Connecting Race, Place and HIV Prevalence: A County-Level Analysis
Vaughan, Adam Stephen
Committee Chair / Thesis Adviser:
Sullivan, Patrick S
Shouse, R. Luke
Health Sciences, Epidemiology
Rollins School of Public Health,
Permanent url: http://holden.library.emory.edu/ark:/25593/bnz0j
Connecting Race, Place and HIV Prevalence: A County-Level
Background: Within the United States, disparities in HIV
prevalence among the three major
racial/ethnic groups (white, blacks and Hispanics) are striking.
The pattern of disparities,
with blacks and whites at the extremes and Hispanics in the middle,
extends across risk
groups and geographies. In an attempt to explain these disparities,
recent analyses have
found statistically identical HIV prevalence rates among
heterosexuals in urban settings
when races live under the same impoverished conditions.
Objective: This analysis aimed to use publicly available data to
further explore associations
between poverty, race and HIV in the United States.
Methods: We used national HIV surveillance data and other
publicly available data to
examine the interaction between race and poverty on county level
HIV prevalence by
urbanization. A negative binomial model, stratified on
urbanization, was constructed with
race-specific county-level HIV prevalence rate as the outcome and
race, poverty and other
social determinants of health as covariates. Prevalence rate ratios
(PRR) were calculated for
black and Hispanic populations, relative to white populations, at
each level of urbanization
and for various levels of poverty.
Results: The results indicated that poverty is associated with
HIV prevalence only in major
metropolitan areas. In non-urban areas, poverty was not associated
with HIV prevalence.
High levels of poverty eliminated or attenuated racial/ethnic
disparities only in counties
belonging to major metropolitan areas. Additionally, the
relationship between HIV
prevalence and poverty varied by race, with blacks having a
stronger association that either
whites or Hispanics.
Discussion: High poverty may explain HIV prevalence rate
disparities in major metropolitan counties.
However, outside of these counties, other factors, including sexual
networks, must be
examined to explain racial/ethnic rate disparities. These results
indicate that future HIV
prevention efforts must consider both location and race.
Table of Contents
TABLE OF CONTENTS -- INTRODUCTION 1 -- METHODS 4 -- RESULTS 7 --
Model selection 7 -- Model results 8 -- DISCUSSION 11 --
Limitations 14 -- Future Directions 16 -- CONCLUSION 18 --
REFERENCES 19 -- TABLES AND FIGURES 23 -- APPENDIX A: SAS CODE 33
Permission granted by the author to
include this thesis or dissertation in this repository. All rights
reserved by the author. Please contact the author for information
regarding the reproduction and use of this thesis or