Number Needed to Treat (NNT)

Definition

The Number Needed to Treat (NNT) is the number of patients you need to treat to prevent one additional bad outcome (death, stroke, etc.). For example, if a drug has an NNT of 5, it means you have to treat 5 people with the drug to prevent one additional bad outcome. More detailed discussion of the nature of the NNT measure can be found in the EBM Note on summarising the effects of therapy in the journal Evidence-Based Medicine 1997;2:103-4.

Calculation

To calculate the NNT, you need to know the Absolute Risk Reduction (ARR); the NNT is the inverse of the ARR:

NNT = 1/ARR

Where ARR = CER (Control Event Rate) - EER (Experimental Event Rate).
NNTs are always rounded up to the nearest whole number.

For a more detailed look at the NNT measure, and an interactive nomogram for converting between ARRs, RRRs and NNTs, see Zapletal E, LeMaitre D, Menard J and Degoulet P, The Number Needed to Treat: a clinically useful nomogram in its proper context, BMJ 1996;312:426-9.

Example

The ARR is therefore the amount by which your therapy reduces the risk of the bad outcome. For example, if your drug reduces the risk of a bad outcome from 50 per cent to 30 per cent, the ARR is:

ARR = CER - EER = 0.5 - 0.3 = 0.2 (20 per cent)

therefore

NNT = 1/ARR = 1/0.2 = 5

The following table is taken from Chapter 3 of the book, How to Practice and Teach Evidence-Based Medicine. It shows an abstract of the diabetes control and complications trial examining the effectiveness of intensive diabetes therapy on the development and progression of neuropathy.

Derivation of clinically useful measures of therapy:
The occurrence of neuropathy Event Rates (diabetic neuropathy) RRR
(CER-EER)/CER
ARR
(CER-EER)
NNT
(1/ARR)
Usual insulin regimen CER Intensive insulin regimen EER
in the actual trial0.0960.028(0.096-0.028)/0.096
= 71%
0.096-0.028
= 0.068
1/0.068
= 14.7 or 15
high hypothetical
case A
0.960.28(0.96-0.28)/0.96
= 71%
0.96-0.28
= 0.68
1/0.68
= 1.47 or 2
low hypothetical
case B
0.00960.0028(0.0096-0.0028)/
0.0096
= 71%
0.0096-0.0028
= 0.0068
1/0.0068
= 147


Converting Odds Ratios to NNTs

The formula for converting ORs to NNTs is:

NNT = (1-(PEER*(1-OR))) / ((1-PEER)*(PEER)*(1-OR))

The formula for converting ORs to NNHs (Numbers Needed to Harm) is:

NNH = ((PEER*(OR-1))+1) / (PEER*(OR-1)*(1-PEER))

This table can be used to convert odds ratios to NNTs:


Deriving NNTs from a treatment's Odds Ratio and
the observed or expected Event Rates of untreated groups or individuals:
CER or PEER Odds Ratios
0.50.550.60.650.70.750.80.85 0.9
1.522.533.544.5510
NNTs for efficacy
NNHs for harm
0.05414652596983104 139209
4322151298763
0.12124273136435473110
23129765442
0.2111314172024304061
1485443332
0.38910121418223046
1165433332
0.4789101215192640
1064433332
0.567891114182538
1065443332
0.7679101316202844
1387655554
0.91215182227344664101
322117161414131311

Page last edited: 14 August 2012