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

12th World AIDS Conference

Abstracts 43465-43469 937

of data among the hospitals and the Ministry of Health is carried out by encrypted
diskettes, in order to assure the confidenciality. Its architecture is composed by
the clinic, epidemiologic and cots subsystems.
Results: Up to date clinic, epidemiologic, economic data about the sickness evolution. Supporting mechanism to direct patients care, as resume of the
clinical evolution and other files. (receipts, exam orders, external care forms,
appointments.) Structured data to support the decision-making process and the
evaluation of already implemented strategies of care.
Conclusions: The system allows:
(1) to facilite the engagement of health authorities on HIV/AIDS patients needs
satisfaction,
(2) to optimize individual patients care and the resources management,
(3) to identify more precisely the evolution of HIV/AIDS characteristics,
(4) to evaluate the impact of health programs in clinical, epidemiological and
economic fields.
S43465 Using an HIV register to monitor the HIV epidemic at
community level
Beth Harley. 604 Chezmont, Bower Road, Wynberg, 7800, South Africa
Issue: The national antenatal HIV survey shows that the Western Cape is the
province with the lowest HIV prevalence rate (3.09%) in South Africa. Whilst this
information is useful for planning at a national level, it is not helpful in planning
for service provision at local municipal clinic level.
Project: The South Peninsula Municipality is a newly restructured municipality
of the Cape Town Metrapole. It has a network of 21 clinics providing promotive/preventive and some curative services. For some years the municipal clinics
have offered free HIV counselling and testing. Everyone tested is entered into a
register along with the result, whether positive or negative. The municipal clinics
also deal with the treatment of TB/STD's and the monitoring and immunisation
of children, which results in HIV positive patients sometimes being referred to
the clinics from other health facilities for follow-up. It was decided that these
patients should also be entered into the register. Death returns of someone in
the community served by the clinic stating AIDS as a cause of death are also
entered into the register. The registers are kept strictly confidential, and reporting
of data at municipal level is confined to using age, gender and area of residence
information only. It is recognised that the register can by no means be considered
to be a complete record of the HIV epidemic in those communities.
Results: With the relatively low HIV prevalence rates in the province, the Western Cape has often been considered to not have an "HIV problem". The clinic
registers, however, indicate that whilst some areas have only a few to no HIV
positive patients recorded in the register, some areas can be considered to be
"problem" areas, with up to 122 HIV positive patients recorded. This information
has been invaluable for motivating for the expansion of the clinic services at some
of the clinics to incorporate a primary health care clinic for the staging, monitoring
and primary care treatment of HIV positive patients.
Lessons Learned: Antenatal prevalence studies are not useful for planning
HIV related health services at municipal clinic level. Whilst the registers kept at
the municipal clinics can by no means be considered to be a complete record
of the HIV epidemic, they have become an invaluable tool for revealing local
variations of HIV prevalence rates, which in turn has assisted in planning and
motivating for expansion of services at clinic level.
43466 Prevalence of HIV infection in people who were attended
in an NGO in Rosario, 1995-1997
Lucrecia Maria Corallo, G.D. Caporalini, F.M. Fernandez, N.H.I. Bertoya,
M.O. Gonzales, L.A. Fernandez. Sanchez de Thompson 72, Rosario (Code
2000), Argentina
In 1994 the decision was taken to create the Biochemistry Department in the Volunteers Against AIDS Institution, in order to offer a new service to the community:
The Ab HIV dosage in an anonymous, free and voluntary way. Since 1995 this
project has been consolidated through peniadic talks in educational institutions,
training events for health professionals and society in general, approaching the
subject of STD (Sexual Transmission Disease) and HIV prevention.
Objectives: To know the prevalence of HIV infection in population that sporadically attend the NGO, Biochemistry Department, in order to have the analysis for
AcHIV determination.
Materials and Methods: The search for Antibodies was done by 3rd Generation Elisa for the selection and Westen Blot for its confirmation in 260 people who
voluntarily attended our institution during the period December 1995-1997.
Results: 153 men and 117 women were tested, of ages ranging from 17 to
62 for men and 17 to 60 for women with an average of 27 and 26 respectively.
252 non reactive determinations and 8 / /reactive ones were obtained; the letter
belonging to the male sex.
Conclusions: There is an increase of attendants as regards consulting and
laboratory analysis If December 1996 is taken as division point, the mak attendance increases 100% and the female 124%, the latter group showing a greater

interest and commitment with their reality. Analysing the relation man/woman for
these two periods a slight decrease is noticed: 1.5 (1996) to 1.4 (1997). It is also
found that initially the attendants reported having had risky behaviour or situations
while at present there is an increase in the attendance of couples starting a new
relationship, thus showing the awareness of living a responsible sexuality.
There is an increase of attendants as regards consulting and laboratory analysis. It is also found that initially the attendants reported having had risky behaviour

or situations while at present there is an increase in the attendance of couples
starting a new relationship, thus showing the awareness of living a responsible
sexuality.
43467 Sentinel surveillance conducted within the frame of the
project of HIV prevention among IDUs in the town of
Svetlogorsk, Gomel region, Belarus
Marina Bezruchenko-Novachuk1, Vladimir Romantsov2. 117, Kirov Street (6th
Floor), 220000 Minks; 2 Svetlogorsk Project Manager, Narcologist Svetlogorsk,
Belarus
Issue: According to the Republican AIDS Prevention Centre, as at October 1,
1997, the city of Svetlogorsk had 1125 individuals registered as infected with HIV.
More than 1000 of them are drug users. The total number of IDUs in Svetlogorsk
is estimated by different experts to be between around 5 000. From January 1,
1997, the implementation of the UNAIDS/WHO/Government of Belarus Project is
under operation. The interim assessment of the project reported it to be effective.
Objectives: In September-October 1997 the sentinel surveillance, based on
the methods of laboratory HIV diagnosis by the blood residue in syringes was
conducted to determine the prevalence of HIV infection among IDUs in the town
of Svetlogorsk during implementing the UNAIDS/WHO/Government of Belarus
Project, to evaluate the progress of the epidemic as a measure of effectiveness
of the outreach prevention activities, to confirm the findings of behaviour studies
conducted in January 1997, which showed that IDUs ran a high risk of getting
infected with HIV (92%).
Method: Voluntary confidential unassociated anonymous testing of the syringe
blood residue, a sentinel group of 200 persons was selected for the sentinel area
(Svetlogorsk).
Results: The results of the studies show that the prevalence of HIV infection
among IDUs in the city of Svetlogorsk is 66.8%. The sentinel group has a high
level of HIV infection prevalence. The information obtained in the course of the
sentinel surveillance does not, of course, reflect the situation characteristic of the
entire population in the city. These findings reflect the prevalence of HIV infection
among those drug users who interact with the Project. However, the age and
sex pattern of this group corresponds to the pattern of the whole drug using
population, thus reflecting, though indirectly, the entire target group.
Conclusion: This method could be a good tool for assessing the effectiveness
of the outreach services provided to the IDUs in the harm reduction projects.
S43468 Time-dependence of predictors of death in HIV-infected
patients
Thomas Perneger1, M. Abrahamowicz2, G. Barett2, S. Yerly3. 'Inst. of Social
and Prev. Medicine, CMU 1211, Geneva 4: 3Laboratory of Virology, Geneva,
Switzerland; 2Mcgill University, Montreal Quebec, Canda
Objectives: To establish whether the predictive value of clinical predictors of
death (viremia, CD4 counts, clinical stage) remains stable or changes over time
after the baseline assessment.
Design: Modeling of survival data in a community-based cohort of HIV-infected
persons.
Methods: Baseline measurements of CD4 cells, HIV RNA, and clinical stage
(A-C) were obtained between 1991 and 1993 in 394 HIV-infected persons followed by the Swiss HIV Cohort Study. Follow-up was completed in 1996 (mean
29 months). Two types of survival models were used in this analysis: 1) the
classic proportional hazards model, which assumes that the relative hazard (RH)
remains stable over time, and 2) models that allow for changes in the RH over
time and for testing their significance (Abrahamowicz et al, J Am Stat Assoc
1996;91:1432-9).
Results: During follow-up, 169 (43%) patients died. Time-dependence patterns
varied by predictor. The RH for log (CD4 cells) and stage B remained stable over
time (both p-values for time-dependence >0.5), confirming that the classic Cox
model was appropriate. In contrast, the RH for clinical stage C decayed progressively from -30 during the first year to about 8 after three years (p = 0.03). The
opposite was seen for the RH for log (viremia), which showed a nonsignificant
trend to increase over time (p = 0.11). Finally, the RH associated with the baseline log (RNA/CD4 ratio) displayed a significant increase over time (p. 0.001).
However, further analyses suggested that the latter effect was due to the upper
quartile of the distribution of the RNA/CD4 ratio.
Conclusion: Some markers (e.g., CD4 cells) have a stable predictive value
over time. Other predictors are better at short-term (e.g., clinical stage C) or
long-term (e.g., RNA/CD4 ratio) forecasting of survival in HIV patients. These
findings have implications for patient counseling, for the optimal frequency of
patient monitoring, and for the choice of surrogate markers of clinical outcomes
in clinical trials.
[43469 A simulation model to investigate the 1996-97 decline in
AIDS incidence in the homosexual/bisexual group in

England and Wales
Daniela De Angelis', W.R. Gilks', B.G. Evans2, N.E. Day', O.N. Gill2. 1MRC
Biostatistics Unit, Cambridge; 2PHLS AIDS & STD Centre, CDSC, London, UK
Background: Latest AIDS projections for England and Wales, published in 1996,
have been accurate until the end of 1995. A substantial decline in AIDS incidence
has been observed since then, especially in the homosexual/bisexual exposure