Abstracts Vol. 1 [International Conference on AIDS (11th: 1996: Vancouver, Canada)]

Track C: Epidemiology and Public Health
Results: 40.5% (268/662) of IDUs were HIV-positive. HIV seroprevalence was greater for
Puerto Ricans (45%) and blacks (44%) than for whites (32%). Most sexual and drug use
partnerships of IDUs were racially/ethnically homogenous (83% for sex and 77% for drug
use). In multivariate logistic regression, along with several individual risk behaviors, Puerto
Rican race/ethnicity (O.R. 1.9; 95% C.I. I.2-3.2), having an IDU in the egocentric risk network who is 5 years older than the subject (O.R. 1.6; 95% C.I. 1. -2.4). having any black
IDU in the egocentric risk network (O.R. 2.2; 95% C.I. 1.3--3.9), and being in the "Seidman 2 -core" of the largest sociometric network component (O.R. 1.8; 95% C.I.. I -3.2) were independent risk factors for HIV seropositivity Simultaneous consideration of individual risk
behaviors and risk network characteristics statistically "explained" the black-white variations
in HIV seroprevalence.The higher prevalence of HIV among Puerto Ricans persisted even
after multivariate adjustments.
Conclusions: Social networks of IDUs (how and with whom IDUs are linked; and largerscale patterns of linkages) affect the likelihood of HIV infection. Since HIV entered minority
IIDU networks prior to white IDU networks (Novick et al., 1989), minority networks seem
to function as "pockets of infection" (Friedman et al., 1995).Therefore., the extent of
racial/ethnic in-group risk relationships may increase the likelihood of HIV infection per risky
act for minority IDUs but limit the transmission of infection across groups. Social networks
explain some, but not all. of the racial/ethnic differences in HIV seroprevalence among
IDUs.
Benny Jose. Institute for AIDS Research, NDRI, I I Beach Street, New York, NY 10013, USA
Telephone: [phone] Fax: [phone] email: [email protected]
Tu.C.2538
RISK FACTORS FOR HIV AND OTHER PATHOGENS IN A COHORT OF INJECTION
DRUG USERS IN TREATMENT, SEATTLE,WASHINGTON
Hansen Gail R, Fields MJ. McGough JR Hagan H. Alexander ER. Seattle King County
Department of Public Health, Seattle,Washington, USA.
Objective: To determine the prevalence and epidemiologic correlates of HIV and other
sexually and parenterally transmitted diseases among a cohort of injection drug users in
treatment.
Methods: Six hundred thirty four injectors recruited from four drug treatment centers
completed an extensive structured interview.They were tested for past or current exposure to primarily parenterally transmitted infections (hepatitis C (HCV) and HTLV II), sexually transmitted infections (herpes simplex 2 (HSV2) and syphilis) and HIV and hepatitis B
(HBV), which are both parenterally and sexually transmitted.
Results: The mean age was 38.3 (~ 8.0) and the mean number of years since beginning
injection was I17. I (~ 9.9).
Pathogen Prevalence 95% Confidence Univariate analysis: Risk Factors [Relative Prevalence]
Interval (CI) (all Cl's exclude 1.0)
HIV i i% 0.4, 2.3 Steady sex partner HIV + [36.5]. Gay/bisexual male
[17.6]
HBV 67.7% 64. 1, 71.6 History of gonorrhea [1. 3], > 4 sex partners in past
6 months[ I1.3], Injected drug > I year [315]. Ever
used another's needle[.3]
HCV 90.3% 87.7, 92.5 Injected drugs > I year* [I 10.6], Ever used another's
needle [1.3]
H ITLVII 10.3% 7.7, I 3.6 Injected drugs > 10 years' [4. 1], Hisiory of syphilis
[4.1], Homeless [2.6]
HSV2 63.7% 58.5,68.7 >4 sex partners in past 6 months[I1.6]. Exchanged
money or drugs for sex [I.4]
Syphilis 4.5% 3.0,6.4 Gay/bisexual male [3.4]. > 4 sex partners in past 6
months [2.8], Exchanged money or drugs for sex [3.4]
*age adjusted
Conclusions: In a low HIV seroprevalence area, sexual and parenteral risk of HIV transmission may be indicated by other infections. HIV prevention programs will benefit from monitoring of other pathogens and their risk factors.
Gail R. Hansen, 106 Prefontaine PI., Seattle, WA 98104 USA.Tel: [phone]; Fax: (206)
205 5243.
Tu.C.2539
THE IMPACT OF HIV INFECTION ON MORTALITY OF INJECTION DRUG USERS
Chitwood Dale D.*, Comerford, M.*, McBride, D.C**, Chitwood, J.S.***, McCoy C.B.*,
Weatherby N.L.*. *University of Miami, FL; **Andrews University, Berrien Spnngs, MI;
***St.Thomas University Miami, FL.
Objective: To assess the impact of HIV infection on mortality among injection drug users
(IDUs).
Methods: 601 IDUs who participated in a longitudinal study to estimate prevalence and
incidence of HIV in Miami, FI. were monitored for mortality for six years (72 moniths) after
enrollment. Death was confirmed by death certificate. Descriptive techniques were used to
estimate and describe mortality and the Cox regression model was used to estimate cumulative survival functions and risk factors associated with mortality
Results: There were 72 documented deaths among the 601 study participants.The mean
age of those who had died was 39.The leading cause of death was AIDS-related (45.7%)
and the second leading cause of death was drug related (34.7%). Among the 37 HIV positive individuals who had died, 86.5% died of HIV related causes; an additional 5.4% died of
drug-related causes. Among those who were HIV negative at the time of death, 65 7% died
of drug related causes.
One-third (34.3%) of the cohort members who were HIV positive died within six years of

baseline compared to 8.4% of the HIV negative members. Using Cox regression modeling
for survival, there was a significant difference in survival between HIV positive and HIV negative IDUs. IDUs who were HIV positive had a mean survival time from baseline of 60
months (95% C.I. 56, 65) while those who were negative had a mean survival of 70 months

Tu.C.2538 -Tu.C.2542
(95% C.I. 69, 71). Gender and age were not associated with survival for either the seropositive or seronegative groups. Ethnicity was not associated with survival among those HIV
positive; Hispanics had a significantly shorter survival time than non-Hispanic whites or
African-Americans among those HIV negative.
Conclusions: AIDS-related death was the leading cause and drug-related death was the
second leading cause of death among this cohort of IDUs. In addition, those IDUs who are
HIV positive are dying at a higher rate and have a shorter survival time than those IDUs
who are HIV negative. IDUs always have been a group at high risk for early mortality and
the AIDS epidemic has increased this risk.
Dale D. Chitwood, Ph.D., 1400 N.W 10th Avenue, Rm 212 (D-93), Miami, FL U.S.A 33136
Telephone: [phone] Fax [phone]
Tu.C.2540
PERCEPTION OF RISK FOR HIV INFECTION AND ENGAGING IN HIGH RISK
BEHAVIOR AMONG INJECTING DRUG USERS IN RIO DE JANEIRO, BRAZIL
McBride. Duane C*, Inciardi J**, Surratt H***, Chitwood D***, Metsch L***, McCoy HV.
'Andrews University, Berrien Spring, MI; **University of Delaware, Newark, DE;
***University of Miami, Miami, FL; ****Florida International University, N. Miami, FL.
Objective: To assess the relationship between perception of risk for HIV infection and
engaging in high risk behaviors in a population of Injecting Drug Users (IDUs) in Rio De
Janiero, Brazil.
Methods: One hundred and thirty-two drug injectors were assessed (March, 1994 -December: 1995) for drug using behaviors, HIV risk behaviors including their perception of
their risk for HIV infection and sero-status.
Results: Two-thirds of the injectors perceived that they had little or no chance for HIV
infection. Only about 19% perceived that they had a high or a sure chance for developing
the infection. However, there was a consistent, moderate, and significant correlation (.30 or
higher) between engaging in such high risk behaviors as the use of used needles and high
numbers of sex partners and perception of risk for HIV infection. In addition, there was a
significant moderate relationship between perception of risk and serostatus.
Conclusion: While other studies have tended to show only a weak correlation between
perception of HIV risk and engaging in high risk behaviors among street injectors in the U.S.
Injectors in Brazil appear to engage in much less denial of risk based on their risk behaviors
than their counterparts in the U.S. Because of this, there may be a greater possibility for
intervening in the Rio population than in population that have a higher denial about the
relationship between risk behaviors and the probability of HIV infection.
Duane McBride, Andrews University, Berrien Spring, MI 49 103 Telephone: 61 6 47 I -3576
FAX 6 I 6-4713108 email:[email protected]
Tu.C.254 I
VIRAL HEPATITIS IN HIV POSITIVE MEN AND WOMEN WITH AN INJECTING DRUG
USE (IDU) HISTORY
Bean-Mayberry, Bevanne*, Kloser P*, Cowan J*, Passanante M**. New Jersey Medical
School, a*Department of Medicine, **Department of Preventive Medicine, Newark, New
Jersey, USA.
Objective: To determine the prevalence of viral hepatitis in our urban population of HIV
positive patients with and without an injecting drug use history
Methods: 259 patient charts were randomly sampled from the patient lists generated during one clinic month. Patient charts were assessed for completeness of hepatitis profiles and
risk factor screening. Fifty chart surveys were deemed inappropriate due to incomplete
documentation, negative HIV tests or missing laboratory results. From the remaining sample,
the following results were noted according to gender and injecting drug use admission by
the patient.
Results:
Total N=209 Women N= 115 Men N=94
IDU 48 (41.7%) 41 (43.6a%)
Non IDU 45 (39.1%) 45 (47.9%)
No Response 22 (I19.1%) 8 (8.5%)
HCV+IDU 37/48 (77.1%) 29/41 (70.7%)
HCV+Non IDU 10/45 (22.2%) 9/41 (22.0%)
HBcAb+HBsAb- 28/1 15 (24.3%) 40/94 (42.6%)
+IDU 22/28 (78.6%) 22/40 (55.0%)
+Non--IDU 4/28 (14.3%) 16/22 (40.0%)
Conclusions: We have a high proportion of individuals who have a history of injection drug
use at our HIV clinic site whether male or female. Both genders experience a comparable
frequency of hepatitis C virus. However, for persons with incomplete or decreased immune
response to hepatitis B, a greater percentage of women with IDU histories (78.6%) had
serum with hepatitis B core antibody and no surface antibody compared to men with IDU
histories (55%).While both are relatively high in fiequency the data may indicate that
women have a cormpromised immune response that is more related to their drug abuse
versus other risk factors. Multiple risk factors in this population will need to be assessed
according to behavior and gender in addition to continual hepatitis screening of all patients.
B A. Bean-Mayber y 618 Irvington Avenue, #2, Maplewood, New Jersey 07040,USA.
Telephone: 201 -76 I-4678. Email: [email protected].
Tu.C.2542
THE IMPACT OF THE HIV/AIDS EPIDEMIC ON DRUG INJECTING ATTITUDES AND
BEHAVIOR
Brown, Christopher C., Boyd, Denise. Arizona Department of Health Services, Office of
HIV/STD Services, Phoenix, Arizona, U.S.A.
Introduction: This qualitative mini-field research project conducted in November and
December I 995 posed the overarching question: How has the HIV/AIDS epidemic
changed drug use attitudes and behaviors among male and female injecting drug users

(IDUs) from two ethnic groups in the Phoenix metropolitan area.
Methods: Using a small convenience sample. 15 Hispanic and White IDUs were interviewed
at two in-patient drug treatment facilities in the Phoenix area. Interviewees were asked a
series of open-ended questions relating to their knowledge of HIV/AIDS and their drug

0
>
o
u
c03
<1
c
N
u
O
Ga
mC
e)
Q)
0
0
cx
358