Evaluation of Kenya’s readiness to transition from sentinel surveillance to routine HIV testing for antenatal clinic-based HIV surveillance

Martin Sirengo, George W. Rutherford, Boaz Otieno-Nyunya, Timothy A. Kellogg, Davies Kimanga, Nicholas Muraguri, Mamo Umuro, Joy Mirjahangir, Ellen Stein, Margaret Ndisha, Andrea A. Kim

Sentinel surveillance for HIV among women attending antenatal clinics using unlinked anonymous testing is a cornerstone of HIV surveillance in sub-Saharan Africa. Increased use of routine antenatal HIV testing allows consideration of using these programmatic data rather than sentinel surveillance data for HIV surveillance. To gauge Kenya's readiness to discontinue sentinel surveillance, we evaluated whether recommended World Health Organization standards were fulfilled by conducting data and administrative reviews of antenatal clinics that offered both routine testing and sentinel surveillance in 2010. The proportion of tests that were HIV-positive among women aged 15-49 years was 6.2 % (95 % confidence interval [CI] 4.6-7.7 %] in sentinel surveillance and 6.5 % (95 % CI 5.1-8.0 %) in routine testing. The agreement of HIV test results between sentinel surveillance and routine testing was 98.0 %, but 24.1 % of specimens that tested positive in sentinel surveillance were recorded as negative in routine testing. Data completeness was moderate, with HIV test results recorded for 87.8 % of women who received routine testing. Additional preparation is required before routine antenatal HIV testing data can supplant sentinel surveillance in Kenya. As the quality of program data has markedly improved since 2010 a repeat evaluation of the use of routine antenatal HIV testing data in lieu of ANC sentinel surveillance is recommended.