zosima

This is ridiculous.

Your meta review has like 80%-90% of the studies on semaglutide and liraglutide. When finding random effect per drug, they even say this:

> Hence, the drugs eligible for meta-regression were semaglutide (subcutaneous injection, once weekly) and liraglutide (subcutaneous injection, once daily), which both showed a dose-dependent treatment effect in terms of weight and BMI.

And the class-effect is much more modest than semaglutide by itself would have shown, exactly as I said. What is the point?

But has only weak/moderate effect on weight loss compared to semaglutide, tirzepatide and retatrutide.

Alzheimer is a lot more difficult to measure than weight and amyloid-beta in the brain obviously significantly harder to target than the GLP-1 receptor in the body.

Also, a review pooling together all GLP-1's (including those that failed in development) and concluding that they as a class have only moderate or weak effect on weight loss, would obviously be badly misleading.

Yes, the new anti-amyloids are not a revolutionary therapy. But they do have a small but definite effect on disease-progression. (something which in a highly noisy indication like Alzheimer's may actually mean a lot more to the patients than it seems by looking at the data. E.g. donepezil and memantine have very modest effects too, but patients and caregivers are still surprisingly positive about them).

And since these were developed using different methods and epitopes, it does not make sense to pool them together when doing a comparison. It's obvious that if you combine modestly beneficial compounds and compounds with zero beneficial effect then the mean effect will be worse than the modestly beneficial compounds.

And when you do so just because they were aimed at the same molecular target, not even considering if anyone claims that the old, unapproved antibody-therapies, with no positive studies to support them, have any effect, what you are doing is somewhere between extreme ignorance and deception.

I wonder. Because if that was the question, why would they include anti-amyloid treatments that were never approved and are not part of any current treatment regimen and dilute the real effect of the approved antibodies? (Not that the approved antibodies are a silver bullet, but clearly they have some positive effect.)

That systematic review is unfortunately severely flawed. They pooled the old anti-amyloid antibodies (which everybody agrees are inefficacious) and were never used outside clinical trials, with the new anti-amyloid antibodies, that does show modest efficacy in reducing the speed of progression.

While these two classes of drugs target the same molecule, they are developed to target amyloid-beta in very different forms. Pooling all those drugs together, and then showing very weak effect, makes absolutely zero sense, and I hope and expect this review to be pulled soon.

The genetic evidence that amyloid-beta is in some way causative of at least a subset of Alzheimer's disease is still extremely strong.

The fact that amyloid-beta binding drugs like Lecanemab and Donanemab decrease progression is also very strong evidence for the amyloid-beta hypothesis.

What seemed to be the hurdle was likely, that targeting amyloid-beta in its pathological form just wasn't easy and developers of pharmaceuticals instead focused (for unknown reasons) on the solid amyloid-beta grains in the brain.

The major reason things are moving so slowly is because of the long development times of pharmaceutical compounds. A new compound will at least have to spend 2-3 years in safety studies before being tested on its first real patient, and then for Alzheimer's, which is a slowly developing disorder, nothing short of a 2-year followup time will show efficacy unless it's truly a miracle compound.

What has closing a government agency to do with free speech?

Did you interpret the calls for the end of censoring and cancellations that all government agencies must continue to exist forever?

And if so is there any resemblance of logic behind that interpretation?

I wouldn't bother commenting if I were hallucinating figures. There is no figure 4a.

If you look at figure 4 in the supplemental material you also see, per your expertise, that covid vaccine protects against traumatic injury. However even adjusting for the protective effect against traumatic injury there is still quite a large protective effect against all-cause mortality. So the beneficial effect of the vaccine is not solely caused by its protective effect against traumatic injury.

Or it could be, bold proposition I know, that there is a difference between the groups that both protects against traumatic injury and protects against all-cause mortality, independently of the vaccine.

This is rather weird. Mortality in immediate connection with the vaccine (index time) would not have been captured here. I would hesitate to draw any conclusion from this paper.

> For all individuals, vaccinated or not, follow-up time zero began 6 months after the index date.

I am speaking about what the paper shows.

There are other sources of evidence for efficacy. This paper is not a very strong source of evidence for efficacy due to some obvious uncontrolled difference between groups.

The reduction in all-cause mortality was independent of covid deaths.

Which seems to suggest that there was big differences between the groups other than the vaccination.

This of course does not change that the vaccine seems mostly safe, but it definitely calls in to question whether the protection against covid death was vaccine-mediated or due to some other difference between the groups.

Therefore this paper is moderately strong evidence for the vaccine being safe, but quite weak evidence for the vaccine being efficacious.

Maybe those you call extremists are now the only sane people.

People who are poorer and have worse health are predisposed to live in cheaper dwellings, many of which are closer to roads and thus more noisy and with more air pollution.

People who are poorer and have worse health, also have an increased incidence of dementia, seemingly independently of the number of particles in their dwellings.

This is not "everyone".

I'd hope we'd fire anyone incompetent that's in a position to control the interest rate.

Yes, and so is big parts of Russia. Attacking one country is not the same as attacking a continent.

No, it's not. Social media is massively censored in many EU countries (and UK).

Russia has attacked Ukraine. Not Europe.

Neither Ukraine nor Israel is part of EU or NATO.

Copyright is not censorship.

Censorship is state/company mandated retraction or blockage of certain information. Copyright is state/company mandated blocking of certain forms of expression.

Copyright permits you to publish any idea you so desire, only that you don't plagiarize someone else while doing so. (Which is always possible, as the fair-use doctrine is a thing)

The vast majority of known blocks are missing on that list.

It's in fact worse than no list as it gives the impression of transparency, where there isn't any.

That sounds really easy and straightforward. And yeah, committee decisions are well known for their technical excellence and far-sightedness.

Most of these are bad for most people. They do something superficially useful, but ultimately blocks innovation and small companies and so leads to large companies being protected from competition.

Blocked by court or state. Private ISPs can't block sites.

And this is true for almost all these legislations. They remove power from individuals and companies, while avoiding to limit EU and state power in any meaningful way.

Yes, this is the narrative, supported by the funding agencies.

Also, I'd like to add that it's almost always humans of European descent who are committing the dastardly deeds.

Different stories, same funding agencies. Or at least same global goals for the funding agencies.

I think tolerance builds up fast enough for that to be a rather boring experiment.

The beautifully named 32024R1689, aka AI Act, prohibits a lot of random stuff. It definitely makes many AI efforts into a legal minefield. It does not just cover live facial recognition in public spaces, which I personally could live without.

"Private parties" refer to non-governmental entities, such as individuals or businesses. You may be acting on a governments behalf, but I am not.

I don't get this attitude. Private parties are me and you. I have many interests and ideas, and now many of those have been forbidden for no discernible reason, while the government is still allowed to spy on us to their hearts content.

I can't really begin to fathom how this is good.

I edited mine after you edited yours.

Here is an article about live/post facial recognition:

https://www.politico.eu/article/eu-ai-facial-recognition-tec...

What is non-factual?

We have another token legislation from EU forbidding private parties to most anything, and carefully inserting loopholes for authorities and government to do as they please.

True, the restrictions on live facial recognition is a bit more severe for law enforcement than usual.

But: A. It's not something most people here care about a lot. Law enforcement are still allowed to use AI to create a file on every citizen. B. It gives them political points, because now people less-in-the-know will think that they are actually protecting privacy, which is again, not true.

They have outlawed _live_ facial recognition in public places. And with exemptions such as e.g. terrorism, which I'm guessing is what UK is going to go for with protesters.