This is somewhat of a fallacy.

The body of knowledge, collectively referred to as science (not the scientific method itself), should indeed the testable. However, one cannot possibly have the time required in their lifetime to test the entire corpus. At some point you need to defer to experts in most fields.

I'm a geophysicist. I cannot even test the body of knowledge within my own field -- it is far too large. But from my own background and abilities, I can usually tell if something is legitimate or snake oil fairly quickly within the body of knowledge of my own field. I have to assume that fields outside of mine are going the same. If I don't, I become a crazy person.