Read on Twitter
Thank you for holding the 
https://abs.twimg.com/emoji/v2/... draggable="false" alt="đŹđ§" title="Flag of United Kingdom" aria-label="Emoji: Flag of United Kingdom"> #Lockdown!
So important, as ~1 in 12 UK people (5.5 million) might have #COVID19. ~1 in 7 in London.
That would mean there are infected people in every park, in every shop, in every street. It might include you!
Thread w/calculations if interested:
So important, as ~1 in 12 UK people (5.5 million) might have #COVID19. ~1 in 7 in London.
That would mean there are infected people in every park, in every shop, in every street. It might include you!
Thread w/calculations if interested:
Assume: 14,333 deaths (8958 hosp deaths + 60% for lag + non hosp)
Infection (not case) Fatality Rate: 1%
Days infection to death ~23.5
So 1.4m infected up to 23.5 days ago
~12 days to double (adj. for recovery) so ~2 doublings since
So infections now: ~5.5m. ~1m recovered
Infection (not case) Fatality Rate: 1%
Days infection to death ~23.5
So 1.4m infected up to 23.5 days ago
~12 days to double (adj. for recovery) so ~2 doublings since
So infections now: ~5.5m. ~1m recovered
Model is v.sensitive to inputs, so low confidence + wide ranges apply.
More deaths --> more infections
Higher IFR% --> fewer infs (don& #39;t need as many for same deaths)
Longer days to death --> more infs (more time to double)
Higher doubling rate --> more infs
More deaths --> more infections
Higher IFR% --> fewer infs (don& #39;t need as many for same deaths)
Longer days to death --> more infs (more time to double)
Higher doubling rate --> more infs
I& #39;ve added 60% to the DHSC hospital deaths to account for reporting lag + non-hospital deaths.
27 Mar reports were 78% too low. Gap should have narrowed now but non-hosp increasing?
https://ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregisteredweeklyinenglandandwalesprovisional/weekending27march2020">https://ons.gov.uk/peoplepop... https://twitter.com/ChrisGiles_/status/1247458186300456960?s=20">https://twitter.com/ChrisGile...
27 Mar reports were 78% too low. Gap should have narrowed now but non-hosp increasing?
https://ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregisteredweeklyinenglandandwalesprovisional/weekending27march2020">https://ons.gov.uk/peoplepop... https://twitter.com/ChrisGiles_/status/1247458186300456960?s=20">https://twitter.com/ChrisGile...
We were adjusting the infection to death days to account for exponential skew. No longer necessary now growth has slowed. Thanks @cheianov as ever.
12 day doubling is assumed from recent case growth adjusted for assumed recoveries + assumption we& #39;re slowing similar to Italy.
12 day doubling is assumed from recent case growth adjusted for assumed recoveries + assumption we& #39;re slowing similar to Italy.
UK Government (bizarrely) don& #39;t report recovery rates. I& #39;m assuming 18 day average illness duration so counting new cases from past 18 days at 91% and extrapolating to assume "live infections" as 85%, given cases are the more serious.
~1m recovered to date.
~1m recovered to date.
This is a top-down model.
For pro versions that work from deaths, see these from @Imperial_JIDEA, @uniGoettingen, @cmmid_lshtm:
https://www.imperial.ac.uk/media/imperial-college/medicine/mrc-gida/2020-03-30-COVID19-Report-13.pdf
https://www.imperial.ac.uk/media/imp... href=" https://cmmid.github.io/topics/covid19/severity/global_cfr_estimates.html
https://cmmid.github.io/topics/co... href=" https://www.uni-goettingen.de/en/606540.html
All">https://www.uni-goettingen.de/en/606540... low given deaths/growth since + non-hospital + reporting lag additions
For pro versions that work from deaths, see these from @Imperial_JIDEA, @uniGoettingen, @cmmid_lshtm:
https://www.imperial.ac.uk/media/imperial-college/medicine/mrc-gida/2020-03-30-COVID19-Report-13.pdf
All">https://www.uni-goettingen.de/en/606540... low given deaths/growth since + non-hospital + reporting lag additions
Another interesting approach via symptom tracking.
Estimates 1.4m symptomatic infections (down from 1.9m).
Then need to add 0-20 yrs and over 69 years infected, all asymptomatic infected (50%?) and change since.
@Join_ZOE @timspector @KingsCollegeLon https://covid.joinzoe.com/post/covid-isolation">https://covid.joinzoe.com/post/covi...
Estimates 1.4m symptomatic infections (down from 1.9m).
Then need to add 0-20 yrs and over 69 years infected, all asymptomatic infected (50%?) and change since.
@Join_ZOE @timspector @KingsCollegeLon https://covid.joinzoe.com/post/covid-isolation">https://covid.joinzoe.com/post/covi...
It& #39;s heartening to see more press references to % infected, not just the "confirmed cases" iceberg tip.
https://unherd.com/2020/04/how-likely-are-you-to-die-of-coronavirus/">https://unherd.com/2020/04/h... @TomChivers
https://news.sky.com/story/coronavirus-the-data-suggests-the-uk-is-on-course-for-many-thousands-of-deaths-11966517
https://news.sky.com/story/cor... href=" https://telegraph.co.uk/news/2020/03/29/least-16m-people-could-infected-coronavirus-uk-new-estimates/
https://telegraph.co.uk/news/2020... href=" https://metro.co.uk/2020/03/30/coronavirus-slowing-uk-thanks-social-distancing-measures-12476442/
https://metro.co.uk/2020/03/3... href=" https://ftalphaville.ft.com/2020/04/04/1586015208000/Imperial-s-Neil-Ferguson---We-don-t-have-a-clear-exit-strategy-/">https://ftalphaville.ft.com/2020/04/0... @jemimajoanna
https://unherd.com/2020/04/how-likely-are-you-to-die-of-coronavirus/">https://unherd.com/2020/04/h... @TomChivers
https://news.sky.com/story/coronavirus-the-data-suggests-the-uk-is-on-course-for-many-thousands-of-deaths-11966517
Illustration implies total UK death toll of ~69k...
But only if IFR is 1% + we see zero new infections.
If our ICU capacity is overrun, the IFR% will spike.
New infections depends on hard #Lockdown + test/trace/quarantine.
Similar to: http://covid19.healthdata.org/united-kingdom
https://covid19.healthdata.org/united-ki... href="https://twitter.com/IHME_UW">@IHME_UW
But only if IFR is 1% + we see zero new infections.
If our ICU capacity is overrun, the IFR% will spike.
New infections depends on hard #Lockdown + test/trace/quarantine.
Similar to: http://covid19.healthdata.org/united-kingdom
A wave is crashing over our ICUs.
We can& #39;t reduce the size of that wave because it& #39;s caused by infections from 2-5 weeks ago.
We can improve our readiness by scaling ICUs + protecting NHS staff.
We can stop the wave getting bigger + longer by hard #Lockdown + test/trace/quar.
We can& #39;t reduce the size of that wave because it& #39;s caused by infections from 2-5 weeks ago.
We can improve our readiness by scaling ICUs + protecting NHS staff.
We can stop the wave getting bigger + longer by hard #Lockdown + test/trace/quar.
It& #39;s tempting to think... "There& #39;s 14k deaths and #Covid19 kills about 1 in 100, so there must be ~1.4 infections".
The problem is you& #39;re nearly right. But those ~1.4m infections were ~3 weeks ago. That& #39;s how long it takes to kill.
They& #39;ve been doubling every ~12 days since
The problem is you& #39;re nearly right. But those ~1.4m infections were ~3 weeks ago. That& #39;s how long it takes to kill.
They& #39;ve been doubling every ~12 days since
This isn& #39;t the best way of estimating % infected!
We& #39;re waiting for people to die, assuming fatality rate to judge how many infections drove those deaths, then using doubling rates to judge infections today.
A better way? Random pop. sample testing! https://twitter.com/JamieWoodhouse/status/1247187400876580864?s=20">https://twitter.com/JamieWood...
We& #39;re waiting for people to die, assuming fatality rate to judge how many infections drove those deaths, then using doubling rates to judge infections today.
A better way? Random pop. sample testing! https://twitter.com/JamieWoodhouse/status/1247187400876580864?s=20">https://twitter.com/JamieWood...
Why is estimating # infected more important than just tested "cases"?
https://abs.twimg.com/emoji/v2/... draggable="false" alt="1âŁ" title="Keycap digit one" aria-label="Emoji: Keycap digit one"> Infected people are doing the infecting. The "cases" are in hospital/dead/immune
https://abs.twimg.com/emoji/v2/... draggable="false" alt="2âŁ" title="Keycap digit two" aria-label="Emoji: Keycap digit two"> If you know # infected and # deaths you can assess how deadly this thing really is (IFR%, not just "case" fatality)
BJ& #39;s 23 Mar speech should have started:
"There are infected people in every train, in every shop, in every park, in every tube... you might even be one of them... please take what I am about to say very seriously..."
This poll shows a dangerous level of complacency across the UK. I suspect this is worse in the real world.
Most people think less than a million are infected! That& #39;s what happens when govt + press only talk about the "confirmed cases" iceberg tip. https://twitter.com/JamieWoodhouse/status/1245291470107672576?s=20">https://twitter.com/JamieWood...
Most people think less than a million are infected! That& #39;s what happens when govt + press only talk about the "confirmed cases" iceberg tip. https://twitter.com/JamieWoodhouse/status/1245291470107672576?s=20">https://twitter.com/JamieWood...
If it is, #Covid19 will kill more than 1%. When you run out of ICU beds / ventilators, many more die as in Italy/Spain.
If we& #39;d started projecting total infected from day one (ideally via. regular pop. sample tests), we might have acted sooner.
We might also have persuaded people to comply better with our weak #Lockdown.
Instead... "a national scandal" @richardhorton1 https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30727-3/fulltext">https://www.thelancet.com/journals/...
We might also have persuaded people to comply better with our weak #Lockdown.
Instead... "a national scandal" @richardhorton1 https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30727-3/fulltext">https://www.thelancet.com/journals/...
A powerful visualisation of deadly UK #Covid19 complacency:
"Right things... at the right time..." or "Too little, too late"?
H/t @Imperial_JIDEA
"Right things... at the right time..." or "Too little, too late"?
H/t @Imperial_JIDEA
Amateur epidemiologists like me are annoying, but as long as we have pro epidemiologists saying things like:
- 1000x deaths or
- "2-5x cases" or
- 734k or
- 50%+ or
- "We have no way of knowing" (we do - pop sample tests!)
It feels useful to have a top down sanity check?
- 1000x deaths or
- "2-5x cases" or
- 734k or
- 50%+ or
- "We have no way of knowing" (we do - pop sample tests!)
It feels useful to have a top down sanity check?
Thanks to data people grappling w/this critical estimate - how many are really infected?
@TomChivers @EdConwaySky @alexwickham @Ashley_J_Kirk @PaulNuki @AlbertoNardelli @nicholascecil @jburnmurdoch @jemimajoanna @MaxCRoser @AdamJKucharski @tomaspueyo @StefanFSchubert @cheianov
@TomChivers @EdConwaySky @alexwickham @Ashley_J_Kirk @PaulNuki @AlbertoNardelli @nicholascecil @jburnmurdoch @jemimajoanna @MaxCRoser @AdamJKucharski @tomaspueyo @StefanFSchubert @cheianov
#Lockdown is hard.
But every day, more people recover. Maybe 1 million have so far inhttps://abs.twimg.com/emoji/v2/... draggable="false" alt="đŹđ§" title="Flag of United Kingdom" aria-label="Emoji: Flag of United Kingdom">!
Every day, the strictness of our #Lockdown determines how many new people are infected.
That balance will determine whether our NHS can cope and how many lives we can save.
Thank you!
But every day, more people recover. Maybe 1 million have so far in
Every day, the strictness of our #Lockdown determines how many new people are infected.
That balance will determine whether our NHS can cope and how many lives we can save.
Thank you!
