Coronavirus: Germany infection rate rises as lockdown eases
bbc.co.uk
bbc.co.uk
There were also 2-3 larger clusters discovered this week. Looking at the nationwide numbers might be not as useful in this case. I'm not saying there is no case for concern here, but right now these are very few data points on a noisy curve, and a similar spike that turned out to be nothing happened a few weeks ago. We really have to wait a few days to actually get some reliable numbers.
> Also, internally leaked government documents a few weeks ago showed that there is an explicit strategy of the government, to generate fear in the population in order to make people comply.
A quick scan seems to indicate a document meant to summarize the situation and possible outcomes including options for actions.
They explicitly propose to be as transparent as possible and to openly discuss the worst case scenario.
Without reading it in depth, this sounds like a sensible thing to write down to prepare the relevant authorities for what's coming.
Relevant quote:
> Wir müssen wegkommen von einer Kommunikation, die auf die Fallsterblichkeitsrate zentriert ist. Bei einer prozentual unerheblich klingenden Fallsterblichkeitsrate, die vor allem die Älteren betrifft, denken sich viele dann unbewusst und uneingestanden: «Naja, so werden wir die Alten los, die unsere Wirtschaft nach unten ziehen, wir sind sowieso schon zu viele auf der Erde, und mit ein bisschen Glück erbe ich so schon ein bisschen früher»
Google Translate> We have to get away from communication centered on the case mortality rate. With a case mortality rate that sounds insignificant in percentage terms, and which affects the elderly in particular, many then think subconsciously and admittedly: «Well, this is how we get rid of the elderly who are pulling our economy down, we are already too many on earth anyway, and with a bit of luck I inherit a little earlier »
And I would say this is quite spot on, for what we actually see on Twitter and elsewhere...
The question is if treatment capacity is overwhelmed.
Otherwise we should have tried to keep it in an acceptable very low level. I personally find it astonishing that we are accepting >100 deaths per day.
https://en.m.wikipedia.org/wiki/Motor_vehicle_fatality_rate_...
(Incidentally, using traffic death numbers as a baseline for a rate of fatalities we should accept is bad anyway because those numbers are also higher than we should tolerate, IMO.)
https://www.asirt.org/safe-travel/road-safety-facts/
38,000 a year ~= 104 / day
(based on 120,000 preventable deaths in 2016: https://www.oecd.org/germany/health-at-a-glance-germany-EN.p... )
We spend enormous amounts of money because we "don't accept" the high fatality rates seen in automobiles. And it works. Check out the charts here for very clear proof: https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in...
And now we're doing the same kind of thing (in, obviously, very different ways) to fight the covid pandemic. And it's working too, though better in some areas than others. Germany in TFA is worried that maybe their R0 might have slipped above 1.0 and started growing again. Outside NYC, the US epidemic is still large and still growing.
And it's the same thing with COVID-19. We can't quarantine everyone indefinitely, this is unlikely to end without us reaching herd immunity, and the quarantine itself causes non-negligible suffering. We have to approach this rationally.
Why is trying to minimize a US death toll on the order of hundreds of thousands[1] at the cost of a 20% GDP hit not a "rational" choice? It may not be your personal choice, but it seems to be to most people. And it's absolutely a well-understood tradeoff by everyone involved. No one went into this blind to the costs.
[1] Maybe only ("only") mid-tens-of-thousands if you use the kind of life-years-saved metric that seems to be en vogue among the "open up" set.
a) they are not going to "save hundreds of thousands of deaths", because quarantine doesn't stop the virus it only slows it down, in other-words quarantine will spread out those deaths in time, but many of those deaths are still going to happen (in due time)
b) 20% GDP hit will inevitably directly and indirectly kill, make homeless and starve a lot of people
c) in 2nd and 3rd world contries the distruption of the global economy to this degree will kill a metric ton of people
d) it's impossible to keep quarantine for extended periods of time, because then it's going to be way more than 20% GDP hit.
The only question is which approach is going to kill more people and do more damage in the long run.
To me it seems obvious that extended general quarantine is going to do an order of magnitude more damage (than doing the sane thing - like isolating at risk groups).
They literally (literally literally!) already have. What do you think would have happened in late March if the lockdowns hadn't bent the curves? Why do you think the high exponent mode (doubling every 3-6 days) stopped, if you don't think the lockdowns did it?
The problem with the "anti-quarantine" arguments is that they confuse the death rates of the status quo (locked-down semi-containment) with those that obtain from an unrestricted economy. That's not how it works.
> The only question is which approach is going to kill more people and do more damage in the long run.
Find a serious person with numbers that support your case. I think you'll be surprised.
If quarantine made the virus go away (which it doesn't) - or it bought enough time to make vaccines or effective treatments, then it would make sense, but it doesn't.
All it does it make it seem like you are saving lives, while in effect the deaths are just spread across a longer timeframe and now you're killing, starving and impoverishing people from the severe and dire economic consequences.
Secondly, why is nobody ever talking about the amount of deaths, starvation, homelessness and impoverishment that will result from economic damage of general quarantine. This always gets completely ignored and negated as if it doesn't exist.
You can't say that quarantine saves lives unless you quantify how many people are killed and are going to be killed by it.
My argument is that this by far will kill, starve and impoverish way more people than whatever number of people will get saved by having access to a ventilator.
For... the TFA is literally about Germany, a nation that had its outbreak R0 drop below one (i.e. their outbreak was shrinking), far, far, far underneath the herd immunity threshold. It's simply not true that Germany didn't prevent any deaths, that analysis is straight wrong. The same holds for most of east asia.
The only reason, quite frankly, that this doesn't seem to be happening in the US is because there is a segment of the population being led to believe that lockdowns are impossible, or that they don't work, or that they're an imposition on fundamental freedoms. So we're doing it badly. But it can absolutely be done well, as nations like Germany show.
If you take 1000 deaths and spread them across a 6months or a year instead of 2weeks, no lives have been "saved".
Similarly the act of spreading it out via general extended quarantine, does massive economic damage that will result in deaths, starvation, homelessness and impoverishment that arguably will completely eclipse whatever lives are actually saved by the quarantine (due to say availability of ventilators and medical care not being overrun).
Maybe that damage isn't as large in some developed countries like Germany who can maybe afford it, but it is in many other countries.
That is not the effect of a lockdown. There is your misunderstanding. The proof is in the linked article (or data for Hong Kong, Korea, Taiwan, Hawaii, all of whom also contained the virus before reaching herd immunity).
If those measures are not feasible long term (this seems to be the case), that means they R0 can be held below zero only briefly. In other words, the spread of the virus is only slowed down (or even reversed briefly, with R0 < 0) and delayed, and in essence the most deaths are simply delayed and not saved.
While the deaths from the quarantine measures itself doesn't seem to ever get accounted for.
And that is the problem - those quarantine measure/economic damage deaths doesn't ever seem to be put into equation as if they don't exist or are negligible.
Which makes me question if quarantine does net-benefit or does more harm in log term when all the externalities and knock on effects are actually accounted for.
Or am I missing something?
Rt is below 1 in all states except Minnesota and Nebraska [1]. So the US epidemic is still large, but already shrinking. With that said, when lockdowns are eased, Rt can increase again.
[1] https://rt.live
https://91-divoc.com/pages/covid-visualization/?chart=states...
Edit: in fact I went back and spot-checked Iowa. That link you posted says that Rt dropped below 1.0 20 days ago. In fact the real world new case rate is almost three times higher than it was 20 days ago! I'll just say it: that site you linked is garbage.
I'm sorry. Those graphs you linked are pretty, but they're just dead wrong.
I'm serious: that site is bad. It's a bad model. You have to stop reading and linking it. The only way to make it "right" is to assume (as you are) wild divergence between measured case incidence and "actual" unmeasured infection statistics, which isn't borne out in any papers I'm aware of from any outbreaks anywhere.
Basically: I'm telling you that measured data makes this model very wrong. And your response is to tell me that this is because measured data is itself extremely (like 3x!) wrong. That's bad science.
Who is we? In the UK, for example, the baseline expected death rate is around 2000 per day. Sure, adding another 500 to that is a big deal, but it is maybe not as big a deal as everyone seems to think it is.
The death toll due to covid is already is visible on all averaged yearly death per day/week charts and we have not even gotten to 1% of population having gotten the disease. That is a big deal.
Only when one year is taken into account. How about decade averages? I bet tobacco is #1 on the list and covid does not even make top 10
Close to a million people die in Germany in any given year, we're always accepting >100 deaths per day. The question in 12 months will be whether the death rate significantly increased or whether it primarily moved the cause of death, so more people died from Covid-19 and less died of atypical pneumonia, old age, cancer, heart disease or the flu.
And of course, we should have kept up the social distancing.
But note this includes estimates from February, with a lot more unknowns, and fails to take into account improved treatments (Remdisivir, Plasma, ventilator use improvements, etc).
Moving forward, given treatment improvements, I think a consensus IFR range would likely be 0.1% to 0.7%.
But this only works with a very low infection rate. Once your testing capacity (which is still, basically everywhere, much smaller than hospitals' ability to treat the same outbreak) is overwhelmed, you lose control and end up back where everyone else is.
People are watching this closely.
It eventually will be if R0 is greater 1.0. The government's position is that it needs to stay below one. The official metric to that effect is the number of new infections per 100,000 people. If it crosses above 50 for any administrative district, lockdown measures ought to be tightened. As of today, there are four above that threshold.
Just because the US and UK can’t get their act together doesn’t mean it’s impossible.
China is still finding new cases so it isn’t eliminated. New Zealand can keep the virus out because it’s an island (but they better be prepared to live without tourism).
It’s all just not very sustainable.
Your statement implies that if we could guarantee hospitals would never get overwhelmed we would happily let this wash over the whole population and kill 1% of us. That should be avoided (and would also hurt the economy just like the lockdown is, it’s not an either/or choice).
But yes, it could end before herd immunity. Or it could be like the flu, and mutate / continue indefinitely.
I gather that is unlikely, coronaviruses don't mutate like the flu.
In fact, according to experts, it is most likely that this coronavirus will decrease in severity over time and become endemic like the other several coronaviruses that are responsible for 15% of common colds. Or, alternatively, occasionally returning like a bad flu year. https://www.statnews.com/2020/02/04/two-scenarios-if-new-cor...
I remember reading in other places this may have happened before, something about a really bad flu year in the 1870's that may have been another coronavirus introducing itself to humanity.
In the Netherlands there are approximately 1.100.000 people between 30 and 34 years old (inclusive). Of those people 150 entered the hospital. That’s 0.01%. Most of these people are not dying (in fact only 3 of them did, which is 0.0003%) and while there isn’t a number of people with the chronic problems it definitely isn’t all of them.
By the way slightly more than 0.1% of the population dies every year due to smoking. I have yet to see the massive safety efforts.
Respectfully, I do not agree. New Zealand went on lockdown March 23 when they jumped overnight from 66 to 102 confirmed cases. They have 1144 as of yesterday.
There are now 1.3 million cases in the U.S., almost that many in Western Europe. There are not even enough tests to know the real number. Antibody tests are indicating that a couple tens of millions of people may have already been exposed.
I know of no precedent for using quarantines to eradicating a disease that is already so widespread. There is no evidence that such a thing is possible. It would have been at some point in the past, when there were some hundreds of known cases, but not any more.
What to do from this moment forward? Did you mean to imply that you thought we could stay locked down until this is eradicated?
I wonder if there are models that can give us covid predictions based on continuing the existing lockdowns versus easing of lockdowns, with a vaccine in 12 months versus a vaccine in 4 years, and estimated economic and social costs of lockdowns. I don't think it's responsible to make these big decisions without objectively weighing all the data, and the data is fuzzy, refined over time. (Although I am pleased at how much data is publicly available.)
Thats exactly what will happen. Based on sampling its much much less like .1% to .5% but there not much of an option. It would be like saying no one should smoke or drive cause people will die.
People are now meeting face to face much more. This is causing the increase.
Now I look like a fake news idiot :D
Some may think we should "sacrifice" them immediately, because living without a mocha latte grande whatever just isn't okay, but if we'd face mass starvation, I'm pretty sure that everybody would agree that we can't keep going. Of course, that's unlikely, but let's not pretend that there's some (sane) group of people out there, that actually say "under no circumstances may any life be put in danger, ever".
The US has just suffered its largest job losses in history. The people who lost these jobs skew heavily toward lower income brackets and little or no savings. The losses are still growing.
How can you say something like this with a straight face?
It's meant to be the worst possible reason to end the lock down, with "otherwise everybody dies" being the best possible reason. Everybody falls somewhere between these extremes. I just wanted to make clear that there is no sane position that responds to the worst possible circumstances by saying "lives must not be lost because of Covid-19, but I'm fine with them being lost because of our reaction to Covid-19".
My personal opinion is somewhat close to yours, I believe. I'm aware that a massively shrinking economy has severe real world consequences.
And while specifically protecting nursing homes and the elderly is obviously being attempted, that's just far harder than one might think: nursing homes have a lot of employees, almost all of which are rather young. While some contacts can be minimised by, for example, switching to external catering, the need for care, cleaning, and medical treatments are rather inflexible.
After that, you're expected to isolate for two weeks no matter what, because any test wouldn't necessarily detect an infection if it's still in the earliest stages.
[1] https://interaktiv.morgenpost.de/corona-deutschland-intensiv...
So Germany is going for total eradication of this virus? What makes them think that’s possible.
They also have a lower population and density.
Is the same thing realistic for an industrialized country in the middle of Europe?
So it is definitelly possible to "emullate an island" even in the middle of Europe, and it seems to have worked for the countriws that did it - daily case numbers are staying low, death toll is really low, so some of these restrictions are slowly being rolled back, with cross border bis and train lines resumming from 15th of May.
You're quoting the journalist, not any German authority.
6000 (peak daily infections) * (1 - 0.85)^Months, and you’d get to <1 daily infections within five months.
And, of course, having already reduced cases by 85% makes tracing the remaining cases far easier. That, plus improvements in understanding the dynamics of the virus and all the countermeasures, now allows for a “budget” so-to-speak, to be spent on some loosening of restrictions and adding whatever is necessary to keep re-introductions at bay.
I've rarely been more unhappy about the state of our politics. There was a lot of public discussion about easing the lockdown and regions randomly made their own decisions, affecting people's behavior even where no decision had been made yet and forcing Merkels hand in the end, who was much more in favor of playing it safe.
We were on course to get the level of containment and traceability that several Asian countries have achieved, if we had only waited a little longer, maybe just a few weeks.
To me it seems we might be letting the economy run itself into the ground. I hope I am wrong.
Staying quarantined until a vaccine is available is likely not a sustainable plan.
I've argued with many people who firmly believe it is worth burning the economy to the ground in order to "save lives". Eventually, the argument distills down to: "Do you care more about corporate profits and bankers and billionaires OR peoples' lives?"
People have to feel like eating at a restaurant is safe for being open to matter.
This is not brought up enough. There's a family of immigrants in my neighborhood from Sierra Leone, and the mom is so terrified of covid-19 that she won't let her kids go out and play, or even go to the local elementary school for free meals (they are on government assistance). We've delivered them food because of this fear. I've tried to explain that kids are at low risk and that it's probably worse to keep them locked inside for months vs. letting them play out side, but alas, the fear is too strong...
"VISCUSI: If you multiply the 1 million lives saved [Trump], and if the lives are worth $10 million each [OHSA], the result is $10 trillion. … Ten trillion dollars is half of the U.S. GDP, which means that in order to justify completely opening businesses back up, the economy would need to lose half of its value."
You'd have to argue why its suddenly OK now.
You can also argue to discount children - where I live, we make them by the dozen. It is less applicable to COVID-19 through.
But VSL the correct framework because its based on the premium the average American at large puts on jobs to compensate for the risk of injury or death.
What could be more democratic?
But $10T is such a large envelope that even if you added the granny discount, you would still come out at an enormous fraction of USA and hence world GDP.
https://twitter.com/AndyBiotech/status/1241741127205572609/p...
But feel free to add the 38% USG pensioner discount (or 100% if you care).
You'll still come out at a very, very large number.
When supply chains stop how will the lowest of the lowest on the ladder cope?
"About 265 million people around the world are forecast to be facing acute food insecurity by the end of this year, a doubling of the 130 million estimated to suffer severe food shortages last year." [0]
Excess death because overrun hospitals should never happen, but the global perspective is horrifying and will never give any easy answers.
[0]: https://www.theguardian.com/world/2020/apr/21/global-hunger-...
It goes down to - do you want to save lives of wealthy - usually old middle class people in 1st world countries who can "afford to quarantine" or do you want to save more lives of people who are going to starve, become homeless or die due to massive economic disruption this extended quarantine is going to bring upon.
It's only middle class pencil pushers who can quarantine at will semi-indefinitely and work from home who think extended general quarantine is even a viable option in the first place.
'Crushing' the curve and getting back to contact tracing levels is one. 'Flattening' and avoiding Lombardy is another.
I was hoping we would have used the time to have massive testing programs by now in the US, to follow Korea's example, but that hasn't happened at all.
Paul Romer's plan is my favorite but the country has seen little progress since he proposed it https://www.newyorker.com/news/q-and-a/paul-romer-on-how-to-...
Places like NZ are looking at total eradication, but the rest of the world will most like just have this spread through our populations before any effective treatment of vaccine is produced. Countries like NZ .. might not ever be able to open their boards again.
If the antibodies don't last long, yes we could see reoccurances for the rest of our lives. It's in our environment. It took 20 years of very big vaccination efforts to eradicate smallpox. Sure we have better technology now, but to think we can eliminate this in less than a year is hubris.
I wrote this post last week, about how we might just be in denial that our life expectancy might now be lower, and how we might need to just come to terms with that:
https://battlepenguin.com/politics/this-is-not-a-time-of-hon...
Flattening the curve just below hospital capacity is shown to be dangerous because you cannot control the infection rate with sufficient precision, causing slipups and second, third, fourth waves necessitating repeated longer lockdowns. All in all this damages the economy more because of the panic it periodically creates, repeated longer lockdowns (albeit maybe not as strict) and the uncertainty. There is always another lockdown looming just a few bad numbers ahead. Flattening the curve is sooo March 2020 ;)
Germany is now trying to feel its way to a best possible compromise of somewhat lesser/better targeted restrictions that still keep R<1.0 and, thus, the virus on track for elimination.
The vast majority of the early countries in Europe and Asia has seen drastic successes. See, for example, https://www.nytimes.com/interactive/2020/world/coronavirus-m... (scroll up slightly or search for "decreasing").
Among the richer countries, it's just the UK, US, and Sweden that are "levelling off".
We can only count deaths, but even that is tricky since people that were given 30 days to live are being counted as COVID-19 victims if they were infected.
The only way to know what the impact of the virus is/has been, is to compare with deaths from the previous years.
Reproductive numbers are important, but I'm not sure governments have really stepped up to the plate to explain how the virus spreads and what people need to do to stop spreading the thing. "Social Distance", "wash your hands" and sometimes "wear a mask" really lacks lacks explanatory power.
Given the vast costs the world has put up with so far to slow the coronavirus down, there is actually room in the conversation to have teams of people travel around and conduct some sort of training session on how not to spread this thing and have people actually act out simulations. There need to be finer grained tools than mass lockdowns and alternatives to tracking everyone with apps.
I've seen a lot of basic confusion on HN about whether masks even help. This is pretty clear evidence of uncertainty in the broad community about how to deal with respiratory viruses. I don't believe that is due to a lack of research; engineers have had the practical tools to research droplets in airflows work for more than a century now.
Even the fact that "SARS with a risk of asymptotic spread" didn't have world leaders slamming borders closed in Jan/Feb hints at a possible general misunderstanding of how virus spread works. There were other considerations at the time, but in retrospect more general education on the subject would have been well worth the time invested.
But these are exactly the sorts of things that shouldn't still be open for discussion after 4 months and global lockdowns. Even among the general public.
And a lot of comments here talk about restrictions deployed simply to avoid an overwhelm of the healthcare system. While that’s important, the longer we can prevent the disease running through the population, the lower we can make the case severity and fatality rate due to advancements in treatment.
A lot of commentary on COVID-19 recently presumed that we can’t control outcomes, only the rate of spread. That may be true at a point in time, but the longer we hold out, the better the outcomes will be on the tail end with advancements in treatment research.
I calced a bit and saw about a 1% chance of getting infected within a month at the office. Maybe acceptable for a month, not for a year (>10% risk than).
Likewise, by the time governments officially started easing restrictions many people had grown weary of long-term social distancing and the weather started getting nicer so people started venturing out more than they had been.
Also people already started to take the restrictions less seriously quite a bit earlier than the official relaxations. And the lockdown was never very strict in Germany in the first place. Some types of businesses had to close, but people could go outside "for no particular reasons", just keep a distance to people not in your own household and if possible don't travel across Germany.
And finally, AFAIK the RKI "infection rate computation" is based at least partly on predictive models.
PS: The actual lockdown rules differed slightly across the German federal states. I think at least Bavaria had stricter rules then what I described, which makes sense because they've been hit harder.
The regulations where eased on some issues, but stricter requirements were introduced in other place, e.g. masks were made mandatory in public transport, super markets in most (all?) states etc.
Does anybody know an opensource epidemic simulator to test the idea?