Could false-positive PCR test results be fueling a Fake Pandemic?
Every diagnostic test approved by the FDA must submit a specificity rating, and the PCR test for detecting COVID-19 is no different. But what does specificity mean, why is it important, and how is this rating not disclosed publicly?
Let's start with an understanding of diagnostic specificity. An article by VeryWellHealth.com explains it in plain language: "Specificity refers to the ability of a test to rule out the presence of a disease in someone who does not have it. In other words, in a test with high specificity, a negative is negative. A test with low specificity can be thought of as being too eager to find a positive result, even when it is not present, and may give a high number of false positives. This could result in a test saying that a healthy person has a disease, even when it is not actually present."
For example, if 1,000 non-infectious healthy people are given a COVID test with a specificity rating of 95%, it would produce 50 false-positive results on average. Here's a great interactive calculator that you can use to see how various ratings effect test results in your community: BMJ Covid-19 Test Calculator
And our friends at the CDC have also prepared two graphical examples to explain false-positives: CDC Calculator Graphic
The CDC states: "Despite the high specificity of antigen tests, false positive results will occur, especially when used in communities where the prevalence of infection is low – a circumstance that is true for all in vitro diagnostic tests. CDC considers low prevalence to be when NAAT positivity over the last 14 days is less than 5% or when there are fewer than 20 new cases of COVID-19 per 100,000 persons within the last 14 days. In general, the lower the prevalence of infection in the community, the higher the rate of false positive test results." (source)
"In general, the lower the prevalence of infection in the community, the higher the rate of false positive test results."
I live in Mendocino County, California where (according to what the CDC describes above) is experiencing a "low prevalence" of the virus. My state's COVID tracker website tells me that the positivity rating for my county is currently at 1.2%, and the state average is 1.7% positivity, which is well below the 5% threshold detailed above.
So how many false-positive test results are there in Mendocino County? To find out that information, we would need the specificity rating of the PCR test being used in my county. I have spent the past 3 days on the phone with the county health officials, county supervisors, the California department of health, and with the testing company hired by the state (Optum Serve). Not a single person will disclose the specificity rating for the test we are using. The response I received from Dr. Andy Coren (Mendocino County Health Officer) was " I do not have this information".
So without that information, we are forced to make an educated guess. I have researched PCR tests that have been given an Emergency Use Authorization by the FDA, and their specificity rating varies from about 95% to 99% (source). To be "fair" I will assume a specificity rating of 98%. So here's what we have: a virus prevalence of 1.2%, a specificity rating of 98%, and an average daily test count of 150 tests. Doing the math gives a likelihood of 3 false-positives on average per day in Mendocino County. Now that might not sound like a lot, but if you look up my county's average daily cases, it has been between 2-4 new cases per day for the past month. So could it be that these "new cases" are actually just false-positives produced by an imperfect test? Is this the "fake pandemic" that we've been hearing about?
If you're looking to help, there are some people you can contact in Mendocino County that should know the specificity rating of the test we use. They won't tell us, but maybe they'll give an answer if more people reach out. You can contact Dr. Andy Coren (Mendocino County Health Officer) at (707) 472-2759 or firstname.lastname@example.org and Ted Williams (Mendocino County 5th District Supervisor) at (707) 937-3500 or email@example.com
We can also do the math on a national level. The positivity rate in the US is currently at 5.2% and we are averaging about 1.5 million tests per day. Assuming an average specificity rating of 98%, there is a likelihood of 28,440 false-positive tests per day. The US is averaging about 70,000 new positive tests per day, so it is possible that 40.6% of our new cases are actually false positives. And that is assuming a specificity rating of 98%, but we know that there are Covid tests with a rating of under 95% which would produce more than double the amount of false-positives than a test with a 98% rating.
More notes on specificity from the CDC: "Sensitivity and specificity of any test for influenza viruses in respiratory specimens might vary by the type of testing method and specific test used, the time from illness onset to specimen collection, the quality of the specimen collected, the respiratory source of the specimen, handling and processing of the specimen, and the time from specimen collection to testing. The post-test probability or predictive values (positive and negative predictive values) of an influenza virus test depend upon the prevalence of circulating seasonal influenza viruses in the patient population, and the specific test characteristics (sensitivity and specificity) compared to a “gold standard” comparison test (molecular assay or viral culture). As with any diagnostic test, results should be evaluated in the context of other clinical and epidemiologic information available to health care providers. Serological testing does not provide timely results to inform clinical management decisions."
When addressing interpretation of test results, the CDC recommends: "In contrast, false-positive [RIDT] results are less likely, but can occur and are more common during periods of low influenza activity. Therefore, when interpreting results of a rapid influenza diagnostic test, clinicians should consider the accuracy of the test in the context of the level of influenza activity in their community"