25 March 2021

How sewage testing can help tackle the spread of Covid-19

Try this one weird trick to track the pandemic.

HONG KONG, CHINA - JANUARY 20: Workers check the sewage system where COVID-19 cases have been confirmed in the Jordan district on January 19, 2021 in Hong Kong, China. Hong Kong issues an isolation order for entire residential blocks for the first time since the COVID-19 pandemic began as the number of cases continued to climb with mandatory testing of residents in over 70 buildings. (Photo by Anthony Kwan/Getty Images)

Where will the next Covid-19 outbreak begin? How many cases will we see? How many people will die? Alongside the morass of mathematical models aimed at forecasting Covid rates, there have also been some, shall we say, sillier-sounding methods of prediction.

In December a leading British psychic informed us that she “started January 2020 with a feeling of dread” and “felt death coming” – though it would’ve been nice if she’d told us that at the time. And although many astrologers failed to notice Covid on their star charts, some did predict that 2020 would be a particularly turbulent year.

Even within the bounds of actual science, we’ve seen some bizarre ideas. For instance, in mid-2020 a study claimed that the ratio of finger lengths on people’s hands predicted the number of Covid deaths in their country (a finding that has not stood up to scrutiny).

Here’s another hare-brained Covid-prediction scheme: rooting around in sewage for evidence of the spread of the coronavirus. Sounds ridiculous – right?

The idea certainly didn’t get off to a good start. An early paper, from May 2020, claimed that sewage-based testing could give you an almost perfect prediction of how many Covid hospitalisations there would be in an area just three days later. That’s better than any psychic – but it’s also too good to be true. As explained at the time by Tom Chivers, this was based on a fumbled statistic: the scientists had essentially calculated the same set of numbers twice and correlated the sets together, giving them an impressive-looking – but completely meaningless – result (they later corrected the error).

Since then, though, the evidence has redeemed the idea of sewage-based testing for Covid-19. It might even turn out to be a crucial part of our strategy for safely reopening the country.

Here’s how it works. We know that the coronavirus’s genetic material (its RNA) can survive in faeces: doctors can reliably find it in stool samples from patients. Luckily for us, that RNA doesn’t disintegrate when it’s sent to the sewers. We can therefore take samples from local drains, or from sewage treatment plants, and send them off to the lab for testing. Repeat the sampling day in, day out, and we can get an idea of the trends of Covid infection in a local area – and whether those trends look like a new spike in infections.

It’s not straightforward. The amount of faecal matter present in the sewers waxes and wanes throughout the day. If it’s raining, there’ll be more water in the drains, diluting the signal of the viral RNA. Naively sampling the wastewater is a recipe for confusion.

But there are some ingenious ways to avoid the pitfalls. Scientists have been using other viruses that are found dependably in human waste – like the rather cute-sounding pepper mild mottle virus – as a benchmark, giving us an indirect flag of how much faeces is present at any given time. Analysts can then adjust their RNA analysis to match.

Contrary to some earlier, overblown claims, sewage rates don’t correlate perfectly with disease: the overall correlation seems to be somewhere in the region of 0.4, on a scale of -1.0 to 1.0. Not super-strong, but far better than nothing. And although more works needs to be done (there are calls to make more of the data available for research), most scientists think wastewater testing will be an important part of the tool-kit for monitoring Covid in future.

This comes with some big advantages. Although the wastewater analysis isn’t accurate enough to tell us the prevalence of Covid in any given area (it’s more about the change over time than the level of infection), it could serve as a non-invasive early-warning system, telling us where we should be aiming our testing, messaging and restrictions. It’s particularly useful for schools as children are too young to receive the current vaccines. It could help us to spot the appearance of new variants of the virus. And it could partly address one of the perennial features of the Covid pandemic, inequality, by drawing our attention to outbreaks in areas with lower uptake of testing and vaccination.

Expect to hear a lot more about wastewater Covid testing in the coming months. Although the government hasn’t yet made good on its plans to add the data to the official UK government’s coronavirus dashboard, you can see similar numbers from Scotland, the Netherlands, Canada and many cities and regions worldwide.

Maybe those who know their scientific history could have predicted – even without psychic powers or horoscopes – that sewage-based testing would end up being important. Not only has it shown promise for tracking other diseases in the past, but the very foundation of the science of epidemiology lies in wastewater. When John Snow mapped out a cholera epidemic in 1850s London, it dawned on him that the disease was in the water, which led to his disabling of the Broad Street Pump in Soho, halting the epidemic in its tracks.

It won’t be as simple for Covid-19, which of course isn’t spread (much) by contaminated water. But it’s there all the same, and – weird and perhaps unpleasant as it may sound – we’re now using that knowledge to our advantage.