HIV Prevalence Estimates in Rural South Africa
Author Information
Author(s): Brian D Rice, Jörg Bätzing-Feigenbaum, Victoria Hosegood, Frank Tanser, Caterina Hill, Till Barnighausen, Kobus Herbst, Tanya Welz, Marie-Louise Newell
Primary Institution: Africa Centre for Health & Population Studies, KwaZulu Natal, South Africa
Hypothesis
How do population-based and antenatal-care HIV prevalence estimates compare among women in a rural South African population?
Conclusion
Population-based surveillance likely under-estimates HIV prevalence, while ANC sentinel surveillance may over-estimate it due to selection biases.
Supporting Evidence
- Population-based HIV prevalence was 25.2% compared to 37.7% for ANC attendees.
- Testing consent rates varied significantly by age and residency.
- Urban residents had the lowest rates of consent to test.
Takeaway
This study shows that testing for HIV in pregnant women at clinics might give a higher number than testing all women in the community, which can be misleading.
Methodology
HIV prevalence was compared between population-based surveys and ANC sentinel surveillance using blood samples from women.
Potential Biases
Potential biases include unrepresentative testing by age, residence, and HIV status.
Limitations
The study may not fully account for biases in testing consent and clinic attendance.
Participant Demographics
Women aged 15 to 49 years in rural KwaZulu-Natal, South Africa.
Statistical Information
P-Value
p<0.05
Confidence Interval
95% CIs: 24.0%, 26.4%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website