[murky]

March 30, 2020 Update: The 202 new confirmed cases reported today include some results that were not previously reported over the past two days. This increase reflects a reporting delay 

So why don't they just correct the prior days' totals to include the newly-reported results from those days? The legend of the graph says that the date of cases is the date of COVID-19 confirmation. Now that is no longer true, and it detracts from the utility of the graph.

Would be nice if they could report by the day the swab was taken.

[murky]

March 30, 2020 Update: The 202 new confirmed cases reported today include some results that were not previously reported over the past two days. This increase reflects a reporting delay 

So why don't they just correct the prior days' totals to include the newly-reported results from those days? The legend of the graph says that the date of cases is the date of COVID-19 confirmation. Now that is no longer true, and it detracts from the utility of the graph.

Would be nice if they could report by the day the swab was taken.

[oldalum]

Would be nice if they could report by the day the swab was taken.

Yes! And even better if they could also mandate reporting to them, and report to the public, the number of negative tests as well.

[oldalum]

Would be nice if they could report by the day the swab was taken.

Yes! And even better if they could also mandate reporting to them, and report to the public, the number of negative tests as well.

[oldalum]

SCC: 42 new cases recorded yesterday. Impossible to look for a trend with the problems in the data, but it sure beats 3 figures!

[oldalum]

SCC: 42 new cases recorded yesterday. Impossible to look for a trend with the problems in the data, but it sure beats 3 figures!

[M T]

Would be nice if they could report by the day the swab was taken.

Yes! And even better if they could also mandate reporting to them, and report to the public, the number of negative tests as well.

That reporting to the health department is required, within an hour, as of March 24.  Actual order here.  The date the specimen was taken is also reported.

The problem is that the health departments are swamped.  I wish they gave more, useful data so we armchair researchers could be better informed about risks to us in the community.   However, I acknowledge that they are swamped.  I don't care for pretty web pages, I want more people to get better medical attention asap.  

I expect they have all the data they can handle, and see the better view than we hope for.   The question is, with their budget & manpower, what is the best for them to do?  Even if you went to them and said "For free, let me build you a better dashboard, with hospitalizations, reported dates, sample dates, outcomes, ages, etc.", that would cost them valuable time they don't have.

Knowing they have the data, I am concerned that today's stricter order was in response to less than optimal infection spread rates.

Ian Larkin, whose interview I linked to here, said that China only got the rate down to 1.3 or so by isolation.  They had to separate known infected individuals (this would include mild cases) from possibly infected individuals (ie, their households) in order to get it down to 0.7.   But that means a  5x greater number of beds for those people, plus a large number of staff to attend to them, plus a large number of PPE & masks.

[M T]

Would be nice if they could report by the day the swab was taken.

Yes! And even better if they could also mandate reporting to them, and report to the public, the number of negative tests as well.

That reporting to the health department is required, within an hour, as of March 24.  Actual order here.  The date the specimen was taken is also reported.

The problem is that the health departments are swamped.  I wish they gave more, useful data so we armchair researchers could be better informed about risks to us in the community.   However, I acknowledge that they are swamped.  I don't care for pretty web pages, I want more people to get better medical attention asap.  

I expect they have all the data they can handle, and see the better view than we hope for.   The question is, with their budget & manpower, what is the best for them to do?  Even if you went to them and said "For free, let me build you a better dashboard, with hospitalizations, reported dates, sample dates, outcomes, ages, etc.", that would cost them valuable time they don't have.

Knowing they have the data, I am concerned that today's stricter order was in response to less than optimal infection spread rates.

Ian Larkin, whose interview I linked to here, said that China only got the rate down to 1.3 or so by isolation.  They had to separate known infected individuals (this would include mild cases) from possibly infected individuals (ie, their households) in order to get it down to 0.7.   But that means a  5x greater number of beds for those people, plus a large number of staff to attend to them, plus a large number of PPE & masks.

[Farm93]

SCC: 42 new cases recorded yesterday. Impossible to look for a trend with the problems in the data, but it sure beats 3 figures!

CA now 3rd in cases by state.   Yes, CA is not testing everyone, but the only NY has really been aggressively testing.

CA is now 6th in deaths.   Given the size of the state and the idea that CA was relatively early to see community spread, 6th is really good news...on a relative basis.

Loads of states with smaller populations are zooming right by CA now.  

Korea, Japan and Singapore will show the value testing and tracing.
Germany will show the value of diligent preparation when a touch late to testing.
Italy and Spain will show the impact of taking big measures after cases were out of control.
Many states in the USA will show the impact of relying on social distancing for several weeks.

Several years from now historians will get to look at the different responses to see which approaches in democracies were best for public health and, if different, which approaches were best for economic growth.

For now - Public health has to be prioritized as at this early stage it seems the longer you wait the more that will die and the more you spend.

[Farm93]

SCC: 42 new cases recorded yesterday. Impossible to look for a trend with the problems in the data, but it sure beats 3 figures!

CA now 3rd in cases by state.   Yes, CA is not testing everyone, but the only NY has really been aggressively testing.

CA is now 6th in deaths.   Given the size of the state and the idea that CA was relatively early to see community spread, 6th is really good news...on a relative basis.

Loads of states with smaller populations are zooming right by CA now.  

Korea, Japan and Singapore will show the value testing and tracing.
Germany will show the value of diligent preparation when a touch late to testing.
Italy and Spain will show the impact of taking big measures after cases were out of control.
Many states in the USA will show the impact of relying on social distancing for several weeks.

Several years from now historians will get to look at the different responses to see which approaches in democracies were best for public health and, if different, which approaches were best for economic growth.

For now - Public health has to be prioritized as at this early stage it seems the longer you wait the more that will die and the more you spend.

[M T]

66 new cases & 2 new deaths as  of 5PM March 31.
Still no hospitalization data, so the only hard data is the 32 deaths.

956 cases.
Naive CFR = 3.3% BUT the number of cases is presumed to be undercounted, while not everyone who is sick will survive.
If CFR = 2%, 32 deaths suggests 1600 cases with symptoms (and another roughly 1600 without symptoms).

Deaths:  (Percentages given of approximate number in each age group.  But these only poorly correspond to CFR.
0-20: 0  (/ ~29 = 0%)
21-30: 1  (/ 105 = 1%)
31-40: 0  (/ 182 = 0%)
41-50: 4  (/ 191 = 2.1%)
51-60: 4  (/ 182 = 2.2%)
61-70: 5  (/ 124 = 4%)
71-80: 9  (/ 86 = 10.5%)
81-90: 8  (/ ~38 = 21.1%)
90+: 1  (/ ~10 = 10%)
Unk: 0 (/ ~10)
F: 8
M: 24

[M T]

66 new cases & 2 new deaths as  of 5PM March 31.
Still no hospitalization data, so the only hard data is the 32 deaths.

956 cases.
Naive CFR = 3.3% BUT the number of cases is presumed to be undercounted, while not everyone who is sick will survive.
If CFR = 2%, 32 deaths suggests 1600 cases with symptoms (and another roughly 1600 without symptoms).

Deaths:  (Percentages given of approximate number in each age group.  But these only poorly correspond to CFR.
0-20: 0  (/ ~29 = 0%)
21-30: 1  (/ 105 = 1%)
31-40: 0  (/ 182 = 0%)
41-50: 4  (/ 191 = 2.1%)
51-60: 4  (/ 182 = 2.2%)
61-70: 5  (/ 124 = 4%)
71-80: 9  (/ 86 = 10.5%)
81-90: 8  (/ ~38 = 21.1%)
90+: 1  (/ ~10 = 10%)
Unk: 0 (/ ~10)
F: 8
M: 24

[oldalum]

from the Merc today:
"The governor said the state of California planned to release hospitalization numbers by county later in the day Wednesday."

So we should have data at least going forward.

[oldalum]

from the Merc today:
"The governor said the state of California planned to release hospitalization numbers by county later in the day Wednesday."

So we should have data at least going forward.

[M T]

The new dashboard gives us new numbers, but we have to be careful as the time of data is different for the different numbers.
Note that you can move your mouse over the graphs to get accurate numbers.
For instance on March 30, Covid-19 patients occupied 97 acute care beds and 75 ICU beds. (total 172)
On March 31, 108 acute care beds and 90 ICU beds.  (total 198)

Something weird though.... The last hospitalization numbers was for March 25, when there were 154 hospitalizations.
In retrospect, it isn't clear to me whether that was the total of all who were ever hospitalized, or current hospitalizations.

These numbers clearly indicate the number in the hospital now.
There weren't just 18 new hospitalizations between March 25 and March 30.  It means a net of +18 hospitalizations.

[M T]

The new dashboard gives us new numbers, but we have to be careful as the time of data is different for the different numbers.
Note that you can move your mouse over the graphs to get accurate numbers.
For instance on March 30, Covid-19 patients occupied 97 acute care beds and 75 ICU beds. (total 172)
On March 31, 108 acute care beds and 90 ICU beds.  (total 198)

Something weird though.... The last hospitalization numbers was for March 25, when there were 154 hospitalizations.
In retrospect, it isn't clear to me whether that was the total of all who were ever hospitalized, or current hospitalizations.

These numbers clearly indicate the number in the hospital now.
There weren't just 18 new hospitalizations between March 25 and March 30.  It means a net of +18 hospitalizations.

[Nan3cy]

Santa Clara County is providing a "dashboard" with useful graphs and stats:  SCC Dashboard

One thing I noticed from my male perspective: males comprise 53% of cases but 76% of deaths.

I wonder if body mass and efficiency has anything to do with it. Larger people need more oxygen to stay alive, and men tend to be larger than women. The only way to see if this hypothesis might be correct would be to see if weight and mortality rates are positively correlated or to compare survival/mortality rates of patients of similar weight, regardless of sex.

My son and his wife are both docs in Nassau County, NY. The hospital where my DIL works has (duh) a number of Covid-19 patients, and she says many/most of them are obese. (Both men and women.)

[Nan3cy]

Santa Clara County is providing a "dashboard" with useful graphs and stats:  SCC Dashboard

One thing I noticed from my male perspective: males comprise 53% of cases but 76% of deaths.

I wonder if body mass and efficiency has anything to do with it. Larger people need more oxygen to stay alive, and men tend to be larger than women. The only way to see if this hypothesis might be correct would be to see if weight and mortality rates are positively correlated or to compare survival/mortality rates of patients of similar weight, regardless of sex.

My son and his wife are both docs in Nassau County, NY. The hospital where my DIL works has (duh) a number of Covid-19 patients, and she says many/most of them are obese. (Both men and women.)

[oldalum]

today's SCC new cases are 54.  Since 3/30, the new cases numbers have been 42, 66, 63, 75, 54. That's an average of 60/day. The 5 days before that, the average was 78/day. I conclude that, over the last 10 days of numbers, the rate has not been increasing. I'd like to think it may have even dropped slightly, but the data are noisy and the cutpoint was arbitrary.

[oldalum]

today's SCC new cases are 54.  Since 3/30, the new cases numbers have been 42, 66, 63, 75, 54. That's an average of 60/day. The 5 days before that, the average was 78/day. I conclude that, over the last 10 days of numbers, the rate has not been increasing. I'd like to think it may have even dropped slightly, but the data are noisy and the cutpoint was arbitrary.

[magnus]

Are the number of tests steady or increasing?  If so, that's good news.

[magnus]

Are the number of tests steady or increasing?  If so, that's good news.

[Leftcoast]

The dashboard showing test counts in Santa Clara is the last of the three shown here:   

https://www.sccgov.org/sites/phd/Disease...board.aspx

Testing rates are steady but still very low-614 tests shown today out of 1.8M in Santa Clara county. Worse, only 9910 county residents have been tested, TOTAL, since the crisis began. The barrier threshold for testing must remain quite high.

[Leftcoast]

The dashboard showing test counts in Santa Clara is the last of the three shown here:   

https://www.sccgov.org/sites/phd/Disease...board.aspx

Testing rates are steady but still very low-614 tests shown today out of 1.8M in Santa Clara county. Worse, only 9910 county residents have been tested, TOTAL, since the crisis began. The barrier threshold for testing must remain quite high.

[M T]

today's SCC new cases are 54.  Since 3/30, the new cases numbers have been 42, 66, 63, 75, 54. That's an average of 60/day. The 5 days before that, the average was 78/day. I conclude that, over the last 10 days of numbers, the rate has not been increasing. I'd like to think it may have even dropped slightly, but the data are noisy and the cutpoint was arbitrary.

There are two different sets of numbers.  The very noisy ones that had the exceptional 202 daily number on the day just before your set of 5, which distorts things greatly.  (If you'd chosen 6 days, the average would have been 83.7 in the most recent set versus 54.2 in the previous set)

Given that we don't have total hospitalization numbers, the better numbers appear to be the positive test results, by the day the results were returned.  They add up to the same total, but don't have what may be delayed reporting form the SCC PHD.  These numbers show a basically flat trajectory for the last week.  Indeed for longer than that.  Why?  This may be a testing capacity issue.

(March 20) 51, 44, 48, 49, (March 24 SiP+incubation) 62, 70, 51, 71, 58, 47, 47, 71, (April 1) 56

The best I can hope for from that data is that the growth factor is approximately 1.0.  While this state is very much better than NYC, at this rate, we'll be in this situation forever.  We want to get the numbers down faster somehow.

---

I had missed this paper when it came out.
A Genomic Survey of SARS-CoV-2 Reveals Multiple Introductions into Northern California without a Predominant Lineage

Basically, it examines cases in northern California and decides that there have been a number of different introductions of the disease into the area.

Phylogenetic analyses revealed at least 8 different SARS-CoV-2 lineages, suggesting multiple independent introductions of the virus into the state. ...
In Santa Clara County, cases with onset dates within two weeks of each other were associated with a large facility with multiple employers, large areas of shared space, and heavy pedestrian traffic. ...
Consistent with this epidemiological data, the genomes (UC13-15) from the three workers in the large facility are distinct from each other.

[M T]

today's SCC new cases are 54.  Since 3/30, the new cases numbers have been 42, 66, 63, 75, 54. That's an average of 60/day. The 5 days before that, the average was 78/day. I conclude that, over the last 10 days of numbers, the rate has not been increasing. I'd like to think it may have even dropped slightly, but the data are noisy and the cutpoint was arbitrary.

There are two different sets of numbers.  The very noisy ones that had the exceptional 202 daily number on the day just before your set of 5, which distorts things greatly.  (If you'd chosen 6 days, the average would have been 83.7 in the most recent set versus 54.2 in the previous set)

Given that we don't have total hospitalization numbers, the better numbers appear to be the positive test results, by the day the results were returned.  They add up to the same total, but don't have what may be delayed reporting form the SCC PHD.  These numbers show a basically flat trajectory for the last week.  Indeed for longer than that.  Why?  This may be a testing capacity issue.

(March 20) 51, 44, 48, 49, (March 24 SiP+incubation) 62, 70, 51, 71, 58, 47, 47, 71, (April 1) 56

The best I can hope for from that data is that the growth factor is approximately 1.0.  While this state is very much better than NYC, at this rate, we'll be in this situation forever.  We want to get the numbers down faster somehow.

---

I had missed this paper when it came out.
A Genomic Survey of SARS-CoV-2 Reveals Multiple Introductions into Northern California without a Predominant Lineage

Basically, it examines cases in northern California and decides that there have been a number of different introductions of the disease into the area.

Phylogenetic analyses revealed at least 8 different SARS-CoV-2 lineages, suggesting multiple independent introductions of the virus into the state. ...
In Santa Clara County, cases with onset dates within two weeks of each other were associated with a large facility with multiple employers, large areas of shared space, and heavy pedestrian traffic. ...
Consistent with this epidemiological data, the genomes (UC13-15) from the three workers in the large facility are distinct from each other.

[oldalum]

The very noisy ones that had the exceptional 202 daily number on the day just before your set of 5, which distorts things greatly.  (If you'd chosen 6 days, the average would have been 83.7 in the most recent set versus 54.2 in the previous set)

I deliberately did not choose 6 days because it would have been wrong: the county made clear that the 202 on 3/30 was not real; it included results from previous days that had not been reported when they should have been (they probably did not get any reporting from some sites on 1 or more preceding days). So I averaged the 202 with the 4 preceding days; to average the 202 with any subsequent days would have been inappropriate because the average would have transferred some of the new cases from before 3/30 to after 3/30 where they did not belong. The "distortion" caused by the extreme 3/30 number is avoided by averaging it with the preceding days.

What I meant by arbitrary is that my choice of a 5-day average was not random, it was dictated by that reporting glitch.

[oldalum]

The very noisy ones that had the exceptional 202 daily number on the day just before your set of 5, which distorts things greatly.  (If you'd chosen 6 days, the average would have been 83.7 in the most recent set versus 54.2 in the previous set)

I deliberately did not choose 6 days because it would have been wrong: the county made clear that the 202 on 3/30 was not real; it included results from previous days that had not been reported when they should have been (they probably did not get any reporting from some sites on 1 or more preceding days). So I averaged the 202 with the 4 preceding days; to average the 202 with any subsequent days would have been inappropriate because the average would have transferred some of the new cases from before 3/30 to after 3/30 where they did not belong. The "distortion" caused by the extreme 3/30 number is avoided by averaging it with the preceding days.

What I meant by arbitrary is that my choice of a 5-day average was not random, it was dictated by that reporting glitch.

[cardcrimson]

[quote="oldalum" pid='275393' dateline='1586041160']
Given that we don't have total hospitalization numbers, the better numbers appear to be the positive test results, by the day the results were returned.  

Posted without comment:
https://www.sccgov.org/sites/phd/Disease...x#hospital

[cardcrimson]

[quote="oldalum" pid='275393' dateline='1586041160']
Given that we don't have total hospitalization numbers, the better numbers appear to be the positive test results, by the day the results were returned.  

Posted without comment:
https://www.sccgov.org/sites/phd/Disease...x#hospital

[M T]

[quote="oldalum" pid='275393' dateline='1586041160']
Given that we don't have total hospitalization numbers, the better numbers appear to be the positive test results, by the day the results were returned.  

Posted without comment:
https://www.sccgov.org/sites/phd/Disease...x#hospital

https://thecardboard.org/board/showthrea...#pid275167

[M T]

[quote="oldalum" pid='275393' dateline='1586041160']
Given that we don't have total hospitalization numbers, the better numbers appear to be the positive test results, by the day the results were returned.  

Posted without comment:
https://www.sccgov.org/sites/phd/Disease...x#hospital

https://thecardboard.org/board/showthrea...#pid275167

[M T]

The relation between the daily new cases from the SCC site and the testing results is becoming clearer.  The testing results are dynamic.  If they have a case for which the results came in 3 days ago, the results are added to today's new cases, and added to the testing results of 3 days ago.

I continue to assert the Testing stats are the best stats we have for watching the situation.  (Total hospitalizations (not daily) would also be good, but we don't have them.)   Earlier today, Dr. Cody indicated there were still problems in getting tests, so the testing numbers have to be viewed carefully and skeptically.

I have the Santa Clara County dashboard up in 2 screens, one from 2 days ago (partial day results for Apr. 2) and one from today (partial day results for Apr. 4).  I will give these numbers as Positive Results/Negative Results/Pending. 
The first set of numbers is April 2; the 2nd set is April 4;  the third set are the deltas between the two

March 24  62/643/1  60/642/1  -2/1/0
March 25 70/679/1  70/679/1   0/0/0
March 26 51/525/1  51/526/0    0/1/-1
March 27 71/593/0  75/650/1    4/57/1
March 28 58/553/0  69/677/0    11/124/0
March 29  47/567/1  52/642/1    5/75/0
March 30  47/434/0  50/478/0    3/44/0
March 31  71/529/1  74/561/1    3/32/0
April 1      56/614/2  60/657/1    4/43/-1
April 2       9/175/0   52/452/0   43/277/0

So, the increases shown in the daily new case numbers is apparently just a reflection of the PHD processing the reports it has been given over the last week or so.    The positive tests are updated as they get processed, using the date when they came in, but not in a FIFO order.

That said, the positive test numbers since March 24 (7 days after SiP order):
60, 70, 51, 75, 69, 52, 50, 74, 60, 52, 31

Caveat:  The number of tests has dropped as well.  Only 202 tests on that last day of which 31 were positive.
That's less than half as many tests as the day before, and 1/3 of the number two days before.
That's a high percentage positive/total tests relative to the two days before it.

It is NOT clear there is a downward trend.  It looks like it, but it could be an artifact of fewer tests available.

We can hope...

[M T]

The relation between the daily new cases from the SCC site and the testing results is becoming clearer.  The testing results are dynamic.  If they have a case for which the results came in 3 days ago, the results are added to today's new cases, and added to the testing results of 3 days ago.

I continue to assert the Testing stats are the best stats we have for watching the situation.  (Total hospitalizations (not daily) would also be good, but we don't have them.)   Earlier today, Dr. Cody indicated there were still problems in getting tests, so the testing numbers have to be viewed carefully and skeptically.

I have the Santa Clara County dashboard up in 2 screens, one from 2 days ago (partial day results for Apr. 2) and one from today (partial day results for Apr. 4).  I will give these numbers as Positive Results/Negative Results/Pending. 
The first set of numbers is April 2; the 2nd set is April 4;  the third set are the deltas between the two

March 24  62/643/1  60/642/1  -2/1/0
March 25 70/679/1  70/679/1   0/0/0
March 26 51/525/1  51/526/0    0/1/-1
March 27 71/593/0  75/650/1    4/57/1
March 28 58/553/0  69/677/0    11/124/0
March 29  47/567/1  52/642/1    5/75/0
March 30  47/434/0  50/478/0    3/44/0
March 31  71/529/1  74/561/1    3/32/0
April 1      56/614/2  60/657/1    4/43/-1
April 2       9/175/0   52/452/0   43/277/0

So, the increases shown in the daily new case numbers is apparently just a reflection of the PHD processing the reports it has been given over the last week or so.    The positive tests are updated as they get processed, using the date when they came in, but not in a FIFO order.

That said, the positive test numbers since March 24 (7 days after SiP order):
60, 70, 51, 75, 69, 52, 50, 74, 60, 52, 31

Caveat:  The number of tests has dropped as well.  Only 202 tests on that last day of which 31 were positive.
That's less than half as many tests as the day before, and 1/3 of the number two days before.
That's a high percentage positive/total tests relative to the two days before it.

It is NOT clear there is a downward trend.  It looks like it, but it could be an artifact of fewer tests available.

We can hope...

[magnus]

The other shining example of social distancing related measures,  WA is showing similar behavior.   New cases going down but testing also dropping too. 

I guess it's better than new cases going up while testing goes down.

[magnus]

The other shining example of social distancing related measures,  WA is showing similar behavior.   New cases going down but testing also dropping too. 

I guess it's better than new cases going up while testing goes down.

[Goose]

The other shining example of social distancing related measures,  WA is showing similar behavior.   New cases going down but testing also dropping too. 

I guess it's better than new cases going up while testing goes down.

Testing is probably going down because people that are candidates for testing are decreasing in number. The two should more or less track if that is the case.

[Goose]

The other shining example of social distancing related measures,  WA is showing similar behavior.   New cases going down but testing also dropping too. 

I guess it's better than new cases going up while testing goes down.

Testing is probably going down because people that are candidates for testing are decreasing in number. The two should more or less track if that is the case.

[magnus]

The other shining example of social distancing related measures,  WA is showing similar behavior.   New cases going down but testing also dropping too. 

I guess it's better than new cases going up while testing goes down.

Testing is probably going down because people that are candidates for testing are decreasing in number. The two should more or less track if that is the case.

That's a good point and i really hope that's the case and not that they've run out of supplies again.

[magnus]

The other shining example of social distancing related measures,  WA is showing similar behavior.   New cases going down but testing also dropping too. 

I guess it's better than new cases going up while testing goes down.

Testing is probably going down because people that are candidates for testing are decreasing in number. The two should more or less track if that is the case.

That's a good point and i really hope that's the case and not that they've run out of supplies again.