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

12th World AIDS Conference

Abstracts 43492-43496 943

surveillance shows rate of around 0.5% in ANMs signifying that HIV has spilled
over in general pop'n as well.
43492 Cost-effectiveness of antiretroviral therapy vs. other HIV
control options in developing countries
J.Dik F. Habbema, C Van Villet, EL Korenromp. Dept. of Public Health, Erasmus
University Rotterdam PO. Box 1738, 3000 DR Rotterdam, The Netherlands
Objective: To compare the health effects and cost-effectiveness (CE) of modern
antiretroviral HIV therapy (ART) with other HIV control measures, under different
patterns of HIV spread.
Methods: The impact and costs of ART and several strategies of condom
promotion and STD treatment were evaluated with the microsimulation model
STDSIM. In a best case scenario, ART was assumed to be received by all AIDS
patients, to prolong life expectancy of symptomatic patients from 1 to 5 years,
and to reduce HIV infectivity to the level of asymptomatic patients for the rest
of life. We did not account for the probably big organizational, compliance and
viral resistance problems with ART. Under four patterns of sexual behaviour,
each giving rise to a different form of the AIDS-epidemic, the effects on HIV
incidence and AIDS mortality and the CE of the control options were calculated
and compared.
Results: At the studied intensities of the control options, increased condom
use in commercial and casual contacts reduced HIV incidence and mortality best.
The projected health impact of ART in this best case scenario was less than
that of targeted condom promotion, but more than that of untargeted improved
STD treatment or increased condom use in stable relationships. Within ranges
of plausible costs, ART proved far less cost-effective in reducing HIV incidence
than the other control measures studied.
Conclusions: The STDSIM model predicts that, in a best case scenario regarding the effectiveness, coverage, compliance and costs of ART, the public
health effects of ART are less to those of targeted condom promotion. In terms of
CE, ART is far inferior to all HIV control options studied, and to many other health
measures yet not fully implemented in resource-poor countries. This conclusion
holds for all patterns of HIV spread studied, and will hold even if costs of ART fall
dramatically.
43493 The demographic impact of AIDS: Projecting the impact
of AIDS on mortality and other demographic measures
Peter Way', J. Stover2. 'International Programs, US Census Bureau
Washington, DC 20233; 2The Futures Group International Glastonbury, CT USA
Objectives: 1) To review national level projected impacts of AIDS on mortality
and life expectancy prepared by different forecasting groups and explaining the
reasons for the differences; 2) to assess which indicators of mortality have the
most stable and easily described relationship to HIV; 3) to assess the sensitivity
of projections to alternative inputs; and 4) to present a methodology for making
estimates of mortality based on a few key assumptions.
Methods: We review population projections prepared by the United Nations
Population Division and the US Bureau of the Census for those countries most
affected by HIV/AIDS. Using demographic analysis techniques, we identify differences in methods and assumptions that explain the differences in the results
of those projections. We examine alternative scenarios of the potential negative
effect of HIV on fertility, as described in several recent community studies, using
the two projection approaches, and explore its potential effect of future population growth. We explore the impact of AIDS mortality on various demographic
measures of mortality, including the impact on infant, child, and adult mortality,
the life expectancy at birth, and the crude death rate. We discuss the utility and
sensistivity of these measures in reflecting the additional mortality resulting from
AIDS, and make recommendations as to the most useful such measures. Finally,
we develop a model for estimating the impact of AIDS on mortality based on a
limited number of demographic and epidemiological parameters.
Results: Alternative projections by the UN and the US. Census Bureau differ
dramatically, but can be attributed primarily to differences in initial epidemic levels
and the future course of the epidemic rather than to differences in demographic
modeling approaches. Life expectancy at birth and the probability of dying during
the productive adult years (45q15) are probably the most sensitive demographic
indicators of excess AIDS mortality. Levels of infant and child mortality will increase due to AIDS but will also be affected by underlying health conditions and
trends. The potential impact of HIV on fertility could have an effect on future
population size and growth, but its impact is far less than the impact of HIV/AIDS
on mortality. The model presented allows the estimation of life expectancy at birth
based on four variables which are easily estimated for any country.
Conclusions: AIDS is substantially reducing levels of life expectancy in many
countries. National levels of HIV infection of 10 percent or less can result in a
doubling in the number of deaths and the crude death rate and a reduction of
20 years in life expectancy. Demographic and epidemiological modeling efforts
are essential in estimating the current and future impact of AIDS since empirical
measures will be lacking in many countries. Empirical studies are still needed to

provide information on the impact of HIV on fertility, survival time after infection
for adults and children, and other epidemiological parameters.

S43494 Using morbidity and mortality data to monitor the impact
of the epidemic
Peter Godwin, S.C. Chawla', E. Eliot3, W. Im-Em4, I. Koithra2. L3/7 DLF Phase
2, Qutab Enclave, Gurgaon; 1Lady Hardinge Medical College, New Delhi;
2 Vishwa Yuvak Kendra, New Delhi, India; 3University of Rouen, Le Havre,
France; 4Mahidol University Bangkok, Thailand
The Issue: New challenges now facing health and social welfare policy makers
and programme designers in Asia are: how best to respond to the increases costs
associated with the unprecedented increase in young adult morbidity and mortality arising from the spread of HIV infection; and the most effective institutional
framework within which to structure these responses. These challenges are enhanced by the "invisibility" of the epidemic to many policy-makers: those infected
with HIV are not easily identified, and many die before actually developing AIDS.
In addition, descriptions of the epidemic rely heavily on modelling, "projections",
and estimates of numbers, rather than on real data.
The Activity: We suggest a methodology for monitoring changing morbidity and
mortality patterns, and review several new, unpublished pilot studies of changes
in morbidity and mortality among adults in three cities in India, and one province
in Northern Thailand. We suggest that this represents a simple and effective way
of monitoring what is really happening with the epidemic, and of planning for
responses.
Results: The pilot studies show that this kind of data is fairly easily available;
and the studies themselves suggest that prime age adult mortality levels are
reaching, or overtaking, child and infant mortality levels in these areas. The
studies suggest an alarming tendency for HIV to be more seriously affecting
adolescent populations than has previously been realised; and hold the potential
to help better understand the duration and pattern of HIV- and AIDS-related
morbidity and mortality.
Conclusions: This kind of monitoring and analysis can help in two particularly
important areas: better understanding of the care needs of those infected and
affected, and cost-effective responses to these needs; and better analysis for
health and welfare systems to anticipate changing demands based on these
needs.
S43495 1Estimates and projections of mother to infant HIV
infection incidence, Rio de Janeiro State, Brazil,
1995-2000
Antonio Costal, J.M.A. Carvalho2, AT. Maciel2, J. Chin3. 'Ladeira Do Castro 123
Apt o 101 2023 0 - 030, Rio de Janeiro, RJ; 2Universidae Federal Do Rio De
Janeiro, Riso De Janeiro, RJ, Brazil; 3University of California at Berkeley,
Berkeley CA, USA
Background: Reported mother to infant transmitted HIV/AIDS cases have increased noticeably since early 1990's in the State of Rio de Janeiro, Brazil.
A better understanding of regional patterns of the main factors underlying HIV
mother to infant transmission is needed, in order to provide adequate preventive
measures.
Objectives: To analyse the role of varying regional adult HIV transmission and
fertility patterns in the projected incidence of mother to infant transmitted HIV
infections in Rio de Janeiro State, from 1995 until 2000.
Methods: A primary hypothetical HIV scenario was developed using EPIMODEL, based on the following assumptions: The HIV epidemic started in 1980;
an estimated HIV prevalence of 50,000 HIV adults in 1995; HIV incidence followed
a gamma 5 curve, peaking in 1995; median progression rates from adult HIV
infection to AIDS and from AIDS to death were 10 years and 1 year, respectively;
mean HIV mother to infant transmission rate was 0.3. Regional age-specific fertility rates and adult female HIV prevalence estimates from 1991 to 1994 (region
1: Rio de Janeiro City; region 2: other counties of the metropolitan area; region
3: inner country towns and rural areas) were added to the primary scenario, providing estimates and projections of annual and cumulative incidence of mother to
infant transmitted HIV infections up to year 2000.
Results: Projected mother to infant transmitted HIV infection cumulative incidence for the years 1995-2000 were 1.8 and 1.2 times higher in region 3 (1,400),
when compared to regions 1 (800) and 2 (1,100), respectively. Total fertility rates
were 1.4 and 1.3 times higher in region 3 (2.73), compared to regions 1 (2.0) and 2
(2.1). The highest estimated adult female HIV prevalence, based on surveillance
data, was associated with region 2.
Conclusion: Higher age specific fertility rates observed in the period 1991-94
make populations living in the inner country towns and rural areas of the State of
Rio de Janeiro, Brazil, more vulnerable to the occurrence of mother to infant transmitted HIV infections. Presumably higher adult female HIV prevalence estimates
also explain more intensive patterns of projected mother to infant transmitted HIV
infection incidence outside the city of Rio de Janeiro, up to year 2000.
43496 HIV/AIDS prevention and awareness intervention in Lira,
Uganda

Marc Sam Opollo', Jimmy A.A. Okello2, Lino Atum2, Betty Atoo2. AIDS CAPINU
c/o Box 10446, Kampala; 2Lira, Uganda
Issue: There is need to make a background survey to asses the needs, risks and
appropriate intervention strategies of the youth at risk of infection with HIV/AIDS
& Sexualy Transmited Diseases due to risky behaviours.
Project: AIDS Care of Prevention Initiative in Northern Uganda (AIDS CAPINU)
provides information, counselling and sensitisation services on HIV/AIDS and dis