Absolute risk reduction

In clinical epidemiology and evidence-based medicine, the absolute risk reduction is a measure that compares the frequency of a clinical outcome in group of patients exposed to a factor compared to a control group of patients. This measure should be contrasted with the relative risk reduction.

Most scientific journal articles fail to report absolute measures which may lead to exaggerated perceptions of results.

Calculations

Two-by-two table for a randomized-controlled trial or cohort study

		Outcome
		Present	Absent
	Experimental group	Cell A	Cell B	Total in the experimental group
	Control group	Cell C	Cell D	Total in the control group
		Total with the outcome	Total without the outcome

${\mbox{Experimental event rate EER)}}=\left({\frac {\mbox{ Cell A}}{{\mbox{Total in the }}experimental{\mbox{ group}}}}\right)$

${\mbox{Control event rate (CER)}}=\left({\frac {\mbox{Cell C}}{{\mbox{Total in the }}control{\mbox{ group}}}}\right)$

${\mbox{Absolute risk reduction}}=\left({\mbox{CER - EER}}\right)$

Confidence intervals

The confidence intervals can be calculated using the method of Daly:

${\mbox{Standard error}}={\sqrt {\left(R_{1}\times {\frac {1-R_{1}}{N_{1}}}\right)+\left(R_{2}\times {\frac {1-R_{2}}{N_{2}}}\right)}}$

${\mbox{Confidence intervals}}={\mbox{Absolute risk reduction}}\pm 1.96\times {\mbox{Standard error}}$

See also

• Randomized controlled trial
• Number needed to treat
• Relative risk reduction