Clinical Sensitivity, Specificity and Epidemiology of SARS-CoV-2 Serological Testing Using the Biozek COVID-19 Test

Abstract

Background: Clinical validation using the Biozek COVID-19 test including sensitivity and specificity and associated patient-reported symptoms with SARS-CoV-2 seropositivity. Methods: 316 sera were analyzed including 47 hospitalized cases, 50 mild cases and 219 negative controls. Results were read visually by two technicians and in case of discrepancy by a third. Models were created between independent variables and IgG seropositivity using multivariable logistic regression analysis. Results: Sensitivity of both IgM and IgG together for hospitalized patients at all time periods was 68.1% (32/47) and 90.0% (27/30) after 10 days or more. From mild/asymptomatic cases the combined IgM and IgG sensitivity was 92.0% (46/50) and 91.8% (45/49) after 10 days or more. In the group of non-COVID-19 cases, the overall specificity was 99.1% (217/219). For IgG alone, the specificity was 99.5% (218/219). In the multivariable analysis loss of smell remained the strongest associated variable with an odds ratio (95%CI): 6.82 (5.61–8.31), p-value < 0.001. Our final prediction model yielded a ROC-AUC of 0.77 (0.74–0.81) showing acceptable discrimination. Conclusions: The Biozek COVID-19 test showed high specificity and good sensitivity 10 days after the first sickness day. Solely IgM positive tests must be interpreted with caution and preferably excluded. In order to capture most symptomatic COVID-19 cases, loss of smell should be included within symptomatic screening policies.

Keywords

• SARS-CoV-2
• COVID-19
• epidemiology
• validation
• loss of smell