According to this study, the absolute risk reduction (ARR) in mortality from a myocardial infarction (MI) after taking a statin medication for 5 years is 4%.
The effectiveness of different therapeutic interventions is frequently reported as relative or absolute risk differences between study groups. Relative comparisons compare the rates of events, such as death or complications, in two study groups, and the differences between groups is usually reported as relative risk (RR), odds ratio, or hazard ratio. Absolute comparisons, however, represent the absolute (that is, total) difference in outcomes between the experimental and control groups. Absolute measures may also be used to calculate the number needed to treat (NNT), an estimate of the number of patients needing to be exposed to an intervention to expect the studied outcome to occur. To perform calculations for diagnostic tests and medical therapeutics, it is useful to create a table (shown).
|
|
Disease
|
|---|
|
|
Yes
|
No
|
|---|
|
Exposed
|
a
|
b
|
|
Not exposed
|
c
|
d
|
The data extracted from this study are shown.
|
|
Death from Myocardial Infarction
|
|---|
|
|
Yes
|
No
|
|---|
|
Statin
|
53
|
1000
|
|
Placebo
|
93
|
1000
|
ARR can then be calculated as follows:
Control event rate (CER) = c/(c + d) = 93/(93 + 1000) = 0.085
Experimental event rate (EER) = a/(a + b) = 53/(53 + 1000) = 0.050
ARR = CER − EER = 0.085 − 0.050 = 0.035
NNT = 1/0.035 = 29
Relative risk reduction (RRR) can be calculated as follows:
RR = EER/CER = 0.050/0.085 = 0.59
RRR = 1 − RR = 1 − 0.59 = 0.41
A disadvantage of relative comparisons is the potential for exaggerated outcomes, especially if the outcomes are uncommon. For example, although statin medications were associated with a 41% RRR in mortality from MI, the ARR was only 4%.