What's the difference between relative and absolute risk?
Say 4 in 100 people in a control group get a disease, versus 2 in 100 in a treated group. The relative risk is 2% ÷ 4% = 0.5, so you can say "risk was cut in half." The absolute risk difference is 4% − 2% = 2 percentage points. Both are correct, but "cut in half" and "2 fewer people out of 100" feel very different.
Number needed to treat (NNT)
NNT tells you how many people need treatment for one person to benefit. It's the inverse of the absolute risk reduction, so the example above is 1 ÷ 0.02 = 50. The number of people treated for one extra harm is the number needed to harm (NNH). In a 2019 Cochrane review, the recombinant shingles vaccine slashed risk with a relative risk of 0.08—yet the absolute difference over three years was 3%, for an NNT of 33.
Hazard ratios and odds ratios
- Hazard ratio (HR): compares the rate at which events occur at any point during follow-up; common in survival analyses and long-term trials.
- Odds ratio (OR): compares the odds of an event happening versus not happening; used mainly in case-control studies, and close to the relative risk when events are rare.
Why baseline risk matters
The same relative risk delivers a bigger absolute benefit in higher-risk groups. A 2014 BMJ meta-analysis estimated that the NNT for H. pylori eradication to prevent stomach cancer was 15 in high-risk Chinese men but 245 in low-risk US women. When you read health news, always check the absolute difference alongside the relative risk—and who the study population was. These numbers usually come from randomized controlled trials and meta-analyses.
FAQ
- Why do headlines usually report relative risk?
- Relative numbers sound more dramatic and stay constant across groups, but without the absolute risk you can't tell how much a treatment actually helps you.
Sources reviewed: 2026-10-25
References
- Laupacis, A., Sackett, D. L., & Roberts, R. S. (1988). An assessment of clinically useful measures of the consequences of treatment. New England Journal of Medicine, 318(26), 1728–1733.
- Gigerenzer, G., et al. (2007). Helping doctors and patients make sense of health statistics. Psychological Science in the Public Interest, 8(2), 53–96.
- Ford, A. C., et al. (2014). Helicobacter pylori eradication therapy to prevent gastric cancer in healthy asymptomatic infected individuals. BMJ, 348, g3174.