Read the odds ratio against its confidence interval: if the interval spans the null value of 1, the study has not established the direction of the effect, and the width limits how firmly you can describe its magnitude.
Start with the endpoints
An odds ratio compares the odds of an outcome between two groups, and the null value is 1 (no difference in odds). The confidence interval gives a range of values consistent with the data at the chosen confidence level. Check three things in order:
- Does the interval include 1? If it does, the data are compatible with no difference, so a statement that the exposure raises or lowers odds is not supported by this result alone.
- How wide is the interval? A wide interval means the estimate is imprecise. The point estimate may sit far from 1, yet the interval may still extend close to or across the null.
- What do the endpoints mean clinically? Translate the lower and upper bounds into the outcome scale you care about. If the interval runs from a trivial difference to a large one, the study cannot pin down the size of the effect.
Write a cautious interpretation
Cochrane Handbook Chapter 15 covers confidence intervals, exact P values, absolute and relative effects, applicability, and responsible review conclusions. Cochrane Handbook Chapter 10 covers effect measures, heterogeneity, and careful interpretation of pooled results. Use them to keep the wording tied to what the interval shows.
A defensible statement names the estimate, the interval, and the resulting limit on the claim. For example: "The odds ratio was elevated, but the confidence interval included 1, so these data do not establish a direction of effect." Avoid asserting a benefit or harm when the interval crosses the null, and avoid describing the point estimate as the effect size when the interval is wide.
Steps for a review or guideline note
- Record the odds ratio, the confidence interval bounds, and the null value of 1 side by side. Result: a clear picture of whether the interval crosses the null.
- Convert the endpoints into absolute terms for your population if the study reports baseline risk. Result: a sense of whether even the extreme bounds matter clinically.
- Draft the interpretation sentence using the interval, then check that every claim about direction and magnitude is contained within the bounds. Result: a statement the interval can support.
- Flag the result as imprecise when the width leaves direction or size unresolved, and note what further evidence would narrow it. Result: a transparent basis for the review's certainty judgment.
