Relative risk

    A relative risk is a ratio of risks between two groups. It shows how many times more or less likely an outcome is with an exposure. However, relative risk hides the baseline risk, so it can look large even when the real-world change is tiny.

    Summary

    What is relative risk?

    Relative risk, also called the risk ratio, compares the probability of an outcome occurring in one group with the probability in another. It is calculated as a simple ratio:

    The risk in the exposed (or intervention) group divided by the risk in the unexposed (or control) group.

    The relative risk indicates how many times more (or less) likely an outcome is with a given exposure, not how many more outcomes will actually happen.[1][2][3]

    • If the relative risk = 1, there is no difference in risk between the exposed/intervention group and the nonexposed/control group.
    • If the relative risk is greater than 1, there is a greater risk of an outcome happening in the exposed/intervention group. A relative risk of 2 is a two-times greater risk, 3 is three-times greater, and so on.
    • If the relative risk is less than 1, there is a lower risk of an outcome happening in the exposed/intervention group. A relative risk of 0.5 is a two-times smaller risk, 0.25 is four-times smaller, and so on.

    How is relative risk different from absolute risk?

    Absolute risk is the plain probability of an event happening in a group, whereas 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. Let’s look at 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 (95% confidence interval 1.2 to 3.3), or a two-times greater risk of feeling scared when exposed to spiders compared to when not exposed to spiders. That 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 indicates that, even if the association between spiders and feeling scared is real, the clinical significance is likely very small.

    While reporting the relative risk can be useful because it summarizes two numbers — the risk in the exposed group and the risk in the control group — into a single ratio, it can also give the impression that something big is happening when, in fact, it is not. This is because the relative risk hides information about the base rate (i.e., the absolute risk in the unexposed/control group).[1][2][3] When a study reports the relative risk or risk ratio, always ask yourself, “What’s the absolute risk?” Knowing the absolute risk can also help make individual decisions about using an intervention.[4] For this reason, relative risk and absolute risk are best presented together.

    References

    1. ^Noordzij M, van Diepen M, Caskey FC, Jager KJRelative risk versus absolute risk: one cannot be interpreted without the other.Nephrol Dial Transplant.(2017 Apr 1)
    2. ^Monaghan TF, Rahman SN, Agudelo CW, Wein AJ, Lazar JM, Everaert K, Dmochowski RRFoundational Statistical Principles in Medical Research: A Tutorial on Odds Ratios, Relative Risk, Absolute Risk, and Number Needed to Treat.Int J Environ Res Public Health.(2021 May 25)
    3. ^Schwartz LM, Woloshin S, Dvorin EL, Welch HGRatio measures in leading medical journals: structured review of accessibility of underlying absolute risks.BMJ.(2006 Dec 16)
    4. ^Fagerlin A, Zikmund-Fisher BJ, Ubel PAHelping patients decide: ten steps to better risk communication.J Natl Cancer Inst.(2011 Oct 5)
    Relative risk - Examine - Examine