ObjectivesKenya is one of the first African countries to scale up a national HIV viral load monitoring program. We sought to assess program scale up using the national database and identify areas for systems strengthening.MethodsData from January 2012 to March 2016 were extracted from Kenya’s national viral load database. Characteristics of 1,108,356 tests were assessed over time, including reason for testing, turnaround times, test results, treatment regimens, and socio-demographic information.ResultsThe number of facilities offering viral load testing increased to ~2,000 with >40,000 tests being conducted per month by 2016. By March 2016, most (84.2%) tests were conducted for routine monitoring purposes and the turnaround time from facility-level sample collection to result dispatch from the lab was 21(24) [median (IQR)] days. Although the proportions of repeat viral load tests increased over time, the volumes were lower than expected. Elevated viral load was much more common in pediatric and adolescent patients (0-<3 years: 43.1%, 3-<10 years: 34.5%, 10-<20 years: 36.6%) than in adults (30-<60 years: 13.3%; p<0.001).ConclusionsCoverage of viral load testing dramatically increased in Kenya to >50% of patients on antiretroviral therapy (ART) by early 2016 and represents a relatively efficient laboratory system. However, strengthening of patient tracking mechanisms and viral load result utilization may be necessary to further improve the system. Additional focus is needed on paediatric/adolescent patients to improve viral suppression in these groups. Kenya’s national viral load database has demonstrated its usefulness in assessing laboratory programs, tracking trends in patient characteristics, monitoring scale-up of new policies and programs, and identifying problem areas for further investigation.

ObjectiveTo describe factors associated with mother-to-child HIV transmission (MTCT) in Kenya and identify opportunities to increase testing/care coverage.DesignCross-sectional analysis of national early infant diagnosis (EID) database.Methods365,841 Kenyan infants were tested for HIV from January 2007-July 2015 and results, demographics, and treatment information were entered into a national database. HIV risk factors were assessed using multivariable logistic regression.Results11.1% of infants tested HIV positive in 2007–2010 and 6.9% in 2014–2015. Greater odds of infection were observed in females (OR: 1.08; 95% CI:1.05–1.11), older children (18–24 months vs. 6 weeks-2 months: 4.26; 95% CI:3.87–4.69), infants whose mothers received no PMTCT intervention (vs. HAART OR: 1.92; 95% CI:1.79–2.06), infants receiving no prophylaxis (vs. nevirapine for 6 weeks OR: 2.76; 95% CI:2.51–3.05), and infants mixed breastfed (vs. exclusive breastfeeding OR: 1.39; 95% CI:1.30–1.49). In 2014–2015, 9.1% of infants had mothers who were not on treatment during pregnancy, 9.8% were not on prophylaxis, and 7.0% were mixed breastfed. Infants exposed to all three risky practices had a seven-fold higher odds of HIV infection compared to those exposed to recommended practices. The highest yield of HIV-positive infants were found through targeted testing of symptomatic infants in pediatric/outpatient departments (>15%); still, most infected infants were identified through PMTCT programs.ConclusionDespite impressive gains in Kenya’s PMTCT program, some HIV-infected infants present late and are not benefitting from PMTCT best practices. Efforts to identify these early and enforce evidence-based practice for PMTCT should be scaled up. Infant testing should be expanded in pediatric/outpatient departments, given high yields in these portals.