As the Covid-19 pandemic marches across America, causing record-breaking numbers of cases, almost every solution for controlling the disease includes more testing, especially as cities and states try to reopen. But with states hitting their limits on testing, we need new tools for understanding Covid-19 transmission. A national wastewater surveillance program offers a cost-effective approach to track Covid-19 across the majority of the U.S. population and provide early warnings of resurgence.

Simply testing individuals with symptoms won’t be enough to track how many people are infected. Given the high degree of asymptomatic spread — even in the absence of superspreading events — we need a national strategy that relies heavily on surveillance. Experts increasingly argue that we should be testing nursing home residents every week. Some universities are planning to test every student multiple times per week. And the federal government is talking about pooled testing as a strategy for testing large groups of individuals at once. The hope is that such testing, when coupled with robust contact tracing, can identify infections before they spread.

Since the start of the pandemic in February, approximately 30 million individual Covid-19 tests have been performed in the U.S. Nobel laureate Paul Romer and others argue that we need the capacity to perform this number of tests every day — about 80 times our current capacity. Even if we could afford the $3 billion daily expense, it would be logistically nearly impossible to achieve, at least without substantial federal engagement and leadership, which does not appear to be imminent.


Testing individuals is the only way to find out who is infected, isolate them, and trace their contacts. But we should not rely on individual testing to understand transmission trends in communities. An alternative approach that complements the current need for testing more individuals is testing entire communities by tracking the coronavirus in wastewater. This is logistically easier, would serve as an early warning signal of the presence of SARS-CoV-2 in a community, and result in nearly continuous tracking of Covid-19 in three-quarters of the U.S. population at a cost of $40 million to $80 million per week.

People infected with SARS-CoV-2 shed the virus in their stool even before they show symptoms of Covid-19. Analyzing sewage for the virus, using methods like the ones used for testing individuals, can predict the community level of infection one week to two weeks in advance of clinical diagnoses, and show increasing and decreasing levels of coronavirus infection and transmission. Testing the wastewater of a nursing home, prison, or dormitory could potentially give an early signal that one or more residents are infected and therefore prompt testing of each resident.


National wastewater surveillance would require at least weekly testing at wastewater treatment plants and in sewers. Although some states, such as Colorado, Ohio, and New York, have taken steps toward implementing wastewater surveillance efforts, the U.S. lacks a national Covid-19 program for it.

Other countries have national programs in place. Australia is working with state and territory authorities and 16 of the largest utilities in the country to implement weekly wastewater monitoring as part of the Collaboration on Sewage Surveillance of SARS-COV-2 project. In Finland, the Finnish Institute for Health and Welfare is overseeing sampling at 28 wastewater treatment plants in a monitoring program that covers 60% of the Finnish population. The Netherlands has implemented weekly testing at 29 plants that cover about 25% of the Dutch population. Pakistan has instituted a program that leverages the existing polio environmental surveillance network to monitor SARS-CoV-2 at 78 wastewater treatment plants across the country.

Although we can learn from the national wastewater surveillance experience of other countries, any program designed for the U.S. will necessarily be larger and more complex. At last count in 2012, there were 14,748 wastewater treatment plants providing sewage collection and treatment to 80% of people living in the U.S. The other 20% are served by septic systems. While regular sampling for all wastewater treatment plants is certainly feasible, even focusing on just the 542 largest ones — those with treatment capacities exceeding 10 million gallons of sewage per day — would cover half the U.S. population.

sewer wastewater sampling
Wastewater sampling can be done in sewer lines downstream of nursing homes, dorms, or other facilities. Courtesy Teledyne ISCO

To be effective, wastewater surveillance needs to use a sensitive and specific testing method that gives reliable and repeatable results that are available within a day. The results need to be quantitative, such as genome copies per liter of wastewater, which can then be translated into an estimate of people infected. Such analytical capabilities already exist in commercial environmental testing laboratories. And efforts are underway to adapt electrochemical biosensors previously developed for virus detection, including microfluidic paper-based devices, for near real-time SARS-CoV-2 quantification in wastewater.

There are still important questions to be answered about national wastewater surveillance for SARS-CoV-2: How should samples be collected? (We recommend combining samples taken from one location over the course of a day, although a sample from one point in time could also suffice.) How often should testing be done? (We recommend two to three times per week.) Where should testing be done? (We recommend testing at wastewater treatment plants and in select locations in the sewer system feeding the plants). Collecting samples in the sewer system could be accomplished by deploying autosamplers at manholes, ideally selecting those downstream of nursing homes and other vulnerable facilities.

We will also need a national surveillance database that allows seamless input of analytical results, visualization of trends, and ultimately predictions of where the next outbreak will be.

The U.S. needs to get started with wastewater surveillance now. The science clearly supports this strategy. Doing so will provide a cost-effective method of understanding the pandemic and an early warning system for increased and decreased transmission as the pandemic waxes and wanes.

An early warning system is essential if we are going to keep ahead of the virus.

Anna Mehrotra is a senior wastewater engineer with CDM Smith Inc., in Boston. David A. Larsen is an environmental epidemiologist and associate professor of public health at Syracuse University in New York. Ashish Jha is the director of the Harvard Global Health Institute and professor of global health at the Harvard T.H. Chan School of Public Health.

  • Woo hoo!! Thank you! I’ve been pleading for this for months. The jails need this testing. I’m curious to see how it would work in schools, do kids defecate at school?

Comments are closed.

A roundup of STAT’s top stories of the day in science and medicine

Privacy Policy