Absolute Risk Reduction vs. Relative Risk ReductionI'm starting a new thread but following up on bfw's post about Absolute Risk Reduction (ARR) vs. Relative Risk Reduction (RRR).
Here's a great article on ARR vs. RRR.
In regards to the Resverlogix pooled ASSURE/SUSTAIN post-hoc MACE analysis quoted from the July Annual Informaton Form, "Results from an additional high risk patient population, those with diabetes mellitus were reported September 2, 2014. In this analysis, patients treated with RVX-208 (n=127) had less cumulative events of 4.71% vs. 20.31% (p=0.008) in the placebo treated group (n=65). These early pooled analyses on MACE are also being examined in other high risk populations within both SUSTAIN and ASSURE patients such as CKD patients. A similar trend was observed in these patients as well. We will continue to utilize this important information to further determine the best path moving forward for future clinical development of RVX-208."
So the
relative risk (RR) is (4.71%)/(20.31%) =
0.2319 for the diabetic patients. The
relative risk reduction (RRR) is (1 - RR) = (1 - 0.2319) =
0.7681. This is the rounded
77% RRR that the Resverlogix news releases and presentations have used. The
absolute risk reduction (ARR) is 20.31% - 4.71% =
15.6%. So in other words, 15 patients (15.6) out of 100 treated experienced a beneficial MACE preventing effect in the 6 month treatment period.
BDAZ