This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 469
Type: Contributed
Date/Time: Wednesday, August 4, 2010 : 8:30 AM to 10:20 AM
Sponsor: Biopharmaceutical Section
Abstract - #307167
Title: A 'Paradox' in the Use of an ANCOVA Model in Determination of Treatment Effect in Clinical Trials
Author(s): Chi-Hse Teng*+ and Larry Z. Shen and Ping Yan
Companies: Amylin Pharmaceuticals and Amylin Pharmaceuticals and Amylin Pharmaceuticals
Address: , , ,
Keywords: ANCOVA model ; Paradox ; Clinical trials ; Lsmeans ; Treatment Effect
Abstract:

An ANCOVA model adjusted for baseline values of response variable and stratification factors (used for randomization) are commonly used to determine the treatment effect in an efficacy endpoint in diabetes clinical trials. The within-treatment effect is often reported to be the LS mean of a treatment derived from the model. However, in some cases, the LS mean of a treatment derived from the model can be out of the range of within-stratification-factor treatment means or LS means from subgroup analyses. This raises the concern about the interpretability and validity of the LS means of the within-treatment effect and the model. We will present a general formula that quantifies the condition under which these circumferences can occur and discuss factors that contribute to this condition. Statistical recommendations including alternative model approaches will be provided.


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 2010 program




2010 JSM Online Program Home

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.