Run a GRADE evidence profile by judging certainty for each outcome separately, recording the reason for every downgrade, then assembling those outcome-level judgements into a single profile in GRADEpro or an equivalent template. GRADE rates certainty of evidence through explicit criteria applied at the outcome level, so the profile is built outcome by outcome before it is assembled.
- List every outcome in the review and set up one row per outcome in your profile template. Expected result: each outcome has its own judgement slot, so certainty is rated for that outcome rather than for the review as a whole.
- For each outcome, apply the GRADE criteria and record a judgement with a written reason for any downgrade. Expected result: every rating decision has a stated justification tied to the studies that informed that outcome.
- Assemble the outcome judgements into the evidence profile using GRADEpro or an equivalent template, and write footnotes that trace each downgrade to its study-level source. Expected result: a completed profile in which a reader can follow any downgrade from the footnote back to the supporting study.
GRADE is described by the GRADE Working Group as a transparent approach to rating certainty of evidence and strength of recommendations, using explicit criteria and outcome-level judgements. Cochrane's GRADE resources describe it as a systematic approach for rating certainty in systematic reviews and evidence syntheses, and include Summary of Findings table guidance and GRADEpro workflows. Those two sources frame the method and the tooling; the specific criteria you apply and the reasons you record come from your own outcome-level assessment of the included studies.
A practical check before you finish: read each footnote and confirm it names the study-level source behind the downgrade. If a footnote only restates the criterion without pointing to the studies, the trace is incomplete and the profile needs another pass.