Free to Use

NNT Calculator

Calculate Number Needed to Treat, Absolute Risk Reduction, and Relative Risk Reduction. Essential for evidence-based clinical decision making and research analysis.

Event rate in the control/untreated group (0–100%)
Event rate in the treatment/intervention group (0–100%)
Optional: calculate confidence interval for NNT

Real-World NNT Examples

💊 Statin Therapy for Heart Attack Prevention

A clinical trial evaluates statin use for secondary prevention. The control group has a 25% event rate (heart attack) over 5 years. The treatment group has a 16% event rate.

CER: 25% | TER: 16%

ARR: 25% − 16% = 9%

RRR: 9% / 25% × 100 = 36%

NNT: 1 / 0.09 = 12 (rounded up from 11.1)

Number needed to treat: 12 patients for 5 years to prevent one heart attack

This means treating 12 patients with statins for 5 years prevents one major cardiovascular event compared to no treatment.

✅ Vaccine Efficacy Trial

A vaccine trial reports a control group infection rate of 7% and a vaccinated group infection rate of 0.7%.

CER: 7% | TER: 0.7%

ARR: 7% − 0.7% = 6.3%

RRR: 6.3% / 7% × 100 = 90%

NNT: 1 / 0.063 = 16 (rounded up from 15.9)

Number needed to vaccinate: 16 people to prevent one infection

A high RRR of 90% combined with an NNT of 16 indicates excellent vaccine efficacy.

⚠️ Treatment Showing Harm (NNH)

A treatment for a chronic condition shows a control group adverse event rate of 2% but the treatment group has an adverse event rate of 8%.

CER: 2% | TER: 8%

Absolute Risk Increase (ARI): 8% − 2% = 6%

NNH: 1 / 0.06 = 17 (rounded up from 16.7)

Number Needed to Harm: 17 patients treated leads to one additional adverse event

When TER exceeds CER, the calculator automatically reports NNH instead of NNT, highlighting potential treatment risks.

Understanding NNT Calculations

The Number Needed to Treat (NNT) is an epidemiological measure that indicates how many patients need to receive a treatment to prevent one additional adverse outcome. It is a clinically intuitive way to express the effectiveness of an intervention, derived from Absolute Risk Reduction (ARR).

ARR = CER − TER
Absolute Risk Reduction: the difference in event rates between control and treatment groups
RRR = (ARR / CER) × 100%
Relative Risk Reduction: the proportional reduction in risk, expressed as a percentage
NNT = 1 / ARR
Number Needed to Treat: inverse of ARR, rounded up to the nearest whole number

How to Calculate Step by Step

1
Identify CER: Obtain the event rate in the control (untreated) group from the clinical trial or study data
2
Identify TER: Obtain the event rate in the treatment (intervention) group
3
Calculate ARR: Subtract TER from CER (ARR = CER − TER). A positive value means the treatment reduces risk
4
Calculate RRR: Divide ARR by CER and multiply by 100 to get the percentage reduction
5
Calculate NNT: Take the reciprocal of ARR (as a decimal) and round up. If ARR = 0, NNT is infinite

NNT vs NNH

Number Needed to Treat (NNT) is used when the treatment reduces the risk of an adverse event (TER < CER). When the treatment actually increases the risk (TER > CER), the calculation becomes the Number Needed to Harm (NNH), indicating how many patients need to be treated for one additional harmful event. This calculator automatically detects which measure is appropriate based on your input values.

Confidence Intervals for NNT

The confidence interval for NNT is derived from the confidence interval of the ARR. First, the standard error of the ARR is calculated using the formula SE(ARR) = √[CER(1−CER)/n₁ + TER(1−TER)/n₂], where n₁ and n₂ are the group sizes. The 95% CI for ARR is then ARR ± 1.96 × SE(ARR). The CI for NNT is the inverse of each bound of the ARR CI. Note: For this calculator, providing group sizes enables more accurate CI calculation.

🩺
NNT & NNH
Automatically determines whether to calculate Number Needed to Treat or Number Needed to Harm based on your input data.
📊
ARR & RRR
Absolute and Relative Risk Reduction computed instantly with clear step-by-step breakdown of every calculation.
🔬
Confidence Intervals
Optional confidence interval calculation for NNT at 90%, 95%, or 99% confidence levels based on group sample sizes.
📖
Evidence-Based
Based on standard epidemiological formulas used in clinical research, systematic reviews, and medical literature.

What is Number Needed to Treat (NNT)?

The Number Needed to Treat (NNT) is one of the most clinically meaningful measures in evidence-based medicine. It represents the average number of patients who need to receive a specific treatment to prevent one additional adverse outcome, compared to a control group receiving no treatment or a placebo. A lower NNT indicates a more effective treatment — an NNT of 1 means every patient benefits, while an NNT of 50 means 50 patients must be treated for one to benefit.

NNT is derived directly from the Absolute Risk Reduction (ARR), which is the simple arithmetic difference between the event rate in the control group (CER) and the event rate in the treatment group (TER). For example, if a treatment reduces the risk of stroke from 10% to 5%, the ARR is 5 percentage points, and the NNT is 1/0.05 = 20. This means 20 patients need to be treated to prevent one stroke — a clinically useful benchmark for comparing treatments.

NNT = 1 / (CER − TER)
Where CER and TER are expressed as decimals (e.g., 20% = 0.20)

NNT values vary widely depending on the condition, treatment, and patient population. Preventive treatments in low-risk populations may have NNTs in the hundreds or thousands, while acute interventions in high-risk settings may have NNTs below 10. For instance, statins for secondary prevention of cardiovascular disease have an NNT of approximately 12 over 5 years, while certain stroke treatments can have an NNT as low as 3 when administered within the appropriate time window.

Interpreting NNT in Clinical Practice

Understanding how to interpret NNT values is crucial for clinical decision making. A systematic approach considers several factors beyond the raw NNT number:

⏱️ Time Horizon Matters

NNT is always associated with a specific time period. An NNT of 12 over 5 years is very different from an NNT of 12 over 1 month. Always check the study duration when comparing NNTs across treatments.

🎯 Baseline Risk

The NNT depends heavily on baseline risk. A treatment with a fixed Relative Risk Reduction will have a much lower (better) NNT in high-risk patients compared to low-risk patients. This is why NNT varies across patient subgroups.

⚖️ Benefit vs. Harm

Always compare NNT with NNH (Number Needed to Harm) for adverse events. A treatment with NNT=20 and NNH=10 means harm is more common than benefit. The NNT/NNH ratio helps weigh risks against benefits.

🔍 Confidence Intervals

Always consider the confidence interval around the NNT. A wide interval indicates imprecision in the estimate. If the 95% CI crosses infinity (includes both positive and negative values), the result is not statistically significant.

NNT, ARR, and RRR — Understanding the Differences

While NNT is a practical clinical measure, it is derived from more fundamental epidemiological metrics. Understanding the relationship between Absolute Risk Reduction (ARR), Relative Risk Reduction (RRR), and NNT helps avoid common misinterpretations:

Absolute Risk Reduction (ARR) is the simple difference in event rates between control and treatment groups. It reflects the actual magnitude of benefit in absolute terms. An ARR of 2% means the treatment reduces the absolute risk by 2 percentage points. This is mathematically related to NNT (NNT = 1/ARR).

Relative Risk Reduction (RRR) is the proportional reduction in risk, expressed as a percentage of the control group rate. A treatment that reduces events from 10% to 5% has an RRR of 50%, while the same treatment in a lower-risk population (2% to 1%) also has an RRR of 50%. The RRR exaggerates the benefit when baseline risk is low — which is why both ARR and NNT provide a more clinically grounded perspective.

For example, a treatment might be reported as reducing the risk of a rare complication by 50% (RRR = 50%), which sounds impressive. But if the control event rate is only 0.2%, the ARR is just 0.1% and the NNT is 1,000. Understanding both the relative and absolute measures gives a more complete picture of treatment effectiveness.

Frequently Asked Questions

What is a good NNT value?
There is no universal threshold for a "good" NNT — it depends on the condition, treatment, and clinical context. Generally, an NNT below 10 is considered highly effective (e.g., antibiotics for H. pylori eradication, NNT ≈ 2). An NNT of 10–50 is moderate (e.g., statins for secondary prevention, NNT ≈ 12). An NNT above 50 may still be worthwhile for serious conditions if the treatment is safe and inexpensive (e.g., aspirin for primary prevention). The ideal NNT must always be weighed against the NNH, cost, and patient preferences.
How is NNT different from Relative Risk Reduction?
Relative Risk Reduction (RRR) tells you the proportional reduction in risk (e.g., "the treatment reduces risk by 50%"), which can sound impressive even when the absolute benefit is small. NNT tells you how many patients you need to treat to help one person, which is more clinically intuitive. For example, a treatment that reduces a rare event from 0.2% to 0.1% has an RRR of 50% but an NNT of 1,000. RRR is useful for understanding the biological effect of a treatment, while NNT is better for clinical decision making and resource allocation. Both measures are complementary and should be reported together.
What does Number Needed to Harm (NNH) mean?
Number Needed to Harm (NNH) is the counterpart to NNT. It indicates how many patients need to receive a treatment to cause one additional adverse event compared to the control group. It is calculated the same way as NNT (1 / absolute risk increase), but applied when the treatment group has a higher event rate than the control group. For example, if a drug causes gastrointestinal bleeding in 3% of treated patients versus 1% of controls, the NNH is 1 / 0.02 = 50. A lower NNH means greater harm. Clinicians should compare NNT with NNH to assess whether the benefits of a treatment outweigh its risks.
Can NNT be negative?
No, NNT itself is always a positive number (or infinite). When a treatment increases the risk of an event (TER > CER), the Absolute Risk Reduction becomes negative. In this case, the calculation yields a Number Needed to Harm (NNH) instead of NNT. Our calculator automatically detects this and displays NNH with appropriate labeling. When ARR = 0 (TER = CER), the NNT is infinite, meaning the treatment has no effect on the outcome — no matter how many patients are treated, you will not prevent any additional events.
How do I calculate the confidence interval for NNT?
The confidence interval for NNT is derived from the confidence interval of the ARR. First, calculate the standard error of the ARR: SE(ARR) = √[(CER × (1−CER)) / n₁ + (TER × (1−TER)) / n₂], where n₁ and n₂ are the number of participants in the control and treatment groups, respectively. Then, the CI for ARR = ARR ± Z × SE(ARR), where Z is 1.96 for 95% CI, 1.645 for 90% CI, or 2.576 for 99% CI. Finally, the CI for NNT is the inverse of each bound of the ARR CI. Note that when the ARR CI includes zero, the NNT CI includes infinity, indicating a non-significant result.
What is the difference between NNT and number needed to screen?
While both are "number needed" measures, they apply to different contexts. NNT (Number Needed to Treat) applies to therapeutic interventions — how many patients need a treatment to prevent one event. Number Needed to Screen (NNS) applies to screening programs — how many people need to be screened to detect one case of disease or prevent one adverse outcome. NNS is typically much larger than NNT because screening involves testing a general population with a relatively low disease prevalence. Both are calculated similarly (1 / absolute risk difference) but answer different clinical questions.

⚠️ Important Medical Disclaimer: This NNT Calculator is for informational and educational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. NNT values from clinical trials should be interpreted in the context of the specific study population, duration, and methodology. Always consult qualified healthcare professionals when making clinical decisions. Do not change medications or treatment plans based solely on calculator results.