Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

2018 coefficients #2

Open
benavege opened this issue Feb 1, 2018 · 4 comments
Open

2018 coefficients #2

benavege opened this issue Feb 1, 2018 · 4 comments

Comments

@benavege
Copy link

benavege commented Feb 1, 2018

Hello,

will the 2018 coefficients be posted here?

thank you!

@kuitang
Copy link

kuitang commented Mar 12, 2019

Hello,

I am also interested in 2018 (and 2019) coefficients. In general, is do you have a script to extract the coefficients and HCC mappings from a given CMS distribution? Thanks!

@KennyMonster
Copy link

I started looking into this on my own, and here's what I think I figured out so far:

I grabbed the 2020 SAS code from CMS, which in turn contains the model coefficients in a SAS transport format (whatever that is). I've been looking all over for a way to get this into CSV.

One solution appears to be to use SAS (of course). Someone did this on the SAS form: https://communities.sas.com/t5/SAS-in-Health-Care-Related/SAS-Transport-Format-Files/td-p/605588 . In turn they were kind enough to post the coefficients CSV. Of course the coefficient names don't match.

A possible DIY option is Pandas which I haven't tested yet: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.read_sas.html . I think the "XPORT" format it mentions is one of the transport format output options. Not positive here though.

Anyhow, apart from coefficient updates, I think the ICD10 HCC reference data also needs to be updated as well, but CMS appears to post a CSV of this on the same download page as the SAS code.

@KennyMonster
Copy link

As an update, turns out the newer models released by CMS were changed in a significant way. As such, just shoving in new coefficients wouldn't make any sense, without also updating the related Python Code.

@kindofluke
Copy link
Contributor

kindofluke commented Mar 13, 2020

Hello from the maintainers. @KennyMonster is exactly right starting in 2019 the model got much more complex. We are gauging community interest in updating this library for 2020. If you have interest please let us know at info @ algorexhealth.com

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants