Differences

This shows you the differences between two versions of the page.

--- documentation:oncology:meeting_notes_2019_aug-06 [2019/08/07 12:30] – mgurley
+++ documentation:oncology:meeting_notes_2019_aug-06 [2019/09/16 04:40] (current) – mgurley
@@ Line 7: / Line 7: @@
     * Michael Gurley shared updated DDL for the OMOP Oncology CDM extension. [[https://github.com/OHDSI/OncologyWG/tree/master/ddl|DDL]].   The update included DDL for the CONCEPT_NUMERIC table and a CSV file and script to load the new table.
     * It was announced that a new release of NAACCR within the OMOP vocabulary had been made.  Available at http://athena.ohdsi.org/.  This new release of NAACCR added support for relationships necessary to support numeric NAACCR concepts and some fixes to the NAACCR treatment items and item codes.
-    *  The group discussed how to handle overlapping NAACCR schemas.  This is technical issue for full vocabulary support of NAACCR.  The current approach is to collapse schemas when a site/histology combination falls into more than one NAACCR schema.  There are cases where the same site/histology combination is present in more than one schema.  In thse case,  NAACCR requires other variables to be looked at in order to reveal the semantics of some NAACCR items and/or NAACCR item codes.  Sometimes a NAACCR variable schema discriminator, one other time Gender.  This is a consequence of NAACCR's design flaw in reusing codes across unique semantic concepts.  The group agreed that collapsing schemas is OK in cases where it does not corrupt semantics but a solution must be come up for the non-collapsible cases.  Once possible solution would be to associate the other NAACCR variable necessary for mapping to a NAACCR schemes within the OMOP vocabulary.
+    *  The group discussed how to handle overlapping NAACCR schemas.  This is technical issue for full vocabulary support of NAACCR.  The current approach is to collapse schemas when a site/histology combination falls into more than one NAACCR schema.  There are cases where the same site/histology combination is present in more than one schema.  In these cases,  NAACCR requires other variables to be looked at in order to reveal the semantics of some NAACCR items and/or NAACCR item codes: sometimes a NAACCR variable schema discriminator 1, 2 or 3 (NAACCR items 3926, 3927, 3928), other times Sex (NAACCR item 220).  There might be others.  This is a consequence of NAACCR's design flaw in reusing codes across unique semantic concepts.  The group agreed that collapsing schemas is OK in cases where it does not corrupt semantics but a solution must be come up for the non-collapsible cases.  Once possible solution would be to associate the other NAACCR variables necessary for mapping to a NAACCR schema within the OMOP vocabulary.
+    * It was announced that a OMOP Oncology Hackathon will take place on August 21st and August 22nd in the offices of IQVIA in New York city.  Details on exact timing and location to follow.