生體相等性試驗在學名藥的審核中扮演重要的角色，一般是針對自願的健康的成人進行雙向交叉設計實驗，根據相關的生體可用率參數進行上述檢定。美國食品藥物管理局（FDA, 1992）建議使用單一生體可用率進行生體相等性檢定，但是同時針對多個生體可用率參數建構檢定應更合理，因此Berger和Hsu（1996）與Wang等人（1999）分別在平行設計與雙向交叉設計下建立生體相等性的多元檢定。本文考慮在雙向交叉設計之下，修正Berger和Hsu（1996）的方法，針對多個生體可用率參數建立聯合信賴域進行生體相等性的多元檢定。除以模擬方式研究所提修正方法的型I誤差率及檢定力表現，最後經由實例分析說明所提方法的應用。 The bioequivalence test plays an important role on the approval of a generic drug. To perform such a test, a two way crossover design is usually conducted which involves healthy adults and some bioavailability parameters are then measured. The U.S. Food and Drug Administration (FDA, 1992) recommends testing the bioequivalence based on one of the bioavailability parameters. However, it would be more reasonable to consider a test based on some parameters simultaneously. Berger and Hsu (1996) and Wang et al. (1999) construct multivariate bioequivalence tests under parallel and crossover design, respectively. This article considers revising the work of Berger and Hsu (1996) and contrasting a multivariate bioequivalence test based on a confidence set for crossover design. The results of a simulation study are presented for the investigation of type I error rate and power of the proposed methods. Finally, the application of the tests is illustrated by using a real data.
