Stay up to date on all things crypto and blockchain
Token Daily is a place to discover trending news and products in crypto and blockchain.
Source: medrxiv.org ·
© Dave Birch (2018) - @dgwbirch
3 months ago
RT @BrennanSpiegel: This graph is amazing. It shows that measuring #SARSCoV2 levels in municipal sewage almost perfectly predicts forthcoming #COVID19 cases with a full week's notice (R=0.994). It's one of several discoveries in this new study from @Yale: https://t.co/hZVdXebx2D. C-19 is #InThePoop https://t.co/shNzTKYoYP
6624
12001
Chandra Duggirala MD - @csentropy
3 months ago
RT @BrennanSpiegel: This graph is amazing. It shows that measuring #SARSCoV2 levels in municipal sewage almost perfectly predicts forthcoming #COVID19 cases with a full week's notice (R=0.994). It's one of several discoveries in this new study from @Yale: https://t.co/hZVdXebx2D. C-19 is #InThePoop https://t.co/shNzTKYoYP
6319
11482
Christopher Wheeler - @Stocktrader
3 months ago
RT @BrennanSpiegel: This graph is amazing. It shows that measuring #SARSCoV2 levels in municipal sewage almost perfectly predicts forthcoming #COVID19 cases with a full week's notice (R=0.994). It's one of several discoveries in this new study from @Yale: https://t.co/hZVdXebx2D. C-19 is #InThePoop https://t.co/shNzTKYoYP
5390
9630
Thomas Lee - @fundstrat
3 months ago
RT @BrennanSpiegel: This graph is amazing. It shows that measuring #SARSCoV2 levels in municipal sewage almost perfectly predicts forthcoming #COVID19 cases with a full week's notice (R=0.994). It's one of several discoveries in this new study from @Yale: https://t.co/hZVdXebx2D. C-19 is #InThePoop https://t.co/shNzTKYoYP
5206
9276
Vinay Gupta - @leashless
3 months ago
RT @BrennanSpiegel: This graph is amazing. It shows that measuring #SARSCoV2 levels in municipal sewage almost perfectly predicts forthcoming #COVID19 cases with a full week's notice (R=0.994). It's one of several discoveries in this new study from @Yale: https://t.co/hZVdXebx2D. C-19 is #InThePoop https://t.co/shNzTKYoYP
3837
6721
OldBoltonian
MS | Physics | Astrophysics | Project Manager, Det 3 months ago
Hi biiingo, your submission has been removed for the following reason(s):
Papers on pre-print services such as arXiv and bioRxiv are not a peer-reviewed and are ineligible per Submission Rule #1b. If the research has been published in a peer-reviewed journal, please link to it in the comments and message the moderators for re-approval.
If you feel this was done in error, or would like further clarification, please don't hesitate to message the mods.
1
autotldr
3 months ago
This is the best tl;dr I could make, original reduced by 96%. (I'm a bot)
SARS-CoV-2 viral RNA was detectable in all samples tested and ranged from 1.7 x 103 virus RNA copies mL-1 to 4.6 x 105 virus RNA copies mL-1.
Regressing the adjusted COVID-19 testing data and SARS-CoV-2 virus RNA concentrations yield a value of 1,305 virus RNA copies/mL per new COVID-19 case report as the outbreak ascended, and 1,240 virus RNA copies/mL per new COVID-19 case report as the outbreak descended.
Average sludge SARS-CoV-2 RNA concentration time course data and average hospital admissions data with LOWESS smoothing; rescaled smoothed SARS-CoV-2 virus RNA concentrations and hospital admissions; smoothed sludge SARS-CoV-2 virus RNA concentration with smoothed COVID19 epidemiology curve; regression between smoothed virus RNA and new COVID-19 cases, slope=1,240 virus RNA copies/new case, R2=0.99; regression between smoothed virus RNA and new COVID-19 cases, slope=1,305 virus RNA copies/new case, R2=0.97.
Extended Summary | FAQ | Feedback | Top keywords: RNA^#1 SARS-CoV-2^#2 COVID-19^#3 concentration^#4 sludge^#5
1
autotldr
3 months ago
This is the best tl;dr I could make, original reduced by 96%. (I'm a bot)
SARS-CoV-2 viral RNA was detectable in all samples tested and ranged from 1.7 x 103 virus RNA copies mL-1 to 4.6 x 105 virus RNA copies mL-1.
Regressing the adjusted COVID-19 testing data and SARS-CoV-2 virus RNA concentrations yield a value of 1,305 virus RNA copies/mL per new COVID-19 case report as the outbreak ascended, and 1,240 virus RNA copies/mL per new COVID-19 case report as the outbreak descended.
Average sludge SARS-CoV-2 RNA concentration time course data and average hospital admissions data with LOWESS smoothing; rescaled smoothed SARS-CoV-2 virus RNA concentrations and hospital admissions; smoothed sludge SARS-CoV-2 virus RNA concentration with smoothed COVID19 epidemiology curve; regression between smoothed virus RNA and new COVID-19 cases, slope=1,240 virus RNA copies/new case, R2=0.99; regression between smoothed virus RNA and new COVID-19 cases, slope=1,305 virus RNA copies/new case, R2=0.97.
Extended Summary | FAQ | Feedback | Top keywords: RNA^#1 SARS-CoV-2^#2 COVID-19^#3 concentration^#4 sludge^#5
2
ClearlyMindless
3 months ago
Does the virus live and replicate in sewage?
Apparently cats and dogs get C19, what about other wildlife? Fish? Does our sewage contact rats that will spread it?
Do we have a big spread risk here?
1
mingy
3 months ago
> Does the virus live and replicate in sewage?
It does not. A virus needs a suitable live host cell to reproduce and those don't exist in sewage. They are detecting viral RNA fragments, not the actual viral particles. Those take much longer to degrade.
3
cornchipnsalsa
3 months ago
Snack sparingly on sewage? I joke but I’m actually concerned about this. Considering the US doesn’t separate water to potable / non-potable. My understanding water is processed and reused. Doesn’t water than have the potential for contamination
1
OrganOCreme
3 months ago
Yes they do, what do you think we are a third world country? We invented modern water filtration in the 19th Century. Turned the river in Chicago backwards and all that.
They only reuse water in Orange County and a few other places and it’s filters and is cleaner than fresh water. They do the exact same thing in Singapore with like half of their water and it’s fine.
That’s right if your reading this in Singapore and drinking a cool glass of water, it has millions of people’s pee in it...
2
Asmonymous
3 months ago
Now where on earth outside of labs do you get a .99 correlation in anything?
2
androstaxys
3 months ago
Basically nowhere?
Seems like data is not available for public/mass analysis.
Two things that make me rein in the high hopes here.
2
BusinessProstitute
3 months ago
Totally! If I were ever presented with that I would verify everything in dataset and calculations until i was absolutely certain I was right. I wouldn’t believe it myself without a ton of scrutiny, no way anyone else would.
2
Covfefetarian
3 months ago
Is the data made public anywhere, so we could replicate the analysis?
(Fellow statistician here)
7
stat-chick
3 months ago
Looks like they regressed the “smoothed” values, which is not appropriate.
2
Thetrashman1812
3 months ago
If I’m correct though (and it’s been a while since I’ve had to read reports like this) isn’t the smoothed data different and higher than the .994 result. Also with how little deviation there is in the numbers what is smoothing actually doing here besides removing some wild ass outliers. Once again haven’t really thought about statistics and scientific research in while so maybe I’m completely off base. These are just my first thoughts after a quick read through of the results
2
NoMadTruffle
3 months ago
R=0.996 is for RNA levels vs hospital admissions after adjusting for 3 days lag, R=0.994 is for RNA levels vs cases after adjusting for 7 days lag. In both cases they used smoothed data. I can understand using smoothed data if they're going for proof of concept to predict future admissions/cases (I'm assuming the LOWESS smoothing method they used is a moving average that only relies on past data points, if not then wouldn't be applicable in real life), but then it should be clearer that the correlation is near perfect only AFTER applying this heavy processing.
Still it would have been better to report R for both non smooth and smooth data. They did plot individual data for hospital admissions and based on that the R would not be anywhere close to 0.9; they didn't even show individual points for cases. The paper hasn't been peer reviewed yet so they might have to modify it quite a bit.
3
-Natsoc-
3 months ago
Usually “perfectly” is not among the vocabulary of scientific studies, this time though it appears to be justified
1
lex52485
3 months ago
Statistician here. Holy shit is right (assuming methodology is sound, which I haven’t looked at yet)
12
Big_Tree_Z
3 months ago
Yeah, some commenters have probably explained it enough already, but an R value that high generally means there’s a very strong correlation.
Another commenter mentioned 98.8% coverage of variance within the results. That’s the R^2 value. Literally square the R.
2
NohPhD
3 months ago
The COVID virus can be found in feces in sewage from infected individuals, even before they are symptomatic.
In lieu of massive individual testing, the detection of and amount of SARS-CoV-2 in sewage can be a useful proxy for public health experts to forecast COVID-19 impact to local governments.
Individual testing is a lagging indicator (yeah, you’ve got C19 or no you didn’t have C19). The sewage test results are a leading indicator, meaning it can be used to predict things like potential patient loads up to a week in advance.
29
lhospitalsrule
3 months ago
And it is a VERY strong relationship, like nearly perfectly correlated. The model explains 98.8% of all variance, which is kind of incredible.
15
NoMadTruffle
3 months ago
Too incredible because they did a lot of heavy processing. They did something like a moving average (but I'm not sure if the LOWESS method they used only takes into account past data, i.e. Could they use this smoothing method in real life to build the line as they go along or not?). And then shifted the lagging curve a few days ahead so they would overlap nicely with each other, THEN used those points in the smoothed out line to calculate R. Fig 2A shows the scatter of true data points for RNA levels and hospital admissions - the R for unsmoothed data would be a much more realistic, lower number. I'm not an expert in modelling future clinical outcomes so maybe this is a common method to do things? Idk but seems disingenuous and easily sensationalised to me.
3
mendelevium34
3 months ago
"SARS-CoV-2 RNA concentrations were a seven-day leading indicator ahead of compiled COVID-19 testing data and led local hospital admissions data by three days. Decisions to implement or relax public health measures and restrictions require timely information on outbreak dynamics in a community."
4
AutoModerator
3 months ago
Thanks for your submission. New posts are pre-screened by the moderation team before being listed. Posts which do not meet our high standards will not be approved - please see our posting guidelines. It may take a number of hours before this post is reviewed, depending on mod availability and the complexity of the post (eg. video content takes more time for us to review).
In the meantime, you may like to make edits to your post so that it is more likely to be approved (for example, adding reliable source links for any claims). If there are problems with the title of your post, it is best you delete it and re-submit with an improved title.
I am a bot, and this action was performed automatically. Please contact the moderators of this subreddit if you have any questions or concerns.
1
Madman200
3 months ago
This article is great. There's so much you can learn about a population from it's sewage, and indeed testing sewage may give a better picture of the scale of a disease outbreak.
However, the study is discussing a US metropolitan area centered around New Haven, Connecticut. It's very possible we might see similar results in Canadian locations but this study isn't making any predictions about cases in Canada.
8
rationalphi
3 months ago
Fun fact: Statistics Canada did a pilot study on detecting levels and trends in regional drug use via waste water testing.
4
travman6
3 months ago
I remember that one, Atlantic Canada took 1st place for cannabis smoking. Trailer park boys is an exaggeration, but only a little bit.
Edmonton took 1st place for Meth.
2
christ344
3 months ago
I don’t know the top comment god but I know like the secretary of his underling, top comment demigod.
1
h3yn0w75
3 months ago
Good summary and graph to outline the findings here : https://twitter.com/brennanspiegel/status/1265119535901732865?s=21
3
Alces_alces_Alces
3 months ago
In the waste water, from your stool and urine, inactive parts of the virus can be found. It is part of how your body removes the waste from mounting an immune response.
17
bythesword86
3 months ago
Thank you, I'm gonna read the paper later but I'm assuming this means the virus is dead while in the sewage?
4
AutoModerator
3 months ago
Thank you for posting to r/CanadaCoronavirus. Please read our rules.
Please remember that all posts and comments should reflect factual, truth-based discussion. The purpose of this subreddit is to share trustworthy resources and ensure Canadians are as informed and educated as possible.
We will not tolerate racism, sexism, or harassment of any kind.
Any comments or posts made contrary to these values will be subject to review by the Mod team
I am a bot, and this action was performed automatically. Please contact the moderators of this subreddit if you have any questions or concerns.
1
Trending this week..
Join Our Newsletter
Post a Job
Teams Hiring Now
-
TokenSoft is the volume leader in compliant token sales.
-
The open protocol for tokenized debt.
-
A secure online platform for buying, selling, and storing digital currencies.
-
A second layer, off-chain scaling proposal for bitcoin.
-
Ensuring the blockchain is inexpensive and accessible to everyone.
-
An open protocol for decentralized exchange on the ethereum blockchain.
- See all