June 30, 2008

Kerry N. Weems
Acting Administrator
Centers for Medicare & Medicaid Services
Hubert H. Humphrey Building
200 Independence Avenue, SW, Room 445-G
Washington, DC 20201

RE: (CMS-1534-P); Medicare Program; Prospective Payment System and Consolidated Billing for Skilled Nursing Facilities for FY 2009; (Vol. 73, No. 89), May 7, 2008.

Dear Mr. Weems:

On behalf of The Hospital & Healthsystem Association of Pennsylvania (HAP), which represents approximately 250 member institutions, including 125 stand-alone hospitals and another 120 hospitals that comprise 32 health systems across the state, we appreciate this opportunity to comment on the Centers for Medicare & Medicaid Services’ (CMS) proposed changes to the skilled nursing facility (SNF) prospective payment system (PPS) for fiscal year (FY) 2009. Our comments focus on our concerns about the proposed methodology for adjusting for case-mix changes related to the proposed re-weighting. We have also responded to CMS’ request for input on its initiatives related to value-based purchasing and the current demonstration on hospital discharges to post-acute care.

PROPOSED CASE-MIX ADJUSTMENT

On January 1, 2006, CMS increased the number of SNF resource utilization groups (RUGs) from 44 to 53 to better account for the higher cost of treating medically complex patients who also need rehabilitation. Using claims data from 2001, CMS projected that 19 percent of SNF cases would fall into the nine new RUGs. However, recent analysis of 2006 payments found that the new RUGs accounted for more than 30 percent of total cases. CMS’ intention when introducing the new RUGs was to implement a budget-neutral change to the SNF PPS, relative to what Medicare would have paid under the former 44-RUG model. In the proposed rule, CMS recommends a one-time, across-the-board reduction of RUGs payments to account for higher than anticipated utilization of the new RUGs. CMS proposes a recalibration of the RUG weights that produces an 8.22 percent reduction, which amounts to a $770 million cut in Medicare payments to SNFs for 2009.

While HAP supported the addition of nine new RUGs for the sickest and costliest SNF patients, who are treated most frequently in hospital-based SNFs, we are very concerned that CMS proposes to make such a significant reduction to offset a greater-than-anticipated level of utilization of the new RUG categories. We believe that both the size and timing of this adjustment is inappropriate. Higher utilization of the new RUG categories reflects real changes in the SNF patient population between 2001 and 2006, not just a forecasting error. Additionally, other policy influences such as the inpatient rehabilitation 75% Rule and heightened medical necessity review of inpatient rehabilitation providers, contributed to an increase in the clinical acuity of SNF patients who fell within the nine new RUGs. Therefore, we recommend that CMS limit the size of this adjustment to only account for coding behavior and spread out any proposed cut over a minimum of a two-year period to minimize instability for SNF providers, especially hospital-based SNFs that already struggle with extremely negative Medicare margins.

Benefits of Prospective Payment for Medicare Providers

There are many benefits of paying Medicare providers based on a prospective payment basis, which is effective and efficient, including:

• The case-mix measure is a well-defined and well-understood metric of provider output and allows for reliable, predictable payment.
• Case-mix systems are fair because they adjust for patient acuity, as well as many other factors.
• Medicare prices are set in advance, and thereby reward efficient providers.
• Providers are not at risk for changes in case-mix that result from underlying changes in patient acuity, referred to as “real case-mix.”

Historically, CMS has made case-mix refinements to more accurately pay for expensive cases, which facilitates access for acutely ill patients because providers are more appropriately paid for these cases. To refine and recalibrate case-mix systems, CMS uses the most recently available data to adjust the case weights of the new system so that payments are budget neutral for a defined set of cases under both the new and the old case-mix systems. Prior to this proposed rule, it was expected that case-mix during the first payment year under a refined system would increase for the following reasons:

• Observed Case-Mix. When case-mix measurement systems are refined, such as the transition from 44 to 53 RUGs in FY 2006, providers must change their coding practices to ensure that the diagnoses that are important to the new case-mix system are accurately captured. This can result in a different-than-expected case-mix because the old data used to develop the new weights and make utilization projections were incomplete.
• Real Case-Mix. Some portion of case-mix increase is real, since it is based on changes in the clinical acuity of patients treated. Real case-mix changes have always been paid for under Medicare to the extent that CMS could measure them.

Departure from Traditional Prospective Payment Approach

In this proposed rule, CMS acts in opposition to several of the noted beneficial principles of prospective payment and, as a result, could undermine provider trust in the accuracy and fairness of prospective payment. CMS would break precedent with 25 years of prospective payment by having skilled nursing providers bear the cost of real case-mix changes based on the false premise that the full change in case-mix measurement from the 44-RUG system to the 53-RUG system should be budget neutral in its first year of implementation.

To develop this proposal, CMS used 2001 data to compare the distribution of RUG days under the 44-RUG and 53-RUG groupers. CMS’ most recent analysis uses 2006 SNF data to adjust the 53-RUG weights to make SNF payments budget neutral across the two systems. In discussing its so-called “forecast error” of $770 million that would adjust for complete difference in case weights under both groupers, CMS fails to measure the portions of case-mix change that are real and observed. It is extremely important for CMS to conduct this analysis and report its findings to the SNF community. Based on such findings, CMS should lower the proposed reduction from an adjustment for the entire case-mix change that occurred in the transition from the 44-RUG system to the 53-RUG system to a reduction for only that portion of the case-mix change that is due to observed case-mix.

To place this transition in context, it is useful to note that, in the past, both the Medicare Payment Advisory Commission (MedPAC) and CMS’ rulemaking on other prospective payment systems distinguished between real and observed case-mix change.

VALUE-BASED PURCHASING

CMS requested input from the SNF providers on how the inpatient value-based purchasing system would work for the SNF PPS and other payment systems. We urge CMS to first study the operation and impact of this policy as it applies to the inpatient PPS before determining whether to proceed with a similar plan for any other payment systems. We encourage CMS to work closely with the affected providers up front to build support, reduce confusion, and coordinate any such effort with providers’ independent quality improvement initiatives. To meaningfully impact quality of care at the patient level, it is critical that selected measures represent a consensus among affected stakeholders and that they first undergo validation for accuracy and reliability. An advisory panel of clinicians and researchers would provide the agency with guidance on which SNF conditions are appropriate for inclusion under this policy. In addition, a two-phase rollout of a value-based payment system similar to the process used for the inpatient PPS would be appropriate for the SNF PPS, with pay-for-reporting first, followed by pay-for-outcomes.

HOSPITAL DISCHARGE TO POST-ACUTE CARE DEMONSTRATION

In the proposed rule, CMS requests input from providers on its current demonstration involving a standardized assessment of Medicare beneficiaries being discharged from inpatient hospitals to post-acute settings. Specifically, CMS is seeking input on how this demonstration on a common patient assessment instrument (the Continuity Assessment Record and Evaluation (CARE) tool) “might advance the use of health information technology in automating the process for collecting and submitting quality data.” While we feel there is long-term potential for the common use of a patient assessment instrument, such as the CARE tool, to facilitate the efficient flow of electronic, secure patient information, it is too early to draw any conclusions. The CARE demonstration is in the beginning stages and electronic information has yet to be transmitted between participating organizations. At the current time, there is only one test site in operation. Additionally, it is unclear whether the demonstration will be able to test web-based, inter-organization transfer of patient information. Therefore, because it is too early to know how successfully hospitals can implement the CARE tool, it is impossible to predict whether the tool would be effective for collecting and submitting quality data.

HAP appreciates the opportunity to submit our comments on the proposed rule. If you have any questions about our comments, please contact Melissa Dehoff, Director, Health Care Continuum finance policy, at (717) 561-5318, or by email at mdehoff@haponline.org.

Sincerely,

PAULA A. BUSSARD
Senior Vice President, Policy & Regulatory Services