6 ms·

Reclassification of earlier U.S. coronavirus deaths causes record 24-hour number

61 points by jdkee 7y ago

Aeolun 7y ago
Wow, that’s a doubling of deaths over 24 hours too. But I hope this is exceptional. I cannot see this being 8000 deaths tomorrow.
- surge 7y ago
  Hopefully, this is the peak we expected, meaning it won't keep going up by very much.
  - vkou 7y ago
    The first peak, that is. It's likely that during the many re-openings to come, we will have other peaks.
    - mlyle 7y ago
      I believe that if we loosen things a little soon, this will be the main one. I think it is likely that we have reduced susceptibility by 15% in New York already. And seasonal factors are swinging to be on our side. If R0 was 2.5, a virus with Rt of 1.9 (*0.85, and then a little seasonal benefit) is still scary, but it's a completely different ball game. On the other hand, if we don't experiment with reopening a little soon, we'll lose the chance to effectively use the time where weather is on our side... Emphasis on "a little". Slightly increase the set of permitted activities, designate an industry or two more as permitted, and continue to encourage mask wearing.
      - jhwang5 7y ago
        Not sure why pessimistic views are downvoted. You are just expressing your opinion and providing a reasonable reasoning.
        mlyle 7y ago
        I'm not sure whether I was downvoted for being pessimistic or optimistic. Most places seem headed towards extended controls and demanding a very high standard for any reopening. I'm in the SF Bay Area which has had a fairly steady daily case count in the face of increased testing, but a declining number of COVID cases in ICU over the past 10 days. The latter is a lagging indicator, so whatever trend is there is going to continue for 10+ more days. IMO, we should slightly loosen controls, but I do not think public health will even consider that before a couple weeks have expired.
      - vkou 7y ago
        Do we have any solid evidence for seasonality being a factor?
        mlyle 7y ago
        Only moderate quality evidence, e.g. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551767 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551767
- zouhair 7y ago
  With some people protesting the quarantine by gathering in the street I highly doubt it.
- bb2018 7y ago
  This number included "probable" deaths that had not been included before. This is an improvement in seeing the totality of the situation. However, if you want to track change you can see deaths by the old methodology are level: https://twitter.com/NateSilver538/status/1250917711586168835 https://twitter.com/NateSilver538/status/1250917711586168835
- jshevek 7y ago
  It's not an real doubling of number of people actual dying within a 24 hour period.
chkaloon 7y ago
Didn't NY just add 3+ thousand that weren't counted before?
- Scipio_Afri 7y ago
  2 days ago. "New York City raised its likely death toll from coronavirus by 3,778 people on Tuesday afternoon" https://www.wsj.com/articles/nursing-homes-remain-a-concern-as-new-yorks-coronavirus-outbreak-appears-to-plateau-11586892250 https://www.wsj.com/articles/nursing-homes-remain-a-concern-...
  - chkaloon 7y ago
    maybe there was a slight delay getting them in this count
    - jMyles 7y ago
      It's not (or at least, not only) a delay, but a change in methodology of classification. Previously, all patients who tested positive for SARS-CoV-2 were counted, regardless of their cause of death. In the new methodology (in order to ensure that false negatives aren't missed), people with a pattern of symptoms are also counted, regardless of their cause of death. So, many, many more people are counted than those who died directly from complications arising from Covid-19. This is not altogether unreasonable; we want to know how many people are dying after contracting the virus. However, it is being reported as though the virus is the cause of death, when nobody doing the reporting is claiming that it is. It's a total mess. Here's the actual data sheet: https://www1.nyc.gov/assets/doh/downloads/pdf/imm/covid-19-daily-data-summary-deaths-04162020-1.pdf https://www1.nyc.gov/assets/doh/downloads/pdf/imm/covid-19-d...
      - sergers 7y ago
        Are they including the people who die at home? They weren't including these results previously as they were overwhelmed with testing the living, so weren't testing suspected deaths at home
vanniv 7y ago
This is a garbage number. New York reclassified a bunch of folks that died over the last few days as COVID deaths under their new methodology, where all deaths are covid deaths if there is even a hint that they might have had COVID. These are not deaths that occurred in a 24-hour period. These are deaths that were reported in a 24-hour period, and covers basically everyone with any respiratory illness who died at home on Tuesday, Wednesday, or Thursday. Trying to compare that to a rate of deaths in hospital of likely or confirmed cases (the methodology used from last week until Tuesday), or hospital deaths of confirmed cases (the methodology used before last week) is meaningless. If you change methodology, you really have to go back and recast historic numbers. But that isn't being done.
- jMyles 7y ago
  ...what an asinine headline. This is open letter worthy from reputable epidemiologists. It's been hard enough to identify and asses data (and expert commentary) without unambiguously deceptive (and panic-riddled) reporting like this.
  - LB232323 7y ago
    The profitable manufacture of fear and justification of government overreach has been a reliable product and goal of sensationalist journalism for a very long time. The Spanish-American War is a prime example. If anything, this crisis has drawn visible lines between those with unwavering trust in mass media and those without. Those who have good reasons to distrust authority have been the least likely to comply with orders and recommendations from both government agencies and mass media outlets. On the other hand, those who benefit from the current power structure have taken to their advice with great enthusiasm. Between these groups are those like yourself who try to sift through the sensationalism for good information. Even then, international authorities like WHO have revealed clear political bias in regards to issues like Taiwanese independence. Who can we trust?
    - jMyles 7y ago
      > The Spanish-American War is a prime example. Where is our John Quincy Adams?
dorkwood 7y ago
I think this is because NY is now reporting probable deaths, and not just those who tested positive before dying.
- cylinder 7y ago
  Dr Birx said the whole country is doing this, and she seems totally fine with it.
maxroham1977 7y ago
That's a wrong number for daily deaths. It factors in previous deaths counts that weren't counted (mainly in NY). Those numbers should have been backdated. Currently it seems like the daily death toll is rising, although it doesn't (at least not so sharply).
- aaron695 7y ago
  This is not what the media is reporting, they say it's in a 24 hour time period. Do you have a link?
neonate 7y ago
https://archive.md/tD5QU https://archive.md/tD5QU
jb775 7y ago
These numbers need to be seriously scrutinized for everyone's sake. Hospitals are now financially incentivized to claim as many Coronavirus hospitalizations/deaths as possible so they get their piece of the stimulus package.
- draw_down 7y ago
  Politicians are incentivized to lie and dissemble in the other direction. Pick your poison.
- NotSammyHagar 7y ago
  That sounds like a conspiratoral worry. Hospitals are all lying? We should just publish overall death rates long with traditional death rates and known cv cases. it's guaranteed there are missed cv tests. In Seattle they have not had enough cv tests to 'bother' testing the sad cases of large numbers of people dying, even in the famous kirkland senior citizen homes.
beezle 7y ago
This was brought up during the daily Cuomo yesterday. The media have not fully explained the new reporting (am I cynical if I say so they get better clicks?) CDC mandated that all localities, not just NYC/NYS, count probable deaths in addition to those where a covid positive test was available. These figures are not part of the 'official' covid death totals but instead are a separate category. The reason this has been done is likely due to the reports coming from NYFD/PD concerning the much higher incidence of finding dead people when called. Those bodies are not being brought to hospitals for autopsie or anything else if there are no suspicious circumstances. By one report I read a few days back, a year ago it was typical to have encounter 25 dead on 911 calls over a 24 hour period. They are now seeing upwards of 200.
asabjorn 7y ago
It is worthwhile to notice that there are as of April 13 there are 11k weekly coronavirus related deaths according to the CDC out of 600k tested [1]. In a normal flu season there are 8k daily flu deaths [2]. The ones that are the worst affected should be in the hospital. The cumulative rate of hospitalization is 12.3 per 100k, which adds up to 39k. So that puts an upper bound on known risk. [3] [1] https://www.cdc.gov/nchs/nvss/vsrr/COVID19/index.htm https://www.cdc.gov/nchs/nvss/vsrr/COVID19/index.htm [2] https://www.cdc.gov/mmwr/volumes/68/wr/mm6826a5.htm https://www.cdc.gov/mmwr/volumes/68/wr/mm6826a5.htm [3] https://gis.cdc.gov/grasp/COVIDNet/COVID19_3.html?fbclid=IwAR02UGPEV4r5LicjyNum1mpmFixTpCPYHJY8Nf3d8QV1iKrKK6ZlGTecifs https://gis.cdc.gov/grasp/COVIDNet/COVID19_3.html?fbclid=IwA...

61 points by jdkee 7y ago

Wow, that’s a doubling of deaths over 24 hours too. But I hope this is exceptional. I cannot see this being 8000 deaths tomorrow.

surge 7y ago
Hopefully, this is the peak we expected, meaning it won't keep going up by very much.
- vkou 7y ago
  The first peak, that is. It's likely that during the many re-openings to come, we will have other peaks.
  - mlyle 7y ago
    I believe that if we loosen things a little soon, this will be the main one. I think it is likely that we have reduced susceptibility by 15% in New York already. And seasonal factors are swinging to be on our side. If R0 was 2.5, a virus with Rt of 1.9 (*0.85, and then a little seasonal benefit) is still scary, but it's a completely different ball game. On the other hand, if we don't experiment with reopening a little soon, we'll lose the chance to effectively use the time where weather is on our side... Emphasis on "a little". Slightly increase the set of permitted activities, designate an industry or two more as permitted, and continue to encourage mask wearing.
    - jhwang5 7y ago
      Not sure why pessimistic views are downvoted. You are just expressing your opinion and providing a reasonable reasoning.
      - mlyle 7y ago
        I'm not sure whether I was downvoted for being pessimistic or optimistic. Most places seem headed towards extended controls and demanding a very high standard for any reopening. I'm in the SF Bay Area which has had a fairly steady daily case count in the face of increased testing, but a declining number of COVID cases in ICU over the past 10 days. The latter is a lagging indicator, so whatever trend is there is going to continue for 10+ more days. IMO, we should slightly loosen controls, but I do not think public health will even consider that before a couple weeks have expired.
    - vkou 7y ago
      Do we have any solid evidence for seasonality being a factor?
      - mlyle 7y ago
        Only moderate quality evidence, e.g. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551767 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551767
zouhair 7y ago
With some people protesting the quarantine by gathering in the street I highly doubt it.
bb2018 7y ago
This number included "probable" deaths that had not been included before. This is an improvement in seeing the totality of the situation. However, if you want to track change you can see deaths by the old methodology are level: https://twitter.com/NateSilver538/status/1250917711586168835 https://twitter.com/NateSilver538/status/1250917711586168835
jshevek 7y ago
It's not an real doubling of number of people actual dying within a 24 hour period.

chkaloon 7y ago

Didn't NY just add 3+ thousand that weren't counted before?

Scipio_Afri 7y ago
2 days ago. "New York City raised its likely death toll from coronavirus by 3,778 people on Tuesday afternoon" https://www.wsj.com/articles/nursing-homes-remain-a-concern-as-new-yorks-coronavirus-outbreak-appears-to-plateau-11586892250 https://www.wsj.com/articles/nursing-homes-remain-a-concern-...
- chkaloon 7y ago
  maybe there was a slight delay getting them in this count
  - jMyles 7y ago
    It's not (or at least, not only) a delay, but a change in methodology of classification. Previously, all patients who tested positive for SARS-CoV-2 were counted, regardless of their cause of death. In the new methodology (in order to ensure that false negatives aren't missed), people with a pattern of symptoms are also counted, regardless of their cause of death. So, many, many more people are counted than those who died directly from complications arising from Covid-19. This is not altogether unreasonable; we want to know how many people are dying after contracting the virus. However, it is being reported as though the virus is the cause of death, when nobody doing the reporting is claiming that it is. It's a total mess. Here's the actual data sheet: https://www1.nyc.gov/assets/doh/downloads/pdf/imm/covid-19-daily-data-summary-deaths-04162020-1.pdf https://www1.nyc.gov/assets/doh/downloads/pdf/imm/covid-19-d...
    - sergers 7y ago
      Are they including the people who die at home? They weren't including these results previously as they were overwhelmed with testing the living, so weren't testing suspected deaths at home

vanniv 7y ago

This is a garbage number. New York reclassified a bunch of folks that died over the last few days as COVID deaths under their new methodology, where all deaths are covid deaths if there is even a hint that they might have had COVID. These are not deaths that occurred in a 24-hour period. These are deaths that were reported in a 24-hour period, and covers basically everyone with any respiratory illness who died at home on Tuesday, Wednesday, or Thursday. Trying to compare that to a rate of deaths in hospital of likely or confirmed cases (the methodology used from last week until Tuesday), or hospital deaths of confirmed cases (the methodology used before last week) is meaningless. If you change methodology, you really have to go back and recast historic numbers. But that isn't being done.

jMyles 7y ago
...what an asinine headline. This is open letter worthy from reputable epidemiologists. It's been hard enough to identify and asses data (and expert commentary) without unambiguously deceptive (and panic-riddled) reporting like this.
- LB232323 7y ago
  The profitable manufacture of fear and justification of government overreach has been a reliable product and goal of sensationalist journalism for a very long time. The Spanish-American War is a prime example. If anything, this crisis has drawn visible lines between those with unwavering trust in mass media and those without. Those who have good reasons to distrust authority have been the least likely to comply with orders and recommendations from both government agencies and mass media outlets. On the other hand, those who benefit from the current power structure have taken to their advice with great enthusiasm. Between these groups are those like yourself who try to sift through the sensationalism for good information. Even then, international authorities like WHO have revealed clear political bias in regards to issues like Taiwanese independence. Who can we trust?
  - jMyles 7y ago
    > The Spanish-American War is a prime example. Where is our John Quincy Adams?

dorkwood 7y ago

I think this is because NY is now reporting probable deaths, and not just those who tested positive before dying.

cylinder 7y ago
Dr Birx said the whole country is doing this, and she seems totally fine with it.

maxroham1977 7y ago

That's a wrong number for daily deaths. It factors in previous deaths counts that weren't counted (mainly in NY). Those numbers should have been backdated. Currently it seems like the daily death toll is rising, although it doesn't (at least not so sharply).

aaron695 7y ago
This is not what the media is reporting, they say it's in a 24 hour time period. Do you have a link?

neonate 7y ago

https://archive.md/tD5QU https://archive.md/tD5QU

jb775 7y ago

These numbers need to be seriously scrutinized for everyone's sake. Hospitals are now financially incentivized to claim as many Coronavirus hospitalizations/deaths as possible so they get their piece of the stimulus package.

draw_down 7y ago
Politicians are incentivized to lie and dissemble in the other direction. Pick your poison.
NotSammyHagar 7y ago
That sounds like a conspiratoral worry. Hospitals are all lying? We should just publish overall death rates long with traditional death rates and known cv cases. it's guaranteed there are missed cv tests. In Seattle they have not had enough cv tests to 'bother' testing the sad cases of large numbers of people dying, even in the famous kirkland senior citizen homes.

beezle 7y ago

This was brought up during the daily Cuomo yesterday. The media have not fully explained the new reporting (am I cynical if I say so they get better clicks?) CDC mandated that all localities, not just NYC/NYS, count probable deaths in addition to those where a covid positive test was available. These figures are not part of the 'official' covid death totals but instead are a separate category. The reason this has been done is likely due to the reports coming from NYFD/PD concerning the much higher incidence of finding dead people when called. Those bodies are not being brought to hospitals for autopsie or anything else if there are no suspicious circumstances. By one report I read a few days back, a year ago it was typical to have encounter 25 dead on 911 calls over a 24 hour period. They are now seeing upwards of 200.

asabjorn 7y ago

It is worthwhile to notice that there are as of April 13 there are 11k weekly coronavirus related deaths according to the CDC out of 600k tested [1]. In a normal flu season there are 8k daily flu deaths [2]. The ones that are the worst affected should be in the hospital. The cumulative rate of hospitalization is 12.3 per 100k, which adds up to 39k. So that puts an upper bound on known risk. [3] [1] https://www.cdc.gov/nchs/nvss/vsrr/COVID19/index.htm https://www.cdc.gov/nchs/nvss/vsrr/COVID19/index.htm [2] https://www.cdc.gov/mmwr/volumes/68/wr/mm6826a5.htm https://www.cdc.gov/mmwr/volumes/68/wr/mm6826a5.htm [3] https://gis.cdc.gov/grasp/COVIDNet/COVID19_3.html?fbclid=IwAR02UGPEV4r5LicjyNum1mpmFixTpCPYHJY8Nf3d8QV1iKrKK6ZlGTecifs https://gis.cdc.gov/grasp/COVIDNet/COVID19_3.html?fbclid=IwA...