In cohort studies, risk ratio is preferred over odds ratio when which condition is met?

• A. When the outcome is rare
• B. When incidence can be directly estimated
• C. In case-control designs
• D. When exposure odds are difficult to estimate

Answer: (B)

In cohort studies, you can observe the actual incidence of the outcome in exposed and unexposed groups over time. This lets you calculate the risk (probability) of the outcome directly in each group, and the risk ratio expresses the relative change in that probability in a simple, intuitive way: risk in the exposed divided by risk in the unexposed. This direct estimation makes the risk ratio the clearest measure of association when incidence can be measured.

The odds ratio, while always calculable in many study designs, uses odds rather than probabilities and can be harder to interpret, especially when the outcome is not rare. It tends to overstate the strength of association when the outcome is common, which is why risk ratio is preferred when you can directly estimate incidence.

So, the best-fit condition is: incidence can be directly estimated. If incidence data aren’t available (as in a case-control study), the odds ratio is typically used instead. When the outcome is rare, the two measures approximate each other, but direct incidence estimation still anchors the preference for the risk ratio.