Senate Finance Committee Chairman Orrin Hatch (R-Utah) urged the Centers for Medicare and Medicaid Services (CMS) Administrator Seema Verma not to finalize the recently developed Home Health Groupings Model (HHGM) in its Calendar Year (CY) 2018 Medicare Home Health Prospective Payment System proposed rule until the agency can fully validate the impact these reforms will have on Medicare spending.
“While I support the agency’s long-term objective of more precisely aligning Medicare payments with patient care needs…I am concerned that CMS may be rushing to finalize complex policy changes too quickly,” Hatch wrote in a letter. “CMS’ efforts to implement policies similar to the Medicare Payment Advisory Commission (MedPAC) recommendations as well as those of the Senate Committee on Finance are a good first step…It is equally as important, however, that these proposed changes are methodologically sound and implemented in a reasonable timeframe.”
The full letter may be found below:
Dear Administrator Verma:
I am pleased that the Centers for Medicare and Medicaid Services (CMS) has embraced the fundamental policy goals outlined in a September 2011 Senate Committee on Finance staff report[1] recommending that CMS examine options to refine the Medicare home health payment system. The committee report prioritized developing a future reimbursement structure that moves away from volume based service utilization toward a new system that prioritizes tying payment to patient health characteristics. It is clear that data-driven changes are needed to improve not only the accuracy of payments to home health agencies (HHAs), but also the quality of skilled care services delivered to Medicare beneficiaries.
While I support the agency’s long-term objective of more precisely aligning Medicare payments with patient care needs, upon reviewing the Calendar Year (CY) 2018 Home Health Prospective Payment System (HH PPS) proposed rule, I am concerned that CMS may be rushing to finalize complex policy changes too quickly. The proposed rule contemplates a brand new Home Health Groupings Model (HHGM) that would use 30-day periods, rather than 60-day episodes – combined with case-mix refinements – to place Medicare patients into more precise payment categories. Given that the CMS proposed changes differ in certain aspects from reform options recommended by the Medicare Payment Advisory Commission (MedPAC), extra time and more robust data analysis is needed to assure the agency’s proposed methodology and implementation strategy achieve our shared goal of improving home health payment accuracy.
I respectfully request CMS to refrain from finalizing the HHGM model in its CY 2018 proposed rule. Additional time is needed not only to provide stakeholders the opportunity to analyze fully the proposed changes and provide substantive input to CMS through an open dialogue, but also to ensure that any payment reforms are implemented in the least disruptive manner possible.
Interestingly, the rule indicates that the overall impact of the proposed HH PPS case-mix adjustment methodology refinements amounts to an estimated negative $950 million (negative 4.3 percent) in payments to HHAs in CY 2019 if implemented in a non-budget neutral manner. That figure drops to an estimated negative $480 million (negative 2.2 percent) in payments to HHAs in CY 2019 if the refinements are implemented in a partially budget-neutral manner.[2]
When CMS proposes regulatory payment reforms of this magnitude, typically the changes are implemented in a fully budget neutral manner – unless specifically directed otherwise by Congress. Respectfully, it is my view that Congress retains the prerogative whether or not to change current Medicare payment law, outlined in the Social Security Act, in a way that generates mandatory savings and returns that money to the United States Treasury. As a result, I am extremely concerned that the agency may be exceeding its regulatory authority in this proposed rule. As the Chairman of the Senate Committee on Finance, I believe the Administration must stand ready to work in a bipartisan fashion with me and the other members of the House and Senate committees of jurisdiction, to design and implement any legislative proposals that data suggest will encourage high-quality, patient centered care while deterring, or where appropriate eliminating, policies that incentivize misuse or abuse of the Medicare program.
I also find it troubling that CMS has not provided complete data and supporting information describing how the proposed rule’s overall revenue impact figures were calculated. First, the proposed negative 4.3 percent payment adjustment described in Table 55 of the rule states that the CMS analysis included “assumptions on [HHA] behavioral responses as a result of the new case-mix adjustment methodology”. Unfortunately the proposed rule does not clarify (1) what HHA behavioral assumptions CMS made when modeling the estimated HHGM financial impacts, and (2) how changes in assumed HHA behavior will directly, or indirectly, influence Medicare spending. It is critical that CMS disclose sufficient data in order for Congress to understand the Administration’s policy intent as well as for stakeholders to develop independent impact evaluations. Without reasonable disclosure of data – such as (1) a state-by-state financial impact analysis and (2) a disaggregation of the proposed rule’s impact based on patient characteristics – the opportunity for robust stakeholder participation and input is diminished. I have heard from multiple stakeholders concerned that they will be unable to fully research, analyze and evaluate the impact of the proposed rule within the 60-day comment period.
In the past, CMS has made honest mistakes when modeling how health care providers might react to certain proposed Medicare payment system changes. In light of these errors, it is imperative that Congress know more about how CMS believes HHAs will respond to its HHGM refinement proposal. For example, Congress enacted a new Medicare home health payment system in the Balanced Budget Act (BBA) of 1997.[3] The payment reforms were implemented in two stages. An interim payment system was phased in starting October 1997 while CMS developed the new prospective payment system. This policy was expected to save $16.2 billion from 1998-2002, but it actually reduced Medicare home health spending by $74.4 billion from 1998-2002.[4]
More recently, CMS implemented a number of changes to the Skilled Nursing Facility (SNF) payment system effective in Fiscal Year (FY) 2011. Specifically CMS modified how SNFs bill for concurrent therapy. The agency expected its changes would decrease billing for higher levels of therapy services. Although CMS intended the FY 2011 revisions to be budget neutral, Medicare payments actually increased by $2.1 billion (16 percent) from the last half of FY 2010 to the first half of FY 2011. Unanticipated billing patterns contributed to a dramatic spending increase which the Department of Health and Human Services, Office of the Inspector General calculated to exceed $4 billion in additional Medicare payments to SNFs in FY 2011.[5] Subsequently CMS took corrective action in FY 2012 to stop the overpayment, but the recalibration was finalized in a prospective, not retrospective, manner – effectively producing a multi-billion dollar windfall to SNFs in FY 2011.[6]
Additionally, in 2007, CMS unveiled the FY 2008 Inpatient Prospective Payment System (IPPS) proposed rule.[7] At the time, CMS was preparing hospitals to transition from the previous diagnosis-related group (DRG) classification system to the current Medicare Severity DRG (MS-DRG) classification system. In this proposed rule, CMS announced plans to apply a 2.4 percent behavioral offset due to anticipated hospital documentation and coding practices that did not reflect real changes in patient acuity, thus driving reimbursements higher.
Because CMS was forthcoming and provided more specific impact amounts that quantified this proposed change, Congress ultimately decided to step in and act legislatively – implementing a slightly different policy to recoup the remaining overpayments in the Transitional Medical Assistance, Abstinence Education, and QI Programs Extension Act (TMA) of 2007.[8] The law implemented a negative 0.6 percent adjustment in FY 2008 and a negative 0.9 percent adjustment in 2009 to inpatient payments. Congress also specified that CMS, to the extent that these adjustments either over-compensated or under-stated the actual documentation and coding changes, should make further prospective and/or retrospective cuts.
Ultimately, behavioral assumptions can trigger payment swings that produce sizeable reimbursement reductions or windfalls in the Medicare payment system. Because errors sometimes do occur in the impact modeling phase, it is vital that CMS conduct a more comprehensive impact analysis prior to the agency finalizing the HHGM proposal as part of the CY 2018 HH PPS proposed rule. For example, did CMS calculate what the HHGM payment adjustments would have totaled if the proposal was implemented in a budget neutral manner? What would the fiscal impacts have been? Has CMS considered how the proposed rule may create unintended HHA behavioral responses not previously contemplated, such as providers capitalizing on the higher profit-margin low-utilization payment adjustment (LUPA) to receive enhanced reimbursement?
Finally, but most important, is my concern about what the proposed HHGM means for my home state of Utah. Although state-level average impacts of the proposed HHGM described in the CY 2018 HH PPS proposed rule are not publicly available, some estimates suggest Utah may see anywhere from a negative 12 percent payment reduction to a negative 14 or 15 percent payment reduction. The proposed rule’s regulatory impact analysis states its rationale: “HHAs that provide a larger percentage of overall visits as therapy visits compared to skilled nursing visits may experience larger decreased in payments under the HHGM.”[9] Without a publicly available state-by-state distributional analysis, however, it is difficult for members of Congress to ascertain the root cause of these estimated financial shifts.
Remarkably, according to the Program for Evaluating Payment Patterns Electronic Report (PEPPER) data from the fourth quarter of CY 2016, Utah maintains a current average number of 8.3 therapy visits per home health episode. That figure is about 1.2 visits higher as compared to the national average of 7 visits per home health episode. Yet the latest Home Health Compare data shows Utah as the state with the second to lowest re-hospitalization rate in the nation at 14.60 percent, second only to Hawaii at 13.90 percent. Utah HHAs consistently perform above average compared to the national re-hospitalization rate of 16.7 percent. It would appear that the HHGM changes are designed to provide increased payments for medically complex patients with higher comorbidities. If the goal is to help sicker, more complex patients get out of the hospital and back into their homes not just more quickly, but also safely, can CMS provide additional details about the impact the proposed HHGM design will have on hospital readmission rates? How is CMS factoring this important data point into its methodology?
Changes to the HH PPS such as moving from the current Home Health Resource Group (HHRG) case-mix classification system to a consolidated HHGM classification system are significant. Given the list of concerns outlined above, I again would urge CMS not to finalize the HHGM refinement model at this time, but instead continue its dialogue with the members of the Senate Committee on Finance and industry stakeholders – many of whom agree with the principle of placing medically complex cases into more meaningful payment categories in order to more accurately reflect the cost to provide skilled care services. CMS’ efforts to implement policies similar to MedPAC’s recommendations as well as those of the Senate Committee on Finance are a good first step toward achieving our common goal of a fair and accurate home health payment system. It is equally as important, however, that these proposed changes are methodologically sound and implemented in a reasonable timeframe.
Thank you for your consideration of these comments.
