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

12th World AIDS Conference

Abstracts 43443-43448 933

43443 A change-point model for reporting delays under change
of AIDS case definition
Farzaneh Tabnak', H.-G. Muller2, J.-L. Wang2, J.-M. Chiou2, R. Sun1.
'California DHS Office of AIDS, 830 S Street Sacramento, CA, 2University of
California, Davis Davis CA, USA
Background & Objective: The 1993 expansion of the CDC AIDS case definition
was expected to have an impact on the time lag of case reporting, but no published methods are available for quantifying the impact. The goal of this study
was to examine the effect of the 1993 change in definition of AIDS on the time
lag of case reporting while accounting for the contribution of covariates.
Method: We propose a change-point model for reporting delays in AIDS which
takes into account the changes in the definition of AIDS. We assume that the
Brookmeyer-Liao model holds only for those time intervals in which the definition
remains unchanged. We applied the method to 1989-1996 California AIDS data (n
= 75,492) on reporting delays and analyzed the changes occurred after the 1993
change of definition in terms of the factors which contribute to reporting delays.
Results: The effect of the 1993 change in definition on the distribution of
reporting delays was highly significant (p-value of less than 0.0001). Region of
residence at the time of diagnosis was the most significant factor in determining
the distribution of reporting delays. Mode of exposure and race/ethnicity were
other significant covariates. There were some significant interactions among the
covariates in the model.
Conclusion: The change-point model introduced here can be applied to other
situations where changes in diagnostic criteria and case definition occur. The
reporting delay distributions obtained can be used to upwardly adjust AIDS
surveillance data. The adjustments will be free of the effect of the 1993 change in
definition while accounting for the significant covariates, and may provide a more
accurate picture of AIDS incidence.
S43444 Epidemiological surveillance systems for HIV/AIDS in the
Western Pacific region
Annette Ghee, S.O. Omi, N.F. Fee, G.P. Poumerol. United Nations Avenue PO.
Box 2939 1000, Manila
Issues: Countries use various approaches to HIV/AIDS surveillance as a way
to improve their understanding of their HIV epidemics. Many countries in Asia
and the Pacific have developed surveillance systems over the last several years.
Learning from this experience, we can select among various available data collection methods to design cost-effective surveillance systems that are appropriate
for a given setting.
Project: The surveillance systems of selected countries served by the WHO
Regional Office for the Western Pacific were reviewed and compared.
Results: The range of methods employed by countries include: case reports,
routine screening (mandatory or voluntary), large scale screening, sentinel surveillance, and ad hoc surveys. Surveillance systems are in place in countries with
several years of experience with HIV epidemics. In Cambodia and the Philippines,
HIV sentinel surveillance is combined with case reports and routine screening.
Similar systems have been recently established in China and Viet Nam. In Australia and New Zealand case reporting gives a good profile of the epidemic's
evolution.
Lessons learned: Information exchange between countries has lead to refinements in surveillance systems. Regular review and modification of surveillance
systems is needed as resources improve and information becomes available.
This helps make accurate information available to planners and decision-makers
at a reasonable cost. HIV reporting, although incomplete is a useful as a complementary source of information and an advocacy tool. Large scale screening is
relatively inefficient and costly yet it persists in several countries.
43445 Mobilizing school and community youth in STD/AIDS
prevention
Josefina Del Gallego', A.R. Parawan2, P.S. Uysingco2. 'Tridev, Room 206, Del
Mundo Bldg., 2121 Taft Ave., Malate Manila; 2 Tridev Specialists Foundation Inc.,
Manila, Philippines
Issue: Among today's youth, one of the most serious reproductive health problems is sexually transmitted diseases (STD), particularly HIV/AIDS. Unprotected,
premarital and extramarital sex (commercial and casual) puts these young people
at risk to STD/AIDS.
Project: The School- and Community-Based Youth Educators and Advocates
for Reproductive Health Project aims to mobilize the participation of youth in 10
high schools and 5 communities in STD/AIDS prevention. The project's main
components are Capability Building, Health Education and Advocacy, and Networking. The main strategies are to establish partnerships with 10 high schools,
through their student-editors, teacher-advisers and guidance counselors, and 5
communities through their elected youth leaders, and to enable these partner
groups to provide education on reproductive health and STD/AIDS prevention
and risk reduction. The project also establishes direct linkages with the referral

and support system, particularly local government, for health and other social
services for the youth.
Results: The project has organized and trained 11 students, 11 teacher-advisers and 11 guidance counselors from 11 partner high schools, and 10 elected
youth leaders from 4 communities in a municipality in Metro Manila, Philippines
Through school publications and classroom-based activities, the 11 high schools
are reaching their combined student population of about 20,000 with informa
tion on STD/AIDS prevention. The partner schools have also organized youth
health clubs among their students to initiate education and advocacy activities
in the campus. The 10 elected youth leaders are integrating STD/AIDS prevention and reproductive health in their community programs for the youth. The
project has also provided communication materials to support the school- and
community-based activities.
Lessons learned: STD/AIDS projects for the youth need to take into account
the different situations of school and community youth, and to tailor interventions
to their specific needs and capabilities.
S43446 Evaluation of the WHO staging classification for HIV
infection using data from a rural Ugandan cohort
Sewava Sammuel Malamba', D. Morgan2, B. Mayanja2, J. Whitworth2. 'MRC -
Uganda Virus Research Institute PO. Box Entebre; 2MRC Programme On AIDS
In Uganda Entebre, Uganda
Background: To evaluate the WHO staging classification of HIV infection using
clinical information from a population based natural history cohort of HIV positive
and negative participants.
Methods: Participants are seen every 3 months when a detailed questionnaire
is administered, a clinical examination performed and samples obtained. Using
a computer algorithm, each visit was categorised according the WHO staging
system. A ranking for the clinical conditions using median survival time to death
was developed and the prognostic strength of the WHO clinical and laboratory
stages were assessed separately and in combination by using Kaplan Meier and
Cox proportional hazard methods.
Results: By the end of 1997, 235 HIV-1 infected individuals contributed to 799.6
person years of observation. There was a strong correlation between the WHO
ranks and the ranks developed using the median survival times to death for each
clinical condition used in the staging (Spearman's rank correlation coefficient =
0.87). Certain conditions (eg. prolonged fever) showed weaker predictive value
than others due to high prevalence in HIV negatives. After adjusting for age and
HIV group (prevalent/incident), Oral candidiasis and chronic diarrhoea were the
strongest predictors of death Hazard Ratio = 4.8 (95% Cl = 2.7-8.5) and 2.8 (95%
CI = 1.6-4.8) respectively.
Conclusion: Application of the WHO staging to our data has shown that the
staging is clinically relevant since it reflects the increasing severity of disease and
predicts survival to death. Some simple modifications taking into account those
conditions with low prediction of survival, due to high population background
prevalence, could considerably improve the staging system.
43447 An HIV surveillance study in Mizoram, India
Anil Purohit1 2, J. Zohmingthanga3, Rohming3, B. Thangdailova3, S. Kalla4,
A. Kiessling2, J. Chakraborty5. 'c/o Judy Suleski, Dept. of Physiology &
Molecular Medicine Medical College of Ohio, 3035 Arlington Avenue, Toledo,
Ohio; 2BIDMC Harvard Medical School, Boston; 5Medical College of Ohio,
Toledo, USA; 3Regional Paramedical Training Institute, Aizawl, India; 4Dept. of
OB/GYN, John Radcliff Hospital, Oxford, UK
Objectives: To obtain data on HIV infection in Mizoram, India.
Design: Retrospective.
Method: Phase I October 1990-January 1991; a limited survey was conducted by testing HIV in blood samples from high risk groups. Phase II February
1991-December 1992; screening was done on blood donors from blood banks,
patients from civil hospitals and STD clinics, and volunteers. Phase III January
1993-September 1993; a maximum number of high risk groups were tested.
Phase IV October 1993-September 1994; mandatory blood screening before
transfusion was strictly maintained. Phase V October 1994-September 1995;
individuals with risky behavior participated in voluntary testing. Phase VI October
1995-September 1996; larger community involvement because of awareness on
HIV/AIDS, involved screening of IDU's, CSW's, blood donors and others. Phase
VII October 1996-September 1997; extensive larger community involvement because of more awareness on HIV/AIDS included from category VI plus spouses
and children of HIV positive individuals.
Results: In Phase I, 9 of 14; Phase II, 15 of 3,294; Phase III, 7 of 3,006; Phase
IV, 22 of 5,181; Phase V, 9 of 4,735, Phase VI, 9 of 5,664; and in Phase VII, 25
of 5,103 individuals tested positive.
Discussion/Conclusion: This study indicates that although HIV infection is not
yet very high in Mizoram, heterosexual transmission and infections in individuals
without risky behavior are major concerns. Extensive surveillance and public
education are needed in Mizoram.
S43448 Completeness of AIDS reporting in Rio de Janeiro:
A linkage study between surveillance system of cases
and hospital information system
Vanja M.B. Ferreira', M.C. Portela2. 'State Health Secretary, Rio de Janeiro, RJ;
Rua Sacopa 109/507, Lagoa 22471-180 Rio de Janeiro; 2National School of

Public Health/Fiocruz, Rio de Janeiro, RJ, Brazil
Background: The Hospital Information System (AIH) is used by the Brazilian
Ministry of Health to control and pay for the care of in-patients. It is a prospective
payment system and it also contains some epidemiological information.
Objectives: To assess the level of under-reporting of AIDS cases in the
municipality of Rio de Janeiro, Brazil.