Where does one find downloadable category data for ICD-9 codes in CSV/XML/JSON or other structured format?

(We are looking for structured data rather than web scraping or writing parsers for semi-structured text.)

CMS offers downloads of subcategories but the category headings for each subcategory are missing.

The CDC offers ICD-9 downloads as well but the updates only go to 2011 and the file is in RTF format as opposed to CSV/XML etc.

A closed question on StackOverflow is filled with efforts that parsed those RTF files - but there appear to be no structured sources that include the latest CMS updates from 2014.

As some background... ICD-9-CM codes are three-to-five digit numeric and, in certain cases, alphanumeric codes. The first three digits in a code are called the “category.” The category describes the general illness, injury, or condition of the patient such as: 123.0 – {Disease} in Chest. The zero after the decimal point is the subcategory.

123 – {Disease} (The first three digits make up the category)

123.0 – {Disease} in Chest (The zero after the decimal point is the subcategory.)
123.00 - … uncomplicated
123.01 – … with complications in cardiac system
123.02 – … with complications in digestive system (the last digit is the subclassification. This gives even further information about the designation outlined in the subcategory. If we were to select 123.02 as our code, we’d read the full code as “{Disease} in chest, with complications in the digestive system.”
  • Bioportal lists several ICDx offerings. Is the one you need amongst them? – Grimaldi Jun 28 '17 at 19:27
  • @Grimaldi I see ICD9 listed but clicking on various options - we don't see any actual list of the data or download links – Praxiteles Jun 29 '17 at 18:33

Here are a couple I found on a quick Google search:

  • That's a great find on Github. It is nice that it adds a Python API. – Praxiteles Jun 30 '17 at 17:52
  • Icd 9 cm is listed on Bioportal to be available through UMLS. – Grimaldi Jun 30 '17 at 18:10

Wikipedia seems to have a decent listing. OK, that is not in CSV/XML/JSON, but you can use OpenRefine to parse the data.

  • We are looking for structured data rather than web scraping or writing parsers for semi-structured text - thus the format specifications in the question – Praxiteles Jun 28 '17 at 11:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.