[BostonCard]

Control serum samples included archived anonymous serum samples obtained from healthy blood donors 18-64 years with no history of SARS-CoV-2 infection (n = 10) and no recent travel history, sera from patients with acute viral respiratory tract infections caused by other coronaviruses (n = 5) or non-coronaviruses (n = 45), and sera from patients positive for dengue virus (n = 9), cytomegalovirus (CMV; n = 2) and Epstein Barr virus (EBV; n = 10). One patient was positive for both CMV and EBV.

By my count that's about 80 samples. Assuming that by 100% specificity they mean that 0 of the 80 were false positives, the 95% confidence interval around that is 95.5% to 100%. If you are using this as a screening test, assuming a low prevalence, you probably need to establish at least a 99% specificity, for which you would need at least 300 samples.

BC

[BostonCard]

Control serum samples included archived anonymous serum samples obtained from healthy blood donors 18-64 years with no history of SARS-CoV-2 infection (n = 10) and no recent travel history, sera from patients with acute viral respiratory tract infections caused by other coronaviruses (n = 5) or non-coronaviruses (n = 45), and sera from patients positive for dengue virus (n = 9), cytomegalovirus (CMV; n = 2) and Epstein Barr virus (EBV; n = 10). One patient was positive for both CMV and EBV.

By my count that's about 80 samples. Assuming that by 100% specificity they mean that 0 of the 80 were false positives, the 95% confidence interval around that is 95.5% to 100%. If you are using this as a screening test, assuming a low prevalence, you probably need to establish at least a 99% specificity, for which you would need at least 300 samples.

BC