GRADE certainty ratings beside your evidence

Build a GRADE evidence table per outcome: study design sets the starting level, the five downgrade and three upgrade criteria adjust it, and the table exports as CSV.

Rate your outcomes Free core tools. Runs in your browser, nothing to install.

GRADE beside the evidence, not in a separate app

Rate the certainty of evidence for each outcome using the GRADE approach: randomized trials start at high certainty, observational studies at low, and your judgements on the standard criteria move the rating from there. Because the table lives in the same project as your risk of bias assessments and meta-analysis, the evidence behind each judgement is one click away, not in another window.

What the tool covers

Why the placement matters

An imprecision judgement needs the pooled confidence interval in front of you. An inconsistency judgement needs the I² and the forest plot. When GRADE lives in a separate tool, you alt-tab and retype; when it lives beside the analysis, you look and rate.

An honest note

GRADEpro GDT is the established tool for guideline panels and Cochrane-style summary of findings tables, and for that work it is the reference implementation. Covexe suits the systematic review where GRADE is one step among many and you want it next to the numbers it depends on, free.

Frequently asked questions

How does the rating work?

Randomized trials start at high certainty and observational studies at low. You rate the five downgrade criteria and three upgrade criteria per outcome, and the final level follows from your judgements.

Can I export the table?

Yes, as CSV, ready to format into a summary of findings table. The ratings also flow into the project report and manuscript draft.

Is this a GRADEpro alternative?

GRADEpro GDT is the established tool for guideline panels and the right choice there. Covexe suits reviews where you want the ratings beside the meta-analysis and risk of bias they depend on.

Is it free?

Yes. GRADE is part of the free core, and no account is required.