Remember the famous line from Bert in Mary Poppins. “Winds in the east. Mist coming in. Like something is brewing, about to begin.”
Somethings brewing for sure and it is coming from Baltimore Md. I’m talking about the abrupt changes CMS is making to the SNF 5-star ratings.
In part 1 of this blog series, we talked about the changes that will affect the health inspection portion of the rating in July. In Part 2 we will break apart the changes to quality measure domain that are slated to go into effect in October.
The What
The 5- star rating is made up of of three domains, health inspection, staffing and quality measures. There are 10 MDS-based quality measures that impact a facility’s rating. One of those measures is the long stay antipsychotic measure.
This is the measure at which CMS’ summer storm is taking aim. While the changes will not go into affect until the October care compare update, it is worth considering the potential effect they will have on the star ratings.
The Why
Currently, the long stay measure, Percent of Residents Who Received an Antipsychotic Medication, reports the percentage of long-stay residents who are receiving antipsychotic drugs in the target period. This data comes directly from the MDS item N0415A1, and is reflective of antipsychotic medications that were recorded as received by a resident during the 7-day lookback period associated with that item.
It is a simple measure and has not changed significantly since it’s inception. In QSO-25-20-NH, CMS has indicated that, “…the use of the minimum data set (MDS) for reporting the number of long-stay residents receiving antipsychotic medications may not accurately reflect the number of residents who are prescribed antipsychotic medications.”
They are correct. That’s not the reason this measure was created. Prescribed and received are two completely different things. A prescription noted on a Part D claim does not mean that the patient actually received, let alone took, the medication.
The How
To address the perceived discord between what the MDS records and what claims are indicating, CMS is updating the measure by including Medicare and Medicaid claims data and Medicare Advantage encounter data to supplement MDS data. According to CMS, Claims and Medicare Advantage encounter data will capture antipsychotic medication use that may be underreported on the MDS.
We have as yet to receive an updated QM manual and technical specifications to know specifically what the new measure will look like. We also need an updated 5-Star user’s guide to know what the cut points will look like for the national decile distributions will look like. What we do have is the Technical Expert Panel (TEP) for Refinement of the Nursing Home (NH) Antipsychotic Medication Measures dated Sept. 2023.
In this TEP report there are two draft options for what this new measure might look like. Both draft options maintain the current measure and add an OR criteria for when a Part D claim overlaps the lookback period in section N. In other words, if the the current measure criteria are met, OR, if a Part D claim indicated an antipsychotic had been prescribed in the same lookback period, the measure would trigger.
Both draft options also add an exclusion for a resident not being continuously enrolled in Medicare Part D for one year before the target date to the month of the target assessment. The only difference between the two options is that draft option 2 allows a denominator exclusion for Schizophrenia, Huntington’s or Tourette’s recorded on the Part D claim.
Not much to see there. Of note, these draft specifications do not mention Medicaid claims data and Medicare Advantage encounter data as the QSO does and do not address the question of whether the resident actually took the medication. Again, a prescription noted on a Part D claim does not mean that the patient actually received, let alone took, the medication.
Also of note, in the QSO, CMS indicates that MDS-reported exclusion diagnoses are validated with claims and encounter data, reducing the number of excluded residents due to the overreporting of schizophrenia diagnoses on the MDS and that these updates leverage data to improve the measure’s accuracy. That seems to be reflected in the TEP draft measures as well.
What Now
There is much to consider in the lead up to this change becoming effective this October. Let’s keep our heads about us and think soberly about what these changes could mean. Absent an official revised technical specification document the best we can do is speculate.
However, in the QSO CMS has indicated that the national percentage of residents receiving an antipsychotic is 14.64% under the existing measure. Under the new measure, this will increase to 16.98% due to increased accuracy. In other words, On October 29, 2025, CMS will incorporate the new measure on Nursing Home Care Compare.
In the meantime, here are some questions I would like to ask CMS, if they ever start the open Door Forum back up again.
- Since a prescription noted on a Part D, or Medicaid or Managed Care claim claim does not indicate that the patient actually received, let alone took, the medication, what will this measure actually measure? What if the Part D claim was initiated in another setting? Why are we being measured on what claims data says was prescribed not the actual administration of the medication?
- How is the claims data being verified. For example, how does CMS know if the hospital assigned diagnosis of schizophrenia meets the clinical guidelines. What if the MA plan fails to add a diagnosis of Tourette’s.
- What if an antipsychotic was prescribed in the hospital but the the nursing home discontinued it and the Part D claim data overlaps the ARD lookback of an MDS completed in the nursing home? The draft measures specifications appear to trigger the measure if either the antipsychotic was coded at N0415A OR if a claim overlaps the ARD lookback of the MDS.
The wind is about to start blowing. 5-Star ratings will change overnight in July and in October and for nothing that the nursing home did or didn’t do differently. 5-Star changes will occur simply because CMS decided to change the way the calculation works.
When the 5-Star user’s guide and Quality Measure manual is updated, perhaps we will have a better understanding. Until then, get ready with as much information as you can. If your rating changes overnight, hospitals, families, managed care contracts all may anticipate that you have some explaining to do. It will be good to know what you are explaining.
