PUBLIC LIBRARY / CALCULATOR

NNT and NNH (numbers needed to treat and harm)

Translates a trial or cohort result into how many patients must be treated to prevent (NNT) or cause (NNH) one additional event compared with control.

Calculate with transparency

Check the population, units and version. The result shows the formula or classification; it does not make an automatic treatment decision.

Preparing the form…

Check the method
Enter the data
Check the result

Method · Limits of application

NNT/NNH/Laupacis 1988; Wald ARR 95%; CI inversion Altman 1998; rounding up

NNT or NNH depends on the event, population, baseline risk and follow-up duration; use the same undesirable binary outcome and time horizon in both groups. Do not transfer the value directly to a different duration or population. Censored time-to-event data require survival methods, not simple counts in this interface. A zero risk difference corresponds to infinite NNT; an interval crossing zero requires interpretation that may span benefit and harm. NNT does not guarantee benefit for one person.

Conditions of use

Check the population, units, inclusion and exclusion criteria, and version in the original source. A result alone does not establish a diagnosis, discharge decision or prescription.

Documented parameters

Events in the control group
Total control-group participants
Events in the treated group (intervention)
Total treated-group participants

4/4 reference cases checked. Numerical tests are not clinical validation.

Sources and documentation

Documented formula

Transcribed from the source or a recorded revision; this is not a clinical recommendation.

RC = events/total in control · RT = events/total in treated · RRA = RC − RT · RRR = RRA / RC · RR = RT / RC.

NNT = 1 / RRA, rounded up. If treatment increases the event (negative RRA), the result is NNH = 1 / |RRA|.

95% CI for RRA by the Wald method: RRA ± 1.96 × √[RC(1 − RC)/n_C + RT(1 − RT)/n_T]. The NNT CI is the inverse of the limits (Altman, 1998); if RRA includes zero, the NNT CI extends from benefit through infinity to harm.

References

Identifier: numero-necessario-para-tratar

Original clinical content: Brazilian Portuguese.

Method version

NNT/NNH/Laupacis 1988; Wald ARR 95%; CI inversion Altman 1998; rounding up

numero-necessario-para-tratar@native-2026-10-03+ccc995c68ea4

Source for the active edition

The bibliographic edition identifies the source. By itself, it proves neither full conformity of the formula nor clinical approval.

Synthetic cases checked (4)

Synthetic cases for checking the implementation, with no patient data.

Case 1

reference-001

Inputs

Events in the control group: 20
Total control-group participants: 100
Events in the treated group (intervention): 10
Total treated-group participants: 100

{
  "ec": 20,
  "nc": 100,
  "et": 10,
  "nt": 100
}

Expected result

{
  "arr": 10,
  "rrr": 50,
  "shown": 10,
  "arrLo": 0.2,
  "arrHi": 19.8
}

Actual result

{
  "arr": 10,
  "rrr": 50,
  "nnt": 10,
  "shown": 10,
  "arrLo": 0.2001799999999998,
  "arrHi": 19.79982
}

Case 2

reference-002

Inputs

Events in the control group: 10
Total control-group participants: 100
Events in the treated group (intervention): 15
Total treated-group participants: 100

{
  "ec": 10,
  "nc": 100,
  "et": 15,
  "nt": 100
}

Expected result

{
  "arr": -5,
  "shown": 20
}

Actual result

{
  "arr": -4.999999999999999,
  "rrr": -49.999999999999986,
  "nnt": 20.000000000000004,
  "shown": 20,
  "arrLo": -14.140663579204082,
  "arrHi": 4.1406635792040865
}

Case 3

reference-003

Inputs

Events in the control group: 10
Total control-group participants: 100
Events in the treated group (intervention): 10
Total treated-group participants: 100

{
  "ec": 10,
  "nc": 100,
  "et": 10,
  "nt": 100
}

Expected result

{
  "main": "∞",
  "verdict": "sem diferença"
}

Actual result

{
  "arr": 0,
  "rrr": 0,
  "nnt": null
}

Case 4

reference-004

Inputs

Events in the control group: 120
Total control-group participants: 100
Events in the treated group (intervention): 10
Total treated-group participants: 100

{
  "ec": 120,
  "nc": 100,
  "et": 10,
  "nt": 100
}

Expected result

{
  "error": true
}

Code and reproducibility

This tool’s code on GitHub

Download code and tests

Sources, version and reproduction instructions

The code tests are technical checks. Clinical approval and professional translation review are recorded separately.

Relationship with cancer research

Support for research — context required

Quantifies effect measures from a study. It may support cancer trials when population, comparator, outcome and time horizon correspond to the investigation.

Applications: Clinical trials and methods