Center Comments on Skilled Nursing Facilities Payments and Oversight, 2023

Centers for Medicare & Medicaid Services
Department of Health and Human Services

June 5, 2023

Re: Medicare and Medicaid Programs; Prospective Payment System and Consolidated Billing for Skilled Nursing Facilities; Updates to the Quality Reporting Program and Value-Based Purchasing Program for Federal Fiscal Year 2024, CMS – 1779-P

Submitted electronically, https://www.regulations.gov

Dear CMS Colleagues:

The Center for Medicare Advocacy (Center) submits comments on proposed rules on the annual update to Medicare Part A payments for skilled nursing facilities (SNFs), 88 Fed. Reg. 21316 (Apr. 10, 2023).

The Center for Medicare Advocacy (Center) is a national, private, non-profit law organization, founded in 1986, that provides education, analysis, advocacy, and legal assistance to help people nationwide, primarily older people and people with disabilities, to obtain necessary health care, therapy, and Medicare. The Center focuses on the needs of Medicare beneficiaries, people with chronic conditions, and those in need of long-term care and provides training regarding Medicare and health care rights throughout the country. We advocate on behalf of beneficiaries in administrative and legislative forums, and serve as legal counsel in litigation of importance to Medicare beneficiaries and others seeking health coverage. These comments are based on our experiences talking with and representing Medicare beneficiaries and their families and advocates.

Overview of Primary Concerns

The Center is particularly concerned about and strongly opposed to (1) the Centers for Medicare & Medicaid Services’s (CMS’s) proposal to use the CoreQ Short-Stay Discharge Measure, developed by the American Health Care Association, in CMS’s Quality Reporting System; and (2) CMS’s proposal to establish a constructive waiver process that will give SNFs an automatic 35% reduction in their civil money penalties even when they do not send a written statement waiving their right to appeal, as required by the regulations for nearly 30 years.

In addition, the Center is concerned about what is missing in the proposed rules: any discussion of, or efforts to address, the dramatic decline in therapy services since the Patient-Driven Payment Model (PDPM) was implemented. The Center is particularly concerned about CMS’s failure to ensure that maintenance coverage of nursing and therapy services in SNFs, as mandated in 2013 by the federal district court in Vermont in the nationwide class action Jimmo v. Sebelius, is fully realized.

Detailed comments about primary concerns

CMS proposes to adopt the CoreQ Short Stay Discharge Measure (NQF #2614) beginning with the FY 2026 SNF QRP. 88 Fed. Reg., 21344. The Center strongly opposes adoption of this measure, which we opposed in response to last year’s NPRM.

The CoreQ Short Stay Discharge Measure was developed by the American Health Care Association, the major nursing home trade association. The four questions are too vague to provide useful information; the entire process excludes many individuals who may be likely to have negative feelings about their nursing home experience; the grading scale includes three “positive” choices, one negative choice, and no neutral middle choice, despite the use of a five-point scale, skewing the results towards a positive conclusion; and the process for selecting discharged residents who receive the questionnaire allows for significant provider gaming.

There are four questions in the short stay discharge measure:

In recommending this facility to your friends and family, how would you rate it overall?
Overall, how would you rate the staff?
How would you rate the care you receive?
How would you rate how well your discharge needs were met?

These four questions are problematic. Question 3 about “the care” is too general and vague. Care in a nursing home encompasses multiple disciplines and components – nursing, therapy, dietary, social work, activities, and more. Residents may have positive experiences in some aspects of their stay, negative experiences in others. Producing a single score is meaningless in this area. Evaluating a nursing home is not like evaluating a cell phone, where a composite score is much easier to determine and the stakes are so much lower.

The questions also depend on information that former residents may not have. How can former residents rate their “discharge needs” (Question 4) if they do not know what the requirements for discharge planning actually are and what discharge services their facility should have provided to them? What would be the basis of an opinion, in the absence of meaningful baseline information?

Under CoreQ, the short-stay discharge questions are sent to former residents who were discharged home or to assisted living. The CoreQ process excludes former residents who may be likely to have been dissatisfied with the care they were receiving. http://coreq.org/ For example, it excludes residents who go to another skilled nursing facility, psychiatric facility, inpatient rehabilitation facility, or long- term care hospital. Why? Those former residents may have chosen to leave the facility for another post-acute facility providing what they perceive as better care. The process also excludes residents who left against medical advice or who were on hospice, again reflecting former residents who may have been dissatisfied with their care at the facility or who believed they received poor care. The process also excludes residents with dementia, eliminating a large category of people, at least some of whom could give meaningful opinions about their nursing home stay. These discriminatory exclusions likely skew the results to former residents who were temporarily in the facility for rehabilitation, went home, and were satisfied on that basis (return to home) alone.

The fact that many facilities have “voluntarily adopted” CoreQ and use it “with ease,” 88 Fed. Reg., 21345, suggests that the tool is useful to facilities. Facilities have historically used satisfaction surveys for marketing purposes. However, these factors – voluntary adoption and ease of use – do not suggest that the tool is equally useful or meaningful to government regulators, who should be looking for an objective measure of resident satisfaction.

The federal survey process already requires surveyor interviews with residents, resident councils, and families. This existing information could be expanded (if needed) and translated into a satisfaction score, which would be far more objective than CoreQ.

CMS’s description of the measure’s testing – reliability testing and validity testing, 88 Fed. Reg., 21346 – does not support CMS’s conclusion that the measure provides meaningful information. The limited number of questions in the tool, and their vagueness, simply make it more likely that the results will be repeated.

CMS describes comments of interested parties and the Technical Expert Panel (TEP), some of whom were critical of CoreQ and whose concerns it does address. I was a member of a TEP that reviewed CoreQ and I was, and remain, extremely critical of the tool.

Another concern is that the CoreQ Users Manual defines 1 to 5 as poor, average, good, very good, excellent, making three of five responses reflect positive responses, one response (“average:) a neutral response, and only one response (“poor”) a negative response.

http://www.coreq.org/CoreQ%20Satisfaction%20Questionnaire%20and%20User%20Manual.pdf. By replacing a neutral middle, as is typical and appropriate in a five-point scale, with a positive “good” score, the grading system inappropriately skews the results towards positive.

Imputation of scores is permitted if the discharged resident answers three of the four questions; CoreQ adds the answers to the three questions and divides by three to impute a score for the fourth question and then calculates the overall score. Because three is a positive response, not a neutral, middle answer, and there is only a single negative response (“poor”), this methodology overstates positive responses.

To calculate the facility’s score, CoreQ adds the scores of residents that are three or higher and then divides by the total number of questionnaires that are considered valid. Again, because three is a positive response, not a neutral, middle answer, and there is only a single negative response (“poor”), this methodology also overstates positive responses.

Although CMS requires facilities to use a CMS-approved vendor, CMS anticipates that SNFs would incur “an increase of 17.5 hours of staff time to assemble and submit the resident information files” to the vendor. Id. 2410. This facility-based activity means that the facility determines which specific discharged residents should be included in the list of former residents that the facility gives to the vendor and which specific discharged residents are within one of the many excluded categories. This facility-based process creates a loophole and an easy way for a facility to game the system, excluding discharged residents whom it believes will have negative comments to share in a survey. Facilities’ ability to exclude discharged residents who might be more likely to give negative feedback is easy and obvious and built into the process. CMS-approved vendors cannot make up for the obvious bias in facilities’ determining which specific discharged residents receive the questionnaire; vendors’ role is limited to administering a flawed survey and essentially ministerial.

CoreQ’s User Manual considers 20 valid responses sufficient for a quality measure calculation, regardless of how many residents were discharged. Facilities’ ability to handpick which discharged residents receive the survey compounds our concern about the number of responses used in the measure and makes this sample too small to be valid.

Independent reviewers have not endorsed CoreQ. To the contrary, in its April 2022 report on nursing homes, the National Academy of Sciences, Engineering, and Medicine Committee wrote, recognizing that facilities use satisfaction surveys in their promotional and advertising activities:

In parallel with federal and state efforts, the nursing home industry has developed and implemented its own measures of resident and family satisfaction. For example, CoreQ, endorsed by the American Health Care Association, has three versions: long-term care residents, long-term care family, and short-stay discharged patients. (Castle et al., 2020; CoreQ, 2019; Schwartz, 2021). Each version consists of three or four general questions that focus less on rating the quality of resident experience and more on summative satisfaction ratings. Another example of an industry-developed tool is NRC Health’s My Inner View Customer Satisfaction Survey (NRC Health, 2021). Many nursing homes promote and advertise high scores from self-designed and administered surveys of their residents. However, consumer advocates and survey methodologists have raised concerns that item wording and the choice of response formats may increase the tendency of respondents to provide socially appropriate response choices and thus provide only minimal variation in the scale (Bowling, 2005; Dillman et al., 2014; Nadash et al., 2019).[1]

The Committee explicitly did not endorse CoreQ, but, instead, recommended adding the federally-developed Consumer Assessment of Healthcare Providers and Systems (CAHPS) measures of resident and family experience (i.e., the nursing home CAHPS surveys) to Care Compare. Recommendation 6A, p. 511.

Civil Money Penalties: Waiver of Hearing, Automatic Reduction of Penalty Amount, 88 Fed. Reg., 21399-21400

Repeating the identical proposal made by the Trump Administration in 2019,[2] which was never issued as final regulations, CMS proposes to create a “constructive waiver” process for nursing facilities that fail to state in writing that they are not appealing deficiencies and corresponding civil money penalties (CMPs) and to give these facilities, nevertheless, an automatic 35% reduction in the CMP. CMS is proposing to revise a policy that has been in place since the final 1994 enforcement regulations were published.

CMS’s rationale for re-proposing this process is the same as the rationale offered by the Trump Administration and is not compelling.

CMS reports that in calendar years 2016-2011, between 2% and 6% of facilities requested a hearing in writing and most facilities (more than 80% in each year) submitted written statements indicating that they were waiving hearings. Only between 7% of facilities (in 2021) and 17% of facilities (in 2022) did not submit written waivers. CMS does not explain why the comparatively small number of facilities that fail to submit written waivers should nevertheless get an automatic 35% reduction in their CMP. CMS notes that the federal rules authorize settlement of a case at any time, 42 C.F.R. §488.30, implying that CMS could settle CMPs for 35% off (or more) at any time. This settlement authority provides no support for the automatic 35% reduction in a CMP for a facility that fails to submit a written waiver of a hearing.

CMS disingenuously contends that the constructive waiver process will save relatively small amounts of money. First it reports savings in administrative costs – $861,676 in administrative costs for facilities (4,308 facilities times the $200 cost for sending the letter waiving the right to appeal) and $772,044 for CMS Locations. 88 Fed. Reg., 21417-21418. CMS then calculates the savings to facilities that will result from the automatic 35% reduction in the CMPs. It reports that, in calendar year 2022, CMS imposed a combined total of $190,967,833 in per day and per instance CMPs and that the median CMP is $4,545, 88 Fed. Reg. 21417. It calculates that the 35% reduction in the total CMP will save facilities $1,438,038 (35% of $4,545 for 904 facilities). The total savings to facilities is estimated to be $2,299,716 ($861,678 in administrative costs and $1,438,038 for the 35% reduction in CMPs).

These estimates may be too low. As of May 2, 2023, QCOR reports that in calendar year 2022, CMS imposed $177,555,874.75 in per day CMPs and $37,160,847.62 in per instance CMPs – a total of $214,716,722.37 (a higher total than the $190,967,833 that CMS reports in the NPRM), and that in calendar year 2022, CMS imposed $160,994,773.89 in per day CMPs and $46,865,189.69 in per instance CMPs – a total of $207,809,963.58 – not the total of $190,967,833 that CMS reports in the NPRM.

The Center notes that an appellate panel of the Departmental Appeals Board recently upheld the dismissal of a facility’s appeal when the facility failed to waive the hearing requirement in a timely way. New Grove Manor, Docket No. A-22-80, Decision No. 3090 (Mar. 2, 2023), https://www.hhs.gov/about/agencies/dab/decisions/board-decisions/2023/board-dab-3090/index.html. In that case, the per instance CMP was $16,845. A 35% reduction would have been a savings of $5,895.75, on top of the $200 cost of sending the letter to CMS waiving the appeal, totaling savings to the facility of $6,095.75.

CMS already imposes comparatively few CMPs because, as a matter of policy, it generally limits CMPs to deficiencies that are cited as actual harm or immediate jeopardy – classifications applied to less than 4% of all deficiencies.[3] Most deficiencies have no financial consequence, no matter how serious the harm to residents.[4]

CMS provides no real rationale for the proposed rule, which creates a financial windfall of millions of dollars for facilities. We strongly oppose the proposed change.

The diminution of therapy services in SNFs

In the final rule establishing PDPM, 83 Fed. Reg. 39162 (Aug. 8, 2018), https://www.govinfo.gov/content/pkg/FR-2018-08-08/pdf/2018-16570.pdf, CMS described the overpayments for therapy services under the prior reimbursement system, Resource Utilization Groups, 83 Fed. Reg. 39183-39185. CMS recounted commenters’ concerns, with which it agreed, that PDPM could result in facilities’ reducing therapy services that residents actually need, in light of the dramatically changed financial incentives in PDPM. Id. 39186. CMS added items to the discharge assessment to allow CMS to monitor the amount and intensity of therapy received under PDPM.

In the final rule published in 2018, CMS indicated that it would give facilities a “non-fatal warning edit” if they exceeded the permissible uses of group and concurrent therapy. Id. 39239. CMS also indicated that it would continue to monitor group and concurrent therapy utilization “and consider making future proposals to address abuses of this proposed policy or flag providers for additional review should an individual provider consistently be found to exceed the proposed threshold after implementation of the proposed PDPM.” Id. 39239.

In the final rule published August 7, 2019, just as PDPM was to go into effect, CMS revised the definition of group therapy to permit groups of two to six residents. It reported:

Anecdotally, we have been told by an industry group that they would advise their facilities to give as much group and concurrent therapy as possible based on the limit we set for group and concurrent therapy, so that if the limit were 50 percent, they would advise their facilities to give 50 percent group and concurrent therapy. This group informed us that they plan to advise their facilities to furnish 25 percent of all therapy as group and concurrent therapy.

84 Fed. Reg. 38728, 38749 (Aug. 7, 2019), https://www.govinfo.gov/content/pkg/FR-2019-08-07/pdf/2019-16485.pdf. Despite this blatant industry statement that it would encourage facilities to exploit whatever therapy rules CMS permitted, CMS essentially ignored the statement and reiterated its longstanding policy that most therapy should be individual therapy:

We note that we do not believe it would be appropriate to automatically provide the maximum amount of group and concurrent therapy permitted under the percent cap set by Medicare without considering the individual clinical needs of each patient. As we stated previously, we expect therapists to determine the frequency, duration, and modality of therapy based on sound clinical reasoning and the individual needs of each patient. Further, as we stated above and in the FY 2020 SNF PPS proposed rule (84 FR 17635), we continue to believe that individual therapy is the preferred mode of therapy provision and should be considered the standard of care in therapy services provided to SNF residents.

Id. 38749-38750. CMS also promised: “We plan to implement a robust monitoring program to assess compliance with the 25 percent cap, and based on our findings, we may propose taking additional action in future rulemaking.” 84 Fed. Reg. 38728, 38748 (Aug. 7, 2019).

As PDPM went into effect in October 2019, the media immediately reported that thousands of therapists had been laid off.[5]

In the proposed annual update to Medicare Part A payment rates, published April 15, 2021, CMS presented “some of the results of our PDPM data monitoring efforts.” 86 Fed. Reg. 19954, 19985 (Apr. 15, 2021), https://www.govinfo.gov/content/pkg/FR-2021-04-15/pdf/2021-07556.pdf. CMS reported a dramatic 30% decline in therapy services received by residents at SNFs, from 91 minutes per resident per day to 62 minutes per resident per day, and the substantial shift from individual therapy to group and concurrent therapy. Id. 19986. CMS noted that PDPM had resulted in significantly higher payments to SNFs in the first year of its implementation, $1.7 billion, or 5%. CMS’s response was to consider recalibrating the Medicare rates, which the Center supported. We noted in our 2021 comments that CMS had immediately and fully recalibrated rates for Fiscal Year 2012 when the transition from RUG-III to RUG-IV resulted in $4.47 billion in overpayments, 12.5%, in Fiscal Year 2011. In 2021, however, CMS chose in the final rules not to recalibrate the Medicare rates, in light of COVID-19 and the public health emergency. CMS did not propose any efforts to ensure that residents received the therapy they were denied in the transition to PDPM.

The Center repeats the recommendations that we made in June 2021 in response to the alarming decline in therapy services:

CMS should analyze the resident Discharge Assessment data since implementation of PDPM and should publicly report its findings.
CMS should add a mandatory financial penalty for facilities that exceed the 25% cap on group or concurrent therapy, with the penalty set at an amount to exceed the cost of compliance with the limitations on group or concurrent therapy.
CMS should identify nursing facilities that dramatically changed the therapy services they provide following implementation of PDPM and direct state survey agencies to conduct surveys at those facilities in order to identify whether they violated the Requirements for Participation, including resident assessment and care planning, professional standards of quality, and provision of care and services. If survey agencies identify noncompliance, CMS should cite appropriate deficiencies and impose enforcement actions, specifically, per day civil money penalties that exceed the cost of compliance.
CMS should consider reinstating a requirement for multiple resident assessments (as in the prior reimbursement system, Resource Utilization Groups) to prevent the gaming (and overstatement) that occurs with the single assessment on the fifth day of a resident’s stay, as now required by PDPM.

In 2022, CMS proposed to recalibrate Medicare rates by 4.6% (less than the percentage identified by CMS), $1.7 billion, with a delayed and phased implementation of the recalibration. 87 Fed. Reg. 22720, 22737-22743 (Apr. 15, 2022), https://www.govinfo.gov/content/pkg/FR-2022-04-15/pdf/2022-07906.pdf.

We reiterate our 2021 recommendations about therapy and the need for CMS to take strong action against facilities that denied residents appropriate therapy services. We urge CMS to take all appropriate actions to ensure that residents actually receive all the therapy services they need, whether for improvement or maintenance.

We also reiterate our 2018 concerns about PDPM’s ignoring the Court-Ordered Settlement in Jimmo, which confirms coverage of maintenance therapy in SNFs. Although PDPM undermines all therapy services for SNF residents, we expressed concern that PDPM would be particularly devastating for maintenance therapy.

On January 24, 2013, Chief Judge Christina Reiss of the federal District Court in Vermont approved a nationwide Settlement, negotiated by a class of Medicare beneficiaries, who challenged Medicare’s “improvement standard,” and the Federal Government. The Settlement confirms that the Medicare law and regulations provide coverage of maintenance therapy services in SNFs (and also home health and outpatient therapy) if such therapy is needed “to maintain the patient’s current condition or to prevent or slow further deterioration . . . so long as the beneficiary requires skilled care for the safe and effective performance of the program.”^[6] On February 16, 2017, the Court issued an additional order and approved a Corrective Statement that CMS was required to reproduce in full on the dedicated webpage that the Court ordered CMS to establish. The Corrective Statement says, in part, “The Jimmo Settlement may reflect a change in practice for those providers, adjudicators, and contractors who may have erroneously believed that the Medicare program covers nursing and therapy services under these benefits only when a beneficiary is expected to improve.”^[7]

To the extent that SNFs are required to change their practices to provide maintenance therapy to residents for whom it is medically necessary, the Medicare reimbursement system must reflect these therapy services. At the third Technical Expert Panel that advised CMS and reviewed Acumen’s research on the new Medicare reimbursement system, which ultimately became PDPM, Acumen staff indicated that they would incorporate Jimmo only if they had hard data documenting which facilities properly implement the Settlement and how much such implementation costs. Since that task was impossible to meet, Acumen’s position was that the requirements of Jimmo would not be included in the proposed system, but could be added at a later time, if and when the reimbursement system is further revised. Accordingly, the original NPRM for PDPM reflected Acumen’s failure to incorporate Jimmo.

That problem continues to this day. Jimmo has now been the law of the land for more than a decade. CMS must ensure that Jimmo’s mandate for maintenance coverage of nursing and therapy services in SNFs is fully implemented in the Medicare reimbursement system. To date, it has not done so.

Additional Comments

Hospital inpatient wage data, 88 Fed. Reg., 21324

CMS proposes to continue using hospital inpatient wage data for SNFs. This policy automatically and artificially inflates SNFs’ reimbursement rates because it is well-known, widely recognized, and publicly reported that nursing facilities typically pay lower salaries to their nursing staff than hospitals pay – 10-20% lower, according to a 2022 article in Health Affairs.^[8]

Although the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000 (BIPA), Pub. L. 106-554, authorizes CMS to establish a SNF wage index, CMS describes the task as “unfeasible:”

[A]uditing all SNF cost reports, similar to the process used to audit inpatient hospital cost reports for purposes of the IPPS [inpatient prospective payment system] wage index, would place a burden on providers in terms of recordkeeping and completion of the cost report worksheet. . . . While we continue to believe that the development of such an audit process could improve SNF cost reports in such a manner as to permit us to establish a SNF-specific wage index, we do not believe this undertaking is feasible at this time.

88 Fed. Reg., 21324. CMS’s concerns are not persuasive. Requiring facilities to complete cost reporting worksheets should not be viewed as burdensome; it is essential to understanding how facilities spend their reimbursement. Moreover, facilities are already required to complete costs reports. The missing action is meaningful auditing of these cost reports.

As the National Consumer Voice for Quality Long Term Care reported in its 2023 report Where Do the Billions of Dollars Go? A Look at Nursing Home Related Party Transactions,[9] there is little evidence that CMS seriously audits Medicare cost reports, when so many cost reports have missing information and contain glaring errors. The result of the lack of federal oversight is that skilled nursing facilities, with impunity, divert billions of dollars from care to related party transactions. See letter of 18 state Attorneys General supporting greater accountability in disclosure of ownership information[10] and litigation filed by the New York State Attorney General against three nursing facilities.[11]

Hospital wages for registered nurses and certified nurse aides are typically far higher than SNF wages, according to the Bureau of Labor Statistics (May 2021).

Category of nursing staff	Hospital salaries	Skilled nursing facility salaries
Registered nurse	$85,020	$72,260
Licensed practical nurse	$48,050	$53,670
Certified nurse aides	$32,090	$35,190

From: Bureau of Labor Statistics (May 2021)
RNs, https://www.bls.gov/oes/current/oes291141.htm
LPNs, https://www.bls.gov/oes/current/oes292061.htm
CNAs, https://www.bls.gov/oes/current/oes311131.htm

CMS’s continued use of hospital wages to calculate SNF Part A payment rates results in significant overpayments to SNFs.

The Center urges CMS to conduct audits of Medicare cost reports, as Congress authorized nearly 25 years ago, for two important purposes: (1) to begin the process of developing a SNF-specific wage index in order to avoid the overpayments that result from using hospital wages; and (2) to make sure that skilled nursing facilities stop diverting Medicare reimbursement from resident care to related parties and inflated profits.

SNF Quality Reporting Program (SNF QRP), 88 Fed. Reg., 21332

CMS proposes to adopt the Discharge Function Score Measure, beginning with the FY 2025 SNF QRP. 88 Fed. Reg., 21337. The proposed measure is based on self-reported resident assessment data, calculates the percentage of residents “who meet or exceed an expected discharge function score,” and is risk-adjusted. Id. 21338. CMS proposes to use the measure in the Value-Based Purchasing Program as well.

As a general matter, the Center opposes measures that are based exclusively on self-reported resident assessment (MDS) data. As CMS knows from Abt’s evaluations of the five-star reporting system[12] and CMS’s repeated need to recalibrate the quality domain because of facilities’ inflated quality measure scores,[13] self-reported measures are not accurate assessments of facilities’ performance. An example is the proposed falls measure for the SNF Value-Based Purchasing Program, which is discussed later in these comments.

In addition, despite limited language in the preamble about “maintenance” of function, the measure is quite clearly focused on improvement. CMS reports that the patient and family/caregiver advocates (PFAs) as well as the Technical Expert Panels emphasized the importance of a resident’s ability to improve in self-care and mobility. Id. 21340-21341.

Medicare covers maintenance therapy and maintenance nursing services in Part A stays, as confirmed by the federal district court in the nationwide class action Jimmo v. Sebelius[14] (discussed above) and addressed by CMS in its Jimmo Settlement website.[15] The Center is concerned that the maintenance standard of coverage is undermined when the entire focus of the discharge measure is improvement and there is no comparable measure for maintenance coverage.

SNF Value-Based Purchasing Program (SNF VBP), 88 Fed. Reg., 21361

The Center supports the new VBP measure Total Nursing Staff Turnover Measure, beginning with FY 2026 SNF VBP. As CMS points out, staff turnover matters because turnover decreases quality of care for residents.

CMS proposes a new measure, Percent of Residents Experiencing One or More Falls with Major Injury (Long-Stay), based on resident assessment (MDS) data, beginning with FY 2027 SNF VBP. Id. 21368-21371.

The problem with this MDS-based measure is its serious inaccuracy. In prior years, the Center has expressed concern to CMS that MDS data on resident falls are inaccurate and understate residents’ falls, citing Prachi Sanghav, Shengyuan Pan, Daryl Caudry, “Assessment of nursing home reporting of major injury falls for quality measurement on nursing home compare,” Health Services Research, p. 5. 2019;00:1-10. https://doi.org/10.1111/1475-6773.13247.

In that study, the first “national-level assessment of how nursing homes self-report major injury fall rates, which are used by CMS for quality measurement and public reporting,” researchers “found substantial underreporting on the specific Minimum Data Set (MDS) item (J1900C) used by NHC [Nursing Home Compare].” Only 57.5% of residents’ major injury falls that were identified in Medicare hospital admissions claims data were reported on residents’ assessment data.

Researchers analyzed 100% of major injury falls in hospital admissions claims data from the Medicare Provider Analysis and Review (MedPAR) for the period January 1, 2011 to September 30, 2015 (150,828 falls). They compared these claims data to facilities’ self-reported MDS data for the same period, focusing on J1900C (major injury during current stay), “as the responses to this question for long-stay residents are used to create an NHC quality measure and are part of the star rating algorithm.” Id. 2. Researchers found

Only 57.5% of the claims were reported on MDS.
More falls were reported on MDS for long-stay residents (62.9%) than for short-stay residents (47.2%).
More falls were reported on MDS for white residents (59.0%) than for nonwhite residents (46.4%).
Long-stay white residents had the highest reporting rate (64.5%), while short-stay nonwhite residents had the lowest reporting rate (37.4%).

Researchers also found poor correlations between claims-based falls rates and quality measure star ratings and overall ratings. At least 75% of the nursing facilities had a four- or five-star quality measure rating and half the facilities had four- and five-star overall ratings. Id. 4-5, Table 4.

See also Prachi Sanghavi and Zihan Chen, “Underreporting of Quality Measures and Associated Facility Characteristics and Racial Disparities in US Nursing Home Ratings,” JAMA Network Open (May 23, 2023), https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2805170, finding “widespread underreporting of major injury falls and pressure ulcers across US nursing homes, and underreporting was associated with the racial and ethnic composition of a facility.”

Even John O’Connor, an editor of the trade press McKnight’s Long-Term Care News, recognizes, in reporting the Sanghavi/Chen article above, that CMS should not use facilities’ self-reported data. In colorful language, he writes:

What this report actually reveals is a deeper problem: industry self-reporting of bad news.
* * *
Should it be terribly surprising that any person or business might fudge a bit when given the opportunity to do so? Hardly.
No, the real problem here is that it should not be facilities reporting these numbers in the first place.
* * *
But until such information can be obtained from an objective source, we would do well to keep an old adage about data collection in mind: Garbage in, garbage out.

John O’Connor, “Here’s what’s really troubling about the latest damning report,” McKnight’s Long-Term Care News (May 29, 2023), https://www.mcknights.com/daily-editors-notes/heres-whats-really-troubling-about-the-latest-damning-report/

CMS proposes to adopt a Validation Process That Applies to SNF VBP Measures that Are Calculated Using MDS Data. Id. 21398. While validating measures that use MDS data makes sense, such a validation project should occur before CMS adopts any measures in the SNF VBP based on MDS data, not, as CMS is proposing, after it has already adopted MDS measures.

The Center has the same concerns about the Discharge Function Score Measure discussed above in the Quality Reporting discussion.

A concern about the Number of Hospitalizations per 1,000 Long-Stay Resident Days Measure, which measures unplanned hospitalizations of Medicare beneficiaries in traditional Medicare beginning with FY 2027 SNF VBP, is its exclusion of residents using Medicare Part C (managed care). When half of all Medicare beneficiaries nationwide rely on Part C,[16] the measure fails to give an accurate picture of hospitalizations.

Thank you for the opportunity to submit comments.

Sincerely,

Toby S. Edelman
Senior Policy Attorney
tedelman@MedicareAdvocacy.org
(202) 293-5760

[1] National Academy of Sciences, Engineering and Medicine, The National Imperative to Improve Nursing Home Quality: Honoring Our Commitment to Residents, Families, and Staff, p. 111 (2022)

[2] 84 Fed. 34737, 34751 (Jul. 18, 2019), https://www.govinfo.gov/content/pkg/FR-2019-07-18/pdf/2019-14946.pdf

[3] CMS, Nursing Home Data Compendium 2015 Edition, Figure 2.2.e. Percentage Distribution of Scope and Severity of Health Deficiencies: United States, 2014, p. 48 (showing 0.9% of deficiencies as immediate jeopardy; 2.3% of deficiencies as actual harm), https://www.cms.gov/Medicare/Provider-Enrollment-and-Certification/CertificationandComplianc/Downloads/nursinghomedatacompendium_508-2015.pdf ;

[4] See the Elder Justice Newsletter, https://medicareadvocacy.org/newsletter-elder-justice-what-no-harm-really-means-for-residents/, jointly issued by the Long Term Care Community Coalition and the Center, highlighting the “no harm” deficiencies reflecting actual harm.

[5] See Alex Kacik, “Therapists decry layoffs amid SNF reimbursement overhaul,” Modern Healthcare (Oct. 2, 2019), https://www.modernhealthcare.com/payment/therapists-decry-layoffs-amid-snf-reimbursement-overhaul; Danielle Brown, ‘Therapist advocates sharing layoff concerns with CMS,” McKnight’s Long-Term Care News (Oct. 3, 2019), https://www.mcknights.com/news/therapist-advocates-sharing-layoff-concerns-with-cms/

[6] Jimmo Settlement ¶6.a (Jan. 24, 2013).

[7] The Corrective Statement is available at https://www.medicareadvocacy.org/medicare-info/improvement-standard/.

[8] Mary Ersek, “The Waltz: To Improve Nursing Home Care, Invest In The Workforce,” Health Affairs (May 26, 2022), https://www.healthaffairs.org/do/10.1377/forefront.20220524.944560

[9] Available at https://theconsumervoice.org/uploads/files/issues/2023-Related-Party-Report.pdf

[10] Available at https://ag.ny.gov/sites/default/files/letters/Letter%20from%20State%20AGs%20to%20CMS%20re%20Proposed%20Rule%20on%20Nursing%20Home%20Transparency.pdf

[12] Abt, “Nursing Home Compare: The First Four Years of the Five-Star Quality Rating System” (Powerpoint, Nov. 2013), https://www.cms.gov/Medicare/Provider-Enrollment-and-Certification/CertificationandComplianc/Downloads/2013-The-First-Four-Years-of-Five-Star.pdf. See slide 14 (“Quality Measures: Proportion receiving 4 or 5 stars has increased, while the proportion receiving 1 or 2 stars has decreased”), slide 16 (showing changes in quality measure rating: 2009-2013). See also “Changes to Nursing Home Compare and the Five Star Quality Rating System” (CMA Alert, Feb. 26, 2015), https://medicareadvocacy.org/changes-to-nursing-home-compare-and-the-five-star-quality-rating-system/ (discussing recalibration of quality measures)

[13] CMS, “Nursing Home Compare ‘3.0’ – Five Star Quality Rating System – Expanded and Strengthened,” S&C: 15-26-NH (Feb. 13, 2015), https://www.cms.gov/Medicare/Provider-Enrollment-and-Certification/SurveyCertificationGenInfo/Downloads/Survey-and-Cert-Letter-15-26.pdf. See also Katie Thomas, “Medicare Star Ratings Allow Nursing Homes to Game the System,” The New York Times (Aug. 25, 2014), https://www.nytimes.com/2014/08/25/business/medicare-star-ratings-allow-nursing-homes-to-game-the-system.html?searchResultPosition=1

[14] Jimmo v. Sebelius, Civil Action No. 5:11-CV-17-CR (D. Vt. Jan. 24, 2013)

[15] https://www.cms.gov/Center/Special-Topic/Jimmo-Center

[16] Jeannie Fuglesten Biniek, Meredith Freed, Anthony Damico, and Tricia Neuman, KFF, “Half of All Eligible Medicare Beneficiaries Are Now Enrolled in Private Medicare Advantage Plans” (May 1, 2023), https://www.kff.org/policy-watch/half-of-all-eligible-medicare-beneficiaries-are-now-enrolled-in-private-medicare-advantage-plans/

Footer

Stay Connected: