Understanding low sensitivity of community‐based HIV rapid testing: experiences from the HPTN 071 (PopART) trial in Zambia and South Africa

Abstract

Introduction: Population-wide HIV testing services (HTS) must be delivered in order to achieve universal antiretroviral treatment (ART) coverage. To accurately deliver HTS at such scale, non-facility-based HIV point-of-care testing (HIV-POCT) is necessary but requires rigorous quality assurance (QA). This study assessed the performance of community-wide HTS in Zambia and South Africa (SA) as part of the HPTN 071 (PopART) study and explores the impact of quality improvement interventions on HTS performance. Methods: Between 2014 and 2016, HIV-POCT was undertaken within households both as part of the randomly selected HPTN 071 research cohort (Population Cohort [PC]) and as part of the intervention provided by community HIV-care providers. HIV-POCT followed national algorithms in both countries. Consenting PC participants provided a venous blood sample in addition to being offered HIV-POCT. We compared results obtained in the PC using a laboratory-based gold standard (GS) testing algorithm and HIV-POCT. Comprehensive QA mechanisms were put in place to support the community-wide testing. Participants who were identified as having a false negative or false positive HIV rapid test were revisited and offered retesting. Results: We initially observed poor sensitivity (45–54%, 95% confidence interval [CI] 31–69) of HIV-POCT in the PC in SA compared to sensitivity in Zambia for the same time period of 95.8% (95% CI 93–98). In both countries, specificity of HIV-POCT was >98%. With enhanced QA interventions and adoption of the same HIV-POCT algorithm, sensitivity in SA improved to a similar level as in Zambia. Conclusions: This is one of the first reports of HIV-POCT performance during wide-scale delivery of HTS compared to a GS laboratory algorithm. HIV-POCT in a real-world setting had a lower sensitivity than anticipated. Appropriate choice of HIV-POCT algorithms, intensive training and supervision, and robust QA mechanisms are necessary to optimize HIV-POCT test performance when testing is delivered at a community level. HIV-POCT in clients who did not disclose that they were on ART may have contributed to false negative HIV-POCT results and should be the topic of future research.

To err is human, to correct is public health: a systematic review examining poor quality testing and misdiagnosis of HIV status

“…Most studies were carried out in Africa ( n  = 48) [5,7,10–14,16,22–25,29,30,32–34,36,37,39,41,43,44,46,47,49,51,52,54–59,61–65,67–72,74,75], followed by in the Americas ( n  = 7) [28,31,42,50,53,60,66], Asia ( n  = 4) [8,35,45,73] and Europe ( n  = 1) [48]. There were also four multi-country/regional studies [26,27,38,40].…”

“…Nine studies reported more than one unit of measure, and three studies did not specify sample size (see supplementary information). The majority of studies occurred in a facility-based setting; studies carried out in community settings included the workplace ( n  = 1) [57], home-based testing ( n  = 2) [14,39] and a mobile setting ( n  = 1) [32]. …”

“…Thirty-seven studies reported using a suboptimal testing strategy that differed from the WHO recommendations [5,8,11–14,16,22–30,32–34,36–39,42–44,49,51,53,59,62,64–66,68,72,75]. Suboptimal testing strategies included using a highly specific first-line test and highly sensitive second-line test [14,33,39,55], using a single RDT for HIV-positive diagnoses [11,66,72], using a high prevalence testing strategy in a low prevalence setting [16,49], using a parallel testing algorithms and a tiebreaker testing strategy (where a third assay is used to resolve discrepant test results and rule in HIV infection) [5,12,13,16,24,25,27–30,32,34,36,37,68].…”

“…User errors, incorrectly performing the test procedure or incorrectly interpreting results, defined as human errors, were reported in 25 studies [7,8,11,14,26–28,31,33,34,37,40,42,46,52,54,57,60,65–68,70,72,73]. Errors identified included users having difficulty with specimen collection [14,28,68], performing RDTs [31,73], interpreting test results [10,24,27,30,32,40,42,48,62,65,66,74], reading test results too early [7] and not using the correct reagents/buffer [7].…”

“…Errors identified included users having difficulty with specimen collection [14,28,68], performing RDTs [31,73], interpreting test results [10,24,27,30,32,40,42,48,62,65,66,74], reading test results too early [7] and not using the correct reagents/buffer [7]. …”