HIT Policy Committee Recommendation One Small (CPOE) Step from Flexibility
What is the 3-1-1-1-0 Recommendation?
The HIT Policy Committee today created a framework for much needed flexibility within the EHR incentive program. Rather than requiring providers to meet all of the 20-plus meaningful use objectives (as is currently required in the CMS proposed rule), the Policy Committee recommended that providers be able to defer a certain portion of meaningful use objectives without jeopardizing their incentive payments. The rationale given for this was that an “all-or-nothing approach may not accommodate legitimate, unanticipated, local circumstances and constraints.”
The American Hospital Association has advocated for this type of flexibility, and I have supported the approach in a recent blog “CMS Proposed Rule: How to Stretch Without Breaking.”
The way 3-1-1-1-0 works is that providers would have the ability to defer a certain number of objectives from 4 of the 5 meaningful use domains: 3 objectives from the quality/efficiency domain, 1 objective from the patient engagement domain, 1 from the care coordination domain, 1 from the population health domain, and 0 from the privacy and security domain. However, a list of 7 objectives would remain mandatory. (See the list at the end of this blog)
When I first heard of this new approach, I was excited by the prospect of finally being able to blog about a positive development in an HIT incentive program that has the potential, if only it were properly structured, to improve care quality and efficiency in rural communities.
Unfortunately, among the mandatory objectives is one that effectively undermines the recommendation’s intent to “accommodate legitimate constraints.” By recommending that computerized provider order entry (CPOE) be mandatory rather than deferrable, the HIT Policy Committee has given with one hand and taken away with the other.
Why CPOE Should be Deferrable
Of all the meaningful use objectives, CPOE is the poster child for deferability. This is because (1) it is the most complicated to achieve[1], (2) it is a capstone application that requires that most other elements of the EHR be in place before it can be implemented, and (3) if implemented without the time required and attention paid to success factors, CPOE is the most likely implementation to produce numerous unintended consequences that will increase medication errors, reduce care quality, and undermine efficiency.
Let’s consider just one of the dozens of CPOE success factors identified as “imperative” in numerous case studies: the importance of implementing an EHR portal (along with the clinical systems that feed the portal) prior to any hospital CPOE implementation. The Advisory Board articulates this success factor as follows:
“CPOE entails a radical change to the day-to-day workflow of most physicians; hospitals are advised to introduce new functionality incrementally, gradually ramping up toward more complicated applications.
“Before deploying physician ordering functionality, hospital builds a user-friendly interface for physicians to retrieve all available patient information; goal is to entice physicians with convenient information access before asking them to actually input information.
“Once up and running, hospital ensures all physicians are trained and using the electronic record before deploying orders functionality; mandate enforced by requiring electronic signature and eliminating paper copies of the record.
“Building widespread portal usage before deploying physician ordering functionality accelerates the timetable for universal order entry, allows hospitals to tackle the initial physician adoption challenges—overcoming the fear of using the computer—before attempting to introduce order entry.”[2]
The takeaway from this is that all or most of a hospital’s clinical information needs to be in electronic form before CPOE can be effectively implemented. Lab, pharmacy, radiology, nurse documentation, EHR portal, e-signature and many other applications are required before even beginning the process of CPOE implementation. Mandating CPOE as part of Stage 1 meaningful use therefore ensures that most hospitals at early stages of EHR adoption will not be able to access any HIT incentives.
On the issue of unintended consequences, the Oregon Health and Science University identifies 9 categories of CPOE-related unintended consequences. The only one I will reference is their category 7: Generation of New Kinds of Errors.
“CPOE systems prevent some types of errors while creating or propagating new ones. New CPOE-related errors result from: problematic electronic data presentations; confusing order option presentations and selection methods; inappropriate text entries; misunderstandings related to test, training, and production versions of the system; and workflow process mismatches. Recognizing current unintended consequences should encourage system designers to optimize human computer interface design, and to exert caution when implementing new alerts.”[3]
These and other unintended consequences can certainly be overcome, but who among us doubts that rushing providers to CPOE without giving them time to implement the requisite base of applications, according to established success factors, will lead to failed implementations and reduced quality and efficiency on an unprecedented scale?
Recommendation
Let’s include early stage providers in the HIT incentive program by making CPOE a deferrable objective.
I would like to hear other perspectives on this important issue, so please post comments, especially if you disagree.
Figure 1: Mandatory Objectives (from HIT Policy Committee presentation: “Proposed Recommendations on MU Notice of Proposed Rule Making.)
[1] “CPOE is the most difficult technology implementation I can think of in the acute care setting” John Glaser, Senior Advisor to David Blumenthal. From Computerized Physician Order Entry: Securing Physician Acceptance, Advisory Board, 2004.
[2] Computerized Physician Order Entry: Securing Physician Acceptance, Advisory Board, 2004
[3] POET Recommendations: Types of Unintended Consequences of CPOE, April 20, 2007
{ 2 trackbacks }
{ 3 comments… read them below or add one }
I agree with your comments on the difficulty of implementing CPOE. As I understand the HIT incentive objective it is that 10% of orders must be placed electronically and that these orders can be placed by MD, DO, RN, PA and NP. Seems like this objective can be met by the orders placed by RNs alone. Then what about the physician orders given to a nurse to be entered electronically? If they count towards the 10% criteria then I believe the stage 1 CPOE objective can be met without requiring a hospital to implement full blown CPOE.
Greg,
You bring up important points. On the second issue (do physician orders entered by a nurse count?), I am interpreting “use of CPOE for orders directly entered by authorized providers” to mean that the provider signing off on (i.e. responsible for initiating) the order must be the one entering the order for it to count. If physician paper orders entered by nurses were to count, then meeting the CPOE 10% threshold would be reasonably attainable through the implementation of pharmacy, eMAR, and nurse documentation systems. But the word “directly” in the proposed rule seems to exclude that interpretation.
On the first issue (what percentage of orders are initiated/signed off on by nurses?), good question. Almost all medication, lab, and radiology orders require physician approval. However, certain consults (RT/OT/Respiratory), as well as nursing orders (turn every two hours, float heels, oral care), are initiated and signed off on by nurses. There’s no question that the vast majority of orders will be initiated by physicians, but orders initiated by nurses may approach or exceed the 10% barrier. I will research this at some of our network’s hospitals to try to get ballpark ratio. Anyone else with information on this please chime in.
Louis
This is follow-up on the above discussion regarding what % of orders are authorized by RNs rather than physicians. The three small rural facilities I’ve talked to about this issue consider the procedures initiated by nurses as “interventions” rather than “orders.” At least at these facilities, all “orders” need to be authorized/signed-off on by physicians.
I realize that those who created the 10% rule felt it was a compromise, but it really only helps providers that are positioned to start a CPOE implementation. Hospitals at early stages of EHR adoption have so much work to do in preparation for CPOE, 10% may as well be 100%.
Louis