Accuracy and usability of saliva and nasal rapid antigen self-testing for detection of SARS-CoV-2 infection in the general population: a head-to-head comparison

Abstract

BackgroundSARS-CoV-2 self-tests may lower the threshold of testing and produce a result quickly. This could support the early detection of infectious cases and reduce further community transmission. However, the diagnostic accuracy of (unsupervised) self-testing with rapid antigen diagnostic tests (Ag-RDTs) is mostly unknown. We therefore conducted a large-scale head-to-head comparison of the diagnostic accuracy of a self-performed SARS-CoV-2 saliva and nasal Ag-RDT, each compared to a molecular reference test, in the general population in the Netherlands.MethodsIn this cross-sectional study we consecutively included individuals aged 16 years and older presenting for SARS-CoV-2 testing at three Dutch public health service test sites irrespective of their indication for testing, vaccination status, and symptomatology. Participants were sampled for molecular testing at the test site and received two self-tests (the Hangzhou AllTest saliva self-test and the SD Biosensor nasal self-test by Roche Diagnostics) to perform at home within a few hours without knowledge of their molecular test result. Information on presence and type of symptoms, user experiences, and results of both self-tests were collected via an online questionnaire. For each self-test, sensitivity, specificity, positive and negative predictive values were determined with molecular testing as reference standard.FindingsThe SARS-CoV-2 molecular reference test positivity rate was 6.5% in the 2,819 participants. Overall sensitivities with 95% confidence intervals were 46.7% (85/182; 39.3%-54.2%) for the saliva Ag-RDT, and 68.9% (124/180; 61.6%-75.6%) for the nasal Ag-RDT. With a viral load cut-off (≥5.2 log10 SARS-CoV-2 E-gene copies/mL) as a proxy of infectiousness, sensitivities increased to 54.9% (78/142; 46.4%-63.3%) for the saliva Ag-RDT and 83.9% (120/143; 76.9%-89.5%) for the nasal Ag-RDT.For the nasal Ag-RDT, sensitivities were 78.5% [71.1%-84.8%] and 22.6% [9.6%-41.1%] in those with and without symptoms at the time of sampling, which increased to 90.4% (113/125; 83.8%-94.9%) and 38.9% (7/18; 17.3%-64.3%) after applying the viral load cut-off. In those with and without prior confirmed SARS-CoV-2, sensitivities were 36.8% [19/372; 16.3%-61.6%] and 72.7% [161/2437; 65.1%-79.4%] for the nasal Ag-RDT, which increased to 100% (7/7; 59.0%-100%) and 83.1% (113/126; 75.7%-89.0%) after applying the viral load cut-off.The diagnostic accuracy of the nasal Ag-RDT did not differ by COVID-19 vaccination status, sex, and age. Specificities were >99%, positive predictive values >70% and negative predictive values >95%, for the saliva Ag-RDT, and >99%, >90%, and >95% for the nasal Ag-RDT, respectively, in most analyses.Interpreting the results was considered (very) easy for both self-tests.InterpretationThe Hangzhou AllTest self-performed saliva Ag-RDT is not reliable for SARS-CoV-2 infection detection overall nor in the studied subgroups. The SD Biosensor self-performed nasal Ag-RDT had high sensitivity in individuals with symptoms and in those without a prior SARS-CoV-2 infection. The overall accuracy in individuals with symptoms was comparable to that found in previous studies with professional sampling for this Ag-RDT. The extremely low sensitivity of the nasal Ag-RDT in asymptomatic individuals and in individuals who had had a prior SARS-CoV-2 infection is an important finding and warrants further investigation.FundingDutch Ministry of Health, Welfare, and Sport.

Accuracy of COVID-19 self-tests with unsupervised nasal or nasal plus oropharyngeal self-sampling in symptomatic individuals in the Omicron period

“…Our pre-Omicron studies, and when less than 5% of participants were confirmatory testers, found sensitivities of 72-83% for three different Ag-RDTs when performed by trained professionals, and 78.5% for the SD Biosensor Ag-RDT when performed with unsupervised self-sampling from the nose. 7,18,19 The sensitivities that we found in the first week of the current study, when Delta was still highly dominant, were similar (Flowflex 87%, MPBio 80%, and Clinitest 83%), although the percentage of confirmatory testers in this study was much higher (21%-24%). However, sensitivities declined to 80%, 70%, and 70%, respectively, in the Omicron period.…”

“…However, we observed similar trends for the BD-Veritor and SD-Biosensor Ag-RDTs with either professional or self-sampling in two previous studies. 5,7 The lower sensitivities of Ag-RDTs in individuals with a prior infection may be explained by generally lower viral loads, with some individuals potentially carrying viral RNA in the absence of a productive infection (i.e. no viral antigen production).…”

“…In our previous SD-Biosensor (Roche) SARS-CoV-2 nasal Ag-RDT self-testing study, we observed a sensitivity of 79% in symptomatic participants. 7 For the present study, we conservatively assumed sensitivities of 70% for all three Ag-RDTs irrespective of self-sampling method, with an error margin of 5%, type I error of 5% and power of 80%. Hence, we aimed for 335 positive RT-PCR tests per Ag-RDT and per self-sampling method.…”

“…6 We previously showed that the SD Biosensor by Roche Diagnostics Ag-RDT with unsupervised nasal selfsampling had a 78.5% sensitivity in symptomatic individuals. 7 However, since end November 2021, the Omicron variant rapidly replaced the Delta variant. Performance of Ag-RDTs for Omicron might be different due to alterations in viral proteins and infection dynamics.…”

Diagnostic accuracy of SARS-CoV-2 rapid antigen self-tests in asymptomatic individuals in the Omicron period: cross sectional study

“…Sensitivities of Ag-RDT -both with professional sampling and self-test sampling -may be lower in asymptomatic individuals than in symptomatic individuals due to differences in viral load distributions. (1,4,11) In addition, it might be more difficult for asymptomatic individuals to self-retrieve sufficient nasal fluid by selfswabbing because of the absence of rhinorrhea. Addition of oropharyngeal self-sampling to the currently applied nasal self-sampling might have led to higher Ag-RDT sensitivities (2,12), but it is very unlikely that the sensitivities would have reached acceptable levels.…”