Monday, March 15, 2010

CCHIT & the ONC IFR - Odds & Ends

As CCHIT published its comments to the ONC IFR, I took another look at the Interim Final Rules and I must admit CCHIT is making some very good points. Read the CCHIT comments here.

Here are a few other odds and ends that I believe will cause problems. I don't quite understand why definitions couldn't have been tightened down a bit more prior to publication. These are pretty elementary things.

Medications & CPOE - Tangled Up

For ambulatory providers Medications are addressed in four different criteria:
170.302(a) - Drug-drug, drug-allergy, drug-formulary checks.
170.302(c) - Maintain active medication list.
170.304(a) - Computerized provider order entry. Enable a user to electronically record, store, retrieve, and manage..... (1) Medications;(2) Laboratory;(3) Radiology/imaging; and(4) Provider referrals
170.304(b) - Electronically exchange prescription information

Theoretically, purchasing a CPOE module and one of the available ePrescribers, like Dr. First for example, should satisfy all four criteria. However, the CPOE module cannot obtain certification for 170.304(a) without managing Med lists, which the ePrescriber already does and must do. So either Dr. First starts managing Lab orders, or Lab order modules start managing prescriptions, or Dr. First and the Lab order vendor need to apply for certification together as one product.

Suggested solution: Remove Medications management from CPOE.

CPOE - No Standards

For ambulatory EHR modules CPOE is defined as follows
170.304(a) - Computerized provider order entry. Enable a user to electronically record, store, retrieve, and manage, at a minimum, the following order types:
(1) Medications;
(2) Laboratory;
(3) Radiology/imaging; and
(4) Provider referrals.

There is no specification of any standards to be used for such recording of orders. This will render CPOE modules, certifying under this premise, useless for electronic orders down the road. I guess buyers need to beware!

Suggested solution: Require that Lab and Radiology orders contain the HL7 required fields for each type of test. Medications, as mentioned above, should be dealt with separately.

Smoking Status - What is that all about?

170.302(f) - Smoking status. Enable a user to electronically record, modify, and retrieve the smoking status of a patient. Smoking status types must include: current smoker, former smoker, or never smoked.

Is anybody envisioning a standalone module that records and displays one item? What would a vendor of such module present for certification? A single screen with three checkboxes and an interface to send the data item somewhere, or a database with two columns (one for patient identifier and one for smoking status)?

Suggested solution: If the insurance industry is hell bent to have physicians record this cryptic and clinically lacking data point, it should be added to 170.302(e) - Record and chart vital signs. This is a temporary solution, and not a very good one, until Meaningful Use requires recording of other History items.

Regarding the EHR Module concept in general, the definition that a Module is anything that satisfies at least one criteria doesn't make sense to me. If we are going to certify Modules, and we should, they need to be defined in a more sensible manner. When I think of an EHR Module, I envision an ePrescriber, or a Registry, not a Smoking Status Recorder.


  1. Hi Margalit,
    Being on CCHIT as a co-chair, I know well the strengths and weaknesses. Every org has room for improvement, but few if any (including CCHIT) are as "evil" as their detractors make out. I appreciate your common sense comments re certification in this blog and other recent ones. Re "modules" while I see your point, it could be difficult to "predefine" those modules rigidly, as different EHRs carve up the pie differently, especially when you get into complex areas like med management. Admittedly "smoking status recorder" would be a very odd "module" though...

    BTW, I notice you're from St. Louis (which was my home for the 1st half of my life). Are you associated with Wash U. (which one of my alma maters)?
    David Tao

  2. Hello David, glad you dropped by... :-)

    Regarding CCHIT, I think ONC went from one extreme to the other. I did not like the first one and I do not like the current one. It is more than foolish to discard years of accumulated knowledge for political correctness reasons. I am also very concerned with NIST's ability to master what needs to be mastered in order to become the absolute authority in certifying clinical software (those pitiful test procedures are still on their website).
    Regarding modules, I'm not sure how to go about it, and I agree that they cannot be rigidly defined. Maybe instead of defining a module as something that can satisfy at least one MU criterion, we define a module as something that can satisfy a Meaningful combination of criteria? I know it's fuzzy and I'm not certain how to proceed here...
    My concern is that between the proliferation of single modules and the proliferation of single module certifying bodies, we will create a royal mess for providers, mainly small practices.

    I have this very uncomfortable feeling that we are over-thinking over-regulating too soon (see my post today).

    I'm not affiliated with Wash U, although I live about 5 minutes away. I took some classes there years ago in a futile attempt to finish my PhD. Very nice school. You should visit your home town once in a while... :-)