Abstract
Early infant HIV diagnosis (EID) is critical to ensuring timely diagnosis of HIV-exposed infants, and treatment in those found to be infected. However estimates of coverage vary considerably, depending on data sources used. We used 4 methods to estimate coverage among a historical cohort of HIV-exposed infants in rural South Africa, between 2010-2016. Three datasets from this analysis are available on the repository.
· Dataset 1 (NHLS): This dataset includes data from NHLS, the National Health Laboratory Service. NHLS is responsible for laboratory testing in public facilities in South Africa, and data on all HIV DNA PCR tests conducted at the 17 healthcare facilities within the Hlabisa health sub-district (regardless of the age of the individual) between 1st June 2010 and 1st July 2017 were downloaded from the NHLS database through a secure file transfer protocol. A deterministic and probabilistic data linkage algorithm, based on first name, surname, sex, date of birth, facility at which the test was conducted and the infant's facility ID was used to identify repeat tests on the same child.
· Dataset 2 (ACDIS): This dataset includes all infants from ACDIS, AHRI's demographic surveillance database, who were born to women with HIV. Maternal HIV status was determined on the basis of either a positive HIV test result in the AHRI serosurvey, or on linkage to TIER.net, the national ART surveillance system. To estimate testing coverage, these HIV-exposed infants were linked to deduplicated PCR test data from the National Health Laboratory Service (NHLS, responsible for laboratory testing in public facilities in South Africa) from healthcare facilities in the Hlabisa health-subdistrict. Linkage was conducted using a deterministic and probabilistic linkage algorithm based on infant's first name, surname, date of birth and sex.
· Dataset 3 (MONARCH): This dataset includes data from the MONARCH trial, which evaluated the impact of a quality improvement intervention package on PMTCT processes. All children in South Africa are given a patient-held medical record at birth, called the Road-to-Health Booklet, in which information including HIV testing is recorded. As part of the trial, booklets belonging to all infants born to women receiving antenatal care at the 7 clinics in the AHRI surveillance area between July 2015 and December 2016 were photographed up to the 6-week postnatal visit. This dataset includes one record per HIV-exposed infant, with variables indicating whether a PCR test was recorded in their Road-to-Health Booklet and when their booklet was photographed.