Clinical Decision Support is one of the criteria needed to demonstrate Meaningful Use and access bonus money from the CMS EHR Incentive Program. Yet, the term is fairly vague, and means different things to different people.
The actual criterion, as defined in the Eligible Professional (EP) Attestation Worksheet (criterion #11), defines the Objective as “Implement one clinical decision support rule relevant to specialty or high clinical priority along with the ability to track compliance with that rule.” The actual Measure is simple: “Implement one clinical decision support rule relevant to specialty or high clinical priority along with the ability to tract compliance to that rule” – the measure only requires a yes/no answer.
Digging a little deeper at the definition for Certification, defined in the NIST certification document (170.304.e), a Certified product must: (1) provide one or more clinical decision support rules (in addition to drug-drug and drug-allergy contraindication checking) that utilize data elements in a patient’s problem list, medication list, demographics and laboratory test results; and (2) generate and indicate automatically, electronically and in real-time notifications and care suggestions based upon clinical decision support rules.
Where do clinical decision support rules come from?
Simple clinical decision support (CDS) rules, as indicated, include drug-drug interaction checking and drug-allergy warnings – but Meaningful Use wants something beyond that for this criterion.
Our view sees that Clinical Decision Support rules are closely tied to Clinical Quality Measures (CQM), which we have described previously. A CQM feature in an EHR would generate a report of all patients needing a particular intervention – for example, patients due for mammography, or diabetic patients due for glycohemoglobin (HbA1c) blood test checking. The CQM feature of an EHR, then, is focused on populations, and creates lists that can be exported and used for outreach. These lists are not about patients being seen today, and are run periodically “in the background.” They rely on medical office personnel doing something with the results – looking through the lists, seeing which patients are already scheduled to be seen, and reaching out to patients who may have “fallen through the cracks” but are in need of some pro-active intervention. Effective population management is based on CQM lists.
One way of thinking of a CDS rule is as an individual-patient prompt of a CQM item. If there is a list, say, of diabetics who need certain lab testing done, then for each patient on that list, there needs to be a real-time prompt – e.g. a dashboard, or a flowsheet – indicating that these tests need to be done when the patient shows up for an appointment. If a diabetic patient, for example, comes to the office complaining of seasonal allergies, on that patient’s dashboard there would be a display noting that the patient is diabetic, and is due (or past due) for particular interventions – this would be an example of CQM-driven Clinical Decision Support.
More advanced CDS
Rules and prompts for evidence-based quality-of-care are plentiful. The problem, at present, is that there are often multiple, competing recommendations from different specialty societies – should mammography be done before the age of 50? Should PSAs be used to screen for prostate cancer? Different societies have slightly different recommendations, and a clear “standard of care” is sometimes fuzzy on some of these items.
For now, CQM-associated CDS rules make the most sense – CQM criteria, after all, are things that have gone through a long, multi-stakeholder process (under the umbrella of the National Quality Forum) and have emerged as a minimum set of evidence-based care-quality measurements.
CDS vs. References
As defined in the Meaningful Use criteria, CDS is meant to be a real-time prompt that occurs at the point of care. That means, for the clinician, when sitting face-to-face with a patient, that some sort of prompt or flowsheet is available with reminders of what things might be missing and are due.
Reference materials, on the other hand, are also important, but are not real-time. Quickly reviewing medical literature (e.g. UpToDate, or the Cochrane Library) can help a clinician figure out a puzzle, or provide current options on treating a particular condition – but this is done “offline” and out of the immediate seeing-the-patients-here-now workflow. A way of flagging or bookmarking a patient’s chart note, however, might be useful – that way, a clinician can tag a particular chart, and then later (after-hours, or when there is a lull) go back to review the case and look up reference material in the literature.
Clinical Decision Support is a somewhat vague term that means different things to different people. In the context of Meaningful Use, it is something beyond drug-drug and drug-allergy interaction checking. Our implementation of CDS ties the decision support rules to Clinical Quality Measures – CQM generates lists, and is about populations; CDS is the individual-patient prompting about the results of these lists (done at the point of care). CDS rules are meant to be real-time, in the workflow of seeing patients now – this is different that tagging an encounter for offline review of reference material and medical literature (though that is important too).
This is the conceptual framework with which we are building our EHR product, as we create a fully-Certified web-based EHR that will allow clinicians to access Meaningful Use money later this year.