Update as of early 3/28/20.
Total U.S. cases -- 104,256 (up by over 18,000 from 3/27)
Active cases -- 100,027
Deaths -- 1,704
CFR -- 1.6%
The case curves are still pointing steeply upward (which most expected), signifying that we remain in the escalation phase. Also, clusters are developing in some cities that appear to have the potential to become major hot spots (e.g., New Orleans). Hopefully, mitigation measures will begin to have an effect on the curves soon. Parenthetically, consider what the numbers would be today had we not started the aggressive mitigation campaign about 10-11 days ago. That has bought us some time, while the country is ramping production/delivery of PPE and ventilators, constructing/converting facilities, which is crucial to keeping the healthcare system from becoming overwhelmed. But, in some places we are approaching capacity of skilled health care workers (as their ranks are depleted by having to quarantine following exposure/becoming symptomatic (I know one doctor and her husband, both symptomatic)-- which is becoming a problem). Many places have already mobilized medical students, residents, doctors/nurses in unrelated specialties, retirees, etc., so that source of supply is already being tapped. Also, the CFR now is up slightly, despite the fact that much more testing has been done. Hopefully that will reverse and decline as even more testing is done.
There has been good news, though.
The lead proponent of the Imperial College model that drove the models being used in the UK and USA concluded that the number of unreported cases that involved few or no symptoms has likely been much higher than was assumed in the original model. That is, he now believes there have been many more cases that have not been discovered/reported (and not resulted in deaths).
If he is right, it would be good in two ways. First, our Case Fatality Rate (CFR) would go down from the originally estimated 2-3%, perhaps below 1% or even lower. Second, it would mean that many tens or hundreds of thousands of Americans have unknowingly already been infected/recovered and likely developed antibody immunity, bringing us meaningfully closer to "herd immunity" but without having strained the health care system and without having taken more lives in the process. And, if we are closer to herd immunity for that reason, we will reach it much earlier than has been estimated, and with less loss of life and less use of healthcare resources and burden. We should hope he is correct.
We won't know if this change is predictive of what will happen, however, until we can do mass testing to find out who/how many have been infected already.
Also, excellent preliminary results continue to be reported regarding off-label use of hydroxychloroquine to lessen severity of disease and length of stay. Hopefully in a few weeks we will have results of a controlled, randomized, blinded clinical trial that will confirm these preliminary reports and observational studies. We should appreciate the patients willing to enroll, as those who receive placebo will be short-changed if the drug actually works (as many doctors are reporting they believe is happening with their other patients). That means, however, that from a statistical perspective, there will be an interim examination of the data and, if the Data Monitoring Committee finds statistical significance of efficacy before scheduled trial completion, they would recommend an early halt of the trial. It would be unethical, under such circumstances, to continue giving placebo to trial enrollees who are randomized to the control arm of the trial. My uninformed guess is that we might hear of an interim look in 2-3 weeks.
Other therapeutic candidates (mostly, if not exclusively, several not previously approved for other indications, unlike hydroxychloroquine is) are also now in testing (e.g., remdesivir, et al). They, however, may have to undergo more testing than hydroxychloroquine in order to study safety more thoroughly.
Another bit of good news is the rapid development of testing capability, particularly point of care testing, which is about to ramp mass production. That would help in multiple ways, from earlier diagnosis and quarantine (reducing spread), to reduced use of PPE and other resources (on patients who, when tested, will be immediately known to be negative, despite symptoms), to facilitating study of spread concentrations and patterns (informing, hopefully, when and where easing of restrictions can safely occur).
Keep your fingers crossed. If you are religious, say a prayer for our courageous health care providers and their families. Our debt to them will be more than we could ever repay.
COVID-19 -- March 28 update
Re: COVID-19 -- March 28 update
COVID-19 and the oil price war will impact the "gassers". I guess all major supply/demand disruptions have "winners and losers".
Read: file:///C:/Users/Dan/Downloads/Nat%20Gas%20-%20Research%20Note%20-%2003-27-20.pdf
If you know anyone with a small business, tell them to spend time looking at what the $2.2 Trillion bailout does for them. It might save their business.
Keep in mind that on the Johns Hopkins University website here:
https://gisanddata.maps.arcgis.com/apps ... 7b48e9ecf6
The red dots on the map are not in proportion to the population. If they were, the red dots would be so small you could not see them.
Population of the U.S. = ~350,000,000 if you count all of the undocumented people living here.
So, only 0.03% of the people living in the U.S. are confirmed cases; 1 in every 3,330 people have been confirmed. The companies that make the test kits are going to make a lot of money.
There may be 10X or 100X more people that have the virus, but most people with healthy immune systems either have no symptoms or the symptoms are very mild and they get over it in about two weeks thinking that they just had a cold. If the actual number of contagious people is 100X more than the number of Confirmed Cases, is continuing the lockdown worth the cost?
I like the concept of "herd immunity". Maybe the Asian Herd is already there. I just hope the U.S. Herd gets there soon because life without baseball will be tough.
The numbers on the Johns Hopkins University website are only as good as what each country is reporting, but it sure looks like the number of people who have recovered is grossly understated. For example, the Diamond Princess Cruise Ship is a separate line item. Of the 712 confirmed cases from the ship over a month ago, only 10 have died. However, the JH website only shows that 597 have recovered and that number has not changed this week. If the virus runs its course in two weeks, what's happened to the other 105? Also, the U.S. and several other countries are sending home mild cases and it looks like they are not following up to see if they have recovered. This is important because the key number is the number of active cases.
Read: file:///C:/Users/Dan/Downloads/Nat%20Gas%20-%20Research%20Note%20-%2003-27-20.pdf
If you know anyone with a small business, tell them to spend time looking at what the $2.2 Trillion bailout does for them. It might save their business.
Keep in mind that on the Johns Hopkins University website here:
https://gisanddata.maps.arcgis.com/apps ... 7b48e9ecf6
The red dots on the map are not in proportion to the population. If they were, the red dots would be so small you could not see them.
Population of the U.S. = ~350,000,000 if you count all of the undocumented people living here.
So, only 0.03% of the people living in the U.S. are confirmed cases; 1 in every 3,330 people have been confirmed. The companies that make the test kits are going to make a lot of money.
There may be 10X or 100X more people that have the virus, but most people with healthy immune systems either have no symptoms or the symptoms are very mild and they get over it in about two weeks thinking that they just had a cold. If the actual number of contagious people is 100X more than the number of Confirmed Cases, is continuing the lockdown worth the cost?
I like the concept of "herd immunity". Maybe the Asian Herd is already there. I just hope the U.S. Herd gets there soon because life without baseball will be tough.
The numbers on the Johns Hopkins University website are only as good as what each country is reporting, but it sure looks like the number of people who have recovered is grossly understated. For example, the Diamond Princess Cruise Ship is a separate line item. Of the 712 confirmed cases from the ship over a month ago, only 10 have died. However, the JH website only shows that 597 have recovered and that number has not changed this week. If the virus runs its course in two weeks, what's happened to the other 105? Also, the U.S. and several other countries are sending home mild cases and it looks like they are not following up to see if they have recovered. This is important because the key number is the number of active cases.
Last edited by dan_s on Sat Mar 28, 2020 1:52 pm, edited 1 time in total.
Dan Steffens
Energy Prospectus Group
Energy Prospectus Group
Re: COVID-19 -- March 28 update
The "gasser" url "cannot be found." file:///C:/Users/Dan/Downloads/Nat%20Gas%20-%20Research%20Note%20-%2003-27-20.pdf
Re: COVID-19 -- March 28 update
Sorry that the link did not work. I have posted it to the EPG website as "Good News for Gassers". When you log on, just scroll down and you will see on the home page.
Dan Steffens
Energy Prospectus Group
Energy Prospectus Group
Re: COVID-19 -- March 28 update
There may be 10X or 100X more people that have the virus, but most people with healthy immune systems either have no symptoms or the symptoms are very mild and they get over it in about two weeks thinking that they just had a cold. If the actual number of contagious people is 100X more than the number of Confirmed Cases, is continuing the lockdown worth the cost?
The change in the model was not how many people "have the virus" or "are contagious." Instead it was in the number of people who would have had, and would no longer have, the virus and thus would no longer be contagious, and therefore also would theoretically (even this is still to be established) be immune for some undetermined period (until the antibodies wear off). But this is only an assumption used in a model, not something that has been proved. As Dr. Birx has said, "there are models, and then there are facts." All models, by definition theoretical, are used to try to predict ultimate numbers. That's all.
There is no evidence whatever that the total number of cases in the U.S. is 10x or 100x the total number thus far reported. Only mass testing will show how many people have been infected.
The relevance of all of this goes to the issue of whether the healthcare system is going to be overwhelmed, significantly damaging our health care workforce and necessitating rationing of healthcare. No public official wants to decide the criteria for who will be denied life-saving medical care. Read the executive summary of this 2007 pandemic planning document. http://www.cidrap.umn.edu/sites/default ... idance.pdf
Very close to 100% of doctors and public health experts in the U.S. (and around the world) agree that without "flattening the curve," which (absent an effective vaccine) can be accomplished only by the mitigation efforts currently imposed, our healthcare system would rapidly become overwhelmed by numbers of serious cases that cannot be handled, for which we are not equipped, and which would damage our health care system substantially, and still would require mass rationing of care. To answer your question, if you were deciding policy, what would be your eligibility criteria for care? Which patients (by age, socio-economic group, condition, whatever) would you decree are to be denied care? And where would you go to replace the doctors, nurses, respiratory therapists, etc., who would die because we lacked adequate PPE but who we expected to care for the sick anyway?
If the virus runs its course in two weeks, It doesn't (see next para). Even if it did, the more important question is WHEN all the cases occur. Do they (at least the first wave) occur in a 30-45 day period (overwhelming our healthcare system), or are we able to spread them out over a longer period, so as to give us time to produce PPE, develop therapies, create more beds, produce more ventilators, and save more lives (including those of healthcare workers)?
As for the active case count, I suspect that many countries use different criteria for "recovered," which directly influences the number of "active" cases they report. In the U.S., the CDC defines recovery from COVID-19 as an absence of fever, with no use of fever-reducing medication, for three full days; improvement in other symptoms, such as coughing and shortness of breath; a period of seven full days since symptoms first appeared. Two negative swab tests on consecutive days are considered as the all-clear – meaning self-isolation can end and a patient can theoretically begin having contact with others, including at work. Recovery time varies generally with severity of disease, and widely, with one recent report of a case where the virus was found in a patient's respiratory tract for a total of 37 days. One source says, "using available preliminary data, the median time from onset to clinical recovery for mild cases is approximately 2 weeks and is 3-6 weeks for patients with severe or critical disease."
Notice, also, that "recovery" does not mean "total" recovery. Many COVID-19 patients show chest CT evidence of "ground glass nodules" in their lungs (it's almost a hallmark of the disease, though it is a finding seen in some other respiratory diseases, too). Many of these patients will never "fully recover," but instead will have life-long, residual scarring in their lung tissue that will compromise lung function.
Three days ago, you guessed that the US active case count will peak at 100,000-120,000 by April 15.
As I type this, the US is approaching 115,000 active cases (vs. about 100,000 yesterday), 2666 (vs. 2463 yesterday) of which are "serious or critical." But the numbers are rising rapidly. It is extremely unlikely that the peak will be 120,000, the upper end of your prediction. Also, the time course of the disease differs by region, when mitigation efforts began, degree of observance of mitigation efforts, etc., so that peaks will occur in different times in different places. So, the health care systems in those places will be stressed at different times and to different degrees.
Predicting when the peak will occur, and where, with any reliability, is dependent upon too many unknowns. Testing on an even wider scale, which is supposedly coming very soon, will apparently inform analysis that will enable more informed policy making re mitigation measures.
One final reminder. As Dr. Birx recently reiterated, a person can test negative one day and positive the next. It happens.
The change in the model was not how many people "have the virus" or "are contagious." Instead it was in the number of people who would have had, and would no longer have, the virus and thus would no longer be contagious, and therefore also would theoretically (even this is still to be established) be immune for some undetermined period (until the antibodies wear off). But this is only an assumption used in a model, not something that has been proved. As Dr. Birx has said, "there are models, and then there are facts." All models, by definition theoretical, are used to try to predict ultimate numbers. That's all.
There is no evidence whatever that the total number of cases in the U.S. is 10x or 100x the total number thus far reported. Only mass testing will show how many people have been infected.
The relevance of all of this goes to the issue of whether the healthcare system is going to be overwhelmed, significantly damaging our health care workforce and necessitating rationing of healthcare. No public official wants to decide the criteria for who will be denied life-saving medical care. Read the executive summary of this 2007 pandemic planning document. http://www.cidrap.umn.edu/sites/default ... idance.pdf
Very close to 100% of doctors and public health experts in the U.S. (and around the world) agree that without "flattening the curve," which (absent an effective vaccine) can be accomplished only by the mitigation efforts currently imposed, our healthcare system would rapidly become overwhelmed by numbers of serious cases that cannot be handled, for which we are not equipped, and which would damage our health care system substantially, and still would require mass rationing of care. To answer your question, if you were deciding policy, what would be your eligibility criteria for care? Which patients (by age, socio-economic group, condition, whatever) would you decree are to be denied care? And where would you go to replace the doctors, nurses, respiratory therapists, etc., who would die because we lacked adequate PPE but who we expected to care for the sick anyway?
If the virus runs its course in two weeks, It doesn't (see next para). Even if it did, the more important question is WHEN all the cases occur. Do they (at least the first wave) occur in a 30-45 day period (overwhelming our healthcare system), or are we able to spread them out over a longer period, so as to give us time to produce PPE, develop therapies, create more beds, produce more ventilators, and save more lives (including those of healthcare workers)?
As for the active case count, I suspect that many countries use different criteria for "recovered," which directly influences the number of "active" cases they report. In the U.S., the CDC defines recovery from COVID-19 as an absence of fever, with no use of fever-reducing medication, for three full days; improvement in other symptoms, such as coughing and shortness of breath; a period of seven full days since symptoms first appeared. Two negative swab tests on consecutive days are considered as the all-clear – meaning self-isolation can end and a patient can theoretically begin having contact with others, including at work. Recovery time varies generally with severity of disease, and widely, with one recent report of a case where the virus was found in a patient's respiratory tract for a total of 37 days. One source says, "using available preliminary data, the median time from onset to clinical recovery for mild cases is approximately 2 weeks and is 3-6 weeks for patients with severe or critical disease."
Notice, also, that "recovery" does not mean "total" recovery. Many COVID-19 patients show chest CT evidence of "ground glass nodules" in their lungs (it's almost a hallmark of the disease, though it is a finding seen in some other respiratory diseases, too). Many of these patients will never "fully recover," but instead will have life-long, residual scarring in their lung tissue that will compromise lung function.
Three days ago, you guessed that the US active case count will peak at 100,000-120,000 by April 15.
As I type this, the US is approaching 115,000 active cases (vs. about 100,000 yesterday), 2666 (vs. 2463 yesterday) of which are "serious or critical." But the numbers are rising rapidly. It is extremely unlikely that the peak will be 120,000, the upper end of your prediction. Also, the time course of the disease differs by region, when mitigation efforts began, degree of observance of mitigation efforts, etc., so that peaks will occur in different times in different places. So, the health care systems in those places will be stressed at different times and to different degrees.
Predicting when the peak will occur, and where, with any reliability, is dependent upon too many unknowns. Testing on an even wider scale, which is supposedly coming very soon, will apparently inform analysis that will enable more informed policy making re mitigation measures.
One final reminder. As Dr. Birx recently reiterated, a person can test negative one day and positive the next. It happens.