Two words: Stage 2 (Part 2 of 3)

Today's post is a continuation from last day, a discussion of the highlights of Meaningful Use Stage 2 by Dr. Farzad Mostashari, ONC Director. Please see Part 1, published Monday, September 3, 2012. We are considering several key concepts in Meaningful Use Stage 2, not new concepts, rather concepts that have come into the limelight in Stage 2. 

Part 2 of 3.

Common datasets, structured and coded data.
If you've done much reading up on Stage 2, you'll have seen these terms used frequently. What are common datasets? Consider that, over the course of your life as a physician, your every patient will see potentially hundreds of different doctors in a variety of settings: family physicians, dentists, eye care providers, hospital and emergency room physicians, and specialists of many descriptions. Since these many doctors will use a host of different software solutions, how do we have any hope of achieving the meaningful exchange of health information? The answer is datasets, data structured to share a common language. If you call it one thing and I call it another, it's the common dataset that tells us we mean the same thing.  ICD-10, SNOMED and LOINC codes are good examples of how information from disparate systems can be communicated effectively, even if the human languages are foreign to each other. 

Another example: as providers exchange CCDs or SCRs, the documents are validated against a NIST standard. It's this standard that allows one EHR to consume the data sent from another, creating a new patient file or adding to an existing one. This consumption of data marks a fundamental difference - that "giant leap" - between receiving then viewing a fax (the good ol' fashioned way) and seeing that new data entered automatically into your electronic health record, no keystrokes, no mouse-clicks (the new speed-you-up way). 

What about structured data? Structured is most easily understand in contrast against free- text. In a free-form field, you would type your notes. It's completely up to you what you type. If you use voice entry, same result. The problem with free text - or unstructured data - is that it is not searchable. New smart technologies are introducing the ability to structure free text but this is far from an available norm today. Structured data is basically organized in such a way that analytics tools can be used to make sense of the data, see trends, determine best outcomes and best practices, for example.

Datasets, structured and coded data remind us that EHRs are not simply about recording results in a neat, legible record, rather about sharing and analyzing data in order to improve healthcare. Core values in health reform: higher quality, lower cost, better patient outcomes.

To be continued ...

Alistair Jackson, M.Ed.

To read Dr. Mostashari's full article, see Meaningful Use Stage 2: A Giant Leap in Data Exchange.