Although clinical trials are conducted with multiple treatment groups, questions of interest are often expressed as pairwise comparisons among the groups. For example, a clinical trial of two dose (D1 and D2) groups and placebo (P ) may have as its objective the effectiveness of each dose, and whether the doses differ in their effectiveness. That is the objective may be formulated in terms of the three pairwise comparisons: D1 − P , D2 − P , and D2 − D1. Other contrasts among the groups may be of interest; e.g. the average of the doses versus placebo. Of course if a trial consists of only two treatment groups, the objective would be formulated as a single comparison of the two groups. In this chapter, we present statistical methods for comparing treatment groups in clinical trials using the R system. Specifically, we present two data sets from clinical trials in Section 3.1 and in Section 3.2, we introduce the associated
Note: to run the R programs in this chapter, the analyst should install the following R packages first: RODBC and bootstrap.