|
Activity Number:
|
173
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #308097 |
|
Title:
|
Instrumental Variables in Pharmacoepidemiology
|
|
Author(s):
|
M. Alan Brookhart*+
|
|
Companies:
|
Brigham and Women's Hospital
|
|
Address:
|
1620 Tremont St, Harvard Medical School, Boston, MA, 02138,
|
|
Keywords:
|
pharmacoepidemiology ; instrumental variables
|
|
Abstract:
|
Post-marketing studies of the comparative safety and effectiveness of prescription drugs are often based on health care claims data files. These files contain drug dispensing and clinical outcome information on entire populations followed over extended periods of time. Because these data were collected for administrative purposes rather than for research, they do not capture many important variables that physicians use to make prescribing decisions. Unmeasured confounding, therefore, is particularly worrisome in observational studies of drugs in health care claims data. Statistical methods based on instrumental variables present one possible approach to this problem, provided that suitable instruments can be identified. I describe several potential instruments for use in observational studies of drugs and discuss some of their limitations.
|
- 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 2007 program |
