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

calculated using data collected through population-based surveys on two
variables: reported condom use in the last high-risk sexual act and reported
'every time' use in all high-risk sex acts in the preceding year.
Surveys using these measures have yielded high levels of consistent condom use
- levels that are believed by some to be over-reported due to social desirability
bias. This paper presents an alternate measure for consistent condom use that
yields lower and potentially realistic levels of consistent condom use.
Methods: In a cross-sectional population based survey among male clients
of sex workers in four Indian states in December 2005, Population Services
International, through the Avahan: India AIDS Initiative, supported by the Bill
and Melinda Gates Foundation, collected information on one additional variable
using the question: In the last year, have there been any instances when you
have not used a condom with a commercial sex worker?" We compared the data
obtained in this survey with a previous cross-sectional population based survey
among male clients of sex workers in four Indian states in March 2004 that used
typical measures of consistent condom use.
Results: While reported condom use in the last high-risk sex act was found to
be the same in both surveys, the new composite measure of consistent condom
users utilized in the survey carried out in December 2005 was found to be
significantly lower (p <.01) than the composite measure used in March 2004.
Conclusions: Adding one more question to typical measures of consistent
condom use may yield more realistic levels of consistent condom use and has
the potential to reduce social desirability bias.

13 and 24 (mean age - 17.2) in selected household which contained one or
more adolescents in a multi-stage cluster sample of enumeration areas in four
wards in the South-East region of Nigeria in addition 24 focus group discussion
(FGDs) where held among 182 in-school and out of school adolescent from all
four wards.
Results: Despite relatively high levels of STD and HIV/AIDS knowledge,
many Nigerian adolescents continue to engage in sexual risk behaviours. Low
academic achievement and parental communication were among the most
reliable factors contributing to sexual risk behaviours while individual and
family- level protective factors appear to moderate the impact of risk. Focus
group discussions among in - and out - of - school adolescents provide further
evidence of the general applicability of the explanatory model to the Nigerian
context and contribute the additional constructs of traditional values, gender
inequality, and severe poverty.
Conclusions: The explanatory model of adolescent problem behaviour
employed in this study was found to be generally applicable to the Nigerian
context, while the additional dimension of qualitative research to aid in the
construction of culturally meaningful youth-defined risk and protective factors
greatly enhance the interpretation of study findings, and may act as a guide for
contextually relevant programmatic approaches.
TUPEO467

I Tuesday
15 August
Poster
Exhibition

Findings from the first large multi-city U.S. Sample
of high-risk, substance using men who have sex
TUPEO465 with men: baseline results from the project mix
Characteristics of Injection drug users who inject behavioral intervention trial
with shared needles in Canada

D. Boulos', Y. Choudhril, M. Hennink2, P. Millson3, C. Morissette4, C. Fyfes,
S. Snelling', J. Wylie7, A. Singh", C. Archibald1, I-Track Study
Team. 'Surveillance and Risk Assessment Division/ CIDPC, Public Health
Agency of Canada, Ottawa, Canada, 2Population and Public Health Services,
Regina Qu'Appelle Health Authority, Regina, Canada, 3HIV Social, Behavioural
and Epidemiological Studies Unit, University of Toronto, Toronto, Canada,
4Institut national de sante publique du Quebec/ Direction de la sante publique
de Montreal-Centre/ Ddpartement de medecine sociale et preventive,
University de Montreal, Montreal, Canada, 5Vancouver Island Health Authority,
Victoria, Canada, 6PHRED Program, Sudbury & District Health Unit, Sudbury,
Canada, 'Cadham Provincial Laboratory/ Departments of Medical Microbiology
and Community Health Sciences, University of Manitoba, Winnipeg, Canada,
8Alberta Health and Wellness, Edmonton, Canada
Background: The sharing of injection drug use (IDU) paraphernalia, especially
needles, is a high risk behaviour for HIV transmission. To help guide HIV
prevention programs, we studied factors associated with needle-sharing among
IDUs in Canada.
Methods: I-Track is an enhanced surveillance project that collects behavioural
data and biological specimens for HIV testing from IDUs in sentinel sites across
Canada. Data collected during 2003-2005 from six sites (Victoria, Edmonton,
Regina, Winnipeg, Toronto and Sudbury) formed the basis of this analysis.
Stepwise logistic regression was used to assess demographic and drug use
behavioural variables associated with the sharing of needles.
Results: Of 1,418 individuals who indicated their needle-sharing status, 12.6%
injected with used needles during the past 6 months. Needle-sharing decreased
with age: 18.0% in IDUs aged <30yrs, 10.8% in those >30yrs. Needle-sharing
also increased with increasing injection frequency: 9.4% in IDUs injecting twice
or less per week, 13.3% in those injecting at least 3 times per week and 19.3%
in those injecting at least daily. Needle-sharing was higher among women than
men (16.2% vs 10.8%), among those with any cocaine-use (14.0% vs 8.2%),
and among IDUs with less stable housing (16.5% vs 11.0%) during the past 6
months. The stepwise logistic regression analysis indicated that more needlesharing was significantly associated with females (OR: 1.64, 95%CI: 1.16 -2.31), younger (<30yrs) IDUs (OR: 1.75, 95%CI: 1.22-2.50), IDUs with less
stable recent housing (OR: 1.83, 95%CI: 1.30-2.60), IDUs with any cocaine
usage (OR: 1.62, 95%CI: 1.03-2.54) and IDUs injecting at least daily (OR:
2.00, 95%CI: 1.37-2.91), relative to IDU's injecting twice or less per week.
Conclusions: Within this IDU population, increased recent needle-sharing
was associated with females, younger IDUs, any cocaine usage and less stable
recent housing arrangements. These associations can be used to guide HIV
prevention programming and to focus activities on groups at highest risk for
needle-sharing.
TUPEO466
Risk and protective factors for HIV/AIDS among
Nigerian adolescents
C. Nwagbo-Nwankwo', A. Amali2, F. Alimi3. 'Civil Resource Development and
Documentation Centre (CIRDDOC), Enugu, Nigeria, 2School Kids Net, Kaduna,
Nigeria, 3School Kids Net, Maiduguri, Nigeria
Background: Available data on adolescent's reproductive and sexual health in
Nigeria paints a daunting picture of HIV risk, including early sexual initiation,
multiple partners, intergenerational sex and low condom use. The present
study examines the association between risks and or protective factors and
sexual risk behaviours among youth in the South-East region of Nigeria, and
the applicability of an explanatory model of adolescent problem behaviour to
the Nigerian context.
Methods: Structured interviews investigating a large number of risk and
protective factors operating at the individual, Family, School, Communities and
societal levels were conducted with 2,931 adolescents between the ages of

G. Mansergh, Project Mix Study Group. CDC, Division of HIV/AIDS Prevention,
Atlanta, United States
Background: Substance use is associated with sexual risk among general
samples of men who have sex with men (MSM). However little is known about
behavior specific to very high-risk men, a prime target for change. Project Mix
- a behavioral intervention trial to reduce risk of substance-using MSM who
are not in drug treatment settings - provides the first opportunity to assess
behavior of a large U.S. sample of MSM who report having recent unprotected
anal sex and anal sex while drunk or high.
Methods: Randomized-control trial in 4 U.S. cities: Chicago, Los Angeles, New
York, San Francisco. Intervention and comparison arms consist of 6 smallgroup sessions led by trained facilitators. Participants will be enrolled through
May 2006 (n=1200).
Results: Baseline analyses are for 849 men enrolled by January 2006
(presentation will be n=1200). The sample was diverse: 50% HIV-positive,
30% Black, 25% Latino, 15% non-U.S. native, and 16% non-gay identified.
Participants reported a median of 5 non-primary male sex partners in the
prior 3 months. Behavior during their most recent anal sex encounter with a
non-primary partner included unprotected anal sex (67% overall; 41% with
an HIV-discordant partner), and alcohol (59%) and drug (60%) use (median
2 substances), including methamphetamine (17%). No city differences were
found for prevalence of unprotected anal sex or overall substance use;
men from Los Angeles and San Franciso, and HIV-positive men, were more
likely to report methamphetamine use during sex. In multivariate analysis,
methamphetamine use during sex by the respondent (OR=2.67, 95%CI=1.49 -4.78) and by his recent partner (OR=2.60, 95%CI=1.38-4.88) were associated
with unprotected anal sex.
Conclusions: A diverse sample of very high-risk substance-using MSM can be
successfully enrolled in a large behavioral intervention trial. Even in this risky
sample, methamphetamine stands out as the drug linked to unprotected anal
sex of both respondents and their partners.
TUPEO468
Three-year trends in voluntary HIV testing in
Namibia: can testing have both productive and
counterproductive effects?
R. Rimall, A. Creel', I. Tweedie2, N. Gorelick', K. Stratten3. 'Johns Hopkins
University, Health, Behavior & Society, Baltimore, United States, 2Johns
Hopkins University, Center for Communication Programs, Baltimore, United
States, 3Johns Hopkins University, Center for Communication Programs,
Windhoek, Namibia
Background: Persuading individuals to get tested for HIV remains an important
public health goal. While resources and access to testing sites are important
determinants of testing, this paper focused on the psychosocial and behavioral
determinants of voluntary HIV testing in Namibia. Psychosocial factors included
knowledge about HIV and self-efficacy. Behavioral factors included total number
of sexual partners, prior condom use, intentions to use condoms, and intentions
to remain abstinent in the future. Individuals were categorized into one of three
groups: those who
(a) had not been tested and do not wish to be tested,
(b) had not been tested but wish to be tested, and
(c) had already been tested.
Research questions raised in this paper were: Have voluntary HIV testing
rates increased in Namibia over time and what factors differentiate the three
groups?
Methods: Two cross-sectional data waves were collected through random
sampling techniques in three districts in Namibia, at baseline in 2003 (N=900;
females=65%; mean age=33.3, SD=14.8) and two years later at midterm
(N=900; females=50%; mean age=34.6, SD=15.7).
Results: There was a significant increase in voluntary HIV testing from

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