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

Monday, July 8, 1996

Mo.C.221 - Mo.C.225

(p<.0000 1). In addition to the primary drugs of heroin and cocaine, on' lhi i to over onehalf of the sample also consumed cannabis and alcohol (lower conu,ipon among women,
p<.0000 I) as well as tranquilizers and methadone (higher consuiimtionr mon women,
p<.007, but non-significant among HIV+). Female IDUs were somewhat mr'e + likely to consume drugs daily (8 I% vs. 78%, p=.06) and inject currently (95% vs %, p.O 1), but once
again, not among HIV+ IDUs. Even though on average, women hav. I i.rter drug
use career (with both hard and soft drugs) than men and are I year ynieri at initiation of
IEDU, there is no difference in the duration of IDU (i.e., women pioc ir, to I)U1 faster than
men).Women were also m ore likely to have ever shared needle', vs. p=.004)
which may in part explain their slightly higher seroprevalence (26% 2, p.05). Finally,
women were more likely to have ever been in treatment (86% vs. i I p t-.000, primarily
due to differential participation in methadone regiments), but no dlieri nce itreatment
history exist among HIV+ women and men (917e).
Conclusions: Female Ii DUs in Zurich distinguish themselves from thei, H ial counterparts in
all the domains tested, even though the absolute magnitude of the di ierencesi were for the
most part modest. Of note, many of the gender discrepancies disa pic,,,,whe n one focuses
the analysis on HIV+ IDUs. Nevertheless, information on sociod-cel rUaphics and drug use
will be informative for programs targeting drug users and drug-usin women in particular
(e.g., employment and housing), whereas history of drug use, risk.N be avio, Ind HIV serology
iare also of piarticular epidemiological interest (e.g., increased vulr ealnity ).
Jeri Waing, Institut fOr Sozia-I und Praventivmedizin, Sumatrastr 30, Ch b800 iZur ich,
Switzerland Tel: (4 I-I) 257 66 49; Fax: (41- I) 257 69 62; Email: jwn i spm.r i zh.ch
Mo.C.22I
PREDICTORS OF HIV INFECTION AMONG WOMEN DRUG USERS IN
LOW AND HIGH SEROPREVALENCE CITIES
Tortu, Stephanie*, Beardsley, mM*, Deren, S*, Williams, M**, Stark,. Iale ei-, R****,
Estrada,A*****, foldstein, M*, McCoy, HV******. *NDRI, NY, N'Y; NOVA, Bethesda,
MD; ***Oregon Health Div, Portland, OR; ****NIDA, Bethesda, MD,, - 'niv. of
Arizona,Tucson, AZ; a*****Florida Intl Univ., FL
Introduction: HIV infection in US women continues to increase, e spe ii l2 n Lrgethnic
i inorities.This study describes and compares sociodemographic aiables, behavioral risk
faitors, and health status variables associated with HIV infection in a niulti-site sample of
women injection drug users (IDUs) and crack smokers.Women fro iscities witl low (LO)
seroprevalence were compared to those from high (HI) seroprevalence dres.
Method: As part of a large, multi-site study, data were collected from act ive IDUs and crack
users in 20 cities across the US and Puerto Rico.Women (n-2338) over 18 yr., recruited
using localized targeted sampling plans, were 62.3% Afr-Am; 18.9% White; 12. I( Latina.The
Risk Behavior Assessment questionnaire (RBA) was used to assess soaodemographic characteistics, drugand sex-related risk factors, and sexually transmitted diseases. Current drug
use was verified by urine testing and/or visual inspection of recent track mairks (for IDUs).
Voluntary HIV testing was conducted. In step I, bivariate analyses (IChi sqs./t-tesls) were
used to identify sig. dffs. between seropositive and seronegative women within LO and HI
cities. In step 2, sig. vars. (p<.OS) were entered into logistic regression analyses to predict
HIV serostatus separately for LO and HI areas.
Selected Results: LO cities (n= 1838: 5.2% were HIV+): Sig. predictors of infection (p<.00 1)
were: Af -Am ethnicity (OR- 5.5); ever in drug trmt. (OR- I.7); ever injected (OR= 3.3). HI
cities (n-500; 30.4% were HIV+): Sig. predictors (p.<.02) were: H.S. Education (OR-.6 I);
and ever exchanged sex (OR- 2.0), while the OR for ever injected was 1.5, p<.07. In LO
and HI cities, ever injected & ever exchanged were not highly correlated (i<.07).
Discussion: The differences noted between LO and HI cities underscore the need for
appropriately targetted prevention efforts for women drug users. Special eflirts to reduce
injection-related risks, such as needle exchange, are needed in all cities, and in ffO cities they
may function so conta n the epidemic. More intensive prevention efforts ate needed for
high risk women in all HI cities, where it is especially critical to target those who exchange
sex. Future research should seek to understand in more detail the context of women's risk
and the characteristics associated with specific geographic areas.
Stephanie Tortu, Ph.D. National Development & Research Institutes, Inc. I I Beach Street,
O NewYork, NY 10013 [phone] Ext. 505
S Mo.C.222
> RISK FACTORS FOR HIV SEROCONVERSION AMONG YOUNG WOMEN IN A
S RURAL COMMUNITY IN THE SOUTHEASTERN UNITED STATES
0
u Dominguez, K*, Ellerbrock TV*, Harrington PE*, Bush T*, Malecki '.,Sin onds RJ**.
CO Centers for Disease Control and Prevention, Atlanta, GA*; Palm Beac h Countyv Public
> Health Unit, Florida Department of Health and Rehabilitative Services'
Objective: To identify behavioural and sociodemographic risk factor (or Cincident HIV infections among women in a rural community with high HIV seroprevalence.
Methods: During I 989-9 I1, 52 (5. I%) of 101 I consecutive pregrnant women who registered
S for prenatal care at a public health clinic in Florida tested HIV positive (HIV +). In 1I993-95.,
3 86 (40%) of the 959 HIV negative (HIV) women were randomly selected nterviewed,
and retested for HIV an average of 3.8 years after initially testing HiV.
cv Results: When initially tested fo HIV, 66i of the 386 women v ere -2, Sye rs of age 89%
s-ceie either AficarnAnmeracarnao Hispanic, 69% had completed < IO eas of school, and
0 83%6 had annual household incomes <$10,000. Upon retesting for HIl, all denied ever
S injecting illicit drugs, but 12 (3. 1%) reported a history of ever using crack cI caine. Moreover,
O at ansting, the aeas number of lfetina sex partners was 3.0, yr n rh.'1 1 mn reportI,, ed reyea exchsangfog sex for money ar drugs. 44 (12 %) had had syphaha, onoarahoaa, 01
o- h anaydia (defined as incident S-tO infectior), since initially testing HIi'Whena etested, II: (3811,) of 386 woaren serocaoanierted to HIV during a totsl of 145,a ie or" 'eas of follow.O ap, giving ain idea nse ate of+ 0.76 pea 100 psirson-years.Ter (9 1.8 ofthe I1 HIV+ women
m were Al-i Aimericain, compaa'ed to 62% ofthe HIV womn a'00 ). hato as associated
m_ iab HIV sea ainnveasion inlaided having incident STD infection (rati\,eto Ran b.t;
-( B P0.01I evea- exchanging sex form mney or drugs (BR, 9.1;' P-Oft') evi aug ain rak cocaine
-: (RB. 6.9: P 0.04), rod hiving sex with a mar who used cryck (RR,3 P. P-.02/. 55% of HIV+
mad 38a of HIV- womer used condoms sometimes or yeary often II -. Ii cIuanla the stein
cal ifter iiay testing HIV- (p0.3 7). By multivariate anailysas, orcin Il l Sian was
in deperdeotly associated with incident STD infection (OR, 7.2: Cl 2 -.17
34

Conclusions: Most of these recently infected women acquired HIV through heterosexual
contat. In the U.S., increasing HIV prevalence and widespread crack-related sexual behaviours may place other rural communities at risk for heterosexually transmitted HIV.
Dr: Kenneth L. Dominguez, Mailstop E-45, DHAP NCHSTP Centers for Disease Control and
Prevention, 1600 Clifton Rd, Atlanta, GA 30333, USA.Tel:[phone]; Fax: [phone]:
email: [email protected]
Mo.C.223
USE OF INTRAVAGINAL PREPARATIONS, PRESENCE OF LACTOBACILLUS IN THE
VAGINA,AND RISK FOR HIV IN ZIMBABWEAN WOMEN
van de Wi er ltnneke H.H.M. ~*, Mason PR*, Ray CSC, Gwanzura L*, Maposhere C*,
Katze'stein Di, Padian NS~*. *Zimbabwe AIDS Prevention Project-University of
Zimblabwe; ~niversity of California at San Francisco; t Stanford University.
Objectives: To study use of intravaginal preparations, and its association with gynecological
problems, presence of lactobacillus in the vaginal flora, and risk for HIV in commercial sex
woer as (C SWs) in Hanare, Zimbabwe.
Methods: 149 CSWs were enrolled and followed up biweekly for 12 weeks.The following
proc dures were carried out at each visit: interview about demographics and sexual behavior,
revieinw of a selfadministered coital log, supply of female and male condoms, and treatment of
S-I's when applicable. A physical exam and microbiologic sampling was carried out at enrollment 6 and 12 weeks, and a blood draw and HIV counseling at enrollment and 12 weeks.
Results: 69. 1% of the women reported ever having inserted preparations r inside the vagina
wher-as 30.2% of the women practiced insertion currently. Herbal preparations were most
frequently mentioned (20.8%) but douching with water and soap (I18.8%), wiping inside the
vaginai with newspaper, cloth or tissues (I 6.8%), douching with dettol or betadine (14.8%),
and inserting cotton wool (I 4. 1%) were also frequently mentioned. Most women who currentl insert preparations do so to dry vaginal secretions (68.9%) or to constrict the vagina
(I 5.6o). Single women with a regular boyfriend were more likely to prepare for sex than
divored women (p = 0.0087) or single women without a boyfriend (p = 0.01I58).There
was strong association between inserting preparations (ever and current) and reported
abno-mal vaginal discharge in the last 3 months (p - 0.0069 and p = 0.0290), and reported
deep pain during sex in the last 3 months (p = 0.0673 and p = 0.0363).The prevalence of
lactobacillus was 25.5%.Women who douche with dettol or betadine (OR = 0.44; p =
0.3 I 4) or insert herbs (OR -= 0.52; p -= 0.3108) seem to be somewhat less likely to have
lactobacillus in their vagina.We also found a strong association between the absence of lactobaaillus in the vagina and being HIV positive (OR = 3.70; p = 0.0090) in this cohort.
Con:clusions: The use of intravaginal preparations is widespread in Zimbabwe. Our data
suggest that the use of such preparations might cause or are prompted by gynecological
prob ers.The use of certain intravaginal preparations, but not all of them, was associated
with the absence of lactobacillus, and the absence of lactobacillus was associated with being
HIV lpositive.We are currently investigating these relationships in more detail in a prospective cohort study of "low-risk" women.
Jannr ke vc n de Wijgert, ZAPP Savvas flats ground floor 14 Baker Avenue, Harare,
Zimbabwe. Phone: [phone] / 739406; Fax: [phone]; E-mail: [email protected]
MoC.224
PREVALENCE OF DOMESTIC VIOLENCE AND CHILDHOOD ABUSE AMONG
WO'1EN WITH HIV AND HIGH RISK UNINFECTED WOMEN
Dearnant C, Cohen M, Barkan S, Richardson J, FitzGerald G,Young M, Holman SAnastos K,
Cohen J. Melnick S for the WIHS Collaborative Study Group
Objective: to determine the prevalence of domestic violence and childhood abuse reported b/women with HIV and women at risk for HIV.
Methods: The Women's Interagency HIV Study (WIHS) is a multi-site cohort study of HIV
infec.ed and a comparable at risk uninfected control group.This abstract reflects the initial
analysis of available data from 1103 (875 HIV + and 228 HIV-) women enrolled prior to
5/ I 5 All reported values were statistically significant with p-values <.00 I.
Results: Sixty seven percent of the women reported sexual or physical violence by a current or past partner and in 31% of these women this occurred within the past one year
Domestic violence was not associated with ethnicity educational level, marital status or HIV
sero',tatus. Higher rates of domestic violence, including abuse within the past year, occurred
in women with: any lifetime drug use (7 I% vs. 4 I1%), IV drug use in past 6 months (83% vs.
65%, more than 10 lifetime sex partners (75% vs. 57%), sex partner at risk for HIV (71%
vs. 5 7%) and exchange of sex for money/drugs/shelter (8 1% vs. 58%).Thirty three percent
had been victims of childhood sexual abuse and I 5% reported childhood physical violence.
Childhood abuse was associated with: IV drug use (50% vs. 35%), exchange of sex for
drug,/money/shelter (56% vs. 29%), adult domestic violence (52% vs. 27%) adult sexual
abuse (35% vs. 22%).
Con:clusions: There is a high prevalence of domestic violence and childhood abuse among
women with HIV and at risk for HIV A history of childhood abuse may identify women at
increased risk for sexual and physical violence and behaviours associated with increased risk
for HIV infection. Effective HIV risk reduction and HIV care should also include domestic
violence identification and intervention.
Cather ine Deaant, M.D., I835 West Harrison, Chicago, II 60612 USATeI: (3 12) 633-5080:
h-ax' 1312) o338333: E-mail: N/A
Mo.C.225
HIV RISK BEHAVIORS AND HIV-RELATED MORBIDITY IN A COHORT OF IN
CARCERATED WOMEN IN MASSACHUSETTS
De ivroat, Arane S.1,2, Zuerle, SI, Stevens, j I, Dean, DI, Ferdinand MM I, jnsdale, BM I
fGeoge. JA, Scheub BC2. I Broorn University School of Medicine, Providence RI, and 2
HIV/AIDS Unit, Lemuel Shattuck Hospil, Boston, MA, USA
Objective: Duss to the nature of the cm-ames for which women are incacerated, the snropres en e If IV us alarmingly hugh is wormen's pmisoos.Ws assessed HIV risk behavions and
h~iV-melated Imnraidity us a cohort of 291I HIV seropostuve (HIV+) women inmates (WI) followsd 1o-ea 5~5 year s at the HIV cin aic of a lange state pison foa moines us Massachusetts.
Methods: Wa' examined date of HIV diagnosis, race, ethnicity, and T cell count decline for a
cobhact at 231 HIV+ WI. In addition, 160 W (I I I HIV+ WI and 49 HIV senonegate WI)