FHA Comments on Inpatient PPS Proposed Rule for FY2013

6/21/12

Page 2

Submitted Electronically

June 21, 2012

Marilyn Tavenner

Acting Administrator

Centers for Medicare & Medicaid Services

Hubert H. Humphrey Building

200 Independence Avenue, S.W., Room 445-G

Washington, D.C. 20201

RE: CMS-1588-P, Medicare Program; Hospital Inpatient Prospective Payment Systems for Acute Care Hospitals and the Long-Term Care Hospital Prospective Payment System and Fiscal Year 2013 Rates; Hospitals’ Resident Caps for Graduate Medical Education Payment Purposes; Quality Reporting Requirements for Specific Providers and for Ambulatory Surgical Centers; Proposed Rule (Vol. 77, No. 92), May 11, 2012

Dear Ms. Tavenner:

The Florida Hospital Association (FHA), representing more than 185 hospitals and health systems, welcomes the opportunity to provide comments to the Centers for Medicare & Medicaid Services (CMS) regarding the proposed rule pertaining to the inpatient prospective payment system (IPPS) for fiscal year (FY) 2013. These comments will focus on the proposed coding and documentation adjustment, inpatient quality reporting and value-based purchasing, the Hospital-Acquired Conditions program, the Hospital Readmissions Reduction program, outlier threshold, labor and delivery beds, Section 5503 residency programs, and timely filing requirements of claims for Medicare Advantage enrollees.

Documentation and Coding Adjustment

In federal FY2008, CMS adopted the Medicare-Severity Diagnosis Related Groups (MS-DRGs) to better recognize severity of illness in Medicare IPPS payment rates. In the final IPPS rule for FY2008, CMS implemented a prospective coding adjustment to the standard amount, phased-in over three years, to neutralize for expected case mix increases due to changes in documentation and coding. Subsequent legislation reduced CMS’ prospective coding adjustment to 0.6 percent in FY2008 and 0.9 percent in FY2009, but gave CMS authority to retroactively correct for differences between the offsets applied during those two years and actual coding improvement impacts.

Based on its analysis of case mix increases for claims paid in FYs 2008 and 2009, CMS estimated that the increase in case mix due to coding improvement in FY2008 was 2.5 percent and that there was an additional 2.9 percent coding-related increase in FY2009, yielding a total coding increase of 5.4 percent. CMS has recouped the purported overpayments for FYs 2008 and 2009 and has adjusted the federal amount prospectively.

In the proposed FY2013 IPPS rule, CMS is proposing an additional -0.8 percent prospective adjustment to correct for coding behavior in FY2010. CMS proposes to use the same methodology for this adjustment as was used for the FY2008 and FY2009 adjustments. The FHA is extremely troubled by this proposal and by the fact that CMS continues to compare hospitals’ documentation and coding practices to their documentation and coding practices under an entirely different system in FY2007.

The concepts and assumptions that underlie CMS’ calculation are flawed and the analysis does not fulfill the legislative mandate. Specifically, it does not differentiate between case mix increases due to changes in coding behavior and case mix changes that reflect real changes in patient characteristics and treatment patterns. In prior years’ rulemaking, hospitals and their associations have repeatedly contested the CMS methodology for determining the impact of coding behavior and CMS’ conclusions drawn from that analysis as being overstated and flawed. CMS still has not addressed those concerns nor has it provided details on its calculation of the FY2013 -0.8 percent adjustment.

We recommend that CMS reverse its proposed -0.8 percent coding adjustment for FY2013. Failure to identify real case mix change is a fatal flaw in CMS’ methodology. We recommend that CMS conduct a serious and meaningful analysis, using the methodology it has previously endorsed in similar settings, to determine changes in real case mix that have occurred since FY2006. Unless/until CMS can provide a meaningful analysis that identifies changes in real case mix, any calculation of coding improvement is misleading and inappropriate.

Value-Based Purchasing (VBP) and Inpatient Quality Reporting (IQR)

FY2015 VBP Program – PSI-90 Measure

The Affordable Care Act (ACA) mandates that measures selected for the VBP program be included on the Hospital Compare Web site for at least one year before the beginning of the measure’s VBP performance period. CMS is proposing to begin the performance period for the PSI-90 composite measure on October 15, 2012. To fulfill its statutory obligation, CMS must have posted hospitals’ PSI-90 measure rates to Hospital Compare by October 15, 2011. Although CMS did post PSI-90 measures to Hospital Compare prior to October 15, 2011, those measures reflect data for only four of the eight measures that make up the complete composite measure (PSI-6, 12, 14, 15). To date, neither the overall composite measure rate nor the remaining component measure rates (PSI-3, 7, 8, 13) have been posted to Hospital Compare.

In addition, CMS is proposing to remove four of the eight component measures that make up the PSI-90 composite measure from the IQR program beginning with FY2015 payment determinations. The measures proposed for removal (PSI-6, 12, 14, 15) are the only measures with data currently available on Hospital Compare. By law, only IQR program measures can be used in the VBP program.

CMS should re-evaluate if the PSI-90 measure is appropriate for inclusion in the FY2015 VBP program. PSI-90 does not meet the ACA mandate that requires VBP program measures be posted to the Hospital Compare Web site for at least one year before the beginning of the measure’s VBP performance period. Because PSI-90 is a composite measure made up of eight individual measures, it is essential that hospitals and independent organizations have access to all of the data necessary to appropriately evaluate and track performance on the proposed measure. In addition, CMS is proposing major revisions to the composite measure for FY2015 payment determinations under the IQR program. If CMS believe these revisions are necessary for the measure, it may not be appropriate to use the measure, as currently designed, to evaluate hospital performance under a pay-for-performance program.

FY2015 VBP Program – CLABSI Measure

The ACA mandates that measures selected for the VBP program be included on the Hospital Compare Web site for at least one year before the beginning of the measure’s VBP performance period. CMS is proposing to begin the performance period for the CLABSI measure on January 26, 2013. To fulfill its statutory obligation, CMS must have posted hospitals’ CLABSI measure rates to Hospital Compare by January 26, 2012. Although CMS did post CLABSI measure data to Hospital Compare prior to January 26, 2012, only three months of performance data (January 2011 – March 2011) was made available. Because the volume of data associated with the CLABSI measure is inherently small, and because only three months of hospital performance data was actually available on Hospital Compare, data for only one-third of all hospitals is currently available on Hospital Compare. Also, CMS is proposing to use the CLABSI measure for only one VBP program year (FY2015).

CMS should re-evaluate the appropriateness of the CLABSI measure for inclusion in the FY2015 VBP program – especially if CMS believes that this measure is not appropriate for continued use under the VBP program. To maintain some level of program stability, CMS should not adopt measures to be used for only one VBP program year.

In addition, because only three months of performance data has been posted to Hospital Compare, the CLABSI measure does not meet the spirit of the intent of the ACA mandate that requires VBP program measures be posted to the Hospital Compare Web site for at least one year before the beginning of the measure’s VBP performance period. This data is not representative of hospital performance on the measure. It is essential that hospitals and independent organizations have access to the amount of data comparable to the actual lengths of the baseline/performance period in order to appropriately evaluate and track performance on the proposed measure. Also, because the volume of data associated with the CLABSI measure is inherently small, hospitals could be disadvantaged by this measure if used under the VBP program.

FY2015 VBP Program – Use of HAC Measures under the VBP Program

The ACA mandates the implementation of a Hospital-Acquired Condition (HAC)-specific payment policy beginning in FY2015. The PSI-90 composite measure and CLABSI measure proposed for use in the FY2015 VBP program are HAC measures. CMS should consider the applicability of these measures to the HAC payment policy rather than the VBP program. If HACs are components of two different payment policies, hospitals could be penalized twice for the same performance.

Similar to the express exclusion of readmissions from VBP (because the ACA authorized a readmissions-specific payment policy), HAC measures should also be excluded from the VBP program. CMS should not adopt HAC measures for use under the VBP program.

FY2015 VBP Program – HCAHPS Additional Questions

We appreciate CMS’ efforts to improve the patient mix adjustment by including two additional questions to HCAHPS regarding emergency department (ED)-originated admissions and evaluation of mental and emotional health. Research indicates that various patient populations including those that have extended length of stay, are depressed or have a high severity of illness tend to give lower HCAHPS scores. Ensuring that an appropriate adjustment factor is in place to account for these patient population differences across hospitals is critical as hospitals are now being paid based on these results. That being said, we have some concerns as to how the questions may be implemented. We understand the data element previously used to determine ED-originated admissions is no longer being collected, hence the proposal for a patient-reported question. However, accuracy can be a major concern as the path for admission can be confusing and ultimately may lead to inaccurate responses. We believe there should be some other administrative-driven data element that can serve as a proxy for the ED-originated admission.

The second question CMS proposes to add would ask a beneficiary to rate his/her overall mental or emotional health on scale ranging from excellent to poor. CMS offers neither a rationale as to why it wants to add this question, nor a statement of what it will do with the data. In addition, CMS does not state whether it has pilot-tested this question with beneficiaries. We have several concerns about this question. First, we are concerned that patient characteristics, which are not adjusted for, may unduly influence how beneficiaries respond to this question. This could result in the potential unintended consequence of hospitals treating the most severely ill patients having systematically lower scores on this question, thereby unfairly disadvantaging these hospitals. In addition, we are concerned that beneficiaries may not feel comfortable answering this question and may perceive it as an invasion of privacy. Finally, we also are concerned that this type of self-diagnosing of psychiatric conditions may have potential unintended consequences for hospitals, such as increased liability risk. Because there are many unanswered questions and we are unsure of the state of pilot testing, we cannot support this proposal at this time.

FY2015 VBP Program – Hospital Inclusion/Exclusion Policies and Domain Reweighting

For FY2015 and subsequent years, CMS is proposing to require hospitals to have domain scores in two out of the four domains in order to be included in the program. For hospitals with only two or three domain scores, the VBP Total Performance Score (TPS) would be reweighted proportionately to the scored domains. For the FYs 2013 and 2014 VBP programs, hospitals must receive domain scores on all domains (two in FY2013 and three in FY2014) in order to be included in the program.

For FY2015 and future VBP program years, CMS should maintain its current policy that requires hospitals to receive domain scores in all domains in order to be included in the program. CMS’ proposed proportional reweighting approach would yield VBP TPSs that would not be comparable across hospitals and could severely disadvantage certain hospitals. Comparable score comparisons are essential under the VBP program because the IPPS payment redistributions under the program are reliant on hospital performance and the resulting VBP TPSs of all hospitals included in the program. Proportional reweighting for scored domains would lead to TPSs that are not comparable across hospitals and would skew program results and IPPS payment redistributions.

Instead of adopting the proposed reweighting scheme, CMS should move forward with establishing the ACA-mandated VBP demonstration project for small hospitals that are excluded from the VBP program based on the current exclusion policy. Through this demonstration, CMS can gain experience with how best to apply a VBP program to hospitals that do not meet the minimum case and measure counts for inclusion in the Hospital VBP program.

FY2015 VBP Program – Medicare Spending per Beneficiary (MSPB) Measure

The FHA recognizes that efficiency measures can be valuable and have a place in the VBP program. However, hospitals need more time to gain experience with the MSPB measure and learn about their potential for performance improvement. This measure was reported on Hospital Compare for the first time in April and is claims-based; therefore, hospitals do not have the same level of familiarity with it as they do other VBP measures. Further, CMS has only provided hospitals with data indicating whether the average spending for patients treated by their hospitals is above or below the national average; CMS has not provided hospitals with access to the raw data that would allow them to independently replicate and verify the agency’s calculation of this measure. Therefore, we urge CMS to delay its decision to finalize the MSPB measure until the FY2014 rulemaking cycle. Doing so would still allow CMS to include the measure in the VBP program for FY 2015.

In addition, we urge CMS to be more transparent with respect to the data underlying this measure. CMS provides very little data on Hospital Compare regarding this measure – it only indicates whether the average spending for patients treated by the hospital is above or below the national average. Because the measure includes a 30-day post-discharge period, it is likely that many other costs beyond inpatient care contribute to the overall metric. Yet, CMS makes this information available only to each hospital in a confidential report, meaning the hospital cannot learn by comparing its performance to other hospitals. We urge CMS to make these data publicly available.