Thread by @DFisman, So sickkids is doing a study to try to figure out whether [...]

So sickkids is doing a study to try to figure out whether masking is harmful in kids??

There are 10,000 important kid-related questions for COVID and this is their research focus?

I can’t even

Psst if you’re looking for a study idea can we please get some seroprevalence estimates from left over pediatric sera? That’s a major gap in Canada currently k thx

This is pretty weird stuff

It’s hard not to see this as a response to some of the raised eyebrows around some of the weirder mask-related assertions in the sk document

But first off, as someone who’s reviewed a lot of clinical epi grants and papers over the past 2 decades: what knowledge gap are they trying to fill here?

Do we think that kids touching masks poses a risk to child and population health that outweighs the benefits of masking?

I think we’ve talked about this before

It’s silly https://twitter.com/DFisman/status/1277582297160650753?s=20

https://twitter.com/DFisman/status/1277582297160650753?s=20

So as a reviewer, I’m going to take away points for this study focussing on a question that nobody is really asking (except apparently at sickkids, and even there not seriously, since they’re requiring children to wear masks in their hospital)

And I agree, no masks for kids under two, who last I checked don’t go to school.

But even more fundamentally, this study suggests that the folks at sickkids, who are driving policy around school opening to a degree that is ______ (see poll below) don’t understand what the push is for masks with COVID.

We are much less interested in masks protecting the wearer than we are in using masks and shields to protect those around the wearer, if the wearer has asymptomatic or presymptomatic COVID.

There’s a bunch of fancy biophysics references available now on cloth
masks and their ability to reduce large droplet cloud and aerosol formation.

For surgical masks this isn’t in dispute: that’s why surgeons wear them when they operate.

But rather than struggle through a biophysics paper, I’d recommend you watch this backyard science demonstration by someone named Uncle Rob

All this leads us to the importance of masks in making superspreader events in close/closed/crowded settings (why hello, Ontario public schools)

It’s pretty clear at this point that COVID has an almost bimodal reproduction number. It’s a disease that often causes no, or few, secondary infections, but sometimes caused many

It’s a weird bimodal R0 with a lot of dead ends, or 1-2 secondary infections, but sometimes super spreading with literally a dozen or more secondaries. This is the Pareto distributed R0 that’s been such a notable characteristic of this disease.

https://www.eurosurveillance.org/content/10.2807/1560-7917.ES.2020.25.4.2000058

Mechanistically that seems like a disease with multiple modes of transmission; some common and less efficient (contact, heavy droplet, fomite) and others rarer but highly efficient...so much so that we wind up with crazy secondary:primary case ratios...

80% of secondaries from 10% of primary infections, per one recent estimate https://www.scientificamerican.com/article/how-superspreading-events-drive-most-covid-19-spread1/

How ‘Superspreading’ Events Drive Most COVID-19 Spread

As few as 10 percent of infected people may drive a whopping 80 percent of cases in specific types of situations

https://www.scientificamerican.com/article/how-superspreading-events-drive-most-covid-19-spread1/

So: a disease that sometimes spreads like its cousin MERS (R0 ~1-2) but sometimes like measles (R0 12-18), with our observed R0 of 2-3 giving a sense of the relative frequency of these scenarios. (I’m oversimplifying here).

As noted yesterday: https://twitter.com/DFisman/status/1291782567914614785?s=20

https://twitter.com/DFisman/status/1291782567914614785?s=20

That’s a lot of goo. Possible with kids. But Occam’s razor is probably happiest if extraordinary superspreader events are from true aerosol/droplet nuclei transmission.

That is, someone coughs out enough small but heavily virus laden droplets that air in a closed space becomes infectious. That would depend on high mucus viral load, combined w droplet generating behaviour (loud talking, singing, coughing), maybe in someone with lung involvement

Pulmonary infection is associated with smaller droplets being generated by cough.

Let’s remember: asymptomatic infection (in a child) with pneumonitis/pneumonia on chest CT was recognized in one if the first family clusters of COVID

https://www.thelancet.com/action/consumeSharedSessionAction?JSESSIONID=aaaVBpH5aA8vP_pnbfnpx&MAID=UzRenEZOb2Y5685oStZMog%3D%3D&SERVER=WZ6myaEXBLF7C%2B0zm%2FX%2BaQ%3D%3D&ORIGIN=145942706&RD=RD&rtc=0

So high viral load in mucus, aerosolized with behavioural and/or anatomical assist. Probably in a presymptomatic individuals as viral loads in mucus are highest at day zero of symptoms, or even in late incubation

Let’s have this occur in a poorly ventilated space (meat packing plant, crowded bar, school) and voila, the stage is set for a cloud that can infect individuals who breathe in that space, probably for a few hours.

I have discussed with @BillHanage and @CarolineColijn that you need to call this the Lemony Snickey model of COVID superspreading, because it depends on a series of unfortunate events, in series.

This model of COVID transmission has already been described by Schijven et al in this brilliant paper. https://www.medrxiv.org/content/10.1101/2020.07.02.20144832v1

I think it dovetails beautifully with @C_Althaus ’s work on Pareto distributed R0

Exposure assessment for airborne transmission of SARS-CoV-2 via breathing, speaking, coughing and...

Background Evidence for indoor airborne transmission of SARS-CoV-2 is accumulating. If SARS-CoV-2 also spreads via aerosols, this has implications for measures taken to limit transmission. Objectives...

https://www.medrxiv.org/content/10.1101/2020.07.02.20144832v1

So back to the masks: inasmuch as COVID depends on tipping points/thresholds, and inasmuch as the generation of clouds of respiratory droplets is impeded by masks (vide supra for Uncle Rob video), and inasmuch as symptom screening in kids ain’t gonna keep COVID out of schools

(Vide supra for sickkids own hospital guidance on why mask wearing is so important even when people don’t have symptoms), and inasmuch as our provincial government seems prepared to fund an extra custodian per 3.5 schools for “safe opening” [sic]

Encouraging mask use for kids really is likely to be an important tool for reducing school-based superspreader events and school-driven COVID resurgence. It probably won’t be enough but every little bit helps

There’s more to say about this weird little sickkids simulation study (obviously not the awesomest optics to base it at BSS, UCC, and Branksome Hall when you’re making guidance for cash-strapped public schools)

Weirdness in recruitment materials that
tells us more about the investigators than the study, and which, if they’re going to be asking subjects about anxiety counts as “contamination”:

“Mask wearing for a prolonged period of time may also be anxiety-provoking or frustrating.”

I’ll leave that to others to comment on.
This study is a weird coda to a weird report, and looks like yet more decision-based evidence making.

“We found that 50% of students touched their masks in a manner that could result in contamination of hands” sort of stuff

Again: doesn’t matter. That’s not what the masks are about. Keep up.

Some kids have sensory issues or medical exemptions. Sure. It’s not about perfection.

PS: nieces and nephews in masking jurisdictions. They do it. It’s fine. Easier if grownups set good examples and don’t create issues when there aren’t any.

Good night

Latest Threads Unrolled: