What is the difference between vocabulary ICD10 and ICD10CM? When mapping our source data I see matches in both vocabularies. Is it correct that when looking for ICD10 conditions both ICD10 and ICD10CM should be used?
ICD-10 is the WHO version. It contains less than 20 k codes. The US created its extension, like many countries did. It is almost 70 k long. If you have a code in ICD-10 and a code in ICD-10-CM they are similar, but not identical. The reason is that the ICD-10-CM relegates some of the functionality to the new codes.
So, if you are looking for ICD-10-CM: It’s in the making. We could add the codes, but without mapping it wouldn’t do you any good. The plan is to finish it in January.
Long email below with details: Bottom line – Children’s Hospital Colorado’s analysis of the coverage of the current ICD-10 codes in OMOP against ICD10 actual usage in both encounter (visit_occurrence) and billing diagnoses at our institution is not encouraging (less than 50% coverage at the visit_occurrence-level). We had hoped that we could use the current ICD-10 codes and get “reasonable” visit-level coverage even though we knew that ICD-10-CM is much larger with US-specific extensions. This seems not to be the case as detailed below. I would be interested in similar analyses by non-pediatric facilities to see if our poor coverage is due to differences in coverage of common pediatric conditions in ICD-10. Wouldn’t be the first time that the little ones get the short stick (so to speak) in terminology efforts.
In PEDSnet, we stopped our November ETL extractions as of 9/30/2015 because of the lack of mappings into ICD-10-CM, making Christian’s delivery date in January an important milestone. We will probably include ICD10 codes in our February extraction, putting the codes as source values and know that we could fail in mapping into source_concept_id and concept_id.
************** Long analysis by Greg Schultz @ Children’s Colorado *******************
From: Greg Schulte Gregory.Schulte@childrenscolorado.org
Date: Wednesday, December 30, 2015 at 2:21 PM
To: “Kahn, Michael” Michael.Kahn@ucdenver.edu
Cc: Sara Deakyne Sara.Deakyne@childrenscolorado.org, Amanda Dempsey AMANDA.DEMPSEY@UCDENVER.EDU, Charles Bailey BAILEYC@email.chop.edu
Subject: RE: ICD10 codes analysis for PEDSnet
Michael,
Please see below ICD10 analysis for PEDSnet.
Total number of distinct ICD10 codes in OMOP : 12,855
Total number of distinct ICD10 codes used in CHCO : 16,560
Total number of CHCO ICD10 Termed/Encounter Diagnoses : 9,743,266 (some of these are mapped from previous ICD9 codes)
Total number of CHCO ICD10 Billing Diagnoses: 821,004
Total number of CHCO ICD10 Termed/Encounter Diagnoses that can be mapped to OMOP ICD10 concepts : 4,671,100 (48%)
Total number of CHCO ICD10 Billing Diagnoses that can be mapped to OMOP ICD10 concepts : 388,976 (47%)
Top 10 ICD10 termed/encounter diagnoses that are not in the OMOP set:
Z51.89 Encounter for other specified aftercare
R27.9 Unspecified lack of coordination
Z00.129 Encounter for routine child health examination without abnormal findings
F80.89 Other developmental disorders of speech and language
R62.50 Unspecified lack of expected normal physiological development in childhood
M62.81 Muscle weakness (generalized)
M25.569 Pain in unspecified knee
R26.9 Unspecified abnormalities of gait and mobility
H66.90 Otitis media, unspecified, unspecified ear
C91.01 Acute lymphoblastic leukemia, in remission
Top 10 ICD10 biling diagnoses that are not in the OMOP set:
Z51.89 Encounter for other specified aftercare
M62.81 Muscle weakness (generalized)
R27.9 Unspecified lack of coordination
R62.50 Unspecified lack of expected normal physiological development in childhood
F80.89 Other developmental disorders of speech and language
R29.898 Other symptoms and signs involving the musculoskeletal system
R26.9 Unspecified abnormalities of gait and mobility
K59.00 Constipation, unspecified
J45.909 Unspecified asthma, uncomplicated
Z00.129 Encounter for routine child health examination without abnormal findings
What is the status of ICD-10-CM mapping? Planned to finish in January, but I do not seem ICD-10-CM concepts or mapping in March 2016 vocabulary.
1 Like
Done. Waiting for release. Release is waiting for finalizing drug_strength of disparate drug vocabularies.
1 Like
It’s out.
In April 2016 vocab release, there are 10,537 source codes that overlap between ICD10 and ICD10CM. Of these 7,633 point to the same SNOMED, which leaves 2,904 concepts that point to different SNOMED codes. Looking at a sample of the codes that map to different SNOMED concepts it does look like the target are similar. For example Q65.9 in ICD10 maps to SNOMED concept 77373, Congenital deformity of hip joint, and ICD10CM maps to 201956, Congenital anomaly of lower limb.
I don’t have a question and from what I see either mapping can be considered correct. Just wanted to point out the anomaly.
1 Like
You are totally right. It is being aligned and fixed as we speak.Thanks for watching out for us.
@Christian_Reich - When mapping US claims data should we use both ICD10 and ICD10CM Vocabularies or just stick to ICD10CM. We are using 20160311 Vocabulary so no overlap issue right now with CONCEPT_CODES between Vocabularies but don’t know if that will hold true forever.
Your statement “Stick to ICD10CM” makes me think that ICD10CM is inclusive of all ICD10 codes, however when checking, there are 1,982 valid codes ICD10 that do not exists in ICD10CM for OMOP Vocabulary v4.5 20-JUN-16 release.
Note: I am looking at v4.5 version, but this should be applicable to v5
select count(*)
FROM
(
select distinct source_code
FROM vocabulary_4_5_jun_16.source_to_concept_map
WHERE source_vocabulary_id = 34 and invalid_reason IS NULL
minus
select distinct source_code
FROM vocabulary_4_5_jun_16.source_to_concept_map
WHERE source_vocabulary_id = 70 and invalid_reason IS NULL
)
returns 1,982 rows
In the vocabulary files I downloaded on 2016-07-13, the concept_relationship table contains hierarchical relationships for ICD10, but not ICD10CM. Are the hierarchical relationships for ICD10CM available? Or were they just not built / added to concept_relationship?
EG: ICD10
“45571628 Iron deficiency anemia D50” Subsumes “45605373 Iron deficiency anemia, unspecified D50.9”
In ICD10CM, the concepts for concept_codes D50 and D50.9 are “1567838 Iron deficiency anemia”
and “35206695 - Iron deficiency anemia, unspecified.”
But the only relationships present in concept_relationship for these ICD10CM concepts is the mapping to/from SNOMED.
You are right. We will add these in the next release.
But why do you need them? The ICD10CM hierarchy leaves much to be desired: It is realized through “microcoding” of the identifiers (bad because if an identifier is changed for some reason the hierarchy has to change as well), it allows only one parent, and it only has a few levels in depth. I wouldn’t use it. Use SNOMED.
Thanks @Christian_Reich. Do you know when the next release will be? Our clients still use the “hierarchy”, warts and all. We have talked with them about its deficiencies, and the long term goal is to move to using SNOMED, but in the short term the hierarchy is required.
This week.
Hi @Christian_Reich: An update on the ETA for the ICD10CM and ICD10PCS hierarchical relationships? The latest vocab download does not have them.
1 Like
+1 for the ICD10PCS hierarchy.
Hi @Christian - Just curious if you have an update on when the hierarchical relationships will be available for ICD10CM and ICD10PCS ?