No significant effect of lockdowns on Covid-19 spread.

[PeterNSteinmetz]

Perhaps. But influenza data is still being reported, in contrast to the earlier claim.

Not quite sure what you mean by “the earlier claim”. But that data on the ILI surveillance so far suggests the 2020-2021 season will be on the low side of normal.

 

[PeterNSteinmetz]

Perhaps they could have compared and contrasted with successful countries?
I read it as a poorly done study.

Well then, what were the authors reasons for their choices ?

 

[PeterNSteinmetz]

Alaska's department of health did a study and found:
"Anchorage's mask mandate and summer 'hunker down' orders helped reduce spread of COVID-19, state report finds"
https://www.adn.com/alaska-news/202...reduce-spread-of-covid-19-state-report-finds/
(which has a link to the study report itself)

Executive summary reads:
"This report summarizes changes in the COVID-19 epidemic in Anchorage following Emergency Orders (EOs) enacted to reduce virus transmission and thus prevent excess severe illnesses and deaths. Following an EO to wear facial covering (masks) in most public locations, self-reported mask use increased, and the growth of the epidemic slowed. After another EO that restricted the number of persons allowed in public venues and the subsequent closure of those venues, daily case counts declined and maintained a declining pattern while these EOs were in effect. The data presented here indicate that the local EOs, a mask mandate, and targeted restrictions on gathering locations in Anchorage appear to have contributed to decreasing SARS-CoV-2 transmission rates."

I can't comment on the details of the study... Just throwing in another data point from one locality.

Quick read suggests no comparison group and essentially is the post hoc, ergo propter hoc argument. I would have to check what the Lyu and Wehby published study said about Alaska.

 

[PeterNSteinmetz]

View attachment 93471


More likely, they excluded data that didn't fit the narrative.

So now you are just speculating and attributing malign motive to some of the leading researchers in the field. Oh dear.

 

[PeterNSteinmetz]

These below:

Please note, “not being discussed” and “data are being lumped in” are different than your statement of not being reported.

I do see the source of some ambiguity here though. I agree that “lumped in” is not quite the correct description.

I think the overall point is that it appears the total ILI situation, including both Covid-19 and other flus, appears to be the same as a bad normal flu season. It is not as though Covid has come along and stacked on top of the normal flu, rather it appears it has largely displaced it.

Frankly, the extent of this surprises me.

 

[PeterNSteinmetz]

Except you are being deliberately misleading again in your choice of thread title, because this study isn't comparing lockdowns versus no lockdowns, but mandatory lockdowns vs voluntary interventions in Sweden and testing, contract tracing, and isolation in South Korea. This is not a study comparing lockdowns to doing nothing.

That is of course an ad hominem attack because you are attributing motive to me. This will be my last comment in this thread on title choice and I will not be revising it. You choose to interpret it a particular way, so be it.

 

[PeterNSteinmetz]

Even a cloth mask helps to protect others]

Anecdotal of course. And the overall scientific evidence on the source effect of cloth masks is quite mixed. The results of the DANMASK study, while not directly addressing source effect do not argue in its favor. And as noted above by David Megginson, the in-vitro studies strongly suggest a cloth mask is even less effective than the surgical masks used in the DANMASK study.

 

[PeterNSteinmetz]

Let me offer the following as perhaps a synthesis regarding the original subject here, namely the effect of coercive lockdowns, as well as social distancing, voluntary measures.

It can simultaneously be the case that

1. Staying away from infected people will seriously reduce your chances of being infected by Covid-19.

2. That people will, at least for a period of time, choose to isolate themselves when such a pandemic is occurring.

3. That in the long run the virus will spread until heard immunity is reached, either through natural infection or vaccines.

4. That coercive lockdowns orders don’t significantly slow the spread in most western nations due to a combination of factors such as people traveling and socializing despite such orders.

Nothing inherently contradictory in all 4 of those given the complexity of the biology of viral illnesses and human behavior. And it strikes me that the data suggest all 4 of these are likely true.

 

[PeterNSteinmetz]

The DANMASK study does not argue in favor of source effect since it didn't address it, neither directly or indirectly. What you wrote above is called doubletalk.

Well, let’s consider that scientific question of what the DANMASK study implies in terms of the source effect. It does not directly address that question, as noted by the authors, but I think does argue indirectly against it.

If a source effect is true, it must be because the mask of the wearer is reducing the total number of virions to which the recipient is exposed, thereby reducing the likelihood of infection.

If there is such a reduction, is it likely that it is only one way? If it is of equal strength both ways, then such an effect should benefit the wearer as well. What the DANMASK study showed is that there is no such significant effect for the wearer of surgical masks, at least as measured when making a recommendation to wear the masks and it was self-reported they were worn.

We also know from the in-vitro studies that the cloth masks are considerably less effective than the surgical masks at impeding the number of droplets expelled.

So for there to be a significant source effect given the results of the DANMASK study it would need to be the case that there is effectively a strong asymmetry for the protective effect for the wearer and the recipient. While that is possible, I don’t think one can argue that the results of the DANMASK study argue in favor of a source effect. Do you? If so, how do those results in particular support the existence of a source effect?

I would put the DANMASK study in the indirect evidence against the source effect hypothesis (not strong direct evidence). Thus I said they do not support such an effect. Sadly we don’t have much good data regarding the source effect.

 

[PeterNSteinmetz]

That is actually not true. The DANMASK study showed there was not a 50% reduction in infection rates among mask wearers. The study was inconclusive regarding protective effect below 50%. Your extrapolation of the data to try to argue against source control is mere speculation.

Let's look at the study itself (available at https://www.acpjournals.org/doi/full/10.7326/M20-6817). There has been a great deal of mis-reporting in the media.

It reported NO significant effect (p<0.05) on the likelihood of infection due to the recommendation to wear a mask and self-reported wearing. The confidence interval for a possible effect ranged from 46% reduction to a 23% increase in the rate of infection. It had an 80% power to detect a 50% reduction and failed to detect that.

What this means in more lay terms is that that there was a 95% chance that any observed differences in the rates of infection were due to chance. However, on the off chance the intervention had some effect, the best estimate would be a 15% reduction, but there is a 95% chance the true effect lay somewhere between a 46% reduction and a 23% increase.

Agreed, we need to be cautious about interpreting null results such as that obtained in the DANMASK study (Edited post above to reflect that I was speaking of statistically significant effects). Agreed, any inference regarding source effect from the DANMASK study is weaker. But I don't think the results of the DANMASK study can be said to argue for a source effect at all.

 

[PeterNSteinmetz]

So you are saying a cloth mask wouldn't stop the majority of what is coming out of the guy's mouth below? I'd hate to be in front of him.

Me too. I am not saying that at all. WRT to mask wearing, I am saying that I think the evidence that the general public wearing cloth mask will slow the spread of COVID-19 is at best mixed. And that the evidence that coercive mask mandates will slow the spread of Covid-19 is even weaker.

And that guy should be coughing or sneazing into his sleeve or a kerchief if he wants to keep his mask clean. Why even wear one like that?

 

[PeterNSteinmetz]

Unfortunately, there are many people like the guy below. I'm not even sure he is sneezing. If he were talking to me, I'd want him to wear his mask. You follow your own choice.

In terms of my own choices. I try not to be in crowds like that at all. High risk situation. If I must be in something like that, I wear an N95 and don't depend on others to exercise good sense. I also try to distance more from people behaving like that.

For the record, if people or businesses (not in mandated areas) ask me to wear a mask, I do so as a courtesy to them.

 

[PeterNSteinmetz]

Masks are inexpensive, and without going into detail, there is a very long list of coercive mandates that have far greater costs and other consequences, so I've never understood why there is so much resistance to this one.

I agree that sadly there is such a very long list of coercive mandates with far greater costs and other consequences.

In terms of the resistance to this, I can only speak speculatively on why there has been as much resistance as there has been generally. But here are my speculations. I think the reasons likely fall into 3 categories:

1. Discomfort. Let's face it, the more effective the mask is, the less comfortable wear. Example: wear a P100 versus a single layer cloth mask. People do find these things uncomfortable and annoying, for a variety of reasons. If the possible benefits have mixed evidence and the potential harm to other people is very indirect, I can sympathize with why people resist wearing them.

2. Potential medical harm. There is some evidence, for example MacIntyre et al. 2015, that cloth mask wearing could increase the likelihood of respiratory infections overall. Not convincing evidence in my opinion, but some. And this does agree with people's feelings that restricting ones breathing is not a good thing. Also there is fairly good evidence that they can cause allergic reactions in susceptible individuals.

3. Political. Obviously won't go into detail on this, but I think a fair number of people feel that it is wrong for the government to coerce them in the presence of this type of mixed evidence. And of course this agrees with the feeling of a fair number of US citizens that they just don't like the government telling them what to do.

I really have no idea how these three categories would rank in terms of importance for the majority of people who resist mask wearing.

 

[PeterNSteinmetz]

Still headed for 3.2M dead in the US unless the official infection rates are wildly off. Need about a 4X difference in reported values vs reality to really notice much of a difference in timeline — many months of change toward “sooner” on that linear graph.

I am curious where you got the 3.2M dead?

My own rough guestimate (as you note, very inaccurate on all of this right now) would go as follows:

Estimates of R0 suggest a herd immunity fraction of about 65%. IFR appears to be on the order of 0.25% (or maybe lower with improved treatments, but that is sort of middle of the range).

So 330M people x 0.65 x 0.25% = 536,250.

Obviously back of the envelope calculation and maybe something more sophisticated predicts higher? (Or maybe that is 1% of the 330M?)

 

[PeterNSteinmetz]

That is actually not true. The DANMASK study showed there was not a 50% reduction in infection rates among mask wearers. The study was inconclusive regarding protective effect below 50%.

This statement is not fully correct and could mislead readers into thinking the DANMASK study says nothing about the protective effect or lack thereof below a 50% reduction. That is incorrect.

If one is going to interpret this study which failed to show a significant level of protection for the wearer, it is best to stick to the best estimator, which is a 15% reduction, and a possible range from a 46% reduction to a 23% increase in the likelihood of infection.

Based on that confidence interval, one can compute that overall likelihood that the surgical mask wearing group had an actual INCREASE of infection, not a decrease. That likelihood is close to about 25%. Not an insignificant chance that the wearing of the surigcal masks actually did harm.

Of course, it was a null result. So there is a 95% chance that any such observations were due to chance alone.

 

[PeterNSteinmetz]

Here is another way to think about the possibility of a source effect. What sort of experiments would demonstrate there is not a source effect, at least in principle?

The authors of the last recent review I had posted basically stated they weren't sure that one could be performed. Certainly such an experiment can be conceived in principle, so the source effect hypothesis is not literally a non-falsifiable hypothesis in the logical sense. But it is rather close if no such experiment can practically be performed.

So I ask supporters of the source effect hypothesis, what sort of data would persuade you that there is no source effect? If one has a belief for which there is no possible data in principle which would refute it, then one has a non-falsifiable belief. Further discussion is pointless.

 

[PeterNSteinmetz]

Stop trying to debate me via PM. It is not appreciated.

Hey, was just trying to politely give you a chance privately to remove your numerous posts which violate the TOS.

(To the moderators, I will remove this once his violating post quoted is removed.)

 

[PeterNSteinmetz]

Source of the charts with data directly from CDC.
https://www.cdc.gov/flu/weekly/index.htm

Thanks for the link.

So isn't Hong Kong an example of much of what @David Megginson said. Namely "Wealthy islands in ... the Pacific with outstanding political leadership, stringent entry/exit requirements, and no land borders. 2. Authoritarian country that can jail anyone who doesn't follow guidelines. " Admittedly closer to china so not in the middle of the Pacific and only recently becoming much more authoritarian.

I think those kind of differences between those and Western countries are what make such trans-national comparisons difficult in terms of supporting an argument for causation.

 

[PeterNSteinmetz]

But they all wore masks right away and prevented millions of infections and continue to have extremely low infection rates (159th on the list of countries for infection per capita). So tell me how do masks not work? Seems to me pretty clear cut that if everyone wore masks right away, all the time, and continued to do so, we would not have many infections.

Maybe. Here are a few of the possible other effects to consider that make the argument for causation more difficult.

Being an isolated isthmus and being able to control entry and exit strictly may be an important component of avoiding an outbreak of infection. New Zealand was similar in this regard. Countries with major land borders or tunnels may have a harder time controlling the spread.

ETA: As @Bob Noel notes, do we know the level of actual compliance?

So I don't know that we can conclude that it was entirely or even primarily the wearing of masks which was causal.

Also I don't know the quality of the masks they were wearing. It is fairly clear from the evidence that the type matters a lot. Healthcare workers wearing N95s properly experience about a 95% reduction in the rate of infection. The evidence for single layer cloth masks is much weaker and if there is an effect it is likely much smaller.

So I would not say "masks not (sic) work". I would say the evidence for cloth mask wearing by the general public slowing the spread of Covid-19 is mixed at best. And the evidence for a mandate to wear them working in a relatively free Western country with larger land borders is even weaker.

 

[PeterNSteinmetz]

No, Hong Kong connects directly to mainland China. Extensive ferry service to many of the islands. It's easy to take a train from Hong Kong to/from Shenzhen. A year ago, my USA passport let me into HK, no visa required.

I stand corrected, thanks. What about since the beginning of the pandemic, how easy is it for people to cross?