BACKGROUND: During the COVID-19 pandemic, the US President's Emergency Plan for AIDS Relief supported oral caregiver-assisted HIV self-testing (CG-HIVST) to address the gap in HIV diagnosis of children. We aimed to investigate caregiver uptake, results return, acceptability, and potential social harms of CG-HIVST. METHODS: This prospective, interventional, study was done at 32 health facilities in Uganda and 15 health facilities in Zambia. Caregivers aged 18 years and older (plus emancipated minors aged 15-17 years in Uganda) living with HIV who were currently accessing HIV care and considered index cases, with no positive responses to an intimate partner violence screen, and with one or more children aged 18 months to 14 years with unknown HIV status were eligible to participate. Eligible caregivers were offered oral HIVST kits to screen their children and primary outcomes were described by caregiver and child characteristics. Following HIVST kit administration, caregivers were surveyed using a standardised questionnaire to document their perceptions, adverse events, and social harm. Primary outcomes were the uptake of HIVST and the number and proportion of returned screening test results, reactive results, reactive screens with confirmatory HIV testing, confirmatory testing with a positive result, and children who were confirmed HIV-positive who were linked to treatment. This study was registered with ClinicalTrials.gov, NCT04774666 and NCT04754386, and is completed. FINDINGS: From Feb 1 to Oct 31, 2021, 12 998 interested caregivers were screened for eligibility, 4023 of whom were eligible. 3903 (97·0%) accepted HIVST kits to screen their child for HIV (1609 [41·2%] in Zambia and 2294 [58·8%] in Uganda). Among caregivers, 3094 (79·3%) of 3903 were female, and 809 (20·7%) were male. 7601 children were enrolled (3779 [49·7%] were female and 3822 [50·3%] were male). 4766 (97·9%) of 4866 test results were returned in Uganda and 2647 (96·8%) of 2735 in Zambia. 119 (1·6%) of 7413 children had reactive HIVST results, requiring confirmatory testing. Of 116 children with confirmatory testing, 43 were confirmed HIV-positive (HIV prevalence 0·7% [n=32] in Uganda and 0·4% [n=11] in Zambia) and 100% were linked to antiretroviral therapy. Adverse events were rare (11 [0·4%] of 2720) and minor, and there were no reports of social harm or violence. Caregivers surveyed reported the HIVST kit was easy to use (2637 [97·0%] of 2718), they would use it again (2650 [99·1%] of 2674), and they would recommend it to other parents (2615 [97·8%] of 2674). INTERPRETATION: Our findings suggest that oral CG-HIVST is acceptable, feasible, and safe, with no reports of social harm, and has the potential to expand access to HIV testing for children while reducing the service delivery burden on health facilities. FUNDING: US President's Emergency Plan for AIDS Relief and Wellcome Trust.

BACKGROUND: :To inform optimal management of HIV viremia on TLD, we examined viral load (VL) outcomes of a large cohort of adult PLHIV on TLD in Nigeria. METHODS: :We conducted a retrospective study of adult PLHIV who had ≥1 VL after initiating TLD during January 2017-February 2023. VLs were categorized as undetectable (≤50 copies/mL), low low-level viremia (LLV, 51-199 copies/mL), high LLV (200-999 copies/mL), virologic nonsuppression (VLNS, ≥1000 copies/mL), and virologic failure (VF, ≥2 consecutive VLNS results). Among patients with ≥2 VLs on TLD, we described how viremia changed over time and examined virologic outcomes after VF. We identified predictors of subsequent VLNS using mixed-effects logistic regression and conducted planned contrasts between levels of VL result and regimen types. RESULTS: :Analysis of 82,984 VL pairs from 47,531 patients demonstrated viral resuppression to ≤50 copies/mL at follow-up VL in 66.7% of those with initial low LLV, 59.1% of those with initial high LLV, and 48.9% of those with initial VLNS. Of 662 patients with a follow-up VL after VF, 94.6% stayed on TLD; of which 57.8% (359/621) were undetectable at next VL without regimen change. Previous low LLV (aOR 1.74, 1.56-1.93), high LLV (aOR 2.35, 2.08-2.65), and VLNS (aOR 6.45, 5.81-7.16) were associated with increasingly higher odds of subsequent VLNS, whereas a previously undetectable VL (aOR 1.08, 0.99-1.71) on TLD was not. CONCLUSIONS: :Despite increased odds of subsequent VLNS, most PLHIV with detectable viremia on TLD, including those with VF, will resuppress to an undetectable VL without a regimen change.