Abstract
We examine injecting and sexual risk correlates of hepatitis B (HBV) and hepatitis C (HCV) seroprevalence among new injecting drug users (IDUs) (age 18–30 years, injecting ≤6 years). Participants were interviewed/serotested (HIVab, HBVcAb, HCVab) in New York City, 2/1999–2/2003. Gender-stratified, multivariate logistic regression was conducted. Participants (N=259) were: 68% male; 81% white. Women were more likely to test HCV seropositive (42% vs. 27%) and men HBV seropositive (24% vs. 12%); HIV seroprevalence was low (3%). Among both men and women, HBV seropositivity was associated with ever selling sex, and HCV seropositivity with ever having had infected (HIV, HBV or HCV) sex partners (among those ever sharing injecting equipment). Among women only, HBV seropositivity was associated with ever having had infected sex partners (regardless of ever sharing injecting equipment), and HCV seropositivity with ≥300 lifetime drug injections. Among men only, HCV seropositivity was associated with ≥40 lifetime number of sex partners (among those never sharing injecting equipment). In this new IDU sample, HBV and HCV seroprevalence differed by gender and were considerably higher than HIV seroprevalence. Early interventions, targeting injecting and sexual risks and including HBV vaccination, are needed among new IDUs to prevent HBV, HCV and, potentially, HIV epidemics.
Keywords: HIV, HBV, HCV, drug injectors, injecting risk, sexual risk
1. Introduction
Many new injecting drug users (IDUs) are at risk of infection with HIV, the hepatitis B virus (HBV), and the hepatitis C virus (HCV) soon after initiating injecting (Friedman et al., 1989; Neaigus et al., 1996; Garfein et al., 1996; Hahn et al., 2001). However, while most new IDUs are uninfected with HIV in the first few years after initiating injecting, infection with HBV and HCV is widespread (Des Jarlais et al., 2003; European Centre for the Epidemiological Monitoring of AIDS. 2003). Not only are HBV and HCV more efficiently transmitted than HIV (Ippolito et al., 1999; Puro et al., 2001), but among new IDUs the probability of exposure to HBV and HCV is generally greater than the probability of exposure to HIV (Des Jarlais et al., 2003). Nevertheless, many of the risk factors for infection with HBV and HCV are also risk factors for infection with HIV. Examining the risk factors for HBV and HCV infection in low HIV-prevalence populations of new IDUs can provide sentinel indicators of HIV risk prior to the outbreak of an HIV epidemic, which can assist in developing early interventions.
Although new IDUs are at both injecting and sexual risk of infection with HIV and HBV, they have increasingly gained access to legal sources of sterile syringes and many have reduced the direct sharing of syringes (Des Jarlais et al., 2000; Emmanuelli and Desenclos, 2005). Nonetheless, sharing drug preparation equipment, such as cookers, is still common (Hagan et al., 2001). Sexual risk has also been more difficult to change (Neaigus et al., 1990; Semaan et al., 2002; Strathdee and Sherman, 2003) and new IDUs remain at sexual risk for HIV and HBV infection. New IDUs may also be at sexual risk of HCV infection─while HCV is mainly transmitted parenterally through blood-to-blood transmission, several studies have found evidence for the sexual transmission of HCV (Thomas et al., 1994; Ndimbie et al., 1996; Rooney and Gilson, 1998; Wejstal, 1999).
In the following, injecting and sexual risk correlates of HBV and HCV seroprevalence are examined in a low HIV-seroprevalence population of new IDUs. Since becoming infected requires both risk behaviors and exposure to infectious risk network members, i.e., those with whom new IDUs inject drugs and/or engage in sex (Neaigus et al., 1994), the analysis considers both risk behaviors and the characteristics of risk networks. Moreover, in addition to examining the main effects of injecting and sexual risks for infection, the overlap of injecting and sexual risks for each pathogen is analyzed by assessing interactions between injecting and sexual risk. The analysis is also stratified by gender, since women may become infected at a faster rate than men soon after initiating injecting. This may reflect both their increased risk of being exposed to older injecting and/or sex partners, who may be more likely to be infectious for HIV and hepatitis B or C (Nelson et al., 1995; Friedman et al., 1999; Doherty et al., 2000), as well as normative factors and social network influence from IDU sex partners that may increase female new IDUs’ risk of sharing drug use equipment (Frajzyngier et al., in press; Neaigus et al. 1995; Evans et al., 2003), and possibly of engaging in unprotected sex. A gender stratified analysis is also appropriate because the risk factors for infection may differ by gender (Solomon et al., 1993).
2. Methods
2.1 Recruitment and procedures
Between February 1999 and February 2003, cross-sectional data were obtained from new IDUs recruited in the Lower East Side/East Village area of New York City for a study of behavioral and network risks for infection with HIV, HBV and HCV. Participants were non-drug-treatment recruited using targeted sampling, street outreach and chain-referral methods (Watters and Biernacki, 1989; Heckathorn, 1997; Sifaneck and Neaigus, 2001). Eligible participants were between 18 and 30 years of age, had initiated drug injecting within the prior six years, and had injected drugs (heroin, cocaine, or methamphetamines) within the prior 30 days. Drug use was verified through urine drug toxicology tests (Roche Abuscreen OnTrak) or a hair assay (radioimmunoassay by Psychemedics). To verify recent injecting, arms or other visible body sites were inspected for injecting marks. Those without visible marks indicating recent injecting were excluded from the study.
After providing their informed consent, eligible participants were administered, in private, a structured face-to-face interview, which was conducted in-person by a trained interviewer at a research site in the recruitment area. Following the interview participants were pre-test counseled for HIV, HBV and HCV and had blood specimens collected by a trained phlebotomist/counselor. Drug treatment and health and social service referrals were available. Those who returned for their test results and were HBV seronegative were referred to the New York City Department of Health HBV vaccination clinics. Participants were paid a small monetary incentive for their time and travel expenses. All procedures involving human subjects were reviewed by the Institutional Review Board at National Development and Research Institutes.
The 6-year cut-off point to define “new injectors” is based on prior research in New York City and elsewhere which has found that among IDUs HIV prevalence is lower for those injecting for 6 or fewer years (Friedman et al., 1989; Robles et al., 1992; Neaigus et al., 1996; Garfein et al., 1996). Thus new injectors represent a susceptible subpopulation of IDUs in which HIV infection may not be widespread, but in which the prevalence of HBV and HCV, and the risk factors for blood-borne and sexually transmitted infections in general, including HIV, may be considerable.
2.2 Measurement and variables
Participants were interviewed about their sociodemographic (age, gender, and race/ethnicity) and other background characteristics, whether they were in drug treatment (methadone maintenance, detoxification or other drug treatment), and their injecting and sexual risk behaviors and risk networks. Injecting and sexual risk variables measure lifetime risk, except for the characteristics of injecting risk networks (last 30 days) and sexual risk behaviors (last six months), for which lifetime data were not collected.
Lifetime injecting risk behaviors included: lifetime frequency of injecting; ever injecting specific drugs (e.g., heroin); ever engaging in receptive syringe sharing; ever sharing a cooker, cotton or rinse-water; ever engaging in receptive syringe-mediated drug-sharing; ever renting a syringe; ever injecting at a commercial multi-user setting (a “shooting gallery”); and ever sharing any injecting equipment (including both syringes and non-syringe injecting equipment used for drug preparation, such as cookers). Injecting risk network variables (based on participants’ responses to questions about the specific people with whom they injected in the prior 30 days) included injecting network size (the number of 30-day injecting risk network members) and having injecting risk network members who, for each characteristic separately: were infected with HIV, HBV or HCV; were men-who-have-sex-with-men (MSM); were known casually; were five or more years older than the participant; and who knew each other (a measure of social network density). This latter variable was derived by asking participants, for each pair of nominated injecting risk network members, whether the members knew each other.
Sexual risk variables included: engaging in unprotected vaginal sex; engaging in any anal sex; the lifetime number of sex partners; ever having sex risk network members who, for each characteristic separately, were: infected with HIV, HBV or HCV; were IDUs; and were MSM.
Ever selling sex for money or drugs, being a MSM, being a woman-who-has-sex-with-women (WSW), and ever having a sexually transmitted disease (STD) were treated as proxy variables for direct injecting or sexual risk exposures.
Several variables were treated as possible confounders of direct injecting and sexual risk exposures, including: the number of years since initiating injecting; age at injecting initiation; calendar year of injecting initiation; the number of years since initiating sexual activity; age at initiating sexual activity, non-white race/ethnicity; age; and ever being incarcerated in jail or a detention center.
Categorical variables were dichotomized as no/yes. Continuous and interval level variables with skewed distributions were dichotomized at the median.
2.3 Laboratory analyses
Blood specimens were tested for HIV-1 antibody (EIA with Western Blot confirmation (Abbott)), antibody to the hepatitis B core antigen (Abbott hepatitis B virus core antigen (recombinant) CORZYME immunoassay), and HCV antibody (Abbott HCV EIA 2.0). A positive test for the HBV core antibody indicates whether an individual has ever been infected with the pathogen and does not, by itself, indicate current infection. The HCV enzyme immunoassay test used in this study has a high positive predictive value of 100% when confirmed by supplemental testing (Abbott EIA 2.0 and HCV MATRIX) (Des Jarlais et al., 2003).
2.4 Statistical analyses
Statistical analyses were stratified by gender and conducted separately for each infection. Of 259 participants, five women and six men self-reported being vaccinated against HBV and were excluded from the analysis of the correlates of ever being HBV infected because it was not possible to determine if they were immune against HBV through being vaccinated (which requires the HBV surface antigen and surface antibody tests as well as the core antibody test). Since only seven participants were HIV infected, the analysis of the correlates of HIV infection was limited to assessing univariate associations.
The analysis of the sample characteristics used the chi-square test or Fisher’s exact test for categorical variables, and the t-test for continuous variables. The univariate analyses of the risk correlates used the Wald chi-square to test the significance of individual logistic regression coefficients. Adjusted odds ratios (AOR) and 95% confidence intervals (95% CI) were estimated using multivariate logistic regression.
The multivariate models were developed using a sequential approach. First, injecting and sexual risk exposure variables that were p < 0.20 in univariate analysis were entered into multivariate models using backward elimination, with variables that were significant at p < 0.05 retained in the models. Second, possible confounders were tested by using stepwise selection against the retained injecting and sexual risk exposure variables. Finally, simultaneous multivariate analyses were conducted that included significant (p < 0.05) injecting and sexual risk exposure variables from the first step, significant confounders from the second step, and interactions between the retained sexual risk variables and ever sharing any injecting equipment. The interaction variables were created by stratifying the retained sexual risk variables by ever sharing any injecting equipment. If the sexual risk correlates remained significant among both those who ever shared any injecting equipment and those who did not, only the main effects for the sexual risk correlates are included in the final models. Statistical significance is two-sided at p < 0.05. All analyses were conducted using SAS v9.
3. Results
3.1 Sample characteristics
The majority (68%) of the 259 participants were male, and most (81%) self-reported being of white race/ethnicity. (Table 1) Males were significantly older than females. Approximately half were low income, homeless, and had not graduated high school. Two thirds had ever been incarcerated, with men significantly more likely to report incarceration. Few were currently in drug treatment.
TABLE 1.
Sociodemographic characteristics, drug use and sex risk history, and baseline seroprevalence of HIV, HBV, and HCV among new drug injectors 18–30 years old, New York City, February 1999 – February 2003
| Characteristics | Total N (%) | Females N (%) | Males N (%) |
|---|---|---|---|
| Total | 259 (100) | 83 (100) | 176 (100) |
| Race/ethnicity | |||
| African-American/Black | 5 (1.9) | 3 (3.6) | 2 (1.1) |
| Hispanic | 30 (11.6) | 9 (10.8) | 21 (11.9) |
| White | 209 (80.7) | 65 (78.3) | 144 (81.8) |
| Other | 15 (5.8) | 6 (7.2) | 9 (5.1) |
| Age (mean (SD)) | 22.8 (3.3) | 21.9 (3.2) | 23.3 (3.2) ** |
| Income in past 6 months less than $5000 | 129 (49.8) | 45 (54.2) | 84 (47.7) |
| Currently homeless | 147 (56.8) | 50 (60.2) | 97 (55.1) |
| High school graduate | 125 (48.3) | 44 (53.0) | 81 (46.0) |
| Ever incarcerated in jail or a detention center | 169 (65.3) | 44 (53.0) | 125 (71.0) ** |
| Currently in drug treatment | 18 (6.9) | 4 (4.8) | 14 (8.0) |
| Years since initiating injecting (mean (SD)) | 2.9 (1.7) | 2.8 (1.6) | 3.0 (1.8) |
| Age at initiating injecting (mean (SD)) | 19.4 (3.6) | 18.7 (3.6) | 19.8 (3.6) ** |
| Ever injected heroin | 244 (94.2) | 77 (92.8) | 167 (94.9) |
| Ever injected cocaine | 176 (68.0) | 57 (68.7) | 119 (67.6) |
| Ever injected speedball | 153 (59.1) | 48 (57.8) | 105 (59.7) |
| Ever injected crack | 57 (22.0) | 16 (19.3) | 41 (23.3) |
| Ever injected methamphetamines | 30 (11.6) | 10 (12.0) | 20 (11.4) |
| Ever engaged in receptive syringe sharing | 88 (34.0) | 30 (36.1) | 58 (33.0) |
| Ever shared cookers/cotton/rinse-water | 129 (50.0) | 48 (57.8) | 81 (46.0) |
| Ever receptive syringe-mediated drug-sharing (e.g., backloading) | 59 (22.8) | 19 (22.3) | 40 (22.7) |
| Ever rented syringes | 8 (3.0) | 1 (1.2) | 7 (4.0) |
| Ever injected at a shooting gallery | 30 (11.6) | 7 (8.4) | 23 (13.1) |
| Ever shared any injecting equipment | 142 (54.8) | 51 (61.4) | 91 (51.7) * |
| Age at first sex (mean (SD)) | 14.3 (2.5) | 14.5 (2.2) | 14.2 (2.6) |
| Ever same gender sexual experience | 66 (25.5) | 40 (48.2) | 26 (14.8) ** |
| Ever sold sex for money or drugs | 28 (10.8) | 15 (18.1) | 13 (7.4) ** |
| HIV seropositive | 7 (2.7) | 3 (3.6) | 4 (2.3) |
| HBV seropositive | 50 (20.2) | 9 (11.5) | 41 (24.1) ** |
| HCV seropositive | 83 (32.0) | 35 (42.2) | 48 (27.3) ** |
| HIV/HBV co-seropositive | 4 (1.5) | 1 (1.2) | 3 (1.7) |
| HIV/HCV co-seropositive | 4 (1.5) | 1 (1.2) | 3 (1.7) |
| HBV/HCV co-seropositive | 34 (13.1) | 8 (9.6) | 26 (14.8) |
| HIV/HBV/HCV co-seropositive | 3 (1.2) | 1 (1.2) | 2 (1.1) |
| HIV ever infected (self-report) | 3 (1.2) | 1 (1.2) | 2 (1.1) |
| HBV ever infected (self-report) | 20 (7.7) | 4 (4.8) | 16 (9.1) |
| HCV ever infected (self-report) | 33 (12.7) | 16 (19.3) | 17 (9.7) |
| Vaccinated against HBV (self-report) | 11 (4.2) | 5 (6.1) | 6 (1.0) |
p<0.20
p < 0.05
On average, participants had initiated injecting 2.9 years prior to the interview and were 19.4 years of age at initiation; females initiated injecting at a significantly younger age. Almost all had ever injected heroin, while two-thirds had ever injected cocaine and over half speedball (a mixture of cocaine and heroin). Many had engaged in unsafe injecting, with a third ever engaging in receptive syringe sharing, and over half ever sharing any injecting equipment. The mean age of sexual initiation was 14.3 years. While a quarter self-reported same gender sexual experience, women were more likely to do so, and were also more likely to report ever selling sex. Other background characteristics, drug use behaviors and injecting risk behaviors are shown in Table 1.
Seven (2.7%) participants (three women and four men) tested HIV seropositive, while 20% and 32%, respectively, were seropositive for HBV and HCV, with men significantly more likely to test HBV seropositive, and women HCV seropositive. The most frequently co- occurring seropositive tests were for HBV and HCV. Among the seropositive, those who self reported ever being infected included 3 of 7 (43%) who tested HIV seropositive, 13 of 50 (26%) who tested HBV seropositive, and 31 of 83 (37%) who tested HCV seropositive.
3.2 Correlates of HIV, HBV and HCV seroprevalence
The univariate results of the correlates of HIV, HBV and HCV seroprevalence are shown in Table 2 (women) and Table 3 (men). In Tables 2 and 3 only variables that were significant at p < 0.20 for any of the three viruses are shown. Where the odds ratios could not be calculated because of zero cells or quasi-complete separation, the significance level is shown but the odds ratios are not applicable (N/A). Table 4 shows the results of the multivariate analysis of HBV and HCV seroprevalence, stratified by gender. For variables that are displayed in the tables as inversely associated, we have reported the results in the direction of an association with testing seropositive.
TABLE 2.
Univariate analysis of HIV, hepatitis B (HBV), and hepatitis C (HCV) seroprevalence among female new drug injectors, 18–30 years old, New York City, February 1999 – February 2003; p<0.20 for any infection
| HIV+ | HBV+ | HCV+ | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Variables | Total N | % | OR | 95%CI | % | OR | 95%CI | % | OR | 95%CI |
| Total | 83 | 3.6 | 11.5 | 42.2 | ||||||
| Injecting risk behaviors (lifetime) | ||||||||||
| Lifetime frequency of injecting | ||||||||||
| < 300 times | 30 | 3.3 | Ref | 6.9 | Ref | 16.7 | Ref | |||
| ≥ 300 times | 53 | 3.8 | 1.1 | 0.1–13.1 | 14.3 | 2.2 | 0.4–11.6 | 56.6 | 6.5 | 2.2–19.6** |
| Ever injected speedball | ||||||||||
| no | 35 | 8.6 | 14.7 | Ref | 42.9 | Ref | ||||
| yes | 48 | 0.0 | N/A* | 9.1 | 0.6 | 0.1–2.4 | 41.7 | 0.9 | 0.4–2.3 | |
| Ever injected crack | ||||||||||
| no | 67 | 4.5 | 14.3 | 41.8 | Ref | |||||
| yes | 16 | 0.0 | N/A | 0.0 | N/A* | 43.8 | 1.1 | 0.4–3.3 | ||
| Ever injected methamphetamines | ||||||||||
| no | 73 | 4.1 | 13.0 | 38.4 | Ref | |||||
| yes | 10 | 0.0 | N/A | 0.0 | N/A | 70.0 | 3.8 | 0.9–15.7* | ||
| Ever receptive syringe-mediated drug-sharing (e.g., backloading) | ||||||||||
| no | 64 | 3.1 | Ref | 8.2 | Ref | 39.1 | Ref | |||
| yes | 19 | 5.3 | 1.7 | 0.1–20.1 | 23.5 | 3.4 | 0.8–14.6* | 52.6 | 1.7 | 0.6–4.9 |
| Injecting risk networks (last 30 days) | ||||||||||
| Injected with people who are MSM | ||||||||||
| no | 76 | 1.3 | Ref | 11.3 | Ref | 42.1 | Ref | |||
| yes | 7 | 28.6 | 30.0 | 2.3–390** | 14.3 | 1.3 | 0.1–12.3 | 42.9 | 1.0 | 0.2–4.9 |
| Injected with people who are 5 or more years older | ||||||||||
| no | 58 | 3.4 | Ref | 9.1 | Ref | 36.2 | Ref | |||
| yes | 25 | 4.0 | 1.2 | 0.1–13.5 | 17.4 | 2.1 | 0.5–8.7 | 56.0 | 2.2 | 0.9–5.8* |
| Sexual risk behaviors (last 6 months) | ||||||||||
| Any anal sex | ||||||||||
| no | 65 | 3.1 | Ref | 13.1 | Ref | 50.0 | Ref | |||
| yes | 18 | 5.6 | 1.8 | 0.2–21.7 | 5.9 | 0.4 | 0.0–3.6 | 40.0 | 1.5 | 0.5–4.3* |
| Sexual risk networks (lifetime) | ||||||||||
| Lifetime number of sex partners | ||||||||||
| < 40 | 70 | 1.4 | Ref | 9.2 | Ref | 41.4 | Ref | |||
| ≥ 40 | 13 | 15.4 | 12.5 | 1.05–150** | 23.1 | 3.0 | 0.6–13.8* | 46.2 | 1.2 | 0.4–4.0 |
| Ever had IDU sex partner | ||||||||||
| no | 10 | 20.0 | Ref | 10.0 | Ref | 40.0 | Ref | |||
| yes | 73 | 1.4 | 0.1 | 0.0–0.7** | 11.8 | 1.2 | 0.1–10.8 | 42.5 | 1.1 | 0.3–4.3 |
| Ever had sex partner with HIV or HBV or HCV infection | ||||||||||
| no | 60 | 5.0 | 7.0 | Ref | 31.7 | Ref | ||||
| yes | 23 | 0.0 | N/A | 23.8 | 4.1 | 0.99–17.3* | 69.6 | 4.9 | 1.7–14.0** | |
| Injecting or sexual risk exposure proxy variables | ||||||||||
| Ever sold sex | ||||||||||
| no | 68 | 0.0 | 7.8 | Ref | 36.8 | Ref | ||||
| yes | 15 | 20.0 | N/A** | 28.6 | 4.7 | 1.1–20.6** | 66.7 | 3.4 | 1.1–11.2** | |
| WSW | ||||||||||
| no | 43 | 0.0 | 7.3 | Ref | 48.8 | Ref | ||||
| yes | 40 | 7.5 | N/A* | 16.2 | 2.4 | 0.6–10.6 | 35.0 | 0.6 | 0.2–1.4 | |
| Reported ever having an STD | ||||||||||
| no | 47 | 0.0 | 8.7 | Ref | 42.6 Ref | |||||
| yes | 36 | 8.3 | N/A* | 15.6 | 1.9 | 0.5–7.9 | 41.7 | 1.0 | 0.4–2.3 | |
| Possible confounder variables | ||||||||||
| Years since injecting initiation > 2 | ||||||||||
| no | 45 | 4.4 | Ref | 14.3 | Ref | 35.6 | Ref | |||
| yes | 38 | 2.6 | 0.6 | 0.1–6.7 | 8.3 | 0.5 | 0.1–2.4 | 50.0 | 1.8 | 0.8–4.4* |
| Non-white race/ethnicity | ||||||||||
| no | 65 | 1.5 | Ref | 13.1 | Ref | 46.2 | Ref | |||
| yes | 18 | 11.1 | 8.0 | 0.7–93.8* | 5.9 | 0.4 | 0.0–3.6 | 27.8 | 0.4 | 0.1–1.4* |
| Age <20 years | ||||||||||
| no | 60 | 3.3 | Ref | 14.3 | Ref | 48.3 | Ref | |||
| yes | 23 | 4.3 | 1.3 | 0.1–15.3 | 4.5 | 0.3 | 0.0–2.4 | 26.1 | 0.4 | 0.1–1.1* |
| Interaction variables included in the final model(in bold) | ||||||||||
| “Ever shared any injecting equipment” and “Ever had sex partner with HIV or HBV or HCV infection” | ||||||||||
| no/no(ref) | 23 | 0.0 | Ref | 4.6 | Ref | 34.8 | Ref | |||
| no/yes | 9 | 0.0 | N/A | 25.0 | 7.0 | 0.5–91.0* | 66.7 | 3.7 | 0.7–19.0* | |
| yes/no | 37 | 8.1 | N/A | 8.6 | 2.0 | 0.2–20.0 | 29.7 | 0.8 | 0.3–2.4 | |
| yes/yes | 14 | 0.0 | N/A | 23.1 | 6.3 | 0.6–68.0* | 71.4 | 4.7 | 1.1–20.0** | |
p<0.20
p<0.05
N/A = Not applicable because of zero cell frequencies or quasi-complete separation of data points Ref = Reference category
TABLE 3.
Univariate analysis of HIV, hepatitis B (HBV), and hepatitis C (HCV) seroprevalence among male new drug injectors, 18–30 years old, New York City, February 1999 – February 2003; p<0.20 for any infection
| HIV+ | HBV+ | HCV+ | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Variables | Total N | % | OR | 95%CI | % | OR | 95%CI | % | OR | 95%CI |
| Total | 176 | 2.3 | 24.1 | 27.3 | ||||||
| Injecting risk behaviors (lifetime) | ||||||||||
| Lifetime frequency of injecting | ||||||||||
| < 300 times | 66 | 4.5 | Ref | 21.5 | Ref | 24.2 | Ref | |||
| ≥ 300 times | 110 | 0.9 | 0.2 | 0.0–1.9* | 25.7 | 1.3 | 0.6–2.6 | 29.1 | 1.3 | 0.6–2.6 |
| Ever injected heroin | ||||||||||
| no | 9 | 0.0 | 22.2 | Ref | 0.0 | |||||
| yes | 167 | 2.4 | N/A | 24.2 | 1.1 | 0.2–5.6 | 28.7 | N/A* | ||
| Ever injected cocaine | ||||||||||
| no | 57 | 1.8 | Ref | 25.0 | Ref | 17.5 | Ref | |||
| yes | 119 | 2.5 | 1.4 | 0.1–14.2 | 23.7 | 0.9 | 0.4–2.0 | 31.9 | 2.2 | 1.1–4.8** |
| Ever injected speedball | ||||||||||
| no | 71 | 4.2 | Ref | 20.0 | Ref | 16.9 | Ref | |||
| yes | 105 | 1.0 | 0.2 | 0.0–2.1 | 27.0 | 1.5 | 0.7–3.2 | 34.3 | 2.6 | 1.2–5.4** |
| Ever injected crack | ||||||||||
| no | 135 | 3.0 | 21.1 | Ref | 25.2 | Ref | ||||
| yes | 41 | 0.0 | N/A | 35.1 | 2.0 | 0.9–4.5* | 34.1 | 1.5 | 0.7–3.3 | |
| Injecting risk networks (last 30 days) | ||||||||||
| Number of injecting network members > 2.4 | ||||||||||
| no | 122 | 3.3 | 28.8 | Ref | 31.1 | Ref | ||||
| yes | 54 | 0.0 | N/A | 13.5 | 0.4 | 0.2–0.9** | 18.5 | 0.5 | 0.2–1.1* | |
| Sexual risk behaviors (last 6 months) | ||||||||||
| Any anal sex | ||||||||||
| no | 150 | 2.0 | Ref | 24.3 | Ref | 25.3 | Ref | |||
| yes | 26 | 3.8 | 2.0 | 0.2–19.6 | 23.1 | 0.9 | 0.3–2.5 | 38.5 | 1.8 | 0.8–4.4* |
| Sexual risk networks (lifetime) | ||||||||||
| Lifetime number of sex partners | ||||||||||
| < 40 | 123 | 2.4 | Ref | 20.5 | Ref | 22.8 | Ref | |||
| ≥ 40 | 53 | 1.9 | 0.8 | 0.1–7.6 | 32.1 | 1.8 | 0.9–3.8* | 37.7 | 2.1 | 1.02–4.1** |
| Ever had IDU sex partner | ||||||||||
| no | 36 | 5.6 | Ref | 25.0 | Ref | 36.1 | Ref | |||
| yes | 140 | 1.4 | 0.2 | 0.0–1.8* | 23.9 | 0.9 | 0.4–2.2 | 25.0 | 0.6 | 0.3–1.3* |
| Ever had sex partner with HIV or HBV or HCV infection | ||||||||||
| no | 146 | 1.4 | Ref | 20.4 | Ref | 22.6 | Ref | |||
| yes | 30 | 6.7 | 5.1 | 0.7–38.0* | 42.9 | 2.9 | 1.2–6.8** | 50.0 | 3.4 | 1.5–7.7** |
| Injecting or sexual risk exposure proxy variables | ||||||||||
| Ever sold sex | ||||||||||
| no | 163 | 1.8 | Ref | 21.0 | Ref | 25.8 | Ref | |||
| yes | 13 | 7.7 | 4.4 | 0.4–46.0 | 61.5 | 6.0 | 1.8–19.6** | 46.2 | 2.5 | 0.8–7.8* |
| MSM | ||||||||||
| no | 150 | 0.7 | Ref | 20.7 | Ref | 26.0 | Ref | |||
| yes | 26 | 11.5 | 19.4 | 1.9–195** | 44.0 | 3.0 | 1.2–7.3** | 34.6 | 1.5 | 0.6–3.7 |
| Reported ever having an STD | ||||||||||
| no | 146 | 1.4 | Ref | 20.4 | Ref | 26.7 | Ref | |||
| yes | 30 | 6.7 | 5.0 | 0.7–38.1* | 42.9 | 2.9 | 1.2–6.8** | 30.0 | 1.2 | 0.5–2.8 |
| Possible confounder variables | ||||||||||
| Years since injecting initiation > 2 | ||||||||||
| no | 89 | 2.2 | Ref | 16.3 | Ref | 22.5 | Ref | |||
| yes | 87 | 2.3 | 1.0 | 0.1–7.4 | 32.1 | 2.4 | 1.2–5.1** | 32.2 | 1.6 | 0.8–3.2* |
| Calendar year of initiation before 1997 | ||||||||||
| no | 111 | 2.7 | Ref | 19.8 | Ref | 23.4 | Ref | |||
| yes | 65 | 1.5 | 0.6 | 0.1–5.5 | 31.3 | 1.8 | 0.9–3.6* | 33.8 | 1.7 | 0.9–3.3* |
| Non-white race/ethnicity | ||||||||||
| no | 144 | 0.0 | 23.0 | Ref | 25.0 | Ref | ||||
| yes | 32 | 12.5 | N/A** | 29.0 | 1.4 | 0.6–3.3 | 37.5 | 1.8 | 0.8–4.0* | |
| Interaction variables included in the final model (in bold) | ||||||||||
| “Ever shared any injecting equipment” and Ever had sex partner with HIV or HBV or HCV infection” | ||||||||||
| no/no(ref) | 75 | 0.0 | Ref | 19.2 | Ref | 22.7 | Ref | |||
| no/yes | 10 | 10.0 | N/A | 60.0 | 6.3 | 1.6–25.0** | 40.0 | 2.3 | 0.6–9.0 | |
| yes/no | 71 | 2.8 | N/A | 21. 7 | 1.2 | 0.5–2.6 | 22.5 | 1.0 | 0.5–2.2 | |
| yes/yes | 20 | 5.0 | N/A | 33.3 | 2.1 | 0.7–6.6 | 55.0 | 4.2 | 1.5–12.0** | |
| “Ever shared any injecting equipment” and”Lifetime number of sex partners ≥ 40” | ||||||||||
| no/no(ref) | 62 | 1.6 | Ref | 18.3 | Ref | 16.1 | Ref | |||
| no/yes | 23 | 0.0 | N/A | 39.1 | 2.9 | 1.0–8.3** | 47.8 | 4.8 | 1.6–14.0** | |
| yes/no | 61 | 3.3 | 2.1 | 0.2–23.0 | 22.8 | 1.3 | 0.5–3.2 | 29.5 | 2.2 | 0.9–5.2* |
| yes/yes | 30 | 3.3 | 2.1 | 0.1–35.0 | 26.7 | 1.6 | 0.6–4.6 | 30.0 | 2.2 | 0.8–6.3 |
p<0.20
p<0.05
N/A = Not applicable because of zero cell frequencies or quasi-complete separation of data points Ref = Reference category
TABLE 4.
Multivariate analysis of risk correlates for testing seropositive for hepatitis B virus (HBV) and hepatitis C virus (HCV) among female and male new drug injectors 18–30 years old, New York City, February 1999 – February 2003 – final models
| HBV | HCV | |
|---|---|---|
| AOR (95%CI) | AOR (95%CI) | |
| Female | ||
| Lifetime frequency of injecting – ≥ 300 times | - | 5.9 (1.9, 18.5) |
| Ever sold sex | 5.2 (1.1, 24.6) | - |
| Ever had sex partner with HIV or HBV or HCV infection | 4.5 (1.1, 20.3) | |
| among those who never shared any injecting equipment | - | 3.0 (0.62, 14.2) |
| among those who ever shared any injecting equipment | - | 5.7 (1.4, 23.0) |
|
| ||
| Male | ||
| Ever sold sex | 7.3 (2.1, 25.6) | - |
| Ever injected speedball | - | 3.2 (1.4, 7.2) |
| Ever had IDU sex partner | - | 0.37 (0.15, 0.91) |
| Number of injecting network members in last 30 days > 2.4 | 0.33 (0.13, 0.84) | 0.33 (0.13, 0.83) |
| Lifetime number of sex partners ≥ 40 | ||
| among those who never shared any injecting equipment | - | 3.9 (1.4, 11.1) |
| among those who ever shared any injecting equipment | - | 1.7 (0.6, 4.5) |
| Ever had sex partner with HIV or HBV or HCV infection | ||
| among those who never shared any injecting equipment | - | 1.8 (0.40, 7.8) |
| among those who ever shared any injecting equipment | - | 8.2 (2.7, 25.2) |
3.2.1 Univariate and multivariate analyses among women
Among women, ever selling sex was significantly (p < 0.05) associated with being seropositive for each of the three viruses. (Table 2) The other variables significantly associated with being HIV seropositive included injecting with MSMs, a higher lifetime number of sex partners, and never having had an IDU sex partner. Ever having had an infected (HIV, HBV or HCV) sex partner and a higher lifetime frequency of injecting were associated with being HCV seropositive. The univariate association of the interaction variable included in the final model for HCV among women is shown, in bold, in Table 3.
In the multivariate analysis, ever selling sex and ever having had an infected sex partner were independently associated with being HBV seropositive A higher lifetime frequency of injecting, and ever having had an infected sex partner (only for those who had ever shared any injecting equipment) were independently associated with testing HCV seropositive. (Table 4)
3.2.2 Univariate and multivariate analyses among men
Being a MSM and of non-white race/ethnicity were significantly associated with HIV infection among male new IDUs. (Table 3) The variables significantly associated with testing HBV seropositive included a lower number of 30-day injecting network members, ever having had an infected sex partner, ever selling sex, being a MSM, ever having an STD, and initiating injecting more than two years prior to the interview. Ever injecting cocaine, ever injecting speedball, a higher lifetime number of sex partners, and ever having had an infected sex partner were significantly associated with testing HCV seropositive. The univariate association of the interaction variables included in the final model for HCV among men are shown, in bold, in Table 3.
In the multivariate analysis among men, ever selling sex and a lower number of 30-day injecting network members were independent correlates of testing HBV seropositive. (Table 4) The independent correlates of testing HCV seropositive included ever injecting speedball, never having had an IDU sex partner, a lower number of 30-day injecting network members, a higher lifetime number of sex partners (only among those who had never shared any injecting equipment), and ever having had an infected sex partner (only among those who had ever shared any injecting equipment).
A correlation analysis was conducted, within each stratum of ever sharing injecting equipment, to determine other sex risk factors that may be associated with the relationship of having a higher lifetime number of sex partners with testing HCV seropositive. The only significant result was among men who had never shared any injecting equipment, among whom having a higher number of sex partners was significantly correlated with ever selling sex (r = 0.24, p < 0.03). Moreover, among men overall, ever selling sex was strongly correlated with being a MSM (r = 0.56, p < 0.001).
3.2.3 Multivariate analysis of confounders
For both female and male new IDUs, none of the possible confounders were significantly associated with either HBV or HCV seropositivity in the multivariate analysis.
4. Discussion
In this study, very few new IDUs were infected with HIV, although HBV and HCV seroprevalence were considerably higher. This suggests that although HIV prevalence is currently low, many of the risk behaviors and risk network characteristics may exist for an HIV epidemic. There may also be gender differences in the rate of infection by different pathogens; while men were more likely to be seropositive for HBV, women were more likely to be seropositive for HCV.
Among women, ever having been HBV infected appears to be associated with sexual risk, including ever having had an infected sex partner and ever selling sex. While among women those with greater parenteral exposure were more likely to have been HCV seropositive, the evidence is less convincing for the sexual transmission of HCV, since the association of ever having had an infected sex partner with being HCV seropositive was significant only among those who had ever shared any injecting equipment. This suggests that among female new IDUs the risk of HCV infection from infected sex partners may be confounded and/or overlapping with sharing injecting equipment, since their sex partners may also be their injecting partners (Neaigus et al., 1995; Evans et al., 2003).
Among men, ever having been HBV infected is also associated with sexual risk, as indicated by ever selling sex. Moreover, among men ever selling sex is associated with being a MSM. Not surprisingly, men with greater parenteral exposure were also more likely to be HCV seropositive, since injecting speedball has been associated with greater injecting risk (Chaisson et al., 1989). The significant association of ever having had an infected sex partner with being HCV seropositive only among those who had ever shared any injecting equipment again suggests, as with women, that the risk of HCV infection from infected sex partners may be confounded and/or overlapping with sharing injecting equipment with injecting partners who are also sex partners. By contrast, the significant association of having a higher lifetime number of sex partners with being HCV seropositive only among those who had never shared any injecting equipment suggests that there may be an independent sexual risk of HCV infection among male new IDUs. Among men, having a higher lifetime number of sex partners is associated with selling sex which, as discussed above, is associated with being a MSM. Although in this study being a MSM may have been underreported and weakened the association of being a MSM with being HCV seropositive, other studies have found that HCV infection among MSM is associated with sexual practices that may involve blood-to-blood transmission, such as unprotected receptive anal sex and receptive anal “fisting” (Ndimbie et al., 1996; Ghosn et al., 2004; Gambotti et al., 2005; Gotz et al., 2005).
The other correlates of testing seropositive among male new IDUs suggest that social network factors may increase as well as reduce risk. Males who have never had an IDU sex partner may have had large, high-turnover injecting networks comprised of acquaintances or strangers, and thereby may have had multiple exposures to HCV (Neaigus et al., 1994; Kral et al., 2001). By contrast, the protective effect of ever having had an IDU sex partner on being HCV seropositive suggests that heterosexual male new IDUs may be at reduced risk of HCV infection because of norms which prescribe that men go first when injecting equipment is shared with female injecting partners (Bennett et al., 2000; Evans et al., 2003). Once an infectious pathogen has entered a larger sociocentric IDU network (comprised of both direct and indirect ties between IDUs), having smaller egocentric injecting networks (the direct ties between an IDU and her/his injecting partners) imbedded within this larger sociocentric network may increase infection risk since high-risk injecting behavior is more likely to occur in smaller egocentric networks, in which strong norms of trust and reciprocity may encourage or reinforce unsafe injecting practices (Neaigus et al., 1995; Friedman et al., 1997; Strathdee et al., 1997). Since the effects of network factors on injecting and sexual infection risk among new IDUs, and IDUs in general, are complex, further study is warranted.
Although HIV seroprevalence was low, the distribution of infection across sociodemographic and sociobehavioral groups may help to identify which new IDU subpopulations are at greatest risk of becoming infected with HIV. Other studies in the United States have also found an increased risk for HIV among young and/or new IDUs who are African-American/black or non-white Hispanics (Neaigus et al., 1996; Holmberg, 1996; Friedman et al., 1999), MSM (Garfein et al., 1996; Des Jarlais et al., 1999; Kral et al., 2001; Strathdee et al., 2001; Shafer et al., 2002), sell sex (Kral et al., 2001), or who are WSWs (Friedman et al., 2003).
Interventions among new IDUs to prevent and control infection with HBV and HCV and to prevent HIV epidemics should include mass screening for these pathogens among new IDUs and expanded coverage of HBV vaccination programs to IDUs. Interventions should also target specific subpopulations that may be at greatest risk. Such interventions should focus on those who may be the most susceptible for becoming infected with HIV, and on new IDUs in overlapping injecting and sexual relationships. Since the overlap of injecting and sexual risk may be rooted in the intersection of IDUs’ injecting and sex risk networks with their social support networks, especially among women (Miller and Neaigus, 2002), interventions are also needed that address underlying societal factors (e.g., unstable housing) that may lead new IDUs to seek out social support from other, potentially infected IDUs. Interventions are especially needed for primary prevention to prevent the initiation into injecting by non-injecting users of “hard” drugs, such as heroin, cocaine and other drugs (Neaigus et al., 2006).
The data have certain limitations. With cross-sectional data the temporal direction between infection and the risk correlates of infection cannot be determined. Participants may also have been selective or inaccurate in their recall of lifetime risk exposures. Another limitation is that the variables measuring recent correlates are conservative estimates of lifetime infection risks. However, these limitations may have been minimized because participants were relatively short-term injectors and had been sexually active for a relatively short period of time. Participants’ self-reports about their injecting and sex partners’ risk characteristics depend on communication among partners about these characteristics and the factors influencing such communication. Nevertheless, engaging in injecting and/or sexual risk behaviors with known infected partners is a plausible risk for becoming infected with HIV, HBV and HCV. Although the tests used for HBV and HCV are indicators of ever being infected rather than current infection, for HCV it is estimated that between 60%–80% of infected adults testing seropositive for the HCV antibody are chronic carriers (Seeff, 2002). The findings also cannot be readily generalized to non-white new IDUs, and the relatively small sample of women may have limited statistical power to detect smaller significant differences among female new IDUs. While the methods used for sampling and recruiting in this study have also been used by many other studies of non-drug-treatment recruited drug users, the sample is non-random, so that generalizations from the study’s findings must be informed by an understanding of the possible limitations of these sampling and recruiting methods.
As in many countries where HIV prevalence among new IDUs is currently low, many new IDUs in this study were or had been infected with HBV or HCV. Those with greater sexual risk exposure were at greater risk of ever being HBV infected, although parenteral transmission may also have occurred. Infection risk for HCV was through unsafe injecting and, possibly, high-risk sexual practices involving blood-borne transmission. Even in new IDU populations with low HIV prevalence, many new IDUs engage in injecting and sexual risk behaviors, and the outbreak of HIV epidemics cannot be ruled out should there be a change in the population probability of exposure to HIV. Early interventions among new IDUs are needed that combine mass screening for infection with these pathogens, information campaigns and HBV vaccination programs, along with interventions targeted at high-risk subpopulations of new IDUs to both prevent HBV and HCV epidemics and potential HIV epidemics.
Acknowledgments
The study was funded by the United States National Institute on Drug Abuse, grant DA09920 “Non-injecting heroin users, new injectors and HIV risk” (Principal Investigator: Alan Neaigus). We would like to thank Gilbert Ildefonso, Stephen J. Sifaneck, Peter Blasko, Jesse de Jesus and other members of the research staff for recruiting participants and data collection, as well as those who agreed to participate in the study.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- Bennett GA, Velleman RD, Barter G, Bradbury C. Gender differences in sharing injecting equipment by drug users in England. AIDS Care. 2000;12:77–87. doi: 10.1080/09540120047495. [DOI] [PubMed] [Google Scholar]
- Chaisson RE, Bacchetti P, Osmond D, Brodie B, Sande MA, Moss AR. Cocaine use and HIV infection in intravenous drug users in San Francisco. JAMA. 1989;261:561–5. [PubMed] [Google Scholar]
- Des Jarlais DC, Friedman SR, Perlis TE, Chapman TF, Sotheran JL, Paone D, Monterroso E, Neaigus A. Risk behavior and HIV infection among new drug injectors in the era of AIDS in New York City. J Acquir Immune Defic Syndr. 1999;20:67–72. doi: 10.1097/00042560-199901010-00010. [DOI] [PubMed] [Google Scholar]
- Des Jarlais DC, Perlis T, Friedman SR, Chapman T, Kwok J, Rockwell R, Paone D, Milliken J, Monterroso E. Behavioral risk reduction in a declining HIV epidemic: injection drug users in New York City, 1990–1997. Am J Public Health. 2000;90:1112–6. doi: 10.2105/ajph.90.7.1112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Des Jarlais DC, Diaz T, Perlis T, Vlahov D, Maslow C, Latka M, Rockwell R, Edwards V, Friedman SR, Monterroso E, Williams I, Garfein RS. Variability in the incidence of human immunodeficiency virus, hepatitis B virus, and hepatitis C virus infection among young injecting drug users in New York City. Am J Epidemiol. 2003;157:467–71. doi: 10.1093/aje/kwf222. [DOI] [PubMed] [Google Scholar]
- Doherty MC, Garfein RS, Monterroso E, Latkin C, Vlahov D. Gender differences in the initiation of injection drug use among young adults. J Urban Health. 2000;77:396–414. doi: 10.1007/BF02386749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Emmanuelli J, Desenclos JC. Harm reduction interventions, behaviours and associated health outcomes in France, 1996–2003. Addiction. 2005;100:1690–1700. doi: 10.1111/j.1360-0443.2005.01271.x. [DOI] [PubMed] [Google Scholar]
- European Centre for the Epidemiological Monitoring of AIDS. HIV/AIDS surveillance in Europe, mid-year report 2003. Institute de Veille Sanitaire; Saint Maurice, France: 2003. [Google Scholar]
- Evans JL, Hahn JA, Page-Shafer K, Lum PJ, Stein ES, Davidson PJ, Moss AR. Gender differences in sexual and injection risk behavior among active young injection drug users in San Francisco (the UFO Study. J Urban Health. 2003;80:137–46. doi: 10.1093/jurban/jtg137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Frajzyngier VM, Neaigus A, Gyarmathy VA, Miller M, Friedman SR. Gender differences in injection risk behaviors at the first injection episode. Drug Alcohol Depend. doi: 10.1016/j.drugalcdep.2006.12.021. in press. [DOI] [PubMed] [Google Scholar]
- Friedman SR, Des Jarlais DC, Neaigus A, Abdul-Quader A, Sotheran JL, Sufian M, Tross S, Goldsmith D. AIDS and the new drug injector. Nature. 1989;339:333–334. doi: 10.1038/339333a0. [DOI] [PubMed] [Google Scholar]
- Friedman SR, Neaigus A, Jose B, Curtis R, Goldstein MF, Ildefonso G, Rothenberg RB, Des Jarlais DC. Sociometric risk networks and risk for HIV infection. Am J Public Health. 1997;87:1289–1296. doi: 10.2105/ajph.87.8.1289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Friedman SR, Curtis R, Neaigus A, Jose B, Des Jarlais DC. Social networks, drug injectors’ lives, and HIV. Plenum; New York: 1999. [Google Scholar]
- Friedman SR, Chapman TF, Perlis TE, Rockwell R, Paone D, Sotheran JL, Des Jarlais DC. Similarities and differences by race/ethnicity in changes of HIV seroprevalence and related behaviors among drug injectors in New York City, 1991–1996. J Acquir Immune Defic Syndr. 1999;22:83–91. doi: 10.1097/00042560-199909010-00011. [DOI] [PubMed] [Google Scholar]
- Friedman SR, Ompad DC, Maslow C, Young R, Case P, Hudson SM, Diaz T, Morse E, Bailey S, Des Jarlais DC, Perlis T, Hollibaugh A, Garfein RS. HIV prevalence, risk behaviors, and high-risk sexual and injection networks among young women injectors who have sex with women. Am J Public Health. 2003;93:902–6. doi: 10.2105/ajph.93.6.902. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gambotti L, Batisse D, Colin-de-Verdiere N, aroque-Astagneau E, Desenclos JC, Dominguez S, Dupont C, Duval X, Gervais A, Ghosn J, Larsen C, Pol S, Serpaggi J, Simon A, Valantin MA, Velter A. Acute hepatitis C infection in HIV positive men who have sex with men in Paris, France, 2001–2004. Euro Surveill. 2005;10:115–117. [PubMed] [Google Scholar]
- Garfein RS, Vlahov D, Galai N, Doherty MC, Nelson KE. Viral infections in short-term injection drug users: The prevalence of the hepatitis C, hepatitis B, human immunodeficiency, and human T- lymphotropic viruses. Am J Public Health. 1996;86:655–661. doi: 10.2105/ajph.86.5.655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ghosn J, Pierre-Francois S, Thibault V, Duvivier C, Tubiana R, Simon A, Valantin MA, Dominguez S, Caumes E, Katlama C. Acute hepatitis C in HIV-infected men who have sex with men. HIV Med. 2004;5:303–306. doi: 10.1111/j.1468-1293.2004.00225.x. [DOI] [PubMed] [Google Scholar]
- Gotz H, van Doornum G, Niesters H, den Hollander J, Thio H, de Zwart O. A cluster of acute hepatitis C virus infection among men who have sex with men--results from contact tracing and public health implications. AIDS. 2005;19:969–974. doi: 10.1097/01.aids.0000171412.61360.f8. [DOI] [PubMed] [Google Scholar]
- Hagan H, Thiede H, Weiss NS, Hopkins SG, Duchin JS, Alexander ER. Sharing of drug preparation equipment as a risk factor for hepatitis C. Am J Public Health. 2001;91:42–46. doi: 10.2105/ajph.91.1.42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hahn JA, Page-Shafer K, Lum PJ, Ochoa K, Moss AR. Hepatitis C virus infection and needle exchange use among young injection drug users in San Francisco. Hepatology. 2001;34:180–187. doi: 10.1053/jhep.2001.25759. [DOI] [PubMed] [Google Scholar]
- Heckathorn DD. Respondent-driven sampling: A new approach to the study of hidden populations. Social Problems. 1997;44:174–199. [Google Scholar]
- Holmberg SD. The estimated prevalence and incidence of HIV in 96 large US metropolitan areas. Am J Public Health. 1996;86:642–54. doi: 10.2105/ajph.86.5.642. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ippolito G, Puro V, Heptonstall J, Jagger J, De CG, Petrosillo N. Occupational human immunodeficiency virus infection in health care workers: worldwide cases through September 1997. Clin Infect Dis. 1999;28:365–383. doi: 10.1086/515101. [DOI] [PubMed] [Google Scholar]
- Kral AH, Bluthenthal RN, Lorvick J, Gee L, Bacchetti P, Edlin BR. Sexual transmission of HIV-1 among injection drug users in San Francisco, USA: Risk-factor analysis. Lancet. 2001;357:1397–1401. doi: 10.1016/S0140-6736(00)04562-1. [DOI] [PubMed] [Google Scholar]
- Miller M, Neaigus A. Sex partner support, drug use and sex risk among HIV-negative non-injecting heroin users. AIDS Care. 2002;14:801–13. doi: 10.1080/0954012021000031877. [DOI] [PubMed] [Google Scholar]
- Ndimbie OK, Kingsley LA, Nedjar S, Rinaldo CR. Hepatitis C virus infection in a male homosexual cohort: risk factor analysis. Genitourin Med. 1996;72:213–16. doi: 10.1136/sti.72.3.213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Neaigus A, Sufian M, Friedman SR, Goldsmith DS, Stepherson B, Mota P, Pascal J, Des Jarlais DC. Effects of outreach intervention on risk reduction among intravenous drug users. AIDS Educ Prev. 1990;2:253–71. [PubMed] [Google Scholar]
- Neaigus A, Friedman SR, Curtis R, Des Jarlais DC, Furst RT, Jose B, Mota P, Stepherson B, Sufian M, Ward T, Wright JW. The relevance of drug injectors’ social networks and risk networks for understanding and preventing HIV infection. Soc Sci Med. 1994;38:67–78. doi: 10.1016/0277-9536(94)90301-8. [DOI] [PubMed] [Google Scholar]
- Neaigus A, Friedman SR, Goldstein MF, Ildefonso G, Curtis R, Jose B. Using dyadic data for a network analysis of HIV infection and risk behaviors among injecting drug users. NIDA Research Monograph. 1995;151:20–37. [PubMed] [Google Scholar]
- Neaigus A, Friedman SR, Jose B, Goldstein MF, Curtis R, Ildefonso G, Des Jarlais DC. High-risk personal networks and syringe sharing as risk factors for HIV infection among new drug injectors. J Acquir Immune Defic Syndr. 1996;11:499–509. doi: 10.1097/00042560-199604150-00011. [DOI] [PubMed] [Google Scholar]
- Neaigus A, Gyarmathy VA, Miller M, Frajzyngier VM, Friedman SR, Des Jarlais DC. Transitions to injecting drug use among noninjecting heroin users: social network influence and individual susceptibility. J Acquir Immune Defic Syndr. 2006;41:493–503. doi: 10.1097/01.qai.0000186391.49205.3b. [DOI] [PubMed] [Google Scholar]
- Nelson KE, Vlahov D, Solomon L, Cohn S, Munoz A. Temporal trends of incident human immunodeficeincy virus infection in a cohort of injecting drug users in Baltimore, MD. Arch Intern Med. 1995;155:1305–11. [PubMed] [Google Scholar]
- Puro V, De CG, Scognamiglio P, Porcasi R, Ippolito G. Risk of HIV and other blood-borne infections in the cardiac setting: patient-to-provider and provider-to-patient transmission. Ann NY Acad Sci. 2001;946:291–309. doi: 10.1111/j.1749-6632.2001.tb03918.x. [DOI] [PubMed] [Google Scholar]
- Robles RR, Colón HM, Sahai H, Matos TD, Marrero CA, Reyes JC. Behavioral risk factors and HIV prevalence among intravenous drug users in Puerto Rico. Am J Epidemiol. 1992;135:531–540. doi: 10.1093/oxfordjournals.aje.a116320. [DOI] [PubMed] [Google Scholar]
- Rooney G, Gilson RJ. Sexual transmission of hepatitis C virus infection. Sex Transm Infect. 1998;74:399–404. doi: 10.1136/sti.74.6.399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Seeff LB. Natural history of chronic hepatitis C. Hepatology. 2002;36:S35–S46. doi: 10.1053/jhep.2002.36806. [DOI] [PubMed] [Google Scholar]
- Semaan S, Des Jarlais DC, Sogolow E, Johnson WD, Hedges LV, Ramirez G, Flores SA, Norman L, Sweat MD, Needle R. A Meta-analysis of the effect of HIV prevention interventions on the sex behaviors of drug users in the United States. J Acquir Immune Defic Syndr. 2002;30(Suppl 1) [PubMed] [Google Scholar]
- Shafer KP, Hahn JA, Lum PJ, Ochoa K, Graves A, Moss A. Prevalence and correlates of HIV infection among young injection drug users in San Francisco. J Acquir Immune Defic Syndr. 2002;31:422–31. doi: 10.1097/00126334-200212010-00009. [DOI] [PubMed] [Google Scholar]
- Sifaneck SJ, Neaigus A. The ethnographic accessing, sampling and screening of hidden populations: Heroin sniffers in New York City. Addiction Research and Theory. 2001;9:519–543. [Google Scholar]
- Solomon L, Astemborski J, Warren D, Munoz A, Cohn S, Vlahov D, Nelson KE. Differences in risk factors for human immunodeficiency virus type 1 seroconversion among male and female intravenous drug users. Am J Epidemiol. 1993;137:892–8. doi: 10.1093/oxfordjournals.aje.a116750. [DOI] [PubMed] [Google Scholar]
- Strathdee SA, Patrick DM, Archibald CP, Ofner M, Cornelisse PG, Rekart M, Schechter MT, O’Shaughnessy MV. Social determinants predict needle-sharing behaviour among injection drug users in Vancouver, Canada. Addiction. 1997;92:1339–47. [PubMed] [Google Scholar]
- Strathdee SA, Galai N, Safaiean M, Celentano DD, Vlahov D, Johnson L, Nelson KE. Sex differences in risk factors for hiv seroconversion among injection drug users: a 10-year perspective. Arch Intern Med. 2001;161:1281–8. doi: 10.1001/archinte.161.10.1281. [DOI] [PubMed] [Google Scholar]
- Strathdee SA, Sherman SG. The role of sexual transmission of HIV infection among injection and non-injection drug users. J Urban Health. 2003;80:7–14. doi: 10.1093/jurban/jtg078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thomas DL, Cannon RO, Shapiro CN, Hook EW, III, Alter MJ, Quinn TC. Hepatitis C, hepatitis B, and human immunodeficiency virus infections among non-intravenous drug-using patients attending clinics for sexually transmitted diseases. J Infect Dis. 1994;169:990–995. doi: 10.1093/infdis/169.5.990. [DOI] [PubMed] [Google Scholar]
- Watters J, Biernacki P. Targeted sampling: Options for the study of hidden populations. Social Problems. 1989;6:416–430. [Google Scholar]
- Wejstal R. Sexual transmission of hepatitis C virus. J Hepatol. 1999;31(Suppl 1):92–95. doi: 10.1016/s0168-8278(99)80382-1. [DOI] [PubMed] [Google Scholar]