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

Tu.C.2474 -Tu.C.2478

Tuesday July 9, 1996

sex with men in the prior year and of these 19% reported sex with a gay man in the prior
3 months.Two thirds of women who had sex with a gay man were teenagers and all selfidentified as bisexual or something other than lesbian.
Conclusions: Findings underscore the need to target HIV prevention service to both
bisexual women and teenage women among the population of women who have sex with
women
S.D. Cochran, Ph.D. CSUN/UCLA, 405 Hilgard Ave., Box 951563 Los Angeles, CA 90095 -1563 fax [phone] email: [email protected]
Tu.C.2474
HIV COUNSELING AND TESTING OF WOMEN PRESENTING FOR ROUTINE GYNECOLOGIC VISITS IN A NEW YORK CITY CLINIC: RECOMMENDATIONS FOR CREATING A MODEL FOR TESTING PATIENTS AT RISK
McGuinness Kathleen, Dozier A, Niera C, LaGuardia K, Biggers S. New York HospitalCornell University Medical Center, New York., NY
Objective: To evaluate HIV counseling and testing (C&T) for the gynecological population,
at two affiliated New York City Clinics.We will assess the risk factors associated with HIV
seropositivity and compare the risk profile of women who accepted HIV testing, and those
who declined.
Methods: From I / 1/95 through I 2/3 1/95 there were 10,960 gynecological visits at NYHCUMC and the NYH-Western Queens clinic.Women seen for C&T were selected by
physician referral, patient self-referral, and random selection.Test acceptance rate and seroprevalence rate were determined. Risk factors assessed were: Injecting drug use (IDU) sex
with IDU, sex while using non- injection drugs, sex for drugs/money(SDM), sex with bisexual men and history of sexual assault. Statistical analysis was performed using chisquare,with Fishers exact test for statistical significance.
Results:
HIV Counseled HIVTested Test Acceptance HIV Positive Seroprevalance
NYH-CUMC 582 324 56% 6 1.8
NYH-Queens 622 401 69% 4.9
Total 1204 725 64% 10 1.4
Of the 1,204 counseled 725 (64%) accepted, and 10 were positive for a seroprevalence
rate of (I.4%).
The only risk factor associated with HIV seropositivity was the exchange of (SDM), (OR
26.8, p=.005). IDU was a risk factor in only I / 10(10%) HIV positive women. None of the
HIV positive women reported sex with IDU or sex with bisexual men.The risk profiles
between these women accepting HIV testing and those declining, revealed no significance
statistical data.Women accepting testing reported a history of sexually transmitted diseases
more often than women declining testing (OR 1.9, p=.05).
Conclusions: In our population, identifying women at risk was difficult.The only traditional
risk faitor associated with HIV seropositivity was exchanging SDM.The patients declining
testing reported an equivalent level of risk to those accepting testing. Universal C &T is not
feasible due to the exceeding number of 10,000 clinic visits last year: A model for HIV C&T
needs to be developed in this population. Recommendations include: 1) providing a self
learning module on HIV infection and risk reduction 2) a simplified consent form 3) integration of HIV counseling with reproductive health care and 4) physician initiated HIV C&T.
Kathleen McGuinness, 525 East 68th Street HT582, NewYork, NY 10021 Telephone: [phone] Fax: [phone]
Tu.C.2475
HIV RISK PROFILE OF DRUG USING WOMEN WHO HAVE SEX WITH WOMEN IN
19 U.S. CITIES
Kral, Alex H-, Lorvick J, Bluthenthal R,Watters J). * Urban Health Study, IHPS, University
of California, San Francisco, U.S.A. n"(Deceased Nov 20, 1995)
Objective: To analyze HIV related risks of women drug users who have sex with women.
Methods: Injection drug (IDU) anrid crack cocaine (CCU) users were recruited from street
settings using targeted sampling techniques in 19 U.S. cities between 1992 and 1994
through the NIDA Cooperative Agreement.The cities were geographically diverse: 4 in the
Northeast, 4 in the South, 4 in the Midwest, and 7 in the West. Eligibility criteria included
being at least 18 years of age and riot in drug treatment at time of interview. Respondents
received an HIV test (ELISA, with Wb confirmation) and were interviewed using a risk
behavior assessment questionnaire.We analyze data on 231 women who reported woman
to-woman sexual (WSW) behavior in the 30 days prior to interview.
Results: The sample was 57% African American, I 6% white, I 1% Latin American, I 1%
Caribbean Hispanic, and 4% other races. One hundred and nineteen were IDUs (7 I of
whom also reported crack use) rind 112 were CCUs (none of whom reported injection
use).Thirteen percent were HIV positive. Sixty-two percent had a history of sex work.
Twenty one percent had multiple female sex partners in the previous 30 days. Only I 6
(7%) of women had used barrier protection during oral sex with women in the 30 days
prior to interview. One hundred and ten women (50%) also had sex with men in the previous 30 days. During vaginal sex with men, only 30% reported always using condorns. Of the
23 women who had anal sex with men in the 30 days prior to interview, only 26% reported always using condoms. In the 30 days prior to interview, 53% of the IDU women had
shared syringes, and 66% had shared other injection supplies. IDUs were more likely than
CCUs to be Latino (33% vs. I 2%; p<O.O001), have a history of hepatitis B (22% vs. 8%;
p<O.O05), have ever been in drug treatment (7 1% vs. 53%; p<.O05), and to have made
over $500 in the last month (67% vs. 5 1%; p<O.OI). CCUs were more likely to have had
multiple female sex partners (27% vs. / 6%; p<0.05), to be African American (76% vs. 39%;
p<0.000 I), to have drank alcohol in the past 48 hours (72% vs. 46%; p<0.0001), and to
have had sex for drugs (58% vs. 38%; p<0.002).
Conclusions: There is a high prevalence of risky sex and drug behaviors among women
IDUs and CCUs who have sex with women. Health education and HIV/AIDS prevention
efforts need to pay attention to specific needs and behaviors of WSWs.
Alex H. Kral, Uran Health Study UCSF, 31 80 I 8th Street, Suite 302, San Francisco, CA
94 1I0 U.S.A.Tel: I-4 15-502-722 I, Fax: I -4 15-476-3406,
e-mail: [email protected]

Tu.C.2476
SEXUAL IDENTITY AND BEHAVIOR AMONG WOMEN WITH FEMALE SEXUAL
PARTNERS:THE WOMEN'S INTERAGENCY HIV STUDY (WIHS)
Barkan S*, DGernrt C,orun.-1, Stonis, Lucille F., Lucey M, Wilson T Kilpatrick S,
Denenberg P, /r.nick. for the WIHS Collaborative Study Group, *New England
Research Institutes, Watertown, MA, USA
Objective:To desrnbe the exual identity and behavior of women (WM) who report having sex with other VWh/l it an effort to enhance recognition and understanding of the range
of behaviors and par tresr choices.These data have implications for the development and
targeting of HIV and other STD education and prevention efforts.
Methods: Data fiom the WIHS, a prospective study of 2080 HIV(+) and 575 HIV(-) WM
enrolled between 10/94 and I I/95 at six clinical consortia in the US provided a unique
opportunity to explore these objectives. All WM received a comprehensive str uctured
interview with detailed questions about sexual behavior with male and female partners and
a question about self-identfication as heterosexual, lesbian/gay or bisexual.This abstract
reflects the initial analysis of data firom I693 [I 364 HIV(+) and 329 HIV(-)] WM enrolled
prior to 5/95
Results: Similar proportions of HIV(+/-) WM identified as bisexual (7%/7%), and heterosexual (88%/83%) while fewer HIV+ WM identified as lesbian (4%/8%). Seventy percent of
HIV'+) and 8 1% of HIV(-) heterosexual WM; 62%/ 82% of (+/-) bisexual WM, and I I%/9%
of (+/-) lesbian WM reported rhaving had sex with a man in the last 6 months and, <1I%/1%
of heterosexual WM, 38%/54% of bisexual WM, and 77%/68% of lesbian WM reported
having had sex with a woman in the last 6 months. In the combined cohorts, 4% of heterosexual, 94% of bisexual and 99% of lesbian WM reported ever having had sex with a
female partner, 25% overall. Self-reported history of most STDs was similar for WM identifying as lesbian in comparison to those identifying as heterosexual.
Conclusions: While significant overlap exists between WM's self identity and their actual
sexual behavior, rigid classification and assumptions regarding the range of sexual behaviors
in these different groups could lead to misclassification of sexual risk particularly among
wornen identifying as lesbian since a large proportion reported having had sex with men.
This would suggest that broadening the conceptualization and targeting of HIV and other
STD education and prevention efforts to all sexually identified groups is critical for reducing
transmission risk.
Lucille F. Stonis, New England Research Institutes, 9 Galen Street, Watertown, MA 02172
Voice: [phone], x407 Fax: [phone] E-mail: lucilles%neri@)mcimail.com
Tu.C.2477
RISK FACTORS ASSOCIATED WITH STD/HIV SEROPOSITIVITY AMONG FEMALE
COMMERCIAL SEX WORKERS (CSWS) - IMPLICATIONS FOR AN INTEGRATED
INTERVENTION STRATEGY
K-G Douglas, *Brathwaite A, *Patterson V, *Wilson I, 'Figueroa P *Ministry of Health,
Jamaica * Association for the Control of Sexually Transmitted Diseases (ACOSTRAD)
Objectives: To determine the prevalence of HIV/STD; to describe the risk factors assocrated with HIV seropositivity among CSWs at initial presentation to a screening service
(Women Health Interventions (WHI),ACOSTRAD proect) specially designed to meet the
health needs of this group of women.
Methods: During the period August 1994 to October 1995, 143 CSWs attended for
screening in Montego Bay. Jamaica. Screening consisted of a detailed sexual history, a full
physical examination, and voluntary laboratory examination for syphilis and HIV. A questionnaire was also administered.
Results: Twenty-five per cent of thfe CSWs were seropositive for HIV infection and 34%
had a positive serological test for syphilis. Significant associations were found between HIV
seropositivity and duration of time in prostitution, 5-7 yrs OR=4.07 C 1= I.35, 12.29; living
alone with no regular partner OR=3.90 CI= I1.36, I.2 I; number of different sex partners
(paying and nonpaying) this last week greater than two OR=2.86 CI 1.25,6.54; number of
paying customers this last week greater than five OR=2.57 Ci I. 10,5.99 p=0.031. Partners'
cocaine use, ever had syphilis", ever had sores in genital and "ever had bruising during
sex" were also significantly correlated with HIV seropositivity No significant associations
were noted with condom use during last sex act, sexual practices, self reported drug use
and past history of other common STDs (gonorrhoea, PID, genital warts and herpes).
Conclusions: The seroprevalernce of HIV among CSWs in Montego Bay is high. Intervention
strategies to reduce the risk of HIV transmission by this population of sex workers mrust
include: (a) programmnes to modify sexual behaviour in order to reduce unprotected sex
wits both steady and paying sex partners, (b) identification and treatment of common STD
conditions.
KG Douglas, c/o Alfred Ninrocks, AIDSCARP 2101 Wilson BIvd.,9700 Arlington,VA, USA,
2220 I;Tel. [phone];
SFax. 703.5169781
Tu.C.2478
BEHAVIOR CHANGES AMONG CRACK-USING RURAL AND URBAN WOMEN
McCoy H.Virginia*, Weatherby NL", Metsch LR, McBride DC *, McCoy CB", Bletzer
KRi, Chitwood DDi-. *Florida International University North Miami, FL; "University of
Miami, Miami, FL; and r*Andrews University, Berrien Springs, MN
Objective: To determine changes over time in high risk sexual and drug-using behaviors
among rural and urban women.
Methods: Rural and urban female drug users and sex partners participated in a prospective
cohort study of HIV prevention interventions which followed gender-specific themes and
focused on urban or rural values.Targeted sampling was used to improve representativeness of the samples. A subsample of 319 females in an urban (Miami, FL) area and an
incomplete subsample of 37 females in a rural area (Collier Co., FL) (to be completed by
May 1996) were analyzed. Cases were selected for analysis which had baseline and followup data and for which participants had used drugs and had sexual intercourse in the previous 30 days. An 80% follow-up rate was achieved for the urban sample; a similar rate is
anticipated for the rural sample.T-tests assessed the change in risk factor statistical means
from baseline to follow-up periods.
Results: Preliminary results showed that there were few differences in demographic characteristics between urban and rural crack users. Roman women were olden had somewhat

347