Abstract Book Vol. 1 [International Conference on AIDS (16th: 2006: Toronto, Canada)]

MOACO201 MOACO203
Assessing a national HIV behavior change campaign An evaluation of a quasi-experimental community
focusing on multiple concurrent partnerships in level HIV intervention among out-of-school youth in
Swaziland 26 communities in Nigeria

D. Halperin', N. Andersson2, M. Mavuso3, B. George4. IUSAID, Southern
Africa HIV-AIDS Program, Mbabane, Swaziland, 2Centro de Investigaci6n de
Enfermedades Tropicales (CIET), Acapulco, Mexico, 3United Nations Family
Planning Association, Mbabane, Swaziland, 4Centers for Disease Control
(CDC), Monitoring and Evaluation, Pretoria, South Africa
Background: Swaziland is currently experiencing the world's most severe HIV
epidemic, at over 40% in pregnant women (2004 survey). Researchers, planners
and donors increasingly consider the pervasive pattern of concurrent multiple
partnerships to be a key epidemiological driver in the spread of HIV, especially
in such high prevalence, generalized epidemics (Halperin and Epstein, Lancet
2004). A government national multi-media campaign proposed to generate
increased awareness on this issue in the hope of eventually reducing HIV risktaking behavior and associated social norms.
Methods: A baseline survey was conducted to explore sexual attitudes and
social norms of behavior among over 2000 adults in 12 randomly selected
communities (4 urban and 8 rural), just prior to the launch in June 2005 of
the government (NERCHA)-led behavior change communication campaign. A
follow-up survey of over 2000 adults in the same communities in June 2006
measured the potential contribution of and processes involved in the campaign
(recognizing that other significant behavior change efforts may also have
impacted on behavioral patterns and norms). In addition, three rounds of
intensive focus group discussions were held in each of the 12 communities (one
for each gender, or 72 groups in total).
Results: The large majority of respondents in the baseline survey and the
first two focus group rounds reported that "lishendes", or multiple concurrent
partners, were "very common" or "common" in their communities. Over
the course of the study, a substantial number of people reported that such
behaviors were detrimental for the HIV-AIDS crisis, and considerable numbers
of them, especially females, felt that "choosing to have only one partner" could/
should be an important prevention approach for the country's fight against the
deadly epidemic.
Conclusions: The study ended up helping guide the development of the
government BCC campaign, and could also help inform similar such behavior
change interventions carried out elsewhere in the region.
MOACO202
Exposure to a community-mobilization intervention
and HIV-related knowledge, attitudes, and
practices, Botswana, 2003
T. Koppenhaver', D. Fleming2, B. Meyerson3, A. Robbins3, G.S. Kebonang4,
T.H. Roels', P.H. Kilmarxs. 1BOTUSA Project, Global AIDS Program, Gaborone,
Botswana, 2BOTUSA Project, Gaborone, Botswana, 3National Alliance of
State and Territorial AIDS Directors (NASTAD), Washington, United States,
4Humana People to People / Total Community Mobilization, Gaborone,
Botswana, 5Centers for Disease Control and Prevention (CDC), Atlanta, United
States
Background: The Total Community Mobilization (TCM) program is a doorto-door intervention which delivers HIV prevention and treatment-related
information to households in Botswana. We examine the association of exposure
to TCM with selected indicators of HIV/AIDS-related knowledge, attitudes, and
practices (KAP).
Methods: Our data come from a cross-sectional household survey, which
interviewed 807 15-to 49-year-old Batswana in seven of Botswana's 22 health
districts in 2003. At this time, TCM was operating in five of these seven health
districts. We used bivariate and multivariate analyses to compare KAP outcomes
among those individuals with reported TCM exposure (defined as having ever
spoken with someone from TCM, either in a group or individually) to those
without.
Results: In the five TCM-active health districts, 23% of respondents reported
exposure to TCM. Controlling for demographics and regularly listening to an
HIV/AIDS education-entertainment radio drama, TCM exposure was significantly
(p<.05) associated with spontaneously mentioning abstinence (adjusted odds
ratio (AOR) 1.99) and condom use (AOR 2.96) as HIV prevention methods. TCM
exposure was similarly associated with knowing that it's possible to prevent
mother-to-child transmission (MTCT) and that AZT is a means of preventing
MTCT, but not with spontaneously mentioning the three modes of transmission
(pregnancy, delivery, and breastfeeding). On four of six separate questions
about stigma, TCM exposure was significantly associated with non-stigmatizing
responses (AORs from 1.44 to 9.12). Further, TCM-exposed individuals were
twice as likely as the unexposed to report having been tested for HIV (AOR
1.96, p<.01). However, TCM-exposed individuals were no more likely to report
abstaining, not having concurrent sex partners, or using condoms with noncohabiting partners.
Conclusions: These findings suggest that TCM has helped to improve HIVprevention knowledge, reduce stigma, and encourage testing in Botswana.
The lack of association between TCM exposure and targeted sexual behaviors
indicates that important behavior-change challenges remain.

J. Anyanti, G. Omoregie, A. Ankomah, S. Adebayo, A. Buba-Vaganda. Society
for Family Health, Research and Evaluation, Abuja, Nigeria
Background: In developing countries, a major issue with HIV interventions
is the failure to attribute programme success to programme interventions.
Promoting Sexual and Reproductive Health and HIV/AIDS (PSRHH) is a seven
year DFID and USAID funded programme, which was designed as an 18 -month quasi experimental design to test whether young persons (15-24) at
intervention sites were more likely to use condoms compared with the control
sites. This paper presents the key findings from the study.
Methods: Thirteen most at risk intervention sites spread across Nigeria's six
health zones were purposively selected along with 13 matched controls. Data
were collected from 1741 respondents prior to implementation and from 1713
respondents 18 months later based on multistage probability sampling design.
The main intervention was the Peer Education Plus (PEP) model consisting
of innovative peer education, edutainment and special interventions among
'influencers' of young persons. Multivariate statistical methods were used
to investigate whether changes in condom use will be significantly higher at
intervention than control sites even after controlling for selected population
characteristics.
Results: While there was no significant change at the control sites, (from 58%
in 2002 to 55% in 2004, p > 0.05), at the intervention sites the proportion
of females that used condoms in last non-marital sex increased significantly
from 54% to 69% (p< 0.05). For males too, the programme was effective in
significantly increasing the proportion of young males who used condoms in
last risky sex from 64% to 75% (p< 0.05). The change in control communities
was not significant.
Conclusions: For both males and females, the PEP interventions were effective
in increasing condom use at the intervention but not the control sites. The
model is therefore recommended as an effective tool and can be scaled up at
similar sites throughout Nigeria and elsewhere.
MOACO204
Changes in sexual behavior and risk of HIV
transmission after two years of antiretroviral
therapy and prevention interventions in rural
Uganda
R. Bunnell', N. Wamai2, J.P. Ekwaru1, D. Moore2, W. Were2, S. Bechangel,
A. Coutinho3, E. Madraa4, J. Mermin'. 'CDC-Uganda, Global AIDS Program,
CDC/HHS, GAP-Uganda, Entebbe, Uganda, 2CDC-Uganda, HBAC, Tororo,
Uganda, 3The AIDS Support Organisation (TASO), Kampala, Uganda, 4Ministry
of Health, AIDS Control Programme, Kampala, Uganda
Background: The long-term impact of antiretroviral therapy (ART) on sexual
HIV transmission risk among HIV-infected persons in Africa is unknown. We
assessed changes in sexual behavior and estimated HIV transmission from HIVinfected adults after 2 years of ART in Tororo, Uganda.
Methods: Between May 2003 and December 2004 we enrolled and followed
926 drug-naive HIV-infected adults in a home-based AIDS care program that
included HIV prevention counseling, voluntary counseling and testing (VCT)
for cohabitating partners and condom provision. At baseline and follow-up,
we assessed participants' HIV plasma viral load and partner-specific sexual
behaviors. We defined risky sex as intercourse with inconsistent or no condom
use with partners of HIV-negative or unknown serostatus in the previous 3
months. We compared risky sex rates using a Poisson regression model and
estimated transmission risk per partner based on established viral load-specific
transmission rates. Sero-conversion rates among cohabitating sero-discordant
partners with laboratory results were assessed at 24 months.
Results: By January 2006, 454 (49%) participants had reached 24 months of
follow-up. Sexual activity increased from 28% at baseline to 38% (p<0.001)
(women: 21%-31%; men: 47-61%) at 24 months. The proportion of sexually
active participants reporting any risky sex decreased from 22% to 15%
(p=.062). Median viral load among those reporting risky sex was 122,500
copies/ml at baseline and undetectable at follow-up. At 2 years, the reduction
in estimated risk of HIV transmission from cohort members was 98%, from
45.7 to 1.0 per 1000 person years. One sero-conversion occurred within 62
cohabitating sero-discordant partners during 1st year of therapy; no seroconversions occurred in the second year.
Conclusions: Providing ART and on-going prevention interventions was
associated with a reduced estimated risk of HIV transmission among HIVinfected Ugandan adults after two years of therapy, despite an increase in
overall sexual activity. Integrated ART and prevention programs may reduce
HIV transmission in Africa.

XVI INTERNATIONAL AIDS CONFERENCE * 13-18 AUGUST 2006 * TORONTO CANADA * ABSTRACT BOOK VOLUME 1