What is NNT?
The Number Needed to Treat (NNT) is a fundamental concept in evidence-based medicine that quantifies the effectiveness of a treatment. It represents the number of patients who need to be treated with a specific intervention for one additional patient to benefit (or one additional adverse event to be prevented) compared to a control group.
NNT was introduced in 1988 by Laupacis, Sackett, and Roberts as a way to translate complex statistical measures into a clinically meaningful number. Unlike relative measures (relative risk reduction), NNT accounts for the baseline event rate, providing a more honest picture of treatment benefit.
A lower NNT indicates a more effective treatment. An NNT of 1 would represent a perfect treatment (every treated patient benefits). Real-world NNTs for well-established treatments typically range from 2 to 50.
NNT Formulas
Where:
- CER = Control Event Rate (proportion of events in the control/placebo group)
- EER = Experimental Event Rate (proportion of events in the treatment group)
- ARR = Absolute Risk Reduction (the absolute difference in event rates)
- RRR = Relative Risk Reduction (the proportional reduction relative to baseline risk)
Interpreting NNT
| NNT Value | Interpretation | Example Context |
|---|---|---|
| 1–5 | Excellent effectiveness | Antibiotics for confirmed bacterial infection |
| 5–15 | Good effectiveness | Statins in high-risk cardiovascular patients |
| 15–40 | Moderate effectiveness | Screening programs, preventive treatments |
| 40–100 | Marginal benefit | Some cancer screening in low-risk populations |
| >100 | Minimal benefit | Treatment may not justify cost/risk |
| Negative (NNH) | Treatment causes harm | Treatment group has worse outcomes |
NNT Visual Explanation
Number Needed to Harm (NNH)
When the treatment group has a higher event rate than the control group (EER > CER), the absolute risk reduction becomes negative, indicating absolute risk increase (ARI). In this case, we calculate the Number Needed to Harm (NNH):
NNH = 1 ÷ ARI
NNH represents how many patients need to be treated for one additional patient to be harmed. A higher NNH indicates a safer treatment (more patients can be treated before one experiences harm). When evaluating a treatment, the ideal scenario is a low NNT (effective) and a high NNH (safe), giving a favorable benefit-to-harm ratio.
Clinical vs Statistical Significance
One of the most important applications of NNT is distinguishing between statistical significance and clinical significance:
- Statistical significance (p < 0.05) tells you the result is unlikely due to chance, but says nothing about the magnitude of the effect.
- Clinical significance is determined by the actual size of the treatment effect, which NNT helps quantify.
A study with a very large sample size might find a statistically significant difference (p = 0.001) with an ARR of only 0.5%, giving an NNT of 200. This means you would need to treat 200 patients for just one to benefit — likely not clinically meaningful despite being statistically robust.
Conversely, a treatment with an NNT of 5 means that for every 5 patients treated, 1 additional patient benefits. This is both statistically and clinically significant. NNT provides the context that p-values alone cannot.
NNT Benchmarks for Common Treatments
| Treatment | Condition | NNT |
|---|---|---|
| Aspirin | Prevent 1 death in acute MI | 42 |
| Statins (5 years) | Prevent 1 cardiovascular event (high risk) | 18 |
| ACE inhibitors | Prevent 1 death in heart failure | 26 |
| Antibiotics | Cure 1 case of strep pharyngitis | 4 |
| tPA for stroke | 1 additional good outcome at 3 months | 8 |
| Mammography (10 years) | Prevent 1 breast cancer death (50–69) | 556 |
| Influenza vaccine | Prevent 1 case of influenza | 12 |
Worked Example
A randomized controlled trial compares a new anticoagulant to placebo for stroke prevention in atrial fibrillation. The study enrolled 400 patients (200 per group) over 2 years:
- Control group: 40 of 200 patients had a stroke (CER = 40/200 = 20%)
- Treatment group: 24 of 200 patients had a stroke (EER = 24/200 = 12%)
NNT = 1 / 0.08 = 12.5 ≈ 13
RRR = 8% / 20% × 100 = 40%
Interpretation: You need to treat 13 patients with the new anticoagulant for 2 years to prevent one additional stroke, compared to no treatment. The treatment reduces the relative risk of stroke by 40%. This represents good clinical effectiveness and would likely justify treatment in most patients with atrial fibrillation.
Frequently Asked Questions
What is a "good" NNT?
It depends on the context. For a life-saving treatment with minimal side effects, an NNT of 20–50 may be acceptable. For a treatment with significant side effects, an NNT below 10 might be required to justify use. Always consider NNT alongside NNH (number needed to harm) and the severity of the condition being treated.
Can NNT be less than 1?
No. An NNT of 1 represents a perfect treatment (every patient benefits). NNT values are always ≥ 1 (for beneficial treatments) or reported as NNH when the treatment causes harm. Fractions between 0 and 1 would imply treating fewer than one patient produces a benefit, which is not meaningful.
Why is NNT always rounded up?
NNT is conventionally rounded up to the next whole number because you cannot treat a fraction of a patient. If the calculated NNT is 12.3, you report it as 13 — you need to treat at least 13 complete patients for one to benefit on average.
How does baseline risk affect NNT?
NNT is heavily influenced by baseline (control) event rate. A treatment that reduces relative risk by 50% will have different NNTs depending on baseline risk: with a 40% baseline rate, NNT = 5; with a 4% baseline rate, NNT = 50. This is why treatments may be recommended for high-risk but not low-risk patients.