No one is advocating for "pay for access" or a means of bypassing trials that facilitate a "flood of [scammers]", so this is a straw man. You'd have to show why any design for relaxing regulation of trials is necessarily bad in aggregate.
The article even cites a case where the bureaucrats are confident in the safety of a trail drug but unable to approve for political reasons.
Not to mention the FDA allows plenty of snake oil supplements, etc. to be sold, so it's not particularly good at the goal you claim it's optimizing for.
FDA doesn’t regulate supplements as medication but food; however Federal consumer protection laws prevent making false claims.
I imagine allowing terminal non-trial participants access is more than just regulatory red-tape.
If understood one of the authors points it was that allowing non-candidates access could speed up approval. I think that is true, but that would require the same regimented process for administering the trial to ensure quality of data. The same would hold if there was an adverse event as well—patients need to be monitored closely to capture data to provide valuable to the clinical research results.
Supplements are regulated as food products as long as the manufacturer doesn't claim they treat or cure anything. The FDA allows them for the same reason they allow fudge brownies or Spam to be sold.
The article even cites a case where the bureaucrats are confident in the safety of a trail drug but unable to approve for political reasons.
Not to mention the FDA allows plenty of snake oil supplements, etc. to be sold, so it's not particularly good at the goal you claim it's optimizing for.