> The negative estimates in the final period arguably suggest different behaviour and/or exposure patterns in the vaccinated and unvaccinated cohorts causing underestimation of the VE. This was likely the result of Omicron spreading rapidly initially through single (super-spreading) events causing many infections among young, vaccinated individuals.

That's from the linked paper. The authors don't claim at all that the vaccine has negative efficacy, and the study is simply not set up to measure that anyway. You cannot draw this conclusion from the data in this paper.

The study indeed did not set out to prove negative efficacy but a negative VE of 75% is a result that absolutely warrants further research.

Assuming that Denmark vaccinated oldest/most vulnerable first, this could be a confounded estimate (because age both means you got the vaccine first, and that your immune system will be less effective against Covid).

Personally, I think it's a statistical fluke, and was going to argue that it must be due to a smaller sample size until I saw the width of the confidence intervals (CI's get narrower as observations increase).

Very odd, and I'll be surprised if that makes it through peer review intact.

It's a statistical fluke that has been observed with Delta as well, particularly with the AstraZeneca vaccine:

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3949410

It stands to reason that the unvaccinated at this point are in large parts hermits who are "protected" by their lifestyle. However, it's also plausible that the non-neutralizing antibodies from the vaccine induce a suboptimal immune response with Omikron.