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

12th World AIDS Conference

Abstracts 43454-43459 935

Conclusions: The Kenya National AIDS and STDs Control Program has developed an effective sentinel surveillance system. The results have been used to
target preventive services appropriate for each region. Seven-year trends indicate
that prevalence may have reached a plateau level in some urban areas, whereas
some rural areas are still experiencing rapid growth.
43454 The impacts of a laboratory reporting system on
surveillance of HIV infection
Teresa Man Yan Choi, S.S. Lee. AIDS Unit 5/F Yaumatei JCC 145 Battery
Street, Yaumatei Kowloon, Hong Kong
Objectives: To determine the impacts of laboratory reporting on HIV surveillance by evaluating the extent of duplicated, under and delayed reporting of the
anonymous voluntary clinical HIV reporting.
Methods: The HIV infections diagnosed in three major HIV confirmation laboratories from Jan 1, 1995 to Sep 30, 1997 were evaluated against the anonymous
physician voluntary reporting forms. The date of HIV confirmation diagnosis was
used as the unique identifier for record matching.
Results: A total of 387 HIV infections were diagnosed by the laboratories
within the study period. For these cases, 38 physician reports were obtained
directly from the government HIV clinic whereas 318 reports were submitted by
other medical doctors. Case matching was done on a quarterly basis. Eighty-two
laboratory reported HIV cases remained unidentified while 51 physician reports
were duplicated. An overall 78.8% were reported by both systems. Missing age in
laboratory reports versus physician reports was 4.1% to 0.3%, missing ethnicity
5% to 0.9%, and routes of transmission 9.8% to 3.6%. Clinical information and
date of birth were not available in the laboratory reports. Non-reporting was associated with diagnosis made in the private sector (OR 7.56, 95% Cl 4.27-13.41).
Ninety-one percent of all physician reports were submitted within 3 months of
laboratory diagnosis. The annual increase in the diagnosis of HIV infection was
45% (95% Cl: 25-67%) as basing on the physician reports which was higher
compared to the 25% (95% Cl 10-42%) modeled using the laboratory data.
Conclusion: Laboratory reporting has strongly enhanced the accuracy of the
HIV surveillance basing on anonymous voluntary reporting system through minimizing (1) double counting through case matching using date of confirmation
diagnosis as the unique identifier; and (2) under-reporting in the private sector.
143455 Epidemiological features and trends of AIDS cases
among female adolescents
Katia R.B. Sanches', A.H. Matida2, D.R.F. Pires2. 1Federal University of Rio De
Janeiro; R Henrique Cordeiro, Rio De Janeiro; 2Health Department Rio De
Janeiro State, RJ, Brazil
Objectives: 1- To exam the epidemiological features of AIDS among female adolescents in Rio de Janeiro state. 2- To provide information for future prevention
programs.
Methods: It was analyzed through the AIDS information system, all cases
among adolescents and young adults (15 to 29 years) reported to the AIDS
Department of Rio de Janeiro state, from 1984 to December, 1997. The group
selected included 29 years old and younger considering the prolonged period
between HIV infection and the development of major symptoms.
Results: By December, 1997, 18,610 cases of AIDS had been reported in
Rio de Janeiro (14,779 males and 3,850 females). Of these, 4,637 (24.5%) were
among adolescents or young adults aged 15-29 years (3527 males and 1110
females). Of all women with AIDS, 28.8% were diagnosed between 15-29 years
of age compared with 23.9% of males (p < 0.001). The distribution according
to exposure categories showed that 592 (53.3%) were heterosexual; 109 (9.8%)
were injecting drug users, 82 (7.4%) were blood transfusion recipients and 327
(29.5%) the exposure category was unknown. Among women injecting drug
users with AIDS, 42.2% were diagnosed between 15-29 years of age. In no
other behavior-related exposure category was the proportion seemingly infected
in adolescence so high.
Conclusion: In the state of Rio de Janeiro, females with AIDS were more likely
to have been infected during adolescence than males. Almost half the women
injecting drug users with AIDS appear to have been infected in adolescence.
These findings have important policy implications for HIV prevention among
adolescents.
43456 Preliminary findings of behavior sentinel surveillance for
HIV risk behaviors - 1997
Myint Zaw, Rai Mra, HIa Htut Lwin, Khin Ohnmar San, Edward Zan. National
AIDS Program, 36 Theinbyu Street, Yangon, Myanmar
The National AIDS Program implemented the HIV Sentinel Surveillance System starting 1992 March and have finished 11 rounds in 5 years. Different HIV
prevalence rates were detected among different risk groups. We can map out
epidemiological pattern of HIV by geographic areas, populations groups and by

time trend and use them to define problem areas and projection land estimation
of HIV/AIDS burden in the country by using these surveillance figure. But we still
need to know what BEHAVIOUR and risk are causing factors for these changes
in HIV trends and patterns. The distinctive feature of behavioral surveillance is
that it can greatly enrich and enhance the epidemiological sentinel surveillance
both qualitatively and quantitatively. The method is cost effective and has the
ability to find out changes in risk BEHAVIOUR patterns otherwise imperceptible

by conventional KAP surveys. The behavioral surveillance was pilot tested in
September and October 1997, in Yangon and Mandalay with all the risk groups.
Questions regarding the pre or extra marital sexual practices, drug BEHAVIOUR
and experiences on STDs were put in the questionnaires and the participants had
to respond them in a confidential manner. Behavioral factors relevant to HIV/AIDS
transmission (i.e. multiple sex partners, condom use, history of STD, injecting
drug use) were obtained from male STD, female STD, CSW, IDUs, and pregnant
women from Mandalay and Yangon cities (n = 2027), Pertinent findings were that
only 4% of male STD, 5% of female STD and 1% of pregnant women were using
condoms. Among military recruits of Yangon and Mandalay, 25% had sexual
experience, 2.0% had extra marital sex and only 0.8% had used condoms persistently. Fourteen percents of IDU also had multiple sex partners. Even in CSW
only 33% used condoms. We need to have more in-depth studies, aggressive
health education campaigns and condom promotion programs targeted towards
the high risk populations and their HIV/AIDS related behaviors.
43457 Investigation of AIDS cases through coroner services in
Sao Paulo
I.A. Guibu', D.D. Barbieri2, C. Pasqualucci3, M.C. Palhares2, N.J.S. Santos2.
'Rua Antonio Carlos 122 D1309-010, Sao Paulo; 2Health Department of Sao
Paulo State Sao Paulo SP; 3University of Sao Paulo, Sao Paulo SR Brazil
Objective: To analyse deaths having AIDS as suspect cause by using serology
and necropsy.
Design: Cross-sectional.
Methods: Blood samples of all suspect AIDS cases arriving at the Sao Paulo
city coroner service were drawn from August to December 1997. The Sao Paulo
coroner service receives non-violent death cases of unknown cause, especially
those that occur up to 72 hours of inpatient care. All necropsy reports, and Elisa
and Immunofluorescence test results were analysed to confirm AIDS or not. All
confirmed cases were looked up in the State data system to verify if they had
been reported.
Results: Approximately 76% of confirmed AIDS cases among men, and 86%
of cases among women had not been previously reported to the disease surveillance system. Approximately 70% of suspect AIDS cases among men and 57%
cases among women were confirmed by serologic tests.
Conclusion: Under reporting of AIDS was found to be quite high among cases
examined at the coroner service. Data from necropsy reports may become an
important source of additional information to the State AIDS disease surveillance
system.
43458 Investigation of cases and deaths by AIDS from death
certificates
Deborah Barbieri, N.J.S. Santos, I.A. Guibo, L.F. Jamal, S. Caminada,
C.S.B. Domingues. Rua Antonio Carlos 122 - Sao Paulo 01309010, Health of
Sao Paulo State Sao Paulo SR Brazil
Objectives: improve statistical data referring to mortality rates by obtaining AIDS
cases through death certificate.
Design: cross-sectional
Methods: all death certificates of the State of SP (except S.P city) filed at
SEADE Foundation from the year of 1995 were examined to see if AIDS was
referred as a cause of death.
Results: approximately 100.000 death certificates were examined, 3170 of
which had AIDS as basic cause. The State Disease Surveillance Service did not
know of 971 cases. The inclusion of these cases represent an additional 11.6%
cases to those reported during the period. Although 722 were known AIDS cases,
the information on the death of the patient had not reached the system. These
cases represent under reporting of 12.2% of the deaths by AIDS reported in 1995.
Conclusion: death by AIDS was found in 205 (32.3%) cities of the state,
showing the growth of the epidemic towards the interior of the states. Death
certificates are an important source of information to decrease under reporting of
cases and deaths due to AIDS. Considering the 5903 deaths reported in 1995
plus the 1693 cases found by death certificates, the total number of death due to
AIDS in the state of Sao Paulo would increase to 7596.
43459 HIV prevalence trends among intravenous drug users in
Yunnan Province between 1992-1997
Jiapeng Zhang. Yunnan Provincial Health and Anti-Epidemic Centre, Kunming,
Yunnan, 650022, China, China
Background: Sentinel surveillance was conducted to observe variation in HIV
prevalence and transmission speed among IVDUs at different time and in different
districts.
Methods: HIV prevalence among IVDUs at 11 sentinel sites distributed in 11
county (city) treatment centers for drug users located in west, south-west,south
and central Yunnan Province was observed systematicly once a year from year

1992 to 1997. Lot Quality Assurance Sampling was used at each site and started
in Sept. 1991. Samples for HIV antibody test were collected successively each
round until reaching the sample size.
Results: HIV prevalence among IVDUs at 11 sites can be divided into three
types: rapidly increasing type, gradually increasing type and stable high prevalence type. The HIV prevalence among IVDUs at each site over the years was:
1. Rapidly increasing type (6 sites): Kunming city (1992-1997): 0; 0; 0; 0.8