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

Track C: Epidemiology and Public Health
Methods: Data from 2-4 visits of 865 women (565 HIV-positive and 300 fIV negative) in
the multicenter HERS project over an I 8 month period were analyzed. C( ervicovaginal
lavage specimens taken at each visit were tested for HPVs by PtR using conssa'ris LI
primers for amplification and one generic and 25 type-specific probes for diagnoses.
Persistence was defined as presence of the same HPV type in paired aiples ftrim an individual and was calculated for pairs of samples for 6 month, 12 monthi int n]8 month visits.
In individuals with multiple HPV types, each type was treated separately Persistence data of
24 HPV types were analyzed together: Robust methods for logistic regression were used to
provide standard errors adjusted by multiple observations on individls.
Results: At recruitment, the HPV prevalence was much higher in HIV ositive than in HIV
negative women (66% vs. 34%, p < 0.00 I). HPVs persisted much longer in IIV positive
women than in HIVnegative wonmen (median duration of HPV peri,,is I I 8 months vs.
6 months).At each time period (6, 12 and 18 months) the persiserece in HIVplositive
women (0.69, 0.58 anrid 0.54 of the HPVs) was significantly greater than that in HIV negative
women (0.49, 0.30 anrid 0.21 of the HPVs). In HIV-infected women, igeir psi sItece was
significantly associated with low CD4 counts ( <200 vs. >500) andir wit virmiiifple simultane
ous HPV infections (4 or more vs. 3 or less).A high viral load in the first specimen of the
pair (high signal vs. low signal) was a strong risk factor for persist ce r otl grocups of
women. Age ( <30 vs. >30) and cervical cancer risk category of HtPV ty l ( lrw crisk, inte r
mediate risk, high risk) were not associated with persistence in either g oup. In ddition. low
CD4 counts were associated with a higher HPV viral load in HPV prosiv a specirmens.
Conclusions: The markedly longer persistence and higher viral load probably account, in
part, for the high disease potential of HPV infections in immunosuppre, ed tHIV positive
women.
KV Shah, Johns Hopkins School of Public Health, 615 N.Wolfe St., Baltis,e. MD 21205
USA Phone: [phone] Fax: [phone] email: kusah(@phrnet.phihu.edu
Tu.C.2467
HEPATITIS CAN IMPORTANT CO-PATHOGEN IN WOMEN WITH HIV INFECTION
Hershow, Ronald C *, Kalish L * Sha B *,Till M *, Cohen M *. Wo ieni s Interagency HIV
Study (WIHS), Chicago Consortium, IL, USA; ** NERI, Watertown, MA, USA
Objective: To identify faictors associated with hepatitis C virus (HCV) infection in Chicago
women enrolled in WIHS from 10/94 to 5/95.The WIHS is a prospes livi, iV niatural history study conducted at 6 US sites.
Methods: We tested 269 Chicago women (231 with, 38 without fI itection) with a second generation HCV antibody (ab) enzyme immuno-assay (EIA). Repeater ily reactive specimens were confirmed with recombinant immunoblot assay (RIBA). b aseline ntervews ind
liver enzyme determinations ensued.
Results: Of 269 women, 108 (40%) were HCV--antibody (ab) (+), 146 (54%) IHCV ab (-),
and 15 (6%) indeterminant by RIBA; the latter were excluded. HCaV a prevalence was 45%
and 26% (p=.03) in those with and without HIVinfection. Among the HIVin-fec ted, HCV ab
prevalence was 32% in women with CD4 counts <200 and 52% in ti osc 200, p.01.
HCV ab prevalence was 9 I% in persons who, since 1985, had ever injecie dsr.,gs (IDU) vs
12% in non-users (p<.001 I). In analyses which adjusted for IDU, 6 other associations were
identified. In a multivariate model with these variables, HIV seroposit virty (p -.02), age > 35
(p=.00 I), < high school education (p=.03), history of gonorrhea (p= 00/) and IDU
(p<.001 ) were independently associated with HCV infection. Sex wi! anr IDU and blood
transfusions were not.The HCV-infected were more likely to have an alan ne arnrnotransferase >50 (36% vs 8%, p <.001), but there were no significant difflerfens Ihetween the
HCV-singly-infected (n= 10) and the HCV/HIV coinfected (n- 95).
Conclusions: HCV infection is highly prevalent in women with or at risk tar t-IIVinfection
and injection drug use is the strongest associated fiactor:The indeendnrit association of
HICV with prior episodes of gonorrhea suggests that sexual transmission contributes to
acquisition in this population.These data do not suggest that HCV hastens HIV disease progression or that HIV potentiates liver injury in the HIV/HCV coinfected.
Ronald C. Hershow, MD UIC School of Public Health Room 504 2 121 W.Trylor Chicago, IL
USA 6061 2 Tel: [phone] Fax: [phone] email:rchersho(uic.edu
Tu.C.2468
SEXUAL ABUSE AND HIV RISK
Paone Denise*, Miller- M*, Shi Q*, Des Jarlais DC*. Beth Israel Medical Center; New York,
NY USA
Objective: To determine the impact of previous sexual abuse on currenit HIV sex and drug
risk behaviors.
Methods: One hour face-to-face interviews were conducted with 130 womern opiate users.
Subjects were recruited fr-om methadone maintenance and syringe excihange progams in
New York City Sexual abuse and family histories, drug use and sexual behaviors for ther past
six months, and self-reported HIV status were ascertained.We constructed a famiy history
scale consisting of parental substance use in family of origin, maternal sexuai abuse history,
sibling sexual abuse history and initiating drugs to escape famiy p or s
Results: The rean age of the 130 womnen interviewed was 36.5 (+i 7.5) years, with the
sexual abuse group being on average two years younger (35.6 vs 37.5). Ihe women were
ethnically diverse: 54% Latina, 32% African American, 21% White. sixty eiglt (52%) reported
a sexual abuse history; more than half (56%) of these reported sexua 'buse prior to noninjecting drug use.When sexual abuse history was entered into a oeigiis regrestsion model
controlling for family history and HIV status, results indicated that womns witts sexual abuse
histories were much more likely to have engaged in sex work in the past six months
(OR=5.2, 95% Cl= 1.8, 15.5) than women without such histories.WIen dt u us e -utcores
were examined, again controlling for both family history and HIV at, 'art reic with sexual
abuse histories were more likely to currently inject (OR=3.4, 95% l's.../), to nlect
cocaine daily or more (OR=2.7, 95% Cl= 1.0, 7.5).
Conclusions: The data indicate that a history of sexual abuse, independent of family pathology and HIV status, has significant influence on current HIV risk behaviors.
D. Paone, First Av. & 16th St., New York, NY 10003, USA Telephone: 2 Tb 718, ext.

5773 Fax 2 I 2-387-3897 email: [email protected]

Tu.C.2467 -Tu.C.2473
Tu.C.2469
STRUCTURE OF AIDS RISK FOR FEMALE SEXUAL PARTNERS OF INJECTION DRUG
USERS
Lewis, Yvonne IL,Young P*, Willianms ML*, Wild. J*. *NOVA Research Company Bethesda,
MD, USA
Objectives:l The project explored the contextual conditions that impact on the experiences,
norms, and values of female sexual partners of injection drug users (IDUs); their choice of
IDUs as their sexual partners: and the factors that guide their decisions-making and patterns
of behavior in particular their HIV-related risk behaviors-vis i vis their sexual partners and
others.
Methods: The project utilized an ethnographic approach designed for a quick, yet in- depth,
collection of descriptive data on I189 female sexual partners of IDUs in 5 cities. In 2-hour
interviews, data were collected on relationship expectations and ideals, attra iction to part
neors, dynamics of relationships, feelings about partners and their drug inrectonr the women's
own drug use, their sex lives, and why the women stayed in the relationships. A mnemonic
coding system based on major -themes in the interviews across sites, was used. Coded
"strips" of descriptive information were then comrrpiled using thIe AFTER computer progra
to idcrntify each descriptive entr y Frequency tables were generated to identity theme,nd
'sred, is the women's experiences and behaviors.
Results: The risks assoiated with HIV/AIDS transmission among female sexual partner, of
I[Us reflect a n~exus of socio-econornic factors, cultural norms, and gender socialization. A
woman's "cioice"-intentional or not-of an IDU as a sexual partner is conditioned by factors
seemingly beyond her control: a history of parental abuse; lack of childhood love and attention; feelirs of unworthiness and vulnerability to rnmale attention; cultural norms and a
socialization process that create gender role expectations for both self and partner: and
lack of economic self-sufficiency
Conclusions: It was concluded that meaningful strategies to reduce HIV-retlated risk behaviort n for this population cannot be effected without an understanding of the link between
risk and individual choice, and the impact on that choice of family history cultural norms,
feelings of self-power gender role expectations, and self-sufficiency
YP Lewis, 4600 East-West Hwy Suite 700, Bethesda, MD 20814, USA Telephone: [phone] I Fax 30I 986-4931 email: ylewis(@aoada.ssw.dhhs.gov
Tu.C.2470
DETERMINANTS OF CONDOM USE AMONG WOMEN AT RISK FOR HIV
Tesselaar Helen, O'Connell A, Stark M, Person B, Cohen A, Walls C, Hembry K
Objective. Amon g women at risk for HIV 1) to determine the association between contraceptive method and condom use with main and casual partners and 2) to assess the relalionship belween psychosocial and demographic factors and condom use with main and
caSUal par ne5s.
Methods: In 1993 -94, the CDC sponsored Prevention of HIV in Women and Infants
Demonstr.tion Project recruited women from community and facility settings from 8 sites
in 6 U.S. cities for a structured interview regarding disease prevention and reproductive
Ihealth. For these analyses, eligible women were from 15 - 34 years old, not HIV+, had been
sexually active within the last 30 days, and had at least one risk factor for HIV infection
(injection drug use, multiple sex partners or a casual sex partner: or a matuin partner who
either injected drugs or had other sex partners.) Two chi-square analyses were conducted
to determine the relationship between women's usual contraceptive method and consistent
condom use (every time for at least I month) with main (n=607) and casual partners
n --606). In addition, two sets of three logistic regressions were conducted with stages of
change (5OC) of condomr use with main (N=891) and casual partners (n=865) as the
depe ndent variables.
Results: Only 19%2 of women used condoms consistently with their main partners and 46%
with their casual partners.Women who used condoms as their usual method of contraception were more likely to report condom use at last intercourse with main or other partners than were women who were surgically sterilized (p<.0 I) or who used the pill,
Norplantfc, Depo Provera~, or an IUD (p<.0 I). Having confidence in her ability to use
condoms under a variety of circumstances (self-efficacy) (p<.05), talking about condoms
with others (p<.05), and perceiving the advantages of condoms were associated with higher stages of condom use (p<.05).
Conclusions: Family planning practitioners should consider advising condoms for contr.ception for worsen at risk for tHIV infection. Interventions are needed to I) increase self effica-y by encouraging women to practice ways to use condoms in difficult situations and 2)
stimulate women to discuss and negotiate condom use.
Helen M. Tesseiaar 800 NE Oregon St, Ste 370 [phone] Portland OR 97232 USA
(50.3) 731 -3095 (FAX)
Tu.C.2473
SEXUAL BEHAVIORS AND SEXUAL IDENTITIES THAT PUT YOUNG LESBIANS AT
RISK FOR HIV AND STD'S
Coctln susan D I), MaysVM.*. *University of California, Los Angeles, USA; OCalifornia
State University Northridge, USA
Objective: Young lesbian and bisexual women may be at greater risk for HIV and other
STI-s than older lesbian and bisexual women due to sexual experimentation with both
men and women (Hunter et al., 1993) and the higher rates of partner change seen among
young adults (CD)C, 1992). Behavior changes in response to DIV and other STDs are not
well documented among this population.
Method: We examined prevalence of sexual behaviors in a two ethnically diverse (African
Atteca Asian American, Latina and White) samples of lesbian and bisexual women, aged
I8 to 25 'sears, recruited in 1993 (N I 81) and 1995 (N - 271) from gay pride festivals
in lsis Angeleos (ounty. Women completed brief surveys assessing both homosexual and
heterosexua sexual behavior in the prior year and self-reports of recent sexually transmitted diseases.
Results: Prev ilerce of sexual behaviors over the two year period were stable. Reports of
prevalent HIV inection were rare (0.5%); approximately 7% reported another STD in the

prior yea: Although few women indicated using protective barriers during sex with other
women, rates of partner change were not high, with women reporting 1.0 median number
of fermale sex partners in the prior year. Nevertheless, 26% of women report having had

O,
4)
ad
0
to
>
cO
to
u
0
0
O
U
c0
c
nt
to
0

346