What is absolute risk?

Absolute risk is the plain probability of an event happening in a group over a time period.^[1][2][3]

When groups are compared, the “absolute risk difference” (also called “risk difference”) is calculated as:

The absolute risk in the exposed (or intervention) group minus the absolute risk in the unexposed (or control) group.

The absolute risk difference is useful for assessing clinical significance and making clinical decisions because it indicates how many extra/fewer events actually occur due to an exposure or intervention.^[1][2][3] The absolute risk difference is also important for understanding a concept called the number needed to treat (NNT), which refers to the number of people that need to be treated with a given intervention in order to observe an effect on one person.^[4]

How is absolute risk different from relative risk?

Relative risk is a ratio of two absolute risks: the absolute risk of the exposed (or intervention) group divided by the absolute risk of the unexposed (or control) group. Due to the way they are calculated, absolute and relative risk values provide different information. Here’s an example:

Imagine that an observational study reports an association between exposure to spiders and an increased risk of feeling scared. The study reports the relative risk as 2.0. That indicates a two-times greater risk of feeling scared when exposed to spiders compared to when not exposed to spiders. Sounds scary! However, when the authors look at the absolute risk of feeling scared in the group not exposed to spiders (also called the “base rate”), they notice that the absolute risk increased from only 1 to 2 per 100,000 (from 0.001% to 0.002%). So, the relative risk doubled, while the absolute risk rose by just 1 in 100,000. This means that even if the association between spiders and feeling scared is real, the clinical significance is likely very small.

This is how relative risk can sometimes give the impression that something big is happening when, in fact, it is not. When evaluating the relative risk (or risk ratio) reported in a study, always ask, “What’s the absolute risk?” because the absolute risk includes information about the base rate (i.e., the absolute risk in the unexposed/control group).^[1][2][3] Knowing the absolute risk can also help you make a personal decision about using an intervention or not.^[5] For this reason, relative risk and absolute risk are best presented together.