Mother-to-child transmission (MTCT) is the main route of transmission for HIV among under 5 children in Brazil. National data indicate that missed opportunities for HIV prevention of MTCT are still common in antenatal care (ANC). We studied variables related to target process indicators in a cohort of HIV exposed children. We used data from 1996 to 2013 related to HIV exposed uninfected and HIV-infected children attended in an HIV reference hospital in Rio de Janeiro, Brazil. Data were collected from baseline questionnaires applied to all children followed-up in the hospital. Gestational and perinatal history were extracted from the mother's ANC card. Infants were categorized according to dates of first HIV care at the unit (1996-2000, 2001-2006 and 2007-2013). Distances between recorded addresses and the nearest maternity/hospital were measured by Euclidean distance, the shortest car route calculated in Google Maps and the route of the available bus line. Of the 599 children who fulfilled the inclusion criteria, 178 (29.7%) were HIV-infected. Approximately 70% of infants exposed to the virus from 1996-2000 were infected, dropping to 15.2% from 2001-2006 and rebounding to 30.1% from 2007-2013. Birth cohort was associated with ANC, and mothers from 2007-2013 had a lower chance of attending ANC (OR = 0.16; 95%CI 0.08-0.30). In addition, when the distance home-birthplace was higher than 9.5 km, there was a lower chance that the mother had attended ANC (OR = 0.35; 95%CI 0.18-0.68). Birth cohort was associated to HIV and ANC, and our data showed that a reduction of ANC might be related to rebound in HIV cases. There seems to have an association between larger distances from home to the birthplace and absence of ANC, which suggests that ANC was being performed in the tertiary units instead of in the primary care facilities as recommended.
Keywords: Brazil; GIS; HIV/Aids; mother-to-child transmission; prevention.