Hacker News new | past | comments | ask | show | jobs | submit login
Excess Deaths in 2020 (kieranhealy.org)
157 points by panic 13 days ago | hide | past | favorite | 209 comments

I'm curious if there's a good way to breakdown how much is caused by COVID vs., say, caused by the lockdown (i.e. delayed elective medical procedures, suicides, etc.). For example the US suicide rate declined from 2018 to 2019, but went back up in 2020. Naively assuming that all of the increase was due to the effects of the lockdowns would attribute ~2,880 suicides to them (or more if in a counterfactual world the suicide rate was going to continue declining if there hadn't been a lockdown). Obviously the real causes of these suicides would be difficult to impossible to determine with certainty, making this a difficult path to explore.

In the UK a study predicts a 7.9-9.6% increase in breast cancer deaths within 5 years of diagnosis due to diagnostic delays caused by COVID. For other cancers their estimates vary, i.e. 15.3%-16.6% for colorectal, 4.8%-5.3% for lung, etc. thelancet.com/journals/lanonc/article/PIIS1470-2045(20)30388-0/fulltext

Basically I guess what I'm really trying to get at is that while COVID is certainly a terrible disease, there seems to have been little to no cost-benefit analysis performed by policy makers regarding the lockdowns and other attempts to combat it.

Suicides declined in 2020 [1]

Cost benefit analysis is important. But it's pretty clear the vast majority of excess deaths were covid related, because there's a very strong correlation between excess deaths and reported covid cases/deaths, even when you look at different states that had different peaks at different times, and no such correlation to other responses like different types of lockdowns. More delayed / indirect / long-term effects might be harder to measure though.


Weird, the first result when I did a search claimed it went up to 14.8 per 100,000 this year, from 13.9 per 100,000 last year according to the Google summary, but the actual page doesn't seem to actually have that data: https://www.americashealthrankings.org/explore/annual/measur...

Looks like the CDC data is still preliminary? So I guess we'll have to wait and see what the actual number is.

EDIT: Looking through the charts on that link it does list a 14.8 for 2020 suicides but doesn't seem to have a source?

Likewise, you should cite a source for your statistics in your original post.

Isnt that true during many major world events that are bad? I cant remember the study but I recall reading that during WW1 and WW2 suicides declined greatly even though it was a terrible time for many.

This is going to depend which place and group of people you talk about. Suicides of Jews in Germany went up by the start of WW2 - source is Richard J Evans; Third Reich Trilogy. He had no stats about non-jews. But, being oppressed and humiliated definitely made some Jewish Germans to kill themselves. (Convolute phrase, but plenty of them considered themselves Germans, so.)

And it was much easier to commit suicide by lettin other people kill you too.

America is might be different, because it did not happened on their soil.

> being oppressed and humiliated definitely made some Jewish Germans to kill themselves. (Convolute phrase, but plenty of them considered themselves Germans, so.) <

They — like my grandparents —considered themselves German because they were German.

The othering of Jews, that they are not really English/Spanish/German/Russian (to name a few examples), has been used to justify the persecution of Jews for the last 2000 years.

Because you don't think too much about your own suffering when you hear about the suffering of others... probably.

Yeah, but some claim overall "deaths of despair" went up [1], which is probably better to use here.

[1] https://www.marketwatch.com/story/deaths-of-despair-during-c...

For example, overdoses and suicides are often quite difficult to distinguish.

Suicidal ideation does appear to have increased: https://www.cdc.gov/mmwr/volumes/69/wr/mm6932a1.htm?s_cid=mm... But it doesn't correlate well to actual suicide rates. It's also quite possible that the emphasis on suicides has in fact raised awareness enough to prevent them. Or perhaps the way people are dealing with the stresses that lead to suicide has changed, resulting in that being expressed in some other way. Or a mix of all these factors.

That's an assumption made off incomplete data.

You could compare the excess deaths for New Zealand, South Korea, Australia and other countries that were able to prevent Sars-CoV-2 from spreading with strict lockdowns.

Diagnostic delays are not caused by lockdowns. They are caused by an overwhelmed health care system or the fear of infection, both of which are caused by high incidence.

You also need to consider the damage that long covid possibly do over the next 30 years. If only 1% of the working age population that got infected suffers permanent disability, then the cost-benefit analysis will show that a low incidence strategy is clearly the better strategy.

Here are the statistics for Australia. They had a strict lockdown when cases where high.


Or even by state. Hawaii had a strict lockdown but low excess mortality.

Arizona and Texas had weak lockdowns but high excess mortality.

There's absolutely no correlation between lockdown policy and mortality. For example, click the link above and compare Florida to California. Florida has been wide open since last summer (over 9 months now) and has performed about the same as California. And you can cherry pick Hawaii, but New York and New Jersey have been among the most strict lockdowns and they find themselves in the top 3.

It seems to help to be an island country or state though, or a geographically isolated region like South Korea or Alaska, as they can screen people coming into the country and there's not a lot of cross-pollination with tight border controls.

> Florida has been wide open since last summer (over 9 months now) and has performed about the same as California.

Remember that Florida has a significantly older population than California too. 2.1% of the population >85 years old vs 1.3%; 17.6% over 65 years old vs 10.6%. COVID deaths are very concentrated in the elderly. The CDC estimates that the 85+ age group has a 8700x higher chance an infection leading to death, and a 95x higher chance of infection leading to hospitalization, than the 5-17 age group. The article appears to be using non-age-adjusted excess deaths. So California and Florida being equal is a strong sign that California did significantly worse.




How much other causation have you looked into? For example population density?

Miami is a small city of 500,000 people.

The Miami metropolitan area population is 6,000,000.


...with a density of 1004 people / sq. mi., compared to the NY metro area with a density of 5313 people / sq. mi..



Yes, NY is clearly is densely populated. I guess I would look for the differences with California.

LA has more people (13.2 million) but appears to be less densely populated with 550 people per sq mile.

I didn’t follow California but I know Florida was offered as a place that not locking down worked.

I can look into it a little more. Florida has a lot of cases, and most came after the first wave. The same with Texas.

This is a falsehood that people have been promoting since last summer

From the WSJ on July 16th, 2020, someone trying to promote the idea that it was only a Blue state problem:


“Here are some real numbers to put things into context, and it's obvious the problem is certain incompetent blue state governors.

As of today:

Florida (population 23M) COVID deaths: 4,782

Texas (population 30M) COVID deaths: 3,561

New York (population 19M) Covid deaths: 25,014

New Jersey (population 9M) COVID deaths: 15,665”


I recently compared these numbers in early March I believe

Florida: 31,523 for 26,741 additional deaths

New Jersey: 23,521 for 7856 additional deaths

New York: 48,410 for 23,396 additional deaths

Texas: 45,274 for 41,713 additional deaths


NY: 48,410/25,014 = 1.9 times more death

FL: 31,523/4,782 = 6.5 times more death

Florida and Texas lead in deaths after the first wave

March 6, 2021

Florida: 31,523 for 26,741 additional deaths

New Jersey: 23,521 for 7856 additional deaths

New York: 48,410 for 23,396 additional deaths

Texas: 45,274 for 41,713 additional deaths

Explain Melbourne, Australia.

It stands to reason that actually locking people up in their homes for several weeks would have a significant effect on viral spread, especially if done early, in a wealthy country where people have a lot of individual living space.

Most lockdowns are not like that, for practical reasons. They are a random basket of restrictions that go into effect when the virus has already spread enough to cause great concern. At that point, it is hard to discern whether any of these restrictions cause a significant effect, much less a net benefit.

So you are saying ineffective lockdowns are ineffective.

I gave a specific example of Melbourne and I was implying a specific timeframe when their lockdown went into effect. They went from having uncontrolled community spread, with tens of thousands of infections, to actually zero. They did it by having a very detailed plan, strict enforcement, high compliance, and an excellent testing regime. They also had clear goals and thresholds for lifting restrictions and communicated them to the public.

I generally agree that ineffective lockdowns mostly just slow down the spread. The spread picks up again exactly in proportion to restrictions being lifted. However, effective lockdowns have actually been able to eradicate the virus completely from whole areas: from Melbourne, from Victoria, from all of Australia, New Zealand, and several Asian countries. You need to be able to quarantine international travel and have high compliance. The places where the virus have been eradicated have to remain on high alert and remain extremely vigilant until they can roll out mass vaccination. I spent 10 months in Australia during the pandemic, 99% of the time stuck in the ACT. The extreme vigilance was contact tracing, social distancing, and some restrictions. There wasn't a mask wearing mandate, ever, because they never allowed uncontrolled community spread.

The US and European countries cannot meet those criteria, so they have the virus simmering until they can roll out massive vaccination. The US and Europe were too slow to take the pandemic seriously because their political leaders rolled in BS and denial for months and seeded their idiot followers with denialism. It's too late for them, and the only way out is vaccination. And yet, some idiots are still rolling out conspiracy theories about vaccination and want to f** that up, too. It's so disappointing.

To add some stats. In Au and NZ we had a big drop in flu cases https://info.flutracking.net/reports-2/new-zealand-reports/

Not sure what that did to the death rate.

You also need to consider the damage that long covid possibly do over the next 30 years.

Yes, and also the long term damage of lockdowns. Some children will be permanently harmed by the loss of education, and many people will be dealing with the effects of social isolation for quite a while.

> caused by the lockdown (i.e. delayed elective medical procedures, suicides, etc.)

That's a dangerous way to phrase it. People would delay the elective medical procedures due to covid, even without lockdown. Overwhelmed hospitals would enforce the same. The covid/lockdown split of excess deaths is very complicated in itself.

I would argue that in a lot of places, including most of the USA, lockdown didn't really happen. We had these lame, unenforced stay-at-home "orders" which people ignored, mask "mandates" that were openly disregarded and even publicly protested, business "closures" that were unevenly enforced. I think you'd have a hard time correlating anything that happened in the USA in 2020 with a lockdown, since nothing was really locked. It seems like a huge exaggeration to call what most of the world did a "lockdown".

Any question of “what impact did a lockdown have” should start with clarifying what constitutes a lockdown.

From that definition, I think a lot of the conclusions people are drawing would be substantially less surprising.

If we’re going to call a polite suggestion to stay home and a potential for small fines which were never or rarely levied a “lockdown” then it’s not surprising it had little impact.

Correct, there's a really big difference between the "lockdown" in most of the US and for example Melbourne.

> Overwhelmed hospitals would enforce the same.

Not necessarily. There is always the option of triaging COVID victims into palliative care (which can be very fast and administered even by non-specialist personnel) and away from intensive care. While that has never been official policy due to the political optics, there have been claims that it was done on a hospital-by-hospital basis in Italy and Sweden for elderly victims. If the voting public were clearly informed that it was an option they could choose as an alternative to lockdowns, it may well have seen significant support from the population.

You think the public would have shown significant support for “just let the infected die”?

And you can add Canada to that list of countries that have had to do that, by the way. Like the others you mention it’s been a forced decision due to the healthcare system becoming overwhelmed.

The public may not have significantly supported that in the first lockdown last year, but there apparently has been support in later waves. In my country during the third wave, the health minister has pleaded that hospitals are being overwhelmed, yet at the same time the data he’s getting from mobile operators show that the public has returned to 2019-levels of mobility. The broad public, not just some outliers unwilling to observe the lockdown. The general attitude now seems to be that vaccination has arrived, so restrictions are no longer appropriate, and if you die before you can get vaccinated, tough luck.

In my country medical help got way more difficult to get in early 2020 despite having comparatively few cases. Hospitals were not overrun. It was purely because of lockdowns and constant scaremongering.

Since natural immunity is durable, an unmitigated pandemic would be over sooner. So in many circumstances the hospitals would be overloaded for less total time and thus lead to less non-covid excess deaths.

Here in Canada, excess deaths in 2020 for the 15 to 64 age range has been abnormally high, starting in March. Yet, unlike older age groups, the level of excess does not follow the pattern of reported COVID infections. That looks like deaths from lockdown and delayed medical care.

I personally know people who, for example, put off getting potential cancers checked out due to fear of covid. Similarly, I had to convince some people to go to the hospital for potentially serious non-covid symptoms who were afraid to do so because of covid.

Source: https://mpidr.shinyapps.io/stmortality/

This analysis presumes that hospital emergency and critical care capacity is flexible (it's not), that we can expect medical professionals to keep on performing at 100% in the face of overwhelming stress (they will not), and that Canada would somehow succeed at what Italy did not back in the first half of 2020 by doing even less mitigation. An argument in favour of no mitigation would have to demonstrate:

1. covid-related mortality and morbidity would not make up for any drop in other excess deaths

2. short-term stops on all non-critical care would, on balance, result in a significant drop in excess deaths over longer-term stops on a smaller percentage of non-critical care

3. hospital systems and their staff would still be intact after such a large wave. This may sound crazy, but while we're trading anecdotes you may be surprised at how many ICU nurses considered quitting because of how hard the latest surge hit. These are the people who dictate how many "ICU beds" we have, not the physical beds themselves!

4. that we can make any kind of meaningful analytical statement on covid mortality vs excess mortality for a specific age group given the lack of reporting data on both. In particular, data from coroner's offices on covid deaths has a huge lag time at present. Looking at case counts only is a poor proxy given you have to model infection -> death/recovery/discharge lag times, geographical variations and healthcare capacity as well. The HMD link merely provides total counts and proportions with no additional commentary. You can find more specific discussions on the subject [1], but they don't cover demographic breakdowns over time and necessarily maintain a high degree of uncertainty.

I don't think it's controversial to say governmental policy here has been sub-optimal at best. But to say that one extreme approach or the other is what we ought to have done based on woefully incomplete data is a stretch too.

[1] https://health-infobase.canada.ca/covid-19/epidemiological-s...

> This analysis presumes that hospital emergency and critical care capacity is flexible (it's not), that we can expect medical professionals to keep on performing at 100% in the face of overwhelming stress

No I'm not. Remember, what I said was: "So in many circumstances the hospitals would be overloaded for less total time and thus lead to less non-covid excess deaths."

Let's assume hospitals are severely overloaded, and are simply unable to treat COVID patients. The pedantic meaning of what I said will certainly be true: non-covid excess deaths will be less.

But the steel-manned version - less excess deaths in total - could still be true if lockdown drags on sufficiently long, and COVID-19 is sufficiently mild, and/or the marginal benefit of hospital care on survival is low enough.

> hospital systems and their staff would still be intact after such a large wave. This may sound crazy, but while we're trading anecdotes you may be surprised at how many ICU nurses considered quitting because of how hard the latest surge hit. These are the people who dictate how many "ICU beds" we have, not the physical beds themselves!

I'm well aware. Frankly, I suspect that problem is made worse by a long, dragged out, pandemic. People can often handle a short period of extreme stress better than they can handle a much longer period of high stress.

> Let's assume hospitals are severely overloaded, and are simply unable to treat COVID patients. The pedantic meaning of what I said will certainly be true: non-covid excess deaths will be less.

Even the pedantic interpretation does not necessarily hold. Though healthcare resources for life-threatening conditions are not strictly zero-sum, many of them would effectively be once covid overwhelms existing ICU and ED capacity. There are plenty of examples of personnel being pulled from e.g. cancer care or cardiology teams to help in covid wards. This directly translates to increased wait times for patients in those categories, much like would happen if people were putting off getting checked out themselves.

Moveover, hospitals and healthcare systems to not just "bounce back" after being severely overloaded. Staffing is already strained, and unless you ban clinicians from taking leave you would see cascading reductions in capacity after a couple months of being overloaded. This is to say nothing of the equipment shortages we had in the early days of the pandemic. Ultimately, patients who would require care for non-covid reasons would still suffer because of a massive backlog. Determining whether this backlog would be larger or more/less harmful when amortized compared to what we have now requires health economics analysis that I am certainly not qualified to perform.

> Frankly, I suspect that problem is made worse by a long, dragged out, pandemic. People can often handle a short period of extreme stress better than they can handle a much longer period of high stress.

This feels right to me too, though I've not the literature to verify it. However, it's an equally valid argument for strict short-term lockdowns a la New Zealand, Australia, Vietnam, Taiwan, etc. That is to say, it does not follow that not mitigating is the only reasonable strategy to prevent a dragged out pandemic.

A better (if grim) post-hoc analysis of Canada would be to compare the Atlantic provinces (less densely populated, strictly enforced interventions), the other smaller provinces (less densely populated, few interventions until recently) and the large provinces (densely populated, moderate but very poorly enforced interventions). My guess is that all three approaches will not be found to be equally effective.

On the other hand Covid-19 would have killed many more people directly if action hadn't been taken to mitigate it's spread. Probably exponentially more people, based on how viruses typically spread.

In other words, the comparison isn't between how many people Covid-19 did kill and how many preventative measures did. It's between how many it would have killed and how many preventative measures did. While there's a lot of uncertainty in how many people Covid-19 would have killed, it's still certainly higher than the preventative measures.

I think, very sadly, that India is giving us an insight now. It must be remembered that the situation in India is into a society with some immunity from earlier infections and also some immunity from vaccination (I think ~10% combined) which muddies the water a bit. But, still the picture of an unmoderated c19 epidemic is horrifying.

Something I haven't found is detailed breakdowns of who is actually being infected this time around.

India did have very strict lockdowns. Yet, very high rates of antibodies were also measured in some areas. Those strict lockdowns may have been successful in shielding some of the population from covid, and that part of the population is getting infected now. India is a very unequal society, with a very wide range of wealth present in the same areas.

In the US in some lockdown states there have been dramatic differences between infection rates between different socioeconomic levels. While in other states that were more open, that has tended to be less true.

> Probably exponentially more people, based on how viruses typically spread.

It will be decades before we know the answer to this, if we ever do. Trying to justify these lockdowns while in the middle of the storm is fools errand. Heads are running way to hot and brains aren’t thinking clearly.

What we’ve done over the last year represents one of the largest uncontrolled public health experiments ever performed. The best part is all the billions caught up in the experiment never consented and many don’t even know it was an experiment.

Society has never, ever had a pandemic plan that includes what we’ve done over the 14 months.

Only through the passage of time, when all the people who instituted these policies are no longer with us and all the costs of these lockdowns will be laid bare, will history truly be able to judge our actions.

(Personally I think history will take an incredibly dim view of lockdowns and those who enacted them)

The lockdowns weren’t unprecedented. We did something similar in 1918, and it backfired in the same ways. As a bonus, last time, a much more deadly strain emerged and ended up ripping through the population killing young adults and middle aged people.

If they hadn’t locked down for the first wave, they probably would have reached herd immunity before the second, more devastating wave. This time, we have a vaccine, so we’ll probably avoid that, at least.

That’s a long-winded way of saying that it’s a safe bet that people will take a dim view of the 2020 lockdown in hindsight.

>We did something similar in 1918,

I looked at this a while back and, as far as I can tell, there were relatively limited interventions in the US. If you read beyond the headline of this link [1], for example, it doesn't seem as if you had much in the way of workplace closures and, in general the "non-pharmaceutical interventions" were for a fairly limited period of time. This study concludes they helped but the data is inconsistent and hard to interpret.

[1] https://www.cidrap.umn.edu/news-perspective/2007/08/study-sa...

"The researchers report that all the cities used at least one of the three main categories of NPIs, and 15 used all three at the same time. The most common combination was school closures and public gathering bans, used by 34 cities for a median of 4 weeks. All the cities except New York, Chicago, and New Haven (Conn.) closed their schools for some period of time; the median was 6 weeks."

You are talking out of both sides of your mouth. How is it that people who think the lockdowns were net beneficial need to wait 'decades' to decide, but you're allowed to conclude now that they were a terrible idea?

It seems to me we should all state our positions now with the information we have available now, and let the future take care of itself.

Because he/she stated it as a personal opinion while the other person said it as a matter of fact that couldn't be questioned.

I'll go a step further and actually suggest that authoritarian lockdown measure haven't done as much as they've been touted as doing. This is pretty easy to see by comparing states with less lockdown measures vs. ones that went full economy shutdown. Most of the "open" states aren't any worse off than the ones who went lockdown crazy and in some cases they're actually better off when you look at spread and death.

> Most of the "open" states aren't any worse off than the ones who went lockdown crazy and in some cases they're actually better off when you look at spread and death.

This could also indicate that states where Covid-19 had a bigger impact responded by enacting stronger mitigations. What makes you prefer your interpretation?

>This could also indicate that states where Covid-19 had a bigger impact responded by enacting stronger mitigations. What makes you prefer your interpretation?

You can say that but there's no data that supports this at all.

What states are you comparing? Massachusetts and NYC locked down pretty hard and their post-April numbers look pretty good compared to, say, Texas, Florida, and South Dakota.

All of them, and since the start of this. I don't know where you're getting your data and why you've decided April numbers are a good measure of how states have done overall. Infection rate is less relevant than deaths anyway. If people are getting infected and recovering it's not really an issue.

NYC(not state) tops the charts for deaths per 100k. Notice how low FL is?


What you can easily notice is that strict lockdown measures don't seem to make any difference at all.

All while the topology in all jurisdictions is so different that it will never be possible to directly compare

In the US, 33-50% of the population apparently had antibodies pre-vaccination, and we apparently did a terrible job keeping it out of nursing homes (so the percentage that caught it that died was higher than it would have been across the whole population).

So, it would have killed 2-3x more people, max. There were strategies that were developed after the 1918 pandemic that probably would have saved more people than the lockdown at much lower cost, but they were ignored by the politicians. The basic idea was to focus all resources on protecting the vulnerable, and let the disease run its course quickly. Effectively protecting the vulnerable during extended lockdowns is extremely difficult, and the COVID lockdown was no exception.

> 2-3x more people, max

Rounding, that's possibly another 1-2 million people. Those are big numbers!

Also one quibble in that most of the US did not have an extended lockdown. Here in GA ours lasted basically a month. As others have said, people voluntarily stayed home, in part because the exponents of letting the disease "run its course" tended to exhibit a level of comfort with millions of deaths that perhaps is not widely shared in the population.

> So, it would have killed 2-3x more people, max.

You are very cavalier with somewhere around 1.1-1.6 million extra deaths, to date. That's your GOOD number?

If that's your anti-intervention argument, I'm unconvinced. Saving a million lives is worth a lot to me.

>In the US, 33-50% of the population apparently had antibodies pre-vaccination

Where did that number come from?

I have read more like 8% (https://www.thelancet.com/article/S0140-6736(20)32009-2/full...)

So - 2 to 3x... Nope. 8 - 10x.

What I want to know is why you have constructed this belief and want to propagate it when it is refuted by a google search in 10 seconds flat?

Your article is from last September, using data from last July. It only takes ten seconds to google, but I guess it takes a few more to check timestamps or think critically about the results.

Using more current antibody data, the CDC released updated estimates yesterday [0] that there have been ~115 million actual COVID infections in the US. That makes the 2x-3x number pretty reasonable (especially since herd immunity is supposed to kick in around 70%).

[0] https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/burd...

I wonder if the theory that previously infected people are reacting more strongly to the first doses is true?

I had no real reaction to the first one and a day of increased fatigue so far with the second.

Anecdotally, many more people are having stronger reactions to the second.

Not sure. Sweden had no strict lockdowns. No mask mandates (only recommendation to use one in public transport during rush hour). They also had a lot of deaths in nursing homes when the COVID first hit. Meanwhile my country (Lithuania) was in strict lockdown last spring and is still in lockdown from the beginning of November. It was made stricter mid-December and only recently some restrictions started to be dropped. So while in strict lockdown we managed to catch-up and surpass Sweden by the number of deaths per million.

Even still, you're balancing the number of people dying from the lockdown against the number of people who would have died if the hospitals would have been completely overwhelmed (a la NYC, Wuhan, India, and Italy in various times of the pandemic.)

The "what ifs" get harder to calculate.

> In the UK a study predicts a 7.9-9.6% increase in breast cancer deaths within 5 years of diagnosis due to diagnostic delays caused by COVID. [...] Basically I guess what I'm really trying to get at is that while COVID is certainly a terrible disease, there seems to have been little to no cost-benefit analysis performed by policy makers regarding the lockdowns and other attempts to combat it.

I see this line of reasoning a lot, even in mainstream media here in Sweden, but it seems flawed to me. As I see it diagnostic delays and delayed elective care is caused by a lack of lockdown or other stringent decease control measures. It’s when hospitals are overrun by people infected with SARS-CoV-2 that they have to delay other care. This happens due to rapid spread of the virus, which happens when lockdown is not imposed.

What’s the argument for the parents opposite line of reasoning?

As a datapoint on Sweden, we have a normally pretty well functioning breast cancer screening system where women between 40-74 are to be offered a screening opportunity every other year.

In 2019 more than 800 000 screenings were done (with 8288 cancer cases discovered in the country that year) and this system has been credited for pushing down the yearly cancer deaths due to earlier discoveries.

In 2020 there was 160 000 less screenings so we can probably expect an increase in number of deadly cases in the coming years.

(Source, https://www.dn.se/sverige/en-av-fem-kvinnor-screenades-inte-... )

Ok, but again: The primary reason for this is the significant spread of SARS-CoV-2 which has overwhelmed hospitals. This in turn could have been prevented with lockdown(s). Right?

>overwhelmed hospitals

Something that has never been properly explained to me is how "overwhelmed" the hospitals were in relation to their normally "overwhelmed" state.

I wrote a script which extracted time series data from one of the hospital bed data APIs in my country around April last year and ran it for a couple of months. I wasn't able to make any useful conclusions from the data other than more patients were coming than going for a certain time periods and some wards had beds reassigned to other wards. It would have been a lot better if I had access to historical data to compare it to previous years.

The hospitals weren't (all) overwhelmed but were focused on managing the pandemic. Fewer appointments scheduled, procedures pushed back, no weekend services, that kind of thing. Perhaps an element of patients huddling at home and being less proactive in their own care.

(Btw I'm not in healthcare. If anything my point is a logical alternative to the idea that overwhelmed covid wards could be the only thing affecting other services. But it's my impression of how things seemed to be, at least in the UK).

In Lithuania health care barely functioned in mid-2020 purely because of the anti-pandemic measures and restrictions not because of being overwhelmed. It's only by the end of the year when the big second wave hit they were stressed. Yet many months lost because of lockdowns and bad decisions.

The health impacts due to social/psychological effects alone of lockdown may be substantial. As an analogue, the annual spike of deaths around the holiday period increases “somewhere between 3% and 9% depending on the demographic group you look at, and between 1% and 10%, depending on the cause of death analyzed.”


> there seems to have been little to no cost-benefit analysis performed by policy makers regarding the lockdowns and other attempts to combat it.

What evidence do you have of this? This seems like a baseless criticism.

The obvious way to do this analysis is to look at all cause death rates in countries that locked down successfully, like china, new zealand, or australia, and see that there was no spike

Here is some more data.

Deaths from all causes has declined rapidly in the U.S. since January. The red dotted line is my best attempt to correct for reporting lag: https://i.imgur.com/z8M0k2h.png

Excess deaths exceeded COVID deaths every week of the pandemic: https://i.imgur.com/MeGMsRx.png

Total excess deaths in the U.S. were about 500,000 for 2020. The excess breaks down as 77% COVID, 16% non-COVID natural causes, and 7% external causes for the whole year. External causes include homicide, suicide, accidents, overdoses/poisonings.

The count of COVID deaths include COVID as underlying cause of death (91%) and COVID as a contributing cause (9%). Even when including the contributing cause, it looks like COVID deaths were significantly undercounted in the first wave March/April/May and somewhat undercounted in the later waves. https://i.imgur.com/YjZeI2E.png

Deaths from most major categories were higher or unchanged: https://i.imgur.com/TDGwVU9.png

Excess deaths by age group: https://i.imgur.com/oKda9Dl.png

Ages 0-14 had lower than normal deaths in 2020.

Non-COVID excess deaths by age group: https://i.imgur.com/8pJnauz.png

>Excess deaths exceeded COVID deaths every week of the pandemic: https://i.imgur.com/MeGMsRx.png

Not entirely sure how to parse that statement, but I think most people would interpret that as "non-COVID excess deaths exceeded COVID deaths", which doesn't match what's on your graph. E.g. for week 15, expected deaths = 55k, deaths - COVID = 63k and all deaths=79k.

The only truthful way to interpret your statement is that "excess deaths (including COVID deaths) exceeded COVID deaths". Which isn't that interesting of a statistic.

> "excess deaths (including COVID deaths) exceeded COVID deaths"

Yes, that's what I meant. I didn't think it was ambiguous. Lots of people think that COVID deaths were exaggerated. There are countries (with not very many COVID deaths) where the number of excess deaths is less than the number of COVID deaths. There are even countries with negative excess deaths for the whole year, even though they had some COVID deaths.

I'd expect lower or negative excess deaths in countries with very effective COVID responses because those responses can also prevent other causes of death. Masks, hand washing, social distancing also prevent the flu and other diseases. Lockdowns and working from home can reduce commuting and traffic-related deaths (though I heard in some places in the U.S. car deaths were up, possibly because drivers were driving faster and more recklessly on the lightly trafficked roads).

It is very interesting. It means that either covid deaths were undercounted, or something else unique to 2020 also killed a large number of people, or both.

It appears as though you're using the first interpretation, and assuming it's true. It is not. I agree it'd be very interesting if true.

The second interpretation (which is the one I called "not interesting") just means there was at least one non COVID death in each interval, and the total deaths never dropped below the average number of deaths. One could reasonably assume there was at least one non-COVID death per week, making that part of the statistic useless, and you could just state the total deaths.

No I'm using the second interpretation. It doesn't mean that there was one non covid death on each interval, but that there was one non covid death above the baseline of the expected number of deaths. Or in other words, deaths were always higher than expected, even after accounting for deaths attributed to covid.

The number of unattributed excess deaths is fewer than the number of covid attributed excess deaths, bit it should be zero, because the null hypothesis is that the only things killing people are "the same stuff as last year" and covid.

>there was one non covid death above the baseline of the expected number of deaths

While a true (and interesting) statement, that's not what the OP said. He said "Excess deaths exceeded COVID deaths", not "Excess deaths minus COVID deaths exceeded expected deaths". Again, the second statement is true (based on his graph), but not all all what he said.

> He said "Excess deaths exceeded COVID deaths"

Yes, and this is the right statement!

Excess deaths minus covid deaths should be zero, because the null hypothesis is that covid is the only thing contributing to the excess deaths.

That there are additional excess deaths beyond covid deaths is interesting (and means, as I originally said, that covid deaths are undercounted, or that some other unknown force is killing people).

The statement you provide as an alternative "Excess deaths minus COVID deaths exceeded expected deaths", doesn't make sense. Excess deaths are already above the expected deaths.

Let me run an example.

Last year, 100 people died each month. This is the expected number of deaths. This year, 150 people died each month. This gives us 50 excess deaths. 40 deaths are attributed to Covid. So, 10 are of unknown origin. This is interesting!

If the excess deaths - covid deaths exceeded expected deaths, you'd have to have 40 covid deaths, 110 deaths of unknown origin, and 100 "expected" deaths, which would also be quite interesting, but abjectly terrifying too (the death rate would need to more than double).

For international data, The Economist has the charts of excess mortality through 2020 for different countries:


The negative number of excess deaths in some Asian countries like Japan, Korea and Taiwan is interesting. I wonder if it's because of the very high rate of mask wearing that these countries have been known for since early in the pandemic. This might have helped reduce the number of deaths from influenza and other respiratory infections so much that they outweigh the deaths from Covid.

That's a great visualization, thank you.

FWIW, an economist published this paper recently estimating "the cost/benefit ratio of lockdowns in Canada, in terms of life-years saved, is between 3.6–282":

> Generally speaking, the ineffectiveness of lockdown stems from voluntary changes in behavior. Lockdown jurisdictions were not able to prevent noncompliance, and non-lockdown jurisdictions benefited from voluntary changes in behavior that mimicked lockdowns. The limited effectiveness of lockdowns explains why, after one year, the unconditional cumulative deaths per million, and the pattern of daily deaths per million, is not negatively correlated with the stringency of lockdown across countries. Using a cost/benefit method proposed by Professor Bryan Caplan, and using two extreme assumptions of lockdown effectiveness, the cost/benefit ratio of lockdowns in Canada, in terms of life-years saved, is between 3.6–282. That is, it is possible that lockdown will go down as one of the greatest peacetime policy failures in Canada’s history.


I thought that the purpose of lockdown (in the UK at least) was to prevent health care from being overwhelmed, at which point deaths would drastically increase.

So we don't really know how many people would have died without lockdown to make a comparison, but it would have been a lot higher.

If the health service is completely overloaded it means people would start dying from all sorts of other things too.

Am I missing the point?

I guess the point is that if lockdowns are ineffective, you are going to reach the overload point regardless. The fact that it did not happen means that I would not have happened anyway.

You’re treating healthcare overload as a binary condition when it is not. It matters how much your healthcare system is overloaded. If the system is 10% over capacity, then 1 out of 11 people aren’t getting proper treatment and run an increased risk of dying. If your system is 50% over capacity then only 2 out of 3 people are getting proper treatment. Neither is great, but they’re objectively not the same levels of risk or outcome.

Even if lockdowns are ineffective at preventing overload, it doesn’t follow that they can’t reduce the severity of that overload, or that the outcome would be the same either way.

That's quite a bit of stretch to say. There are thousands scenarios where medical system would collapse completely and people would either make it on their own or die. With secondary effect of everybody else with life-threatening condition dying too.

Would it be end of humankind? Nope, but I sure as hell don't want to be part of such selfish civilization just that cashflow is maintained

> but I sure as hell don't want to be part of such selfish civilization just that cashflow is maintained

so where does pharma with no liability, running their own trials and effectively being handed blank checks during the pandemic fit into this?

In the UK at least, you can see really clearly where lockdowns start and cases & hospital admissions drop dramatically from their peak to really low. Deaths went from >1000 to <50 per day. And when lockdowns ended they went right back up again.

> I thought that the purpose of lockdown (in the UK at least) was to prevent health care from being overwhelmed

Which never really pans out (field hospitals, hospital ship sent to NYC, etc.) - they got strained and perhaps local spots overwhelmed, but never doomsday scenarios we hear bandied about without any evidence. His point is that lockdown doesn't really change behavior and places that don't have strong mandates end up largely practicing safer behavior anyway.

I think there's also a strong argument that lockdown ends up decreasing the time spent outdoors, which promotes spread.

> Which never really pans out (field hospitals, hospital ship sent to NYC, etc.) - they got strained and perhaps local spots overwhelmed, but never doomsday scenarios we hear bandied about without any evidence.

"The thing you were trying to prevent didn't happen, therefore you didn't need to try to prevent it" is really curious reasoning. "It didn't happen, therefore it was because we did this things" is also a weak claim, but the reverse seems far sillier. Especially if you look at the hot spots where things got worst, like Italy or India. Instead of being glad we avoided that, we're whining that we had to do anything at all.

> His point is that lockdown doesn't really change behavior and places that don't have strong mandates end up largely practicing safer behavior anyway.

Again, this is a curious claim. If lockdown doesn't really change behavior, how can lockdown be responsible for the things it's being blamed for like isolation and economic impact? That is then the fault of the pandemic, and the natural behavior changes that resulted, instead!

Note how toilet paper was gone from grocery stores BEFORE any official action was taken.

If you're looking for a doomsday scenario, what's happening in India right now is probably the worst we've seen since the pandemic started. Essentially every hospital in every major city is full and there are patients dying outside.

Yup. Having insufficient bed space and medical supplies will kill people that might otherwise have survived a Covid-19 infection.

Anything and everything must be done in each country to prevent the hospitals from filling up.

Was also looking real bad in Brazil a few weeks ago.

Still is. I expect Brazil to exceed the US death toll in about 100 days.

They've got a fundamentally different society and lockdown is a luxury policy that only rich nations (and people) can afford.

> They've got a fundamentally different society

You have to say more than this.

Didn't India also aggressively lock down in the beginning as well? Not sure what point the Op was trying to make here.

That’s exactly the kind of paper you get when economists are publishing outside their specialty. Most interesting example was he excluded any data that was counter to his theory and used highly overfit data. “Whether Atkeson (February 2021) has correctly modeled the Covid-19 virus is not at issue.”

They can’t exactly do cost benefit analysis without models to compare, but their cherry picking models to support their assumptions and then happy that their assumptions where validated by their models.

The point he's making in that section is that by accounting for voluntary changes in human behavior (or failing to account for it as many models did) will drastically affect the results of any model that attempts to predict the rate of spread.

Human behavior is also influenced by messaging. You can’t assume economically rational actors in the real world when modeling actual behavior as people don’t know actual risks etc.

In a non lockdown world where schools are open and attendance is mandatory... well it’s clear to the average person the government doesn’t think people need to stay home. Sure, old people are dying in hospitals but I am 40 so it’s nothing to worry about etc.

However, shut everything down and people get the message that this is serious irrespective of the actual dangers they personally face.

> shut everything down and people get the message that this is serious

the Pajama Class say this and then wonder why their "messaging" doesn't go over well with anyone who's not able to log into zoom and chat it up with their comfortable, salaried, urban-elitist compatriots

I am sorry that reality doesn’t line up with your preconceived notions so you feel the need to lash out with name calling etc.

Consider sitting, back taking a deep breath, and then stepping away from the computer to do something you enjoy.

> reality doesn’t line up with your preconceived notions

I'm claiming this precise thing, that you're arrogantly presuming that everyone can just hole up in their house like the elite class of people who do nothing but berate the (almost always) lower class people who do not have the privilege to stay home.

> that you're arrogantly presuming that everyone

Gp didn't claim that. You're putting words in their mouth and lashing out for things they never said.

A reasonable interpretation of their claims is that even if not everyone can hole up, government policies that are intense encourage everyone to hole up as much as they can.

It's also worth mentioning that many, many working class people did not enjoy the fact that they had to take on risk in order to survive. Pretending everything is fine simply offloads even more risk onto those people.

You do realize that normative claims like "they ought", "government should", etc. are precisely the faculty-lounge, haughty idealism, that's worth absolutely nothing to people as they are, spoken from a position of privilege, condescension and fake expertise?

Notice how teachers act as though they're above everyone else? Refusing to work, whereas normal people that aren't of means largely kept working, and didn't act like paranoid agoraphobes exaggerating their actual risk?

It's no secret that most people overestimate their risk by an order(if not multiple) of magnitude.

This has nothing to do with what I or the other poster said. I'll reiterate that many people who were forced to work were not happy about that fact.

You're somewhere between pretending those people don't exist and claiming that those lower class people who didn't want to work but had to are also morally bankrupt. It's neither coherent nor respectful of those you claim to be defending.

Per your bio, you work at Google. You're divorced from the experience of struggling people.

I'm reiterating the experiences of my friends who work in the service industry and whom in some cases have been unable to pay rent reliably the past year.

Having to make the choice between going to work at a job where you have health concerns and losing your home is not a fun situation. You're correct that this is not a choice I've personally had to face, but it is a choice that many people have had to deal with. The issue, that you keep ignoring, is that people who "kept working" did so in many cases not by choice, but by necessity. Yes, the privilege that some (including myself!) had to continue working safely is nice. But you shouldn't use that to discount the opinions of people who had to continue working unsafely, but did not want to.

And you are continuing to ignore those people. Anecdotally, those people make up the majority of those in service industries that I know. They weren't happy to have to continue working. They were scared. Were it economically feasible, most of them would have preferred to stay home and not work. But that option wasn't available to them, because not working would mean losing their homes too.

And it is ridiculous for you to try and weaponize my privilege to ignore those people's opinions.

That is an extremely presumptuous statement to make. You know nothing of their background, upbringing, path to where they are today. I'm not sure what you're trying to accomplish in this thread anymore, but it certainly is not constructive.

> That is an extremely presumptuous statement to make.

No, especially in the context of the pandemic, it's easy to start talking about pandemic policy as though "everyone should just do X" where X is something extremely convenient, profitable, even pleasant for tech workers and other white collar jobs.

You say that yet his statements directly line up with what a friend working in a grocery store said. I don’t know if you’ve ever actually talked with anyone working in Amazon’s warehouse, but I can tell you it’s a very hard job that’s better paying than most people working there can get elsewhere. Some felt it was important, many of others actually quit over COVID.

On the other hand pointlessly attacking people just makes you seem foolish.

> and then happy that their assumptions where validated by their models

This was done in the pro-lockdown measures camp all the time, epidemiologists, etc. love using models based on assumptions that are used to show scenarios as evidence of the original assumptions.

They're arguing simultaneously "lockdowns had no impact on people's behavior and therefore had no benefit" and "lockdowns had massive negative costs". Lo and behold that is the exact mistake the paper is making! It's basing the cost analysis off the following question:

>Suppose you could either live a year of life in the COVID era, or X months under normal conditions. What’s the value of X that makes the AVERAGE American indifferent?

Yet you'll notice this is not about lockdown versus non-lockdown this is "normal" versus COVID. Moreover, the number they use for X is just made-up. Talk about the pot calling the kettle black - this is the exact mistake the author goes on and on about for the other side of the cost/benefit equation.

This paper is super sketchy in other ways - it's clearly a personal diatribe, not an attempt to be accurate. Like:

> Lockdown is a formal,state-mandated “one size fits all” version of the social norm “keep your distancefrom people who are sick."

Is the author unaware of asymptomatic transmission? If all we had to do was keep away from sick people, then this would have been a SARS-1 outcome.

> As a result, the ICL model made some dire predictions ... For instance: “In the (unlikely) absence of any control measures or spontaneous changes in individual behaviour ... In total,in an unmitigated epidemic, we would predict approximately 510,000 deaths in GB and 2.2 million in the US...”

The US has achieved 1/4 of those deaths with behavioral changes (from lockdowns or not), so 2 million without any behavioral changes seems like a good estimate from early data.

Fig 2B is clearly over-fit on past data. I doubt it will hold up well.

> "lockdowns had no impact on people's behavior and therefore had no benefit" and "lockdowns had massive negative costs"

I think you're overlooking the idea that "lockdowns", as we know them, do little other than impose massive costs and anxieties on society, while failing to achieve their main objective i.e. complete immobilization and eradication.

So you end up with all the costs of the policy with little upside because the dynamics of spread seem very heterogeneous, mass-spread seems to happen in relatively few people in very particular scenarios (crowded indoor areas - notice that lockdown ends up pushing the entire society, especially the least healthy and most vulnerable, indoors).

It depends some on how lockdowns are implemented. Where I live, for instance, they outlaw indoor dining but still allow outdoor dining, which seems to me to be exactly what they should do.

For some reason Europe saw a lot fewer excess deaths in 2020.

For example, in Germany it was only ~ +3%, and Sweden, which didn't lock down and had no mask mandates, had a lower death rate than in 2012:


Germany was one of the countries that I remember handling it pretty well in the early days though. Italy, Spain, Belgium, and to some extent UK and France were the ones towards the top of the Worldometers per capita deaths IIRC.

What I wonder from that table though is why does Germany seem to have a higher death rate than peer countries over the prior few years? They are in line with Estonia and noticably higher than all the other countries listed.

Well, the first outbreaks in Europe were in Italy and Spain, too, so I'd say they had more of an excuse than the rest of us.

A big contributor (at least according to local epidemiologists) was the numbers here in Belgium were over reported, whereas most countries under reported. There was also an issue getting accurate numbers, because (mostly privately run) elderly homes weren't looking into it carefully enough to distinguish between Covid and other causes. That said, balls were definitely dropped.

Especially for Germany, the statistics are not complete for 2020. They will be in a few months hopefully.

Europe is light years ahead in medical emergency services, when compared to US, this my limited observation, I am EMT volunteer in US.

That's interesting. I'm a layman at best, but I've always heard that the US health care infrastructure is really good, as long as you are eligible for care in the first place. Given the potential astronomical cost of a sudden uninsured health emergency, it would seem that the money is there for it to be that way at least. I'm not american so I've only heard anecdotes though.

I meant in terms of emergency services response in general and transport/care from incident to hospital. I can't speak much about care in hospitals. But initial response to COVID was substantially worse in US then in Europe, in my subjective perspective.

It seems to vary also a lot by region. Take for instance this local news article from yesterday about excess death in some east German states: https://www.dnn.de/Region/Mitteldeutschland/Uebersterblichke...

Among other things it says the excess death in last December among members of a large health insurer (AOK) rose to 250% of the previous year in various heavy hit regions like Zwickau, Görlitz and Sächsische Schweiz ("So stieg die Übersterblichkeit im vergangenen Dezember unter den sächsischen AOK-Mitgliedern in den Landkreisen Zwickau, Görlitz und Sächsische Schweiz auf 250 Prozent des Vorjahresniveau.").

Sure, but if you look at the general trend in Sweden the death rate had been in a steady decline for many years so 2020 was a clear trend breaker.

Some call Sweden a nanny state and it's true in some regards as there had been changes to regulations (esp laws regarding smoking in bars,etc), comprehensive cancer screening initiatives, road construction laws,etc to decrease the impact of common mortality causes.

That's interesting, why is that?

Higher obesity in US maybe?

I doubt the entire US became sharply more obese in the past year, though, so much so that it also died from it.

The suggestion wasn't that they were dying from obesity, but from its correlation with covid. Not sure if it can account for the increase, but in conjunction with pretty terrible handling of the crisis during 2020 I could imagine it at least.

That makes sense, I misunderstood, thanks.

the US was already more obese than Germany -- and obesity seems to correlate with higher covid-19 mortality

I'm assuming tokai is saying that because obese people are more likely to have complications from Covid19, the US saw a higher increase in excess deaths, given that the obesity rate is higher than eg Germany.

This also explains some of the racial disparities in COVID deaths since the obesity rate for blacks is 49.6% vs 44.8% for Hispanics, 42.2% for whites, and 17.4% for Asians

Oh, I see, I thought they meant that the excess deaths were caused by obesity, not by COVID19 leading to higher rates of death among the obese. That makes more sense.

But the pandemic could have hit the more obese American population harder.

Probably a combination of:

- US letting it rip through vulnerable populations with the promise of "herd immunity" being right around the corner

- Fewer hospital beds available in less populous areas.

- Taking COVID seriously became a culture war issue. So while some countries might've had its youth catch it while mostly keeping it away from older populous, the US had the opposite: less vulnerable youth took it relatively seriously, with lots of 60+ flaunting safety measures they felt were oppressive

How does that explain Sweden? Or Germany. Or any of of the stats the person above you was replying to.

I'm not sure it's really a good comparison to pick a subset of European countries and compare them to the whole US. Some countries seem to have handled it better than others, which we could definitely learn from, but I don't think it is a US vs Europe problem like some other commenters seem to be implying. We could also compare Italy, Belgium, Hungary to Vermont.

In a year when Geico is giving their customers an across the board rebate on their premiums because driving miles are down so much, and driving accidents are the third or fourth leading cause of death in a non-pandemic year, I think we'd need to adjust downward the baseline. I wonder if these calculations take that into account?

'Tragic': Driving Was Down In 2020, But Traffic Fatality Rates Surged

The National Safety Council (NSC) says deaths from motor vehicles rose 8% last year, with as many as 42,060 people dying in vehicle crashes.

When comparing traffic deaths to the number of miles driven, the rate of fatalities rose 24% — the highest spike in nearly a century, NSC says.


Enforcement went down to basically zero in my area. Now that the officers are vaccinated, they have started aggressively enforcing moving violations.

Remember that kinetic energy is proportional to velocity squared. So if less traffic is letting people drive faster, that could easily result in more deaths. I live in a city, and average speeds certainly seem faster than pre-pandemic traffic.

Also, working at home may be allowing people to avoid traffic because hours are less strict. Again, that would allow traffic to move faster on average.

Thanks! That's wild!

I'm suprised by the number of people arguing for and against the lockdowns using just deaths statistics.

Lockdowns of the duration that we experienced should be illegal because they take away the citizens' liberties to an extent that no government should be able to take them away. It has nothing to do with health.

>Lockdowns of the duration that we experienced should be illegal

I fully agree with you. If people want to sit at home it should be a personal choice. However people do like to control other peoples lives, that is the bottom line and they will use any argument to justify it.

The thing is though that, by and large, there never were "lockdowns" in the US. I'm sure there are exceptions but most people were free to go out in public if they wanted to (perhaps with masks/distancing), many types of "essential" businesses remained open, etc.

I know in my area, there was zero enforcement of travel restrictions, zero enforcement of stay-at-home, zero government enforcement of mask wearing (shops generally had to kick people out), and very spotty enforcement of business closures. There was no lockdown. It was totally voluntary, and luckily enough people voluntarily did the right thing that hospitals didn't overflow.

Or people voluntarily did things like wearing masks/distancing indoors that presumably helped. I was more conservative personally but I'm not sure how much, say, driving to adjacent states for outdoor activities made a difference.

If I read your comment correctly, whether or not the lockdowns are legal should Not have anything to do with health. So what you are saying is that you don’t care if it’s a health risk - nobody should ever be able to order you to stay in your home by fear of fines or imprisonments etc.

IMO that would be fine so long as we had a citizenship that was well educated and the topic of virus spread had not been politicized to the point that certain groups believe the virus itself is a conspiracy.

Why does it only go until 2015?

This seems like data that would be better understood as per capita all cause mortality versus time over the last century.

Anyone know where to find that plot?

Wouldn't you end up confounding a lot of different issues with that? Lifestyles and medicine has radically changed a fair bit in the last century, and so you'd end up with a bunch of confounding factors.


This is close to the analysis of interest.

It’s true there are so many confounding issues, e.g. obesity rates have never been higher and are a huge cofactors, population inversion, etc.

Interesting that way fewer people died after the 1918 pandemic. I assume it "weeded out" all the people who were likely to die soon anyway, but what's more interesting is that the negative bar seems to be longer than the positive one, which would somehow mean that the pandemic saved lives? Very odd.

1918 was also during WWI. While they mark that with the pandemic, are they including US service member deaths during the war in death totals? I'm trying to find a breakdown by year, but the total US deaths from the war was around 100k.

Even taking them as evenly distributed over 1914-1918 (inaccurate) that's 20k extra deaths in each of those years.

Assuming it's not just noise, could be that technology developed or cultural norms changed because of the pandemic saved some lives down the line.

Also, weren't deaths in the first year of life much more common back then? Maybe there were just significantly less people having kids right after the pandemic.

Antibiotics became widely used in the last 100 years. We’ve learned a lot as a species in that time.

Exactly. And we've gotten better with cancer, but we've gotten worse with obesity. There's a bunch of layers to apply, so I don't know how factoring in the excess mortality from 1957 or 1942 is particularly useful if you're looking at how much upset COVID causes society.

Why? You want to compare pre-COVID to COVID.

It's interesting that in "normal" years, deaths go up in the middle of the winter and down in the middle of the summer.

Look up Hope-Simpson's work on influenza epidemiology, he observed seasonality like you say, but there are certain regimes that correspond to latitudes.

I'd like to see a version of this analysis that incorporates all-cause mortality from a much larger time period; say the past fifty years rather than the past five.

While that would be interesting, it would be a less direct measure of COVID's impact, as other variables are introduced.

The tricky thing about discussing covid-19 mortality is that there are always confounding variables. The disease is deadlier for people with pre-existing conditions, particularly cardiovascular.

A common line of argument online is "it's just the flu, and we don't panic like this for flu season every year." Observing a spike in total deaths puts that line of thought to the test.


Because it would give people a better intuitive sense of how bad COVID is in a larger civilizational/historical context.

Over the past 12+ months, people have accepted enormous infringements on freedoms in their everyday lives. Many of the public health measures that we take as a matter of course in the "new normal" - lockdowns, social distancing, extended closing of churches, etc.- are things that would have been almost unthinkable as recently as January of 2020.

If you look far enough back, to some point in the twentieth century, you'll find a time at which the baseline infectious disease risk in everyday life was higher than it has been in the COVID-era US. But people at that time went to crowded theaters, ate in restaurants, packed together in churches and schools, and so on. Restricting these things could have decreased their all-cause mortality, but they clearly found that option unacceptable.

Note that this isn't an issue of whether people "believe in science" or "believe in data". It's an issue of risk tolerance. People a few generations ago understood the Germ Theory of Disease perfectly well, and they knew that they could reduce the spread of infectious disease by implementing the kinds of measures we've implemented for COVID. And yet they didn't implement these measures, except in very minor and temporary ways. We've become far more risk-averse than they were. And the question of whether that's a good thing is not an empirical question.

I don't understand your argument. You're saying that because people didn't mind dying from infectous diseases at a time where scraping your hand could prove deadly, we shouldn't mind it today either?

It's not a question of what people "mind" or "don't mind", it's a question of what level of risk people will accept in order to live a normal life. Hiding away from people may reduce risk in a lot of ways, but most people find it unacceptable.

Before COVID, I was under the impression that most Americans found "Give me liberty or give me death" an admirable sentiment. But it turns out that I was mistaken about that.

> Before COVID, I was under the impression that most Americans found "Give me liberty or give me death" an admirable sentiment.

Are you under the impression that you were experiencing pure liberty before COVID? Because it's pretty obvious to me that there was an acceptable level of restraints that society generally agreed to, and those levels changed due to the pandemic. It's a fun saying, but it's in no way, shape, or form realistic.

Are you suggesting that since we didn't have full liberty before the pandemic we should be ok with less liberty during the pandemic and into the future? Because by accepting these measures now I can guarantee they'll become the new norm just like the changes after 9/11.

> Because by accepting these measures now I can guarantee they'll become the new norm just like the changes after 9/11.

What measures are you referring to with this? Masks clearly aren't going to stay around, and there's a push to reopen everything and get life back to normal as soon as possible.

> Before COVID, I was under the impression that most Americans found "Give me liberty or give me death" an admirable sentiment. But it turns out that I was mistaken about that.

America is country with most strict safety standards in pretty much anything I ever heard of. Including kids playground structures.

50 years ago was 1970. I agree that just comparing last year's death rate directly to rates from the past 50 years would be a bad analysis, but I think this is some hyperbole too.

That doesn’t seem like an empirical question at all. Whether it’s a good thing depends on what you value.

In a vacuum it would give you more of a sense about how much of an outlier this year was. But obviously enough other things have changed over the last 50 years that the numbers would stop being comparable. Perhaps looking at a 5 year sliding window to see how big the peaks usually are relative to their surrounding years could get at this with a bit less confounding.

Which is exactly what the article did in its first graph.

Unless I'm missing something they only use a single 5 year window to look at the 2020 peak. I meant if you slide the window back over the past 50 years, what other peaks show up and how much larger are they then the other years in their window?

Ah, I misunderstood your point. You're missing nothing.

Lies, damn lies and statistics.

Big problem with this type of calculation: it is based on the prior n-years of data. These figures do not consider that the just this year, the leading edge of Baby Boomers are achieving their average life expectancy and dying. There needs to be some corrective math that takes the birth rate at (current year - average life expectancy) into account.

We don't have a system where you hit 75 years old and drop dead. Everybody got a year older, and ~21% of those were baby boomers. The statistical impact of that on a year-by-year basis is minimal.

On average, that’s exactly the system we have. Nice attempt at sounding authoritative while attempting to invert a fundamental reality, that more births obviously results in more deaths.

I doubt a rising wave of baby boomer deaths is going to have a significant change on those statistics, but if your objection is correct, the death rate should continue to rise, so in a couple of years we can look back and see if you were right. Or look at mortality by age.

One thing I would like to know, and I don’t know enough about mortality measures so maybe someone who does could chip in, would be essentially what is the probability of observing a year like 2020 with respect to all cause mortality. Is it in 99th-percentile or are we looking at an observation within a standard deviation of the expected value,etc.

Short answer: In the U.S., 12 standard deviations.

In the U.S., deaths were 16%-17% higher than expected. The death rate was 17.8% higher than in 2019. Excess deaths were higher in 1918, but every other year since 1900 is lower.

Using 1946-2019 as our sample, the average yearly change in death rate is -0.2% with a standard deviation of about 1.47 (percentage points). The biggest increase in this year range was 3.1% (2.2σ).

If we go back before the antibiotic era, the biggest one-year increases are in 1918 (29.3%) and 1929 (7.2%). Note that statistics before 1933 don't include all states.

For some other ways of looking at the data, here's an economists blog post: https://economistwritingeveryday.com/2021/01/06/excess-morta...

(He was writing in January, so had less complete stats.)

Thanks! That blog post was an interesting read, age adjusted mortality rates was something that hadn't occurred to me as someone unfamiliar with this type of data.

As a data point in my country (Poland) there haven't been that many deaths since WW2.

The pre-war population of this country was around 34mln - a figure reached again in the early 70s - while currently it's slightly less than 38mln, so this event can definitely be seen as rare.

Given the range of values over 5 years it seems pretty unlikely for most states, but might not be out of the question for say Hawaii or Vermont say. Unlikely, but not impossible.

One factor to bear in mind is the 5 year period includes a particularly nasty flu season in 2017-18 so the last 5 year period includes a high outlier.

The thing is over the long term all kinds of outlier events pop up, like this one, but it's hard to think of any with this magnitude. 9/11 killed 3k people in one go, but that's only one tenth of the excess deaths in New York from Covid in a single month. Even the 1906 Earthquake that devastated San Francisco only killed less than 1% of the city's population. World War 2 had an appreciable effect of course at 300k deaths over about 3 years.

If we're including other pandemics of course there's the Spanish Flu which also killed about half a million Americans or more, against a population a third the size.

The true covid death rate has been tricky to record in the UK, a positive test result within 28 days prior to death is recorded as a covid death, even if the death was caused by something else. We obviously don't know enough about COVID to know if it is contributing all the COVID registered deaths.

The UK government publishes other ways to measure COVID deaths. the "positive test result with 28 days of death" results in the lowest count:

Deaths in England and Wales in 2020

    80,830  Covid-19 as underlying cause of death or contributing factor
    77,161  Increase in deaths (from all causes) from 2019 to 2020
    73,444  Covid-19 as underlying cause of death on death certificate
    70,013  Any death within 28 days of a positive Covid test




All this being with lockdowns, which are demonstrably effective at slowing viral spread. Just imagine what these charts would look like _without_ lockdowns.

This comment section is a pretty strong argument that we are living in a post-truth future.


You're made for each other. Excess deaths capture the deaths from lockdown related causes too, like my grandmother, who died when her social life was indefinitely stopped, and teenagers that committed suicide due to lockdowns

Excess deaths don't tell us how deadly COVID really was, they tell us how deadly COVID plus the COVID response really were.

Someday historians will argue: which was worse? The disease or the treatment? I don't think it's as clear cut in reality as you or your "friend" that you're mocking think.

Maybe both of you should try listening to the people you disagree with

The problem with blaming the response, is that a strong response with very strict lockdowns is sometimes a result of a weak early response. That is, if you hesitate with doing lockdowns, you'll eventually be forced to do a much more severe and longer lasting lockdown anyway. A weak response will also overrun the hospitals, which makes it difficult for them to diagnose and treat other diseases.

Norway has had no significant excess mortality through the entire pandemic. In recent months it has been well below 0%. It has been consistently been better than Sweden, which has had less restrictions and a similar culture.


The restrictions in Norway has been less severe than many places with higher excess mortality. Why is that? My theory is that Norwegians have been very compliant with advice and restrictions set by the government, and the government has been pro-active with enacting restrictions. This has helped contain and suppress the spread of the virus to smaller geographical regions, so that other areas can avoid lockdowns. And it reduces the duration of outbreaks, meaning you can lift restrictions again more quickly.

So my theory is that fighting against restrictions is extremely counter-productive. It only serves to create a culture of ignoring government advice, and makes politicians hesitate to enact necessary restrictions early, with the end result being more restrictions in total, not less.

Sweden had an abnormally low death rate in 2019; Norway did not. Averaged over the past two years, Sweden's death rate in 2019 and 2020 is just 1.2% above 2018, as compared to Norway's 0.48% below 2018.

It's quite likely that many people who would have died in 2019 in a normal year died in 2020 instead.

https://www.statista.com/statistics/611743/number-of-deaths-... https://www.statista.com/statistics/525353/sweden-number-of-...

Suicides were down in 2020 in the US. Australia had long and strict lockdowns, so has Vietnam. But they have negative excess deaths in 2020. There's no evidence whatsoever that lockdowns increased deaths, with the possible exception of an increase in fatal traffic accidents due to more speeding.

> Excess deaths capture the deaths from lockdown related causes too, like my grandmother, who died when her social life was indefinitely stopped, and teenagers that committed suicide due to lockdowns

If only we had an easy mathematical tool to calculate how many people in the excess mortality numbers didn't die of COVID. We could call it "subtraction".

But whether someone died of Covid or not is also another area of dispute. The "it's just a flu" crowd can cite dubious websites saying "Oh the government is fudging the stats, there were many deaths being recorded as Covid-deaths even if they're not (Stupid example: man flies off motorcycle into wall, dies. Autopsy shows he has the virus in his blood, the conspiracy theorists will say the Illuminati/Bill Gates/the government will record that as a death due to Covid).

Inversely, there's also reports of governments not testing people who died for the virus, to make their numbers look good so they don't end up in some "Do not travel" list, or because they don't want their population angry and removing them from power. Even India is still fudging its numbers lately.

At least it's hard to dispute the status of being dead, and "excess deaths" is a usable number.

It would be interesting if we could "reload from a save point" and try surviving the pandemic without lockdowns. IMO the number of COVID deaths would be higher. Of course some people think otherwise would hypothize the opposite.

I guess a glib response would be to say "Look at fucking India right now". And it's a bit binary to argue e.g. "Sweden didn't have a lockdown, look, it wasn't that badthere.". Uh, even without a lockdown, Swedes changed their behavior (a lot of work from home, cafes and restaurants became emptier), and if we reloaded from that save point and didn't impose lockdowns, the cautious people would still stay home. And I wonder how many of the "But muh freedom!" moaners would be name-calling responsible organizers who cancelled concerts and festivals as "fascist nazis", etc.

> teenagers that committed suicide due to lockdowns

This turned out to be false expectation. Turns out that while teenagers were unhappy and did not enjoyed the situation, they were able to adapt without destroying themselves.

I love it when people who ignore all the data and intelligent discussion around a subject argue that people aren't listening to them.

I see the 2020 mortality no different from other years: https://github.com/klahnakoski/mo-statcan#example---covid-19

I’m always fairly torn by these numbers (I’ve done a lot of research and tracked this[1]). There’s a lot of confounding factors...

The lockdowns increased the rate of suicides substantially. Largely suicides occur in people 65+, so a 100% increase in suicides would be some 150k deaths in the same age range as COVID.


Murder is also up some 50% and there were riots...

Not disputing the excess deaths, just the causes aren’t all that clear imo.

[1] https://github.com/lettergram/covid19-analysis

This is nonsense. Suicides are down in 2020. It was a virus that killed people, the deaths were reported. Conspiracy theories are so exhausting

The government has an incentive to report deaths as covid deaths. I'm simply stating it is not 100% clear what causes the deaths. The reason I suspect a higher suicide rate than otherwise estimated is that the number of people with suicidal ideation has doubled.

NOTE, I was also considering unintentional suicide (overdoses), which were dramatically up (38%).




> The government has an incentive to report deaths as covid deaths.

What is this incentive? How did governments benefit from more covid deaths?

It makes their policies appear sensible.

You say that with such certainy but the official number I've seen is a 10% increase. Also there is no measure for misery.

it would be helpful if you provided your sources

> The lockdowns increased the rate of suicides substantially

why did you link to stats that only go to 2018?

Suicides are down in 2020[1]

1. https://www.axios.com/suicide-decreased-in-2020-pandemmic-me...

Succesful suicides were down, but attempted suicides and overdoses were both much higher.

Your suicide statistic link doesn't have any data for the 2020 year. And in my region suicides are down. https://calgary.ctvnews.ca/suicide-deaths-in-alberta-in-2020...

The problem with "what about other increases in cause of deaths" is that you have to find a LOT of these reasons to even begin to compete with Covid-19 as a cause. When the increase in deaths is in the hundreds of thousands it is not adequate to point to something that potentially increased the deaths by hundreds or thousands and say that's enough evidence that the Covid numbers are grossly inflated. You need to provide evidence backing your claims for hundreds of thousands of deaths.

Looking at the NIMH link you sent, I'm not sure where you're pulling that data from as 65+ isn't in the chart but some quick maths shows it would be about 9k people, so your assumed 100% increase would only lead to an extra 9k deaths from suicide in that group. The total for all age groups is 48k, so you'd need a 200% increase in all ages to get to that number.

However, if we look at this, based on CDC data: https://jamanetwork.com/journals/jama/fullarticle/2778234 "From 2019 to 2020 ... suicide deaths declined by 5.6%." So you really need some strong evidence to back that suicides went up at all, let alone by 100%. It would also be very difficult to blame false attribution of suicide deaths to Covid-19 without equally strong evidence as there are usually pretty strong clues to the cause of death in suicides.

The riots & associated high levels of aggression between social groups did not lead to a high number of deaths, definitely not enough to have a significant impact on total deaths, unless you have a good source for that claim.

Murder rates seem to be up by about 36.7% across the US, but even then that increase is not enough to cause a significant bump in the numbers. There are normally about 870 deaths per 100,000 in the US due to all causes. Murder is about 5 in 100,000 (2018) so a 36.7% increase in that 5 gives 7, an increase in 2 per 100,000. That's an increase in the death rate of 0.23% due to increased murder. The CDC number I just checked is a 15.9% increase in deaths for 2020.

https://www.npr.org/2021/01/06/953254623/massive-1-year-rise... https://www.cdc.gov/mmwr/volumes/70/wr/mm7014e1.htm https://www.cdc.gov/mmwr/volumes/69/wr/mm6942e2.htm

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact