Coding in EMRs – we are doing it wrong
I was in an pan-Canadian EMR Benefits Evaluation meeting today hosted by Canada Health Infoway. A little heated debate came up around the value of coding data. Some fell on the side of don’t bother coding, use free text and fix it later”. I fall on the other side of trying to do it right the first time. Clinicians can code and I think they should, so they can get some of the advanced feedback and functions of EMRs
I am a supporter of free text -d on’t get me wrong. Narrative is very important to our current recording practice. However, selective coding does have value in places like medications, allergies, problem lists (labs should have attached codes from the lab automatically).
However, if it takes the physician 30 seconds to properly code an item on the problem or if a user needs to memorize SNOMED CT codes like 191736004 then we are designing our EMRs wrong.
Rather than saying don’t bother coding, I suggest we should be asking:
How can we make the User Experience so elegant that coding is seamless?
Does coding = semantic tagging?
Rasool
10 Mar 12 at 9:13 am
Hi
Yes. More specifically coding in this context includes some standardized reference sets from a terminology like SNOMED CT. So when coding a diagnosis you ensure that you document that diagnosis so that it is tagged in the background with a common nomenclature that can be acted upon (e.g. with clinical decision support) and compared within and between practices.
priceless
10 Mar 12 at 1:22 pm