2021-16311. Medicare Program; FY 2022 Hospice Wage Index and Payment Rate Update, Hospice Conditions of Participation Updates, Hospice and Home Health Quality Reporting Program Requirements

AGENCY:

Centers for Medicare & Medicaid Services (CMS), HHS.

ACTION:

Final rule.

SUMMARY:

This final rule updates the hospice wage index, payment rates, and aggregate cap amount for Fiscal Year 2022. This rule makes changes to the labor shares of the hospice payment rates and finalizes clarifying regulations text changes to the election statement addendum that was implemented on October 1, 2020. In addition, this rule makes permanent selected regulatory blanket waivers that were issued to Medicare-participating hospice agencies during the COVID-19 public health emergency (PHE) and updates the hospice conditions of participation. This rule updates the Hospice Quality Reporting Program and finalizes changes beginning with the January 2022 public reporting for the Home Health Quality Reporting Program to address exceptions related to the COVID-19 PHE.

DATES:

These regulations are effective on October 1, 2021.

FOR FURTHER INFORMATION CONTACT:

For general questions about hospice payment policy, send your inquiry via email to [e-mail].

For questions regarding the CAHPS® Hospice Survey, contact Lori Teichman at [phone], Lauren Fuentes at [phone], and Debra Dean-Whittaker at [phone].

For questions regarding the hospice conditions of participation, contact Mary Rossi-Coajou at [phone] and CAPT James Cowher at [phone].

For questions regarding home health public reporting, contact Charles Padgett [phone].

For questions regarding the hospice quality reporting program, contact Cindy Massuda at [phone].

SUPPLEMENTARY INFORMATION:

I. Executive Summary

A. Purpose

This rule updates the hospice wage index, payment rates, and cap amount for fiscal year (FY) 2022 as required under section 1814(i) of the Social Security Act (the Act). In addition, this rule rebases the labor shares of the hospice payment rates and finalizes clarifying regulations text changes to the election statement addendum requirements finalized in the FY 2020 Hospice Wage Index and Payment Rate Update final rule (84 FR 38484). This rule also provides a summary of comments received regarding hospice utilization and spending patterns. This rule makes permanent selected regulatory blanket waivers for hospice agencies during the COVID-19 public health emergency (PHE) and provides revisions to the hospice conditions of participation (CoPs). This rule finalizes changes to the Hospice Quality Reporting Program (HQRP), summarizes the comments to the requests for information on advancing to digital quality measurement and the use of Fast Healthcare Interoperability Resources (FHIR) and the White House Executive Order related to health equity in the HQRP. Finally, this rule finalizes changes to the Home Health Quality Reporting Program (HH QRP) to address the January 2022 refresh in accordance with sections 1895(b)(3)(B)(v)(III) and 1899B(f) of the Act.

B. Summary of the Major Provisions

Section III.A of this final rule includes a summary of comments from the public, including hospice providers as well as patients and advocates, regarding the presented analysis in the FY 2022 hospice proposed rule on hospice utilization, spending patterns and non-hospice spending during a hospice election.

Section III.B of this final rule rebases and revises the labor shares for continuous home care (CHC), routine home care (RHC), inpatient respite care (IRC), and general inpatient care (GIP) using 2018 Medicare cost report (MCR) data for freestanding hospice facilities.

Section III.C of this rule updates the hospice wage index and makes the application of the updated wage data budget neutral for all four levels of hospice care and discusses the FY 2022 hospice payment update percentage of 2.0 percent, updates to the hospice payment rates, as well as the updates to the hospice cap amount for FY 2022 by the hospice payment update percentage of 2.0 percent.

Section III.D finalizes clarifying regulations text changes regarding the election statement addendum requirements that were finalized in the FY 2020 Hospice Wage Index and Rate Update final rule (84 FR 38484).

Section III.E makes permanent selected regulatory blanket waivers that were issued to Medicare-participating hospice agencies during the COVID-19 PHE. We are revising the hospice aide requirements to allow the use of the pseudo-patient for conducting hospice aide competency evaluations. We are also revising the hospice aide supervision requirements to address situations when deficient practice is noted and remediation is needed related to both deficient and related skills, in accordance with § 418.76(c).

In section III.F of this rule, we finalize proposals to the HQRP including the addition of claims-based Hospice Care Index (HCI) measure, and Hospice Visits in the Last Days of Life (HVLDL) measure for public reporting; removal of the seven Hospice Item Set (HIS) measures because a more broadly applicable measure, the NQF #3235 HIS Comprehensive Assessment Measure for the particular topic is available and already publicly reported; and further development of the Hospice Outcome and Patient Evaluation (HOPE) assessment instrument. We also finalize the public reporting change for one refresh cycle to report less than the standard quarters of data due to the COVID-19 PHE exemptions; use 2 years (8 quarters) of data for the claims-based measures in order to report on small providers; and add the Consumer Assessment of Healthcare Providers and Systems (CAHPS®) Hospice Survey Star ratings. Additionally, we summarize the comments on the requests for information (RFI) on advancing to digital quality measurement and the use of FHIR and on addressing the White House Executive Order related to health equity in the HQRP.

Finally, in section III.G of this rule, we are finalizing our proposal to the HH QRP so that, beginning with the January 2022 through the July 2024 public reporting refresh cycle, we will report fewer quarters of data due to COVID-19 PHE exceptions granted on March 27, 2020. We included the HH QRP policy in this rulemaking in order to resume public reporting for the HH QRP with the January 2022 refresh of Care Compare. To accommodate the excepted HH QRP of 2020 Q1 and Q2, we resume public reporting using 3 out of 4 quarters of data for the January 2022 refresh. In order to finalize this proposal in time to release the required preview report related to the January 2022 refresh, which we release 3 months prior to any given refresh (October 2021), we needed the rule containing this proposal to finalize by October 2021.

C. Summary of Impacts

The overall economic impact of this final rule is estimated to be $480 million in increased payments to hospices for FY 2022.

II. Background

A. Hospice Care

Hospice care is a comprehensive, holistic approach to treatment that recognizes the impending death of a terminally ill individual and warrants a change in the focus from curative care to palliative care for relief of pain and for symptom management. Medicare regulations define “palliative care” as patient and family-centered care that optimizes quality of life by anticipating, preventing, and treating suffering. Palliative care throughout the continuum of illness involves addressing physical, intellectual, emotional, social, and spiritual needs and to facilitate patient autonomy, access to information, and choice (42 CFR 418.3). Palliative care is at the core of hospice philosophy and care practices, and is a critical component of the Medicare hospice benefit.

The goal of hospice care is to help terminally ill individuals continue life with minimal disruption to normal activities while remaining primarily in the home environment. A hospice uses an interdisciplinary approach to deliver medical, nursing, social, psychological, emotional, and spiritual services through a collaboration of professionals and other caregivers, with the goal of making the beneficiary as physically and emotionally comfortable as possible. Hospice is compassionate beneficiary and family/caregiver-centered care for those who are terminally ill.

As referenced in our regulations at § 418.22(b)(1), to be eligible for Medicare hospice services, the patient's attending physician (if any) and the hospice medical director must certify that the individual is “terminally ill,” as defined in section 1861(dd)(3)(A) of the Social Security Act (the Act) and our regulations at § 418.3; that is, the individual has a medical prognosis that his or her life expectancy is 6 months or less if the illness runs its normal course. The regulations at § 418.22(b)(2) require that clinical information and other documentation that support the medical prognosis accompany the certification and be filed in the medical record with it and those at § 418.22(b)(3) require that the certification and recertification forms include a brief narrative explanation of the clinical findings that support a life expectancy of 6 months or less.

Under the Medicare hospice benefit, the election of hospice care is a patient choice and once a terminally ill patient elects to receive hospice care, a hospice interdisciplinary group is essential in the seamless provision of primarily home-based services. The hospice interdisciplinary group works with the beneficiary, family, and caregivers to develop a coordinated, comprehensive care plan; reduce unnecessary diagnostics or ineffective therapies; and maintain ongoing communication with individuals and their families about changes in their condition. The beneficiary's care plan will shift over time to meet the changing needs of the individual, family, and caregiver(s) as the individual approaches the end of life.

If, in the judgment of the hospice interdisciplinary team, which includes the hospice physician, the patient's symptoms cannot be effectively managed at home, then the patient is eligible for general inpatient care (GIP), a more medically intense level of care. GIP must be provided in a Medicare-certified hospice freestanding facility, skilled nursing facility, or hospital. GIP is provided to ensure that any new or worsening symptoms are intensively addressed so that the beneficiary can return to his or her home and continue to receive routine home care. Limited, short-term, intermittent, inpatient respite care (IRC) is also available because of the absence or need for relief of the family or other caregivers. Additionally, an individual can receive continuous home care (CHC) during a period of crisis in which an individual requires continuous care to achieve palliation or management of acute medical symptoms so that the individual can remain at home. Continuous home care may be covered for as much as 24 hours a day, and these periods must be predominantly nursing care, in accordance with the regulations at § 418.204. A minimum of 8 hours of nursing care, or nursing and aide care, must be furnished on a particular day to qualify for the continuous home care rate (§ 418.302(e)(4)).

Hospices must comply with applicable civil rights laws,^[1] including section 504 of the Rehabilitation Act of 1973 and the Americans with Disabilities Act, under which covered entities must take appropriate steps to ensure effective communication with patients and patient care representatives with disabilities, including the provisions of auxiliary aids and services at no cost to the individual. Additionally, they must take reasonable steps to ensure meaningful access for individuals with limited English proficiency, consistent with Title VI of the Civil Rights Act of 1964. Further information about these requirements may be found at: http://www.hhs.gov/​ocr/​civilrights.

B. Services Covered by the Medicare Hospice Benefit

Coverage under the Medicare hospice benefit requires that hospice services must be reasonable and necessary for the palliation and management of the terminal illness and related conditions. Section 1861(dd)(1) of the Act establishes the services that are to be rendered by a Medicare-certified hospice program. These covered services include: Nursing care; physical therapy; occupational therapy; speech-language pathology therapy; medical social services; home health aide services (called hospice aide services); physician services; homemaker services; medical supplies (including drugs and biologicals); medical appliances; counseling services (including dietary counseling); short-term inpatient care in a hospital, nursing facility, or hospice inpatient facility (including both respite care and procedures necessary for pain control and acute or chronic symptom management); continuous home care during periods of crisis, and only as necessary to maintain the terminally ill individual at home; and any other item or service which is specified in the plan of care and for which payment may otherwise be made under Medicare, in accordance with Title XVIII of the Act.

Section 1814(a)(7)(B) of the Act requires that a written plan for providing hospice care to a beneficiary who is a hospice patient be established before care is provided by, or under arrangements made by, the hospice program; and that the written plan be periodically reviewed by the beneficiary's attending physician (if any), the hospice medical director, and an interdisciplinary group (section 1861(dd)(2)(B) of the Act). The services offered under the Medicare hospice benefit must be available to beneficiaries as needed, 24 hours a day, 7 days a week (section 1861(dd)(2)(A)(i) of the Act).

Upon the implementation of the hospice benefit, the Congress also expected hospices to continue to use volunteer services, though Medicare does not pay for these volunteer services (section 1861(dd)(2)(E) of the Act). As stated in the Fiscal Year (FY) 1983 Hospice Wage Index and Rate Update proposed rule (48 FR 38149), the hospice must have an interdisciplinary group composed of paid hospice employees as well as hospice volunteers, and that “the hospice benefit and the resulting Medicare reimbursement is not intended to diminish the voluntary spirit of hospices.” This expectation supports the hospice philosophy of community based, holistic, comprehensive, and compassionate end of life care.

C. Medicare Payment for Hospice Care

Sections 1812(d), 1813(a)(4), 1814(a)(7), 1814(i), and 1861(dd) of the Act, and the regulations in 42 CFR part 418, establish eligibility requirements, payment standards and procedures; define covered services; and delineate the conditions a hospice must meet to be approved for participation in the Medicare program. Part 418, subpart G, provides for a per diem payment based on one of four prospectively-determined rate categories of hospice care (routine home care (RHC), CHC, IRC, and GIP), based on each day a qualified Medicare beneficiary is under hospice care (once the individual has elected). This per diem payment is meant to cover all of the hospice services and items needed to manage the beneficiary's care, as required by section 1861(dd)(1) of the Act.

While payments made to hospices are to cover all items, services, and drugs for the palliation and management of the terminal illness and related conditions, Federal funds cannot be used for the prohibited activities, even in the context of a per diem payment. While recent news reports ^[2] have brought to light the potential role hospices could play in medical aid in dying (MAID) where such practices have been legalized in certain states, we wish to remind hospices that The Assisted Suicide Funding Restriction Act of 1997 (Pub. L. 105-12) prohibits the use of Federal funds to provide or pay for any health care item or service or health benefit coverage for the purpose of causing, or assisting to cause, the death of any individual including mercy killing, euthanasia, or assisted suicide. However, the prohibition does not pertain to the provision of an item or service for the purpose of alleviating pain or discomfort, even if such use may increase the risk of death, so long as the item or service is not furnished for the specific purpose of causing or accelerating death.

1. Omnibus Budget Reconciliation Act of 1989

Section 6005(a) of the Omnibus Budget Reconciliation Act of 1989 (Pub. L. 101-239) amended section 1814(i)(1)(C) of the Act and provided changes in the methodology concerning updating the daily payment rates based on the hospital market basket percentage increase applied to the payment rates in effect during the previous Federal fiscal year.

2. Balanced Budget Act of 1997

Section 4441(a) of the Balanced Budget Act of 1997 (BBA) (Pub. L. 105-33) established that updates to the hospice payment rates beginning FY 2002 and subsequent FYs be the hospital market basket percentage increase for the FY. Section 4442 of the BBA amended section 1814(i)(2) of the Act, effective for services furnished on or after October 1, 1997, to require that hospices submit claims for payment for hospice care furnished in an individual's home only on the basis of the geographic location at which the service is furnished. Previously, local wage index values were applied based on the geographic location of the hospice provider, regardless of where the hospice care was furnished. Section 4443 of the BBA amended sections 1812(a)(4) and 1812(d)(1) of the Act to provide for hospice benefit periods of two 90-day periods, followed by an unlimited number of 60-day periods.

3. FY 1998 Hospice Wage Index Final Rule

The FY 1998 Hospice Wage Index final rule (62 FR 42860), implemented a new methodology for calculating the hospice wage index and instituted an annual Budget Neutrality Adjustment Factor (BNAF) so aggregate Medicare payments to hospices would remain budget neutral to payments calculated using the 1983 wage index.

4. FY 2010 Hospice Wage Index Final Rule

The FY 2010 Hospice Wage Index and Rate Update final rule (74 FR 39384) instituted an incremental 7-year phase-out of the BNAF beginning in FY 2010 through FY 2016. The BNAF phase-out reduced the amount of the BNAF increase applied to the hospice wage index value, but was not a reduction in the hospice wage index value itself or in the hospice payment rates.

5. The Affordable Care Act

Starting with FY 2013 (and in subsequent FYs), the market basket percentage update under the hospice payment system referenced in sections 1814(i)(1)(C)(ii)(VII) and 1814(i)(1)(C)(iii) of the Act are subject to annual reductions related to changes in economy-wide productivity, as specified in section 1814(i)(1)(C)(iv) of the Act.

In addition, sections 1814(i)(5)(A) through (C) of the Act, as added by section 3132(a) of the Patient Protection and Affordable Care Act (PPACA) (Pub. L. 111-148), required hospices to begin submitting quality data, based on measures specified by the Secretary of the Department of Health and Human Services (the Secretary), for FY 2014 and subsequent FYs. Since FY 2014, hospices that fail to report quality data have their market basket percentage increase reduced by 2 percentage points. Note that with the passage of the Consolidated Appropriations Act, 2021 (hereafter referred to as CAA 2021) (Pub. L. 116 260), the reduction changes to 4 percentage points beginning in FY 2024.

Section 1814(a)(7)(D)(i) of the Act, as added by section 3132(b)(2) of the

PPACA, required, effective January 1, 2011, that a hospice physician or nurse practitioner have a face-to-face encounter with the beneficiary to determine continued eligibility of the beneficiary's hospice care prior to the 180th day recertification and each subsequent recertification, and to attest that such visit took place. When implementing this provision, the Centers for Medicare & Medicaid Services (CMS) finalized in the FY 2011 Hospice Wage Index final rule (75 FR 70435) that the 180th day recertification and subsequent recertifications would correspond to the beneficiary's third or subsequent benefit periods. Further, section 1814(i)(6) of the Act, as added by section 3132(a)(1)(B) of the PPACA, authorized the Secretary to collect additional data and information determined appropriate to revise payments for hospice care and other purposes. The types of data and information suggested in the PPACA could capture accurate resource utilization, which could be collected on claims, cost reports, and possibly other mechanisms, as the Secretary determined to be appropriate. The data collected could be used to revise the methodology for determining the payment rates for RHC and other services included in hospice care, no earlier than October 1, 2013, as described in section 1814(i)(6)(D) of the Act. In addition, CMS was required to consult with hospice programs and the Medicare Payment Advisory Commission (MedPAC) regarding additional data collection and payment revision options.

6. FY 2012 Hospice Wage Index Final Rule

In the FY 2012 Hospice Wage Index final rule (76 FR 47308 through 47314) it was announced that beginning in 2012, the hospice aggregate cap would be calculated using the patient-by-patient proportional methodology, within certain limits. Existing hospices had the option of having their cap calculated through the original streamlined methodology, also within certain limits. As of FY 2012, new hospices have their cap determinations calculated using the patient-by-patient proportional methodology.

7. IMPACT Act of 2014

The Improving Medicare Post-Acute Care Transformation Act of 2014 (IMPACT Act) (Pub. L. 113-185) became law on October 6, 2014. Section 3(a) of the IMPACT Act mandated that all Medicare certified hospices be surveyed every 3 years beginning April 6, 2015 and ending September 30, 2025. In addition, section 3(c) of the IMPACT Act requires medical review of hospice cases involving beneficiaries receiving more than 180 days of care in select hospices that show a preponderance of such patients; section 3(d) of the IMPACT Act contains a new provision mandating that the cap amount for accounting years that end after September 30, 2016, and before October 1, 2025 be updated by the hospice payment percentage update rather than using the consumer price index for urban consumers (CPI-U) for medical care expenditures.

8. FY 2015 Hospice Wage Index and Payment Rate Update Final Rule

The FY 2015 Hospice Wage Index and Rate Update final rule (79 FR 50452) finalized a requirement that the Notice of Election (NOE) be filed within 5 calendar days after the effective date of hospice election. If the NOE is filed beyond this 5-day period, hospice providers are liable for the services furnished during the days from the effective date of hospice election to the date of NOE filing (79 FR 50474). As with the NOE, the claims processing system must be notified of a beneficiary's discharge from hospice or hospice benefit revocation within 5 calendar days after the effective date of the discharge/revocation (unless the hospice has already filed a final claim) through the submission of a final claim or a Notice of Termination or Revocation (NOTR).

The FY 2015 Hospice Wage Index and Rate Update final rule (79 FR 50479) also finalized a requirement that the election form include the beneficiary's choice of attending physician and that the beneficiary provide the hospice with a signed document when he or she chooses to change attending physicians.

In addition, the FY 2015 Hospice Wage Index and Rate Update final rule (79 FR 50496) provided background, described eligibility criteria, identified survey respondents, and otherwise implemented the Hospice Experience of Care Survey for informal caregivers. Hospice providers were required to begin using this survey for hospice patients as of 2015.

Finally, the FY 2015 Hospice Wage Index and Rate Update final rule required providers to complete their aggregate cap determination not sooner than 3 months after the end of the cap year, and not later than 5 months after, and remit any overpayments. Those hospices that fail to submit their aggregate cap determinations on a timely basis will have their payments suspended until the determination is completed and received by the Medicare contractor (79 FR 50503).

9. FY 2016 Hospice Wage Index and Payment Rate Update Final Rule

In the FY 2016 Hospice Wage Index and Rate Update final rule (80 FR 47142), CMS finalized two different payment rates for RHC: A higher per diem base payment rate for the first 60 days of hospice care and a reduced per diem base payment rate for subsequent days of hospice care. CMS also finalized a service intensity add-on (SIA) payment payable for certain services during the last 7 days of the beneficiary's life. A service intensity add-on payment will be made for the social worker visits and nursing visits provided by a registered nurse (RN), when provided during routine home care in the last 7 days of life. The SIA payment is in addition to the routine home care rate. The SIA payment is provided for visits of a minimum of 15 minutes and a maximum of 4 hours per day (80 FR 47172).

In addition to the hospice payment reform changes discussed, the FY 2016 Hospice Wage Index and Rate Update final rule implemented changes mandated by the IMPACT Act, in which the cap amount for accounting years that end after September 30, 2016 and before October 1, 2025 would be updated by the hospice payment update percentage rather than using the CPI-U (80 FR 47186). In addition, we finalized a provision to align the cap accounting year for both the inpatient cap and the hospice aggregate cap with the FY for FY 2017 and thereafter. Finally, the FY 2016 Hospice Wage Index and Rate Update final rule (80 FR 47144) clarified that hospices would have to report all diagnoses on the hospice claim as a part of the ongoing data collection efforts for possible future hospice payment refinements.

10. FY 2017 Hospice Wage Index and Payment Rate Update Final Rule

In the FY 2017 Hospice Wage Index and Rate Update final rule (81 FR 52160), CMS finalized several new policies and requirements related to the Hospice Quality Reporting Program (HQRP). First, CMS codified the policy that if the National Quality Forum (NQF) made non-substantive changes to specifications for HQRP measures as part of the NQF's re-endorsement process, CMS would continue to utilize the measure in its new endorsed status, without going through new notice-and-comment rulemaking. CMS would continue to use rulemaking to adopt substantive updates made by the NQF to the endorsed measures adopted for the HQRP; determinations about what constitutes a substantive versus non-substantive change would be made on a measure-by-measure basis. Second, we finalized two new quality measures for the HQRP for the FY 2019 payment determination and subsequent years: Hospice Visits when Death is Imminent Measure Pair and Hospice and Palliative Care Composite Process Measure-Comprehensive Assessment at Admission (81 FR 52173). The data collection mechanism for both of these measures is the Hospice Item Set (HIS), and the measures were effective April 1, 2017. Regarding the Consumer Assessment of Healthcare Providers and Systems (CAHPS®) Hospice Survey, CMS finalized a policy that hospices that receive their CMS Certification Number (CCN) after January 1, 2017 for the FY 2019 Annual Payment Update (APU) and January 1, 2018 for the FY 2020 APU will be exempted from the Hospice CAHPS® requirements due to newness (81 FR 52182). The exemption is determined by CMS and is for 1 year only.

11. FY 2020 Hospice Wage Index and Payment Rate Update Final Rule

In the FY 2020 Hospice Wage Index and Rate Update final rule (84 FR 38484), we finalized rebased payment rates for CHC and GIP and set those rates equal to their average estimated FY 2019 costs per day. We also rebased IRC per diem rates equal to the estimated FY 2019 average costs per day, with a reduction of 5 percent to the FY 2019 average cost per day to account for coinsurance. We finalized the FY 2020 proposal to reduce the RHC payment rates by 2.72 percent to offset the increases to CHC, IRC, and GIP payment rates to implement this policy in a budget-neutral manner in accordance with section 1814(i)(6) of the Act (84 FR 38496).

In addition, we finalized a policy to use the current year's pre-floor, pre-reclassified hospital inpatient wage index as the wage adjustment to the labor portion of the hospice rates. Finally, in the FY 2020 Hospice Wage Index and Rate Update final rule (84 FR 38505), we finalized modifications to the hospice election statement content requirements at § 418.24(b) by requiring hospices, upon request, to furnish an election statement addendum effective beginning in FY 2021. The addendum must list those items, services, and drugs the hospice has determined to be unrelated to the terminal illness and related conditions, increasing coverage transparency for beneficiaries under a hospice election.

12. Consolidated Appropriations Act, 2021

Division CC, section 404 of Consolidated Appropriations Act, 2021 (CAA 2021) amended section 1814(i)(2)(B) of the Act and extended the provision that currently mandates the hospice cap be updated by the hospice payment update percentage (hospital market basket update reduced by the productivity adjustment) rather than the CPI-U for accounting years that end after September 30, 2016 and before October 1, 2030. Prior to enactment of this provision, the hospice cap update was set to revert to the original methodology of updating the annual cap amount by the CPI-U beginning on October 1, 2025. Division CC, section 407 of CAA 2021 revises section 1814(i)(5)(A)(i) to increase the payment reduction for hospices who fail to meet hospice quality measure reporting requirements from two percent to four percent beginning with FY 2024.

III. Provisions of the Final Rule

A. Hospice Utilization and Spending Patterns

In the FY 2022 proposed rule (86 FR 19700), CMS provided data analysis on hospice utilization trends from FY 2010 through FY 2019. The analysis included data on the number of beneficiaries using the hospice benefit, live discharges, reported diagnoses on hospice claims, Medicare hospice spending, and Parts A, B and D non-hospice spending during a hospice election. The proposed rule also solicited comments from the public, hospice providers, patients and advocates regarding hospice utilization and spending patterns. We also solicited comments regarding skilled visits in the last week of life, particularly, what factors determine how and when visits are made as an individual approaches the end of life and how hospices make determinations as to what items, services and drugs are related versus unrelated to the terminal illness and related conditions. That is, how do hospices define what is unrelated to the terminal illness and related conditions when establishing a hospice plan of care.

Likewise, we solicited comments on what other factors may influence whether or how certain services are furnished to hospice beneficiaries. Finally, we requested feedback from stakeholder as to whether the hospice election statement addendum has changed the way hospices make care decisions and how the addendum is used to prompt discussions with beneficiaries and non-hospice providers to ensure that the care needs of beneficiaries who have elected the hospice benefit are met. A summary of these comments and our response to those comments appear below:

1. Hospice Utilization and Spending Patterns

Several commenters thanked CMS for continuing to incorporate monitoring and data analysis into its proposed hospice payment rule. Many commenters stated that while the structure of the hospice benefit and approach to care at the end of life remain unchanged, changes in the characteristics of patients served (particularly the shift from predominantly cancer patients to those with end-stage neurological and other conditions) is largely responsible for driving changes in utilization trends and hospice practice over recent decades. Many commenters suggested that CMS provide more detailed analysis of physician billing as it relates to non-hospice spending and a few commenters suggested that CMS release additional data connected to CMS' Part D spending analysis to better inform stakeholders and assist in helping to determine what factors may be contributing to these increased Part D expenditures during a hospice election.

2. Skilled Visits in the Last Days of Life

One commenter stated that the service intensity add-on (SIA) payment has been one of the greatest improvements in the hospice benefit in recent years. Many commenters recommended that CMS modify the SIA payments to include any visits which could be counted toward end-of-life care, not just skilled visits (for example, chaplain and spiritual care or hospice aide).

3. Items, Services, and Drugs Related and Unrelated to the Terminal Illness and Related Conditions

Several commenters stated that the determination of relatedness, as applied to coverage decisions connected to terminal prognosis, is a clinical decision specific to the unique clinical circumstances of each patient. Several commenters stated that they work in collaboration with their respective IDGs to determine the items, services, and drugs that are related versus unrelated once the comprehensive assessment is completed.

4. Election Statement Addendum

Several commenters stated that the addendum has not changed their practices for determining what is related or unrelated under the hospice benefit, but has enhanced the upfront communication with patients and representatives during the admission process. One commenter stated that their hospice revisited the way relatedness is defined, and realized that many diagnoses that were previously thought to be unrelated were related. Another commenter stated that very few patients and their representatives have requested the addendum and that the burden of implementation of the addendum outweighs the benefits.

We appreciate the comments provided regarding the analysis presented in the proposed rule. We plan continue to monitor hospice trends and vulnerabilities within the hospice benefit. We will consider these comments and suggestions for ongoing monitoring analyses, program integrity efforts, and for potential future rulemaking.

B. FY 2022 Labor Shares

1. Background

The labor share for CHC and RHC of 68.71 percent was established with the FY 1984 Hospice benefit implementation based on the wage/nonwage proportions specified in Medicare's limit on home health agency costs (48 FR 38155 through 38156). The labor shares for IRC and GIP are currently 54.13 percent and 64.01 percent, respectively. These proportions were based on skilled nursing facility wage and nonwage cost limits and skilled nursing facility costs per day (48 FR 38155 through 38156; 56 FR 26917).

In the FY 2022 proposed rule (86 FR 19717 through 19719), we proposed to rebase and revise the labor shares for CHC, RHC, IRC and GIP using Medicare cost report (MCR) data for freestanding hospices (collected via CMS Form 1984-14, OMB NO. 0938-0758) for 2018. We proposed to continue to establish separate labor shares for CHC, RHC, IRC, and GIP and base them on the calculated compensation cost weights for each level of care from the 2018 MCR data. We describe our proposed methodology for deriving the compensation cost weights for each level of care using the MCR data below as well as a summary of the comments received and our responses.

Twenty unique stakeholders submitted their comments on the proposal to rebase the hospice labor shares. In response to public comments, we are adopting the revised hospice labor shares calculated as we proposed with a slight modification to the methodology.

Comment: A few commenters supported the proposal to rebase the labor share for the four levels of care based on the 2018 MCR data. One commenter supported the proposed methodology of using actual hospice cost report data calculated using all applicable costs as well as including only providers who performed each level of care normalizing for outliers. Another commenter stated it was appropriate that the hospice labor shares be based on data for hospice providers, rather than home health agencies and skilled nursing facilities. Several commenters stated that basing the hospice labor shares on recent MCR data for hospice providers will improve payment accuracy.

One commenter strongly encouraged CMS not to revise the labor share using the 2018 MCR for freestanding hospices. One commenter opposed the proposed labor shares, stating that the data in the cost report do not provide adequate or appropriate measures of labor expenses. One commenter agreed with the increased labor share for CHC and for IRC, but did not agree with lowering the labor share for RHC and GIP. One commenter acknowledged the rationale for using hospice cost report data, but stated that this will reduce reimbursement for many of their members, particularly those who provide more GIP than average.

Response: We believe that our proposal to revise the labor shares based on MCR data for hospice providers is a technical improvement to the current labor shares and appreciate the support from the commenters.

We disagree with commenters that the hospice MCR data does not provide adequate or appropriate measure of labor expenses. The MCR data captures detailed labor and non-labor expenses for patient (including but not limited to nursing, physician, therapy and medical supply expenses) and non-patient expenses (such as administrative and general) by level of care. We would note that the freestanding hospice MCR data was used to rebase the hospice payment rates effective for FY 2020 (84 FR 38487 to 38496). In addition, we remind providers that when submitting the MCR data they must certify the cost report that “to the best of [their] knowledge and belief, [the] report and statement are true, correct, complete and prepared from the books and records of the provider in accordance with applicable instructions, except as noted.”

Comment: Several commenters expressed concern regarding the impact of COVID-19 on labor costs. Commenters stated that while they do not yet know the full extent of the impact on labor costs, they expect it to be significant. They stated that the PHE could considerably change the labor share in the next several years of cost report data, as the use of cost reports has a 2-year delay in data. These commenters stated that the impact of COVID-19 on the labor component of the rates cannot be captured in cost report data that is at least 2 years old. The commenters requested consideration of the impact of COVID-19 when setting labor shares for future years.

Several other commenters stated that hospices face significant challenges in the labor market, particularly for nurses. They stated that more nurses are retiring, competition for available nurses is fierce, and many hospices are paying premium salaries and bonuses to recruit and retain qualified nursing staff. One commenter stated that the hospice per diem structure severely limits the amounts they can spend on staff. One commenter stated during the pandemic more time has been needed to train and retrain on infection control standards, as well as changes in communication due to practice changes.

One commenter stated that it is difficult to attract nurses to their geographic area because of the increase in the median home price between January 2021 and May 2021. The commenter stated that they are forced to outsource many nursing functions at high cost, along with paying retention bonuses to current staff. The commenter stated that these labor market challenges will have an impact on the labor shares, which will not necessarily be reflected when the cost report data used is 2 years old. One commenter urged CMS to give special consideration to challenges faced by rural health care providers with specific attention given to the impact workforce shortages have in setting reimbursement rates related to the labor shares.

Response: We acknowledge and appreciate the commenters' concerns regarding labor costs and understand the challenges created by the PHE. We believe using updated labor shares based on 2018 data is a technical improvement over the current labor shares as they reflect recent cost data for freestanding hospice providers. The current labor shares were primarily based on data from the early 1980s. The proposed labor shares reflect the skilled care (including the number of visits) provided under the hospice per diem payment rates for each level of care. For example, the higher labor share for CHC compared to RHC reflects the higher number of visits per day provided with CHC relative to RHC. The current labor shares did not reflect this differential in utilization as the same labor share was used for both levels of care.

We plan on reviewing the 2020 hospice MCR data when complete information is available that will allow us to consider whether the hospice labor shares based on 2018 data are still appropriate. Any future revisions to the hospice labor shares will be proposed and subject to public comments in future rulemaking.

Comment: Several commenters expressed concerns about the frequency of updating the labor shares in the future. A few of these commenters requested that CMS provide further clarification of the frequency of updates to the labor shares with hospice cost report data. One commenter stated that it is important that CMS address this frequency so that hospices and cost report preparers can ensure that the data submitted on the cost report can be used for the labor share calculations.

Response: We acknowledge the commenters' concern that the proposed rule did not explicitly state when we plan to propose any revisions to the hospice labor shares beyond FY 2022.

The labor shares for other PPS systems (for example, IPPS, SNF, IRF, IPF, and LTCH) are typically rebased every four to five years. We tentatively plan to rebase the hospice labor shares on a similar schedule as the other payment systems under Medicare. However, in light of the COVID-19 PHE, we plan to monitor the upcoming MCR data to see if a more frequent revision to the hospice labor shares is necessary in order to reflect the most recent cost structures of hospice providers. We note that any future revisions to the hospice labor shares will be proposed and subject to public comments in future rulemaking.

Comment: A few commenters stated that while they understand the desire and rationale for using hospice data to revise the hospice labor shares (and to make other policy changes), they believe it is important to recognize that the data inputs utilized must be appropriate to the task. The commenters stated that the hospice cost report in its current form does not suit all data purposes for hospice policy changes, and does not fully support calculation of the hospice payment rate labor shares.

One commenter noted that the hospice cost report for freestanding providers is being proposed to be used for the first time to determine the labor component of the rates for each level of care. While the commenter commended CMS for using hospice-specific data, they were also concerned about the accuracy of the data submitted by providers.

One commenter stated concern that due to hospice MCRs not being audited, as well as some sections of the cost report offering multiple methods of reporting, there is a general lack of consistency in the way that the reports are completed by hospice providers that will necessarily distort the average labor figures. The commenter was also concerned that it's not likely that most payroll applications used by hospice providers can correctly allocate costs by level of care, so due to different methods applied by hospice providers to estimate this, the labor costs will also be impacted.

One commenter stated that there are no checks and balances on whether cost reporting data are accurate. They claimed that classifying costs across the four levels of care can contain inaccuracies, particularly when staff allocate time to various levels of care in the same working day. The commenter stated that there are no regulations that require cost reports to be completed by an outside or otherwise qualified accounting firm, and many hospices are doing their own costs reports without complete understanding of how to allocate specific costs and which box is appropriate for particular costs. They stated that the number of hospices that do not pass level 1 edits is also of concern.

One commenter stated that they do not believe hospice cost reports are historically very accurate. They stated that in many healthcare systems someone from the accounting department completed the cost report form with very little input from the hospice program. The commenter stated that they never had an opportunity to review the cost report prior to submission to verify the information was accurate and that they believe this is a common occurrence across the country. Therefore, the commenter stated that they do not believe that cost reports capture labor costs very accurately.

A few commenters stated that if data from the hospice cost report is to be used for calculating the labor component by level of care, revisions to the cost report should be proposed to address current inconsistent, but acceptable, reporting practices. Further, the commenters stated that these changes should be instituted to ensure greater accuracy of the data being used to establish labor shares for GIP and IRC. A few commenters stated that these changes should be implemented as quickly as possible, and once they are in place CMS should undertake a recalculation of the labor shares.

Response: The freestanding hospice MCR form used for the proposed labor shares (CMS-1984-14; OMB NO. 0938-0758) was revised effective for cost reporting periods beginning on or after October 1, 2014 in response to section 1814(i)(6) of the Act, as added by section 3132(a)(1)(B) of the PPACA, which authorized the Secretary to collect additional data and information determined appropriate to revise payments for hospice care and other purposes. The types of data and information suggested in the PPACA could capture accurate resource utilization, which could be collected on claims, cost reports, and possibly other mechanisms, as the Secretary determined to be appropriate.

CMS form 1984-14 was proposed and subject to public comments. Hospice providers previously completed MCR form (CMS-1984-89, OMB NO. 0938-0758). The revised MCR enabled CMS to collect more detailed data regarding labor costs by level of care. The prior MCR did not collect total costs by level of care or detailed costs by level of care (such as labor and nonlabor).

We disagree with the commenter that the cost report in its current form does not support the calculation of the hospice payment rate labor shares. Providers are required to report detailed patient costs (including but not limited to nursing, physician, therapy, and medical supplies) and non-patient costs for each level of care. These costs are further subdivided into labor and non-labor costs.

Our proposal to use the 2018 MCR data recognizes that providers have had 4 years to familiarize themselves with the form and, thereby, improve the accuracy of the data. We note that based on comments received during the CMS-1984-14; OMB NO. 0938-0758 clearance process, the implementation of the MCR form was delayed to October 1, 2014. In addition, as stated previously, providers must certify the cost report that “to the best of [their] knowledge and believe, [the] report and statement are true, correct, complete and prepared from the books and records of the provider in accordance with applicable instructions, except as noted.” Nonetheless, we recognize that data can be misreported at times and, therefore, our proposal for revising the labor shares included applying several edits to remove possible outlier data—a common statistical practice.

We continue to encourage hospice providers to report accurate and complete data on the cost reports. We will evaluate and consider any future changes to the hospice cost report that will allow for the collection of data that may improve the calculation of the hospice labor shares. In addition, we will monitor the compensation cost weights reported by hospices over time to determine if changes to the labor share are appropriate. Any future changes to the cost report or labor shares would be subject to public comments.

While we acknowledge that hospice providers can use different methodologies for reporting data, we believe that our proposed methodology allows for these differences and still results in a reasonable and accurate measure of the cost structures of hospice facilities.

The proposed labor shares are based on MCR data for freestanding hospice facilities. As stated in the proposed rule, we did explore the possibility of using facility-based hospice MCR data to calculate the compensation cost weights; however, very few providers passed the Level I edits (as described in more detail below) and so these reports were not usable.

Comment: One commenter stated that the finances of freestanding hospices are significantly different than those of hospices based at hospitals, home health agencies and nursing homes; therefore, data from freestanding hospices should not be allowed to represent the industry as a whole.

Response: As stated in the FY 2022 Hospice Wage Index and Rate Update proposed rule (86 FR 19717), we did explore the possibility of using facility-based hospice MCR data to calculate the compensation cost weights; however, very few providers passed the Level I edits and so these reports were not usable. We also plan to continue to review the 2020 hospital-based hospice MCR data to see if the reporting of the detailed expense data by level of care has improved for possible incorporation into the labor share calculations. We would note that the freestanding hospice providers account for about 85 percent of hospice providers and therefore, we believe our proposal to use only the freestanding hospice MCR data to revise the labor shares is reasonable and a technical improvement over the current labor shares.

2. Methodology for Calculating Compensation Costs

We proposed to derive a compensation cost weight for each level of care that consists of five major components: (1) Direct patient care salaries and contract labor costs, (2) direct patient care benefits costs, (3) other patient care salaries, (4) overhead salaries, and (5) overhead benefits costs. For each level of care, we proposed to use the same methodology to derive the components; however, for the (1) direct patient care salaries and (3) other patient care salaries, we proposed to use the MCR worksheet that is specific to that level of care (that is, Worksheet A-1 for CHC, Worksheet A-2 for RHC, Worksheet A-3 for IRC, and Worksheet A-4 for GIP).

a. Direct Patient Care Salaries and Contract Labor Costs

Direct patient care salaries and contract labor costs are costs associated with medical services provided by medical personnel including but not limited to physician services, nurse practitioners, RNs, and hospice aides. We proposed to define direct patient care salaries and contract labor costs to be equal to costs reported on Worksheet A-1 (for CHC) or Worksheet A-2 (for RHC) or Worksheet A-3 (for IRC) or Worksheet A-4 (for GIP), column 7, for lines 26 through 37.

Comment: One specific concern of the commenters regarding the proposed methodology was on the data used from Worksheet A-1 and A-2 column 7, lines 26 through 37 for total labor costs associated with each respective level of care. The commenters stated that certain costs are not consistently reported by hospices despite these costs being in compliance with cost reporting instructions. For example, the commenters provided that some hospices track mileage allowances enabling them to be reported on Worksheet A-1 and A-2 while other hospices allocate these mileage reimbursement costs via Worksheet B and B-1 using miles traveled. The commenters asked CMS whether any consideration was given to this inconsistent, but acceptable, reporting for mileage allowances.

Response: We appreciate the commenter's concern. The proposed methodology for calculating the labor shares cited by the commenter of using Worksheet A-1 and A-2 column 7, lines 26 through 37 for total labor costs reflects only one component of the proposed calculation of the labor share. As discussed in the FY 2022 Hospice proposed rule (86 FR 19718) and above, we proposed to derive Direct patient care salaries and contract labor costs using (for CHC as an example) Worksheet A-1 column 7, lines 26 through 37 on the cost report, which would capture any staff transportation costs reported in these cost centers on Worksheet A-1.

Also included in the compensation costs for each level of care, as discussed in the FY 2022 Hospice proposed rule (86 FR 19718) and below, is a proportion overhead salaries and benefits. The overhead salaries includes those reported in the staff transportation cost center (reported in Worksheet A, column 1, line 12) and the overhead benefits for the staff transportation cost center (Worksheet B, column 3, line 12).

Therefore, after consideration of public comments, we believe that our proposed methodology is capturing both the direct patient care costs reported on Worksheet A-1 and any overhead salaries and overhead benefits related to staff transportation costs that are allocated on Worksheet B. We believe that the non-salary non-benefit costs for staff transportation that are allocated on Worksheet B (for example, cost of owning or renting vehicles) should not be included in the labor share of the hospice payment rate that is adjusted by the wage index, as they are not compensation costs, nor do they vary with the local labor market.

b. Direct Patient Care Benefits Costs

We proposed that direct patient care benefits costs for CHC are equal to Worksheet B, column 3, line 50, for RHC are equal to Worksheet B, column 3, line 51, for IRC are equal to Worksheet B, column 3, line 52, and for GIP are equal to Worksheet B, column 3, line 53.

c. Other Patient Care Salaries

Other patient care salaries are those salaries attributable to patient services including but not limited to patient transportation, labs, and imaging services. These salaries reflecting all levels of care are reported on Worksheet A, column 1, lines 38 through 46 and then are further disaggregated for CHC, RHC, IRC, and GIP on Worksheets A-1, A-2, A-3, and A-4, respectively, on column 1 (salaries), lines 38 through 46. Our analysis, however, found that many providers were not reporting salaries on the detailed level of care worksheets (A-1, A-2, A-3, A-4, column 1), but rather reporting total costs (reflecting salary and nonsalary costs) for these services for each level of care on Worksheets A-1, A-2, A-3, A-4, column 7. Therefore, we proposed to estimate other patient care salaries attributable to CHC, RHC, IRC, and GIP by first calculating the ratio of total facility (reflecting all levels of care) other patient care salaries (Worksheet A, column 1, lines 38 through 46) to total facility other patient care total costs (Worksheet A, column 7, lines 38 through 46). For CHC, we proposed to then multiply this ratio by other patient care total costs for CHC (Worksheet A-1 column 7, lines 38 through 46). For RHC, we proposed to multiply this ratio by total other patient care costs for RHC (Worksheet A-2, column 7, lines 38 through 46). For IRC, we proposed to multiply this ratio by total other patient care costs for IRC (Worksheet A-3, column 7, lines 38 through 46). For GIP, we proposed to multiply this ratio by total other patient care costs for GIP (Worksheet A-4, column 7, lines 38 through 46). This proposed methodology assumes that the proportion of salary costs to total costs for other patient care services is consistent for each of the four levels of care.

Comment: One commenter stated that the proposed methodology for calculating compensation costs omits two of the required disciplines in a hospice patient's interdisciplinary team. They stated that social workers and counselors provide direct patient care along with nurses and hospice aides in both routine home care and general inpatient care. The commenter claimed that the proposed methodology only captures salaries and benefits of physicians, nurse practitioners, RNs and hospice aides. The commenter stated that this disregards the essence of the hospice interdisciplinary team which cares for the patient and family as a unit of care. Social workers and counselors serve both the patient and their family. Their salaries and benefits must also be captured in the methodology. The commenter stated that it is unclear in the proposed rule whether they are included in “Other Patient Care Salaries” since only mentioned are patient transportation, labs and imaging services.

Response: As stated in the FY 2022 hospice proposed rule (86 FR 19717 through 19719) as well as above, we proposed that Direct patient care salaries and contract labor costs be equal to costs reported on Worksheet A-1 (for CHC) or Worksheet A-2 (for RHC) or Worksheet A-3 (for IRC) or Worksheet A-4 (for GIP), column 7, for lines 26 through 37 (86 FR 19718). These lines include Medical Social Services (line 33), Spiritual Counseling (line 34), Dietary Counseling (line 25), and Counseling Other (line 36). Therefore, we proposed to include direct patient care salaries and contract labor for social workers and counselors in the calculation of the labor shares.

d. Overhead Salaries

The MCR captures total overhead costs (including but not limited to administrative and general, plant operations and maintenance, and housekeeping) attributable to each of the four levels of care. To estimate overhead salaries for each level of care, we first proposed to calculate noncapital nonbenefit overhead costs for each level of care to be equal to Worksheet B, column 18, less the sum of Worksheet B, columns 0 through 3, for line 50 (CHC), or line 51 (RHC) or line 52 (IRC) or line 53 (GIP). We then proposed to multiply these noncapital nonbenefit overhead costs for each level of care times the ratio of total facility overhead salaries (Worksheet A, column 1, lines 4 through 16) to total facility noncapital nonbenefit overhead costs (which is equal to Worksheet B, column 18 (total costs), line 101 less the sum of Worksheet B, columns 0 (direct patient care costs), column 1 (fixed capital), column 2 (moveable capital) and column 3 (employee benefits), line 101).

e. Overhead Benefits Costs

To estimate overhead benefits costs for each level of care, we proposed a similar methodology to overhead salaries. For each level of care, we proposed to calculate noncapital overhead costs for each level of care to be equal to Worksheet B, column 18, less the sum of Worksheet B, columns 0 through 2, for line 50 (CHC), or line 51 (RHC) or line 52 (IRC) or line 53 (GIP). We then proposed to multiply these noncapital overhead costs for each level of care times the ratio of total facility overhead benefits (Worksheet B, column 3, lines 4 through 16) to total facility noncapital overhead costs (Worksheet B, column 18, line 101 less the sum of Worksheet B, columns 0 through 2, line 101). This proposed methodology assumes the ratio of total overhead benefit costs to total noncapital overhead costs is consistent among all four levels of care.

Comment: Another specific concern raised by the commenters was that there are inconsistencies in reporting medical supply and pharmacy costs on line 10 and line 14 of Worksheet A. They stated that some hospices use Worksheets A-1, A-2, A-3, and A-4 to report all or most of these costs whereas others use lines 10 and lines 14 and report costs as overhead costs. The commenters recommended that CMS look further into reporting all pharmacy and medical supply costs as direct patient care costs on future cost reports. The commenter stated that other acceptable cost reporting methods may be applicable; however, a Level 1 edit is not currently produced if costs are reported in one of the two acceptable locations.

Response: As described in the FY 2022 hospice proposed rule (86 FR 19717 through 19719), our proposed calculation to derive the hospice labor shares uses the sum of five categories of compensation costs. The estimated compensation costs related to medical supply and pharmacy costs would be reflected in the Other Patient Care Salaries, Overhead Salaries, and Overhead Benefits categories. We proposed that total costs for CHC be equal to Worksheet B, column 18, line 50, for RHC are equal to Worksheet B, column 18, line 51, for IRC would be equal to Worksheet B, column 18, line 52, and for GIP are equal to Worksheet B, column 18, line 53. These total costs would reflect medical supply and pharmacy costs when reported on Worksheet A line 10 and 14 or when reported on Worksheet A-1, A-2, A-3, and A-4. Therefore, we believe our proposed methodology captures these costs appropriately. However, we will consider this comment when requesting any future revisions to the Level 1 edits applied to the hospice cost report.

Comment: One commenter had concerns with the inconsistent reporting of certain types of overhead expenses among hospices. They stated in some instances, Medical Directors are employees and salaries would be reported; however, other hospices contract for this position. The commenter stated that the contracted payments for Medicare Directors are not included in the proposed calculation of overhead salaries. The commenter asked whether any consideration was made regarding this inconsistency or other common inconsistencies in the nature of the expenses.

Response: We appreciate the commenter's concern and conducted an additional review of our proposed methodology for appropriately capturing overhead costs in the labor shares.

As noted by the commenter, salaries and benefit costs for employed Medical Directors would be reported in Worksheet A, column 1, line 15 (salaries) and Worksheet B, column 3, line 15 (benefits), which are both included in our proposed methodology as these expenses are reported in overhead salaries and overhead benefits. As described in the proposed rule (86 FR 19718) and above, we include a proportion overhead salaries and overhead benefits in the compensation cost weights for each level of care.

However, after performing a detailed review of the calculation, we acknowledge that Medical Director contract labor costs would be reported in Worksheet A, column 2, line 15, which we do not include in the proposed compensation cost weight. In addition to Physician Administrative Services (line 15), we identified one additional overhead cost center where contract labor costs for patient care are reported and not reflected in the labor shares for each level of care: Nursing Administration (line 9). We believe these cost centers (Physician Administrative Services and Nursing Administration) are labor-intensive and vary with the local labor market and, thus, we believe contract labor costs for these services should be included in the labor shares for each level of care. Therefore, in response to public comment, we are revising our methodology for calculating overhead benefits attributable to each level of care. We are including in total facility overhead benefits those costs reported in Worksheet A, column 2, lines 9 and 15. A proportion of overhead benefit costs are allocated to each level of care using our methodology as stated above and in the proposed rule (86 FR 19718). This revision to our labor share methodology results in upward revisions to the proposed labor shares for each of the levels of care (between 0.6 percentage point and 1.1 percentage point). The labor shares showing the revised methodology are provided in Table 1.

f. Total Compensation Costs and Total Costs

To calculate the compensation costs for each provider, we proposed to then sum each of the costs estimated in steps (1) through (5) to derive total compensation costs for CHC, RHC, IRC, and GIP. We proposed that total costs for CHC are equal to Worksheet B, column 18, line 50, for RHC are equal to Worksheet B, column 18, line 51, for IRC are equal to Worksheet B, column 18, line 52, and for GIP are equal to Worksheet B, column 18, line 53.

3. Methodology for Deriving Compensation Cost Weights

To derive the compensation cost weights for each level of care, we first proposed to begin with a sample of providers who met new Level I edit conditions that required freestanding hospices to fill out certain parts of their cost reports effective for freestanding hospice cost reports with a reporting period that ended on or after December 31, 2017.^[3] Specifically, we required the following costs to be greater than zero: Fixed capital costs (Worksheet B, column 0, line 1), movable capital costs (Worksheet B, column 0, line 2), employee benefits (Worksheet B, column 0, line 3), administrative and general (Worksheet B, column 0, line 4), volunteer service coordination (Worksheet B, column 0, line 13), pharmacy and drugs charged to patients (sum of Worksheet B, column 0, line 14 and Worksheet A, column 7, line 42.50), registered nurse costs (Worksheet A, column 7, line 28), medical social service costs (Worksheet A, column 7, line 33), hospice aide and homemaker services costs (Worksheet A, column 7, line 37), and durable medical equipment (Worksheet A, column 7, line 38). Applying these Level I edits to the 2018 freestanding hospice MCRs resulted in 3,345 providers that passed the edits (four were excluded).

Then, for each level of care separately, we proposed to further trim the sample of MCRs. We outline our proposed trimming methodology using CHC as an example. Specifically, for CHC, we proposed that total CHC costs (Worksheet B, column 18, line 50) and CHC compensation costs to be greater than zero. We also proposed that CHC direct patient care salaries and contract labor costs per day is greater than 1. We also proposed to exclude those providers whose CHC compensation costs were greater than total CHC costs.

For the IRC and GIP compensation cost weights, we proposed to only use those MCRs from providers that provided inpatient services in their facility. Therefore, we proposed to exclude providers that reported costs greater than zero on Worksheet A-3, column 7, line 25 (Inpatient Care—Contracted) for IRC and Worksheet A-4, column 7, line 25 (Inpatient Care—Contracted) for GIP. The facilities that remained after this trim reported detailed direct patient care costs and other patient care costs for which we could then derive direct patient care salaries and other patient care salaries per the methodology described earlier.

Comment: One commenter stated that many of the hospice cost reports filed in 2018 failed to report contracted GIP days and contracted IRC care days on Worksheet S-1. Instead, they included all these days on line 23 and 33 of Worksheet S-1 but failed to report contracted days on line 40 and 41 of Worksheet S-1. The commenter stated that the failure to report contracted days on lines 40 and 41 would avoid a Level 1 edit if costs were not reported on Worksheets A-3 and A-4, line 25. The commenter stated that they understand that this reporting is inaccurate; however, there is no existing Level 1 edit that would catch it. The commenter questioned how CMS is determining that the inpatient costs are related solely to a freestanding inpatient unit on Worksheet A-4. The commenter claimed that if it is solely because no costs are reported on line 25, this assumption is in error. The commenter also claimed that if it is based on no days being reported as contracted on Worksheet S-1, this assumption is also in error. The commenter was concerned that costs—and accordingly labor component costs—are based on a small population with high risk of error.

One commenter stated that with only those cost reports from providers that have a hospice inpatient unit being used to determine the GIP and inpatient respite labor costs, they are concerned because one of their two affiliated hospices does have an inpatient unit, and yet they sometimes refer patients to contracted facilities for these levels of care as well. The commenter stated that it appears that the percentage of hospice cost reports used for determining GIP and respite total costs and labor-component costs is based on a small population of hospice providers with a significant risk of error; therefore, the commenter recommended that CMS rethink its approach for GIP and respite labor costs.

One commenter stated that their hospice utilizes general inpatient contracts, as they do not have our own facility. Thus, inpatient services on line 25 are not captured.

Response: We appreciate the commenters' concerns on the accuracy of the IRC and GIP cost data on the MCR. As stated in the FY 2022 Hospice proposed rule (86 FR 19718 through 19719) and above, for purposes of calculating the IRC and GIP compensation cost weights, we excluded providers that reported costs greater than zero on Worksheet A-3, column 7, line 25 (Inpatient Care—Contracted) for IRC and Worksheet A-4, column 7, line 25 (Inpatient Care—Contracted) for GIP. Then, for each level of care separately, we further trimmed the sample of cost reports. Specifically, for IRC, we required total IRC costs (Worksheet B, column 18, line 52) and IRC compensation costs to be greater than zero. We also required that IRC direct patient care salaries and contract labor costs per day would be greater than 1. We also excluded those providers whose IRC compensation costs were greater than total IRC costs. We then simultaneously removed those providers whose total IRC costs per day fall in the top and bottom one percent of total IRC costs per day for all IRC providers as well remove those providers whose compensation cost weight falls in the top and bottom five percent of compensation cost weights for all IRC providers.

We did not exclude providers based on the reporting of contracted inpatient days as reported on Worksheet S-1. In response to the public comment, we did test applying an additional edit that would exclude providers who reported contracted inpatient days on Worksheet S-1 as part of our basic trims. This excluded two providers and had no impact on the compensation cost weights for both IRC and GIP when rounded to a tenth of a percentage point. We encourage providers to report their cost report data accurately and timely.

Comment: Another specific concern stated by the commenters was that the determination of the labor share for GIP and IRC is based on Worksheet A-3 and A-4; however, any hospices reporting costs on line 25 (contracted services) were not included in the sample used for setting the labor share. The commenters recognize that the inclusion of any costs on line 25 would distort the labor component for these inpatient services; however, the commenters' experience indicates that most hospices with inpatient units also contract for some inpatient days with outside providers for a variety of reasons. The commenters stated that many of these hospices providers have some of the best accounting records in the industry and the proposed methodology for calculating the labor components eliminates the costs of these facilities from consideration. The commenters stated that the proposed rule indicates that 20 percent of IRC and 28 percent of GIP providers were included in the calculation. The commenters requested that CMS provide the final number of hospices with inpatient units that were used in the calculation of the labor components for both levels of care, and the total universe of IRC and GIP providers. One commenter also stated that they were interested in how the percentage of hospices that operate inpatient facilities can be increased and all costs, including contracted costs, can be included.

Response: The proposed hospice labor shares for the IRC level of care and GIC level of care (after trimming for outliers) is based on costs for 416 and 295 providers, respectively. These providers reflected approximately 53,000 IRP days of which about 47,000 were Medicare and approximately 136,000 GIC days of which about 108,000 were Medicare. Although this a smaller sample of providers than used for the other proposed labor shares for RHC (2,919 providers) and CHC (1,240 providers), we believe this is a technical improvement to the current labor shares that were primarily based on skilled nursing facility costs from the early 1980s. Our proposed methodology utilizes freestanding hospice cost report data reflecting the skilled hospice care provided in 2018 and the associated direct and indirect costs required to provide these services in 2018. We encourage all providers to report the cost report data accurately and timely so we can include more providers' cost report data in the labor share calculations. We will monitor the cost report data to determine whether the proposed updated labor shares are still appropriate.

Comment: Another specific concern raised by commenters was that the cost reports should be amended to allow for a greater breakdown of costs for contracted vs. hospice-administered inpatient services. Specifically, one commenter stated that when the cost report was revised in 2014, some industry experts recommended that CMS develop two separate worksheets for IRC and GIC. The first worksheet would represent costs associated with freestanding units operated by the hospice and the second worksheet would be for costs associated with contracted services. The commenter stated CMS should see value in potentially adding these worksheets if, in fact, it intends to calculate labor components for these levels of care based on cost report data going forward. The commenter also recommended that CMS could add a question to the cost report asking whether the hospice operates a freestanding inpatient and/or inpatient respite care facility. A “no” answer would require reporting contracted days and contracted costs or produce a Level 1 edit. The commenter stated that this would better allow CMS to isolate the costs of those facilities that truly operate an inpatient unit.

One commenter requested that CMS work with stakeholders and the hospice community to identify the best approaches, and separate worksheets, for GIP and inpatient respite costs, including both hospices that operate a freestanding facility and hospices that have contracted beds.

Response: We appreciate the commenters request for future changes to the hospice cost report to allow us to better isolate costs of those facilities that operate an inpatient unit. As stated above, we believe that our current method for calculating the IRC and GIP compensation cost weights provides an accurate measure of the labor shares for these levels of care. We will consider this comment when working on any future modifications to the hospice cost report. We will also continue to monitor the hospice labor shares as more recent data become available. We note that any future revisions to the hospice labor shares will be proposed and subject to public comments in future rulemaking.

Finally, as proposed, to derive the compensation cost weights for each level of care for each provider, we divide compensation costs for each level of care by total costs for each level of care. We then trim the data for each level of care separately to remove outliers. Following our example for CHC, we simultaneously remove those providers whose total CHC costs per day fall in the top and bottom one percent of total CHC costs per day for all CHC providers as well remove those providers whose compensation cost weight falls in the top and bottom five percent of compensation cost weights for all CHC providers. We then sum the CHC compensation costs and total CHC costs of the remaining providers, yielding a proposed compensation cost weight for CHC.

Since we limited our sample for IRC and GIP compensation cost weights to those hospices providing inpatient services in their facility, we conducted sensitivity analysis to test for the representative of this sample by reweighting compensation cost weights using data from the universe of freestanding providers that reported either IRC or GIP total costs. For example, we calculated reweighted compensation cost weights by ownership-type (proprietary, government and nonprofit), by size (based on RHC days) and by region. Our reweighted compensation cost weights for IRC and GIP were similar (less than one percentage point in absolute terms) to our proposed compensation cost weights for IRC and GIP (as shown in Table 1) and, therefore, we believe our sample is representative of freestanding hospices providing inpatient hospice care.

Comment: One commenter requested that clarification as to how CMS will adjust the labor share if certain types of hospices are found to provide more services and thus, likely have a larger labor share but contribute fewer cost reports.

Response: As described in the FY 2022 Hospice proposed rule (86 FR 17919) and above, the proposed compensation cost weights are equal to the sum of the compensation costs divided by the sum of the total costs for those remaining providers after trimming for outliers. Therefore, hospice providers with larger costs (reflecting larger utilization) would have a larger weight in the proposed labor shares. We would note that Medicare days, in aggregate, account for over 80 percent of total facility days. As stated previously, we will continue to monitor the labor shares over time and propose revisions to these shares to reflect a more recent cost structure and mix of providers.

Comment: One commenter stated that given the inherent differences in the provision of the hospice benefit between different types of hospice providers, they would recommend that CMS monitor any significant disparities in the distribution of labor and non-labor inputs across the hospice industry by program characteristics. The commenter stated that they would become concerned, for instance, if data indicates that some providers offer significantly fewer hours of professional interdisciplinary team (IDT) care yet make up a disproportionate percentage of providers filing cost reports. This could lead to unintended negative consequences for those providers fulfilling the true spirit and intent of the benefit. Put simply, if cost reports and other data indicate a widening gap in labor inputs between for-profit and not-for-profit providers, then CMS should investigate this trend further.

Response: We appreciate the commenter's concern regarding labor hours provided by type of facility. As we are able to obtain more recent cost report data, we will monitor the labor shares by ownership-type over time.

Comment: One commenter stated that if the labor shares are going to have a greater weight on CHC, hospices should be allowed to use it effectively. The commenter recommended that the current continuous care timeframe change from midnight to midnight to a new time frame of noon to noon and that visits from other providers such as chaplains and home health aides count toward the continuous care timeframe.

Response: While this comment is outside the scope of this rule as we did not make any proposals relating to our CHC policy, we thank the commenter for their recommendations and will take them under consideration for future rulemaking.

Final Decision: In summary, in response to public comments, we are adopting the revised hospice labor shares calculated as we proposed with a slight modification to the methodology to derive the overhead benefit calculations as described previously. Table 1 provides the finalized labor share for each level of care based on the compensation cost weights we derived using our revised methodology. As we proposed, the labor shares are rounded to three decimal places consistent with the labor shares used in other Prospective Payment Systems (PPS) (such as the inpatient prospective payment system (IPPS) and the Home Health Agency PPS). The revised labor shares will be implemented in a budget neutral manner through the use of labor share standardization factors.

We also received six comments on the use of the labor share standardization factor including hospices, national industry associations. A summary of these comments and our responses to those comments appear below:

Comment: A few commenters requested more information regarding the labor share standardization factor; specifically, its purpose, and any anticipated future use of the factor.

Response: The labor share standardization factor is applied to the FY 2022 hospice payment rates so that the aggregate payments do not increase or decrease due to changes in the labor share values. We proposed to implement the proposed hospice labor shares in a budget neutral manner which is consistent with our policy of implementing updates to the hospice wage index in a budget neutral manner as well as updates in other perspective payment systems such as the annual recalibration of the case-mix weights in home health and updates to the home health wage index. In order to calculate the labor share standardization factor, we simulate total payments using FY 2020 hospice utilization claims data with the FY 2022 hospice wage index and the current labor shares and compare it to our simulation of total payments using the FY 2022 hospice wage index with the final revised labor shares. By dividing total payments for each level of care (RHC days 1 through 60, RHC days 61+, CHC, IRC, and GIP) using the FY 2022 wage index, current labor shares and payment rates for each level of care by the total payments for each level of care using the final revised labor shares and FY 2022 wage index and payment.

Final Decision: We are finalizing the proposal to implement the hospice labor shares in a budget neutral manner through the use of the labor share standardization factors, so that the aggregate payments do not increase or decrease due to changes in the labor share values.

C. FY 2022 Hospice Wage Index and Rate Update

1. FY 2022 Hospice Wage Index

The hospice wage index is used to adjust payment rates for hospices under the Medicare program to reflect local differences in area wage levels, based on the location where services are furnished. The hospice wage index utilizes the wage adjustment factors used by the Secretary for purposes of section 1886(d)(3)(E) of the Act for hospital wage adjustments. Our regulations at § 418.306(c) require each labor market to be established using the most current hospital wage data available, including any changes made by the Office of Management and Budget (OMB) to the Metropolitan Statistical Areas (MSAs) definitions.

In general, OMB issues major revisions to statistical areas every 10 years, based on the results of the decennial census. However, OMB occasionally issues minor updates and revisions to statistical areas in the years between the decennial censuses. On March 6, 2020, OMB issued Bulletin No. 20-01, which provided updates to and superseded OMB Bulletin No. 18-04 that was issued on September 14, 2018. The attachments to OMB Bulletin No. 20-01 provided detailed information on the update to statistical areas since September 14, 2018, and were based on the application of the 2010 Standards for Delineating Metropolitan and Micropolitan Statistical Areas to Census Bureau population estimates for July 1, 2017 and July 1, 2018. (For a copy of this bulletin, we refer readers to the following website: https://www.whitehouse.gov/​wp-content/​uploads/​2020/​03/​Bulletin-20-01.pdf). In OMB Bulletin No. 20-01, OMB announced one new Micropolitan Statistical Area, one new component of an existing Combined Statistical Are and changes to New England City and Town Area (NECTA) delineations. In the FY 2021 Hospice Wage Index final rule (85 FR 47070) we stated that if appropriate, we would propose any updates from OMB Bulletin No. 20-01 in future rulemaking. After reviewing OMB Bulletin No. 20-01, we have determined that the changes in Bulletin 20-01 encompassed delineation changes that would not affect the Medicare wage index for FY 2022. Specifically, the updates consisted of changes to NECTA delineations and the redesignation of a single rural county into a newly created Micropolitan Statistical Area. The Medicare wage index does not utilize NECTA definitions, and, as most recently discussed in the FY 2021 Hospice Wage Index final rule (85 FR 47070), we include hospitals located in Micropolitan Statistical areas in each state's rural wage index. Therefore, while we proposed to adopt the updates set forth in OMB Bulletin No. 20-01 consistent with our longstanding policy of adopting OMB delineation updates, we note that specific wage index updates would not be necessary for FY 2022 as a result of adopting these OMB updates. In other words, these OMB updates would not affect any geographic areas for purposes of the wage index calculation for FY 2022.

In the FY 2020 Hospice Wage Index final rule (84 FR 38484), we finalized the proposal to use the current FY's hospital wage index data to calculate the hospice wage index values. In the FY 2021 Hospice Wage Index final rule (85 FR 47070), we finalized the proposal to adopt the revised OMB delineations with a 5 percent cap on wage index decreases, where the estimated reduction in a geographic area's wage index would be capped at 5 percent in FY 2021 and no cap would be applied to wage index decreases for the second year (FY 2022). For FY 2022, the final hospice wage index will be based on the FY 2022 hospital pre-floor, pre-reclassified wage index for hospital cost reporting periods beginning on or after October 1, 2017 and before October 1, 2018 (FY 2018 cost report data). The final FY 2022 hospice wage index will not include a cap on wage index decreases and would not take into account any geographic reclassification of hospitals, including those in accordance with section 1886(d)(8)(B) or 1886(d)(10) of the Act. The appropriate wage index value is applied to the labor portion of the hospice payment rate based on the geographic area in which the beneficiary resides when receiving RHC or CHC. The appropriate wage index value is applied to the labor portion of the payment rate based on the geographic location of the facility for beneficiaries receiving GIP or IRC.

In the FY 2006 Hospice Wage Index final rule (70 FR 45135), we adopted the policy that, for urban labor markets without a hospital from which hospital wage index data could be derived, all of the Core-Based Statistical Areas (CBSAs) within the state would be used to calculate a statewide urban average pre-floor, pre-reclassified hospital wage index value to use as a reasonable proxy for these areas. For FY 2022, the only CBSA without a hospital from which hospital wage data can be derived is 25980, Hinesville-Fort Stewart, Georgia. The FY 2022 final wage index value for Hinesville-Fort Stewart, Georgia is 0.8635.

There exist some geographic areas where there were no hospitals, and thus, no hospital wage data on which to base the calculation of the hospice wage index. In the FY 2008 Hospice Wage Index final rule (72 FR 50217 through 50218), we implemented a methodology to update the hospice wage index for rural areas without hospital wage data. In cases where there was a rural area without rural hospital wage data, we use the average pre-floor, pre-reclassified hospital wage index data from all contiguous CBSAs, to represent a reasonable proxy for the rural area. The term “contiguous” means sharing a border (72 FR 50217). Currently, the only rural area without a hospital from which hospital wage data could be derived is Puerto Rico. However, for rural Puerto Rico, we would not apply this methodology due to the distinct economic circumstances that exist there (for example, due to the close proximity to one another of almost all of Puerto Rico's various urban and non-urban areas, this methodology would produce a wage index for rural Puerto Rico that is higher than that in half of its urban areas); instead, we would continue to use the most recent wage index previously available for that area. For FY 2022, we proposed to continue to use the most recent pre-floor, pre-reclassified hospital wage index value available for Puerto Rico, which is 0.4047, subsequently adjusted by the hospice floor.

As described in the August 8, 1997 Hospice Wage Index final rule (62 FR 42860), the pre-floor and pre-reclassified hospital wage index is used as the raw wage index for the hospice benefit. These raw wage index values are subject to application of the hospice floor to compute the hospice wage index used to determine payments to hospices. As previously discussed, the adjusted pre-floor, pre-reclassified hospital wage index values below 0.8 will be further adjusted by a 15 percent increase subject to a maximum wage index value of 0.8. For example, if County A has a pre-floor, pre-reclassified hospital wage index value of 0.3994, we would multiply 0.3994 by 1.15, which equals 0.4593. Since 0.4593 is not greater than 0.8, then County A's hospice wage index would be 0.4593. In another example, if County B has a pre-floor, pre-reclassified hospital wage index value of 0.7440, we would multiply 0.7440 by 1.15, which equals 0.8556. Because 0.8556 is greater than 0.8, County B's hospice wage index would be 0.8.

The final hospice wage index applicable for FY 2022 (October 1, 2021 through September 30, 2022) is available on our website at: https://www.cms.gov/​Medicare/​Medicare-Fee-for-Service-Payment/​Hospice/​Hospice-Wage-Index.html.

We received seven comments on the proposed FY 2022 hospice wage index from various stakeholders including hospices, and national industry associations. A summary of these comments and our responses to those comments appear below:

Comment: One commenter expressed concern that hospices in Montgomery County, Maryland are at a long-term competitive disadvantage due to what they refer to as a Medicare hospice Federal payment inequity involving CBSAs specifically when Metropolitan Divisions are present. The commenter stated that that hospices in Montgomery County should be reimbursed at the same level as hospices in the Washington, DC area because Montgomery County has a similar cost of living and cost of doing business compared to Washington, DC and shares the same labor market when competing for labor. This commenter recommended several solutions to resolve this issue, including applying the outmigration hospital adjustment which is a hospital wage adjustment based on commuting patterns referenced in section 505 of the Medicare Prescription Drug, Improvement and Modernization Act of 2003 to the hospice wage index; allowing hospices serving patients in MSAs that are large enough to be subdivided into metropolitan divisions to opt for the higher wage index valuation within the MSA's respective CBSAs or providing a 1-year limited increase in hospice wage index payments in the Montgomery County Metropolitan Divisions as a short-term fix to this problem.

Response: We thank the commenter for these recommendations. However, we continue to believe that the OMB's geographic area delineations represent a useful proxy for differentiating between labor markets and that the geographic area delineations are appropriate for use in determining Medicare hospice payments. Additionally, we do not believe that we have the authority to apply the outmigration hospital adjustment to the hospice wage index because it is specific to the commuting patterns of hospital employees. We also do not believe it would be appropriate to allow hospices to opt for or be assigned a higher CBSA designation based on subdivided metropolitan divisions. Finally, in the FY 2021 Hospice Wage Index and Payment Rate Update final rule (85 FR 47079), we finalized a 1-year transition 5 percent cap on wage index decreases for fiscal year (FY) 2021 only. We believe that this transition was sufficient in order to mitigate the resulting short-term instability and negative impacts on certain providers after the implementation of the new OMB labor market delineations. We do not believe that a 1-year limited increase in hospice wage index payments for hospices specifically in the Montgomery County Metropolitan Divisions is appropriate at this time.

Based on the OMB's current delineations, Montgomery County belongs in a separate CBSA from the areas defined in the Washington-Arlington-Alexandria, DC-VA CBSA. Unlike inpatient prospective payment system (IPPS) hospitals, inpatient rehabilitation facilities (IRFs), and skilled nursing facilities (SNFs), where each provider uses a single CBSA, hospice agencies may be reimbursed based on more than one wage index. Payments are based upon the location of the beneficiary for routine and continuous home care or the location of the facility for respite and general inpatient care. Hospices in Montgomery County, Maryland may provide RHC and CHC to patients in the “Washington-Arlington-Alexandria, DC-VA” CBSA and to patients in the “Baltimore-Columbia-Towson, Maryland” CBSA. We have used CBSAs for determining hospice payments since FY 2006. Additionally, other provider types, such as IPPS hospitals, home health agencies (HHAs), SNFs, IRFs, and the dialysis facilities all use CBSAs to define their labor market areas. We believe that using the most current OMB delineations provides a more accurate representation of geographic variation in wage levels and do not believe it would be appropriate to allow hospices to be assigned a higher CBSA designation or to allow 1-year limited increase in hospice wage index payments for hospices only in the Montgomery County Metropolitan Divisions.

Comment: One commenter recommended CMS institute a policy that no hospice be paid below the rural floor for their state, allow hospices and other post-acute providers to utilize a reclassification board similar to hospitals, and consider working with the Congress on policies to reform the wage index such as revisiting MedPAC's 2007 proposal which recommended that the Congress repeal the existing hospital wage index statute, including reclassifications and exceptions, and give the Secretary authority to establish new wage index systems. In chapter 6 of the June 2007 Report to Congress, MedPAC recommended the new wage index should: Use wage data from all employers and industry-specific occupational weights, adjust for geographic differences in the ratio of benefits to wages, adjust at the county level and smooth large differences between counties, and be implemented so that large changes in wage index values are phased in over a transition period.^[4] Another commenter recommended that CMS develop and implement a wage index model that is consistent across all provider types so that all types of providers have a level playing field from which to compete for personnel.

Response: We appreciate the commenters' recommendations; however, these comments are outside the scope of the proposed rule. Any changes to the way we adjust hospice payments to account for geographic wage differences, beyond the wage index proposals discussed in the FY 2022 Hospice Wage Index and Rate Update proposed rule, would have to go through notice and comment rulemaking. While CMS and other stakeholders have explored potential alternatives to the current CBSA-based labor market system, no consensus has been achieved regarding how best to implement a replacement system.

Additionally, the regulations that govern hospice reimbursement do not provide a mechanism for allowing hospices to seek geographic reclassification or to utilize the rural floor provisions that exist for IPPS hospitals. The reclassification provision found in section 1886(d)(10) of the Act is specific to hospitals. Section 4410(a) of the Balanced Budget Act of 1997 (Pub. L. 105-33) provides that the area wage index applicable to any hospital that is located in an urban area of a state may not be less than the area wage index applicable to hospitals located in rural areas in that state. This rural floor provision is also specific to hospitals. Because the reclassification provision and the hospital rural floor applies only to hospitals, and not to hospices, we continue to believe the use of the pre-floor and pre-reclassified hospital wage index results in the most appropriate adjustment to the labor portion of the hospice payment rates. We remind stakeholders that the hospice wage index does include the hospice floor which is applicable to all CBSAs, both rural and urban. Pre-floor, pre-reclassified hospital wage index values below 0.8 are adjusted by a 15 percent increase subject to a maximum wage index value of 0.8.

Comment: A few commenters stated that providers should be protected against substantial payment reductions due to dramatic reductions in wage index values from one year to the next. One commenter recommended that CMS maintain the 5 percent cap that was put in place for FY 2021 or lower the cap to 3 percent to protect hospice providers who are already operating with negative or razor thin operating margins. Another commenter expressed concern regarding the adoption of the New Brunswick-Lakewood, NJ CBSA and recommended CMS adopt a transition policy that holds the FY 2022 and FY 2023 wage index for all affected facilities harmless from any reduction relative to their FY 2021 wage index.

Response: We appreciate the concerns sent in by the commenters regarding the impact of wages index changes from year to year as well as the concerns from providers who have been impacted by the implementation of the New Brunswick-Lakewood, NJ CBSA designation. While, we understand the commenters' concern regarding the potential financial impact, we believe that the OMB delineations for Metropolitan and Micropolitan Statistical Areas are appropriate for use in accounting for wage area differences and that the values computed under the delineations result in more appropriate payments to providers by more accurately accounting for and reflecting the differences in area wage levels. In the FY 2021 Hospice Wage Index and Payment Rate Update final rule (85 FR 47079), we finalized a 1-year transition for fiscal year (FY) 2021 only, to mitigate the resulting short-term instability and negative impacts on certain providers and to provide time for providers to adjust to their new labor market delineations. We believe that the 1-year 5 percent cap transitional policy provided for FY 2021 was an adequate safeguard against any significant payment reductions, allowed for sufficient time to make operational changes for future fiscal years, and provided a reasonable balance between mitigating some short-term instability in hospice payments and improving the accuracy of the payment adjustment for differences in area wage levels.

We note that certain changes to wage index policy may significantly affect Medicare payments. These changes may arise from revisions to the OMB delineations of statistical areas resulting from the decennial census data, periodic updates to the OMB delineations in the years between the decennial censuses, or other wage index policy changes. While we consider how best to address these potential scenarios in a consistent and thoughtful manner, we reiterate that our policy principles with regard to the wage index include generally using the most current data and information available and providing that data and information, as well as any approaches to addressing any significant effects on Medicare payments resulting from these potential scenarios, in notice and comment rulemaking.

Final Decision: We are finalizing our proposal to use the FY 2022 pre-floor, pre-reclassified hospital wage index data as the basis for the FY 2022 hospice wage index. The wage index applicable for FY 2022 is available on our website at https://www.cms.gov/​Medicare/​Medicare-Fee-for-Service-Payment/​Hospice/​Hospice-Wage-Index. The hospice wage index for FY 2022 is effective October 1, 2021 through September 30, 2022.

2. FY 2022 Hospice Payment Update Percentage

Section 4441(a) of the BBA (Pub. L. 105-33) amended section 1814(i)(1)(C)(ii)(VI) of the Act to establish updates to hospice rates for FYs 1998 through 2002. Hospice rates were to be updated by a factor equal to the inpatient hospital market basket percentage increase set out under section 1886(b)(3)(B)(iii) of the Act, minus 1 percentage point. Payment rates for FYs since 2002 have been updated according to section 1814(i)(1)(C)(ii)(VII) of the Act, which states that the update to the payment rates for subsequent FYs must be the inpatient market basket percentage increase for that FY. CMS currently uses 2014-based IPPS operating and capital market baskets to update the market basket percentage. In the FY 2022 IPPS proposed rule ^[5] we proposed to rebase and revise the IPPS market baskets to reflect a 2018 base year. We refer stakeholders to the FY 2022 IPPS proposed rule for further information (86 FR 25416 through 25428).

Section 3401(g) of the Affordable Care Act mandated that, starting with FY 2013 (and in subsequent FYs), the hospice payment update percentage would be annually reduced by changes in economy-wide productivity as specified in section 1886(b)(3)(B)(xi)(II) of the Act. The statute defines the productivity adjustment to be equal to the 10-year moving average of changes in annual economy-wide private nonfarm business multifactor productivity (MFP).

In the FY 2022 Hospice Wage Index and Payment Rate Update proposed rule (86 FR 19720), we proposed the market basket percentage increase of 2.5 percent for FY 2022 using the most current estimate of the inpatient hospital market basket (based on IHS Global Inc.'s fourth-quarter 2020 forecast with historical data through the third quarter 2020). Due to the requirements at sections 1886(b)(3)(B)(xi)(II) and 1814(i)(1)(C)(v) of the Act, the proposed inpatient hospital market basket update for FY 2022 of 2.5 percent was reduced by a productivity adjustment as mandated by Affordable Care Act (estimated in the proposed rule to be 0.2 percentage point for FY 2022). Therefore, the proposed hospice payment update percentage for FY 2022 was 2.3 percent.

We also stated if more recent data became available after the publication of the proposed rule and before the publication of the final rule (for example, more recent estimates of the inpatient hospital market basket update and/or productivity adjustment), we would use such data to determine the hospice payment update percentage for FY 2022 in the final rule. For this final rule, based on IHS Global Inc.'s (IGI) second quarter 2021 forecast with historical data through the first quarter 2021 of the inpatient hospital market basket update, the market basket percentage increase for FY 2022 is 2.7 percent. The productivity adjustment for FY 2022, based on IGI's second quarter 2021 forecast, is 0.7 percent. Therefore, the hospice payment update percentage for FY 2022, based on more recent data, is 2.0 percent.

Currently, the labor portion of the hospice payment rates are as follows: For RHC, 68.71 percent; for CHC, 68.71 percent; for GIP, 64.01 percent; and for IRC, 54.13 percent. As discussed in section III.B of this rule, we are finalizing to rebase and revise the labor shares for CHC, RHC, GIP and IRC using MCR data for freestanding hospices (CMS Form 1984-14, OMB Control Number 0938-0758) for 2018. We are finalizing the labor portion of the payment rates to be for CHC, 75.2 percent; for RHC, 66.0 percent; for GIP, 63.5 percent; and for IRC, 61.0 percent. The non-labor portion is equal to 100 percent minus the labor portion for each level of care. Therefore, we are finalizing the non-labor portion of the payment rates to be as follows: For CHC, 24.8 percent; RHC, 34 percent; for GIP, 36.5 percent; and For IRC, 39.0 percent.

Comment: We received seven comments in support of the proposed hospice update percentage of 2.3 percent. However, in its comment, MedPAC “concluded that the aggregate level of payments could be reduced and would still be sufficient to cover hospice providers' costs and preserve beneficiaries' access to care.” Therefore, MedPAC recommended a zero percent update for FY 2022 for all hospice providers.

Response: We appreciate the support from commenters as well as MedPAC's concerns. However, section 1814(i)(1)(C)(iii) of the Act requires the Secretary, for years subsequent to the first fiscal year in which payment revisions described in paragraph (6)(D) are implemented, to update the payment rates by the market basket percentage increase (as defined in section 1886(b)(3)(B)(iii)) of the Act for the fiscal year; section 1814(i)(1)(C)(iv)(I) of the Act requires that subsequent to such increase, the payment rates be reduced by the productivity adjustment described in section 1886(b)(3)(B)(xi)(II) of the Act.

Final Decision: We are finalizing the hospice payment update percentage of 2.0 percent for FY 2022. Based on IHS Global, Inc.'s more recent forecast of the inpatient hospital market basket update and the productivity adjustment, the hospice payment update percentage for FY 2022 will be 2.0 percent for hospices that submit the required quality data and 0.0 percent (FY 2022 hospice payment update of 2.0 percent minus 2.0 percentage points) for hospices that do not submit the required data.

3. FY 2022 Hospice Payment Rates

There are four payment categories that are distinguished by the location and intensity of the hospice services provided. The base payments are adjusted for geographic differences in wages by multiplying the labor share, which varies by category, of each base rate by the applicable hospice wage index. A hospice is paid the RHC rate for each day the beneficiary is enrolled in hospice, unless the hospice provides CHC, IRC, or GIP. CHC is provided during a period of patient crisis to maintain the patient at home; IRC is short-term care to allow the usual caregiver to rest and be relieved from caregiving; and GIP is to treat symptoms that cannot be managed in another setting.

As discussed in the FY 2016 Hospice Wage Index and Rate Update final rule (80 FR 47172), we implemented two different RHC payment rates, one RHC rate for the first 60 days and a second RHC rate for days 61 and beyond. In addition, in that final rule, we implemented a SIA payment for RHC when direct patient care is provided by an RN or social worker during the last 7 days of the beneficiary's life. The SIA payment is equal to the CHC hourly rate multiplied by the hours of nursing or social work provided (up to 4 hours total) that occurred on the day of service, if certain criteria are met. To maintain budget neutrality, as required under section 1814(i)(6)(D)(ii) of the Act, the new RHC rates were adjusted by a service intensity add-on budget neutrality factor (SBNF). The SBNF is used to reduce the overall RHC rate to ensure that SIA payments are budget-neutral. At the beginning of every fiscal year, SIA utilization is compared to the prior year in order calculate a budget neutrality adjustment.

In the FY 2017 Hospice Wage Index and Rate Update final rule (81 FR 52156), we initiated a policy of applying a wage index standardization factor to hospice payments to eliminate the aggregate effect of annual variations in hospital wage data. Typically, the wage index standardization factor is calculated using the most recent, complete hospice claims data available. However, due to the COVID-19 PHE, we looked at using the previous fiscal year's hospice claims data (FY 2019) to determine if there were significant differences between utilizing 2019 and 2020 claims data. The difference between using FY 2019 and FY 2020 hospice claims data was minimal. Therefore, we will continue our practice of using the most recent, complete hospice claims data available; that is, we used FY 2020 claims data for the FY 2022 payment rate updates.

To calculate the wage index standardization factor, we simulate total payments using FY 2020 hospice utilization claims data with the FY 2021 wage index (pre-floor, pre-reclassified hospital wage index with the hospice floor, and a 5 percent cap on wage index decreases) and FY 2021 payment rates (that include the current labor shares) and compare it to our simulation of total payments using the FY 2022 hospice wage index (with hospice floor, without the 5 percent cap on wage index decreases) and FY 2021 payment rates (that include the current labor shares). By dividing total payments for each level of care (RHC days 1 through 60, RHC days 61+, CHC, IRC, and GIP) using the FY 2021 wage index and payment rates for each level of care by the total payments using the FY 2022 wage index and FY 2021 payment rates, we obtain a wage index standardization factor for each level of care. As stated above, in order to calculate the labor share standardization factor, we simulate total payments using FY 2020 hospice utilization claims data with the FY 2022 hospice wage index and the current labor shares and compare it to our simulation of total payments using the FY 2022 hospice wage index with the final revised labor shares. By dividing total payments for each level of care (RHC days 1 through 60, RHC days 61+, CHC, IRC, and GIP) using the current labor shares and FY 2022 wage index and payment rates for each level of care by the total payments for each level of care using the final revised labor shares and FY 2022 wage index and payment rates for each level of care, we obtain a labor share standardization factor for each level of care. The wage index and labor share standardization factors for each level of care are shown in the Tables 2 and 3.

The FY 2022 RHC rates are shown in Table 2. The FY 2022 payment rates for CHC, IRC, and GIP are shown in Table 3.

Sections 1814(i)(5)(A) through (C) of the Act require that hospices submit quality data, based on measures to be specified by the Secretary. In the FY 2012 Hospice Wage Index and Rate Update final rule (76 FR 47320 through 47324), we implemented a HQRP as required by those sections. Hospices were required to begin collecting quality data in October 2012, and submit that quality data in 2013. Section 1814(i)(5)(A)(i) of the Act requires that beginning with FY 2014 and each subsequent FY, the Secretary shall reduce the market basket update by 2 percentage points for any hospice that does not comply with the quality data submission requirements with respect to that FY. The FY 2022 rates for hospices that do not submit the required quality data would be updated by the FY 2022 hospice payment update percentage of 2.0 percent minus 2 percentage points. These rates are shown in Tables 4 and 5.

Final Decision: We are implementing the updates to hospice payment rates as discussed in the proposed rule.

4. Hospice Cap Amount for FY 2022

As discussed in the FY 2016 Hospice Wage Index and Rate Update final rule (80 FR 47183), we implemented changes mandated by the IMPACT Act of 2014 (Pub. L. 113-185). Specifically, the IMPACT Act requires that, for accounting years that end after September 30, 2016 and before October 1, 2025, the hospice cap be updated by the hospice payment update percentage rather than using the CPI-U. Division CC, section 404 of the CAA 2021 has extended the accounting years impacted by the adjustment made to the hospice cap calculation until 2030. Therefore, for accounting years that end after September 30, 2016 and before October 1, 2030, the hospice cap amount is updated by the hospice payment update percentage rather than using the CPI-U. As a result of the changes mandated by Division CC, section 404 of the CAA 2021, we proposed conforming regulation text changes at § 418.309 to reflect the new language added to section 1814(i)(2)(B) of the Act.

The hospice cap amount for the FY 2022 cap year will be $31,297.61, which is equal to the FY 2021 cap amount ($30,683.93) updated by the FY 2022 hospice payment update percentage of 2.0 percent.

Comment: Generally, commenters supported the update to the cap amount. We received a comment indicating some hospice agencies never hit the cap amount and recommend for CMS to utilize available claims and quality data to target hospices with questionable practices to avoid exceeding the cap amount.

Response: We appreciate the concern and recommendation. We encourage those who have concerns about fraud, waste, or abuse to report these to CMS Center for Program Integrity. Resources can be found at https://www.cms.gov/​About-CMS/​Components/​CPI.

Comment: MedPAC recommended the hospice cap amount be reduced by 20 percent as a way to focus payment reductions on providers with particularly high margins. MedPAC also recommended wage adjusting the hospice cap amount to make it more equitable across providers.

Response: We appreciate MedPAC's comments; however, we are required by law to update the hospice cap amount from the preceding year by the hospice payment update percentage, in accordance with section 1814(i)(2)(B)(ii) of the Act. Therefore, we do not have the statutory authority to reduce the aggregate cap amount nor the statutory authority to wage-adjust the cap amount.

Final Decision: We are finalizing the update to the hospice cap amount for FY 2022 in accordance with statutorily-mandated requirements as well as the conforming regulation text changes at § 418.309.

D. Clarifying Regulation Text Changes for the Hospice Election Statement Addendum

In the FY 2020 Hospice Wage Index and Payment Rate Update final rule (84 FR 38484), we finalized modifications to the hospice election statement content requirements at § 418.24(b) to increase coverage transparency for patients under a hospice election. These changes included a new condition for payment requiring a hospice, upon request, to provide the beneficiary (or representative) an election statement addendum (hereafter called “the addendum”) outlining the items, services, and drugs that the hospice has determined are unrelated to the terminal illness and related conditions. We stated in the final rule that the addendum is intended to complement the Hospice Conditions of Participation (CoPs) at § 418.52(c)(7) and (8), which require hospices to verbally inform beneficiaries, at the time of hospice election, of the services covered under the Medicare hospice benefit, as well as the limitations of such services (84 FR 38509). The requirements at §§ 418.24(b) and 418.52(a) ensure that beneficiaries are aware of any items, services, or drugs they would have to seek outside of the benefit, as well as their potential out-of-pocket costs for hospice care, such as co-payments and/or coinsurance.

Section 418.24(c) sets forth the elements that must be included on the addendum:

1. The addendum must be titled “Patient Notification of Hospice Non-Covered Items, Services, and Drugs”;

2. Name of the hospice;

3. Beneficiary's name and hospice medical record identifier;

4. Identification of the beneficiary's terminal illness and related conditions;

5. A list of the beneficiary's current diagnoses/conditions present on hospice admission (or upon plan of care update, as applicable) and the associated items, services, and drugs, not covered by the hospice because they have been determined by the hospice to be unrelated to the terminal illness and related conditions;

6. A written clinical explanation, in language the beneficiary and his or her representative can understand, as to why the identified conditions, items, services, and drugs are considered unrelated to the terminal illness and related conditions and not needed for pain or symptom management. This clinical explanation must be accompanied by a general statement that the decision as to what conditions, items, services, or drugs are unrelated is made for each individual patient, and that the beneficiary should share this clinical explanation with other health care providers from which he or she seeks services unrelated to his or her terminal illness and related conditions;

7. References to any relevant clinical practice, policy, or coverage guidelines;

8. Information on the following:

a. Purpose of the addendum

b. patient's right to immediate advocacy

9. Name and signature of the Medicare hospice beneficiary (or representative) and date signed, along with a statement that signing this addendum (or its updates) is only acknowledgement of receipt of the addendum (or its updates) and not necessarily the beneficiary's agreement with the hospice's determinations.

The hospice is required to furnish the addendum in writing in an accessible format, so the beneficiary (or representative) can understand the information provided, make treatment decisions based on that information, and share such information with non-hospice providers rendering un-related items and services to the beneficiary. Therefore, the format of the addendum must be usable for the beneficiary and/or representative. Although we stated in the FY 2020 Hospice Wage Index and Payment Rate Update that hospices may develop their own election statement addendum (84 FR 38507), we posted a modified model election statement and addendum on the Hospice web page,^[6] along with the publication of the FY 2021 Hospice Wage Index and Payment Rate Update final rule (85 FR 47070). The intent was to provide an illustrative example so hospices can modify and develop their own forms to meet the content requirements. In the FY 2021 Hospice Wage Index and Payment Rate Update final rule, we stated that most often we would expect the addendum would be in a hard copy format the beneficiary or representative can keep for his or her own records, similar to how hospices are required by the hospice CoPs at § 418.52(a)(1) to provide the individual a copy of the notice of patient rights and responsibilities (85 FR 47091). The hospice CoPs at § 418.104(a)(2) state that the patient's record must include “signed copies of the notice of patient rights in accordance with § 418.52.” Likewise, since the addendum is part of the election statement as set forth in § 418.24(b)(6), then it is required to be part of the patient's record (if requested by the beneficiary or representative). The signed addendum is only acknowledgement of the beneficiary's (or representative's) receipt of the addendum (or its updates) and the payment requirement is considered met if there is a signed addendum (and any signed updates) in the requesting beneficiary's medical record with the hospice. We believe that a signed addendum indicates the hospice discussed the addendum and its contents with the beneficiary (or representative). Additionally, in the event that a beneficiary (or representative) does not request the addendum, we expect hospices to document, in some fashion, that an addendum has been discussed with the patient (or representative) at the time of election, similar to how other patient and family discussions are documented in the hospice's clinical record. It is necessary for the hospice to document that the addendum was discussed and whether or not it was requested, in order to prevent potential claims denials related to any absence of an addendum (or addendum updates) in the medical record.

Though we did not propose any changes to the election statement addendum content requirements at § 418.24(c), or the October 1, 2020 effective date, in the FY 2021 Hospice Wage Index and Payment Rate Update proposed rule, we solicited comments on the usefulness of the modified model election statement and addendum posted on the Hospice Center web page (85 FR 20949). In the FY 2021 Hospice Wage Index and Payment Rate Update final rule (85 FR 47093), we responded to comments received, and stated that, as finalized in the FY 2020 Hospice Wage Index and Payment Rate Update final rule, the hospice election statement addendum will remain a condition for payment that is met when there is a signed addendum (and its updates) in the beneficiary's hospice medical record.

Since its implementation on October 1, 2020, CMS has received additional inquiries from stakeholders asking for clarification on certain aspects of the addendum. We appreciate and understand the importance of provider input and involvement in ensuring that this document is effective in increasing coverage transparency for beneficiaries. Therefore, in the FY 2022 proposed rule (86 FR 19724) we provided clarification on, and proposed modifications to, certain signature and timing requirements and proposed corresponding clarifying regulations text changes.

Currently the regulations at § 418.24(c) require that if a beneficiary or his or her representative requests the addendum at the time of the initial hospice election (that is, at the time of admission to hospice), the hospice must provide this information, in writing, to the individual (or representative) within 5 days from the date of the election. In the FY 2022 hospice proposed rule, we noted that hospices have reported that beneficiaries or representatives sometimes do not request the addendum at the time of election, but rather within the 5 days after the effective date of the election (86 FR 19724). In these situations, the regulations require the hospice to provide the addendum within 3 days, as the beneficiary requested the addendum during the course of care. However, in accordance with § 418.54(b), the hospice IDG, in consultation with the individual's attending physician (if any), must complete the hospice comprehensive assessment no later than 5 calendar days after the election of hospice care. We stated that in some instances, this may mean that the hospice must furnish the addendum prior to completion of the comprehensive assessment. The comprehensive assessment includes all areas of hospice care related to the palliation and management of a beneficiary's terminal illness. This assessment is necessary because it provides an overview of the items, services and drugs that the patient is already utilizing as well as helps determine what the hospice may need to add in order to treat the patient throughout the dying process. If the addendum is completed prior to the comprehensive assessment, the hospice may not have a complete patient profile, which could potentially result in the hospice incorrectly anticipating the extent of covered and non-covered services and lead to an inaccurate election statement addendum. Hospice providers are only able to discern what items, services, and drugs they will not cover once they have a beneficiary's comprehensive assessment. We proposed allowing the hospice to furnish the addendum within 5 days from the date of a beneficiary or representative request, if the request is within 5 days from the date of a hospice election. For example, if the patient elects hospice on December 1st and requests the addendum on December 3rd, the hospice would have until December 8th to furnish the addendum.

Additionally, we acknowledged that hospices have noted that there is not a timeframe in regulations regarding the patient signature on the addendum. Section 418.24(c)(9) requires the beneficiary's signature (or his/her representative's signature) as well as the date the document was signed. We noted in the FY 2021 Hospice Wage Index & Payment Rate Update final rule that because the beneficiary signature is an acknowledgement of receipt of the addendum, this means the beneficiary would sign the addendum when the hospice provides it, in writing, to the beneficiary or representative (85 FR 47092). Obtaining the required signatures on the election statement has been a longstanding regulatory requirement. Therefore, we stated that we expect that hospices already have processes and procedures in place to ensure that required signatures are obtained, either from the beneficiary, or from the representative in the event the beneficiary is unable to sign, and we anticipate that hospices would use the same procedures for obtaining signatures on the addendum. We did note that we understand that some beneficiaries or representatives may request an emailed addendum or request more time to review the addendum before signing, in which case the date that the hospice furnished the addendum to the beneficiary (or representative) may differ from the date that the beneficiary or representative signs the addendum. This means the hospice may furnish the addendum within the required timeframe; however, the signature date may be beyond the required timeframe. Therefore, we proposed to clarify in regulation that the “date furnished” must be within the required timeframe (that is, 3 or 5 days of the beneficiary or representative request, depending on when such request was made), rather than the signature date. At § 418.24(c)(10), we proposed that the hospice would include the “date furnished” in the patient's medical record and on the addendum itself.

In the FY 2021 Hospice Wage Index and Payment Rate Update final rule, we addressed a concern regarding a potential situation wherein the beneficiary or representative refuses to sign the addendum (85 FR 47088). We reiterated that the signature on the addendum is only acknowledgement of receipt and not a tacit indication of agreement with its contents, and that we expect the hospice to inform the beneficiary of the purpose of the addendum and rationale for the signature. However, we recognized that there might be rare instances in which the beneficiary (or representative) refuses to sign the addendum, and noted that we would consider whether this issue would require future rulemaking. In the proposed rule, we stated that we have subsequently received this question from stakeholders post implementation, and therefore, clarified that if a patient or representative refuses to sign the addendum, the hospice must document clearly in the medical record (and on the addendum itself) the reason the addendum is not signed in order to mitigate a claims denial for this condition for payment. We stated that in such a case, although the beneficiary has refused to sign the addendum, the “date furnished” must still be within the required timeframe (that is, within 3 or 5 days of the beneficiary or representative request, depending on when such request was made), and noted in the chart and on the addendum itself (86 FR 19725).

We also noted that stakeholders again requested that CMS clarify whether a non-hospice provider is required to sign the addendum in the event that the non-hospice provider requests the addendum rather than the beneficiary or representative. We reiterated that if only a non-hospice provider or Medicare contractor requests the addendum (and not the beneficiary or representative) we would not expect a signed copy in the patient's medical record. We stated that hospices can develop processes (including how to document such requests from non-hospice providers and Medicare contractors) to address circumstances in which the non-hospice provider or Medicare contractor requests the addendum, and the beneficiary or representative does not (86 FR 19725). As such, we proposed to clarify in regulation that if a non-hospice provider requests the addendum, rather than the beneficiary or representative, the non-hospice provider is not required to sign the addendum.

We also discussed that there may be instances in which the beneficiary or representative requests the addendum and the beneficiary dies, revokes, or is discharged prior to signing the addendum (86 FR 19725). While we stated in the FY 2020 Hospice Wage Index and Payment Rate Update final rule, that if the beneficiary requests the election statement addendum at the time of hospice election but dies within 5 days, the hospice would not be required to furnish the addendum as the requirement would be deemed as being met in this circumstance (84 FR 38521), this policy was not codified in regulation. Therefore, we proposed conforming regulations text changes at § 418.24(c) to reflect this policy. Furthermore, we proposed to clarify at § 418.24(d)(4) that if the patient dies, revokes election, or is discharged within the required timeframe (3 or 5 days after a request, depending upon when such request was made), but the hospice has not yet furnished the addendum, the hospice is not required to furnish the addendum. Similarly, we proposed to clarify at § 418.24(d)(5) that in the event that a beneficiary requests the addendum and the hospice furnishes the addendum within 3 or 5 days (depending upon when the request for the addendum was made), but the beneficiary dies, revokes, or is discharged prior to signing the addendum, a signature from the individual (or representative) is no longer required. We stated that we would continue to expect that the hospice would note the “date furnished” in the patient's medical record and on the addendum, if the hospice has already completed the addendum, as well as an explanation in the patient's medical record noting that the patient died, revoked, or was discharged prior to signing the addendum (86 FR 19725).

Finally, we proposed conforming regulations text changes at § 418.24(c) in alignment with subregulatory guidance indicating that hospices have “3 days,” rather than “72 hours” to meet the requirement when a patient requests the addendum during the course of a hospice election. We proposed that hospices must furnish the addendum no later than 3 calendar days after a beneficiary's (or representative's) request during the course of a hospice election. This means that hospice providers must furnish the addendum to the beneficiary or representative on or before the third day after the date of the request. For example, if a beneficiary (or representative) requests the addendum on February 22nd, then the hospice will have until February 25th to furnish the addendum, regardless of what time the addendum was requested on February 22nd. The intent of this clarification is to better align with the requirement for furnishing an election statement addendum when the addendum is requested within 5 days of the date of election, which also uses “days” rather than “hours”.

Thirty-one unique stakeholders submitted their comments on the proposed clarifications to the election statement addendum. A few commenters requested additional clarification on certain topics and offered recommendations for the election statement addendum. These comments along with our responses are summarized below.

Comment: The majority of commenters supported the clarifications and proposed regulation text changes regarding the election statement addendum. Commenters thanked CMS for these regulatory changes, stating that these clarifications will facilitate administration of the addendum and reduce hospice burden.

Response: We thank commenters for their feedback.

Comment: Some commenters recommended that the timeframe to furnish the addendum to the beneficiary (or representative) when requested after the first 5 days of a hospice election be changed from 3 days to 5 days. Other commenters recommended that CMS change the requirement from 3 calendar days to 3 business days. One commenter requested clarification that the day of request is considered day zero. Another commenter mentioned that providing the addendum within 3 days is burdensome to beneficiaries (or representatives), because they felt pressured to meet with hospice staff to provide their signature for the requested addendum.

Response: We did not propose to change the timeline for furnishing the addendum when a beneficiary requests the addendum during the course of a hospice election (that is, after the first five days of a hospice election date), and we continue to believe that 3 days is an adequate amount of time for the hospice to furnish the addendum. As we stated in the FY 2020 hospice final rule, because the hospice has already completed the comprehensive assessment and has begun providing care, we believe that this represents a sufficient timeframe for reviewing the patient record and completing the addendum if this information is requested during the course of hospice care (84 FR 38511).

Additionally, as the plan of care should identify the conditions or symptoms that the hospice determines to be “unrelated,” this information should be readily accessible to the hospice in order to allow for the timely completion of the addendum. Hospices should update the addendum to include such conditions, items, services, and drugs they determine to be unrelated throughout the course of a hospice election. Hospices are able to create their own process when it comes to updating and providing the requested addendum to the beneficiary (or representative). Furthermore, we believe 3 calendar days, rather than 3 business days continues to be appropriate, as hospice care is provided around the clock rather than only during business days and hours.

In the proposed rule, we provided an example acknowledging the day of the request as day zero. We stated that when the request is within 5 days from the date of a hospice election, and the patient elects hospice on December 1st and requests the addendum on December 3rd, the hospice would have until December 8th to furnish the addendum (86 FR 19724), making December 1st as day zero in this example. Moreover, because we proposed to change the timeframe requirements to correspond with the “date furnished” rather than the “signature date,” we disagree that this timeframe would be burdensome to beneficiaries. We noted in the FY 2021 Hospice Wage Index & Payment Rate Update final rule that because the beneficiary signature is an acknowledgement of receipt of the addendum, this means the beneficiary would sign the addendum when the hospice provides it, in writing, to the beneficiary or representative (85 FR 47092). Obtaining the required signatures on the election statement has been a longstanding regulatory requirement (84 FR 38484); however, we did acknowledge in the proposed rule that there may be time constraints and/or circumstances that would prevent a beneficiary from signing and returning the addendum to the hospice by a specified deadline. We proposed to require that the “date furnished” be within the required timeframe, rather than the signature date, to mitigate any undue strain on the beneficiary or representative in returning the addendum to the hospice by a specified date.

Comment: Some commenters expressed concern that the request from a non-hospice provider for the election statement addendum does not require a signature. Commenters stated that hospices would have no proof that the addendum was provided to the non-hospice provider without the provider's signature.

Response: If a non-hospice provider requests the addendum, the hospice must furnish the addendum, however, the non-hospice provider is not required to sign the addendum. We remind commenters that the intent of the addendum is to ensure that hospice beneficiaries and their representatives are fully informed of any items or services for which they must assume financial responsibility. Consequently, if only a non-hospice provider or Medicare contractor request the addendum (and not the beneficiary or representative) CMS would not expect a signed copy in the patient's medical record. Hospices can develop processes (including how to document such requests from non-hospice providers and Medicare contractors) to address circumstances in which the non-hospice provider or Medicare contractor requests the addendum, and the beneficiary or representative does not, as a means of demonstrating that the addendum was furnished to a non-hospice provider and/or Medicare contractor upon request.

Comment: A commenter asked CMS to define whether or not a mailed copy of the form would be acceptable. The commenter stated that they believe their patients and their representatives would welcome this option; however, it is unclear whether mailing the form is acceptable for CMS.

Response: There is nothing precluding hospices from furnishing an addendum through mail. We expect that hospices would take steps in working with patients and their representatives to better understand which methods (that is, in person, mail, etc.) of delivery would work best in furnishing the addendum. Some beneficiaries or representatives may have time constraints that prevent them from signing and returning the addendum by a certain deadline, in which case, the date that the hospice furnishes the addendum to the beneficiary may differ from the date that the beneficiary (or representative) signs the addendum. Hospices would need to make sure the “date furnished' on the addendum is within the required timeframe (3 or 5 days, depending upon when the request was made). Furthermore, we expect that hospices will have processes in place when they are obtaining a signed addendum from a beneficiary or representative.

Comment: Many commenters requested making the proposed clarifications to the hospice election statement addendum retroactive to the implementation date of October 1, 2020. One commenter requested delaying the effective date of the proposed clarification for the hospice election statement addendum to provide time for software updates in addition to reporting and system alerts.

Response: We do not believe that making these clarifications retroactive or delaying the effective date is necessary. To date we have not received reports of claims denials resulting from the implementation of the election statement addendum and the current regulations at § 418.24. Furthermore, many of these clarifying regulations text changes have been previously addressed in sub-regulatory guidance. As such, the implementation of these clarifications on October 1, 2021 would not cause a burden for software updates.

Comment: Many commenters encouraged CMS to update the model hospice election statement addendum on the CMS hospice center web page to illustrate these clarifications.

Response: We will post an updated model election statement addendum on the Hospice web page,^[7] along with the publication of this FY 2022 Hospice Wage Index and Payment Rate Update final rule. This is an illustrative example for hospices to modify and develop their own forms that meet the content requirements at § 418.24.

Comment: Some commenters stated that it is redundant to require the hospice to note on the addendum and in the medical record the reason that a beneficiary did not provide their signature.

Response: We recognize the commenters' concerns and agree that it is appropriate for the hospice to document only on the addendum itself the reason that an addendum is un-signed. This could include not only a beneficiary refusing to sign, but also death, discharge, or revocation prior to the hospice obtaining the signature. However, while a hospice can choose to document the reason for an unsigned addendum in the medical record, as well as on the addendum, it is not required.

Comment: Many commenters offered suggestions regarding additional aspects of the election statement addendum for which we did not propose clarifying changes. Some commenters recommended that CMS align the late penalty for the addendum with the penalty for late submission of the NOE. Other commenters stated that denying the whole hospice claim when the addendum is furnished late is excessive. A commenter stated that as currently structured, the penalty is a negative incentive to furnish the addendum in a timely manner if a hospice misses the initial required timeframe. Some commenters mentioned there was confusion regarding billing when an addendum is furnished late. Other commenters recommended using a code to indicate billed but not covered hospice days when the addendum is furnished late. A few commenters stated they believe the addendum and the ABN have the potential to decrease transparency and increase confusion for hospice patients, whereas, other commenters recommended expanding the usage of the addendum, which included combining the ABN and addendum, and to include drugs or services which the hospice has determined to be medically unreasonable or no longer necessary. One commenter recommended that CMS explore ways to educate hospice providers about how they can inform their beneficiaries (or representative) when items, services, or drugs are considered related, but non-covered due to reasons such as not reasonable or necessary for the palliation and management of the terminal illness and related conditions. Moreover, a commenter recommended developing an exceptions process for when hospice providers are unable to provide an addendum because of `exceptional circumstances' that are beyond the control of the hospice. Lastly, one commenter suggested that since an electronically sent addendum could be tracked, a signature should not be required.

Response: While these comments are out of scope of the proposed rule, we appreciate and welcome all feedback related to the late penalty; ABN and expansion of the addendum; signatures; exceptional circumstances; and educating hospice providers. While we did not propose any of these recommendations we could consider them for future rulemaking. We understand the possibility of conflating the differences between the ABN and the hospice election statement addendum. The ABN transfers potential financial liability to the Medicare beneficiary in certain instances, whereas the addendum (upon request) informs terminally ill beneficiaries (or their representative) only of items, services, or drugs the hospice will not be providing because the hospice has determined them to be unrelated to the terminal illness and related conditions. We refer readers to FY 2020 Hospice Wage Index and Payment Rate Update final rule (84 FR 38512) to learn more about the usage of the ABN. The hospice CoPs at § 418.56(b) require hospices to educate each patient and their primary caregivers(s) on services identified on the plan of care and document the patient's (or representative's) level of understanding involvement and agreement with the plan of care. We expect that hospices would use the same methods when educating patients (or representatives) about the addendum and non-covered items, services and drugs, which the hospice has determined are not reasonable or necessary for the palliation and management of the terminal illness and related conditions.

The hospice CoPs at §  418.52(a)(1) require that in advance of receiving care, patients are informed about their rights, and hospices must provide the patient (or representative) with verbal and written notice of the patient's rights and responsibilities in a language and manner the patient understands. Likewise, the hospice CoPs at §  418.52(a)(3) requires that hospices obtain the patient's or representative's signature confirming that he or she has received a copy of the notice of rights and responsibilities. So, it is not unreasonable to require that the electronically sent addendum also be signed to ensure that the patient is aware of the important information about hospice non-covered items, services, and drugs. We do not have a policy for `exceptional circumstances' (that is floods, hurricanes, etc.) but we will consider addressing this policy in future rulemaking.

Final Decision: We are finalizing the clarifications and addendum regulation text changes at § 418.24(c) as proposed, with the exception of requiring the reason that the addendum is not signed to be documented in the patient's medical record. This explanation must be clearly noted on the addendum itself, but is not required to be documented in both places. Based on comments, we are amending the regulation text at § 418.24 to state that if the beneficiary dies, revokes election, is discharged prior to signing the addendum, or refuses to sign the addendum, the addendum would not be required to be signed in order for the hospice to receive payment. The hospice must note (on the addendum itself) the reason the addendum was not signed and the addendum would become part of the patient's medical record. These changes will be effective on October 1, 2021.

E. Hospice Waivers Made Permanent Conditions of Participation

1. Background

In order to support provider and supplier communities due to the COVID-19 PHE, CMS has issued an unprecedented number of regulatory waivers under our statutory authority set forth at section 1135 of the Act. Under section 1135 of the Act, the Secretary may temporarily waive or modify certain Medicare, Medicaid, and Children's Health Insurance Program (CHIP) requirements to ensure that sufficient health care items and services are available to meet the needs of individuals enrolled in the programs in the emergency area and time periods, and that providers who furnish such services in good faith, but who are unable to comply with one or more requirements as described under section 1135(b) of the Act, can be reimbursed and exempted from sanctions for violations of waived provisions (absent any determination of fraud or abuse). The intent of these waivers was to expand healthcare system capacity while continuing to maintain public and patient safety, and to hold harmless providers and suppliers unable to comply with existing regulations after a good faith effort.

While some of these waivers simply delay certain administrative deadlines, others directly affect the provision of patient care. The utilization and application of these waivers pushed us to consider whether permanent changes would be beneficial to patients, providers, and professionals. We identified selected waivers as appropriate candidates for formal regulatory changes. Those changes and their respective histories and background information are discussed in the rule. We are also finalizing regulatory changes that are not directly related to PHE waivers that will clarify or align some policies that have been raised as concerns by stakeholders.

We are finalizing the following revisions to the hospice CoPs.

2. Hospice Aide Training and Evaluation—Using Pseudo-Patients

Hospice aides deliver a significant portion of direct care. Aides are usually trained by an employer, such as a hospice, HHA or nursing home and may already be certified as an aide prior to being hired. The competency of new aides must be evaluated by the hospice to ensure appropriate care can be provided by the aide. Aide competency evaluations should be conducted in a way that identifies and meets training needs of the aide as well as the patient's needs. These evaluations are a critical part of providing safe, quality care.

The current hospice aide competency standard regulations at § 418.76(c)(1) requires the aide to be evaluated by observing an aide's performance of the task with a patient. We are finalizing similar changes to hospice aide competency standards to those already made with respect to HHAs (see § 484.80(c)) in our hospice regulations at § 418.76(c)(1)). Additionally, we are finalizing definitions for both “pseudo-patient” and “simulation” at § 418.3. Therefore, we are finalizing changes to permit skill competencies to be assessed by observing an aide performing the skill with either a patient or a pseudo-patient as part of a simulation. The final definitions are as follows:

• “Pseudo-patient” means a person trained to participate in a role-play situation, or a computer-based mannequin device. A pseudo-patient must be capable of responding to and interacting with the hospice aide trainee, and must demonstrate the general characteristics of the primary patient population served by the hospice in key areas such as age, frailty, functional status, cognitive status and care goals.
• “Simulation” means a training and assessment technique that mimics the reality of the homecare environment, including environmental distractions and constraints that evoke or replicate substantial aspects of the real world in a fully interactive fashion, in order to teach and assess proficiency in performing skills, and to promote decision making and critical thinking.

These changes will allow hospices to utilize pseudo-patients, such as a person trained to participate in a role-play situation or a computer-based mannequin device, instead of actual patients, in the competency testing of hospice aides for those tasks that must be observed being performed on a patient. This could increase the speed of performing competency testing and would allow new aides to begin serving patients more quickly while still protecting patient health and safety.

3. Hospice Aide Training and Evaluation—Targeting Correction of Deficiencies

We are also amending the requirement at § 418.76(h)(1)(iii) to specify that if an area of concern is verified by the hospice during the on-site visit, then the hospice must conduct, and the hospice aide must complete, a competency evaluation of the deficient skill and all related skill(s) in accordance with § 418.76(c). This change will permit the hospice to focus on the hospice aides' specific deficient and related skill(s) instead of completing another full competency evaluation. We believe when a deficient area(s) in the aide's care is assessed by the RN, there may be additional related competencies that may also lead to additional deficient practice areas and thus would require that those skills be included in the targeted competency evaluation.

We received a total of 32 comments pertaining to the proposed revision to the CoPs. Commenters included individuals, hospice agencies, state hospice associations, national provider organizations, and patient advocacy groups. The response to those comments follows:

Comment: Commenters were overwhelmingly supportive of the provisions to permit the use of pseudo-patients and simulation when conducting hospice aide competency training and for retraining of deficient skills. Several commenters indicated that the changes will facilitate a more time-efficient process in the evaluation of aide skills. Another commenter stated the changes improve the efficiency of onboarding new staff in a safe and effective manner.

Response: We appreciate these comments and agree that the utilization of pseudo-patients and simulation will facilitate more timely completion of training requirements for newly hired hospice aides as well as allowing hospices to target specific competency training for hospice aides noted to have deficient skill(s) on the supervisory visit. We believe that this will benefit the hospice and the patient by allowing new aide trainees and aides requiring remedial training and competency testing to begin serving patients more quickly while protecting patient health and safety.

Comment: Several commenters stated that the use of pseudo-patients and simulation techniques are common in healthcare and a standard of practice in many formal nursing assistant programs. These commenters also state that hospices can adequately assess an aide's skills through these means during competency training. Another commenter indicated that the use of pseudo-patients and simulation will support patient privacy.

Response: We appreciate the commenters highlighting the use of pseudo-patients and simulation techniques in other healthcare setting and agree that the use of these techniques is standard of practice in many formal nursing assistant programs. We believe patient privacy is a fundamental right for those persons receiving hospice care. We agree that permitting competency testing of hospice aides utilizing a pseudo-patient will support patient privacy while also assuring a competently trained hospice aide workforce that provide high quality patient care.

Comment: While the majority of commenters supported the proposed changes; one commenter did not support the use of the pseudo-patient or targeted competency testing. The commenter suggested that more research and data are required on the use of pseudo-patients and changes to competency requirements prior to making a policy decision. The commenter also stated that data and research should support that using a non-patient in training is safe when aides subsequently provide care. Additionally, the commenter raised concerns regarding instances when multiple areas of deficient practice are noted and if a full competency would be done these instances.

Response: We appreciate the commenter's concern and the request for additional research in this area. We believe, and other commenters noted, that the use of pseudo-patients and simulation is an accepted standard of practice for training in healthcare, including nurse aide training programs. These same requirements were implemented for home health aide supervision in 2019 (see 84 FR 51732 and the associated regulations at § 484.80(c)(1)), without any reported adverse impacts noted to-date in CMS survey data or complaints being reported to CMS. Both the use of the pseudo-patient and targeted aide training align requirements between these two providers, home health and hospice, affording the opportunity for efficiency in implementation for many agencies that are Medicare certified to provide both services.

When deficient aide skills are noted during a supervisory visit, the RN determines the deficient skills and all related skills that may be impacted. The supervising RN then determines the scope of the competency testing required, which may include a full competency testing of all skills if warranted, such as when multiple areas of deficient practice are noted.

Comment: One commenter recommended CMS broaden its view of nurses to include licensed practical nurses (LPNs) for conducting aide supervisory visits. The commenter indicated that this change would provide greater staffing flexibility for hospices given workforce shortages among essential workers.

Response: We appreciate the recommendation to permit greater flexibility for hospices in regards to staffing of essential workers. However, we have previously addressed this matter in prior rulemaking (see Medicare and Medicaid Programs: Hospice Conditions of Participation; final rule; 73 FR 32131 issued June 5, 2008) and believe the rationale for requiring a RN for conducting supervisory visits continues to be warranted. Registered nurses, through their education, training, and role in provision of hospice care, are best positioned to assess the adequacy of the aide services in relationship to the needs of the patient and family to a greater degree than LPNs, or licensed vocational nurses (LVNs). Ideally, the supervising RN is both responsible for supervision of the aide services as well as being primarily responsible for the patient's nursing care. This allows the RN to develop a complete picture of the patient and family and of the aide's services.

Comment: Many commenters stated that focusing the competency training on specific deficient skills provided greater efficiency for hospices. One commenter indicated that comprehensive competency testing can take up to a full 8-hour day and a targeted approach will save time related to this requirement. Another commenter stated that completing a full competency test takes the focus away from the identified deficiency and is not effective. A third commenter stated that topic-specific evaluations will significantly reduce time and allow hospices to concentrate on the specific deficient skills with additional practice and training.

Response: We appreciate the support for this comment and agree that a targeted approach is both more efficient and will permit greater focus on remediating the deficient skills.

Comment: Many commenters requested clarification related to the use of technology under the Medicare hospice benefit during the PHE. These commenters requested that CMS further clarify that technology-based visits are permissible outside of a PHE under the same circumstances and conditions as under a PHE, provided applicable HIPAA requirements are met, and requested that CMS establish modifiers that can be used on claims to designate such visits.

Response: While comments on this topic are out of scope for this rulemaking, we do believe the subject is important to address, given the number of comments on this topic. On April 6, 2020, we published an interim final rule “Medicare and Medicaid Programs; Policy and Regulatory Revisions in Response to the COVID-19 Public Health Emergency” (85 FR 19230). This rule provided individuals and entities that provide services to Medicare beneficiaries needed flexibilities to respond effectively to the serious public health threats posed by the spread of COVID-19. The rule implemented temporary changes to the hospice payment requirements to provide broad flexibilities to furnish services using telecommunications technology in order to avoid exposure risks to health care providers, patients, and the community during the PHE. These changes will expire at the end of the COVID-19 PHE. The use of telehealth for conducting the required hospice face-to-face (F2F) encounter is statutorily limited to the PHE for COVID-19 in accordance with section 1814(a)(1)(7)(D)(i) of the Act, as amended by section 3706 of the Coronavirus Aid, Relief, and Economic Security Act (Pub. L. 116-136).

The CoPs are not relevant to payment questions regarding the use of technology, such as telehealth, in the provision of hospice services. The standard of practice for hospice is that care and services are provided on an in-person basis based on needs identified in the comprehensive assessment and services ordered by the IDG and outlined in the plan of care. While nothing in the COPs prevent hospices from augmenting in-person visits with technological means, such as telehealth, these are not intended to change the standard of practice or replace in-person visits. Additionally, for the duration of the PHE, we expect that it would be up to the clinical judgment of hospice as to whether such technology can meet the patient's/caregiver's/family's needs and the use of technology should be included on the plan of care for the patient and family.

We will continue to evaluate the impact of the COVID-19 PHE. At this point, we are still assessing the impact of all waivers and flexibilities on beneficiaries and the delivery of healthcare services under the PHE. While the impact of some waiver and flexibilities may be more apparent at this time, such as the waivers related to hospice aide supervision, flexibilities associated with other aspects of care are more complex requiring additional time for a complete understanding of their impact. We will continue to evaluate the flexibilities to determine if additional changes are warranted in the future.

Final Rule Action: We are finalizing as proposed at § 418.76(c)(1) our policy that hospices may conduct competency testing by observing an aide's performance of the task with a patient or pseudo-patient. Additionally, we are finalizing as proposed at § 418.3 the definitions of “pseudo-patient” and “simulation”.

We are also finalizing as proposed the requirement at § 418.76(h)(1)(iii) to specify that if an area of concern is verified by the hospice during the on-site visit, then the hospice must conduct, and the hospice aide must complete, a competency evaluation of the deficient skill and all related skill(s) in accordance with § 418.76(c).

F. Updates to the Hospice Quality Reporting Program

1. Background and Statutory Authority

The Hospice Quality Reporting Program (HQRP) specifies reporting requirements for both the Hospice Item Set (HIS) and Consumer Assessment of Healthcare Providers and Systems (CAHPS®) Hospice Survey. Section 1814(i)(5) of the Act requires the Secretary to establish and maintain a quality reporting program for hospices. Section 1814(i)(5)(A)(i) of the Act was amended by section 407(b) of Division CC, Title IV of the CAA 2021 (Pub. L. 116-260) to change the payment reduction for failing to meet hospice quality reporting requirements from 2 to 4 percentage points. This policy will apply beginning with FY 2024 annual payment update (APU). Specifically, the Act requires that, beginning with FY 2014 through FY 2023, the Secretary shall reduce the market basket update by 2 percentage points and beginning with the FY 2024 APU and for each subsequent year, the Secretary shall reduce the market basket update by 4 percentage points for any hospice that does not comply with the quality data submission requirements for that FY. We noted this revised statutory requirement in our proposed rule (86 FR 19726) and are codifying the revision at § 418.306(b)(2).

In addition, section 407(a)(2) of the CAA 2021 removes the prohibition on public disclosure of hospice surveys performed be a national accreditation agency in section 1865(b) of the Act, thus allowing the Secretary to disclose such accreditation surveys. In addition, section 407(a)(1) of the CAA 2021 adds new requirements in a newly added section 1822(a)(2) to require each state and local survey agency, and each national accreditation body with an approved hospice accreditation program, to submit information regarding any survey or certification made with respect to a hospice program. Such information shall include any inspection report made by such survey agency or body with respect to such survey or certification, any enforcement actions taken as a result of such survey or certification, and any other information determined appropriate by the Secretary. This information will be published publicly on our website, such as Care Compare, in a manner that is easily accessible, readily understandable, and searchable no later than October 1, 2022. In addition, national accreditation bodies with approved hospice accreditation programs are required to use the same survey form used by state and local survey agencies, which is currently the Form CMS-2567, on or after October 1, 2021.

Depending on the amount of the annual update for a particular year, a reduction of 2 percentage points through FY 2023 or 4 percentage points beginning in FY 2024 could result in the annual market basket update being less than zero percent for a FY and may result in payment rates that are less than payment rates for the preceding FY. Any reduction based on failure to comply with the reporting requirements, as required by section 1814(i)(5)(B) of the Act, would apply only for the specified year. Any such reduction would not be cumulative nor be taken into account in computing the payment amount for subsequent FYs. We are revising the regulations text at § 418.306(b)(2) under a “good cause” waiver of proposed rulemaking as this change was noted in the proposed rule and is a statutory requirement of the CAA of 2021. Under the Administrative Procedure Act (APA) (5 U.S.C. 553), the agency is not required to conduct notice and comment rulemaking for a change that is statutory. Section V. of this final rule further details this waiver of proposed rulemaking. Thus, 42 CFR 418.306(b)(2) has been revised to follow the CAA of 2021 updates for the survey agencies.

Section 1814(i)(5)(C) of the Act requires that each hospice submit data to the Secretary on quality measures specified by the Secretary. The data must be submitted in a form, manner, and at a time specified by the Secretary. Any measures selected by the Secretary must have been endorsed by the consensus-based entity which holds a performance measurement contract with the Secretary under section 1890(a) of the Act. This contract is currently held by the National Quality Forum (NQF). However, section 1814(i)(5)(D)(ii) of the Act provides that in the case of a specified area or medical topic determined appropriate by the Secretary for which a feasible and practical measure has not been endorsed by the consensus-based entity, the Secretary may specify measures that are not endorsed, as long as due consideration is given to measures that have been endorsed or adopted by a consensus-based organization identified by the Secretary. Section 1814(i)(5)(D)(iii) of the Act requires that the Secretary publish selected measures applicable with respect to FY 2014 no later than October 1, 2012.

In the FY 2014 Hospice Wage Index and Payment Rate Update final rule (78 FR 48234), and in compliance with section 1814(i)(5)(C) of the Act, we finalized the specific collection of data items that support the seven NQF-endorsed hospice measures described in Table 6. In addition, we finalized the Hospice Visits When Death is Imminent measure pair (HVWDII, Measure 1 and Measure 2) in the FY 2017 Hospice Wage Index and Payment Rate Update final rule, effective April 1, 2017. We refer the public to the FY 2017 Hospice Wage Index and Payment Rate Update final rule (81 FR 52144) for a detailed discussion.

The CAHPS Hospice Survey is a component of the CMS HQRP, which is used to collect data on the experiences of hospice patients and their family caregivers listed in their hospice records. Readers who want more information about the development of the survey, originally called the Hospice Experience of Care Survey, may refer to 79 FR 50452 and 78 FR 48261. National implementation of the CAHPS Hospice Survey commenced January 1, 2015, as stated in the FY 2015 Hospice Wage Index and Payment Rate Update final rule (79 FR 50452).

The CAHPS Hospice Survey measures received NQF endorsement on October 26, 2016 and was re-endorsed November 20, 2020 (NQF #2651). NQF endorsed six composite measures and two overall measures from the CAHPS Hospice Survey. Along with nine HIS-based quality measures, the CAHPS Hospice Survey measures are publicly reported on a designated CMS website that is currently Care Compare. Beginning no earlier than May 2022, the Hospice Visits in Last Days of Life measure and the Hospice Care Index will also be publicly reported on the CMS website. Table 6 lists all quality measures planned for FY 2022 for HQRP.

The Hospice and Palliative Care Composite Process Measure—HIS-Comprehensive Assessment at Admission measure (hereafter referred to as “the HIS Comprehensive Assessment Measure”) underwent an off-cycle review by the NQF Palliative and End-of-Life Standing Committee and successfully received NQF endorsement in July 2017 (NQF 3235). The HIS Comprehensive Assessment Measure captures whether multiple key care processes were delivered upon patients' admissions to hospice in one measure as described in the Table 6. NQF 3235 does not require NQF's endorsements of the previous components to remain valid. Thus, if the components included in NQF 3235 do not individually maintain endorsement, the endorsement status of NQF 3235, as a single measure, will not change.

In the FY 2016 Hospice Wage Index and Rate Update final rule (80 FR 47142), we finalized the policy for retention of HQRP measures adopted for previous payment determinations and seven factors for measure removal. In that same final rule, we discussed that we will issue public notice, through rulemaking, of measures under consideration for removal, suspension, or replacement. However, if there is reason to believe continued collection of a measure raises potential safety concerns, we will take immediate action to remove the measure from the HQRP and will not wait for the annual rulemaking cycle. Such measures will be promptly removed and we will immediately notify hospices and the public of our decision through the usual HQRP communication channels, including but not limited to listening sessions, email notification, Open Door Forums, HQRP Forums, and Web postings. In such instances, the removal of a measure will be formally announced in the next annual rulemaking cycle.

In the FY 2019 Hospice Wage Index and Rate Update final rule (83 FR 38622), we also adopted an eighth factor for removal of a measure. This factor aims to promote improved health outcomes for beneficiaries while minimizing the overall costs associated with the program. These costs are multifaceted and include the burden associated with complying with the program. The finalized reasons for removing quality measures are:

1. Measure performance among hospices is so high and unvarying that meaningful distinctions in improvements in performance can no longer be made;

2. Performance or improvement on a measure does not result in better patient outcomes;

3. A measure does not align with current clinical guidelines or practice;

4. A more broadly applicable measure (across settings, populations, or conditions) for the particular topic is available;

5. A measure that is more proximal in time to desired patient outcomes for the particular topic is available;

6. A measure that is more strongly associated with desired patient outcomes for the particular topic is available;

7. Collection or public reporting of a measure leads to negative unintended consequences; or

8. The costs associated with a measure outweighs the benefit of its continued use in the program.

On August 31, 2020, we added correcting language to the FY 2016 Hospice Wage Index and Payment Rate Update and Hospice Quality Reporting Requirements; Correcting Amendment (85 FR 53679) hereafter referred to as the FY 2021 HQRP Correcting Amendment. In this final rule, we made correcting amendments to 42 CFR 418.312 to correct technical errors identified in the FY 2016 Hospice Wage Index and Payment Rate Update final rule. Specifically, the FY 2021 HQRP Correcting Amendment (85 FR 53679) adds paragraph (i) to § 418.312 to reflect our exemptions and extensions requirements, which were referenced in the preamble but inadvertently omitted from the regulations text. Thus, these exemptions or extensions can occur when a hospice encounters certain extraordinary circumstances.

As stated in the FY 2019 Hospice Wage Index and Rate Update final rule (83 FR 38622), we launched the Meaningful Measures initiative (which identifies high priority areas for quality measurement and improvement) to improve outcomes for patients, their families, and providers while also reducing burden on clinicians and providers. More information about the Meaningful Measures initiative can be found at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​QualityInitiativesGenInfo/​MMF/​General-info-Sub-Page.html.

In the FY 2020 Hospice Wage Index and Payment Rate Update final rule (84 FR 38484), we discussed our interest in developing quality measures using claims data, to expand data sources for quality measure development. While we acknowledged in that rule the limitations with using claims data as a source for measure development, there are several advantages to using claims data as part of a robust HQRP as discussed previously in the FY 2020 rule. We also discussed developing the Hospice Outcomes & Patient Evaluation (HOPE), a new patient assessment instrument that is planned to replace the HIS. See an update on HOPE development in section III.F.6, “Update regarding the Hospice Outcomes & Patient Evaluation (HOPE) development”.

We also discussed our interest in outcome quality measure development. Unlike process measures, outcome measures capture the results of care as experienced by patients, which can include aspects of a patient's health status and their experiences in the health system. The portfolio of quality measures in the HQRP will include outcome measures that reflect the results of care.

We received comments from various stakeholders on the proposals and updates including a consumer advocacy group, health care providers, hospice provider organizations, hospice trade groups, including those focused on rural providers, consultants, EHR vendors, and MedPAC.

Comment: We received a comment that we are making many updates in this rule and the resources for them are significant, especially during the COVID-19 Public Health Emergency (PHE). They ask us to consider a more gradual transition to new quality initiatives, staggered and prioritized.

Response: We are mindful of the burden related to our updates. We purposely made no updates or proposals in the FY 2021 final rule during the COVID-19 PHE. For FY 2022, two of the four measures we proposed to add were claims-based measures which do not increase burden to providers. We also proposed to remove multiple measures thus leading to a net decrease of total measures. Under our proposal, the HQRP will go from 10 measures down to 4 measures with two of these measures being claims-based measures, and the two already publicly reported measures of the CAHPS Hospice Survey and NQF #3235, the HIS-Comprehensive Assessment Measure. The public reporting has been thoughtfully considered as discussed in this rule so that providers can access their data earlier and prepare for public reporting in FY 2022, no sooner than May 2022. We also consider this work in coordination with planned future HOPE implementation and ensuring that the HQRP now covers the entire hospice stay with these 4 measures rather than just admission and discharge.

2. Removal of the Seven “Hospice Item Set Process Measures” From HQRP Beginning FY 2022

In the FY 2014 Hospice Wage Index and Payment Rate Update final rule (78 FR 48234), and in compliance with section 1814(i)(5)(C) of the Act, we finalized the specific collection of standardized data items, known as the HIS, that support the following NQF-endorsed measures:

• NQF #1617 Patients Treated with an Opioid who are Given a Bowel Regimen
• NQF #1634 Pain Screening
• NQF #1637 Pain Assessment
• NQF #1638 Dyspnea Treatment
• NQF #1639 Dyspnea Screening
• NQF #1641 Treatment Preferences
• NQF #1647 Beliefs/Values Addressed (if desired by the patient)

These measures were adopted to increase public awareness of key components of hospice care, such as pain and symptom management and non-clinical care needs. Consistent with our policy for measure retention and removal, finalized in the FY 2016 Hospice Wage Index and Rate Update final rule (80 FR 47142), we reviewed these measures against the factors for removal. Our analysis found that they meet factor 4: “a more broadly applicable measure (across settings, populations, or conditions) for the particular topic is available.” We determined that the HIS Comprehensive Assessment Measure, discussed in detail in the FY 2017 Hospice Wage Index and Payment Rate Update final rule (81 FR 52144), is a more broadly applicable measure and continues to provide, in a single measure, meaningful differences between hospices regarding overall quality in addressing the physical, psychosocial, and spiritual factors of hospice care upon admission.

The HIS Comprehensive Assessment Measure's “all or none” criterion requires hospices to perform all seven care processes in order to receive credit. In this way, it is different from an average-based composite measure and sets a higher bar for performance. This single measure differentiates hospices and holds them accountable for completing all seven process measures to ensure core services of the hospice comprehensive assessment are completed for all hospice patients. Therefore, the HIS Comprehensive Assessment Measure continues to encourage hospices to improve and maintain high performance in all seven processes simultaneously, rather than rely on its component measures to demonstrate quality hospice care in a way that may be hard to interpret for consumers. The individual measures show performance for only one process and do not demonstrate whether the hospice provides high-quality care overall, as an organization. For example, a hospice may perform extremely well assessing treatment preferences, but poorly on addressing pain. High-quality hospice care not only manages pain and symptoms of the terminal illness, but assesses non-clinical needs of the patient and family caregivers, which is a hallmark of patient-centered care. Since the HIS Comprehensive Assessment Measure captures all seven processes collectively, we believe that public display of the individual component measures is not necessary.

The interdisciplinary, holistic scope of the HIS Comprehensive Assessment Measure aligns with the public's expectations for hospice care. In addition, the measure supports alignment across our programs and with other public and private initiatives. The seven individual components address care processes around hospice admission that are clinically recommended or required in the hospice CoPs. The Medicare Hospice CoPs require that hospice comprehensive assessments identify patients' physical, psychosocial, emotional, and spiritual needs and address them to promote the hospice patient's comfort throughout the end-of-life process. Furthermore, the person-centered, family, and caregiver perspective align with the domains identified by the CoPs and the National Consensus Project ^[8] as patients and their family caregivers also place value on physical symptom management and spiritual/psychosocial care as important factors at the end-of-life. The HIS Comprehensive Assessment Measure is a composite measure that serves to ensure all hospice patients receive a comprehensive assessment for both physical and psychosocial needs at admission.

In addition, MedPAC's Report to Congress: Medicare Payment Policy ^[9] in recent years noted that the HIS Comprehensive Assessment Measure differentiates the hospice's overall ability to address care processes better than the seven individual HIS process measures. In this way, it provides consumers viewing data on Care Compare with a streamlined way to assess the extent to which a hospice follows care processes. In this final rule, we are not making any revisions to the HIS Comprehensive Assessment Measure because the single measure continues to show sufficient variability and therefore provides value to patients, their families, and providers.

Because the HIS Comprehensive Assessment Measure is a more broadly applicable measure, we are finalizing our proposal to remove the seven individual HIS process measures from the HQRP, no longer publicly reporting them as individual measures on Care Compare beginning with FY 2022. In addition, we proposed and finalize in this rule to remove the “7 measures that make up the HIS Comprehensive Assessment Measure” section of Care Compare, which displays the seven HIS measures. We proposed and are finalizing these changes to remove the seven HIS process measures as individual measures from HQRP no earlier than May 2022.

Although we would remove the seven individual HIS process measures, it does not change the requirement to submit the HIS admission assessment. Since the HIS Comprehensive Assessment Measure is a composite of the seven HIS process measures, the burden and requirement to report the HIS data remain unchanged in the time, manner, and form finalized in the FY 2017 Hospice Wage Index and Rate Update final rule (81 FR 52144). Hospices which do not report HIS data used for the HIS Comprehensive Assessment Measure will not meet the requirements for compliance with the HQRP.

We solicited public comment on the proposal to remove the seven HIS process quality measures as individual measures from the HQRP no earlier than May 2022, and to continue including the seven HIS process measures in the confidential quality measure (QM) Reports which are available to hospices. The seven HIS process measures are also available by visiting the data catalogue at https://data.cms.gov/​provider-data/​topics/​hospice-care. We sought public comment on the technical correction to the regulation at § 418.312(b) effective October 1, 2021.

We received several comments on the proposal to remove the seven “Hospice Item Set process measures” from the HQRP beginning FY 2022. A summary of the comments and our responses to those comments appears below:

Comment: The majority of commenters supported the removal of the seven HIS process measures. Several commenters opposed removing the seven HIS process measures, at least prior to implementation of HOPE. These commenters believed that the existing process measures provide more valuable and transparent information about hospice performance than the HIS Comprehensive Assessment composite measure. Finally, some commenters recommended both removing the seven individual HIS process measures and retiring the HIS Comprehensive Assessment measure. These commenters suggested that retiring the composite measure would reduce provider burden.

Response: We appreciate the support for this proposal. In response to the concerns raised by those opposing the removal of seven HIS process measures, we would like to emphasize that all but one of the seven HIS measures are topped out individually and one HIS measure is almost topped out and shows insignificant variability between hospices. The 7 HIS measures credited hospices when any of these measures were performed regardless of the individual patient. In contrast, the HIS Comprehensive Assessment Measure measures whether a hospice assesses each patient on the 7 HIS measures. This distinction is important since it explains why the individual HIS measures can be topped out but when measured together as a group, or composite, that is required on each patient in order to get credit for the measure, the HIS Comprehensive Assessment Measure shows variability and meets public reporting standards. This distinction explains why most hospices receive the maximum possible score on each of the 7 HIS measures, but not on the HIS Comprehensive Assessment Measure. As such, the individual measures have a limited ability to differentiate hospices. In contrast, the HIS Comprehensive Assessment Measure shows that hospices need to improve on providing a comprehensive set of assessments on each patient at admission and supports why it continues to be a useful HQRP measure.

While we consider it a success that hospices are assessing the care processes included in the 7 HIS measures, hospices have improved since 2014 to the point that these 7 individual HIS measures no longer differentiate quality of care between hospices and need to be retired as individual quality measures and thereby removed from the HQRP. Now that we reached that milestone, we need to recognize that there is a need to focus on assessing the 7 HIS measures to each patient at admission, which is what the HIS Comprehensive Assessment Measure addresses. It more closely aligns with the intent of the Hospice CoPs at Title 42 Part 418.54 that require a comprehensive assessment on each patient. This is why the HIS Comprehensive Assessment Measure provides valuable and transparent information about hospice performance. Patients electing to receive hospice services should expect quality care and a comprehensive assessment of their needs at admission, which the HIS Comprehensive Assessment Measure reflects. While the transition from the HIS to HOPE will eventually enable the HQRP to be more robust, we should not wait to seek improvement on this composite measure as an indicator of quality. This supports why we must remove the 7 HIS measures now in favor of the one more meaningful measure.

Finally, we support minimizing provider burden while maintaining quality measures that provide valuable information to providers and consumers about hospice quality. The variability shown in the HIS Comprehensive Assessment measure continues to provide useful information that allows patients and families to differentiate hospices and help select the best providers for their care.

Comment: MedPAC recommended that CMS consider removing the HIS Comprehensive Assessment Measure because the scores suggest the composite measure is limited in distinguishing provider quality. The comment suggested that the HIS Comprehensive Assessment measure would be likely to top out due to high scoring trends among hospices.

Response: We appreciate MedPAC raising this concern. We recognize that the HIS Comprehensive Assessment Measure reflects high scores and is improving over time, which may cause the measure to also become topped out in the future.^[10] However, we believe that the single measure currently continues to show sufficient variability to differentiate hospices and therefore provides value to patients, their families, and providers. Further, the HIS Comprehensive Assessment Measure reflects the Hospice CoPs for comprehensive assessments performed at admission, which is a critical time to determine the plan of care. Its removal would not only leave HQRP without this important admission quality of care measure but also result in HQRP having only two claims-based measures, HCI and HVLDL, and the CAHPS Hospice Survey. It is these four quality measures, the HIS Comprehensive Assessment Measure, HCI, HVLDL, and CAHPS Hospice Survey that make up the FY 2022 HQRP requirements. These four measures cover hospice care throughout the hospice stay. The HIS Comprehensive Assessment Measure covers care at admission. HCI covers care throughout the hospice stay. HVLDL covers care during discharge and the CAHPS Hospice Survey covers the caregiver experience of hospice care. They complement each other and further support the need for each measure in the HQRP. We will continue to monitor the HIS Comprehensive Assessment Measure performance and consider if removal or refinements would be appropriate in the future.

Final Decision: In this final rule, we are not making any revisions to the HIS Comprehensive Assessment Measure. We are finalizing our proposal to remove the seven individual HIS process measures from the HQRP, no longer publicly reporting them as individual measures on Care Compare beginning with FY 2022. In addition, we will remove the “7 measures that make up the HIS Comprehensive Assessment Measure” section of Care Compare, which displays the seven HIS measures. These will be effective no earlier than May 2022. Hospice providers, must report HIS data used for the HIS Comprehensive Assessment Measure, in order to meet the requirements for compliance with the HQRP.

3. Addition of a “Claims-Based Index Measure”, the Hospice Care Index

We proposed the addition of a new hospice quality measure, called the Hospice Care Index (HCI), to HQRP. The HCI will provide more information to better reflect several processes of care during a hospice stay, and better empower patients and family caregivers to make informed health care decisions. The HCI is a single measure comprising ten indicators calculated from Medicare claims data. The index design of the HCI simultaneously monitors all ten indicators. Collectively these indicators represent different aspects of hospice service and thereby characterize hospices comprehensively, rather than on just a single care dimension. Therefore, the HCI composite yields a more reliable provider ranking.

The HCI indicators, through the composite, will add new information to HQRP that was either directly recommended for CMS to publicly report by Federal stakeholders ^[11 12] or identified as areas for improvement during information gathering activities. Furthermore, each indicator represents either a domain of hospice care recommended by leading hospice and quality experts ^[13] for CMS to publicly report, or a requirement included in the hospice CoPs. The indicators required to calculate the single composite are discussed in the “Specifications for the HCI Indicators Selected” section. These specifications list all the information required to calculate each indicator, including the numerator and denominator definitions, different thresholds for receiving credit toward the overall HCI score, and explanations for those thresholds. Indicators reflect practices or outcomes hospices should pursue, thereby awarding points based on the criterion. The HCI scoring example in Table 8 illustrates how points are awarded based on meeting the criterion of the indicator. For example, Gaps in Skilled Nursing Visits have a criterion of “lower than the 90th percentile,” and supports the hospice CoPs that require an assessment of the patient and caregiver needs as well as implementation of the plans of care. Other indicators, such as nurse visits (RN and LPN) on weekends or near death, have a criterion of “higher than the 10th percentile,” identifying hospice care delivery during the most vulnerable periods during a hospice stay.

Each indicator equally affects the single HCI score, reflecting the equal importance of each aspect of care delivered from admission to discharge. A hospice is awarded a point for meeting each criterion for each of the 10 indicators. The sum of the points earned from meeting the criterion of each indictor results in the hospice's HCI score, with 10 as the highest possible score. The ten indicators, aggregated into a single HCI score, convey a broad overview of the quality of the provision of hospice care services and validates well with CAHPS Willingness to Recommend and Rating of this Hospice. Skilled nursing visit data for indicators 2, 8, and 9 (described below) uses revenue center code 055X, which includes both RN and LPN visits for consistency with other indications for HCI.

The HCI will help to identify whether hospices have aggregate performance trends that indicate higher or lower quality of care relative to other hospices. Together with other measures already publicly reported in the HQRP, HCI scores will help patients and family caregivers choose between hospice providers based on the factors that matter most to them. Additionally, creating a comprehensive quality measure capturing a variety of related care processes and outcomes in a single metric will provide consumers and providers an efficient way to assess the overall quality of hospice care, which can be used to meaningfully and easily compare hospice providers to make a better-informed health care decision.

The HCI will complement the existing HIS Comprehensive Measure and does not replace any existing reported measures. Both the HCI and the HIS Comprehensive Measure are composite measures in that they act as single measures that capture multiple areas of hospice care. Because the indicators comprising the HCI differ in data source from the HIS Comprehensive Measure, the HCI and the HIS Comprehensive Measure can together provide a meaningful and efficient way to inform patients and family caregivers while supporting their selection of hospice care providers. As a claims-based measure, the HCI measure will not impose any requirements for collection of new information. To learn more about the background of the HCI, please watch this video: https://youtu.be/​by68E9E2cZc.

a. Measure Importance

The FY 2019 Hospice Wage Index and Payment Rate Update final rule (83 FR 38622) introduced the Meaningful Measure Initiative to hospice providers to identify high priority areas for quality measurement and improvement. The Meaningful Measure Initiative areas are intended to increase measure alignment across programs and other public and private initiatives. Additionally, the initiative points to high priority areas where there may be informational gaps in available quality measures. The initiative helps guide our efforts to develop and implement quality measures to fill those gaps and develop those concepts towards quality measures that meet the standards for public reporting. The goal of HQRP quality measure development is to identify measures from a variety of data sources that provide a window into hospice care services throughout the dying process, fit well with the hospice business model, and meet the objectives of the Meaningful Measures initiative.

To that end, the HCI will add value to the HQRP by filling informational gaps in aspects of hospice service not addressed by the current measure set. Consistent with the Meaningful Measure Initiative, we conducted a number of information gathering activities to identify informational gaps. Our information gathering activities included soliciting feedback from hospice stakeholders such as providers and family caregivers; seeking input from hospice and quality experts through a Technical Expert Panel (TEP); interviews with hospice quality experts; considering public comments received in response to previous solicitations on claims-based hospice quality initiatives; and a review of quality measurement recommendations offered by the HHS Office of Inspector General (OIG), MedPAC, and the peer-reviewed literature.

We found that hospices currently underutilize HQRP measures to inform their quality improvement, mainly because of gaps in relevant quality information within the HQRP measure set. In particular, the existing HQRP measure set, calculated using data collected from the HIS and the CAHPS Hospice survey, does not assess quality of hospice care during a hospice election (between admission and discharge). Moreover, the current measure set does not directly address the full range of hospice services or outcomes. Therefore, we have identified a need for a new quality measure to address this gap and reflect care delivery processes during the hospice stay using available data without increasing data collection burden.

Claims data are the best available data source for measuring care during the hospice stay and present an opportunity to bridge the quality measurement gap that currently exists between the HIS and CAHPS Hospice Survey. Medicare claims are administrative records of health care services provided and payments which Medicare (and beneficiaries as applicable) made for those services. Claims are a rich and comprehensive source of many care processes and aspects of health care utilization. As such, they are a valuable source of information that can be used to measure the quality of care provided to beneficiaries for several reasons:

• Claims data are readily available and eliminates provider burden for implementation, as opposed to data collection through patient assessments or surveys, which require additional effort from clinicians, patients, and family caregivers before they can be submitted and used by CMS.
• Claims data are collected based on the actual care delivered, providing a more direct reflection of care delivery decisions and actions than patient assessments or surveys.
• Claims data are considered a reliable source of standardized data about the services provided, because providers must comply with Medicare payment and claims processing policy.

CMS already publicly reports several pieces of information derived from hospice claims data in the HQRP on Care Compare, including (i) the levels of care provided by the hospice, (ii) the primary diagnoses of patients served by the hospice, (iii) the location of hospice service provided, and (iv) the hospice's average daily census.

In the FY2018 Hospice Wage Index & Payment Rate proposed rule (82 FR 20750), we solicited public comment on two high-priority claims-based measure concepts being considered at the time, one which looked at transitions from hospice and another which examined access to higher levels of hospice care. In response to this solicitation, CMS received public comments highlighting the potential limitations of a single concept claims-based measure. In particular, a single-concept claims-based measure may not adequately account for all relevant circumstances that might influence a hospice's performance. While external circumstances could justify a hospice's poor performance on a single claims-based indicator, it would be unlikely for external circumstances to impact multiple claims-based indicators considered simultaneously. Therefore, the result of a multi-indicator claims-based index, such as HCI, is more likely to differentiate hospices than a single claims-based indicator. Taking this public feedback into consideration, we designed the HCI and developed specifications based on simulated reporting periods.

b. Specifications for the HCI Indicators Selected

Specifications for the ten indicators required to calculate the single HCI score are described in this section. These component indicators reflect various elements and outcomes of care provided between admission and discharge. The HCI uses information from all ten indicators to collectively represent a hospice's ability to address patients' needs, best practices hospices should observe, and/or care outcomes that matter to consumers. Each indicator is a key component of the HCI measure that we proposed, and all ten are necessary to derive the HCI score. We use analytics, based on a variety of data files, to specify the indicators and measure. These data files include:

• Medicare fee-for-service (FFS) hospice claims with through dates on and between October 1, 2016 and September 30, 2019 to determine information such as hospice days by level of care, provision of visits, live discharges, hospice payments, and dates of hospice election.
• Medicare fee-for-service inpatient claims with through dates on and between January 1, 2016 and December 31, 2019 to determine dates of hospitalization.
• Medicare beneficiary summary file to determine dates of death.
• Provider of Services (POS) File to examine trends in the scores of the HCI and its indicators, including by decade by which the hospice was certified for Medicare, ownership status, facility type, census regions, and urban/rural status.
• CAHPS Hospice Survey to examine alignment between the survey outcomes and the HCI.

We acquired all claims data from the Chronic Conditions Warehouse (CCW) Virtual Research Data Center (VRDC). We obtained the hospice claims and the Medicare beneficiary summary file in May 2020, and the inpatient data in August 2020. We obtained the POS file data via: https://www.cms.gov/​Research-Statistics-Data-and-Systems/​Downloadable-Public-Use-Files/​Provider-of-Services. We obtained the Hospice-aggregate CAHPS Hospice Survey outcome data via: https://data.cms.gov/​provider-data. We performed analyses using Stata/MP Version 16.1.

Table 7 indicates the number of hospice days, hospice claims, beneficiaries enrolled in hospices and hospices with at least one claim represented in each year of our analysis. Analysis for each year was based on the FY calendar. For example, FY 2019 covers claims with dates of services on or between October 1, 2018 and September 30, 2019. For these analyses, we exclude claims from hospices with 19 or fewer discharges ^[14] within a FY. The table reports the sample size before and after exclusion.^[15]

The rest of this section presents the component indicators and their specifications. Although we describe each component indicator separately, the HCI is a composite that can only be calculated using all 10 indicators combined. We believe that, composed of this set of ten indicators, the HCI will strengthen the HQRP by comprehensively, reflecting hospices' performance across all ten indicators.

(1). Indicator One: Continuous Home Care (CHC) or General Inpatient (GIP) Provided

Medicare Hospice Conditions of Participation (CoPs) require hospices to be able to provide both CHC and GIP levels of care, if needed to manage more intense symptoms.^[16 17] However, a 2013 OIG report ^[18] found that 953 hospice programs did not provide any GIP level of care services, and it was unclear if dying patients at such hospices were receiving appropriate pain control or symptoms management (a similar concern exists for hospice services at the CHC level). To consider the provision of adequate services needed to manage patients' symptoms, the HCI measure includes an indicator for whether hospice programs provided any CHC or GIP service days. This indicator identifies hospices that provided at least one day of hospice care under the CHC or the GIP levels of care during the period examined. The provision of CHC and GIP is identified on hospice claims by the presence of revenue center codes 0652 (CHC) and 0656 (GIP).

The specifications for Indicator One, CHC or GIP services provided, are as follows:

• Numerator: The total number of CHC or GIP services days provided by the hospice within a reporting period.
• Denominator: The total number of hospice service days provided by the hospice at any level of care within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if they provided at least one CHC or GIP service day within a reporting period.

(2). Indicator Two: Gaps in Skilled Nursing Visits

The OIG has found instances of infrequent visits by nurses to hospice patients.^[19] To assess patients' receipt of nurse visits as outlined in the plan of care, one HCI indicator examines hospices that have a high rate of patients who are not seen at least once a week by nursing staff. This indicator includes both RN and LPN visits to recognize the frequency of skilled nursing visits and to maintain consistency in HCI when using revenue center code 055X.

This indicator identifies whether a hospice is below the 90th percentile in terms of how often hospice stays of at least 30 days contain at least one gap of eight or more days without a nursing visit. Days of hospice service are identified based on the presence of revenue center codes 0651 (routine home care (RHC)), 0652 (CHC), 0655 (inpatient respite care (IRC)), and 0656 (GIP) on hospice claims. We identify the dates billed for RHC, IRC, and GIP by examining the corresponding revenue center date (which identifies the first day in the sequence of days by level of care) and the revenue center units (which identify the number of days (including the first day) in the sequence of days by level of care). We identify the dates billed for CHC by examining the revenue center date.^[20] We define a hospice stay by a sequence of consecutive days for a particular beneficiary that are billed under the hospice benefit. A gap of at least 1 day without hospice ends the sequence. For this indicator, we identified hospice stays that included 30 or more consecutive days of hospice. Once we identified those hospice stays, we examined the timing of the provision of nursing visits within those stays. We identified nursing visits if we observed any of the following criteria:

• The presence of revenue center code 055x (Skilled Nursing) on the hospice claim. The date of the visit is recorded in the corresponding revenue center date.
• The presence of revenue code 0652 (CHC) on the hospice claim. Days billed as CHC require more than half the hours provided be nursing hours.
• The presence of revenue code 0656 (GIP) on the hospice claim. We assume that days billed as GIP will include nursing visits. We make that assumption instead of looking at the visits directly because Medicare does not require hospices to record all visits on the claim for the GIP level of care.

If within a hospice stay, we find eight or more consecutive days where no nursing visits are provided, no CHC is provided, and no GIP is provided, then we identify the hospice stay as having a gap in nursing visits greater than 7 days. This indicator helps the HCI to capture patients' receipt of skilled nursing visits and direct patient care, which is an important aspect of hospice care. For each hospice, we divide the number of stays with at least one gap of eight or more days without a nursing visit (for stays of 30 or more days) by the number of stays of 30 or more days. We only consider the days within the period being examined.

The specifications for Indicator Two, Gaps in Skilled Nursing Visits, are as follows:

• Numerator: The number of elections with the hospice where the patient experienced at least one gap between nursing visits exceeding 7 days, excluding hospice elections where the patient elected hospice for less than 30 days within a reporting period.
• Denominator: The total number of elections with the hospice, excluding hospice elections where the patient elected hospice for less than 30 days within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their individual hospice score for gaps in skilled nursing visits greater than 7 days falls below the 90th percentile ranking among hospices nationally.

(3). Indicator Three: Early Live Discharges

Prior work has identified various concerning patterns of live discharge from hospice. High rates of live discharge suggest concerns in hospices' care processes, their advance care planning to prevent hospitalizations, or their discharge processes.^[21] As MedPAC noted,^[22] “Hospice providers are expected to have some rate of live discharges because some patients change their mind about using the hospice benefit and dis-enroll from hospice or their condition improves and they no longer meet the hospice eligibility criteria. However, providers with substantially higher percent of live discharge than their peers could signal a potential concern with quality of care or program integrity. An unusually high rate of live discharges could indicate that a hospice provider is not meeting the needs of patients and families or is admitting patients who do not meet the eligibility criteria.”

Our live discharge indicators included in the HCI, like MedPAC's, comprise discharges for all reasons. They include instances where the patient was no longer found terminally ill and revocations due to the patient's choice. MedPAC explains their rationale for including all discharge as follows: ^[23]

“Some stakeholders argue that live discharges initiated by the beneficiary—such as when the beneficiary revokes his or her hospice enrollment—should not be included in a live-discharge measure because, some stakeholders assert, these discharges reflect beneficiary preferences and are not in the hospice's control. Because beneficiaries may choose to revoke hospice for a variety of reasons, which in some cases are related to the hospice provider's business practices or quality of care, we include revocations in our analysis.”

This indicator identifies whether a hospice is below the 90th percentile in terms of the percentage of live discharges that occur within 7 days of hospice admission during the fiscal year examined. Live discharges occur when the patient discharge status code on a hospice claim does not equal a code from the following list: “30”, “40”, “41”, “42”, “50”, “51”. We measure whether a live discharge occurs during the first 7 days of hospice by looking at a patient's lifetime length of stay in hospice.^[24] For each hospice, we divide the number of live discharges in the first 7 days of hospice by the number of live discharges. Live discharges are assigned to a particular reporting period based on the date of the live discharge (which corresponds to the through date on the claim indicating the live discharge).

The specifications for Indicator Three, Early Live Discharges, are as follows:

• Numerator: The total number of live discharges from the hospice occurring within the first 7 days of hospice within a reporting period.
• Denominator: The total number of all live discharge from the hospice within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their individual percentage of live discharges on or before the seventh day of hospice falls below the 90th percentile ranking among hospices nationally.

(4). Indicator Four: Late Live Discharges

The rate of live discharge that occurred 180 days or more after hospice enrollment identifies another potentially concerning pattern of live discharge from hospice. Both indicator three and indicator four of the HCI recognize concerning patterns of live discharge impacting patient experience and quality of care. MedPAC, in descriptive analyses of hospices exceeding the Medicare annual payment cap, noted that “if some hospices have rates of discharging patients alive that are substantially higher than most other hospices it raises concerns that some hospices may be pursuing business models that seek out patients likely to have long stays who may not meet the hospice eligibility criteria”.^[25] Because of quality implications for hospices who pursue such business models, the live discharge after long hospice enrollments was included in the index.

This indicator identifies whether a hospice is below the 90th percentile in terms of the percentage of live discharges that occur on or after the 180th day of hospice. Live discharges occur when the patient discharge status code does not equal a value from the following list: “30”, “40”, “41”, “42”, “50”, “51”. We measure whether a live discharge occurs on or after the 180th day of hospice by looking at a patient's lifetime length of stay in hospice. For each hospice, we divide the number of live discharges that occur on or after the 180th day of hospice by the number of live discharges. Live discharges are assigned to a particular reporting period based on the date of the live discharge (which corresponds to the through date on the claim).

The specifications for Indicator Four, Late Live Discharges, are as follows:

• Numerator: The total number of live discharges from the hospice occurring on or after 180 days of enrollment in hospice within a reporting period.
• Denominator: The total number of all live discharge from the hospice within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their individual hospice score for live discharges on or after the 180th day of hospice falls below the 90th percentile ranking among hospices nationally.

(5). Indicator Five: Burdensome Transitions (Type 1)—Live Discharges From Hospice Followed by Hospitalization and Subsequent Hospice Readmission

The Type 1 burdensome transitions reflects hospice live discharge with a hospital admission within 2 days of hospice discharge, and then hospice readmission within 2 days of hospital discharge. This pattern of transitions may lead to fragmented care and may be associated with concerning care processes. For example, Type 1 burdensome transitions may arise from a deficiency in advance care planning to prevent hospitalizations or a discharge process that does not appropriately identify a hospice patient whose conditions are stabilized prior to discharge.^[26]

This indicator identifies whether a hospice is below the 90th percentile in terms of the percentage of live discharges that are followed by a hospitalization (within 2 days of hospice discharge) and then followed by a hospice readmission (within 2 days of hospitalization) during the FY examined. Live discharges occur when the patient discharge status code does not equal a value from the following list: “30”, “40”, “41”, “42”, “50”, “51”. Hospitalizations are found by looking at all fee-for-service Medicare inpatient claims. Overlapping inpatient claims were combined to determine the full length of a hospitalization (looking at the earliest from date and latest through date from a series of overlapping inpatient claims for a beneficiary). In order to be counted, the “from” date of the hospitalization had to occur no more than 2 days after the date of hospice live discharge.^[27] From there, we found all beneficiaries that ended their hospitalization and were readmitted back to hospice no more than 2 days after the last date of the hospitalization. To calculate the percentage, for each hospice we divided the number of live discharges that are followed by a hospitalization (within 2 days of hospice discharge) and then followed by a hospice readmission (within 2 days of hospitalization) in a given reporting period by the number of live discharges in that same period.

The specifications for Indicator Five, Burdensome Transitions Type 1, are as follows:

• Numerator: The total number of live discharges from the hospice followed by hospital admission within 2 days, then hospice readmission within 2 days of hospital discharge within a reporting period.
• Denominator: The total number of all live discharge from the hospice within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their individual hospice score for Type 1 burdensome transitions falls below the 90th percentile ranking among hospices nationally.

(6). Indicator Six: Burdensome Transitions (Type 2)—Live Discharges From Hospice Followed by Hospitalization With the Patient Dying in the Hospital

Death in a hospital following live discharge in another concerning pattern in hospice use. Thus, we believe that indicators five and indicator six of the HCI are necessary to differentiate concerning behaviors affecting patient care. This indicator reflects hospice live discharge followed by hospitalization within 2 days with the patient dying in the hospital, referred to as Type 2 burdensome transitions. This pattern of transitions may be associated with a discharge process that does not appropriately assess the stability of a hospice patient's conditions prior to live discharge.^[28]

This indicator identifies whether a hospice is below the 90th percentile in terms of the percentage of live discharges that are followed by a hospitalization (within two days of hospice discharge) and then the patient dies in the hospital. Live discharges occur when the patient discharge status code does not equal a value from the following list: “30”, “40”, “41”, “42”, “50”, “51”. Hospitalizations are found by looking at all inpatient claims. Overlapping inpatient claims were combined to determine a full length of a hospitalization (looking at the earliest from date and latest through date from a series of overlapping inpatient claims). To be counted, the “from” date of the hospitalization had to occur no more than 2 days after the date of hospice live discharge. From there, we identified all beneficiaries whose date of death is listed as occurring during the dates of the hospitalization. To calculate the percentage, for each hospice we divided the number of live discharges that are followed by a hospitalization (within 2 days of hospice discharge) and then the patient dies in the hospital in a given FY by the number of live discharges in that same reporting period.

The specifications for Indicator Six, Burdensome Transitions Type 2, are as follows:

• Numerator: The total number of live discharges from the hospice followed by a hospitalization within 2 days of live discharge with death in the hospital within a reporting year.
• Denominator: The total number of all live discharge from the hospice within a reporting year.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their individual hospice score for Type 2 burdensome transitions falls below the 90th percentile ranking among hospices nationally.

(7). Indicator Seven: Per-Beneficiary Medicare Spending

Estimates of per-beneficiary spending are endorsed by NQF (#2158) ^[29] and publicly reported by CMS for other care settings. Because the Medicare hospice benefit pays a per diem rate, an important determinant of per-beneficiary spending is the length of election. MedPAC reported that nearly half of Medicare hospice expenditures are for patients that have had at least 180 or more days on hospice, and expressed a concern that some programs do not appropriately discharge patients whose medical condition makes them no longer eligible for hospice services, or, that hospices selectively enroll patients with non-cancer diagnoses and longer predicted lengths of stay in hospice.^[30] The other determinant of per-beneficiary spending is the level of care at which services are billed. In a 2016 report, the OIG has expressed concern at the potentially inappropriate billing of GIP care.^[31] For these reasons the HCI includes one indicator for per-beneficiary spending; lower rates of per beneficiary spending may identify hospices that provide efficient care at a lower cost to Medicare.

This indicator identifies whether a hospice is below the 90th percentile in terms of the average Medicare hospice payments per beneficiary. Hospice payments per beneficiary are determined by summing together all payments on hospice claims for a particular reporting year for a particular hospice. The number of beneficiaries a hospice serves in a particular year is determined by counting the number of unique beneficiaries on all hospice claims in the same period for a particular hospice. Medicare spending per beneficiary is then calculated by dividing the total payments by the total number of unique beneficiaries.

The specifications for Indicator Seven, Per-Beneficiary Medicare Spending, are as follows:

• Numerator: Total Medicare hospice payments received by a hospice within a reporting period.
• Denominator: Total number of beneficiaries electing hospice with the hospice within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their average Medicare spending per beneficiary falls below the 90th percentile ranking among hospices nationally.

(8). Indicator Eight: Skilled Nursing Care Minutes per Routine Home Care (RHC) Day

Medicare Hospice CoPs require a member of the interdisciplinary team to ensure ongoing assessment of patient and caregiver needs.^[32] Nursing services require initial and ongoing assessment of patient family needs to ensure the successful preparation, implementation, and refinements for the plan of care. This also includes patient and caregiver education and training as appropriate to their responsibilities for the care and services identified in the plan of care. This indicator includes both RN and LPN visits to recognize the frequency of skilled nursing visits and to maintain consistency in HCI when using revenue center code 055X.

This indicator identifies whether a hospice is above the 10th percentile in terms of the average number of skilled nursing minutes provided on RHC days during the reporting period examined. We identify RHC days by the presence of revenue code 0651 on the hospice claim. We identify the dates of RHC service by the corresponding revenue center date (which identifies the first day of RHC) and the revenue center units (which identifies the number of days of RHC (including the first day of RHC)). We identify skilled nursing visits by the presence of revenue code 055x (Skilled Nursing) on the claim. We count skilled nursing visits where the corresponding revenue center date overlaps with one of the days of RHC previously identified. We then count the minutes of skilled nursing visits by taking the corresponding revenue center units (that is, one unit is 15 minutes) and multiplying by 15. For each hospice, we sum together all skilled nursing minutes provided on RHC days and divide by the sum of RHC days.

The specifications for Indicator Eight, Skilled Nurse Care Minutes per RHC Day, are as follows:

• Numerator: Total skilled nursing minutes provided by a hospice on all RHC service days within a reporting period.
• Denominator: The total number of RHC days provided by a hospice within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their individual hospice score for Skilled Nursing Minutes per RHC day falls above the 10th percentile ranking among hospices nationally.

(9). Indicator Nine: Skilled Nursing Minutes on Weekends

Our regulations at § 418.100(c)(2) require that “[n]ursing services, physician services, and drugs and biologicals . . . be made routinely available on a 24-hour basis seven days a week”.^[33] Fewer observed hospice services on weekends (relative to that provided on weekdays) is not itself an indication of a lack of access. In fact, on weekends, patients' caregivers are more likely to be around and could prefer privacy from hospice staff. However, patterns of variation across providers could signal less service provider availability and access for patients on weekends. Thus, the HCI includes this indicator to further differentiate whether care is available to patients on weekends. To assess hospice service availability, this indicator includes minutes of care provided by skilled nurses on weekend RHC days. This indicator includes both RN and LPN visits to recognize the frequency of skilled nursing visits and to maintain consistency in HCI when using revenue center code 055X.

This indicator identifies whether a hospice is at or above the 10th percentile in terms of the percentage of skilled nursing minutes performed on weekends compared to all days during the reporting period examined. We identify RHC days by the presence of revenue code 0651 on the hospice claim. We identify the dates of RHC service by the corresponding revenue center date (which identifies the first day of RHC) and the revenue center units (which identifies the number of days of RHC (including the first day of RHC)). We identify skilled nursing visits by the presence of revenue code 055x (Skilled Nursing) on the claim. We count skilled nursing visits where the corresponding revenue center date overlaps with one of the days of RHC previously identified. We then count the minutes of skilled nursing visits by taking the corresponding revenue center units and multiplying by 15. For each hospice, we sum together all skilled nursing minutes provided on RHC days that occur on a Saturday or Sunday and divide by the sum of all skilled nursing minutes provided on all RHC days.

The specifications for Indicator Nine, Skilled Nursing Minutes on Weekends, are as follows:

• Numerator: Total sum of minutes provided by the hospice during skilled nursing visits during RHC services days occurring on Saturdays or Sunday within a reporting period.
• Denominator: Total skilled nursing minutes provided by the hospice during RHC service days within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their individual hospice score for percentage of skilled nursing minutes provided during the weekend is above the 10th percentile ranking among hospices nationally.

(10). Indicator Ten: Visits Near Death

The end of life is typically the period in the terminal illness trajectory with the highest symptom burden. Particularly during the last few days before death, patients (and caregivers) experience many physical and emotional symptoms, necessitating close care and attention from the hospice team and drawing increasingly on hospice team resources.^{[34 35 36]} Physical symptoms of actively dying can often be identified within three days of death in some patients.^[37]

This indicator identifies whether a hospice is at or above the 10th percentile in terms of the percentage of beneficiaries with a RN, LPN, and/or medical social services visit in the last 3 days of life. For this indicator, we first determine if a beneficiary was in hospice for at least 1 day during their last 3 days of life by comparing days of hospice enrollment from hospice claims to their date of death. We identify skilled nursing visits and medical social service visits by the presence of revenue code 055x (Skilled Nursing) and 056x (Medical Social Services) on the claim. We identify the dates of those visits by the revenue center date for those revenue codes. Additionally, we assume that days billed as GIP (revenue code 0656) will include skilled nursing visits. We make that assumption instead of looking at the visits directly because Medicare does not require hospices to record all visits on the claim for the GIP level of care. For each hospice, we divide the number of beneficiaries with skilled nursing or medical social service visits on a hospice claim during the last 3 days of life by the number of beneficiaries with at least 1 day of hospice during the last 3 days of life. In the proposed rule, the denominator description is discussed accurately, as the number of beneficiaries with at least one day of hospice during the last three days of life within a reporting period. However, the specification summary inaccurately reflected the number of decedent beneficiaries served by the hospice within a reporting period. In this final rule, we correct this error and replace the description of the denominator accurately as the number of beneficiaries with at least 1 day of hospice during the last 3 days of life within a reporting period.

The specifications for Indicator Ten, Visits Near Death, are as follows:

• Numerator: The number of decedent beneficiaries receiving a visit by a skilled nurse or social worker for the hospice in the last 3 days of the beneficiary's life within a reporting period.
• Denominator: The number of beneficiaries with at least 1 day of hospice during the last 3 days of life within a reporting period.
• Index Earned Point Criterion: Hospices earn a point towards the HCI if their individual hospice score for percentage of decedents receiving a visit by a skilled nurse or social worker in the last 3 days of life falls above the 10th percentile ranking among hospices nationally.

(11). Hospice Care Index Scoring Example

As discussed during the NQF's January 2021 MAP meeting, the HCI summarizes information from ten indicators with each indicator representing key components of the hospice care received, recognizing care delivery and processes. Hospices receive a single HCI score, which reflects the information from all ten indicators. Specifically, a hospice's HCI score is based on its collective performance for the ten performance indicators detailed earlier, all of which must be included to calculate the score and meaningfully distinguish between hospices' relative performance. The HCI's component indicators are assigned a criterion determined by statistical analysis of an individual hospice's indicator score relative to national hospice performance. Table 8 illustrates how a hypothetical hospice's score is determined across all ten indicators, and how the ten indicators' scores determine the overall HCI score.

c. Measure Reportability, Variability, and Validity

As part of developing the HCI, we conducted reportability, variability, and validity testing using claims data from FY 2019. Reportability analyses found a high proportion of hospices (over 85 percent) that would yield reportable measure scores over 1 year (for more on reportability analysis, see section (2) Update on Use of Q4 2019 Data and Data Freeze for Refreshes in 2021.). Variability analyses confirmed that HCI demonstrates sufficient ability to differentiate hospices. Hospices' scores on the HCI can range from zero to ten. During measure testing, we observed that hospices achieved scores between three and ten. In testing, 37.1 percent of hospices scored ten out of ten, 30.4 percent scored nine out of ten, 17.9 percent scored eight out of ten, 9.6 percent scored seven out of ten, and 5.0 percent scored six or lower, as shown in Figure 1.

Validity analyses showed that hospices' HCI scores align with family caregivers' perceptions of hospice quality, as measured by CAHPS Hospice survey responses (NQF endorsed quality measure #2651). Hospices with higher HCI scores generally achieve better caregiver ratings as measured by CAHPS Hospice scores, and hospices with lower HCI scores generally achieve poorer CAHPS Hospice scores. As measured by Pearson's correlation coefficients, the correlation between the CAHPS hospice overall rating and the HCI is +0.0675, and the correlation between the CAHPS hospice recommendation outcome and the HCI score is +0.0916. As such, HCI scores are consistent with CAHPS Hospice caregiver ratings, supporting the index as a valid measurement of hospice care.

We also conducted a stability analysis by comparing index scores calculated for the same hospice using claims from Federal FY 2017 and 2019. The analysis found that 82.8 percent of providers' scores changed by, at most, one point over the 2 years. These results serve as evidence of the measure's reliability by indicating that a hospice's HCI scores would not normally fluctuate a great deal from one year to the next.

d. Stakeholder Support

A TEP convened by our measure development contractor, in April 2020, provided input on this measure. Additionally, during the summer of 2020, CMS convened five listening sessions with national hospice provider organizations to discuss the HCI concept with the goals of engaging stakeholders and receiving feedback early in the measure's development. In October 2020, our contractor convened a workgroup of family caregivers whose family members have received hospice care to provide input on this measure concept from the family and caregiver perspective. Finally, the NQF Measures Application Partnership (MAP) met on January 11, 2021 and provided input to CMS. The MAP conditionally supported the HCI for rulemaking contingent on NQF endorsement. The “2020-2021 MAP 2020 Final Recommendations” can be found at: http://www.qualityforum.org/​WorkArea/​linkit.aspx?​LinkIdentifier=​id&​ItemID=​94893.

Stakeholders were generally supportive of a quality measure based on multiple indicators using claims data for public reporting. Several hospice providers expressed support for the measure's ability to demonstrate greater variation in hospice performance than the component indicators taken individually. Hospice caregivers also welcomed the addition of new quality measures to the HQRP to better differentiate between hospices. In particular, family caregivers stated that there might be a need for several HCI indicators, such as nursing availability on weekends and average Medicare per-beneficiary spending, to be included on Care Compare as additional information.

Some stakeholders raised concerns that claims data may not adequately express the quality of care provided, and may be better suited as an indicator for program integrity or compliance issues. Hospice providers suggested that claims may lack sufficient information to adequately reflect individual patient needs or the full array of hospice practices. In particular, claims do not fully capture patients' clinical conditions, patient and caregiver preferences, or hospice activities such as telehealth, chaplain visits, and specialized services such as massage or music therapy. After much consideration of the input received, we believe the benefits of adopting the HCI outweigh its limitations. The HCI is not intended to account for all potentially valuable aspects of hospice care, nor is it expected to entirely close the information gaps presently found in the HQRP. Rather, the HCI will serve as a useful measure to add value to the HQRP by providing more information to patients and family caregivers and better empowering them to make informed health care decisions. We view the HCI as an opportunity to add value to the HQRP, augmenting the current measure set with an index of indicators compiled from currently available claims data. This will provide new and useful information to patients and family caregivers without further burden to them, or to providers.

Stakeholders also suggested several valuable exploratory analyses, improvements for the indicators presented, and ideas for eventual public display for CMS to consider. We further refined the HCI based on this feedback, focusing on those indicators with the strongest consistency with CAHPS Hospice scores and/or which quality experts have identified as salient issues for measurement and observation. We also revised and refined how the HCI will be publicly displayed on Care Compare in response to family caregiver input.

e. Form, Manner and Timing of Data Collection and Submission

The data source for this HCI measure will be Medicare claims data that are already collected and submitted to CMS. We proposed and finalizing in the rule to begin reporting this measure using existing data items no earlier than May 2022. For more details, see section (3). Publicly Report the Hospice Care Index and Hospice Visits in the Last Days of Life Claims-based Measures.

In addition, to help hospices understand the HCI and their hospice's performance, we will revise the confidential QM report to include claims-based measure scores, including agency and national rates through the Certification and Survey Provider Enhanced Reports (CASPER) or its replacement system. The QM report will also include results of the individual indicators used to calculate the single HCI score, and provide details on the indicators and HCI overall score to support hospices in interpreting the information. The HCI indicators will be available by visiting the Provider Data Catalog at https://data.cms.gov/​provider-data/​topics/​hospice-care.

We solicited public comment on the proposal to add the composite HCI measure to the HQRP starting in FY 2022. We also solicited comments on the proposal to add the HCI to the program for public reporting beginning no earlier than May 2022. We received many comments on these proposals. A summary of the comments we received regarding HCI and our responses to those comments appear below:

Comment: Several commenters expressed the importance of HCI for beneficiary and families that will give them information about care processes and add value to the available information about hospices that identifies aberrant practice when comparing hospices.

Response: We appreciate the support by comments recognizing the value HCI brings to consumers by providing more information not previously available about hospices. The HCI will add value to the HQRP by filling measurement gaps using existing data sources.

Comment: Many commenters appreciate the need for CMS to address program integrity or identify hospices with aberrant practices, and encouraged CMS to develop different measures that better reflect the holistic, interdisciplinary nature of hospice. Other comments also suggested that data already provided in PEPPER reports should not be included in HCI or that CMS should share the indicators in the PEPPER reports rather than implement the HCI quality measure to provide hospices the opportunity to implement continuous quality improvement activities.

Response: We recognize commenters' concern that HQRP measures reflect quality of care rather than program integrity issues. We believe HCI does reflect hospice quality because the HCI indicators were identified as quality issues by the Office of Inspector General,^[38,39,40] the Medicare Payment Advisory Commission,^{[41 42 43]} by peer reviewed articles, and our technical expert panel (TEP). Further, HCI like the other HQRP quality measures validates well with the CAHPS Hospice Survey “willingness to recommend”, which signifies a quality measure useful for public reporting.

We also appreciate the suggestions to include HCI indicators in PEPPER reports rather than implement HCI. However, unlike PEPPER reports that are issued to hospices to support their compliance efforts related to potential improper payments, as part of the HQRP, the HCI will become information on Care Compare that beneficiaries, caregivers, or other stakeholders may consider as they make choices about end-of-life care.

Comment: Several comments suggested that CMS differentiate circumstances in which a patient refused a service measured by the HCI from circumstances in which the hospice did not offer the service to the patient. Other comments highlighted the possible impact of claims-based measures on rural and small providers because they may not capture care in rural communities or possibly identified as an outlier due to low volume.

Response: CMS acknowledges that patients have the right to refuse hospice services, and that some refusals are expected and appropriate. CMS expects hospices to honor patient wishes on a case-by-case basis. Thus, we do not anticipate service refusals to be concentrated among particular hospices, and as such do not expect refused visits to have an outsized effect on any hospice's performance on this measure. Several existing measures, such as the HIS-based HVWDII measure and its replacement HVLDL, also do not differentiate refused visits.

We also appreciate the comments expressing concern about the impact these measures may have on small and/or rural hospices. We recognize that there are many regional variations in care delivery trends. We will monitor HCI score trends to identify whether any regional or size-based variations suggest a need for measure revision. However, population-based measures such as indicators on the HCI allow for hospice variation for an indicator while offering opportunities to earn points on other indicators. The points are earned without weighting to recognize the tradeoffs for each indicator's specifications.

Comment: Several comments recommended that CMS not implement HCI because the indicators seem to emphasize medical services, focused heavily on services provided by RNs/LPNs, or do not account for the full interdisciplinary group (for example, claims do not account for spiritual care). Some commenters questioned whether services provided by LPNs would be accounted for in the HCI indicators and many commenters requested that CMS clarify whether code 055X would be further differentiated between RN visits versus LPN visits for the indicators.

Response: We recognize that claims data do not include all the disciplines involved in the delivery of hospice care, such as the frequency and length of chaplain visits. While changing the data included in claims is outside the scope of this proposed measure, we believe that using the claims data that currently exists still provides new and useful information not currently available to patients, families, and caregivers with the existing HQRP measures. As we showed with the HVLDL claims-based measure, RN services correlate well with CAHPS data and therefore are important services to reflect hospice quality of care. The HCI serves as a useful step in addressing HQRP data gaps and providing useful information to consumers, even if it does not account for all potentially valuable data currently missing from HQRP. CMS will monitor data availability as well as measure performance, and may re-specify the measure if needed. If additional data points become available, CMS will consider modifying the measure in light of the new data. CMS' sub-regulatory Quality Measure Users' Manual on the CMS HQRP Current Measures web page will include specifications for each indicator and scoring for HVLDL, and the HIS Comprehensive Assessment measure (NQF #3235).

We appreciate the comments and request for clarification on whether LPNs are included in visits. Both RN and LPN visits are included on the hospice claim under revenue code 055X and as such, the HCI does include LPN visits for the indicator for all indicators that use revenue code 055X for consistency. This does not constitute a change to the requirements of the CoPs.

Comment: Several commenters stated that the HCI should focus on whether hospices are prepared to provide key services, rather than whether claims for those services were billed during a given reporting period. One way to approach this would be to use state survey data to identify hospices that are deficient and do not have contracts to provide GIP. This information would provide additional context to the claims data of whether a hospice provided CHC or GIP.

Response: We appreciate commenters' interest in having the HCI reflect how prepared hospices are to provide key services to patients. We believe that by measuring whether hospices actually provided CHC and GIP, the HCI will recognize the extent to which hospices both kept patients at home and recognized the need for inpatient care when necessary. In this way, these billing categories reflect actions taken to meet patients' needs during the reporting period. While we recognize the additional context that state survey data would provide, we believe the claims data used to calculate the HCI will provide valuable information to consumers on their own.

Comment: We received several comments out of scope of the proposal suggesting CMS allow for use of the spiritual care HCPCS code approved for Veteran Administration use. Some commenters requested that CMS expand billing codes for telehealth visits and recognize telehealth services within the HCI. Other commenters expressed concern that the HCI indicators do not take patient preferences into account, and that the HCI might incentivize hospices to standardize the types and amount of services provided rather than considering personal patient circumstances.

Response: We appreciate commenters' concerns that hospice providers continue to recognize and address the unique circumstances of hospice patients. At this time, the HCPCS code for spiritual care is not used on the hospice claim form (no revenue center exists to correspond to such code), and as such, cannot be applied to the HCI. Additionally, we did not propose to expand billing codes for telehealth services or patient preferences, and as such cannot include such services in the HCI. However, if additional Medicare hospice claims data points become available, we may consider modifying the measure in light of the new data. We are concerned hospices believe HCI may incentivize hospices to standardize the types or amount of services provided to patients and not individualize beneficiary care on a case-by-case basis at the end of life. CMS will continue to monitor for any aberrant behavior in regard to HCI and the care provided by hospices.

Comment: Several commenters would like more time and information to replicate the analysis for HCI. The commenters suggest a delay in publicly reporting or no earlier than May 2022, which would to allow time for internal analysis.

Response: We appreciate commenters' concerns that hospice providers do not believe they could replicate the indicators without more information. However, in the preamble of the FY 2022 Hospice proposed rule (86 FR 19700) and in this final rule is a description for each indicator including the rationale, numerator, denominator, exclusion criterion, and data sources. We believe the information provided in the proposed and final rule allows for commenters to replicate, with their own claims data, the indicators, thresholds, and points earned. The sub-regulatory Quality Measure Users' Manual will be posted on the HQRP Current Measures web page to provide measure specifications. We believe this information provides the detail needed, as with prior versions of the Quality Measure Users' Manual, to model and analyze HCI and its indicators. As discussed later in this section of the preamble, hospices will have access to preview reports in advance of publicly reporting HCI.

Comment: Many commenters offered suggestions to modify specific HCI indicators and expressed concerns about specific indicators rather than the HCI as a whole. Several commenters suggested that CMS adjust the thresholds for specific services, such as gaps in skilled nursing visits, and phase in the thresholds over time. Some commenters questioned how well the HCI differentiates between high-quality, average, and low-quality hospices. They encouraged CMS to conduct further analyses before finalizing the measure.

Response: We appreciate commenters' suggestions for modifications to the indicators, additional analyses to conduct, and requests to monitor the indicators. We also appreciate the concern that we avoid duplicating measures in the development of new measures based on assessment data, claims, or other available data sources. We conducted multiple analyses during the development of HCI to validate these indicators and determine thresholds before selecting them for inclusion in the final HCI measure. We also shared the measure concept publicly and solicited stakeholder feedback, which we considered before finalizing the measure specifications. Our analyses showed that the HCI as currently defined does differentiate between hospices, as the range of HCI scores across hospices was found to be sufficiently large to highlight very high performing hospices, as well as identify the need for improvement in others. Additionally, the distribution of HCI scores aligns with caregivers' perceptions of hospice quality. As such, we have determined that the ten HCI indicators, taken together as currently defined, reflect a holistic view of hospice performance trends during a patient's stay.

Comment: Several commenters expressed concerns that the HCI will overlap with, or be duplicative of, HOPE-based measures.

Response: We appreciate commenters' concerns regarding the administrative burden in quality reporting. Because the HCI relies on claims data that are already collected by CMS, reporting claims-based measures places no additional burden for hospice providers or other stakeholders. In addition, the HCI and HOPE will complement each other, providing related but distinct information to providers and consumers to compare hospices.

Comment: Some commenters expressed concern that the HCI will become “topped out,” with 85 percent of hospices scoring a 7 or better, limiting the measure's ability to differentiate between hospices.

Response: We appreciate commenters' concerns that HQRP measures will not be able to adequately differentiate hospices if they become “topped out.” We also understand why commenters might expect process measures to be prone to “topping out.” CMS has taken this into consideration in designing the HCI measure. The design of the HCI ensures that the measure is very unlikely to become topped out. Each HCI indicator is scored based on comparative performance, with hospices receiving a point based on their performance relative to a national percentile threshold. Using percentile rankings derived from national performance, it is very unlikely for all hospices to receive the same score. Our analyses suggest that the scoring criteria ensure distributions of HCI scores that allow for differentiation between hospices in any given year. However, CMS will continue to monitor the HCI after implementation to ensure the measure reflects hospice quality, differentiates between hospices, and does not become topped out.

Final Decision: We are finalizing the proposal to add composite HCI measures to the HQRP as of FY 2022 and will monitor the measure. As discussed later in this section of the preamble, we will publicly report no earlier than May 2022.

4. Update on the Hospice Visits in the Last Days of Life (HVLDL) and Hospice Item Set V3.00

On August 13, 2020, we sought public comment in an information collection request to remove Section O “Service Utilization” (hereafter referred to as Section O) of the HIS discharge assessment. Removal of Section O is the sole change from HIS V2.01 and in effect eliminate the HVWDII quality measure pair. In Paperwork Reduction Act package (PRA), CMS-10390 (OMB control number: 0938-1153), we provided the HVLDL specifications and also proposed to replace the HVWDII measure pair with the HVLDL. This means that we will no longer report HVWDII with patient stays and will start publicly reporting HVLDL no earlier than May 2022. The Office of Management and Budget (OMB) approved the collection of information to remove Section O of the HIS expiring on February 29, 2024, (OMB Control Number: 0938-1153, CMS-10390). We direct the public to review the PRA at https://www.cms.gov/​regulations-and-guidancelegislationpaperworkreductionactof1995pra-listing/​cms-10390 and HVWDII report at https://www.cms.gov/​files/​document/​hqrphospice-visits-when-death-imminent-testing-re-specification-reportoctober-2020.pdf. As a claims-based measure, the HVLDL measure would not impose any new requirements for the collection of information.

The HVLDL measure, as a replacement, will continue to fill an important area in hospice care previously filled by the HVWDII measure pair. We discussed the analysis with a TEP convened by our measure development contractor in November 2019 and with the MAP, hosted by the NQF in December 2019 ^[44] for inclusion in the HQRP. During these meetings, the discussions reflecting on the analysis generally supported the replacement of HVWDII with a claims-based HVLDL measure. The November 2019 TEP report can be found in the downloads section at Hospice QRP Provider Engagement Opportunities and final recommendations and presentation of the HVLDL measure before NQF's MAP can be found at Quality Forum—Post-Acute Care, https://www.qualityforum.org/​Publications/​2020/​02/​MAP_​2020_​Considerations_​for_​Implementing_​Measures_​Final_​Report_​-_​PAC_​LTC.aspx.

OMB approved the proposal to replace the HVWDII measure with the HVLDL measure and remove Section O from the discharge assessment on February 16, 2021. The HIS V3.00 became effective on February 16, 2021 and expires on February 29, 2024; OMB control number 0938-1153.

We received several comments regarding the updates to the Hospice Visits in the Last Days of Life (HVLDL) and Hospice Item Set V3.00. A summary of the comments we received and our responses those comments are below:

Comment: Several comments support the re-specified HVLDL claims-based measure and the resulting reduction of burden, but expressed concern that the measure is limited to RN and medical social worker. Commenters stated that the measure should recognize the full spectrum of disciplines involved in hospice care. Some commenters requested that LPNs count for the measure, in addition to RNs. Other commenters stated that chaplain or spiritual services may be as important to patients as nursing services.

Response: As discussed in the CMS-10390 Supporting Statement published October, 23, 2020 and HIS V3.00 approved by OMB on February 16, 2021, we pursed a re-specification of the HVWDII measure concept using Medicare claims data because claims data also capture RN and medical social worker visits by hospice. While CMS agrees that all patient visits are meaningful, based on our analyses, we found that RN and medical social worker visits correlate well with the CAHPS quality measures for “would recommend” the hospice. HVLDL indicates the hospice provider's proportion of patients who have received visits from an RN or medical social worker (in-person) on at least two out of the final three days of the patient's life. While all patient visits are meaningful, only patients with visits on two different days during the last three days of life will count towards the numerator for this measure. These visits can be made by either the RN, the medical social worker, or both. We were interested in re-specifying the visit measure to better align with the SIA because, as we discussed in previous rules, patient needs typically surge as the end of life approaches and more intensive services are warranted. The provision of care would proportionately escalate to meet the increased clinical, emotional, and other needs of the patient and family.

Comment: Some commenters stated that the measure specifications would not adequately capture hospices' care activities. Some commenters suggested that the measure should allow for two visits occurring on the same day to meet the measure qualifications, as visits on the same day could address different patient needs, representing meaningful care on the part of the hospice. Other commenters requested that this measure recognize visits offered during CHC or GIP care. Some commenters stated that the measure should recognize telehealth visits in the last days of life, as circumstances such as the recent COVID-19 PHE may make in-person visits impossible or undesirable for patients or families.

Response: We agree that hospice care is interdisciplinary care delivered by clinical and non-clinical staff supporting the patient's plan of care. We also support hospices providing necessary visits in the last days of life such that two visits occurring on the same day may be necessary. However, as discussed in the CMS-10390 Supporting Statement published October 23, 2020 and HIS V3.00 approved by OMB on February 16, 2021, our analysis comparing HVWDII and HVLDL with CAHPS “would recommend” scores demonstrates that HVLDL results in higher validity and variability testing results compared to HVWDII. We found a stronger correlation coefficient with CAHPS “would recommend” scores for HVLDL than for HVWDII. This means that when visits by RNs or medical social workers occurred in at least two of the last three days of life, family and caregivers agree or positively correlate that they would recommend the hospice, more often when compared to HVWDII, on average. The literature strongly supported the focus on RNs and medical social workers in the revised measure.

Actively dying is a critical and unique time when in-person, skilled care is typically needed. HVLDL is defined for in-person visits. As with all quality measures, we are encouraging quality of care and as such hospices are expected to use in-person visits when visits are needed during these critical last days of life. We agree there are benefits to telehealth visits that supplement, not replace, in-person visits. If claims data are revised to include other disciplines, we may consider whether to include them in this measure. This measure does not recognize visits during CHC and GIP because these higher levels of care inherently require skilled visits per the COPs in accordance with § 418.110 and § 418.302.

Comment: Several comments requested that CMS clarify how “the last three days of life” would be calculated. Commenters expressed concern that definitions were unclear.

Response: The exclusion criteria used for HVWDII and now HVLDL criteria remain the same. The calculation of the last three days remain unchanged from the last three days documented in Section O of the HIS V2.00 that was used to calculate the HVWDII. Information defining the last three days has been included in the HIS Manuals since 2017. These specifications will now be contained in the revised HQRP QM User's Manual V4.00 located on the CMS HQRP Current measures web page. This information was also posted in the document “Common Questions HQRP Claims-Based Measures_Feb.2021” located in the Downloads section of the Hospice Item Set web page at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​Hospice-Item-Set-HIS.

Specifically these three days are “indicated by the day of death, the day prior to death, and two days prior to death.” The day of death is the same as the date provided in A0270, Discharge Date. (or the day of death); One day prior to death is calculated as A0270 minus 1, and two days prior to death is calculated as A0270 minus 2. Full HVLDL specifications are also publicly available on the HQRP website at: https://www.cms.gov/​files/​document/​hospice-visits-last-days-life-hvldl-measure-specifications.pdf.

5. Proposal To Revise § 418.312(b) Submission of Hospice Quality Reporting Program Data

To address the inclusion of administrative data, such as Medicare claims used for hospice claims-based measures like the HVLDL and HCI in the HQRP and correct technical errors identified in the FY 2016 and 2019 Hospice Wage Index and Payment Rate Update final rules, we proposed and finalize in this rule the regulation at § 418.312(b) by adding paragraphs (b)(1) through (3). Paragraph (b)(1) will include the existing language on the standardized set of admission and discharge items. Paragraph (b)(2) would require collection of Administrative Data, such as Medicare claims data, used for hospice quality measures to capture services throughout the hospice stay. And these data automatically meet the HQRP requirements for § 418.306(b)(2).

Paragraph (b)(3) is a technical correction to address errors identified in the FY 2016 and FY 2019 Hospice Wage Index and Payment Rate Update final rules, (80 FR 47186 and 83 FR 38636). In the FY 2016 Hospice final rule (80 FR 47186) adopted seven factors for measure removal, and in the FY 2019 Hospice final rule (83 FR 38636) adopted the eighth factor for measure removal. In those final rules, we referenced the measure removal factors in the preamble but inadvertently omitted them from the regulations text. Thus, these measure removal factors identify how measures are removed from the HQRP. Section 418.312(b)(3) would include the eight measure removal factors as follows:

CMS may remove a quality measure from the Hospice QRP based on one or more of the following factors:

(1) Measure performance among hospices is so high and unvarying that meaningful distinctions in improvements in performance can no longer be made.

(2) Performance or improvement on a measure does not result in better patient outcomes.

(3) A measure does not align with current clinical guidelines or practice.

(4) The availability of a more broadly applicable (across settings, populations, or conditions) measure for the particular topic.

(5) The availability of a measure that is more proximal in time to desired patient outcomes for the particular topic.

(6) The availability of a measure that is more strongly associated with desired patient outcomes for the particular topic.

(7) Collection or public reporting of a measure leads to negative unintended consequences other than patient harm.

(8) The costs associated with a measure outweigh the benefit of its continued use in the program.

We did not receive comments on this proposal. We are finalizing in this rule the regulation at § 418.312(b) to add paragraphs (b)(1) through (3) to include administrative data as part of the HQRP, and correct technical errors identified in the FY 2016 and 2019 Hospice Wage Index and Payment Rate Update final rules.

6. Update Regarding the Hospice Outcomes & Patient Evaluation (HOPE) Development

As finalized in the FY 2020 Hospice Wage Index and Payment Rate Update and Hospice Quality Reporting Requirements final rule (84 FR 38484), we are developing a hospice patient assessment instrument identified as HOPE. This tool is intended to help hospices better understand care needs throughout the patient's dying process and contribute to the patient's plan of care. It will assess patients in real-time, based on interactions with the patient. HOPE will support quality improvement activities and calculate outcome and other types of quality measures in a way that mitigates burden on hospice providers and patients. Our two primary objectives for HOPE are to provide quality data for the HQRP requirements through standardized data collection, and to provide additional clinical data that could inform future payment refinements.

We anticipate that HOPE will replace the HIS. While the HIS is a standardized mechanism for abstracting medical record data, it is not a patient assessment tool because HIS data are not collected during a patient assessment. HIS data collection “consists of selecting responses to HIS items in conjunction with patient assessment activities or via abstraction from the patient's clinical record.” (HIS Manual v.2.01). In contrast, HOPE is a patient assessment instrument, designed to capture patient and family care needs in real-time during patient interactions throughout the patient's hospice stay, with the flexibility to accommodate patients with varying clinical needs. HOPE will enable CMS and hospices to understand the care needs of people through the dying process, supporting provider care planning and quality improvement efforts, and ensuring the safety and comfort of individuals enrolled in hospice nationwide. HOPE will include key items from the HIS and demographics like gender and race. This approach to include key aspects of demographics supports hospice feedback provided in the FYs 2017 and 2018 Hospice Wage Index and Payment Rate Update final rule (81 FR 52171 and 82 FR 36669) and CMS' goals for a hospice assessment instrument, as stated in the FY 2018 Hospice Wage Index and Payment Rate Update final rule. The HOPE assessment instrument would facilitate communication among providers and measure the care of patient populations across settings. While the standardized patient assessment data elements for certain post-acute care providers required under the IMPACT Act of 2014 is not applicable to hospices, it makes reasonable sense to include some of those standardized elements that appropriately and feasibly apply to hospice. Some patients may move through the healthcare system to hospice so capturing and tracking key demographic and social risk factor items that apply to hospice may help CMS achieve our goals for continuity of care, overall patient care and well-being, interoperability, and health equity that are also discussed in this rule.

The draft of HOPE has undergone cognitive and pilot testing, and will undergo field testing to establish reliability, validity, and feasibility of the assessment instrument. We anticipate proposing HOPE in future rulemaking after testing is complete.

We will continue development of HOPE in accordance with the Blueprint for the CMS Measures Management System. Development of HOPE is grounded in extensive information gathering activities to identify and refine hospice assessment domains and candidate assessment items. We appreciate the industry's and national associations' engagement in providing input through information sharing activities, including listening sessions, expert interviews, key stakeholder interviews, and focus groups to support HOPE development. As CMS proceeds with field testing HOPE, we will continue to engage with stakeholders through sub-regulatory channels. In particular, we will continue to host HQRP Forums to allow hospices and other interested parties to engage with us on the latest updates and ask questions on the development of HOPE and related quality measures. We also have a dedicated email account, [e-mail], for comments about HOPE. We will use field test results to create a final version of HOPE to propose in future rulemaking for national implementation. We will continue to engage all stakeholders throughout this process. We appreciate the support for HOPE and reiterate our commitment to providing updates and engaging stakeholders through sub-regulatory means. Future updates and engagement opportunities regarding HOPE can be found at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​HOPE.html.

We received many comments about the HOPE update. A summary of these comments and our responses appear below:

Comments: Several commenters encouraged CMS to thoughtfully consider the implementation timeline for HOPE and the collection demographic and social risk factor data. The comments pointed out that the process for providers to adapt to the new tool requires at least 6 months or more. They noted the implementation of a new assessment instrument would be burdensome on both providers and EMR vendors.

Several commenters noted the potential for overlap in quality measures from HOPE and HCI or future measures. They encouraged CMS to eliminate any duplicative measures from HCI and HOPE, and to consider using HOPE data as the source for publicly reported information once it is implemented.

Response: We thank commenters for raising points for CMS to consider in advance of HOPE implementation. We appreciate commenters' concern for provider and vendor burden in implementing a new tool and encourage all key stakeholders to continue to stay informed and engaged through the HQRP Forums, Quarterly Updates, and listserv notifications.

7. Update on Quality Measure Development for Future Years

In the FY 2017 Hospice Wage Index and Payment Rate Update final rule (81 FR 52160), we finalized new policies and requirements related to the HQRP, including how we would provide updates related to the development of new quality measures. Information on the current HQRP quality measures can be found at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​Current-Measures. In this proposed rule, we are continuing to provide updates for both HOPE-based and claims-based quality measure development.

To support new measure development, our contractor convened TEP meetings in 2020 to provide feedback on several measure concepts. In 2020, the TEP explored potential quality measure constructs that could be derived from HOPE and their specifications. Specifically, for HOPE-based measure development, the TEP focused on pain and other symptom outcome measure concepts that could be calculated from HOPE. Input from initial TEP workgroups held in spring 2020 informed follow-up information-gathering activities related to pain in general and neuropathic pain in particular. The 2020 Information Gathering Summary report is available at https://www.cms.gov/​files/​document/​12042020-information-gathering-oy1508.pdf. During fall 2020, the TEP reviewed measure concepts focusing on pain and symptom outcomes that could be calculated from HOPE items.

The TEP supported further exploration and development of these measures. As described in the 2020 TEP Summary Report, the TEP generally supports the following measure concepts that are calculated using HOPE items: Timely Reduction of Pain Impact, Reduction in Pain Severity, and Timely Reduction of Symptoms. The candidate measure Timely Reduction of Pain Impact reports the percentage of patients who experienced a reduction in the impact of moderate or severe pain. HOPE items assessing Symptom Impact, and Patient Desired Tolerance Level for Symptoms or Patient Preferences for Symptom Management were used to calculate this measure. The candidate measure Reduction in Pain Severity reports the percentage of patients who had a reduction in reported pain severity. The primary HOPE items used to calculate this measure include Pain Screening, Pain Active Problem, and Patient Desired Tolerance Level for Symptoms or Patient Preferences for Symptom Management. The last candidate measure discussed by the TEP was Timely Reduction of Symptoms which measures the percentage of patients who experience a reduction in the impact of symptoms other than pain. HOPE items assessing Symptom Impact, and Patient Desired Tolerance Level for Symptoms or Patient Preferences for Symptom Management were used to calculate this measure. HOPE items for all three measure are collected at multiple time points across a patient's stay, including at Admission, Symptom Reassessment, Level of Care Change, and Recertification. Overall, the TEP supported each candidate measure and agreed that they were viable for distinguishing hospice quality. We continue to develop all three candidate quality measures.

We are interested in exploring patient preferences for symptom management, addressing patient spiritual and psychosocial needs, and medication management in outcomes of care in development of quality measures. We sought public comment on methods, instruments, or brief summaries on hospice quality initiatives related to goal attainment, patient preferences, spiritual needs, psychosocial needs, and medication management.

Information about the TEP feedback on these quality measures concepts and future measure concepts can be obtained via: https://www.cms.gov/​files/​document/​2020-hqrp-tep-summary-report.pdf. Related to the outcome measures and in order to have HOPE pain and symptom measures in the program as soon as possible, we plan to develop process measures, including on pain and symptom management. These process measures may support or complement the outcome measures. We solicit comments on current HOPE-based quality measure development and recommendations for future process and outcome measure constructs.

In the FY 2020 Hospice Wage Index and Payment Rate Update final rule (84 FR 38484) and as discussed later in this section of the preamble, we are interested in claims-based quality measures in order to leverage the multiple data sources currently available to support quality measure development. Specifically, we intend to develop additional claims-based measures that may enable beneficiaries and their family caregivers to make more informed choices about hospice care and to hold hospices more accountable for the care they provide. As discussed in this section, the HVLDL and HCI claims-based measures support the Meaningful Measures initiative and address gaps in HQRP. Additional claim-based measure concepts we are considering for development include hospice services on weekends, transitions after hospice live discharge, Medicare expenditures per beneficiary (including the share of non-hospice spending during hospice election, and the share for hospice care prior to the last year of life), and post-mortem visits as measures of hospice quality. We intend to submit additional claims-based measures for future consideration and solicit public comment.

We solicited public comment on the aforementioned HOPE- and claims-based quality measures to distinguish between high- and low-quality hospices, support healthcare providers in quality improvement efforts, and provide support to hospice consumers in helping to select a hospice provider. We also solicited public comment on how the candidate measures may achieve those goals.

We are also considering developing hybrid quality measures that would be calculated using claims, assessment (HOPE), or other data sources. Hybrid quality measures allow for a more comprehensive set of information about care processes and outcomes than cannot be calculated using claims data alone. Assessment data can be used to support risk-adjustment. We sought public comment on quality measure concepts and considerations for developing hybrid measures based on a combination of data sources.

We received many comments on future quality measure development aspects. A summary of these comment and our responses to those comments appear below:

Comment: We received several comments suggesting concepts for future quality measures in the HQRP such as measures related to postmortem service, plan of care goal achievement, spiritual care, psychosocial care, veteran services, volunteer activities, visit activity at the time of admission, change of level of care, change of physical location, safety culture, and workforce engagement, and patient and family care needs. Comments urge CMS to monitor duplication of measures when HOPE-based and other future measures are under development. Many commenters emphasized the need to engage providers to share information and for CMS to seek feedback when developing quality measures.

We received many comments expressing the need for HCPCS codes for all hospice disciplines, including spiritual care professionals. These comments also suggested including these disciplines in future claims-based measures to recognize the multi-disciplinary nature of hospice care.

Many commenters noted their concern about the distinction between performance measures and quality of care measures. Commenters emphasized that performance measures should be used to measure program integrity, but should not be publicly reported. Several commenters encouraged CMS to use quality claims-based data and other data sources for hybrid measure, consider the implications of claims-based measures to measure quality, use of survey data if feasible, explore outcome measures related to pain and other symptom management, and explore goal achievement. Several comments suggest CMS explore statewide or regional approaches to measure quality rather than using national analysis and perform rigorous data validation by hospice providers for claims-based measures.

Response: We thank all the commenters for their thoughtful suggestions and feedback related to future of quality measure development for the HQRP. We appreciate suggestions for new quality measures, as well as comments about the public reporting of quality measures. CMS will take these comments under advisement for future consideration of quality measures and the Meaningful Measures System Blueprint. We encourage all key stakeholders to continue to stay informed and engaged through the HQRP Forums, Open Door Forums, Quarterly Updates, and listserv notifications.

8. CAHPS Hospice Survey Participation Requirements for the FY 2023 APU and Subsequent Years

a. Background and Description of the CAHPS Hospice Survey

The CAHPS Hospice Survey is a component of the CMS HQRP which is used to collect data on the experiences of hospice patients and the primary caregivers listed in their hospice records. Readers who want more information about the development of the survey, originally called the Hospice Experience of Care Survey, may refer to 79 FR 50452 and 78 FR 48261. National implementation of the CAHPS Hospice Survey commenced January 1, 2015 as stated in the FY 2015 Hospice Wage Index and Payment Rate Update final rule (79 FR 50452).

b. Overview of the “CAHPS Hospice Survey Measures”

The CAHPS Hospice Survey measures was re-endorsed by NQF on November 20, 2020. The re-endorsement can be found on the NQF website at: https://www.qualityforum.org/​Measures_​Reports_​Tools.aspx. Use the QPS tool and search for NQF number 2651. The survey received its initial NQF endorsement on October 26, 2016 (NQF #2651). We adopted 8 survey based measures for the CY 2018 data collection period and for subsequent years. These eight measures are publicly reported on a designated CMS website, Care Compare, https://www.medicare.gov/​care-compare/​.

c. Data Sources

We previously finalized the participation requirements for the CAHPS Hospice Survey, (84 FR 38484). We propose no changes to these requirements going forward.

d. Public Reporting of CAHPS Hospice Survey Results

We began public reporting of the results of the CAHPS Hospice Survey on Hospice Compare as of February 2018. Prior to the COVID-19 PHE, we reported the most recent 8 quarters of data on the basis of a rolling average, with the most recent quarter of data being added and the oldest quarter of data removed from the averages for each data refresh. Given the exemptions provided due to COVID-19 PHE in the March 27, 2020 Guidance Memorandum,^[45] public reporting will continue to be the most recent 8 quarters of data, excluding the exempted quarters; Quarter 1 and Quarter 2 of CY 2020. More information about this is detailed in the section entitled: Proposal for Public Reporting CAHPS-based measures with Fewer than Standard Numbers of Quarters Due to the COVID-19 PHE Exemptions

e. Volume-Based Exemption for CAHPS Hospice Survey Data Collection and Reporting Requirements

We previously finalized a volume-based exemption for CAHPS Hospice Survey Data Collection and Reporting requirements for FY 2021 and every year thereafter (84 FR 38526).

We propose no changes to this exemption. The exemption request form is available on the official CAHPS Hospice Survey website: http://www.hospiceCAHPSsurvey.org. Hospices that intend to claim the size exemption are required to submit to CMS their completed exemption request form by December 31, of the data collection year.

Hospices that served a total of fewer than 50 survey-eligible decedent/caregiver pairs in the year prior to the data collection year are eligible to apply for the size exemption. Hospices may apply for a size exemption by submitting the size exemption request form. The size exemption is only valid for the year on the size exemption request form. If the hospice remains eligible for the size exemption, the hospice must complete the size exemption request form for every applicable FY APU period, as shown in table 9.

f. Newness Exemption for CAHPS Hospice Survey Data Collection and Public Reporting Requirements

We previously finalized a one-time newness exemption for hospices that meet the criteria as stated in the FY 2017 Hospice Wage Index and Payment Rate Update final rule (81 FR 52181). In the FY 2019 Hospice Wage Index and Payment Rate Update final rule (83 FR 38642), we continued the newness exemption for FY 2023, and all subsequent years. We encourage hospices to keep the letter they receive providing them with their CMS Certification Number (CCN). The letter can be used to show when you received your number.

g. Survey Participation Requirements

We previously finalized survey participation requirements for FY 2022 through FY 2025 as stated in the FY 2018 and FY 2019 Hospice Wage Index and Payment Rate Update final rules (82 FR 36670 and 83 FR 38642 through 38643). We also continued those requirements in all subsequent years (84 FR 38526). Table 10 restates the data submission dates for FY 2023 through FY 2025.

For further information about the CAHPS Hospice Survey, we encourage hospices and other entities to visit: https://www.hospiceCAHPSsurvey.org. For direct questions, contact the CAHPS Hospice Survey Team at [e-mail] or call 1-[phone].

h. Proposal to Add CAHPS Hospice Survey Star Ratings to Public Reporting

CMS currently publishes CAHPS star ratings for several of its public reporting programs including Home Health CAHPS and Hospital CAHPS. The intention in doing so is to provide a simple, easy to understand, method for summarizing CAHPS scores. Star ratings benefit the public in that they can be easier for some to understand than absolute measure scores, and they make comparisons between hospices more straightforward. The public's familiarity with a 1 through 5 star rating system, given its use by other programs, is also a benefit to using this system.

In the proposed rule, we proposed to introduce Star Ratings for public reporting of CAHPS Hospice Survey results on the Care Compare or successor websites no sooner than FY 2022. We proposed that the calculation and display of the CAHPS Hospice Survey Star Ratings be similar to that of other CAHPS Star Ratings programs such as Hospital CAHPS and Home Health CAHPS. The stars would range from one star (worst) to five stars (best). We proposed that the stars be calculated based on “top-box” scores for each of the eight CAHPS Hospice Survey measures. Specifically, individual-level responses to survey items would be scored such that the most favorable response is scored as 100 and all other responses are scored as 0. A hospice-level score for a given survey item would then be calculated as the average of the individual-level responses, with adjustment for differences in case mix and mode of survey administration. For a measure composed of multiple items, the hospice-level measure score would be the average of the hospice-level scores for each item within the measure. Similar to other CAHPS programs, we proposed that the cut-points used to determine the stars be constructed using statistical clustering procedures that minimize the score differences within a star category and maximize the differences across star categories.

We proposed to use a two-stage approach to calculate these cut-points. In the first stage, we would determine initial cut-points by calculating the clustering algorithm among hospices with 30 or more completed surveys over 2 quarters (that is, 6 months); restricting these calculations to hospices that meet a minimum sample size promotes stability of cut-points. Depending on whether hospices that meet this minimum sample size have different score patterns than smaller hospices, the initial cut-points may be too high or too low. To ensure that cut-points reflect the full distribution of measure performance, in the second stage, we proposed to compare mean measure scores for the bigger hospices used in the first stage to all other hospices, and update cut-points by adjusting the initial cut-points to reflect the normalized difference between bigger and smaller hospices. This two-stage approach allows for calculation of stable cut-points that reflect the full range of hospice performance. We proposed that hospice star ratings for each measure be assigned based on where the hospice-level measure score falls within these cut-points.

We further proposed to calculate a summary or overall CAHPS Hospice Survey Star Rating by averaging the Star Ratings across the 8 measures, with a weight of 1/2 for Rating of the Hospice, a weight of 1/2 for Willingness to Recommend the Hospice, and a weight of 1 for each of the other measures, and then rounding to a whole number. We proposed that only the overall Star Rating be publicly reported and that hospices must have a minimum of 75 completed surveys in order to be assigned a Star Rating. Finally, we proposed to publish the details of the Star Ratings methodology on the CAHPS Hospice Survey website, www.hospicecahpssurvey.org. CMS requires no additional resources to create and display CAHPS star ratings.

We solicited comments on these proposals for CAHPS Star Ratings and the public reporting of star ratings no sooner than FY 2022.

Comment: Many commenters expressed concern about the timeframe for implementing CAHPS Hospice Survey star ratings. They suggested that the display of star ratings be delayed because CMS needs to provide additional opportunities for providers to learn about and comment on the details of the methodology. In addition, some commenters wanted CMS to consider creating a single star rating based on both CAHPS and other measures, such as the HOPE tool.

Response: As stated in the proposed rule, we will display CAHPS Hospice Survey star ratings no sooner than FY 2022. Prior to finalizing a timeline, CMS will provide multiple opportunities to share information and receive comments from stakeholders. This could include a special open door forum or other venues for interaction. CMS proposed a CAHPS-only star rating since other portions of Care Compare also display a CAHPS-only star rating (for example, Hospital CAHPS and Home Health CAHPS). We will take the recommendation of a single star rating into consideration for the future.

Comment: A few commenters requested specifically for an explanation for using top-box scoring of individual level responses for the star ratings. They note that other star ratings use a 0-100 linear-scaled score.

Response: CMS analyzed existing data to inform the development of star ratings in the hospice setting. We examined star ratings using linear means and, separately, top-box scores. For CAHPS Hospice Survey data, using top-box scores resulted in wider star rating categories that make the star ratings less sensitive to small changes in scores. For this reason, we proposed to calculate CAHPS Hospice star ratings using top-box scores.

Comment: Several commenters raised a concern regarding whether relatively high levels and tight distribution of performance on CAHPS Hospice Survey measures will result in hospices with high scores receiving 3 or fewer stars. Some commenters were concerned about the comparative nature of CAHPS star ratings and a few called for an alternative methodology that would rate hospices against a benchmark.

Response: Our analyses of existing CAHPS Hospice Survey data demonstrate that hospices with high scores would overwhelmingly receive 4 and 5 stars. Clustering methodology assigns cut points by minimizing differences within star categories and maximizing differences across star categories. This methodology does not force a set number of hospices into each star category. Using a benchmark rather than the clustering approach represents a major shift from our current practice. The current methodology has been successful for other provider types. We do not believe it is necessary to drastically change our methodology for the CAHPS Hospice Survey.

Comment: Some commenters raised questions about using 75 completed surveys as the threshold for public reporting of stars. They were concerned that this number is nearly double the number of survey responses required from home health agencies (40 completes) and more than double the number of responses a hospice must currently have for CAHPS® Hospice Survey measures to be reported (30 completes). They requested a justification for using this number. One commenter stated that given the survey response rate, a hospice would have more than 200 completed surveys in order for star ratings to be displayed. This was a concern for many commenters because it would mean that star ratings would be available only for large hospices. Some commenters suggested that CMS formulate a methodology that would include smaller hospices in star ratings. Additionally, several commenters noted that the proposed rule does not state how many hospices will meet the 75 completes threshold.

Response: CMS seeks to balance the goal of reporting star ratings for as many hospices as possible with the need to ensure that the star ratings can be stably estimated and distinguish between hospices' performance. If a hospice does not have enough survey completes to reliably measure performance, the star ratings would be picking up more noise than true performance. Our analyses have determined that the optimal balance between these two goals is at 75 completed surveys per hospice. We expect that approximately 70 percent of hospices with publicly reported CAHPS Hospice Survey measure scores meet the threshold of 75 completed surveys.

Comment: Several commenters expressed concerns that the public will not interpret the star ratings correctly. They also called for more explanatory information on the Care Compare website.

Response: The star rating approach proposed for CAHPS Hospice Survey measures is similar to what has been used for Medicare Advantage and Part D plan measures and Hospital CAHPS measures successfully for many years. These other settings utilize a clustering algorithm such that providers within a cluster are more alike than providers across clusters. The proposed CAHPS Hospice Survey stars will adopt a similar overall approach, although using top-box scores rather than linear means, based on our analyses of existing data. Consumers have generally welcomed star ratings. We will make explanatory information available to consumers, while recognizing that keeping the interface as streamlined as possible improves the usability of the site for consumers.

Comment: Several commenters stated concerns that the public might misinterpret the lack of star ratings for smaller hospices as being evidence of poor quality care. They called for customer research on how the public would interpret the absence of star ratings as well as research on the extent to which the public understands how star ratings are calculated.

Response: Star Ratings are easy for consumers to understand and interpret and are used in a variety of settings. We will explore alternatives for presenting additional information about star ratings on the Care Compare website so that consumers may be informed about why smaller hospices may not have stars.

Comment: A few commenters requested more details about if and how we will include patient-mix adjustment.

Response: Star ratings are based on CAHPS Hospice Survey measure scores, which are adjusted for case mix and mode of survey administration. Detailed information regarding adjustment of measure scores is available at https://hospicecahpssurvey.org/​en/​scoring-and-analysis/​.

Comment: Several commenters raised issues about the eight quarters of data included in public reporting. They believe that this is too long and that it makes it difficult for hospices to use publicly-reported data for quality improvement.

Response: CMS seeks to balance the goal of publicly reporting measure scores for as many hospices as possible with the need to ensure that measure scores can be stably estimated and distinguish between hospices' performance. Rolling up eight quarters of data instead of four ensures that measure scores are available for many more hospices, which improves the usefulness of the Compare web tools for hospice consumers. The eight quarter approach does not result in a delay of when data become available (since the most recent quarters of data are included in the rolled-up score), but it does ensure more accurate measurement. The decision to use eight quarters of rolling data for hospices reflects the size of hospices, which differ in size and other dimensions from other types of entities, such as hospitals and Medicare Advantage contracts, for which CMS publicly reports scores and star ratings. We note that hospices should be able to receive timely reports and data directly from their survey vendors. We encourage hospices who want to use CAHPS data for quality improvement to talk to their vendors about the reports and data that may be available shortly after data collection.

Comment: A commenter stated that the preview report timeframe is too short and that hospices should receive preview data at least 1 year prior to its publication in order to analyze performance and implement quality improvement.

Response: As stated previously, we recommend that hospices use data from their vendors for quality improvement, rather than wait for publicly-reported data. If we were to provide preview data a year in advance, the publicly reported data would be too old to be a meaningful reflection of the hospice's performance. We believe additional delays in public reporting of data is not in the interest of the public using Care Compare.

Comment: Many commenters expressed concern about publicly reporting data that was collected and/or delivered during the COVID-19 PHE. They commented that these data could be skewed by the public health emergency.

Response: We will not include data from Q1 and Q2 2020 in Star Rating calculations, as hospices were exempted from submitting these quarters of data to CMS due to the COVID-19 PHE.

Comment: Several commenters stated that the CAHPS Hospice Survey is unlike other CAHPS surveys in that the respondents are family members or friends of the deceased—not the patients themselves. They believe this is a key difference between the hospice survey and other CAHPS surveys and called for more information on the Care Compare site to make sure consumers are not misled.

Response: Although Care Compare already notes that for Hospice CAHPS the user is comparing “. . . hospices based on results from a national survey that asks a family member or friend of a hospice patient about their hospice care experience,” we will consider whether there are additional ways to highlight this.

Comment: Some commenters objected to the comparative nature of the CAHPS Hospice Survey star ratings, preferring instead, a rating based upon an external criteria rather than one that compares hospices to each other. As a few commenters noted, “Each hospice is afforded the opportunity to achieve excellent ratings on the CAHPS Hospice Survey. Similarly, this same right should be afforded hospices under the Star Rating system through a clear portrayal Star Rating of performance to consumers and the public that reflects how most respondents scored the hospice, not how the hospice fares compared to all other hospices.” One commenter also suggested that star ratings calculations be made available to hospices before they are publicly reported.

Response: Similar to other CMS CAHPS star ratings, we propose that the cut-points used to determine CAHPS Hospice Survey stars be constructed using statistical clustering procedures that minimize the score differences within a star category and maximize the differences across star categories. This ensures that star assignments clearly differentiate performance across groups of hospices. Such comparative star ratings, as proposed by CMS, help consumers identify high and low performing hospices. With respect to making calculations available before they are publicly reported, we do plan to provide star ratings calculations in preview reports prior to their display.

Comment: Several commenters noted that CMS is currently conducting a pilot test of a revised CAHPS Hospice Survey questionnaire and wondered whether the release of a new questionnaire would coincide with the introduction of star ratings. They also questioned whether CMS expected that use of a revised questionnaire would increase the number of hospices that achieve 75 completed questionnaires and would, therefore, be included in star ratings.

Response: We are currently conducting an experiment to test a new version of the survey, including the web mode of administration which may have an impact on response rates and the number of survey completes. Results of this experiment will help to inform changes to the survey in the future. We anticipate that star ratings will be released prior to a new version of the survey. Star ratings will continue to be calculated and released as we phase in the new survey version.

Comment: Many commenters questioned the weighting of the components of the star ratings, particularly the decision to weigh the two global questions (Overall Rating and Willingness to Recommend) at 50 percent of the weight for each composite measure.

Response: The Willingness to Recommend and Overall Rating measures are highly correlated with one another, as both provide global assessments of hospice care. Given this, weighting each of the two measures at 100 percent would over-emphasize global assessments of care relative to the other aspects of care assessed by CAHPS Hospice Survey measures. CMS maintains its proposal to weight Willingness to Recommend and Overall Rating at 50 percent each for the purpose of calculating an overall CAHPS Hospice Survey star rating. This approach parallels the one used by CMS for calculating star ratings for hospitals.

Comment: A few commenters questioned whether it is CMS's intent for the CAHPS® to be the sole star rating vehicle for hospice care or whether there would be another star rating for HOPE measures when it is implemented?

Response: The FY 2022 proposal contemplated a CAHPS-only measure in the short-term. At this time, it is premature to determine whether the HOPE tool should be used to create star ratings, either separately from CAHPS or in combination with CAHPS. The HOPE tool is now under development. We will consider other star ratings as applicable.

Comment: One commenter recommended that CMS award star ratings in FY 2022, but suppress public reporting in Care Compare until the August 2023 refresh when all the data will be after the COVID-exempted quarters.

Response: As mentioned previously, we plan to display stars no sooner than FY 2022. We will take into consideration the option of starting the stars display when all data will be after the COVID-exempted quarters.

Comment: One commenter strongly suggested that there should be a “not applicable” response option available for each question in the questionnaire. Indeed, they noted that “Questions such as “How often did your family member get the help he or she needed for trouble breathing” or “How often did your family member get the help he or she needed for constipation” are difficult for family members to answer if their loved one did not experience issues with those symptoms.”

Response: On the questionnaire, the respondent is asked if their family member experienced the symptom. If they did not experience the symptom, the instructions say to skip to another question. Under these circumstances a “not applicable” is not needed.

Comment: A few commenters stated that the survey is too long. One commenter suggested that we should identify the key 1 or 2 questions in each survey domain and use them instead.

Response: We are currently conducting an experiment to test a shorter version of the CAHPS Hospice Survey. Results of this experiment will help to inform changes to the survey in the future.

Final Decision: After consideration of the public comments, we are finalizing our proposal to display Hospice CAHPS Star ratings no sooner than FY 2022. We plan to provide opportunities for interaction with stakeholders to discuss our plans and methodology and to receive feedback prior to the start of star ratings display. We will also explore the feasibility of conducting a dry run of the star ratings with reporting to hospices via preview reports, which would occur prior to the start of the public display of the ratings.

9. Form, Manner, and Timing of Quality Data Submission

a. Statutory Penalty for Failure To Report

Section 1814(i)(5)(C) of the Act requires that each hospice submit data to the Secretary on quality measures specified by the Secretary. Such data must be submitted in a form and manner, and at a time specified by the Secretary. Section 1814(i)(5)(A)(i) of the Act was amended by the CAA 2021 and the payment reduction for failing to meet hospice quality reporting requirements is increased from 2 percent to 4 percent beginning with FY 2024. The Act requires that, beginning with FY 2014 through FY 2023, the Secretary shall reduce the market basket update by 2 percentage points and then beginning in FY 2024 and for each subsequent year, the Secretary shall reduce the market basket update by 4 percentage points for any hospice that does not comply with the quality data submission requirements for that FY. We received a few comments on this policy. A summary of these comment and our responses to those comments appear below:

Comment: We received several comments objecting to the increase in the percentage penalty for failure to provide quality reporting data.

Response: We thank the commenters for their views, but as noted, this provision is required by section 407(b) of the CAA and does not permit any discretion on the part of the Secretary to implement it.

Comment: Several commenters requested that CMS communicate widely and display prominently notices and information about the increase in the penalty for failure to comply with HQRP requirements. They suggested using multiple avenues of communication including the HQRP website and MLN Connects.

Response: We agree that communicating widely is critically important, to ensure as many hospices as possible are aware not only of the increase in penalty, but also clearly understand the HQRP reporting requirements and the APU process. We will consider using multiple avenues for communication, including this rule, the Medicare Claims Manual, the HQRP website, such as the HQRP Requirements and Best Practices web page at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​HQRP-Requirements-and-Best-Practices and the Training and Education Library page at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​Hospice-Quality-Reporting-Training-Training-and-Education-Library. We will also consider opportunities to communicate through webinars, Open Door Forums, and other resources as relevant.

Comment: A few commenters did not agree with the CAA 2021 provision that removes the prohibition on public disclosure of hospice surveys performed by a national accreditation agency in section 1865(b) of the Act, thus allowing the Secretary to disclose such accreditation surveys. Many commenters also noted the special focused program that requires each state and local survey agency, and each national accreditation body with an approved hospice accreditation program, to submit information respecting any survey or certification made with respect to a hospice program.

Response: The proposed regulatory policies to implement the hospice survey and enforcement provisions in section 407 of CAA, 2021 were included in CY 2022 Home Health Prospective Payment System proposed rule with the comment period found here: https://www.govinfo.gov/​content/​pkg/​FR-2021-07-07/​pdf/​2021-13763.pdf. We encourage commenters to provide us input and comments on these provisions in response to that rule. The link to the Federal Register can be found here: CMS-1747-P CY 2022 Home Health Prospective Payment System Rate Update. Note: The comment period closes on August 27, 2021.

b. Compliance

HQRP Compliance requires understanding three timeframes for both HIS and CAHPS. (1) The relevant Reporting Year, payment FY and the Reference Year. The “Reporting Year” (HIS)/“Data Collection Year” (CAHPS). This timeframe is based on the CY. It is the same CY for both HIS and CAHPS. If the CAHPS Data Collection year is CY 2022, then the HIS reporting year is also CY 2022. (2) The APU is subsequently applied to FY payments based on compliance in the corresponding Reporting Year/Data Collection Year. (3) For the CAHPS Hospice Survey, the Reference Year is the CY prior to the Data Collection Year. The Reference Year applies to hospices submitting a size exemption from the CAHPS survey (there is no similar exemption for HIS). For example, for the CY 2022 data collection year, the Reference Year, is CY 2021. This means providers seeking a size exemption for CAHPS in CY 2022 would base it on their hospice size in CY 2021. Submission requirements are codified in § 418.312.

For every CY, all Medicare-certified hospices are required to submit HIS and CAHPS data according to the requirements in § 418.312. Table 11 summarizes the three timeframes. It illustrates how the CY interacts with the FY payments, covering the CY 2020 through CY 2023 data collection periods and the corresponding APU application from FY 2022 through FY 2025.

As illustrated in Table 11, CY 2020 data submissions compliance impacts the FY 2022 APU. CY 2021 data submissions compliance impacts the FY 2023 APU. CY 2022 data submissions compliance impacts FY 2024 APU. This CY data submission impacting FY APU pattern follows for subsequent years.

c. Submission Data and Requirements

As finalized in the FY 2016 Hospice Wage Index and Payment Rate Update final rule (80 FR 47192), hospices' compliance with HIS requirements beginning with the FY 2020 APU determination (that is, based on HIS-Admission and Discharge records submitted in CY 2018) are based on a timeliness threshold of 90 percent. This means CMS requires that hospices submit 90 percent of all required HIS records within 30-days of the event (that is, patient's admission or discharge). The 90-percent threshold is hereafter referred to as the timeliness compliance threshold. Ninety percent of all required HIS records must be submitted and accepted within the 30-day submission deadline to avoid the statutorily-mandated payment penalty.

To comply with CMS' quality reporting requirements for CAHPS, hospices are required to collect data monthly using the CAHPS Hospice Survey. Hospices comply by utilizing a CMS-approved third-party vendor. Approved Hospice CAHPS vendors must successfully submit data on the hospice's behalf to the CAHPS Hospice Survey Data Center. A list of the approved vendors can be found on the CAHPS Hospice Survey website: www.hospicecahpssurvey.org. Table 12. HQRP Compliance Checklist illustrates the APU and timeliness threshold requirements.

Most hospices that fail to meet HQRP requirements do so because they miss the 90 percent threshold. We offer many training and education opportunities through our website, which are available 24/7, 365 days per year, to enable hospice staff to learn at the pace and time of their choice. We want hospices to be successful with meeting the HQRP requirements. We encourage hospices to use this website at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​Hospice-Quality-Reporting-Training-Training-and-Education-Library.

For more information about HQRP Requirements, please visit the frequently-updated HQRP website and especially the Best Practice, Education and Training Library, and Help Desk web pages at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting. We also encourage members of the public to go to the HQRP web page and sign-up for the Hospice Quality ListServ to stay informed about HQRP.

d. Update on Transition to iQIES

In the FY 2020 Hospice Wage Index and Payment Rate Update final rule (84 FR 38484), we finalized the proposal to migrate our systems for submitting and processing assessment data. Hospices are currently required to submit HIS data to CMS using the Quality Improvement and Evaluation System (QIES) Assessment and the Submission Processing (ASAP) system. The FY 2020 Hospice Wage Index and Payment Rate Update final rule (84 FR 38484) finalized the proposal to migrate to a new internet Quality Improvement and Evaluation System (iQIES) that will enable us to make real-time upgrades. We are designating that system as the data submission system for the Hospice QRP. We will notify the public about any system migration updates using subregulatory mechanisms such as web page postings, listserv messaging, and webinars.

We received several on the transition to iQIES. A summary of these comment and our responses to those comment appear below:

Comment: Several commenters requested 6-month minimum notice prior to the transition of hospice to the iQIES system. Some of these commenters further requested that CMS provide announcements about the upcoming implementation of hospice in iQIES through all CMS and MAC communication platforms to ensure wide penetration of the message, and ensure a smooth transition given lessons from the transition of other settings to iQIES.

Response: We appreciate that providers will benefit from advanced notice regarding the transition of hospice to the iQIES systems. We plan to communicate with the provider community via sub-regulatory means about the upcoming transition as the timing becomes clear, and will provide sufficient time and appropriate information for a smooth transition.

10. Public Display of “Quality Measures” and Other Hospice Data for the HQRP

a. Background

Under section 1814(i)(5)(E) of the Act, the Secretary is required to establish procedures for making any quality data submitted by hospices available to the public. These procedures shall ensure that individual hospices have the opportunity to review their data prior to these data being made public on our designated public website. To meet the Act's requirement for making quality measure data public, we launched Hospice Compare in August 2017. This website allows consumers, providers, and other stakeholders to search for all Medicare-certified hospice providers and view their information and quality measure scores. In September 2020, CMS transitioned Hospice Compare to the Care Compare website. Hospice Compare was discontinued in December 2020. Care Compare supports all Medicare settings and fulfills the Act's requirements for the HQRP. For more information about Care Compare, please see the Update on the Hospice Quality Reporting Requirements for FY 2022 in section D.

Since 2017, we have increased and improved available information about the care hospices provide for consumers. To indicate the quality of care hospices provide, we first posted the seven HIS Measures (NQF #1641, NQF #1647, NQF #1634, NQF #1637, NQF #1639, NQF #1638, and NQF #1617) in 2017, and then added the CAHPS Hospice Survey measure (NQF #2651) and the HIS Comprehensive Assessment at Admission (NQF #3235) in 2018. In 2019, we added the Hospice Visits When Death is Imminent (Measure 1) to the website.

As discussed previously, we are finalizing our proposal to remove the seven HIS Measures from public reporting on Care Compare no earlier than May 2022. The Hospice Item Set V3.00 PRA Submission replaced the HVWDII measure with a more robust version: The claims-based measure HVLDL. We will publicly report the HVLDL no earlier than May 2022. We are also finalizing our proposal to publicly report the HCI, another claims-based measure no earlier than May 2022. In addition to the publicly-reported quality measure data, in 2019 we added to public reporting, information about the hospices' characteristics, taking raw data available from the Medicare Public Use File and other publicly-available government data sources and making them more consumer friendly and accessible for people seeking hospice care for themselves or family members, (83 FR 38649). This publicly reported information currently includes diagnoses, location of care, and levels of care provided.

b. Data Collection and Reporting During a Public Health Emergency

(1). Background: COVID-19 Public Health Emergency Temporary Exemption and Its Impact on the Public Reporting Schedule

Under authority of section 319 of the Public Health Service (PHS) Act, the Secretary declared a PHE effective as of January 27, 2020. On March 13, 2020, the President declared a national state of emergency under the Stafford Act, effective March 1, 2020, allowing the Secretary to invoke section 1135(b) of the Act (42 U.S.C. 1320b-5) to waive or modify the requirements of titles XVIII, XIX, and XXI of the Act and regulations to the extent necessary to address the COVID-19 PHE. Many waivers and modifications were made effective as of March 1, 2020 ^[46 47] in accordance with the president's declaration. On March 27, 2020, we sent a guidance memorandum under the subject title, “Exceptions and Extensions for Quality Reporting Requirements for Acute Care Hospitals, PPS-Exempt Cancer Hospitals, Inpatient Psychiatric Facilities, Skilled Nursing Facilities, Home Health Agencies, Hospices, Inpatient Rehabilitation Facilities, Long-Term Care Hospitals, Ambulatory Surgical Centers, Renal Dialysis Facilities, and MIPS Eligible Clinicians Affected by COVID-19” ^[48] to the Medicare Learning Network (MLN) Connects Newsletter and Other Program-Specific Listserv Recipients,^[49] hereafter referred to as the March 27, 2020 CMS Guidance Memorandum. In that memo, which applies to HIS and CAHPS Hospice Survey, CMS granted an exemption to the HQRP reporting requirements for Quarter 4 (Q4) 2019 (October 1, 2019 through December 31, 2019), Quarter 1 (Q1) 2020 (January 1, 2020 through March 30, 2020), and Quarter 2 (Q2) 2020 (April 1, 2020 through June 30, 2020). We discuss the impact to the HIS here, and the impact to the CAHPS Hospice Survey further in section F.10.b.4. For HIS, the quarters are defined based on submission of HIS admission or discharge assessments.

The exemption has impacted the public reporting schedule. Since launching Hospice Compare in 2017, HIS-measures have been reported using 4 quarters of data. The 4 quarters included are the most recent data that have gone through Review and Correct processes, have been issued in a provider preview report, and have time allotted for addressing requests for data suppression before being publicly reported. As discussed in the FY 2017 Hospice Wage Index and Payment Rate Update final rule (81 FR 52183), CMS requires at least 4 quarters of data to establish the scientific acceptability for our HIS-based quality measures. For CAHPS-based measures, we have reported CAHPS measures using eight rolling quarters of data on Hospice Compare since 2018. In the FY 2017 Hospice Wage Index and Payment Rate Update final rule (81 FR 52143), we stated that we would continue CAHPS reporting with eight rolling quarters on an ongoing basis. This original public reporting schedule included the exempted quarters of Q4 2019 and Q1 and Q2 2020 in six refreshes for HIS and 11 refreshes for CAHPS. Table 13 displays the original schedule for public reporting prior to the COVID-19 PHE.

During the spring and summer of 2020, we conducted testing to inform decisions about publicly reporting data for those refreshes which include exempt data. The testing helped us develop a plan for posting data as early as possible, for as many hospices as possible, and with scientific acceptability similar to standard threshold for public reporting. The following sections provide the results of our testing and explain how we used the results to develop a plan that we believe allows us to achieve these objectives as best as possible.

(2). Update on Use of Q4 2019 Data and Data Freeze for Refreshes in 2021

In the March 27, 2020 Guidance Memorandum, we stated that we should not include any post-acute care (PAC) quality data that are greatly impacted by the exemption in the quality reporting programs. Given the timing of the COVID-19 PHE onset, we determined that we would use any data that was submitted for Q4 2019. We conducted analyses of those data to ensure that their use was appropriate. In the original schedule (Table 13) the November 2020 refresh includes Q4 2019 data for HIS- and CAHPS-based measures (Q1 through Q4 2019 for HIS data and Q1 2018 through Q4 2019 for CAHPS data) and is the last refresh before Q1 2020 data are included. Before proceeding with the November 2020 refresh, we conducted testing to ensure that, even though we made an exception to reporting requirements for Q4 2019 in March 2020, public reporting would still allow us to publicly report data for a similar number of hospice providers, as compared to standard reporting. Specifically, we compared submission rates in Q4 2019 to average annual rates (Q4 2018 through Q3 2019) to assess the extent to which hospices had taken advantage of the exemption, and thus the extent to which data and measure scores might be affected. We observed that the HIS data submission rate for Q4 2019 was in fact 1.8 percent higher than the previous CY (Q4 2018). For the CAHPS Hospice Survey, 2.1 percent more hospices submitted data in Q4 2019 than in Q4 2018. We note that Q4 2019 ended before the onset of the COVID-19 PHE in the United States (U.S.). Thus, we proceeded with including these data in measure calculations for the November 2020 refresh.

As for Q1 and Q2 2020, we determined that we would not use HIS or CAHPS data from these quarters for public reporting given the timing of the COVID-19 PHE onset. All refreshes, during which we decided to hold these data constant, included more than 2 quarters of data that were affected by the CMS-issued COVID reporting exceptions; thus we did not have an adequate amount of data to reliably calculate and publicly display provider measures scores. Consequently, we determined to freeze the data displayed, that is, holding data constant after the November 2020 refresh without subsequently updating the data through November 2021. This decision was communicated to the public in a Public Reporting Tip Sheet, which is located at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​HQRP-Requirements-and-Best-Practices.

(3). Public Reporting of HIS-based Measures With Fewer Than Standard Numbers of Quarters Due to COVID-19 PHE Exemption in February 2022

As noted previously, we used Q4 2019 data for public reporting in November 2020 and froze that data for the February, May, August, and November 2021 refreshes. This addressed five of the six COVID-19 PHE-affected quarters for HIS-based measures, and five of the 11 COVID-19 PHE-affected quarters of CAHPS-based measures.

Because November 2020 refresh data will become increasingly out-of-date and thus less useful for consumers, we analyzed whether it would be possible to use fewer quarters of data for the last refresh affected by the exemption (February 2022) and thus more quickly resume public reporting with updated quality data. Using fewer quarters of more recent data, the first option, would require that (1) a sufficient percentage of providers would still likely have enough assessment data to report quality measures (reportability); and (2) fewer quarters would likely produce similar measure scores for hospices, and thus not unfairly represent the quality of care hospices provide during the period reported in a given refresh (reliability). To assess these criteria, we conducted reportability and reliability analysis using 3 quarters of data in a refresh, instead of the standard 4 quarters of data for reporting HIS-based measures. Specifically, we used historical data to calculate HIS-based quality measures under two scenarios:

• Standard Public Reporting (SPR) Scenario: We used data from the four quarters of CY 2019, which represent CY 2020 public reporting in the absence of the temporary exemption from the submission of PAC quality data, as the basis for comparing simulated alternatives. For HIS-based measures, we used quarters Q1 through Q4 2019.
• COVID-19 PHE Affected Reporting (CAR) Scenario: We calculated quality measures using Q2 2019, Q3 2019, and Q4 2019 data, to simulate using only Q3 2020, Q4 2020, and Q1 2021 data for public reporting.

The HIS Comprehensive Assessment Measure is based on the receipt of care processes at the time of admission. Therefore for the COVID-19 Affected Reporting (CAR) Scenario, we excluded data for patient stays with admission dates in Q1 2019.

For each scenario, we calculated the reportability as the percent of hospices meeting the 20-case minimum for public reporting (the public reporting threshold). To test the reliability of restricting the providers included in the Standard Public Reporting (SPR) Scenario to those included in the CAR Scenario, we performed three tests. First, we evaluated measure correlation using the Pearson and Spearman correlation coefficients, which assess the alignment of hospices' HIS Comprehensive Assessment Measure scores between scenarios. Second, for each scenario, we conducted a split-half reliability analysis and estimated intra-class correlation (ICC) scores, where higher scores imply better internal reliability. Modest differences in ICC scores between scenarios would suggest that using fewer quarters of data does not impact the internal reliability of the results. Third, we estimated reliability scores. A higher value in these scores indicates that HIS Comprehensive Assessment Measure values are relatively consistent for patients admitted to the same hospice and variation in the measure reflects true differences across providers.

Testing results show that the CAR scenario—specifically using 3 quarters of data for the HIS Comprehensive Assessment Measure—demonstrates acceptable levels of reportability and reliability. As displayed in Table 14, the number of providers who met the public reporting threshold for the HIS Comprehensive Assessment Measure decreases by 236 (or by 5.2 percentage points) when reporting three versus four quarters of data. In the FY 2014 Hospice Wage Index and Payment Rate Update final rule (78 FR 48234) we stated that reportability of 71 percent through 90 percent is acceptable. Therefore using 3 quarters of data for the HIS Comprehensive Assessment Measure would achieve acceptable reportability shown in Table 14.

Table 14 indicates that the reliability of the HIS Comprehensive Assessment Measure scores is similar for the CAR and SPR scenarios. Testing also yielded correlation coefficients above 0.9, indicating a high degree of agreement between hospices' HIS Comprehensive Assessment Measure scores when using 3 or 4 quarters of data. The results also show that th e HIS Comprehensive Assessment Measure's ICC for CAR and SPR scenarios are similar, with only a 0.02 difference. This implies high internal reliability of the measure in both scenarios. The median reliability scores for the HIS Comprehensive Assessment Measure are also very similar in both CAR and SPR scenarios. This indicates that scores estimated using 3 quarters of data continue to capture provider-level differences and that admission-level scores remain consistent within hospices.

In Table 15, we explore changes in hospices' relative rankings between the SPR and CAR scenarios. For each scenario, we divided hospices in quintiles based on their HIS Comprehensive Assessment Measure score, such that higher scores are in a higher quintile. Changes in a hospices' quintile from the SPR to CAR scenario would indicate a re-ranking of hospices when using 3 quarters compared to 4 quarters. Over 93 percent of hospices remain in the same quintile, suggesting that the ranking of hospices is fairly stable between the SPR and CAR scenarios.

We also used the results presented in Table 16 to assess the option of reporting Q4 2019, Q3 2020, Q4 2020, and Q1 2021 for the February 2022 refresh. This option maintains requirements in the FY 2017 Hospice Wage Index and Payment Update final rule for publicly reporting 4 quarters of data, but it requires using some data that are more than 2 years old. Also, the relatively high number of hospices that meet the public reporting threshold in the CAR scenario, relative to the SPR scenario, with just 3 quarters of data justify the use of 3 quarters in the unusual circumstances of the COVID-19 PHE and its associated exemptions.

We are finalizing our proposal that, in the COVID-19 PHE, we would use 3 quarters of HIS data for the final affected refresh, the February 2022 public reporting refresh of Care Compare for the Hospice setting. Using 3 quarters of data for the February 2022 refresh would allow us to begin displaying Q3 2020, Q4 2020, and Q1 2021 data in February 2022, rather than continue displaying November 2020 data (Q1 2019 through Q4 2019). We believe that updating the data in February 2022 by more than a year relative to the November 2020 freeze data would assist consumers by providing more relevant quality data and allow hospices to demonstrate more recent performance. Our testing results indicate we can achieve these positive impacts while maintaining high standards for reportability and reliability. Table 16 summarizes the comparison between the original schedule for public reporting with the revised schedule (that is, frozen data) and with the proposed schedule that is, using 3 quarters in the February 2022 refresh.

We solicited public comment on this proposal to use 3 quarters of HIS data for the February 2022 public reporting refresh. We received many comments this proposal on related questions about publicly reporting claims-based measures using data from the COVID-19 PHE. A summary of the comments received regarding public reporting and our responses those comments appear below.

Comment: We received several comments supporting our proposal to begin public reporting in February 2022 using Q3 and Q4 of 2020 and Q1 of 2021. These commenters also suggested that CMS post a statement that the data displayed include care provided during the COVID-19 PHE on Care Compare until August 2023. One commenter opposed the public reporting of any quality data collected during the COVID-19 PHE (not just the Q1 and Q2 2020 which were subject to the exemptions), because of the impact COVID-19 had on hospice processes and operations.

Response: We appreciate the commenters' support for this proposal. In response to the commenter who did not support this proposal, we would like to emphasize that, while we recognize that the impact of COVID-19 has impacted the hospice community, we also believe that we have a responsibility to consumers to make informed decisions about selecting care. Providing information for decision-making is all the more important during and in the wake of a COVID-19 PHE, when our health as a nation has been shaken.

We disagree with commenters that notices should be posted on Care Compare regarding the inclusion of data from the COVID-19 PHE as such notice would not help consumers distinguish between hospices in their region. Instead, we will continue to post national averages for quality measures, and will add state scores for all measures no earlier than May 2022. This information will help consumers understand relative performance at national and local levels in light of the COVID-19 PHE.

Given the overall positive response to our proposal, we believe that the proposed approach balances fairness to providers with a commitment to transparency and information for consumers.

Comment: Several commenters expressed concern about publicly reporting claims-based measures using data from care provided during the COVID-19 PHE. Specifically, they stated that claims from the COVID-19 PHE would not reflect typical hospice services. Comments specific to HCI noted that abnormalities due to the COVID-19 PHE would affect all of the indicators, while those for HVLDL indicated that the number of in-person visits likely fell during the COVID-19 PHE due to patient and caregiver preferences, with implications for quality measurement. The commenters recommended that CMS post a notice on Care Compare to ensure consumers understand the context, with particular attention to the fact that telehealth visits are not captured in claims reporting.

Response: We appreciate commenters' concerns about publicly reporting claims from the COVID-19 PHE. As stated earlier, we pre-emptively issued the March 27, 2020 CMS Guidance Memorandum making 2019 Q4 and Q1 and Q2 2020 exempt from reporting requirements. In that Memorandum, we stated that we would not include any post acute care (PAC) quality data that are greatly impacted by the exemption in the quality reporting programs. Given the timing of the COVID-19 PHE onset in the U.S., we determined that we would use data that were submitted for Q4 2019. We will apply the principles of this Memorandum to new claims-based measures for hospice. Thus, we will publicly report claims data for care delivered in Q4 2019 and Q3 2020 onward, but we will not publicly report claims data for care delivered Q1 and Q2 of 2020. This approach aligns with what we are doing for the other PAC setting Quality Reporting Programs, including home health (see section III.G).

We acknowledge that the COVID-19 PHE did not end at the beginning of Q3 2020. Our testing indicates that claims data from the COVID-19 PHE are generally stable. Although the number of visits in did visibly decline in 2020, we remain committed to re-initiating publicly reporting of claims data beginning in Q3 2020 for the following reasons: (i) We believe that we have an important commitment to consumers of hospice care to empower them to make informed decisions. This is particularly important during the COVID-19 PHE; (ii) With annual reporting of claims data, we can reasonably state that the COVID-19 PHE affected hospices nationally in a similar way. Given that HCI is scored relative to the national average, scores will be accounted for as part of the measure calculation. To the extent there have been regional differences, we will also provide state scores for both HCI and HVLDL no earlier than May 2022, so that consumers can benchmark to more local realities.

We respectfully disagree with commenters who have requested that we post a notice on Care Compare alerting consumers to potential abnormalities in claims data wholly or partially coming from COVID-19 PHE (excluding Q1 and Q2 2020). Despite the COVID-19 PHE, we would expect that hospices would still provide comprehensive care to hospice patients during the pandemic, and believe that telehealth visits are not full substitutes for care provided in person, particularly in the case of the visits measured in the HVLDL and HCI measures. We acknowledge that there may have been an increase in refusals during the COVID-19 PHE. However, this increase would likely impact hospices in a region similarly, and thus will not impact a hospice's score relative to local competitors. We will include state average scores to further ensure any regional differences in the impact of the COVID-19 PHE on hospices are captured for consumers. For these reasons, adding disclaimer text as suggested would not help consumers seeking information make decisions about care options.

Final Decision: We are finalizing our proposal to resume public reporting of HIS quality measures in February 2022 using data from Q3 and Q4 of 2020 and Q1 of 2021.

(4). Proposal for Public Reporting of “CAHPS Hospice Survey-based Measures” Due to COVID-19 PHE Exemption

Prior to COVID-19 PHE, the CAHPS Hospice Survey publicly reported the most recent eight rolling quarters of data. We propose to continue to report the most recent 8 quarters of available data after the freeze, but not to include the data from the exempted quarters of Q1 and Q2 of 2020 as issued in the March 27, 2020 Guidance Memorandum with the effected quarters. The optional data submission for Q4 2019 results in publicly reporting of that data since the CAHPS Hospice Survey from that quarter were not impacted. The data submitted for Q4 2019 referred to deaths that occurred prior to COIVD-19. For the CAHPS Hospice Survey, 2.1 percent more hospices submitted data in Q4 2019 than in the same quarter a year earlier.

Like HIS, our goal is to report as much of the most recent CAHPS Hospice Survey data as possible, to display data for as many hospices as possible, and to maintain the reliability of the data.

Similar to HIS, the CAHPS Hospice Survey reviewed the data for reportability using fewer quarters than normal. However, we found that using fewer than 8 quarters of data would have two important negative impacts on public reporting. First, it would reduce the proportion of hospices that would have CAHPS Hospice Survey data displayed on Care Compare. An analysis of the 8 quarters of data from Q1 2018 through Q4 2019 (publicly reported in November 2020) shows there were 5,041 active hospices. Of these hospices: 2,941 (58.3 percent) had 30+ completes for those 8 quarters, and had scores publicly reported. Fewer hospices, 2,328 (46.2 percent), would have had 30+ completes if 4 quarters of data were used to calculate scores and 1,970 (39.1 percent) would have 30+ completes if 3 quarters were used to calculate scores. In addition, the overall reliability of the CAHPS scores would decline with fewer quarters of data. For these reasons, we determined the best course of action would be to continue to publicly report the most recent 8 quarters of data, but exempting Q1 and Q2 2020. This will allow us to maximize the number of hospices that will have CAHPS scores displayed on Care Compare, protect the reliability of the data, and report as much of the most recent data as possible.

CMS froze CAHPS data starting with the November 2020 refresh and concluding with the November 2021 refresh. We propose that starting with the February 2022 refresh, CMS will display the most recent 8 quarters of CAHPS Hospice Survey data, excluding Q1 and Q2 2020. We will resume public reporting by displaying 3 quarters of post-exemption data, plus five quarters of pre-exemption data. (Please see Table 18.) We propose that in each refresh subsequent to February 2022, we will report one more post-exemption quarter of data and one fewer pre-exemption quarter of data until we reach eight quarters of post-exemption data in May of 2023. We further propose that as of August 2023, we will resume reporting a rolling average of the most recent 8 quarters of data. Table 18 specifies the quarters for each refresh. This will allow us to report the maximum amount of new data, maintain reliability of the data, and permit the maximum number of hospices to receive scores. In addition, Table 18 shows the proposed CAHPS public reporting schedule during and after the data freeze.

We sought public comment on this proposal to publicly report the most-recently available 8 quarters of CAHPS data starting with the February 2022 refresh and going through the May 2023 refresh on Care Compare because we cannot publicly report Q1 2020 and Q2 2020 data due to the COVID-19 PHE.

Comment: One commenter agreed with our proposal to report the eight most recent quarters of data for the CAHPS Hospice Survey, skipping the exempted quarters. They also requested that Care Compare provide information to users explaining that the published data included pre-COVID quarters. They wanted this continued until all publicly-reported data is from after the exempted quarters.

Response: We thank the commenter and will take this into consideration as information for Care Compare is developed. We will work with colleagues to provide information on Care Compare that alerts users the composition of the data.

Final Decision: We are finalizing our proposal to publicly report the most-recently available 8 quarters of CAHPS data starting with the February 2022 refresh and going through the May 2023 refresh on Care Compare because we cannot publicly report Q1 2020 and Q2 2020 data due to the COVID-19 PHE.

c. Quality Measures To Be Displayed on Care Compare in FY 2022 and Beyond

(1). Removal of the Seven “Hospice Item Set Process Measures” From Public Reporting

We are finalizing our proposal to remove the seven HIS process measures from the HQRP as individual measures, and no longer applying them to the FY 2024 APU and thereafter. We are finalizing our proposal to remove the seven HIS process measures no earlier than May 2022 refresh from public reporting on Care Compare and from the Preview Reports but continue to have it publicly available in the data catalogue at https://data.cms.gov/​provider-data/​topics/​hospice-care.

We solicited public comment on this proposal to remove the seven HIS process measures from public reporting on Care Compare. We received several comments from various stakeholders. A summary of the comments we received on this proposal and our responses to those comments appear below.

Comment: The majority of commenters supported the removal of the seven HIS process measures no earlier than May 2022. However, a number of comments suggested that CMS continue providing the option for consumers to view detailed information about the individual measures that make up the HIS Comprehensive Assessment measure for transparency. One commenter who opposed the proposal to remove the seven HIS measures expressed concern that such a removal runs counter to the objectives of Care Compare to provide a personalized experience. Some comments expressed concern about the public's ability to be aware of and find the seven HIS measure scores in the Provider Data Catalogue.

Response: CMS does not believe that the public display of the individual process measures on Care Compare will add value for consumers. The individual measures show performance for only one process and do not demonstrate whether the hospice provides high-quality care overall, as an organization. Conversely, the HIS Comprehensive Assessment Measure, which is a single composite measure, differentiates hospices by holding them accountable for completing all seven process measures to ensure these core hospice services are completed for all patients. This interdisciplinary, holistic scope of the HIS Comprehensive Assessment Measure better aligns with the public's expectations for hospice care. We maintain transparency since stakeholders, who are interested in the seven HIS measures, will have access to the Provider Data Catalogue where they can find all HIS component measure scores.

We respectfully disagree that having the seven HIS measures listed is more transparent and understandable for consumers than a concise summary: Market research conducted by our teams has found that “less is more” for Care Compare consumers, who become overwhelmed by too much information. In fact, these findings were one of the primary reasons we have transitioned from Hospice Compare and the other individual compare sites to Care Compare.

We appreciate the concern that consumers may not know about the component measure scores in the Provider Data Catalogue. As we prepare to update Care Compare for the removal of the seven measures, we will consider ways to make consumers of Care Compare aware of this additional data, if they are interested in viewing them.

Comment: Several commenters expressed concern about the public's ability to understand the meaning of the HIS Comprehensive Measure without being able to see the seven component measures. These commenters provided general and specific suggestions about how to display the HIS Comprehensive Measure on Care Compare if the seven HIS measures are removed. Several other commenters also suggested posting a disclaimer that the HIS Comprehensive measure only comes from the admission item set and may not be reflective of subsequent care.

Response: We appreciate that the presentation of the seven HIS measures helped consumers understand the content of the HIS Comprehensive Measure. As we prepare to update Care Compare for their removal, we will consider ways to revise the measure description for the HIS Comprehensive Measure on Care Compare so that it adequately explains the elements contained in the measure.

As for the request to notify consumers that the measure is based on admission alone, we do not believe this would help consumers use the measure to compare and select hospices, as intended. The HIS Comprehensive Measure, like any given quality measure, is one part of a portfolio of measures intended to provide a holistic view of care. No single quality measure within the portfolio is expected, or necessarily intended, to provide that view on its own. As we determine the most appropriate way to display the measure, we will ensure that the scope of the HIS Comprehensive Measure is clear for consumers, who can use the information with other information on the website to make their decisions.

Comment: A number of commenters suggested that CMS continue providing the option for hospices to view detailed information about the individual measures that make up the HIS Comprehensive Assessment measure to support quality improvement.

Response: We will ensure that the confidential QM reports continue to include the seven HIS process measures, in addition to the HIS Comprehensive Assessment Measure. This helps hospices apply quality improvement processes to continue improving their performance on the HIS Comprehensive Assessment Measure.

Final Decision: We are finalizing our proposal to remove the seven HIS process measures no earlier than the May 2022 refresh from public reporting on Care Compare and from the Preview Reports but continue to have them publicly available in the data catalogue.

(2). Calculating and Publicly Reporting “Claims-Based Measure” as Part of the HQRP

In the HIS V3.00 Paperwork Reduction Act Submission (OMB control number: 0938-1153, CMS-10390), we finalized a proposal to adopt HVLDL into the HQRP for FY 2021. We are also proposing in this rule to adopt the HCI into the HQRP for FY2022. In this section, we presented three proposals related to calculating and reporting claims-based measures, with specific application to HVLDL and HCI. First, we are finalizing our proposal to extract claims data to calculate claims-based measures at least 90 days after the last discharge date in the applicable period, which we will use for quality measure calculations and public reporting on Care Compare. For example, if the last discharge date in the applicable period for a measure is December 31, 2022, for data collection January 1, 2022, through December 31, 2022, we would create the data extract on approximately March 31, 2023, at the earliest. We would use those data to calculate and publicly report the claims-based measures for the CY2022 reporting period. This is similar to those finalized in other PAC settings, including the CY 2017 Home Health Prospective Payment System final rule (81 FR 76702), FY 2017 Inpatient Rehabilitation Facility Prospective Payment System final rule (81 FR 52056), and the FY 2017 Long Term Care Hospital Prospective Payment System final rule (81 FR 56762).

We are finalizing the proposed timeframe which allows us to balance providing timely information to the public with calculating the claims-based measures using as complete a data set as possible. We recognize the approximately 90-day “run-out” period is shorter than the Medicare program's current timely claims filing policy under which providers have up to 1 year from the date of discharge to submit claims. However, several months lead-time is necessary after acquiring the data to conduct the claims-based calculations. If we were to delay our data extraction point to 12 months after the last date of the last discharge in the applicable period, we would not be able to deliver the calculations to hospices sooner than 18 to 24 months after the last discharge.

To implement this process, hospices would not be able to submit corrections to the underlying claims snapshot or add claims (for those claims-based measures) to this data set at the conclusion of the 90-day period following the last date of discharge used in the applicable period. Therefore, we would consider the hospice claims data to be complete for purposes of calculating the claims-based measures at this point. Thus, it is important that hospices ensure the completeness and correctness of their claims prior to the claims “snapshot.”

Second, we are finalizing our proposal to update the claims-based measures used for the HQRP annually. Specifically, we will refresh claims-based measure scores on Care Compare, in preview reports, and in the confidential CASPER QM preview reports annually. This periodicity of updates aligns with most claims-based measures across PAC settings.

Third, we are finalizing our proposal to calculate claims-based measure scores based on one or more years of data. We considered several factors to determine the number of years to include in measure calculations. Using only 1 year (4 quarters) of data, as is currently done for HIS-based quality measures reported on Care Compare, allows us to share with the public only the most up-to-date information and best reflects current realities. Having only the most recent data can also help incentivize hospices with lower scores to make changes and have the results of their effort be reflected in better scores.

At the same time, we want to report measures scores to the public for as many hospices as possible, including small hospices. Currently, only Medicare-certified hospices with more than 20 patient stays each year have quality measure results publicly available on Care Compare. This public reporting threshold protects the privacy of patients who seek care at smaller hospices. However, due to the threshold, at least some hospices will not achieve the minimum patient stays within 1 year. This means that their scores will not be displayed on Care Compare, and consumers will not have information about them to inform their decisions about selecting a hospice. Using more years of data allows more of these hospices to meet this threshold.

We conducted reportability testing for HCI and HVLDL to help us consider how best to balance the need for recent data with the need for transparency in reporting the HQRP claims-based measures. Specifically, we conducted a simulation using 2 years of data. We then calculated the change in the number of hospices which achieved the minimum reporting standard. We also compared the measure scores of the hospices that meet the reporting threshold when we use 2 years of data with hospices that meet the threshold using only 1 year of data.

Results for both HCI and HVLDL indicate that using 2 years of data increases reportability. For HVLDL, combining 2 years of data (FY 2018 to FY 2019) allows an additional 326 hospices to share measure scores, or 33.8 percent of the hospices that do not meet the reporting threshold in FY 2019 alone. For HCI, combining 2 years of data (FY 2018 to FY 2019 data) allows an additional 277 to report HCI measure scores on Care Compare, or 43.2 percent of the hospices that do not meet the reporting threshold in FY 2019 alone.

Our simulations indicate that the hospices that only meet the reporting threshold when using 2 years of data have performance scores substantially lower than average. For HVLDL, where higher scores indicate better quality of care, the national average score was 65.5 percent in FY 2019, where 965 hospices did not meet the reportability threshold. After pooling data using FY 2018 to FY 2019, 326 additional hospices met the reportability threshold, or 33.8 percent of those previously missing. Those addition 326 hospices had an average HVLDL score of just 43.3 percent, about 20 percentage points lower than the hospices meeting the reportability threshold using FY 2019 alone national average score for this HVLDL measure.

The results for HCI similarly show that the hospices with reportable data when using two-pooled years of data had lower HCI scores compared to the national average when using just FY 2019 data. Higher HCI scores indicate better performance. As Figure 2 shows, a larger numbers of hospices among the 277 hospices that only meet the reporting threshold when using 2 years of data had HCI scores between four and eight, while a larger number of hospices in the FY 2019 population had a perfect score of 10.

Given these findings, we are finalizing our proposal to use 2 years of data to publicly report HCI and HVLDL in 2022. The use of 2 years or 8 quarters of quality data is already publicly reported for the quality measures related to the CAHPS Hospice Survey so hospices are familiar with this approach. We plan to consider multiple years of data, like the 2 years of data, for other claims-based measures proposed in subsequent years. We believe it is important to support consumers by sharing information on the performance of hospices that have lower scores, and to incentivize those hospices to improve. The results demonstrate that using multiple years of data help include more hospices that have lower performance rates for HVLDL and HCI in public reporting on Care Compare. While using more years of data would allow us to report measures for even more hospices, it would involve sharing data that are no longer relevant, and display scores that do not reflect recent hospice improvement efforts.

We solicited public comment on these proposals related to the use of 2 years of data for claims-based measures and public reporting of claims measures in general and their application to HVLDL and HCI specifically. We received several comments from various stakeholders on this proposal. A summary of the comments we received on this proposal and our responses to those comments appear below:

Comment: A few commenters expressed concern that hospices would not be able to view data close to real time, which might inhibit the ability to use the score to inform continuous quality improvement.

Response: We agree that there is a lag time between the delivery of care and the calculation and reporting of the claims-based quality measures, including HCI. However, the time is needed. After the data extract is created after the 90-day run-off, it takes several months to incorporate other data needed for the calculations. We then need to generate and check the calculations before posting for confidential reporting. Our proposal for using the 90-day run-off strikes a balance between allowing time for hospices to make corrections to their claims, while also seeking to post more rather than less up-to-date information. We have streamlined our processes as much as possible, and time is needed to go through these steps to ensure accurate publication of quality measure data.

Comment: Several commenters requested that CMS issue confidential reports with hospices' claims-based measure scores in CASPER to help hospices understand and validate their scores before they are publicly reported.

Response: Section 1814(i)(5)(E) of the Act requires that the Secretary establish procedures for making HQRP data available to the public and ensure that hospices have the opportunity to review HQRP data before their release to the public. We will provide this opportunity to review for claims-based measures in a process similar to HIS-based measures. Hospices can review and correct their HIS data before the Data Correction Deadline; for claims data, hospices will be able to ensure that the data are accurate through the end of the 90-day run-off period. Subsequently, as with HIS-based measures, we will implement a 30-day preview period for claims-based measures, which will serve as the final opportunity for hospices to review their data and alert CMS about any errors in the measure calculations they identify. Should a hospice believe they have found an error with an HIS or claims-based measure calculation as displayed in their preview reports, they can request a review, and we will suppress if the review finds the calculation problematic. We refer readers to the HQRP website at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​Public-Reporting-HIS-Preview-Reports-and-Requests-for-CMS-Review-of-HIS-Data,, which we will revise to include further information on public reporting of claims as well as HIS data. This page covers information about for accessing reports and an email address should hospices have questions regarding any of the above-mentioned reports or processes.

In addition to the Preview Report, we will also include claims-based measure scores in the Hospice Agency-Level QM Report in CASPER. This report is intended to support quality improvement for hospices. Measure scores will be updated annually in the QM Report as they will in the Preview Report and on Care Compare and the Provider Data Catalogue.

Comment: We received several comments with a request for CMS to consider quarterly as opposed to annual reporting of claims-based measures to best support continuous quality improvement activities.

Response: Our proposal to update annually reflects our understanding that claims measures reflect business practices that are slow to change. For example, for HCI, as we discussed in the proposed rule, we compared index scores calculated for the same hospice using annual claims from Federal FY 2017 and 2019. The analysis found that 83% of hospices had HCI scores that were 0-1 percentage points different in FY2019 relative to their FY2017 scores. These results indicate that a hospice's HCI scores would not normally fluctuate a great deal from one year to the next, and that they will fluctuate even less from quarter to quarter. Thus, quarterly updates would not necessarily provide meaningful support to hospices seeking to improve their quality of care. Instead, progress on HCI will occur over longer time frames, and annual updates are sufficient to support hospices' efforts to improve.

Other PAC settings show similar findings regarding the stability of claims measures compared to assessment scores, which we update quarterly. In the home health setting, for example, national median scores for OASIS-based measures tend to increase, while the acute care hospitalization measure remains steady (Figure 3).

At the same time, reporting claims-based measures does require additional labor. Given the findings about stability in claims measure scores, and the cost of updating more frequently, all PAC settings update claims-based measures annually. Hospital claims-based measures are also updated annually. The HQRP seeks to align with the other settings.

Given the findings and considerations, we believe that our proposal to provide annual updates is appropriate. However, we will remain open to reconsidering the frequency of reporting claims across all PAC settings in the future, should data after implementation indicate that such change is warranted.

Comment: One commenter expressed concern that CMS would obtain the data from cost reports, which would not allow them time to understand or preview the measures before they were publicly reported.

Response: We will not pull claims data for calculating the measures from cost reports. Instead, it will come from our research database that contains Medicare files including fee-for-service claims data. As stated, data source and timing will allow time for hospices to preview their measure scores before they are publicly reported.

Comment: We received comments in support of the proposal to use two years of data for publicly reporting HVLDL and HCI. One of these commenters expressed support for making the reporting more inclusive of smaller hospices, to encourage them to also improve the quality of care they provide. Other commenters suggested using a 1-year time frame, so as to make the measure score more reflective of current operations and performance, and thus more understandable and useful for providers and consumers. Some commenters recommended adding a disclaimer that the data are two years old and do not reflect the current status of hospice performance.

Response: We agree that there are benefits to reporting just one year of data. However, we also believe that we must strike a balance between the benefits of reporting fewer years of more timely data with the need to be more inclusive of smaller hospices, which MedPAC has found have higher live discharge rates than larger hospices.^[50] In other settings, some claims-based measures also use two or even three years of data for reporting. For example, as part of the Home Health Quality Reporting Program, the Potentially Preventable 30-Day Post-Discharge Readmission measure is reported using three years of data, while Medicare Spending Per Beneficiary and Discharge to Community measures are reported using two years of data. We also considered using three years of data for HVLDL and HCI, and determined that three years did not yield the same benefit (that is, inclusion of hospices) relative to cost (that is, lag in reporting), and thus proposed using two years of data. With two years of data, 50 percent of the data come from the more recent year, and hospices should still be able to see their scores change as their performance improves.

Final Decision: We are finalizing as proposed our proposals to use 90-day run-off data to calculate claims-based measures, to update claims-based measure scores annually, and to use eight quarters of data to report HVLDL and HCI.

(3). Publicly Report the Hospice Care Index and “Hospice Visits in the Last Days of Life” Claims-Based Measures

As discussed previously, we are finalizing our proposal to publicly report the HCI and HVLDL using 2 years, which is 8 quarters of Medicare claims data. We will publicly report the HCI and HVLDL beginning no earlier than May 2022, and to include it in the Preview Reports no sooner than the May 2022 refresh. The publicly-reported version of HCI on Care Compare will only include the final HCI score, and not the component indicators. The Preview Reports will reflect the HCI as publicly reported.

We solicited public comment on this proposal for HCI and HVLDL public reporting on Care Compare no sooner than May 2022. A summary of the comments we received on this proposal and our responses to those comments appear below:

Comment: Many commenters requested clarification on the reporting period for initial reporting. They also requested clarification on the logistics of the reporting process—in particular, when specifications would be available.

Response: We appreciate the opportunity to provide clarification. If released in May 2022 using eight quarters of data, the HCI and HVLDL measure reporting period would begin with FY2021 (Q1, Q2, and Q3 2021 and Q4 2020). The next four quarters would be Q3 2020 and Q2, Q3, and Q4 of 2019—that is, past quarters adding up to eight quarters but omitting Q1 and Q2 of 2020, which were exempt from quality reporting (please see section 10.b.(2) above, “Update on Use of Q4 2019 Data and Data Freeze for Refreshes in 2021”). As provided in sections III F(3). “Addition of a “claims-based index measure”, the Hospice Care Index” and III F(4). “Update on the Hospice Visits in the Last Days of Life (HVLDL) and Hospice Item Set V3.00”, we gave sufficient information in the proposed rule and this final rule to calculate HCI and HVLDL and access specifications. The HQRP will post a revised QM Users' Manual that contains HCI and HVLDL no later than October 1, 2021 at: https://www.cms.gov/​Medicare/​Quality-Initiatives-Patient-Assessment-Instruments/​Hospice-Quality-Reporting/​Current-Measures.

Comment: We received several comments expressing concern about the timing for publicly reporting HVLDL and HCI on Care Compare and the Provider Data Catalogue. Commenters requested sufficient time to understand the measures, set up monitoring systems (sometimes with vendor support), assess trends in their performance relative to national benchmarks, and develop plans for quality improvement, as CMS normally provides. One noted that this time is needed in particular because visits on claims have not previously impacted hospice quality scores or payment. Others noted that the delay could allow time for additional analysis of the measure, and for more transparency about the rationale for it. Many of these commenters requested that CMS wait a year (until 2023) to publicly report the measures, while also requesting to confidential reports with the claims-based measures as soon as possible. One commenter requested a minimum of 6 months from the date final specifications are available for EMR and other vendors to respond to any changes in the HQRP.

Response: As stated in section III F(3)(e). “Form, Manner and Timing of Data Collection and Submission”, we have provided and will consolidate in the Users' Manual specifications for HCI and HVLDL in time to meet commenters' stated needs. In addition, we will provide hospices with confidential reporting of their HVLDL and HCI measure scores in the Agency-Level QM report after this rule is finalized—after August 2021. This would allow sufficient time to complete the activities related, which is what we normally aim to give providers to understand and prepare for public reporting of a new measure, if we publicly report in May 2022. We believe that the QM report and Provider Preview report will provide an indication on how well the hospice is performing as well as opportunities to provide CMS feedback on technical issues with the measures. To further support the hospice community, we will also provide education, training, and additional opportunities for hospices to receive information about the measures through open door forums or other venues.

Although these measures represent the first time that hospices are held accountable for visits information in claims, the measures reflect ideas about best practice and compliance that hospices have already known. While we are committed to provide time for understanding and preparation, we are not committed to ensuring that all hospices achieve high scores on the new measures before publicly reporting them. For these reasons, we believe that no additional dry run period is warranted.

Comment: A commenter suggested that CMS should not use claims data from a time period before a measure is finalized through rulemaking.

Response: Our practice across all PAC settings has been to allow the use of claims data originating from before the finalization of a proposal to adopt a claims-based measure. For example, for the Home Health QRP, we finalized the Potentially Preventable 30-Day Post-Discharge Readmission Measure in the CY 2017 Home Health QRP Rule (81 FR 76770 through 76775) for reporting with three consecutive years of claims data beginning with the CY 2018 Home Health QRP.

Comment: Commenters recommended using simple language to describe HVLDL on Care Compare, to ensure that the average consumer will understand it. For HVLDL, one commenter suggested that CMS notify consumers that the measure does not capture visits from chaplains, volunteers, hospice aides, and complementary therapies, among others. For HCI, several commenters expressed concern about CMS's ability to help consumers interpret it in a way that helps support informed decision-making. For example, an average consumer might misinterpret higher scores for live discharges or avoidance of general inpatient care as favorable.

Response: We also believe in the importance of using simple language on Care Compare to ensure consumers can easily use and appropriately interpret quality information that we provide for their decision-making. As with any measure included in the HQRP, we are committed to providing all users with the necessary information to understand the intent and application of measures in the HQRP. Before we publicly report this measure, we will provide resources to aid the public in interpreting publicly displayed quality data. For HVLDL specifically, we will list the multi-disciplinary team member visits that are included in the measure as part of the measure description displayed on Care Compare.

For the public display of HCI, our measure development contractor convened two small caregiver workgroups to gather impressions and input on the value of HCI for consumers. The caregivers were generally receptive and positive about the HCI as an additional measure for the Hospice QRP, and expressed interest in the indicator-level information as well as the index score to better understand the hospice. Their response confirmed our understanding that the data included in HCI will be useful for patients and families as they compare and select hospice providers. Based on the caregivers' feedback, we proposed reporting the HCI as a single score to report on Care Compare, while providing the indicator scores in the Provider Data Catalog (PDC). We will continue to apply ideas shared by the Caregiver Workgroup participants as we refine plans for the measure's public display to minimize the risk of misinterpretation.

Final Decision: We are finalizing as proposed to publicly report the HCI and HVLDL beginning no earlier than May 2022, and to include it in the Preview Reports no sooner than the May 2022 refresh.

(4). Update on Publicly Reporting for the “Hospice Visits When Death is Imminent (HVWDII) Measure 1” and the “Hospice Visits in the Last Days of Life (HVLDL) Measure”

As discussed earlier, the HIS V3.00 PRA Submission, CMS-10390 (OMB control number: 0938-1153), finalized the proposal to replace the HVWDII measure pair with a re-specified version called HVLDL, which is a single measure based on Medicare claims. Relatedly, in the HIS V3.00 PRA Submission, CMS-10390 (OMB control number: 0938-1153), we finalized the proposal to remove Section O from the HIS. As stated in section 1814(i)(5)(E) of the Act, we establish procedures for making all quality data submitted by hospices under § 418.312 available to the public. Thus, we would have continued to publicly report HVWDII Measure 1 data through the November 2021 refresh. Because of the data freeze, HVWDII Measure 1 data from the November 2020 refresh, covering HIS admissions during Q1 through Q4 2019, will be publicly displayed for all calendar year 2021 refreshes. We may retain the November 2020 refresh for HVWDII Measure 1 for one or more refreshes in 2022, when there will be no HIS Section O data, if doing so will allow us to consolidate changes and thus operate more efficiently.

D. Update on Transition From Hospice Compare to Care Compare and Provider Data Catalog

In September 2020, we launched Care Compare, a streamlined redesign of eight existing CMS healthcare compare tools available on Medicare.gov, including Hospice Compare. Care Compare provides a single user-friendly interface that patients and family caregivers can use to make informed decisions about healthcare based on cost, quality of care, volume of services, and other data. With just one click, patients can find information that is easy to understand about doctors, hospitals, nursing homes, and other health care services instead of searching through multiple tools.

For the last six years, Medicare's Hospice Compare has served as the cornerstone for publicizing quality care information for patients, family caregivers, consumers, and the healthcare community. The new website builds on the eMedicare initiative to deliver simple tools and information to current and future Medicare beneficiaries. Drawing on lessons learned through research and stakeholder feedback, Care Compare includes features and functionalities that appeal to Hospice Compare consumers. By offering an accessible and user-friendly interface and a simple design that is optimized for mobile and tablet use, it is easier than ever to find information that is important to patients when shopping for healthcare. Enhancements for mobile use will give practical benefits like accessing the tool using a smartphone that can initiate phone calls to providers simply by clicking on the provider's phone number.

In conjunction with the Care Compare launch, we have made additional improvements to other CMS data tools, to help Medicare beneficiaries compare costs. Specifically, the Provider Data Catalog (PDC) better serves innovators and stakeholders who are interested in detailed CMS data and use interactive and downloadable datasets like those currently available on data.Medicare.gov. The PDC now makes quality datasets available through an improved Application Programming Interface (API), allowing innovators in the field to easily access and analyze the CMS publicly-reported data and make it useful for patients.

e. Update on Additional Information on Hospices for Public Reporting

In the FY 2019 Hospice Wage Index and Payment Rate Update and Hospice Quality Reporting Requirements final rule (83 FR 38622), we finalized plans to publicly post information from the Medicare Provider Utilization and Payment Data: Hospice Public Use File (PUF) and other publicly-available CMS data to Hospice Compare or another CMS website. Hospice PUF data are available for CY 2014 through CY 2016. Beginning with CY 2017 data, hospice PUF data are public as part of the Post-Acute Care and Hospice Provider Utilization and Payment PUF (hereafter PAC PUF). For more information, please visit the PAC PUF web page at: https://www.cms.gov/​Research-Statistics-Data-and-Systems/​Statistics-Trends-and-Reports/​Medicare-Provider-Charge-Data/​PAC2017. Both the Hospice and PAC PUFs provide information on services provided to Medicare beneficiaries by hospice providers. Specifically, they contain information on utilization, payment (Medicare payment and standard payment), submitted charges, primary diagnoses, sites of service, and beneficiary demographics organized by CCN (6-digit provider identification number) and state.

PUF data, along with clear text explaining the purpose and uses of this information and suggesting consumers discuss this information with their healthcare provider, first displayed in a consumer-friendly format on Hospice Compare in May 2019. Beginning May 2021, we will begin to display additional information from the PAC PUF on Care Compare. This additional information includes hospices' beneficiary characteristics such as the percentage of patients enrolled in Medicare Advantage. In addition, consumers will see whether a hospice provided services to Medicare Advantage enrollees or patients who have coverage under both Medicaid and Medicare, also called dual eligible patients. The data for these additional characteristics are pulled directly from the PAC PUF file and provide potential hospice service patients and family caregivers with more detail prior to selecting a hospice.

As finalized in the FY 2019 Hospice Wage Index and Payment Update final rule (83 FR 38622), we also improved access to publicly-available information about hospices' compliance with Hospice QRP requirements. Specifically, we already post the annual Hospice APU Compliant List on the HQRP Requirements and Best Practices web page. This document displays the CCN, name, and address of every hospice that successfully met quality reporting program requirements for the fiscal year. Hospices are only considered compliant if they meet the standards for HIS and CAHPS reporting, as codified in § 418.312. Consumers can now access the Hospice APU compliance file from Care Compare, enabling them to determine if a particular hospice is compliant with CMS' quality reporting requirements.

G. January 2022 HH QRP Public Reporting Display Schedule with Fewer than Standard Number of Quarters Due to COVID-19 Public Health Emergency Exemptions

1. Background and Statutory Authority

We include this Home Health proposal in this rule because we plan to resume public reporting for the HH QRP with the January 2022 refresh of Care Compare. In order to accommodate the exception of 2020 Q1 and Q2 data, we are proposing to resume public reporting using 3 out of 4 quarters of data for the January 2022 refresh. In order to finalize this proposal in time to release the required preview report related to the refresh, which we release 3 months prior to any given refresh (October 2021), we need the rule containing this proposal to finalize by October 2021.

The HH QRP is authorized by section 1895(b)(3)(B)(v) of the Act. Section 1895(b)(3)(B)(v)(II) of the Act requires that for 2007 and subsequent years, each HHA submit to the Secretary in a form and manner, and at a time, specified by the Secretary, such data that the Secretary determines are appropriate for the measurement of health care quality. To the extent that an HHA does not submit data in accordance with this clause, the Secretary shall reduce the home health market basket percentage increase applicable to the HHA for such year by 2 percentage points. As provided at section 1895(b)(3)(B)(vi) of the Act, depending on the market basket percentage increase applicable for a particular year, the reduction of that increase by 2 percentage points for failure to comply with the requirements of the HH QRP and further reduction of the increase by the productivity adjustment (except in 2018 and 2020) described in section 1886(b)(3)(B)(xi)(II) of the Act may result in the home health market basket percentage increase being less than 0.0 percent for a year, and may result in payment rates under the Home Health PPS for a year being less than payment rates for the preceding year. For more information on the policies we have adopted for the HH QRP, we refer readers to the following rules:

• CY 2007 HH PPS final rule (71 FR 65888 through 65891).
• CY 2008 HH PPS final rule (72 FR 49861 through 49864).
• CY 2009 HH PPS update notice (73 FR 65356).
• CY 2010 HH PPS final rule (74 FR 58096 through 58098).
• CY 2011 HH PPS final rule (75 FR 70400 through 70407).
• CY 2012 HH PPS final rule (76 FR 68574).
• CY 2013 HH PPS final rule (77 FR 67092).
• CY 2014 HH PPS final rule (78 FR 72297).
• CY 2015 HH PPS final rule (79 FR 66073 through 66074).
• CY 2016 HH PPS final rule (80 FR 68690 through 68695).
• CY 2017 HH PPS final rule (81 FR 76752).
• CY 2018 HH PPS final rule (82 FR 51711 through 51712).
• CY 2019 HH PPS final rule with comment period (83 FR 56547).
• CY 2020 HH PPS final rule (84 FR 60554 through 60611).
• CY 2021 HH PPS final rule (85 FR 70326 through 70328).

2. Public Display of Home Health Quality Data for the HH QRP

Section 1895(b)(3)(B)(v)(III) of the Act requires the Secretary to establish procedures for making HH QRP data, including data submitted under sections 1899B(c)(1) and 1899B(d)(1) of the Act, available to the public. Such public display procedures must ensure that HHAs have the opportunity to review the data that will be made public with respect to each HHA prior to such data being made public. Section 1899B(g) of the Act requires that data and information regarding PAC provider performance on quality measures and resource use or other measures be made publicly available beginning not later than 2 years after the applicable specified “application date”.

We established our HH QRP Public Display Policy in the CY 2016 HH PPS final rule (80 FR 68709 through 68710). In that final rule, we noted that the procedures for HHAs to review and correct their data on a quarterly basis is performed through CASPER along with our procedure to post the data for the public on our Care Compare website. We have communicated our public display schedule, which supports our Public Display Policy, on our websites whereby the quarters of data included are announced.

3. Proposal To Modify HH QRP Public Reporting To Address CMS' Guidance To Except Data During the COVID-19 PHE Beginning January 2022 Through July 2024

We proposed to modify our public display schedule to display fewer quarters of data than what we previously finalized for certain HH QRP measures for the January 2022 refresh. Under authority of section 319 of the PHS Act, the Secretary declared a PHE effective as of January 27, 2020. On March 13, 2020, the President declared a national state of emergency under the Stafford Act, effective March 1, 2020, allowing the Secretary to invoke section 1135(b) of the Act (42 U.S.C. 1320b-5) to waive or modify the requirements of titles XVIII, XIX, and XXI of the Act and regulations to the extent necessary to address the COVID-19 PHE. Many waivers and modifications were made effective as of March 1, 2020 in accordance with the President's declaration.^[51]

On March 27, 2020, we sent a guidance memorandum under the subject title, “Exceptions and Extensions for Quality Reporting Requirements for Acute Care Hospitals, PPS-Exempt Cancer Hospitals, Inpatient Psychiatric Facilities, Skilled Nursing Facilities, Home Health Agencies (HHAs), Hospices, Inpatient Rehabilitation Facilities, Long-Term Care Hospitals, Ambulatory Surgical Centers, Renal Dialysis Facilities, and MIPS Eligible Clinicians Affected by COVID-19” to the MLN Connects Newsletter and Other Program-Specific Listserv Recipients,^[52] hereafter referred to as the March 27, 2020 CMS Guidance Memorandum. In the March 27, 2020 CMS Guidance Memo, we granted an exception to the HH QRP reporting requirements under the HH QRP exceptions and extension requirements for Quarter 4 (Q4) 2019 (October 1, 2019 through December 31, 2019), Q1 2020 (January 1, 2020 through March 30, 2020), and Q2 2020 (April 1, 2020 through June 30, 2020). The HH QRP exception applied to the HH QRP Outcome and Assessment Information Set (OASIS)-based measures, claims-based measures, and HH CAHPS Survey. We discuss the impact to the OASIS and claims here, and discuss to the HH CAHPS further in section III.G. 4, Update on Use of Q4 2019 HH QRP Data and Data Freeze for Refreshes in 2021. For the OASIS, the exempted quarters are based upon admission and discharge assessments.

A subset of the HH QRP measures has been publicly displayed on Home Health Compare (HH Compare) since 2003. Under the current HH QRP public display policy, Home Health Compare uses 4 quarters of data to publicly display OASIS-based measures, and 4 or more quarters of data to publicly display claims-based measures. We use four rolling quarters of data to publicly display Home Health Care Consumer Assessment of Healthcare Providers and Systems (HH CAHPS) Survey measures on Care Compare. As of September 2020, HH QRP OASIS, claims-based, and HH CAHPS Survey measures are reported on the www.medicare.gov' s Care Compare website. As of December 2020, the data is no longer reported on the www.medicare.gov' s Home Health Compare website.

The exception granted under the March 27, 2020 CMS Guidance Memo impacted the HH QRP public display schedule. We proposed resuming publicly displaying HH QRP claims-based measures in January 2022 based upon the quarters of data specified for each of the claims-based measures. Table 20 displays the original schedule for public reporting of OASIS and HH CAHPS Survey measures prior to the Q1 and Q2 2020 data impacted by the COVID-19 PHE.

During the spring and summer of 2020, we conducted testing to inform decisions about publicly displaying HH QRP data for those refreshes which include data from the exception period of October 1, 2019 through June 30, 2020 (hereafter “excepted data”). The testing helped us develop a plan for displaying HH QRP data that are as up-to-date as possible and that also meet scientifically-acceptable standards for publicly displaying those data. We believe that the plan allows us to provide consumers with helpful information on the quality of home health care, while also making the necessary adjustments to accommodate the exception granted to HHAs. The following sections provide the results of our testing for OASIS and claims and explain how we used the results to inform a proposal for accommodating excepted data in public reporting. HH CAHPS discussion is further in section III.G.4.

4. Update on Use of Q4 2019 HH QRP Data and Data Freeze for Refreshes in 2021

In the March 27, 2020 Guidance Memorandum, we stated that we should not include any PAC quality data that are greatly impacted by the exception granted in the quality reporting programs. Given the timing of the COVID-19 PHE onset, we determined that we would not use HH QRP OASIS, claims, or HH CAHPS data from Q1 and Q2 of 2020 for public reporting, and that we would assess the impact of the COVID-19 PHE on HH QRP data from Q4 2019. In the original schedule (Table 20), the October 2020 refresh included Q4 2019 measure based on OASIS and HH CAHPS data and is the last refresh before Q1 2020 data are included.

Before proceeding with the October 2020 refresh, we conducted testing to ensure that publicly displaying Q4 2019 data would still meet our standards despite granting an exception to HH QRP reporting requirements for Q4 2019. Specifically, we compared submission rates in Q4 2019 to average rates in other quarters to assess the extent to which HHAs had taken advantage of the exception, and thus the extent to which data and measure scores might be affected. We observed that the quality data submission rate for Q4 2019 was in fact 0.4 percent higher than the previous calendar year (Q4 2018). We note that Q4 2019 ended before the onset of the COVID-19 pandemic in the U.S. Thus, we proceeded with including Q4 2019 data in measure calculations for the October 2020 refresh.

Because we excepted HHAs from the HH QRP reporting requirements for Q1 and Q2 2020, we did not use OASIS, claims, or HH CAHPS data from these quarters. All refreshes, during which we decided to hold this data constant, included more than 2 quarters of data that were affected by the CMS-issued COVID reporting exceptions, thus we did not have an adequate amount of data to reliably calculate and publicly display provider measures scores. Consequently, we determined to freeze the data displayed, that is, holding data constant after the October 2020 refresh without subsequently updating the data through October 2021. We communicated this in a Public Reporting Tip Sheet, which is located at: https://www.cms.gov/​files/​document/​hhqrp-pr-tip-sheet081320final-cx-508.pdf.

5. Application of the COVID-19 PHE Affected Reporting (CAR) Scenario To Publicly Display Certain HH QRP Measures (Beginning in January 2022 Through July 2024)

We also proposed to use the CAR scenario for refreshes for January 2022 for OASIS and for refreshes from January 2022 through July 2024 for some claims-based measures. There are several forthcoming HH QRP refreshes for which the original public reporting schedule included other quarters from the quality data submission exception. These refreshes for claims-based measures, OASIS-based measures, and for HH CAHPS Survey measures are outlined in Table 20.

Because October 2020 refresh data will become increasingly out-of-date and thus less useful for the public, we analyzed whether it would be possible to use fewer quarters of data for one or more refreshes and thus reduce the number of refreshes that continue to display October 2020 data. Using fewer quarters of more up-to-date data requires that: (1) A sufficient percentage of HHAs would still likely have enough OASIS data to report quality measures (reportability); and (2) using fewer quarters of data to calculate measures would likely produce similar measure scores for HHAs, and thus not unfairly represent the quality of care HHAs provided during the period reported in a given refresh (reliability).

To assess these criteria, we conducted reportability and reliability analysis excluding the COVID-19 affected quarters of data in a refresh instead of the standard number of quarters of data for reporting for each HH QRP measure to model the impact of not using Q1 or Q2 2020 Specifically, we used historical data to calculate HH quality measures under two scenarios:

• Standard Public Reporting (SPR) Scenario: We used HH QRP data from CY 2017 through 2019 to build the standard reported measures, to represent as a proxy CY 2020 public reporting in the absence of the temporary exemptions from the submission of OASIS quality data, as the basis for comparing simulated alternatives. This entails using 4 quarters of CY 2019 HH QRP data to model the OASIS based measures that are normally calculated using 4 quarters of data. This also entailed using 4 quarters of HH QRP data from CY 2019 for the all-cause hospitalization and emergency department use claims-based measures, 8 quarters of HH QRP data from CY2018 and CY2019 for Medicare spending per beneficiary (MSPB) and discharge to community (DTC) claims-based measures; and or 12 quarters from January 2017 to December 2019 for the potentially preventable readmission claims-based measure.
• COVID-19 Affected Reporting (CAR) Scenario: We calculated OASIS-based measures using 3 quarters of HH QRP CY 2019 data to simulate using only Q3 2020, Q4 2020, and Q1 2021 data for public reporting. We calculated claims-based measures using HH QRP CY 2017 to 2019 data, to simulate using the most recent data while excluding the same quarters (Q1 and Q2) that are relevant from the COVID-19 PHE exception. We used 3 quarters of HH QRP data from CY 2019 for the all-cause hospitalization and emergency department use claims-based measures and 6 quarters of data from HH QRP CY 2018 and CY 2019 were used for both the Medicare spending per beneficiary and discharge to community claims-based measures. We used 10 quarters of HH QRP data from CY 2017 to 2019 to calculate the CAR scenario for the potentially preventable readmissions claims-based measure. For both claims and OASIS-based measures, the quarters used in our analysis were the most recently available data that exclude the same quarters (Q1 and Q2) as that are relevant from the COVID-19 PHE exception, and thus take seasonality into consideration.

The OASIS-based measures are based on the start of care and calculated using admission dates. Therefore, under the CAR scenario we excluded data for OASIS-based measures for HHA patient stays with admission dates in Q1 and Q2 2019. To assess performance in these scenarios, we calculated the reportability as the percent of HHAs meeting the 20-case minimum for public reporting (the public reporting threshold, or “PRT”). We evaluated measure reliability using the Pearson and Spearman correlation coefficients, which assess the alignment of HHs measure scores between scenarios. To calculate the reliability results, we restricted the HHAs included in the SPR Scenario to those included in the CAR Scenario.

Testing results showed that using the CAR scenario would achieve scientifically acceptable quality measure scores for the HH QRP. As displayed in Table 21, the percentage of HHAs that met the public display threshold for the OASIS-based measure decreases by 5.5 percentage points or less for all but one QM, the Influenza Immunization for the Current Flu Season in the CAR scenario versus SPR scenario. CMS has traditionally used a reportability threshold of 70 percent, meaning at least 70 percent of HHAs are able to report at least 20 episodes for a given measure, as the standard to determine whether a measure should be publicly reported. By this standard, we consider a decrease of 5.5 percentage points or less scientifically acceptable. The change in reportability for the Influenza Immunization for the Current Flu Season measure is related to the seasonality of this measure, which includes cases that occur during the flu season only.

Under the CAR scenario, the January 2022 refresh data would cover Q3 and Q4 of 2020 and Q1 of 2021, which occur during the flu season. This simulation included Q2 through Q4 of 2019, which crosses the flu season. Thus, the reportability of the actual data used is likely to be better than this simulation. Therefore, in general, using CAR scenario for the OASIS and claims-based measures would achieve acceptable reportability for the HH QRP measures. Testing also yielded correlation coefficients above 0.85, indicating a high degree of agreement between HH measure scores when using the CAR scenario or the SPR scenario.

We proposed to use the CAR scenario for the last of the refreshes affecting OASIS-based measures, which will occur in January 2022. We also proposed to use the CAR scenario for refreshes from January 2022 through July 2024 for some claims-based measures.

Our proposal to adopt the CAR scenario for the January 2022 refresh would allow us to begin displaying recent data in January 2022, rather than continue displaying October 2020 data (Q1 2019 through Q4 2019). We believe that updating the data in January 2022 by more than a year relative to the October 2020 freeze data can assist the public by providing more relevant quality data and allow CMS to display more recent HHA performance. Similarly, using fewer than standard numbers of quarters for claims-based measures that typically use eight or twelve months of data for reporting between January 2022 and July 2024 will allow us to begin providing more relevant data sooner. Our testing results indicate we can achieve these positive impacts while maintaining high standards for reportability and reliability. Table 22 and Table 23 summarize the comparison between the original schedule for public reporting with the revised schedule (that is, frozen data) and also with the proposed public display schedule under the CAR scenario (that is, using 3 quarters in the January 2022 refresh), for OASIS- and claims-based measures respectively.

We solicited public comments on the proposal to use the CAR scenario to publicly report HH OASIS in January 2022 and claims-based measures beginning with the January 2022 through July 2024 refreshes. A summary of the comments we received on this proposal and our responses to those comments appear below:

Comment: We received many comments supporting HH QRP reporting to resume beginning January 2022. One commenter suggested including a statement that data cover care provided during the COVID-19 PHE for eight quarters.

Response: We thank commenters for their support of this proposal on public reporting for refreshes affected by the exceptions. However, we do not agree with the commenter who suggested including a statement on Care Compare regarding the inclusion of data from the COVID-19 PHE because such an announcement will not help consumers distinguish between HHAs in their region. Instead, we will continue to post state and national averages for HH QRP measures. This information will help consumers understand relative performance at national and local levels in light of the COVID-19 PHE.

Given the overall positive response to our proposal, we believe that the proposed approach balances fairness to providers with a commitment to transparency and information for consumers.

Final Decision: We are finalizing our proposal to use the CAR scenario for refreshes for January 2022 for OASIS-based measures and for refreshes from January 2022 through July 2024 for some claims-based measures.

6. Update to the Public Display of HH CAHPS Measures Due to the COVID-19 PHE Exception

Since April 2012, we have publicly displayed four quarters of HH CAHPS data every quarter, in the months of January, April, July, and October. The COVID-19 PHE Exception applied to Q1 and Q2 of 2020. Those excepted quarters cannot be publicly displayed and resulted in the freezing of the public display using Q1 2019 through Q4 2019 data for the refreshes that would have occurred from October 2020 through October 2021, as shown in Table 24. Beginning with January 2022, we will resume reporting four quarters of HH CAHPS data. The data for the January 2022 refresh are Q3 2020 through Q2 2021. These are the same quarters that would have been publicly displayed despite the COVID-19 PHE. Table 24 summarizes this discussion.

IV. Requests for Information

A. Fast Healthcare Interoperability Resources (FHIR) in Support of Digital Quality Measurement in Post-Acute Care Quality Reporting Programs—Request for Information

Through the proposed rule, we sought input on the following steps that would enable transformation of CMS' quality measurement enterprise to be fully digital (86 FR 19765):

a. What EHR/IT systems do you use and do you participate in a health information exchange (HIE)?

b. How do you currently share information with other providers and are there specific industry best practices for integrating SDOH screening into EHR's?

c. What ways could we incentivize or reward innovative uses of health information technology (IT) that could reduce burden for post-acute care settings, including but not limited to hospices?

d. What additional resources or tools would post-acute care settings, including but not limited to hospices and health IT vendors find helpful to support testing, implementation, collection, and reporting of all measures using FHIR standards via secure APIs to reinforce the sharing of patient health information between care settings?

e. Would vendors, including those that service post-acute care settings, including but not limited to hospices, be interested in or willing to participate in pilots or models of alternative approaches to quality measurement that would align standards for quality measure data collection across care settings to improve care coordination, such as sharing patient data via secure FHIR API as the basis for calculating and reporting digital measures?

f. What could be the potential use of FHIR dQMs that could be adopted across all QRPs?

We plan to continue working with other agencies and stakeholders to coordinate and to inform our transformation to dQMs leveraging health IT standards. While we stated that we would not be responding to specific comments submitted in response to this Request for Information in the FY 2022 Hospice Wage Index final rule, we will actively consider all input as we develop future regulatory proposals or future sub-regulatory policy guidance. Any updates to specific program requirements related to quality measurement and reporting provisions would be addressed through separate and future notice- and-comment rulemaking, as necessary.

Comments: We received many comments expressing support for the adoption of a standardized definition of dQM in the hospice setting and the use of Fast Healthcare Interoperability Resources (FHIR) to support quality measurements in the HQRP. Many commenters noted that there is a great deal of variation among FHIR systems, which could impede the adoption of a standard system across hospices. Commenters also expressed issues surrounding interoperability capabilities of EHR vendor systems noting that currently, some EHR vendors do not include features important for interoperability as a part of their base product, which would represent additional costs for hospices which can lead to affordability issues for many providers. Furthermore, commenters noted that interoperability challenges lead to complications when sharing health information with other providers. They encouraged HHS to continue pursuing adoption of FHIR APIs for health IT vendors.

We also received several comments responding to how CMS should incentivize the use of HIT. Commenters noted that hospices were not included in the EHR Incentive Program, which provided grants to hospices to develop HIT systems. We received many comments emphasizing that financial incentives would encourage providers to adopt new HIT systems and work to reduce burden using FHIR and EHR. Commenters also encouraged CMS to provide early testing and education for providers on HIT and to provide a structured FHIR transition framework for key stakeholders.

We also received several comments explaining the various EHR/HIT systems currently in use, as well as discussions surrounding health information exchange with other providers.

Response: While we stated that we would not be responding to specific comments submitted in response to this RFI in the FY 2022 Hospice Wage Index final rule, we appreciate all of the comments and interest in this topic. We will continue to take all concerns, comments, and suggestions into account as we consider Fast Healthcare Interoperability Resources (FHIR) in support of Digital Quality Measurement in Post-Acute Care Quality Reporting Programs.

B. Closing the Health Equity Gap in Post-Acute Care Quality Reporting Programs—Request for Information

While hospice is not included in the Improving Medicare Post-Acute Care Transformation Act of 2014 (IMPACT Act of 2014) (Pub. L. 113-185), we sought comment on the possibility of revising measure development, and the collection of other data that address gaps in health equity in HQRP (86 FR 19766). Any potential health equity data collection or measure reporting within a CMS program that might result from public comments received in response to this solicitation would be addressed through a separate notice-and-comment rulemaking in the future. We invited public comment on the following:

• Recommendations for quality measures, or measurement domains that address health equity, for use in the HQRP.
• Suggested parts of SDOH standardized patient assessment data elements adoption that could apply to hospice in alignment with national data collection and interoperable exchange standards. This could include collecting information on race, ethnicity, and certain SDOH, including preferred language, interpreter services, health literacy, transportation and social isolation. This could also include guidance on any additional items, including standardized patient assessment and data elements that could be used to assess health equity in the care of hospice patients, for use in the HQRP.
• Ways CMS can promote health equity in outcomes among hospice patients. We were also interested in feedback regarding whether including facility-level quality measure results stratified by social risk factors and social determinants of health (and relevant proxies, such as dual eligibility for Medicare and Medicaid, and race) in confidential feedback reports could allow facilities to identify gaps in the quality of care they provide. (For example, methods similar or analogous to the CMS Disparity Methods which provide hospital-level confidential results stratified by dual eligibility for condition-specific readmission measures currently included in the Hospital Readmission Reduction Program (84 FR 42496 through 42500)).
• Methods that commenters or their organizations use in employing data to reduce disparities and improve patient outcomes, including the source(s) of data used, as appropriate.
• Given the importance of structured data and health IT standards for the capture, use, and exchange of relevant health data for improving health equity, the existing challenges providers' encounter for effective capture, use, and exchange of health information, such as data on race, ethnicity, and other social determinants of health, to support care delivery and decision making.

While we stated that we would not be responding to specific comments submitted in response to this RFI in the FY 2022 Hospice Wage Index final rule, we appreciate all of the comments and interest in this topic. We will continue to take all concerns, comments, and suggestions into account as we continue work to address and develop policies on this important topic. It is our hope to provide additional stratified information to providers related to race and ethnicity if feasible. The provision of stratified measure results will allow hospices to understand how they are performing with respect to certain patient risk groups, to support these providers in their efforts to ensure equity for all of their patients, and to identify opportunities for improvements in health outcomes.

2. Public Comments Summarized

We received many comments about the use of standardized patient assessment data in the hospice setting to assess health equity and social determinants of health (SDOH). Many commenters noted a 2019 Abt Associates and RAND Corporation study which excluded hospices from the standardized data elements for patient assessment denominator, citing that hospice patients have a different goal of care which does not align with standardized data elements for patient assessment. Commenters encouraged CMS to only utilize certain aspects of standardized data elements for patient assessment (specifically, Z-codes 55-65) in collecting health equity data. We also received some comments which expressed that standardized data elements for patient assessment does not currently capture the current understanding of SDOH.

We also received feedback from several commenters about additional factors which should be considered when collecting data about health equity and disparities. We noted several categories, including: culture, spiritual beliefs, food insecurity, access to interpreter services, health literacy, caregiving, housing scarcity, marital status, and socioeconomic status. Commenters encouraged CMS to stratify quality measures by demographic data, social risk factors, and social determinants of health.

We also noted a comment encouraging CMS to implement a best-practice assessment for the collection of demographic and SDOH data. A commenter noted that there is not a standard initial nursing or social worker assessment that currently screens for SDOH.

One commenter also expressed a desire to include permanent telehealth provisions in the QRP, as that would help improve rural healthcare access.

We appreciate all the comments and interest in this topic. We believe that this input is very valuable in the continuing development of the CMS health equity quality measurement efforts. We will continue to take all concerns, comments, and suggestions into consideration for future development and expansion of our health equity quality measurement efforts.

V. Waiver of Proposed Rulemaking

We ordinarily publish a notice of proposed rulemaking in the Federal Register and invite public comment on the proposed rule before the provisions of the rule are finalized, either as proposed or as amended in response to public comments, and take effect, in accordance with the Administrative Procedure Act (APA) (Pub. L. 79-404), 5 U.S.C. 553, and, where applicable, section 1871 of the Act. Specifically, 5 U.S.C. 553 requires the agency to publish a notice of the proposed rule in the Federal Register that includes a reference to the legal authority under which the rule is proposed, and the terms and substance of the proposed rule or a description of the subjects and issues involved. Further, 5 U.S.C. 553 requires the agency to give interested parties the opportunity to participate in the rulemaking through public comment before the provisions of the rule take effect. Similarly, section 1871(b)(1) of the Act requires the Secretary to provide for notice of the proposed rule in the Federal Register and a period of not less than 60 days for public comment for rulemaking carrying out the administration of the insurance programs under title XVIII of the Act. Section 1871(b)(2)(C) of the Act and 5 U.S.C. 553 authorize the agency to waive these procedures, however, if the agency for good cause finds that notice and comment procedures are impracticable, unnecessary, or contrary to the public interest and incorporates a statement of the finding and its reasons in the rule issued.

We are revising the provisions at § 418.306(b)(2) to change the payment reduction for failing to meet hospice quality reporting requirements from 2 to 4 percentage points. This policy will apply beginning with FY 2024 annual payment update (APU). Specifically, the Act requires that, beginning with FY 2014 through FY 2023, the Secretary shall reduce the market basket update by 2 percentage points and beginning with the FY 2024 APU and for each subsequent year, the Secretary shall reduce the market basket update by 4 percentage points for any hospice that does not comply with the quality data submission requirements for that FY. We noted this revised statutory requirement in our proposed rule (86 FR 19726) and are codifying the revision at § 418.306(b)(2). While we received comments, this update is statutorily required and self-implementing. Notice and comment are unnecessary because we are conforming the regulation to statute and there is no discretion on the part of the Secretary.

VI. Collection of Information Requirements

Under the Paperwork Reduction Act of 1995, we are required to provide 30-day notice in the Federal Register and solicit public comment before a collection of information requirement is submitted to OMB for review and approval. In order to fairly evaluate whether an information collection should be approved by OMB, section 3506(c)(2)(A) of the Paperwork Reduction Act of 1995 requires that we solicit comment on the following issues:

• The need for the information collection and its usefulness in carrying out the proper functions of our agency.
• The accuracy of our estimate of the information collection burden.
• The quality, utility, and clarity of the information to be collected.
• Recommendations to minimize the information collection burden on the affected public, including automated collection techniques.

A. ICRs Regarding Hospice QRP

We are revising the provisions at § 418.306(b)(2) to change the payment reduction for failing to meet hospice quality reporting requirements from 2 to 4 percentage points. This policy will apply beginning with FY 2024 annual payment update (APU). Specifically, the Act requires that, beginning with FY 2014 through FY 2023, the Secretary shall reduce the market basket update by 2 percentage points and beginning with the FY 2024 APU and for each subsequent year, the Secretary shall reduce the market basket update by 4 percentage points for any hospice that does not comply with the quality data submission requirements for that FY. We noted this revised statutory requirement in our proposed rule (86 FR 19726) and are codifying the revision at § 418.306(b)(2). While we received comments, this update is statutorily required and self-implementing. Notice and comment are unnecessary because we are conforming the regulation to statute and there is no discretion on the part of the Secretary. The HQRP proposals would not change provider burden or costs.

• For the proposal to remove the 7 HIS measures from the HQRP, we do not propose any changes to the requirement to submit the HIS admission assessment since we continue to collect the data for these 7 HIS measures in order to calculate the more broadly applicable NQF # 3235, the Hospice and Palliative Care Composite Process Measure—HIS-Comprehensive Assessment Measure at Admission.
• The proposal to add the HCI also would not change provider burden or costs since it is a claims-based measure that CMS calculates from the Medicare claims data.
• Likewise, the proposal to publicly report the claims-based HVLDL quality measure would not result in reduced provider burden and related costs. The reduction in provider burden and costs occurred when we replaced the HIS-based HVWDII quality measure via the HIS-information collection request (ICR) -CMS-10390 (OMB Control Number: 0938-1153 (Expiration date: February 29, 2024).
• Finally, the Home Health proposal would not change provider burden or costs since it only affects the number of quarters used in the calculation of certain claims-based measures for the public display for certain refresh cycles.

B. ICRs Regarding Hospice CoPs

We are revising the provisions at § 418.76(c)(1) that requires the hospice aide to be evaluated by observing an aide's performance of the task with a patient. This revision is subject to the PRA; however, the information collection burden associated with the existing requirements at § 418.76(c)(1) are accounted for under the information collection request currently approved OMB control number 0938-1067 (Expiration date: March 31, 2024). We requested public comment in determining if the time and effort necessary to comply with implementing the use of the pseudo-patient for hospice aide training at § 418.76(c)(1) would reduce burden on the provider. While comments were overwhelmingly supportive, we did not receive any comments that would support burden changes.

We are also revising the provisions at § 418.76(h)(1)(iii) to state that if an area of concern is verified by the hospice during the on-site visit, then the hospice must conduct, and the hospice aide must complete, a competency evaluation related to the deficient and related skill(s) in accordance with § 418.76(c). While many commenters indicated that the proposed changes increase efficiency of training, none provided specific information or data to describe a change in burden. Additionally, we believe that both the requirements at § 418.76(h) are exempt from the PRA. In accordance with the implementing regulations of the PRA at 5 CFR 1320.3(b)(2), we believe competency evaluations are a usual and customary business practice and we state as such in the information collection request associated with the Hospice CoPs—CMS-10277 (OMB control number 0938-1067). Therefore, we are not seeking OMB approval for any information collection or recordkeeping activities that may be conducted in connection with the revisions to § 418.76(h).

VII. Regulatory Impact Analysis

A. Statement of Need

This final rule meets the requirements of our regulations at § 418.306(c) and (d), which require annual issuance, in the Federal Register, of the hospice wage index based on the most current available CMS hospital wage data, including any changes to the definitions of CBSAs or previously used MSAs, as well as any changes to the methodology for determining the per diem payment rates. This final rule also updates payment rates for each of the categories of hospice care, described in § 418.302(b), for FY 2022 as required under section 1814(i)(1)(C)(ii)(VII) of the Act. The payment rate updates are subject to changes in economy-wide productivity as specified in section 1886(b)(3)(B)(xi)(II) of the Act. Lastly, section 3004 of the Affordable Care Act amended the Act to authorize a quality reporting program for hospices, and this rule discusses changes in the requirements for the HQRP in accordance with section 1814(i)(5) of the Act.

B. Overall Impacts

We estimate that the aggregate impact of the payment provisions in this rule will result in an increase of $480 million in payments to hospices, resulting from the hospice payment update percentage of 2.0 percent for FY 2022. The impact analysis of this rule represents the projected effects of the changes in hospice payments from FY 2021 to FY 2022. Using the most recent complete data available at the time of rulemaking, in this case FY 2020 hospice claims data as of January 15, 2021, we apply the current FY 2021 wage index with the current labor shares. Using the same FY 2020 data, we apply the FY 2022 wage index and the current labor share values to simulate FY 2022 payments. We then apply a budget neutrality adjustment so that the aggregate simulated payments do not increase or decrease due to changes in the wage index. Then, using the same FY 2020 data, we apply the FY 2022 wage index and the current labor share values to simulate FY 2022 payments and compare simulated payments using the FY 2022 wage index and the proposed revised labor shares. We then apply a budget neutrality adjustment so that the aggregate simulated payments do not increase or decrease due to changes in the labor share values.

Certain events may limit the scope or accuracy of our impact analysis, because such an analysis is susceptible to forecasting errors due to other changes in the forecasted impact time period. The nature of the Medicare program is such that the changes may interact, and the complexity of the interaction of these changes could make it difficult to predict accurately the full scope of the impact upon hospices.

We have examined the impacts of this rule as required by Executive Order 12866 on Regulatory Planning and Review (September 30, 1993), Executive Order 13563 on Improving Regulation and Regulatory Review (January 18, 2011), the Regulatory Flexibility Act (RFA) (September 19, 1980, Pub. L. 96-354), section 1102(b) of the Social Security Act, section 202 of the Unfunded Mandates Reform Act of 1995 (March 22, 1995; Pub. L. 104-4), Executive Order 13132 on Federalism (August 4, 1999), and the Congressional Review Act (5 U.S.C. 804(2)).

Executive Orders 12866 and 13563 direct agencies to assess all costs and benefits of available regulatory alternatives and, if regulation is necessary, to select regulatory approaches that maximize net benefits (including potential economic, environmental, public health and safety effects, distributive impacts, and equity). Section 3(f) of Executive Order 12866 defines a “significant regulatory action” as an action that is likely to result in a rule: (1) Having an annual effect on the economy of $100 million or more in any 1 year, or adversely and materially affecting a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or state, local or tribal governments or communities (also referred to as “economically significant”); (2) creating a serious inconsistency or otherwise interfering with an action taken or planned by another agency; (3) materially altering the budgetary impacts of entitlement grants, user fees, or loan programs or the rights and obligations of recipients thereof; or (4) raising novel legal or policy issues arising out of legal mandates, the President's priorities, or the principles set forth in the Executive Order.

In accordance with the provisions of Executive Order 12866, this regulation was reviewed by OMB.

A regulatory impact analysis (RIA) must be prepared for major rules with economically significant effects ($100 million or more in any 1 year). Based on our estimates, OMB's Office of Information and Regulatory Affairs has determined that this rulemaking is “economically significant” as measured by the $100 million threshold, and hence also a major rule under Subtitle E of the Small Business Regulatory Enforcement Fairness Act of 1996 (also known as the Congressional Review Act), 5 U.S.C. 804(2). Accordingly, we have prepared a RIA that, to the best of our ability presents the costs and benefits of the rulemaking.

C. Detailed Economic Analysis

1. Hospice Payment Update for FY 2022

The FY 2022 hospice payment impacts appear in Table 25. We tabulate the resulting payments according to the classifications (for example, provider type, geographic region, facility size), and compare the difference between current and future payments to determine the overall impact. The first column shows the breakdown of all hospices by provider type and control (non-profit, for-profit, government, other), facility location, facility size. The second column shows the number of hospices in each of the categories in the first column. The third column shows the effect of using the FY 2022 updated wage index data. This represents the effect of moving from the FY 2021 hospice wage index to the FY 2022 hospice wage index. The fourth column shows the effect of the final rebased labor shares. The aggregate impact of the changes in column three and four is zero percent, due to the hospice wage index standardization factor and the labor share standardization factor. However, there are distributional effects of the FY 2022 hospice wage index. The fifth column shows the effect of the hospice payment update percentage as mandated by section 1814(i)(1)(C) of the Act, and is consistent for all providers. The 2.0 hospice payment update percentage is based on the 2.7 percent inpatient hospital market basket update, reduced by a 0.7 percentage point productivity adjustment. The sixth column shows the effect of all the proposed changes on FY 2022 hospice payments. It is projected that aggregate payments would increase by 2.0 percent; assuming hospices do not change their billing practices. As illustrated in Table 25, the combined effects of all the proposals vary by specific types of providers and by location.

In addition, we are providing a provider-specific impact analysis file, which is available on our website at https://www.cms.gov/​Medicare/​Medicare-Fee-for-Service-Payment/​Hospice/​Hospice-Regulations-and-Notices.html. We note that simulated payments are based on utilization in FY 2020 as seen on Medicare hospice claims (accessed from the CCW in May 2021) and only include payments related to the level of care and do not include payments related to the service intensity add-on.

As illustrated in Table 25, the combined effects of all the proposals vary by specific types of providers and by location.

2. Regulatory Review Cost Estimation

If regulations impose administrative costs on private entities, such as the time needed to read and interpret this rule, we should estimate the cost associated with regulatory review. Due to the uncertainty involved with accurately quantifying the number of entities that will review the rule, we assume that the total number of unique commenters on last year's proposed rule will be the number of reviewers of this rule. We acknowledge that this assumption may understate or overstate the costs of reviewing this rule. It is possible that not all commenters reviewed last year's rule in detail, and it is also possible that some reviewers chose not to comment on the proposed rule. For these reasons we thought that the number of past commenters would be a fair estimate of the number of reviewers of this final rule. We also recognize that different types of entities are in many cases affected by mutually exclusive sections of the final rule, and therefore, for the purposes of our estimate we assume that each reviewer reads approximately 50 percent of the rule.

Using the wage information from the Bureau of Labor Statistics (BLS) for medical and health service managers (Code 11-9111); we estimate that the cost of reviewing this rule is $114.24 per hour, including overhead and fringe benefits (https://www.bls.gov/​oes/​current/​oes_​nat.htm). This final rule consists of approximately 72,000 words. Assuming an average reading speed of 250 words per minute, it would take approximately 2.4 hours for the staff to review half of it. For each hospice that reviews the rule, the estimated cost is $274.18 (2.4 hour × $114.24). Therefore, we estimate that the total cost of reviewing this regulation is $14,531.54 ($274.18 × 53 reviewers).

D. Accounting Statement and Table

As required by OMB Circular A-4 (available at https://www.whitehouse.gov/​sites/​whitehouse.gov/​files/​omb/​circulars/​A4/​a-4.pdf), in Table 26, we have prepared an accounting statement showing the classification of the expenditures associated with the provisions of this final rule. Table 26 provides our best estimate of the possible changes in Medicare payments under the hospice benefit as a result of the policies in this rule. This estimate is based on the data for 4,995 hospices in our impact analysis file, which was constructed using FY 2020 claims available in May 2021. All expenditures are classified as transfers to hospices.

E. Regulatory Flexibility Act (RFA)

The RFA requires agencies to analyze options for regulatory relief of small businesses if a rule has a significant impact on a substantial number of small entities. The great majority of hospitals and most other health care providers and suppliers are small entities by meeting the Small Business Administration (SBA) definition of a small business (in the service sector, having revenues of less than $8.0 million to $41.5 million in any 1 year), or being nonprofit organizations. For purposes of the RFA, we consider all hospices as small entities as that term is used in the RFA. The Department of Health and Human Services practice in interpreting the RFA is to consider effects economically “significant” only if greater than 5 percent of providers reach a threshold of 3 to 5 percent or more of total revenue or total costs. The effect of the FY 2022 hospice payment update percentage results in an overall increase in estimated hospice payments of 2.0 percent, or $480 million. The distributional effects of the final FY 2022 hospice wage index do not result in a greater than 5 percent of hospices experiencing decreases in payments of 3 percent or more of total revenue. Therefore, the Secretary has certified that this rule will not create a significant economic impact on a substantial number of small entities.

In addition, section 1102(b) of the Act requires us to prepare a regulatory impact analysis if a rule may have a significant impact on the operations of a substantial number of small rural hospitals. This analysis must conform to the provisions of section 604 of the RFA. For purposes of section 1102(b) of the Act, we define a small rural hospital as a hospital that is located outside of a MSA and has fewer than 100 beds. This rule will only affect hospices. Therefore, the Secretary has certified that this rule will not have a significant impact on the operations of a substantial number of small rural hospitals (see Table 25).

F. Unfunded Mandates Reform Act (UMRA)

Section 202 of the Unfunded Mandates Reform Act of 1995 (UMRA) also requires that agencies assess anticipated costs and benefits before issuing any rule whose mandates require spending in any 1 year of $100 million in 1995 dollars, updated annually for inflation. In 2021, that threshold is approximately $158 million. This rule is not anticipated to have an effect on state, local, or tribal governments, in the aggregate, or on the private sector of $158 million or more in any 1 year.

G. Federalism

Executive Order 13132 establishes certain requirements that an agency must meet when it promulgates a proposed rule (and subsequent final rule) that imposes substantial direct requirement costs on state and local governments, preempts state law, or otherwise has Federalism implications. We have reviewed this rule under these criteria of Executive Order 13132, and have determined that it will not impose substantial direct costs on state or local governments.

H. Conclusion

We estimate that aggregate payments to hospices in FY 2022 will increase by $480 million as a result of the market basket update, compared to payments in FY 2021. We estimate that in FY 2022, hospices in urban areas will experience, on average, 2.0 percent increase in estimated payments compared to FY 2021. While hospices in rural areas will experience, on average, 2.2 percent increase in estimated payments compared to FY 2021. Hospices providing services in the Outlying and South Atlantic regions would experience the largest estimated increases in payments of 2.9 percent and 2.5 percent, respectively. Hospices serving patients in areas in the New England and Middle Atlantic regions would experience, on average, the lowest estimated increase of 1.2 percent in FY 2022 payments.

This final regulation is subject to the Congressional Review Act provisions of the Small Business Regulatory Enforcement Fairness Act of 1996 (5 U.S.C. 801 et seq.) and has been transmitted to the Congress and the Comptroller General for review.

I, Chiquita Brooks-LaSure, Administrator of the Centers for Medicare & Medicaid Services, approved this document on July 23, 2021.

List of Subjects in 42 CFR Part 418

• Health facilities
• Hospice care
• Medicare
• Reporting and recordkeeping requirements

For the reasons set forth in the preamble, the Centers for Medicare & Medicaid Services amends 42 CFR chapter IV as set forth below.

PART 418—HOSPICE CARE

1. The authority citation for part 418 continues to read as follows:

Authority: 42 U.S.C. 1302 and 1395hh.

2. Section 418.3 is amended by adding definitions for “Pseudo-patient” and “Simulation” in alphabetical order to read as follows:

§ 418.3

Definitions.

* * * * *

Pseudo-patient means a person trained to participate in a role-play situation, or a computer-based mannequin device. A pseudo-patient must be capable of responding to and interacting with the hospice aide trainee, and must demonstrate the general characteristics of the primary patient population served by the hospice in key areas such as age, frailty, functional status, cognitive status and care goals.

* * * * *

Simulation means a training and assessment technique that mimics the reality of the homecare environment, including environmental distractions and constraints that evoke or replicate substantial aspects of the real world in a fully interactive fashion, in order to teach and assess proficiency in performing skills, and to promote decision making and critical thinking.

* * * * *

3. Section 418.24 is amended by:

a. Revising paragraphs (c) introductory text and (c)(9);

b. Adding paragraph (c)(10);

c. Redesignating paragraphs (d) through (g) as paragraphs (e) through (h); and

d. Adding a new paragraph (d).

The revisions and additions read as follows:

§ 418.24

Election of hospice care.

* * * * *

(c) Content of hospice election statement addendum. For hospice elections beginning on or after October 1, 2020, in the event that the hospice determines there are conditions, items, services, or drugs that are unrelated to the individual's terminal illness and related conditions, the individual (or representative), non-hospice providers furnishing such items, services, or drugs, or Medicare contractors may request a written list as an addendum to the election statement. The election statement addendum must include the following:

* * * * *

(9) Name and signature of the individual (or representative) and date signed, along with a statement that signing this addendum (or its updates) is only acknowledgement of receipt of the addendum (or its updates) and not the individual's (or representative's) agreement with the hospice's determinations. If the beneficiary (or representative) refuses to sign the addendum, the hospice must document on the addendum the reason the addendum was not signed and the addendum would become part of the patient's medical record. If a non-hospice provider or Medicare contractor requests the addendum, the non-hospice provider or Medicare contractor are not required to sign the addendum.

(10) Date the hospice furnished the addendum.

(d) Timeframes for the hospice election statement addendum. (1) If the addendum is requested within the first 5 days of a hospice election (that is, in the first 5 days of the hospice election date), the hospice must provide this information, in writing, to the individual (or representative), non-hospice provider, or Medicare contractor within 5 days from the date of the request.

(2) If the addendum is requested during the course of hospice care (that is, after the first 5 days of the hospice election date), the hospice must provide this information, in writing, within 3 days of the request to the requesting individual (or representative), non-hospice provider, or Medicare contractor.

(3) If there are any changes to the plan of care during the course of hospice care, the hospice must update the addendum and provide these updates, in writing, to the individual (or representative) in order to communicate these changes to the individual (or representative).

(4) If the individual dies, revokes, or is discharged within the required timeframe for furnishing the addendum (as outlined in paragraphs (d)(1) and (2) of this section, and before the hospice has furnished the addendum, the addendum would not be required to be furnished to the individual (or representative). The hospice must note the reason the addendum was not furnished to the patient and the addendum would become part of the patient's medical record if the hospice has completed it at the time of discharge, revocation, or death.

(5) If the beneficiary dies, revokes, or is discharged prior to signing the addendum (as outlined in paragraphs (d)(1) and (2) of this section), the addendum would not be required to be signed in order for the hospice to receive payment. The hospice must note (on the addendum itself) the reason the addendum was not signed and the addendum would become part of the patient's medical record.

* * * * *

4. Section 418.76 is amended by revising paragraphs (c)(1) and (h)(1)(iii) to read as follows:

§ 418.76

Condition of participation: Hospice aide and homemaker services.

* * * * *

(c) * * *

(1) The competency evaluation must address each of the subjects listed in paragraph (b)(3) of this section. Subject areas specified under paragraphs (b)(3)(i), (iii), (ix), (x), and (xi) of this section must be evaluated by observing an aide's performance of the task with a patient or pseudo-patient. The remaining subject areas may be evaluated through written examination, oral examination, or after observation of a hospice aide with a patient or a pseudo-patient during a simulation.

* * * * *

(h) * * *

(1) * * *

(iii) If an area of concern is verified by the hospice during the on-site visit, then the hospice must conduct, and the hospice aide must complete, a competency evaluation of the deficient skill and all related skill(s) in accordance with paragraph (c) of this section.

* * * * *

5. Section 418.306 is amended by revising paragraph (b)(2) to read as follows:

§ 418.306

Annual update of the payment rates and adjustment for area wage differences.

* * * * *

(b) * * *

(2) For fiscal years 2014 and through 2023, in accordance with section 1814(i)(5)(A)(i) of the Act, in the case of a Medicare-certified hospice that does not submit hospice quality data, as specified by the Secretary, the payment rates are equal to the rates for the previous fiscal year increased by the applicable hospice payment update percentage increase, minus 2 percentage points. Beginning with fiscal year 2024 and subsequent fiscal years, the reduction increases to 4 percentage points. Any reduction of the percentage change will apply only to the fiscal year involved and will not be taken into account in computing the payment amounts for a subsequent fiscal year.

* * * * *

6. Section 418.309 is amended by revising paragraphs (a)(1) and (2) to read as follows:

§ 418.309

Hospice aggregate cap.

* * * * *

(a) * * *

(1) For accounting years that end on or before September 30, 2016 and end on or after October 1, 2030, the cap amount is adjusted for inflation by using the percentage change in the medical care expenditure category of the Consumer Price Index (CPI) for urban consumers that is published by the Bureau of Labor Statistics. This adjustment is made using the change in the CPI from March 1984 to the fifth month of the cap year.

(2) For accounting years that end after September 30, 2016, and before October 1, 2030, the cap amount is the cap amount for the preceding accounting year updated by the percentage update to payment rates for hospice care for services furnished during the fiscal year beginning on the October 1 preceding the beginning of the accounting year as determined pursuant to section 1814(i)(1)(C) of the Act (including the application of any productivity or other adjustments to the hospice percentage update).

* * * * *

7. Section 418.312 is amended by revising paragraph (b) to read as follows:

§ 418.312

Data submission requirements under the hospice quality reporting program.

* * * * *

(b) Submission of Hospice Quality Reporting Program data. (1) Standardized set of admission and discharge items Hospices are required to complete and submit an admission Hospice Item Set (HIS) and a discharge HIS for each patient to capture patient-level data, regardless of payer or patient age. The HIS is a standardized set of items intended to capture patient-level data.

(2) Administrative data, such as Medicare claims data, used for hospice quality measures to capture services throughout the hospice stay, are required and fulfill the HQRP requirements for § 418.306(b).

(3) CMS may remove a quality measure from the Hospice QRP based on one or more of the following factors:

(i) Measure performance among hospices is so high and unvarying that meaningful distinctions in improvements in performance can no longer be made.

(ii) Performance or improvement on a measure does not result in better patient outcomes.

(iii) A measure does not align with current clinical guidelines or practice.

(iv) The availability of a more broadly applicable (across settings, populations, or conditions) measure for the particular topic.

(v) The availability of a measure that is more proximal in time to desired patient outcomes for the particular topic.

(vi) The availability of a measure that is more strongly associated with desired patient outcomes for the particular topic.

(vii) Collection or public reporting of a measure leads to negative unintended consequences other than patient harm.

(viii) The costs associated with a measure outweigh the benefit of its continued use in the program.

* * * * *

Dated: July 27, 2021.

Xavier Becerra,

Secretary,

Department of Health and Human Services.

Footnotes

1.  Hospices are also subject to additional Federal civil rights laws, including the Age Discrimination Act, Section 1557 of the Affordable Care Act, and conscience and religious freedom laws.

2.  Nelson, R., Should Medical Aid in Dying Be Part of Hospice Care? Medscape Nurses. February 26, 2020. https://www.medscape.com/​viewarticle/​925769#vp_​1.

3.  Medicare Department of Health and Human Services (DHHS) Provider Reimbursement Manual—Part 2, Provider Cost Reporting Forms and Instructions, Chapter 43, Form CMS-1984-14. April 13, 2018. https://www.cms.gov/​Regulations-and-Guidance/​Guidance/​Transmittals/​2018Downloads/​R3P243.pdf.

4.  Report to Congress, Promoting Greater Efficiency in Medicare. MedPAC. June 2007. http://www.medpac.gov/​docs/​default-source/​reports/​Jun07_​EntireReport.pdf.

5.  IPPS Regulations and Notices. https://www.cms.gov/​Medicare/​Medicare-Fee-for-Service-Payment/​AcuteInpatientPPS/​IPPS-Regulations-and-Notices.

6.  Hospice web page. https://www.cms.gov/​Medicare/​Medicare-Fee-for-Service-Payment/​Hospice/​index.

7.  Hospice web page: https://www.cms.gov/​Medicare/​Medicare-Fee-for-Service-Payment/​Hospice/​index.

8.  The National Consensus Project Guidelines expand on the eight domains of palliative care in the 3rd edition and include clinical and organizational strategies, screening and assessment elements, practice examples, tools and resources. The guidelines were developed by the National Consensus Project for Quality Palliative Care, comprising 16 national organizations with extensive expertise in and experience with palliative care and hospice, and were published by the National Coalition for Hospice and Palliative Care. Journal of Hospice & Palliative Nursing: December 2018—Volume 20—Issue 6—p 507.

9.  MedPAC. (2020). Chapter 12: Hospice Services. http://medpac.gov/​docs/​default-source/​reports/​mar20_​medpac_​ch12_​sec.pdf.

10.  MedPAC. (2020). Chapter 11: Hospice Services. http://www.medpac.gov/​docs/​default-source/​reports/​mar21_​medpac_​report_​to_​the_​congress_​sec.pdf?​sfvrsn=​0.

11.  2019: Vulnerabilities in Hospice Care (Office of the Inspector General).

12.  Report to Congress: Medicare Payment Policy (March 2019) MEDPAC.

13.  2019: Vulnerabilities in Hospice Care (Office of the Inspector General).

14.  We count discharges as any claim with a discharge status code other than “30” (which is defined as “Still Patient”).

15.  Another exclusion was made prior to reporting the numbers in Table B.1. We exclude all claims for a beneficiary if a beneficiary ever had two overlapping hospice days on separate claims. For FY 2019 this removes 5,212,319 hospice days that come from 218,420 claims and 33,009 beneficiaries.

16.  See Special coverage requirements, Title 42, Chapter IV, Subchapter B, Part 418, § 418.204. https://www.ecfr.gov/​cgi-bin/​text-idx?​rgn=​div5;​node=​42%3A3.0.1.1.5#se42.3.418_​1204.

17.  See Payment procedures for hospice care, Title 42, Chapter IV, Subchapter B, Part 418, § 418.302. https://www.ecfr.gov/​cgi-bin/​text-idx?​rgn=​div5;​node=​42%3A3.0.1.1.5#se42.3.418_​1302.

18.  Office of Inspector General. (2013). Medicare Hospice: Use of General lnpatient Care. https://oig.hhs.gov/​oei/​reports/​oei-02-10-00490.pdf.

19.  Office of Inspector General. (2019). Hospice Deficiencies Pose Risks to Medicare Beneficiaries. https://oig.hhs.gov/​oei/​reports/​oei-02-17-00020.pdf?​utm_​source=​summary-page&​utm_​medium=​web&​utm_​campaign=​OEI-02-17-00020-PDF.

20.  Hospices bill each day of CHC on a separate line item on the hospice claim.

21.  Teno J.M., Bowman, J., Plotzke, M., Gozalo, P.L., Christian, T., Miller, S.C., Williams, C., & Mor, V. (2015). Characteristics of hospice programs with problematic live discharges. Journal of Pain and Symptom Management, 50, 548-552. doi: 10.1016/j.jpainsymman.2015.05.001.

22.  MedPAC. (2020). Chapter 12: Hospice Services. http://medpac.gov/​docs/​default-source/​reports/​mar20_​medpac_​ch12_​sec.pdf.

23.  MedPAC. (2020). Chapter 12: Hospice Services. http://medpac.gov/​docs/​default-source/​reports/​mar20_​medpac_​ch12_​sec.pdf.

24.  That is, we are measuring the first seven days of hospice over a patient's lifetime and potentially across multiple hospice elections and fiscal years.

25.  MedPAC. (2020). Chapter 12: Hospice Services. http://medpac.gov/​docs/​default-source/​reports/​mar20_​medpac_​ch12_​sec.pdf.

26.  For example, see: Teno J.M., Bowman, J., Plotzke, M., Gozalo, P.L., Christian, T., Miller, S.C., Williams, C., & Mor, V. (2015). Characteristics of hospice programs with problematic live discharges. Journal of Pain and Symptom Management, 50, 548-552. doi: 10.1016/j.jpainsymman.2015.05.001.

27.  For example, if the hospice discharge occurred on a Sunday, the hospitalization had to occur on Sunday, Monday, or Tuesday to be counted.

28.  For example, see: Teno J.M., Bowman, J., Plotzke, M., Gozalo, P.L., Christian, T., Miller, S.C., Williams, C., & Mor, V. (2015). Characteristics of hospice programs with problematic live discharges. Journal of Pain and Symptom Management, 50, 548-552. doi: 10.1016/j.jpainsymman.2015.05.001.

29.  National Quality Forum. (2013). #2158 Payment-Standardized Medicare Spending Per Beneficiary (MSPB). https://www.qualityforum.org/​Projects/​c-d/​Cost_​and_​Resource_​Project/​2158.aspx.

30.  MedPAC. (2020). Chapter 12: Hospice Services. http://medpac.gov/​docs/​default-source/​reports/​mar20_​medpac_​ch12_​sec.pdf.

31.  Office of Inspector General. (2016). Hospices Inappropriately Billed Medicare Over $250 Million for General Inpatient Care. https://oig.hhs.gov/​oei/​reports/​oei-02-10-00491.pdf.

32.  See Condition of participation: Interdisciplinary group, care planning, and coordination of services, Title 42, Chapter IV, Subchapter B, Part 418, § 418.56 (https://www.ecfr.gov/​cgi-bin/​text-idx?​rgn=​div5;​node=​42%3A3.0.1.1.5#se42.3.418_​156) and Condition of participation: Hospice aide and homemaker services, Title 42, Chapter IV, Subchapter B, Part 418, § 418.76 (https://www.ecfr.gov/​cgi-bin/​text-idx?​rgn=​div5;​node=​42%3A3.0.1.1.5#se42.3.418_​176).

33.  See § 418.100 (https://www.ecfr.gov/​cgi-bin/​text-idx?​rgn=​div5;​node=​42%3A3.0.1.1.5#se42.3.418_​1100).

34.  de la Cruz, M., et al. (2015). Delirium, agitation, and symptom distress within the final seven days of life among cancer patients receiving hospice care. Palliative & Supportive Care, 13(2): 211-216. doi: 10.1017/S1478951513001144.

35.  Dellon, E.P., et al. (2010). Family caregiver perspectives on symptoms and treatments for patients dying from complications of cystic fibrosis. Journal of Pain & Symptom Management, 40(6): 829-837. doi: 10.1016/j.jpainsymman.2010.03.024.

36.  Kehl, K.A., et al. (2013). A systematic review of the prevalence of signs of impending death and symptoms in the last 2 weeks of life. American Journal of Hospice & Palliative Care, 30(6): 601-616. doi: 10.1177/1049909112468222.

37.  Hui D et al. (2014). Clinical Signs of Impending Death in Cancer Patients. The Oncologist. 19(6):681-687. doi:10.1634/theoncologist.2013-0457.

38.  Department of Health and Human Services, Office of Inspector General. (2013). Medicare hospice: Use of general inpatient care. Accessible via: https://oig.hhs.gov/​oei/​reports/​oei-02-10-00490.asp.

39.  Department of Health and Human Services, Office of Inspector General. (2016). Medicare hospice: Hospices Inappropriately Billed Medicare Over $250 Million for General Inpatient Care. Accessible via: https://oig.hhs.gov/​oei/​reports/​oei-02-10-00491.asp.

40.  Department of Health and Human Services, Office of Inspector General. (2019). Registered Nurses Did Not Always Visit Medicare Beneficiaries Homes at Least Once Every 14 Days to Assess the Quality of Care and Services Provided by Hospice Aides. Accessible via: https://oig.hhs.gov/​oas/​reports/​region9/​91803022.pdf.

41.  Medicare Payment Advisory Commission. March 2009 Report to the Congress: Medicare Payment Policy, “Chapter 6: Reforming Medicare's hospice benefit.” February 27, 2009. Accessible via: http://www.medpac.gov/​docs/​default-source/​reports/​Mar09_​Ch06.pdf?​sfvrsn=​0.

42.  Medicare Payment Advisory Commission. March 2011 Report to the Congress: Medicare Payment Policy, “Chapter 11: Hospice.” March 15, 2011. Accessible via: http://www.medpac.gov/​docs/​default-source/​reports/​Mar11_​Ch11.pdf?​sfvrsn=​0.

43.  Medicare Payment Advisory Commission. March 2020 Report to the Congress: Medicare Payment Policy Text, “Chapter 12: Hospice Services.” March 13, 2020. Accessible via: http://www.medpac.gov/​docs/​default-source/​reports/​mar20_​medpac_​ch12_​sec.pdf?​sfvrsn=​0.

44.  National Quality Forum. (2020). MAP 2020 Considerations for Implementing Measures Final Report—PAC LTC. http://www.qualityforum.org/​Publications/​2020/​02/​MAP_​2020_​Considerations_​for_​Implementing_​Measures_​Final_​Report_​-_​PAC_​LTC.aspx.

45.  https://www.cms.gov/​files/​document/​guidance-memo-exceptions-and-extensions-quality-reporting-and-value-based-purchasing-programs.pdf.

46.  Azar, A. M. (2020 March 15). Waiver or Modification of Requirements Under Section 1135 of the Social Security Act. Public Health Emergency. https://www.phe.gov/​emergency/​news/​healthactions/​section1135/​Pages/​covid19-13March20.aspx.

47.  https://www.phe.gov/​emergency/​news/​healthactions/​section1135/​Pages/​covid19-13March20.aspx.

48.  https://www.cms.gov/​files/​document/​guidance-memo-exceptions-and-extensions-quality-reporting-and-value-based-purchasing-programs.pdf.

49.  (2020, March 27). Exceptions and Extensions for Quality Reporting Requirements for Acute Care Hospitals, PPS-Exempt Cancer Hospitals, Inpatient Psychiatric Facilities, Skilled Nursing Facilities, Home Health Agencies, Hospices, Inpatient Rehabilitation Facilities, Long-Term Care Hospitals, Ambulatory Surgical Centers, Renal Dialysis Facilities, and MIPS Eligible Clinicians Affected by COVID-19. Centers for Medicare & Medicaid Services. . https://www.cms.gov/​files/​document/​guidance-memo-exceptions-and-extensions-quality-reporting-and-value-based-purchasing-programs.pdf.

50.  MedPAC. 2020. Report to the Congress: Medicare Payment Policy | March 2020. http://medpac.gov/​docs/​default-source/​reports/​mar20_​medpac_​ch12_​sec.pdf?​sfvrsn=​0. Accessed June 13, 2021.

51.  Azar, A. M. (2020 March 15). Waiver or Modification of Requirements Under Section 1135 of the Social Security Act. Public Health Emergency. https://www.phe.gov/​emergency/​news/​healthactions/​section1135/​Pages/​covid19-13March20.aspx.

52.  (2020, March 27). Exceptions and Extensions for Quality Reporting Requirements for Acute Care Hospitals, PPS-Exempt Cancer Hospitals, Inpatient Psychiatric Facilities, Skilled Nursing Facilities, Home Health Agencies, Hospices, Inpatient Rehabilitation Facilities, Long-Term Care Hospitals, Ambulatory Surgical Centers, Renal Dialysis Facilities, and MIPS Eligible Clinicians Affected by COVID-19. Centers for Medicare & Medicaid Services. .https://www.cms.gov/​files/​document/​guidance-memo-exceptions-and-extensions-quality-reporting-and-value-based-purchasing-programs.pdf.