Mathwords logoMathwords

Relative Risk — Definition, Formula & Example

Relative risk is a number that tells you how many times more likely (or less likely) an outcome is for one group than for another. It is also called the risk ratio.

The relative risk (RR) is the ratio of the probability of an event in an exposed group to the probability of the same event in an unexposed group: RR=pE/pURR = p_E / p_U, defined when pU>0p_U > 0. It satisfies RR≥0RR \ge 0. A value of RR=1RR = 1 indicates no association, RR>1RR > 1 indicates higher risk in the exposed group, and RR<1RR < 1 indicates lower risk.

Key Formula

RR=pEpU=a/(a+b)c/(c+d)RR = \frac{p_E}{p_U} = \frac{a/(a+b)}{c/(c+d)}
Where:
  • aa = Number in the exposed group who have the outcome
  • bb = Number in the exposed group who do not have the outcome
  • cc = Number in the unexposed group who have the outcome
  • dd = Number in the unexposed group who do not have the outcome
  • pE,pUp_E, p_U = Risk (proportion with the outcome) in the exposed and unexposed groups

Worked Example

Problem: A study follows 200 drivers who regularly text while driving and 400 drivers who do not. Over one year, 30 of the texting drivers and 20 of the non-texting drivers have a crash. Find the relative risk of a crash for texting drivers.
Step 1: Find the risk in the exposed group (texting drivers).
pE=30200=0.15p_E = \frac{30}{200} = 0.15
Step 2: Find the risk in the unexposed group (non-texting drivers).
pU=20400=0.05p_U = \frac{20}{400} = 0.05
Step 3: Divide the exposed risk by the unexposed risk.
RR=0.150.05=3RR = \frac{0.15}{0.05} = 3
Answer: The relative risk is 3. Texting drivers had 3 times the crash risk of non-texting drivers, which is a 200% increase in risk.

Why It Matters

Relative risk is a standard way medical studies summarize a 2-by-2 table from a cohort study or experiment, such as a vaccine trial. Researchers report a confidence interval for RR, usually built on the log scale. If a 95% interval contains 1, the difference between the groups is not significant at the 5% level.

Common Mistakes

Mistake: Reading RR = 3 as a "300% increase" in risk.
Correction: An RR of 3 means the risk is 3 times as large. The increase is (3 − 1) × 100% = 200%. Subtract 1 before converting to a percent change.

Related Terms

  • Relative Frequency — Proportion of times an event occurs in a group
  • Confidence Interval — Gives a range of plausible values for the true RR
  • Null Hypothesis — Usually states RR equals 1, meaning no association
  • Hypothesis Testing — Decides whether an observed RR differs significantly from 1
  • p-value — Measures how surprising the observed RR is if RR equals 1
  • Standard Error — Measures the sampling variability of the estimated log RR
  • Odds Ratio — Similar comparison of two groups using odds instead of risk