Record imprecision as a GRADE footnote when the confidence interval crosses your importance threshold or the sample and event count fall short of the optimal information size, then state the downgrade level for that outcome. GRADE rates certainty at the outcome level, so the footnote belongs to the specific outcome estimate, not to the review as a whole.
Steps for the footnote
- Set the importance threshold for the outcome before judging the estimate, for example the smallest effect that would change a clinical decision. Mark whether the confidence interval crosses that threshold. If it does, the estimate is compatible with both worthwhile and trivial effects, which is the imprecision signal.
- Compare the sample size and number of events with the optimal information size for that outcome. When the total sample or event count falls short of that size, record the shortfall as a second imprecision reason in the same footnote.
- Write the footnote with three parts: the outcome, the specific reason (interval crosses the threshold, or sample and event count below the optimal information size), and the downgrade level you applied, such as one level or two levels.
Example footnote wording
"Downgraded one level for imprecision: the 95% CI for [outcome] crosses the threshold for a clinically important effect, and the event count is below the optimal information size."
What to keep consistent
Apply the same threshold and information-size reasoning across outcomes in the review so readers can compare footnotes. Cochrane Handbook Chapter 15 covers confidence intervals, exact P values, absolute and relative effects, and certainty judgements, and the GRADE Working Group describes outcome-level certainty rating with explicit criteria. Use those two sources to justify the threshold and the downgrade level you record.