Effect of point-of-care CD4 cell count results on linkage to care and antiretroviral initiation during a home-based HIV testing campaign: a non-blinded, cluster-randomised trial

Summary Background HIV disease staging with referral laboratory-based CD4 cell count testing is a key barrier to the initiation of antiretroviral treatment (ART). Point-of-care CD4 cell counts can improve linkage to HIV care among people living with HIV, but its effect has not been assessed with a randomised controlled trial in the context of home-based HIV counselling and testing (HBCT). Methods We did a two-arm, cluster-randomised, controlled efficacy trial in two districts of western Kenya with ongoing HBCT. Housing compounds were randomly assigned (1:1) to point-of-care CD4 cell counts (366 compounds with 417 participants) or standard-of-care (318 compounds with 353 participants) CD4 cell counts done at one of three referral laboratories serving the study catchment area. In each compound, we enrolled people with HIV not engaged in care in the previous 6 months. All participants received post-test counselling and referral for HIV care. Point-of-care test participants received additional counselling on the result, including ART eligibility if CD4 was less than 350 cells per μL, the cutoff in Kenyan guidelines. Participants were interviewed 6 months after enrolment to ascertain whether they sought HIV care, verified through chart reviews at 23 local clinics. The prevalence of loss to follow-up at 6 months (LTFU) was listed as the main outcome in the study protocol. We analysed linkage to care at 6 months (defined as 1–LTFU) as the primary outcome. All analyses were by intention to treat. This trial is registered at ClinicalTrials.gov , number NCT02515149. Findings We enrolled 770 participants between July 1, 2013, and Feb 28, 2014. 692 (90%) had verified linkage to care status and 78 (10%) were lost to follow-up. Of 371 participants in the point-of-care group, 215 (58%) had linked to care within 6 months versus 108 (34%) of 321 in the standard-of-care group (Cox proportional multivariable hazard ratio [HR] 2·14, 95% CI 1·67–2·74; log rank p