Urassa M, Isingo R, Kumogola Y, Mwidunda P, Halelwa M, Changalucha J, Mngara J, Zaba B, Boerma T, Calleja T, Slaymaker E.
Unpublished report for WHO SURVART initiative
Background: Many ante-natal clinics that provided unlinked anonymous data for national HIV sentinel surveillance have started providing services for the prevention of mother to child transmission (PMTCT). Unlike data from unlinked anonymous testing, estimates of HIV prevalence based on voluntary test results from these clinics may be biased by refusals to undergo voluntary counselling and testing (VCT) 1. Other biases may arise if PMTCT clinics attract women from outside the usual clinic catchment area. HIV prevalence trends in non-PMTCT clinics which continue to provide unlinked anonymous surveillance data may also be affected by women choosing to go to clinics away from their area of residence. The Mwanza region of Tanzania furnishes suitable background data to investigate this subject, as two rounds of expanded ANC surveillance have already been conducted in this area.
Study design and objectives: To investigate how partial coverage of PMTCT affects pregnant women’s choice of ANC, and whether characteristics of women are associated with the type of clinic chosen and whether the same factors are associated with HIV infection. Specific aims were:
- Acceptance of testing: to monitor refusal rates for interviews and serological tests in different clinics
- HIV prevalence trends: to examine trends in HIV prevalence in clinics that previously participated in anonymous unlinked ANC surveillance
- Geography of HIV infection: to map the geographical patterns of HIV infection in the study area
- Effect of clinic choice: to study the relationship between VCT uptake and factors associated with HIV infection identified in those clinics with anonymous surveillance
- Changes in characteristics of women: to identify trends in socio-demographic profiles and sexual risk behaviour in women attending clinics that participated in all surveillance rounds
- Characteristics of women attending different types of clinic: to compare the residential distribution, socio-demographic background and sexual behaviour characteristics of women attending ANC that offer PMTCT services, ANC that offer only syphilis treatment, and ANC with no diagnostic services
- Representativeness of ANC data: to evaluate the extent to which the serological and behavioural data collected in ANC serving Kisesa ward represent the community characteristics as measured in the most recent community-based survey
- Factors associated with HIV infection: To describe behavioural and socio-demographic correlates of infection
- Coverage of PMTCT: to estimate what percentage of HIV infected women received VCT at ANC and to identify clinics with high unmet need for HIV testing of pregnant women.