Define risk ratio and odds ratio and indicate when each is most appropriate.

• A. Risk ratio is used in cohort studies; odds ratio is used in case-control studies or for rare outcomes
• B. Risk ratio is used in case-control studies; odds ratio is used in cohort studies and for common outcomes
• C. Risk ratio and odds ratio are interchangeable across designs
• D. Odds ratio is used only in cross-sectional studies; risk ratio in cohort studies

Answer: (A)

When comparing how often an outcome occurs in different groups, the way you collect data guides which measure you should use. In a cohort study, you follow people over time and can directly observe the incidence of the outcome in those exposed and unexposed. That makes the risk ratio, which compares the probability (risk) of the outcome between the two groups, the natural and most informative measure. In a case-control study, you start with people who have the outcome and those who don’t and look back to see who was exposed. Because you don’t have information on the true incidence of the outcome, you can’t calculate the risk ratio directly; instead you use the odds of exposure among cases versus controls, yielding the odds ratio. Additionally, when the outcome is rare, the odds ratio closely approximates the risk ratio, which is why the odds ratio is commonly used in analyses or models even outside case-control designs. So the statement that risk ratio is used in cohort studies and odds ratio is used in case-control studies (or when the outcome is rare) is the best characterization.