#15,260
The two main measures of the accuracy of any diagnostic test are sensitivity and specificity.
- Sensitivity is defined as the ability of a test to correctly identify individuals who have a given disease or condition.
- Specificity is defined as the ability of a test to exclude someone from having a disease or illness (false positives).
It is also worth noting, that after more than 7 years of research, as of last October no fully validated laboratory antibody test had been developed for MERS-CoV (see EID Journal: Sensitivity and Specificity Of MERS-CoV Antibody Testing).Understandably - as nations look to rollback their lockdowns and revive their economies - much hope has been pinned on the wide availability of rapid testing for active SARS-CoV-2 infection, and for antibodies that indicate past infection and likely immunity.
While we continue to hear that testing capacity is rapidly growing, the accuracy of these tests is rarely mentioned. Since many of these recently deployed tests have not yet been verified or approved by the FDA, we only have the manufacturer's estimate of sensitivity and specificity.
The early problems with the CDC's test kits are well known (see Problematic Lab Testing For The Novel Coronavirus), and in recent weeks we've seen disturbing reports of less-than-reliable antibody tests for COVID-19 being sold, both in the United States, and around the world (see New Test Hopes Dashed as U.K. Finds Antibody Kits Don’t Deliver).All of which brings us to a new study, published yesterday in the Annals of Internal Medicine, that finds a high rate of false negative results, particularly early in the pre-symptomatic phase of infection.
However, even on the 1st day of symptoms, the false-negative rate was 38%.Even more sobering, three days into symptoms the probability of seeing a false-negative result was still 20%, and began to rise again by the 5th day. The authors caution: `Care must be taken in interpreting RT-PCR tests for SARS-CoV-2 infection—particularly early in the course of infection. . .'
Follow the link to read the full report.
Original Research 13 May 2020
Variation in False-Negative Rate of Reverse Transcriptase Polymerase Chain Reaction–Based SARS-CoV-2 Tests by Time Since Exposure
Lauren M. Kucirka, MD, PhD, Stephen A. Lauer, PhD, … View all authors
Author, Article and Disclosure Information
https://doi.org/10.7326/M20-1495
A second study, this time published on the BioRxiv preprint sever for Biology (a non-peer-reviewed site), looks at the accuracy of the highly touted Abbott ID NOW machine. Their results (which are disputed by the manufacturer), also suggest a less-than-stellar sensitivity rate for the Abbott test.
First a link to (and abstract from) this study, followed (in fairness) by Abbott Lab's full response.
Performance of the rapid Nucleic Acid Amplification by Abbott ID NOW COVID-19 in nasopharyngeal swabs transported in viral media and dry nasal swabs, in a New York City academic institution
Atreyee Basu, Tatyana Zinger, Kenneth Inglima, Kar-mun Woo, Onome Atie, Lauren Yurasits, Benjamin See, Maria E. Aguero-Rosenfeld
doi: https://doi.org/10.1101/2020.05.11.089896
This article is a preprint and has not been certified by peer review [what does this mean?]
Abstract
The recent emergence of the SARS-CoV-2 pandemic has posed formidable challenges for clinical laboratories seeking reliable laboratory diagnostic confirmation. The swift advance of the crisis in the United States has led to Emergency Use Authorization (EUA) facilitating the availability of molecular diagnostic assays without the more rigorous scrutiny to which tests are normally subjected to prior to FDA approval.
The need to identify the COVID-19 positive cases quickly and accurately has propelled the release of a variety of assays intended to meet the urgent demand. Several Nucleic Acid Amplification Tests (NAAT) platforms are currently available. Our laboratory currently uses two real time RT-PCR platforms, the Roche Cobas SARS-CoV2 and the Cepheid Xpert Xpress SARS-CoV-2. Both platforms demonstrate comparable performance; however the run times for each assay are 3.5 hours and 45 minutes, respectively.
In search for a platform with shorter turnaround time, we sought to evaluate the recently released Abbott ID NOW COVID-19 assay which is capable of producing positive results in as little as 5 minutes. We present here the result comparisons between Abbot ID NOW COVID-19 and Cepheid Xpert Xpress SARS-CoV-2 using nasopharyngeal swabs transported in VTM as well as dry nasal swabs for the Abbott assay.
Regardless of method of collection and sample type, Abbot ID NOW COVID-19 missed a third of the samples detected positive by Cepheid Xpert Xpress when using NP swabs in VTM and over 48% when using dry nasal swabs.
The Abbott response disputes these findings, citing a 91% accuracy in a recent study conducted by Washington State University.
ABBOTT RESPONSE TO THE NYU STUDY
MAY 13 2020
The NYU study results are not consistent with other studies. For example, in another recent study in Washington state, out of approximately 1,000 subjects there were 23 positive for COVID-19 on a lab test and ID NOW picked up 21, demonstrating 91% sensitivity and 100% specificity. We have many questions for the study authors.
While no test is perfect, Abbott's ID NOW is delivering reliable results when and where they're needed most. The test is performing as expected by the more than 1,000+ sites using ID NOW for COVID-19.
The world needs a variety of tests in labs and at point of care, and as many as possible, if we are to help reduce the risk people have every single day of contracting the virus. ID NOW is an important tool in that equation. Risk reduction is the goal, which is why we're developing as many tests as we can across all of our diagnostics platforms.
Even at a respectable 91% sensitivity rate, a negative test can't be used to rule out infection. In presymptomatic, or mildly symptomatic cases, viral shedding may be below detectable levels, and sample collection methods and swab types may further degrade accuracy.
None of these problems are new, or exclusive to SARS-CoV-2. We've seen similar problems with laboratory testing for MERS (see AJIC:Intermittent Positive Testing For MERS-CoV), and avian flu as well.COVID-19 testing continues to improve, and while their results may never become 100% reliable, when they are viewed in context with the current community spread of the virus - and the patient's risk factors and physical presentation - they can be a highly effective tool for both diagnosis and pandemic response.
But despite the hype and the hope, testing alone isn't going to be a panacea for a pandemic.
