Seven months after the pandemic kicked into high gear in the US, contact tracing still isn’t working to slow the spread of Covid-19. It’s not the fault of the contact tracers: With wait times for diagnostic test results as long as weeks, the window for tracers to identify Covid-19 patients before they have a chance to infect others is vanishingly small.
Now, though, there is an approach that could make contact tracing more effective: rapid antigen testing.
In late August, the US government spent $760 million on 150 million antigen tests from the Illinois-based company Abbott Laboratories, which should be distributed to various state and tribal governments through the rest of the year. Several other versions are being sold. But it’s clear that these tests won’t actually help slow the spread of Covid-19 if they’re used without a plan.
Diagnostic PCR tests, which analyze samples of a person’s saliva or nasal secretions for genetic evidence of the coronavirus, have run into delays at every turn in the US. Clinics have run into bottlenecks of specialized cotton swabs, plastic tubes, and chemical reagents. Samples run into shipping delays as they’re sent to certified labs for processing. And the labs themselves have a limited ability to test samples on dedicated machines.
Antigen testing, by contrast, can be done on the spot. Instead of DNA, these diagnostic tests look for proteins on the outer coating of the virus. They’re not as sensitive as PCR tests, meaning they can only detect viral proteins when a person has a high viral load and is therefore most infectious, but because they’re cheaper—to the tune of $5 per test—individuals can get tested repeatedly over days or weeks. They require equipment the size of a credit card instead of an entire lab, and results are ready in 15 minutes.
That’s music to contact tracers’ ears. Their job is to call everyone a Covid-19 patient contacted in the 48 hours before they started showing symptoms. Contacts—which the US Centers for Disease Control and Prevention defines as those having had physical touch or having been within six feet for more than 15 minutes—of an index case would need to quarantine for 14 days to avoid spreading the virus.
Immediate results could mean more accurate tracing: The longer the delay in an index case’s test results, the harder it is for them to remember their contacts. And because antigen tests are cheaper, they could be used more frequently, making it easier to spot exactly when a person becomes infectious. If a person tested negative one day but positive a few days later, contact tracers would only need to find contacts in the days between the two tests.
Antigen tests aren’t perfect, of course; they are less accurate than PCR tests, with high false positive rates. But because they’re so cheap, it’s possible to double-check the results against other rapid tests.
“Each company has their own,” explains Anthony Lemmo, the chief executive officer and president of Biodot, a California-based company that manufactures diagnostic tests. If one antigen test shows that a person has Covid-19, they could take another that looks for a different part of the SARS-CoV-2 protein casing. If that test is negative, the first was likely a false positive. If both are positive, the person can assume the antigen test was correct, and opt for a PCR test to confirm.
The US government has invested heavily in antigen testing in the hopes it can solve the problems that PCR testing and contact tracing haven’t been able to. But simply having antigen tests alone isn’t enough. They have to be distributed in a way that makes sense for a containment strategy, says Michael Mina, an epidemiologist at the Chan School of Public Health at Harvard.
In a perfect world, everyone would be tested twice weekly to see if they had developed high levels of the virus, even if they aren’t showing symptoms. If they do, they could have confirmatory testing, and isolate accordingly, and then their contacts notified to quarantine. This is what’s happening on some college campuses that have reopened.
The US is way too big for everyone to do that, though. It’d take more than 34 billion tests to track the population for an entire year—not to mention the extra tests needed to confirm positive results. But there are ways to be smarter about it.
“We just need to achieve what I consider to be herd effects,” says Mina. Right now, on average, every person with Covid-19 infects a few more. This number, called R0 (pronounced R-naught) needs to be below one in order for the pandemic to die out, he explains. “Our simulation models suggest that even if we just had half of a population using a test twice a week, or even once a week, that’d be enough to drop the average number of new infections per case to get it below one,” says Mina.
That effort, though, requires coordination. Providing tests to states is a good step, but these tests are only useful if local governments can identify and ensure access to the populations that would benefit most from repeated testing. Rather than providing a single test from Abbott, Mina suggests packaging several companies’ tests together, so it’d be easy to take a second test if one did pop up positive. Another option would be to combine this kind of contact tracing and testing with other surveillance, like wastewater monitoring.
Fighting a pandemic is like fighting a war with a microscopic enemy. The US federal government keeps looking for the silver bullet to try to stop the pandemic. But wars are not won by bullets; it takes competent soldiers to deploy them.