Cochrane Review

A Cochrane Review is a systematic review of primary research in healthcare and health policy produced by the Cochrane Collaboration, following a structured, transparent, and pre specified methodology set out in Cochrane's methodological guidance, which for intervention reviews is the Cochrane Handbook for Systematic Reviews of Interventions. Cochrane Reviews are published in the Cochrane Database of Systematic Reviews and can be updated when new evidence or methods are likely to change their findings.

Cochrane Reviews are considered the gold standard in evidence synthesis for healthcare. Their rigorous methodology, including comprehensive searching, duplicate study selection and data extraction, standardized risk of bias assessment, and pre specified analysis plans, minimizes the risk of bias and enhances the reliability of their conclusions. Healthcare guideline developers, health technology assessment agencies, and policy makers worldwide prioritize Cochrane Reviews when evaluating the evidence for clinical interventions. For researchers, Cochrane Reviews provide a model of best practice in systematic review methodology and a benchmark against which other reviews are compared.

The production of a Cochrane Review follows a detailed sequence. First, the review team submits a review proposal to Cochrane and, once it is accepted, publishes a protocol in the CDSR, specifying the research question, eligibility criteria, search strategy, data extraction plan, risk of bias assessment method, and statistical analysis approach. The protocol is peer reviewed before publication.

The search strategy is designed to be comprehensive, searching multiple databases (CENTRAL, MEDLINE, Embase, and often additional databases relevant to the topic), trial registries, grey literature sources, and reference lists. The goal is to identify all relevant studies regardless of language, publication status, or results, minimizing publication and language bias.

Study selection and data extraction are performed independently by at least two reviewers, with disagreements resolved by discussion or a third reviewer. Risk of bias is assessed with the tools the Cochrane Handbook recommends: RoB 2 for randomized trials (many earlier reviews used the original Cochrane risk of bias tool) and ROBINS-I for non randomized studies of interventions. The judgments are presented in risk of bias tables and summary figures.

When studies are sufficiently similar, meta analysis is conducted using random effects or fixed effect models, with heterogeneity assessed using the I squared statistic and chi square test. Subgroup analyses and sensitivity analyses are pre specified in the protocol. The certainty of evidence is evaluated using the GRADE framework, which rates evidence as high, moderate, low, or very low based on risk of bias, inconsistency, indirectness, imprecision, and publication bias.

The completed review undergoes editorial and external peer review, managed by Cochrane's Central Editorial Service, before publication. Cochrane once expected reviews to be updated about every two years; it now decides whether to update a review based on need, that is, whether the question still matters and whether new studies or methods would meaningfully change the findings. A small number of high priority reviews are maintained as living systematic reviews.

Cochrane Reviews should be consulted as a primary evidence source for any clinical question about the effectiveness or safety of health interventions. Researchers planning a new systematic review should first check whether a relevant Cochrane Review exists, to avoid duplication and to identify any gaps in the existing evidence. The Cochrane methodology can also serve as a template for high quality reviews conducted outside the Cochrane framework.

A common error is assuming that a Cochrane Review is definitive and current without checking its update status. Some reviews have not been updated for several years and may not include important recent evidence. Another mistake is ignoring the GRADE assessments provided in Cochrane Reviews, which indicate the certainty of the evidence and are essential for understanding how much confidence to place in the conclusions.

Researchers sometimes also cite a Cochrane Review's conclusions without examining its limitations, such as the number and quality of included studies, the degree of heterogeneity, and the applicability of the evidence to their specific clinical context. Finally, there is sometimes a misperception that only Cochrane Reviews follow rigorous methodology; many non Cochrane systematic reviews achieve comparable quality when they follow rigorous conduct standards, register a protocol, report transparently using PRISMA 2020, and use validated risk of bias tools.

The Cochrane Review "Corticosteroids for pneumonia" (Stern and colleagues, 2017) included 17 randomized controlled trials with 2,264 participants: 13 trials in 1,954 adults and 4 trials in 310 children, all in hospital with community acquired pneumonia. Two review authors independently assessed risk of bias and extracted data, and trials were pooled as risk ratios with a fixed effect model where possible. In adults with severe pneumonia, corticosteroids reduced mortality (risk ratio 0.58, 95% CI: 0.40 to 0.84; moderate quality evidence), but in adults with non severe pneumonia there was no clear effect on mortality (risk ratio 0.95, 95% CI: 0.45 to 2.00). Early clinical failure was reduced in both severe and non severe pneumonia (high quality evidence), while hyperglycemia was more common with corticosteroids (risk ratio 1.72, 95% CI: 1.38 to 2.14). The authors concluded that corticosteroids reduced mortality and morbidity in adults with severe pneumonia, and morbidity but not mortality in non severe pneumonia, and that the harms did not seem to outweigh the benefits. Because its searches ran to March 2017, a reader today should check whether the review has been updated and whether later trials change the picture.