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

XIV International AIDS Conference

Abstracts WeOrC1272-WeOrC1310 93

Presenting author: Leng Bun Hor, #170, Preah Sihanouk Blvd., Phnom Penh,
Cambodia, Tel.: [phone], Fax: [phone], E-mail: Ihor@bigpond.
com.kh
WeOrC1272 Declining sexually transmitted disease and HIV
prevalences among antenatal clinic (ANC)
attenders in Nairobi, Kenya, from 1992-2001
S. Moses1, E.N. Ngugi2, A. Costigan3, C. Kariuki3, FA. Plummer1.
I Department of Medical Microbiology, University of Manitoba, Department of
Medical Microbiology, University of Manitoba, Room 503, 730 William Avenue,
Winnipeg, MB R3E OW3, Canada; 2Department of Community Health, University
of Nairobi, Nairobi, Kenya; 3Department of Medical Microbiology, University of
Nairobi, Nairobi, Kenya
Background: An HIV/STD surveillance program was established in 3 health centres to monitor the impact of an intervention program in the clinics and their catchment areas consisting of: 1) improved primary level STD services; and 2) community HIV/STD prevention programs for female sex workers.
Methods: Eight survey rounds were completed from 1992-2001 with samples of
about 200-300 female ANC attenders per centre per round. A questionnaire was
administered, cervical specimens were collected for gonorrheal culture, and blood
was drawn for syphilis, HIV & chlamydia serology.
Results: Trends in STD/HIV prevalence are shown. Declines are statistically significant (P < 0.001) for gonorrhea, chlamydia and syphilis, and for HIV in women
under 20 from Surveys 2-8.
Survey Gonorrhea Chlamydia Syphilis HIV (all) HIV (<20 years)
(%) (%(% )%) (%) (%)
1 (3/92-9/92) 4.8 (32/667) 31.2 (104/333) 4.9 (33/668) 14.4 (96/668) 16.3 (26/160)
2 (10/92-10/93) 4.0 (34/853) 30.5 (168/551) 5.1 (44/858) 19.0 (163/859) 21.1 (40/190)
3 (11/93-11/94) 3.3 (30/902) 20.3 (156/767) 4.2 (38/909) 17.5 (159/910) 20.1 (42/209)
4 (12/94-3/96) 3.6 (32/894) 21.8 (104/777) 3.9 (35/895) 17.3 (155/895) 14.0 (26/186)
5 (5/96-4/97) 3.9 (35/906) unavailable 3.7 (34/912) 18.0 (164/913) 15.1 (36/239)
6 (2/99-9/99) 1.5 (7/477) 15.7 (64/407) 2.5 (12/479) 19.6 (94/480) 16.2 (18/111)
7 (5/00-12/00) 1.3 (12/916) pending 1.6 (15/916) 18.4 (169/920) 11.4 (23/201)
8 (5/01-10/01) 1.2 (11/886) pending 1.7 (15/885) 14.7 (130/887) 8.2 (20/244)
Behavioural changes were also observed, including significant declines in women
reporting selling sex, declines in numbers of reported sex partners, and increases
in condom knowledge and use.
Conclusion: The prevalences of the "conventional" STDs, and HIV infection in
young women have declined, probably due in part to these interventions.
Presenting author: Stephen Moses, Department of Medical Microbiology, University of Manitoba, Room 503, 730 William Avenue, Winnipeg, MB R3E 0W3,
Canada, Tel.: [phone], Fax: [phone], E-mail: smoses@cc.
umanitoba.ca
WeOrC1308 Proposals for estimating HIV incidence in the
United States from HIV reporting data and the
detuned assay
J. Karon1, E.H. Kaplan2, R. Brookmeyer3. 1Division of HIV/AIDS, Surveillance &
Epidemiology, Centers for Disease Control and Prevention, Atlanta, Georgia,
United States; 2Yale School of Management & Department of Epidemiology and
Public Health, Yale School of Medicine, New Haven, CT United States;
3Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health,
Baltimore, MD, United States
Background: Development of the detuned assay has led the US Centers for
Disease Control and Prevention to consider new approaches for estimating HIV
incidence in the
United States from new HIV diagnoses reported to HIV surveillance.
Methods: Two statistical models have been developed to account for the fact that
individuals choose whether (and how frequently) to seek HIV testing, that test results are only reported if HIV+ on the sensitive antibody test, and that additional
information on prior (HIV-negative) testing dates can be obtained via surveys.
Both models account for heterogeneity in testing frequency and assign a weight
to each detected recently infected person. Model II uses detuned assay results;
Model I also uses time since last negative test for some seroconverters, regardless of detuned assay results. The performance of these models was assessed
by simulation.
Results: Assuming that persons test twice per year on average, 100 simulations
gave the following results for 1-year incidence estimates and standard deviations
of the estimates:
Mean Seroconversions Model 1 Mean (StDev) Model 2 Mean (StDev)
100.9 82.0 (18.0) 115.9 (16.3)
247.2 203.6 (38.3) 285.7 (25.4)

Simulations presuming that prior test dates would not be available for 25% of the
infected persons revealed that while Model I remained accurate, Model II became
badly biased.
Conclusions: These analyses show that it may be feasible to estimate HIV incidence from statistical models based on data from HIV surveillance and detuned
assay results.

Further refinements that address issues such as non-reporting bias (due to those
at risk who do not seek HIV testing), samples that are not tested using the detuned assay, and changes in HIV test-seeking behavior leading to testing must be
investigated before a final system for estimating HIV incidence can be designed
and implemented.
Presenting author: Edward Kaplan, Yale School of Management, Box 208200,
New Haven, Connecticut 06520-8200, United States, Tel.: [phone], Fax:
[phone], E-mail: [email protected]
WeOrC1309 Modelling the spread of HIV-1: the basis for the
2001 UNAIDS/WHO country specific estimates and
projections of adult HIV prevalence
N.C. Grassly1, G.P. Garnett1, U.N. AIDS Epidemiology Reference Group2.
1Imperial College, London, United Kingdom; 2 UNAIDS, Geneva, Switzerland
Background: The country-specific estimates of HIV-1 prevalence and AIDS mortality produced by the Joint United Nations Programme on HIV/AIDS (UNAIDS)
are widely used for both planning and advocacy purposes. To date, UNAIDS have
used a simple curve fitting procedure, based on a gamma distribution of HIV incidence. However, this method lacks flexibility, is not designed to make predictions,
and does not describe the mechanisms underlying the spread of infection. In interpreting the progress of the epidemic a transparent and flexible model is required
to represent the course of the epidemic and to make short-term projections.
Methods: A simple, flexible and easily comprehended model, which can be used
globally and fitted to sentinel surveillance data was developed in a collaboration
between UNAIDS and an advisory group. Prevalence is modelled using an intuitive epidemiological description including a start date, an initial rate of effective
contacts, a proportion of the population at risk and a parameter representing the
behavioural response. These 4 parameters are fitted while externally supplied
values are used for background rates of fertility and mortality and progress from
HIV infection to AIDS.
Results: The model has been successfully applied for a large number of epidemics from various countries and risk groups within countries. Confidence in
the validity of the epidemic curve depends upon the availability of seroprevalence
data. Through maximum likelihood methods short term projections of the future
course of the epidemic can also be generated.
Conclusions: Through the integration of a number of different modelling approaches a simple and universal model for the HIV epidemic has been generated
which has been applied to the country specific estimates of HIV prevalence. Future developments of this model should include more emphasis on the impact of
interventions and on methods of short term projection.
Presenting author: Geoffrey Garnett, Dept Infectious Disaese Epidemiology, Imperial College, St Mary's Campus, Norfolk Place, London, United Kingdom, Tel.:
[phone], Fax: [phone], E-mail: [email protected]
WeOrC 310 Modelling the spread of HIV infection in four cities
of sub-Saharan Africa
B. Auvert1, B. Ferry2. 1lnserm U88, Inserm U88, 14, rue du Val d'Osnee, 94415
Saint-Maurice Cedex, France; 2IRD, Paris, France
Backgrounds: The rate of spread of HIV is heterogeneous in sub-Sahran Africa.
HIV prevalence among the general population varies from less than 2% to more
than 30%. This heterogeneity is not yet understood. The multicenter study carried out in four African cities has provided new epidemiological information. The
objective of the current study was to use these new data to identify the potential
mechanism(s) responsible for this heterogeneity
Methods: A stochastic simulation model (Simul-AIDS) was used. This model allowed a detail description of demographic features, sexual behavior and a generic
STI. In this model male circumcision reduced the female to male transmission of
HIV and of the generic STI. HIV transmission was facilitated by the presence of
the generic STI. In a first analysis, factors having an strong impact on the spread
of HIV were identified. In a second analysis, numerical values for these factors
were chosen accordingly to the data of the multicenter study.
Results: Heterogeneity in rate of spread of HIV was obtained by varying independently sexual behavior, probability of sexual transmission of the generic STI
or of HIV. When using data from the multicenter study, male circumcision (MC)
was the only factor able to reproduce the heterogeneous rate of spread of HIV. In
particular, the model was able to reproduce the situation of Yaounde (Cameroon)
where the sexual activity is higher and the HIV spread is lower than in Kisumu
(Kenya) and in Ndola (Zambia). The model was also able to reproduce explosive
epidemic in situation where the men are not circumcised.
Conclusions: Modelling of the spread of HIV in sub-Saharan Africa should carefully take into account the factors other than sexual behavior and sexual mixing in
order to reproduce the reality. The interaction between male circumcision, other
STIs and HIV is an important mechanism to understand the spread of HIV in
sub-Saharan Africa.

Presenting author: Bertran Auvert, Inserm U88, 14, rue du Val d'Osnee, 94415
Saint-Maurice Cedex, France, Tel.: [phone], Fax: [phone],
E-mail: [email protected]