OHDSI Home | Forums | Wiki | Github

Historical HCPCS codes?

Vocabulary Users
#1

Does anybody know the answers to any of the following questions?

  • Where can I get all (or most) of the historical HCPCS codes?
  • UMLS has HCPCS codes, but it seems that one has to download all of the releases back to ~2002 to get them. Is this correct? Anybody know of an efficient way to do this, if so?

I am trying to map all (ever existing) HCPCS codes for drugs that have been released, if possible. Thanks in advance if anyone has any ideas. If I am missing something obvious, I am happy to learn about it.

#2

@Mark_Danese Have you looked the NCI Cancer Research Network (CRN) Cancer Therapy Look-up Tables?

https://crn.cancer.gov/resources/codes.html

From their site:
“These tables contain over 8,000 NDC and 500 procedure cancer treatment-related codes. The tables are updated when new products or codes are identified. The codes are stratified by class of product (e.g., chemotherapy, hormone therapy, immunotherapy) and code type (e.g., ICD-9, HCPCS).”

It looks like one of the CRN sources is the NCI Procedure Codes for SEER-Medicare Analyses.

https://healthcaredelivery.cancer.gov/seermedicare/considerations/procedure_codes.html

From their site:
"The tables contain codes for procedures that are frequently included in SEER-Medicare analyses.’

I am not sure how far back these resources go for listing HPCS codes goes but the CRN has a contact for collaboration questions:

https://crn.cancer.gov/collaboration/

I am planning to contact the CRN about using the Cancer Therapy Look-up Tables in an algorithm to abstract/derive higher level oncology treatments from lower level observational events.

#3

Thanks Michael. I have seen the CRN page (which is really nice) and the SEER Medicare page. I don’t know where it comes from, but it is definitely useful. We have not validated our list against that one. It is yet another NCI resource that isn’t linked to other NCI resources.

I would really like it if I could validate the HCPCS codes against the actual historical HCPCS data, if possible. They we would have a pipeline we could update.

#4

@Mark_Danese:

What do you mean by “validate a HCPCS code”?

#5

Sorry. Bad choice of words. I would like to be able to identify all of the historical HCPCS codes for drugs. I am creating a master list of all “chemotherapy” agents, and there are a number of older codes that I need to support older datasets.

I am actually comparing what we find against a master list compiled by NCI, so I am validating our approach against their list. And, of course, it works both ways. If we identify relevant codes not on their list, we can add them. But once we have it all programmed, then we only have to worry about new codes.

#6

@Mark_Danese
Can you explain a little about the use case(s) that drive the necessity of having a master list of “chemotherapy” agents? I am sympathetic to this project. I would like to be able to reliably pull from a patient’s pile of prescriptions and administrations the ones performed as an oncology treatment. Organizing the welter, so to speak. Your “chemotherapy” agents list would seen to be aligned with this need.

#7

Our use case seems to be similar to yours. In order to do analyses of SEER Medicare data (or any other oncology data set), we need to know what treatments (if any) were received over time. In claims data, we have to infer first line, second line, etc treatment based on the pattern of reimbursement claims (generally NDC and HCPCS codes). We have our own software application that we use to create code sets (and more complicated temporal logic) to find these things in electronic data.

We would like to have as complete a listing as possible so that people who want to do studies of SEER Medicare data can do feasibility assessments for studies – for example “How many patients diagnosed between 2010 and 2013 with colorectal cancer received irinotecan within 90 days of diagnosis?”

We work primarily with the source vocabulary, so having all of the codes is helpful for complete ascertainment. I mentioned it here because I think it would also help OHDSI to have all of the historical HCPCS codes so they can be mapped to RxNorm.

#8

Will do.

We do that already. We should be find going forward.

t