Does nonlocal women’s attendance at antenatal clinics distort HIV prevalence surveillance estimates in pregnant women in Zimbabwe?

OBJECTIVE The objective was to assess whether HIV prevalence measured among women attending antenatal clinics (ANCs) are representative of prevalence in the local area, or whether estimates may be biased by some women's choice to attend ANCs away from their residential location. We tested the hypothesis that HIV prevalence in towns and periurban areas is underestimated in ANC sentinel surveillance data in Zimbabwe. METHODS National unlinked anonymous HIV surveillance was conducted at 19 ANCs in Zimbabwe in 2000, 2001, 2002, 2004, 2006, 2009, and 2012. This data was used to compare HIV prevalence and nonlocal attendance levels at ANCs at city, town, periurban, and rural clinics in aggregate and also for individual ANCs. RESULTS In 2000, HIV prevalence at town ANCs (36.6%, 95% CI 34.4-38.9%) slightly underestimated prevalence among urban women attending these clinics (40.7%, 95% CI 37.6-43.9%). However, there was no distortion in HIV prevalence at either the aggregate clinic location or at individual clinics in more recent surveillance rounds. HIV prevalence was consistently higher in towns and periurban areas than in rural areas. Nonlocal attendance was high at town (26-39%) and periurban (53-95%) ANCs but low at city clinics (<10%). However, rural women attending ANCs in towns and periurban areas had higher HIV prevalence than rural women attending rural clinics, and were younger, more likely to be single, and less likely to be housewives. CONCLUSIONS: In Zimbabwe, HIV prevalence among ANC attendees provides reliable estimates of HIV prevalence in pregnant women in the local area.