Thank you for your posts. I send people to them all the time. One literally said, "She is pure gold."
I'm working as a volunteer on a COVID-related issue: getting Evusheld to the immunocompromised. Gov't bought 875,000 doses 6 months ago and only 1/8 of it has been distributed to patients. Problem seems to be that too few patients and doctors know of its usefulness. I hope you'll put in a little mention of Evusheld and what it does in a blog post now and then, to raise awareness.
"If you’re high risk, please consider getting a prescription of Paxlovid" Just a couple of questions: 1) How is "high risk" defined in this case? I'm 74 but have no other risk factors, wear an N95 mask in all public indoor situations, and have had both boosters. Am I considered "high risk"? 2) Is Paxlovid something I should have on hand just in case I get infected, or should I ask my PC doc for it if and only if I test positive?
We're off on a two week European trip in two weeks, so I'm wondering whether or not I should just have some on hand. Thanks.
Stephen, I wonder if part of the answer is how easy it would be for you to get Paxlovid in Europe (and, possibly, whether US insurance would pay for it). I may one day face the same question (being 74 and no other risks); am I protected 'enough' being double-boosted? Also I wonder if it prevents immune systems from fighting the virus on a future encounter (given already being vaccinated). How much more do we need to know to inform decisions like you pose?
Thanks, that answers my questions. That wasn't clear in the original context. I was aware that Paxlovid wasn't OTC but I didn't know whether it could be prescribed for prophylactic use.
Very disheartening that vaxed and boosted folks are on the wrong side of the odds - even still being careful. Great fodder for those who avoid vaccines. Thanks as always for the update.
We seem to be near the point where we can compare clinically-serious illness among vaccinated (in various subsets) vs. unvaccinated, even adjusted for gender and age. But are we able to say by how much being vaccinated reduces the risk of getting or transmitting Covid (understanding there are nuances e.g. wearing a good mask etc.)? I recently attended a conference (virtually, tho have been double-boosted) where in-person attendees were required to either submit proof of vaccination or "a test within 72 hours." How much evidence is there that either one of these actions would decrease the probability of transmitting the virus? (I estimate <25% were masked based on session videos)
Thanks for this, KB. The paper in medRxiv looked only at neutralizing antibodies (as you point out) - but I'm having trouble understanding how that translates into their conclusion that BA.4 and BA.5 'have potential to result in a new infection wave," **IF** people have been previously infected and/or vaccinated and their T-cells are working? The medRxiv website says 125 media posts have already been made on this paper!! For example, should we expect that long-ago infection or vaccination with measles or polio is still protecting us? Would we have neutralizing antibodies if tested? Or is our lingering protection (if any) down to our T-cell memory? I also have questions about the statistical test used in their paper but should probably discuss that elsewhere...
Thanks, KB! I'll dive into the Cell paper. Also an audio interview in this week's New Engl Journal on immunity markers. We need more clarity and/or better communication on the clinical meaning of neutralizing antibodies because people (and popular journalists) are reading papers like this, implying that vax, boosting +/- Omicron breakthrough is at best modest at protecting against BA.4/5 based on neutralizing antibodies (https://www.science.org/doi/10.1126/sciimmunol.abq2427). Meanwhile, I hope the smallpox vaccine I got 70 years ago is holding out.
EXCELLENT Post! Question: early on, the statistics were that males did much worse with COVID-19 infections than women in most age groups. What are the current stats for women vs. men for things like: [1] current stats overall? [2] age-adjusted overall? [3] those protected by full vaccination? [4[ Etc.. THANK YOU!
I don't think we're out of the woods yet and undoubtedly with this large a reservoir of infected people and animals, the virus will continue to select for more transmittable variants, despite our countermeasures. To repost an earlier remark, and thinking numerically about it. It seems to me that the pandemic, far from being over, rang the planet like a bell setting up disequilibria everywhere and it will take a while, maybe three to five years for things to settle down to a new normal (which won't be the old one - it's gone) even if the virus were to disappear today. In effect, the planet itself has long covid.
Precisely. I see turbulence introduced across the board in human systems and it will likely take a long time for them to damp out (even assuming we don't experience more waves).
Have you heard of the musical duo Flanders & Swann? One of their songs is "The Gasman Cometh", about a hapless homeowner trying to get their gas tap fixed.
T'was on the Monday morning, the gas man came to call,
The gas tap wouldn't turn, I wasn't getting gas at all;
He tore out all the skirting boards to try and find the main,
And I had to call a carpenter to put them back again!
Oh, it all makes work for the working man to do...
T'was on the Tuesday morning, the carpenter came round:
He hammered and he chiseled and he said "Look what I've found!
Your joists are full of dry rot, but I'll put them all to rights!"
Then he nailed right through a cable and out went all the lights!
Oh, it all makes work for the working man to do...
T'was on a Wednesday morning, the electrician came:
He called me 'Mr. Sanderson', which isn't quite the name;
He couldn't reach the fuse box without standing on the bin,
And his foot went through a window, so I called the glazier in!
Oh, it all makes work for the working man to do...
T'was on the Thursday morning, the glazier came along
With his blowtorch, and his putty, and his merry glazier's song;
He put another pane in, it took no time at all
But I had to get a painter in to come and paint the wall!
Oh, it all makes work for the working man to do...
T'was on a Friday morning the painter made a start,
With undercoats and overcoats he painted every part,
Every nook and every cranny; but I found when he was gone
He'd painted over the gas tap and I couldn't turn it on!
Oh, it all makes work for the working man to do...
On Saturday and Sunday they do no work at all;
So it was on the Monday morning that the gas man came to call!
I feel like that's kind of the situation we're in now. Every intervention, while necessary, puts a strain on something else, in ways we don't fully understand.
Well put. One problem we might be facing is that highly articulated systems get increasingly unstable unless there is substantial "give" built into the system (think of expansion seams on Bridges). Our systems are perhaps lacking in the give necessary to function well in environments of imperfect information, lack of redundancy, lack of capacity and so forth. We're definitely learning and will do better next time I imagine.
Yeah, I think one way of putting more give into the system would have been to have more graceful, discreet versions of various mandates. For instance, a good infrastructure for promoting masks on, say, airplanes would be to run "open tabs" like at bars. The airline would keep your credit card on file, and if they catch you in breach of their mask policy they'd just run your card.
Thanks for another informative post! A few comments:
1) Given that daily cases are vastly undercounted, by at least one order of magnitude, what is the value in continued publication of the surge chart? Individuals who use it to assess risk will quickly be misled, increasing the likelihood of poor decisions and infection.
2) Given that it’s possible to adjust the daily case chart based on sampling so that it more closely reflects reality, why doesn’t the CDC do this? They could footnote their assumptions for full disclosure. Instead, they continue to publish a chart that makes it appear they have the pandemic under control, which they do not. This chart is worse than worthless, it leads to risky behavior and unnecessary transmission. Either fix the chart for accuracy or retire it.
3). The NYT recently published an article stating that once symptoms begin, vaccinated individuals who use home tests are more likely to test false-negative for several days. I know several vaccinated and boosted people who become symptomatic, test negative at home, wait a day or two, and test negative at home again. They do everything right, assume they don’t have Covid, and go back to work or school, right at the moment they are becoming most infectious. So while it’s great that vaccines and boosters are preventing severe illness and death, are they inadvertently increasing transmission? While this early-false-negative phenomena has always been true for home tests, its even more pronounced with omicron, especially for vaccinated/boosted individuals where the vaccines "mask" a true positive in the early days. Here’s the NYT article (click “show more” to see full article): https://www.nytimes.com/explain/2022/coronavirus-questions#at-home-covid-test-negative
4) Are vaccines and boosters really preventing hospitalization and death today? Or did previous surges take out the “low hanging fruit” (most vulnerable) and those who are left are healthier and/or have natural immunity?
1. Look at the rate of change in the curves. Case rates increasing are still a good indicator.
2. CDC is an almost pure academic institution. Their choices of metrics are based on how their internal conventions.
3. There's explicit CDC guidance on how long to isolate following symptoms and/or a positive test of any sort. 3-6 days after symptom onset, especially in vaccinated persons, the amount of detectable antigen is decreased to near background levels. If the persons in question consider the guidance and follow it, when they're out of isolation, they are virus free. If they retest on a positive and break isolation early, there remains a chance they're still infectous. I can't access the NYT article (paywall) but the background study they are likely talking about did not reach the conclusion you present.
4. There's likely some truth to the concept that the population was culled significantly, but there are case control studies that support the concept that hospitalization, significant disease and increased mortality are reduced in vaccinated individuals.
#1 people look at absolute level, not some fancy “rate of change” concept from high school calculus. The chart makes today’s transmission level (and risk) look low. Based on anecdotal evidence, today’s surge is probably just as bad as January 22’s peak, if not worse
#2 no disrespect, but who cares? Know your audience. If the chart is public, the public is your audience. That is the lens through which the chart’s value must be judged. What information is being communicated to the public? Is it correct? What actions will the public take in response?
#3 the large print instructions on the home test say something like “test today, test 24 hours later.” When someone is freaked out about Covid, they don’t stop to read the fine print or visit CDC website. They don’t quarantine 5-7 days before testing. Your “isolate 3-6 days” is nowhere on the CDC website, nor did I find language where the CDC breaks out isolation and testing guidance based on vaccination status (although quarantine guidance is broken out based on vaccination status). Nor did I see the phrase “virus free” (try saying that to someone suffering from long Covid). As a side note, the CDC website funnels readers based on whether they seek to “isolate” or “quarantine” when in fact most Americans use these terms interchangeably or are unaware of the nuances as to how these terms differ. Why not use simple and direct language like “exposed?” and “got Covid?” (Again: know you’re audience)
Since you are unable to access the NYT article (paywall), here is a direct quote: "The harder your immune system is working to tamp down the virus, particularly an immune system supercharged by vaccine antibodies, the more likely you are to get an early negative result on a rapid test, even if you're infected."
#4 the “death chart” lines for vaccinated and unvaccinated are nearly touching *today*. Unless the chart is greatly enlarged and underlying data points are provided, when one line appears to nearly touch the other, from an intuitive sense, it cannot be 17x. Also, as vaccines wane and natural immunity increases, the distinction between “vaccinated” and “unvaccinated” becomes blurred
I'm unable to reply right now: Too busy. However, I will follow up ASAP. One thought: I spent over 18 months avoiding news media stories on COVID. Too many are snippets taken out of context, where the context is important. The NYT quote is interesting but incomplete immunologically.
Thank you for keeping up with all this. Your posts help me to protect myself, despite mask-fatigue, and I pass the info on. I am double-boosted and intend to continue with further immunization when they become available. Dying on a ventilator does not attract.
Can you address what is known about flying now that there's no mask mandate? Any new empirical evidence? My spouse just got COVID after air travel, and that was his primary exposure. He and his colleague both took the same plane on Thursday (that sat on a runway for a while), and they haven't seen each other since. They came down with it today (4 days later) on antigen test. They were wearing KN95s. Are people with N95s actually avoiding it? I have to go to a conference in 3 weeks. If I actually avoid COVID from my spouse, I'm wondering if a N95 will be good enough or if I'm just as likely to get COVID.
In your paragraph “battle of Omicron” you noted BA.4 and BA.5 are gaining traction and they are reinfecting folks with Omicron. Hence their ability to spread quickly. The key question is what about the folks that aren’t getting infected with “old Omicron” and or the new variants. Are they not getting infected because they are not exposed, or their immune system does better? Since we cannot do random controlled tests and have to rely on vagaries of observed data, this is hard to answer, but it is a critical question despite structural data issues.
Second, it is important to be careful with models that are conjecture, for example the real COVID rate, even if they have fancy math. Additionally, what we don’t get when a COVID positive test is reported is “how sick” is the person.
“Massive” case surge needs to be defined. Massive in that your chances are 1 in 10 of going to a movie and leaving with COVID or 1 in 10,000. Risk needs to be made real.
I have been a big proponent of vaccines – love the MRNA technology. At this point my suggestion is to stop reporting separately on the unvaccinated. It isn’t helping and there are now data complications. We don’t separately report on folks with diabetes.
Yes I do a ton of analytics without borders and have been since 1977 when I joined IBM at age 23. This work have covered planning and scheduling to colon cancer.
This is getting more and more frustrating as time goes on. I'm 77 years old, vulnerable and really want to continue living for a "few" more years. There's no one to trust anymore. Except you. Protection from four shots is waning, leaving us more vulnerable to new variants. Guess I should continue to mask up, avoid crowds and strangers until our medical professionals figure this out.
As best I can determine, if I wear a nanofiber respirator mask at all times when I am indoors that has claimed protection of 98%, and if I fit it properly so that I can not feel any leaks around the edges even when I puff a breath, then I should be completely safe against the covid virus, regardless of how many people near me are emitting the virus into the indoor air, regardless of my personal risk level, and regardless of the fact that I am indeed vaccinated and twice boosted. Am I missing something? Why don’t 98% masks suffice?
Probably the answer depends on stuff like: What does "98%" mean (how does whatever they do to certify the 98% translate into questions like "What is the probability of me inhaling 'some' SARS-Cov-2 if I'm sitting next to a person on a plane who has asymptomatic Covid and not wearing a mask for 4 hours?" Which is dependent on more stuff like how many particles the person is transmitting, whether we're talking, etc. And then what's the probability of me becoming infected based on all those considerations? And what does "98" mean? That over some standardized time period 2 out of 100 viral particles get through? Because if it's "with each breath," that's different than "over one hour."
Case rates among the boosted are also increasing bc the population of people who are most likely to be boosted -- the elderly and people with weakened immune systems -- are also more vulnerable to infection due to age/health. I'd imagine we're going to approach a leveling off soon where we see more hospitalizations/deaths are among the boosted than the unvaxxed, as a function of the a priori higher risk presented by age and co-morbidities.
I don't think I can agree completely with your premise. Yes, the older population has higher vaccination rates but I think what we're seeing, as @Katelyn has said, is a shift in who's getting NAAT tests vs who's relying solely on home tests and luck (not consulting a doc when the at-home rapid-antigen tests are positive). The testing system in this country was broken initially, and the termination of free tests and mass-testing sites has not worked in favor of assembling reliable statistics.
I'm about to get back to my 4 hr/night schedule from training... We're also renovating a house and I'm one of the workers. Over 250k reviewed and prereview papers so far that I can access. Skimming the surface is an understatement.
Once again - thank you for all your data driven, fact based work! I’m still wondering about the more difficult to answer question around longer term immunity (B and T cells). I know it’s not easy to measure (like antibody levels) but I really do wonder why there hasn’t been more data that shows long term durability.
Several studies have addressed cellular immunity and its persistence. At least one such study suggested B- and T-cell training did persist even if the actual B-cell initiation on exposure was much more tied to initiating training, where those classes of lymphocytes were trained and able to recognize, then subsequently start producing an active immune response in a matter of hours to days.
Wonderfully thoughtful post, Katelyn - thank you! Denominators question: In the Y-axes labeled incidence/100,000 population - is that the whole population (of whatever region)? If so, where can we find charts where the denominators are various statuses of prior infection, vaccinated, not vaccinated, etc.? Some preprint papers sort of get at this question. Does getting that information depend on rigorous analysis of large datasets of people's entire continuum of healthcare? For example, in health insurance claims or large healthcare systems in which members access essentially all of their healthcare? Setting up standardized queries takes a lot of work by people who deeply understand the nuances of using real-world (that is, non-clinical trial) evidence, but once set up could be run periodically for updates. Or is this already happening?
One of the biggest problems we've faced in this pandemic has been the scarcity of reliable data. We've never comprehensively tested the population either completely, or statistically. PCR testing has varied from lab to lab. Reliable antibody testing has only recently evolved, and most available antibody tests even today can differentiate between nucleocapsid and spike/receptor binding domain antibodies.
And then there's reporting. Some states have stopped reporting to CDC at all, others have decided what they will and won't report. Others have decided to report numbers in the categories CDC's requested but use different criteria than the recommended ones to report a positive case, thus skewing their result reporting. There are documented cases where death determinations have been changed to not reflect COVID cases if any other potential cause of death could rationally be listed.
I've been working a little bit with real-world data but have had to back off recently due to some family business. I hope to get back to some of the coding needed to do the work on a predictive system that assimilates historical data and and produces a slightly more robust tool to predict vulnerability. Finding a good and complete data source is a real problem.
As someone who has Covid at Easter I am wondering if when people get reinfected after omicron, do we know if their course of illness is similar to what they had before? I “just” had a cold though I tested positive for 2 weeks. Would another bout likely be similar?
There's clinical consensus that generally, the significant variants at play now are producing markedly less severe symptoms in most people, especially in those who were vaccinated, or were vaccinated either before or after vaccination. The majority of cases I've seen or heard about over the last several months reported the equivalent of a bad "head cold". If you tested positive for 2 weeks, what tests were you using? Generally, after testing positive with a home kit, there's no point in testing again 'til about a week after symptom onset.
I've seen at least one study noting that omicron failed to produce an acceptable neutralizing immune response, so your post-omicron immunity is likely poorly represented.
Thank you for your posts. I send people to them all the time. One literally said, "She is pure gold."
I'm working as a volunteer on a COVID-related issue: getting Evusheld to the immunocompromised. Gov't bought 875,000 doses 6 months ago and only 1/8 of it has been distributed to patients. Problem seems to be that too few patients and doctors know of its usefulness. I hope you'll put in a little mention of Evusheld and what it does in a blog post now and then, to raise awareness.
Our site is at rrelyea.github.io/evusheld/
It's a free public service -- no registration, no charge, just go there and get the information you need.
"If you’re high risk, please consider getting a prescription of Paxlovid" Just a couple of questions: 1) How is "high risk" defined in this case? I'm 74 but have no other risk factors, wear an N95 mask in all public indoor situations, and have had both boosters. Am I considered "high risk"? 2) Is Paxlovid something I should have on hand just in case I get infected, or should I ask my PC doc for it if and only if I test positive?
We're off on a two week European trip in two weeks, so I'm wondering whether or not I should just have some on hand. Thanks.
Stephen, I wonder if part of the answer is how easy it would be for you to get Paxlovid in Europe (and, possibly, whether US insurance would pay for it). I may one day face the same question (being 74 and no other risks); am I protected 'enough' being double-boosted? Also I wonder if it prevents immune systems from fighting the virus on a future encounter (given already being vaccinated). How much more do we need to know to inform decisions like you pose?
Thanks, that answers my questions. That wasn't clear in the original context. I was aware that Paxlovid wasn't OTC but I didn't know whether it could be prescribed for prophylactic use.
Very disheartening that vaxed and boosted folks are on the wrong side of the odds - even still being careful. Great fodder for those who avoid vaccines. Thanks as always for the update.
We seem to be near the point where we can compare clinically-serious illness among vaccinated (in various subsets) vs. unvaccinated, even adjusted for gender and age. But are we able to say by how much being vaccinated reduces the risk of getting or transmitting Covid (understanding there are nuances e.g. wearing a good mask etc.)? I recently attended a conference (virtually, tho have been double-boosted) where in-person attendees were required to either submit proof of vaccination or "a test within 72 hours." How much evidence is there that either one of these actions would decrease the probability of transmitting the virus? (I estimate <25% were masked based on session videos)
Thanks for this, KB. The paper in medRxiv looked only at neutralizing antibodies (as you point out) - but I'm having trouble understanding how that translates into their conclusion that BA.4 and BA.5 'have potential to result in a new infection wave," **IF** people have been previously infected and/or vaccinated and their T-cells are working? The medRxiv website says 125 media posts have already been made on this paper!! For example, should we expect that long-ago infection or vaccination with measles or polio is still protecting us? Would we have neutralizing antibodies if tested? Or is our lingering protection (if any) down to our T-cell memory? I also have questions about the statistical test used in their paper but should probably discuss that elsewhere...
Thanks, KB! I'll dive into the Cell paper. Also an audio interview in this week's New Engl Journal on immunity markers. We need more clarity and/or better communication on the clinical meaning of neutralizing antibodies because people (and popular journalists) are reading papers like this, implying that vax, boosting +/- Omicron breakthrough is at best modest at protecting against BA.4/5 based on neutralizing antibodies (https://www.science.org/doi/10.1126/sciimmunol.abq2427). Meanwhile, I hope the smallpox vaccine I got 70 years ago is holding out.
EXCELLENT Post! Question: early on, the statistics were that males did much worse with COVID-19 infections than women in most age groups. What are the current stats for women vs. men for things like: [1] current stats overall? [2] age-adjusted overall? [3] those protected by full vaccination? [4[ Etc.. THANK YOU!
I don't think we're out of the woods yet and undoubtedly with this large a reservoir of infected people and animals, the virus will continue to select for more transmittable variants, despite our countermeasures. To repost an earlier remark, and thinking numerically about it. It seems to me that the pandemic, far from being over, rang the planet like a bell setting up disequilibria everywhere and it will take a while, maybe three to five years for things to settle down to a new normal (which won't be the old one - it's gone) even if the virus were to disappear today. In effect, the planet itself has long covid.
Do you think that things like gun violence, civil unrest, monkeypox, supply chain disruptions, inflation, etc could be "long covid writ large"?
Precisely. I see turbulence introduced across the board in human systems and it will likely take a long time for them to damp out (even assuming we don't experience more waves).
Have you heard of the musical duo Flanders & Swann? One of their songs is "The Gasman Cometh", about a hapless homeowner trying to get their gas tap fixed.
T'was on the Monday morning, the gas man came to call,
The gas tap wouldn't turn, I wasn't getting gas at all;
He tore out all the skirting boards to try and find the main,
And I had to call a carpenter to put them back again!
Oh, it all makes work for the working man to do...
T'was on the Tuesday morning, the carpenter came round:
He hammered and he chiseled and he said "Look what I've found!
Your joists are full of dry rot, but I'll put them all to rights!"
Then he nailed right through a cable and out went all the lights!
Oh, it all makes work for the working man to do...
T'was on a Wednesday morning, the electrician came:
He called me 'Mr. Sanderson', which isn't quite the name;
He couldn't reach the fuse box without standing on the bin,
And his foot went through a window, so I called the glazier in!
Oh, it all makes work for the working man to do...
T'was on the Thursday morning, the glazier came along
With his blowtorch, and his putty, and his merry glazier's song;
He put another pane in, it took no time at all
But I had to get a painter in to come and paint the wall!
Oh, it all makes work for the working man to do...
T'was on a Friday morning the painter made a start,
With undercoats and overcoats he painted every part,
Every nook and every cranny; but I found when he was gone
He'd painted over the gas tap and I couldn't turn it on!
Oh, it all makes work for the working man to do...
On Saturday and Sunday they do no work at all;
So it was on the Monday morning that the gas man came to call!
I feel like that's kind of the situation we're in now. Every intervention, while necessary, puts a strain on something else, in ways we don't fully understand.
Well put. One problem we might be facing is that highly articulated systems get increasingly unstable unless there is substantial "give" built into the system (think of expansion seams on Bridges). Our systems are perhaps lacking in the give necessary to function well in environments of imperfect information, lack of redundancy, lack of capacity and so forth. We're definitely learning and will do better next time I imagine.
Yeah, I think one way of putting more give into the system would have been to have more graceful, discreet versions of various mandates. For instance, a good infrastructure for promoting masks on, say, airplanes would be to run "open tabs" like at bars. The airline would keep your credit card on file, and if they catch you in breach of their mask policy they'd just run your card.
Thanks for another informative post! A few comments:
1) Given that daily cases are vastly undercounted, by at least one order of magnitude, what is the value in continued publication of the surge chart? Individuals who use it to assess risk will quickly be misled, increasing the likelihood of poor decisions and infection.
2) Given that it’s possible to adjust the daily case chart based on sampling so that it more closely reflects reality, why doesn’t the CDC do this? They could footnote their assumptions for full disclosure. Instead, they continue to publish a chart that makes it appear they have the pandemic under control, which they do not. This chart is worse than worthless, it leads to risky behavior and unnecessary transmission. Either fix the chart for accuracy or retire it.
3). The NYT recently published an article stating that once symptoms begin, vaccinated individuals who use home tests are more likely to test false-negative for several days. I know several vaccinated and boosted people who become symptomatic, test negative at home, wait a day or two, and test negative at home again. They do everything right, assume they don’t have Covid, and go back to work or school, right at the moment they are becoming most infectious. So while it’s great that vaccines and boosters are preventing severe illness and death, are they inadvertently increasing transmission? While this early-false-negative phenomena has always been true for home tests, its even more pronounced with omicron, especially for vaccinated/boosted individuals where the vaccines "mask" a true positive in the early days. Here’s the NYT article (click “show more” to see full article): https://www.nytimes.com/explain/2022/coronavirus-questions#at-home-covid-test-negative
4) Are vaccines and boosters really preventing hospitalization and death today? Or did previous surges take out the “low hanging fruit” (most vulnerable) and those who are left are healthier and/or have natural immunity?
1. Look at the rate of change in the curves. Case rates increasing are still a good indicator.
2. CDC is an almost pure academic institution. Their choices of metrics are based on how their internal conventions.
3. There's explicit CDC guidance on how long to isolate following symptoms and/or a positive test of any sort. 3-6 days after symptom onset, especially in vaccinated persons, the amount of detectable antigen is decreased to near background levels. If the persons in question consider the guidance and follow it, when they're out of isolation, they are virus free. If they retest on a positive and break isolation early, there remains a chance they're still infectous. I can't access the NYT article (paywall) but the background study they are likely talking about did not reach the conclusion you present.
4. There's likely some truth to the concept that the population was culled significantly, but there are case control studies that support the concept that hospitalization, significant disease and increased mortality are reduced in vaccinated individuals.
Thank you for your thoughtful reply.
#1 people look at absolute level, not some fancy “rate of change” concept from high school calculus. The chart makes today’s transmission level (and risk) look low. Based on anecdotal evidence, today’s surge is probably just as bad as January 22’s peak, if not worse
#2 no disrespect, but who cares? Know your audience. If the chart is public, the public is your audience. That is the lens through which the chart’s value must be judged. What information is being communicated to the public? Is it correct? What actions will the public take in response?
#3 the large print instructions on the home test say something like “test today, test 24 hours later.” When someone is freaked out about Covid, they don’t stop to read the fine print or visit CDC website. They don’t quarantine 5-7 days before testing. Your “isolate 3-6 days” is nowhere on the CDC website, nor did I find language where the CDC breaks out isolation and testing guidance based on vaccination status (although quarantine guidance is broken out based on vaccination status). Nor did I see the phrase “virus free” (try saying that to someone suffering from long Covid). As a side note, the CDC website funnels readers based on whether they seek to “isolate” or “quarantine” when in fact most Americans use these terms interchangeably or are unaware of the nuances as to how these terms differ. Why not use simple and direct language like “exposed?” and “got Covid?” (Again: know you’re audience)
Since you are unable to access the NYT article (paywall), here is a direct quote: "The harder your immune system is working to tamp down the virus, particularly an immune system supercharged by vaccine antibodies, the more likely you are to get an early negative result on a rapid test, even if you're infected."
#4 the “death chart” lines for vaccinated and unvaccinated are nearly touching *today*. Unless the chart is greatly enlarged and underlying data points are provided, when one line appears to nearly touch the other, from an intuitive sense, it cannot be 17x. Also, as vaccines wane and natural immunity increases, the distinction between “vaccinated” and “unvaccinated” becomes blurred
I'm unable to reply right now: Too busy. However, I will follow up ASAP. One thought: I spent over 18 months avoiding news media stories on COVID. Too many are snippets taken out of context, where the context is important. The NYT quote is interesting but incomplete immunologically.
Thank you for keeping up with all this. Your posts help me to protect myself, despite mask-fatigue, and I pass the info on. I am double-boosted and intend to continue with further immunization when they become available. Dying on a ventilator does not attract.
Can you address what is known about flying now that there's no mask mandate? Any new empirical evidence? My spouse just got COVID after air travel, and that was his primary exposure. He and his colleague both took the same plane on Thursday (that sat on a runway for a while), and they haven't seen each other since. They came down with it today (4 days later) on antigen test. They were wearing KN95s. Are people with N95s actually avoiding it? I have to go to a conference in 3 weeks. If I actually avoid COVID from my spouse, I'm wondering if a N95 will be good enough or if I'm just as likely to get COVID.
In your paragraph “battle of Omicron” you noted BA.4 and BA.5 are gaining traction and they are reinfecting folks with Omicron. Hence their ability to spread quickly. The key question is what about the folks that aren’t getting infected with “old Omicron” and or the new variants. Are they not getting infected because they are not exposed, or their immune system does better? Since we cannot do random controlled tests and have to rely on vagaries of observed data, this is hard to answer, but it is a critical question despite structural data issues.
Second, it is important to be careful with models that are conjecture, for example the real COVID rate, even if they have fancy math. Additionally, what we don’t get when a COVID positive test is reported is “how sick” is the person.
“Massive” case surge needs to be defined. Massive in that your chances are 1 in 10 of going to a movie and leaving with COVID or 1 in 10,000. Risk needs to be made real.
I have been a big proponent of vaccines – love the MRNA technology. At this point my suggestion is to stop reporting separately on the unvaccinated. It isn’t helping and there are now data complications. We don’t separately report on folks with diabetes.
Yes I do a ton of analytics without borders and have been since 1977 when I joined IBM at age 23. This work have covered planning and scheduling to colon cancer.
Again, love your newsletters
This is getting more and more frustrating as time goes on. I'm 77 years old, vulnerable and really want to continue living for a "few" more years. There's no one to trust anymore. Except you. Protection from four shots is waning, leaving us more vulnerable to new variants. Guess I should continue to mask up, avoid crowds and strangers until our medical professionals figure this out.
As best I can determine, if I wear a nanofiber respirator mask at all times when I am indoors that has claimed protection of 98%, and if I fit it properly so that I can not feel any leaks around the edges even when I puff a breath, then I should be completely safe against the covid virus, regardless of how many people near me are emitting the virus into the indoor air, regardless of my personal risk level, and regardless of the fact that I am indeed vaccinated and twice boosted. Am I missing something? Why don’t 98% masks suffice?
Probably the answer depends on stuff like: What does "98%" mean (how does whatever they do to certify the 98% translate into questions like "What is the probability of me inhaling 'some' SARS-Cov-2 if I'm sitting next to a person on a plane who has asymptomatic Covid and not wearing a mask for 4 hours?" Which is dependent on more stuff like how many particles the person is transmitting, whether we're talking, etc. And then what's the probability of me becoming infected based on all those considerations? And what does "98" mean? That over some standardized time period 2 out of 100 viral particles get through? Because if it's "with each breath," that's different than "over one hour."
Case rates among the boosted are also increasing bc the population of people who are most likely to be boosted -- the elderly and people with weakened immune systems -- are also more vulnerable to infection due to age/health. I'd imagine we're going to approach a leveling off soon where we see more hospitalizations/deaths are among the boosted than the unvaxxed, as a function of the a priori higher risk presented by age and co-morbidities.
I don't think I can agree completely with your premise. Yes, the older population has higher vaccination rates but I think what we're seeing, as @Katelyn has said, is a shift in who's getting NAAT tests vs who's relying solely on home tests and luck (not consulting a doc when the at-home rapid-antigen tests are positive). The testing system in this country was broken initially, and the termination of free tests and mass-testing sites has not worked in favor of assembling reliable statistics.
Thanks for filling in those holes and the preprint reference. I'm literally too busy these days to keep up with all the new literature!
I'm about to get back to my 4 hr/night schedule from training... We're also renovating a house and I'm one of the workers. Over 250k reviewed and prereview papers so far that I can access. Skimming the surface is an understatement.
Once again - thank you for all your data driven, fact based work! I’m still wondering about the more difficult to answer question around longer term immunity (B and T cells). I know it’s not easy to measure (like antibody levels) but I really do wonder why there hasn’t been more data that shows long term durability.
Several studies have addressed cellular immunity and its persistence. At least one such study suggested B- and T-cell training did persist even if the actual B-cell initiation on exposure was much more tied to initiating training, where those classes of lymphocytes were trained and able to recognize, then subsequently start producing an active immune response in a matter of hours to days.
Question - is there a point on the case curve where most of the spread is in households vs the community? Is this something anyone has tried to model?
Wonderfully thoughtful post, Katelyn - thank you! Denominators question: In the Y-axes labeled incidence/100,000 population - is that the whole population (of whatever region)? If so, where can we find charts where the denominators are various statuses of prior infection, vaccinated, not vaccinated, etc.? Some preprint papers sort of get at this question. Does getting that information depend on rigorous analysis of large datasets of people's entire continuum of healthcare? For example, in health insurance claims or large healthcare systems in which members access essentially all of their healthcare? Setting up standardized queries takes a lot of work by people who deeply understand the nuances of using real-world (that is, non-clinical trial) evidence, but once set up could be run periodically for updates. Or is this already happening?
One of the biggest problems we've faced in this pandemic has been the scarcity of reliable data. We've never comprehensively tested the population either completely, or statistically. PCR testing has varied from lab to lab. Reliable antibody testing has only recently evolved, and most available antibody tests even today can differentiate between nucleocapsid and spike/receptor binding domain antibodies.
And then there's reporting. Some states have stopped reporting to CDC at all, others have decided what they will and won't report. Others have decided to report numbers in the categories CDC's requested but use different criteria than the recommended ones to report a positive case, thus skewing their result reporting. There are documented cases where death determinations have been changed to not reflect COVID cases if any other potential cause of death could rationally be listed.
I've been working a little bit with real-world data but have had to back off recently due to some family business. I hope to get back to some of the coding needed to do the work on a predictive system that assimilates historical data and and produces a slightly more robust tool to predict vulnerability. Finding a good and complete data source is a real problem.
As someone who has Covid at Easter I am wondering if when people get reinfected after omicron, do we know if their course of illness is similar to what they had before? I “just” had a cold though I tested positive for 2 weeks. Would another bout likely be similar?
There's clinical consensus that generally, the significant variants at play now are producing markedly less severe symptoms in most people, especially in those who were vaccinated, or were vaccinated either before or after vaccination. The majority of cases I've seen or heard about over the last several months reported the equivalent of a bad "head cold". If you tested positive for 2 weeks, what tests were you using? Generally, after testing positive with a home kit, there's no point in testing again 'til about a week after symptom onset.
I've seen at least one study noting that omicron failed to produce an acceptable neutralizing immune response, so your post-omicron immunity is likely poorly represented.