Hacker News Reader

Comment by hogFeast | Hacker News Reader

And lockdowns can cause mortality too. In England, the waiting list for operations (these are procedures that a doctor has determined are required medically, some are elective but England doesn't do many electives because of cost) is expected to hit 12m in a few years...the total adult population in England is 56m. Ignoring the economic turmoil that they cause, ignoring the social turmoil.

There are no very good choices. But the high cost of lockdowns should dictate that they only be used when the alternative is extremely large numbers of people dying (at least in the UK, this was probably met at some point last year but hasn't been met for the majority of the Covid period).

View on HN

I agree that the NHS is fucked, but that's because of:

* COVID itself

* its funding and delivery model making it vulnerable to politicians acting in bad faith

far more than restrictions on gatherings.

By the way, an "elective" surgery is one that can be scheduled in advance. Most elective surgeries are not optional. https://www.hopkinsmedicine.org/health/treatment-tests-and-t...

Right, and how do you explain what has happened in Scotland then? The SNP gave ample rope to the NHS, and they promptly hung themselves with it. I agree there are structural issues but those aren't separable from the funding issues. A few months ago, the Tories said the NHS was getting £20bn in extra funding, and journalists were being briefed by NHS lobbyists/BMA that weekend that it wasn't enough...at some point, the public has to recognise that the resources problem isn't separable from the structural problems.

My comment about electives was for the US audience. The US does a huge amount of quality-of-life, elective surgeries that don't happen in the UK (i.e. giving an 85-year old a knee replacement). So when you write something like: the waiting list is long, people from the US might not understand that the mix of surgeries in the UK is very different from there, and it largely represents people who need prompt care (for example, I have a relative who is a high bowel cancer risk and needs a regular screen, in the US this would likely happen annually but hasn't happened for about three years now because they were overdue in early 2020 and then everything stopped).

How many people is "extremely large" in your opinion? And who?

Or to put it another way: Would you be okay to go, while we keep everything open? How about your parent, or grand-parent?

Isn't this exactly what measuring excess mortality is for? But it's not reported anymore since there is none, and media has instead shifted to infection rate.

If you were old and had little time left, and had a compromised immune system that made any slight infection possibly lethal, and you had the choice to close down society to lower your risk of infection, would you do it?

What? There has been significant excess mortality, as summarised here: https://www.economist.com/graphic-detail/coronavirus-excess-...

My own country alone has seen nearly 300 000 excess deaths compared to the annual average, with the waves of excess deaths neatly tracking those of known COVID infection waves.

This is exactly the problem, you're using an absolute number without reference "look big number = scary" and you are simply comparing to an average, which doesn't say anything about whether or not this is an unusual or especially different situation.

"The temperature outside is 2 degrees higher than average". Ok what conclusions can you draw from that? Very little. Just because you make a big scary number without reference, doesn't allow you to jump to a conclusion out of fear, it's just manipulative misleading misinterpretation to make it look as scary as possible, to disable people's critical thinking and support a point that actually doesn't have any basis.

No, I'm providing what you asked for: A measurement of excess deaths.

That is, not just the number of people who died, but the number calculated to have died over and above the normal seasonal average.

If those excess death calculations are perfectly acceptable to calculate the effect of bad flu seasons and other mortality-affecting events, why are they suddenly not okay now?

Yeah like every other article they use a formula to calculate what is "excessive" which is not explained at all, and is used as the whole basis of the article. I took a peak in the github repo and I can't really read R well enough to make anything of it, but from a quick look and reading the article "take the number of people who die from any cause in a given region and period, and then compare it with a historical baseline from recent years" it could be a well defined prediction, or it could just be an average. It's unclear as the method section here is completely unknown, and I'm sceptical how you could ever calculate a fixed number prediction without obscuring the standard deviation.

Why not just show standard deviation, median, average, percentile. Show the standard deviation in the graph.

You know that's just an article providing an overview, right? It's not the official source of excess mortality statistics.

For that, each country maintains an official group that has access to death certificates and other records when preparing statistics. The figure I quoted, for instance, is from the South African Medical Research Council, who publish their methodology. There's also EuroMOMO and many others.

Did you really think that statisticians and scientists have just magically forgotten about how to do basic statistics, for decades, when preparing these reports?

Why should I put any weight to statistical results that are made with some unknown "formula" and is not presented in a standard way.

Instead of using the normal measures of mean, average, standard deviation and percentiles, we just used "formula" and voila, here is answer. Ok I'm going to continue to be sceptical of such results..

It's not unknown, all the groups publish their methodology in the open.

I'm now convinced you just want to be a contrarian, rather than learning about a system you don't understand.

You need at least to include the standard deviation for that to be of any importance, and at least for 10 years in the past

Now you're just shifting the goal posts.

Excess death tracking is hardly a new thing, most countries have been doing it for decades to measure the effect of things like flu outbreaks. Therefore when those same statistics being done using the same approach suddenly show huge climbs in 2020 and 2021 it means something.

Stop trying to minimise the pandemic. If your arguments about how to deal with it are rational they should stand up to the reality of how serious it really is, rather than a pretend reality where people haven't been dying in large numbers and filling up hospitals for two years.

I'm not trying to minimise the pandemic, I'm pointing out obvious ways that it is exaggerated, such as lacking basic statistical scrutiny and throwing around huge scary numbers with no comparison or explanation.

Another example from the media (now I am shifting goal posts, but defending my point of proving that it's exaggerated and manipulative): Why are they reporting running totals of number of deahts? And making big headlines of when they reach "new grim milestones"? How is that not a completely irrelevant manipulative misinterpretation? A running total of deaths, really? When will that ever go down? And using that to support a case that "things are getting worse" somehow.

Nope. You may think you are, but in reality you're ignorant of how these statistics are calculated and you that's why you think you're smart for finding what you think are mistakes. This has, actually, all been taken into account, and more, for excess mortality statistics.

I'd recommend you spend some more time learning how the scientists and statisticians calculating these figures actually work, and maybe speak to a few, before commenting further.

It's not my job to find out how they made their secret calculations, it's up to them to show it if they want to present credible results, otherwise it should be ignored.

> Isn't this exactly what measuring excess mortality is for? But it's not reported anymore since there is none

It isn't reported because it takes more explanation than the media likes to give, and direct COVID-19 death counts are less complicated, not because “there is none” which is just false:

https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm

https://www.destatis.de/EN/Press/2021/10/PE21_481_12.html

Direct covid-19 deaths are irrelevant in relation to deciding about lockdowns, that's deliberately making an issue more complex than it is, in order to obscure it. If covid deaths are simply a proportion of expected death, which are not higher than usual, then there is no need for any measures at all.

> expected death, which are not higher than usual

Again, as stated in GP (with supporting data) in neither Germany (the context of the article) nor the US (the context and source of much of the political debate in the thread) is it true that actual deaths are not higher than expected from pre-pandemic experience.

Replying to sibling: There's PLENTY of excess mortality: https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm

Summer of this year was close to normal again, though it is of course easy to imagine it wouldn't have been in a zero-restrictions scenario.

That doesn't say anything, it only counts death during one single year: 2020. And concludes that the death rate is excessive during the flu season, compared to not in the flu season.

It expects death rate to be constant over the year, which is wrong.

You need to take the total number of deaths per capita, for 2020, and compare that to the same number for 2019, 2018, 2017 etc. And look at the standard deviation.

What on earth are you looking at? The default view is US weekly excess deaths for 2018 through October 2021, and uses historical estimates from 2013 to present to estimate per-week thresholds: it's most certainly NOT using a single threshold per year. Go read the methodology section, please!

You can clearly see bumps due to flu season in previous years, but because flu season happens every year since before 2013, this becomes a regular seasonal effect and don't show up as 'excess' except in late 2018. We also happen to know that flu was greatly suppressed by anti-Covid measures these last two years.

You are correct, I looked at it again, it's actually using historical data.

But I'm still not sure what I'm looking at, overlaying a single year over an average is always going to look "excessive". I want to see previous years as well, why not just extend the graph? That's the one single most important number that you never get so see. Here they have buried it in a complex algorithm which looks honestly suspicious at first glance, upper bound for excessive death is basically the same as average. Is standard deviation really that low?

EDIT: I just want to see how many people died per capita in 2020, 2019, 2018, 2017 and so forth, why is that not shown anywhere?

> EDIT: I just want to see how many people died per capita in 2020, 2019, 2018, 2017 and so forth, why is that not shown anywhere?

Because that statistic makes the whole thing seem very mellow. Same thing as average life expectancy by year; the change from 2019 to 2020/21 is very slight.

Yeah but then the whole thing is very mellow, and showing that to people is the right thing to do, and not misinterpreting to blow something out of proportion to spread panic. But I guess it doesn't get enough clicks, and it's too late now to point out that the emperor is naked...

Based on your earlier misreading of the data, please notice that you have a conclusion that you're projecting into the data. As they say, you can torture the data until it speaks. But it's very easy to look up yearly flu deaths (~40k, up to 60k in the exceptional 2017-18 season), and see that there's an order of magnitude difference between that and what we see here.

In other words, Is the emperor actually naked, or are you simply fishing for the conclusion you want to hear?

I don't think you need to outline a number in advance (if you are familiar with the UK's response, one very big problem has been that forecasts of deaths has consistently undershot estimates even when England wasn't in lockdown and had no mask mandate). We had excess mortality around 10k/week in April, that probably did require a lockdown (to be clear though, if the govt had handled the care home situation better than maybe not necessary) but the bump in December last year was very minor. I am not sure what you mean by who...whoever gets Covid.

Yes, no issues at all. I support lockdowns and have had relatives who became very ill because of lockdowns (delayed/ineffective hospital treatment, I am actually not well myself either for the same reason). I think trying to argue that someone should only support/oppose something because they have something to gain personally is the reason why politics in some countries is totally fucked (the UK is probably the worst for this, somewhere around 2017/18 most of the media became tabloids and every politics story contained some hyperbolic bullshit from some random person claiming their life would be destroyed by X minor policy change...it is just a waste of time and energy for all involved).

Reply on news.ycombinator.com