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

934 Abstracts 43449-43453

12th World AIDS Conference

Design: Retrospective, database cross-linked study.
Methods: Data related to any AIDS procedure performed at hospitals of
the municipality of Rio de Janeiro during the year of 1996 were extracted from a
database of the Hospital Information System (AIH) and compared to the database
containing the cases reported to the surveillance system of the municipality of Rio
de Janeiro (SINAN). Statistical analysis was performed using the SAS system.
Results: A total of 3789 observations were analyzed, corresponding to 1946
patients. Out from these individuals, 1114 (57.2%) were reported to the SINAN
as being AIDS cases. However, despite the fact that 832 (42.8%) patients were
diagnosed as AIDS, they were not detected by the surveillance system. The
under-reported cases were similar to reported ones with respect to sex, but were
significantly different in what concerned age, mainly because reporting of children
tended to be lower than adults (x2 p < 0.006). Diagnosis of disorders of the
autonomous nervous system was also significantly less reported (x2 p < 0.001).
Conclusions: Physicians should realize their critical role in reporting AIDS
cases to the surveillance system. In addition, the strategy of comparing database
banks should be used routinely in order to assess the real level of AIDS cases,
therefore resulting in a better allocation of health resources.
43449 Trends of HIV infection among pregnant women in Togo
Teyi A. Lawson1, Kpante Bassabi2, Mireille David2, Michel Etchepare2. 1BP
13146 Lome; 2PNLS/MST, Lome, Togo
Objective: To appreciate trends of the HIV infection among pregnant women in
the different regions in Togo.
Methods: From 1994 to 1997, a serosurveillance (anonymous testing) has
been regulary carried out (every hear) in five sites of the contry in the group of
pregnant women. This surveillance is led according to a sentry system. Unlinked
aliquots of sera collected for syphilis testing were obtained from 250 to 500 eligible
women consecutively selected. Sera are sent to the central laboratory and were
screened for HIV antibodies by WHO's strategy II. For the uncertain results, a
third Elisa or a Western blot is carried out.
Results: Evolution of the HIV prevalence among pregnant women from 1994
to 1997

1994 (CI) 1995 (CI) 1996 (CI)
Sokode 4.5% (2.7%-6.5%) 4.8% (2.9%-6.7%) 3.8% (2.2%-5.4%)
Kara 4.5% (2.6%-6.4%) 3.7% (2%-10%) 5.5% (3.5%-7.5%)
Dapaong 2.4% (1.1%-3.7%) 2.5% (1.1-3.9%) 4.1% (2.3%5.8%)

1997 (CI)
4.6% (2.7%-6.5%)
8.2% (5.6%-10.8%)
3% (1.5%-4.5%)

In one site out of three, Kara (because the results of the two sites are not yet
available), there is a significant increase of the prevalence which switched from
4.5% in 1994 to 8.2% in 1997 (CI calculated at 95%); but in the two others sites
(Sokode, Dapaong), it has not increased at lot and remain comparable to the
adult general population prevalence estimated (3.5% in 1996).
Conclusion: This increase in the prevalence of HIV among pregnant women
in Kara expresses an anxious situation and needs an operational research as
to this situation in this region must be controled and an urgent and appropriate
action must be taken to control.
S43450 Monitoring the HIV/AIDS epidemic in Belgium in the era
of new treatments
Andre Sasse, J. Van Der Heyden, V. Batter, A. Stroobant. 16 J. Wytsman Street,
1050 Brussels, Belgium1 Scientific Institute of Public Health, Brussels, Belgium
Objective: To survey the epidemic in Belgium by means of HIV and AIDS cases
reporting.
Methods: Approximately 600.000 screening tests are yearly performed in Belgium, excluding the tests related to blood donations. All serums with a positive
result are submitted for confirmation to the AIDS Reference Laboratories. For
each confirmed test, the laboratories send a form to the patient's clinician requiring data on age, sex, nationality, residence and possible route of transmission.
Data are validated for deleting duplicate records and included in a HIV database
since 1985, at the Scientific Institute of Public Health-Louis Pasteur. A database
including newly diagnosed AIDS cases reported by clinicians has been started in
1984. In 1990, HIV and AIDS databases were integrated. A follow up survey is
conducted each year to collect data on last consultation and possible death of
reported AIDS cases.
Results: The AIDS incidence in Belgium, adjusted for reporting delays, reached
a plateau between 1991 and 1995 (range: 244-257 diagnosed cases/year). A
25% drop was observed in 1996 (185 reported cases). HIV infections, although
roughly stable during the period 1986-1996 (range 719-977 diagnosed HIV
cases/year), show a slight decrease during the last years: 943, 801, 762, and
719 new cases were diagnosed in 1993, 1994, 1995, and 1996 respectively.
The reporting delays are negligible. The case-fatality rate decreased significantly
in 1996: among the AIDS patients diagnosed during the period 1990-1994, an
average of 44% (39-47%) died before the end of the year following that of the
diagnosis versus 32% among the patients diagnosed in 1995. Reporting delays
can not explain this change; the use of new treatments in 1996 is the most
plausible explanation. The analysis of the 1997's data may contribute to describe
further the effects of the use of new treatments.
Conclusions: The surveillance by means of a HIV and AIDS cases reporting
allows to continue with the monitoring of the HIV epidemic in the era of more
effective HIV treatment. It may also contribute to describe the effects at a country
level of the generalized use of new HIV therapeutics.

43451 Evaluation of STD/HIV/AIDS surveillance systems in five
Caribbean countries
Bilali Camara, H.-U. Wagner, C.J. Hopedales, M. Lewis, M. de Groulard.
Caribbean Epidemiology Centre (CAREC), 16-18 Jamaica Boulevard,
Federation Park, Port of Spain, Trinidad
Background: Strengths and weaknesses of STD/HIV/AIDS surveillance were assessed by the Caribbean Epidemiology Centre in 5 countries (CMC): Barbados,
Grenada, Guyana, St. Vincent & The Gren., Trinidad & Tob.
Method: Using evaluation guidelines, teams reviewed records and conducted
interviews to assess reporting regulations, formats, flow, data analysis, system
sensitivity, dissemination of results and their use.
Results: AIDS case reporting is by name and linked to HIV testing. All CMC
had a report form, but there were problems with its use. Case definitions were not
well known to clinicians. Most reported cases were tested because of symptoms.
Sensitivity estimates for AIDS (% of cases captured by system) ranged from 35%
to 90%, and from 10% to 60% for asymptomatic HIV-infections. High sensitivity
in 3/5 CMC was associated with centralized testing, and good treatment and
support services. Low sensitivity was associated with limited access to voluntary
testing and "friendly" treatment services, concerns about confidentiality, high private sector testing, and (mainly private) providers' low reporting compliance. 2/5
CMC had conducted HIV sentinel surveillance within the last 3 years. No recent
data were available on likely core groups (CSWs, drug users, MSM). In 4/5 CMC,
STD cases were reported by public STD clinics, 2/4 showing a decline, possibly
because of changed care seeking behaviour. One country did some contact tracing for STD but not HIV/AIDS. Named reporting was used to identify duplicates
but rarely for supportive follow-up. Because of focus on AIDS reporting, limited
analysis of existing data or its dissemination, some important recent trends were
not recognized or used for planning.
Discussion: Surveillance of AIDS has been improved. Evidence of rapidly
growing incidence was used for political advocacy. However, available data on
vulnerable groups were hardly used to redirect strategies. Based on the findings
of the evaluation, workplans were developed to strengthen surveillance of HIV
and STDs, and the measurement of sociodemographic and behavioural markers.
These are being implemented with CAREC's assistance.
4342 Increasing prevalence of HIV infection in
Nigeria
DanlamiArabs Rukujei, I.E. Derex-Brigs, A.C. Zwandor, M. Gboun. c/o Dr K.
Oforsu-Barko CPA UNAIDS UNDP Office Lagos, Nigeria
Objective: To monitor HIV prevalance among women attending antenatal clinics.
Design: HIV sentinel sero-prevalance survey.
Methods: At fifty eight preselected sentinel sites in twenty states of the federation, women attending antenatal clinic for the first time during current pregnancy
were enlisted. Over a period of eight weeks, alliquotes of blood taken for routine
tests were collected using unlinked anonymous methods. Double ELISA technique was used for sample testing. Samples that were reactive to both methods
were reported as being seropositive. Data was analysed using EPI-lnfo Version
6. Results were compared with that of 1991 and 1993 surveys.
Results: There was a wide variation of prevalance rates between states (range
0.1%-34.5%). The Nationa Prevalance rate was computed as 4.5%. Prevalance
rates were 1.4% and 3.8% in 1991 and 1993 respectively.
There is an alarming increase of HIV infection in Nigeria with 221% increase
between 1991 and 1995. With an estimated over two million people infected (more
than the total population of some African countries), a pathetic situation prevails.
Despite the rising cost of conducting sentinel surveiliance, it remains a viable tool
for monitoring the trend of HIV in populations and needs to be continued.
43453 The HIV sentinel surveillance system in Kenya:
Monitoring HIV levels and trends
Godfrey Baltazar1, T.M. Okeyo1, J. Stover2, A.G. Johnston3. 1 Kenya National
AIDS & STDs Control Programme, Ministry of Health, PO. Box 19361, Nairobi,
Kenya; 2 The Futures Group International; 3Research Triangle Institute, Res.
Triangle PK, USA
Background: The objective of the HIV sentinel surveillance program in Kenya is
to provide a basis for estimating the extent and pattern of HIV infection in Kenya
in order to monitor the epidemic and design effective programme strategies.
Methods: Sentinel surveillance is conducted for STD patients and antenatal
clinic (ANC) attendees. The system has been operational since 1990 with ANC
data available for 13 urban sites annually from 1990 to 1997. In 1994 six periurban and rural sites were added, for which data is available through 1997. For
ANC sites, blood is drawn and tested from 200-300 women attending the clinic
for the first antenatal visit during the last quarter of the year.
Results: HIV prevalence among pregnant women in urban areas has been
high, with some annual fluctuations from 1990 to 1996 in the port city of Mombasa
(12-16%) and in areas of Western Kenya around Lake Victoria and the border

with Uganda (20-30%). Prevalence has been increasing rapidly in areas immediately north and south of Lake Victoria in Western Kenya and along the road from
the Uganda border to Mombasa, including Nairobi. In these areas prevalence
has increased from 2-10% in 1990 to 4-20% in 1997. In Nairobi, prevalence has
increased steadily from 6% in 1990 to 25% in 1997. Prevalence has remained
lower (4-6%) in areas north and east of Nairobi. Peri-urban sites in areas of high
urban prevalence show levels as high as the urban areas.