My point is not whether the FDA okay'd this particular protocol--it's that there wasn't an already-FDA-approved medication around for this particular situation (i.e. administering to a rabies patient after symptoms have developed). My point is that there was little evidence supporting it's use before they decided to try it (in my book, four previous survivors is still "little evidence").

But tying it back to that other thread, here is a case where doctors are pushing the envelope in the treatment of rabies, but the "doing" is ahead of the "science" (at least the formalized clinical trial or other formal study design). Which is why I thought it was appropriate that the definition of science to be broader and not just limited to formal tests of hypotheses. What these physicians did will certainly add to medical knowledge, which probably would never have happened if they had to set up a formal clinical trial and waited for enough rabies patients to enroll in some study.

That said, it is certainly possible that sometime in the future, there is enough additional evidence that points to this protocol not being effective; that these survivors shared some particular unique traits or circumstances that allowed them to survive that had little to do with the protocol.