Improved injection network ascertainment with supplementary elicitation techniques.

Prior research indicates that injection drug users forget substantial proportions of their injection partners when asked to recall them. Such under-reporting hampers both ascertainment of the injection networks that underlie transmission of blood-borne pathogens and contact-tracing efforts for disease control. We report here an evaluation of supplementary elicitation techniques--a set of prompting strategies and recall cues--employed in a study that involved tracing of injection partners. Sixty-one index drug injectors in Seattle participated in the study. The supplementary partner elicitation techniques enhanced recall of injection partners substantially and identified persons relevant to transmission of blood-borne pathogens. As a set, the supplementary techniques elicited additional partners from 70% of injectors, and the additional partners elicited represent a 75% increase on average. Drug injectors who recalled many partners before administration of the supplementary techniques tended to report more additional partners in response to the supplementary techniques than injectors who freely recalled few partners. In addition, partners elicited by the supplementary techniques were as likely as freely recalled partners to test positive for hepatitis C virus antibody and engage in risk behaviour with indexes. Furthermore, we found that the supplementary techniques boosted connectivity in the observed injection network considerably.

Social structural and behavioral underpinnings of hyperendemic hepatitis C virus transmission in drug injectors.

BACKGROUND: Hepatitis C virus (HCV) is hyperendemic in drug injectors, yet social structural and behavioral factors underlying transmission are not well established. METHODS: We conducted a case-control study of HCV seroconversion in drug injectors, focusing on transmission within networks. Incident case subjects (n=17) and seronegative control subjects (n=42) reported injection and sex partners and referred as many as 5 for interviewing and blood testing. We performed nucleotide sequencing of HCV isolates from infected individuals. RESULTS: Seventy-eight percent of recent injection partnerships involved behavior that could transmit HCV. Case subjects and control subjects were similar demographically and behaviorally. Case subjects, however, had more HCV-infected partners and consequently engaged in injection risk behavior with more infected partners. The injection network was mostly connected, dense, and cyclic, but the sexual network was highly fragmented. Although participants generally injected with partners of similar age, most HCV-uninfected participants recently had injected with infected partners. In at least 1 of 4 pairs of genetically linked infections, transmission appeared to be due to sharing of injection equipment other than syringes. Except for transmission pairs, network distance between incident case subjects and genetic distance between their HCV variants were uncorrelated. CONCLUSIONS: Without dramatic reductions in injection risk behaviors, shattering of cohesive injection networks, and/or broad coverage of an effective vaccine, HCV will likely remain hyperendemic in drug injectors.