Choice of effect measure for epidemiological data

The debate concerning the choice of effect measure for epidemiologic data h as been renewed in the literature, and it suggests some continuing disagree ment between the pertinent clinical and statistical criteria. In this artic le, some defining characteristics of the main choices of effect measure [ri sk difference (RD), relative risk (RR), and odds ratio (OR)] for binary dat a are presented and compared, with consideration of both the clinical and s tatistical perspectives. Relationships of these measures to the relative ri sk reduction (RRR) and number needed to treat (NNT) are also discussed. A n umerical comparison of models of constant RD, RR, and OR is made to assess when and by how much they might differ in practice. Typically the models sh ow only small numerical differences, unless extreme extrapolation is involv ed. The RD and RR models can predict impossible event rates, either less th an zero or greater than 100%. Each measure has potential theoretical justif ication. RD and RR may enjoy some advantages for communication of risk, but OR may be preferred for data analysis. A clear distinction should be maint ained between the objectives of data analysis and subsequent risk communica tion, and different effect measures may be needed for each. (C) 2000 Elsevi er Science Inc. All rights reserved.