When hospitalized, I tend to think of myself as an inpatient.

In fairness, I have spent a good deal of my professional career studying
the oral and written communications patterns that physicians use in
communicating with one another. There are limited parallels, to putting
things in context, to establishing dominance in medical interactions.

I'm not trying to be top cat, but I will insist on being a cat that is
not trivialized. In practice, I often intersperse certain speech or
logic patterns -- not necessarily vocabulary -- into the interaction
when a history is being taken, which can cause distinctive doubletakes,
ranging from "WHAT did you say you did for a living" to simply noting
the clinician, not even being particularly aware of doing so, start
addressing me as Doctor.

This can become high performance art. I must say that I thoroughly
enjoyed dealing with frustration at the time of my first angioplasty.
Really, I was tolerant of the sequence of medical students and interns
coming, one after the other, to practice history taking.

When an especially pompous cardiology fellow, with a gaggle of junior in
trail, came in to inquire "And how are we today? And what brings us to
the hospital?", I responded:

"Without performing a history and physical, and any indicated testing, I
cannot give an opinion on how you are. It would be a reasonable
presumption that you were brought to the hospital by the terms of your
contract.

"As far as myself (shifting tones), this is a (as I was at the time)
39-year old well developed, well nourished computer science status post
one month of onset of unstable angina of one-block effort, unresponsive
to medical management. Initial presentation was significant for
posterior pain with radiation into the left arm and sternum, exacerbated
by exercise and relieved with organic nitrates. ECG was nonspecific
other than QT prolongation, with an isoelectric ST segment. Cardiac
enzymes were within normal limits. ICU monitoring revealed occasional
bigeminy but no significant ST abnormalities..."

And on and on, system by system, test by test, rejected differential
diagnosis after rejected differential diagnosis, until coming to the key
tests that were positive. He sputtered "Mr. Berkowitz, you certainly
seem to know a lot about your disease."

Ignoring that, I switched into my most pompous professor mode, and
commented, "I'm sure you noted the pattern of idiopathic hypokalemia.
Would you enlighten us, Doctor (this is the standard way a senior puts a
junior on the spot) about your differential diagnosis and treatment plan
for this possibly significant metabolic abnormality?"

It's more accurate to say "I'm not a physician, but I simulate them on
computers."  My major professional areas, which overlap to a degree, are
high-performance networking and medical informatics. I started in
biochemistry and moved into computer science, but I routinely read
current journals, attend continuing medical education, and interact with
physicians on a professional level.

Going back to 1970 or so, I developed various expert systems to assist
physicians or even substitute in some parts of the process. My interests
tend to be now both in expert prescribing systems and emergency/disaster
interest, with a strong interest in clinical pharmacology.

To build an expert system that will "sound right" to a physician, you
have to understand how physicians communicate with one another.  An
excellent introduction is in Michael Crichton's paperback _Five
Patients_, which is obsolescent in some areas (telemedicine and
intensive care), but has an excellent discussion of physician
communication.  A good backup for this is some study of cultural
anthropology and understanding the difference between high-context and
low-context human communications.

Inter-physician communications are extremely high-context. When
discussing history, physical, or diagnostic workup, everyone involved
has a generally common mental map. One clinician may ask a question, for
example, about the result of a part of an examination. The other
clinician may respond with an apparently unrelated laboratory result.

The two, however, are quite related in the underlying context. The first
physician was asking about a physical exam finding that might suggest
one disease, and the second clinician realizes what the first one has in
mind. Rather than answering directly, he or she might cite a laboratory
value that rules out the first hypothesis. Crichton gives an example of
two physicians ruling out diabetic complications without ever mentioning
diabetes.

BEVERAGE WARNING....

Ex-wife #2 developed abdominal pain, and we rushed off to the ER. Our
personal internist happened to be there, and suggested he also do a
pelvic.

Unfortunately, Michael had sprained his wrist a couple of weeks ago,
playing tennis. It was his exploring hand. Unfortunately, during the
pelvic, there was a loud CRAAACK from his wrist, followed by a scream
from my wife. Nothing broken, but he had a tendon whip back and forth
and cause an involuntary jerk of his whole hand.

It happened that the internist, my wife, and myself happened to use the
same hand surgeon. My wife said that before Michael would be allowed to
touch her there again, he would have to have a note clearing him from
Bruce, the hand surgeon.

When next we were at the hand surgeon's office, she told him the story.
We had forgotten that before he retrained in hand surgery, he was a
board-certified OB/GYN.  He laughed so hard that he had tears running
down his face, and then called the internist to lecture him (and, given
the tone of his voice and continuing hysterics, the rest of the office0.