JSM 2004 - Toronto

Abstract #300540

This is the preliminary program for the 2004 Joint Statistical Meetings in Toronto, Canada. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2004); 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 2004 Program page
Activity Number: 438
Type: Contributed
Date/Time: Thursday, August 12, 2004 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract - #300540
Title: Nonparametric Confidence Intervals for Tmax: Methods Comparison for 3x3 Crossovers
Author(s): Susan A. Willavize*+ and Elizabeth A. Morgenthien
Companies: Pfizer Inc. and Quincunx Group Biostatistical Consulting
Address: PGRD, Groton Laboratories--MS8260-2221, Groton, CT, 06340,
Keywords: 3x3 crossover trial ; nonparametric confidence interval ; Hodges-Lehmann confidence interval ; Tmax
Abstract:

Tmax, the time associated with the maximum plasma drug concentration, is a continuous random variable in theory, but actually discrete, because it is equated to a nominal sampling time. The European guidance on bioavailability and bioequivalence (2001) indicates that, when Tmax is analyzed, a nonparametric confidence interval (CI) method should be used for the pairwise comparisons of interest. For 3x3 crossover designs, we propose a new median-scaled (MS) method of adjusting for periods in a Hodges-Lehmann type CI. A simulation study was done comparing MS with a previously described stratified ranks method. The normal theory (NT) and the nonparametric CI approaches (without adjustment for periods) were also compared. Concentration data were simulated using (Pharsight's) TS-2 software. Results show that sparse sampling and period effects cause increases in CI length. NT can be liberal (i.e., less than nominal coverage) if there is a true treatment effect. All the nonparametric methods are conservative and among them MS is least conservative and has shortest CIs. SR is most conservative and has very long CIs.

  • 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 2004 program

JSM 2004 For information, contact jsm@amstat.org or phone (888) 231-3473. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2004