Odds Ratio

Odds are a different way to express chance: the number with the event divided by the number without it, not divided by the total. If 20 of 100 people have the event, the risk is 0.20 but the odds are 20 to 80, which equals 0.25. The odds ratio (OR) divides the odds in one group by the odds in another. Like the risk ratio, an OR of 1 means no effect, below 1 is protective, and above 1 is harmful. Why have a second ratio at all? Because the OR can be calculated even when true risk cannot be measured. This matters most in case-control studies, where investigators choose how many cases and controls to enroll, so risks are not meaningful, and in logistic regression, whose natural output is an odds ratio. These two situations are extremely common, which is why odds ratios fill the literature. The key nuance is the rare-disease assumption. The OR approximates the RR only when the outcome is rare, roughly under 10%. When the outcome is common, the OR sits further from 1 than the RR and therefore exaggerates the effect. So an odds ratio of 2.5 for a common outcome does not mean the risk is 2.5 times higher. Readers routinely misinterpret odds ratios as if they were risk ratios, overstating effects, especially for frequent outcomes. In practice, report the OR with its confidence interval (an interval including 1 is non-significant), state the study design so readers know why an OR was used, and when the outcome is common, be careful about translating the OR into everyday risk language. If a true risk is available and the outcome is common, a risk ratio or an adjusted risk difference communicates the effect more honestly to clinicians and patients.