[BostonCard]

One serious problem with the argument that reopening doesn't save jobs because people stay home anyway (don't change their behavior) is that if people stay home anyway, "reopening" has no BAD effects either. Both things can't be true. What could be true is that reopening has more bad effects than good ones, but it isn't an  all or nothing proposition.

This is a good point, but it does treat the population as a homogeneous group that elects whether to do every activity based on either whether it is allowed or not or whether the pandemic is out of control or not (and to be fair, the original argument that I've made kind of makes that implicit assumption).  In reality, of course, different individuals will have different reactions to different activities based partly on what is allowed and partly on how widespread the virus is.  A good example of this is the recent thread that GR started about movie theaters.  I read that thread thinking, you can't pay me to go to a movie theater any time soon, and that's even while I acknowledge that under many guidelines (limited capacity to allow for social distancing), watching a two hour movie is a moderate risk (indoors, prolonged contact, but socially distant, with masks, and no talking).  On the other hand, GR, who is an avowed cinephile despite being in a higher risk group than I am, can't wait to go back to the theater to watch a movie.  So, yeah, reopening theaters will mean some increased commercial activity from people like GR, but nowhere near "normal" because of people like me.  The big question is whether the voluntary reductions in allowed activity (from people like me, who aren't going to the movie theater) is enough to limit spread.

BC

[BostonCard]

One serious problem with the argument that reopening doesn't save jobs because people stay home anyway (don't change their behavior) is that if people stay home anyway, "reopening" has no BAD effects either. Both things can't be true. What could be true is that reopening has more bad effects than good ones, but it isn't an  all or nothing proposition.

This is a good point, but it does treat the population as a homogeneous group that elects whether to do every activity based on either whether it is allowed or not or whether the pandemic is out of control or not (and to be fair, the original argument that I've made kind of makes that implicit assumption).  In reality, of course, different individuals will have different reactions to different activities based partly on what is allowed and partly on how widespread the virus is.  A good example of this is the recent thread that GR started about movie theaters.  I read that thread thinking, you can't pay me to go to a movie theater any time soon, and that's even while I acknowledge that under many guidelines (limited capacity to allow for social distancing), watching a two hour movie is a moderate risk (indoors, prolonged contact, but socially distant, with masks, and no talking).  On the other hand, GR, who is an avowed cinephile despite being in a higher risk group than I am, can't wait to go back to the theater to watch a movie.  So, yeah, reopening theaters will mean some increased commercial activity from people like GR, but nowhere near "normal" because of people like me.  The big question is whether the voluntary reductions in allowed activity (from people like me, who aren't going to the movie theater) is enough to limit spread.

BC

[teejers1]

P.S.  On all those SCC deaths you cite, can you tell me how many people under the age of 20, with no comorbidities, died from Covid?  How about under 30?  How about under 40?  50?  60?  70?  Why do we not have that information (when clearly the county does)?  You may think that's irrelevant; I do not.

If only SCC had a dashboard you could get that info from:
https://www.sccgov.org/sites/covid19/Pag...eaths.aspx

You can even download the underlying datasets:
https://data.sccgov.org/browse?category=COVID-19

1/363 (0.3%) was a death in an individual under 20.
4/363 (1.1%; 0.3% + 0.8%) occurred in individuals under 30.
11/363 (3%) of deaths occurred in individuals under 40.
25/363 (6.9%) of deaths occurred in individuals under 50.
58/363 (16%) of deaths occurred in inidviduals under 60.
119/363 (32.8%) of deaths occurred in patients under 70.

SCC doesn't do the joint probability of comorbidities and age, but 13% of patients were known not to have comorbidities (an additional 6.6% had an unknown comorbidity status).  Obviously comorbidities and age aren't independent so you can't just estimate them based on the joint probabilities, but a substantial majority of those deaths occurred in patients with comorbidities, so deaths in young patients without comorbidities are very rare indeed.

This pattern is consistent with data that has been reported elsewhere, which is to say that there is a strong association between age and comborbidities and risk of fatal COVID-19.

BC

C'mon, BC, you don't think I know about that dashboard?  That's why I asked why the joint numbers are not reported.  It's mathematically possible, is it not, that every single person who died without a comorbidity was over the age of 70?  (I could be wrong . . . was never a math guy.  [BTW, I don't expect that to be the case, but it would be interesting to know]).

And Dabig, as harsh as it sounds, yes, it does matter what the profile is of those perishing from a societal weighing of costs and benefits.  You may think that's harsh, cold, what have you - but there's a pretty strong argument that it is irresponsible for policymakers NOT to be considering the decedent profiles when making policies.  I love how people are so "caring" for decedents they do not know, but seemingly uncaring for the 200X as many people who lose jobs with all the ancillary drawbacks from that.  As for the rest of your short-form summary, that barely deserves a response.  But I will say this: with SCC's profile of cases and hospitalizations, the schools - especially north county - should be opening (at least try).  It's coming . . . slowly . . . and we'll see.  [BTW, if I were Dear Bureaucrat, I don't think I'd open indoor bars.  Youth soccer?  College soccer?  Absolutely they should be playing.  And really, most kids' stuff should be going forward. . .]

[teejers1]

P.S.  On all those SCC deaths you cite, can you tell me how many people under the age of 20, with no comorbidities, died from Covid?  How about under 30?  How about under 40?  50?  60?  70?  Why do we not have that information (when clearly the county does)?  You may think that's irrelevant; I do not.

If only SCC had a dashboard you could get that info from:
https://www.sccgov.org/sites/covid19/Pag...eaths.aspx

You can even download the underlying datasets:
https://data.sccgov.org/browse?category=COVID-19

1/363 (0.3%) was a death in an individual under 20.
4/363 (1.1%; 0.3% + 0.8%) occurred in individuals under 30.
11/363 (3%) of deaths occurred in individuals under 40.
25/363 (6.9%) of deaths occurred in individuals under 50.
58/363 (16%) of deaths occurred in inidviduals under 60.
119/363 (32.8%) of deaths occurred in patients under 70.

SCC doesn't do the joint probability of comorbidities and age, but 13% of patients were known not to have comorbidities (an additional 6.6% had an unknown comorbidity status).  Obviously comorbidities and age aren't independent so you can't just estimate them based on the joint probabilities, but a substantial majority of those deaths occurred in patients with comorbidities, so deaths in young patients without comorbidities are very rare indeed.

This pattern is consistent with data that has been reported elsewhere, which is to say that there is a strong association between age and comborbidities and risk of fatal COVID-19.

BC

C'mon, BC, you don't think I know about that dashboard?  That's why I asked why the joint numbers are not reported.  It's mathematically possible, is it not, that every single person who died without a comorbidity was over the age of 70?  (I could be wrong . . . was never a math guy.  [BTW, I don't expect that to be the case, but it would be interesting to know]).

And Dabig, as harsh as it sounds, yes, it does matter what the profile is of those perishing from a societal weighing of costs and benefits.  You may think that's harsh, cold, what have you - but there's a pretty strong argument that it is irresponsible for policymakers NOT to be considering the decedent profiles when making policies.  I love how people are so "caring" for decedents they do not know, but seemingly uncaring for the 200X as many people who lose jobs with all the ancillary drawbacks from that.  As for the rest of your short-form summary, that barely deserves a response.  But I will say this: with SCC's profile of cases and hospitalizations, the schools - especially north county - should be opening (at least try).  It's coming . . . slowly . . . and we'll see.  [BTW, if I were Dear Bureaucrat, I don't think I'd open indoor bars.  Youth soccer?  College soccer?  Absolutely they should be playing.  And really, most kids' stuff should be going forward. . .]

[BostonCard]

My point is that you are asking for a specific set of data that is clearly beyond the scope of what you can reasonably expect them to provide in an easy to use format.  Does it matter all that much if out of the 11 deaths in patients under 40, only 2 of them were in people without comorbidities or if it was 6?

BC

[BostonCard]

My point is that you are asking for a specific set of data that is clearly beyond the scope of what you can reasonably expect them to provide in an easy to use format.  Does it matter all that much if out of the 11 deaths in patients under 40, only 2 of them were in people without comorbidities or if it was 6?

BC

[teejers1]

My point is that you are asking for a specific set of data that is clearly beyond the scope of what you can reasonably expect them to provide in an easy to use format.  Does it matter all that much if out of the 11 deaths in patients under 40, only 2 of them were in people without comorbidities or if it was 6?

BC

Really?  I would not have thought so (how hard is it to merge data on deaths by age and comorbidities by age?).

And to your specific question, no.  But it might matter if it were 0 for everyone up through their 50s.

[teejers1]

My point is that you are asking for a specific set of data that is clearly beyond the scope of what you can reasonably expect them to provide in an easy to use format.  Does it matter all that much if out of the 11 deaths in patients under 40, only 2 of them were in people without comorbidities or if it was 6?

BC

Really?  I would not have thought so (how hard is it to merge data on deaths by age and comorbidities by age?).

And to your specific question, no.  But it might matter if it were 0 for everyone up through their 50s.

[BostonCard]

My point is that you are asking for a specific set of data that is clearly beyond the scope of what you can reasonably expect them to provide in an easy to use format.  Does it matter all that much if out of the 11 deaths in patients under 40, only 2 of them were in people without comorbidities or if it was 6?



BC

Really?  I would not have thought so (how hard is it to merge data on deaths by age and comorbidities by age?).



And to your specific question, no.  But it might matter if it were 0 for everyone up through their 50s.

Well sure.  But you want the joint analysis of age and comborbidities; someone else wants the joint analysis of race and sex; someone else wants to look at the interplay of date and city to see how the pandemic has moved around the county.  Just because SCC has the data and can do the analysis and you want them to doesn't obligate them to.

Personally, I think they should release the raw data so that people can do whatever analysis they want to do with it.  My guess is that there is some concern about the data enabling the identification of individuals, which, if you added comorbidities might enable someone to figure out someone's medical history.  I think this is a bit of an overstated concern, but then I've been surprised about how surprisingly easy it is to re-identify deidentified data.

BC

[BostonCard]

My point is that you are asking for a specific set of data that is clearly beyond the scope of what you can reasonably expect them to provide in an easy to use format.  Does it matter all that much if out of the 11 deaths in patients under 40, only 2 of them were in people without comorbidities or if it was 6?



BC

Really?  I would not have thought so (how hard is it to merge data on deaths by age and comorbidities by age?).



And to your specific question, no.  But it might matter if it were 0 for everyone up through their 50s.

Well sure.  But you want the joint analysis of age and comborbidities; someone else wants the joint analysis of race and sex; someone else wants to look at the interplay of date and city to see how the pandemic has moved around the county.  Just because SCC has the data and can do the analysis and you want them to doesn't obligate them to.

Personally, I think they should release the raw data so that people can do whatever analysis they want to do with it.  My guess is that there is some concern about the data enabling the identification of individuals, which, if you added comorbidities might enable someone to figure out someone's medical history.  I think this is a bit of an overstated concern, but then I've been surprised about how surprisingly easy it is to re-identify deidentified data.

BC

[Goose]

Well sure.  But you want the joint analysis of age and comborbidities; someone else wants the joint analysis of race and sex; someone else wants to look at the interplay of date and city to see how the pandemic has moved around the county.  Just because SCC has the data and can do the analysis and you want them to doesn't obligate them to.

FWIW, San Mateo County public health complained in the early summer that analysis of this type was not possible for them to do, blaming lack of personnel and lack of expertise. SMC said that the State, to whom all this data must be reported by law, normally did this type of analysis but has not done so. I would assume that SCC, although possibly better resourced, is in the same boat. No point in preparing to do analysis that the State will do for you, unless they don't.

[Goose]

Well sure.  But you want the joint analysis of age and comborbidities; someone else wants the joint analysis of race and sex; someone else wants to look at the interplay of date and city to see how the pandemic has moved around the county.  Just because SCC has the data and can do the analysis and you want them to doesn't obligate them to.

FWIW, San Mateo County public health complained in the early summer that analysis of this type was not possible for them to do, blaming lack of personnel and lack of expertise. SMC said that the State, to whom all this data must be reported by law, normally did this type of analysis but has not done so. I would assume that SCC, although possibly better resourced, is in the same boat. No point in preparing to do analysis that the State will do for you, unless they don't.

[BostonCard]

Anyone with basic stats training could do the analysis.  I think the more pressing issue is whether the juice was worth the squeeze, as they say.  Is the work necessary to produce it, which is not much, but is not nothing either, worth it just for Teejers to be happy.

BC

[BostonCard]

Anyone with basic stats training could do the analysis.  I think the more pressing issue is whether the juice was worth the squeeze, as they say.  Is the work necessary to produce it, which is not much, but is not nothing either, worth it just for Teejers to be happy.

BC

[teejers1]

Anyone with basic stats training could do the analysis.  I think the more pressing issue is whether the juice was worth the squeeze, as they say.  Is the work necessary to produce it, which is not much, but is not nothing either, worth it just for Teejers to be happy.

BC

If it's just Teejers, then obviously not.

But I'm surprised that anyone would find the info un-interesting.  
And you may not know if the juice is worth the squeeze until you see the results.

[teejers1]

Anyone with basic stats training could do the analysis.  I think the more pressing issue is whether the juice was worth the squeeze, as they say.  Is the work necessary to produce it, which is not much, but is not nothing either, worth it just for Teejers to be happy.

BC

If it's just Teejers, then obviously not.

But I'm surprised that anyone would find the info un-interesting.  
And you may not know if the juice is worth the squeeze until you see the results.

[Goose]

Anyone with basic stats training could do the analysis.  I think the more pressing issue is whether the juice was worth the squeeze, as they say.  Is the work necessary to produce it, which is not much, but is not nothing either, worth it just for Teejers to be happy.

BC

If it's just Teejers, then obviously not.

But I'm surprised that anyone would find the info un-interesting.  
And you may not know if the juice is worth the squeeze until you see the results.

Anyone with basic stats training could do the computation. Recognizing what (if anything) it means may be significantly harder. Recognizing that what Teejers wants is just "interesting" but some other set of data is actually very enlightening is harder still. This type of analysis needs to be done, and based on Dr. Morrow's comments, it isn't (at least several months ago). Granted, some academic institution may publish something really great eventually, or not. The State gets all the data. Nobody else does AFAIK. The State should at least be trying to tease out any meaning it may have.

[Goose]

Anyone with basic stats training could do the analysis.  I think the more pressing issue is whether the juice was worth the squeeze, as they say.  Is the work necessary to produce it, which is not much, but is not nothing either, worth it just for Teejers to be happy.

BC

If it's just Teejers, then obviously not.

But I'm surprised that anyone would find the info un-interesting.  
And you may not know if the juice is worth the squeeze until you see the results.

Anyone with basic stats training could do the computation. Recognizing what (if anything) it means may be significantly harder. Recognizing that what Teejers wants is just "interesting" but some other set of data is actually very enlightening is harder still. This type of analysis needs to be done, and based on Dr. Morrow's comments, it isn't (at least several months ago). Granted, some academic institution may publish something really great eventually, or not. The State gets all the data. Nobody else does AFAIK. The State should at least be trying to tease out any meaning it may have.

[M T]

Anyone with basic stats training could do the computation. Recognizing what (if anything) it means may be significantly harder. Recognizing that what Teejers wants is just "interesting" but some other set of data is actually very enlightening is harder still. This type of analysis needs to be done, and based on Dr. Morrow's comments, it isn't (at least several months ago). Granted, some academic institution may publish something really great eventually, or not. The State gets all the data. Nobody else does AFAIK. The State should at least be trying to tease out any meaning it may have.

It is the doctors and the county health departments that interact with the patients at this point.  The state only has access to the data doctors and county health departments gather.  If nobody asks the patients about, say, eyeglass usage, vitamin supplements, school attendance, mask usage, or whether cases ate at restaurants, the data isn't there to process.

Unfortunately, the county health departments are oversubscribed just trying to get people to test & quarantine.

Among the feedback I've given to my county supervisor is that I'm frustrated that with thousands of cases, the county is not revealing characteristics of those that get infected.  It would be easy for the county to gather information with surveys of a week's worth of cases (N > 500) to give insight onto many potential lifestyle features that may help other citizens to know what activities may be associated with avoiding or risking infection.   Virtually all they are revealing is aggregate gender, age, and zip code.

[M T]

Anyone with basic stats training could do the computation. Recognizing what (if anything) it means may be significantly harder. Recognizing that what Teejers wants is just "interesting" but some other set of data is actually very enlightening is harder still. This type of analysis needs to be done, and based on Dr. Morrow's comments, it isn't (at least several months ago). Granted, some academic institution may publish something really great eventually, or not. The State gets all the data. Nobody else does AFAIK. The State should at least be trying to tease out any meaning it may have.

It is the doctors and the county health departments that interact with the patients at this point.  The state only has access to the data doctors and county health departments gather.  If nobody asks the patients about, say, eyeglass usage, vitamin supplements, school attendance, mask usage, or whether cases ate at restaurants, the data isn't there to process.

Unfortunately, the county health departments are oversubscribed just trying to get people to test & quarantine.

Among the feedback I've given to my county supervisor is that I'm frustrated that with thousands of cases, the county is not revealing characteristics of those that get infected.  It would be easy for the county to gather information with surveys of a week's worth of cases (N > 500) to give insight onto many potential lifestyle features that may help other citizens to know what activities may be associated with avoiding or risking infection.   Virtually all they are revealing is aggregate gender, age, and zip code.

[Goose]

Among the feedback I've given to my county supervisor is that I'm frustrated that with thousands of cases, the county is not revealing characteristics of those that get infected.  It would be easy for the county to gather information with surveys of a week's worth of cases (N > 500) to give insight onto many potential lifestyle features that may help other citizens to know what activities may be associated with avoiding or risking infection.   Virtually all they are revealing is aggregate gender, age, and zip code.

I am quite certain the County actually does have a lot of this "characteristic" information based on what (little) SMC has actually revealed. While the counties (like everybody else) are novices with COVED-19, they have significant experience tracking other diseases (on a much smaller scale). They know what data to gather, at least to a large extent. While vastly overloaded, they are not novices in epidemiology. Normally, the State does a lot of the analysis because in normal times each county data set is too small and they can't afford to pay full time analysis folks for a very few cases. Dr. Morrow says it isn't happening at the State level. Could be sour grapes, but I myself expect it is largely true. Like many other parts of our PHS, we haven't had to rely on these folks to do their jobs, so their budgets have been cut, the good people have left, and the vestigial capability remaining just isn't up to the task. I suspect most people didn't even know the State did such things.

As a side note, I suggest people read
https://www.smchealth.org/health-officer...-statement

It is a bit dated, but it does provide some valuable perspective on how San Mateo County sees it's problems. While I don't agree with everything Scott Morrow has done, I think he is entirely rational and (unlike some other Health Department Heads) explains his positions, even in cases he didn't get his preferred outcome.

[Goose]

Among the feedback I've given to my county supervisor is that I'm frustrated that with thousands of cases, the county is not revealing characteristics of those that get infected.  It would be easy for the county to gather information with surveys of a week's worth of cases (N > 500) to give insight onto many potential lifestyle features that may help other citizens to know what activities may be associated with avoiding or risking infection.   Virtually all they are revealing is aggregate gender, age, and zip code.

I am quite certain the County actually does have a lot of this "characteristic" information based on what (little) SMC has actually revealed. While the counties (like everybody else) are novices with COVED-19, they have significant experience tracking other diseases (on a much smaller scale). They know what data to gather, at least to a large extent. While vastly overloaded, they are not novices in epidemiology. Normally, the State does a lot of the analysis because in normal times each county data set is too small and they can't afford to pay full time analysis folks for a very few cases. Dr. Morrow says it isn't happening at the State level. Could be sour grapes, but I myself expect it is largely true. Like many other parts of our PHS, we haven't had to rely on these folks to do their jobs, so their budgets have been cut, the good people have left, and the vestigial capability remaining just isn't up to the task. I suspect most people didn't even know the State did such things.

As a side note, I suggest people read
https://www.smchealth.org/health-officer...-statement

It is a bit dated, but it does provide some valuable perspective on how San Mateo County sees it's problems. While I don't agree with everything Scott Morrow has done, I think he is entirely rational and (unlike some other Health Department Heads) explains his positions, even in cases he didn't get his preferred outcome.

[dabigv13]

[Tweet]https://twitter.com/EricTopol/status/1317849521691402240?s=20[/tweet]

We're in caution warranted guys!

[dabigv13]

[Tweet]https://twitter.com/EricTopol/status/1317849521691402240?s=20[/tweet]

We're in caution warranted guys!