Our long-term goals are to contribute to the fight against HIV/AIDS by developing useful statistical methods for the design, conduct, and analysis of research studies aimed at the prevention of HIV infection, the treatment of persons with HIV, and better understanding the inter-relationship between HIV, the immune system, and therapeutic interventions. During the period of the proposed research, we will focus on the following specific aims: 1. To develop statistical methods for prevention studies, with special emphasis on (1.1) estimating the timing of perinatal and vertical transmission of HIV based on diagnostic tests with imperfect specificity and time-dependent sensitivity, and (1.2)statistical methods for HIV preventive vaccine trials, especially in assessing vaccine efficacy and viral transmissibility in the presence of several viral subtypes; 2. To develop statistical methods for the analysis and design of structured interruption trials of host immune suppression, including: (2.1) estimation of the subdistribution of time to host immune suppression in the presence of incomplete follow-up of subjects in post-treatment evaluation phase, (2.2) methods for assessing the consequences of premature treatment interruption, including the time course of viral rebound, the probability of successful rechallenge, and in subjects who are successfully rechallenged, the distribution of time to host immune suppression, and (2.3) the design of structured interruption trials; 3. To develop statistical methods for analyzing HIV/AIDS data subject to interval and informatively missing observations, including (3.1) methods for assessing a primary response variable whose observation depends on the values of intermediate markers, (3.2) regression methods in the presence of interval-censored covariates, and (3.3) models for defining guidelines for the real-time monitoring of viral load for treatment failure, and methods for identifying factors that are associated with viral rebound; and 4. To develop methods for the analysis of Phase I studies, especially methods for designing and analyzing dose escalation studies for trials with bivariate outcome measures.
Showing the most recent 10 out of 128 publications
