|
Activity Number:
|
280
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #305301 |
|
Title:
|
Handling Missing Data in the Analysis of Exposure-Adjusted Incidence Rates with Applications to Safety Endpoints in Clinical Trials
|
|
Author(s):
|
Adeniyi J. Adewale*+ and Adam B. Polis
|
|
Companies:
|
Merck & Co., Inc. and Merck & Co., Inc.
|
|
Address:
|
Merck Research Laboratories, UG1C-36, North Wales, PA, 19454,
|
|
Keywords:
|
Exposure adjusted incidence rates ; Safety endpoint ; Missing data ; Multiple imputation ; Propensity score ; Clinical trials
|
|
Abstract:
|
Analysis of exposure adjusted incidence rates of adverse experiences are often necessary in clinical trials with long-term follow-up in order to mitigate against the effect of possible differences in duration of follow-up between treatment groups. The incidence rate is calculated as the ratio of the count of the event of interest to the total person-time follow-up for a treatment group. When an adverse experience is known to occur but its date of occurrence is unknown, handling of such cases can potentially influence the rate estimates. We describe various mechanisms of occurrence of this problem. We discuss strategies for handling missing follow-up time ranging from ignoring them in the computation of total person-time to propensity score based multiple imputation. The properties of these strategies are evaluated using simulations. Finally, we considered an application.
|
