Contents
Academy · Research Data, Analytics & AI · Number needed to treat
Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit
A 50% relative risk reduction may appear substantial until the baseline risk to which it applies is known. The number needed to treat makes that baseline risk explicit.
Number needed to treat (NNT) is the number of patients who would need to receive a treatment, instead of the comparator, for one additional patient to benefit (or avoid one additional adverse outcome). It is calculated as the inverse of the absolute risk reduction, and it translates a relative statistic into a clinically interpretable one.
Learning objectives
- How NNT is calculated from absolute, not relative, risk reduction.
- Why a relative risk reduction can look identical for a treatment with very different real-world impact.
- Why NNT needs a stated time frame and comparator to be interpretable.
Common misconceptions
- “A relative risk reduction tells you how useful a treatment is.” Cook and Sackett show that the same relative risk reduction can correspond to very different absolute benefits depending on the baseline risk of the population: a 50% relative reduction applied to a 40% baseline risk prevents far more events, and implies a far lower NNT, than the same 50% relative reduction applied to a 2% baseline risk.
- “A lower NNT always indicates a better treatment.” NNT must be compared with the number needed to harm (NNH, the same calculation applied to an adverse effect) and interpreted over the specific time frame and population for which it was calculated; a low NNT for benefit accompanied by a low NNH for serious harm does not clearly favour the treatment.
Calculation of the number needed to treat from absolute risk reduction
Cook and Sackett define NNT as the reciprocal of the absolute risk reduction (the difference between the event rate in the control group and the event rate in the treatment group): NNT = 1 / (control event rate − treatment event rate)1. An NNT of 20 means that, on average, 20 patients need to receive the treatment instead of the comparator for one additional patient to benefit, over the time period measured in the underlying trial.
Limitations of relative risk reduction as a sole measure
Relative risk reduction is calculated independently of baseline risk, which is precisely why it can obscure clinical impact: the worked illustration by Cook and Sackett shows that identical relative risk reductions, applied to different baseline risks, produce very different absolute risk reductions and correspondingly different NNTs1. A treatment reported only by its relative risk reduction may therefore appear equally effective whether it benefits 1 in 10 patients or 1 in 500.
Information required to interpret the number needed to treat
NNT is meaningful only in relation to the specific comparator, outcome, population and time frame for which it was calculated; the same treatment can have a very different NNT against placebo than against an active comparator, or over one year than over five. Cook and Sackett also recommend reporting NNT with its confidence interval, since, as with any estimate, it carries uncertainty that a single point figure does not convey1.
Key points
Key points
NNT is the inverse of absolute risk reduction, and it is more clinically interpretable than relative risk reduction precisely because it incorporates baseline risk — but it needs a stated comparator, outcome, population and time frame, with a confidence interval, to be properly interpreted.
Practical application
Practical application
- The number needed to treat (NNT) should be reported alongside relative risk reduction, not in place of absolute figures. Both relative and absolute risk reduction should be given, together with the resulting NNT and its confidence interval.
- The comparator, outcome, population and time frame should be stated explicitly alongside any NNT. An NNT without these specifics is not interpretable or comparable to another NNT.
- The number needed to harm (NNH) should be reported alongside NNT where a relevant harm exists. The number needed to harm for a significant adverse effect should be presented alongside the number needed to treat for benefit, so that the balance between them is evident.
Frequently asked questions
Yes, when a treatment increases rather than decreases risk relative to the comparator, in which case the calculation instead yields the number needed to harm (NNH).
Only with caution. Differences in population risk, outcome definition, comparator and follow-up time all affect NNT, so direct comparison across unrelated studies is usually not meaningful.
No. NNT reflects efficacy alone; cost, adverse-effect burden (reflected in NNH) and patient preference all determine whether a lower NNT corresponds to a better clinical choice.
References
- Cook RJ, Sackett DL. The number needed to treat: a clinically useful measure of treatment effect. BMJ. 1995;310(6977):452-454. read 10 October 2026. View on PubMed Central
Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.
How to cite this article
Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.
Pubrica Academy. (2026, October 10). Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit. Pubrica. https://pubrica.com/academy/concepts-definitions/number-needed-to-treat/
Pubrica Academy. “Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit.” Pubrica, 10 Oct. 2026, pubrica.com/academy/concepts-definitions/number-needed-to-treat/.
Pubrica Academy (2026) Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit. Pubrica. Available at: https://pubrica.com/academy/concepts-definitions/number-needed-to-treat/ (Accessed: 15 October 2026).
Pubrica Academy. “Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit.” Pubrica. October 10, 2026. https://pubrica.com/academy/concepts-definitions/number-needed-to-treat/
Pubrica Academy, ‘Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit’ (Pubrica, 10 October 2026) <https://pubrica.com/academy/concepts-definitions/number-needed-to-treat/> accessed 15 October 2026.
Pubrica Academy. Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit. [Internet]. Pubrica; 2026 Oct 10 [cited 2026 Oct 15]. Available from: https://pubrica.com/academy/concepts-definitions/number-needed-to-treat/
Pubrica Academy, “Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit,” Pubrica, Oct. 10, 2026. [Online]. Available: https://pubrica.com/academy/concepts-definitions/number-needed-to-treat/
@misc{number_needed_to_treat_2026,
author = {{Pubrica Academy}},
title = {Number Needed to Treat: Translating Relative Risk Reduction into Absolute Clinical Benefit},
howpublished = {Pubrica},
year = {2026},
month = {oct},
url = {https://pubrica.com/academy/concepts-definitions/number-needed-to-treat/},
note = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.
NNT reporting check
For use when reporting a treatment effect for a binary outcome.
[ ] Absolute risk reduction reported, not only relative risk reduction [ ] NNT calculated and reported with confidence interval [ ] Comparator, outcome, population and time frame stated explicitly [ ] NNH reported alongside NNT where a relevant harm exists
Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.
How Pubrica can help
Pubrica calculates NNT and NNH and places them in their correct clinical context, so that the clinical impact of a treatment effect is clear and properly bounded.