KidDoc said:

---

nortex97 said:

---

I'm just a guy forced to work from home now, but there is a lot of skepticism about their tests; you can trust them and their figures if you'd like to do so. No worries.

When you rush to approve/produce a rapid test kit in weeks you get things like 30% sensitivity. Same thing with the Chinese kits (or much worse).

https://www.reddit.com/r/CoronavirusDownunder/comments/fiedl5/south_korean_covid19_test_kits_not_accurate_falls/

https://fortunascorner.com/2020/03/25/80-of-coronavirus-test-kits-gifted-to-czechs-by-china-faulty/

---

These are excellent points that most people are not thinking about. The tests we use are not FDA approved which means we have no idea what the sensitivity is. Most of the ID docs I have read think there is likely 70% sensitivity which means 30% of the negatives are actually positives. If/when we get widespread testing if this data point holds true we may get a significant spread due to people testing negatives and spreading it anyway.

---

Actually, it means 30% of positive individuals test negative. That is completely different from 30% of the negatives being positive. Given in some areas (outside of NYC) we are seeing 90% of the tests come back negative, it means the difference between 15% of being positive and close to 40% being positive.

BlackGoldAg2011 said:

---

nortex97 said:

---

I'm just a guy forced to work from home now, but there is a lot of skepticism about their tests; you can trust them and their figures if you'd like to do so. No worries.

When you rush to approve/produce a rapid test kit in weeks you get things like 30% sensitivity. Same thing with the Chinese kits (or much worse).

https://www.reddit.com/r/CoronavirusDownunder/comments/fiedl5/south_korean_covid19_test_kits_not_accurate_falls/

https://fortunascorner.com/2020/03/25/80-of-coronavirus-test-kits-gifted-to-czechs-by-china-faulty/

---

sorry if i came across as argumentative, i don't "want to trust them". I want the best data available. If there was reason to doubt the data i wanted to know, and this discussion led me to this study (https://www.mdmag.com/medical-news/comparing-rt-pcr-and-chest-ct-for-diagnosing-covid19)
which gives some solid data to suggest the negative tests could have a rather alarming false-negative rate. I don't know enough about the korean testing to say one way or the other if this applies to them, but it is actual data to suggest the possibility that they are alarmingly under-identifying. I asked the way i did because too often people try to discount data simply because it doesn't agree with their initial position. i'm not trying to accuse you of that, but i also don't know a thing about any of you other than an online persona.

that being said i think at this point we can all agree that we don't have a solid handle on "the denominator" and it could be a pretty wide range of numbers.

but even with that, i just wanted to point out that we have a lot of data pointing to the possibility that this could be much worse than flu unchecked. we may decide it's not when we have more data in and the benefit of hindsight, but from a forecasting perspective right now, the data available (good or bad) points towards this warrants a stronger reaction from us than the seasonal flu

---

Ok I gotta run after this for a bit but I don't take offense at internet posts ever, let alone yours, no worries. I respect the term 'possibility' in your post here certainly. I also think that possibility is looking less and less likely. The public discourse is focused on the news out of places like NYC/Italy/New Orleans. The likely possibility to me is that...it's actually not much worse, if at all, than the flu. It may, quite likely, have a different impact on population groups (like type A blood type, or the elderly) it's first one or two cycles through the population, and it may wind up not popping up much again since it mutates pretty slowly and we'll have a vaccine in a year or two.

My opinion is just that it might actually not be as bad as many current 'doomsayers' are emotionally attached to believing. Some have a religious belief in how terrible it is.

The problem is we are not shutting everything down. we shut down bars, gyms, theaters and a few other non essential things. Everything else and every other job is still being done around the country. People are driving where they need to go. shopping in public. unless you go door to door and weld doors shut you don't have a quarantine. as long as people are interacting, you are just maybe slowing the spread but since you don't know who has it, it can still spread until someone shows up with symptoms. How long have they been infecting people. how many things have they come in contact with?

BiochemAg97 said:

---

PJYoung said:

---

nortex97 said:

---

South Korea is largely packed into one huge urban sprawl a la NYC. I don't think they represent anything near an accurate data point vs. the whole US. I also don't really trust the tests they used.

---

You think they had more positives than they showed even tho they controlled it better than anybody else in the world? As in their clusters were shut down and isolated and the infection rate went to almost nothing.

I find it hard to believe they had this large population of positives that they didn't find but I guess anything is possible.

---

The other explanation is they had bad tests, had 3x the cases, the CFR is significantly lower that what people are saying because the asymptomatic cases are 50%+ of the total cases, and the infection burned through the population already, resulting in the decline they are seeing.


Hard to tell, and wanting tests that were decent is part of the delay in getting testing in the US. I guess the question is which is better, testing soon with a lot of false negatives so people think they don't have it and continue to spread it, or delaying testing until you have a good test and have a bunch of people not knowing if they have it and continuing to spread it.

---

If you're asking if I choose between what happened in South Korea and what is happening here I would choose SK every time.

nortex97 said:

---

BlackGoldAg2011 said:

---

nortex97 said:

---

I'm just a guy forced to work from home now, but there is a lot of skepticism about their tests; you can trust them and their figures if you'd like to do so. No worries.

When you rush to approve/produce a rapid test kit in weeks you get things like 30% sensitivity. Same thing with the Chinese kits (or much worse).

https://www.reddit.com/r/CoronavirusDownunder/comments/fiedl5/south_korean_covid19_test_kits_not_accurate_falls/

https://fortunascorner.com/2020/03/25/80-of-coronavirus-test-kits-gifted-to-czechs-by-china-faulty/

---

sorry if i came across as argumentative, i don't "want to trust them". I want the best data available. If there was reason to doubt the data i wanted to know, and this discussion led me to this study (https://www.mdmag.com/medical-news/comparing-rt-pcr-and-chest-ct-for-diagnosing-covid19)
which gives some solid data to suggest the negative tests could have a rather alarming false-negative rate. I don't know enough about the korean testing to say one way or the other if this applies to them, but it is actual data to suggest the possibility that they are alarmingly under-identifying. I asked the way i did because too often people try to discount data simply because it doesn't agree with their initial position. i'm not trying to accuse you of that, but i also don't know a thing about any of you other than an online persona.

that being said i think at this point we can all agree that we don't have a solid handle on "the denominator" and it could be a pretty wide range of numbers.

but even with that, i just wanted to point out that we have a lot of data pointing to the possibility that this could be much worse than flu unchecked. we may decide it's not when we have more data in and the benefit of hindsight, but from a forecasting perspective right now, the data available (good or bad) points towards this warrants a stronger reaction from us than the seasonal flu

---

My opinion is just that it might actually not be as bad as many current 'doomsayers' are emotionally attached to believing. Some have a religious belief in how terrible it is.

---

I question why China went to the extreme measures they did for something that isn't that serious then. I don't know how you rectify that. Was it just a massive overreaction by them? Locking down 150 million or whatever it was for well over 2 months? Building those hospitals just for this virus?

BiochemAg97 said:

---

nortex97 said:

---

I don't think they have as many folks testing now, and besides, when you have your whole population wearing masks transmission will be lower anyway. But no, I don't think they are lying about their numbers a la China. It's just that their numbers aren't really as good/meaningful/accurate at all as many think they are.

Most of us have had a kid/ourselves test negative for strep many times, but doc smells the breath and diagnoses as positive etc., you get the pills/shot. There's no such 'sanity check' for these kits. Some are under the misnomer that all diagnostic tests work really well, and that even CT scans are fool proof etc, and that's just...not how it works.

Until we get to genomic testing, a lot of this is just three blind mice with calculators and keyboards is my contention, and I think the truth is that the spread is vastly larger than folks think it is (but innocuous to almost everyone).

---

Genomic testing? You know most if not all the tests in the US are testing for the presence of the viral genome.


What will be more interesting is a good reliable test for antibody to the virus that can tell who had it and who didn't.

---

I agree, once you have the antibody, you should be free to proceed with your life.

Testing negative today, even if the test was 100% accurate, is no guarantee that you didn't pick it up on the way home from the test, at the grocery store the next day or somewhere else. That negative test was only a valid data point at the time the test was taken.

A positive test, assuming 100% accuracy again, on the other hand means that the clock has started and after 2-3 weeks of isolation, you should be good to go if all goes well.

dpeterson said:

---

BiochemAg97 said:

---

nortex97 said:

---

I don't think they have as many folks testing now, and besides, when you have your whole population wearing masks transmission will be lower anyway. But no, I don't think they are lying about their numbers a la China. It's just that their numbers aren't really as good/meaningful/accurate at all as many think they are.

Most of us have had a kid/ourselves test negative for strep many times, but doc smells the breath and diagnoses as positive etc., you get the pills/shot. There's no such 'sanity check' for these kits. Some are under the misnomer that all diagnostic tests work really well, and that even CT scans are fool proof etc, and that's just...not how it works.

Until we get to genomic testing, a lot of this is just three blind mice with calculators and keyboards is my contention, and I think the truth is that the spread is vastly larger than folks think it is (but innocuous to almost everyone).

---

Genomic testing? You know most if not all the tests in the US are testing for the presence of the viral genome.


What will be more interesting is a good reliable test for antibody to the virus that can tell who had it and who didn't.

---

I agree, once you have the antibody, you should be free to proceed with your life.

Testing negative today, even if the test was 100% accurate, is no guarantee that you didn't pick it up on the way home from the test, at the grocery store the next day or somewhere else. That negative test was only a valid data point at the time the test was taken.

A positive test, assuming 100% accuracy again, on the other hand means that the clock has started and after 2-3 weeks of isolation, you should be good to go if all goes well.

---

Why do you think you are free to proceed once you have the antibody? This is not how routine Coronovirus works it tends to fade after a few months. Comparing this to SARS & MERS those seem to have decent titers that start to drop after 12 months which would be awesome if COVID does the same but we just do not know.

The apparent return of China to pre-pandemic work is reassuring though but we are all waiting and worried about the next wave once immunity begins to decrease.

pocketrockets06 said:

---

Your alternative explanation is really contraindicated by the data. If the cases were dying out because CFR is low, most are asymptomatic and it burned through the population, you would expect the deaths (even at the low CFR) to be more uniformly distributed across the country. But they're not - the deaths are concentrated in the places where the measured cases are concentrated. This should be our best indicator that the denominator is not appreciably larger than the tested cases (at least for South Korea, not true for the US).

---

I don't think that is necessarily true. If the worst cases were more likely to test positive (higher viral loads or testing criteria or other factors), then you would still have deaths concentrating with positive cases. Also, I haven't looked at the distribution of positive tests/cfr compared to population/population density. I assume there is correlation there which would also lead to a concentration of deaths.

Not saying I agree with the alternate theory, but you can get to the other side of the peak either by stopping the spread or getting to herd immunity. Although with increased virulence (R0), you need higher immune % for effective herd immunity.

KidDoc said:

---

dpeterson said:

---

BiochemAg97 said:

---

nortex97 said:

---

I don't think they have as many folks testing now, and besides, when you have your whole population wearing masks transmission will be lower anyway. But no, I don't think they are lying about their numbers a la China. It's just that their numbers aren't really as good/meaningful/accurate at all as many think they are.

Most of us have had a kid/ourselves test negative for strep many times, but doc smells the breath and diagnoses as positive etc., you get the pills/shot. There's no such 'sanity check' for these kits. Some are under the misnomer that all diagnostic tests work really well, and that even CT scans are fool proof etc, and that's just...not how it works.

Until we get to genomic testing, a lot of this is just three blind mice with calculators and keyboards is my contention, and I think the truth is that the spread is vastly larger than folks think it is (but innocuous to almost everyone).

---

Genomic testing? You know most if not all the tests in the US are testing for the presence of the viral genome.


What will be more interesting is a good reliable test for antibody to the virus that can tell who had it and who didn't.

---

I agree, once you have the antibody, you should be free to proceed with your life.

Testing negative today, even if the test was 100% accurate, is no guarantee that you didn't pick it up on the way home from the test, at the grocery store the next day or somewhere else. That negative test was only a valid data point at the time the test was taken.

A positive test, assuming 100% accuracy again, on the other hand means that the clock has started and after 2-3 weeks of isolation, you should be good to go if all goes well.

---

Why do you think you are free to proceed once you have the antibody? This is not how routine Coronovirus works it tends to fade after a few months. Comparing this to SARS & MERS those seem to have decent titers that start to drop after 12 months which would be awesome if COVID does the same but we just do not know.

The apparent return of China to pre-pandemic work is reassuring though but we are all waiting and worried about the next wave once immunity begins to decrease.

---

My point was that a negative test is all but useless if they are given to the population over time. A negative result is valid for the point in time it was administered and useless after that.

If having the antibodies still might mean weeks before it's safe to return to public life then I'm more than happy to defer to the experts. I still believe it to be the best test and the safest return to public life until a vaccine is available, even if it takes a couple of weeks.

KidDoc said:

---

dpeterson said:

---

BiochemAg97 said:

---

nortex97 said:

---

I don't think they have as many folks testing now, and besides, when you have your whole population wearing masks transmission will be lower anyway. But no, I don't think they are lying about their numbers a la China. It's just that their numbers aren't really as good/meaningful/accurate at all as many think they are.

Most of us have had a kid/ourselves test negative for strep many times, but doc smells the breath and diagnoses as positive etc., you get the pills/shot. There's no such 'sanity check' for these kits. Some are under the misnomer that all diagnostic tests work really well, and that even CT scans are fool proof etc, and that's just...not how it works.

Until we get to genomic testing, a lot of this is just three blind mice with calculators and keyboards is my contention, and I think the truth is that the spread is vastly larger than folks think it is (but innocuous to almost everyone).

---

Genomic testing? You know most if not all the tests in the US are testing for the presence of the viral genome.


What will be more interesting is a good reliable test for antibody to the virus that can tell who had it and who didn't.

---

I agree, once you have the antibody, you should be free to proceed with your life.

Testing negative today, even if the test was 100% accurate, is no guarantee that you didn't pick it up on the way home from the test, at the grocery store the next day or somewhere else. That negative test was only a valid data point at the time the test was taken.

A positive test, assuming 100% accuracy again, on the other hand means that the clock has started and after 2-3 weeks of isolation, you should be good to go if all goes well.

---

Why do you think you are free to proceed once you have the antibody? This is not how routine Coronovirus works it tends to fade after a few months. Comparing this to SARS & MERS those seem to have decent titers that start to drop after 12 months which would be awesome if COVID does the same but we just do not know.

The apparent return of China to pre-pandemic work is reassuring though but we are all waiting and worried about the next wave once immunity begins to decrease.

---

Since this one is most closely related to SARS, seems like that would be a more valid comparison.

BlackGoldAg2011 said:

---

Since so many people seem so insistent on wanting to compare this disease to flu but none seem willing to actually do the work to see that comparison and would rather just throw virtual stones in spite of all the knowledge our TexAgs medical community keeps providing, I decided to just go ahead and give them the numbers they so think they desire. All data is based on 2018-2019 US flu season, data sourced here

First up is total cases by week. Total lab confirmed flu cases reported to the CDC in the provided data fell well short of their 16.5 MM case estimate for the season. So for these plots I took the lab confirmed cases for each week and multiplied them by 13.7 to match the total case load for the year.


(updated 3/27/2020)
This is probably the least fair look because it inflates the Flu lab confirmed cases but leaves the COVID cases as just lab confirmed. Even this though shows the need to "flatten the curve". Looking to the log scale plot, if left exponential, in 3 weeks COVID would overtake total estimated flu cases and in a month would surpass the entire estimated 2018-2019 flu case load.

Since the first look didn't quite seem like a fair comparison, especially since we are seeing our resident medical professionals confirm what we know, that a lot of cases are not getting lab tests, this second look is just looking at total lab confirmed cases submitted to the CDC from all labs


(updated 3/27/2020)
This look is better but still not ideal because this includes all the local/small commercial labs that to date have been unable to run tests on COVID cases. I think so far everyone agrees we are vastly under testing COVID cases to understand our full case load. But even with this issue, COVID will overtake total lab confirmed flu cases in a week. Also worth noting, since flu is not novel and is technically always around, the flu curve starts at 285k cases in the first week, COVID being novel and having an actual start date starts at 15.

Third look is only flu tests run by CDC labs. This should be a closer comparison in the terms of percentage of cases tested since so far CDC has been the only one really able to test COVID cases. It may not be completely fair to flu though because there may be less of a push to have CDC test flu cases since there is less need to do so. That may not be true but is possible and must be acknowledged

(updated 3/27/2020)
No log scale needed for this one... COVID will pass total CDC flu labs fr the whole 2018-2019 season by the end of this week.
3/27/2020 update notes: So it has been brought to my attention I may bee looking at this plot wrong. What I am calling "CDC labs" the fluview site actually is labeling "all public health labs so this might actually include more than just CDC. But we also started including clinical labs in the COVID numbers here in the last week. so due to the discrepancy i'm just going to cut the data off on this graph at the end of week 5 (3/20/2020) and stop updating it. But even at the end of week 5 the point is still clear

ok, ok, i hear you, but this is just case load and tells us nothing about true impact, what about deaths? we all know due to under testing cases our CFR is super inflated right? flu kills so many more people after all. well glad you asked, i've got those numbers for you too. Now I couldn't find deaths by week so i took the total composite CFR from 2018-2019 flu of 0.1% (technically 1 in 1009 but i rounded) and multiplied each weeks case load by that CFR to estimate deaths by week. the actual weeks may be off slightly but the total numbers should be right. Since there are solid arguments for why our COVID CFR may be artificially inflated due to under testing we will be looking at total deaths in the US

First up, the estimated deaths calculated from the total estimated flu cases.



(updated 3/27/2020)
lots of reasons this may not be a fair comparison, but even with its slow start COVID will pass total estimated flu deaths within a week.
3/27/2020 update notes: one point worth noting here, as as of today there still one full day missing from COVID's "week 6" data so by the end of today, COVID will be brushing up against the total estimate flu deaths


One last look. What I am seeing from our resident TexAgs medical staff is that we are seeing deaths in "unconfirmed" COVID cases due to the lag/shortage in testing, so for a final look lets look at total COVID death count vs the estimated weekly flu deaths calculated from the total lab confirmed flu cases rather than the total estimate:

(updated 3/27/2020)
from this data set, COVID will pass the flu death count estimate by the end of saturday 3/28/2020
3/27/2020 update notes: with one full day of data still missing from week 6, confirmed COVID deaths have already surpassed my calculated estimate of "confirmed flu deaths" for the entire 2018-2019 flu season.

i'll keep these graphs updated on a regular basis in this first post, and will indicate if i add any observations or change any conclusions.

---

Updated my plots and added a few notes. the most stand out new note to me is the very last one. The last few plots to me are the most relevant to the discussion at this point anyways. there are a lot of valid questions being posed as to what our real case load is, and therefore what the real CFR should be. Until this wraps up and we have the benefit of some hindsight we will likely not know the answer to that. What we do have a decent handle on however is the total death count, and that is not dependent on an accurate count of total cases including all of those asymptomatic ones out there. and by that measure, COVID has already passed the "lab confirmed" flu deaths, and over the next week will pass the total flu death count despite flu having a head start, starting in week 1 with 285 deaths. and with our current track, it would take a radical "flattening of the curve" to not surpass total flu deaths for the whole flu season. our current rate of increase would put us at that point in the next 2-3 weeks.

That's exactly why I'm using last year's (2018-2019) flu data

pocketrockets06 said:

---

Your alternative explanation is really contraindicated by the data. If the cases were dying out because CFR is low, most are asymptomatic and it burned through the population, you would expect the deaths (even at the low CFR) to be more uniformly distributed across the country. But they're not - the deaths are concentrated in the places where the measured cases are concentrated. This should be our best indicator that the denominator is not appreciably larger than the tested cases (at least for South Korea, not true for the US).

---

I'm waiting for the CDC mortality data to come out for Weeks 10, 11, and 12. There was a huge spike in influenza-like-illness doctor visits beginning in Week 10 for people age 25+. I suspect there will be an obvious spike in pneumonia-related deaths over Weeks 10-12 in previous years, even in states with low numbers of known cases now. I also expect a noticeable bump in overall mortality numbers for those weeks, accounting for C19-related cardiac arrest, etc.

Question for someone better at statistics than I am. Regarding the mortality rate: everything I've read about the sensitivity of the tests being relatively low has pointed toward there being this large group of infected people who have tested negative and that the denominator is actually much larger than we think and based on the presumed actual demonator the mortality rate is actually x which is much lower than is reported. But they don't increase the numerator. Presumably, the sensitivity is low across the board and there are a number of people who have been hospitalized or died from covid who have tested negative as well who aren't being counted in the statistics. Is there a way to account for that?

littledude said:

---

Question for someone better at statistics than I am. Regarding the mortality rate: everything I've read about the sensitivity of the tests being relatively low has pointed toward there being this large group of infected people who have tested negative and that the denominator is actually much larger than we think and based on the presumed actual demonator the mortality rate is actually x which is much lower than is reported. But they don't increase the numerator. Presumably, the sensitivity is low across the board and there are a number of people who have been hospitalized or died from covid who have tested negative as well who aren't being counted in the statistics. Is there a way to account for that?

---

Once we know the sensitivity and specificity of each test you can estimate. Due to the urgent need the test bypassed FDA which is where those things are calculated.

Thought this was a good article, https://spectator.us/deadly-coronavirus-still-far-clear-covid-19/?fbclid=IwAR1DgOVHtAPlz0ClxP4-eV4MT1pgiHt7wa_j_PInykpkxexPA7V7h34_jAI

KidDoc said:

---

littledude said:

---

Question for someone better at statistics than I am. Regarding the mortality rate: everything I've read about the sensitivity of the tests being relatively low has pointed toward there being this large group of infected people who have tested negative and that the denominator is actually much larger than we think and based on the presumed actual demonator the mortality rate is actually x which is much lower than is reported. But they don't increase the numerator. Presumably, the sensitivity is low across the board and there are a number of people who have been hospitalized or died from covid who have tested negative as well who aren't being counted in the statistics. Is there a way to account for that?

---

Once we know the sensitivity and specificity of each test you can estimate. Due to the urgent need the test bypassed FDA which is where those things are calculated.

---

For what it's worth, I see a lot of 70% sensitivity and everyone is freaking out about it. As a reference, the upper bound for the sensitivity of rapid flu test everyone gets at the local clinic during flu season, with some data suggesting as low as 10% sensitivity. That is all included on the rapid test data sheet.

It is not clear to me if that was the stats from the crappy China tests and the flawed CDC tests which all seemed to have a 1 in 3 failure. Or if it was stats from the current batch of PCR tests in the US. There was some effort put in to make sure the PCR tests were going to be positive on all known sequences of SARS-CoV2 so it is possible those are better. Part of the problem in not being able to find the data is these are EUA tests and not approved tests so, like was mentioned, we don't have that information available.

BlackGoldAg2011 said:

---

Since so many people seem so insistent on wanting to compare this disease to flu but none seem willing to actually do the work to see that comparison and would rather just throw virtual stones in spite of all the knowledge our TexAgs medical community keeps providing, I decided to just go ahead and give them the numbers they so think they desire. All data is based on 2018-2019 US flu season, data sourced here

4/1/20 update notes: Iooking through the data i realized i had been using the wrong flu numbers for total case load and estimated death count. I was using number of cases that sought medical attention not symptomatic cases. This took the 2018-2019 flu totals from 16.5MM to 34.1MM cases and deaths from 16.5k to 34.1k. I have updated all of the graphs to reflect this info. Also, i adjusted the COVID curves in all of the graphs to use daily data expressed in fractional weeks to smooth the curves better rather than taking partial weeks and applying the data to the end of the week.

First up is total cases by week. Total lab confirmed flu cases reported to the CDC in the provided data fell well short of their 16.5 MM case estimate for the season. So for these plots I took the lab confirmed cases for each week and multiplied them by 13.7 to match the total case load for the year.


(updated 4/1/2020)
This is probably the least fair look because it inflates the Flu lab confirmed cases but leaves the COVID cases as just lab confirmed. Even this though shows the need to "flatten the curve". Looking to the log scale plot, if left exponential, in 3 weeks COVID would overtake total estimated flu cases and in a month would surpass the entire estimated 2018-2019 flu case load.


Since the first look didn't quite seem like a fair comparison, especially since we are seeing our resident medical professionals confirm what we know, that a lot of cases are not getting lab tests, this second look is just looking at total lab confirmed cases submitted to the CDC from all labs


(updated 4/1/2020)
This look is better but still not ideal because this includes all the local/small commercial labs that to date have been unable to run tests on COVID cases. I think so far everyone agrees we are vastly under testing COVID cases to understand our full case load. But even with this issue, COVID will overtake total lab confirmed flu cases in a week. Also worth noting, since flu is not novel and is technically always around, the flu curve starts at 285k cases in the first week, COVID being novel and having an actual start date starts at 15.
4/1/20 update notes: total lab confirmed cases have officially passed 2018-2019 lab confirmed flu cases at equivalent points in the season, and in another few weeks should have eclipsed the season total

Third look is only flu tests run by CDC labs. This should be a closer comparison in the terms of percentage of cases tested since so far CDC has been the only one really able to test COVID cases. It may not be completely fair to flu though because there may be less of a push to have CDC test flu cases since there is less need to do so. That may not be true but is possible and must be acknowledged

(updated 3/27/2020)
No log scale needed for this one... COVID will pass total CDC flu labs fr the whole 2018-2019 season by the end of this week.
3/27/2020 update notes: So it has been brought to my attention I may bee looking at this plot wrong. What I am calling "CDC labs" the fluview site actually is labeling "all public health labs so this might actually include more than just CDC. But we also started including clinical labs in the COVID numbers here in the last week. so due to the discrepancy i'm just going to cut the data off on this graph at the end of week 5 (3/20/2020) and stop updating it. But even at the end of week 5 the point is still clear

ok, ok, i hear you, but this is just case load and tells us nothing about true impact, what about deaths? we all know due to under testing cases our CFR is super inflated right? flu kills so many more people after all. well glad you asked, i've got those numbers for you too. Now I couldn't find deaths by week so i took the total composite CFR from 2018-2019 flu of 0.1% (technically 1 in 1009 but i rounded) and multiplied each weeks case load by that CFR to estimate deaths by week. the actual weeks may be off slightly but the total numbers should be right. Since there are solid arguments for why our COVID CFR may be artificially inflated due to under testing we will be looking at total deaths in the US

First up, the estimated deaths calculated from the total estimated flu cases.


(updated 4/1/2020)
lots of reasons this may not be a fair comparison, but even with its slow start COVID will pass total estimated flu deaths within a week.
3/27/2020 update notes: one point worth noting here, as as of today there still one full day missing from COVID's "week 6" data so by the end of today, COVID will be brushing up against the total estimate flu deaths
4/1/2020 update notes: at this point, basically no matter what we have done to the curve, confirmed COVID deaths will pass total estimated flu deaths at the equivalent point in the season by the end of this week despite Flu's massive head start.

One last look. What I am seeing from our resident TexAgs medical staff is that we are seeing deaths in "unconfirmed" COVID cases due to the lag/shortage in testing, so for a final look lets look at total COVID death count vs the estimated weekly flu deaths calculated from the total lab confirmed flu cases rather than the total estimate:

(updated 4/1/2020)
from this data set, COVID will pass the flu death count estimate by the end of saturday 3/28/2020
3/27/2020 update notes: with one full day of data still missing from week 6, confirmed COVID deaths have already surpassed my calculated estimate of "confirmed flu deaths" for the entire 2018-2019 flu season.
4/1/2020 update notes: I will stop updating this one going forward too as COVID has so surpassed the flu curve that this plot is no longer relevant.

4/1/2020 update notes: I decided to add another look to approximate what our actual real caseload is assuming several different CFRs since we all agree we don't actually know the "denominator". Here is that plot:

Methodology: To come up with these estimated case load curves, I assumed the death count trails the case count by 2 weeks, so I took the death total for each day, and use that and the stated CFR to estimate what the total case load was 14 days earlier. For the 2 most current weeks, where 2 week out death numbers are not available, I looked at what the recent trend of the ration of estimated cases to confirmed cases was doing and carried that trend out for 2 weeks and applied that multiplier to the confirmed case count to get the estimated case count. and my observations from this look are:

• First, 0.02% CFR is currently the absolute lower limit for possibilities, because this means that every single person in the USA currently has been infected by COVID-19. So not the lowest feasible CFR, but the lowest that is physically possible.
• If the CFR is the same as the flu (0.1%), we have already surpassed the worst flu year in the last decade by 44% and just breached the upper bound of the estimated H1N1 case load from the 2009 pandemic at only 6.5 weeks into the pandemic. Also, if not a singled person more gets COVID (unlikely), we will top out at 65k deaths around 2 weeks from now.

i'll keep these graphs updated on a regular basis in this first post, and will indicate if i add any observations or change any conclusions.

---

updated plots to reflect most recent data. also, added a plot to look at a couple of different CFR scenarios. and the short version of that look is that this COVID most likely fits one of two descriptions. it either has a similar mortality to flu but is VASTLY more contagious (like greater than 1/3-1/2 of the total pop will get it before then end of April contagious) or we are closer to actual case count than we think and this is significantly more deadly than flu, and somewhat more contagious.

Interesting data. Thanks for that. Looks pretty clear we probably aren't at .02% CFR.

Looking at the graph, it seems you are assuming the caseload follows the confirmed curve, just by a multiplier to adjust to the CFR. While I don't think we are at a point where most of the country has been infected and cleared, how would you think the graphs change if you assume some of the exponential growth in confirmed cases is due to rapid increase in testing rather than the growth being entirely due to increasing cases.

Say we started seeing a roll over in cases earlier, but the confirmed cases kept growing from the continued testing. The CFR might be lower without being at everyone in the country infected already.


Also, I would add it makes sense that we would have more infections that the flu because there is no vaccine and no immunity as there is with flu.

BiochemAg97 said:

---

Interesting data. Thanks for that. Looks pretty clear we probably aren't at .02% CFR.

Looking at the graph, it seems you are assuming the caseload follows the confirmed curve, just by a multiplier to adjust to the CFR. While I don't think we are at a point where most of the country has been infected and cleared, how would you think the graphs change if you assume some of the exponential growth in confirmed cases is due to rapid increase in testing rather than the growth being entirely due to increasing cases.

Say we started seeing a roll over in cases earlier, but the confirmed cases kept growing from the continued testing. The CFR might be lower without being at everyone in the country infected already.


Also, I would add it makes sense that we would have more infections that the flu because there is no vaccine and no immunity as there is with flu.

---

So to the part I bolded, not quite but I'll explain what I did a little better. For the estimate case load 14 days ago and earlier, the curve is not tied in any way to the confirmed cases data. For that calculation I took the confirmed/known death count and for each assumed CFR, assumed it was tied to the real case count 14 days earlier (which is why that calculation stops 14 days ago, I run out of death data). From that point, to carry the curve forward, I looked at how each calculated total case curve was trending as a ratio compared to confirmed cases over the last 2 weeks (in every case the ratio was trending down towards 1 which you would expect as testing catches up) and carried that ratio trend forward to calculate the most recent two weeks based of estimated total case load off of the confirmed cases. So the last 13 data points of each curve are tied to confirmed case counts but that is it. My hope here is to start establishing some bounds on what a realistic number might be for CFR based on what that CFR would require actual infraction numbers to be in order to be accurate. For instance that .02% is obviously low since no one believes that every person in the country has been infected at this point. Even a .1% seems like it will be too low when you look at where that would put the actual case count going forward in the next few weeks based on the current trend.

My understanding is one big difference between COVID and the flu is the medical staff are much more at risk of getting infected with this thing. They all have gotten their flu shots and have typically built up some pretty decent immunity to the average flu strain, but without any vaccine for COVID, the folks treating patients and keeping our hospitals running are more vulnerable and so there is attrition as this thing knocks some medical staff out. I saw Italy had reported about a week ago that 37 doctors had died from COVID complications, and that New York doctor seemed to mention having to manage staff losses, and also re-integrating staff who had recovered.

As of 6 days ago

CNN.com edition
Web results
51 Italian doctors who contracted coronavirus have died since the start of the pandemic

BlackGoldAg2011 said:

---

BiochemAg97 said:

---

Interesting data. Thanks for that. Looks pretty clear we probably aren't at .02% CFR.

Looking at the graph, it seems you are assuming the caseload follows the confirmed curve, just by a multiplier to adjust to the CFR. While I don't think we are at a point where most of the country has been infected and cleared, how would you think the graphs change if you assume some of the exponential growth in confirmed cases is due to rapid increase in testing rather than the growth being entirely due to increasing cases.

Say we started seeing a roll over in cases earlier, but the confirmed cases kept growing from the continued testing. The CFR might be lower without being at everyone in the country infected already.


Also, I would add it makes sense that we would have more infections that the flu because there is no vaccine and no immunity as there is with flu.

---

So to the part I bolded, not quite but I'll explain what I did a little better. For the estimate case load 14 days ago and earlier, the curve is not tied in any way to the confirmed cases data. For that calculation I took the confirmed/known death count and for each assumed CFR, assumed it was tied to the real case count 14 days earlier (which is why that calculation stops 14 days ago, I run out of death data). From that point, to carry the curve forward, I looked at how each calculated total case curve was trending as a ratio compared to confirmed cases over the last 2 weeks (in every case the ratio was trending down towards 1 which you would expect as testing catches up) and carried that ratio trend forward to calculate the most recent two weeks based of estimated total case load off of the confirmed cases. So the last 13 data points of each curve are tied to confirmed case counts but that is it. My hope here is to start establishing some bounds on what a realistic number might be for CFR based on what that CFR would require actual infraction numbers to be in order to be accurate. For instance that .02% is obviously low since no one believes that every person in the country has been infected at this point. Even a .1% seems like it will be too low when you look at where that would put the actual case count going forward in the next few weeks based on the current trend.

---

Thanks for the explanation.

updated. here is in my opinion the most interesting/significant update


my original methodology for calculating an estimated true case load for the most recent 2 weeks broke down with the most recent data, so i adjusted how i handled that and changed the methodology description below to match the new graphs



(updated 4/3/2020)
Methodology: To come up with these estimated true case load curves, I assumed the death count trails the true case count by 2 weeks, so I took the death total for each day, and use that and the assumed CFR to estimate what the total true case load was 14 days earlier. For the 2 most current weeks, where 2 week out death numbers are not available, I used forecasts for the death count over the next two weeks to calculate and estimated true case load. i indicated on each curve where the calculation starts being based on forecast rather than real data. since we are seeing death count bend over slightly and not lie on an exponential curve over the last 12 days, I used two possible curve fits for the death count forecast, a 2nd order polynomial and a power law curve. My personal opinion is the power law curve will ultimately be a better forecast due to its better match to the shape of the last 12 days worth of the real curve, but I went ahead and included the poly curve as it also is a decent fit and should represent the most optimistic (while being realistic) outlook on future death count numbers. I also added a CFR of 1.75% since that is what S. Korea is trending towards at nearly 1% of their population tested and 60% of their total cases having resolved already. and my observations from this look are:

• First, 0.022% and .0082% CFRs for power law and poly curves respectively iare currently the absolute lower limits for possibilities, because this means that every single person in the USA currently has been infected by COVID-19. So not the lowest feasible CFR, but the lowest that is physically possible.
• in all but the really deadly CFR scenarios, we have already surpassed last year's flu numbers in estimated true case counts 7 weeks into the season despite flu having a 600k case head start.
• I also added some population % lines at points that are relative to thresholds of infection percentage estimates from past pandemics.

death forecast curves for reference:

Have you read this? Essentially the same idea you're doing - assume a severity and fit it to the observed deaths to estimate total infected.

As you note, as deaths increase it becomes falsifiable by population limits.

I wouldn't use a poly fit, there's no possibility to have that shape (i.e., it won't reverse...). Log, power, or exponential will be better guides.

no i haven't read it yet, but just glancing through it, it looks very interesting and I will.

As for the poly fit, I know there is no physical way for it to fit the poly curve long term, but with so many variables unknown and changing (like our ever adjusting lock-down restrictions setting new artificial expansion boundaries) the final curve is most likely to fit different function shapes at different points along it with no curve being a good fit across the entire time. so in my view at least, it is possible that when looking at the time period of the next few weeks it could fit a poly curve better than the others. more importantly though, just eyeballing the fit, it looks to me to be a decent lower boundary for what death count could do in the next two weeks (with exponential being the upper boundary). that being said, i still contend the power law is more accurate but wanted to include the poly fit just for completeness and transparency.

Gotcha. Yeah right now trying best fit this is like trying to plot a car's speed in traffic.

Would you mind replotting the assumed severity spread with the minimum y axis at 1000?

updated