TABLE 3.
Panel | No. of samples | No. (%) of betacoronavirus IgG-positive samples | A450-A620 cutoff | No. of negative samples, % specificity (95% CI), for ELISA 1 (sera, 1:100) | No. of negative samples, % specificity (95% CI), for ELISA 2 (sera, 1:2) | No. of negative/bl samples, % specificity (95% CI), for Euroimmun NCP IgG |
---|---|---|---|---|---|---|
Germany | 139 | 51 (36.7) | 0.200 | 139, 100.0 (96.8–100.0) | 139, 100.0 (96.8–100.0) | 139, 100.0 (96.8–100.0) |
0.400 | 139, 100.0 (96.8–100.0) | 139, 100.0 (96.8–100.0) | ||||
Ghana A | 131 | 33 (25.2) | 0.200 | 128, 97.7 (93.2–99.5) | 129, 98.5 (94.3–99.9) | 97, 74.0 (65.9–80.8) |
0.400 | 129, 98.5 (94.3–99.9) | 131, 100.0 (96.6–100.0) | ||||
Ghana B | 145 | 34 (23.4) | 0.200 | 142, 97.9 (93.8–99.6) | 145, 100.0 (96.9–100.0) | 132, 91.0 (85.1–94.8) |
0.400 | 145, 100.0 (96.9–100.0) | 145, 100.0 (96.9–100.0) | ||||
Madagascar | 166 | 30 (18.1) | 0.200 | 165, 99.4 (96.3–100.0) | 166, 100.0 (97.3–100.0) | 164, 98.8 (95.4–100.0) |
0.400 | 166, 100.0 (97.3–100.0) | 166, 100.0 (97.3–100.0) | ||||
Nigeria | 149 | 79 (53.0) | 0.200 | 143, 96.0 (91.3–98.3) | 149, 100.0 (97.0–100.0) | 107, 71.8 (64.1–78.4) |
0.400 | 146, 98.0 (94.0–99.6) | 149, 100.0 (97.0–100.0) | ||||
Colombia | 40 | 12 (30.0) | 0.200 | 40, 100.0 (89.6–100.0) | 40, 100.0 (89.6–100.0) | 40, 100.0 (89.6–100.0) |
0.400 | 40, 100.0 (89.6–100.0) | 40, 100.0 (89.6–100.0) | ||||
Lao PDR | 20 | 5 (25.0) | 0.200 | 20, 100.0 (81.0–100.0) | 20, 100.0 (81.0–100.0) | 20, 100.0 (81.0–100.0) |
0.400 | 20, 100.0 (81.0–100.0) | 20, 100.0 (81.0–100.0) |
Serum/plasma samples collected from symptom-free donors before 2019 in Europe, Africa, South America, and Asia were analyzed with SARS-CoV-2 IgG FcγR ELISAs 1 and 2 and the Euroimmun anti-SARS-CoV-2 NCP ELISA (IgG). Specificities and 95% confidence intervals (CIs) were calculated for two alternative cutoff values (A450-A620 = 0.2 and 0.4). bl, borderline.