PRISMA Checklist

What PRISMA is

PRISMA is the reporting guideline for systematic reviews and meta-analyses. The current version, PRISMA 2020, has a 27-item checklist (several items with sub-items), a separate checklist for abstracts, and the PRISMA flow diagram showing how records were identified, screened and included. Nearly every journal that publishes systematic reviews expects PRISMA, usually with the completed checklist and flow diagram submitted alongside the manuscript.

When it applies

Use PRISMA for any systematic review, with or without a meta-analysis. Extensions cover related designs, including scoping reviews (PRISMA-ScR), network meta-analyses (PRISMA-NMA), reviews of individual participant data (PRISMA-IPD) and review protocols (PRISMA-P).

What the checklist covers

• Title and abstract: identified as a systematic review, with a structured abstract.
• Introduction: rationale and an explicit statement of the objectives or question.
• Methods: eligibility criteria, every information source with the date last searched, the full search strategy for at least one database, the selection and data collection processes, risk of bias assessment, effect measures, synthesis methods, and assessment of reporting bias and certainty of evidence.
• Results: the study selection process (the flow diagram), study characteristics, risk of bias in each study, results of syntheses, and certainty of the evidence.
• Discussion: interpretation, limitations of the evidence and of the review process, and implications.
• Other information: registration and protocol, support, competing interests, and availability of data and code.

What authors most often miss

• The full search strategy, reported only as a list of keywords rather than the exact query as run.
• Registration: no PROSPERO (or other registry) number, or a protocol that does not match the methods reported.
• Flow diagram numbers that do not add up between stages.
• Risk of bias assessed but not used in the synthesis or interpretation.
• Certainty of evidence (for example with GRADE) not assessed, so conclusions read stronger than the evidence supports.

Ophthalmology: eyes, not just studies

When pooling ophthalmology studies, note whether each included study analysed eyes or patients, and how studies that counted both eyes were handled in the synthesis. Mixing per-eye and per-patient data without comment is a common reason for reviewer queries.