Use absolute risk reduction instead of number needed to treat when understanding is the goal

Summarize treatment benefit as an absolute change instead of only as an NNT sentence when readers need to understand how much risk changes. For example, replace “100 people must be treated for one to benefit” with “the chance falls by 10 per 1000 people,” or with a treated-versus-untreated statement such as “from 10% to 5%.”

Absolute risk reduction tells readers the size of the change directly. Number needed to treat asks them to mentally invert a treatment count into an effect size.

Mechanism: Absolute change makes the treatment effect visible in the same units as the underlying risk, so readers can judge magnitude without translating from a reciprocal summary.

Evidence: Compared with NNT, absolute risk reduction was better understood and was perceived as larger, while showing little or no difference in persuasiveness. (Akl et al., 2011)

• User Goal: Explain how much an intervention changes risk.
• Task: Help readers identify which treatment is more effective or estimate risk after treatment.
• Data: One intervention effect summarized from treatment and control risks.
• Chart Setting: The effect is presented as a single numeric summary in a label, caption, table cell, or text callout.
• Audience: Readers interpreting treatment effects.
• Success Criterion: Readers correctly interpret the size of the risk reduction.

Break it when: The communication goal is willingness to adopt the intervention rather than understanding the size of the effect. Why: The review found little or no difference between ARR and NNT for persuasiveness.

Sacrifice: You give up a compact clinician-facing summary some readers may expect. Risk: Small absolute differences can look unimpressive when baseline risk is low. Mitigation: Include the baseline risk or both treated and untreated risks with the absolute change.

Mistake: Use NNT as the only effect summary when readers must judge how much risk changes. Why it fails: NNT does not directly show the magnitude of the risk reduction.

Failure Sign: Readers can repeat the NNT statement but cannot restate the underlying risk change. Quick Check: Look for effect labels that mention only “need to treat” and omit the absolute change. Stronger Test: Ask readers which of two treatments is more effective or what the risk is after treatment; if they do better with an ARR version, switch to ARR.

• Replace the NNT sentence with an absolute reduction statement.
• Add the treated and untreated risks in the same line when space allows.
• Express the absolute change as a percentage-point change or as a frequency reduction.
• Keep NNT out of the primary label when the main task is understanding effect size.