Excess deaths and SARS-CoV2 vaccinations in Scotland
drowningindatadotblog.wordpress.comI often flag and downvote stories like these, because I do not believe they are consistent with HN's mission to feed intellectual curiosity. It is clear we have a very vocal contingent of vaccine deniers, and I do think it is useful to hear their arguments, but I believe that's best done in the context of source stories that actually convey some new understanding.
From my observations, the best lens for understanding vaccine denial (or other related forms of Covid denial) is religious belief. Articles like these are medium-quality grains of opinion supporting that belief. Outside Covid, such things are obviously not in scope for HN, but the virus has captured our attention.
If I were the mod, I'd have an explicit policy that Covid stories need to come from high quality sources. Dissenting or unusual perspectives are fine, but these "just asking questions" pieces with confusing (at best) statistics aren't. And I'd honestly consider most mainstream press pieces not to clear the bar either. There are some excellent science journalists and bloggers out there (Derek Lowe is one of my favorites of the latter). We should be listening to them, I think.
Another issue with these "just asking questions" posts (some call them JAQing off) is that they often aren't in good faith and even when they are they create the false impression of an even debate on things that really are mostly settled.
I think these subjects that require a lot of care and can have a lot of impact based solely on rhetoric are really not what blogs and rapid response commentary excels at.
I'm happy at least there is enough detachment here for it to be realized, in most other places this would be drowned out in the echo of bickering.
For very controversial topics I would expect high-quality sources to protect their reputational downside by taking conservative stances. In that case I think it’s fine to see some medium-quality outsider sources and use my own noodle. That satisfies my own intellectual curiosity.
I see your point but from my perspective, in most places there's a trickle of good stuff and a flood of junk. I try to offset that by occasionally reading a virology or epidemiology paper myself and listening to TWiV when I get the time. The thing that's striking to me is how different that picture is from middlebrow sources, in which I include a lot of the mainstream press. For example, their scare stories every time a new variant comes out is cookie-cutter journalism (and is basically crying wolf for when something notable does happen, like delta).
> From my observations, the best lens for understanding vaccine denial (or other related forms of Covid denial) is religious belief.
How about the popular opinion that "vaccines are always good and should never be questioned no matter what"? Do you treat that as religious belief as well?
Sure, if that's the belief that's actually driving decisions. It certainly isn't the way I approach things. I eagerly sought a vaccine (my appointment was 10:20am on the first day I was eligible) because the safety and efficacy was demonstrated by data from trials conducted with a high level of scientific rigor, plus observational studies from places that had gotten there early. It was thrilling to see the early data out of Israel that showed such a strong vaccine effect from their age-stratified rollout.
And to further support my point, I wouldn't recommend anyone get CureVac - from the data we have now, that looks like a dud, even though it is an mRNA vaccine with some similarities to the others.
My guess is that that there is a deeper underlying question: should we trust the scientists and institutions who do these trials. And on that my personal belief is that yes we should, but we should also be skeptical and hold them accountable. They make mistakes - the recent aducanumab decision looks like a big one - so I am definitely not arguing we should uncritically accept everything they do and say. But when they do it right, as has happened with vaccines in the US, we should celebrate, not undermine, that.
I do appreciate that trust in people and institutions is a personal thing and there are reasons for it to vary.
>[...] because I do not believe they are consistent with HN's mission to feed intellectual curiosity
Let others decide, whatever fills your curiosity is different than mine and that's a good thing. You are not the arbiter of knowledge.
I am not, but I can downvote things that I think degrade the quality of discussion, which in my personal opinion includes a great many Covid threads. Others of course can submit and upvote stuff they find is interesting. If you have good stuff that challenges the conventional wisdom, please post it, as I actually seek out that kind of material (for example, Alina Chan's Twitter feed, which in my opinion is the best skeptical source on the lab leak theory).
I may just be exceptionally stupid, but is this a gag article? I mean that in all seriousness.
Graph of daily vs weekly deaths look completely differently, numbers are present in absence of any context (how are vaccine rates measured? Does it only count it after the two week build up period?), and the article cited for proof of spike protein levels doesn't actually cite spike protein levels, but lipid levels.
It gets worse. Anyone who tracked COVID at all knows that there is a several week gap between getting COVID and when people die from it. So vaccinations today are expected to have zero impact on the death rate over the next month.
The result is that the decline in the death rate starts at about where you'd expect it to start if the vaccine did what it is supposed to do. And the people dying when the vaccine rollout started are people who caught it before the vaccine was available and only then reached the point where the disease killed them.
In a world filled with garbage analyses, this one is exceptionally bad. I flagged it because this isn't the kind of weird garbage that I come to HN to see.
And it may get even worse than that: in a world with dozens of countries vaccinated, statistically one is expected to show correlations where none exists.
Another aspect: sociological effects of people thinking enough people are vaccinated prior to herd immunity.
What a misleading article.
1) The weekly excess death tend to always largely fluctuate as far as I know.
2) If there really is a causation between excess death and vaccination, shouldn't the number of weekly excess death increase when the relative in crease in the amount of vaccinated people is higher (e.g. the steps between the measurement points is larger)?
3) The drop at the end is also easily explainable, it's a statistics over the first vaccination and vaccination takes some time to take effect and increasingly more people over the same time got vaccinated and once a large amount of people are vaccinated the excess death due COVID will notable fall. So not really that surprising.
Did I miss something?
1) If you have a large enough population sample, excess deaths fluctuate less because of the law of large numbers
The US year-on-year excess death charts are much smoother than those for a small country like Belgium for example for this reason.
So yes, Scotland has too few people for you to see much in this data due to all the noise, humans easily see patterns where there aren't any.
2) This sort of nonsense generally posits some vague association between vaccination and negative outcomes and isn't too interested in trying to imagine an actual mechanism that would inevitably be debunked.
3) It's also noteworthy that over this same period the case rates subsided in the UK considerably. Whether I'm vaccinated against the disease or not, if nobody I meet has the infectious disease I'm not going to catch it from them.
IMNSHO a much more significant effect, which I measured over the whole UK since that's the data I had, was that the ratio between those dying aged 90+ and those dying at my age fell dramatically a few weeks after they were vaccinated - exactly as you'd expect if vaccinations work.
Before the uptick due to the Delta variant, the best overall week for UK COVID deaths seemed to be at the end of May, but if you were a young adult, say age 25, death rates were actually still higher than last summer - the overall rate was low because rates were now far better for the elderly. Because of course they'd been vaccinated. Next week vaccination will be offered to all English adults who have not previously qualified (e.g. because they were too young and had no other vulnerability) so by August the proof will be in whether that arrests further transmission or not.
Countries that pursued elimination (e.g. New Zealand) are still doing better, because, vaccine or no vaccine, if nobody has the virus then nobody can die from it.
So are vaccinations causing excess deaths? Just remember that correlations to not imply causation. To understand why, imagine repeating this analysis using a sliding window with a width of one year. In the last year, you will see an increase in excess deaths and vaccinations. It does not imply vaccinations cause excess deaths. The causation goes the other way... the excess deaths caused the vaccinations
yep... Lets say someone says bananas help with covid, so when there is an increase in excess deaths, people start eating it, afraid of going bananas. You just cant later see the bananas graph and say "oh bananas!". :)
That’s right. You can’t get at causation by passively looking at data. You need an intervention. One stage of a clinical trial looks at safety by giving some people the vaccine and other people a placebo. You then look for excess deaths in the vaccinated people. The author is ignoring crucial clinical trial interventions and instead using shoddy correlations to passively imply causations that do not exist.
The people that volunteered for the clinical trials took the real risk and this article is a disservice to their altruism
There are a good number of examples that show the problem with correlation being miss understood.
https://www.google.com/amp/s/blog.statwolf.com/does-eating-c...
Can you really simplify something this complex this much?
What happens if we, say, include people who died of covid despite getting the vaccine? Would this still stand?
Edit: read spadez other links. This is basically monty python witch scene plus some graphs and citations to make it look legit.
As far as I know, this also doesn't show a correlation. This is just right now showing a coincidence no?
Wouldn't you need to compute the correlation coefficient or at least plot a scatter graph?
A follow up to the post..
https://drowningindatadotblog.wordpress.com/2021/05/27/linki...
And there is also one from CDC data..
Interesting thanks for sharing not sure why this is being downvoted but I'd consider myself a layperson on this matter. Could someone help explain why this post above is worthy of downvotes?
I scrolled though a bit wondering that too. Seems that buried in the post is a graph that shows it actually was a dramatic plunge in deaths correlated with the vaccination. I say ‘buried’ because it is surrounded by graphs and text concerned about a relatively much smaller increase in uncategorized deaths. It looks intellectually lazy. Similar stuff has been posted many times and this doesn’t appear to add to the discussion of COVID. The links may confirm anti-COVID biases and confuse laymen, but does not seem likely to withstand rigorous analysis.
And that much smaller increase is because a death may initially be coded as uncategorized and later revised. So at any point in time, the most recent data you look at will show this phenomena - COVID, vaccine or not.
>a graph that shows it actually was a dramatic plunge in deaths correlated with the vaccination.
Please share the link to the page and section heading under which this graph is present.
Also this page seem to be updated https://austingwalters.com/covid19-vaccine-risks/ with the information about observation of increase in R00-R99 deaths for current time.
>but does not seem likely to withstand rigorous analysis.
But where is this rigorous analysis?
Well, take the second link for example: https://austingwalters.com/covid19-vaccine-risks/
1) The chart shown mixes up lipid concentrations for protein spikes from the original cited paper and data
2) R00-R99 coded deaths will always be higher for "new" data. Over time, these are revised and coded differently. So if you look backwards from today, it will always seem like this number has gone way up the last month or so. The author notes this in his post (good), but leaves the original one in its place (also fine, but just be aware).
The author's hypothesis is not supported by data, and the arguments put forth are based on misunderstanding or misinterpreting things.
Also if the fear here is the mRNA vaccines are dangerous, there's always the more traditional Janssen (J&J) vaccine which does not use mRNA.
This page seem to be updated https://austingwalters.com/covid19-vaccine-risks/ with the information about observation of increase in R00-R99 deaths for current time. But the article does not appear to believe that it changes their reading of the data.
It's not worthy of downvotes. But people vote with their emotions, not based on any rational criteria.
The OP also seems to be the submitter - that explains it. People are downvoting because they don't like the article ( and you can't just downvote a submission, strangely.)
If you have enough fake internet points you can flag a submission.
Does one flag low quality or flawed content?
I'm not sure that's against the rules.
> If a story is spam or off-topic, flag it.
Seems to pass that.
What we should expect, here, is a decrease of excess death.
If we were seeing that it would seem logical to attribute this effect to the vaccination.
Well it is a WordPress.com article...