Date: Thu, 7 Apr 2005 10:29:35 -0700
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject: Re: Stat Opinion Question
Content-type: text/plain; charset=US-ASCII
"Worden, Nina" <nina.worden@SUMMITMARKETING.COM> wrote:
> If you have a patient that has numerous conditions and you want to put
> them in a group that represents the primary condition that's effecting
> their health, how do you go about making that determination? What
> tool/process/method would you apply to the data? I realize this is an
> opinion question and will most likely receive various answers which is
> OK considering I'm looking for ideas.
If they're all under the same general ICD-9 grouping, then there
ought to be a general condition you can use as a label.
But I'll differ from Venita slightly. I'm not a doctor, and I don't
even get to play one on television. As a statistician, it's not my
place to be making assessments about primary condition. If the patient
ID has a set of condition codes for heart arrhythmia and kidney stones
and Reynaud's syndrome and Dominican Republic syndrome and Peyronie's
disease, then I certainly don't think that *I* should be making that
'primary condition' call. Work with an internist on this.
David Cassell, CSC
Senior computing specialist