M A J O R A R T I C L E
Changes in Blood -borne Infection Risk Among
Injection Drug Users
Shruti H. Mehta,1Jacqueline Astemborski,1Gregory D. Kirk,1,2Steffanie A. Strathdee,3Kenrad E. Nelson,1
David Vlahov,1,4and David L. Thomas1,2
1Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, and2Department of Medicine, Johns Hopkins School of Medicine,
Johns Hopkins University, Baltimore, Maryland;3Department of Medicine, University of California San Diego School of Medicine, San Diego, California;
and4Center for Urban Epidemiologic Studies, New York Academy of Medicine, New York, New York
HCV infection prevalence among injection drug users (IDUs) recruited over 4 periods: 1988–1989, 1994–1995,
1998, and 2005–2008. We calculated HIV and HCV infection incidence within the first year of follow-up among
IDUs whose test results were negative for these viruses at baseline (n 5 2061 and n 5 373, respectively). We used
Poisson regression to compare trends across groups.
Results.HIV infection incidence declined significantly from 5.5 cases/100 person-years (py) in the 1988–1989
group to 2.0 cases/100 py in the 1994–1995 group to 0 cases/100 py in the 1998 and 2005–2008 groups.
Concurrently, HCV infection incidence declined but remained robust (22.0 cases/100 py in the 1988–1989 cohort to
17.2 cases/100 py in the 1994–1995 cohort, 17.9 cases/100 py in the 1998 cohort, and 7.8 cases/100 py in the 2005–
2008 cohort; P 5 .07). Likewise, HCV infection prevalence declined, but chiefly in younger IDUs. For persons
aged ,39 years, relative to the 1988–1989 cohort, all groups exhibited significant declines (adjusted prevalence ratio
[PR] for the 2005–08 cohort, .73; 95% confidence interval [CI], .65–.81). However, for persons aged >39 years, only
the 2005–2008 cohort exhibited declining prevalence compared with the 1988–1989 cohort (adjusted PR, .87; 95%
Conclusions.Although efforts to reduce blood-borne infection incidence have had impact, this work will need to
be intensified for the most transmissible viruses, such as HCV.
Population-level hepatitis C virus (HCV) infection incidence is a surrogate for community
We characterized trends in human immunodeficiency virus (HIV) and HCV infection incidence and
Nearly 30 years into the HIV epidemic, injection drug
users (IDUs) remain at high risk for HIV infection. Data
from surveillance systems and cohort studies have col-
lectively suggested that HIV infection incidence among
IDUs has declined [1–11], a trend attributed at least in
part to harm-reduction measures including needle-
exchange programs (NEPs) and substance-abuse treat-
ment. Hepatitis C virus (HCV) is nearly 10 times more
transmissible by needlestick than is HIV . Preva-
lence estimates of HCV infection among IDUs have
been reported to exceed 50% in most IDU populations,
ranging as high as 95% . Sharing a needle even once
is enough to transmit or acquire HCV . Harm-
reduction measures that have led to declines in HIV
infection incidence have not been as successful for HCV
infection [15, 16].
HCV transmission is almost exclusively parenteral,
in contrast to the multiple routes of transmission for
HIV. Furthermore, drug use is much more likely to
result in exposure to HCV than to HIV, and each ex-
posure is more likely to result in transmission, making
population-level HCV infection incidence a surrogate
Received 2 June 2010; accepted 23 November 2010.
Presented inpart: 17thConference on Retroviruses and Opportunistic Infections in,
AIDS Conference in Mexico City, Mexico, 3–8 August 2008, Poster MOPE0390.
Potential conflicts of interest: none reported.
Reprints or correspondence: Shruti H. Mehta, PhD, MPH, Department of
Epidemiology, Johns Hopkins Bloomberg School of Public Health, Johns Hopkins
University, 615 N Wolfe Street, E6537, Baltimore, MD, 21205 (shmehta@
The Journal of Infectious Diseases
? The Author 2011. Published by Oxford University Press on behalf of the Infectious
Diseases Society of America. All rights reserved. For Permissions, please e-mail:
Blood Borne Infection in IDUs
d JID 2011:203 (1 March)
for drug-related risk behavior in the community [17, 18]. There
are few recent reports on HCV infection incidence among
community-based IDUs, but some data suggest that high-risk
drug behaviors persist among IDUs. In 2009, the US Centers for
Disease Control and Prevention published a report on HIV-
US cities . Overall, one-third of IDUs reported sharing in-
jection equipment in the preceding year and fewer than one-
third had participated in an HIV behavioral intervention.
Current estimates of HCV infection incidence would provide
further context for these reports.
we previously demonstrated declining HIV infection incidence
from1988 through2007[10,20].We havealsoreportedonhigh
HCV infection prevalence and incidence among this study’s
initial cohort, recruited in 1988–1989 . We have recruited
of this analysis was to characterize trends in HIV and HCV
infection incidence as well as HCV infection prevalence among
IDUs over 4 recruitment periods spanning 20 years.
All participants provided written informed consent, and the
study was approved by the Johns Hopkins Bloomberg School of
Public Health Institutional Review Board.
The AIDS Linked to the Intravenous Experience (ALIVE) study
follows a community-based cohort of IDUs in Baltimore that
has been described elsewhere . In 1988–1989, 2946 IDUs
were enrolled through community outreach and were followed
up at 6-month intervals. All participants acknowledged non-
medical injection-drug use within the preceding 11 years, were
>18 years of age, and were free of AIDS at entry into the study.
In a similar manner, and from the same community, additional
persons were recruited into this cohort in 1994–1995 (n 5 391),
1998 (n 5 244), and 2005–2008 (n 5 875). Some recruitment
criteriachangedovertime.Inthe fourthperiod, personswereno
longer required to be AIDS-free at entry. To replenish with
active injectors, in 1994–1995, persons had to have injected in
the preceding 3 years, and in 1998 and 2005–2008, in the pre-
We calculated HIV infection incidence for each recruitment
cohort among persons who were HIV antibody negative at the
baseline visit and who had a follow-up visit within 1 year of
baseline (n 5 2061). Individuals who were lost-to-follow-up or
did not have at least one follow-up visit within 1 year of baseline
the first follow-up visit that occurred within 1 year of baseline.
Incidence was calculated to reflect only time until the first semi-
annual visit that occurred among this group. HCV infection
incidence analysis was performed similarly among those in-
dividuals who were HCV antibody negative at the baseline visit
and had a follow-up visit within 1 year of baseline (n 5 373).
Individuals were eligible for the HCV infection prevalence
analysis only if a serum sample was available for us to test for
HCV antibodies at baseline. From the original cohort, we ran-
domly selected 250 persons, 4 of whom had no sample. The
random sample was comparable to the full population with
respect to demographics and risk behaviors (data not shown).
Serum samples were not available for 12 participants in the
1994–1995 cohort (3.1%), 12 (4.9%) in the 1998 cohort, and 1
person (.1%) of the 2005–2008 cohort. Final sample sizes for the
HCV infection prevalence analysis were 246from the 1988–1989
recruitment group, 379 from 1994–1995, 232 from 1998, and
874 from 2005–2008.
At all study visits, each participant underwent a blood draw and
answered a questionnaire that included a combination of
interviewer-administered and audio computer-assisted self-
interview (ACASI) questions. Blood samples were stored frozen
at 270?C in a repository for future testing. We performed HCV
antibody testing on frozen specimens using a second-genera-
tion or later enzyme immunoassay (Ortho Diagnostics).
At the baseline visit, we elicited information on demographics,
lifetime medical history, and lifetime drug-use and sexual his-
tory. Participants were asked about when they started using and
injecting drugs; lifetime experience with needle sharing, at-
tending shooting galleries (venues where IDUs can rent, bor-
row, or purchase injection equipment and where injection
equipment is typically used repeatedly), and being in drug
treatment of any form, including methadone maintenance and
detoxification; and number of sexual partners in the preceding
10 years. At baseline and follow-up visits, we collected data on
behaviors in the preceding 6 months, including frequency
(categorized as less than daily and at least daily) and types
(heroin, cocaine, and speedball) of drugs injected; number of
needle-sharing partners (eg, the number of persons with whom
needles were shared or number of persons from whom the
participant borrowed or rented a needle that had been pre-
viously used); shooting-gallery attendance; any drug treatment,
specifically methadone maintenance; and number and types of
sexual partners. Data on NEP attendance was available only
after 1998, so it was not included. Data on sensitive in-
formation (eg, drug use) was collected using ACASI after 1998.
We compared characteristics of participants across the 4 re-
cruitment periods using v2tests for categorical variables and the
Kruskal-Wallis test for continuous variables. We calculated in-
cidence asthenumberofnewHIV orHCVinfectionsdividedby
d JID 2011:203 (1 March)
d Mehta et al.
person-years (py) of follow-up between baseline and the first
of baseline) for persons initially HIV or HCV antibody negative.
We did not perform multivariate analysis for HIV because we
observed no infections in either the 1998 or 2005–2008 cohorts.
We used Poisson regression to calculate incidence-rate ratios of
HCV infection by recruitmentcohort after adjustment. We built
models sequentially to assess the impact of different factors.
Initially, we included differences across recruitment cohorts to
ensure that the differences observed were not explained by
changes inthe populations overtime.First,weincludedonlyage
and time since first injection followed by demographic variables
(sex, race, and educational attainment) and HIV serostatus. We
constructed subsequent models to determine whether any of the
changes observed were explained by changes in recent drug-use
or sexual risk behaviors over time. Drug-related risk behaviors
included frequency of injection, needle sharing, shooting gallery
attendance, and drug treatment in the preceding 6 months;
sexual risk behaviors included number of sexual partners and
self-reported sexually transmitted infections in the preceding 6
months. We also constructed models varying the order in which
variables were included (eg, drug behaviors before de-
mographics). We compared HCV infection prevalence across
the 4 cohorts using the v2test for trend. We used Poisson re-
gression with robust variance estimation to calculate prevalence
differences in confounders across recruitment cohorts and the
impact of changes in risk behavior. The model-building strategy
was similar to that described for HCV infection incidence. The
primary difference was that in the prevalence models, we in-
cluded lifetime risk behaviors rather than recent risk behaviors.
We assessed effect modification by age and years since first re-
ported injection on the association between recruitment cohort
and HCV infection prevalence by including interaction terms in
regression models. All statistical analyses were performed using
Stata software (version 10.1; StataCorp).
Table 1 illustrates characteristics of the 4 recruitment cohorts at
enrollment. Over time, the cohorts were older and the median
duration of injection drug use was longer. Compared with the
1988–1989 cohort, subsequent recruitments included more
women. The first 3 cohorts had a higher proportion of African
Americans compared with the most recent recruitment (P ,
.001 for all). There were no statistically significant differences in
marital status or educational attainment over time. The 2 later
cohorts had a higher proportion of individuals with no formal
income.HIV infection prevalence fluctuated from 23%in 1988–
89 to 11 % in 1994–95 to 31% in 1998 and 23% in 2005–08.
There were some variations in injection practices over time,
someofwhichreflectrecruitment differences.The proportion of
participants who reported ever sharing needles over time de-
creased slightly; however, the proportion who reported ever
attending a shooting gallery increased, as did the proportion
who reported ever being in drug treatment (P , .001). The
number of sexual partners in the preceding 10 years decreased
over time. The proportion not injecting within 6 months of
baseline was highest in the original and 2005–2008 recruitments.
The proportion of participants injecting only heroin was higher
in the 1994–1995 and 1998 cohorts than in those from other
Trends in HIV and Hepatitis C Virus Infection Incidence
HIV infection incidence declined significantly over time from
5.5 cases/100 py in the 1988–1989 group to 2.0 cases/100 py in
the 1994–1995 group to 0 cases/100 py in the 1998 and 2005–
2008 groups (Figure 1). HCV infection incidence decreased over
time from 22.0 cases/100 py in the 1988–1989 group to 17.2
cases/100 py in the 1994–1995 group, 17.8 cases/100 py in the
1998 group, and 7.8 cases/100 py in the 2005–2008 group
(P 5 .07, v2test for trend) (Figure 1). The decline between the
1988–1989 and 1994 recruitment groups was not statistically
significant (incidence rate ratio [IRR], .79; 95% confidence in-
terval [CI], .28–2.19; Table 2). Relative to 1988–89, there were
also statistically nonsignificant declines in HCV infection in-
cidence inthe1998and2005–2008groups.There was nodecline
between the 1994–1995 and 1998 groups (IRR, 1.04; 95% CI,
.28–3.87) and a statistically nonsignificant decline between the
1998 and 2005–2008 groups (IRR, .43; 95% CI, .11–1.75). These
differences strengthened in magnitude after adjustment for de-
mographics, duration of injection, and HIV serostatus; however,
none were statistically significant (Table 2, models 2 and 3).
Neither drug-related nor sexual-related risk behavior explained
a large proportion of the change in HCV infection incidence
adjusting for drug-related risk behavior was .38 (95% CI, .09–
1.62), whereas after adjusting for drug-related risk behavior the
IRR was .43 (95% CI, .09–2.18).
Trends in Hepatitis C Virus Infection Prevalence
We observed similar trends in the association between re-
cruitment cohort and HCV infection prevalence across strata of
age and years since first injection. A statistically significant de-
cline in HCV infection prevalence was observed among persons
who were younger (,39 years of age) and had shorter injection
history (,15 years) but not among those who were older and
injecting for longer. Among those who had been injecting
for ,5 years at entry, the HCV infection prevalence was 70% in
the 1988–1989 group, 65% in the 1994–1995 group, 50% in the
1998 group, and 52% in the 2005–2008 group (P 5 .02). Fur-
thermore, .80% prevalence was reached by 5–9 years of in-
jecting in the1988–1989 group versus 15–19 years of injecting in
the 2005–2008 group. Figure 2 shows HCV infection prevalence
Blood Borne Infection in IDUs
d JID 2011:203 (1 March)
by age. Because differences across cohorts over time were het-
erogeneous across strata of age, all analyses included an in-
teraction between age (stratified at the median of 39 years) and
The pattern of decline in HCV infection prevalence over time
among individualsaged,39years persistedafter adjustment for
demographic factors and injection duration differences across
recruitment cohorts (Figure 3A). All cohorts exhibited statisti-
cally significant differences in HCV infection prevalence relative
to the 1988–1989 cohort (model 2); there was also a decline
between the 1998 and 1994–1995 cohorts (PR, .82; 95% CI,
.69–.97; P 5 .02) and a statistically nonsignificant increase
between the 1998 and 2005–2008 cohorts (PR, 1.08; 95% CI,
.91–1.29; P 5 .38). These changes were not largely explained by
differences in lifetime drug-related or sexual-related risk be-
havior across the cohorts, as adjustment for these behaviors did
not substantially attenuate the PRs (model 3).
Among individuals aged >39 years, after adjustment for de-
mographic and time since injection differences, there were no
statistically significant differences between the 1994–1995 or the
1988 cohort and the 1988–1989 cohort (Figure 3B), but the
2005–2008 cohort had significantly lower HCV infection prev-
alence compared with the 1988–1989 cohort (PR, .86; 95% CI,
.76–.98). Only a small proportion of this decline was explained
by changes in drug-related risk behavior over time (model 3).
In this cohort of IDUs, we observed a dramatic decline in HIV
infection incidence over 2 decades; no new infections occurred
within the first year of follow-up in cohorts recruited in 1998
forward. Reductions in HCV infection incidence and prevalence
were also observed over the same period, most notably among
persons who had started injection recently and were younger.
However,similarprevalences ofHCV infectionovertimeamong
improvements delay but do not prevent HCV infection at the
populationlevel.Collectively,these datasupportintensifying the
harm-reduction strategies that have markedly reduced HIV
transmission to reduce further the risk of HCV infection.
Table 1. Description of study population by calendar period of recruitment*
1988-89 (n5246) 1994-95 (n5379)1998 (n5232) 2005-08 (n5874)P value
Median age (interquartile range)34 (30 – 38) 37 (32 – 42)40 (31 – 40)43 (36 – 48)
Duration of injection drug use12.9 (6.9 – 18.8)15.4 (8.0 – 23.0) 17.9 (10.0 – 25.5)18.5 (10.2 – 27.3)
Male gender201 (81.7)253 (66.8)150 (64.7) 562 (64.3)
African-American216 (87.8) 360 (95.0)219 (94.8) 573 (65.6)
>High school education
154 (62.6) 255 (67.3)151 (66.2) 549 (62.9)0.4068
114 (46.5)177 (47.0) 95 (41.7)367 (42.0) 0.2796
5001 – 10,000
Lifetime risk behaviors
56 (22.8) 42 (11.1)71 (30.6) 201 (23.0)
Ever shared needles 236 (96.0) 312 (82.5)165 (71.7) 755 (86.5)
Ever attend shooting gallery 114 (46.3) 222 (58.7)140 (60.6) 759 (86.8)
Ever been in drug treatment 125 (50.8)232 (61.4) 143 (61.9) 770 (88.2)
Sexual partners in prior 10 years
Recent risk behaviors
Injection drug use in prior year225 (91.5) 374 (98.9) 232 (100)874 (100)
Drug injected in prior six months*
Shared needles in prior six months149 (60.6) 130 (39.3)89 (39.4) 562 (64.5)
Attended shooting gallery in prior six months 70 (28.5)42 (12.7) 33 (14.5)213 (24.4)
NOTE.* Numbers do not add up due to missing values; p ,0.05 for chi-squared test for trend for all but male gender, African-American race, marital status and
d JID 2011:203 (1 March)
d Mehta et al.
Large-scale expansion of NEPs and opiate substitution
treatment programs appear to have reduced HIV transmission
among IDUs [6, 7, 23]. Accordingly, in this Baltimore IDU
cohort, we have seen marked reductions in HIV infection in-
cidence over 2 decades [10, 20]. In this analysis, we also detected
a decline in HCV infection incidence as well as HCV infection
prevalence among those who were younger or had recently
started injection. Importantly, we observed that HCV acquisi-
tion may be delayed by up to 10 years among IDUs compared
with that in the late 1980s when the epidemic was at peak. These
results are consistent with a recent meta-analysis by Hagan
and colleagues  that suggested that time to HCV infection
has lengthened in developed countries as well as a number of
reports suggesting declines among younger injectors and new
initiates [25–30]. Although we hypothesize that these declines
are due to expanded harm-reduction efforts and reduction in
drug-related risk behavior, our data did not demonstrate that
changes in self-reported injection behavior had substantial
impact. Of note, we were not able to account for changes in
Despite reductions in HCV infection prevalence among
in those who were older and had longer injection histories, with
the exception of a slight reduction among the most recent co-
hort compared with the earliest. It is possible that we failed to
observe a difference over time among older IDUs with longer
history of injection because it will simply take longer for re-
ductions in HCV infection incidence to impact prevalence
among older individuals, given that many of the individuals in
expanded harm-reduction strategies. This hypothesis is sup-
ported by the significantly lower prevalence among older IDUs
in the most recent recruitment cohort.
However, it is important to consider other reasons why
analogous reductions in HCV infection among older persons
with a history of drug injection have not been detected. HCV is
an order of magnitude more transmissible than HIV by a single
needlestick . Thus, measures that reduce the likelihood of
viral exposure or that reduce the number of viruses in each
exposure might be sufficient to affect HIV infection incidence
without having a commensurate effect on HCV infection. In-
terventions such as NEPs and opiate substitution may reach
IDUs too late in their injecting careers to have a significant
impact on HCV infection incidence. Furthermore, although
these measures may reduce the frequency of needle sharing that
may be sufficient to impact HIV transmission, they may not
completely eliminate risk behavior, making them insufficient to
virus and hepatitis C virus infection by recruitment cohort in the AIDS
Linked to the Intravenous Experience (ALIVE) cohort, 1988–2009.
Incidence per 100 person-years of human immunodeficiency
Table 2. Incidence rate ratios of HCV infection by calendar period of enrollment*
Incidence rate ratio (95% CI)
1988–89 1994–951998 2005–08
Model 1: unadjustedREF 0.79 (0.28 – 2.19) 0.82 (0.27 – 2.47)0.36 (0.12 – 1.09)
Model 2: adjusted for age, time
REF 0.80 (0.27 – 2.33) 0.98 (0.31 – 3.06)0.55 (0.14 – 2.14)
Model 3: adjusted age, time since
1st injection, gender, race, HIV,
REF0.70 (0.21 – 2.38)0.95 (0.26 – 3.45)0.38 (0.09 – 1.62)
Model 4: adjusted age, time since
1stinjection, gender, race, HIV,
education, income, drug-related
REF0.98 (0.28 – 3.37) 0.99 (0.27 – 3.63)0.43 (0.09 – 2.18)
Model 5: adjusted age, time since
1stinjection, gender, race, HIV,
education, income, drug and
sex-related risk behaviors***
REF1.00 (0.29 – 3.42) 0.90 (0.26 – 3.20)0.36 (0.07 – 1.87)
NOTE. *Results from Poisson regression; age was included in all models as a continuous variable.
**Summary drug-related risk behavior including recent (past six months) frequency of drug injection, needle sharing, shooting gallery attendance and drug
***Summary sex-related risk behavior includes recent number of partners in the last 6 months categorized as 0, 1, and .2 and any sexually transmitted infection.
Blood Borne Infection in IDUs
d JID 2011:203 (1 March)
effectively prevent HCV transmission throughout an IDU’s in-
Data from this cohort and many others indicate the risk of
HCV-related liver disease sharply rises in persons aged .40
years [31–33]. It is difficult to disentangle the effects of age and
duration of HCV infection on disease progression, and despite
some public health benefits of delaying HCV infection, it will
likelynot besufficienttoprevent the long-term complicationsof
disease in a large proportion of IDUs. Even with some reduction
in transmission, of the 2005–2008 cohort, .80% of individuals
>39 years old are infected with HCV and at risk for developing
any of the complications associated with chronic liver disease.
Furthermore, this level of prevalence points to a large reservoir
of HCV infection among this population, which significantly
hampers prevention efforts.
Efforts need to be intensified on both the prevention and
treatment fronts to reduce the reservoir of HCV-infected IDUs.
For many persons, the interval between initiation and injection
simply remains too brief for prevention strategies to be suc-
cessful. Targeting very young IDUs or drug users who have not
yet transitioned to injection is critical. However, because
transmission continues to occur even among older IDUs, pre-
vention strategies must target IDUs at all stages and ages, and it
may be that different strategies are needed for different sub-
populations. The other method for reducing the size of the
reservoir is HCV infection treatment. Current treatment regi-
mens have limited efficacy among certain subpopulations. In
particular, our population in Baltimore is predominantly in-
fected with genotype 1 and African American, with a high
prevalence of HIV co-infection, all of which are factors associ-
ated with reduced treatment success [34–36]. The impact of
treatment in reducing HCV burden is likely to have been min-
imal, because we and others have demonstrated that few
IDUs are engaged in care for HCV infection and even fewer
successfully clear their HCV infections . With new, more
efficacious therapeutics on the horizon , it is critical that
strategies to improve uptake and completion of HCV infection
treatment of IDUs be implemented, because treatment is the
only option for the large numbers of IDUs already infected with
Wewerelimited in thisanalysis by nothaving HCV testing on
the full baseline recruitment cohort; however, the random
sample did appear to be similar to the full baseline cohort with
respect to key covariates that would be associated with HCV
infection prevalence. Our analysis was further limited by the
small number of persons who were HCV antibody negative at
baseline; this is a common problem in epidemiologic studies
among IDUs. All of our behavioral data were collected via self-
report, a method that has inherent limitations, some of which
may have impaired our ability to assess how much impact be-
havior change had, particularly in the analysis of HCV infection
prevalence. We cannot rule out the possibility of variability in
unmeasured individual factors or secular changes across the
cohorts that could have impacted prevalence and incidence.
Eligibility criteria slightly changed over time, and we observed
that a number of factors, including HIV serostatus, differed
across recruitment cohorts. Although we adjusted for all mea-
sured confounders, we cannot rule out the possibility of residual
confounding. Finally, prevalence estimates are impacted by both
incidence and mortality, and it is possible that some of the early
changes in prevalence observed actually reflect the high levels of
mortality due to drug overdose and AIDS in the era prior to the
1988–2008 (n 5 1731).
Hepatitis C virus infection prevalence by age at entry and recruitment cohort in the AIDS Linked to the Intravenous Experience (ALIVE) cohort,
d JID 2011:203 (1 March)
d Mehta et al.
use of highly active antiretroviral therapy (HAART). However,
continued declines even after 1996 suggest that some of this
difference may be because of reduced incidence as well. Finally,
behavioral data were collected by interviewer-administered
questionnaire prior to 1998 and via ACASI after 1998, which
may have further impacted differences.
These limitations notwithstanding, the data collectively sug-
gest that harm-reduction strategies that have been successful for
HIV infection may also be contributing to declines in HCV
acquisition. However, additional, more intensive strategies,
particularly those that target new initiates into drug injection,
are needed tosignificantly impact community-leveldrug-related
risk. Furthermore, HCV infection prevalence and incidence re-
main up to 10-fold higher than those of HIV infection in this
population, reinforcing not only the continued need for pre-
ventive measures but alsotheneed toexpand careand treatment
to those already infected.
This study was supported by Public Health Service Grants from the
National Institute on Drug Abuse (DA04334, DA12568, DA16078 and
DA013868). The funders had no role in the study design, data collection and
analysis, decision to publish, or preparation of the manuscript.
We thank the ALIVE study staff and participants, without whom this
would not have been possible.
1. Fleming PL, Ward JW, Karon JM, Hanson DL, De Cock KM. Declines
in AIDS incidence and deaths in the USA: A signal change in the
epidemic. AIDS 1998; 12:S55–61.
2. Centers for Disease Control and Prevention. HIV and AIDS–United
States, 1981–2000. MMWR Morb Mortal Wkly Rep 2001; 50:430–4.
3. Longshore D, Anglin MD. Seroprevalence stable and seroconversions
rare among injection drug users in Los Angeles. J Acquir Immune Defic
Syndr 1992; 5:212–3.
4. Watters JK. Trends in risk behavior and HIV seroprevalence in het-
erosexual injection drug users in San Francisco, 1986–1992. J Acquir
Immune Defic Syndr 1994; 7:1276–81.
5. Des Jarlais C, Perlis T, Friedman SR, et al. 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.
6. Des Jarlais DC, Marmor M, Friedmann P, et al. HIV incidence among
injection drug users in New York City, 1992–1997: Evidence for a de-
clining epidemic. Am J Public Health 2000; 90:352–9.
7. Kral AH, Lorvick J, Gee L, et al. Trends in human immunodeficiency
virus seroincidence among street-recruited injection drug users in San
Francisco, 1987–1998. Am J Epidemiol 2003; 157:915–22.
8. Moss AR, Vranizan K, Gorter R, Bacchetti P, Watters J, Osmond D.
HIV seroconversion in intravenous drug users in San Francisco, 1985–
1990. AIDS 1994; 8:223–31.
9. Wiebel WW, Jimenez A, Johnson W, et al. Risk behavior and HIV
seroincidence among out-of-treatment injection drug users: A four-
year prospective study. J Acquir Immune Defic Syndr Hum Retrovirol
10. Nelson KE, Galai N, Safaeian M, Strathdee SA, Celentano DD, Vlahov
D. Temporal trendsinthe incidence ofhuman immunodeficiency virus
infection and risk behavior among injection drug users in Baltimore,
Maryland, 1988–1998. Am J Epidemiol 2002; 156:641–53.
11. Van Den Berg C, Smit C, Van Brussel G, Coutinho R, Prins M. Full
participation in harm reduction programmes is associated with de-
creased risk for human immunodeficiency virus and hepatitis C virus:
Evidence from the Amsterdam Cohort Studies among drug users.
Addiction 2007; 102:1454–62.
12. Sulkowski MS, Ray SC, Thomas DL. Needlestick transmission of
hepatitis C. JAMA 2002; 287:2406–13.
13. Aceijas C, Rhodes T. Global estimates of prevalence of HCV infection
among injecting drug users. Int J Drug Policy 2007; 18:352–8.
14. Garfein RS, Vlahov D, Galai N, Doherty MC, Nelson KE. Viral in-
fections in short-term injection drug users: The prevalence of the
cohort for persons aged ,39 years and persons aged >39 years. Results
are from Poisson regression with robust variance estimates are stratified
by age because there was a statistically significant interaction between
age and recruitment cohort (P ,.01). Panel A reflects ages of ,39 years
and panel B reflects ages of >39 years. The reference group for all
models is the 1988–1989 cohort. P values for all prevalence ratios in
panel A are statistically significant. Only the P value for the 2005–2008
cohort in model 2 of panel A is statistically significant. Age was included
as a continuous variable in both models to account for residual
confounding. Drug-related risk behaviors include lifetime history of
needle sharing, shooting gallery attendance, and drug treatment; sexual
risk behavior includes number of sexual partners in the preceding 10
Prevalence ratios of hepatitis C virus infection by recruitment
Blood Borne Infection in IDUs
d JID 2011:203 (1 March)
hepatitis C, hepatitis B, human immunodeficiency, and human T- Download full-text
lymphotropic viruses. Am J Public Health 1996; 86:655–61.
15. Wright NM, Tompkins CN. A review of the evidence for the effec-
tiveness of primary prevention interventions for hepatitis C among
injecting drug users. Harm Reduct J 2006; 3:27.
16. Palmateer N, Kimber J, Hickman M, Hutchinson S, Rhodes T,
Goldberg D. Evidence for the effectiveness of sterile injecting equip-
ment provision in preventing hepatitis C and human immunodefi-
ciency virus transmission among injecting drug users: A review of
reviews. Addiction 2010; 105:860–1.
17. Vickerman P, Hickman M, May M, Kretzschmar M, Wiessing L. Can
hepatitis C virus prevalence be used as a measure of injection-related
human immunodeficiency virus risk in populations of injecting drug
users? An ecological analysis. Addiction 2010; 105:311–8.
18. Des Jarlais DC, Arasteh K, McKnight C, Hagan H, Perlman D,
Friedman SR. Using hepatitis C virus and herpes simplex virus-2 to
track HIV among injecting drug users in New York City. Drug Alcohol
Depend 2009; 101:88–91.
19. Centers for Disease Control and Prevention. HIV-associated behaviors
among injecting-drug users–23 cities, United States, May 2005–
February 2006. MMWR Morb Mortal Wkly Rep 2009; 58:329–32.
20. Mehta SH, Galai N, Astemborski J, et al. HIV incidence among in-
jection drug users in Baltimore, Maryland (1988–2004). J Acquir Im-
mune Defic Syndr 2006; 43:368–72.
21. Villano SA, Vlahov D, Nelson KE, Lyles CM, Cohn S, Thomas DL.
Incidence and risk factors for hepatitis C among injection drug users in
Baltimore, Maryland. J Clin Microbiol 1997; 35:3274–7.
22. Vlahov D, Anthony JC, Munoz A, et al. The ALIVE study, a longitu-
dinal study of HIV-1 infection in intravenous drug users: Description
of methods and characteristics of participants. NIDA Res Monogr
23. Des Jarlais DC, Perlis T, Arasteh K, et al. HIV incidence among in-
jection drug users in New York City, 1990 to 2002: Use of serologic test
algorithm to assess expansion of HIV prevention services. Am J Public
Health 2005; 95:1439–44.
24. Hagan H, Pouget ER, Des Jarlais DC, Lelutiu-Weinberger C. Meta-
regression of hepatitis C virus infection in relation to time since onset
of illicit drug injection: The influence of time and place. Am J Epi-
demiol 2008; 168:1099–109.
25. Burt RD, Hagan H, Garfein RS, Sabin K, Weinbaum C, Thiede H.
Trends in hepatitis B virus, hepatitis C virus, and human immuno-
deficiency virus prevalence, risk behaviors, and preventive measures
among Seattle injection drug users aged 18–30 years, 1994–2004.
J Urban Health 2007; 84:436–54.
26. Des Jarlais DC, Perlis T, ArastehK, et al. Reductions in hepatitis C virus
and HIV infections among injecting drug users in New York City,
1990–2001. AIDS 2005; 19:S20–5.
27. Tseng FC, O’Brien TR, Zhang M, et al. Seroprevalence of hepatitis C
virus and hepatitis B virus among San Francisco injection drug users,
1998 to 2000. Hepatology 2007; 46:666–71.
28. Goldberg D, Burns S, Taylor A, Cameron S, Hargreaves D, Hutchinson
S. Trends in HCV prevalence among injecting drug users in Glasgow
and Edinburgh during the era of needle/syringe exchange. Scand
J Infect Dis 2001; 33:457–61.
29. Crofts N, Aitken CK. Incidence of bloodborne virus infection and risk
behaviours in a cohort of injecting drug users in Victoria, 1990–1995.
Med J Aust 1997; 167:17–20.
30. van den Berg CH, Smit C, Bakker M, et al. Major decline of hepatitis C
virus incidence rate over two decades in a cohort of drug users. Eur J
Epidemiol 2007; 22:183–93.
31. Thomas DL, Astemborski J, Rai RM, et al. The natural history of
hepatitis C virus infection: Host, viral, and environmental factors.
JAMA 2000; 284:450–6.
32. Poynard T, Bedossa P, Opolon P. Natural history of liver fibrosis
progression in patients with chronic hepatitis C. The OBSVIRC, MET-
AVIR,CLINIVIR, andDOSVIRCgroups. Lancet 1997; 349:825–32.
33. Grebely J, Raffa JD, Lai C. Impact of hepatitis C virus infection on
all-cause and liver-related mortality in a large community-based
cohort of inner city residents [published online ahead of print 25
February, 2010]. J Viral Hepat 2011; 18:32–41. doi: 10.1111/j.1365-
34. Manns MP, McHutchison JG, Gordon SC, et al. Peginterferon alfa-2b
plus ribavirin compared with interferon alfa-2bplus ribavirin for initial
treatment of chronic hepatitis C: A randomised trial. Lancet 2001;
35. Chung RT, Andersen J, Volberding P, et al. Peginterferon alfa-2a plus
ribavirin versus interferon alfa-2a plus ribavirin for chronic hepatitis C
in HIV-coinfected persons. N Engl J Med 2004; 351:451–9.
36. Jeffers LJ, Cassidy W, Howell CD, Hu S, Reddy KR. Peginterferon alfa-
2a (40 kd) and ribavirin for black American patients with chronic HCV
genotype 1. Hepatology 2004; 39:1702–8.
37. Mehta SH, Lucas GM, Mirel LB, et al. Limited effectiveness of anti-
viral treatment for hepatitis C in an urban HIV clinic. AIDS 2006;
38. Mehta SH, Genberg BL, Astemborski J, et al. Limited uptake of hep-
atitis C treatment among injection drug users. J Community Health
39. Grebely J, Raffa JD, Lai C, et al. Low uptake of treatment for hepatitis C
virus infection in a large community-based study of inner city resi-
dents. J Viral Hepat 2009; 16:352–8.
40. Hagan H, Latka MH, Campbell JV, et al. Eligibility for treatment of
hepatitis C virus infection among young injection drug users in 3 US
cities. Clin Infect Dis 2006; 42:669–72.
41. Foster GR, Hezode C, Bronowicki JP, et al. Activity of telaprevir alone
or in combination with peginterferon alfa-2A and ribavirin in treat-
ment-naı ¨ve genotype 2 and 3 hepatitis-C infected patients: Interim
results of study C209. In: Scientific Programme of the 44th Annual
Meeting of the European Association for the Study of the Liver
(Copenhagen, Denmark). Geneva Switzerland: European Association
for the Study of the Liver, 2009.
d JID 2011:203 (1 March)
d Mehta et al.