Date:         Thu, 7 Apr 2005 10:29:35 -0700
Reply-To:     cassell.david@EPAMAIL.EPA.GOV
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject:      Re: Stat Opinion Question
In-Reply-To:  <1B2D45DABE9FD411822F00E018C49EEFBA9634@SRHD_EXCH>
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
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician