In this episode of Ask Dr. Mingle, Dr. Dan Mingle answers a great question from Gail about deduplication of data from multiple EHRs for an accurate eCQM submission. Later, Dr. Mingle answers questions about MIPS measures #5, #8, and #130.

Use the links below to jump to a specific question and answer, or click play on the video to listen now.

• Deduplicating Data for eCQM Reporting from Multiple EHRs
• MIPS Measures #5 and #8
• MIPS Measure #130

Question One: eCQM Reporting from Multiple EHRs

Gail asks: “My team has been discussing the challenges of eCQM reporting from multiple EHRs, and special concerns about deduplication. I think we understand the basics of matching and deduplication of patients. My question is the next step. Once you know that patient John Doe is on the two files and you have to get rid of one, how do you merge the data to capture the correct outcome on the eCQMs that you are reporting?”

Great question, Gail.

It’s more straight-forward than it appears.

It’s really simply emulating at the organization level with multiple medical record systems what you’d see at the practice level with one medical record.

It’s a common occurrence that patients have multiple visits, multiple episodes, multiple measurements, multiple opportunities to provide the indicated intervention.

At Mingle Health, we run into it if we are serving:

• One provider with his/her panel of patients in one EHR
• A group practice with one EHR
• An ACO or a community healthcare system with multiple EHRs

The starting point for any quality measurement project is to get a complete accounting of the patients with the data that generally indicates measure eligibility.

This, for us, is usually a complete set of insurance claims for the time period.

The next hurdle is to get all of the relevant clinical data that indicates performance. Each piece of clinical data needs to include adequate patient identifiers to match it back to the eligible patients and a date stamp indicating when it was done.

If we have data as described, we don’t care in which practice or by what provider’s actions it happened.

We expect a large set of redundant data.

Once we have all of the relevant data and we’ve normalized it into a single set of data with each piece tagged to the patient it pertains to and the date it was done, we can then ignore where it came from.

In all cases, the measure specifications tell us which particular data element determines performance.

• Sometimes, only the most recent data is factored into the analytics.
• Sometimes, the best measurement in the set gets factored in.
• Sometimes, if the action occurred anywhere, at least once in the performance period, performance is met.

When you report as a large group, like an ACO, every provider/practice gets performance credit when any provider/practice meets performance.

It means we need to have all of the data about all of the potentially involved patients from all of the practices. Holes in the data mean holes in the performance and/or counts against data completeness requirement.

We never individually “choose” the data we include. We never discard or disregard data.

When there are multiple data points, each, potentially, meriting a different performance score, the measure specifications always tell us which one takes precedence.

Question Two: MIPS Measures #5 and #8

Julie asks: “Quick question on MIPS measures #5 and #8. Can we use code 3022F to indicate when a patient’s LVEF is greater than 40% and the patient does not meet the denominator qualifications for these measures? Or would we leave these unanswered?”

For those two measures, we want some way to exclude patients with only mild left ventricular dysfunction. The ICD-10 diagnoses for Left Ventricular Systolic Dysfunction (“LVSD” or “heart failure”) are not adequately specific. You need only report on those with moderate or severe LVSD and you must exclude those with only mild LVSD.

There are a number of different ways that can be accomplished.

3022F is an effective way to exclude mild left ventricular dysfunction, or M1150, or G8923 is an effective way to include moderate to severe left ventricular dysfunction.

We can also use a numeric entry, if you have one, for LVEF (left ventricular ejection fraction) from your EHR (we’ll include those LVEF < 40 and/or exclude those with LVEF >= 40).

Or, if you routinely grade the severity of heart failure in your problem list, we can use that. (We’ll include those with moderate to severe disease and/or exclude those with mild disease.)

Our goal is to capitalize on the strengths that already exist in your records. Or, if what you have is not adequate, help you implement a more effective solution with the least disruption possible.

Question Three: MIPS Measure #130

Rick asks: “I have question about MIPS measure #130: ‘Documentation of Current Medications in the Medical Record.’ The measure specifications say ‘Eligible clinician attests to documenting in the medical record they obtained, updated, or reviewed the patient’s current medications.’ Or ‘Documentation of a medical reason for not…’

I’m trying to determine if the documentation has listed the medications, and if the clinician signs the note, if that is enough of an attestation. Or would we need a separate attestation statement? And, if our EHR has a “hard stop” to reconcile medications, does this hard stop action count as attestation?”

Great question!

The answer is in the specifications that you quoted: “Eligible clinician attests…”

Appropriate use of this measure demands an attestation that indicates “Performance” as “Met”.

There should always be a medication list in the electronic chart, unless the patient is on no medications.

That list can be right or wrong. You test its accuracy or inaccuracy.

Unfortunately, doing the work to test its accuracy creates no reliable tags in the EHR database that show or prove that the list was tested or verified:

• If you are ePrescribing, a prescription indicates an action but does not confirm validation of the list.
• Adding or subtracting a medication from the list shows attention to that medication, but not necessarily in the context of a complete validation-review.
• Signing a note that contains the medication list does not demand validation or review.

And my bias is that there should be only one list in the chart.

If there is a list in a note at every visit, it introduces duplicate work and potential error to the extent that the medications might have been wrong at the time of the visit, or the list has changed or become outdated since the visit.

Ideally, the note does not present a new list but refers to the EHR-documented list.

If you are going to say in your written note: “See medication list”, you’d be well served to expand that declaration to say: “See the medication list that the patient confirmed is complete and accurate.”

Even better, you’d have a few options to reflect:

• “Caregiver verified the list”
• “List validated against the pill bottles patient brought to the visit”
• “Pharmacy was called to confirm the list”

I’d be very uncomfortable with a “hard stop” in the note generated by the EHR on medication verification.

I think a specific attestation is called for. The corollary to this being that there has to be a way to indicate that verification was not performed. There will be times that you don’t or can’t verify the meds. And if you can’t complete the note without claiming verification, it forces the clinician to lie.

Send us your value-based care questions!

If you’d like to ask a question about the APP transition, MIPS, ACO quality reporting, or any other Alternative Payment Model, you can reach out to us in three ways:

You can leave your questions in a YouTube comment under any episode of Ask Dr. Mingle.

On LinkedIn, leave your questions in a comment on any of our posts.

And you can reach out directly by sending an email to hello@minglehealth.com.