XVIII International AIDS Conference


Back to the PAG
Sign In

Spatial epidemiology of HIV among injection drug users in Tijuana, Mexico

K.C. Brouwer1, M.L. Rusch1, J.R. Weeks2, R. Lozada3,4, A. Vera1, C. Magis-Rodríguez5, S.A. Strathdee1

1University of California at San Diego, La Jolla, United States, 2San Diego State University, Department of Geography, San Diego, United States, 3Patronato Pro-COMUSIDA, A.C., Tijuana, Mexico, 4Prevencasa A.C., Tijuana, Mexico, 5Centro Nacional para la Prevención y el Control del VIH/SIDA - CENSIDA, México City, Mexico

Background: Co-occurrence of the drug and sex trades in Tijuana, Mexico is contributing to increasing prevalence and incidence of HIV/STIs. However, HIV prevalence has remained relatively stable in injection drug users (IDUs) thus far. We assessed trends in the spatial epidemiology of HIV among Tijuana IDUs to better understand transmission dynamics and the epidemic's trajectory.
Methods: From 4/06-12/09, 1056 IDUs injecting within the last month were recruited using respondent-driven sampling. Over 18 months, participants underwent semi-annual surveys, mapping, and testing for HIV, tuberculosis, and syphilis. Using Average Nearest Neighbor and Getis-Ord Gi* spatial statistics, distributions where participants lived, worked, bought and injected drugs were compared with distributions of HIV and environmental, demographic, and behavioral factors.
Results: Median age was 37 years; 86% were male. Crude HIV prevalence was 10.2% among females and 3.5% among males. The 47 HIV cases most strongly clustered by drug injection site (Z-Score -6.173, p< 0.001), which spanned 246 km2; however, we identified a 16 km2 HIV hotspot adjacent to the U.S./Mexico border, encompassing the red-light district. Participants who were female, homeless, knew someone with HIV, or injected in shooting galleries also clustered by injection site (p< 0.001). HIV strongly clustered with both acute and lifetime syphilis infection (p< 0.001), but not tuberculosis. All HIV seroconverters (N=11) injected within a 2.5 block radius immediately prior to seroconversion. Standard deviational ellipses to monitor directional trends suggested a largely static HIV epidemic until July-December 2008, when cases spread from the border area to the southeast.
Conclusions: Geographical distributions of HIV cases can provide insight into transmission mechanisms and epidemic trajectories. Additional studies are needed to determine whether recent HIV dispersion is related to intensified community violence and policing that spiked in the latter half of 2008 and whether this diffusion forebodes increased HIV transmission.

Back - Back to the Programme-at-a-Glance

Contact Us | Site map © 2010 International AIDS Society