Success With The Use of Rapid Versus ELISA Testing in the Implementation of PMTCT Programs

1
XV International AIDS Conference
WePeE6825
Success With The Use of Rapid Versus
ELISA Testing in the Implementation of PMTCT
Programs J.Nkhoma1, A. Moses1, C.Zimba1, G.
Mwale1, J. Murotho1, W.Dzinyemba1, S.Kamanga1, D.
Jones1, F.E.A. Martinson1, D. Chilongozi1, S.Napra
vnik2, I. F. Hoffman2, C. van der
Horst2. 1University of North Carolina Project,
Lilongwe, Malawi 2University of North Carolina
at Chapel Hill, Chapel Hill, NC
Abstract
Results
TABLE 2. Acceptance of VCT by method of test
Background The University of North Carolina
(UNC) Project, with funding from the Elizabeth
Glaser Pediatric AIDS Foundation (EGPAF) and
UNICEF- Malawi, put in place a Prevention of
Mother-to-Child Transmission (PMTCT) of HIV
program within 4 Lilongwe sub-urban government
public health centers.The program was
incorporated into the existing maternal and child
health care delivery system with the aim of
contributing to the national goal of reducing the
risk of mother-to-child HIV transmission.The
program reaches over 18,000 pregnant mothers who
attend antenatal clinic and offers voluntary
counseling and testing and single-dose Nevirapine
intervention (HIVNET 012 regime) to the
mother-infant pairs.The program started operating
in April 2002 with one site and increased to 4 by
April 2003.ELISA HIV tests were used initially,
requiring women to return after one week for
post-test counseling.In July 2003, rapid on-site
testing with same-day post-test counseling was
introduced using two parallel tests (Unigold and
Determine). ELISA tests are done on 10 of all
tests for quality control. Methodology We
compared antenatal women who received ELISA with
those offered same day rapid tests to determine
(1) VCT acceptance rate and (2) the proportion
that returned for post test counseling.Data were
collected on mother and baby clinic visit forms
daily and analyzed monthly. Results A total of
11,724 clients who had consented to be tested
were tested between January and December 2003.
Over the 12 months, 78 of the women who were
offered rapid HIV testing accepted VCT compared
with 36 of those who were offered ELISA tests.
Also, post-test counseling rate was 97 in the
rapid test arm compared to 78 who had an ELISA
test. Conclusions On-site rapid testing with
same-day results improves VCT uptake and post-
test counseling significantly when compared with
the ELISA test and the need to return for HIV
test results. Recommendations When deciding
between the two options, rapid test is more
effective in increasing the number of clients
opting for VCT and chances of the clients getting
post test results and counseling.
Clients offered VCT Clients accepting VCT
RR p-Value Test Method and
getting tested (95 CI) N
N () ELISA Test 10,243 3,718
(36.3) 1. Rapid Test 10,230
8,006 (78.3) 2.16 (2.10, 2.21)
lt0.0001 Total 20,473 11,724 (57.3)
The risk ratio (RR) and 95 confidence interval
(95 CI) for accepting VCT contrasting the rapid
test with the ELISA test was 2.16 (95 CI 2.10,
2.21) indicating that women offered the rapid
test were two times more likely to accept VCT in
contrast to those offered the ELISA test (exact
2- sided Fishers p value was plt 0.0001).
TABLE 3. HIV prevalence amongst clients by method
of test

• HIV Result ELISA Test
  Rapid Test Total
  
  
• Positive 720
  1,213
  1,933
• Negative 2,998
  6,793
  9,791
• HIV Prevalence 19.37 15.51
  16.49
• 95 CI (18.11, 20.67)
  (14.37, 15.96)
  (15.82, 17.17)

Background
Mother-to-Child-Transmission(MTCT)of HIV
infection has become a major focus due to high
prevalence of HIV in pregnant women (21.7 in
urban Malawi and 16.9 in Lilongwe antenatal
clinics). The National prevalence is estimated to
be 19.8 (2003 HIV sentinel report).Malawis
estimated MTCT is around 35 (NAC, 2002).
Concerns have been raised in the reproductive
health program over limited activities to
address this high antenatal women HIV
prevalence.It was against this background that
UNC Project with funding from EGPAF and in
collaboration with Ministry of Health and
Population (MOHP)initiated a PMTCT program
within 4 Lilongwe sub-urban government facilities
of Bottom HospitalKawale, Area 18 and Area 25
health centers.The program was incorporated into
the existing Maternal and Child Health care
delivery system with the aim of reducing the risk
of Mother-to-Child-Transmission of HIV in the
Lilongwe sub urban community.This was to be
achieved through integration of HIV/AIDS
education,Voluntary Counseling and Testing and
the administration of single dose Nevirapine
therapy to Mother-Infant pairs into the existing
Maternal and Child Health care delivery system.
  The program started operating in April 2002
with one site (Bottom Hospital) and was the first
to partner with the MOHP in the Central Region.
Nationwide, itwas the 4th site after Embangweni
in Mzimba,Chiradzulu and Thyolo districts. By
April 2003, all 4 centers had become operational
following a gradual phase - in approach.The
program currently reaches over 18,000 pregnant
mothers who attend antenatal clinics in these
facilities every year. NOTE The National
PMTCT program was launched on June 12, 2003 well
after Lilongwe district program was
initiated.
TABLE 4. Proportion of clients tested and
returning for results by method of test amongst
both HIV positive and negative clients

• Returning for results ELISA Test Rapid
  Test Total RR
  p-Value
• (95 CI)
• Yes
  2929 7,868
  10,797
• No
  789 138
  927
• Proportion of clients 21
  2 8
  (10.75, 14.09)
• NOT returning for results (789 / 3,718)
  (138 / 8,006) (927 / 11,724) 12.31
  lt0.0001

TABLE 5. Proportion of clients tested and
returning for results by method of test amongst
HIV positive clients only

• Returning for results ELISA Test
  Rapid Test Total RR
  p-Value
• (95 CI)
• Yes
  561 1,177
  1738
• No
  159 36
  195
• Proportion of clients 22
  3 10
  (5.56, 9.96)
• NOT returning for results (159 / 720)
  (36 / 1,213) (195 / 1,933)
  7.44 lt0.0001

The risk ratio for NOT returning for results
contrasting ELISA with the rapid test among all
those tested was 12.31 (95 CI 10.75, 14.09)
indicating that women tested with ELISA were over
12 times more likely to NOT return for their
results in contrast to those tested with the
Rapid test (Fishers exact 2-sided test
plt0.0001, Table 4). Among HIV infected clients,
the RR for NOT returning for results was 7.44
(95 CI 5.56, 9.96), with a p value lt 0.0001
(Table 5).
Methods
FIGURE 1. VCT in relation to clients tested and
post tested by method of test
All newly registered antenatal mothers at the 4
facilities were sensitized and counseled in a
group setting followed by individual counseling
prior to being offered VCT. Also, previously
registered antenatal attendees,not previously
tested for HIV infection during this pregnancy
were offered the opportunity to participate if
they so desired.
Before July, 2003, ELISA tests on venous blood
were used, requiring women to come a week later
for their results. Since July, 2003, Rapid
on-site testing using venous blood with same day
results has been introduced. Rapid tests used
were Determine plus () UNIGOLD using a parallel
testing approach.Concordant positive () results
were recorded as HIV positive(). Similarly,
concordant HIV negative () were recorded as HIV
negative(-). Discordant results were confirmed
using ELISA tests on the collected venous
samples. Data were collected during each mothers
clinic visit and recorded on the daily log form.
FIGURE 2. The proportion of clients accepting
VCT, tested and post tested by method of test
Laboratory setting depicting Rapid tests being
done in one of the UNC Project Laboratories at
Lilongwe Central Hospital, Malawi
We compared antenatal women who received ELISA
test with those who were offered same day rapid
tests to determine 1) VCT acceptance rate and 2)
The proportion that returned for post test
counseling/ results. Statistical analyses
Standard prevalence estimates, risk ratios, and
associated 95 confidence intervals were
calculated. The 2-sided Fishers exact test was
used to test for statistical significance
(alpha0.05).
Results
Overall, 20,473 women were approached for VCT,
with 50 (n10,243) offered ELISA test and 50
(n 10,230) offered rapid testing. Of these
women, 57 (n 11,724) agreed to VCT. Client
demographic and clinical characteristics were
similar among the women tested with the ELISA
test and the rapid test (Table 1).
TABLE 1. Characteristics of clients accepting VCT
by method of test
Conclusions
Client characteristics
ELISA Test
Rapid Test


(n 3718)
(n 8006) Mean age in years (SD) 25.8
(4.1) 26.2 (4.3) Median gravidity (1st, 3rd
quartile) 4 (3, 6) 4 (3, 7) Median parity (1st,
3rd quartile) 3 (2, 5) 3 (2, 6) Mean
gestational age at the start of ANC care in
weeks (SD) 23.2 (3.7) 23.3 (3.9)

• On site rapid testing with same day results
  doubles VCT uptake compared with the ELISA
  testing.
• Rapid testing improves percentage of clients
  receiving post-test counseling and results (78
  for ELISA test versus 97 for Rapid test).

Recommendations

• Use of rapid test is a more effective tool in
  promoting VCT uptake in the antenatal setting.
• Use of rapid test improves the proportion of
  those receiving their results.
• Receiving results with a positive message will go
  a long way in influencing behavioral change our
  desired goal on preventive strategy against HIV
  spread.
• It will be interesting for future evaluations to
  determine if mothers having ELISA testing have
  similar postnatal attendance experiences as those
  having Rapid testing.

Acknowledgement
References
This program is sponsored by the Elizabeth Glaser
Pediatric AIDS Foundation in Washington DC in
the U.S and UNICEF Malawi Office. The Malawi
Ministry of Health at Lilongwe Central Hospital,
Bottom Hospital and The Lilongwe District Health
Office provides infrastructure and other
technical support to run the program.
1. Annual Report, Malawi National AIDS Commission
(NAC)December, 2002. 2..HIV Sentinel
Surveillance Report Ministry of Health and
Population- MalawiNovember,2003.