Equivalence testing with dental clinical trials

Clinical trials whose purpose are to show equivalence of two or more treatments have traditionally utilized methods for demonstrating superiority, and, if no statistical differences are found, the treatments are assumed to be equivalent. Methods for directly demonstrating equivalence have been developed by the pharmaceutical industry to establish the bioequivalence between drug formulations. This approach should have significant future implications for dental research. It is presently well-characterized in the statistical literature, but is virtually unknown within the dental literature and has had limited documented application with dental clinical trials. The purpose of this paper is to consider equivalence testing with dental clinical trials. Equivalence and superiority testing are contrasted, and comparisons between parallel and crossover designs with equivalence testing are discussed. To exemplify equivalency trial methodology for dental research application, several examples are proposed, including, in detail, an implant overdenture clinical trial to test the equivalence of treatment outcome between a two- and a four-implant overdenture treatment. Sample size calculations for otherwise-identical crossover and parallel equivalence studies are demonstrated. Results show that for a power of 0.90, a sample size of 24 is required for the crossover trial and over 150 subjects for the parallel design. For this dental research application, the equivalence/crossover design methodology is shown to be an efficient and appropriate approach.