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

192 Abstracts WeOrE1281-WeOrE1320

XIV International AIDS Conference

and/or has been completed in eight countries. An analysis of the prevalence of of
physical and sexual violence has been conducted in Brazil, Peru and Thailand.
An analysis of the intersections between different forms of violence and abuse is
ongoing.
Results Lifetime and current prevalence of partner physical and sexual violence
(preliminary results) are listed in the table (page 191).
Conclusion: The findings provide minimum estimates of the prevalence of physical and sexual VAW. They highlight the extent to which physical and sexual violence within intimate partnerships is widespread and threaten women's ability to
negotiate condom use. Within HIV prevention it is critical that interventions openly
discuss and challenge VAW, and work with the VAW movement to challenge the
gender-inequalities and social expectations that fuel both HIV and VAW.
Presenting author: Henrica A.EM. Jansen, WHO, GWH, 1211 Geneva 27,
Switzerland, Tel.: +41- 22 - 791 3106, Fax: +41 22 - 791 4382, E-mail: jansenh
@who.ch
WeOrE1281 Istimela: gang rape in South Africa
K. Wood, N. Jama, R. Jewkes, M. Nduna, J. Levin. Medical Research Council,
Medical Research Council, Private Bag X385, Pretoria 0001, South Africa
Introduction: Gang rape is a form of sexual violence which carries substantial
risk of HIV transmission as sex is unprotected, multiple men are involved and
women are often injured. One third of rapes reported to the police in South Africa
involve more than one man. This paper describes research findings relating to
this phenomenon in two provinces of South Africa.
Methods: Ethnographic data was collected during long-term fieldwork with township youth in the Eastern Cape, and during facilitation of Stepping Stones HIV prevention workshops with youth in the North-West province. Quantitative data was
collected using an interviewer-administered questionnaire (sample: 122 men).
Results: Gang rape, often referred to as istimela (train) or streamline, is most
commonly done by groups of friends. While sometimes practised opportunistically
against a randomly chosen woman, it is often committed as a way of 'disciplining'
or humiliating a young woman known to them, often an (ex-)/girlfriend of one of
the group, for perceived transgressions (such as sexual infidelities). Young men
who participate often do not view their actions as part of the crime of 'rape' because they argue that they have sexual access to these women, that a woman's
silei ene entails 'consent', that it often involves trickery and subtle coercion rather
than outright violence, and because drunk women (who are often targetted) are
thought to be fair game. In the survey sample 15.6% (19) of the men reported
having participated in an istimela. Logistic discriminant analysis showed that the
only important factor associated with it was alcohol consumption: the lower the
alcohol consumption, the less the chance of having been involved (OR=0.56,
p=0.012). Reporting was not associated with age, education or socio-economic
status.
Conclusions: Gang rape is a relatively common form of sexual violence entailing
high HIV risk which needs to be addressed in prevention programmes.
Presenting author: Katharine Wood, Medical Research Council, Private Bag
X385, Pretoria 0001, South Africa, Tel.: [phone], Fax: [phone],
E-mail: woodkm @yahoo.com
WeOrE12821 Sexual violence, pressure and HIV in rural
Mwanza, Tanzania
M.L. Plummer', D. Wight2, J. Wamoyi1, G. Mshana1, D. Ross', Z. Salamba1.
INIMR/AMREF/LSHTM, PO Box 11936, Mwanza, Tanzania; 2MRC-UK,
Glasgow, United Kingdom
Background: There is a need to collect accurate data on unwanted sex in a rural
African context, to better understand how it may contribute to HIV transmission.
Methods: The paper draws on data from: a self-completion questionnaire with
6,077 primary school adolescents; 72 in-depth interviews with HIV+, pregnant
or randomly selected primary school adolescents; 6 series of 3-4 focus group
discussions (FGD) with same-sex groups in their late teens or early 20s; and participant observation (P0) carried out by young researchers in 7 villages (total 110
person-weeks).
Results: FGD and P0 were the most useful methods, identifying 3 categories. (a)
Publicly acknowledged 'rape' is limited to violent sexual assaults which become
subject to quasi-judicial responses such as a fine or beating, because the victim
is not believed to have encouraged it. (b) More prevalent appears to be sexual
violence that is not publicly defined as "rape". If a woman took a gift from a man
prior to an assault, she is likely to be blamed for it. Also, she may not report it if an
authority or a friend acted as an intermediary facilitating the assault. (c) In some
instances, violence does not occur but the woman only has sex due to pressure
from relatives, the threat of violence, or the threat of being shamed because of
previously having accepted a gift from the man. Qualitative researchers estimate
that a quarter of young women have been pressured into sex after accepting a
gift, and half have been pressured into it by intermediaries. Any statistical link

between unwanted sex and HIV may relate to a lack of vaginal lubrication caused
by female arousal, and/or the degree of violence, as these are likely to contribute
to abrasions. Condom use is rare even in consensual sex, so a lack of condom
use per se is not likely to contribute to higher HIV transmission.
Conclusions: Force is used in some sexual encounters, and non-violent pressure is even commoner. Both may contribute to HIV transmission.

Presenting author: Mary Plummer, PO Box 11936, Mwanza, Tanzania, Tel.:
[phone], Fax: [phone], E-mail: [email protected]
WeOrE12831 Relationship between violences experienced by
women and the heterosexual transmission of
HIV-AIDS
D. Damant1, L. Binet1, G. Trottier', J. Lindsay', L. Noel2. 1Universit6 Laval,
Local 0439, Universite Laval, Cite Universitaire, Sainte-Foy, G K 7P4, Canada;
2Centre de sant6 publique, Quebec, Canada
This presentation gives the results of a research on the relation between violence
experienced by women and the heterosexual transmission of HIV-AIDS. Violence
is treated as a risk factor in social and individual vulnerability that interferes in
heterosexual sexual relations and affects the course of life in general. Violence
is also considered as a health determinant. Biographical interviews were held
with 24 women belonging to groups at risk to contract HIV. The study takes into
account the context in which these women live and their social position.
Following a brief presentation of the study group and a summary of the methodology, the results will be given. They differentiate the forms of violence experienced,
processes of accumulated violent experiences and the role of violence in relation
to other deliberate acts within the framework of sexual relations, acts that vary
with the type of sex partner. The results also cover various means of negotiating
a safe sexual relation, on the production of the feeling of safety despite the nonuse of a condom and on the sense given to an unprotected relation. Finally, the
results identify instances when protection is impossible. The conclusion outlines
ways of teaching HIV prevention in a social milieu associated with marginality.
Presenting author: Dominique Damant, Local 0439, Universite Laval, Cite Universitaire, Sainte-Foy, G1K 7P4, Canada, Tel.: [phone], Fax: [phone]
3309, E-mail: [email protected]
WeOrE1284 Marital violence in India: women's heightened
vulnerability to HIV/AIDS
V.F. Go1, C.J. Sethulakshmi2, M.E. Bentley3, S. Sivaram', A.K. Srikrishnan2,
D.D. Celentanol, S. Solomon2. 1Johns Hopkins University Bloomberg School of
Public Health, Johns Hopkins Bloomberg School of Public Health, 615 North
Wolfe Street, E6007, Baltimore, Maryland 21205, United States, United States;
2 YRG CARE, Chennai,/ndia; 3 University of North Carolina, ChapelHill, United
States
Background: We examined how marital gender norms and power dynamics affect women's vulnerability to HIV/AIDS in Chennai slum communities.
Methods: As part of a multi-site international behavioral HIV intervention trial,
in-depth interviews with men and women (20 each) and focus groups of men and
women (7 groups each) were conducted in 2 randomly selected slums in Chennai. Interviews were taped, transcribed, translated, coded and analyzed to identify
recurrent themes.
Results: Mean age was 28.0; most (98%) women were married. Participants
noted that husbands hold decision-making power in economic, social and sexual
spheres. Clear patterns of violence were present; respondents reported that husbands regularly beat wives in most marriages. Women described being slapped,
kicked, having their head hit against the floor and being burned with lit cigarettes;
some were struck with objects such as ladles or stones. Slapping or hitting the
face was the most frequent type of violence and physical sequelae such as recurring headaches and blurred vision were noted. Disobeying husband or elders,
neglecting household chores, refusing sex and suspected infidelity often triggered
violence. Drinking often preceded explosive outbreaks of violence. Some women
modified their behavior to avoid physical violence by engaging in silent passivity
during verbal arguments and acquiescing to unwanted sex. Although most respondents felt wife beating was the norm, the acceptable intensity of violence
varied by gender.
Conclusions: In Chennai slum communities, gender norms sanction often violent expressions of husbands' locus of control in the household. As women try to
minimize exposure to violence, their ability to insist on monogamy, negotiate condom use or refuse sex is limited. HIV interventions should incorporate the social
context in which they operate. As in other sites in this trial, prevention messages
targeting men rather than women may effectively reduce women's risk of HIV.
Presenting author: Vivian Go, Johns Hopkins Bloomberg School of Public
Health, 615 North Wolfe Street, E6007, Baltimore, Maryland 21205, United
States, United States, Tel.: [phone], Fax: [phone], E-mail: [email protected]
WeOrE1320 The need for harm reduction in U.S. prisons: an
activist's view
A. Reynolds. H/V and HCV in Prison Committee, 187 Golden Gate Ave., San

Francisco, CA, 94102, United States
Issues: The United States has the world's second largest prison system with over
2 million men and women incarcerated and an additional 4.3 million people in jail
or on probation or parole. Although HIV and HCV rates are 8 to 10 times higher in
prison than in the general public, there has been no effort to implement harm reduction programs into the U.S. prison system. Condom distribution occurs in only
2 state and 5 jail systems. There are no bleach distribution or syringe exchange