Unfortunately this is the status quo, not shenanigans. The data the FDA relies on to approve drugs and vaccines is considered a trade secret and is generally never made public.
Yeah fair you can FOIA it, if you have the money and are willing to wait multiple years. But the data is never released without doing that. The point is there is no way for the average person to know whether the drugs they are prescribed actually work or not, nor is there any way for doctors to know whether the drugs they're prescribing actually work or not, because whatever is published in the academic literature doesn't reflect the trial data. E.g.
"The fact that what gets published are the trials with positive results was most convincingly shown by a group of researchers at the Oregon Health and Science University, who followed up on our initial analysis of the FDA data by comparing the conclusions reached by the FDA with those reported by the drug companies in journal articles. Of 38 drug-company clinical trials that the FDA viewed as having positive results, all but one was published. In the same documents, the FDA described 36 trials as having negative or questionable results. Most of these negative trials were not published at all, and of the few that were published, most were described in the journal articles as showing positive results -- despite the fact that the FDA had concluded that they had not." Source: The Emperor's New Drugs, p. 67
Drug companies consider the raw data a trade secret, so they are never going to give it to you themselves.
So, releasing the data is hard by default because of what it would require - you are talking hundreds of thousands of pages for a single drug.
It would take a very large permanent staff to do it.
Or a new set of requirements placed on pharma for how data is submitted to reduce that burden.
I'm not opposed, mind you, just saying what happens now would make it hard for the FDA to achieve "release by default" in practice[1]
Changing the rules here would also make things more adversarial in practice than they are now (for better or worse).
I think the value in the end is probably worth it, but i don't think it's as obvious a win as it is on paper. It's very easy to say how things should be abstractly. It's very hard to make systems that work in practice, and it's never as theoretically nice as we want them to be, nor can you really force them to be (which is often the proposed solution).
[1] you will also likely have to play whack-a-mole for a while
(not saying it's not worth it mind you, just that it's not easy). IE you will now be given 5x the data just because they know it will take you a while to sort through it. It will take a few iterations and lawsuits to get regulation/etc right to cause this not to happen.
