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

XIV International AIDS Conference

Abstracts ThOrD1400-ThOrD1426 489

Results: Less than half of women (47.3%) report receiving counseling, and only
74.3% report being offered an HIV test. Once offered a test, the majority (90.0%)
accept. 135 surveyed providers (81.5% were MDs) report that more than 90% of
patients are counseled in 96.1% of practices, and offered testing in 91.9%. Focus group respondents describe widely differing VCT experiences, perceptions of
whether counseling and test-offering occurred were subject to interpretation, and
knowledge of legal rights to receive and voluntarily accept the test were poorly
understood.
Conclusions: Rates of HIV counseling and test offering, are far shy of universal,
and opportunities for intervention may be lost. An evaluation of compliance to the
mandate is subject to interpretation, estimation, and recall bias. The results lend
support to a revised policy of routine HIV testing where women have right of refusal. At a minimum, state or national health agencies should provide monitoring
to encourage provider accountability and compliance with the law. Clients should
be educated about their rights to receive and make an informed choice about
VCT
Presenting author: Elizabeth Montgomery, UZ-UCSF Collaborative Programme,
15 Phillips Ave, Belgravia, Harare, Zimbabwe, Tel.: [phone], Fax: +263 4
704 897, E-mail: liz@ uz-ucsf.co.zw
ThOrD1 400 On-site HIV testing in residential drug treatment
programs: results from a nationwide United States
survey
S.M. Strauss1, D.C. Des Jarlais2, Z.R Vassilev1, J. Astonel. 'NDRI, NDRI, 71
West 23rd Street, 8th floor, New York, New York 10010, United States; 2Beth
Israel Medical Center, New York, United States
Background: Drug treatment programs are uniquely situated to provide drug
users with HIV counseling and testing, but not all types of programs make testing
available to their patients. While some organizational and institutional factors associated with HIV counseling and testing in outpatient drug treatment programs
have been identified, the research involving residential treatment units has been
quite limited. This study identifies program characteristics that differentiate residential treatment programs in the U.S. that make HIV testing available to their
patients on-site (47.8%) and those that do not (52.2%).
Method: The analyses involve t-tests and chi-square analyses, and use data collected in telephone interviews with program managers in a random nationwide
U.S. sample (N=136) of residential drug treatment programs.
Results: Residential programs having a greater number of staff that provide direct patient services are significantly more likely to make HIV testing available to
their patients on-site. Programs having a medical orientation are also significantly
more likely to provide on-site HIV testing. In particular, programs that provide this
testing are significantly more likely to (1) be operated by a hospital; (2) view their
agency as being their patients' primary medical provider; (3) provide patients with
medical care either on-site or at another part of the agency; and (4) have medical
staff either on-site or at another part of the agency.
Conclusions: In view of the critical importance of HIV testing in this population,
the ready availability of training in HIV counseling in most U.S. cities, and the
availability of a simplified testing protocol involving saliva samples (eliminating
the need for phlebotomy), programs that do not have a medical orientation need
to be encouraged to make HIV testing available in their programs.
Presenting author: Shiela Strauss, NDRI, 71 West 23rd Street, 8th floor, New
York, New York 10010, United States, Tel.: [phone], Fax: [phone]
0894, E-mail: [email protected]
ThOrD1 401 An evaluation of California's neighborhood
interventions geared to high-risk testing (NIGHT)
outreach program
H. Rasmussen1, M. Chen1, R. Myrick2, S. Truax1. I Department of Health
Services, Office of AIDS, 611 North Seventh Street, Suite A, Sacramento, CA
95814, United States; 2Universitywide AIDS Research Program, University of
California, Office of the President, Oakland, United States
Background: In 1996, the California State Office of AIDS (OA) began implementing an innovative statewide HIV counseling and testing (C&T) referral initiative
(NIGHT) targeting high-risk populations. Through the use of mobile vans, NIGHT
offers street outreach in geographic locations frequented by high-risk individuals
and provides HIV C&T services and referrals. This study evaluates the extent to
which NIGHT is achieving its goals.
Methods: We used data collected on individuals who tested for HIV between
07/01/97 and 12/31/01 in counties/cities that have implemented NIGHT. We compared demographic characteristics, HIV serostatus, and HIV-related risk behaviors of clients who were referred by NIGHT to that of clients who were referred to
testing by other sources.
Results: Between 07/01/97 and 12/31/01, a total of 483,463 HIV tests were
conducted in counties/cities that have implemented NIGHT. Of these, 104,850

(21.7%) were referred by NIGHT NIGHT-referred tests accounted for 18.1% of
the overall HIV tests in 2000/2001, up from 11.0% in 1997/1998. Of the 104,850
tests referred by NIGHT, 62,321 (59.4%) were for people of color, 35,448 (33.8%)
women of child-bearing age, 24,222 (23.1%) injection drug users (IDUs), 46,938
(44.8%) stimulant drug users, 10,676 (10.2%) men who have sex with men, and
12,692 (12.1%) sex workers. About 1.1% (1,174) of the 104,850 NIGHT-referred
tests were HIV positive. Compared to the 378,613 HIV tests referred by other

sources, NIGHT-referred tests had significant higher percentage of African Americans (28.2% vs. 13.3%, p<0.001), IDUs (23.1% vs. 11.0%, p<0.001), stimulant drug users (44.8% vs. 25.3%, p<0.001), and sex workers (12.1% vs. 5.1%,
p<0.001).
Conclusion: NIGHT is achieving its goal of referring individuals at high level of
risk for HIV to HIV C&T services. Street outreach is particularly effective in increasing HIV testing among drug users, sex workers and high-risk African American clients.
Presenting author: Mi Chen, 611 North Seventh Street, Suite A, Sacramento,
CA 95814, United States, Tel.: [phone], Fax: [phone], E-mail:
[email protected]
ThOrD1402 Routes to combined HIV/TB care: VCT Vs.
Household
H.M. Ayles1, D. Chikwamphul, A. Mwale1, V. Halumamba', P. Mitimingi1,
P Godfrey-Faussett2. 1 Zambart project, 3, Purves Road, kensal Green, London,
Zambia; 2London School of Hygiene and Tropical Medicine, London, United
Kingdom
Background: TB and HIV are causing a major combined epidemic. There is considerable momentum to tackle them together by linking HIV counselling and care
with TB prevention and treatment.
Methods: Combined HIV/TB pilot sites were set up in Lusaka in conjunction with
government and NGO health providers. Access to care was via voluntary counselling and testing (VCT) centres where services included TB case finding, TB
preventive therapy (TBPT), STI management, treatment of opportunistic infections and referral systems to other care in addition to regular VCT activity At the
same time TB patients in 2 of the communities were approached directly and offered an intervention for their household that included TB case finding, TBPT and
VCT in the home.
Results: Over a 2 year period 6 TB/HIV VCT centres were established. 10,944
individuals underwent VCT, 3,071 tested positive for HIV and 1142 started TBPT.
Adherence to TBPT was only 23.5%. 78 additional cases of TB were diagnosed.
In the household studies, 250 households were recruited consisting of 865 adults.
80% of adults received counselling and 317 tested for HIV, 156 (49.2%) were
positive. 49 started on TBPT with an adherence of 42.1%. 48 additional cases
of TB were detected. Disclosure of results within households was good and TB
outcomes were improved.
Conclusions: These two routes cannot be directly compared but they do offer insight into combined HIV/TB care. A combined approach is possible both via VCT
and household route. By involving the entire household we can improve TB outcome, improve TBPT adherence and show enhanced discussion of HIV within
households and disclosure of HIV status. We suggest that alternative routes
should be evaluated for increased quality of the intervention, especially as an
HIV prevention strategy.
Presenting author: Helen Ayles, 3, Purves Road, kensal Green, London, United
Kingdom, Tel.: [phone], E-mail: [email protected]
ThOrD1426 Preferential in-utero transmission of HIV-1 subtype
C compared to subtype A or D
B. Renaifo1, T.Z. Vitamin and HIV Study Group2, P Gilbert3, B. Chaplin1,
G. Msamanga2, D. Mwakagile2, W. Fawzi4, M. Essex'. 'Harvard AIDS Institute,
Harvard AIDS Institute, 651 Huntington Av, Boston, MA 02115, United States
Territory; 2Muhimbili Medical Centre, Dar es Salaam, Tanzania; 3Fred
Hutchinson Cancer Research Center, Seattle, United States Territory; 4Harvard
School of Public Health, Boston, United States Territory
Background: HIV-1 subtype C has become the most prevalent subtype in the
HIV pandemic, accounting for the majority of all circulating subtypes in the world
with subtype C transmitted HIV-1 to their infants significantly more frequently than
other subtypes. Since in-utero, intra-partum or breastfeeding represent different
stages for MTCT, we sought to establish whether transmission during these periods was similar among subtypes, as such information might be useful for scheduling prophylaxis interventions.
Methods: Infants were tested by PCR at birth, six weeks and at subsequent
three month intervals until 18 months of age and thereafter until interruption of
breastfeeding. C2-C5 envelope sequences were used for subtype classification.
Kruskal-Wallis and exact matched conditional logistic regression analysis (MLR)
were used to assess differences in transmission among subtypes.
Results: Of 253 infants positive for HIV-1, 101 infections were caused by subtype A, 73 by subtype C, 53 by subtype D, 22 by recombinants and 4 samples
could not be typed. The distribution of infections during in-utero, intra-partum or
breastfeeding periods was different among infants infected with subtype A, C, or
D (p=0.032). Subtype C infections occurred earlier with a larger component of

in-utero infections, while subtype D infections showed a larger fraction of breastfeeding infections (OR:3.8; 95% C1.1.33-10.8). This association was independent
of maternal CD4 counts.
Conclusion: In comparison of subtypes A and D, a significant number of MTCT
subtype C infections took place in-utero. Prophylaxis initiated at delivery may be
less effective in protecting infants born in geographical regions with expanding
subtype C epidemics.