Absolute Risk Reduction Calculator: Formula and Example

An absolute risk reduction calculator finds the difference between the event rates in a control group and a treatment group, expressed as a percentage-point reduction. Enter both event rates as percentages to get the absolute risk reduction, along with the relative risk reduction and number needed to treat when the reduction is positive.

Formula

Absolute risk reduction (ARR) = control event rate − treatment event rate. Here, c is the percentage of people in the control group who experienced the event, and t is the corresponding percentage in the treatment group. Both rates must use the same scale, such as percentages from 0 to 100. The result is in percentage points.

Relative risk reduction (RRR) = (c − t) ÷ c × 100. This expresses the reduction relative to the control event rate, as a percentage. Number needed to treat (NNT) = 100 ÷ ARR when ARR is entered in percentage points. NNT conventionally represents people treated to prevent one additional event over the study’s stated follow-up period. The calculator displays the raw result to two decimal places; practical NNTs are generally reported as whole people, with the rounding convention stated.

Worked Example

Using the block’s default values—control event rate 20% and treatment event rate 15%:

1. ARR = 20% − 15% = 5.00 percentage points.
2. RRR = (20 − 15) ÷ 20 × 100 = 25.00%.
3. NNT = 100 ÷ (20 − 15) = 20.00 people.

In this example, the event occurred in 5 fewer people per 100 in the treatment group than in the control group. The NNT is 20 for the outcome and follow-up period represented by these rates; it is not a timeless or universal property of a treatment.

What Absolute Risk Reduction Means

ARR describes the observed difference in event probability between two groups. It is called “absolute” because it reports the difference directly, rather than dividing that difference by the control group’s risk. A 5-percentage-point reduction means a change from, for example, 20% to 15%. It does not mean a 5% relative reduction.

ARR is useful because it helps put a relative effect into context. If a treatment lowers a risk from 2% to 1%, the ARR is 1 percentage point and the RRR is 50%. If it lowers risk from 40% to 20%, the ARR is 20 percentage points and the RRR is also 50%. The same relative reduction can therefore correspond to very different absolute differences and NNTs.

In a study, these rates are usually calculated as the number of participants with the defined event divided by the number of participants in each group, then multiplied by 100. The event could be a benefit, such as recovery, or an unwanted outcome. Always identify what event the rate describes: a lower rate is beneficial only when the event itself is undesirable.

How to Use and Interpret the Calculator

• Enter the control-group event rate and treatment-group event rate on the same percentage scale. For example, type 12 for 12%, not 0.12.
• Check that both figures describe the same outcome, population, and follow-up duration.
• Read ARR as a percentage-point difference, RRR as a relative percentage reduction, and NNT as people treated per additional event prevented.
• If the treatment rate is higher than the control rate, ARR will be negative. That indicates an absolute increase in the event, not a reduction; the displayed NNT will also be negative and should not be interpreted as a conventional NNT.

The calculator uses percentages, not raw participant counts. If you have counts, first calculate each group’s event rate separately. For example, 30 events among 200 control participants is 15%, while 20 among 200 treatment participants is 10%; ARR is 5 percentage points. Unequal group sizes are fine when rates are calculated with the correct denominator for each group.

ARR, RRR, and NNT: Important Differences

These measures answer different questions. ARR gives the size of the difference between the two event rates. RRR tells you how large that difference is compared with the control risk. NNT reframes a positive ARR as the number of people who would need treatment, over the specified period, for one fewer event on average.

For the example, saying “risk fell by 25%” refers to the relative reduction. Saying “risk fell by 5 percentage points, from 20% to 15%” gives the absolute reduction and the underlying rates. The second statement is more informative by itself because it shows the baseline risk as well as the change.

NNT is not a guarantee that exactly one person in every group of 20 will benefit. It is an average derived from group-level rates. It also changes when baseline risk, follow-up duration, outcome definition, or population changes. A valid interpretation should preserve those details rather than presenting an NNT alone.

Common Mistakes and Limitations

• Confusing percentage with percentage points: A change from 20% to 15% is a 5-percentage-point ARR and a 25% RRR.
• Mixing formats: Enter both rates as values such as 20 and 15, not one as 0.20 and the other as 15.
• Ignoring follow-up time: A 5-point reduction over one year is not automatically comparable with a 5-point reduction over ten years.
• Rounding too early: Calculate rates from the original counts when available, keep precision through the calculations, and round only for display.
• Assuming the difference proves a cause: A calculator only subtracts the entered rates. It does not assess study design, uncertainty, bias, or whether the groups were comparable.

Study estimates have uncertainty. Confidence intervals, sample size, and the study’s methods matter when judging whether an observed difference is precise or likely to apply elsewhere. This simple calculator does not compute confidence intervals or account for competing risks, changing follow-up, or differences between participants. It is a calculation aid, not a substitute for careful interpretation of the source data.

Frequently Asked Questions

How do you calculate absolute risk reduction?
Subtract the treatment-group event rate from the control-group event rate. If rates are 18% and 12%, respectively, ARR is 6 percentage points.

Is absolute risk reduction the same as relative risk reduction?
No. ARR is the direct difference between rates. RRR divides that difference by the control rate, then expresses the result as a percentage.

Can absolute risk reduction be negative?
Yes. If the treatment-group event rate is higher, control rate minus treatment rate is negative. This indicates an absolute increase in the event in the treatment group.

How do I calculate NNT from ARR?
When ARR is a percentage-point value, divide 100 by ARR. For a 5-point ARR, NNT is 20. NNT applies to the outcome and follow-up period used to calculate the rates.

Should I enter percentages or decimals?
Enter percentage numbers on a 0-to-100 scale: enter 15 for 15%, not 0.15. Use the same format for both groups.

Related

• Bore Hole Volume Calculator and Formula
• Arc Length Calculator: Formula and Examples
• Angle of Elevation and Depression Calculator: How It Works

in c | Control event rate | % | 20 | 0
in t | Treatment event rate | % | 15 | 0
out arr | Absolute risk reduction | percentage points | c - t | 2
out rrr | Relative risk reduction | % | (c - t) / c * 100 | 2
out nnt | Number needed to treat | people | 100 / (c - t) | 2