When is relative risk appropriate for an epidemiologic cohort study?

Relative risk is appropriately applied to categorical data derived from an epidemiologic cohort study.

In this context, it measures the strength of an association by comparing the incidence of an event in an exposed group to the incidence of that same event in a baseline (unexposed) group. Because cohort studies follow groups over time to observe the development of outcomes, they allow for the direct calculation of incidence, making relative risk a suitable metric for quantifying the relationship between exposure and outcome.1

A relative risk of 1 indicates no association, while a value greater or less than 1 indicates an association, such as an increased risk or a protective effect, respectively. While relative risk quantifies the strength of the association, it does not test for statistical significance; researchers typically use a confidence interval to determine if the relationship is statistically significant.1