Foundational statistical principles in medical research: A tutorial on odds ratios, relative risk, absolute risk, and number needed to treat

Abstract

Evidence-based medicine is predicated on the integration of best available research evidence with clinical expertise and patient values to inform care. In medical research, several distinct measures are commonly used to describe the associations between variables, and a sound understanding of these pervasive measures is foundational in the clinician’s ability to interpret, synthe-size, and apply available evidence from the medical literature. Accordingly, this article aims to pro-vide an educational tutorial/topic primer on some of the most ubiquitous measures of association and risk quantification in medical research, including odds ratios, relative risk, absolute risk, and number needed to treat, using several real-world examples from the medical literature.