Deaths due to Covid-19 compared with deaths from influenza and pneumonia | Hacker News Reader

Deaths due to Covid-19 compared with deaths from influenza and pneumonia

ons.gov.uk

54 pointsDanBC114 comments

I think it is important to keep in mind that Covid-19 is still contained to some extent due to various measures. The influenza numbers are for the most part without any behaviour changes or rules (the influenza season ended roughly when Covid started to spread). The Covid-19 deaths would be higher if it were able to spread unhindered through the entire population.

And even in this somewhat unfair comparison Covid caused three times as many deaths.

Influenza is contained because there is some form of herd immunity already. Remember COVID-19 is a novel coronavirus, which is why it's spread so much so fast, no one is immune at first. But to suggest that the covid death numbers will be this high year after year after year is unlikely in my opinion, as is shown by the fact that many places with big first waves (north italy, sweden, london) don't have a second wave (especially in death numbers).

Influenza is also new every year. There have been other Coronaviruses before.

Not exactly new. Partial immunity to many strains of flu is common and widespread. Not so much to SARS family coronaviruses. (Which are very different from other coronaviruses, much more than flu strains between each other.)

Nevertheless, isn't the latest finding that many people already seem to be partially immune - perhaps because there are some similarities to other Corona viruses that the immune system can latch on to?

London was under lockdown until June, everyone is already acting like it will be locked down again in the next week or so, and has had strict mask and hygiene and social distancing rules for six months, had a government direction for everyone "who can" work from home to do so for ~4.5 of the last six months (it was briefly rescinded then reinstated), crash built hospitals in convention centres to try to handle overload from hospitals and still is currently at the same new case rate as April[0]. Fortunately death rates are still fairly low. I don't know if a vaccine/other interventions will prevent it being a yearly disaster but it certainly currently is.

[0]: https://data.london.gov.uk/dataset/coronavirus--covid-19--ca...

My friends in London tell me mask wearing is not widely respected or enforced, which is contributing towards the next lockdown.

It's, also, important to keep in mind that Influenza has a vaccination, which is very cheap and many elderly people take it. Yet, many die from it (and it doesn't make any headlines).

We are at the end just comparing apples to oranges. It would be really interesting to make such comparisons if we hadn't any vaccination for Influenza.

Also Influenza is not a novel virus, so if 40k people die of Influenza this year, doesn't mean that it couldn't have killed more people, because in fact, there was another 40k who died the year before and therefore don't die again, and another 40k who died 2 years ago, etc.

Basically Influenza is endemic and the deaths are evenly distributed over long periods of time, whereas a novel virus will obviously have a huge initial death count in year 1, but subsequent years will see equally many less deaths logically.

You can approximate the effect of a new Influenza without vaccines if you look at the numbers for the year 1918. That definitely made headlines (In this special case that is an interesting story by itself). But the effect influenza had in 1918 is not comparable to the normal flu.

So if people insist that Covid-19 is just like the flu, agree with them and remind them of a new flu without vaccines, just like the Spanish Flu.

It’s also contained because unleashing it would totally overwhelm almost any nation’s hospitals, cause a cascading death toll because no treatment could be given.

>It’s also contained because unleashing it would totally overwhelm almost any nation’s hospitals, cause a cascading death toll because no treatment could be given.

This is clearly not true because it didn't overwhelm Sweden's hospitals, or Japan's, or South Dakota's, even though they didn't have lockdowns, just voluntary social distancing.

Not as clear as you make it sound: Swedes did change their behaviour/travel patterns in response to Covid-19, the Japanese instituted mask etiquette a century ago, and no one lives in South Dakota.

No, it's pretty clear. Relatively minor changes in behaviour (Swedes were famously out chatting in cafes and shopping at a time when many other places were under universal curfews and business closures) cannot explain the huge deviation from what the models predicted, at least not without making the models entirely unfalsifiable. Meanwhile masks are useless. Look at the graphs of almost any country that has a mask mandate and when that mandate was introduced - no inflection point is visible. Nowhere! If masks worked, at least some places would show a clear change in growth rate $incubation_period days after the mask mandate was enforced, but it never happens.

As for "nobody lives in South Dakota", well, tell the people who live there that.

And don't forget Belarus. The usual excuse for that is "you can't trust anything out of Belarus", although of course many people live there and no overflowing hospitals were reported by any foreigners there either.

There's a very clear and visible pattern here, which is a really deep desire to believe that epidemiology as a field knows what it's doing, that the models work, that lockdowns weren't useless. But the data is as clear as day. No correlation between lockdowns and health outcomes. No lockdowns yet no ICU exhaustion in Sweden despite Sweden having one of the lowest ICU capacities in Europe - the exact opposite of what we were told to expect. Mask mandates have no visible impact on the spread of 'infection' (PCR positives, not quite the same thing).

I intentionally mentioned behaviour changes first in my comment, we can't fully separate rules and lockdowns from voluntary behavioural changes. In the end all that matters is the changed behaviour, not whether it was enforced or not.

Full lockdowns are also not the best way to handle increasing numbers now, they were necessary earlier this year because we didn't know much about the virus and which situations are the most dangerous. With the knowledge we have today we can use more specific interventions to contain the virus.

Well it probably depends on the amount of IC beds / ventilators / trained personnel available per location. I know that here in The Netherlands they narrowly avoided a triage situation where your condition decided if were going to get treated.

A second order effect that might not be obvious is that during the first wave a lot of non-emergency procedures were postponed, a lot of these non-emergency procedures can still be quite urgent and can have negative consequences for the people who had to wait.

Maybe in Netherlands non-emergency treatments were postponed, but in Hungary even some emergency treatments were deprioritized by the government. I tell my parents every times I speak to them to be very careful to not get sick, as they can’t expect the same level of treatment that they had a year ago.

Att the start of the pandemic Sweden had the next-lowest number of ICU places per capita in Europe (Portugal had the lowest). The capacity was doubled, field hospitals were built... and not used since the hospitals could handle the inflow. This was partly due to some very controversial and in my opinion unethical guidelines which mandated morphing instead of oxygen for certain groups of elderly (i.e. involuntary euthanasia) but it is unknown if those policies had much effect on the park ICU occupation.

Apart from some bad mistakes in the beginning where the infection was allowed to spread in care facilities Sweden has weathered the epidemic (as in the local version of the pandemic) no worse than surrounding countries and certainly no worse than e.g. New York.

no worse than surrounding countries

Deaths per 1k population:

                total       last month
    Sweden      0.59        0.0058
    Denmark     0.11        0.0053
    Finland     0.063       0.0016
    Norway      0.051       0.0020

I’m tired of seeing scoreboard style comparisons everywhere as some sort of gotcha. Any scientific analysis worth its salt will look into the values to start explaining the numbers and put it all into context.

For instance, can we explain where Sweden’s deaths occurred? Large care homes (100+ residents) that were ill prepared? Are they expected to have a large amount or average amount of excess deaths at the end of 2020? Why is it tracking average and why are some lockdown heavy countries tracking much higher (because lack of health care due to shutdowns?) What even represents a covid death or case?

What makes presenting a list of numbers a “gotcha”? It’s a better starting point for discussion than a random unsupported assertion. I would interpret a complaint about introducing numbers and measurement as evidence that the complainant has a weak position that can’t stand objective scrutiny. Numbers aren’t everything, but they’re definitely a something...

> For instance, can we explain where Sweden’s deaths occurred? Large care homes (100+ residents) that were ill prepared?

A large problem in Sweden is that many of the staff in care homes do not speak Swedish or don't speak/read it well. This has led to the instructions on how and when to use PPE not being followed or not being followed well which led to a large number of infections in care homes [1]. Since people in care homes tend to be part of high-risk groups for severe COVID-19 this had led to a large number of deaths which could have been prevented. I stated this in the original message ('Apart from some bad mistakes in the beginning where the infection was allowed to spread in care facilities').

[1] https://www.sydsvenskan.se/2020-07-10/svag-svenska-kan-ha-pa...

> This was partly due to some very controversial and in my opinion unethical guidelines which mandated morphing instead of oxygen for certain groups of elderly (i.e. involuntary euthanasia)

This is an extremely serious accusation. I live in Sweden and have not seen any reference to this practice mentioned in media.

I saw something related to it in two outlets, the BBC and YLE (finnish news).

https://www.bbc.com/news/world-europe-52704836

https://www.wsj.com/articles/coronavirus-is-taking-a-high-to...

https://yle.fi/uutiset/3-11433976

The intent was to give palliative care, not literally murder people . Although this source does sound the alarm that the line is thin.

In essence, palliative care for cancer does not translate into the same effect with COVID-19.

https://www.dagensmedicin.se/artiklar/2020/06/15/palliativ-v...

Seriously?

http://www.aftonbladet.se/nyheter/kolumnister/a/dO3ddB/en-pa...

http://www.aftonbladet.se/nyheter/a/GGk664/de-nast-intill-av...

http://www.aftonbladet.se/nyheter/a/4q9V1g/kritiken-mot-aldr...

https://www.dagensmedicin.se/artiklar/2020/06/15/palliativ-v...

https://www.expressen.se/debatt/min-pappas-dod-i-covid-19-bl...

https://www.aftonbladet.se/ledare/a/Ln0r9Q/morfin-istallet-f...

https://www.exakt24.se/lakare-upprord-efter-lackt-journalant...

https://www.exakt24.se/lakare-forskrackt-dodshjalp-istallet-...

https://lakartidningen.se/aktuellt/nyheter/2020/05/geriatrik...

https://www.varldenidag.se/nyheter/lakare-kraver-syrgas-till...

https://www.riksdagen.se/sv/dokument-lagar/dokument/skriftli...

If you have not heard of this it is because you're following the wrong media, clear and simple. Even Aftonbladet published on this, there was an open letter in Läkartidningen (doctor's journal), Dagens Medicin wrote on it, the 'alternative' media were all over this, questions were asked in Riksdagen.

Why did you not know? Also, why did you not simply search around a bit to verify the statement before posting your reply?

Applying palliative care on the recommendation of Socialstyrelsen is not euthanasia.

Japan has a culture of wearing masks when conditions prevail (normally pollution related but also disease). They also tend to be not into physical contact in social settings, and are more likely to be in compliance to societal laws.

That's an additional factor if you get to the limits of your healthcare system. At that point Covid gets more lethal because many people don't get adequate care, and every other critical illness also get more lethal because the hospitals are overflowing.

Sweden showed that not to be the case. It's the only real control I know of for the hypothesis that you present as fact.

This is pretty damming to anyone who claims "its just the flu":

“More than three times as many deaths were recorded between January and August this year where COVID-19 was the underlying cause compared to influenza and pneumonia."

“The mortality rate for COVID-19 is also significantly higher than influenza and pneumonia rates for both 2020 and the five-year average."

“Since 1959, which is when ONS monthly death records began, the number of deaths due to influenza and pneumonia in the first eight months of every year have been lower than the number of COVID-19 deaths seen, so far, in 2020.”

The number of deaths from flu, pneumonia, and Covid are about 50% higher than the year 2000 (Jan-Aug) and about 25% higher than the year 1983. Do those numbers justify draconian shutdowns and bankrupting half of humanity? I'm not so sure. In the early days, we believed that Covid was 10-20 times deadlier than the flu, and that's what was used to justify the draconian measures.

Note also these numbers are for the UK, one of the top worst countries hit in terms of deaths per million in the world.

Also, Covid does not seem to have the same seasonality that the flu has, which peaks in cold winter months. It may therefore not be altogether accurate to merely compare Jan-Aug. We may need the whole year's data to better compare.

> In the early days, we believed that Covid was 10-20 times deadlier than the flu, and that's what was used to justify the draconian measures.

I'm looking at this [1] data, Table 4, specifically the row for April.

Deaths in April

Pneumonia+Influenza, 5yr avg: 2,354.8

Pneumonia+Influenza, 2020: 1,933

COVID: 29,128

[1] https://www.ons.gov.uk/file?uri=%2fpeoplepopulationandcommun...

April was peak, pre-lockdown, full-blown pandemic conditions. That's a 12x difference. Bad. Shit.

The influenza season in the UK is typically October-May with the peak period December-March. So the 2020 number for influenza is a bit flawed given that it's missing 25% of the typical peak period and 3/8 of the whole influenza season.

You'd have to compare infection rates for the season peak to matter. Infection rates for influenza probably being an order of magnitude worse than Coronavirus in April, probably. But whatever, the mortality rate data is there in the table, too, if you care to look. Influenza death rate in January 5yr average was 4218, which is about 1.8x worse than April. That absolutely does not make up the 12x difference here.

The numbers for Jan-Aug don't include December for any previous year, so there's no flaw there.

The flaw is in comparing COVID-19's "season" with influenza's season while leaving out a full quarter of a typical influenza season.

To be clear: the flaw doesn't flip my interpretation of the results or anything, but it's still a glaring flaw that doesn't seem to be addressed in the article.

Okay, now open Table 5 on that same spreadsheet. And take a look at the numbers for the year 2000.

# of deaths Jan-Aug (due to either flu, pneumonia or Covid)

2000: 40,436 2020: 14,013 + 48,168

Yes, this is significant and certainly Covid seems to be able to spread across a vulnerable population much more quickly than the flu but it doesn't seem to be orders of magnitude deadlier on balance. I get why people thought it was 10-20x in the beginning but it seems as we've learned more it is not the plague we thought it was.

Also note the numbers for August have already flipped so that 4x more people are dying of flu and pneumonia now than of Covid.

When you average over a long timespan that includes unprecedented global steps to limit this thing. Also, essentially nobody was infected with COVID in January, so you are just flat out being disingenuous. You could as well go back to November if you really want to make the numbers look good for your argument.

> I get why people thought it was 10-20x in the beginning

It was 10x in the beginning. It averaged 12x for an entire month. It was 15x for numerous days in a row, and this is despite it infecting far fewer people than the flu. It was why we had a massive global lockdown, mask mandates, social distancing, the whole shitshow, in case you missed that.

At this point I am inclined to disengage because you, specifically, have been all over every single COVID thread with the same misinformation and distortions. Frankly, it's tiring. It doesn't matter what the numbers are anymore. Never did. "Just a flu". smh.

You can perform the same exact calculations over the numbers in Sweden, where there were no mask mandates and no shut downs and no school cancellations and they had the exact same epidemic curve as the UK. In fact they did better than the UK on a per capita basis. Like the UK, Sweden had huge outbreaks in vulnerable populations which occurred in nursing homes and assisted living facilities. Once they got those outbreaks under control, their death rates plummeted. So two countries with vastly different approaches had the same outcome. That underscores the fact that Covid is more infectious than the flu and can spread more rapidly.

The virus is getting less deadly. You can have many hypotheses for why that is: better treatments, younger patients, viral mutations, or just better testing which is giving us a clearer picture of the actual threat.

But you can't deny that YTD, the data does not show a disease that is anywhere close to 10-12x the flu. The fact that the flu has already surpassed Covid in monthly deaths (despite being less infectious) by the month of August at a rate of 4x Covid should tell you something, and we're not even in peak flu season yet.

Like others have mentioned, if this disease is capable of spreading much more rapidly than the flu, then it may knock out many of the same people who would have been vulnerable to the flu in a subsequent year. It may be that better treatments cause the death rate to continue to plummet. Is it fair to compare a disease we have been fighting for thousands of years to a novel virus we had no idea about 8 months ago? Probably not. As best practices and treatments get established it may very well turn out that Covid remains less deadly than the flu in perpetuity.

> Covid does not seem to have the same seasonality that the flu has,

[citation needed]

Considering that covid-19 is a pandemic, and has not yet reached the point where its endemic, we can't actually know that.

> The number of deaths from flu, pneumonia, and Covid are about 50% higher than the year 2000 (Jan-Aug)

Correct, welcome to averages.

In april the total mortality was between 50 and 125% higher than normal (per week, looking at total deaths, not counting where covid was mentioned.) That peak was neatly 3-4 weeks after lockdown was imposed. So to answer your question "justify the draconian measures" do you want to re-run that experiment? because I don't like the look of those odds. Looking at the USA, where they've tried it, it doesn't look too hot either.

see raw data here https://www.ons.gov.uk/peoplepopulationandcommunity/birthsde...

> Do those numbers justify draconian shutdowns and bankrupting half of humanity?

These are the numbers with "draconian" counter measures. We can only imagine the numbers without them. And that is what should be compared with the numbers from flu and pneumonia. Luckily, we don't have these numbers.

The confusion lies in "just the flu". Actual Influenza is in fact quite dangerous.

However, we need to compare COVID-19 to Influenza to get some perspective. I would argue that Influenza is worse than COVID, for the simple reason that it kills more young people and children. The average age of a COVID fatality is higher than life expectancy. Influenza destroys more years of life.

We also need to make a distinction between the time frame in which the virus was new and killed a disproportionate amount of susceptible people, versus the time frame we are in now, where the virus is once again kills less people than Influenza - as the data clearly shows, though it isn't highlighted in the publication.

>However, we need to compare COVID-19 to Influenza to get some perspective. I would argue that Influenza is worse than COVID, for the simple reason that it kills more young people and children. The average age of a COVID fatality is higher than life expectancy. Influenza destroys more years of life.

I think that's a very good point. And also one of the reasons the perspectives differ so much.

However we don't know yet about the long-term effect of COVID on younger people. It could still very well lead to a decreased quality of life for young people as we've seen in other posts about scarring of the lungs and permanent fatigue.

Granted influenza can cause similar after effects (like pericarditis and permanent fatigue), but it appears to be less frequent.

> Actual Influenza is in fact quite dangerous.

Agreed, but we don't destroy economies over it, we just get on with things and accept it for what it is.

Except influenza is nowhere near spreading as fast. If let unchecked, many millions of people would be dead - this has been avoided only by control measures, and we're not past it yet.

If it's similar seasonality to Spanish flu, expect 4 waves. This is start of the second wave, likely biggest.

Sweden coped fine with minimal intervention from the government.

The rest of your comment is pure speculation. It now seems to be the given narrative that there will be a second wave despite coronaviruses having different characteristics from flu. I would argue that the current second waves in Europe are really the first wave continuing after lockdown restricting were lifted (plus a lot more testing). The number of deaths isn't rising much.

> Sweden coped fine with minimal intervention from the government.

Did it, though? https://web.archive.org/web/20201008174215if_/https://www.ec...

> If let unchecked, many millions of people would be dead - this has been avoided only by control measures, and we're not past it yet.

This is an unfalsifiable claim, considering that virtually every country has done something to inhibit spread. It's also meaningless, because nobody is saying that nothing should be done to inhibit spread.

The question is, what should be done? How many people-years can we realistically save, and at what cost? Every year, almost 3 million Americans will die of any cause. If we take a CFR of 0.4%, and a "herd immunity" threshold of 70%, the upper bound of deaths in the US from COVID-19 is under a million. The average age of a COVID-19 fatality is higher than life expectancy, which means that half of these deaths would've occured under any circumstance, in the near future. We can see high excess mortality right now, but we haven't seen how that translates into lowered mortality in the future.

Sweden would be the control for the hypothesis, and while it wasn't no measures, there was no lockdown.

Just look at the daily and monthly numbers. April was absolutely horrific, with daily deaths spiking at 15x and April at more than 10x. Lockdowns flattened the curve, care improved, mask wearing improved, and overall better hygiene, particularly at nursing homes. The overall death rate is going down because the age demographic of infection has changed as well. Basically, it's hitting more younger people and older people are taking more precautions.

COVID is not to be trifled with.

>Lockdowns flattened the curve, care improved, mask wearing improved, and overall better hygiene, particularly at nursing homes.

Comparing places with and without lockdowns there's actually little evidence that lockdowns led to a reduction in mortality: https://www.jpost.com/israel-news/social-distancing-more-imp.... What was associated with a reduction in mortality was social distancing.

This is a really difficult causal inference to make. There are so many confounding variables, from quality of healthcare to age demographic of infection, etc. It's also noisy. A nursing home getting infected and many older folks dying can completely spike the numbers. So I pretty much don't trust their statistics, especially if they are reporting 3 significant figures like "7.49 days doubling the number of fatalities". Even just misaligning the starts of the exponential growth phase of curves (does it begin at 1 case per 100k, 5, 100?) can throw this way off.

Lockdown was due to being unprepared. It's a long time since nearly all nations opened up again, though measures vary wildly and is researched on.

Surely it's much more complicated than this? Assuming the COVID-19 assigned deaths are accurately labelled, how many COVID-19 deaths would have been pneumonia/influenza deaths? How many pneumonia/influenza deaths from 2021, 2022, ... have been brought forward to 2020? How much has the lockdown contributed to poor immune systems? Which age groups are affected more, were they already at life expectancy? It's obviously not the flu but it seems clear it will be years before we understand what the impact really is.

From TFA:

"The analysis of COVID-19, influenza and pneumonia deaths in this bulletin focuses on deaths where these conditions were the underlying cause of death (deaths “due to”), rather than deaths where the conditions were either the underlying cause or mentioned as a contributing factor (deaths “involving”). Data for deaths where COVID-19, influenza and pneumonia were a contributing factor are available in the accompanying dataset."

They also point to this document for more details: https://www.ons.gov.uk/peoplepopulationandcommunity/birthsde...

I made a number of points and I'm not sure which point this quote is meant to address as it doesn't appear to address any of them?

There is no way to know if somebody would have died from the flu/pneumonia if they didn't first die from COVID-19. Excess death calculations provide a hint but are confounded by additional variables created by the lockdown (eg people not getting exercise, not getting checkups, not eating as well, depression, etc.)

Furthermore excess death calculations can only take into account < one year of data at present. We don't know if it takes off 0.5, 1, 2, 3 or more years of your life expectancy ie. would you likely have died of a respiratory illness in 2021, 2022, etc. - these are important questions, I've seen estimates saying 8 years - which is a very compelling argument for lockdown - but would 0.5 or 1 year have the same gravity?

   > This is pretty damming to anyone who claims "its just
   > the flu"

Any discussion about the COVID deaths that does not take age into account is not worth the time. For the younger population it may actually look better than the flu.

Most of "the younger population" you refer to doesn't live isolated away from everyone else, so their own personal health isn't the only factor to consider.

But we need rules for the benefits of the society as whole not just for teens.

How did they determine the numbers?

This chart from Sweden says its not worse than influenza: https://pbs.twimg.com/media/Ejd0IpdVcAASTqR?format=jpg&name=...

Looking at the source data, the numbers seem roughly right: https://knoema.de/atlas/Schweden/Sterblichkeitsrate

I don't know what to think anymore.

Well the UK chart also shows that:

- Influenza and Pneumonia are counted as one, because the majority of Pneumonia deaths are due to Influenza

- Influenza and Pneumonia deaths have been SUBSTANTIALLY lower ever single month in 2020 than the five year average

- Of the deaths where both influenza and pneumonia, and COVID-19 were mentioned on the death certificate, the underlying cause of death was counted as COVID-19 in 95.8%

So basically this alone already highlights the gross overestimation of COVID-19 deaths. I feels plain wrong to attribute almost all deaths to COVID-19 when the person also had Influenza when it is evident that we've seen so many less Influenza deaths this year. Clearly we're not giving Influenza enough credit in being the cause of death which is wrong IMHO.

Also further below when you look at the age distribution, the vast majority of deaths happened in 80+ year olds, even more in 84+ year olds, which makes one really think. COVID-19 is not nearly as much of a threat to the vast majority of the population as we are made to believe. It seems the people who are dying from COVID-19 are those who have reached an age which is well beyond the average life expectancy of a first world country.

Not saying that COVID-19 isn't a threat, it is, but it's not like the Spanish flu which was killing predominantly young people or the plague which was wiping people of all age groups.

EDIT: Typos corrected.

It's also well worth factoring in what Amnesty have just reported about the state of care homes, their residents access to medical care and DNARs in the UK during this period.

https://www.amnesty.org/en/latest/news/2020/10/uk-older-peop...

> It seems the people who are dying from COVID-19 are those who have reached an age which is well beyond the average life expectancy of a first world country.

Some data in support of this:

„According to the CDC’s data, the survival rate for COVID-19 is as follows:

0-19: 99.997%

20-49: 99.98%

50-69: 99.5%

70 & over: 94.6%

Life expectancy in the USA per CDC: 78.7 (2018)“ [0]

[0] https://twitter.com/drsimonegold/status/1313948710523424768?...

I belive the average age of people who die from COVID is higher than the average of age of people who die

Doesn't it make sense that influenza numbers would be far lower this year given all the health security measures employed for covid?

It would, but when you look at the data for the last 3 months Influenza has killed more people than COVID-19, so apparently that logic is not true.

Note that despite what you might think, Swedes do socially distance. So if they have what looks like a bad flu season despite all the changes to everyday life you can assume that it's at the very least worse than influenza for which no such restrictions are done.

I've seen a joke that is along the lines of, "Swedes are looking forward to the relaxing of the 2m social distancing rule, so that they can resume the 10m social distancing of the past."

I'm of the opinion that comparing countries is mostly a pointless exercise, because there are so many variables involved.

Agreed. But you can compare Sweden to itself and it's pretty clear that Sweden had a much stronger reduced social life during Covid compared to the mentioned flu seasons.

Are there instances of hospitals corridors turning into morgues in recent years due to Influenza?

If Covid-19 is not worse than the Influenza we should have similar impact due to influenza.

I was not able to confirm that Influenza creates similar strain to healthcare infrastructure as Covid-19 which makes me think that the Covid-19 situation in Sweden is more nuanced than the numbers say.

Afaik such things happen, although I don't know about Sweden. But many reports that induced panic were just things that happen from time to time, only the population was not used to seeing it (no reporting). There were other aspects, too, like morgues closing out of fear or because of regulation (can't handle the dead because of risk of infection).

Edit: HN doesn't let me post more comments atm, so here is my reply to "needs citations":

They "need" citations? Why do they need them, because it is my duty to convince you? Why?

If you are interested, you can ask if I have citations, or you can google yourself. But there is no "need" as such, because it is not my duty to convince you.

These things made the rounds at the high time early in the pandemic. Maybe if I have time I'll google them later. Maybe you could google for "Morgues closing corona" or something like that.

New York especially afaik has a problem with Morgues, even in normal times there are not enough of them, and it is not easy to build more because of regulations.

Strong claims, all these points need citations. It would be interesting that medical professionals did not notice that if true.

Even if this data was correct, the difference is that in 2020 there were pretty stringent measures in place to reduce the spread of Covid (which were not in place in the past for major influenza outbreaks). We do not have a clean counterfactual to what the number of deaths would be in the absence of those measures, but the height of the outbreak in Italy can perhaps serve as some approximation, and there the excess mortality was orders of magnitude above previous years.

Not sure if you misread, but MrBuddyCasino's comment was about Sweden.

Have a look at those dates along the bottom of the charts.

COVID dates are in spring, Flu is the middle of winter. TBF we don't have data on what COVID is like in the winter so the jury is still (a little bit) out on what that looks like but this comparison is misleading.

I bet if you plotted Flu for those dates (i.e. in spring) you would see significantly fewer deaths than from COVID.

Doesn't seem right to me. Euromomo does show a bump for Sweden, though it is still smaller than some other countries that did lockdowns:

https://www.euromomo.eu/graphs-and-maps

Also, several countries show no excess mortality.

Compare Sweden to its immediate neighbors like Norway, Finland, and Denmark and the cost of Sweden's failed response becomes pretty obvious.

They didn't have to shut down the country though and since August there's 6 or less deaths a day.

https://www.worldometers.info/coronavirus/country/sweden/

Norway has a death every 3 days or so right now.

This is an oversimplification. What specifics do you claim puts Sweden closer to the other Nordics than other European contries?

For example, Sweden had school breaks around the time of the large outbreaks in northern Italy and Austria, leading to many ski vacations to those areas. Did the other Nordics?

To be clear, I don't proclaim this little example to be an explanation for why Sweden had more cases, only to shine some light on how complicated things can be. It is not clear that Sweden's response "failed" and it certainly isn't "pretty obvious".

> For example, Sweden had school breaks around the time of the large outbreaks in northern Italy and Austria, leading to many ski vacations to those areas. Did the other Nordics?

Yes, Norway and Denmark had ski vacations at the same time with a lot of travel to the Alps. All three countries had very similar initial conditions and trajectories for the epidemic until early April. The differences in outcomes were clearly related to policy decisions made in March and April.

You are cherry-picking countries based on geographic proximity, but that's not really a meaningful factor. Germany is right next to France, France had the stricter lockdown, yet Germany has no excess mortality.

Meanwhile, Norway has almost twice the per-capita GDP of the other nordic countries. Finland has no large metropolitan area. Sweden has more people in elderly care. Lots of factors to consider, but geographic proximity isn't really one.

> France had the stricter lockdown

You are begging the question. France might have stricter rules, but didn't enforce them. German police enforced (and enforces) the rules that they have in terms of contact tracing, quarantine/isolation, indoors mask rules, regional/local lockdowns etc far more rapidly and stringently than France, Spain, UK or US. (Not as well as East Asian countries, but in European comparison the stereotype of following orders kind of works.

Also Sweden does not only have geographic proximity to Norway and Denmark, but similar holiday travel behaviour, similar wealth level and business travel pattern, similar cultural mores of proximity. A far far more even comparison than with France, say. (And yes Finland has cities, ever heard of Helsinki with 1.4m people in the metro area?)

So we both agree that geographic proximity alone is not a good factor to compare by? Good.

It then makes sense to compare Sweden against all other European countries instead of just those few neighboring countries.

Compare the last few years then. Sweden did not have a bad flu seasons unlike the neighbours, who did.

Why show five graphs between 1990-2000, and zero for the next twenty years?

Those are the worst flu seasons (in recent history) in Sweden.

According to what? Even if that's the case, a glance at the second graph should show that skipping twenty years is intentionally presenting misleading information.

While interpreting this it's worth noting that the influenza season in the UK is typically October-May with the peak period December-March. So the 2020 numbers for influenza are missing 25% of the typical peak period and 3/8 of the whole influenza season.

I'm not a statistician or a scientist, but I don't it does any good to lump everything together over the past ~9 months and compare it to Influenza and Pneumonia.

Just think about it.

We barely knew anything about covid or how to treat it at the start of the year. Now, it seems like we have a better understanding and how to potentially treat it.

Compare this to Influenza/Pneumonia, which I'm just guessing we've been studying for much longer and already have treatments for.

You could probably add some sort of moving average 50 days and 200 days to provide a clearer picture of how we are currently doing.

Why make this point and ask for a moving average? The fact that it is new and unknown is part of why it is deadly, and that is reflected in the statistics. Why try to erase one takeaways from this set of statistics?

Like I said in another comment, I could have been more clear on my initial comment.

> Now, it seems like we have a better understanding and how to potentially treat it.

I think you're being a bit optimistic here. There've been some procedural improvements. We have a few drugs which appear to shorten hospital stay but not reduce death rate. We have dexamethasone, which gives something like a 20% death rate reduction in hospitalized cases, but is not otherwise super-useful.

There've been modest improvements in treatment, but no silver bullet yet.

I don't think you can lump everything together

Why not? Do you think the number of previously recorded deaths will suddenly decrease because we've found a better treatment? These people died, they should be included in the statistics.

I think the point here is that once we learned how to handle Covid, it will not be more dangerous than flu.

The main question for me is if someone who got Covid and recovered can get sick again. I haven't seen any answer for that question I could trust. There are some headlines from time to time "got Covid again after recovery", but given the quality of media reporting it means nothing.

I edited my comment. I could have been more clear.

I should have added that I think it depends on what the creator of the graphics/information is trying to show. If it's a pure death rate comparison, then sure you could lump everything together.

However, if you want to show a more real-time comparison, you choose to segment the data based on time.

> Compare this to Influenza/Pneumonia, which I'm just guessing we've been studying for much longer and already have treatments for.

Do we? It's mostly the same symptomatic treatment afaik, influenza antivirals are barely effective. The real difference is that there is a vaccine.

So did they test every dying person on influenza, and if they died, cause of death was "influenza"? What if they tested positive for both influenza and Covid-19, what cause would be listed?

I'm just not convinced the comparison is very meaningful, as I suspect many more people were tested for Covid-19 than for influenza. The testing strategy may have been completely different, making the comparison meaningless.

Very difficult to compare this data. Flu deaths will be down this year because of all the additional precautions taken around transmission, which may disproportionately affect flu virality (ie. perhaps COVID still manages to transmit despite precautions while flu does not.) Then if you say okay compare a previous flu year, you’d have to compare with the same conditions, which in a typical year would be no masks, barriers, social distancing, etc. And I think we all know how much worse COVID would be if we weren’t taking any precautions - we were warned for months about it.

So I’d imagine the initial observations that COVID is about 3-4x as deadly as the flu (and not 10x+ as some of these numbers suggest) are probably accurate enough.

Even just comparing total number of deaths this year to last year is pretty gruesome. From the same website as the original article - https://www.ons.gov.uk/peoplepopulationandcommunity/birthsde... - the second last paragraph of section 2 reads:

"Looking at the year-to-date (using the most up-to-date data we have available), the number of deaths up to 25 September 2020 was 453,771, which is 53,888 more than the five-year average."

That’s just in England and Wales.

Shouldn't the comparisons be based from percentages? Also - even with that I'd assume that the more the people, the more dynamic world we live in then the rate of transmission will increase (then the more deaths)

Interesting that it's not a full year. I wonder if the numbers will be significantly different after more time has passed (i.e. 'the flu season')

No data from Ch1na?

Some things to consider from looking at the raw data:

The 5-year average minimum of deaths from "Influenza and pneumonia" is about 50 cases per day. The average for 2020 is about half that. Adding up COVID fatalities and "Influeanza and pneumonia" after July comes to about 50 cases per day. In other words, there is no excess mortality from COVID-19 since July.

> there is no excess mortality from COVID-19 since July

Correct, and that correlates neatly with the number of recorded cases of covid19. One could even draw the conclusion that if less people have covid-19, less people are likely to die from it.

The death rate has plummeted not merely the infection rate. Even where "second waves" are occurring around Europe the death rate for this second wave is substantially lower.

https://metro.co.uk/2020/09/20/second-wave-of-cases-not-caus...

yes, from this we can conclude one of two things:

1) covid is causing less acute problems 2) The percentage of daily tests that are positive is much lower, meaning that there are nowhere near as many uncounted cases in the wild.

I know which my money is on. Its also easy to prove.

> Correct, and that correlates neatly with the number of recorded cases of covid19.

Not quite. The number of recorded cases in the UK from one month ago is equal to the number of recorded cases at the height of the first wave, yet the fatalities are 10x lower.

I do expect excess mortality due to COVID-19 in the coming months, but not to the same extent as during the first wave. We need to re-evaluate the risk profile based on those numbers, not the numbers of the first wave.

> One could even draw the conclusion that if less people have covid-19, less people are likely to die from it.

One could also draw the conclusion that the more people have already died from COVID-19, the less people are likely to die from it in the future, because fewer susceptible people remain in the pool. Unlike with Influenza, it takes many decades of lifetime to develop a serious risk of dying from COVID.

Largely because we locked down the entire country and had largely stopped community transmission by then I imagine. Now that lockdown has been lifted and what restrictions there are are increasingly being ignored we are seeing increased community transmission, a concomitant increase in hospital admissions and should start seeing an increase in deaths over the next couple of weeks. Hopefully this wave of deaths will be lower because we are better prepared and have better treatments, and also because the rise in infections is being driven by young people.

> a concomitant increase in hospital admissions and should start seeing an increase in deaths over the next couple of weeks.

For the past couple months people have been predicting an increase in deaths to follow the increase in cases, but it hasn't happened.

Yeah but there wasn't really an increase in covid related hospital admissions until recently. Given the issues with the uks testing (and reporting) regime it's rather difficult to draw any conclusions about infection rates from the most widely cited set of figures which are numbers tested (or tests mailed out, or tests manufactured or possibly some other definition of test known only to Matt Hancock). The ONS estimates based on random sampling however show that the infection rate only started increasing again at the beginning of September, which explains why we've only seen hospital admissions rising over the last couple of weeks and deaths haven't climbed much yet. Also, so far, it seems not to have started spreading through care homes again like earlier in the year, which is one of the things which seems to have driven up the death rate so high.

Thanks, I was already browsing the data set to look at the same thing. When, as a statistician, you already observe that "In 2020 deaths due to influenza and pneumonia were consistently lower than the five-year average in all months from January to August" you need to include the total figure for deaths of Influenza+Pneumonia+Covid19 to arrive at a meaningful comparison against previous years.

Reply on news.ycombinator.com