A lot of talk about restaurants.

I'm not sure people are aware of how bad things are looking for Dublin, and Ireland more generally, and restaurants seem like an arbitrary thing to target.

Here's a thread on all of it.
In the US, restaurant data is a very accurate predictor of trends.

It's not that everything hinges on restaurants, or even what staff do to make them safer, it's that people interact more when they make additional choices to socialise indoors (pub, restaurant, cafe, gym).
You see here in April, hardly anybody was going out to eat in restaurants in US. The red squiggle takes you to the first fortnight in April.

By middle of June people were starting to venture out a lot more to eat and dining-out rose closer to pre-pandemic levels by September.
In the 2nd week of April, daily cases were over 30,000 and reached a peak of near 50,000.

They dropped back to 20,000 coinciding with people avoiding restaurants, with a 3 week lag.
In the last week of May, there was a large increase in dining-out.

3 weeks later, cases started to once again rise dramatically.

Near the tail end of August, dining-out started to drop significantly and cases are dropping now in September.
When people discuss restaurants, they aren't framing the discussion correctly.

It's got very little to do with large outbreaks at any restaurant.

Even less to do with practices in individual restaurants, who are universally doing an excellent job on hygiene and table layout.
The reason restaurant usage can predict rises and falls in cases, is the process of going out to eat inherently involves societal interaction and increasing close contacts.

A lot of people like to drink with their meal, and that brings in public transport.
A taxi up-and-back involves interacting with 1 or 2 more people.

A bus up-and-back can involve anywhere from 6 to 8 interactions.

The restaurant itself can involve 8 or 10 with nearby tables deemed a close contact.
When people sit at home on a couch, they interact with nobody outside the house.

When people go out to eat, they can be near 10, 20 people easily.

But doesn't that describe any activity like going to work?
Yes. All of what I just said applies to anything leaving the house.

Getting the Luas to work, for example, you can be near the same amount of people.

But restaurant use is predictive because it's doubling that same process by choice, not by necessity of needing to go to work.
In Spain and Israel, they have much more advanced lockdowns than the one proposed for Dublin, because they didn't act decisively weeks ago.

The 14-day incidence in Spain is 292.2 per 100,000 and Israel leads the world at 573.2 per 100,000.

Things are desperate for them.
We are trying something different in Dublin by not waiting until we're completely screwed to act.

This isn't about the restaurants not doing a good job on all the measures.

It's about this:

The more unnecessary journeys and interactions a society has, the fewer cumulative close contacts, is proven to work in containing epidemic spread.

That's the base justification for closing restaurants and pubs and other areas people congregate indoors during a surge.
It's an open-ended question on when to pull the trigger on a measure like this.

Do you wait until it reaches Spain or Israel level? Do you wait until 150.0 per 100,000 or 200.0 per 100,000 recent infections? Do you wait until hospitals have 200 admitted? 300?
There is now 80 confirmed cases in hospital with 44 hospital admissions in 7 days and now 14 in ICU.

The last 48 hours testing has saw 546 positives.

If all of it continues at same pace we will see so much death in Ireland in the coming weeks. 10+ deaths every day in October.
The interventions they're reaching for, if successful, will turn this tide and then people will say they were never needed to begin with.

Drastic intervention today is badly needed because if something big doesn't change, we will end up like Spain.

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Danny Boy

Danny Boy Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @Care2much18

28 Oct
Here is a complete breakdown of cases, hospital admissions, ICU admissions and deaths by age for the last 3 months ish (August 1st to October 27th) in Ireland.

I've included a comparison to the first 3 months of pandemic (to May 31st) so you can see the progression.
February 29th - May 31st:

0-4 years old:

Cases: 153
Hospital admitted: 19 (10.2%)
ICU admitted: 0 (0.0%)
Deaths: 0 (0%)

August 1st - October 27th:

0-4 years old:

Cases: 1,052
Hospital admitted: 19 (1.8%)
ICU admitted: 1 (0.1%)
Deaths: 0 (0%)
So, for babies and toddlers, you can see in the early pandemic 10% were admitted but that has dropped to 1.8% in last 3 months.

This partly because Covid19 was very new in early March and throughout Europe babies were admitted on a precautionary 'better safe than sorry' basis.
Read 22 tweets
28 Oct
This clip is absolutely disgraceful but I'll counter these misleading claims, in case anyone believes them.

Vaccine-derived polio is a real thing but of course they didn't address a) how rare it actually happens or b) how it happens, instead they just tried to scare people. /1
Vaccine-derived poliovirus (VDPV) rarely causes outbreaks, when it does it's called Circulating vaccine-derived poliovirus (cVDPV).

cVDPV can and does paralyse children.

This happens rarely in deprived areas of Africa with bad sanitation, when children excrete vaccine-virus.
It takes a very long time for genetic changes to occur but rarely a child can continue excreting vaccine-virus for a year, allowing the mutation.

One example of long-term excretion actually happened in Ireland, way back in 2002, and thankfully had a very good outcome.
Read 10 tweets
27 Oct
Northern Ireland's infection levels are beginning to drop from the peak, which is much-needed given how high they were.

There is rapid improvement in Derry & Strabane, with a fall of nearly 600.0 from their peak 14-day incidence.
14-day cases per 100,000:

NI: 721.8
ROI: 309.9

Derry City & Strabane: 1,207.9
Mid-Ulster: 1,025.9
Belfast: 1,009.3
Cavan 967.5
Meath: 667.0
Lisburn & Castelreagh: 666.1
Antrim & Newtonabbey: 624.5
Causeway & Glens: 561.5
Newry & Mourne: 544.4
Fermanagh & Omagh: 484.5
When you look at their dashboard, it says "102% of beds occupied". I'm not sure what that means other than a typo, but it was 94% the previous day.

Hospital picture:


Admitted: 360
ICU: 38


Admitted: 354
ICU: 38
Read 5 tweets
27 Oct
If Irish media insist on giving this "herd immunity" guy a platform in the interests of balance, there should be fact checking at the bottom of his articles because some of his claims are stupidly misleading.

Fine I'll do it myself. /1


"The HSE published (December 2018) the recorded data which showed a Flu mortality (CFR) of 2.14 per cent; that is almost 10-fold higher than Covid-19. Undoubtedly, this is an overestimate but there was a vaccine"

Gonna file this under "completely misleading bullshit".
2018 Flu season:

Notified cases: 11,889
Notified deaths: 255
CFR: 2.14%

Dr. Feeley is then saying this Flu CFR is 10-fold higher than Covid-19 fatality rate.

Interesting argument, and by interesting, I mean embarrassing sleight of hand trickery.
Read 11 tweets
27 Oct
Belgium's positivity rates are reflecting their new policy of not testing asymptomatic close contacts.

As a result, positivity rates in Liège are running well over 33% and the age-specific positivity in those over 80-years-old is even higher.

It is looking desperate. /1
In terms of ICU, it was reported that Belgium asked Netherlands to accept some of their patients but Netherlands declined, on the basis they themselves already had to ask Germany to accept some of theirs.

Maximum capacity in Belgium is 2,000 but the rate of admission is the key.
There's 809 admitted into ICU in Belgium, which is a doubling rate of 8 days with +50 daily.

Belgian experts anticipate ICU will be overwhelmed like Lombardy by November 9th if they can't get the admission rates down.

Germany will accept Belgian patients into ICU, which helps.
Read 6 tweets
26 Oct
This is popping up everywhere as an example of an expert getting a prediction laughably wrong.

Make no mistake, it IS laughably wrong - today.

But you have to consider what this tweet was saying back then, not now, and when you do that, it's not laughable at all. /1
The tweet was written on September 14th.

3 population-level interventions were implemented since this tweet was written:

-Level 3 imposed in Dublin, September 18th
-Level 3 imposed in Ireland, October 6th
-Level 5 imposed in Ireland, October 21st
Tweet projected 5,000 cases a day in Dublin alone at the end of October with no intervention.

That is a considerable distance from Irelands 1,283 cases peak notified on October 18th and obviously leaves him wide open to mockery.

But there were interventions.

3 of them.
Read 12 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!