Considering the worst case situation, if this is unsafe, to be death, and best case is a slightly better treatment, I think thorough testing and a long approval process is certainly in order.
Huh?
Suppose that it costs $1 and a quick test shows that it seems to help "a bit". Why not let folks skip a Coke for on the chance that the benefit is real while "thorough testing" goes on?
Costs and benefits apply to testing as well as deployment.
You missed the part about the worst case being death. What if using the device provides short term relief but makes 15 percent of patients die within half a year after its usage?
I read the response that pointed out that the worst case here was slightly earlier death and realized that in many cases the worst case is exactly what I said it was, someone is out the cost of a Coke.
The fact that some medical "treatments" will kill you is not necessary useful when evaluating a different treatment.
It's important to evaluate things based on their actual costs and benefits, not the costs and benefits of other things.
The FDA takes a lot more heat when someone dies from a drug they approve than when 100 people die from not having access to a drug that they drag their feet on.