Okay, thanks for the quick responses! We will analyze our data in Item 419 and figure out if the data is parseable. Not surprisingly we have populated those two code groups differently. Sigh.
The good news is that I have all the Iowa variables mapped to our custom table and Alex will help me work on the date conversions next, but that sparks a new question. If a date field only contains a year, say 1996, would you prefer that we null it out or default it to a generic value like 1/1/YYYY or 7/1/YYYY? Still hoping for a 9/1 delivery, but we may be at risk. Best, Kris (414) 416-2389 (cell) Kristen Osinski, MS Business Analyst, Biomedical Informatics Clinical & Translational Science Institute Medical College of Wisconsin [email protected]<mailto:[email protected]> (414) 805-7245 (office) From: Dan Connolly <[email protected]> Date: Friday, August 28, 2020 at 3:04 PM To: "McDowell, Bradley D" <[email protected]>, "Osinski, Kristen" <[email protected]>, "[email protected]" <[email protected]> Subject: Re: [External] DROC 82 Tumor Table Mapping ATTENTION: This email originated from a sender outside of MCW. Use caution when clicking on links or opening attachments. ________________________________ It was semi-conscious. That is: I asked about it (July 7<https://urldefense.com/v3/__https:/informatics.gpcnetwork.org/trac/Project/ticket/749*comment:34__;Iw!!H8mHWRdzp34!ob5kGdi9WxZS55Uaqikeg2KBAFXZAwhMtMsyHuVyVuYruF3R2fBpm6wf8Df8Sxg$>) and you said OK, but I doubt you realized the impact. Also noted as https://github.com/kumc-bmi/naaccr-tumor-data/issues/19<https://urldefense.com/v3/__https:/github.com/kumc-bmi/naaccr-tumor-data/issues/19__;!!H8mHWRdzp34!ob5kGdi9WxZS55Uaqikeg2KBAFXZAwhMtMsyHuVyVuYruF3R2fBpm6wf2X2ASTw$> -- Dan ________________________________ From: Gpc-dev <[email protected]> on behalf of McDowell, Bradley D <[email protected]> Sent: Friday, August 28, 2020 1:07 PM To: Osinski, Kristen <[email protected]>; [email protected] <[email protected]> Subject: RE: [External] DROC 82 Tumor Table Mapping Hi Kris, That’s interesting… NAACCR has always had these three redundant fields for ICD-O-2, and you might have the same issue for the ICD-O-3 fields as well (numbers 521-523). It appears that 419 and 521 remain part of the NAACCR dictionary, but we did not include them in the tumor table. (I honestly can’t recall if that was a conscious decision.) I wonder if your institution’s tumor registry decided to populate 419 only and leave 420 and 430 blank since they’re redundant. Could you split your morphology variable into separate histology and behavior values and use those to populate 420 and 430? The first four characters correspond to histology and the last character corresponds to behavior (separated by a “/”). Thanks! Brad From: Osinski, Kristen <[email protected]> Sent: Friday, August 28, 2020 12:03 PM To: [email protected] Cc: McDowell, Bradley D <[email protected]> Subject: [External] DROC 82 Tumor Table Mapping Hi Brad, I have found what may be a versioning mismatch in our NAACCR table vs. your Version 18 model for DROC 82. We have mappings for NAACCR Item 419 [Morph--Type&Behav ICD-O-2 ] but the new version appears to have broken Item 419 into 2 separate variables for Histology Type (420) and Behavior (430). Do you have a preference on which new variable I should map our Item 419 data to? Best, Kris (414) 416-2389 (cell) Kristen Osinski, MS Business Analyst, Biomedical Informatics Clinical & Translational Science Institute, Suite M1350 Medical College of Wisconsin 8701 Watertown Plank Road Milwaukee, WI 53226 [email protected]<mailto:[email protected]> (414) 805-7245 [signature_491657025]<https://ctsi.mcw.edu/>[signature_46665966]<https://ctsi.mcw.edu/about/history/partners/> Please cite the NIH CTSA Grant; 2UL1TR001436, 2TL1TR001437, 2KL2TR001438, and acknowledge support.
_______________________________________________ Gpc-dev mailing list [email protected] http://listserv.kumc.edu/mailman/listinfo/gpc-dev
