The COVID-19 Discussion Thread (OTB)

[Swift]

Our lockdown - 1 week severe and then 1 week mild - has now been removed as we have had no new cases. We are almost back to what they were on before the case 'escaped' quarantine. The main difference is that whenever you enter any premises, other than a private house, you must either scan your phone over a QR code or sign a register. This includes picking up takeaway. The records are maintained to enable tracing any possible contacts and are erased after 14 days.

Australia seems to have done a remarkable job controlling this virus.

[ozduck]

Australia seems to have done a remarkable job controlling this virus.

Taiwan seems to have dealt with it best of all but Australia certainly has kept it under control with only the outbreak in Victoria causing major problems before being overcome. All the countries that have done well, like NZ also, seem to have relied on strict quarantine measures and strong lockdowns when necessary.

[Jens]

Taiwan seems to have dealt with it best of all but Australia certainly has kept it under control with only the outbreak in Victoria causing major problems before being overcome. All the countries that have done well, like NZ also, seem to have relied on strict quarantine measures and strong lockdowns when necessary.

It's sort of interesting, I was just looking at the cumulative statistics for the Western Pacific region.

In terms of cumulative cases (by 100,000 population), the top countries in the region are Singapore, Malaysia, Vietnam, Japan, Korea, and Australia, in that order.

But for cumulative deaths, the order is: Philippines, Japan, Australia, Korea/Malaysia.

Japan has many more cases than Australia (327 versus 113), but the number of deaths is less than double (5 versus 3.5). I guess maybe more elderly people were infected in Australia. I think that is part of the reason why Singapore has the highest cases/population in the region but only 0.5 deaths/100,000, that there were many young migrant workers who were infected.

[Jens]

Oh, and by the way, the data I used is from here: https://www.who.int/docs/default-sou...&download=true

[ozduck]

It's sort of interesting, I was just looking at the cumulative statistics for the Western Pacific region.

In terms of cumulative cases (by 100,000 population), the top countries in the region are Singapore, Malaysia, Vietnam, Japan, Korea, and Australia, in that order.

But for cumulative deaths, the order is: Philippines, Japan, Australia, Korea/Malaysia.

Japan has many more cases than Australia (327 versus 113), but the number of deaths is less than double (5 versus 3.5). I guess maybe more elderly people were infected in Australia. I think that is part of the reason why Singapore has the highest cases/population in the region but only 0.5 deaths/100,000, that there were many young migrant workers who were infected.

You are right, the higher death rate was because some Nursing Homes were hit badly in the outbreak that occurred in the State of Victoria. Three quarters of all deaths were in those homes. The second highest rate of deaths was among Cruise passengers and crew.

Unfortunately Victoria is the outlier state. With roughly 1/4 (6.2 Million) of Australia's population (25.5 Million) they have had 90% of all deaths - 820 out of 909 Australia wide. My state with 2.7 Million has had 9 deaths - and they were all Cruise ship passengers and crew who became infected on their overseas voyage. We have had no deaths due to Covid since May last year. In, fact there have been only about 5 deaths in States other than Victoria since last May.

[Gillianren]

A friend of mine has been diagnosed. She has been traveling, which was of course extremely ill-advised, but I didn't say anything, because we're not that close and I already had one huge fight with her recently. So I'm sorry she's got it, but . . . .

[schlaugh]

My wife and I received our second Moderna dose on Saturday. I had fever and chills for a few hours that night and it was done. But my wife...oh dear. She was down and out all day on Sunday; aches, bouncing temps, chills, and a headache that would not give up. By Monday - all good!'

Coincidentally, xkcd published a cartoon on Monday which drew an analogy about how the vaccine works and comparing that process to destroying the Empire's Death Star.

https://xkcd.com/2425/

[grant hutchison]

Pulled across from another thread:

The USA has now passed half a million Covid-19 deaths. Damn.

Something quite peculiar is going on with the US death rate at present. The US and UK both experienced big spikes in case numbers peaking at the start of January. The UK deaths then peaked about three weeks later, as one might expect from the nature of the disease, at which point we had more than half again the per capita death rate of the USA. Today, the UK is pretty much back to matching the US, on a seven-day rolling average of the per capita rate. This is because the UK death rate has declined along pretty much the same trajectory as the case rate did, three weeks previously--whereas the US seven-day average death rate has lolloped along on an almost horizontal line, just below 1:100,000, since the second week in January.
This link should take you to the comparative death graphs; this link to the cases. (Both from the Financial Times, an open-access page which I believe is accessible from outside the UK.) In terms of the natural history of Covid-19 the reason for this isn't at all clear to me, and I haven't yet turned up a convincing explanation. It would appear the USA is smearing out its Covid deaths over a much longer period.

Grant Hutchison

[SeanF]

This link should take you to the comparative death graphs; this link to the cases. (Both from the Financial Times, an open-access page which I believe is accessible from outside the UK.) In terms of the natural history of Covid-19 the reason for this isn't at all clear to me, and I haven't yet turned up a convincing explanation. It would appear the USA is smearing out its Covid deaths over a much longer period.

Grant Hutchison

The death rate for the US on your link doesn't match this link, at least over the recent month or so. I wonder what the difference is...?

[grant hutchison]

The death rate for the US on your link doesn't match this link, at least over the recent month or so. I wonder what the difference is...?

Ooooh, that's interesting.
The FT data come from the COVID Tracking Project. The data in your link are from Johns Hopkins. (Although there seems to be an option to switch data source to the COVID Tracking Project, I couldn't wrestle the interface into submission.)
I haven't examined the "death" criteria in detail from the two sources, but my first impression is that the difference is because of the way a "Covid death" is defined. In the UK we parse the data into people who died within 28 days of a positive Covid test, and people who died with Covid listed on their death certificates--these obviously overlap to a large extent, but not completely. And there's a delay in the arrival of the second category of death data. Our daily reported figures come from matching Covid test data to registered deaths, which happens as soon as the death is recorded, so generally lag the deaths by just a few days; the "Covid on the death certificate" data have a further lag because of the way these data are entered and validated, and they generally produces a significant increment in the earlier reported death rate.
I'd guess in the USA this would all be further complicated by variation in death registration regulations from state to state, but that's just an assumption on my part.

Grant Hutchison

[Gillianren]

No, you're almost certainly right. There is little standardization of anything by state in this country, even when that would make more sense.

[SeanF]

Ooooh, that's interesting.
The FT data come from the COVID Tracking Project. The data in your link are from Johns Hopkins. (Although there seems to be an option to switch data source to the COVID Tracking Project, I couldn't wrestle the interface into submission.)
I haven't examined the "death" criteria in detail from the two sources, but my first impression is that the difference is because of the way a "Covid death" is defined. In the UK we parse the data into people who died within 28 days of a positive Covid test, and people who died with Covid listed on their death certificates--these obviously overlap to a large extent, but not completely. And there's a delay in the arrival of the second category of death data. Our daily reported figures come from matching Covid test data to registered deaths, which happens as soon as the death is recorded, so generally lag the deaths by just a few days; the "Covid on the death certificate" data have a further lag because of the way these data are entered and validated, and they generally produces a significant increment in the earlier reported death rate.
I'd guess in the USA this would all be further complicated by variation in death registration regulations from state to state, but that's just an assumption on my part.

Grant Hutchison

That's all true, but (except for some difference between the two source datasets) I don't think it can explain what we're seeing.

Delays in reporting should just delay the curve, not flatten it. (I'm making the educated guess here that neither of our sites retroactively change published data. If deaths that happened last week get reported today, they go into the graph today. Not a week ago.) And, presumably, the delays in reporting didn't just start a month ago.

Differences in reporting would cause differences in the curve, but those differences (again, presumably) didn't suddenly come into effect a month ago, either.

The fact that the curves more-or-less match up until mid-January means something has to have changed in mid-January, right?

[schlaugh]

Ooooh, that's interesting.
The FT data come from the COVID Tracking Project. The data in your link are from Johns Hopkins. (Although there seems to be an option to switch data source to the COVID Tracking Project, I couldn't wrestle the interface into submission.)
I haven't examined the "death" criteria in detail from the two sources, but my first impression is that the difference is because of the way a "Covid death" is defined. In the UK we parse the data into people who died within 28 days of a positive Covid test, and people who died with Covid listed on their death certificates--these obviously overlap to a large extent, but not completely. And there's a delay in the arrival of the second category of death data. Our daily reported figures come from matching Covid test data to registered deaths, which happens as soon as the death is recorded, so generally lag the deaths by just a few days; the "Covid on the death certificate" data have a further lag because of the way these data are entered and validated, and they generally produces a significant increment in the earlier reported death rate.
I'd guess in the USA this would all be further complicated by variation in death registration regulations from state to state, but that's just an assumption on my part.

Grant Hutchison

My mother-in-law's Florida death certificate stated she died from complications of Parkinson's Disease. It also listed arteriosclerotic heart disease, chronic kidney disease. and COVID-19 as underlying conditions. I'd be curious to know if her death was tallied in the data as COVID-related but I doubt there's any way to find out.

[grant hutchison]

That's all true, but (except for some difference between the two source datasets) I don't think it can explain what we're seeing.

Delays in reporting should just delay the curve, not flatten it. (I'm making the educated guess here that neither of our sites retroactively change published data. If deaths that happened last week get reported today, they go into the graph today. Not a week ago.) And, presumably, the delays in reporting didn't just start a month ago.

Routinely, in the UK, data are added retrospectively. This is done transparently if you look at data from source; less so if you look at what secondary sources offer. For instance, the Scottish graphs I look at every day show cases by date of test, and deaths by date of death. Necessarily, that means the right end of the "cases" graph mutates for a few days (as results come in), and the "deaths" graph for a week or two (as people register deaths and the certificates are analysed and logged). So the daily figures (what gets reported at noon each day) never match the graphs.
And there are certainly occasions when there are sudden transitions in the behaviour of the graphs, because there are sudden transitions in the way data are recorded. For instance, we made a transition in the UK last year from logging only Covid-tested deaths to including deaths in which Covid was mentioned in the death certificate. Then we made another change, setting a time limit on how recent a positive Covid test had to be for a death to be deemed a "Covid death". (Last autumn it was impossible to survive Covid in the UK, because any positive test, ever in your life, would have triggered a "Covid death" flag on your eventual demise in a sky-diving accident.) And we made another similar revision to our "Covid patients in hospital" criteria, which gave the appearance of the hospitals pretty much emptying overnight.

In the UK, the compilation of the stats is the responsibility of a very small number of organizations who feed into a single set of official figures. So we see step transitions and more subtle changes in trends, but we don't see one organization graphing one set of data while another produces graphs of a different set of data. I get the impression that may be different in the USA, and I'm suggesting that what we're seeing is two organizations using different approaches to their data, and that (perhaps) one of them made a transition that the other did not, back in January. I know how I would go about checking that hypothesis in the UK, because I know how things work and who is responsible for what, but I have no inkling of how you'd investigate that in the USA.

Grant Hutchison

[CaliA]

Oh, and by the way, the data I used is from here: https://www.who.int/docs/default-source/coronaviruse/situation-reports/employee monitoring software/20210216_weekly_epi_update_27.pdf?sfvrsn=94da8979_ 5&download=true

I've already been ill with Covid, thank goodness it came in a mild form.
At some point, after the holidays in our country, the cases seemed less. All of a sudden the numbers doubled again, and it is not yet clear why.
However the numbers around the world are far from encouraging....

[grant hutchison]

My mother-in-law's Florida death certificate stated she died from complications of Parkinson's Disease. It also listed arteriosclerotic heart disease, chronic kidney disease. and COVID-19 as underlying conditions. I'd be curious to know if her death was tallied in the data as COVID-related but I doubt there's any way to find out.

Not for the specific case of your mother-in-law, but there will be a protocol, somewhere, describing the procedure by which Florida records deaths as being Covid-related. But I can't even guess if such a thing would be a matter of public record in the USA, let alone where you might go to read it.

Grant Hutchison

[Trebuchet]

Pulled across from another thread:

Something quite peculiar is going on with the US death rate at present. The US and UK both experienced big spikes in case numbers peaking at the start of January. The UK deaths then peaked about three weeks later, as one might expect from the nature of the disease, at which point we had more than half again the per capita death rate of the USA. Today, the UK is pretty much back to matching the US, on a seven-day rolling average of the per capita rate. This is because the UK death rate has declined along pretty much the same trajectory as the case rate did, three weeks previously--whereas the US seven-day average death rate has lolloped along on an almost horizontal line, just below 1:100,000, since the second week in January.
This link should take you to the comparative death graphs; this link to the cases. (Both from the Financial Times, an open-access page which I believe is accessible from outside the UK.) In terms of the natural history of Covid-19 the reason for this isn't at all clear to me, and I haven't yet turned up a convincing explanation. It would appear the USA is smearing out its Covid deaths over a much longer period.

Grant Hutchison

I've been logging Worldometer data for the USA since the beginning, and calculating 7-day averages for deaths and new cases. The 7-day average for deaths has dropped from 3129 two weeks ago to 2091 yesterday. I haven't looked at your links yet but perhaps that's related.

[schlaugh]

Not for the specific case of your mother-in-law, but there will be a protocol, somewhere, describing the procedure by which Florida records deaths as being Covid-related. But I can't even guess if such a thing would be a matter of public record in the USA, let alone where you might go to read it.

Grant Hutchison

Yeah my brief digging on this topic uncovered a political minefield. So that's that.

[grant hutchison]

I've been logging Worldometer data for the USA since the beginning, and calculating 7-day averages for deaths and new cases. The 7-day average for deaths has dropped from 3129 two weeks ago to 2091 yesterday. I haven't looked at your links yet but perhaps that's related.

Well, Worldometer claims its data are used by Johns Hopkins, which are the same data used in SeanF's link.
Whereas the COVID Tracking Project is producing different data, used in the graphs I linked to.

It'll be interesting to see how the Financial Times (using the COVID Tracking Project data) reconcil their US data when that project shuts down at the start of March. FT are very number-savvy, and I'm pretty sure will produce some sort of explanatory note relating to their choice of data sources.

Meanwhile, reading through the COVID Tracking Project's announcement of imminent shut-down seems to confirm what I suspected about the variation in US data sources. The implication seems to be that there was a lack of standardized data coming from federal public health authorities, and that situation is now being rectified.

From Worldometer:

For the COVID-19 data, we collect data from official reports, directly from Government's communication channels or indirectly, through local media sources when deemed reliable. We provide the source of each data update in the "Latest Updates" (News) section. Timely updates are made possible thanks to the participation of users around the world and to the dedication of a team of analysts and researchers who validate data from an ever-growing list of over 5,000 sources.

From the COVID Tracking Project:

Every day for almost a year, hundreds of COVID Tracking Project contributors from all walks of life have compiled, published, and interpreted vitally important COVID-19 data as a service to their fellow Americans.

I'm sure one could tease out the cause of the recent increasing discrepancy in death data for the USA, if only by getting in contact with the two organizations. But this certainly highlights why epidemiologists emphasize the difficulty of comparing death data between countries, given that two organizations analysing the same data for the same country are producing divergent results.

Grant Hutchison

[Trebuchet]

My mother-in-law's Florida death certificate stated she died from complications of Parkinson's Disease. It also listed arteriosclerotic heart disease, chronic kidney disease. and COVID-19 as underlying conditions. I'd be curious to know if her death was tallied in the data as COVID-related but I doubt there's any way to find out.

Yeah my brief digging on this topic uncovered a political minefield. So that's that.

Sent you a PM on that!

[Jens]

I've already been ill with Covid, thank goodness it came in a mild form.
At some point, after the holidays in our country, the cases seemed less. All of a sudden the numbers doubled again, and it is not yet clear why.
However the numbers around the world are far from encouraging....

Welcome to the Cosmoquest forum.

I'm just wondering, what country are you from? Also, it seems that in general around the world, the numbers are dropping (though they are still quite high, of course).

[Extravoice]

COVID case numbers have dropped dramatically in this area lately. I find it hard to believe there was a dramatic shift in human behavior, and the vaccination rate here is still quite low. I don’t understand it, but I’ll take it.

[Jens]

COVID case numbers have dropped dramatically in this area lately. I find it hard to believe there was a dramatic shift in human behavior, and the vaccination rate here is still quite low. I don’t understand it, but I’ll take it.

I don’t claim to understand it well either. But the same thing happens with influenza, so perhaps small changes in R0 end up leading to dramatic changes.

[Swift]

COVID case numbers have dropped dramatically in this area lately. I find it hard to believe there was a dramatic shift in human behavior, and the vaccination rate here is still quite low. I don’t understand it, but I’ll take it.

I think a lot of the huge uptick was from bad human behavior during the Thanksgiving and Christmas holidays (traveling and getting together in groups) and we are only now coming out of that huge spreading event. But even with numbers coming down, we are still just about as high as we were last spring, early in the pandemic, when we thought "this is really bad".

[Extravoice]

...But even with numbers coming down, we are still just about as high as we were last spring, early in the pandemic, when we thought "this is really bad".

Yeah, I’m pretty sure if you dig through the “pandemic” thread, I posted an estimate of 160,000 total dead in the US. We thought that was really bad.

[Jim]

My wife had her first shot last Friday, I get my second today, and I was just able to schedule our daughter the kindergarten teacher's first for this Saturday.

[Extravoice]

I’m scheduled for my first shot on March 7. After weeks of “no appointments available”, I was shocked when some appeared this morning.
That was followed by several cycles of “Choose an appointment - Enter your info - Sorry, that appointment is no longer available” before I got a confirmation.

Unfortunately, Mrs. Extravoice has to wait for the next group before she can try.

[Gillianren]

A friend looks about to lose both his parents; his nieces and nephews have also lost their mom.

[Swift]

A friend looks about to lose both his parents; his nieces and nephews have also lost their mom.

With over 28 million cases and over half a million deaths in the US, we have to have reached the point where pretty much everyone knows people who have had Covid-19 and someone who has died.

I'm probably up to about a dozen people I know who have had it, and one who has died (a friend's father-in-law).

[Roger E. Moore]

Got my COVID-19 vaccination today, about 2 hours ago. Had to drive 100+ miles round trip, but got it done. Have to do it again in 2-3 weeks. Feel fine so far.