How to calculate the Number Needed to Treat (NNT)?

How to Calculate Number Needed to Treat (NNT)

The Number Needed to Treat (NNT) is calculated by taking the inverse of the Absolute Risk Reduction (ARR): NNT = 1/ARR. 1

Basic Calculation Method

To calculate the NNT, follow these steps:

1. Calculate the Absolute Risk Reduction (ARR):

   • ARR = Control Event Rate (CER) - Experimental Event Rate (EER)
   • CER = Number of events in control group ÷ Total number in control group
   • EER = Number of events in treatment group ÷ Total number in treatment group

2. Calculate NNT:

   • NNT = 1 ÷ ARR
   • Always round up to the next whole number (since you cannot treat a fraction of a patient)

Example Calculation

Using data from the NASCET trial 1:

• Symptomatic patients with 70-99% carotid stenosis
• Control group event rate (medical therapy): 26%
• Treatment group event rate (CEA): 9%
• ARR = 26% - 9% = 17%
• NNT = 1 ÷ 0.17 = 5.88, rounded up to 6

This means you need to treat 6 patients with carotid endarterectomy instead of medical therapy to prevent one ipsilateral stroke.

Important Considerations for NNT Calculation

1. Time Horizon

• Always specify the time period over which the NNT applies 2
• Example: "NNT of 12 over 1 year" means treating 12 patients for 1 year prevents one event

2. Confidence Intervals

• Report confidence intervals for NNT to indicate precision 2
• CI for NNT = 1 ÷ (CI for ARR)
• Note: When ARR confidence interval includes zero, the NNT confidence interval will include infinity

3. Baseline Risk

• NNT varies with baseline risk - always report the control group event rate 2
• Higher baseline risk populations generally have lower NNTs (more efficient treatment)

4. Special Cases

For Time-to-Event Data:

• Use survival analysis methods rather than simple proportions
• Calculate ARR from survival curves at a specific time point

For Meta-Analyses:

• Calculate the combined ARR first, then take the inverse to get NNT
• Do not average individual NNTs (this produces incorrect results) 3

NNT Interpretation

Clinical Significance

• Lower NNT values indicate more effective treatments
• Generally:
  • NNT < 10: Very effective treatment
  • NNT 10-20: Effective treatment
  • NNT > 20: Less effective treatment, but may still be worthwhile depending on outcome severity

NNT to Harm (NNH)

• Similar calculation but for adverse events: NNH = 1 ÷ ARI (Absolute Risk Increase)
• Comparing NNT and NNH helps balance benefits and risks
• Likelihood to be Helped or Harmed (LHH) = NNH ÷ NNT 4

Common Pitfalls to Avoid

1. Not specifying time horizon - NNT without a time frame is meaningless
2. Averaging NNTs from different studies - mathematically incorrect
3. Not considering baseline risk - NNT varies with population risk
4. Applying NNT from one population to a different population with different baseline risk
5. Not reporting confidence intervals - fails to communicate precision of estimate

By following these guidelines, you can accurately calculate and interpret NNT to make informed clinical decisions based on absolute treatment benefits.