search

CSSEGISandData / COVID-19

Novel Coronavirus (COVID-19) Cases, provided by JHU CSSE
https://systems.jhu.edu/research/public-health/ncov/
29.12k stars 18.42k forks source link

Data for Pneumonia or Hospitalization #1247

Open ghost opened 4 years ago

ghost commented 4 years ago

Is there any data relative to the severity of the cases. Not sure of the measurement but maybe how many cases by location and age had pneumonia. Maybe it is easier to count the number of people hospitalized?

Ultimately I would like to make a histogram by age of the number of cases the are severe/critical. Would also like to segment out by location the severity of cases.

I think it is not clear how many people are getting severely sick as a percentage of the number infected.

desaian commented 4 years ago

The problem is that the US Healthcare system is too fragmented and has many information silos. This makes it hard to get hospitalization data which is probably more important than case number and deaths. Without a hospitalization number, we can't gauge the need for masks, other PPE and ventilators. The Army needs to take over the hospitals, the data infrastructure, and physical supply chains. This is my perspective as a physician.

VMBoehm commented 4 years ago

The ECDC provides the number of hospitalizations as meta data, but they don't seem to make it available for download (maybe because they know its incomplete). I find it weird though that the number of hospitalizations should be more incomplete than the number of positive tests. Why does one get reported more reliably to public health institutes than the other? Does anyone know more? Could the ECDC be asked to make it available?

desaian commented 4 years ago

I wasn't looking at the European data. But in the US, ICD-10 hospitalization data is held as a trade secret by insurers and Hospitals/Healthcare systems. Certain things are reportable at the county level. Getting reliable county data can be a problem. Anup

Sent from Yahoo Mail on Android

On Tue, Mar 24, 2020 at 1:38 PM, Vanessa Böhmnotifications@github.com wrote:

The ECDC provides the number of hospitalizations as meta data, but they don't seem to make it available for download (maybe because they know its incomplete). I find it weird though that the number of hospitalizations should be more incomplete than the number of positive tests. Why does one get reported more reliably to public health institutes than the other? Does anyone know more? Could the ECDC be asked to make it available?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

ghost commented 4 years ago

Is it truly a trade secret? Seems like the government does not want us to know. Frankly, it doesn't seem too hard to count how many people are hospitalized.......They seem to be able to charge you for everything during a stay.

desaian commented 4 years ago

Either the end consumer (i.e. patient) or Healthcare system / insurance company can provide the information. Maybe an app that aggregates data straight from the end consumer and their family members would be a solution? https://www.advisory.com/daily-briefing/2019/02/12/data-rules

On Tue, Mar 24, 2020 at 6:03 PM, MNagle10notifications@github.com wrote:

Is it truly a trade secret? Seems like the government does not want us to know. Frankly, it doesn't seem too hard to count how many people are hospitalized.......They seem to be able to charge you for everything during a stay.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

VMBoehm commented 4 years ago

having an app to collect this data sounds like a great idea! The best data would probably come from hospitals or counties - information from patients might be biased because certain age groups are more likely to use an app. Would anyone be interested in looking into this? I have not much experience with app design, but I know a lot of people are looking for hospitalization data at this point. Data protection might be an issue here, too...

desaian commented 4 years ago

Read this and send it to all your friends: https://drive.google.com/file/d/1R3NSRHnMPPKLjWQLWIYOCUjhiDDDE2r9/view?usp=drivesdk

Sent from Yahoo Mail on Android

On Wed, Mar 25, 2020 at 4:17 PM, Vanessa Böhmnotifications@github.com wrote:

having an app to collect this data sounds like a great idea! The best data would probably come from hospitals or counties - information from patients might be biased because certain age groups are more likely to use an app. Would anyone be interested in looking into this? I have not much experience with app design, but I know a lot of people are looking for hospitalization data at this point. Data protection might be an issue here, too...

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

desaian commented 4 years ago

This is why we can't count on the counties and the hospitals to share data.

Sent from Yahoo Mail on Android

On Wed, Mar 25, 2020 at 4:17 PM, Vanessa Böhmnotifications@github.com wrote:

having an app to collect this data sounds like a great idea! The best data would probably come from hospitals or counties - information from patients might be biased because certain age groups are more likely to use an app. Would anyone be interested in looking into this? I have not much experience with app design, but I know a lot of people are looking for hospitalization data at this point. Data protection might be an issue here, too...

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

VMBoehm commented 4 years ago

well, there are other countries where municipalities and counties can share this data

desaian commented 4 years ago

Oh no! We've actually gotten to this point. Anup Desai (@Desai2Anup) Tweeted:This is what you get when you have a third world healthcare system in a first world country. I hope the DNC and Biden are taking note. https://t.co/xjdGrf7j9v https://twitter.com/Desai2Anup/status/1242975180575535104?s=20 Sent from Yahoo Mail on Android

On Wed, Mar 25, 2020 at 5:40 PM, Vanessa Böhmnotifications@github.com wrote:

well, there are other countries where municipalities and counties can share this data

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

VMBoehm commented 4 years ago

lol. I'm just saying that such an app could be used worldwide - if some entities in the US could upload data that would be great. if not, it might still be useful for other countries. I'll halt here. I don't want to start any arguments, I just want realiable data to analyze :)

desaian commented 4 years ago

I think if we go to patients and their families... That would be a more reliable source than the government and the Healthcare system. make something reportable directly to the app buy a unique identifier so that we don't have redundancies in the data. And make sure they include some indicator of disease severity. For example, do they have oxygen or are they ventilated.

Sent from Yahoo Mail on Android

On Wed, Mar 25, 2020 at 5:54 PM, Vanessa Böhmnotifications@github.com wrote:

lol. I'm just saying that such an app could be used worldwide - if some entities in the US could upload data that would be great. if not, it might still be useful for other countries. I'll halt here. I don't want to start any arguments, I just want realiable data to analyze :)

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.