane Handbook for Systematic Reviews of Interventions, Chapter 4: Searching for and selecting studies — Cochrane, 2025
- Cochrane Handbook for Systematic Reviews of Interventions, Chapter 10: Analysing data and undertaking meta-analyses — Cochrane, 2024
- The PRISMA 2020 statement: an updated guideline for reporting systematic reviews — BMJ, 2021
- MeSH: Systematic Review (publication type) — U.S. National Library of Medicine
- Time to challenge the spurious hierarchy of systematic over narrative reviews? — European Journal of Clinical Investigation, 2018
- Derleme Yazmanın Temel İlkeleri — Archives of Neuropsychiatry, 2022