"confer" is something you use for a proven link with an interventional study, not for some random link you found within the thousands of variables you went fishing for in your observation study dataset.
"confer" is something you use for a proven link with an interventional study, not for some random link you found within the thousands of variables you went fishing for in your observation study dataset.
It's likely partial because vaccines only confer resistance against one subset of virii and there may be nonviral Alzheimer's too. in any case causality is not really in question in that study
Going in detail to the claim that 'appear to confer' is biased, even if we take such a strict understanding of confer, compare the semantic effect of affirming and negating the sentence with and without 'appear to'
1. 'the vaccines conferred protection'
2. 'the vaccines did not confer protection'
3. 'the vaccines appeared to confer protection'
4. 'the vaccines did not appear to confer protection'
Which of the sentence above is closest to the truth, and which is most inaccurate?
Double blind randomized controled trial or it didn't happen.
Perhaps people with early symptoms are not given the vacine because it's too expensive and not mandatory?
Perhaps people with the vaccine are younger becase it's a recent vaccine?
Perhaps people that has no money for the vaccina had no money for therapies to slow the sympthoms?
'The first double-blinded randomised clinical trial to test the effectiveness of antivirals against dementia is now under way. A group of researchers mostly based at Columbia University are testing whether valacyclovir, an antiviral used against HSV1, can slow down cognitive decline in people with early stage Alzheimer’s. Between 2018 and 2024, the researchers recruited 120 patients and treated half with the antiviral. They expect to publish their findings later this year. John Hardy, whose research forms the basis of the dominant amyloid theory of Alzheimer’s, and who has been a critic of the virus theory, says that a positive result in this trial would begin to convince him otherwise. If Dr Geldsetzer and his team can secure the funding, a similar trial of the shingles vaccine may soon follow.'
https://taggs.hhs.gov/Content/Data/HHS_Grants_Terminated.pdf
The problem is mixing the real protection of the vaccine against shingles that has been tested in RTC and new aplications that have not been tested enough.
If there are too many false provisional anouncements, people will not thrust medicine.
The article then proceeds on the history of the scientific debate on whether the HSV1 virus has a causal link to Alzheimer's and the level of acceptance of the hypothesis.
Finally, the article concludes describing how an RCT is taking place to find whether an antiviral which is used to treat HSV1 has an effect on the development of dementia.
It all ties together with the theme of the article, which is about the possibility that viruses may trigger Alzheimer's and the reason why research is being carried out in that direction. That open question is literally expressed in the headline.
I honestly cannot understand the problem you are seeing in the article. How else would you express the same information?
This is just cargo culting. Read the original source paper and you'll understand why the causal relationship makes sense without an RCT.
I found https://www.nature.com/articles/s41591-024-03201-5
They compare people that got the old vaccines in 2014-2017 with people that got the new vaccine in 2018-2020. Some strange parts:
They say they are analyzing 6 years, but for the second group the 6 years period has not ended. This makes the comparison difficult.
These periods includes the covid-19 epidemic in 2020, with lock downs, changes in death rates and new vaccines. This makes any comparison difficult.
The comparison in figure "1 e" have very different initial slopes. I'd expect dementia to be a slow accumulation of small problems in the brain, so I'd expect an initial equal slope and after some time a difference. They start the comparison after 3 month, but I expect a slower effect.
Figure "1 f" has the same initial slope difference, but there is also a strange slope change in d=1750 and the red line is more noisy probably due to the unfinished 6 years period).
They rule out the objections in your top post — as well as whether people chose to get the vaccine, which would be a huge confounder for a disease that affects cognition.