It happened to be the same day that scientists 10,000 miles away in Botswana and South Africa began alerting the world to a new and highly mutated version of the virus that causes Covid-19.
Alexandria Boehm, a professor of civil engineering at Stanford University, read about the usual pattern of mutations in the yet-to-be-named variant and sprang into action.
For more than a year, Boehm and her team of 45 people at the Sewer Coronavirus Alert Network, or SCAN, have been collecting and testing daily sludge samples from wastewater processing plants across Northern California, hunting for fragments of the new coronavirus.
Wastewater-based epidemiology has proven to be so reliable in dozens of pilot projects across the US that the government has invested millions to create the National Wastewater Surveillance System, or NWSS, a network of 400 testing sites spread across 19 states that is coordinated by the US Centers for Disease Control and Prevention.
Boehm’s SCAN is part of that network, which has been quietly operating behind the scenes, generating data for public health departments across the country, since September 2020.
For the first time, the CDC will publish data that looks at how much coronavirus is turning up in the country’s wastewater. It will add this testing data to its Covid-19 dashboard.
Tests show that there’s been a decrease in the amount of virus at two-thirds of the NWSS’s 400 sites from levels measured 15 days ago, according to a CDC official familiar with the planning.
More than 500 sites will begin submitting data in the coming weeks, the CDC says. A media briefing on the surveillance system is scheduled for Friday.
Data from anywhere with a sewer connection
SARS-CoV-2, the virus that causes Covid-19, is encased in an oily envelope. After it invades our bodies and begins to furiously clone itself, some of those copies are shed into our intestines, where the fatty parts of the virus stick to the fats in stool. When we poop, genetic material from the virus gets flushed down the toilet into the wastewater stream, where it can be detected by the same kinds of tests labs use to detect the virus from nasal swabs: real time polymerase chain reaction tests, or RT-PCR.
This kind of testing is highly sensitive. It can pick up the presence of the virus when just one person out of 100,000 in a given area, or sewershed, is infected.
And because wastewater testing doesn’t depend on people to realize they’re sick and seek out a test, or even to have symptoms at all, it’s often the earliest warning a community has that wave of Covid-19 infections is on the way.
The CDC estimates that it takes five to seven days after a toilet flushes to get the wastewater data to its COVID Tracker, and and the samples typically turn positive in an area four to six days before clinical cases show up.
“As long as people are using a toilet that’s connected to a sewer, we can get information on those cases in that community,” said Amy Kirby, a CDC microbiologist who leads the NWSS project.
Government investment has taken what had been a little-known branch of public health and brought it into the mainstream.
“It’s really exploded the field,” said Colleen Naughton, a civil engineer at the University of California at Merced who runs the @CovidPoops19 Twitter account. “We did wastewater monitoring for other pathogens before this, like poliovirus, but it’s really, the amount of people involved in everything has really increased exponentially.”
Jumping into action
Speed is essential to making wastewater testing useful. So when Boehm heard the gravity behind the alerts of a new variant emerging 10,000 miles away, she didn’t delay.
“It all happened very, very quickly, right around Thanksgiving and the holidays, which is a really inconvenient time for things to happen,” she said.
SCAN — a partnership between Stanford, Emory University and the University of Michigan — worked long hours to change the monitoring system over to a new test.
The Monday before Thanksgiving, Boehm alerted her team to the new variant. On Tuesday, they downloaded the handful of gene sequences for the new variant that had been sent to GISAID, a website used by researchers around the world to share information about the new coronavirus. They started to design a test that could pick up several of its telltale changes, including amino acids that were deleted from a chunk of sites in its spike protein and the addition of three amino acids at another place on its spike.
Boehm ordered supplies: the chemical probes they would need to run their new test. She knew that delivery could take weeks, but she says they got lucky and got the new chemicals in just seven days.
But it wasn’t just luck. She had done this many times over the past year, first for the Alpha variant and then Beta, Gamma, Delta and Mu.
While they waited for the supplies, Boehm dusted off an older test her team had been using to find signs of the Alpha variant. Like Omicron, Alpha is missing certain amino acids from a key location on its spike protein. These deletions cause a pattern on lab tests called an s-gene target failure.
“We knew Alpha was basically gone. It’s extinct in our region,” she said. That meant any hits on the Alpha tests were probably the new variant.
On Thursday, Thanksgiving, they got their first hit. A sample in Merced tested positive on the old test.
On Friday, the World Health Organization gave the variant they were hunting a name, Omicron, and officially listed it as a variant of concern.
The following Tuesday, another sample — this time from Sacramento — was positive, and on Wednesday, Gov. Gavin Newsom and the CDC announced that the first American had tested positive for Omicron in San Francisco.
The woman had recently traveled from South Africa, landing in San Francisco on Nov. 22. She didn’t seek out testing until nearly a week later, on Monday, Nov. 28, four days after the first wastewater sample in Merced was positive.
In the days that followed, more positive samples turned up from Sacramento.
And in Santa Clara County, they got their first hit in Palo Alto on Dec. 7, the same day a person living in same sewershed became that county’s first case.
What started as a trickle of positive samples quickly became a steady stream. On Dec. 16, the Santa Clara County Health Department hosted a news conference warn the public based, in large part, on what they were seeing in the wastewater data.
“When I look around the corner ahead, what I see is a deluge of Omicron. What I see is one of the most challenging moments that we’ve had yet in the pandemic,” said Dr. Sara Cody, director of public health for Santa Clara County.
The limitations of wastewater testing
The CDC’s Kirby says the system worked well for the arrival of Omicron in California. Colorado, Houston, and New York City. She published a snapshot of those efforts in the January 21 Morbidity and Mortality Weekly Report.
“What we hear most often is that our health department partners say this gives them confidence to know what’s going on in their communities. So truly, are cases going up or down? And that’s because it is completely independent of health care and clinical testing,” she says.
Kirby says it’s also a useful tool to let counties and cities know where to direct limited resources.
“So if you have, for example, mobile testing capacity, and you want to decide which communities could most use that additional testing capacity, wastewater surveillance is really useful for that,” she says.
There are some blind spots in the system. Although 80% of homes in the US are connected to sewers, the other 20% or so rely on septic systems. These homes, which are mostly in rural areas, wouldn’t be covered by the testing.
Wastewater testing is also harder to interpret and less useful in areas where people come and go often, like tourist destinations.
It’s also difficult to compare data between sewersheds because different areas use different sampling methods. Some take samples directly from the water with thin cotton swabs encased in 3-D printed submarines, while others collect and sample biosolids. Sewersheds can be very different sizes, which also complicated comparisons. The CDC’s new dashboard shows data at the ZIP code level.
Finally, this kind of testing can’t signal when a community is free from the virus because the threshold of detection — how many people have to be positive in an area to show up in a water sample — isn’t known.
For these reasons, the CDC says wastewater surveillance is best used along with case-based surveillance.
To avoid direct comparisons, Kirby says, the dashboard will compare each site to its own past results. The main metric will be the percent change concentration at the same site over the past 15 days. The other metric people will be able to see is how many detections there were over that same 15 days, how many samples tested positive at all.
“As you can imagine, right now, that’s not a very useful metric, because everywhere is positive. But as cases go down and we see much lower rates, that will become the metric you’ll want to follow to see if SARS-CoV-2 is re-emerging in your community,” Kirby said.
Kirby says wastewater monitoring will be around long after Covid is gone, too. By the end of the year, the CDC plans to expand the number of pathogens tracked on the dashboard to include influenza, a fungal superbug called Candida auris, and foodborne threats like E. coli and salmonella.