Appropriate use of PCR needed for a focused response to the pandemic

On January 13, 2021, the World Health Organization (WHO) issued a little-noticed technical report that calls into question many of the policies that we have adopted to control the spread of the SARS-CoV-2 virus. At the heart of all the data on COVID-19 cases is the reverse transcriptase-polymerase chain reaction (RT-PCR) test, which can detect the virus in nasal swabs. The test is used, sensibly, to guide medical treatment for patients who display COVID-19 symptoms. Less sensibly, it is often used to identify people with no symptoms at all as positive “cases” and impose long quarantine periods on them.

The WHO’s guidance on the RT-PCR test emphasizes two things that have long been known in the scientific literature and public health practice but inexplicably ignored in COVID policy for almost a year. First, they point out that a positive COVID test does not necessarily mean that someone has any capacity of infecting someone else with the virus. Therefore, it instructs laboratories to report a key statistic that indicates how likely a positive test result actually constitutes infectious COVID-19. And second, the WHO warns against relying on a single test for patients without clinical COVID-19 symptoms.

Regarding infectiousness, the PCR test is not designed to identify active infectious disease but rather genetic material (dead, alive or partial) from the virus. PCR amplifies this material in samples to find traces of COVID-19, so while it often identifies people with active, infectious disease, it can also indicate people as “positive” erroneously. Dead COVID-19 RNA in the nose or mouth of someone who was never sick could create a positive PCR result. Recovered patients who test negative and are non-infectious can still come up positive repeatedly in the following months. These are neither new cases nor infectious ones needing quarantine but could be incorrectly counted as such.

Multiple studies show that the number of amplification or duplication “cycles” performed via PCR to amplify the sample has a relationship with infectiousness — at a certain point, the more cycles needed to get positivity from a sample the generally less viral replication the sample shows. An article in the journal Clinical Infectious Diseases found that among positive PCR samples with a cycle count over 35 – a common lab occurrence – only 3 percent of samples showed viral replication. Simply put, the cycle number next to the test is associated with the chances of infectiousness and should be available to the patient and the public. Thankfully, the WHO now explicitly recommends this.

Setting aside the critical issue of purely false positives from contamination, reporting errors and lab deficiencies, there are many dangers of these “weak” positive diagnoses (as the WHO terms them), which can waste resources and upend a family’s life. Among the harms include the obvious financial harm from needless isolation, social harm from misdirected public resources and psychological harm stemming from a positive diagnosis of COVID, and the stigma associated with it. And since the falsely identified “case” is not actually infectious, there is no corresponding benefit in slowing disease spread by a forced quarantine period for the person and their contacts.

Dr. Anthony Fauci himself told This Week in Virology in July, “If you get a cycle threshold of 35 or more … the chances of it being replication-competent are minuscule.” Why then has our national testing standard never reflected this? PCR providers should work with other labs to perform a random viral culture on those who received positive results, to validate their tests in terms of being an indicator of infectiousness. Other states should emulate Florida in requiring laboratories to report cycle times to providers and to public health officials so they can provide better advice to patients and make more nuanced decisions about mandatory quarantine orders.

The WHO’s second recommendation makes good sense as well. When the PCR test reports a weakly positive signal – a sample that requires so many amplification cycles that the likelihood of an infectious sample is unlikely – there should be an additional test conducted on the same patient before any determination by either clinicians or public health officials about COVID-19 infection is made.

There are two possibilities indicated by a weak positive test. First, the patient is in the early stages of infection and the number of copies of the virus is growing. In that case, a second test, with a sample drawn 24 hours later, should produce a stronger positive signal. Second, if the patient is not infectious, a second test will register either the same weak positive signal or a negative signal. We can thus overcome the limitations of false positivity by repeat testing by not exclusively relying on single PCR results, by combining results with symptom information for clinical decision-making, and considering risk factors.

The questionable quality of COVID-19 testing data, alongside loose COVID-19 case definitions, means we cannot effectively work to build an efficient, effective system to address the disease at the individual or community level. Data that are available now are inadequate to inform local and statewide policymakers, business owners, school administrators and the public at large. As we collectively understand trends in infectiousness and fine tune our testing system, we can act more effectively and efficiently, and reduce disruption to people’s lives, schools and businesses. 

Tom Nicholson holds an appointment as associate in research at Duke University’s Sanford School of Public Policy and the Duke Center for International Development. He is executive director of a global health non-profit organization, Advance Access & Delivery, which supports airborne infectious disease programs around the world and at the U.S.-Mexico border. In this capacity he is also an advisor to the United Nations Office for Project Services.

Jay Bhattacharya, MD, PhD is a professor of medicine at Stanford University and a research fellow at the National Bureau of Economics Research. His research work focuses on the epidemiology of COVID-19 and on effects of COVID-19 lockdown policy. He is the co-author of the Great Barrington declaration.