Tim, Hey, I asked that same question not long ago.
Yes, I do dimly recall it now that you mention it. No satisfactory answer, I presume?
I look forward to using the parser.
After you strip all the formatting out, the NCHS RTF files have a very regular format, thank goodness.
BTW you better patent that method for parsing RTF ICD-9 files :-D
Yes, it would seem to be at least as novel as many of Microsoft's recent patent applications. But I can't afford $5000 per day patent attorneys to help me describe something completely obvious and routine in a manner that makes it seem novel, or at least sufficinetly incomprehensible that the patent examiners can't be sure if it is novel, or non-obvious, or not.
Tim C
Tim Churches wrote:
Does anyone know where a set of US ICD-9-CM codes and descriptions as plain text i.e. in a format which can be imported into databse - can be obatined at no cost? The data do not have to be re-distributable, just available on teh Internet for free. I have been able to find a free set of US ICD-9-CM files in RTF (Rich Text Format) format provided by the NCHS (National center for health Statistics), but they are laid out for printing, and would need a lot of error-prone parsing to render them as a database file. Various companies offer ASCII-format ICD-9-CM files, but only for a fee. Note that I am looking for ICD-9-CM, not ICD-9.
Sorry, forget it - I realised that it is quite easy to parse RTF files. About 100 lines of Python and I nearly have a perfect set of text from the RTF files available via anonymus FTP from NCHS. I'll include the parsing code in the packages we will be releasing by Xmas (2004).
Tim C
smime.p7s
Description: S/MIME Cryptographic Signature
