Pooled number of deaths by age group in EU | Hacker News Reader

Pooled number of deaths by age group in EU

euromomo.eu

78 pointst0bia_s79 comments

My company Design Systems International (https://designsystems.international/) designed this site during COVID in 2020 when the interest in mortality data suddenly exploded. Happy to see it on the front page!

Is there a lot of random statistical noise in these esimates of 'normal background mortality'? Prior to Covid you mostly saw these 'excess death' figures in relation to heat waves IIRC, e.g. if 3,000 people were estimated to have died from a heat wave, that figure was obtained by subtracting the actual mortality from the expected mortality over a time period of the heat wave, then reporting that as the death count.

The attribution seems a bit challenging in retrospect. Do people make death forecasts, like with weather forecasts, and use that as a test of forecasting skill and the assumption of relatively non-random average outcomes under normal conditions?

P.S. check out the CDC mortality dashboard:

https://www.cdc.gov/nchs/nvss/vsrr/mortality-dashboard.htm

It certainly is challenging, but there are a lot of statistical methods to gain confidence in those estimates.

E.g; Heat-related mortality in Europe during the summer of 2022 - https://www.nature.com/articles/s41591-023-02419-z

They looked at the 2022 outlier year in terms of heat across Europe, and then found strong correlations with the hottest weeks in the hottest locations in terms of heat-related deaths from the trends established over the prior 5 years. Seems pretty robust and they did a number of tests to ensure it was a reliable result.

To put in perspective, you should compare numbers with mortality caused by cold weather, which is significantly bigger (about 8 times more).

It's 4 594 098 deaths by cold and 489 075 deaths by heat during years 2000-2019, source: https://www.thelancet.com/journals/lanplh/article/PIIS2542-5...

Sure, but my original post was restricted to Europe and per that Lancet study, it's only maybe 3.5x more (660k cold vs. 180k hot) -- hence why the 2022 year stood out so much.

The authors of the prior study estimated 70,000 excess deaths due to heat in 2022 alone which is nearly 40% of the total for the 20 year period from 2000-2019 -- or about 8x the 'background rate' of excess heat deaths. If we are seeing warming and years like 2022 are going to be more common, that's a really, really big deal.

>death forecasts

Sounds cool. I'd like to see that on the evening news TBH.

The title should be "in Europe", not "in EU". Though it's peculiar that Israel is included if the data is supposed to be about the European continent.

I was wondering what the spike in The 15-44 graph near the end of 2023 was... then I saw that Israel is included. RIP.

If the data from Gaza were included, a new scale would have to be invented. R.I.P.

https://www.euromomo.eu/how-it-works/background-data/

For whatever reason, Israel participates in this by providing the necessary data.

Cyprus is also included while it's on the Asian continental plate

Not sure if this post is meant to highlight continued excess mortality despite the official end of covid?

According to wastewater monitoring, covid is thriving: https://biobot.io/data/

Numbers in under 45 age groups goes up in 2021. Safe and effective.

Interesting, didn't realize that the winter of 2023 was about as bad or worse than the winter of 2022. It seems this is the first normal winter since 2019.

In America, more people have already died of covid in January than normally die of the flu in a whole year. At some point "baseline" is going to rise to incorporate this new thing.

it’s been about 3x as bad as flu in January: https://www.cdc.gov/nchs/nvss/vsrr/covid19/index.htm

edit: which is higher than some recent annual flu totals they have, but considerably lower than estimates of annual flu deaths I’ve seen elsewhere? I feel like I’m missing some context for interpreting these numbers.

Measures to mitigate COVID are also effective (arguably more effective) on the flu. So the years where society was trying to stop COVID had much lower than usual flu rates.

In many countries, including the US, death reporting is a bit of a mess. Those numbers will likely continue to tick up as numbers come in.

In Ireland, at one point in the pandemic, a lot of attention was paid to rip.ie, an obituaries website, because the non-hospital official death count was _months_ in arrears; the mechanisms around death reporting were just not designed for realtime analysis.

I’m just used to seeing annual flu death counts that look more like this: https://www.cdc.gov/flu/about/burden/index.html. My first guess would be that it’s an issue of “confirmed” vs. “estimated” - and I’m sure actual covid-related deaths can similarly be estimated to exceed confirmed these days - but I don’t know exactly what the methodology is for each.

Ah, yeah, these are estimated numbers. Estimated numbers aren't usually used for covid (or at least weren't during the pandemic; this may change); really everything was a bit too uncertain for decent estimates for the first year or so, and there would be political sensitivities to using an estimate as a headline figure.

For some sense of the scale, the WHO estimates 15 million deaths globally by end 2021, vs 5.5 million confirmed in the same period. But generally the WHO uses the confirmed figure; the estimate would rile up conspiracy theorists. I'd expect, given the situation in the US, that the CDC would be even more politically sensitive.

CDC has always provided estimated excess deaths throughout the pandemic. They have correction factors for each jurisdiction by week, according to the typical delays in reporting.

I recall watching a video years ago about coffins for some reason. It mentioned how in the US the coffin industry has to make much bigger coffins since people are much bigger. The reporter somewhat jokingly asked "Is there a coffin season?" like chocolate is for Valentines/Hallow'en or toys for Christmas. The owner of the coffin business said "Yes flu season is our busy time." their sales double. That documentary was years before COVID hit.

Well, Ukraine cured covid.

That's a good way to remind oneself how much our perception of a problem is manipulated by the news. That works both ways. They can blackhole a major event, or put a giant microscope on top of a tiny event. You may decide to trust or distrust the news, but you can't decide to hear about something you don't know is happening, and likewise, even if you think another "it's the end of the world" coverage is bullshit, it will still occupy the debate, and therefore your mind.

That's one theory. Another is that people just have a finite amount of emotion they can spend worrying about something before they go back to their lives. Even if objectively there is still something worth worrying about.

More like 95% of people have caught covid and are vaxxed so at this point they’re just not interested in letting it dictate their lives. Much different place this year in terms of immunity and exasperation than prior years.

No grand conspiracies are needed.

If this is the case, it’s interesting that people are apparently now not bothered by an excess death rate near that of COVID peaks.

The things they show you on primetime news is what will bother you. ;)

Turns out uninformed is better than misinformed. That way, at least you are free to make your own opinions and decisions based on your observations and lived experience.

I agree. We had 3 years of narrative shaping and statistics.

That’s a big oversimplification of what people were worried about during the prior years.

While taking substantial precautions to limit the spread there was that many deaths and hospitals were on the brink of collapse. More spread would have started a terrible rationing of care leading to a much higher spike in excess mortality, especially in immune naive populations.

Now despite almost everyone moving on from precautions, due to substantial immunity, we’re no longer on the precipice of spiraling out of control.

Yeah it’s not nearly as newsworthy because we’re not on the edge of a cliff anymore. It wasn’t just the death rate, it was the way it could have exploded.

That’s a hard argument to make, at least here in the UK, because it looks like the strain of restarting normal healthcare with a 2+ year backlog of untreated health issues will doom the NHS.

Also - you can argue that fewer covid precautions may have resulted in a collapsed healthcare system. However, in reality, the NHS occupancy rates for 2020-2021 were double-digit less than previous years.

The most consistent / comparable stats over time seem to come from the wastewater monitoring. Those graphs are currently scary for the US and in multiple states they're going through the biggest outbreak since the beginning.

Anyone confused about the horizontal axis’ tick labels? Usually the range befor e.g. “5” is 4-5. But here it would be 2003-2004 in case of “2005”. I never consciously thought about this but is this way of indexing the standard?

They range from 1 to 51/52/53, so they're week numbers. The description of the graph mentions the data is based on weekly result, though it would probably be better if the hover popup would make the date range covered a bit clearer.

Anybody who knows why there might have been a 10-fold increase of excess deaths in the 0-14 years group (as seen in the cumulative plots) right at the beginning of 2024 compared with other years?

Later influenza season maybe? 2023's curve seemed to lag 2022's abit, which spilled over to 2024?

Hmmm the peaks seem to have become a lot higher after 2020. Too bad it doesn't let us see before 2018.

Why is there higher mortality in winter?

Especially in group 0-14 (higher than during covid) and 15-45 (slightly less than during covid but still more than pre-covid).

Respiratory illnesses peak in winter. There was always a push for flu vaccinations in the lead up to winter, and now covid in the mix too. https://www.hopkinsmedicine.org/health/conditions-and-diseas...

Why is that in turn? I’ve heard people refer to low vitamin D levels but I’m sceptic if that is the reason (would be so easy to fix)

because that's how the phase locking shook out.

One potential mechanism is people staying indoors more in winter, making all kinds of illnesses spread more easily.

Don’t you dare open that window!

ve must have ze luftink!

Would be interesting to know if Germans have less of a flu season.

Cold affects the immune response.

https://time.com/6238816/why-viruses-spread-in-cold-weather/

Cold temps actually make your nose less effective at fighting off viral infections before they move to your lungs etc. https://www.jacionline.org/article/S0091-6749(22)01423-3/ful...

Lower indoor humidity levels also contribute.

Heat kills germs, and from memory SARS-CoV-2 is particularly vulnerable.

Didn't stop COVID from rampaging through plenty of hot places like India.

They have unexpectedly lower excess death from initial covid... Until delta variant comes which isn't fazed at all by summer of 2021.

Indeed, it varies by strain. I was remarking on the initial variant, but I have not followed up on the later variants.

Just means there's more than just heat at play, which is sorta obvious.

Not quite. The cold temporarily weakens our immune systems, particularly in our upper respiratory tract. Couple that with increased proximity to other people in many places when people are more likely to use public transport.

One does not negate the other.

I've not seen studies suggesting that the heat in warmer climates is warm enough to kill off germs in a way that would be relevant. I have seen studies showing the impact of cold on our immune systems. Feel free to provide sources though.

> Mathematical models developed by Seifer & Elbaum (2020) aimed to assess the inactivation of SARS-CoV-2 at physiologically relevant temperatures, i.e. around 40⁰C, and at mild extremes of up to 60⁰C. In cell culture SARS-CoV-2 showed a 99.99% fold reduction in activity after incubation at 56⁰C for 15 minutes, which was in line with the predictions made by the mathematical model, lending some credence to the work. At a normal body temperature of 37⁰C, the model predicts 99.99% inactivation after 48 hours, while a mildly elevated temperature of 39⁰C is expected to induce 99% inactivation at 14 hours, increasing to 99.99% inactivation after 28 hours.

- https://www.news-medical.net/life-sciences/The-Thermal-Inact...

Take that for what it's worth but I honestly thought that this was well known and uncontroversial.

There are other factors, such as the dryness of the heat. But a surface under sunlight in a hot climate could easily hit over 40⁰C

I don’t think you need a hot climate for that. Most surfaces absorb a lot more energy from sunlight than air does, making them hotter than air temperature.

See for example https://web.archive.org/web/20150916052900/http://www.tom-mo..., which has car surfaces hitting 60°C while the air is about 28°C

Higher transmissibility of certain viruses could be one reason. If more people are catching it, it's not hard to imagine that more people are getting sick from it, and infecting others. From one study:

> Our study shows an important dependency of SARS-CoV-2 transmission to weather/climate, with each 1°C reduction in mean regional temperature below 10°C leading to a 0.16-unit increase in R0 , or each 1 g/m3 reduction in AH leading to a 0.15-unit increase in R0. Northern hemisphere countries experienced a new wave of SARS-CoV-2 infections during autumnal transition from summer to winter (colder temperatures and lower absolute humidity levels), while still actively maintaining control strategies. When planning to adjust the level of restrictions on social activities, public health strategies need to account for the increased transmissibility of SARS-CoV-2 when/where cold and dry winter conditions are prevalent.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8211690/

It's more likely to be lower levels of vitamin D in winter. Viral transmission is very poorly understood.

Apart from respiratory infections: cardiovascular events. There was a famous paper about guys dropping dead while/after cleaning snow from the sidewalks/driveways (Illinois?). Not necessary from the physical shovelling but from just riding a domestic snowplow. It is good enough be exposed to cold: vessels constrict, blood pressure rises, you may drop dead when old, obese, unfit and with atherosclerosis.

*edit*: article stating that the spikes in mortality were within 3 days of a cold blast. https://academic.oup.com/aje/article/155/1/80/134292

While I appreciate the data, I think it's presented in an overly dramatic way because the y-axis does not start at zero.

That is the point of a "excess death" graph though. Filter out the "DC" component, like on a oscilloscope.

Sucks that Poland did not include itself in these

So many people die at 26..

That vaccine worked out really well for under 45's didn't it?

What happened week 40, 2023, in the age group 15-44 yo? Is it a data entry error?

EDIT: Ok Israel is in the data. That is a sad revealation and the headline is wrong.

that's such a sad statistic, people in the prime of their life, mainly caused by being out enjoying life at the wrong time

Also probably worth mentioning that the graph doesn't start at 0.

Israel? Where did you read this?

It's one of the 27 participating countries listed at the top of the page.

Unfortunate that they apparently dont appear to include Palestinians in that statistic.

It wasn't included because it isn't part of Europe's sphere of influence unlike Israel

Reply on news.ycombinator.com