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

942 Abstracts 43488-43491

12th World AIDS Conference

Objective: To reconstruct and forecast the spread of the HIV/AIDS epidemic in
Italy. To compare and validate two statistical methods both developed to estimate
HIV/AIDS incidence and prevalence, while focusing attention on different aspects
of the epidemic.
Methods: An Empirical Bayesian Back-Calculation (EB-BC), and the Mover
Stayer (M-S) model, a stochastic dynamic compartmental model, have both been
applied to incident AIDS cases reported up to September 1997 and diagnosed
until March 1997 among adults in Italy. In order to obtain comparable results, the
two methods were applied using homogeneous and updated input parameters
describing the natural history and the surveillance aspects of the epidemic in
Italy.
Results: Through the M-S model the cumulative incidence of HIV infections
and AIDS cases until the end of 1997 were estimated, respectively at 103,965
and 46,380, and the estimated HIV infection curve peaked in 1988 with 14,942
infections. For the same period the EB-BC provided estimates of the cumulative
incidences of HIV infection and AIDS diagnosis respectively as 133,087 and
39,819, and the estimated peak of the HIV infection curve is located in 1986
with 18,165 infections. Moreover the EB-BC suggested that the HIV epidemic
reached its maximum spread between 1986 and 1989 in each transmission category, except for heterosexually-infected women among whom, the estimated HIV
incidence was still increasing at the end of the study period.
Conclusions: The consistency of the results suggests that the two methods,
while focusing on different aspects of the epidemic, are both able to correctly
estimate the HIV/AIDS spread. The synergic use of the EB-BC and of the M-S
model is therefore encouraged since, by providing reliable HIV/AIDS stage-specific estimates by transmission category, can be useful for targeting and planning
the prevention interventions, and provide a valuable source of information for
developing and evaluating the effectiveness of therapies.
S43488 Estimated future economic impact of HIV/AIDS in the
European Union study for the EU Concerted Action on
Multinational AIDS Scenarios
Maarten Postmal, E.J. Beck2, R.M. Leidl3, K. Tolley4, J.C. Jager1. 1National
Institute of public health and the Environment, PO Box 1; 3720 BA Bilthoven,
The Netherlands; 2St Mary's Hospital London; 3University of Nottingham,
Nottingham, UK; 4 University of Ulm, Ulm, Germany
Objective: To project future needs of hospital care for HIV/AIDS in the European
Union (EU). Data National studies on the utilization of hospital care for AIDS
patients were selected from France (YA Flori), Greece (H Kornarou), Italy (AM
Tramarin), the Netherlands (MGW Dijkgraaf), Spain (F Antohanzas) and the UK
(EJ Beck). Utilization figures referred to the period 1990-94. Baseline hospital
services utilization was estimated for inpatient days, outpatient visits and day-care
days.
Methods Baseline hospital care utilization was expressed per person-year and
severity stage (chronic versus terminal hospital care). Cross-country differences
in baseline utilization were analyzed with respect to differences in health-care
systems in EU-countries (for example, per capita numbers of beds, doctors and
nurses). Multinational scenario analysis was used for linking baseline figures
with standardardized epidemiological HIV/AIDS-modelling, assessing both lifetime and annual hospital care needs. The reference scenario shows a future
assuming no major changes in epidemiology, treatment and care. Alternative
scenarios show the impact of prevention and new HIV/AIDS therapies, such as
combination triple-therapy.
Results: Highest per capita hospital inpatient days for AIDS were found for
France, highest outpatient contacts per capita for Italy. By the end of the century annual hospital bed needs for HIV/AIDS in the whole EU will be inbetween
minus 30% (HIV-therapy scenario, for example, combination triple-therapy) and
plus 60% (AIDS-therapy scenario) of estimated 8100 beds for 1995. Outpatient
hospital care was projected to grow faster than inpatient in all scenarios.
Conclusions: Differences in baseline hospital care utilization appear to be in
line with differences in health-care systems across EU-countries. Projected annual hospital beds for AIDS by the end of the century might represent up to 0.65%
of all hospital beds available in the EU. Scenarios on the impact of combination
triple-therapy indicate significant-though partially transient-decreases in annual
hospital inpatient care, whereas lifetime needs might further increase.
EU Concerted Action on Multinational AIDS Scenarios (Contract Number
BMH1-CT-941723)
43489 Estimation and projection of HIV/AIDS in Vietnam,
1996-2000
Tran Hien Nguyen1, Hoang Thuy Long2, Pham Kim Chi2. 1Department of
Epidemiology Hanoi Medical College 1 Ton That Tung Hanoi; 2National Institute
Hygiene Epidemiology 1 Yersin Hanoi, Vietnam
Objective: to estimate HIV/AIDS in 1996 and project for the year 2000 and to
identify the reported/estimated ratio of HIV/AIDS in Vietnam.
Design: Retrospective study

Methods: Direct method was used to estimate HIV infections in each out of 61
provinces in Vietnam in 1996 using available data about the sizes of populations
(IDUs, CSWs, STDs, adult aged 15-49) and HIV seroprevalence in these populations. Projection for the year 2000 was made using EPIMODEL, version 2.01
with assumtions that the year when extensive spread of HIV transmission started
was 1988; estimated number of HIV infection in 1996 in three senarios: 22,000;
25,000 and 28,000; the position on the HIV epidemic curve of in 1996 is 152, the

shape of the HIV prevalence curve is gamma 8 and the mean progression rate
from HIV to AIDS in Vietnam is 6 years
Results: By the end of 1996, 22,000-30,000 HIV infections 1,700-2,300
AIDS cases, and 1,200-1,600 died of AIDS were estimated cumulatively. The
male/female ratio is 2.0-2.4. The reported/estimated HIV cases ratio is 19%-26%.
The reported/estimated AIDS cases ratio is 34%-48%. In 1997, 38,000 to 52,000
HIV infections were estimated cumulatively, among them, 3,700-5,000 developed
AIDS and 2,700-3,700 died of AIDS. By the year 2,000, the cumulative number
of HIV infections will reach about 130,000-180,000, among them, 22,000-30,000
will develop AIDS and 17,000-24,000 die of AIDS.
Conclusion: The official number of HIV/AIDS cases in Vietnam were underreported, only about one fifth for HIV infection and one third for AIDS cases were
reported. The estimate of HIV infection to the year 2000 will be three times higher
than that in 1997.
43490 The impact of partner referral of STD patients on the
spread of HIV in Nairobi: Evaluation with the model
STDSIM
J. Dik F. Habbema1, J.O. Ndinya-Achola2, E.I. Meester3, C. Van Vliet3,
C.P.B. Van der Ploeg3. 1Dr. Molewaterplein 50 3015 GE Rotterdam;
3Department of Public Health, Erasmus University of Rotterdam, Rotterdam,
The Netherlands; 2Dept. of Medical Microbiology, University of Nairobi, Nairobi
Kenya
Background: An important strategy to control HIV is to reduce the spread of classical STDs in the population. Partner referral may contribute to this by reducing
the risk of reinfection of cured STD cases by (asymptomatic) infected partners.
The impact of partner referral of STD patients on HIV in the setting of Nairobi was
evaluated using the model STDSIM.
Methods: The microsimulation model STDSIM was quantified using data on
demography, sexual behaviour and health care from Nairobi. Projections were
done to evaluate the effectiveness of partner referral among STD cases attending
health care facilities. The effect of referring only steady partners of an STD patient
was compared to the effect of referring both steady and non-commercial casual
partners.
Evaluations were performed for various fractions of partners referred. Different
strategies of focusing partner referral were explored: complete partner tracing
of a number of STD patients vs. incomplete partner tracing of more patients.
Outcomes are population prevalence and incidence of classical STDs and HIV.
Results: In the Nairobi context, referral of only steady partners of STD patients
proved to be rather ineffective in reducing population prevalences and incidences
of HIV. Tracing of casual partners besides steady partners resulted in more pronounced reductions of STDs and HIV, even with small fractions of partners being
referred. Complete partner tracing and treatment of a number of STD patients
was projected to be more effective than treating only some partners of more
patients.
Conclusion: The STDSIM model projects that encouragement of partner referral of only the steady partners of STD patients is hardly effective in reducing
HIV in the setting of Nairobi. The main reason is that commercial sex plays a
pivotal role in the spread of HIV in Nairobi. However, when tracing and treating all
partners of STD patients is strived for, partner referral can significantly contribute
to reducing HIV prevalence and incidence.
43491 Changing pattern of HIV infection in some risk groups:
1987-1996, West Bengal, India
Subir Kumar Dey1, N.K.P. Pal2, N.B. Bhattacharjee2, G.P. Pal3. IP-41/1 Natabr
Pal Road, City -Howrah State West Bengal, Pincode-7111105; 2School of
Tropical Medicine Calcutta WB; 3North Bengal Medical College Sushrutanagar
WB, India
Objectives: To analyse HIV prevalence among some high risk groups and antenatal mothers (ANM) in and around Calcutta over a 10 year period and also
compare with high risk groups combined (HRGC) in rural North Bengal Region
(NBR) around Siliguri, Long 89' East and Lat 27 North, 579 Km. away.
Methods: At inception in m87'87, HIV Ab screening in our state were done
in certain high risk groups like CSWs (mainly from 3 redlight areas of Calcutta),
STD patients and IDUs in Calcutta. The changing pattern of HIV sero+vity rates
in these groups were studied from '87 to '96 along with groups not supposed to
be at risk, like ANM in Calcutta. The data were compared to a rural surveillance
in '96 and '97 among HRGC in NBR. Tests done -Sera of the subjects were
screened for HIV Ab by ELISA/SPOT and confirmed by W.B.Results: Calcutta
Yearwise Seropositivity rate percent

Group
CSWs
STD pts.
IDUs '87-'91
ANMs '87-'91

~

'87-'90 1991
0 0.53
0.01 0.08
- 1.29
- 0

1992
0.72
0.13
1.51
0

~

1993
1.41
0.49
0
0.68

1994
1.57
0.58
1.15
0

~

1995
3.56
0.86
1.65
0

~

1996
11.27
2.13
3.52
0

Cumulative Prevalence
In NBR among HRGC it is 9.6% in '96 & 10% in '97. Main route of HIV
transmission is heterosexual in our state, unlike Manipur.
Conclusion: Rising pattern among CSWs and STD pts calls for more effective measures. High rates even in rural areas demands urgent intervention
programmes, hitherto localised in urban settings '97 Interim results of sentinel