The Bayesian Hierarchical Model for Estimating the Size of HIV At-Risk Populations in Bangladesh

• Activity Number: 461
• Type: Topic Contributed
• Date/Time: Wednesday, August 4, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Le Bao*+ and Adrian E. Raftery and Amala Reddy
• Companies: University of Washington and University of Washington and UNAIDS Regional Support Team Asia-Pacific
• Keywords: Bayesian hierarchical model ; population size estimation ; concentrated epidemic ; HIV

Abstract

In most countries in the world outside of sub-Saharan Africa, HIV is largely concentrated in sub-populations whose behavior put them at high risk for contracting and transmitting HIV, e.g. injecting drug users, commercial sex workers, men who have sex with men. The size estimation of those sub-populations is important for assessing the overall HIV prevalence and conducting effective intervention. In this article, we present a hierarchical model for making local and national size estimates of HIV at-risk populations. It incorporates multiple commonly used data sources including mapping data, survey, intervention, capture-recapture, estimate/guestimate from other organizations, and experts' opinions. It is the first statistical model for assessing the uncertainty of the HIV at-risk population size. The proposed model is applied to the 2004 Bangladesh size estimation.