Seasonal effect?

Covid-19 discussion, bring your own statistics
User avatar
sTeamTraen
Dorkwood
Posts: 1211
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: Seasonal effect?

Post by sTeamTraen » Thu Nov 19, 2020 9:45 am

snoozeofreason wrote:
Sun Oct 25, 2020 4:55 pm
More Heneghan. This time explaining The ten worst Covid data failures in an article co-authored with Tom Jefferson in the Spectator, where Heneghan seems to be a regular presence.
Is there a single person with contrarian views on COVID-19 transmission (as opposed to how we should respond, where I think there is a genuine debate to be had, e.g. on schools) who isn't a thinly disguised libertarian?
snoozeofreason wrote:
Sun Oct 25, 2020 4:55 pm
The figure of 85,000 doesn't appear in the Radio Sweden article, and there's no indication of how Heneghan and Jefferson have arrived at it. But it is suspiciously close to the figure you would get if you took the 500,000 UK deaths mentioned in the first sentence (after adding the zero they missed out) and then just decide to divide by 6 because Sweden's population is about 1/6 of the UK. I am not sure it works like this. As the More or Less program has pointed out a few times, you would expect the progress of the virus in Sweden to behave more like that in Norway or Finland - both of which have vastly lower death rates - than the UK.
Sweden announced 96 new deaths yesterday and has been running at levels of new cases per 100k somewhere between the UK and Italy for a couple of weeks. In the first wave they kept their ICUs at relatively low occupancy levels by just giving many people over 80 morphine, but those numbers are also going up fast in some reasons. I know a number of Swedish scientists and they are in despair at the preventable car crash that is about to unfold. Tegnell seems to still be doubling down, which is sadly predictable from what we know of how people cling to beliefs.
Something something hammer something something nail

User avatar
sTeamTraen
Dorkwood
Posts: 1211
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: Seasonal effect?

Post by sTeamTraen » Thu Nov 19, 2020 11:26 am

sTeamTraen wrote:
Thu Nov 19, 2020 9:45 am
but those numbers are also going up fast in some reasons.
in some *regions*. #DYA
Something something hammer something something nail

User avatar
shpalman
After Pie
Posts: 2290
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: Seasonal effect?

Post by shpalman » Thu Nov 19, 2020 11:34 am

I was wondering if some hotter places would have had an extra summer wave as everyone stayed indoors with the air conditioning on, and indeed the US had that, but then they unlocked almost as soon as the first wave of cases stopped going up.
molto tricky

User avatar
sTeamTraen
Dorkwood
Posts: 1211
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: Seasonal effect?

Post by sTeamTraen » Thu Nov 19, 2020 1:49 pm

shpalman wrote:
Thu Nov 19, 2020 11:34 am
I was wondering if some hotter places would have had an extra summer wave as everyone stayed indoors with the air conditioning on, and indeed the US had that, but then they unlocked almost as soon as the first wave of cases stopped going up.
Realistically we are all looking at data from about 100 places where we can have even minimum confidence in the numbers (50 US states plus 50 other countries where the testing programme might, if we're lucky, be catching and reporting something like half of the infections and 80% of the deaths). When you add in the still-low overall prevalence in absolute terms and the variability in climate, culture, and health systems, I don't think you're going to end up with much more than just-so stories.
Something something hammer something something nail

Herainestold
Snowbonk
Posts: 446
Joined: Mon Nov 25, 2019 1:23 pm

Re: Seasonal effect?

Post by Herainestold » Thu Nov 19, 2020 3:53 pm

Woodchopper wrote:
Wed Nov 18, 2020 11:13 pm
Thread on seasonal effects
https://twitter.com/trvrb/status/132743 ... 02337?s=21
Some interesting thoughts..
thanks for posting

User avatar
snoozeofreason
Clardic Fug
Posts: 190
Joined: Fri Nov 15, 2019 1:22 pm

Re: Seasonal effect?

Post by snoozeofreason » Mon Nov 23, 2020 5:43 pm

I'm not sure whether we should carry on posting stuff about Heneghan in this thread, or create a separate thread about him. But he's got another article in the Spectator written with his pal Tom Jefferson, and prompted by a Danish RCT on face masks.
Heneghan and Jefferson wrote:In the end, there was no statistically significant difference between those who wore masks and those who did not when it came to being infected by Covid-19. 1.8 per cent of those wearing masks caught Covid, compared to 2.1 per cent of the control group. As a result, it seems that any effect masks have on preventing the spread of the disease in the community is small.
[my bolding]

It seems that a lot of people on social media have been pointing out that the bolded conclusion doesn't follow from the its premise, because the way masks prevent the spread of disease is mainly by preventing the wearer from infecting others, rather than by preventing the wearer from getting infected. The Spectator has been moved to add a rather tetchy correction to the bottom of the article (best read out in a Sybil Fawlty voice).
Spectator wrote:Due to the large number of people passing comment on the article on social media without reading it, we have updated the headline to emphasise that the study is about facemask wearers. Covid data can be found on our data hub: data.spectator.co.uk
As is generally the case with newspaper corrections, there's no way of knowing how the original article read, but presumably they had a misleading headline to go with Heneghan's misleading text.

To the credit of its authors, the abstract of the original Danish paper was quite upfront about its limitations.
Bundgaard et al. wrote:Limitation:

Inconclusive results, missing data, variable adherence, patient-reported findings on home tests, no blinding, and no assessment of whether masks could decrease disease transmission from mask wearers to others.
[my bolding again]
In six days the Lord made the heavens and the earth, the sea, and all that is in them. The human body was knocked up pretty late on the Friday afternoon, with a deadline looming. How well do you expect it to work?

User avatar
bob sterman
Fuzzable
Posts: 309
Joined: Mon Nov 11, 2019 10:25 pm

Re: Seasonal effect?

Post by bob sterman » Mon Nov 23, 2020 6:09 pm

snoozeofreason wrote:
Mon Nov 23, 2020 5:43 pm
As is generally the case with newspaper corrections, there's no way of knowing how the original article read, but presumably they had a misleading headline to go with Heneghan's misleading text.
The Wayback Machine has what looks like the first version of the article from 19 November 2020, 7:45am with the original misleading headline...

Landmark Danish study shows face masks have no significant effect
https://web.archive.org/web/20201119080 ... -covid-19-

User avatar
snoozeofreason
Clardic Fug
Posts: 190
Joined: Fri Nov 15, 2019 1:22 pm

Re: Seasonal effect?

Post by snoozeofreason » Mon Nov 23, 2020 6:37 pm

bob sterman wrote:
Mon Nov 23, 2020 6:09 pm
snoozeofreason wrote:
Mon Nov 23, 2020 5:43 pm
As is generally the case with newspaper corrections, there's no way of knowing how the original article read, but presumably they had a misleading headline to go with Heneghan's misleading text.
The Wayback Machine has what looks like the first version of the article from 19 November 2020, 7:45am with the original misleading headline...

Landmark Danish study shows face masks have no significant effect
https://web.archive.org/web/20201119080 ... -covid-19-
Thus confusing "significant" with "statistically significant" as well as confusing the risk of catching the virus with the risk of spreading it. Given the consequences of Covid, a fairly small decrease in the risk of catching or spreading it would be significant in the everyday "does this matter?" sense of the word, even if the study lacked the power to detect it.
In six days the Lord made the heavens and the earth, the sea, and all that is in them. The human body was knocked up pretty late on the Friday afternoon, with a deadline looming. How well do you expect it to work?

Herainestold
Snowbonk
Posts: 446
Joined: Mon Nov 25, 2019 1:23 pm

Re: Seasonal effect?

Post by Herainestold » Mon Nov 23, 2020 7:01 pm

So how would you design an RCT to see if masks worked as source control rather than protecting the wearer? I can't think of how to do it.

User avatar
bob sterman
Fuzzable
Posts: 309
Joined: Mon Nov 11, 2019 10:25 pm

Re: Seasonal effect?

Post by bob sterman » Mon Nov 23, 2020 7:28 pm

Herainestold wrote:
Mon Nov 23, 2020 7:01 pm
So how would you design an RCT to see if masks worked as source control rather than protecting the wearer? I can't think of how to do it.
I can think of a way to do it, but I can't think of a way to do it ethically!

Herainestold
Snowbonk
Posts: 446
Joined: Mon Nov 25, 2019 1:23 pm

Re: Seasonal effect?

Post by Herainestold » Mon Nov 23, 2020 7:57 pm

bob sterman wrote:
Mon Nov 23, 2020 7:28 pm
Herainestold wrote:
Mon Nov 23, 2020 7:01 pm
So how would you design an RCT to see if masks worked as source control rather than protecting the wearer? I can't think of how to do it.
I can think of a way to do it, but I can't think of a way to do it ethically!
Exactly

User avatar
snoozeofreason
Clardic Fug
Posts: 190
Joined: Fri Nov 15, 2019 1:22 pm

Re: Seasonal effect?

Post by snoozeofreason » Mon Nov 23, 2020 8:32 pm

I'd imagine this is one of those circumstances where you have to accept that any evidence you have is likely to be observational, and weigh up the risk of acting on such evidence if it's wrong, versus ignoring it if it's right. I don't think there would be much to complain about if Heneghan was making an argument along those lines, but he isn't.
In six days the Lord made the heavens and the earth, the sea, and all that is in them. The human body was knocked up pretty late on the Friday afternoon, with a deadline looming. How well do you expect it to work?

User avatar
Woodchopper
After Pie
Posts: 2105
Joined: Sat Oct 12, 2019 9:05 am

Re: Seasonal effect?

Post by Woodchopper » Mon Nov 23, 2020 8:43 pm

snoozeofreason wrote:
Mon Nov 23, 2020 8:32 pm
I'd imagine this is one of those circumstances where you have to accept that any evidence you have is likely to be observational,
Yes, find some good natural experiments. For example compare similar US counties that had different mask regulations.

Herainestold
Snowbonk
Posts: 446
Joined: Mon Nov 25, 2019 1:23 pm

Re: Seasonal effect?

Post by Herainestold » Mon Nov 23, 2020 10:29 pm

Woodchopper wrote:
Mon Nov 23, 2020 8:43 pm
snoozeofreason wrote:
Mon Nov 23, 2020 8:32 pm
I'd imagine this is one of those circumstances where you have to accept that any evidence you have is likely to be observational,
Yes, find some good natural experiments. For example compare similar US counties that had different mask regulations.
I believe that has been done, but I couldn't find the paper.

User avatar
jdc
Hilda Ogden
Posts: 614
Joined: Wed Sep 25, 2019 4:31 pm
Location: Your Mum
Contact:

Re: Seasonal effect?

Post by jdc » Mon Nov 23, 2020 11:24 pm

bob sterman wrote:
Mon Nov 23, 2020 7:28 pm
Herainestold wrote:
Mon Nov 23, 2020 7:01 pm
So how would you design an RCT to see if masks worked as source control rather than protecting the wearer? I can't think of how to do it.
I can think of a way to do it, but I can't think of a way to do it ethically!
In theory, could you do it the same as Danmask but test contacts of mask wearers and controls for infection rather than testing the mask wearers and controls? Obviously completely impractical, but you can't expect me to worry about every detail.

User avatar
jdc
Hilda Ogden
Posts: 614
Joined: Wed Sep 25, 2019 4:31 pm
Location: Your Mum
Contact:

Re: Seasonal effect?

Post by jdc » Mon Nov 23, 2020 11:38 pm

Herainestold wrote:
Mon Nov 23, 2020 10:29 pm
Woodchopper wrote:
Mon Nov 23, 2020 8:43 pm
snoozeofreason wrote:
Mon Nov 23, 2020 8:32 pm
I'd imagine this is one of those circumstances where you have to accept that any evidence you have is likely to be observational,
Yes, find some good natural experiments. For example compare similar US counties that had different mask regulations.
I believe that has been done, but I couldn't find the paper.
Nor can I at the moment.

I did find https://www.nature.com/articles/s41598-020-72798-7 (Efficacy of masks and face coverings in controlling outward aerosol particle emission from expiratory activities)

Herainestold
Snowbonk
Posts: 446
Joined: Mon Nov 25, 2019 1:23 pm

Re: Seasonal effect?

Post by Herainestold » Tue Nov 24, 2020 12:04 am

jdc wrote:
Mon Nov 23, 2020 11:24 pm
bob sterman wrote:
Mon Nov 23, 2020 7:28 pm
Herainestold wrote:
Mon Nov 23, 2020 7:01 pm
So how would you design an RCT to see if masks worked as source control rather than protecting the wearer? I can't think of how to do it.
I can think of a way to do it, but I can't think of a way to do it ethically!
In theory, could you do it the same as Danmask but test contacts of mask wearers and controls for infection rather than testing the mask wearers and controls? Obviously completely impractical, but you can't expect me to worry about every detail.
If everybody had a viable contact tracing app on their phone you could trace all the contacts of the experimental participants and then test them.

Post Reply