Skip site navigation (1) Skip section navigation (2)

Re: surrogate key or not?

From: Karsten Hilbert <Karsten(dot)Hilbert(at)gmx(dot)net>
To: pgsql-sql(at)postgresql(dot)org
Subject: Re: surrogate key or not?
Date: 2004-07-26 20:34:41
Message-ID: 20040726223441.T699@hermes.hilbert.loc (view raw, whole thread or download thread mbox)
Lists: pgsql-sql
> This reminds me of a project I worked on many years ago, I was pretty much
> fresh out of university writing a system for a large medical practice -
> itwas principally for accounting purposes. I made lots of suggestions like
> Josh's, only to get replies like Karsten's. I the progammer wanted to codify
> everything so as to enable data analysis (ie linking complaints and
> diagnosis, etc) but the doctors wern't interested.
Likely it wasn't their intent with the system. That doesn't
mean it can't be done ... (not that I think that with todays
tools it sufficiently can but ... :->  ICPC probably comes closest
to that). Anyway, I am highly interested in increasing the data
quality in my records. However, to be able to care for a
patient *when needed* I won't allow software I write to
*force* coding upon me. Trust me I have and am still
researching coding, classifying, structuring of medical data
at an ongoing basis which funnily constantly improves my daily
abilities as a doctor. Now, back to GnuMed, we *do* allow to
code arbitrary pieces of narrative with arbitrarily many codes
from arbitrarily many coding systems. Same for classifying
(rather typing) data.

> They just wanted to write free text comments. And the reason for
> it (as far as I can tell) is the distinction between
> general practice and reseach (such as epidemiology).
Rather it is the difference between reality and theory. In
reality you are dealing with tens of patients with 1-5
problems almost neither of which are "quite right" if you go
by the textbooks. Nevertheless everyone expects you'll never
forget/do wrong a thing.

> So (GPs) are not so much searching for new knowlege in their patients records,
> as applying the knowlege gained from research (done by researchers) to treat
> individual patients.
One thing we ARE looking for in our records is the ability to
find groups of patients by arbitrary criteria since one day
I'll have to find all my patients whose father took a statine,
whose second-born child suffered a bout of neutropenia 2 weeks
after birth and who started being on the pill at age 14.
Because they'll have a 3fold increased risk of lung embolus.
Unless monitored for clotting factors every 6 months. Which I
will have to do from now on. Get my point ?  :-)

> Karsten might beg to differ, I don't know, but the "long flaky text" comment
> triggered some old (and fond) memories ;-)
I totally understand what you are saying... We do cling to old
trusted things. Also, I used a bit of hyperbole to get my
point across.

GPG key ID E4071346 @
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346

In response to


pgsql-sql by date

Next:From: Kenneth GonsalvesDate: 2004-07-27 06:03:50
Subject: Re: surrogate key or not?
Previous:From: Andreas HaumerDate: 2004-07-26 15:03:44
Subject: Re: Trigger functions with dynamic SQL

Privacy Policy | About PostgreSQL
Copyright © 1996-2017 The PostgreSQL Global Development Group