Individual and neighborhood characteristics associated with HIV among black and latino adults who use drugs and unaware of their HIV-positive status, New York city, 2000–2004

Abstract

With mounting evidence of how neighborhood socioeconomic context influences individual behavior, investigation of neighborhood social context and sex/drug use risk behavior could help explain and provide insight into solutions to solve persistent racial disparities in HIV. Intervieweradministered surveys and HIV testing among street-recruited individuals who reported illicit drug use in New York City were conducted from 2000 to 2004. Individuals were geocoded to census tracts, and generalized estimating equations were used to determine correlates of being newly diagnosed with HIVat study enrollment. Analyses were completed in 2014. Of the 920 participants, 10.5% were HIV-positive, and among those, 45% were diagnosed at study enrollment. After restricting the sample to those who self-reported negative HIV status (n = 867), 72% were male, 65% Latino, and 5.1%tested HIV-positive. After adjustment, those testing HIV-positive were more likely to report male same-sex partnership (p < 0.01) and less likely to be homeless compared with those confirmed HIV-negative (p < 0.01). Neighborhood-adjusted models indicated those from neighborhoods with less deprivation (p < 0.05), and a higher proportion of owner-occupied homes (p < 0.01) were more likely to test HIV-positive. Additionally, Black individuals who used drugs and were from neighborhoods with a higher proportion of Black residents were more likely to be newly diagnosed compared to Latino individuals who used drugs and were from neighborhoods with lower proportions of Black residents (p < 0.05). These data suggest that HIV prevention and treatment efforts should continue widening its reach to those unaware of their HIV infection, namely men who have sex with men, heavy, drug-involved Black communities, and both Black and Latino communities from relatively less disadvantaged neighborhoods.