Online Program Home
My Program

Abstract Details

Activity Number: 607
Type: Contributed
Date/Time: Wednesday, August 3, 2016 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #321245 View Presentation
Title: Reversed Hazard Parametric Regression Model to Analyze Left-Censored HIV Data
Author(s): Md Akhtar Hossain* and Hrishikesh Chakraborty
Companies: University of South Carolina and University of South Carolina
Keywords: Reversed hazard ; Left censoring ; Parametric regression ; HIV ; Viral load

Survival analysis primarily focuses on understanding the relationship between the lifetime event and the covariates. Though most lifetime regression models are developed to analyze right censored data, it is not uncommon to encounter left censored lifetime data in practice. Previous studies suggested models based on reversed hazard rates as a more appropriate method to handle left censored lifetime data. For example, a parametric inverted Weibull regression, based on reversed hazard rate, was recently suggested for left censored lifetime data analysis. Our study extends reverse hazard parametric models for other lifetime distributions including inverted Gaussian, log-normal, log-logistic, Gompertz, exponential, and gamma distributions. The model performances were assessed using Monte Carlo simulations and were validated using left censored HIV viral load (VL) data from adult HIV patients in South Carolina. The inverted Weibull reversed hazard rate regression has shown superior diagnostic characteristics compared with models using other distributions to model the risk of transitioning from undetectable to detectable HIV VL level.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association