Abstract #300692

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2003 Program page
JSM 2003 Abstract #300692
Activity Number: 349
Type: Invited
Date/Time: Wednesday, August 6, 2003 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract - #300692
Title: Multiple Comparison Adjustments when Two or More Coprimary Endpoints Must All be Statistically Significant
Author(s): Walter W. Offen*+ and Jeffrey D. Helterbrand
Companies: Lilly Corporate Center and Eli Lilly & Company
Address: Dropcode 2233, Indianapolis, IN, 46285-0001,
Keywords: multiplicity ; resampling
Abstract:

In recent years there have been numerous disorders for which regulators have insisted on two or more coprimary endpoints, all of which must achieve statistical significance at a nominal alpha level of 0.05 in order for the sponsor to conclude efficacy of the study compound. Examples include Alzheimer's Disease, Erectile Dysfunction, Migraine, and ophthalmology. It is clear that requiring two or more significant coprimary endpoints decreases a trial's statistical power for a given alternative hypothesis and sample size. The degree of this decrease is a function of the correlation between the endpoints. However, the effect upon Type I error is more controversial. We show that under what can be referred to as the complete null space, testing each endpoint at a nominal alpha level of 0.05 maintains the overall Type I error rate at 0.05. However, under the no-effect null hypothesis, the true Type I error rate is less than 0.05. If one is concerned that the study drug is a placebo, then the no-effect null is the appropriate consideration leading to an appropriate upwards adjustment to the nominal alpha levels. Resampling-based methods are proposed to determine the appropriate adjustment.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2003 program

JSM 2003 For information, contact meetings@amstat.org or phone (703) 684-1221. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2003