Plan your review.
Define your PICO question, draft a structured protocol, and record your eligibility criteria before you begin.
Systematic reviews.
In one place.
Search the literature, screen studies and analyse your findings. A shared workspace for your team, with AI assistance throughout the review.
Search your chosen databases
- PubMed
- Europe PMC
- OpenAlex
- CrossRef
- ClinicalTrials.gov
- Semantic Scholar
- DOAJ
Keep your review organised
Store your protocol, studies, screening decisions and analysis in one project.
Your team checks AI suggestions against the original papers and approves the data, assessments and report.
Search and organise.
Search connected databases and bring records together. Identify duplicates using reference identifiers and title matching.
Screen with AI support.
Check screening suggestions against your criteria and the original record. Resolve disagreements and record your decision.
Extract and assess.
Check extracted values against the original papers and review risk of bias for each result.
Analyse your findings.
Analyse compatible, verified results. Inspect forest plots and heterogeneity, and review evidence certainty for each outcome.
Prepare your report.
Generate PRISMA flow diagrams and report drafts. Export review data, references, and figures for further review.
One review, step by step
Demo · Simulated data
P
Population
Adults with type 2 diabetes
I
Intervention
GLP-1 receptor agonists
C
Comparator
Placebo
O
Outcome
HbA1c change at 24 weeks
Research question
GLP-1 receptor agonists versus placebo for HbA1c change in adults with type 2 diabetes
1 / 5
Systematic review guides
All guidesSystematic Review vs. Meta-Analysis: What's the Difference?
A systematic review is a process; a meta-analysis is an optional statistical step inside one. Learn when to pool, when not to, and how to read forest plots.
Read guideHow to Read a Forest Plot: A Practical Guide
Learn how to read a forest plot: squares, whiskers, the pooled diamond, effect measures, weights, heterogeneity, prediction intervals, and common misreadings.
Read guideAI-Assisted Abstract Screening: How It Works and When to Trust It
How AI-assisted abstract screening works in systematic reviews, where it fails, and how to combine it with human adjudication and transparent reporting.
Read guideQuestions about AutoEvidence
What is AutoEvidence?
AutoEvidence is a workspace for systematic reviews and meta-analyses. It brings together your research question, search, screening, data extraction, analysis, and reporting, with AI assistance and researcher oversight.
What does the AI do, and what do I review?
AI can assist with protocol drafts, abstract screening, data extraction, and assessments. You can compare screening decisions and review disagreements. AI output still needs your assessment, and you remain responsible for the review's scientific decisions.
Which databases can I search?
The available sources are PubMed, Europe PMC, OpenAlex, CrossRef, ClinicalTrials.gov, Semantic Scholar, and DOAJ. PubMed is the base search, and you can select additional sources. Coverage and availability depend on each source.
How does AutoEvidence support PRISMA reporting?
AutoEvidence provides structured protocol drafts, tracks screening decisions and exclusion reasons, and generates PRISMA 2020 flow diagrams and report drafts. Review completeness and check the content against your methods and original sources before using it in a manuscript.
Can I run a meta-analysis in the platform?
Yes. The platform includes fixed-effects and random-effects models, forest plots, funnel plots, and heterogeneity statistics. Pooling requires at least two compatible studies with verified, approved inputs. Your team decides whether the studies and methods are suitable for the question.
What can I export?
Download review data as JSON or CSV, references as RIS or BibTeX, and diagrams as SVG. These exports support further checking and work in your reference manager or manuscript.
Have another question? Get in touch
Start your next review
Bring your research question.
Set up a project and draft your protocol.