4 ms·

Silicon Valley Live Hospital Bed Data

50 points by sahin-boydas 7y ago

ex3ndr 7y ago
Wow so good
GordonS 7y ago
I was confused about where the actual data was... then after several seconds the charts began to appear. I used PowerBI in one project before, with charts embedded in a web page, and this was my experience too - they just took forever to load! On top of that, the documentation around embedding charts was pretty poor, and the licensing model was very confusing - I don't think we ever did really understand it!
- ampdepolymerase 7y ago
  With the new .NET Blazor stack I think this will be the new normal. All rendering will be streamed on-demand through WebSockets. Great for real-time stuff but takes some optimization for initial page load performance.
  - GordonS 7y ago
    Not sure what you mean; when I used embedded PowerBI (~2 years ago, I think), it didn't use WebSockets.
bb123 7y ago
This is interesting. It would appear Santa Clara has quite a bit more capacity for when things get worse.
- redis_mlc 7y ago
  No, the numbers are very bad. There's approx. 860 acute and ICU beds empty (from the text chart) for Santa Clara County, which has about 3-4 million people. So that will be full in about 1-2 weeks. In addition, the California emergency stockpiles were neglected, so all that material rotted. But it gets worse. We're in lockdown to "flatten the curve" on how many ventilators are needed. But it turns out 2/3 of corona ventilator patients die in 2-3 weeks, and the rest are too ill to leave the hospital. So it calls into question why we are in lockdown if ventilators are in fact irrelevant. That needs to be re-evaluated from a public health policy. It looks to me like corona should be treated like a regular flu, and the only lockdown should be for plane flights. Ventilators Are No Panacea For Critically Ill COVID-19 Patients https://www.npr.org/sections/health-shots/2020/04/02/826105278/ventilators-are-no-panacea-for-critically-ill-covid-19-patients https://www.npr.org/sections/health-shots/2020/04/02/8261052...
  - berdario 7y ago
    That still means that (after a few grueling weeks) 1/3rd of the patients could survive. Moreover there's all the non-covid19 patients that could also survive if they wouldn't find the ICU fulls. That's not irrelevant. Btw, you're right that we're not capitalizing on the lockdown as much as we could: we're supposed to build out the testing, contract tracing and early detection infrastructure (temperature checks at every office and grocery shops... Italy just started) Shutting down plane flights is actually not that useful: we should just do what South Korea is doing and test every single arrival, and then we could keep that infrastructure in place (obviously, if the number of flights make it untenable to test every arrival, that should be heavily capped)
    - redis_mlc 7y ago
      No, 1/3 are not expected to survive. If you read the article, they're expecting 90% to either die or have organ damage. So out of 50 patients, they's expecting 2 patients to walk out of the hospital with some kind of normal life. Ventlators don't appear to tbe the answer beyond, "Well, we tried to do something. Too bad they don't work." The additional problem is that using ineffective ventilators kills your hospital staff for little gain. In SARs-1 (2002) in Toronto, more hospital staff got infected or died than in the general population. Hospital ventilation systems circulate airborne diseases. Now that we have data that ventilators don't work, we need to look at what the next steps are. There's no point in destroying the economy waiting for ventilators when they're not a treatment for corona virus.
      - yodon 7y ago
        Your analysis that ventilators don't work may be valid. Your conclusion that therefor trying to flatten the curve is useless is dangerously invalid. Most people who come down with Covid-19 don't require ventilators. Flattening the curve buys time for all measures, including development of other treatment approaches.
        redis_mlc 7y ago
_1tan 7y ago
A similar overview from Germany: https://www.intensivregister.de/#/intensivregister https://www.intensivregister.de/#/intensivregister Sadly the data doesn't seem to be readily accessible.
awaythrower 7y ago
Ignore the rtPCR test result data because SARS-CoV-2 still isn't a mandatory report condition (which is ridiculous) and so private labs don't report positive (or negative) results to any government regulators. The majority of tests are currently being conducted by private labs for people with insurance and only mostly critical cases are being tested, so basically no one infected is included in the statistics. The only ones being tested by the county are typically homeless, undocumented or other people who show up critically-ill to county hospitals. Santa Clara county is about 2-3 weeks behind NY, which means the peak there will be roughly in 4-6 weeks. It's going to be a major disaster and emergency healthcare will be essentially unusable until mid June.
- veggieburglar 7y ago
  Do you mean 2-3 weeks behind in testing capability? Santa Clara County was the first in the nation to go under SIP protocols officially.
  - zaidf 7y ago
    Yeah the narrative that it will peak in the Bay Area in June makes very little sense to me. Even ppl on the front lines are saying otherwise: https://twitter.com/Bob_Wachter/status/1245536549757841409?s=20 https://twitter.com/Bob_Wachter/status/1245536549757841409?s...

50 points by sahin-boydas 7y ago

Wow so good

I was confused about where the actual data was... then after several seconds the charts began to appear. I used PowerBI in one project before, with charts embedded in a web page, and this was my experience too - they just took forever to load! On top of that, the documentation around embedding charts was pretty poor, and the licensing model was very confusing - I don't think we ever did really understand it!

ampdepolymerase 7y ago
With the new .NET Blazor stack I think this will be the new normal. All rendering will be streamed on-demand through WebSockets. Great for real-time stuff but takes some optimization for initial page load performance.
- GordonS 7y ago
  Not sure what you mean; when I used embedded PowerBI (~2 years ago, I think), it didn't use WebSockets.

bb123 7y ago

This is interesting. It would appear Santa Clara has quite a bit more capacity for when things get worse.

redis_mlc 7y ago
No, the numbers are very bad. There's approx. 860 acute and ICU beds empty (from the text chart) for Santa Clara County, which has about 3-4 million people. So that will be full in about 1-2 weeks. In addition, the California emergency stockpiles were neglected, so all that material rotted. But it gets worse. We're in lockdown to "flatten the curve" on how many ventilators are needed. But it turns out 2/3 of corona ventilator patients die in 2-3 weeks, and the rest are too ill to leave the hospital. So it calls into question why we are in lockdown if ventilators are in fact irrelevant. That needs to be re-evaluated from a public health policy. It looks to me like corona should be treated like a regular flu, and the only lockdown should be for plane flights. Ventilators Are No Panacea For Critically Ill COVID-19 Patients https://www.npr.org/sections/health-shots/2020/04/02/826105278/ventilators-are-no-panacea-for-critically-ill-covid-19-patients https://www.npr.org/sections/health-shots/2020/04/02/8261052...
- berdario 7y ago
  That still means that (after a few grueling weeks) 1/3rd of the patients could survive. Moreover there's all the non-covid19 patients that could also survive if they wouldn't find the ICU fulls. That's not irrelevant. Btw, you're right that we're not capitalizing on the lockdown as much as we could: we're supposed to build out the testing, contract tracing and early detection infrastructure (temperature checks at every office and grocery shops... Italy just started) Shutting down plane flights is actually not that useful: we should just do what South Korea is doing and test every single arrival, and then we could keep that infrastructure in place (obviously, if the number of flights make it untenable to test every arrival, that should be heavily capped)
  - redis_mlc 7y ago
    No, 1/3 are not expected to survive. If you read the article, they're expecting 90% to either die or have organ damage. So out of 50 patients, they's expecting 2 patients to walk out of the hospital with some kind of normal life. Ventlators don't appear to tbe the answer beyond, "Well, we tried to do something. Too bad they don't work." The additional problem is that using ineffective ventilators kills your hospital staff for little gain. In SARs-1 (2002) in Toronto, more hospital staff got infected or died than in the general population. Hospital ventilation systems circulate airborne diseases. Now that we have data that ventilators don't work, we need to look at what the next steps are. There's no point in destroying the economy waiting for ventilators when they're not a treatment for corona virus.
    - yodon 7y ago
      Your analysis that ventilators don't work may be valid. Your conclusion that therefor trying to flatten the curve is useless is dangerously invalid. Most people who come down with Covid-19 don't require ventilators. Flattening the curve buys time for all measures, including development of other treatment approaches.
      - redis_mlc 7y ago

_1tan 7y ago

A similar overview from Germany: https://www.intensivregister.de/#/intensivregister https://www.intensivregister.de/#/intensivregister Sadly the data doesn't seem to be readily accessible.

awaythrower 7y ago

Ignore the rtPCR test result data because SARS-CoV-2 still isn't a mandatory report condition (which is ridiculous) and so private labs don't report positive (or negative) results to any government regulators. The majority of tests are currently being conducted by private labs for people with insurance and only mostly critical cases are being tested, so basically no one infected is included in the statistics. The only ones being tested by the county are typically homeless, undocumented or other people who show up critically-ill to county hospitals. Santa Clara county is about 2-3 weeks behind NY, which means the peak there will be roughly in 4-6 weeks. It's going to be a major disaster and emergency healthcare will be essentially unusable until mid June.

veggieburglar 7y ago
Do you mean 2-3 weeks behind in testing capability? Santa Clara County was the first in the nation to go under SIP protocols officially.
- zaidf 7y ago
  Yeah the narrative that it will peak in the Bay Area in June makes very little sense to me. Even ppl on the front lines are saying otherwise: https://twitter.com/Bob_Wachter/status/1245536549757841409?s=20 https://twitter.com/Bob_Wachter/status/1245536549757841409?s...