Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

12th World AIDS Conference

Abstracts 23189-23192 379

response instruments and methods for evaluating and responding to HIV and
other health risks associated with injecting drug use. A Rapid Assessment and
Response (RAR) methodology and guide, and a survey instrument, have been
developed to collect data on HIV risk behaviour and prevalence from injecting
drug users in 20 cities. Testing the RAR highlighted the importance of community participation in the development of responses. Training in the application of
the RAR guide has been undertaken and further data collection using the RAR
guide and the survey instrument will continue through 1998/99. These data will
inform interventions to reduce HIV/AIDS and other health consequences related
to injecting drug use.
Lessons Learned: HIV epidemics associated with injecting drug use can be
avoided or reversed if assessment and response are rapid, appropriate and
involve the target communities.
23189 "Doctors" and "Patients" - Risk behaviors among IDUs
who give injections to or receive injections from other
drug users
Alex Kral2, R.N. Bluthenthal2, J. Lorvick2, E.A. Erringer2, B.R. Edlin2. 13180
18th Street Suite 302 San Francisco, CA; 2Urban Health Study, USCF, San
Francisco, CA, USA
Objective: To assess prevalence and risk factors associated with injection drug
users (IDUs) who give injections to or receive injections from other drug users.
Background: Video-taped injection episodes of IDUs have shown that blood
contact frequently occurs within the context of giving or receiving injections, posing risk for transmission of HIV and viral hepatitis (Flynn 1996). After review of
the literature, we found no studies on the prevalence of these activities.
Methods: IDUs were street-recruited, interviewed using a standard quantitative
instrument, and provided with HIV testing and counseling in the San Francisco
Bay Area from August 1996 through January 1997. "Doctors" are IDUs who
reported giving injections to other IDUs in the past month. "Patients" are IDUs
who reported receiving injections from other people in the past month.
Results: Of 1,166 IDUs interviewed, 283 (24%) reported being patients and
427 (37%) reported being doctors, including 132 (11%) who reported both. Patients and doctors were significantly more likely (p < 0.001) than other IDUs to
report sharing syringes (33% of patients, 21% of doctors, and 7% of others),
cookers (71%, 55%, and 24%), filters (60%, 47%, and 19%) and rinse-water
(45%, 35%, and 13%). In a logistic regression model which controlled for race,
HIV status, homelessness, crack cocaine, and syringe exchange use, patients
(Adjusted Odds Ratio = 4.29; 95% Confidence Interval = 2.76, 6.65) and doctors
(AOR = 1.91; 95% CI = 1.20, 3.04) were more likely than other IDUs to report
having shared syringes.
Conclusions: Giving and receiving injections is common among IDUs in the
SF Bay Area. IDUs who engage in these behaviors are more likely to engage in
other injection-related risk behaviors. Interventions need to address these issues
and provide practical solutions, including teaching IDUs how to properly inject
themselves and how to avoid blood contact when giving or receiving injections.
Epidemiological studies are needed to better understand the infectious disease
risks associated with giving and receiving injections.
23190 1 A decade of HIV prevention among injection drug users
in Toronto, Canada
Peggy Millson1, T. Myers2, L. Calzavara2, C. Major3, M. Fearon3, J. Rankin4,
J. Rigby5. 'Hodd, Room G314, 2075 Bayview Ave., Toronto Ontario M4N 3M5;
2HIV Studies Unit, Fac. of Medicine, Univ of Toronto, Toronto, ON; 3HIV Lab,
Ontario Ministry of Health, Toronto; 5Dept, of Comm. Hlth. and Epid., Dalhousie
U, Halifax, NS, Canada; 4Alcohol and Drug Treatment Services, Sydney, NSW,
Australia
Background: Planning for the first official HIV-prevention programme for injection drug users began in Toronto in 1988. Repeated cross-sectional studies of
out-of-treatment drug users have shown a moderate increase in seroprevalence
since then, but not the very rapid rise seen in Vancouver, for example. Recent
data from a study based at the public health department's needle exchange, were
compared to data from previous studies in the same area conducted since 1989,
to examine population trends in HIV prevalence and reported risk behaviours.
Methods: Clients attending the needle exchanges' fixed site were asked to
complete a questionnaire, with staff assistance if required, and to provide dried
blood spot and saliva samples for anonymous, unlinked testing. Participants had
to have injected within the preceding 2 months, and were paid $10. Findings
were compared to earlier studies (1989, 1991-94) which were similar in design
but used a longer questionnaire designed as part of the WHO Multicity study of
HIV and Injection Drug Use.
Results: Crude seroprevalence rates among out-of-treatment IDU were as follows: 1989- 4.3% (95% C.I. 3-6%); 1991-92- 5.7% (95% C.I. 4-10%), 1992-93 -4.8% (95% C.I. 3-8%), 1993-94- 9.0% (95% C.I. 6-12%); 1997-98- 9.5% (95%
C.I. 3-14%). Both heroin and cocaine have been widely used throughout this
period. Smoking of crack cocaine has also been common, with a more recent rise

in injection of crack. Data on numbers of sharing partners from the 1997 data suggest although over 30% of IDU continue to share to some degree, this is usually
reported to be uncommon (once or twice in 6 months), and is commonly limited
to one other person. This may confirm the key informant observation that the
drug scene in Toronto is relatively diffuse, with prevention services decentralized
to meet this need, including street outreach on foot and using mobile vans.

Conclusions: The pattern of HIV prevalence among Toronto injection drug
users suggests a gradual upward trend. While the relatively diffuse nature of
Toronto's drug scene may make service delivery more complex, it may also
reduce risky contacts among injectors. Recent trends including a rise in injection
of crack cocaine, and a reported rise in homelessness and marginalization may
both pose serious threats for maintaining relatively low levels of transmission.
Further research is required to determine the impact of these and other factors to
assist prevention.
23191 1Needle exchange use and HIV risk behaviors among
adolescent and young adult recently initiated injection
drug users (the REACH I study)
Meg C. Doherty1, R.S. Gafein2, B. Junge2, D. Vlahov2. 159 Hampstead Road #1
Jamaica Plain MA; 2Johns Hopkins School of Public Health, USA
Background: To describe factors associated with needle exchange program
(NEP) use in a cohort of young, recent-onset injection drug users (IDUs) in
Baltimore, Maryland.
Methods: Between 1994-96, interviews were conducted to assess lifetime
and current injection practices and sexual behaviors among 18 to 25 year old
IDUs recruited via street outreach. Concurrently, the first Baltimore NEP opened
in 1994. Serum was tested for antibodies to HIV and data were analyzed using
logistic regression methods.
Results: At study enrollment, among 229 IDUs, mean age at initiation of IDU
was 20 years, mean duration IDU was 3 years, 33 (14%) were HIV-positive, and
51 (22%) were in the NEP. More HIV+ persons compared to HIV-persons used the
NEP (39% vs 20%, p = 0.01); yet due to concurrent enrollment periods, prevalent
HIV infections and lifetime risk behaviors preceded first NEP use. Therefore,
when controlling for HIV serostatus, demographic factors associated with NEP
use included: age >23 (OR = 1.8; 95% CI, 0.9-3.4), African-American race - a
possible marker for proximity to the NEP as it was located in primarily AfricanAmerican neighborhoods (OR = 4.0; 95% CI, 1.3 -11.7), and prior paraphernalia
arrest (OR = 2.0, 95% CI, 0.9-4.5). Injection practices associated with NEP use
included injecting >5 x/day (OR = 3.0; 95% CI, 1.4-6.5) and cocaine injection
(OR = 1.97; 95% CI, 1.0-4.0). Harm reduction measures associated with NEP
use included having needles on one's person at the time of interview (OR =
7.4; 95% CI, 3.0-18.5) and a trend towards reduced needle sharing (OR = 0.65,
95% CI, 0.3-1.3). Trading sex for money or drugs was increased among NEP
users (OR = 2.0; 95% CI, 1.0-3.8). In a multivariate model, having needles on
one's person (OR = 8.6; 95% CI, 3.0-25.2), being African-American (OR = 6.2;
95% CI, 1.9-20.6), and injecting >5 x/day (OR = 2.6; 95% CI, 1.0-6.5) were
independently associated with NEP use when controlling for HIV infection.
Conclusions: Among young, recently initiated IDUs, NEP use was higher
among those with greater lifetime histories of risky behaviors; however, NEP
users also reported increased levels of current risk reduction behaviors. Followed
over time, these NEP users may further reduce risk behaviors. The fact that IDUs,
with a history of paraphernalia arrest, felt free to carry needles should increase
their likelihood of practicing clean and safe injections everytime-fulfilling a goal of
needle exchange programs.
23192 Incidence of HIV infection among injection drug users in
Philadelphia
David Metzger1, G.E. Woody2, N. Galai3, H.A. Navaline1. University of
Pennsylvania 3900 Chestnut Street Philadelphia PA 19104; 2University of
Pennsylvania/VA Medical Center Philadelphia PA, USA; 3Ben-Guiron University
Beer-Sheva, Israel
Background: The assessment of factors associated with HIV incidence among
high risk populations is necessary in order to both target and evaluate prevention
interventions. The objectives of these analyses are to identify factors associated
with HIV incidence among injection drug users in Philadelphia.
Design: Longitudinal study of risk behaviors and HIV infection.
Results: The Risk Assessment Project prospectively followed 347 initially
seronegative injection drug users (IDUs) from Philadelphia. This sample ranged
in age from 20 to 68 years (median = 39), 24% were female, and 62% were
African-American. Serologic and behavioral assessments were completed at six
month intervals. A total of 33 (9.5%) participants became infected with HIV and 54
(15.6%) were determined to have died between 1989 and 1996. A total of 1259
person years of follow-up were accrued during the study interval. Crude incidence
for the entire cohort was 3 infections per 100 person years of follow-up. Incidence
rates were significantly higher during the first three years of the study period. In examining participant characteristics, the highest rate of seroconversion was found
among those who were recruited while they were out of methadone treatment
(4.71 per 100 person years); the crude conversion rate for African-Americans
was found to be 3.66, and 3.03 for males. Using a multivariate Poisson-based
regression model and controlling for gender and age (found non-significant in
univariate analyses), relative incidence rates were 3.61 for African Americans vs.
non-African Americans and.33 for those recruited while they were in methadone

treatment compared to those enrolled while not in treatment.
Conclusion: HIV incidence rates among the RAP cohort suggest significant
protection from HIV infection associated with methadone treatment and elevated
risk among African-American injection drug users in Philadelphia. These factors
can be seen as useful in targeting prevention interventions to those at highest
risk of HIV infection.