In Pursuit of Quality Analytics

Dignity Health’s data repository defines big—every night it grows by some 500 GB. Encompassing 39 hospitals across four western states, Dignity (the former Catholic Healthcare West) is using that data warehouse to create an analytics infrastructure to drive quality improvement efforts among an initial group of 600 physicians, themselves scattered across disparate locations.

Get access to this article and thousands more...

All Health Data Management articles are archived after 7 days. REGISTER NOW for unlimited access to all recently archived articles, as well as thousands of searchable stories. Registered Members also gain access to exclusive industry white paper downloads, web seminars, podcasts, e-books, and conference discounts. Qualified members may also choose to receive our free monthly magazine and any of our e-newsletters covering the latest breaking news, opinions from industry leaders, developing trends and specialized topics like EHR's, revenue cycle management, health insurance exchanges, analytics, and more!

Already Registered?

Forgot Password/Need Help?
Comments (2)
It's hard to get terribly excited about a database that grows by 500Gb every day, but the analyst can't readily and reliably retrieve (as per the article) total colonoscopies because they're documented as "a procedure, part of a history, as a problem, as an order, or it just might come back on a claim."

(I'd also think you'd want to exclude H/O colonoscopy, as well as "a problem" if no procedure was performed, or "an order" of there's no confirmatory procedure note/report...)

As the saying goes, it's not the size of the [database] that matters. It's the data quality that determines its usefulness.

I've worked with (not for) CHW / Dignity and they're great folks. It'd be fun to help take their database to the next level of utility.
Posted by Mark L | Friday, April 26 2013 at 2:12PM ET
Natural language processing (NLP) like MediSapien allows extraction of medical concepts defined by SNOMED trapped in different expressions of unstructured documentation from diverse EHRs, transcription or other legacy text. Congestive heart failure may have 40 different expressions that can be assimilated as one SNOMED code. The text and codes, SNOMED, RxNorm, LOINC, and ICD-9 and 10, can be stored in a clinical data repository and feed many analytic applications.
Posted by | Saturday, April 27 2013 at 7:02AM ET
Add Your Comments:
Not Registered?
You must be registered to post a comment. Click here to register.
Already registered? Log in here
Please note you must now log in with your email address and password.

Already a subscriber? Log in here
Please note you must now log in with your email address and password.