Thanks for the input, Dan.

The code is being ported to R (the original was written in MS-Access VBA). As 
soon as porting is done, I will post the two tables (and their contents) and 
the relatively short code.

The problem is that in the case of "sometimes", the organs involved are covered 
by peritoneum/pleura/pericardium (a category of tissue called a "serous 
membrane", which secretes a small amount of lubricating fluid allowing the 
organs to move about without friction. The serous membrane also acts as a 
defense mechanism - albeit a modest one - against infection following the 
puncture of a hollow organ).

Mesothelioma (in the histology range 9050-55) is a tumor of the serous 
membranes almost invariably caused by asbestos exposure (the actor Steve 
McQueen, of "Bullitt" fame, died of this cancer). It can certainly occur in the 
places where it is flagged. (It would spread to the SURFACE of the heart, 
colon, lung or intestine, but not develop INSIDE it.)

However, mesothelioma, while one of the less common tumors, is so well known 
(thanks to asbestos litigation) that when one codes it, it may be reasonable to 
assume that the histology result is not in error. If the anatomic location is 
compatible with the tumor morphology/behavior, it should be flagged as valid. 
In other words, I would interpret the "and sometimes" as valid.
-------

From: Dan Connolly 
[mailto:[email protected]]<mailto:[mailto:[email protected]]>
Sent: Thursday, February 19, 2015 4:37 PM
To: Nadkarni, Prakash; 
[email protected]<mailto:[email protected]>
Subject: RE: Possible bugs in SEER_Recode.sql

The table-driven approach sounds interesting. I trust you'll share the code in 
due course.

As to the exclusions, they're just transcribed by machine* from the seer 
table<http://seer.cancer.gov/siterecode/icdo3_d01272003/>. The entries you note 
are marked ", and sometimes..." so it's a question of whether you want false 
positives or false negatives, I guess. Did you check the other ", and 
sometimes..." entries?


* seer_recode.py is not published... sigh... but it's on elephant. See 
MultiSiteDev<https://informatics.gpcnetwork.org/trac/Project/wiki/MultiSiteDev>.

--
Dan


Prakash



________________________________
Notice: This UI Health Care e-mail (including attachments) is covered by the 
Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and 
may be legally privileged.  If you are not the intended recipient, you are 
hereby notified that any retention, dissemination, distribution, or copying of 
this communication is strictly prohibited.  Please reply to the sender that you 
have received the message in error, then delete it.  Thank you.
________________________________
_______________________________________________
Gpc-dev mailing list
[email protected]
http://listserv.kumc.edu/mailman/listinfo/gpc-dev

Reply via email to