Programme Supplement [International Conference on AIDS (16th: 2006: Toronto, Canada)]

Late Breaker Abstracts

lowest SLI quartile (44.6%) and 3.61% HIV versus 1.08% in the
rest; HIV rates were higher in the former even for the same half of
SLI (lower 4.03%, upper 2.38%) than in the latter (lower 1.09%,
upper 1.06%). Sentinel surveillance method (2005 HIV rates:
antenatal clinic 3%, STI clinic 22.8%) gave 112,635 adults with
HIV in Guntur district, 2.5 times (sensitivity analysis 2.0-3.2) the
45,925 population-based estimate adjusted for under-represented
high-risk groups. A correction factor from this comparison dropped
the sentinel surveillance HIV estimate of 1.44 million adults in
Andhra Pradesh to 0.51 (0.40-0.64) million.
Conclusions: The estimation method used in India grossly
overestimates HIV in south Indian states, which have a similar
pattern of preferential use of public health system by lower
socioeconomic strata and referral of HIV patients to the public
system. The current HIV estimate of 5.2 million for India needs
substantial downward revision.
THLB0108
HIV prevalence and incidence are no longer falling in
Uganda - a case for renewed prevention efforts: evidence
from a rural population cohort 1989-2005, and from ANC
surveillance
L.A. Shafer', S. Biraro', A. Kamali', H. Grosskurth', W. Kirungi2,
E. Madraa2, A. Opio3. 'MRC/UVRI Uganda Research Unit on AIDS,
Entebbe, Uganda, 2National AIDS Control Programme, Ministry
of Health, Uganda, Kampala, Uganda, 3National AIDS Control
Programme, Ministry of Health, Uganda, Entebbe, Uganda
Background: Throughout the 1990s, Uganda has successfully
controlled its HIV epidemic, with falling prevalence and incidence
rates. Recent evidence, however, indicates that this decline may
not be continuing. Here we report recent epidemiologic trends.
Methods: Two sources of data were used. A rural cohort,
established in 1989, provided HIV prevalence, incidence and"
sexual behaviour data. ANC surveillance provided prevalence
among pregnant women.
Results: In the rural cohort, prevalence reached a low of 5.6
(95%C.I. 4.9-6.3) among males and 6.9 (95%C.I. 6.2-7.7) among
females in 2000. Since then, prevalence has increased to 6.5% in
men and 8.8% in women in 2004, and the trend appears to have
continued in 2005. Incidence estimates follow similar trends (fig).
HIV incidence in Rural Population Cohort. Masaka
O 47
1990 1992 1994 199 1998 2000 2002 2004
[incidence3]
Though the recent upward trend of HIV prevalence and incidence
in this population is not yet statistically significant, similar results
were found in some ANC surveillance sites: HIV prevalence declined
from 1992-2000. Since 2002, the declining trend continued in
7 of 24 sites, but in 10 sites there appears to be a rise in HIV
prevalence. Analysis of behavioural data from various sources
in Uganda provides some evidence that during the period of the
change in the course of the epidemic, there was a corresponding
change in risky behaviour (see abstract Biraro et al).
Conclusions: Uganda has been highly successful in controlling
its HIV epidemic. Factors influencing the recent trends of
the epidemic are still unclear, but may include increased risk
behaviour, the natural epidemiologic cycle and others. To solidify
Uganda's success, the ongoing efforts in HIV prevention need to
be re-emphasized.

Track D
THLBO401
Mapping transactional sex on the Trans-Africa Highway
A. Ferguson', C. Morris'. 1Constella-Futures and University
of Manitoba, Nairobi, Kenya, 2University of Manitoba, Medical
Microbiology, Nairobi, Kenya
Background: As HIV prevalence declines in Kenya and Uganda,
the role of vulnerable groups in maintaining the epidemic becomes
increasingly important. The interaction of two vulnerable groups,
female sex workers (FSW) and truckers, on the major transport
artery linking the East African Coast and the interior, is thought
to have been a major diffusion hearth for HIV, but the volumes
of interaction and the geographical aspects of the 'hot spots'
along the route have never been properly investigated. We report
the results of a study that measures the levels of interaction
between highway-based FSWs and clients, identifies and maps
the hot spots on the route between Mombasa and Kampala and
uses a simple model to predict the annual numbers of new HIV
cases generated by transactional sex on the highway.
Methods: Each of the stopovers was surveyed and mapped using
GIS. Censuses of overnight trucks, FSWs and bar clients were
carried out. At the larger stopovers, samples of FSWs completed
structured diaries for 28 consecutive days to provide data on the
volumes of transactional sex and other information.
Results: 47 stopovers were mapped and surveyed. An average
of 3,000 trucks park overnight and an estimated 8,000 FSWs
are active at these stopovers. From 578 competed diaries,
FSWs recorded a monthly mean of 13 different partners in 24
liaisons and 53 sexual acts. Cond6m use was 90% with casual
clients and 65% with regular clients. 30% of liaisons were with
truckers. Using the AVERT model, about 4,000 annual primary
infections are predicted from transactional sex on this section
of the highway.
Conclusions: While condom use is at a fairly high level, the high
volumes of transational sex continue transmission of HIV. The
bridge population of FSW clients is much wider than expected.
The socio-economic conditions and power relations between
FSWs and clients underlie the problem.
THLBO402
Policy implications of the first systematic tracking of
financial flows for HIV/AIDS in Ukraine
Y. Chechulin', G. Gotsadze2, C. Chanfreau', L. HUnited
Statesk4, V. Saldanha5, A. Nitsoy6, V. Galayda', M. Bhawalkar3,
T. Dmytraczenko3. 'Consultant, Abt Associates Inc., Kiev,
Ukraine, 2International Curatio Foundation, Tbilisi, Georgia, 3Abt
Associates Inc, Bethesda, United States, 4ICF "International HIV/
AIDS Alliance in Ukraine", Kiev, Ukraine, 5UNAIDS, Kiev, Ukraine,
6Consultant, ICF "International HIV/AIDS Alliance in Ukraine",
Kiev, Ukraine, 'Consultant, World Bank, Kiev, Ukraine
Background: Ukraine has the most severe HIV/AIDS epidemic
in Europe. Appropriate monitoring is necessary to support
evidence-based policy decision-making and the requirements of
donors who are important financiers of the HIV/AIDS response.
Ukraine National Health Accounts HIV/AIDS subanalysis 2003 -2004 is country's first attempt to track systematically HIV/AIDS
financial flows.
Methods: The International Classification of Health Accounts
was adapted to Ukrainian context to organize HIV/AIDS financial
flows from public, private and donor sources to financing agents,
providers and functions. Data was collected through official
statistical system, available HIV studies and expert interviews.
Methodology used in data collection and partitioning was
developed and agreed by multi-sector national working group
(government, NGO, etc.).

Late Breaker
Abstracts

----------------------
Affiliated
Events
Roadmaps
Programme
Contributors
Index

XVI INTERNATIONAL AIDS CONFERENCE * 13-18 AUGUST 2006 * TORONTO CANADA * PROGRAMME SUPPLEMENT