NPPEST

[fred trotter]

Just wanted to see if I could post here.

Also wanted to end the crickets...

FWIW, I have started an external project to test the quality of the NPPES data sets... 

https://github.com/ftrotter/NPPEST

It has some fun code, like the code required to test to see if an NPI is valid...

-FT

[Mark Silverberg]

Fred - thanks for the link to that resource! As an open data geek and someone who has tried to work with NPPES for an open data project (for a state health department), I would love to see fresher and better quality data. I dont know if the issues I came across were due to provider entry or manual data entry by a clerk but I have come across a number of issues, mainly around geocoding practice and business location addresses: ranging from zip codes that have a number transcribed to state codes that were incorrect.

It might be interesting to run simulations on (subsets) of the NPPES data to see how prevalent these issues are.. if there is some feedback loop that can be followed. If not, all this QA from external stakeholders might be futile.

Happy to help in anyway I can.

- @Skram

[Alan Viars]

Fred:

Cool.  Do you have a URL where the service is running?

Alan Viars

On Sunday, January 26, 2014 10:33:20 PM UTC-5, fred trotter wrote:

[Alan Viars]

Hello Mark:

Thanks for your insight.

Fixing "staleness" of data is a big problem.  And yes there are some erroneous address records in the NPESS database.  From the public file, I have created a CSV containing original addresses plus verification/correction information for domestic addresses.  Hopefully in another couple of weeks I'll have some server space and I'll make these files accessible.   

One enhancement to try and make keeping NPPES data more up to date is make the experience more like "about.me" so everyone gets a customized profile page.  Setup this way, providers may be more apt to want to update the information.  In my personal opinion, allowing PECOS data to inform/override NPPES address information would also improve data quality.

Just in case you or anyone reading this is unfamiliar, PECOS is the system of Medicare enrollment for providers.  Not all providers participate in Medicare.  In PECOS providers must "revalidate" every 5 years so the data is less stale.

.-Alan Viars

[Mark Silverberg]

Alan - Thank you for responding. All your action items and suggestions (including your personal opinion) are agreeable to me. Please let us know, through this group, how we (the public) can help, if at all, in addition to providing feedback and ideas. 

Mark

[Janos G. Hajagos]

Alan,

As a researcher my interest is in accuracy and historical completeness. Like many on this list I have considerable experience hacking the NPPES. 

I should be able to get a history of a provider. If I am analyzing a doctor teaming graph from 2012 I would like to know where the provider was in 2012. If I am now analyzing the 2013 data I would like to know where their address was in 2013. 

The other sugar that would be nice if the NPPES also contained geocoded addresses. 

A separate idea would be if the states were able to voluntarily submit the provider license information with an identifier that CMS has, e.g, SSN or NPI. For the provider whose license match the database would contain the date the license was validated. The likelihood of this happening is close to 0 but the potential cost saving and utility having an integrated database with accurate license data would be huge. 

Janos 

Not sent from a desktop computer

--
You received this message because you are subscribed to the Google Groups "NPPES Modernization Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nppes+un...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Alan Viars]

Hello Janis:

Although not an official CMS release, I have applied address correction and geo-coding routines to the November 2013 release of the public NPPES data.  Its just an expanded version of the public data set.  I will post the URL to that here in a week or so.  (I'm waiting on HHS sanctioned server space.)

Many addresses are "stale" as you know and we are working on a multi-pronged plan to try and address this.

Best,

Alan Viars