A headline says that an outcome is “three times as likely,” but the study reports an odds ratio of three. Those words may not describe the same result. Odds and probability are related measures with different denominators.
The Cochrane Handbook’s chapter on effect measures explains the distinction. Risk is the proportion experiencing an outcome; odds compare those who experience it with those who do not. When outcomes are common, interpreting an odds ratio as a risk ratio can exaggerate the apparent change.
Work through two imaginary groups
Suppose an outcome occurs in 20 of 100 people in group A and 40 of 100 in group B. These are invented numbers for arithmetic, not a research result.
- Group A’s risk is 20/100, or 20%.
- Group B’s risk is 40/100, or 40%.
- The risk ratio is 40% divided by 20%: 2.
- The odds are 20/80 in A and 40/60 in B.
- The odds ratio is (40/60) divided by (20/80), approximately 2.67.
The same two groups therefore produce a risk ratio of 2 and an odds ratio of about 2.67. Neither calculation is faulty. Calling the second figure “2.67 times the risk” would change its meaning.
Carry the label into the sentence
When reading coverage, find the measure used in the original paper or technical note. Look for the underlying group risks where they are available, the study design, and whether the reported figure is adjusted. Do not attempt to recover an individual person’s risk from a ratio alone.
The BMJ article When can odds ratios mislead? explores why the difference matters. An adjusted association also does not, by itself, establish that one factor caused the outcome.
A careful summary can retain the phrase “higher odds” and explain the baseline rather than replacing it with a dramatic probability claim. The extra few words keep the statistic attached to the question it actually answers.

