International Journal of Drug Policy - 2014

Volume 25 Issue 3 May 2014

Issue link:

Contents of this Issue


Page 26 of 153

N. Kori et al. / International Journal of Drug Policy 25 (2014) 525–532 527 Fig. 1. Location of HIV incidence hotspot in relation to density of drug injection sites in Tijuana, Mexico. or on the streets. If participants answered that they speak "flu- ent/native," "very well," or "ok" English on the questionnaire, then they were defined as speaking "some English." Statistical analysis Only participants who provided mappable data for the location where they most frequently injected drugs were included in the present study (n = 1031/1056). Group comparisons were assessed using Pearson chi-square tests for categorical variables and non- parametric Mann–Whitney U tests for continuous variables. Binary logistic regression was used to assess predictors of injecting in the HIV 'hotspot.' Bivariate analyses were first conducted to deter- mine individual, social, and environmental correlates. During this step, we found different correlates by sex, so we created sepa- rate models for male and female IDUs. Correlation statistics were run between the independent variables to check for collinearity before running the multivariate logistic regression. Variables with significance P ≤ 0.10 were considered for inclusion in the multi- variate regression model. Forward stepwise procedures were then performed to determine which predictors remained in the final, reduced model for both men and women using P < 0.05 as the cutoff. A backward stepwise approach was also used to confirm the for- ward, model-building results which reached the same final model. All statistical analyses were conducted using SAS v9.3 software (SAS Institute Inc., Cary, NC). Results Characteristics of the study population The median age of the 1031 participants included in the present analysis was 36 years; 85% were male; and 13% were homeless in the past six months. The most frequently injected drugs in the past 6 months were heroin alone (57%) and methamphetamine and heroin mixed (40%), and did not differ significantly by injec- tion location. Although the HIV transmission hotspot was small (1.95 km 2 ), substance use was highly concentrated in this area with nearly half (48%) of participants reporting most often inject- ing drugs within the hotspot. Compared to those injecting outside the HIV hotspot, a greater proportion injecting within the HIV hotspot were female (21% vs. 9%); reported sex work as their principal source of income in the past year (P < 0.001); traveled >5 km between where they live and inject (P = 0.006); had ever been deported from the U.S. (P = 0.001); and had lifetime or active syphilis infection (P < 0.001) (Table 1). Interactions with police fea- tured prominently in the hotspot, with a higher proportion of IDUs in the HIV hotspot reporting police affecting injection practices, get- ting arrested for carrying unused needles/syringes, being beaten by police, or having their syringes taken by police in the past six months. Those injecting inside the hotspot had higher perceived risk of getting HIV compared to those injecting outside this area (P = 0.005), and they more often engaged in risky sexual behaviors such as sex work, unprotected sex, and being under the influence of drugs during sex. Of the 490 IDUs who reported regularly injecting drugs within the HIV incidence hotspot, 6.7% were HIV positive (vs. 1.67% outside of this area). Harm reduction services in Tijuana are particularly focused in this area and are reflected by higher aware- ness (34% vs. 16%) of needle exchange services and utilization of these services (27% vs. 8%) in the hotspot compared to outside of the area (Table 1). Forty-six percent of IDUs in the hotspot had known someone personally who has HIV/AIDS and several knew as many as 50 people (vs. 20 outside the hotspot) who had died of AIDS. Because these variables are closely linked to the definition of the HIV incidence hotspot and the availability

Articles in this issue

view archives of International Journal of Drug Policy - 2014 - Volume 25 Issue 3 May 2014