A Retrospective Look at Covid-19 Testing Performance

Presented by:
Mayla Ward, Leah Albaugh, & Martin Ringman (Western Washington University)

In this research paper, we analyze how the sensitivity (the likelihood that those with the disease test positive), specificity (the likelihood that those without the disease test negative), and disease prevalence affect the Covid-19 testing performance. We use a set of different sensitivity and specificity values with a range of prevalence values to evaluate test performance in terms of positive predictive value (PPV, the likelihood that those who test positive have the disease) and negative predictive value (NPV, the likelihood that those who test negative do not have the disease). Our analysis indicates that the disease prevalence has a significant impact on the sensitivity and specificity values necessary to achieve high PPV and NPV. Furthermore, we find that a prevalence of less than 0.01 (1%) would require an unrealistically high specificity value to achieve a strong PPV.