Letters Section Editor: Robert M. Golub, MD, Senior Editor.
In Reply: We agree with Dr Cummings' analysis that ORs are not a good approximation to RRs when the risk of the outcome is large. The Table shows the original ORs with RRs for the same outcomes. Risk ratios were estimated by conditional Poisson regression with bootstrapped 95% confidence intervals (2000 replications sampling by study).1,2 We also report the RR for the alternative outcome (eg, reject vs accept or revise) because it is not the reciprocal of the original RR. The OR of the alternative outcome is the reciprocal of the OR of the original outcome.
Schroter S, Tite L, Hutchings A, Black N. Effects of Differences Between Peer Reviewers Suggested by Authors and by Editors—Reply. JAMA. 2006;296(10):1231-1232. doi:10.1001/jama.296.10.1231-b