I don’t think you can extrapolate the success of a lockdown in a remote island nation like New Zealand and a country where the virus is already endemic. The lockdown could be all costs and no benefits.
…when did I do that?
Where would this graph even have been true anyway? Oh, New Zealand, the isolated island nation.
…what does this even mean?
We aren’t an “isolated island nation” for starters. We are an archipelago.
Secondly, what part of “success looks like over-reaction” isn’t clear to you? The numbers here are illustrative of the point: not the actual point.
… there seems to be a weird kinda tangent in the thread about NZ going on, the graph I posted was illustrative of a point, nothing more. If people want to discuss the NZ response more feel free to start a thread about it, but I won’t be responding more here.
You (and SAGE) are misunderstanding the purpose of modelling, which is risk management.
The decision makers need to know what all the possibilities are, along with an estimate of their likelihood. Otherwise, how can they make an informed decision?
A government needs to take into account many things that scientists do not have to take into account, as puzzlegal pointed out above:
They can’t make reasonable decisions about risk management unless they can consider all likely scenarios – and the likely consequences of actions they can take in those scenarios.
Simply assuming the worst case always, regardless of its likelihood, and regardless of the practical consequences of that assumption, is simply not the way to manage risk.
What’s being lost sight of here is common sense in the real world.
Modeling worst case scenarios isn’t the same as assuming that’s what will happen, though.
But failing to model anything other than worst case scenarios, and failing to consider the likelihood of each scenario, is a failure of modelling.
And even if – if - the Omicron variant turns out to be less severe, hospitals could still fill up with cases, cutting loose people with other ailments to go back home and suffer them there. If I had cancer or a heart attack, Omicron being less severe would mean nothing to me if they’re just going to turn me away because the beds are filled with Covid cases. Already Hawaii has quickly gone from low double digits of new cases daily to more than 700 reported on Friday and Saturday to almost 1000 today. The hospitals are getting nervous. (Dunno how many of the new cases are Omicron, but Omicron is here.)
But I don’t see in that conversation any evidence at all that anyone was being asked to ONLY consider the worst case.
If you’re not presenting scenarios other than the worst case, then you are suggesting that’s the only one to be considered.
Situation: Someone knocks on your front door.
Possibilities: That person has come to murder you. They have come to rob you. They have come to beat you up because of your political views.
Possible actions: Phone emergency services, hide under the bed, pre-emptively attack the person knocking on your front door.
Not modelled: That the person is delivering your Amazon parcel, is a neighbor come to introduce themselves, etc.
Not considered: Likelihood of various scenarios, and consequences of actions.
…if you have the luxury of time, sure. Spend as much as you like working through every single scenario.
But what SAGE has done here is entirely reasonable. Because they aren’t just trying to win a debate on the internet. They are tasked with the responsibility of presenting information that will decide if people live or people die. Omicron is moving dangerously fast. If you make the wrong call it will have deadly consequences. Those tasked with reviewing the information know it’s worst case. This isn’t the problem you are making it out to be.
An article referring to Imperial College research that cannot yet confirm that omicron is less severe, but can confirm that it is quicker to spread.
“We find no evidence (for both risk of hospitalization attendance and symptom status) of Omicron having different severity from Delta, though data on hospitalizations are still very limited,” the researchers state in the report.
These are independent of any government considerations. No obvious reason to skew one way or the other.
And they reference factors that I also touched upon earlier when trying to compare SA and other countries.
“While the lower observed IFR [infection fatality rate] in the early weeks of the Omicron wave in South Africa is better than the alternative, the most likely explanation lies in increased immunity among those being infected; more time and careful comparisons controlling for age, prior immunity, detection bias, lag period, hospital capacity, and numerous other factors will be required to infer anything about intrinsic virulence,”
Of course, if someone has been knocking on doors in the neighborhood and beating up people, or worse, I’d certainly respond to a knock differently.
In this forinstance, especially if the police say they are overwhelmed and can’t respond to calls for help…
This. You don’t want to find out during the worst case scenario that you haven’t prepared. Up until now, this virus has been able to surprise us time and again with a worse reality than we’d bargained for. Thanks for the graph, it’s absolutely applicable to COVID everywhere. And this reaction is a reality in itself which influences policy and leads to compromising on measures to keep things palatable for the public.
I don’t get the people who keep saying that NZ is so different, AFAICT they have a very good prime minister, a population that’s willing to do what’s necessary and the best information on a government website I’ve seen so far. Yes being an island nation might be an advantage. But they also fucked up to a lesser extent than most other affluent nations. Might be a good idea to look to them instead of bashing them.
This. 1,000 times this. Apparently that concept remains difficult to grasp, but that’s what the entire policy of testing/isolating/vaccinating is based on. Reduce the number of cases so you reduce the chance of mutation into something worse. The more it spreads, the more likely this is to happen.
Modelling a worse case scenario isn’t the same as actually reacting and taking action. Modelling a worst case scenario is simply knowing what COULD happen.
No, that just is not logical. Preparation is all about discussing the worst case scenario. You do fire drills because the building really could burn down. Doing a fire drill does not tell people “the building is always burning down.”
Um… this is not a drill.
Well, not really. Proper preparation woild be to model all scenarios, assign probabilities to them, then do a risk/benefit analysis of options for controling outcomes, including the probability that an intervention will actuall6 change things.
Modelling only the worst-case scenarios is a prescription for over-reaction. This makes sense from a public choice perspective, because politicians are risk-averse and would rather over-react than be accused of killing people. But their incentives and interests are not necessarily the same as what’s good for the people.
Also conspicuously absent from all modelling are economic, social, educational and psychological costs of the alternatives. By looking only through the lens of public health we ignore a lot of costs of the various actions. How many lost education-years for children are worth giving up to slow the spread? How many points of GDP are worth giving up? How much supply chain disruption should we tolerate to quell the pandemic? Bear in mind that all of these have human costs, and some carry risks to human life that could be as severe as the pandemic. For example, if parts of Europe were to run out of gas in January it could be a mass casualty event.
These are questions we not only haven’t answered, but aren’t even asking.
South Africa update:
-
Cases are now clearly falling, especially in Gauteng where Omicron first hit.
-
Hospitalizations still remain low, deaths remain low.
-
The fatality rate for people who’ve been admitted to hospital is a fraction of Delta for all age groups, and the average length of a hospital stay is about 3.5 days, compared with 8 days for Delta.
-
Covid exposure and vaccination rates in South Africa are lower than in the UK and most of Europe.
-
Lockdown remains at the lowest level – masks in public places and sanitising hands in shops mandatory, no large indoor gatherings.
-
The Ministerial Advisory Council on Covid has formally recommended to the government that all quarantine and contact tracing should be discontinued, as they serve no useful purpose any more.
Hey GW - just to say, thanks for the updates. From here (UK) your summaries of how the Omicron wave has developed have been very interesting.
Also a question. Is it known with any degree of certainty what proportion of the population in Gauteng have had an Omicron infection? Or does the likelihood of large numbers of asymptomatic cases just rule out any certainty? Are there a range of estimates out there that you could pass on?
j