Number needed to treat is defined as:

Number needed to treat is defined as:

• A. The reciprocal of absolute risk reduction
• B. The inverse of relative risk
• C. The difference in event rates
• D. The number of patients needed to treat to cause one adverse event

Answer: (A)

The main idea is that the number needed to treat (NNT) tells us how many patients must be treated to prevent one additional adverse outcome (or achieve one additional beneficial outcome) on average. It is calculated as the reciprocal of the absolute risk reduction (ARR). ARR is the difference between the event rate in the control group and the event rate in the treatment group: ARR = CER − EER. So NNT = 1 / ARR. For example, if the control risk is 10% and the treatment risk is 4%, ARR is 6% (0.06), and NNT is about 16.7, meaning roughly 17 patients need treatment to prevent one event. The other ideas don’t fit because: the inverse of relative risk is not NNT (it’s 1/RR, a risk-ratio-related measure, not a count of patients needed). The difference in event rates is ARR itself, not its reciprocal. Saying “to cause one adverse event” describes a number needed to harm (NNH), not the number needed to treat (NNT). Therefore, the definition matches the reciprocal of absolute risk reduction.

The main idea is that the number needed to treat (NNT) tells us how many patients must be treated to prevent one additional adverse outcome (or achieve one additional beneficial outcome) on average. It is calculated as the reciprocal of the absolute risk reduction (ARR). ARR is the difference between the event rate in the control group and the event rate in the treatment group: ARR = CER − EER. So NNT = 1 / ARR. For example, if the control risk is 10% and the treatment risk is 4%, ARR is 6% (0.06), and NNT is about 16.7, meaning roughly 17 patients need treatment to prevent one event.

The other ideas don’t fit because: the inverse of relative risk is not NNT (it’s 1/RR, a risk-ratio-related measure, not a count of patients needed). The difference in event rates is ARR itself, not its reciprocal. Saying “to cause one adverse event” describes a number needed to harm (NNH), not the number needed to treat (NNT). Therefore, the definition matches the reciprocal of absolute risk reduction.