Meta-Analysis of Survey Data to Estimate the Size of the Population of Men Who Have Sex with Men for Calculating HIV Rates

• Activity Number: 455
• Type: Topic Contributed
• Date/Time: Wednesday, August 4, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Christopher H. Johnson*+ and David Purcell and Amy Lansky and Joseph Prejean and Paul Denning and Nicole Crepaz and Zaneta Gaul and Renee Stein and Zaneta Gaul
• Companies: CDC and CDC and CDC and CDC and CDC and CDC and CDC and CDC and CDC
• Keywords: surveys ; meta-analysis ; surveillance ; random-effects models ; HIV

Abstract

A majority of HIV cases in the United States have occurred among men who have sex with men (MSM). Lacking an authoritative data source on the size of this population, CDC had not previously calculated HIV rates for this risk group. We recently reported that MSM account for 4.0% (3.1% - 4.9%) of U.S. men and 45% of the 2007 HIV/AIDS diagnoses, yielding a case rate among MSM of 692/100,000 -- 60 times as high as the rates for other men and for women. This presentation focuses on challenging methodological issues in that calculation. A systematic review of potentially relevant published articles identified 7 nationally representative surveys that measured same-sex behavior. We used meta-analytic procedures as extended to survey data by Rao (2008) to estimate MSM proportions, considering fixed- and random-effects models. We discuss methods for making estimates for subgroups by race and age.