Abstract Book Vol. 2 [International Conference on AIDS (14th: 2002: Barcelona, Spain)]

178 Abstracts WePeD6377-WePeD6380

XIV International AIDS Conference

ical status of their sexual partners. Patients reporting SBR were compared with
others by using survival analysis and Cox models.
Results: An increase in SBR with casual partners was observed from 1999 (6.4%
in 1999 to 19.4% after 1999) after a decrease between 1997 and 1999. Thirty-one
patients reported 42 SBR. The median time from enrolment to the first SBR was
6.6 months (range 3.0 - 44.1 months). Reporting an SBR was more frequent in
patients having casual partners, being under 33 years, having had no or a short
symptomatic primary HIV-1 infection. Plasma viral load was above the threshold
of 200/500 copies in 43% of cases of SBR (95% of these cases were treated with
HAART at the time of SBR). However patients reporting an SBR did not differ
from the others according to the viral load (measured at the visit preceding the
SBR), suggesting that having a good (or a bad) response to HAART was not a
major determinant for subsequent SBR.
Conclusion: Even if no relation was found between being under the viral load
threshold and reporting an SBR at the next visit, the fear about a re-emerging
HIV epidemic is still of concern. The young and those who have sex with casual
partners remain a group with a higher risk for HIV-1 transmission.
Presenting author: Loic Desquilbet, Inserm u.292, 82 rue du general leclerc, 94
276 le kremlin bicetre cedex, France, Tel.: [phone], Fax: [phone]
20 75, E-mail: [email protected]
WePeD6377 Who's really at risk for HIV: Exploring risk
perception among adolescents in South Africa
N. Rutenberg1, L. Brown2, K. Macintyre2, A. Karim2. 1 Population
Council/Horizons, Population Council, 4301 Connecticut Avenue NW Suite 280,
Washington, DC, United States; 2Tulane University, New Orleans, United States
Background: Perceived risk of HIV infection is an important antecedent for adopting protective behavior. Yet measures of risk perception, particularly among adolescents, are often simplistic and do not differentiate between "accurate" and "inaccurate" perceptions. In our longitudinal study of adolescents in Kwa-Zulu Natal,
South Africa, we explore HIV risk perception including the interplay between reports of risk perception, the reasons underlying that perception, HIV knowledge,
and sexual risk behavior.
Methods: Structured interviews about schooling, work, HIV, sexual and reproductive behavior were conducted with 3052 adolescents between the ages of 14
and 22 in selected households which contained one or more adolescents in a
multi-stage cluster sample of enumeration areas in Durban Metro and Mtunzini
districts.
Results: 76% of adolescents report being at no risk for HIV infection in the next
12 months, 13% at low risk and 11% at moderate/high risk. The most common
reasons for perceiving that one was not at risk of HIV were consistent condom
use (46%), a monogamous relationship or partner (33%), and abstinence (16%).
However, among those who felt they had no HIV risk and had ever had sex (42%
of no risk group), 42% reported unprotected sex and 22% multiple partnerships in
past year. In contrast, those who felt they were at moderate/high risk cited unprotected sex (36%), multiple partners (29%), and accidents (19%) as putting them
at risk. 55% of this group reported unprotected sex and 24% multiple partnerships
in the past year.
Conclusions: While risk perception, the calculation of that risk, and behavior
are consistent for some adolescents, a significant proportion of adolescentsparticularly among those that consider themselves at low risk of HIV infection
in the next 12 months-have inaccurate perceptions of their HIV risk. This may
be due to lack of information or denial. Programs should aim to help adolescents
accurately assess their risk.
Presenting author: Naomi Rutenberg, Population Council, 4301 Connecticut Avenue NW Suite 280, Washington, DC, United States, Tel.: [phone], Fax:
[phone], E-mail: [email protected]
WePeD6378I HIV/AIDS behavioral study in Niger
C. Chappell1, D. Didier-Blanchard', R. Borwankar, L. Thomas1, H. Hanafi1,
M. Sani2, C. Vondrasek3. 'AED, AED, 1875 Connecticut Ave Suite 900,
Washington DC, 20009, United States; 2CARE-Niger, Niamey, Niger;
3 JHU/CCP/SFPS, Abidjan, Cote d'lvoire
Background: This study of Niger's urban populations examined demographics,
media habits, attitudes and sexual behaviors concerning HIV/AIDS. While data
indicate a <5% general HIV prevalence rate, the true level of infection may be
higher, given the limited health surveillance systems and nomadic lifestyles. This
study examines Niger's HIV/AIDS-related behaviors and the need for health promotion interventions.
Methods: The survey was fielded by CARE-Niger in May 2001 with direction
from AED on behalf of JHU/CCP under the USAID-funded Family Health and
AIDS project. A sample of 1200 urban, francophone respondents, aged 15 - 49,
was interviewed using a household level sampling strategy. Respondents were
proportionately sampled by location, age and gender. Descriptive, bivariate and

multivariate data analyses were run on SPSS software.
Results: The sample indicates high media access through established channels,
placing a priority on media as a source of HIV/AIDS information. Infection via
sharing personal objects is the most commonly identified transmission route. Little association is made between sexual behavior and risk of HIV infection. Men
are less likely than women to cite that condoms can prevent infection and that unprotected sex is a risk factor. 97% know what a condom is. Only 31% have ever

used a condom. Among the sexually active respondents, 51% have ever used a
condom, and 57% of that subgroup used a condom during their last sexual liaison.
60% of the total sample report ever being sexually active, most of whom claim to
be in monogamous relationships and hence have a low HIV-risk self-perception.
Conclusions: Given media access and acceptance, targeted campaigns could
be effective in promoting desired positive health behaviors. Campaigns should
use local opinion leaders to address misconceptions about HIV transmission, include risk-focused education, promote positive behaviors, and stress the importance of voluntary testing.
Presenting author: Christina Chappell, AED, 1875 Connecticut Ave Suite 900,
Washington DC, 20009, United States, Tel.: [phone], Fax: [phone],
E-mail: [email protected]
WePeD6379 Determinants of HIV status for populations
accessing California HIV counseling and testing:
Similar behaviors equal different risks
D.L. Sykes, C.M. Heusner, S.R. Truax, H. Rasmussen. California Office of AIDS,
Office of AIDS, PO. Box 942732, Sacramento, CA 94234, United States
Background: The HIV Counseling Information System tracks risk profile, HIV status, and demographic information of clients accessing California HIV counseling
and testing (HIV C&T) at publicly funded sites. From these data, this study delineates variations in patterns of risk behavior that are predictive of HIV infection
among specific gender/race/sexual orientation groups.
Methods: Data from 533,000 HIV C&T client visits (1998 - 2000) were included
in the analyses. Repeat HIV positives were excluded. Client visits were assigned
to one of 18 mutually exclusive groups defined by gender, sexual orientation, and
race. For each group, significant risk variables from univariate logistic regression
analyses were included in multivariate logistic regression analyses using HIV status as the outcome variable.
Results: Several distinct patterns of risk emerged. For females and heterosexual
males, reporting an HIV positive sex partner was by far the most predictive of
HIV infection (ORs ranged between 3.38 - 24.90). However, for men who have
sex with men, anal receptive sex was the most predictive of HIV infection (ORs
1.70 - 2.66). Other risk factors were significantly related to HIV infection only
among specific subgroups: Stimulant use was predictive among African American females and heterosexual males (OR 1.69 and 1.48, respectively); Injection
drug use for African American and White females and heterosexual males (ORs
1.59 - 2.37); and STD for Latino and White males (ORs 1.15 - 1.55). Latina and
Asian females' only significant predictive risk for HIV infection was reporting an
HIV positive sex partner (OR 21.08 and 24.90, respectively).
Conclusions: These data provide evidence that equivalent behaviors are not
equally risky among different subgroups. Delineating specific behavior patterns
associated with
HIV infection for specific groups may provide useful information for both targeting
populations at risk and for designing prevention programs.
Presenting author: Deanna Sykes, Office of AIDS, P.O. Box 942732, Sacramento, CA 94234, United States, Tel.: [phone], Fax: [phone], E-mail: [email protected]
WePeD6380 How do adverse childhood experiences and
acculturation influence HIV risks among Latinas?
J.V. Carmonal, G.E. Wyatt', G.J. Romero2, M.D. Newcomb3, M.B. Tucker1,
H.A. Wayment4. 'University of California Los Angeles, UCLA, NPI, Dept. of
Biobehavioral Sciences, 760 Westwood Plaza, #C9-539, Los Angeles, CA
90024-1759, United States; 2California State Assembly, Sacramento, United
States; 3University of Southern California, Los Angeles, United States;
4 University of Northern Arizona, Flagstaff, United States
Background: Latino women are among the fastest growing subgroup at risk for
HIV infection. These trends emphasize the need to better understand the multiple
factors that influence Latinas' sexual practices. This study examined early experiences, acculturation and other demographic variables, later adult victimization,
and communication factors that may place these women at higher risk for HIV
infection.
Method: A random community sample of Latinas, 18 to 50 years of age were
interviewed face-to-face by bilingual, bicultural interviewers.
Results: Structural equation modeling procedures indicated that women who
reported more unintended pregnancies were older, younger at first intercourse,
and experienced child abuse. Risky sexual behavior in primary relationships was
more likely among women who were younger, married or cohabiting, and women
who were more able to communicate about sex. Younger women and women
who were more able to communicate about sex were more likely to use a barrier
method of contraception. Women able to communicate about sex were more acculturated and married or cohabiting. Finally, women who reported adult abuse
experienced earlier child abuse and reported less religiosity.

Conclusions: It seems clear from these findings that older women are at risk for
HIV infection and unintended pregnancy and that special efforts must be made
to reach older Latinas with prevention messages and realistic prevention strategies. Further, although gender roles are evolving in all cultures within the U.S.,
these kinds of findings make it clear that prevention efforts must also be directed
at men, as the burden of changing attitudes about condom use and other contraceptive aims cannot be the sole responsibility of women. It is critical for HIV