Compare the conclusion's wording with the outcome-level GRADE certainty, then revise the conclusion so its strength matches the rated evidence. GRADE rates certainty of evidence and strength of recommendations using explicit criteria and outcome-level judgements, so the mismatch usually sits between a hedged certainty rating and a conclusion that reads as firm.
- List each outcome in the review and record its GRADE certainty rating (for example high, moderate, low, or very low) alongside the conclusion sentence that relies on it. The expected result is a side-by-side table showing which conclusions rest on low-certainty outcomes.
- Rewrite each conclusion sentence so its verbs and qualifiers match the rated certainty, and note the outcome it addresses. The expected result is wording that a reader can trace back to a specific rating.
- Record the alignment check and the revised sentences in the review's documentation, so reviewers and editors can see what changed and why.
Cochrane Handbook Chapter 15 addresses responsible review conclusions and covers confidence intervals, exact P values, absolute and relative effects, applicability, indirectness, and certainty. Use that guidance to check whether a conclusion overstates an effect or ignores indirectness when the rating is low.
PRISMA 2020 supports transparent reporting of why a review was done, what methods were used, and what studies were found and included. Use its checklist to keep the revised conclusion consistent with the reported methods and included studies.
When the certainty rating and the conclusion still conflict after revision, describe the result as unresolved and check the outcome-level judgement against the GRADE criteria before finalizing the wording.