Wastewater-based epidemiology (WBE) has recently gained significant attention. And rightfully so, as it’s being conducted at over 3,000 sites in 58 countries across the globe. But in the U.S., active wastewater surveillance for SARS-CoV-2 is only done in some states and to varying degrees. Last Friday, the CDC showcased our system by adding wastewater surveillance to the COVID-19
Tremendous overview ! I'm so proud that the Metropolitan Council here in the Mpls-St Paul metro area, in partnership with the University of Minnesota, started this testing program many weeks ago. Only recently were our data revealed for public viewing. Sewage analysis is a terrific "tool in the pandemic tool bag" and just clever as hell in its simplicity, selectivity, and sensitivity. As some of us have opined locally, "Unless many people are driving across our borders into Wisconsin or Iowa to have all their bowel movements, we have a good handle on CHANGES in the magnitude of viral RNA shedding from a big portion of Minnesota's 5.6 million population. It's the changes over time and not "absolute viral particle concentrations per se" that are the useful data. Thanks for showing our plot as an example -- we may have lousy winter weather and an NFL team that's never won the Super Bowl, but we do have first-rate sewage analysts. And . . . . our plot is obviously and definitely encouraging now.
Wouldn't it be awesome if wastewater data could be used to nip outbreaks in the bud?
"There's an uptick in Covid found in the wastewater in such-and-such a place, let's boost everyone there right now and send lots of high quality masks, antivirals and Evusheld"
Alternatively I could totally see Lloyds of London offering an exotic financial derivative to municipalities based on wastewater data. Quick, somebody find me an epidemiologist and an actuary to figure this out
Vincent Racaniello, a well known virologist at Columbia University, hosts an informative podcast called TWIV (This Week in Virology). February 8th's epidsode is entitled, "A Game of Thrones" and features fairly detailed discussion of waste water monitoring to follow dynamic changes in "genetic signals" present in large sewerages. I found it fascinating -- your mileage may vary -- there are a few details in the protocol used.
My one beef with the "Vince and Amy Show" is that they come close to making dogmatic advice statements about various vaccine decisions even though neither is a medical professional. They get numerous clinical questions from their "Fans". The good news is that a frequent guest "personality" is Dan Griffin, MD, PhD. That guy is a very experienced ID specialist and researcher. Being in hands-on consulting practice at a big hospital, he knows plenty of stuff that is germane to understanding our pandemic.
Thanks again for another informative post! Perhaps for a future topic, do we have any data (from other countries maybe) on omicron-omicron reinfection? Not sure we can hope for info on the new subvariant yet. (speaking from the perspective of an academic teaching a large enrollment class soon and a mom of an under-5 kid who brought us omicron last month....) Since indoor mask mandates will apparently be lifting very soon, it would be helpful to have an idea of when to brace for more childcare closures and disruptions.
Heard back. County has been involved in several tests. "...new study to begin in April is led by the Ga Dept of Health and promises to be the most valuable study to predict COVID in the community. "
It's fascinating to think that the entire profession of epidemiology started with identifying a disease via pumps and drinking water, with John Snow and the Broad Street pump.
i agree! just incredible. i hope you follow Neoliberal John Snow on Twitter. The admin applies events today as if we were in the John Snow days. Gives me a good laugh every time.
They are going to begin testing our wastewater in our small mountain community thanks to funds provided by our state health department. It will be interesting to compare the actual pcr test results with the wastewater lab analysis. Our community is heavily visited by tourists, and visitors are usually here for a few days. It really won't be a true assessment of our full-time community since we have so many tourists in the area. However, I think it will be worthwhile because the wastewater results will help the local health department (and surrounding communities/local hospital 45 minutes away) be prepared for what they may begin to see pop up in the local population and hopefully give them time to prepare for whatever variant might be the variant of the month. Science keeps evolving! One more resource in the toolkit to find a way to best learn and combat all community diseases, not just COVID.
Thanks for a fascinating article! I know you've been over this before, but NYS just dropped their mask mandate. Would you be able to review advice about how people as individuals should decide whether to mask or not given the current circumstances?
I ran the indoor air idea by a scientist on a sub I'm on, and scientist said that indoor air sampling is harder to get data from than wastewater. Apparently you can't really just take air samples because there aren't enough virons per cubic foot, so you need to look at a place where they concentrate, such as an air filter in the ventilation system. But air filters collect lots of other junk too (dust, etc), so finding the virus bits is like finding needles in a haystack. Seems to me like wastewater sampling has a bunch of analogous problems, but I did not get a chance to ask about that.
OK I stand semi-corrected, I must agree that 85 is more than 15. But FWIW, this reminds me of the famous polling error when a telephone survey predicted Hoover would defeat FDR, not recognizing that telephones were a marker of wealth. The fifteen percent with septic tanks are more rural and probably more "red", not a random sample.
My point is that 100% would be ideal, 95% would be almost as good, 90% pretty darn close, and 85% isn't that far off. For the purposes of pandemic surveillance of wastewater, I don't see the absence of those remaining 15% significantly affecting the trends of viral detection; such surveillance doesn't strike me as that granular.
I think this is an excellent way for sentinel detection of pathogens at ports of entry. Testing wastewater at ports of entry could give us an early warning to a disease entering the country.
Last I read, SARS-CoV 2 has been an issue for potential transmission to people from hamsters and ferrets. Other animals can be fomites for transmission between people. https://www.wormsandgermsblog.com/ is a good place for COVID epidemiology involving animals.
One reported case of SARS-CoV 2 infection in a person, then suspected transmission to household cat. Veterinarian that examined cat later tested positive. Too many variables to say for sure whether it was actual transmission from person to cat to person.
This should be used to settle the waning immunity debate. A 2-3 week early warning should be enough to do a mass vaccination in a specific area.
It can also be used for financial stability. Big agricultural companies use "weather derivatives" - financial instruments based on heating & cooling degree days to hedge against losses from droughts. No reason we can't create some entity that could offer something similar based on wastewater surveillance
So pleased to see you doing a post on this, and interested to learn about the drawbacks - not as obvious as the advantages. So what can one infer about what the epidemic is doing in our community from the wastewater data? A great leading indicator for emerging surges. But what can one make of the data as the surge recedes? In San Diego, the wastewater viral levels seem to be lagging the data on reported cases https://searchcovid.info/dashboards/wastewater-surveillance/ - because the virus continues to be shed through feces for some weeks after becoming infected, I suppose?
Wondering about the uptick in home testing kits and how that will impact our ability to detect COVID variations? Will WBE be one way to help with early detection of dangerous COVID variations?
WBE will likely provide more info on variants as genotyping can procede from the original samples. At least in the US, we've not performed genotyping on nearly enough samples even from lab-derived testing, to have contributed significantly to variant detection. We start doing it on limited numbers of samples after we're alerted something's coming.
What we need, at the national level, is a reporting system that allows reporting of at-home testing that can be incorporated into at least provisional incidence evaluations.
I was kind of wondering whether we could just have separate bins where people dispose of the non-recyclable parts of their rapid tests - according to whether they're positive or negative. And then weigh them.
I would like to learn more about Omicron BA.2. I keep hearing that positive cases are coming down and that mask mandates are being revoked. Yet, it seems we have a potential set back coming our way quickly, especially for those who have not gotten Covid yet. When will we know more?
Thanks for asking! I keep hearing that the BA.2 variant is more transmissible than B.1. For those who are frontline, vaccinated, boosted, and have not had Covid yet, how concerned should we be? Should we expect a new wave of infections? If so, is there (or when would there be) a timeline of how that might look?
Omicron in both its forms seems to have a moderate disregard for immunity derived from appropriate vaccination and boosting, ("up-to-date"), save that the symptoms are generally milder than someone who is unvaccinated would experience. BA.2 appears to be up to 3 times as transmissible as BA.1, but also tends to produce milder disease than BA.1 Some areas of the country are seeing a continuation of the BA..1 surge with BA.2 present; other areas that are seeing a rapid drop-off may not have been exposed yet to BA.2 and we might see another spike when it gets to those regions. There is a presumptive degree of immunity derived from infection by BA.1 but the emergence of BA.2 has not provided sufficient time to clarify that completely.
Thanks for taking the time to reply. I will be interested to see what milder disease as compared to BA.1 actually looks like, given that even mild BA.1 can still be rough for some vaccinated people, and that long covid continues to be a concern. But that is good news!
That's been the gotcha. "Relatively mild" has been much more a media statement than a clinical evaluation for hospitalized patients. Still, with omicron (assume BA.1), hospital and ICU stays are statistically shorter and the number of relative deaths is down, although, with the markedly increased transmissibility, the number of cases was much higher, resulting higher overall numbers. I'm waiting to see my first BA.2 case (or even suspected), to see what "milder" looks like.
Great article. I am waiting for your analysis of the report suggesting that all of our "lock-down" mitigation efforts had a less than 1% impact on cases or deaths or something - the one out of Johns Hopkins Economics.
I'm fascinated that we're willing to look at the effectiveness of mitigation measures (mislabeled as 'lockdowns') for a public health emergency, evaluated by economists. If I'm to make an evaluation of the utility of such measures, and I need an economic impact, I won't make sweeping and unsupported groupings as noted in the report. Claiming the respectability of Hopkins for this was a deception.
These ideologically driven economists should make themselves useful and design an exotic financial derivative based on wastewater surveillance. If they're actually being intellectually honest.
But we didn't HAVE lockdowns! Those were in China, folks. That fact alone seems like a fatal flaw in the stufdy to me -- or you could say that using that term is a sign of a paper not written in good faith. We had mandated masking, school closings, restrictions on the size of gatherings, that sort of thing. Of course it makes sense to gather and publish data about the effectiveness of these measures, but wtf is up with talking about "lockdowns"? And if you're going to evaluate the effectiveness of the various measures that were taken, you obviously have to take into account what level of compliance the public exhibited. For instance, if masks are "mandatory" in indoor public settings but half the public is wears no mask or a chin diaper, then it is impossible to tell how much masking indoors reduces infections. Duh.
It seems pretty circular to invent a "stringency index" and then base one's entire analysis on that. It's also a bit of a straw-man. Some lockdowns work, others don't. It depends on how compliant the population is, household sizes, etc. The point of lockdowns was to help prevent hospitals from becoming overwhelmed, which they mostly did, especially early on.
Tremendous overview ! I'm so proud that the Metropolitan Council here in the Mpls-St Paul metro area, in partnership with the University of Minnesota, started this testing program many weeks ago. Only recently were our data revealed for public viewing. Sewage analysis is a terrific "tool in the pandemic tool bag" and just clever as hell in its simplicity, selectivity, and sensitivity. As some of us have opined locally, "Unless many people are driving across our borders into Wisconsin or Iowa to have all their bowel movements, we have a good handle on CHANGES in the magnitude of viral RNA shedding from a big portion of Minnesota's 5.6 million population. It's the changes over time and not "absolute viral particle concentrations per se" that are the useful data. Thanks for showing our plot as an example -- we may have lousy winter weather and an NFL team that's never won the Super Bowl, but we do have first-rate sewage analysts. And . . . . our plot is obviously and definitely encouraging now.
Wouldn't it be awesome if wastewater data could be used to nip outbreaks in the bud?
"There's an uptick in Covid found in the wastewater in such-and-such a place, let's boost everyone there right now and send lots of high quality masks, antivirals and Evusheld"
Alternatively I could totally see Lloyds of London offering an exotic financial derivative to municipalities based on wastewater data. Quick, somebody find me an epidemiologist and an actuary to figure this out
Vincent Racaniello, a well known virologist at Columbia University, hosts an informative podcast called TWIV (This Week in Virology). February 8th's epidsode is entitled, "A Game of Thrones" and features fairly detailed discussion of waste water monitoring to follow dynamic changes in "genetic signals" present in large sewerages. I found it fascinating -- your mileage may vary -- there are a few details in the protocol used.
My one beef with the "Vince and Amy Show" is that they come close to making dogmatic advice statements about various vaccine decisions even though neither is a medical professional. They get numerous clinical questions from their "Fans". The good news is that a frequent guest "personality" is Dan Griffin, MD, PhD. That guy is a very experienced ID specialist and researcher. Being in hands-on consulting practice at a big hospital, he knows plenty of stuff that is germane to understanding our pandemic.
Thanks again for another informative post! Perhaps for a future topic, do we have any data (from other countries maybe) on omicron-omicron reinfection? Not sure we can hope for info on the new subvariant yet. (speaking from the perspective of an academic teaching a large enrollment class soon and a mom of an under-5 kid who brought us omicron last month....) Since indoor mask mandates will apparently be lifting very soon, it would be helpful to have an idea of when to brace for more childcare closures and disruptions.
Perfect! I just emailed our county commission chair about this. Included link to this post. Fingers crossed.
Heard back. County has been involved in several tests. "...new study to begin in April is led by the Ga Dept of Health and promises to be the most valuable study to predict COVID in the community. "
This good.
It's fascinating to think that the entire profession of epidemiology started with identifying a disease via pumps and drinking water, with John Snow and the Broad Street pump.
i agree! just incredible. i hope you follow Neoliberal John Snow on Twitter. The admin applies events today as if we were in the John Snow days. Gives me a good laugh every time.
Just found it! OMG, he's amazing. I'm already laughing (and wincing). Thanks for that!
Thanks. Another Twitter feed I have to read! That's hilarious.
They are going to begin testing our wastewater in our small mountain community thanks to funds provided by our state health department. It will be interesting to compare the actual pcr test results with the wastewater lab analysis. Our community is heavily visited by tourists, and visitors are usually here for a few days. It really won't be a true assessment of our full-time community since we have so many tourists in the area. However, I think it will be worthwhile because the wastewater results will help the local health department (and surrounding communities/local hospital 45 minutes away) be prepared for what they may begin to see pop up in the local population and hopefully give them time to prepare for whatever variant might be the variant of the month. Science keeps evolving! One more resource in the toolkit to find a way to best learn and combat all community diseases, not just COVID.
Thanks for a fascinating article! I know you've been over this before, but NYS just dropped their mask mandate. Would you be able to review advice about how people as individuals should decide whether to mask or not given the current circumstances?
Off topic but sending love and imaginary wine to YLE and the rest of our under-5 club today.
Two questions:
- would testing for multiple respiratory and enteral viruses help to calibrate the accuracy of wastewater testing?
- for ordinary citizens trying to assess risk in their daily lives, wouldn't the Holy Grail be sampling indoor air?
https://www.thermofisher.com/us/en/home/industrial/environmental/in-air-pathogen-surveillance/renvo-sars-cov-2-environmental-test.html
I ran the indoor air idea by a scientist on a sub I'm on, and scientist said that indoor air sampling is harder to get data from than wastewater. Apparently you can't really just take air samples because there aren't enough virons per cubic foot, so you need to look at a place where they concentrate, such as an air filter in the ventilation system. But air filters collect lots of other junk too (dust, etc), so finding the virus bits is like finding needles in a haystack. Seems to me like wastewater sampling has a bunch of analogous problems, but I did not get a chance to ask about that.
A fourth limitation on the reliability of wastewater testing: it misses the part of the population using septic tanks so not connected to the sewers.
Implicit in Dr. Jetelina's post today, "Eighty-five percent of the U.S. population has a piped sewer connection that could be sampled."
Still, 85% is quite high as far as opportunities to sample go. At that high a level, I don't think it rises to the level of a "limitation."
OK I stand semi-corrected, I must agree that 85 is more than 15. But FWIW, this reminds me of the famous polling error when a telephone survey predicted Hoover would defeat FDR, not recognizing that telephones were a marker of wealth. The fifteen percent with septic tanks are more rural and probably more "red", not a random sample.
Understood, and I semi-agree. 😉
My point is that 100% would be ideal, 95% would be almost as good, 90% pretty darn close, and 85% isn't that far off. For the purposes of pandemic surveillance of wastewater, I don't see the absence of those remaining 15% significantly affecting the trends of viral detection; such surveillance doesn't strike me as that granular.
As always, I could be mistaken.
Excellent review. And now that CDC is offering WBE data for select sites, we've another resource.
I think this is an excellent way for sentinel detection of pathogens at ports of entry. Testing wastewater at ports of entry could give us an early warning to a disease entering the country.
Could also be used to allocate antiviral supply.
I didn't think about that. Very good point.
I see you're a DVM. I wish we were doing more surveillance of animals. Any guesses as to how prevalent Covid is among horses, for instance?
Last I read, SARS-CoV 2 has been an issue for potential transmission to people from hamsters and ferrets. Other animals can be fomites for transmission between people. https://www.wormsandgermsblog.com/ is a good place for COVID epidemiology involving animals.
One reported case of SARS-CoV 2 infection in a person, then suspected transmission to household cat. Veterinarian that examined cat later tested positive. Too many variables to say for sure whether it was actual transmission from person to cat to person.
This should be used to settle the waning immunity debate. A 2-3 week early warning should be enough to do a mass vaccination in a specific area.
It can also be used for financial stability. Big agricultural companies use "weather derivatives" - financial instruments based on heating & cooling degree days to hedge against losses from droughts. No reason we can't create some entity that could offer something similar based on wastewater surveillance
So pleased to see you doing a post on this, and interested to learn about the drawbacks - not as obvious as the advantages. So what can one infer about what the epidemic is doing in our community from the wastewater data? A great leading indicator for emerging surges. But what can one make of the data as the surge recedes? In San Diego, the wastewater viral levels seem to be lagging the data on reported cases https://searchcovid.info/dashboards/wastewater-surveillance/ - because the virus continues to be shed through feces for some weeks after becoming infected, I suppose?
Wondering about the uptick in home testing kits and how that will impact our ability to detect COVID variations? Will WBE be one way to help with early detection of dangerous COVID variations?
WBE will likely provide more info on variants as genotyping can procede from the original samples. At least in the US, we've not performed genotyping on nearly enough samples even from lab-derived testing, to have contributed significantly to variant detection. We start doing it on limited numbers of samples after we're alerted something's coming.
What we need, at the national level, is a reporting system that allows reporting of at-home testing that can be incorporated into at least provisional incidence evaluations.
I was kind of wondering whether we could just have separate bins where people dispose of the non-recyclable parts of their rapid tests - according to whether they're positive or negative. And then weigh them.
Since the rapid tests are not all the same, that'd not yield much data, I'm afraid.
But it would provide a rough estimate, wouldn't it?
5 lb worth of discarded positive tests
95 lb worth of negative tests
95% positivity rate
Or go by the frequency with which the bins need to be emptied, based on how fast they fill up
There's also something to be said for keeping the discarded tests out of landfills, is there not?
I would like to learn more about Omicron BA.2. I keep hearing that positive cases are coming down and that mask mandates are being revoked. Yet, it seems we have a potential set back coming our way quickly, especially for those who have not gotten Covid yet. When will we know more?
What questions do you have specifically?
Thanks for asking! I keep hearing that the BA.2 variant is more transmissible than B.1. For those who are frontline, vaccinated, boosted, and have not had Covid yet, how concerned should we be? Should we expect a new wave of infections? If so, is there (or when would there be) a timeline of how that might look?
Omicron in both its forms seems to have a moderate disregard for immunity derived from appropriate vaccination and boosting, ("up-to-date"), save that the symptoms are generally milder than someone who is unvaccinated would experience. BA.2 appears to be up to 3 times as transmissible as BA.1, but also tends to produce milder disease than BA.1 Some areas of the country are seeing a continuation of the BA..1 surge with BA.2 present; other areas that are seeing a rapid drop-off may not have been exposed yet to BA.2 and we might see another spike when it gets to those regions. There is a presumptive degree of immunity derived from infection by BA.1 but the emergence of BA.2 has not provided sufficient time to clarify that completely.
Thanks for taking the time to reply. I will be interested to see what milder disease as compared to BA.1 actually looks like, given that even mild BA.1 can still be rough for some vaccinated people, and that long covid continues to be a concern. But that is good news!
That's been the gotcha. "Relatively mild" has been much more a media statement than a clinical evaluation for hospitalized patients. Still, with omicron (assume BA.1), hospital and ICU stays are statistically shorter and the number of relative deaths is down, although, with the markedly increased transmissibility, the number of cases was much higher, resulting higher overall numbers. I'm waiting to see my first BA.2 case (or even suspected), to see what "milder" looks like.
Great article. I am waiting for your analysis of the report suggesting that all of our "lock-down" mitigation efforts had a less than 1% impact on cases or deaths or something - the one out of Johns Hopkins Economics.
I'm fascinated that we're willing to look at the effectiveness of mitigation measures (mislabeled as 'lockdowns') for a public health emergency, evaluated by economists. If I'm to make an evaluation of the utility of such measures, and I need an economic impact, I won't make sweeping and unsupported groupings as noted in the report. Claiming the respectability of Hopkins for this was a deception.
These ideologically driven economists should make themselves useful and design an exotic financial derivative based on wastewater surveillance. If they're actually being intellectually honest.
But we didn't HAVE lockdowns! Those were in China, folks. That fact alone seems like a fatal flaw in the stufdy to me -- or you could say that using that term is a sign of a paper not written in good faith. We had mandated masking, school closings, restrictions on the size of gatherings, that sort of thing. Of course it makes sense to gather and publish data about the effectiveness of these measures, but wtf is up with talking about "lockdowns"? And if you're going to evaluate the effectiveness of the various measures that were taken, you obviously have to take into account what level of compliance the public exhibited. For instance, if masks are "mandatory" in indoor public settings but half the public is wears no mask or a chin diaper, then it is impossible to tell how much masking indoors reduces infections. Duh.
Can you cite that article specifically, please?
It seems pretty circular to invent a "stringency index" and then base one's entire analysis on that. It's also a bit of a straw-man. Some lockdowns work, others don't. It depends on how compliant the population is, household sizes, etc. The point of lockdowns was to help prevent hospitals from becoming overwhelmed, which they mostly did, especially early on.
And the Forbes version of the medpage article: https://www.forbes.com/sites/brucelee/2022/02/06/did-so-called-johns-hopkins-study-really-show-lockdowns-were-ineffective-against-covid-19/?sh=5e42090e1225