|
Activity Number:
|
233
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 13, 2002 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section*
|
| Abstract - #300088 |
|
Title:
|
Individual Treatment Variability in Clinical Experiments and the Implications for Genetic Studies
|
|
Author(s):
|
Gary Gadbury*+
|
|
Affiliation(s):
|
University of Missouri, Rolla
|
|
Address:
|
202 Rolla Building, Rolla, Missouri, 65409,
|
|
Keywords:
|
causation ; potential outcomes ; treatment heterogeneity ; treatment effect ; pharmacogenomics
|
|
Abstract:
|
Most statistical characterizations of a treatment effect focus on the average effect of the treatment over an entire population. However, average effects may provide inadequate information when a substantial subject-treatment interaction is present in the population. It is even possible that a nonnegligible proportion of individuals in the population experience an unfavorable treatment effect even though the treatment might appear to be beneficial when considering population averages. This paper examines the extent to which information about subject-treatment interaction can be extracted using observed data from common randomized experiments comparing two treatments. Both continuous and binary outcomes are considered. Although subject-treatment interaction is a nonidentifiable quantity, we show that mathematical bounds for it can be estimated from observed data. These bounds lead to estimated bounds for the probability of an unfavorable treatment effect. Use of the estimated bounds is illustrated with some examples, and the implications for pharmacogenomic studies are discussed.
|
