ANC surveillance research

1
ANC surveillance research

• Yusufu Kumogola, Emma Slaymaker, Raphael Isingo,
  Julius Mngara, Basia Zaba, John Changalucha and
  Mark Urassa

TAZAMA / NACP seminar, Dar-es-Salaam, September
19th 2008
2
Structure of presentation

1. Added value of surveillance research topics
   investigated, methods and clinics used
2. Specific aims and results from rounds 1 2
3. Specific aims and results from round 3
4. Plans for round 4
5. Policy implications of findings and suggestions
   for NACP surveillance

3
Overview of surveillance research

• GENERAL AIMS
• To put the Kisesa sero-survey findings into a
  wider context
• To generate HIV surveillance data to complement
  NACP activities
• To discover biases affecting national ANC
  surveillance
• To evaluate and extend service availability in
  ANC
• To explore possibilities of adding extra data to
  national surveillance
• SPECIAL TOPICS
• 2000 2002 establish baseline and investigate
  sexual behaviour
• 2006 travel to clinics and type of service
  provided
• 2008 use of family planning in the context of
  PMTCT

4
How the ANC clinics were chosen

• The 2000/02 studies used a convenience sample
  11 ANC clinics in Magu district and eastern part
  of Mwanza city which already did routine syphilis
  tests or in which it was feasible to introduce
  syphilis testing
• The 2006 study added all other ANC clinics in
  Magu district and eastern Mwanza that had started
  providing syphilis tests (2 clinics) or VCT (1
  clinic)
• The 2006 study also added 11 other clinics in
  which lab tests were not available, but which
  were located within a 20 km radius of clinics
  providing HIV or syphilis tests women in these
  clinics were interviewed but not tested
• The 2008 study is using all the clinics that were
  able to provide lab tests in the previous rounds

5
Our basic questionnaire

• Background date of birth, residence, education,
  parity, date of last birth, survival of last born
• Clinic choice previous clinic attendance,
  transport, reason for using this clinic
• Father of the baby is she married to father, his
  age and residence, does he have other wives /
  girlfriends
• Sexual behaviour age at first sex, age at first
  marriage, other partners apart from father of the
  baby
• Test history ever had VCT or syphilis test
  before

6
Use stickers to link data and specimens
no names used !
Questionnaire completed in clinic
RPR test done in clinic
HIV test done at NIMR
7
Aims of 2000/02 survey

• Establish HIV infection levels and trends in
  urban, roadside and remote clinics
• Measure extent of co-infection with syphilis by
  type of clinic
• Describe patterns of sexual behaviour in young
  pregnant women
• Identify behavioural risk factors for HIV
  infection

8
Findings from 2000/02 rounds
HIV prevalence was higher in rural roadside
clinics and Mwanza city (10 to 13) than in
remote rural clinics and Magu town (6 to 9) But
syphilis was more prevalent in remote rural and
roadside clinics (15 to 21) compared to city
and town clinics (9 to 10) suggesting higher
use of antibiotics in urban areas?
9
HIV prevalence in ANC women by years of sexual
activity before and after marriage
Women who spent more years sexually active before
getting married were at higher risk of HIV
infection. Simple questions about age at first
sex and age at first marriage can provide useful
data for community advocacy
10
Aims of 2006 study

1. To measure the proportion of women accepting VCT
   in ante-natal clinics that offered the service
2. To find out which kind of women received VCT
   during pregnancy
3. To identify clinics with high unmet need for VCT
   services for pregnant women
4. To establish the extent of travel to non-local
   ANC clinics in this population.
5. To assess whether travel to non-local clinics
   affects HIV prevalence estimates

11
Location of clinics used in 2006 ANC surveillance
study and other health facilities in Mwanza
region
12
Use of VCT

• 88 of women attending VCT clinics accepted
  counselling
• About 70 of women attending clinics that did not
  provide VCT were attending their nearest clinic
  40 of those attending VCT clinics were attending
  nearest clinic
• Important predictors of VCT use (after allowing
  for clinic location) were
• urban residence (AOR 8.6, CI 7.2 10.2)
• primary or higher education (AOR 1.8, CI 1.4
  2.4)
• never married (AOR 1.5, CI 1.1 2.0)
• age group 20-29 (AOR 1.2, CI 1.0 1.4)

13
Results of surveillance HIV prevalence by clinic
location
14
HIV prevalence by PMTCT provision
15
(No Transcript)
16
HIV prevalence in women whose nearest clinic
provided syphilis testing but not VCT
17
(No Transcript)
18
How far do women travel?  
19
Aims of 2008 ANC survey follow-up

• Describe Family Planning (FP) use before
  pregnancy and in post partum interval
• Assess if prior FP use differs by HIV status
• Assess if post partum FP use differs by HIV
  status in women who had VCT
• Test the efficacy of added FP counselling added
  to VCT and tailored to status
• Evaluate uptake of PMTCT by infected women who
  had VCT

20
NACP surveillance advice

• There is no evidence of new bias introduction
  into surveillance estimates of HIV prevalence,
  due to clinic choice by pregnant women
• To maintain bias-free surveillance in PMTCT
  clinic, do not include tests on women who have
  been referred from other clinics
• Questions on residence will help map HIV
  prevalence with less dependence on clinic
  location
• Questions on re-marriage, survival of last born
  child, and length of birth interval are easy to
  ask in all ANC and can be used to identify
  pregnant women who are at high risk of HIV
  infection

21
Policy implications

• High prevalence of syphilis in rural areas
  suggests screening and treatment is an important
  priority treatment is cheap and impact on infant
  mortality is high
• Extending syphilis testing to rural areas also
  provides more opportunities for anonymous HIV
  surveillance
• Volume of HIV positive tests per week in clinics
  doing only anonymous testing is a good guide for
  prioritising VCT roll-out
• Women who are identified as high risk in ANC
  clinics that do not offer VCT should be referred
  for PMTCT