|
Activity Number:
|
380
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
ENAR
|
| Abstract - #307531 |
|
Title:
|
Modeling Plasma HIV Viral Load by a Piecewise Polynomial Linear Mixed Model
|
|
Author(s):
|
Hsiao-Chuan Tien*+ and Pai-Lien Chen
|
|
Companies:
|
The University of North Carolina at Chapel Hill and Family Health International
|
|
Address:
|
CB 7420 Department of Biostatistics, Chapel Hill, NC, 27599 7420,
|
|
Keywords:
|
piecewise polynomial linear mixed model ; Markov chain Monte Carlo
|
|
Abstract:
|
The copies of plasma HIV viral load is an important virologic marker for HIV infection. The level of viral load at the set point becomes a crucial index of predicting disease progression. In this study we propose a piecewise polynomial mixed effect model to estimate the level of viral load at set point and to model the plasma HIV viral load process. Our proposed approach is able to integrate the variability of the time of reaching the viral peak level and the viral setpoint among individuals and to incorporate other covariates into the model for evaluating their impact on the level of viral load. The Markov chain Monte Carlo (MCMC) algorithm is used to obtain the estimations of those parameters. Data from a HIV cohort study in Africa is used to illustrate the model.
|
