The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
89
|
Type:
|
Roundtables
|
Date/Time:
|
Monday, July 30, 2012 : 7:00 AM to 8:15 AM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract - #304207 |
Title:
|
Multiple Testing and Modeling in Post-Market Medical Studies
|
Author(s):
|
S. Stanley Young*+
|
Companies:
|
National Institute of Statistical Sciences
|
Address:
|
P.O. Box 14006, Research Triangle Park, NC, , USA
|
Keywords:
|
multiple testing ;
multiple modeling ;
false positives
|
Abstract:
|
Post-marketing medical medical studies can have a profound impact on drug sales, yet they can be subject to unintentional (or intentional) manipulation. There is the obvious problem of multiple testing. Suppose 5 drugs are studied. It is a fact that one will appear most dire. Suppose 20 side effects are studied. With 5 drugs and 20 side effects it is a situation ripe for false positives. Something is very likely to be "statistically significant". Observational studies are typically "adjusted" to remove the effects of confounding variables. Each time a variable is moved into or out of the model, the estimated effect of the drug is changed. There is a need to understand how model selection can influence study claims. The purpose of this roundtable is to discuss the degree to which multiple testing and multiple modeling might be a problem and what might be done to address these issues.
|
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 2012 program
|
2012 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.