A quit-vaping review says low-certainty evidence: read the rating with the result
Distinguish a GRADE certainty assessment from an effect size, a personal probability and a verdict on every outcome in a review.
No Vape · Updated

Locate the exact row
A review may have a Summary of findings table with separate rows for quitting, other use outcomes and reported harms. Find the row that matches the claim you are checking. Do not move a rating from one row to another just because they concern the same program.
Cochrane’s handbook describes four certainty categories: high, moderate, low and very low. It also describes considerations involving study bias, inconsistency, indirectness, imprecision and publication bias. These are part of an evidence assessment, not a scale of how committed a participant was. A reader should retain the authors’ stated rating rather than invent a new one from the headline.
Cochrane: Cochrane Handbook, Chapter 14: Grading the certainty of evidence
Read the explanation before substituting your own story
Look for footnotes or comments explaining the rating. Is the concern about limited information, differences between studies, or whether the evidence directly answers the question? More than one concern may apply. Keep the wording tied to the actual row rather than assuming low certainty always means the same problem.
For a hypothetical example, a review describes a quitting estimate as low certainty because of imprecision and study limitations. A separate row about another outcome has a different rating. Your note should preserve those two assessments. The example is a reading task, not a claim about a named treatment or a current review’s findings.
Separate certainty from the size of an estimate
A large-looking estimate and a low-certainty rating can appear together because they address different questions. The estimate describes the reported comparison; the rating addresses confidence in the evidence behind it. Conversely, a higher-certainty rating does not turn a small estimated difference into a large one.
Do not translate “low” into “half as likely to work” or divide a percentage by an imagined score. GRADE categories do not supply that calculation. Also keep the confidence interval visible where reported. An interval concerns precision under an analysis; certainty can include concerns beyond precision. Neither number assigns you an individual outcome.
Write a bounded sentence you can bring to a conversation
Try a three-part note: “This comparison reports this outcome; the review rates the evidence at this level; its stated reasons are these.” Add the table or section link. If the explanation is missing from the material you can access, record that gap instead of treating a social-media summary as the full appraisal.
A certainty rating helps you describe what you read. It does not by itself select care, establish that an intervention never helps, or certify the opposite of an uncertain result. If an individual health decision depends on the review, bring the source and your question to a qualified professional. You can continue ordinary planning without forcing every uncertain paper into a yes-or-no recommendation.
Sources and scope
Sources inform the guidance. The worksheets and examples are No Vape editorial suggestions, not tested treatments.
Written with AI assistance and checked against the cited sources. No clinical review is claimed. This guide covers practical planning, not individual medical treatment. How we research and write