|
Activity Number:
|
343
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #309311 |
|
Title:
|
Use of Beta Regression To Model Adherence to HIV Medication
|
|
Author(s):
|
Che Smith*+ and Yin Yin and Sharon Murray and Steven Kathman
|
|
Companies:
|
The University of North Carolina at Chapel Hill and GlaxoSmithKline and GlaxoSmithKline and GlaxoSmithKline
|
|
Address:
|
226 Southpoint Crossing Dr, Durham, NC, 27713,
|
|
Keywords:
|
beta ; regression ; adherence
|
|
Abstract:
|
Knowledge of patient adherence to prescribed medication is of great value in HIV clinical trials to help understand and predict trends in viral load. It is also important to be able to make predictions as early as possible to save time and money. Using data from a phase III trial of 51 patients taking highly active antiretroviral therapy (HAART), we fit and compare several statistical models to represent patient adherence to medication. We fit a beta distribution with parameters estimated using the method of moments. Models were constructed using the CAPABILITY and NLMIXED procedures in SAS, which rely on maximum likelihood estimation of parameters. The method of moments provides the best fit of the adherence data.
|
