Share on Facebook Share on Twitter

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Rapid antigen tests
#1
People still use the term "COVID test" without specifying what kind of a test it is.   Those that want results to be negative (regardless of infection) say they want positive confirmation.   To me, for PCR tests, that seems unwarranted.   I suspect some may use techniques to interfere with this testing.

If you acquire a rapid antigen test (BinaxNOW was $14 at Walmart when I bought some last week), you will see that it is actually two tests in a box.  The tests are to be taken 1.5 to 3 days apart.  If EITHER test is positive, then you almost definitely have COVID.  If BOTH tests are negative, then likely you don't have COVID. (What does "likely" mean? See below)

While the test has an EUA only for symptomatic individuals, the disease has a high percentage of asymptomatic infections.  So it is being used to detect those people.

Here are a couple of studies comparing the results of simultaneous testing by BinaxNOW and PCR-RT.
 1. "Performance characteristics of a rapid SARS-CoV-2 antigen detection assay at a public plaza testing site in San Francisco"
 2. "Effectiveness of Abbott BinaxNOW Rapid Antigen Test for Detection of SARS-CoV-2 Infections in Outbreak among Horse Racetrack Workers, California, USA"

First, a bit about PCR tests.   While most people just see a thumbs-up or thumbs-down result for PCR tests (yep, you have COVID; nope, you don't), the PCR test actually can give a Ct (cycle threshold) score that basically gives a measure of the amount of virus in the sample  (note that the process of taking the sample may impact whether that sample is indicative of the virus in the body).   The higher the Ct score, the less virus.  

As I understand it, no one knows how much virus it takes to infect others around you, or (more probably) the amount of time for you to infect someone around you. However, study 1 asserts without reference "we defined a threshold for high virus levels corresponding to the range thought to be the most transmissible: a cycle threshold of 30".

The first study (in SF) has images showing BinaxNOW results on lab-grown samples (Figure 1) at various levels of virus. The also have an image (Figure 2) showing the positive/negative results on 26 people who tested positive by PCR-RT ranked by Ct values. These show the BinaxNOW test always (N=10) negative for Ct>34 and always (N=13) positive for Ct<27.

One could say that the BinaxNOW properly indicated an infection in 16 of 27 (59%) people with PCR-RT confirmed COVID. But I now see that as too simplistic. The positive predictive agreement (59%) will vary depending on the distribution of the levels of virus in the tested population. IF (and I'm not sure it is correct) the apparently less-severe nature of Omicron in the vaccinated population means that people have lower levels of virus, then BinaxNOW probably will have a much lower agreement with PCR-RT testing.

The second study (Racetrack) had more disagreement when comparing BinaxNOW vs Ct. See their Figure 1. One test with a Ct score of 17 was a negative BinaxNOW results.

In both studies, two samples (for BinaxNOW and PCR) were taken by the same person, so that shouldn't matter. Perhaps which sample was taken first might matter. So, why the difference? One might be in the interpretation of the resultant BinaxNOW test. The first study established a standard for reading the results that is more formal than what is in the test kit. The first study also took images of the results and used a colorimeter to evaluate the result.

I'll also mention that when I was looking for instructions on how to perform the BinaxNOW test, I first found some training video produced by some firefighters. I figured they would be good. Well, no. They inserted the swabs deeper than recommended, and then failed to run the swabs around the inside of the nose but instead just twirled the stick. Sigh...
Reply


Messages In This Thread
Rapid antigen tests - by M_T - 12-24-2021, 11:56 AM
RE: Rapid antigen tests - by M_T - 12-24-2021, 12:50 PM

Forum Jump:


Users browsing this thread: