CFD Edits

[Jim Hofferkamp]

I’ve had several people test the metafile and no problems so far. The only issue that has come up is related to the edit ICD Rev No and cause of death. The edit is correct, the tables it uses were updated after last years call for data so some registries may find invalid codes in their data that didn’t show up last year.

I’m hoping to have the metafile posted next week (week of 9/26/16). However, if you would really like to get a look at it ASAP, send me an email and I’ll send it to you.

For US registries submitting to NPCR…if you run your file through the NPCR Call for Data edits, you should be fine with the NAACCR edits. The combined edit set has a few additional edits related to the new geocoded county fields. Other than that it is pretty much the same as the NPCR edit set.

[Recinda Sherman]

This year the county-tract combo edit is no longer optional.

If you experience a large number of edit failures from the county-tract combo edit, there are likely 2 problems:

1) your geocoded tract was not placed in the proper place or is wrong (many registries have found junk in the field–letters, etc).To resolve this, the geocoded data will need to be uploaded in your database again.
2) your reported county differs from the geocoded county. The registry will need to overwrite (at least for the purposes of call for data) the reported county (Item 90) with the geocoded county data. The decade of diagnosis should match the geocoded county decade (1995-1999 dx cases should use the Geocoded County 1990; 2000-2009 dx cases should use the Geocoded County 2000; 2010-2015 dx cases should use the Geocoded County 2010).
NOTE: The edit will only catch cases where a tract does not match a county. So you will need to overwrite County at DX (Item 90) with the geocoded county for ALL cases that are geocoded; not just the cases that do not pass edits. This is because tract 001100 can be in multiple counties. It might pass the edit but if the reported county does not match geocoded county, it would result in the incorrect calculation of poverty code, etc.
NOTE: For cases that are not geocoded, do not overwrite County at DX with blanks. We do not want to lose information.

[Jim Hofferkamp]

The edits Recinda is referring to are:
Census Tract 2010, State, County, 2010-2019 (NPCR)
Census Tract 2000, State, County, 2000-2009 (NPCR)

[Anonymous]

Will the CFD edits metafile need to be updated as a result of the NAACCR v16c metafile release?

[Jim Hofferkamp]

There will not be an update to the CFD metafile due to the v16c update. None of the updates to the v16 full metafile impact the edits in the cfd metafile.

[Jim Hofferkamp]

I had what was essentially the same question come in from several central registries…

For our submission we need to Use the “Coodinated” Configuration?? And the NAACCR CFD v16 meta file. Want to do both NPCR & NAACCR Submission.

My response is below…

Probably the easiest thing to do is pull one file with all of the data items requested by both NAACCR and NPCR. Run that file through the Coordinated edit set. Once the file is clean, run that through NAACCRprep. You can then opt to create a submission file for both NAACCR and for NPCR. If everything worked correctly with NAACCRprep, you shouldn’t have to run any additional edits. If you want to double check, you can run the CINA edits on the NAACCR submission file and the NPCR edits on the NPCR file.

This year NAACCRPrep has the ability to populate your NHIA and NAPIIA data items as well. If you have already run the algorithm to populate these fields, there is no need to do it again.

If you do want NAACCRprep to calculate the fields, you will have to include the patient names in the file you run through NAACCRprep. You could just select file type C when you pull your initial file. The names and any other confidential data not requested by NAACCR or NPCR will not be included in the submission file that is created by NAACCRprep.

[[email protected]]

Current CINA edits include several edits such as N1535 that were created and first applied to data in v12A but are now being applied to earlier cases in the CINA data file. This is causing an exceptional amount of manual review. Was this intended? While we agree that data cleaning is good, it is unfortunate to have these applied in a year when we are trying to get reports from facilities who are still struggling with v18 delays and when CCR staff have their attention split to v21 preparation deadlines as well. Is any relief possible?

[Author]

Posts