CMS published the CY 2021 Home Health Payment System Rate final rule in the Federal Register on November 4th. As expected, the agency held the Behavioral Adjustments for next year to the same level as in the first year of PDGM. Where CMS had originally projected an expected behavioral change of -8.39% for the first year of PDGM, they cut the rate in half since they expected that it may take agencies more time before fully implementing behavioral changes.
Although the second installment of the other half was possible, NAHC and other industry advocates recommended that CMS eliminate the behavioral adjustment altogether. William A. Dombi, president of the National Association for Home Care & Hospice (NAHC), expressed disappointment on that point in an email to the organization’s members. “We believe sufficient information is available to recognize that the behavioral changes assumed have not occurred. We will continue to work to get a mid-year rate change to reflect the absence of the rate behavior that CMS assumed.”
CMS notes in the rule regarding behavioral adjustments that “…we believed it would be premature to propose any changes to the CY 2021 payment rate based on the data available at the time of CY 2021 rulemaking and in light of the ongoing COVID-19 PHE.” As it stands, CMS kept all of the PDGM methodology, LUPA thresholds and case-mix weights the same in CY 2021 as they were in the first year of PDGM.
In my July 2020 blog, I looked at the several areas of expected PDGM behavioral changes. With another quarter of data under our belts and with the better understanding and management of COVID-19, I compared the PDGM period starts for the first three quarters of CY 2020 using the SHP National database reflecting just under five million Medicare 30-day periods.
In regards to Comorbidity adjustments, the third quarter appears to have leveled off to the run rate we are observing year-to-date. The 14.8% “High” level is a 6.7 percentage point increase over what CMS had calculated in their base year PDGM model. This confirms one of the behavioral adjustments CMS had contemplated.
Functional Impairment levels is another PDGM component with a large variance compared to the CMS’ original design for PDGM. Where CMS pegged the “High” impairment level at 31.5%, the run rate YTD is 12.3 percentage points higher and is not materially different across the first three quarters.
Although CMS did not specifically point it out, it is the behavioral change that is increasing overall case-mix weights the most. In regards to changes in diagnosis coding related to Clinical Grouping, there is not much evidence that this has occurred. While there is a slight skew toward higher case-mix weight groups, some of this may be due to the fact there has been fewer elective surgeries due to COVID-19 in CY 2020.
Contrary to the CMS expected behavior changes in avoiding LUPAs, the LUPA rates continue to be three percentage points higher than what CMS was predicting in the first year. Rates have leveled off in the third quarter after the peak observed in March due to COVID-19.
|LUPA Rate||January||February||March||April||May||June||July||August||September||YTD 2020|
CMS has the authority to make higher or lower adjustments to the 30-day payment rate necessary to offset behavior changes and to maintain budget neutrality. Keep in mind they would distinguish budget neutrality to ensure that the aggregate expenditures during CY 2020 under the new case-mix methodology and 30-day unit of payment would be the same as if the 153-group model was still in place in CY 2020.
Many agencies saw a reduction in their admission volume March through June of this year due to COVID-19. In the final rule CMS wrote “future payment adjustments…must be based on the difference in aggregate payments between the assumed versus actual behavior change and not because of utilization changes resulting from the COVID-19 PHE.” If and when CMS does make any future adjustments, it will be important they provide for data transparency, much like they did with the PDGM Limited Data Set, to help substantiate the basis of their calculations.