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

12th World AIDS Conference

Abstracts 23455-23459 433

Conclusion: Brief (2 to 4 session) standardized interventions appear to be effective in assisting women with HIV sex and drug risk behavior change. However,
specialized women-centered interventions that focus on sexual risks - including
how drug use impacts the sexual behavior of women - are needed so that greater
risk reduction can be achieved.
S23455 1 The relationship between HIV risk behavior and partner
violence among women on methadone
Nabila EL - Bassel1, L. Gilbert2, V. Rajah2, V. Frye2, RF. Schilling2. CUSSW
622 West 113 th Street New York NY 10025 USA;2 Columbia University, New
York, NY USA
Objective: To examine the phenomena of violence by an intimate partner and
HIV risk behavior among women on methadone. The study explores how interpersonal context, gender inequalities, victim's drug abuse and that of the perpetrator
contribute to drug and sexual HIV risks in women.
Methods: Data were collected through focus groups with 73 women recruited
from methadone clinics. A woman qualified for the study if she reported any
physical or sexual abuse during the prior year. The majority were Latina (57%)
and African-American (25%). Phenomenological and feminist approaches were
used to guide the focus groups. Data were analyzed using a grounded theory
approach.
Results: Themes generated from the focus groups: 1) rape and sexual assault
occur under four situations: when the woman was sleeping, intoxicated, or overdosed, or she refused to share drugs; 2) physically forced to have unprotected
sex when she had no desire for sex as the result of crack use; 3) physically
abused if she insisted on using the female/male condom; 4) forced to trade sex
to supply drugs for herself and her partner; 5) forced to share drugs and needles; 6) physically abused when she disclosed having extra-relationship affairs or
contracting an STD.
Conclusion: Qualitative research on the context of the connection between
HIV risk and partner violence is needed in order to inform the design of HIV
prevention strategies for drug-using women.
23456 HIV-related symptoms by injection drug use (IDU) after
accounting for CD4~ count and viral load in women
Anne Rompalo, D.K. Smith1, D. Warren1, E.E. Schoenbaum2, P. Schuman3,
C.J. Carpenter4, J. Astemborski5. Rm 1167, 720 Rutland Ave., Baltimore, MD;
1Centers for Disease Control, Atlanta, GA; 2Montefiore Medical Center, Bronx,
NY; 3 Wayne State University, Detroit, MI; The Miriam Hospital, Providence, RI;
5Johns Hopkins University, Baltimore, MD, USA
Objectives: To compare the prevalence of HIV-related symptoms, physical exam
findings, and hematologic variables among women whose risk for HIV is IDU
versus sexual contact (SC), and evaluate the influence of HIV plasma viral load
and CD4 cell count on clinical manifestations according to risk.
Methods: Participants of the HIV Epidemiology Research Study (HERS) (a
multi-centered, prospective, controlled study of HIV infection in women) were administered a risk behavior and symptom interview, physical exam, and had hematologic testing including CD4 cell counts done on study entry. Plasma HIV-1 viral
RNA were performed on stored sera using a second-generation branched-chain
DNA signal amplification assay (Chiron Corp) with a quantification limit of 500
copies/mL. CD4 counts were categorized as <200, 200-500, and >500; HIV viral
loads were categorized as <500, 501-9,999, 10,000-29,999, and >30,000. Data
are available for 1159 HIV infected women: 604 were IDUs and 555 were Scs.
Results: In comparing SC women to IDU women, IDUs were significantly (p
S.05) more likely to present with memory loss, weight loss, history of sepsis
and endocarditis; were less likely to report vaginal herpes history; and were
more likely to have an enlarged liver, hematocrit value <34 and platelet counts
150,000/ml controlling for CD4 count category, HIV viral load category, or both
groupings simultaneously. Women with higher HIV viral load categories were
significantly more likely to report diarrhea, fever, thrush, have hairy leukoplakia,
oral candidiasis, HCT <34, and low platelet counts controlling for both CD4 and
risk categories (p <.05).
Conclusions: Overall, these cross-sectional data suggest that clinical manifestations of HIV infection is similar between women in IDU and SC risk groups,
regardless of CD4 cell counts or HIV viral load categories. The few differences
observed probably reflect effects of IDU rather than HIV infection.
S23457 Continuing risk behaviors of HIV positive women
Lisa R. Metsch, B. McCoy Clyde, Lai Shenghan, A. McAnany Heather. 11400
NW O10th Avenue Room 1110 Miami Florida 33136; 2University of Miami, USA
Objectives: This study investigates continuing drug use and sexual risk behaviors among HIV seropositive chronic drug users in Miami, Florida and the impact
of gender on behavior change.
Methods: Data were collected from a street-recruited sample of HIV positive

injection and non-injection drug users using a standardized risk behavior assessment questionnaire. We focus on 498 injection and non-injection drug users who
tested HIV positive at baseline and then completed at least one follow-up visit
after participating in post-test counseling where the vast majority learned their
HIV status for the first time. We compare respondents' risk behaviors at baseline
to their first follow-up visit at least six months after they had received news of
their HIV status during post-test counseling. During these follow-up visits, respon
dents were administered a risk assessment which elicited the same information
collected at the baseline interview.
Results: From baseline to follow-up, injection drug using women practicing unsafe sex decreased from 74.5% to 45.9%, and unsafe needle practices (re-use of
needle/syringes, and paraphernalia) decreased from 68.0% to 32.6%. However,
approximately one-half (45.8%) of injection and non injection drug using (50.4%)
women continue unsafe sexual activity and needle practices. For gender impact,
women compared to men were less likely to change sexual risks and more likely
to change injection risks.
Conclusions: While HIV prevention activities constitute the major thrust of the
public health's effort to control this spreading epidemic, persons already infected
may continue risky drug use and sexual behaviors which further impact their
health and the infection of others. These findings underscore the need to develop
effective risk reduction intervention models targeting HIV positive drug users,
especially women.
23458 A population-based, door-to-door survey of HIV, STD and
risk behaviors among young women in California, USA
Juan Ruiz1, F. Molitor2, W. McFarland3. Young Women's Survey Group; 1830 S
Street Sacramento CA 95814; 2CA Dept Health Services Sacramento CA;
3AIDS Office San Francisco DPH San Francisco CA, USA
Objective: To assess the prevalence of HIV, STD and related risk behaviors
among women aged 18-29 who reside in low income neighborhoods in five
Northern California counties (USA).
Methods: In a randomized, population-based, door-to-door survey approximately 2,500 young women were administered a standardized questionnaire in
English or Spanish. Blood and urine were tested for HIV, hepatitis B and C,
syphilis, gonorrhea, and chlamydia. All inhabited dwellings from randomly selected blocks were identified. At least 3 attempts were made to locate and recruit
eligible women at each dwelling. The participation rate exceeded 70%. Data were
collected from April, 1996 to January, 1998.
Results: 42% of participants were Hispanic; 32% were African American; 16%
were European American; and 3.7% Asian American. Among specimens tested
to date, HIV positivity was 0.32%, hepatitis B 0.93%, hepatitis C 4.0%, syphilis
0.86%, gonorrhea 0.61%, and chlamydia 3.12%. 3.2% of women reported injection drug use. The mean number of sex partners in the previous 6 months was
2.5. Only 38% of women used a condom the last time they had sex with new or
casual partners.
Conclusion: Although logistically difficult and expensive, population-based
surveys are needed to provide representative information on HIV risk among
marginalized or difficult to reach populations. Young minority women living in
low income neighborhoods may not access existing HIV prevention programs
and messages. Outreach programs appropriate for Latina and African American
women are needed in diverse Northern California communities.
23459 A method of adjusting HIV prevalences derived from
pregnant women to allow for the effect of fertility
Angus Nicoll1, J. Stephenson2, A. Griffoen2, E. Boisson3, S.J. Cliffe1,
P.A. Rogers4. 1PHLS AIDS & STD Centre, CDSC, London; 2Department of
STDS, University College, London, London; 3London School of Hygiene &
Tropical Medicine, London; 4PHLS Statistics Unit London, UK
Objective: To devise, apply and validate a method for adjusting HIV seroprevalence derived from pregnant women for estimating population prevalences among
all women of child-bearing age.
Design: A methodological study using 1) observed fertility data from known
HIV infected women, 2) fertility data among all (infected and uninfected) women,
3) the results of unlinked anonymous HIV surveys of pregnant women.
Methods: Fertility (live births and terminations of pregnancy) rates were calculated from 505 known HIV infected women in the United Kingdom; before and
after HIV diagnosis and according to their route of infection (exposure category).
Comparisons were made with routinely available birth and termination rates for
the whole population to derive Relative Inclusion Ratios (RIRs), the relative probabilities that HIV infected women would be included in a unlinked antenatal or
termination survey compared to an uninfected women. Validation was by applying the birth and termination RIRs to predict differences in HIV prevalence
between antenatal and termination clinic groups in unlinked anonymous surveys
and comparing the results to what was empirically observed.
Results: The derived RIRs for live-births were close to unity; 1.03 (95% C.I.
0.90-1.17) for London and 0.8 (0.71-0.89) for elsewhere. Sensitivity analysis indicated that RIRs could divert from unity according to alterations in the proportion
of infections diagnosed and the balance of exposure category groups among HIV
infected women. RIRs in London were twice as high for terminations, principally
because of higher termination rates among women who knew they were HIV infected. In London hospitals the ratio of observed HIV prevalences among women
having terminations and those having live births was 2.07:1 (0.62% to 0.30%),
very close to what was predicted from comparing the RIRs for women having live

births and terminations.
Conclusions: HIV seroprevalences from pregnant women are likely to need
adjustment for differential fertility rates among HIV infected women if they are
to be applied to whole populations. A general method for adjustment has been
devised and validated. Routinely monitoring fertility rates of HIV infected women
is a useful adjunct to unlinked anonymous surveillance.