Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

12th World AIDS Conference

Abstracts 43110-43114 869

Results: From February-September 1997, a total of 25,790 clients were tested.
Of the 25,790 clients tested on Capillus 5,772 (22.4%) were reactive and of the
5,772 reactive to Capillus, 701 (12.1%) were nonreactive on Serocard, thus needing the "tiebreaker" test. Of 701 who were reactive on Capillus and nonreactive on
Serocard, 24 (3.5%) were reactive on Multispot. Of 492 HIV+ samples re-tested
at NBB, all were confirmed positive. Of 1,019 HIV- samples, 1,012 (99.3%) were
confirmed HIV by NBB. The 7 samples reported HIV- with the AIC RT but tested
HIV+ at NBB were later retested on the AIC RT and found HIV positive; 5 of these
7 errors occurred in Kampala, AIC's busiest site. Investigation of these cases
suggested that these errors were caused by faulty recording and not following
proper test procedures.
Lessons Learned: A rapid HIV testing algorithm is feasible to introduce in
developing countries and can yield highly reliable results, though accuracy was
slightly reduced at the high volume setting. Close laboratory supervision and
random re-testing for quality assurance is recommended. Provision of same day
test results using a rapid testing algorithm is feasible, practical, and more cost
efficient than a waiting period algorithm as all persons tested learn their results.
43110 Partner violence and women's decision to have an HIV
test
Julie Maher', G.M. Seeman1, J. Peterson2, L. Dalhberg1, B. Seals1,
G. Shelley1, M.L. Kamb1. 1CDC 1600 Clifton Road E46, Atlanta, Georgia 30333;
2St. Michael's Medical Center, Newark, NJ, USA
Background: Reports of partner violence after HIV partner notification (PN) have
led some policymakers to reconsider HIV PN procedures. Our objective was to
determine whether partner violence was associated with women's HIV testing
decisions.
Methods: We enrolled women attending a New Jersey sexually transmitted
disease (STD) clinic, who were aged -16, had not been HIV-tested in three
months, and were not known to be HIV-positive. Women were offered HIV testing
and interviewed about STDs, sexual behavior, and violence. Partner violence was
defined as being slapped, pushed, kicked, or otherwise physically hurt by a sex
partner.
Results: Between 12/96 and 3/97, 167 (83% of eligibles) women participated:
median age was 26, 91% were African American, 71% had yearly income of
US$10,000, and 89% had a main partner. Forty-nine (29%) declined HIV testing. Twenty-two (13%) reported partner violence in the past 12 months: 18%
(9/49) of test decliners vs. 11% (13/118) of acceptors (p = 0.201). Twelve (7%)
reported partner violence in the past 3 months: 4% (2/49) of test decliners vs. 8%
(10/118) of acceptors (p = 0.512). When asked, no one reported her declining
testing was related to concern that the health department might notify her partner
or fear that her partner might harm her if she had a positive HIV test. Declining
testing was significantly associated with having a main partner and yearly income
of -US$ 10,000.
Conclusions: Nearly one in seven women reported partner violence; however,
preliminary results suggest that partner violence was not significantly associated
with women's HIV testing decisions. Staff should be aware that partner violence
may exist regardless of women's HIV testing decisions and should be prepared
to make appropriate referrals.
143111 Prenatal HIV counseling and testing: Rates and
determinants among women at public health clinics
Kathleen Ethier, R.A. Fox-Tierney, J.R. Ickovics. Yale University 135 College St
Suite 323 New Haven CT, USA
Background and Objectives: US Public Health Service (USPHS) guidelines
recommend universal HIV counseling and voluntary testing of pregnant women
to promote the reduction of perinatal transmission of HIV. The objectives of this
study were to assess the effectiveness of the implementation of these guidelines
among community health care centers and hospital clinics in Connecticut; we
assessed rates of HIV counseling and testing (C & T) among prenatal patients and
examined characteristics that define patients who accepted HIV testing during
pregnancy.
Methods: Rates of HIV counseling and testing and demographics were collected on all prenatal patients at each of 9 participating prenatal clinics in
Connecticut between March, 1995 and January, 1997 (n = 2135).
Results: Across sites, 90.1% of prenatal patients received HIV counseling
during their current pregnancy. Counseling rates ranged from 42.0% to 100% and
differed significantly by site (x2 (9) = 227.47, p -.001). Of those who were counseled, 76.5% accepted an HIV test. Testing rates ranged from 52.5% to 92.4%
and also differed asignificantly by site (x2 (9) = 66.8, p -.001). Test acceptance
rates differed by race and testing history; Hispanic women were more likely than
either Black or White women to accept an HIV test (OR = 1.74, 1.38-2.2). Women
who had not been previously tested (OR = 1.4, 1.03-2.03) were more likely than
those who had been previously tested to accept an HIV test during pregnancy.
There were no differences in test acceptance rates by age, reports of HIV risk or
insurance status.

Conclusion: As suggested by the USPHS, counseling rates were high across
prenatal settings. However, of women who were counseled, only 76.5% accepted
an HIV test. Clinic policies, access to services, provider attitudes, counselor
style, and site-based differences in client population may all contribute to HIV
counseling rates and test acceptance. More research is necessary to identify the
factors that affect the implementation of USPHS guidelines, and to increase rates
of HIV C&T in prenatal settings.

43112 Reason for receiving an HIV test: Experience in a regional
population seropositive for HIV 1985-1997
Dale Guenter1, D. Johns2, M.J. Gill2. University of Calgary, 213, 906-8 Avenue
SW, Calgary, Alberta; 2Southern Alberta Clinic, Calgary, AB, Canada
Objectives: To determine the reasons that individuals diagnosed HIV seropositive
in a large, well characterised regional population had received their first positive
HIV test, and to analyse for significant differences in testing rationale between
demographic subgroups.
Design: Descriptive study
Methods: A total of 1294 patients with documented HIV infection presented to
the referral clinic for HIV care in Calgary, Canada from the period November 1985
through August 1997. The reason for having been tested for HIV was requested
at the initial visit, along with background demographic characteristics and clinical
data. Relative risk analysis and chi square testing were used to compare reasons
for testing among demographic subgroups.
Results: Data on reason for testing was available for 1272 (98%) of the clinic
patients over the study period. Patients diagnosed after 1992 were more likely to
have been tested because of late HIV-related illness (CD4 - 300) than patients
testing before 1989 (RR = 3.47, 95% C.I. 2.32, 5.21, p - 0.0001). Twenty-eight
percent of patients diagnosed after the age of 40 were tested because of late
HIV-related illness, significantly more than those 40 years and under (RR = 1.96,
95% C.I. 1.49, 2.56, p - 0.0001). IDU and gay men who tested positive had
a similar distribution of reasons for testing. However, HIV positive bisexual men
were more likely to have tested because of late HIV-related illness than either IDU
or gay men (RR bisexual vs. gay = 1.77, 95% C.I. 1.31, 2.39, p = 0.0002). Women
were more commonly diagnosed because of high risk contact with a known HIV
positive individual than were men (RR = 3.02, 95% C.I. 2.05, 4.44, p - 0.0001).
Conclusions: The majority of our patients (74%) were diagnosed before the
onset of significant clinical features of HIV disease, reflecting effective screening
practice. Bisexual men and people over age 40 may not receive appropriate public
health information, or fail to acknowledge their risk of infection. We observed more
infections identified due to illness in recent years, challenging the effectiveness
of current efforts to convince high risk groups about screening practice. Contact
tracing in our study population appears to identify women who are HIV positive
more readily than men.
43113 What you don't know can hurt you: The influence of prior
HIV testing on serostatus results at repeat testing
Judith Levy1, S.E. Fox2, M. Valle2. UIC-SPH 2121 W Taylor, Chicago, IL 60612;
2Univ. of Illinois at Chicago, Chicago, IL, USA
Objective: To examine the influence of prior HIV testing on serostatus results at
repeat testing.
Design: Case Control and In-depth Interviews.
Methods: Using snowball sampling of local drug-using networks, trained outreach workers recruited 334 not-in-treatment, active injecting drug users (IDUs)
for HIV counseling testing, and partner notification. In-depth interviews administered at time of testing were used to collect information about the person's prior
testing experience, risk-behaviors, and attitude toward informing others of potential exposure. Odds ratios were calculated to determine the association between
prior testing and receiving a positive test result at time 2. Qualitative data from
the in-depth interviews were analyzed to identify attitudes toward receiving test
results.
Results: A total of 58 (17%) tested HIV-postive. Of the 334, 107 (32%) had no
history of prior HIV testing, 165 (50%) had tested sero-negative, and 62 (19%)
had been tested but had not returned for their resutls. Previously tested IDUs
who had not obtained their results at time I were over 12 times more likely to test
positive at time 2 than those who had received a negative test result (p -.01).
Reasons for not wanting to be informed of the test outcome include perceived
inability to cope, fear of social rejection, and the desire to avoid adverse legal and
social consequences.
Conclusions: Findings suggest the importance of developing strategies to
encourage IDUs to obtain their test results in order to seek medical treatment if
needed. This effort includes finding means to overcome the barriers to wanting
to learn their test outcome.
43114 Locally sustainable administration of HIV counseling and
testing to young couples in rural regions of western
Kenya
Mark Damesyn1, E.R. Stiehm2, C.G. Neumann3, D. Morisky3, W.O. Omwomo4.
110833 Le Conte Ave., 71-254 CHS, mail code 177220 Los Angeles, CA;
2Principal Investigator/University of California, LA, Dept of Pediatric Immun. Los
Angeles, CA; 3Co-Principal Investigator/UCLA Dept. of Community Health
Sciences, Los Angeles, CA, USA; 4Study Coordinator/Diocese of Maseno West,
Siaya, Kenya

Objectives/(Issues): The Siaya AIDS Prevention study is designed to evaluate methods of preventing HIV infection utilizing pre-existing local health care
systems in Western Kenya. An intervention based on counseling and testing of
young couples was chosen based on previous studies. Local community health
workers were trained in counseling and saliva sample collection. Demand for HIV
counseling and testing was measured as the outcome of interest.
Project/Methods/Results: Households for interview were selected through
multi-stage cluster sampling. The Participants were interviewed by specially