I'm not sure I understand the significance of "average" for a dx. When the doctor evaluates a patient, it is their symptoms, health history, etc. that has consideration. For a diagnosis the evaulation begins with a differential diagnosis with highest probibility as an under lying cause for the malady. Eliminate the highest probability and that may be algorithmetic based and the probability represents a mean. but it doesn't stop there... then most likely cause, diagnosis, etc. If unfound the malady exists but cause would be listed as unknown. There is no unversal norm! If there were a universal norm for an illness there wouldn't be any need to engage in a differential diagnosis, continue research, different medication, different symptoms but the same illness...the list goes no and on....
I fail to see the problem you are trying to put across here? If a Doctor has 10 patients with the same condition and they all die within a week from usual protocol measures, are you saying a Doctor shouldn't think a change in protocol is required for the 11th patient? taking what you would call a risk, but in fact offers more hope? I don't believe Cardiologists take unnecessary risks with any patient, it simply wouldn't help either their career or the confidence in the hospital. If a patient is suffering, then something has to be done and as far as I know, the relevant tests are performed to try and establish the underlying cause. When I had my MI, my Cardiologist didn't simply look at charts or take averages, I had blood tests, echo and ECG. This 'confirmed' I was having MI and emergency stenting was required. There are times when historical data is important and necessary, this is used when deciding on the best drug regimes for example. You need feedback to establish best treatments and without historical data each patient would have no starting point and be subjected to a variety of trials. Would you rather have a medication proved to help millions of patients, or be tested on many which are untested as yet and could possibly give no help for your condition?
So I guess I'm not too sure what you are trying to say and perhaps you need to be just a bit more specific ?