Skip to main content
. 2022 Nov 8;10(6):e03923-22. doi: 10.1128/spectrum.03923-22

TABLE 2.

Overall diagnostic performance of RT-PCR and rapid antigen test performed on nasopharyngeal or salivary samplesa

Patient group Diagnostic test Prevalence (%) Sensitivity [95% CI] (%)b Specificity [95% CI] (%) PPV [95% CI] (%) NPV [95% CI] (%)
Health care workers (n = 453) Nasopharyngeal RT-PCR 38 98 [94 to 99] n.a.c n.a.c 98 [96 to 99]
Salivary RT-PCR 34 87 [81 to 91] n.a.c n.a.c 92 [89 to 95]
Nasopharyngeal RAT 26 66 [59 to 73] 100 [98 to 100] 100 [97 to 100] 100 [78 to 86]
Salivary RAT 1 2 [0.5 to 5] 99 [97 to 100] 60 [23 to 93] 61 [56 to 65]
Hospitalized patients (n = 22) Nasopharyngeal RT-PCR 68 94 [72 to 99] n.a.c n.a.c 86 [49 to 99]
Salivary RT-PCR 50 69 [44 to 86] n.a.c n.a.c 54 [28 to 79]
Nasopharyngeal RAT 23 31 [14 to 56] 100 [61 to 100] 100 [57 to 100] 35 [17 to 59]
Salivary RAT 4.5 6 [1 to 28] 100 [61 to 100] 100 [5 to 100] 29 [14 to 50]

aCI, confidence interval; PPV, positive predictive value; NP, nasopharyngeal; NPV, negative predictive value, n.a., not applicable; RAT, rapid antigen testing; RT, reverse transcription.

bDiagnostic performances were evaluated based on a composite outcome considering NP PCR and/or saliva PCR as a COVID-19 case.

cCalculation of specificity and PPV is not applicable, since a positive NP PCR and/or saliva PCR are part of the definition of the composite outcome.