Characteristics and Health Service Utilization in Two Samples of Young Injection Drug Users Recruited Using Direct and Referral Methods in San Francisco, California

Abstract

Background: Targeted sampling (TS) and respondent-driven sampling (RDS) methods are methods used to recruit 'hard-to-reach' populations. We examine similarities and differences in population characteristics and health service utilization, in two independent samples of young adult injection drug users (IDU). Results:Compared to the TS sample, the RDS sample was more heterogeneous demographically: a higher proportion of women, and racial/ethnic diversity; a significantly larger proportion of the RDS sample (vs. TS) previously received an HIV test (98.6% vs. 88.5%) whereas the TS sample had a significantly higher prevalence of HCV (58.7% vs. 15.1%).Conclusion: TS and RDS reached a similar cross section of the young adult IDU population in San Francisco; however, some notable differences were observed, especially in HCV infection prevalence. The findings highlight the need for continued efforts to improve sampling strategies aimed high risk IDU, so as to inform both public health surveillance, research and prevention interventions.