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

Serum left over from the syphilis testing is anonymised and stored at -200 for
surveillance and research purposes. We tested samples collected in 2000-2002
for HIV (Serodia with confirmation by an ELISA) and HSV-2 infection (HSV
specific IgG test of Kalon Biological Ltd). Results were linked to interview data
using a unique identifier code. Risk factors for HSV-2 infection were explored
with univariate and multivariate analyses.
Results: A total of 938 samples were available for testing. The prevalence of
HSV-2 infection was 54.5% and of HIV infection 27.4%. In women aged 15 -19 years the prevalence of HSV-2 infection was 38.8%; it increased to 75.0%
in women > 29 years. The HSV-2 prevalence decreased from 60.5% in 2000
to 46.7% in 2002 (test for trend: p=0.0003). There was a strong association
with HIV: prevalence of HIV infection was 42.4% in HSV-2 positive women
and 9.6% in HSV-2 negative women. In multivariate analysis HSV-2 infection
remained significantly associated with age, HIV infection and year the sample
was taken.
Conclusions: The prevalence of HSV-2 infection, an incurable STI that
increases the susceptibility to HIV infection, is high among FSW in Cambodia.
But the trend of decreasing prevalence of HSV-2 infection between 2000 and
2002 suggests that the programme of 100% condom use had an effect on HSV2 infection besides its effect on HIV.
TUPE0 29 5
Incident HIV infections attributable to sexually
transmitted and vaginal infections among women in
Zimbabwe and Uganda
J. van de Wijgert', C. Morrison2, J. Wang3, J. Brown4, B. van der Po015,
M. Munjoma6, H. Bakka', B. Richardson8, N. Padian4, R. Salata9. IIATEC
Foundation and Center for Poverty-related Communicable Diseases, Academic
Medical Center, University of Amsterdam, Amsterdam, Netherlands, 2Family
Health International, Research Triangle Park, NC, United States, 3Fred
Hutchinson Cancer Research Center, Seattle, WA, United States, 4University
of California at San Francisco, San Francisco, CA, United States, 'Indiana
University, Indianapolis, IN, United States, 'University of Zimbabwe, Harare,
Zimbabwe, 7Joint Clinical Research Centre, Kampala, Uganda, 'University of
Washington, Seattle, WA, United States, 9Case Western University, Cleveland,
OH, United States
Background: The proportions of incident HIV attributable to sexually
transmitted infections (STI) and vaginal infections were calculated using data
from a large prospective cohort study in Zimbabwe (ZM) and Uganda (UG). This
analysis is unique in that 202 incident HIV infections were observed in a general
population sample, all infections were laboratory-confirmed, and data on many
potential confounding covariates were collected.
Methods: HIV-uninfected family planning clinic attendees (n=4,531) were
followed every 3 months for 15-24 months between 1999 and 2004 to determine
HIV-1 incidence. At each visit, participants underwent pelvic exams, testing for
HIV, STIs and vaginal infections, and were interviewed about risk factors for
HIV. Hazard ratios (HR) for HIV were estimated for each infection, adjusted
for study site, age, schooling, living with partner, hormonal contraceptive use,
pregnancy, number of live births, condom use, risk behavior of the woman, and
risk behavior of her partner(s). Population attributable fractions (PAFs) were
calculated using the adjusted HR and prevalence of infection.
Results: In the combined ZM+UG model, time-varying HSV-2 by EIA (HR=4.17;
95% confidence interval (CI) 2.81-6.17), N. gonorrhoeae by PCR (HR=5.76;
95% CI 3.64-9.13), bacterial vaginosis by Nugent scoring (HR=1.67; 95%
CI 1.24-2.24), and candidiasis by wet mount (HR=1.67; 95% CI 1.13-2.49)
were all significantly associated with HIV acquisition; C. trachomatis by PCR
(HR=1.75; 95% CI 0.98-3.13) and T. vaginalis by wet mount (HR=1.33; 95%
CI 0.66-2.69) were not. The PAFs for HSV-2 were 64.0% (ZM+UG), 65.8%
(ZM) and 56.0% (UG); N. gonorrhoeae 5.9% (ZM+UG), 5.3% (ZM) and 4.3%
(UG); C. trachomatis 1.1% (ZM+UG) and 1.2% (ZM); bacterial vaginosis
13.1% (ZM+UG), 15.8% (ZM) and 6.3% (UG); and candidiasis 5.8% (ZM+UG)
and 11.6% (ZM). No incident HIV infections were attributable to T. vaginalis.
Conclusions: A large proportion of new HIV infections in Zimbabwean and
Ugandan women were attributable to STIs (particularly HSV-2) and vaginal
infections.
TUPE0296
Factors associated with use of male condom in
adolescent from Peru
A. Cotrina-Rabanal, P. Garcia, C. Carcamo. Universidad Peruana Cayetano
Heredia, Lima, Peru
Background: To determine the level of condom use and the factors associated
with condom use between adolescents.
Methods: A cross sectional study, performed in 2004 using a household based
survey. The survey included three cities with high rate of urban areas and high
proportion of young people, socio-culturally representative of the three major
regions (coast, highlands and jungle) of Peru. The study was designated to
include unmarried adolescents between 15 to 19 years old who were permanent

residents of each of the three cities. The sampling approach involved a three
stage cluster sample. The key variables were measured via a standardized
self-reported questionnaire. The effective sample size of 15 19 year old for
the present investigation and the proportion of designated participants with
useable data are 2180 and 95.8%, respectively. The study protocol was
reviewed and approved by the institutional review board (IRB) of Universidad
Peruana Cayetano Heredia in Peru.
Results: "Never use a condom" were reported in 62-76% of females and
20% of males. 83-86% of females reported "never bought a condom". The

greater part of males got the condoms from friends; females got them from
the Health Public System. Adolescents reported the main reason for condom
use is the availability. The regression analysis disclosed that: be male (OR=3,
p=0.00); "insufficient sexual education reported at school" (OR=0.6, p=0.00);
"parental communication about condoms" (OR=2.8, p=0.00) and "know where
get condoms" (OR=4.4, p=0.04) were associated with condom use. 70 -80%
of adolescents reported the condom vending machines can improve the condom
use.
Conclusions: Parental communication and the education at schools can
improve "safe sex behavior" between adolescent. The results from this study
may have important implications to develop effective intervention programs
(like condom vending machines) to the prevention of STD and HIV.
TUPE0297
HIV infection and other sexually transmitted
infections among female sex workers in a mining
township in Yunnan, China
N. Wang', Z. Yang', H. Gao', Q. Duan', R. Zhao', L. Lu2, Y. Pu3, W. Ni3,
Z. Wu4. 1Chinese Center for Disease Control & Prevention, National Center for
AIDS/STDs Control & Prevention, Beijing, China, 2Yunnan Provincial Center
for Disease Control & Prevention, Kunming, China, 3Gejiu Center for AIDS/
STDs Control & Prevention, Gejiu, China, 4China Comprehensive Integrated
Programs for Research on AIDS, National Center for AIDS/STDs Control &
Prevention, Chinese Center for Disease Control & Prevention, Beijing, China
Background: The province of Yunnan has a high prevalence of HIV and the
highest proportion of illegal drug users in China. Therefore, understanding HIV
transmission patterns in this region is integral to controlling HIV transmission
throughout China.
Methods: A cross-sectional prevalence survey on sexually transmitted
infections (STIs) was conducted amongst 237 female sex workers (FSWs) aged
>16 years during December 2005. Demographic and behavioral information
was collected by self-administered questionnaire. Participants provided
tampon swabs for Trichomonas vaginalis and yeast detection, urine samples
for morphine detection, and blood specimens for HIV-1, syphilis, and herpes
simplex virus-2 (HSV-2) antibody testing.
Results: Forty-nine participants (20.7%) were infected with HIV [95% CI,
15.5%-25.8%]. Prevalence figures for syphilis, HSV-2, trichomonas, and vaginal
yeast were 3.8% (95% CI, 1.7%-7.1%), 63.7%(95% CI, 57.6%-69.8%),
4.8%(95% CI, 2.4%-8.4%) and 11.3%(95% CI, 7.5%-16.1%), respectively.
156 participants (65.8%) had evidence of prior HSV-2 infection or syphilis,
and 42 (17.72%) were HIV-positive with evidence of prior HSV-2 infection or
syphilis. Factors associated with HIV infection were illegal drug use (OR, 29.0;
95% CI, 12.8-65.6), intravenous drug use (OR, 7.8; CI, 2.7-22.7), HSV-2
infection (OR, 4.3; 95% CI, 1.8-10.2), lower venue level (p<0.01), older age
group (p<0.001), self-reported risk of HIV infection (p<0.01), and registered
permanent residence in the local township (p<0.001). Fifty-two participants
(21.9%) were urine morphine positive. Of these, 19.2% and 71.1% selfreported never having used illegal drugs and not having used illegal drugs in
the past 3 months, respectively. 67.4% and 79.7% of participants have lived
in this town for >1 year and >0.5 year, respectively. Duration of town residence
was associated with drug use (p<0.01).
Conclusions: HIV and other STIs are highly prevalent amongst FSWs in
this region of China, and most HIV infections are concentrated in illegal drug
users.
TUPE0298
STI coinfections among Toronto's HIV-infected
population
C. Achonu, E. Gournis, N. Whittingham, 3.-A. Ackery, B. Yaffe,
R. Shahin. Toronto Public Health, Communicable Disease Control, Toronto,
Canada
Background: The diagnosis of subsequent sexually transmitted infections
(STIs) among HIV-infected individuals is concerning as it signals continued
high risk behaviour. HIV-infected individuals usually receive counselling on the
need for safer sex practices. Understanding the degree of additional STIs within
this population provides an important indication of ongoing high risk sexual
behaviour. The objective of this study was to understand the pattern of new
STIs within Toronto's HIV-infected population over time.
Methods: All HIV, chlamydia, gonorrhea and infectious syphilis cases reported
to Toronto Public Health (TPH) from 1992 to 2004 were extracted from
Ontario's communicable disease database (RDIS). Coinfections were identified
by matching individual record numbers from HIV, chlamydia, gonorrhea and
infectious syphilis reports. Only STIs reported one incubation period or later
following HIV diagnosis were counted as new coinfections. Demographic and
risk exposure data obtained from STI co-infected HIV cases were analyzed
using descriptive methods.
Results: From 1992 to 2004, there were 7,536 HIV cases reported to TPH.
Among these cases, a total of 1061 STI new coinfections were reported by 599

(8%) individuals. Thirty-six percent (n=219) of all newly co-infected individuals
reported more than one STI. The most commonly reported STI coinfection was
gonorrhea. co-infected individuals were predominantly males between 25 to 39
years at HIV diagnosis whose primary HIV exposure was sex with other males
(MSM). From 1992 to 2004, the rate of new STI coinfections among HIV cases
increased 10-fold from 4.6 to 50.8 coinfections per 1000 person-years.
Conclusions: A small proportion of HIV cases reported from 1992 to 2004
were subsequently diagnosed with STIs. However, an increased rate of STI
coinfection among Toronto's HIV population is concerning and increases the

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