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

374 Abstracts 23165-23169

12th World AIDS Conference

age, residence, ethnicity education and reproductive history. Seropositive mothers and seronegative controls matched on age and parity were further interviewed
about other risk factors. To identify potential risk factors, odds ratios (OR) with
95% confidence intervals (CI) were calculated. Logistic regression was used to
adjust odds ratios for confounding variables.
Results: 168 HIV-1 (0.6%) and 336 HIV-2 (1.1%) seropositive women were
identified. 16 of these women were dually reactive compared with 0.2% and 1.4%
respectively in 1988. Women with HIV-1 infection were significantly younger than
women with HIV-2 infection. There were significant differences in prevalence in
different regions and ethnic groups. From logistic regression analysis, the risk
factors that were statistically significant for HIV infection included; women who
had been married more than once, had history of post coital bleed, had a husband
that smokes, was TPRA positive, had a husband that had traveled to Cote d'lvoire
and had ever used a condom.
Conclusion: The prevalence of HIV- 1 infection has increased since the late
1980s but HIV-2 prevalence has remained at the same level. Of the potential risk
factors investigated, the same 6 variables were statistically significant for HIV-1
and HIV-2 and there were no major difference in the magnitude of the Odds
Ratios between the two HIV types.
23165 1 Discordant couples! A silent high risk group in HIV
transmission, a case of Uganda
Mary Byangire Amanyire1, L. Marum1, C. Rwabukwali2. 1 IPH Makerere
University, PO Box 7062, Kampala; 2Sociology Dept, Makerere University,
Kampala, Uganda
To provide information on attitudes of married couples towards couple counseling,
testing and discordant results. (Objective)
Methodology: A descriptive cross sectional study was carried out in a periurban area using simple random sampling, 30 couples who had lived together
for more than 3 years were selected. A structured questionnaire was used to
interview the couple. A mastersheet was used to code and analyse findings.
Results: Three couples (10%) who had lived together for less than 5 years
and had only one child each had gone for Couple counseling and testing (CCT)
before marriage. Three (10%), couples who had lived together for more than 15
years had each gone for HIV testing individually without their partners' knowledge.
Eighteen (60%) of the men said they would immediately divorce a wife who was
discordant positive while 12 (40%) felt they would completely abstain from any
sexual relationship but no divorce if they have children. Again, 21 (70%) of the
men said they would never inform their wives if found positive and they would
make sure the wife is equally infected. As for the women, 25 (83%) said they
would allow their husbands to use condoms if the men were discordant positive
especially for the sake of the children, 3 (10%) would abstain from sex while 2
(6%) would divorce. These were preliminary findings.
Children seem to be an important factor in guiding the dicision making for
discordant couples. Furthermore, the couples that got married during the epidemic period are more likely to take HIV CCT than those who married before
the epidemic. Again, men are more likely to abandon their partners in a positive
discordant relationship.
23166 HIV discordant and high risk HIV negative couples:
Preliminary results from a study at the AIDS Information
Centre (AIC) in Kampala, Uganda
Elizabeth Marum1, R. Downing2, B. Biryawaho3, J. Kalule,4, J. Mubangizi4,
D. Smith5, M.G. Alwano Edyegu4. Centers for Disease Control & Prevention,
USAID, PO BOX 7007, Kampala; 2CDC/Uganda Virus Research Institute,
Entebbe; 3Uganda Virus Research Institute, Entebbe; 4AIDS Information Centre,
Kampala, Uganda; 5Centers for Disease & Prevention, Atlanta, GA, USA
Background: AIC has been providing voluntary and anonymous counseling and
testing (CT) services since 1990. In 1997 in Kampala, 24% of clients attended with
their partners; 10.2% of couples were HIV discordant. Among those concordantly
HIV negative, 12% reported risk factors similar to couples where one or both
were infected. To determine if additional risk-reduction counselling and medical
services promote the adoption of risk reduction behaviors, AIC designed a study
for both discordant and "high risk negative" couples.
Methods: All couples attending AIC in Kampala thus far in 1998 have been
screened for eligibility; those eligible and consenting give blood and urine specimens to: confirm HIV and syphilis serology; test for pregnancy, determine CD4
and CD8 cell counts and HIV subtype where appropriate.
Results: 120 couples were screened in the first 7 working days of 1998. 20
couples (16.7%) were HIV discordant and 18 couples (15%) were concordantly
negative but met the criteria for "high risk". 25 of these 38 eligible couples consented to join the study and an additional 5 couples asked for time to consider.
Among consenting couples to date, 12 are discordant (female HIV+ in 7 cases)
and 13 are high risk concordant negative. Mean age in females was 26.9, range
16-42; in males was 35.4, range 22-47.5/49 (10.2%) clients (3 females and 2
males) were reactive for syphilis, including one HIV discordant couples who both

appeared to be actively infected with syphilis. Mean CD4 cell counts in infected
females was 425 cells/ul compared to 244 cells/ul in males (p = 0.27).
Conclusions: The early response to the study is encouraging and the target of
250 couples should be reached in 6 months. Preliminary results suggest that men
in these couples tend to be older and infected males have lower CD4 counts than
their younger partners. The high rate of consent among HIV negative but high

risk couples suggests that when informed of their high risk characteristics, such
couples are eager to receive additional services to help them avoid HIV infection.
231671 Widow inheritance in HIV infection: Experience in rural
Uganda
Bogere George Bosa. c/o Rest Nalongo, 0.0. Box 2971, Kampala, Uganda
Issue: Women vulnerable to HIV/AIDS need lifeskills education.
Project: Namulonge AIDS Project was set up in 1988. Here 10 village elders,
9 village councillors, 1 parish priest and 7 catechists were trained as village counsellors and sex educators on how to give AIDS sensitisation and basic counselling
to their people, empowerment of women with lifeskills and supporting them on
how to talk with relatives about the dangers of widow inheritance.
Results: 9 widows rejected being inherited in 1990. 1993 saw widows going
against the practice and 39 widows dropped the culture in 1996. Men have
discovered the taboo associated with widow inheritance and have begun to shun
it.
Lessons Learned: Women should be empowered with lifeskills. Also the community has to be given AIDS awareness to restrain themselves from their cultural
practices to minimise HIV infection.
123168 Gender differences in the risk of HIV infection among
young adult recently initiated injection drug users (the
REACH I study)
Meg C. Doherty1, R.S. Gafein2, D. Brown2, C. Fuller2, D. Vlahov2. 159
Hampstead Road #1 Jamaica Plain MA; 2Johns Hopkins School of Public
Health, USA
Background: To examine the relation between gender and HIV-associated high
risk behaviors among adolescent and young adult recent-onset injection drug
users (IDUs) in Baltimore, Maryland.
Methods: Between 1994-96, interviews were conducted to assess circumstances at initiation of IDU, current IDU, and sexual behaviors among 18 to 25
year old IDUs recruited via street outreach. Serum was tested for antibodies to
HIV. Data were analyzed using stratified and logistic regression methods, with
significance at alpha = 0.05.
Results: At enrollment, mean duration of IDU was 3 years and 33 (14%) of
229 IDUs were HIV-positive. Women were more likely to be infected than men
(17% vs 11%, p < 0.1). At initiation, compared to men, women were more likely
to have had another person ('helper') inject the drug (88% vs. 78%), a helper
who was HIV+ (12% vs. 3%), and a helper who was a female friend (75% vs.
33%). Stratified analysis of HIV infection by gender indicated that for women, ever
injecting cocaine, injecting >5 x/day, backloading and smoking crack daily were
significant. Among men, ever injecting cocaine, having >2 persons help inject the
drug prior to learning self-injection ('trainer'), sharing needles, and smoking crack
daily were significant for HIV. Regarding lifetime sexual behaviors among women,
HIV was associated with having >100 sex partners, and trading anal or oral sex
for money or drugs after starting to inject. For men, sex behaviors associated with
HIV included a history of rape, being gay/bisexual, having >100 sex partners,
and trading anal or oral sex after starting IDU. Separate multivariate models to
predict associations with HIV infection were fit for women and men. For females,
trading anal sex (OR = 32.8; 95% CI, 2.4-462), ever injecting cocaine (OR = 7.8;
95% CI, 1.6-39.3), and smoking crack daily (OR = 3.3; 95% CI, 1.1-9.5) were
independently associated with HIV serostatus. Alternatively for males, trading
anal sex (OR = 11.2; 95% Cl, 1.7-72.3), smoking crack daily (OR = 8.7; 95% Cl,
1.2-66.3), and sharing needles (OR = 4.4; 95% Cl, 1.0-19.7) were independently
associated with HIV serostatus.
Conclusions: Among young recent-onset IDUs, women and men display differential HIV risk behaviors. While trading sex and smoking crack are strongly
correlated with HIV for both genders, women who injected cocaine were at an
increased risk, while males had an added risk from sharing needles. These risk
differentials may reflect differences in the economics of drug use, needle accessibility and strategies to reduce sexual exposures. These findings suggest that
HIV prevention efforts should be sensitive to gender differences in their target
populations.
123169 Very low HIV seroconversions among injecting drug
users participating in a HBV vaccination study. Extension
of an ongoing multicenter study
Paolo Mezzelani1, F. Lugoboni2, L. Venturini3, C. Bossi4, D. Bantiero4,
M. Faccini5, S.R. Friedman6. 1Clinica Medica Policlinico Universita via Delle
Menegone, Verona; 2Sert Bussolengo, Bussolengo VR; 3University of Verona,
Verona; 4Sert Legnago; Legnago VR; 5Sert Zervio, Zervio VR, Italy; 6National
Development & Research Institute, Inc., New York, USA
Objective: To ascertain HIV seroconversions among injecting drug users (IDUs)
participating in a hepatitis B vaccination program. This represents the extension

of an ongoing multicenter study.
Methods: An HBV vaccination program was offered to 557 HBV and HIV
negative IDUs attending 3 large Centers for Drug Users of the Veneto Region of
Italy. Among them 457 accepted and 120 declined. Both groups were followed
over time and tested for at least twice for HIV with a mean of 4.3 tests over 13697
person-months for the vaccination group and a mean of 3.6 tests in 3937 person