Performance of point-of-care birth HIV testing in primary health care clinics: An observational cohort study

BACKGROUND: Failure to timely diagnose HIV in infants is a major barrier for scaling-up paediatric antiretroviral treatment (ART). WHO recommends birth testing for earlier diagnosis and to improve test coverage, but current diagnosis takes 2–3 weeks to complete, thereby limiting the ability of care givers to provide follow-on care, especially in low-resource settings. We evaluated the benefit of implementing rapid diagnosis of HIV at birth in primary health care maternity wards in Mozambique. METHODS AND FINDINGS: Infants born to HIV-infected mothers delivering consecutively at eight primary health care clinics were tested within 24 hours of delivery using on-site POC (Alere q HIV1/2 Detect) and standard laboratory (Roche COBAS AmpliPrep/TaqMan HIV-1 qualitative assay v2.0) testing. Infants were also tested at 4–6 weeks of age with both assays. Of 2,350 HIV-exposed infants enrolled in this implementation research study, 33 tested HIV-positive at birth on both assays. Sensitivity and specificity of POC testing compared with laboratory testing at birth were 100% (95% CI 89·4–100·0) and 100% (95% CI 99·8–100·0), respectively. At 4–6 weeks of age, 61 infants were identified as HIV-positive; of these 29 (47·5%) had a positive test at birth. Testing at both birth and 4–6 weeks identified 71 HIV-positive infants compared with 61 infants by testing at 4–6 weeks alone, a 16% increase. Two infants tested positive at birth but tested HIV-negative during follow-up. CONCLUSIONS: Adding POC birth testing to the 4–6 week screen may increase access to HIV diagnosis and expedite ART initiation in primary health care settings within low resource settings. Guidance on appropriate confirmatory HIV testing algorithms for birth testing is needed.