Number Needed to Treat (NNT) Calculator
Calculate the number of patients who must be treated to prevent one additional bad outcome, from the control and treatment event rates. A key measure of clinical effectiveness.
Last updated: September 2026
Number Needed to Treat
13 patients
Formula below · 2 sources (pubmed.ncbi.nlm.nih.gov, cebm.ox.ac.uk) · Updated Sep 2026
Compare with similar
About this calculator
The number needed to treat (NNT) is how many patients must get the treatment instead of the control, over the trial's follow-up period, to prevent one extra bad outcome. It is the reciprocal of the absolute risk reduction:
ARR = control event rate − treatment event rate NNT = 1 ÷ ARR (with the rates as proportions, or 100 ÷ ARR in percentage points)
With the defaults, 20% of control patients and 12% of treated patients have the event: ARR = 8 percentage points, 1 ÷ 0.08 = 12.5, rounded up to an NNT of 13. NNT is conventionally rounded up to a whole patient; the last tile shows the unrounded value.
Relative vs absolute
The same treatment has a relative risk reduction of (20 − 12) ÷ 20 = 40%. Relative effects sound large and stay the same across risk levels, but NNT depends on the baseline risk: a 40% reduction in an event with a 2% baseline gives ARR 0.8 points and NNT 125. That is why NNT is useful for judging benefit for a particular patient group.
Harm
If the treatment group has more events, the reciprocal is the number needed to harm (NNH): how many patients must be treated for one extra adverse outcome.
Always quote NNT with its time frame (an NNT of 13 over 5 years is not the same as over 6 months) and, from the trial, its confidence interval.
How to use
Enter the event rate in the control group and in the treatment group, as percentages over the same follow-up. Example 1 (the defaults): 20% vs 12%: ARR 8 points, NNT = 100 ÷ 8 = 12.5, rounded up to 13. Example 2: 30% vs 20%: ARR 10 points, NNT = 10. Example 3: 4% vs 3%: ARR 1 point, NNT = 100; the relative risk reduction is still 25%. Example 4: 20% vs 28%: the treatment group has 8 points more events, so NNH = 13.
Frequently asked questions
What does NNT mean?
An NNT of 13 means that, on average, 13 people must be treated for the trial period for one of them to avoid the outcome who otherwise would have had it. The other 12 either would not have had it anyway or have it despite treatment.
Why is NNT based on absolute rather than relative risk reduction?
Because the benefit for a patient depends on how likely the event was to begin with. A 50% relative reduction of a 1% risk saves 1 in 200; of a 20% risk, 1 in 10.
What is a good NNT?
It depends on how serious the outcome is, the time frame, and the treatment's cost and side effects. An NNT of 50 can be worthwhile for preventing death with a cheap, safe drug, but not for a minor symptom with a risky one.
Why round NNT up?
You cannot treat part of a patient, and rounding up avoids overstating the benefit. 12.5 becomes 13.
What is NNH?
The number needed to harm: 1 ÷ the absolute risk increase of an adverse outcome. If 8% more treated patients get a side effect, NNH = 13.
Related calculators
Sources & references
- Laupacis A, Sackett DL, Roberts RS (1988). An assessment of clinically useful measures of the consequences of treatment. N Engl J Med 318:1728-33
- Cook RJ, Sackett DL (1995). The number needed to treat: a clinically useful measure of treatment effect. BMJ 310:452-4
- Centre for Evidence-Based Medicine (Oxford) — Number needed to treat
Think a result is wrong? Report a problem with this calculator.