The “curfew” of 10 pm in SD County that was announced just last night for all restaurants has some aspect of the analog approach that I was suggesting, except that IMO it should also apply to bars that don’t serve food. They too should be allowed to be open – until curfew.
Actually, if I were making the rules, I’d move back the opening time for bars and restaurants, rather than the closing time, to reduce total daily patrons by the same amount, while leaving “night life” more or less in place. It could accomplish the same thing. Say, bars not open until 8 pm, closed by 12, or something like that. Restaurants open from 6 pm to 11 pm, perhaps.
The other possibility would be to leave things as they were as of yesterday (June 30), but allow bars to be open only on Wed - Sat; restaurants open Wed - Sun; no change in hours.
If this doesn’t reduce new infections sufficiently, the hours or days are reduced further. But neither restaurants nor bars are closed completely.
When people drink alcohol (in bars) they tend to make not the best decisions in many aspects, e.g. mask, social distancing etc.
With a quite deadly disease you don’t have the time/luxury of fine tuning such things as it will cost many additional and unnecessary lives and so it is understandable that Newsom is taking a much more direct approach.
The objective is to reduce new infections by reducing exposure time and the number of people exposed. I just don’t think the control has to be an “on-off” switch.
The publications around viral exposure time and severity (or even “initiation” of disease) are far from clear. At this point we don’t know what viral load and exposure will have an effect (and most likely it is different from person to person). Until we don’t have better scientific evidence it’s better to plan for the worst case scenario and behave accordingly
In the case of covid, the process variable might well be the 14-day running average of the ratio of positive tests (new cases) to total tests. The set point might be 3%.
Another – and perhaps better – process variable might be the 7-day running average of all hospital beds in the region that are in use. The set point might be 70% of them.
A mathematical model for the process being controlled is needed. Here’s an example, one of many, for diseases:
Feedback control provides an intelligent way to reduce the peak value and spread out (“flatten”) the “curve” (that has the shape illustrated by the plot of i(t) at the end the MAA article), without exceeding hospital capacity and at the same time minimizing the extent of social damage due to restrictions.
Newsom’s simplistic Threshold => On/Off approach to feedback control, however, is crude and frankly pretty dumb. And it stings. They need an engineer or mathematician to help develop a more scientific approach to accomplish the exact same objectives with less whiplash. We have some of the greatest minds in both of these fields in CA.
This is what I’ve been arguing here or maybe elsewhere (or both). Or trying to. I’ve argued that we should have been and should now be increasing hospital headroom as quickly as possible, to prepare for the worst case. Because covid isn’t going away, and the seasonal flu is on its way as well – and as usual – come fall.
So much money has been spent on damage control for what’s already been screwed up this year. Why not spend some on being prepared for what’s next?
It’s even more disheartening when you begin to understand the scope of the failure and the planning and safeguards that were already in place. Just one perspective
I am fully aware of all the mathematical background - even every rice cooker is based on it. My point is that this principle (if you look into details) will always over- and undershoot and every overshoot would cost many human lives, so that means you are willing to sacrifice lives so that a few bars can stay open ? Newsom’s approach is not simplistic - he just values human lives more than you and doesn’t want to take any risk.
I find this kind of affront objectionable. It’s hard to have a meaningful adult discussion with someone who constantly makes personal attacks like this.
And I think that a more sophisticated feedback control of disease spread would actually save more lives than the Newsom On/Off switch approach.
It’s a long article, but well worth a read. Yes, the failures to manage covid spread worldwide are in large part on our heads, because the world had always before looked to the US for leadership and guidance in crises like this. The analogy to flight (being as he’s an amateur pilot) is interesting, but his detailed overview of the course of events since early January is compelling. Simply put: We’ve been and still are rudderless at the top. The deletion of (and inability or unwillingness to restore) planning and safeguards that were already in place is damning.
I am always surprised that Americans still think “other countries” look to America to lead in crisis - this has drastically changed over the years - one of the reasons why nearly every country (yes, even Italy, Spain etc) have a better grasp in the Covid crisis is that they never looked for leadership from the US. Have those countries made mistakes ? - No doubt but they have learned from it while the US is a giant clusterfuck with no leadership, vision, and afraid of drastic measurements because of convenience.
With no vaccine, no clear policies for handling the pandemic and no direction from the current administration, each State has been left to their own devices. With 40 million people and no help from the occupant of the White House, Newsome has been in uncharted territory and is having to make decisions for the greater good of all 40 million Californians as well as the 5th largest economy on his own. Personally, I think he’s doen a pretty good job managing the situation.
It’s a portion of those 40 million Californians who have chosen to live on the edge putting all of us in jeopardy. Right now he’s having to react and manage the consequences of people defying social distancing and mask requriements and spiking the outbreak. San Diego had 100 more positive cases on Thursday than it had on Wednesday. We’re now up to 22 cluster outbreaks, 3 times the State requirements for reopening and maintaining the opening. Luckily, our hospital bed availability is still not at capacity, but, remember, hospital admissions often lag positive test results. The hospital occupancy level could change fairly rapidly over the next week.
I’m willing to cut Newsom some slack for the time being. I think he’s handled the situation as well as could be expected under the circumstances.
He was responding to the content of the article I posted yesterday that detailed the botched U.S. government response to CV-19. In it the author described the pandemic work that had been done by both the Bush and Obama administrations and how intertwined it was with WHO and our NATO allies. It wasn’t “look to” in terms of the U.S. solving the problem, it was more our NATO allies knew that there were substantial procedures already in place, coupled with the tight relationship the U.S. held with WHO at that time, they were waiting for the U.S. to activate the existing response plan so that there could be a single systematic plan in place for all countries to follow and a specific plan for resource management and distribution. That CLEARLY didn’t happen. The European countries were on their own and have proven that they were more than capable of handling a response. Unfortunately, the 50 states comprising the U.S. haven’t fared as well.
I myself wouldn’t attempt to design a feedback control system for containing the covid. But in very general terms, the amount of correction is a function of the “error” between the set point and the value of the process variable. (I’ve mentioned two possibilities for these, but there are others.) In SD county, the 14-day positivity ratio began to rise around the middle of June. With well-designed feedback control, we wouldn’t have waited for it to go from 3% then to 5% now. Restaurants and bars were “fully open” in mid-June, but cutbacks would have started almost immediately when the positivity ratio began to rise, The amount of change would be related to the amount of “error”, thereby reducing new cases and stabilizing things – and saving lives. The cutbacks would have been based on the amount of “error”, but the fed-back control would not be swinging from “totally open” to “totally closed”. That’s both excessive delay and gross over-correction, in my opinion. Well-designed feedback control leads to stability, greatly reducing, and in many cases virtually eliminating, large oscillations in the process variable. Would this work with covid? Well, like I said, I didn’t design a feedback control so I don’t know. In my first post about this, I was just suggesting it as a thought for an alternative, in response to a post by ElsieDee (in terms of different regions applying different restrictions, leading to movement from strict to less strict areas). Look back to what I said: “Here’s a thought in terms of an alternative.” And then I posted articles describing details of the math and models, “for your consideration”. That’s all. My opinion, though, is that On/Off control has resulted in over-corrections in both directions, and that that approach is very crude and has it’s own problems, including people traveling into areas with lesser “prohibition”. Less difference in restrictions among regions would mean less incentive for movement, I thought. (And I think this could also be applied at the national level.)
The rate at which coronavirus tests in California are coming back positive continues its alarming climb — jumping 51% over the last two weeks, a Los Angeles Times analysis found.
An increasing rate of positive test results is an indication disease transmission is worsening.
On Thursday, a Times analysis of state data showed that over the last seven days, 6.9% of coronavirus tests came back positive. A week earlier, that number was 5.6%; and a week before that, it was 4.6%. The last time the so-called positivity rate was this high was May 1, when the rate was 7%.
The very first data point (on March 27) of a rolling 14-day average of the “positivity rate” was 8% in SD County. That was attributed, in part and after the fact (not by me!) to the status of people being tested, which initially consisted only of those showing symptoms. Later, when testing was open to the public and the number of tests greatly expanded, that percentage came down, eventually leveling off at about 3% (even less, actually). Until very recently.
So (and I hope people don’t get out their torches and pitchforks) perhaps more of the people being tested in the past three weeks are those presenting symptoms, once again? I still believe that randomized testing is more meaningful, but of course people showing symptoms definitely have to be tested. It’s really not clear what the best balance is there. Widespread voluntary testing isn’t really the same as fully random, although it’s closer.
But yes, the increases are alarming. We’ve gone up 50% in SD County in the past two weeks or so. A correction is long overdue – long being measured in days after the increase started.
Also, (ok, you can light your torches now) let’s not forget that in addition to opening up more of the economy in that timeframe, including bars, there were some large gatherings of people, not all of whom were following the mask and spacing needs.
With a long incubation time (but still highly infectious) and high (but unknown level) of asymptomatic cases coupled with a 2-4 weeks delay of impact on the hospitalization numbers (which are also rather “rough” as you need a much more nuanced “binning” of case severity) I really don’t see how your proposal would even remotely work without causing many additional, unnecessary deaths. With such a disease as Covid there is in the moment nothing else as the best method as the different variations done in Europe and Asia of complete close down and tracking of centers of infection until the numbers are really low before reopen. And thereby Newsom doing not a bad job - I hope he is more consequential on those people not wearing mask, social distancing etc. and that he waits longer with reopening restaurants and bars until the numbers are really low.
Whether or not the approach I suggested would succeed without increasing deaths is something that control theory experts working with epidemiologists might be in a position to assess, not me. I only think it’s worth taking a look into it.
However, let me add this. Except when hospital capacity is exceeded, is it not reasonable to assume that eventual deaths are a function of the number of people currently infected (including pre-symptomatic, asymptomatic, and unreported cases)? If you’re at least willing to accept that assumption, then the following article may be of interest:
As for Newsom, as I said before, I think he means well. I voted for him and generally thought he’d been doing a pretty good job as governor, until now.
More people without symptoms are getting tested, which makes the significant increase in the positivity rate even more alarming.
It was clearly a mistake to allow indoor dining. Newsom fucked up on that one. As he continues to fuck up by not imposing statewide standards so as to eliminate the incentive to travel.
A friend here in Berkeley made plans to go to Amador County to have lunch with a friend in a cafe in a town that had had zero cases, but had to cancel as four of the cafe’s employees had been infected.