Odds Ratios and Multiple Regression Models, Why and How to Use Them

Abstract

In observational studies odds ratios (ORs) and multiple regressions models are commonly used for respectively the surrogate measurements of relative risks and the assessments of independent risk factors. In clinical trials both of them can be used for different purposes. Odds ratios unlike chi-square tests provide a direct insight in the strength of the relationship: odds ratios describe the probability that patients with a certain treatment will have the event compared to those without. Multiple regression models can reduce the data spread due to certain patient characteristics like differences in baseline values, and thus, improve the precision of the treatment comparison. Despite these advantages these methods are not routinely used for the evaluation of clinical trials. The current chapter was written (1) to emphasize the great potential of odds ratios and multiple regression models in clinical trials, (2) to illustrate the ease of use, and (3) to familiarize the non-mathematical readership of this book with these important methods for clinical trials.