How Big is a Big Odds Ratio? Interpreting the Magnitudes of Odds Ratios in Epidemiological Studies

• Published in: Communications in statistics. Simulation and computation Vol. 39; no. 4; pp. 860 - 864
• Main Authors: Chen, Henian, Cohen, Patricia, Chen, Sophie
• Format: Journal Article
• Language: English
• Published: Colchester Taylor & Francis Group 01.04.2010; Taylor & Francis; Taylor & Francis Ltd
• Subjects: Applications; Biology, psychology, social sciences; Computation; Computer simulation; Effect size; Epidemiological study; Epidemiology; Equivalence; Exact sciences and technology; Logistic regression; Mathematical analysis; Mathematics; Medical sciences; Numerical analysis; Numerical analysis. Scientific computation; Numerical methods in probability and statistics; Odds ratio; Probability and statistics; Risk factor; Sciences and techniques of general use; Statistics; Biometrics; Statistical method; Logistic regression; Medical science; 62P10; Effect size; Risk factor; Numerical simulation; Odds ratio; 62H20; Epidemiology; Epidemiological study
• ISSN: 0361-0918; 1532-4141
• DOI: 10.1080/03610911003650383

The odds ratio (OR) is probably the most widely used index of effect size in epidemiological studies. The difficulty of interpreting the OR has troubled many clinical researchers and epidemiologists for a long time. We propose a new method for interpreting the size of the OR by relating it to differences in a normal standard deviate. Our calculations indicate that OR = 1.68, 3.47, and 6.71 are equivalent to Cohen's d = 0.2 (small), 0.5 (medium), and 0.8 (large), respectively, when disease rate is 1% in the nonexposed group; Cohen's d < 0.2 when OR <1.5, and Cohen's d > 0.8 when OR > 5.