Your shopping cart is empty.

Numbers needed to treat (NNTs) - calculation, interpretation, advantages and disadvantages

Epidemiology: Number needed to treat

The number needed to treat (NNT) can be thought of as the number of patients that need to be treated in order for one to benefit. It provides an attractive means of summarising the results of a clinical trial in a single figure, because the meaning of a sentence such as ‘20 patients need to be treated to avoid one additional death over a five year period’ is easily understood by both doctors and patients.1

Calculation

The NNT is the inverse of the absolute risk reduction – the difference between the proportion or rate of events in the active treatment intervention group (Pa) and the proportion of events in the control group (Pc):
Number needed to treat =1/(Pa-Pc)

The worked example below is taken from the examiners comments from the June 2001 sitting of the MFPH Part A Examination.

Example:
The following 2×2 contingency table was obtained in a randomised controlled trial:

There are also methods available for calculating NNT using odds ratios and relative risk reduction.2,3

Alongside the NNT, the control treatment and intervention should be outlined, as well as the dose and duration of the intervention, the outcome, and the period over which observations were made.

Interpretation

The ideal NNT would be 1, where all the patients in the treatment group have improved, but no-one has in the control arm. In theory, the higher the NNT, the less effective is treatment, because more people need to receive the treatment to see a benefit in one. However, the value of an NNT should be interpreted in light of the clinical contact. For example, an NNT of between 2-5 would normally indicate an effective therapy, such as a pain killer for acute pain. On the other hand, an NNT of 1 might be seen when treating a sensitive bacterial infection with antibiotics, whilst an NNT of 40+ might be useful in other situations, such as using aspirin after a heart attack.

Treatments
NNT can be used to help us choose between two treatment options. If, for example, the NNT for drug A is lower than that for drug B, it suggests that it may be more effective and – all other things being equal – choosing A rather than B would make sense.

If the drug or intervention is harmful (Pa – Pc), and thus the NNT, will be negative. This is sometimes referred to as the ‘Number Needed to Harm’ (NNTH). This can also be used to describe adverse effects, for example as a result of the treatment under study.

Preventive measures
It is important to distinguish between treatments and preventative measures. In trials of prophylactic treatments, ideally fewer events will hopefully occur in the treatment arm versus the control group, so (Pa – Pc) and the NNT will be negative.

Alternatively, in the case of preventive measures, the formula can be switched around to provide an NNT with a positive sign e.g. 1/(Pc – Pa)

Advantages

  • Useful summary of trial results
  • Useful to inform decision-making about individual patients and treatment options
  • Relatively easy to calculate

Disadvantages

  • NNT is based on the most probable value in a normally distributed population – it does not take into account an individual patient’s baseline risk
  • The clinical meaning of an NNT is subject to interpretation. For example, an NNT of 100 over five years to avoid one clinical event might be seen by some doctors as a health benefit, whereas others will consider the benefit as only moderate or even slight.

Past papers

NNT is addressed by the following MFPH Part A past paper questions:

  • June 2001 – Paper 1A, Question 1
  • January 2008 – Paper 1A, Question 1
  • June 2008 – Paper 2A, Question 3

References

  1. Cook RJ, Sackett DL. The number needed to treat: a clinically useful measure of treatment effect. BMJ 1995;310:452-4.
  2. http://www.jr2.ox.ac.uk/bandolier/band36/b36-2.html -Accesssed 17/12/08
  3. Chatellier G, ZapletalE, Lemaitre D, Menard J, Degoulet P. The number needed to treat: a clinically useful nomogram in its proper context BMJ 1996; 312:426 - 429

© Helen Barratt 2009