Number Needed to Treat/Harm - The Core Formulas
-
Number Needed to Treat (NNT): Average number of patients who need to be treated to prevent one additional bad outcome.
- Formula: $NNT = 1 / ARR$
- A low NNT indicates a more effective intervention.
-
Number Needed to Harm (NNH): Average number of people who need to be exposed to a risk factor to cause one additional harmful event.
- Formula: $NNH = 1 / ARI$
- A high NNH indicates a safer intervention.
-
Core Calculations:
- ARR (Absolute Risk Reduction): $CER - EER$
- ARI (Absolute Risk Increase): $EER - CER$
- CER: Control Event Rate
- EER: Experimental Event Rate
⭐ An NNT of 1 means every single patient treated derives benefit. An NNT of 100 means that for every 100 patients treated, only one benefits.

Cost-Effectiveness and NNT - The Economic Equation
- Incremental Cost-Effectiveness Ratio (ICER): The additional cost for one unit of health outcome (e.g., per QALY gained) when comparing interventions.
- Quality-Adjusted Life Year (QALY): A metric combining quantity and quality of life. 1 QALY = 1 year in perfect health.
- The cost to prevent one adverse event is calculated as:
- $Cost_{prevent} = NNT \times Cost_{treatment}$
| Feature | Drug A | Drug B |
|---|---|---|
| NNT | 10 | 20 |
| Cost/Dose | $5 | $2 |
| Cost to Prevent 1 Event | $50 | $40 |
⭐ When comparing treatments, the one with the lower "Cost per event prevented" is more cost-effective, even if its NNT is higher. In the table, Drug B is superior economically.
Clinical Interpretation - Reading the Fine Print
- Confidence Interval (CI) is Key: A wide CI suggests less precision. If the CI for an NNT crosses infinity or includes zero effect, the result is not statistically significant.
- Context-Specific: NNT is not universally applicable. It is specific to the patient population and the study's timeframe.
- Single-Outcome Limitation: NNT/NNH focuses only on one discrete outcome, ignoring potential side effects or ancillary benefits.
- Baseline Risk Influence: In populations with very low baseline risk, the calculated NNT may be very high, making the intervention appear less impactful.
⭐ If the 95% CI for an NNT includes negative values (crossing into the NNH range), one cannot definitively conclude benefit or harm; the result is statistically insignificant.
High-Yield Points - ⚡ Biggest Takeaways
- NNT (Number Needed to Treat) is the number of patients you must treat to prevent one additional adverse outcome.
- A lower NNT indicates a more effective intervention.
- NNH (Number Needed to Harm) is the number treated for one additional harmful outcome to occur.
- A higher NNH signifies a safer treatment profile.
- Cost-effectiveness weighs the total cost against health benefits (e.g., cost per QALY gained).
- The cost to prevent one outcome = NNT × cost of treatment per patient.
Continue reading on Oncourse
Sign up for free to access the full lesson, plus unlimited questions, flashcards, AI-powered notes, and more.
CONTINUE READING — FREEor get the app