Modeling Seroadaptation and Sexual Behavior with a Simultaneously Multilevel and Multivariate Longitudinal Count Model

• Activity Number: 629
• Type: Invited
• Date/Time: Thursday, August 2, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Yuda Zhu and Robert E Weiss*+
• Companies: University of California at Los Angeles Fielding School of Public Health and University of California at Los Angeles Fielding School of Public Health
• Keywords: Bayesian hierarchical model ; Count data ; HIV behavioral intervention ; multilevel ; serosorting

Abstract

Longitudinal trials to assess sexual behavioral interventions collect complex multilevel multivariate data on each subject at each time point. Numbers of partners and numbers of sex acts with each partner are reported longitudinally. Partners are divided into HIV positive and negative and sex acts with each partner are further differentiated as protected and unprotected. Eligibility criteria limit enrollment to participants with some minimal level of risky sexual behavior tied directly to the outcome of interest. The combination of these factors makes it dicult to quantify sexual behaviors and the effects of intervention. We propose a multivariate multilevel count model that simultaneously models the number of partners, acts within partners, and accounts for recruitment eligibility, and provides a more accurate and complete model of sexual behavior. We examine risk reducing behavior that depends on the HIV serostatus of the partner referred to as seroadaptive behavior. We show how to quantify specific forms of seroadaptive behavior and to distinguish it from nonselective risk reduction behavior.