Utility of PMTCT Program Data for HIV Surveillance In Botswana-Preliminary Results

1
Utility of PMTCT Program Data for HIV
Surveillance In Botswana-Preliminary Results

• Presented by Dr K. Seipone

2
Republic of Botswana

• Landlocked in southern Africa
• Area 581,700 sq km (226,900 sq mi) - size of
  Kenya or France or Texas
• Population 1.7 million mostly concentrated in
  the east
• Health care system decentralized in 24 health
  districts
• About 96 of women receive ANC and 97 deliver in
  a health facility

3
Background

• Overall 33.7 of pregnant women infected (2007)
• Approximately 14,000 HIV infected women deliver
  every year
• Successful PMTCT program with only about 4
  infected children
• Following Pediatric AIDS Clinical Trials Group
  (PACTG) 076, the PMTCT program was piloted in
  1999, nationally rolled out in 2001
• Currently, PMTCT is integrated in ANC settings
  and provided in all health facilities

4
PMTCT Coverage
5
HIV Surveillance in Botswana

• Botswana conduct antenatal HIV sentinel
  surveillance among pregnant women since 1992 with
  2 districts, now includes all 24 (27 soon) health
  districts
• Second generation HIV Surveillance started in
  2003.
• ANC sentinel surveillance in 270 selected public
  facilities
• ANC SITE HEALTH DISTRICT
• Botswana AIDS Impact Survey (BAIS), in 2001 (no
  HIV testing), 2004 and 2008(Include HIV testing)
• Included incidence and HIV DR among pregnant
  women since 2005
• More than 95 of pregnant women accept to test
  for HIV in the PMTCT program between 2005-2007

6
ANC Sentinel Surveillance
7
Health District

• Decentralized and comprises
• one referral/General/primary hospitals in each
  district
• Clinics with maternity
• Clinics without maternity
• Health post
• Private clinics

8
(No Transcript)
9
Objectives

• To assess the demographic similarity between
  PMTCT clients and ANC sentinel population
• To compare HIV prevalence estimates from PMTCT
  program data with those from ANC surveillance
  data
• To identify determinants for differences between
  PMTCT and ANC-related HIV estimates

10
Method

• Data covered period between July-September
  2005-2007
• Data collected from 2 health facilities with the
  highest ANC attendances in each of the 24 health
  districts in both urban and rural areas
• ANC Previous ANC HIV sentinel records
• PMTCT data was collected from facility ANC
  register (Photography)
• PMTCT study sitesANC SS sites
• Age restricted to 15-49 years

11

• Bivariate and multivariate analysis were done to
  compare HIV prevalence estimate
• Variables of interest were age, HIV status,
  area, time

12
1. Data quality
Results
13

• 18106 records reviewed
• 16,424 records were usable (91)
• 1,682 records were unusable (9)
• 212 unusable records were among ANC records
  (13)
• 103 from urban area
• 109 from rural area
• 1470 were among PMTCT records (87)
• 801 from urban area
• 669 from rural area
• Plt0001

14
(No Transcript)
15
(No Transcript)
16
2. Comparison of HIV Prevalence between PMTCT and
ANC SS Data
17
(No Transcript)
18
Comparison of HIV Prevalence between PMTCT and
ANC Data By Area- Rural N()
ANC PMTCT Total P value
HIV- 2375(65.2) 2936(65.5) 5311(65.4)
HIV 1270(34.8) 1544(34.5) 2814(34.6)
3645 4480 8125 0.73
19
Comparison of HIV Prevalence between PMTCT and
ANC Data By Area- Urban N()
ANC PMTCT Total P value
HIV- 2656(65.1) 2835(67.2) 5491(66.2)
HIV 1427(34.9) 1381(32.8) 2808(33.8)
4083 4216 8299 0.9
20
(No Transcript)
21
(No Transcript)
22
Conclusion

• PMTCT program reached more pregnant women aged
  15-49 years than ANC SS during the same period
• ANC SS data quality is superior to that of PMTCT
  data
• ANC SS data not significantly different from that
  of PMTCT program
• PMTCT data can be used for HIV surveillance in
  Botswana

23
Recommendations

• Botswana to improve the quality of PMTCT program
  data
• Botswana to use PMTCT data for HIV SS for HIV
  surveillance to address sustainability and
  efficiency
• Space the conduct of ANC SS to 5 years to
  calibrate PMTCT data
• Redirect ANC SS resources to improve PMTCT data
  quality in selected sentinel surveillance health
  facilities

24
Acknowledgement

• Dr Stephane Bodika, CDC
• Florindo de la Hose Gomez, MOH
• MRS Mpho Mmelesi, MOH
• Dr Negussie Taffa, CDC Botswana
• Dr Thierry Roels, CDC Botswana
• Others partners CDC-Botswana (BOTUSA), PEPFAR,
  BHP, WHO, UNICEF