International Journal of Drug Policy - 2014

Volume 25 Issue 3 May 2014

Issue link: http://digitalreprints.elsevier.com/i/364061

Contents of this Issue

Navigation

Page 100 of 153

D. Kao et al. / International Journal of Drug Policy 25 (2014) 598–607 601 Fig. 1. Study participants. Data sources: ESRI Data & Maps and StreetMap USA 2006; United States Census Bureau. Ulmer, Hurvitz, & Drewnowski, 2012). In preparation, we also tested service areas with various thresholds (e.g. 5 minute, 15 minute, and 30 minute) and overlaid these layers with the city's "super neighborhoods," i.e. geographic areas designated by the city of Houston for local planning and governance (http://www.houstontx.gov/superneighborhoods/index.html). Based on this process, we determined in most cases, a 10-minute driving radius resulted in adequate coverage of the partici- pant's home super neighborhood as well as the adjacent super neighborhoods. Control measures Several variables were also included in the multivariate models as controls, including: drug user group (former, methadone, and current); the participant's age at the time of the interview (con- tinuous); the age at which the participant reported first injecting heroin (continuous); educational attainment (less than high school vs. high school graduate or equivalent); fulltime employment sta- tus (yes or no); and marital status (married vs. not married). Analysis The ArcGIS 10.0 software (Esri, 2011a) was used to manage the spatial data, geocode the participants and facilities, and construct the spatial accessibility measures. Once the spatial accessibility measures were created, bivariate and logistic regression analyses were then used to explore and test the relationships between spa- tial accessibility and the drug use-related outcomes. Using Stata 11.2 (StataCorp, 2009), separate analyses were conducted for the full sample (N = 219), with a particular interest on the current user group (n = 75). Results Descriptives As shown in Table 1, the current users in our sample tended to be younger and less educated than the methadone and former user groups. In particular, only about half of current users (52%) had graduated from high school, compared to 68% of the respondents in both methadone and former user groups. Higher proportions of the respondents in our methadone group were employed fulltime and married compared to the other two groups. Most notably, the average age at which one first injected heroin differed significantly among the three groups, with the current users in our sample start- ing at an older age than both the methadone and former user groups (20.2 years vs. 17.7 and 18.6, respectively). With regards to the drug-related outcomes, current users were significantly less likely to exhibit internal LOC, compared to the other two groups. Less than half of the current user group had a LOC score of 4 or higher, compared to 96% of the former users and 84% of the methadone group. Interestingly, former users were the most likely to report a "very high" chance of injecting in the future (45%), followed by current users (35%). The methadone group was the least likely (20%) to report a "very high" chance of injecting in the future. A significantly greater percentage of current users (42%)

Articles in this issue

Links on this page

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