Measuring the Size of a Treatment Effect: Relative Risk Reduction, Absolute Risk Reduction, and Number Needed to Treat

Hussein Hollands, MD, MS (epid) Peter J. Kertes, MD, CM, FRCS(C) The purpose of this review article is to outline a few concepts that are useful for interpreting the results of a randomized clinical trial (RCT). Specifically, we will focus on measures of the size of a treatment effect— relative risk ratio (RRR), absolute risk ratio, and number needed to treat (NNT). These concepts are important, specifically with respect to decisions to implement a treatment modality in clinical practice based on the results of a RCT. To solidify these ideas, we will work through some practical examples in ophthalmology. In this manuscript, we will assume that we are interpreting results from a well-designed RCT with a clinically important dichotomous ‘‘negative’’ outcome measure (eg, 15-letter loss of visual acuity) comparing a new treatment modality with a control.