Keep the numerator and denominator visible
Rate = number of events / total number of observations. Use the same event definition and observation period for both groups. These examples are fictional learning data, not results from a treatment or a laboratory study.
Relationships between the calculations
| Measure | Formula |
|---|---|
| Control risk | c / n₀ |
| Comparison risk | e / n₁ |
| Risk difference | e/n₁ − c/n₀ |
| Risk ratio | (e/n₁) / (c/n₀) |
| Relative reduction | 1 − risk ratio |
Why “50% less” is not enough
A reduction from 2% to 1% and a reduction from 20% to 10% are both relative reductions of 50%. Their absolute differences are 1 and 10 percentage points. When a headline gives a relative change, find the baseline risk and observation period.
The limits of the calculation
The tool only compares the rates you enter. It does not calculate confidence intervals, adjust for other factors or analyse causation. If the control group has no events, the risk ratio is undefined; it is not replaced with a number. Differences in how the groups were selected may produce differences in their rates. This simple calculation alone is insufficient for a health decision.
Compare event rates
Select the button to calculate.
Sources and provenance
Source check / web edition: 14 October 2026. Historical documents retain their own original dates.