Number Needed to Treat Calculator
How many people must be treated for one to benefit — the number that turns a relative risk reduction into something a person can actually weigh.
NNT is the reciprocal of the absolute risk reduction: 1 ÷ (control event rate − treatment event rate).
How the number needed to treat calculator works
NNT is the reciprocal of the absolute risk reduction: 1 ÷ (control event rate − treatment event rate). If a treatment cuts the event rate from 10% to 8%, the absolute reduction is 2 points and the NNT is 50.
That same treatment would be reported as a "20% relative reduction", which sounds far more impressive. The NNT is the honest translation, and if the treatment causes harm the same arithmetic gives the number needed to harm.
Formula: NNT = 1 / (CER − EER)
Worked examples
| Inputs | Number needed to treat | Note |
|---|---|---|
| 10% down to 8% | 50 | NNT = 50, but a 20% relative reduction |
| A large effect | 5 | NNT = 5 |
| A rare outcome halved | 1,000 | NNT = 1,000 for the same 50% relative cut |
FAQFrequently asked questions
What is the number needed to treat?
How many people must receive a treatment for one additional person to benefit. Lower is better.
What is a good NNT?
It depends entirely on the outcome and the harm. An NNT of 5 for pain relief is excellent; an NNT of 500 to prevent a death may still be worth it.
Why is it better than relative risk reduction?
Because it accounts for the baseline. A 50% relative reduction gives an NNT of 4 on a 50% risk and 1,000 on a 0.2% risk — the same headline, wildly different value.
What is the number needed to harm?
The same arithmetic applied to side effects. Weighing NNT against NNH is how a treatment decision is actually made.
Should NNT be rounded up?
Yes, conventionally — you cannot treat a fraction of a person, and rounding up is the conservative direction.
Where these figures come from
- NIST Digital Library of Mathematical Functions — reference definitions for elementary and special functions
- Wolfram MathWorld — definitions and formulas for every topic on this page
- NIST/SEMATECH e-Handbook of Statistical Methods — the statistical formulas (mean, variance, z, confidence intervals, sample size)
- National curriculum in England — Mathematics — the terms and methods taught in UK schools
Last checked: September 2026. Formulas are fixed by mathematics and do not change with tax years or regulations.