Skip Navigation

Social Structural and Behavioral Underpinnings of Hyperendemic Hepatitis C Virus Transmission in Drug Injectors

  1. Devon D. Brewer1,
  2. Holly Hagan4,
  3. Daniel G. Sullivan2,
  4. Stephen Q. Muth5,
  5. Eileen S. Hough3,
  6. Nathan A. Feuerborn2 and
  7. David R. Gretch2
  1. 1Interdisciplinary Scientific Research,
  2. 2Department of Laboratory Medicine, University of Washington, and
  3. 3Public Health–Seattle and King County, Seattle;
  4. 4Center for Drug Use and HIV Research and Institute for AIDS Research, National Development and Research Institutes, New York, New York;
  5. 5Quintus-ential Solutions, Colorado Springs, Colorado
  1. Reprints or correspondence: Dr. Devon D. Brewer, Interdisciplinary Scientific Research, PO Box 15110, Seattle, WA 98115 (http://www.interscientific.net/contact.html)
 
Next Section

Abstract

BackgroundHepatitis C virus (HCV) is hyperendemic in drug injectors, yet social structural and behavioral factors underlying transmission are not well established

MethodsWe 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

ResultsSeventy-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

ConclusionsWithout 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

Hepatitis C virus (HCV) infection is hyperendemic among drug injectors worldwide [1–3]. The annual incidence of HCV infection in seronegative injectors has ranged from ∼10% to ∼40% [2, 4–6]. Sharing of syringes and other injection paraphernalia is associated with HCV seroconversion [4, 5, 7, 8]. Several research teams have also studied social structural influences on disease transmission in injection drug users (IDUs), investigating aspects of injection networks and the relationship between prevalent HIV infection and network position [9–13]. For other infectious diseases, properties of contact networks covary with the speed and scope of transmission; for instance, the connectivity, cyclicity, density, and concurrency of sexual networks are associated with the incidence of sexually transmitted disease [14–18]

Building on this foundation, we examined the social structural and behavioral factors underlying hyperendemic HCV transmission in IDUs. We measured associations between network and behavioral variables and incident infection. We also assessed injection and sexual network structures for their transmission potential and described patterns of mixing by serostatus and age. Finally, we documented the risk behaviors in partnerships with genetically linked infections—reflecting transmission—and investigated the correspondence between genetic and network distance among incident case subjects

Previous SectionNext Section

Methods

Study Design

We conducted a case-control study of HCV seroconversion in IDUs, focusing on transmission within networks. Case subjects were IDUs with incident HCV infection, defined as either previously HCV antibody (anti-HCV)–negative individuals who were anti-HCV positive at their most recent test or new injectors (<8 months since their first drug injection) who were anti-HCV positive at study enrollment. Control subjects were anti-HCV–negative IDUs. Both case subjects and control subjects referred their injection and sex partners for interviewing and blood testing. Informed consent was obtained from all participants, and study procedures were approved by the Institutional Review Boards of the University of Washington and the Washington State Department of Health

Index Individuals

For this study, we recruited index individuals from a large prospective cohort of drug injectors in Seattle [7, 19]. Individuals were eligible to participate if they had injected drugs at least once during the preceding 12 months, were ⩾14 years of age, and were English speaking. Cohort study participants were scheduled for HCV antibody testing and interviews about their risk behaviors every 6 months. Those meeting case criteria were invited to participate in the present study. For each case subject participating as an index individual, we recruited up to 4 anti-HCV–negative study participants similar in age

Seventeen case subjects and 42 control subjects were enrolled between December 2000 and January 2002. Two index individuals had missing or indeterminate anti-HCV status and were included in network analyses only. These 61 index individuals successfully referred 146 contacts (133 injection partners and 13 injection/sex partners). Twenty-one contacts had already been enrolled as a case subject or control subject, and 14 persons were successfully referred as a contact by multiple index individuals

Procedure

Interviews of index individuals and referral of partners Interviewers elicited each index individual’s injection and sex partners for the period during which the index individual could have possibly acquired HCV (typically 12 months, given the testing interval in the cohort study and window period of the anti-HCV test; range, 5–12 months). Most (68%) of the interviews of the index individuals were conducted on the ending date of the recall period. However, 17% (10) of the index individuals (including 5 case subjects and 4 control subjects) were interviewed >90 days (maximum, 489 days) after the end of the recall period, partly because we sought to include as many incident case subjects as possible. Index individuals’ recall periods were 67% concurrent. That is, the sum of the pairwise temporal overlap between index individuals’ date-specific recall periods was 67% of the sum of the maximum possible overlap of their recall periods if the precise dates could be shifted with the durations kept constant. Therefore, most reported partnerships occurred within the same, relatively short period (mid-2000–mid-2001). Interviewers elicited index individuals’ injection and sex partners separately. Injection partners were defined as persons with whom an index individual had injected drugs, regardless of whether they shared needles [20]. Sex partners were defined as anal, oral, and vaginal sex partners, regardless of the context in which sexual contact occurred [20]. To enhance recall of partners, interviewers administered supplementary elicitation techniques after index individuals freely recalled partners, and these techniques boosted reporting substantially (D.D.B., H.H., and E.H., unpublished data)

Next, index individuals identified partners whom they could refer to the study. Interviewers then asked the index individual to refer all such partners (if <6); if an index individual had >5 partners whom he or she could refer, the interviewer designated a systematic random sample of 5 for referral. Interviewers gathered first names/nicknames and physical and demographic descriptions of the selected partners; for partners not selected, only first names/nicknames were collected. Index individuals received $15 for participating in the interview and $10 for each partner successfully referred. Index individuals received vouchers to give to partners whom they were to refer

Interviewing and testing of partnersWhen partners came to the study site, interviewers determined whether the partner matched (by name, physical description, and voucher number) the individual reported by the index individual (see the Appendix [online only]). Partners referred to the study within 21 days of the index individual’s interview received $5 for redeeming a voucher and an additional $20 for participating in an interview and testing. Seventy-three percent of referrals were successful (146 partners referred successfully of 200 partners sought). Each enrolled partner completed an interviewer-administered questionnaire about the partner’s injection and sexual risk behavior with the index individual and provided a blood specimen. The recall period for most risk-behavior questions in the interviews of partners matched that for the referring index individual. Interviewers did not elicit partners’ other partners

The key measures derived from interviews of partners include whether the partner reported injecting with the referring index individual and the index individuals’ receptive syringe sharing and receptive injection risk. Receptive syringe sharing indicates whether the partner reported the index individual had used a needle/syringe the partner had previously used during the specific recall period. Receptive injection risk indicates whether the partner reported the index individual had engaged, during the specific recall period, in any of the following behaviors that may involve risk of acquisition: (1) used a needle/syringe after the partner had used it (regardless of bleaching); (2) used a “cooker” (i.e., a container for heating drugs into solution) after the partner had used it; (3) used a “cotton” (i.e., a filter to block undissolved contaminants when drawing drug solution into the syringe) after the partner had used it; (4) used the same rinse water (for cleaning the syringe after injection) or same rinse water container as the partner; or (5) divided drugs with the partner using an unsterile, previously used needle (“backloading” or “frontloading”). “Don’t know” and missing partner responses were classified as “no” in analysis. The receptive injection risk measure may be slightly liberal, because the questions regarding shared use of rinse water and syringe-mediated sharing of drugs did not specify the order in which participants had used the rinse water or syringe. Our use of partners’ reports for these measures may lessen any bias arising from social desirability

Laboratory proceduresWithin 3 h after collection, blood specimens from all participants (index individuals and partners) were sent to the laboratory to be centrifuged and then aliquoted for storage at −70°C. Anti-HCV testing was performed with EIA (version 2.0; Abbott Laboratories); reactive specimens were retested in duplicate by EIA and were interpreted as being anti-HCV positive if either or both specimens were reactive. We evaluated samples with a low signal:cutoff ratio [21] by using the Recombinant Immunoblot Assay (Chiron). All participants received pre- and posttest HCV counseling

The stored specimens from participants who tested anti-HCV positive were submitted for DNA sequencing of both the HCV envelope 1 (E1) gene and the hypervariable region 1 (HVR1) of the envelope 2 (E2) gene. Viral RNA was extracted from 160 μl of serum with the QIAamp viral RNA isolation kit (Qiagen). cDNA was synthesized using oligonucleotide primers and Moloney murine leukemia virus reverse transcriptase [22]. Nested polymerase chain reaction (PCR) was performed with Advantage High Fidelity 2 DNA polymerase (BD Biosciences) [23]. Purified PCR products were directly sequenced with an Applied Biosystems automated sequencer (model 377) (see the Appendix [online only])

Nucleotide sequences were optimally aligned with the CLUSTAL W program [24]. Phylogenetic analysis was performed with programs from the PHYLIP package (version 3.5c) [25]. We estimated nucleotide distances between all pairs of sequences with the DNADIST program (Kimura 2-parameter option) and generated a neighbor-joining tree [26]. Genotype was determined by comparing isolate sequences to HCV sequences of known genotype from Genbank (http://www.ncbi.nlm.nih.gov/Genbank/). Sequencing was possible for only 31 of 58 anti-HCV–positive participants (including 10 of 17 incident case subjects), due to either insufficient blood specimens, low viral loads, or cleared infections

Statistical Analysis

Before performing network analyses, we identified participants and reported partners as uniquely as possible by using participants’ anonymous study codes, multiply mentioned uncommon street nicknames of partners, and multiply mentioned uncommon first names of partners (see the Appendix [online only]). We treated each mention of a relatively common nickname or first name not unduplicated by anonymous code as a different individual

We calculated univariate descriptive statistics to summarize participants’ demographic characteristics, risk behaviors, and partner distributions. We computed the distribution of components (disjoint sets of participants connected directly or indirectly via partnerships) [27] separately for the injection network and the sexual network

To assess the tendency of injection partners to be anti-HCV concordant, we computed measures of network autocorrelation (Geary’s C and Moran’s I [28–30]) and associated randomization test probability values based on 10,000 permutations. Geary’s C ranges from 0 to 2 and is 1 when no autocorrelation is present; values <1 represent positive autocorrelation (i.e., clustering of participants with similar characteristics). Moran’s I ranges from −1 to 1 and is -1/(n-1) when no autocorrelation is present; values >0 represent positive autocorrelation. We also used these measures to examine the tendency of participants to inject with partners of similar age (as reflected by absolute age difference in years). Furthermore, we computed matrix correlations and associated randomization tests (10,000 permutations each) [31] between pairs of incident case subjects’ injection network (geodesic) [27] distances and genetic distances of their HCV variants

To compare case subjects and control subjects in terms of risk behaviors with their injection partners and positions in the injection network (membership in the main component), we computed Pearson (point-biserial and ϕ) correlation coefficients (r) odds ratios (ORs), and corresponding probability values. Data management and analysis were performed with Microsoft Access 97, SPSS (version 7.5), UCINET 6 for Windows [32], NetDraw [33], and custom programs written in FreeBasic (http://www.freebasic.net)

Previous SectionNext Section

Results

Characteristics of participantsCase subjects and control subjects were very similar demographically and behaviorally (table 1). Participants were mostly young, white, and heterosexual injectors who primarily injected heroin or amphetamines. Almost half of participants were homeless

Figure 1
View larger version:
Figure 1

Main component of injection network after reduction (see the “Injection and sexual network structure” subsection of Results), as rendered by a spring-embedder algorithm [33] with slight manual adjustments for clarity. Nodes denote index individuals (triangles) and partners (some successfully referred, others not) who were not index individuals (circles); shading denotes whether the participant’s anti–hepatitis C virus status was unknown (white) negative (light gray) or positive (dark gray)

Figure 2
View larger version:
Figure 2

Main component of injection network, as rendered by a spring-embedder algorithm [33] with slight manual adjustments for clarity. Nodes denote index individuals (triangles) and successfully referred partners who were not index individuals (circles); shading denotes whether the participant’s age was 10–19 years (white) 20–29 years (light gray) 30–39 years (medium gray) or ⩾40 years (dark gray)

Figure 3
View larger version:
Figure 3

Phylogenetic tree based on nucleotide sequencing of the hepatitis C virus (HCV) envelope 1 gene and the HCV hypervariable region 1 gene, for 31 HCV-infected participants. Numbers denote participants, and letter suffixes denote the injection network component in which the participant was located (X = main component). Ovals surround transmission pairs, whose pairings were reliable at the 99% confidence level (as determined by bootstrapping [50]) and reflected very small DNA distances (see the “Transmission pairs” subsection of Results). (Scale bar, .05 DNA distance)

Figure 4
View larger version:
Figure 4

Risk behaviors in transmission pairs, as reported by 1 or both partners in the pair. Each pair consisted of a male and a female. Unless otherwise noted, reports refer to behaviors in which the pair engaged with each other during the specific recall period

Table 1
View larger version:
Table 1

Demographic and behavioral characteristics of participants

Case-control comparisons of injection risk behaviors Table 2 shows the comparisons between case subjects and control subjects on several injection risk behaviors with their interviewed injection partners. Case subjects and control subjects engaged in receptive syringe sharing and receptive injection risk with similar proportions of their partners. However, case subjects had nonsignificantly more HCV-infected partners, on average, and consequently tended to have had more HCV-infected partners with whom they engaged in receptive syringe sharing and receptive injection risk

Table 2
View larger version:
Table 2

Injection risk behaviors with injection partners: comparisons between case subjects and control subjects

Injection and sexual network structureIndex individuals recalled a mean of 18 injection partners (median, 12 [SD, 18]; interquartile range [IQR], 5–26; range, 1–74). Of the partners interviewed, 88% confirmed that they had injected with the index individual during the specific recall period, 19% reported syringe sharing with the index individual during that period, and 78% reported engaging in any injection risk with the index individual during that period. The overall reported injection network comprised 15 components, and the largest component included 78% (128/165) of all participants. The 5 next largest components each included 2%–4% of participants. Figure 1 depicts the largest component of the injection network after removal of uninterviewed partners who were mentioned only once or not uniquely identified. This component is dense and pervaded by cycles. Of the reported partnerships, 78% (146/188) were identified by anonymous code (when the partner was successfully referred), and the rest were identified by uncommon names (when the partner was not referred)

Forty-seven sexually active index individuals recalled a mean of 3.7 sex partners (median, 3 [SD, 3.8]; IQR, 1–4; range, 1–17). The corresponding sexual network comprised 45 components; the largest included only 1% of participants

Distribution of HCV infection and age across the injection networkHCV infection was evenly distributed across the injection network of participants (Geary’s C=0.80 [P>.05]; Moran’s I=.05 [P>.05]). Figure 1 shows this fairly uniform scattering of anti-HCV–positive IDUs in the main component, with the nodes (individuals) shaded by serostatus. Similar proportions of incident case subjects (81% [13/16]) and control subjects (74% [32/43]) were in the main component of the injection network (Pearson [ϕ] r .07; OR, 1.5 [P>.05]). The thorough mixing of HCV-infected injectors in this network is further evidenced by control subjects’ network proximity to HCV-infected participants. At least 65% (28/43) of control subjects had injected with anti-HCV–positive participants; 79% (34/43) were within 2 steps (partners of partners) of HCV-infected participants, and 86% (37/43) were within 3 steps of them. Participants tended to inject with individuals of similar age (Geary’s C=0.48 [P<.0001]; Moran’s I=.36 [P<.0001]). Figure 2 displays this moderate assortative mixing by age

Transmission pairsFigure 3 shows the phylogenetic tree for infected participants with sequenced isolates. Sixty-five percent of infected participants had genotype 1a, 23% had genotype 3a, 6% had genotype 1b, and 6% had genotype 2b. Four pairs of participants (8 individuals) had HCV infections that were genetically closely related (median [range] of DNA distances: E1 and HVR1 region, .016 [.010–.021]; E1 region, .008 [.003–.012]; HVR1 region, .049 [.022–.084]), indicating it was highly probable 1 in each pair infected the other or both were infected by another person. By comparison, 3 infected participants each provided 2 specimens during the study (testing interval, 69–323 days), and the DNA distances between their isolate pairs were similarly small (median [range]: E1/HVR1, .010 [.000–.0586]; E1, .005 [.000–.038]; HVR1, .030 [.000–.154]). Three transmission pairs involved an index individual and an injection partner, and the fourth pair was separated in the injection network by no more than 3 intervening participants. Figure 4 lists the reported risk behaviors the 3 former pairs engaged in with each other. One of 3 transmission partnerships involved only injection risk other than syringe sharing. In these 3 transmission pairs, no other interviewed partner of an index individual was HCV infected, preventing partnership-level comparisons. For the fourth pair, all 4 interviewed partners of 1 participant (a female index individual) in the pair denied both syringe sharing and having sex with the index individual, and 2 reported “indirect sharing” behaviors. In the cohort study, the index individual in this pair reported, for the specific recall period, 1 receptive syringe sharing partner, 3 cooker/cotton-sharing partners, and 1 male sex partner

Correspondence between genetic distance and networkdistanceFour genotypes were represented in the main component, and 2 genotypes were represented in a small component (figure 3). Among the 7 incident case subjects with genotype 1a who were in the main component, genetic distance and injection network (geodesic) distance were positively associated (r=.42 [P=.11]). This correlation was driven by the presence of 1 transmission pair; when this pair was excluded, the correlation dropped to .07

Previous SectionNext Section

Discussion

In our case-control study of HCV seroconversion and social networks in young drug injectors, nearly one-quarter of reported injection partnerships involved syringe sharing during a 12-month period, and almost four-fifths involved injection behavior that could transmit HCV. Incident case subjects and seronegative control subjects were demographically and behaviorally similar, including the proportion of injection partners with whom they engaged in syringe sharing and any injection risk behavior. Case subjects, however, had somewhat more HCV-infected partners than did control subjects and consequently engaged in injection risk behavior with a greater number and proportion of infected partners. The injection network was fairly connected, dense, and cyclic, but the sexual network was highly fragmented and was unable to serve as a scaffold for sustained transmission by itself. HCV-infected participants were relatively evenly distributed across the injection network, although injectors displayed moderate assortative mixing by age with their partners. As anticipated previously [2], most HCV-uninfected injectors were quite close, in network terms, to infected injectors, typically having injected with at least 1 HCV-infected partner during the preceding year. Four pairs of injectors had genetically closely related infections, indicating transmission either from 1 member of the pair to the other or from a third individual to both individuals. In 3 pairs, the linked infections were confirmed by reported injection contact between the 2 members of a pair. Transmission in at least 1 of the 4 pairs appeared to be due to sharing of injection equipment other than syringes. Apart from transmission pairs, though, network distance between incident case subjects and genetic distance between their HCV variants were uncorrelated

Our observation that sharing of injection equipment other than syringes was the probable mode of transmission in at least 1 transmission pair extends previous work that indicates that such “indirect sharing” is independently associated with HCV seroconversion in drug injectors [4, 5, 7, 8]. Indeed, in a late 1990s Seattle cohort of IDUs we found that the population attributable fraction of HCV infections due to shared cookers and/or cottons was 13% [7]

The injection network that we observed was mostly connected, dense, and cyclic, which comports with prior observations of needle sharing [9, 34], injection [12, 13], and combined injection and sexual networks [11, 35]. The economics, illegality, logistics, and social aspects of illicit drug injection probably contribute to such network structures. Although age is a moderate correlate of prevalent HCV infection in IDUs in the present (data not shown) and other studies [19], the assortative mixing by age in the injection network was not strong enough to prevent an almost even distribution of HCV-infected participants across the injection network. Because case subjects and control subjects were similar in their injection risk behavior with their partners overall, seroconversion was mostly an accident of network position—that is, injecting with more individuals who happened to be HCV infected. With turnover in injection partners, we expect most HCV-uninfected IDUs will eventually be in similar positions in the injection network and subsequently acquire HCV. Therefore, our data suggest that, without massive reductions in all injection risk behaviors, shattering of cohesive injection networks, and/or broad coverage with an effective vaccine, HCV will remain hyperendemic in similar populations of drug injectors

We found a larger fraction of closely related pairs of infections that were confirmed by reported injection contact between the persons involved (3/4) than Aitken et al. did (12/66) [12]. This difference may be partly due to our longer recall period for eliciting partners and sequencing envelope regions of the HCV genome that experience higher rates of mutation [36, 37] than the core and nonstructural protein 5a regions they sequenced [12]. We also found no correspondence between genetic and network distance in incident case subjects after excluding distances between the transmission pair of incident case subjects. This result resembles Aitken et al.’s [12] observation of no consistent relationship in prevalent cases, although they did not exclude closely related infections from analysis. Our focus on incident cases made the 2 measures more temporally comparable and presumably increased the chances of finding a meaningful correspondence. The lack of correspondence in both studies is likely the consequence of the multiplicity of circulating genotypes and variants throughout the injection network, incomplete network ascertainment [20, 38], spontaneous clearance of infection [39–41] after transmission to others, reinfection/superinfection/mixed infection [42–45], and/or possible changes over time in or rapid evolution of the dominant variant within infected individuals [46, 47]. Hence, genetic distance—except when very small—is probably a poor proxy for network distance

Our study has several limitations. The network data are incomplete, primarily because we did not elicit partners’ partners and lacked sufficient identifying information to “unduplicate” all reported partners and secondarily because of incomplete reporting, despite our use of supplementary elicitation techniques [20]. Therefore, the “true” injection network structure may be more connected, dense, and cyclic—and thus more likely to propagate infection—than that which we observed. The design-based incomplete ascertainment also prevented meaningful analysis of several positional and structural properties. Nevertheless, underascertainment of the network was independent of participants’ ages, behaviors, serostatuses, and HCV variants. We did not assess risk behaviors for all elicited partnerships, but only for those involving successful partner referral. For these partnerships, partners reported the frequency of risk behaviors in relative (i.e., “never,” “rarely,” “sometimes,” “usually,” “always”) rather than absolute terms, preventing more detailed dose-response analyses. Our analyses treated the injection and sexual networks as static entities rather than as the dynamic structures that they are [48], because we did not elicit specific partnership dates. Most partnerships, however, occurred within the same, relatively short time period. Moreover, the sample was fairly small, and we gathered no data on HIV status, anal sex, administering injections to/receiving injections from others, injection with previously discarded syringes, and nonsexual exposures other than illicit drug injection

With adjustments to address such limitations, the kind of research design we implemented—tracing risk networks of incident case subjects and control subjects, gathering detailed data on exposures within partnerships, and sequencing infected persons’ isolates—may serve as one model for investigating modes of infectious disease transmission and the corresponding social structural substrates. This approach combines advantages of other designs (prospective cohort, network description of population affected, contact tracing, and molecular analysis), bringing the investigation to levels that are crucial for understanding transmission—the dyad and network. Furthermore, this approach has the potential to narrow the investigative scope to particular events within dyads, if testing methods for detecting very recently acquired infections are used. Such integrated designs could be effectively employed to establish the key modes of transmission of a pathogen in settings where they have been studied only at the individual and ecological levels [49]

Previous SectionNext Section

Acknowledgments

We thank Susan Barkan, for assisting with project management; Stanley Brown, Heather Haynes, Tamarind Keating, Ben Masaoka, and Teresa Oakland, for assistance with data collection; and John M. Roberts Jr., for helpful discussions

Previous SectionNext Section

Footnotes

  • ↵Presented in part: 25th International Sunbelt Social Networks Conference, Redondo Beach, CA, 16–20 February 2005; Workshop on Network Epidemiology, Stockholm, 7 November 2005

    Potential conflicts of interest: none reported

    Financial support: Association of Schools of Public Health (grant S425-16/16); National Institute on Drug Abuse (grants DA08023, DA11447, and DA15026)

  • Received February 11, 2006.
  • Accepted March 9, 2006.
Previous SectionNext Section

Appendix Details of the Procedure

Validation of referred partnersWe took several steps to minimize the possibility that reported partnerships involving successfully referred partners were fabricated. First, for index individuals who indicated they could locate >5 partners, only 5 were randomly selected for referral. Therefore, such index individuals could not plan in advance who their fabricated partners would be. Second, we collected detailed physical/identifying information from index individuals on the partners whom they were to refer. We compared this information with the persons presenting as index individuals’ partners, and we rejected those whose physical/identifying information did not match that reported by the index individual (this occurred infrequently). Third, we asked referred partners how they got their vouchers and information about the persons who referred them. If persons presenting as partners either reported not knowing the index previously or reported information inconsistent with the description of the index, they were rejected

Laboratory proceduresOligonucleotide primers used in the first round of PCR amplification were 5′-GCGTCCGGGTTCTGGAAGACGGCGTGAACTATGCAACAGG-3′ (corresponding to nucleotides 802–841 of the HCV1 genome) and 5′-AGGCTTTCATTGCAGTTCAAGGCCGTGCTATTGATGTGCC-3′ (corresponding to nucleotides 1600–1639 of the HCV1 genome). The 805-bp product of the second round of PCR was generated using primers 5′-AAGACGGCGTGAACTATGCAACAGGGAACCTTCCTGGTTG-3′ (corresponding to nucleotides 821–856 of the HCV1 genome) and 5′-AGTTCAAGGCCGTGCTATTGATGTGCCAACTGCCGTTGGT-3′ (corresponding to nucleotides 1626–1587 of the HCV1 genome). PCR products were purified by agarose-gel electrophoresis with the Qiaex gel-purification kit (Qiagen). The purified PCR products were directly sequenced using the second-round-PCR primers

Identification of reported partnersWe identified index individuals and reported partners as uniquely as possible, by (1) participants’ unique anonymous codes and interviewers’ familiarity with participants’ identities; (2) nicknames that were mentioned multiple times and that we judged to be uncommon (e.g., names similar to “Seahorse,” “Mink,” and “Grime”) and therefore very likely to refer to the same individual in this local setting; (3) uncommon first names mentioned multiple times (< 0.5 persons expected to have that first name among the total number of partners mentioned by index individuals, as estimated from the Social Security Administration’s first-name database stratified by decade of birth [http://www.ssa.gov/OACT/babynames/], weighted by the frequency of interviewed partners by birth decade, and accounting for whether a name was used for females, males, or both sexes); and (4) any links between these types of information in the data

Previous Section
 

References

  1. ↵
    1. Crofts N,
    2. Dore G,
    3. Locarnini S
    1. Hocking J,
    2. Crofts N,
    3. Aitken CK,
    4. MacDonald M
    . Epidemiology among injecting drug users. In: Crofts N, Dore G, Locarnini S, editors. Hepatitis C: an Australian perspective. Melbourne: IP Communications; 2001. p. 260-95.
  2. ↵
    1. Hagan H
    . Hepatitis C virus transmission dynamics in injection drug users. Subst Use Misuse 1998;33:1197-212.
    MedlineWeb of Science
  3. ↵
    1. Alter MJ,
    2. Moyer LA
    . The importance of preventing hepatitis C virus infection among injection drug users in the United States. J Acquir Immune Defic Syndr Hum Retrovirol 1998;18 Suppl 1:S6-10.
    Medline
  4. ↵
    1. Thorpe LE,
    2. Ouellet LJ,
    3. Hershow R,
    4. et al
    . Risk of hepatitis C virus infection among young adult injection drug users who share injection equipment. Am J Epidemiol 2002;155:645-53.
    Abstract/FREE Full Text
  5. ↵
    1. Hahn JA,
    2. Page-Shafer K,
    3. Lum PJ,
    4. et al
    . Hepatitis C virus seroconversion among young injection drug users: relationships and risks. J Infect Dis 2002;186:1558-64.
    Abstract/FREE Full Text
  6. ↵
    1. Garten RJ,
    2. Lai S,
    3. Zhang J,
    4. et al
    . Rapid transmission of hepatitis C virus among young injecting heroin users in Southern China. Int J Epidemiol 2004;33:182-8.
    Abstract/FREE Full Text
  7. ↵
    1. Hagan H,
    2. Thiede H,
    3. Weiss NS,
    4. Hopkins SG,
    5. Duchin JS
    . Sharing of drug preparation equipment as a risk factor for hepatitis C. Am J Public Health 2001;91:42-6.
    Abstract
  8. ↵
    1. Lucidarme D,
    2. Bruandet A,
    3. Ilef D,
    4. et al
    . Incidence and risk factors of HCV and HIV infections in a cohort of intravenous drug users in the North and East of France. Epidemiol Infect 2004;132:699-708.
    CrossRefMedline
  9. ↵
    1. Rothenberg RB,
    2. Potterat JJ,
    3. Woodhouse DE,
    4. Muth SQ,
    5. Darrow WW
    . Social network dynamics and HIV transmission. AIDS 1998;12:1529-36.
    CrossRefMedlineWeb of Science
  10. ↵
    1. Woodhouse DE,
    2. Rothenberg RB,
    3. Potterat JJ,
    4. et al
    . Mapping a social network of heterosexuals at high risk for HIV infection. AIDS 1994;8:1331-6.
    CrossRefMedlineWeb of Science
  11. ↵
    1. Friedman SR,
    2. Neaigus A,
    3. Jose B,
    4. et al
    . Sociometric risk networks and risk for HIV infection. Am J Public Health 1997;87:1289-96.
    Abstract/FREE Full Text
  12. ↵
    1. Aitken CK,
    2. McCaw RF,
    3. Bowden DS,
    4. et al
    . Molecular epidemiology of hepatitis C virus in a social network of injection drug users. J Infect Dis 2004;190:1586-95.
    Abstract/FREE Full Text
  13. ↵
    1. Willems JCEW,
    2. Iguchi MY,
    3. Lidz V,
    4. Bux DAJ
    . Change in drug-using networks of injecting drug users during methadone treatment: a pilot study using snowball recruitment and intensive interviews. Subst Use Misuse 1997;32:1539-54.
    MedlineWeb of Science
  14. ↵
    1. Rothenberg RB,
    2. Sterk C,
    3. Toomey KE,
    4. et al
    . Using social network and ethnographic tools to evaluate syphilis transmission. Sex Transm Dis 1998;25:154-60.
    MedlineWeb of Science
  15. ↵
    1. Potterat JJ,
    2. Phillips-Plummer L,
    3. Muth SQ,
    4. et al
    . Risk network structure in the early epidemic phase of HIV transmission in Colorado Springs. Sex Transm Infect 2002;78:i159-63.
    Abstract/FREE Full Text
  16. ↵
    1. Potterat JJ,
    2. Muth SQ,
    3. Rothenberg RB,
    4. et al
    . Sexual network structure as an indicator of epidemic phase. Sex Transm Infect 2002;78:i152-8.
    Abstract/FREE Full Text
  17. ↵
    1. Rothenberg R,
    2. Kimbrouh L,
    3. Lewis-Hardy R,
    4. et al
    . Social network methods for endemic foci of syphilis. Sex Transm Dis 2000;27:12-8.
    MedlineWeb of Science
  18. ↵
    1. Potterat JJ,
    2. Rothenberg RB,
    3. Muth SQ
    . Network structural dynamics and infectious disease propagation. Int J STD AIDS 1999;10:182-5.
    Abstract/FREE Full Text
  19. ↵
    1. Hagan H,
    2. Thiede H,
    3. Des Jarlais DC
    . Hepatitis C virus infection among injection drug users: survival analysis of time to seroconversion. Epidemiology 2004;15:543-9.
    CrossRefMedlineWeb of Science
  20. ↵
    1. Brewer DD,
    2. Garrett SB
    . Evaluation of interviewing techniques to enhance recall of sexual and drug injection partners. Sex Transm Dis 2001;28:666-77.
    MedlineWeb of Science
  21. ↵
    1. Alter MJ,
    2. Kuhnert WL,
    3. Finelli L
    . Guidelines for laboratory testing and result reporting of antibody to hepatitis C virus. Centers for Disease Control and Prevention. MMWR Recomm Rep 2003;52:1-13, 15.
    Medline
  22. ↵
    1. Sullivan DG,
    2. Kim SS,
    3. Wilson JJ,
    4. Stehman-Breen C,
    5. Gretch DR
    . Investigating hepatitis C virus heterogeneity in a high prevalence setting using heteroduplex tracking analysis. J Virol Methods 2001;96:5-16.
    CrossRefMedlineWeb of Science
  23. ↵
    1. Sullivan DG,
    2. Wilson JJ,
    3. Carithers RL Jr.,
    4. Perkins JD,
    5. Gretch DR
    . Multigene tracking of hepatitis C virus quasispecies after liver transplantation: correlation of genetic diversification in the envelope region with asymptomatic or mild disease patterns. J Virol 1998;72:10036-43.
    Abstract/FREE Full Text
  24. ↵
    1. Higgins DG,
    2. Bleasby AJ,
    3. Fuchs R
    . CLUSTAL V: improved software for multiple sequence alignment. Comput Appl Biosci 1992;8:189-91.
    Abstract/FREE Full Text
  25. ↵
    1. Felsentein J
    . PHYLIP (Phylogeny Inference Package) version 3.5c. Seattle: University of Washington; 1993.
  26. ↵
    1. Saitou N,
    2. Nei M
    . The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 1987;4:406-25.
    Abstract
  27. ↵
    1. Wasserman RF,
    2. Faust K
    . Social network analysis: methods and applications. Cambridge: Cambridge University Press; 1994.
  28. ↵
    1. Cliff A,
    2. Ord JK
    . Spatial processes: models and applications. London: Pion; 1981.
  29. ↵
    1. Geary RC
    . The contiguity ratio and statistical mapping. Incorporated Statistician 1954;5:115-45.
    CrossRef
  30. ↵
    1. Moran PAP
    . Notes on continuous stochastic phenomena. Biometrika 1950;37:17-23.
    FREE Full Text
  31. ↵
    1. Hubert LJ,
    2. Schultz J
    . Quadratic assignment as a general data analysis strategy. Br J Math Stat Psychol 1975;29:190-241.
    Web of Science
  32. ↵
    1. Borgatti SP,
    2. Everett MG,
    3. Freeman LC
    . UCINET for Windows: software for social network analysis. Harvard, MA: Analytic Technologies; 2002.
  33. ↵
    1. Borgatti SP
    . NetDraw: graph visualization software. Harvard, MA: Analytic Technologies; 2002.
  34. ↵
    1. Rothenberg R,
    2. Baldwin J,
    3. Trotter R,
    4. Muth S
    . The risk environment for HIV transmission: results from the Atlanta and Flagstaff network studies. J Urban Health 2001;78:419-32.
    Medline
  35. ↵
    1. Rothenberg RB,
    2. Long DM,
    3. Sterk CE,
    4. et al
    . The Atlanta urban networks study: a blueprint for endemic transmission. AIDS 2000;14:2191-200.
    CrossRefMedlineWeb of Science
  36. ↵
    1. Rispeter K,
    2. Lu M,
    3. Behrens SE,
    4. Fumiko C,
    5. Yoshida T
    . Hepatitis C virus variability: sequence analysis of an isolate after 10 years of chronic infection. Virus Genes 2000;21:179-88.
    CrossRefMedlineWeb of Science
  37. ↵
    1. Christie JM,
    2. Chapel H,
    3. Chapman RW,
    4. Rosenberg WM
    . Immune selection and genetic sequence variation in core and envelope regions of hepatitis C virus. Hepatology 1999;30:1037-44.
    CrossRefMedlineWeb of Science
  38. ↵
    1. Brewer DD,
    2. Garrett SB,
    3. Kulasingam S
    . Forgetting as a cause of incomplete reporting of sexual and drug injection partners. Sex Transm Dis 1999;26:166-76.
    MedlineWeb of Science
  39. ↵
    1. Keating S,
    2. Coughlan S,
    3. Connell J,
    4. Sweeney B,
    5. Keenan E
    . Hepatitis C viral clearance in an intravenous drug-using cohort in the Dublin area. Ir J Med Sci 2005;174:37-41.
    MedlineWeb of Science
  40. ↵
    1. Beld M,
    2. Penning M,
    3. van Putten M,
    4. et al
    . Quantitative antibody responses to structural (Core) and nonstructural (NS3, NS4, and NS5) hepatitis C virus proteins among seroconverting injecting drug users: impact of epitope variation and relationship to detection of HCV RNA in blood. Hepatology 1999;29:1288-98.
    CrossRefMedlineWeb of Science
  41. ↵
    1. Villano SA,
    2. Vlahov D,
    3. Nelson KE,
    4. Cohn S,
    5. Thomas DL
    . Persistence of viremia and the importance of long-term follow-up after acute hepatitis C infection. Hepatology 1999;29:908-14.
    CrossRefMedlineWeb of Science
  42. ↵
    1. Herring BL,
    2. Page-Shafer K,
    3. Tobler LH,
    4. Delwart EL
    . Frequent hepatitis C virus superinfection in injection drug users. J Infect Dis 2004;190:1396-403.
    Abstract/FREE Full Text
  43. ↵
    1. Herring BL,
    2. Tsui R,
    3. Peddada L,
    4. Busch M,
    5. Delwart EL
    . Wide range of quasispecies diversity during primary hepatitis C virus infection. J Virol 2005;79:4340-6.
    Abstract/FREE Full Text
  44. ↵
    1. Backmund M,
    2. Meyer K,
    3. Edlin BR
    . Infrequent reinfection after successful treatment for hepatitis C virus infection in injection drug users. Clin Infect Dis 2004;39:1540-3.
    Abstract/FREE Full Text
  45. ↵
    1. Dalgard O
    . Follow-up studies of treatment for hepatitis C virus infection among injection drug users. Clin Infect Dis 2005;40 Suppl 5:S336-8.
    CrossRefMedlineWeb of Science
  46. ↵
    1. Aitken C,
    2. McCaw R,
    3. Jardine D,
    4. et al
    . Change in hepatitis C virus genotype in injecting drug users. J Med Virol 2004;74:543-5.
    CrossRefMedlineWeb of Science
  47. ↵
    1. Dove L,
    2. Phung Y,
    3. Bzowej N,
    4. Kim M,
    5. Monto A
    . Viral evolution of hepatitis C in injection drug users. J Viral Hepat 2005;12:574-83.
    CrossRefMedlineWeb of Science
  48. ↵
    1. Moody J
    . The importance of relationship timing in diffusion: indirect connectivity and STD infection risk. Soc Forces 2002;81:25-56.
    Abstract/FREE Full Text
  49. ↵
    1. Brewer DD,
    2. Rothenberg RB,
    3. Potterat JJ,
    4. Brody S,
    5. Gisselquist D
    . HIV epidemiology in sub-Saharan Africa: rich in conjecture, poor in data. Int J STD AIDS 2004;15:63-5.
    FREE Full Text
  50. ↵
    1. Felsentein J
    . Confidence limits on phylogenesis: an approach using the bootstrap. Evolution 1985;39:783-91.
    CrossRefWeb of Science
| Table of Contents

This Article

  1. J Infect Dis. (2006) 194 (6): 764-772. doi: 10.1086/505585
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most