Posted: Jun 16, 2010 5:26 pm
by DST70
A universal 1:1 set mapping symptoms to disease would surely be the Holy Grail of medicine; albeit utterly unobtainable.

But the problem in any case is that if it's signs and symptoms that show variance from case to case, this needs not only to be accounted for, but accounted for in a way that makes it irrelevant to the treatment of a given disease based on pathological causes and pathological effects. How do you do that?

And what do you do about illness that has no obvious underlying physical pathology, and therefore where you don't have obvious access to any discrete disease entity?