Two Stage Sampling for Simultaneously Testing Main and Side Effects in Clinical Trials

✍ Scribed by Dr. P. Bauer

• Publisher: John Wiley and Sons
• Year: 1986
• Tongue: English
• Weight: 506 KB
• Volume: 28
• Category: Article
• ISSN: 0323-3847
• DOI: 10.1002/bimj.4710280717

✦ Synopsis

Summay

I n clinical trials for the comparison of two treatmente it ~B B ~B reawnable to atop the etudy if either one treatment hae worked out to be markedly superior in the main effect, or one to be severely inferior with reapect to an adveree side effect. Two etegs scrmpling plane are considered for eimulteneouely teating a main and side effect, Beeumed to follow a bivariete normal distribution with known variances, but unknown oomlation. The teat procedure keepe the global aignifiosnce tevel under the null hypotheaia of no differencea in main and aide effecte. The critical valnea are chosen under the aide condition, that the probability for ending at the first or second stage with a rejection of the elementary null hypotheaia for the main effect ie controlled, when a partioular conetellation of differencea in mean holde; analogouely the probability of ending with a rejeotion of the null hypotheeee for the aide effect, given certain treatment differences, is controlled too.

Plane "optimal" with mpect to eample eize are given.