[Goose]

There are over 3000 counties in the United States.  If each is trying to do contact tracing and failing, maybe it isn't because there are 3000 sets of incompetent efforts. 



What are the features that might be making it ineffective?

  1.  Too many cases.   You can't catch every rain drop in a storm.  However, I believe almost all current hot spots were not so bad 1.5 or 2 months ago.

It is true that the case count can overwhelm the tracing capability. However, we aren't being told that is the case. San Mateo County and San Francisco County have explicitly stated it is NOT the case in their jurisdictions. SCC has been a bit coy. For sure I expect LA County isn't keeping up.

  2.  Lack of cooperation.  I'm sure there are some cases that don't want to cooperate.  I'd bet that is less than 10%.

I would take that bet. As others have pointed out the contact tracers rely on the telephone. If you are like most of us, you just don't answer unless you recognize the caller. You aren't "intentionally" not cooperating, but it has the same result. At least in NY, the unwillingness to help by giving out contacts has been an issue.

  3.  Hands tied: HIPAA.   I think that telling Bill that he was exposed by Adam would be helpful, and bring it home to people.  But I don't think that is most of the problem.

I think this is a non-issue. If you reach the Bill on the phone, it probably doesn't matter to him who gave him the disease. Knowing who it was may jog the contacts memory, but beyond that I don't think it matters.

  4.  Closing the barn door too late.  It is abundantly clear that contact tracing can't be effective with the kinds of delays we are seeing in doing the tracing after the delays in noticing the disease, testing for it, getting the results back, and then getting the info to the county, then they get contact info, and the tracers then actually get in touch with the contact.  The infectee would already be sick and know it.  You may catch some asymptomatic cases, but clearly that isn't helping.

You are correct that the delays will greatly reduce the effectiveness of contact tracing. But that isn't the problem. If it were, we would still see lots of cases where the person was listed as a contact (if only in retrospect). At least in SCC that isn't happening, as the number of cases who are contacts is now 19.2% and still dropping. SF and SM are unknowns, because they don't report this data.

  4.  Attacking the wrong problem.   We are trying to tell Bill that he was exposed by Adam, but by then, Bill has already been exposing others.  We should be asking Bill who he might have exposed, and telling them.

Not quite. We tell Bill that he was exposed, and that he must quarantine and must get tested immediately. If he does test positive, he becomes a case and the others he has exposed become contacts by natural causes. If he doesn't test positive, we got lucky. He isn't shedding virus and couldn't have infected anybody yet. If Bill does start shedding virus later it won't matter because he is in quarantine. If Bill has already got sick and been tested (because of excessive delay in contact tracing process) that is certainly bad, but the contact tracers should record that statistic as well to show the effect of slow reporting.

  5.  Misunderstanding of the spread mechanism.  When I think about the seemingly universal failure of contact tracing in the US, I keep coming back to this.  No doubt that if you stand next to someone for 15 minutes who is shedding virus, you have a good chance of getting the disease.  But is this a good model of who CAN AND DOES get the disease?   If Adam walks in a crowd or into a grocery store, how many non-close contacts may get infected by him.   The half-life of the virus in an aerosol is an hour.   That's plenty of active virus for hours after he walks in for whomever inhales it.   Alternatively, the model that we are safe outdoors isn't really true.



Maybe this is like bolts of lightning.   Stand out in a field in a lightning storm.  If you stand there longer, your chances of getting hit get higher.  If there is more lightning strikes, your chances of getting hit are higher.    But, even if there is only one lightning strike, and you happen to be in the wrong place at the wrong time, you can get hit.    If there is a crowd of 1000 people in that field, there's a roughly 1000x higher chance that one of them will be hit.  



Maybe all it takes is one viron getting to a cell that somehow is not where the immune system can notice it & respond to it effectively.  That's the lightning strike.  It starts churning out virus, and spreads.  



In the various well-documented spreads, it was never that everyone within some radius got the disease. It was spotty even up close.  But there were people that got it many times further than the nearest one that didn't get it.



I don't know exactly how the probabilities would work.  The obvious thing is that the guy that took 450 breaths within 6' of you would be exposed 45x as much as the person that walked by for 20 seconds and took only 10 breaths.  But maybe you don't shed virons equally with every exhalation and they don't exist in some cloud with a density proportional to the distance from your mouth.  Maybe sometimes you "clear" your throat, or you say "Suffering succotash" and expel more virons in the aerosol from your breath or speech.  The warmth of your breath may take the aerosols up into the air above your heads, and where it cools & happens to ride an air current into someone's breathing space, nobody knows.



(Hmmm... I wonder if anyone has looked at the height of people that get the disease.  I wonder if tall people are more likely to have their heads in clouds of virus-laden aerosols than shorter people.)

The question you are asking comes down to who is a contact. If you got the virus via a stray viron from a guy riding by on a bike, contact tracing isn't going to work in that event. However, it does come down to probabilities. We do know from the events in Taiwan and Korea that contact tracing can work. Germany has also had success in this effort. You don't need to get everybody. You just need to get a high enough percentage that you can drive R below 1. It hasn't been easy, but the contact tracers have been relentless. Consider the gay nightclub guy in Korea. Despite people lying on their contact information, they mostly got traced down.

What is different in all these other nations is that the people do isolate when they test positive and that they do provide contact data (at least in the main). The contacts do get tested and do quarantine. The contact tracers do visit your place of quarantine. If you get caught not meeting your responsibilities, you get locked up. We don't have/do any of those things, and that is why we are failing IMHO.

Unless and until people know the have to cooperate and do so, either willingly or otherwise, contact tracing isn't going to work. Being very slow in contact tracking could be an issue, but as of now we have no evidence that it is a problem. If contact tracing was working well but too slowly, the tracers would know it.

[Goose]

There are over 3000 counties in the United States.  If each is trying to do contact tracing and failing, maybe it isn't because there are 3000 sets of incompetent efforts. 



What are the features that might be making it ineffective?

  1.  Too many cases.   You can't catch every rain drop in a storm.  However, I believe almost all current hot spots were not so bad 1.5 or 2 months ago.

It is true that the case count can overwhelm the tracing capability. However, we aren't being told that is the case. San Mateo County and San Francisco County have explicitly stated it is NOT the case in their jurisdictions. SCC has been a bit coy. For sure I expect LA County isn't keeping up.

  2.  Lack of cooperation.  I'm sure there are some cases that don't want to cooperate.  I'd bet that is less than 10%.

I would take that bet. As others have pointed out the contact tracers rely on the telephone. If you are like most of us, you just don't answer unless you recognize the caller. You aren't "intentionally" not cooperating, but it has the same result. At least in NY, the unwillingness to help by giving out contacts has been an issue.

  3.  Hands tied: HIPAA.   I think that telling Bill that he was exposed by Adam would be helpful, and bring it home to people.  But I don't think that is most of the problem.

I think this is a non-issue. If you reach the Bill on the phone, it probably doesn't matter to him who gave him the disease. Knowing who it was may jog the contacts memory, but beyond that I don't think it matters.

  4.  Closing the barn door too late.  It is abundantly clear that contact tracing can't be effective with the kinds of delays we are seeing in doing the tracing after the delays in noticing the disease, testing for it, getting the results back, and then getting the info to the county, then they get contact info, and the tracers then actually get in touch with the contact.  The infectee would already be sick and know it.  You may catch some asymptomatic cases, but clearly that isn't helping.

You are correct that the delays will greatly reduce the effectiveness of contact tracing. But that isn't the problem. If it were, we would still see lots of cases where the person was listed as a contact (if only in retrospect). At least in SCC that isn't happening, as the number of cases who are contacts is now 19.2% and still dropping. SF and SM are unknowns, because they don't report this data.

  4.  Attacking the wrong problem.   We are trying to tell Bill that he was exposed by Adam, but by then, Bill has already been exposing others.  We should be asking Bill who he might have exposed, and telling them.

Not quite. We tell Bill that he was exposed, and that he must quarantine and must get tested immediately. If he does test positive, he becomes a case and the others he has exposed become contacts by natural causes. If he doesn't test positive, we got lucky. He isn't shedding virus and couldn't have infected anybody yet. If Bill does start shedding virus later it won't matter because he is in quarantine. If Bill has already got sick and been tested (because of excessive delay in contact tracing process) that is certainly bad, but the contact tracers should record that statistic as well to show the effect of slow reporting.

  5.  Misunderstanding of the spread mechanism.  When I think about the seemingly universal failure of contact tracing in the US, I keep coming back to this.  No doubt that if you stand next to someone for 15 minutes who is shedding virus, you have a good chance of getting the disease.  But is this a good model of who CAN AND DOES get the disease?   If Adam walks in a crowd or into a grocery store, how many non-close contacts may get infected by him.   The half-life of the virus in an aerosol is an hour.   That's plenty of active virus for hours after he walks in for whomever inhales it.   Alternatively, the model that we are safe outdoors isn't really true.



Maybe this is like bolts of lightning.   Stand out in a field in a lightning storm.  If you stand there longer, your chances of getting hit get higher.  If there is more lightning strikes, your chances of getting hit are higher.    But, even if there is only one lightning strike, and you happen to be in the wrong place at the wrong time, you can get hit.    If there is a crowd of 1000 people in that field, there's a roughly 1000x higher chance that one of them will be hit.  



Maybe all it takes is one viron getting to a cell that somehow is not where the immune system can notice it & respond to it effectively.  That's the lightning strike.  It starts churning out virus, and spreads.  



In the various well-documented spreads, it was never that everyone within some radius got the disease. It was spotty even up close.  But there were people that got it many times further than the nearest one that didn't get it.



I don't know exactly how the probabilities would work.  The obvious thing is that the guy that took 450 breaths within 6' of you would be exposed 45x as much as the person that walked by for 20 seconds and took only 10 breaths.  But maybe you don't shed virons equally with every exhalation and they don't exist in some cloud with a density proportional to the distance from your mouth.  Maybe sometimes you "clear" your throat, or you say "Suffering succotash" and expel more virons in the aerosol from your breath or speech.  The warmth of your breath may take the aerosols up into the air above your heads, and where it cools & happens to ride an air current into someone's breathing space, nobody knows.



(Hmmm... I wonder if anyone has looked at the height of people that get the disease.  I wonder if tall people are more likely to have their heads in clouds of virus-laden aerosols than shorter people.)

The question you are asking comes down to who is a contact. If you got the virus via a stray viron from a guy riding by on a bike, contact tracing isn't going to work in that event. However, it does come down to probabilities. We do know from the events in Taiwan and Korea that contact tracing can work. Germany has also had success in this effort. You don't need to get everybody. You just need to get a high enough percentage that you can drive R below 1. It hasn't been easy, but the contact tracers have been relentless. Consider the gay nightclub guy in Korea. Despite people lying on their contact information, they mostly got traced down.

What is different in all these other nations is that the people do isolate when they test positive and that they do provide contact data (at least in the main). The contacts do get tested and do quarantine. The contact tracers do visit your place of quarantine. If you get caught not meeting your responsibilities, you get locked up. We don't have/do any of those things, and that is why we are failing IMHO.

Unless and until people know the have to cooperate and do so, either willingly or otherwise, contact tracing isn't going to work. Being very slow in contact tracking could be an issue, but as of now we have no evidence that it is a problem. If contact tracing was working well but too slowly, the tracers would know it.