Do COVID-19 tests exaggerate the number of positive cases?
Claims that PCR tests run too many cycles have traction among skeptics, but don't address the complexities of testing
ROCHESTER, Minn. — Fears that COVID-19 testing has exaggerated the number of infected people were discussed during a recent media call with scientists from Mayo Clinic Laboratories.
Beginning last summer and re-emerging in January, skeptics have asserted that COVID-19 testing uses too many rounds of so-called amplification cycles while searching for viruses.
They liken it to a process of scraping the barrel for signs of illness, over-scrutiny that can ultimately misidentify harmless remnant particles from previous illness or exposure as active infections.
The argument has been deployed to suggest fewer people are contracting COVID-19 than the public is told, that the official case counts are alarmist, and that the public is being given an exaggerated story about the pandemic.
The same message took on political overtones earlier this year when a World Health Organization letter concerning the need for careful adherence to PCR protocols was released, and subsequently misinterpreted in widely shared Facebook posts critical of so-called cycle threshold values.
Some postings implied the WHO letter was timed so that a new U.S. president might inherit fewer cases.
The standard lab method for determining COVID-19 infection does use up to 40 genetic amplification cycles, according to Dr. Matt Binnicker, director of Clinical Virology for Mayo Clinic Laboratories, who spoke to the media recently as part of a periodic update from Mayo Clinic Laboratories.
Binnicker was one of several experts consulted by factcheck.org earlier this year in refuting claims about PCR testing over-sensitivity.
Binnicker says positive cases determined after higher numbers of cycles can indeed pick up only small amounts of virus, but that this process is needed to detect active infections regardless due to the effects of poorly collected samples and a host of other variables that degrade the virus.
"It's basically at what point during the test does it say this (specimen) is positive for the virus," Binnicker said in explaining cycle thresholds. "If the cycle threshold value is high, let's say 35 ... that would indicate there was a really low amount of virus in the sample."
"If the cycle threshold was low, like 10-15 cycles, that would mean there was a lot of virus present, and it didn't take very long before the instrument said 'there's something here.'"
Binnicker adds, however, that far from this being evidence of a false positive, a positive finding of COVID-19 following a high cycle threshold could simply reflect the timing of the collection, as opposed to weak or fragmentary levels of virus.
"For an individual who has just started to develop symptoms," he says, "... it might imply they are early in their disease course, and that if we tested them a day or two later" the sample would come back positive at a lower level of cycles.
Binnicker said other variables explaining a high cycle threshold include, whether the nurse or physician "did a really good job collecting the sample," how the sample was collected, and which among the ten tests used at Mayo was performed.
"We can't necessarily assume someone with a high cycle threshold is not infectious," echoed Dr. Bobbi Pritt, chairperson for the Division of Clinical Microbiology at Mayo Clinic.
Pritt said the cycle threshold needed to find infection could vary depending on whether a sample was collected from the throat, via a nasopharyngeal swab, as well as the transport media used, amount of specimen collected, and duration from time of collection.
Pritt described a paper she co authored last summer for the College of American Pathologists in which known amounts of identical viral specimens were sent to more than 700 different laboratories across the U.S., returning "a difference in cycle threshold values of more than 12 cycles," she said.
"Some publications have proposed a cycle threshold should be used — say, 31. But if that same specimen were sent to different labs, 31 could become 41, or it could come back 21."
Can PCR tests determine infectiousness?
A separate question relating to PCR cycles, Binnicker says, is whether a positive result triggered by a low number of cycles can be used to single out those who are more infectious than others. If so, that could be used to identify so-called "super-spreaders."
Binnicker says he opposes that use of the tests for the same reason high PCR cycle thresholds can't be considered determinative of over diagnosis — the interference of other collection variables.
Since developing one of the first COVID-19 diagnostic tests following the Centers for Disease Control and Prevention test in March of 2020, Mayo Clinic has run more than 560,000 diagnostic COVID-19 tests for patients in southeastern Minnesota, identifying over 35,000 cases in the process.
Areas of the Mayo Clinic Laboratories usually used for HIV and hepatitis testing were reassigned for COVID-19 testing during the early weeks of the pandemic, and "went from 22 to over 200 employees," virtually overnight, according to Pritt. A lab that previously ran during business hours only became one running tests on COVID-19 samples 24 hours a day, seven days a week.
The clinic is currently developing an array of gene sequencing tests to determine the specific variant of COVID-19 contracted, for purposes of public health surveillance as well as patient care, although current tests can detect every variant. A testing issue that remains on the horizon is the potential for antibody tests to one day be utilized as proof of safety for travel.
Currently, proof of vaccination is the prevailing mechanism envisioned as a COVID-19 "passport." That said, serology tests can definitively show if a person is immune to the illness.
Binnicker says some serology tests can provide proof of vaccination, while other serology tests can provide proof of previous illness
"It's one layer of information some countries may require (for travel)," he says. "My own personal opinion is that the results of serology tests are sometimes difficult to interpret," he adds.
"If someone is negative for antibodies, that doesn't mean they aren't protected from COVID-19, as there are other immune responses besides antibodies that can mean someone is protected, like T cells."
That said, Binnicker believes negative PCR and molecular tests — as well as vaccine status — will all become more relevant as nations begin efforts to reopen travel.