Spring 2022 -- Moving Online?

[aprof]

Deaths lag cases by ~2.5 weeks for obvious reasons. We are at or just past peak deaths right now so those numbers are grim, but using deaths to predict the trajectory of the next few weeks or even the current level of community transmission is not accurate. Cases are now falling rapidly, though this may slow down a little with the BA.2 variant. While there are likely to be other (hopefully smaller) waves, it's fairly safe to say that this one is receding. 

We need to quickly adapt our personal and societal risk models to the current situation, both when it is getting worse and when it is improving. My view is that students need the informal interactions and mentoring both from peers and faculty that they cannot get in virtual mode.  I am very glad that we have remained in person this AY despite worries from a handful of faculty and students.

[Aster]

Deaths lag cases by ~2.5 weeks for obvious reasons. We are at or just past peak deaths right now so those numbers are grim, but using deaths to predict the trajectory of the next few weeks or even the current level of community transmission is not accurate. Cases are now falling rapidly, though this may slow down a little with the BA.2 variant. While there are likely to be other (hopefully smaller) waves, it's fairly safe to say that this one is receding. 

Sort of. Yes, there is a lag. Yes, it's 1-3 weeks. It's not a big deal. Just because data is reported in weekly dumps does not mean that we look for patterns in weekly dumps.

Also yes, hospitalization and death rates are much more accurate than positivity reporting. Positivity reporting accuracy went out the window once self-test kits became universally available a few weeks ago. Hospitalizations and death rates do not meaningfully suffer that new loss bias.

Also yes, tracking week-by-week statistics is like using daily weather reports to evaluate climate change. Near-time interpretation of patterns often turn out not to be patterns.

There's lots COVID right now. Lots of sick, and lots of death. One of my coworker's spouses works in a local critical care unit. Her caseload is still sky-high with COVID patients, and we're in the one of the most vaccinated parts of the state. That's not 2-week-old data, that's the state of of one of our regional critical care wards as of three days ago.

Continue to quarantine, continue to offer remote learning where you can, and continue to protect yourselves from infection.

[Puget]

It really isn't in doubt that cases are dropping rapidly in nearly every part of the US now. You don't have to believe the testing data, you can look at the sewage data. Hospitalizations are also down about 25% in the last two weeks nationally. The NYT dashboard is a good place to track these things. That isn't to say you shouldn't still be cautious if you are at high risk, but things are clearly improving, and rapidly so.

[Caracal]

There's lots COVID right now. Lots of sick, and lots of death. One of my coworker's spouses works in a local critical care unit. Her caseload is still sky-high with COVID patients, and we're in the one of the most vaccinated parts of the state. That's not 2-week-old data, that's the state of of one of our regional critical care wards as of three days ago.

Continue to quarantine, continue to offer remote learning where you can, and continue to protect yourselves from infection.

I don't think that advice really helps many of us now. The problem is that it sounds great to say "protect yourself from infection," but what does that mean? I'm obviously wearing masks in public indoor spaces. I'm not eating indoors at restaurants or bars. However, we went to family gathering a few weeks ago. Everyone who was old enough was vaccinated and got rapid tests, but it was still inside and there were a fair number of people there. We've started taking our kid to indoor places like movie theaters now that he's vaccinated. Obviously this isn't the way to completely minimize the risk of infection, but these things are important or add a lot to the quality of our lives. If I'm infected, its very unlikely I'll get seriously ill. I'm sympathetic to people who are at more risk for various reasons, but extreme caution on my part is not going to end the pandemic.

Same thing with remote classes. I hate teaching remotely. I find it depressing. So do my students. There really hasn't been much transmission in classrooms. And again, my risk is pretty low.

[pgher]

It really isn't in doubt that cases are dropping rapidly in nearly every part of the US now. You don't have to believe the testing data, you can look at the sewage data. Hospitalizations are also down about 25% in the last two weeks nationally. The NYT dashboard is a good place to track these things. That isn't to say you shouldn't still be cautious if you are at high risk, but things are clearly improving, and rapidly so.

True, but at least in my state, the case rate is WAY above previous peaks. Down from the most recent peak doesn't mean low.

[Puget]

It really isn't in doubt that cases are dropping rapidly in nearly every part of the US now. You don't have to believe the testing data, you can look at the sewage data. Hospitalizations are also down about 25% in the last two weeks nationally. The NYT dashboard is a good place to track these things. That isn't to say you shouldn't still be cautious if you are at high risk, but things are clearly improving, and rapidly so.

True, but at least in my state, the case rate is WAY above previous peaks. Down from the most recent peak doesn't mean low.

Sure, of course you need to pay attention to local conditions. Where I am we're back below last winter's peak and our campus positive rate is back down to where it was last semester, about 0.3%. Even if it was higher though, the big difference from previous peaks is we are all vaccinated and boosted now, so risk of serious illness is very low. Obviously someone who is immune compromised is going to have to be more cautious, but for most of us the risk is much lower even if the case count is higher.

[namazu]

It really isn't in doubt that cases are dropping rapidly in nearly every part of the US now. You don't have to believe the testing data, you can look at the sewage data. Hospitalizations are also down about 25% in the last two weeks nationally. The NYT dashboard is a good place to track these things. That isn't to say you shouldn't still be cautious if you are at high risk, but things are clearly improving, and rapidly so.

True, but at least in my state, the case rate is WAY above previous peaks. Down from the most recent peak doesn't mean low.

Sure, of course you need to pay attention to local conditions. Where I am we're back below last winter's peak and our campus positive rate is back down to where it was last semester, about 0.3%. Even if it was higher though, the big difference from previous peaks is we are all vaccinated and boosted now, so risk of serious illness is very low. Obviously someone who is immune compromised is going to have to be more cautious, but for most of us the risk is much lower even if the case count is higher.

Sounds nice!  Around here, campus screening test positivity was north of 10% last week (with the caveat that this screening testing mostly applies to people who still aren't vaccinated, as far as I can tell).  This is already a big improvement over the previous week.  We've clearly crested this wave, but in our area it's still a non-negligible concern.

A lot of (mostly fully-vaccinated and boosted) faculty have vulnerable family members (e.g. spouse scheduled to start cancer treatment who can't risk getting infected) and/or kids in daycare (where frequent closures are a real problem). There's been some flexibility granted from above (for both faculty and students) even once in-person classes resumed after an online start to the term, which I think a lot of people need/appreciate right now.

[Caracal]

It really isn't in doubt that cases are dropping rapidly in nearly every part of the US now. You don't have to believe the testing data, you can look at the sewage data. Hospitalizations are also down about 25% in the last two weeks nationally. The NYT dashboard is a good place to track these things. That isn't to say you shouldn't still be cautious if you are at high risk, but things are clearly improving, and rapidly so.

True, but at least in my state, the case rate is WAY above previous peaks. Down from the most recent peak doesn't mean low.

Sure, of course you need to pay attention to local conditions. Where I am we're back below last winter's peak and our campus positive rate is back down to where it was last semester, about 0.3%. Even if it was higher though, the big difference from previous peaks is we are all vaccinated and boosted now, so risk of serious illness is very low. Obviously someone who is immune compromised is going to have to be more cautious, but for most of us the risk is much lower even if the case count is higher.

Sounds nice!  Around here, campus screening test positivity was north of 10% last week (with the caveat that this screening testing mostly applies to people who still aren't vaccinated, as far as I can tell).  This is already a big improvement over the previous week.  We've clearly crested this wave, but in our area it's still a non-negligible concern.

A lot of (mostly fully-vaccinated and boosted) faculty have vulnerable family members (e.g. spouse scheduled to start cancer treatment who can't risk getting infected) and/or kids in daycare (where frequent closures are a real problem). There's been some flexibility granted from above (for both faculty and students) even once in-person classes resumed after an online start to the term, which I think a lot of people need/appreciate right now.

Yeah, I think that's the key. I get the impression that at some places you have this idea that there are large swathes of the faculty who would love to be online forever and any concessions to special circumstances can't be allowed. I really don't see that. Most of us want to teach in person and don't need to be frogmarched back into the classroom, but will feel a lot better about our workplaces if there's an understanding that while most of us are low risk right now there are some people who have real concerns.

[Aster]

I do not recommend that anyone seriously look at case numbers, or even positivity case ratios. Data scientists and health officials will tell you that their reliability went down the toilet once that self-test kits became ubiquitous in the month of January. Most people in the U.S. are no longer going to public testing centers. They're doing self-tests. And very few of those self-tests get reported to any health departments.

Many of the covid testing centers are now closing, or scaling back their operating hours. But they're not doing that because less people are sick. They're doing it primarily because people can test themselves for covid at home. Some dumb$%^ politicians are looking at empty testing centers and now proclaiming that "covid is over!" Well, that's a lie. All you have to do is look at the overall hospitalizations and death rates. For those much more reliable statistics, the overall numbers are dropping, but they're dropping much, much slower than the (biased) positivity test case reporting.

Also, be careful with how covid statistics are reported. When you read numbers like "25% fewer hospitalizations", that may sound great at first blush. But if a 25% drop in hospitalizations was baselined at some record-high benchmark (which is how many are currently reporting it), a 25% drop might only mean that you're hospitalizations are still heavily overloaded, but just a little less now. So they still might be sky-high.

There are easy ways to quickly reference visual graphics and raw data on your area's covid-related hospitalizations and death rates. Google has this all set up for covid. When you open up google, just type in your county or state name along with "covid", and then select the "statistics" icon. if your state and county are participating in these reports, you'll get graphs, data tables, and directed weblinks to local, state, and national reporting sites. If your state and county are not participating, you'll still get access to national data.

[Puget]

I do not recommend that anyone seriously look at case numbers, or even positivity case ratios. Data scientists and health officials will tell you that their reliability went down the toilet once that self-test kits became ubiquitous in the month of January. Most people in the U.S. are no longer going to public testing centers. They're doing self-tests. And very few of those self-tests get reported to any health departments.

Many of the covid testing centers are now closing, or scaling back their operating hours. But they're not doing that because less people are sick. They're doing it primarily because people can test themselves for covid at home. Some dumb$%^ politicians are looking at empty testing centers and now proclaiming that "covid is over!" Well, that's a lie. All you have to do is look at the overall hospitalizations and death rates. For those much more reliable statistics, the overall numbers are dropping, but they're dropping much, much slower than the (biased) positivity test case reporting.

Also, be careful with how covid statistics are reported. When you read numbers like "25% fewer hospitalizations", that may sound great at first blush. But if a 25% drop in hospitalizations was baselined at some record-high benchmark (which is how many are currently reporting it), a 25% drop might only mean that you're hospitalizations are still heavily overloaded, but just a little less now. So they still might be sky-high.

There are easy ways to quickly reference visual graphics and raw data on your area's covid-related hospitalizations and death rates. Google has this all set up for covid. When you open up google, just type in your county or state name along with "covid", and then select the "statistics" icon. if your state and county are participating in these reports, you'll get graphs, data tables, and directed weblinks to local, state, and national reporting sites. If your state and county are not participating, you'll still get access to national data.

As I said before, you don't have to trust case numbers, you can look at the sewage data, which is telling the exact same story right now.
You can also look here for hospital capacity tracking: https://covidcircuitbreaker.shinyapps.io/circuitbreaker/
Currently, relatively few counties have capacity problems.

Look, I'm all for sensible precautions, but an interesting tendency has developed for certain people to discount any good news, no matter how strong the data supporting it is. This isn't productive-- we need to be able to adjust our risk assessment and conditions change, in both directions.

[aprof]

Even better than using severely lagging death numbers (or even case numbers, which may be biased as you say and also lag "on the ground" reality by a few days), wastewater data shows a significant downtrend, with national numbers returning to about Dec. 1 levels after seeing a peak around NYE.
https://biobot.io/data/

By the way, these wastewater data align very well with Google trends for "covid test". An indirect metric to be sure, but it gets around the issue of people testing at home and not reporting.

Not to say that we are in the clear by any means, but things are improving quite a bit.

[Caracal]

I do not recommend that anyone seriously look at case numbers, or even positivity case ratios. Data scientists and health officials will tell you that their reliability went down the toilet once that self-test kits became ubiquitous in the month of January. Most people in the U.S. are no longer going to public testing centers. They're doing self-tests. And very few of those self-tests get reported to any health departments.

Many of the covid testing centers are now closing, or scaling back their operating hours. But they're not doing that because less people are sick. They're doing it primarily because people can test themselves for covid at home. Some dumb$%^ politicians are looking at empty testing centers and now proclaiming that "covid is over!" Well, that's a lie. All you have to do is look at the overall hospitalizations and death rates. For those much more reliable statistics, the overall numbers are dropping, but they're dropping much, much slower than the (biased) positivity test case reporting.

Also, be careful with how covid statistics are reported. When you read numbers like "25% fewer hospitalizations", that may sound great at first blush. But if a 25% drop in hospitalizations was baselined at some record-high benchmark (which is how many are currently reporting it), a 25% drop might only mean that you're hospitalizations are still heavily overloaded, but just a little less now. So they still might be sky-high.

There are easy ways to quickly reference visual graphics and raw data on your area's covid-related hospitalizations and death rates. Google has this all set up for covid. When you open up google, just type in your county or state name along with "covid", and then select the "statistics" icon. if your state and county are participating in these reports, you'll get graphs, data tables, and directed weblinks to local, state, and national reporting sites. If your state and county are not participating, you'll still get access to national data.

Yeah, this is not a good way to interpret data. First of all,  test numbers have always severely underestimated actual infections-at varying levels depending on availability of testing and other factors. Home tests have become more widely available,  but not that widely available and there were plenty of them before so trying to argue that the drop in cases is mostly because of home tests is a reach.

More importantly, hospitalizations and deaths always lag, and go down more slowly than cases even when they catch up. Even when fewer people are being admitted to hospitals, the most seriously ill will still be there for a while and, unfortunately, some of those people will die weeks later. This has been the pattern repeatedly. So, arguing that case numbers are actually higher because hospitalizations are going down more slowly makes no sense. They are going down the way we'd expect them to.

Besides, the same thing has happened in lots of other countries where omicron arrived earlier. Rapid rise, rapid fall. Britain has very good surveillance systems that allow for collection of more data than just reported cases and they all show a big drop as well. Lots of Covid of course, but things are improving.

[Aster]

Health officials, including the head of the CDC, are basing their local health advisories on hospitalizations and death rates. Because those are actual concerns. If you don't believe me, Rochelle Walensky literally said it earlier this week. As I don't think anyone of the forum is an infectious disease expert, I'm forced to go with the expert's recommendations.

But that's not to say that case counts are totally useless and that sewage isn't a cool novel approach to geek-out with.

[Caracal]

Health officials, including the head of the CDC, are basing their local health advisories on hospitalizations and death rates. Because those are actual concerns. If you don't believe me, Rochelle Walensky literally said it earlier this week. As I don't think anyone of the forum is an infectious disease expert, I'm forced to go with the expert's recommendations.

But that's not to say that case counts are totally useless and that sewage isn't a cool novel approach to geek-out with.

Right, that's reasonable in terms of local health advisories. But that's about trying to look at the impact Covid is having and the current strain it is putting on health systems. If hospitalizations are still quite high, you might not want to do things like loosen mask mandates if that might cause case numbers to go down more slowly resulting in more hospitalizations on the downward part of the curve. On the other side, you may be able to have case numbers that would have been quite bad with delta, but with omicron combined with vaccine and disease induced immunity, they won't end up leading to that many hospitalizations.

That's different from claiming that the falling case numbers shouldn't be trusted because hospitalizations aren't falling as rapidly. That's not what Walensky is saying.

[Mobius]

A lot of folks' identities are wrapped up in being extremely cautious about Covid and engaging in Covid shaming. It's going to be rough for them as mask mandates were that last real public health-related restriction.