Hopes of a COVID-19 summer pardon have been dashed as cases rose in June. As infections increase, so does the need for testing.
In late June, Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, announced that health officials were considering conducting pooled tests for COVID-19 in response to the recent increase. On July 18, the Food and Drug Administration issued an emergency use authorization to Quest Diagnostics DGX,
That would allow the laboratory analysis company to pool samples from up to four patients to detect the coronavirus.
So what is pooled evidence and why does the US need it?
Test groups, not individuals
The basic idea of pooled testing is that it allows public health officials to evaluate small groups, called groups, of people who use only one test. This means that you can evaluate more people faster, using fewer tests and for less money.
Instead of testing one person at a time, samples from multiple people would be mixed and analyzed as one. If the test is negative, everyone in the group is clean. If positive, each group member is tested individually.
For example, imagine a workplace with 20 people, one of whom is infected. The pooled tests divide the 20 people into four groups of five. Samples are taken from each person and mixed with samples from other people in their group. Each pooled sample is then analyzed, four tests in total.
One of the groups will be positive because it contains only one infected person. The five people in this group are reassessed individually and the sick person is identified.
In total, nine tests are used and the 20 workers have been selected. If they had all been examined individually, 20 tests would have been performed.
In practice, the number of tests saved by grouping varies according to the fraction of infected people. Higher infection rates mean that more groups are positive again, more people need to be retested, and group savings are lower.
For example, pooled tests on suspected COVID-19 cases in Los Angeles, where as of June 27 nearly 9% of those tested are positive, would reduce the use of the test by approximately 50%. But in Montana, where only 0.1% of the population is infected, a joint testing effort to assess the entire state would reduce test use by almost 90%.
No matter the exact amount, at a cost of about $ 100 per test, the savings add up.
Works?
Pooled tests have long been used to detect disease in large, asymptomatic populations. State laboratories have used it to detect chlamydia and gonorrhea, and the Red Cross has used it to test donated blood for hepatitis B and C, Zika virus, and HIV.
And some places have already started using pooled tests to detect active coronavirus infections.
Stanford University used pooled tests to monitor the early spread of COVID-19 in the Bay Area. The Nebraska Public Health Laboratory collected and analyzed samples from groups of five and roughly doubled the number of suspected COVID-19 patients that it could test weekly.
Some people have expressed concern that dilution of pooling samples could reduce the precision of PCR testing for viral RNA in samples. Fortunately, researchers have already shown that pooled tests are as accurate as individual tests in groups of up to eight people.
Another concern is the prevalence of infection. When infection rates rise above 15%, too many pools are positive and the pool is no longer efficient. Fortunately, nationally, only about 5% of tests in suspected cases are yielding positive results. Even in extreme hot spots like Texas, where 10% or more of suspected cases test positive, pooled tests could still be helpful.
But the true strength of pooled testing would be seen if health officials begin to implement large-scale tests of the general population.
In most cases, patients are currently evaluated only when a healthcare provider suspects that they may be infected with SARS-CoV-2. However, asymptomatic and presymptomatic patients with COVID-19 are common and are probably an important source of COVID-19 transmission. For this reason, experts have recommended screening everyone for infections.
In the general population, the prevalence of new infections is estimated to be less than 2%, even in the most affected states. At these rates, pooled tests could reduce test costs by 80% or more.
To date, the U.S. has conducted around 30 million COVID-19 tests, covering just under 10% of its population. However, some experts have asked that the entire population of the United States be evaluated every three months or even every two weeks.
Considering that American labs no longer have reagent tests, it seems doubtful that companies will soon be able to produce enough material to run these individual tests, assuming that companies or governments can even pay for all of them. The widespread use of pooled tests can help evaluate more people using fewer laboratory tests.
To increase capacity, the FDA recently released a guide that lays out a pathway to authorizing pooled tests so that more laboratories and test manufacturers can start doing so.
Another potential obstacle is cost. If labs charge much more to analyze a pooled sample than an individual one, any cost savings will dissipate.
The cheapest and fastest test methods allow for more widespread and frequent detection. Frequency is critical to controlling a rapidly spreading disease. Also, repeating the tests helps detect cases that may have been previously lost. Repeating the test helps, because research suggests that PCR tests miss about 20% of infected cases, regardless of whether they are pooled or not.
The recent authorization to begin grouped testing is an important step in the right direction. Clustering is an effective way to bring the US closer to the goal of universal and repeated testing that will help contain the spread of this unrelenting disease.
Darius Lakdawalla is Professor of Pharmaceutical Development and Regulatory Innovation Quintiles, Schaeffer Center for Health Policy and Economics, University of Southern California. Erin Trish is associate director of the Schaeffer Center for Health Policy and Economics and an assistant professor of pharmaceutical and health economics at the University of Southern California.
This is an updated version of an article originally published on July 2, 2020.
.
