Yes, they have to do this. The cost of a few bad drug approvals that cause large-scale loss of confidence in the pharma industry would be devastating- including consequential effects on good medications.
We can imagine lots of alternatives to what the FDA currently does - a partial approval pipeline immediately comes to mind in which the degree to which a drug is approved is commensurate to our confidence in the safety/efficacy of the drug. A different system could have different levels of approval on efficacy and safety. A different system could attempt to maximize lives saved (or utility years) and would look very different yet.
There are a lot of different approaches that could be taken that would definitely be more effective in the short term and I too wish we could take the most effective and logical approach. The problem is people are not always logical, especially on a large scale.
A proposal like you mentioned puts a lot of trust in the general public to be capable of making informed decisions based on detailed approval levels and risk profiles and not react to negative outcomes or side effects by blaming the FDA, pharmaceutical science, or the medical field.
It feels like putting that kind of complication into something as important and life affecting as pharmaceuticals is just asking for trouble when we live in a world where the grades for maple syrup needed to be simplified and large numbers of people think the President has direct control of gas prices.
Especially in the age of social media, it only takes a small increase in the number of people reacting illogically and lashing out to snowball, create loud counter narratives, and threaten the view of the FDA as a trusted authority and its ability to provide strong oversight. The result being increasing numbers of people harmed because they chose to avoid care or seek out sham alternative treatments.
We can already see these issues cropping up with regard to things like vaccines where factors like COVID vaccines being rolled out under early approval (an unfortunately necessary special case) has hurt public perception of the FDA.
The unhappy fact is that any approach that does not take into account the need to protect people from themselves, make efforts to minimize bad PR, and ensure the FDA appears as completely as possible to never make mistakes will result in an erosion of trust and even worse outcomes in the long term.
