April AMRPA Magazine

April 2019 • Vol. 22, No.4

April 2019 • Vol. 22, No. 4

The official publication of the American Medical Rehabilitation Providers Association (AMRPA) Richard Kathrins, PhD Chair, AMRPA Board of Directors, President & CEO, Bacharach Institute for Rehabilitation John Ferraro, MS AMRPA Executive Director Kate Beller, JD AMRPA Executive Vice President for Government Relations and Policy Development

Table of Contents Letter from the Chair

4

Legislative Update

6

Changes to CMS’ Extrapolation Guidance Both Help and Hurt Providers 10 Effective Case Management

13

MedPAC Commissioners Table an Episode-Based Unified PAC Payment System (For Now)

17

AMRPA Submits Comments on CMS’ 2020 Medicare Advantage and Part D Advance Notice Part II and Draft Call Letter

20

AMRPA Responds To RTI on the Development of a Unified Post-Acute Care Prospective Payment System

26

OIG Estimates $84 Million in Improper Payments for SNF Services Due to Violations of Three-Day Rule

40

Issue Brief Examines Medicare Advantage Market Concentration and Potential Competition

42

Brian McGowan AMRPA Design and Layout

CMS Office of the Actuary Releases 2018-2027 National Health Expenditure Projections

44

AMRPA Magazine, Volume 22, Number 4

NCHS Releases Long-term Care Provider and Services Trends in the United States

47

Reductions in Post-Acute Care Spending Under BPCI Were Driven

by Shift from Higher-to Lower- Cost Discharge Settings

49

Compliance with CMS Three-Hour Rule Alone Does Not Have a Substantive Impact on TBI Outcomes, Study Finds

50

Latest Research Findings

51

Carolyn Zollar, MA, JD AMRPA Senior Policy Counsel Mimi Zhang AMRPA Senior Policy and Research Analyst Patricia Sullivan AMRPA Senior Editor Lovelyn Robinson AMRPA Researcher and Editor

AMRPA Magazine is published monthly by the American Medical Rehabilitation Providers Association (AMRPA).

AMRPA is the national voluntary trade association representing inpatient rehabilitation hospitals and units, hospital outpatient departments and settings independent of the hospital, such as comprehensive outpatient rehabilitation facilities, rehabilitation agencies and skilled nursing facilities.

SUBSCRIPTION RATES: Member institutions receive the AMRPA magazine as part of their membership dues.

Send subscription requests to AMRPA, 529 14th St., NW, Washington, DC 20045 USA. Make checks payable to AMRPA. ADVERTISING RATES: Full page = $1,500; Half page = $1,000; Third page = $750. Ads may be B&W or full color.

Contact Brian McGowan, bmcgowan@kellencompany.com for additional specs and acceptable submission format.

Advertising Contact: Julia Scott, AMRPA, 529 14th St., NW, Washington, DC 20045 USA, Phone: +1-202-207-1110, Email: jscott@amrpa.org.

Statements of fact and opinion are the responsibility of the authors alone and do not imply an opinion on the part of the officers or the members of AMRPA. All content ©2019

by American Medical Rehabilitation Providers Association.

All rights reserved. Materials may not reproduced in any form without written permission. Design and layout services provided by Kellen Company.

POSTMASTER: Send address changes to Kellen Company, Attn: AMRPA Magazine Circulation 529 14th St., NW, Suite 750, Washington, DC 20045

AMRPA Magazine / April 2019

3

Letter from the Chair

Leadership, Dedication and Vision Our association is fortunate for many reasons. We have a highly engaged membership and a staff of professionals who keep us educated about current and future challenges.

Our Washington-based team has a history of developing policies that help ensure that those in need have access to high quality rehabilitative care.

Richard Kathrins, PhD, President & CEO, Bacharach Institute for Rehabilitation RKathrins@bacharach.org

Leading our association’s Washington team for the past 20 years has been Carolyn Zollar. She has been our association’s founding Executive Vice President for Governmental Relations and Policy Development. Many of you have met Ms. Zollar over the years and some have worked with her on various association committees.

Most of us have also been educated by Ms. Zollar on a vast array of legislative and regulatory issues. Representing AMRPA, she has traveled the country and traversed the halls of Congress and various federal regulatory agencies.

Before her tenure at AMRPA, Ms. Zollar worked for the U.S. Senate, the U.S. House of Representatives, as a consultant for the White House, and in private law practice.

She has been a member of multiple expert panels including the RAND Corporation’s Technical Expert Panel that developed the inpatient rehabilitation facility prospective payment system (IRF PPS). She also participated in a Secretary of Health and Human Services’ study on sub-acute care. We have all benefited from her vast experience.

I have had the privilege of working closely with Ms. Zollar. She is an amazing individual who is able to breakdown complex regulatory and legislative issues and translate them into a language that we can all understand. At times she has done this work in mere days or hours so that our members could receive complex and critical information in a timely manner.

Ms. Zollar is transitioning to a new role with our association, which means she is stepping down as our association’s Executive Vice President. She will continue to serve as counsel to our association. After so many years as the face of our association in Washington and across the country, it will be hard for us to make the transition.

We are glad that Ms. Zollar will continue to serve our members and advocate for our stakeholders.

A few weeks ago we welcomed Kate Beller as our association’s new Executive Vice President of Government Relations and Policy Development. Prior to coming to AMRPA, Ms. Beller worked for several law and lobbying firms including Powers Pyles Sutter & Verville; Hogan Lovells and Capitol Counsel. She received a baccalaureate from Duke University and law degree from Georgetown University. I am sure you all will join me in wishing Ms. Beller all the success in her new role. I know Ms. Beller looks forward to meeting all of you and learning about your issues and unique perspectives.

With only 500 words it is hard to express my appreciation for Carolyn Zollar’s leadership, dedication and vision for our field. It is comforting to me, and it should be to you, to know she is only a phone call away. Ms. Beller’s appointment and Ms. Zollar’s continued participation provides the foundation for a smooth transition of leadership for the association’s important work in the area of government relations and policy development.

4

AMRPA Magazine / April 2019

FOR ADULTS UP TO 700 LBS INTRODUCING THE WORLD'S MOST ADVANCED TREADMILL SYSTEM WITH DYNAMIC BODY-WEIGHT SUPPORT & FALL PROTECTION

INFOTAINMENT DISPLAY

VIRTUAL WALKS, COGNITIVE TASKS, GOAL SETTING, HEART RATE MONITORING, STEP PACING & OTHER METRICS

www.aretechllc.com I info@aretechllc.com I 800-710-0370 AMRPA Magazine

/ April 2019

5

Legislative Update

President Trump Releases Fiscal Year (FY) 2020 Budget Message and Priorities On March 13, 2019, the Office of Management and Budget (OMB) issued President Trump’s Fiscal Year (FY) 2020 Budget – “A Budget for a Better America: Promises Kept.

Taxpayers First.” More detailed budget documents for the Department of Health and Human Services (HHS) released on March 18 provide more details regarding specific Medicare proposals of interest to AMRPA.

Martha M. Kendrick, Esq., Partner, Akin Gump Strauss Hauer & Feld LLP

Highlights:

6

»»

President Trump released details on his FY 2020 Budget, which includes $2.7 trillion in federal spending cuts, with the Department of Health and Human Services facing a proposed 12 percent decrease in comparison to FY 2019.

»»

The Budget specifically highlighted inpatient rehabilitation services as a target for its Medicare prior authorization proposal.

»»

Health and Human Services Secretary Azar defended the administration’s proposed spending reductions to Medicare, Medicaid and other discretionary health programs when testifying before lawmakers who questioned whether the proposals would reduce health care access to care.

»»

The Medicare Payment Advisory Commission (MedPAC) recommends reducing the FY 2020 Medicare payment rate for inpatient rehabilitation facilities by 5 percent, among other post-acute care policies, in its annual March Report to Congress.

»»

Medicare-for-all legislation introduced, as policymakers and presidential contenders debate the policy and impact on the health care system.

AMRPA Magazine / April 2019

The president’s plan proposes to balance the budget in 15 years. The budget achieves this by making $2.7 trillion in spending cuts, including reducing domestic spending by 5 percent compared to FY 2019 levels. The budget cuts $457 billion from Medicare over 10 years by reducing “waste, fraud, abuse,” which includes policies that affect hospital and provider payments.

Of interest to AMRPA, the budget proposes to “address excessive payment for postacute care providers by establishing a unified payment system based on patients’ clinical needs rather than the site of care.” The budget tables show savings from the reform beginning in FY 2020. The HHS Budget-in-Brief explains that the proposal will provide lower annual Medicare payment updates to SNFs, HHAs, and IRFs beginning in FY 2020 through FY 2024. In FY 2025, HHS would implement a unified post-acute care payment system for SNFs, HHAs, IRFs and LTCHs. The payment rates would be budget neutral in FY 2025, risk adjusted, and established prospectively annually. According to HHS, the episode grouping and pricing would be based on the average cost for providing post-acute care services for a diagnosis, similar to the DRG methodology for inpatient hospitals. The proposal provides the secretary with authority to adjust payments based on quality of care, geographic differences in labor and other costs, as well as other factors as determined appropriate. The administration estimates the proposal will save $101.2 billion over 10 years.

The president also proposes to expand Medicare’s existing authority to conduct prior authorization for all Medicare fee-for-service (FFS) items and services at high risk of fraud and abuse. While the initial budget did not specifically identity these high-risk items and services, the subsequent and more detailed release calls out inpatient rehabilitation services as an area where the administration proposes to implement prior authorization. As discussed at the Spring Leadership Conference, AMRPA anticipated IRFs to be targeted as a result of the recent Office of Inspector General (OIG) report on IRF improper payments. The administration estimates the proposal will generate $6.3 billion in savings over 10 years.

The president also proposes to make several changes to the appeals system: changing the Medicare Appeal Council’s standard of review from a de novo to an appellatelevel standard of review; establishing an administrative fee applied post-adjudication for unfavorable Medicare appeals (other than beneficiary appeals); authorizing the

Office of Medicare Hearings and Appeals to issue decisions on the record without holding a hearing if no material facts are in dispute; increasing the minimum amount in controversy necessary for an adjudication of an appeal by an Administrative Law Judge (ALJ) to $1,630 in 2019 and updated annually thereafter; allowing the Office of Medicare Hearings and Appeals to use Medicare magistrates for the appealed claims below the amount in controversy threshold; limiting the right of appellants (other than beneficiaries) to appeal a redetermination of a claim that was denied because of lack of documentation; allowing remand of an appeal to the first level when new documents are submitted into the record during the second or later levels of review; and requiring an attestation that the appeal is being submitted in good faith that the provider is entitled to Medicare reimbursement. Interestingly, the administration estimates that the combined policies will have no budget impact.

Additionally, the administration also proposes an old-favorite; namely to limit Medicare bad debt payments by reducing Medicare reimbursement of bad debt from 65 percent to 25 percent over three years beginning in FY 2020. The administration estimates the proposal will save $38.5 billion over 10 years. The Budget expresses continued support for Affordable Care Act (ACA) repeal and replace. The president proposes to use existing authorities through the Center for Medicare and Medicaid Innovation (CMMI) to establish bundled payment arrangements for certain “high value” devices. The HHS Budget explains, “these devices include technologies that could significantly reduce time and costs in a post-acute care setting but providers otherwise consider them impractical solely in the scope of Medicare payments to facilities.”

The president outlines additional administration budget priorities including: funding for a border wall; ending sanctuary cities; investing in defense capabilities such as space, artificial intelligence and hypersonics; continuing to fund opioid addiction prevention, treatment, research and recovery; holding colleges and universities accountable for student outcomes; modernizing federal information technology; and increasing funding by 10 percent compared to FY 2019 levels to invest $80 billion in services for veterans including rehabilitation services, education and employment assistance, and suicide prevention.

Presidential budgets tend be considered more aspirational than likely to pass, particularly with a Democrat-controlled House and a challenging 2020 election map for Senate Republicans.

However, it does highlight President Trump’s funding priorities, which could set up another Budget showdown like we saw in 2018, and it offers a road map for Congress to initiate cuts in many areas.

On a bipartisan, bicameral basis, congressional leaders are working to reach agreement on a two-year budget deal that would increase discretionary spending caps. Any increases are expected to be modest and in line with inflation at approximately two percent. Absent a budget deal, discretionary spending caps will go into effect at the start of FY 2020 (October 1, 2019), resulting in $126 billion in cuts. President Trump’s budget proposal rejects the idea of increasing the spending caps.

The administration’s position limits the possibility of a budget deal and suggests potential trouble in advancing FY 2020 appropriations bills.

MedPAC Focuses on Post-Acute Care On March 7 and 8, the Medicare Payment Advisory Commission (MedPAC) held two sessions on post-acute care issues. During the first session, “Evaluating an episode-based payment system for post-acute care,” staff discussed the commission’s ongoing work on a unified post-acute care prospective payment system (PAC PPS) and compared a stay-based PPS design with an episode-based PPS design. Staff stated that while an episodebased design is less likely to lead to contribute to unnecessary volume, the design could result in unintended adverse consequences on areas such as patient selection and withholding of care.

The second session, “Mandated report: Changes in post-acute and hospice care following the implementation of the longterm care hospital dual payment-rate structure,” provided an update on the commission’s mandated report on the changes made under the Pathway for SGR Reform Act of 2013, which established a dual payment-rate structure for long-term care hospitals (LTCHs). Among other things, staff found that spending on hospice services increased from about $15 billion in 2012 to almost $18 billion in 2018. Additionally, the number of hospice providers increased. According to the presentation, LTCH margins have decreased, but cases that meet the patientlevel criteria have remained profitable. The final report will be published in June 2019.

MedPAC Releases Annual March Report to Congress On March 15, the Medicare Payment Advisory Commission (MedPAC) released its March 2019 Report to the Congress. The report includes MedPAC’s analyses of payment adequacy in fee-for-service (FFS) Medicare and reviews the status of Medicare Advantage (MA) and Medicare Part D. The commission also recommends that Congress replace the four current hospital quality payment programs with a single program, the hospital value incentive program (HVIP).

In Chapter 7 on “Cross-cutting issues in post-acute care,” the commission examines the post-acute care (PAC) space and concludes that payments are “unnecessarily high” in the Skilled Nursing Facility (SNF), Home Health Agency (HHA), and Inpatient Rehabilitation Hospital (IRF) settings. The chapter calls for implementing “substantial changes” to the HH and SNF payment systems and for continuing to improve setting-specific payment systems until implementation of a unified PAC prospective payment system.

In Chapter 8 on Skilled Nursing Facilities, MedPAC includes two recommendations: 1) The secretary should proceed to revise the skilled nursing facility prospective payment system in fiscal year 2020 and should annually recalibrate the relative weights of the case-mix groups to maintain alignment of payments and costs; 2) The Congress should eliminate the fiscal year 2020 update to the Medicare base payment rates for skilled nursing facilities.

AMRPA Magazine / April 2019

7

In Chapter 9 on Home Health Agencies, MedPAC includes one recommendation: For 2020, the Congress should reduce the calendar year 2019 Medicare base payment rate for home health agencies by 5 percent.

In Chapter 10 on Inpatient Rehabilitation Facilities, MedPAC includes one recommendation: For 2020, the Congress should reduce the fiscal year 2019 Medicare base payment rate for inpatient rehabilitation facilities by 5 percent.

In Chapter 11 on Long-Term Care Hospitals, MedPAC includes one recommendation: For 2020, the Secretary should increase the fiscal year 2019 Medicare base payment rates for long-term care hospitals by 2 percent.

Medicare-for-All Legislation On February 27, Democrats unveiled the Medicare for All Act of 2019 (H.R. 1384). Introduced by Rep. Pramila Jayapal (D-WA07), co-chair of the Congressional Progressive Caucus, the bill would move the health care system to a single-payer model that would provide universal coverage within two years. More moderate Democrats, meanwhile, are backing a proposal (S. 470) from Sen. Debbie Stabenow (D-MI) that would allow Americans age 50 to 64 to buy into Medicare. The House Rules and Budget Committees are expected to hold hearings on Medicare for All proposals this month, but other panels – such as the Energy and Commerce Committee and the Ways and Means Committee – have yet to commit to hold hearings on the issue.

On February 14, Reps. Kathy Castor (D-FL), Gus Bilirakis (RFL), Anna Eshoo (D-CA), and Jaime Herrera Beutler (R-WA) reintroduced the Advancing Care for Exceptional (ACE) Kids Act (H.R. 1226). The legislation was passed in the House last year as part of a Medicaid package but was not taken up by the Senate before the end of the 115th Congress.

Congress Aims to Address Health Care Costs and Transparency for Patients; Surprise Billing Legislation is First Step Momentum continues to build for Congressional action to protect patients against “surprise billing,” where a patient unexpectedly receives a medical bill from a provider that was outof-network after seeking care at an in-network facility.

In the Senate, a bipartisan group of senators is working together to introduced legislation by the end of March. The group includes Sen. Bill Cassidy (R-LA) and Sen. Maggie Hassan (D-NH), who had authored separate proposals in the 115th Congress.

Sens. Michael Bennet (D-CO), Tom Carper (D-DE), Todd Young (R-IN), and Chuck Grassley (R-IA), Chairman of the Senate Finance Committee, have joined the effort. The Senators solicited and received more than two dozen comments from stakeholders

8

AMRPA Magazine / April 2019

including hospitals, physicians, and insurers. Senators on the Health, Education, Labor, and Pensions (HELP) Committee are pressing for a hearing on the issue. HELP Committee staff have been engaging with the administration on the surprising billing issue and legislation may move as part of a larger package to address health care costs.

Rep. Lloyd Doggett (D-TX), chair of the Ways and Means Health Subcommittee, reintroduced his legislation, the “End Surprise Billing Act of 2019” (H.R. 861). The bill would require hospitals to provide advance written notice and seek consent for outof-network items and services to insured individuals; prohibit hospitals from charging above in-network rates if they fail to give advance notice and obtain consent at least 24 hours prior to providing treatment; and limit out-of-network charges for emergency services provided by hospitals.

President Trump and his administration support efforts to hold patients harmless and find a solution to surprise billing.

Administration staff have indicated that independently of Congress’s efforts to address surprise billing, HHS is also working to identify solutions. The administration believes that it has different tools than Congress to ensure patients are not subject to surprise billing. The administration expects to release more details about their plan or policy ideas late this summer or in the fall. Providers are coalescing around adoption of an arbitration model; however, insurers and other stakeholders oppose arbitration, meaning Congress will likely need to act in order to hold patients harmless. *** Although the president’s Budget Blueprint is often considered “dead on arrival,” it does contain policies and “pay-fors” that lawmakers could use for future legislation. As noted above, Congress will have to make several difficult funding decisions in the coming months and weeks. With a ballooning deficit, any increase in spending will likely be balanced with funding reductions. As expected, the post-acute field is a target for budgetary cuts. The recently released MedPAC report, which calls for inpatient rehabilitation facilities to be reduced by 5 percent, is also concerning.

Many of you just visited with your Senators and Member of Congress at AMRPA’s Leadership Conference and Fly-In in March, so we ask you to please remember to cultivate any new relationships and express the rehabilitation field’s concern with the President’s proposals and the MedPAC report. Those of you who could not be with us recently on Capitol Hill, please still engage now! We would be happy to support you and provide you with any additional information you may need!

2019 FALL

2019

Fall Educational Conference Coronado Bay Resort & Expo Loews San Diego, CA | October 14-16, 2019

EDUCATIONAL CONFERENCE & EXPO Loews Coronado Bay Resort San Diego, CA October 14-16, 2019

Abstract Deadline: May 6, 2019

What would you like to discuss this year?

Submit your abstracts for AMRPA’s 17th Annual Educational Conference & Expo!

AMRPA encourages you to share your knowledge, case studies, and experience in these 5 categories:  Business Operations & Leadership Management  Clinical Care Delivery: A Team Approach  Regulatory, Legislative, and Accreditation Matters  Marketing and Relationship Management  Other

Submit your ideas today!

For more information about how to submit your abstract, visit the AMRPA website:

amrpa.org/abstract-submission Questions about submitting your abstract?

Contact Julia Scott, AMRPA Member Services Coordinator, at jscott@amrpa.org for assistance.

#AMRPA

AMRPA Magazine / April 2019

9

Changes to CMS’ Extrapolation Guidance Both Help and Hurt Providers

Peter W. Thomas, Principal, The Powers Law Firm

Christina A. Hughes Counsel, The Powers Law Firm

In a zealous effort to recover overpayments, the Centers for Medicare and Medicaid Services (CMS) continues to utilize multiple types of contractors and various programs and procedures to identify improperly paid claims. Extrapolation is one of the tools that CMS makes available to its contractors. Over approximately the past 10 years, various contractors have tested the effectiveness of using extrapolation. As the use of this process has increased, providers have periodically challenged this sanction, which typically results in eye-popping demands for repayment to the government. The results of these challenges have been mixed.

Past success in challenging the use of extrapolation has centered on one of two avenues: challenging whether the contractors’ decision to use extrapolation is appropriate, or challenging the methodology used by the contractor to extrapolate claims. In the past, providers have run into difficulty under both of these approaches, often based on the vague guidelines and broad discretion provided to the contractors under Medicare guidance.

Effective as of January 2, 2019, CMS has made several important changes to the way its contractors may carry out extrapolation and overpayment estimation against providers.

Some of these changes are quite positive for providers and some will make it even more difficult for providers to prevail through an extrapolation challenge. Most importantly, the revisions to the Program Integrity Manual (PIM) provide much more detail about how Medicare contractors should perform statistical sampling than previously available.

The changes apply to all Medicare contractors eligible to use extrapolation: Medicare Administrative Contractors (MACs), Recovery Audit Contractors (RACs), Supplemental Medical Review Contractors (SMRCs), and Unified Program Integrity Contractors (UPICs).

The Former Guidance As noted, the PIM previously provided little guidance on what standards to follow when extrapolating overpayment findings. The guidance was limited to generally identifying the broad steps contractors should take, without elaboration. The lack of specificity sometimes created opportunities for providers to successfully challenge extrapolations, but it also left ample ambiguity for contractors to prevail. Providers’ success in the past was in part due to the lack of clarity from CMS as to what constituted an appropriate extrapolation methodology.

The New Clarifications On September 28, 2018, CMS issued Transmittal 828, making revisions to the PIM. Under the new guidance, as under the old guidance, a contractor must determine that a sustained or high level of payment error exists or that documented educational

10 AMRPA Magazine / April 2019

intervention has failed in order to use extrapolation. However, the guidance now provides more specific requirements. For instance, it states that extrapolation shall be used when a sustained or high level of payment error exists; extrapolation may be used after documented educational intervention (such as in the Targeted Probe and Educate (“TPE”) program). Thus, extrapolation should not be used if there is not a sustained or high level of payment error or evidence that documented educational intervention has failed.

The PIM now specifies that a “high error rate” exists when the sample review error rate is greater than or equal to 50 percent.

This is a significant difference from error rates Medicare auditors have previously used to justify a high error rate (sometimes as low as ten percent), and may provide some significant relief from the punitive effects of extrapolation. (This is by far the best news for providers contained in this new guidance.)

However, in determining whether a sustained level of payment error exists, the PIM states that the contractor may look to the provider’s history of noncompliance for the same or similar billing issues or a historical pattern of noncompliant billing practice.

This sets up providers who have gone through multiple rounds of TPE audits for a high risk of extrapolation. This is particularly disingenuous considering the lengthy, years-long backlog at the Office of Medicare Hearings and Appeals, which prevents providers from using favorable and timely ALJ decisions to rebut the appropriateness of similar, subsequent denials from Medicare contractors.

In addition to finally providing a definition of what constitutes a “high error rate,” the revised PIM places a finer point on the identification of the claims “universe.” Specifically, it defines the sampling frame as the listing of sample units, derived from the universe, from which the sample is selected and acknowledges that the universe may include items that are not used in the sample frame. This can happen for a number of reasons, including, but not limited to, prior inclusion of the claims/claim lines in a prior review, nonpayment of some of the claims/claim lines, or elimination of some sample units due to the definitions used in the sample frame.

The revised PIM now requires the contractor to include a list of all sample units, all universe elements incorporated in the sample units, and the elements of the universe such that the sample units may be reassembled during the replication process.

While somewhat complicated, this additional information will help providers assess the accuracy of the audit results, and also provides an additional means to attack the validity of the extrapolation.

The revised guidance now imposes limits on certain thresholds of recovery. Under the revised PIM, a Medicare contractor must now seek additional approval from CMS to issue a finding letter when the extrapolated overpayment exceeds $500,000 or is an amount that is greater than 25% of the provider’s Medicare revenue for the previous 12 months. In seeking approval, the contractor must provide CMS with a summary of its investigation, prior history, medical review results, and the extrapolated overpayment amount for CMS’ review. The implication is that CMS will

consider these factors (particularly the provider’s past history) and potentially disallow the use of extrapolation by the contractor before an extrapolation is even imposed on the provider.

Finally, CMS formally acknowledges in the revised guidance that, if claims are reversed from the initial claim determination during the administrative appeals process, Medicare contractors are required to adjust the overpayment estimate. The guidance goes further to specify that the revision must be done with the input of the statistical expert who created the initial methodology (or another statistical expert if the creator of the initial methodology is no longer available.) Given the high rate of reversal of claim denials in inpatient rehabilitation hospitals’ (IRFs’) favor, confirmation that administrative appeal results will impact the final extrapolation amount is a very positive development.

Caveats and Conclusions The revised PIM aims for greater transparency of the extrapolation process and clarifies many existing ambiguities.

However, the PIM continues to state that “[f]ailure by the contractor to follow one or more of the requirements contained herein may result in review by CMS of their performance, but should not be construed as necessarily affecting the validity of the statistical sampling and/or the projection of the overpayment.” PIM, CMS Pub. 100-8, ch. 8, § 8.4.1.1.

Consistent with past experience, CMS also states that a provider appealing an extrapolation “must demonstrate actual error in the methodology that affects the overpayment amount.”

Id. (emphasis added). Furthermore, the new guidance explicitly states that challenges to a sampling methodology based on whether the sample was drawn in a statistically valid manner or whether the sample size is too small “are without merit” when not accompanied by other challenges to the sampling methodology. Id. at § 8.4.4.3. This suggests a steeper hill to climb for providers to invalidate an extrapolation.

Accordingly, providers should remember that, while the guidance does provide some protections against inappropriate extrapolation, a Medicare contractor’s failure to comply strictly with these guidelines may not always yield a successful challenge to the extrapolation. Even when the guidance seems most definitive, CMS continues to hedge its bets, indicating multiple times throughout the revised guidance that failure by a contractor to adhere to the standards set forth is not necessarily grounds for invalidating an overpayment estimate.

In addition, the guidance now makes extrapolation mandatory under certain circumstances and explicitly precludes many challenges to overpayment estimates. But there is real value in the new definition of “high error rate,” which is far more generous than past thresholds used by contractors, as well as the requirement that CMS approve recovery of estimates above certain thresholds. Finally, clarifications in CMS’ preferred sampling methodology will be useful to providers in analyzing extrapolations imposed against them and may assist providers in challenging extrapolations in the future.

AMRPA Magazine / April 2019 11

Find new and exciting opportunities in AMRPA’s Career Center.

Our newly updated Career Center provides services and resources to help the medical rehabilitation field meet their professional goals.

All rehabilitation professionals may browse and apply for jobs at no cost, and AMRPA members will receive discounted rates for posting positions.

Visit our Career Center Here:

careercenter.amrpa.org

Begin by creating your free Career Cast account, which can be found on the top right hand corner of the website.

From there, you can upload and manage multiple resumes, browse through hundreds of job postings, and even research salaries of the positions in question! AMRPA members and affiliates may also purchase Posting Packages at a standard, premium, or platinum level.

AMRPA members will receive a 50% discount on all job postings.

For questions about our Career Center, please contact Anna Kruskop, AMRPA Member Services Associate, at akruskop@amrpa.org or 202-207-1120.

12 AMRPA Magazine / April 2019

Effective Case Management

Lisa Werner, MBA, MS, SLP Director of Consulting Services, Fleming-AOD, Inc.

// Over the years, Medicare Advantage programs and commercial insurance companies have made it increasingly difficult to obtain authorization for inpatient rehabilitation stays.

Every day at work I see miracles happening all around me. I am sure you can relate. Most people grow accustomed to seeing it, so you may not be as impressed by the subtle ways that health care providers use their skills to make a significant impact on the lives of the patients they treat or on their families. This past week, I witnessed such a moment and it got my wheels turning, thinking about how case managers advocate for appropriate lengths of stay for our patients. It occurs to me that some providers have case managers who seek authorization for continued stay who may not have an inpatient rehabilitation background to draw from that will help them put together a strong argument for getting more rehab days. Let’s talk about case management, particularly as it relates to utilization review and securing the ideal length of stay.

Over the years, Medicare Advantage programs and commercial insurance companies have made it increasingly difficult to obtain authorization for inpatient rehabilitation stays.

Providers have defended the need for and the value of an inpatient rehabilitation stay. To get a patient admitted, it takes a well-written pre-admission screening. It should include a summary of the patient’s medical management while in acute care, current status and a strong justification for continued medical management needs. It should also contain a functional assessment that describes the patient’s prior level of function, current function in terms that reflect the burden of care, and what you anticipate the patient’s function will be as a result of the rehab intervention.

For the sake of this article, we must assume that you were successful in obtaining authorization for admission. I know that it is also difficult to extend authorizations for the length of time that will allow the patient to achieve the maximum benefit of the inpatient rehabilitation stay. We need to ensure that we maintain, throughout the length of stay, the posture that we adopt in order to obtain prior authorization. Ensure that the case manager or utilization review coordinator tasked with securing more days is able to speak to the insurance company liaison in a manner that tells the patients’ whole story.

Do not assume that the therapy notes tell the story about burden of care. It is not necessary to tell the story in FIM scores, which has been a concern raised by some providers with the shift away from FIM and to section GG coding in October 2019. What is necessary is to tell the whole story about the care rendered.

We need to discuss cases with the therapists. Find out if the patient who walked 100 feet really walked 100 feet. Was that with or without a rest break? Was that with or without a wheelchair follow? Identify the quality of the gait pattern. Indicate where cueing or support was needed. Note the types and quality of intervention that were provided.

AMRPA Magazine / April 2019 13

Find out if the patient was moderate assistance with dressing. Did they require a device to assist? Did that include all of the items that we include in dressing such as socks and shoes? Find out if the patient who is likely to make a slower return to his prior level of function is progressing. Note where you found evidence of documented progress in the therapy notes. Sometimes changes come slowly at first and then pick up steam. Utilizing a functional assessment such as FIM or section GG may not reflect changes, but notes written by the therapists do. Use those notes to your advantage as you prepare to submit documentation for extending the patient’s stay.

Take advantage of peer-to-peer opportunities. I have heard that the insurance plans do not make this convenient, but it is our right and, for the good of the patient, it is something that we should strive to complete. Make sure that the rehabilitation physicians are versed in how to present a case. Be prepared to discuss how the patient is benefiting from the program, what the anticipated improvement will be in the next week (or more), and include the intangible benefits that may not be clearly spelled out in the documentation.

The rehabilitation physician knows why the setting is the most appropriate for the patient and can articulate the medical conditions that require ongoing care and management, functional status and potential, and risks and potential complications that need to be managed in the rehabilitation setting. Some insurance plans will stipulate that only the acute care hospital or discharging physician can request the peer-to-peer review. Be prepared to engage the discharge team to advocate for the patient’s rehabilitation admission.

14 AMRPA Magazine / April 2019

Encourage families to advocate for the patient. Have the case manager or utilization review manager educate the family on how to talk to the insurance company. The family is the most powerful voice for the patient. They will not be inclined to reach out to the insurance company and will require your help to know that it is both within their right and effective to do so.

If you are reading this and you are the rehabilitation administrator coordinating with a case manager who never worked in the capacity of rehabilitation nurse or therapist, take a few minutes and think about the things that you have learned from years of experience on the rehabilitation floor. Jot these down. Share them with the case manager. Look for opportunities to teach case managers about what makes rehabilitation care different. Teach the case manager how to quantify progress in a meaningful way and how gains, even small ones, impact caregiver burden and quality of life. The best case managers will seek every opportunity to get the best outcomes for the patients, including arguing a case for more days until they have exhausted every avenue. Your patients deserve a case manager like that, so use the resources at your disposal to create one.

Did You Know?

All of our webinars are available online!

Missed out on a recent AMRPA webinar? Not to fear!

All AMRPA webinars are available On Demand for purchase almost immediately following its recording.

W Eour B selection I N A h Rere: Browse amrpa.org/Education/Webinars/OnDemand-Webinars *AMRPA members receive a discount on all webinar recordings.

AMRPA Magazine / April 2019 15

2019 Fall Educational Conference & Expo Loews Coronado Bay Resort San Diego, CA October 14-16, 2019

2020 Fall Educational Conference & Expo Renaissance Dallas Hotel Dallas, TX October 11-14, 2020

2021 Fall Educational Conference & Expo JW Marriott Turnberry Isle Miami, FL October 24-27, 2021

2022 Fall Educational Conference & Expo St. Louis Union Station Hotel St. Louis, MO October 9-12, 2022

16 AMRPA Magazine / April 2019

MedPAC Commissioners Table an Episode-Based Unified PAC Payment System (For Now) At the March public meeting of the Medicare Payment Advisory Commission (MedPAC), the congressional advisory body discussed the viability of an episode-based model for a unified post-acute care (PAC) prospective payment system (PPS). This approach has been previously discussed by the commission as a “PAC-only bundle.”

Mimi Zhang, AMRPA Senior Policy and Research Analyst

Highlights: »»

While an episode-based PAC PPS model could dampen volumedriven utilization, MedPAC is challenged by developing accurate episodic payment rates

»»

Commissioners express support for continuing with a stay-based unified PAC PPS model at this time

In 2016, the commission recommended a PAC PPS prototype that used stays – either an institutional PAC stay or a home health episode – as the unit of service. Institutional PAC includes inpatient rehabilitation hospitals and units, long-term acute care hospitals and skilled nursing facilities. In 2017, commission sought to improve the stay-based model to further discourage fee-for-service incentives and volume-based utilization. Hence over the past year, commission staff have been developing an episode-based PPS model and they presented their findings for the commissioners’ review at this meeting.

Background According to MedPAC, a stay-based unified PAC PPS does little to dampen fee-forservice incentives for volume or encourage PAC providers to offer a continuum of care that would reduce beneficiary care transitions. In contrast, an episode-based PAC PPS that spans multiple “stays” would encourage providers to deliver an efficient mix of care and, if designed appropriately, also encourage institutional providers to offer a continuum of care. This would benefit beneficiaries by reducing the number of transitions that a patient may experience over their course of care. Currently, a third of Medicare PAC stays are part of a sequence of care, where patients transition from one PAC setting to another or extend their care, such as in back-to-back home health stays.

In an episode-based design, a single payment would be made for the combination of stays that make up the PAC episodes. The table below presents other criteria MedPAC considered evaluating in a stay-based versus episode-based design.

Table 1. Comparison of stay-based and episode-based PAC PPS model Aspect

Stay-based design

Episode-based design

Payment accuracy

Accurate for most patient groups

Accurate for most patient groups; less accurate for short or long episodes

Patient selection

Less likely

More likely

Stinting on care

Less likely

More likely

Unnecessary volume

More likely

Less likely

Care coordination

More handoffs

Fewer handoffs

Implementation and administration

Easier

More complicated

Source: MedPAC

AMRPA Magazine / April 2019 17

Model Design In developing an episode-based model, MedPAC took a similar approach to its stay-based model to arrive at a model that establishes payments based on patient and stay characteristics.

Design features common to both models include:

and long PAC episodes became more wildly out of alignment with their costs. According to MedPAC, the variance is so great that even a single outlier payment policy would not be able to adequately correct under- and over-payments.

A home health adjuster (given this setting's much lower costs compared to institutional PAC); Separate simulations to establish payments for routine and therapy care and non-therapy ancillary services, such as drugs, because the benefits that Medicare covers differ slightly across PAC settings; Not including services such as acute care hospitalization, outpatient therapy, and physician services; Keeping payments budget-neutral to the level of aggregate PAC spending in 2017.

One change in the episode-based model was how MedPAC estimated routine costs. This time, MedPAC used contemporaneous cost report data and claims information, whereas it previously used survey-based routine cost data collected in the Post-Acute Care Payment Reform Demonstration (PAC PRD).

MedPAC updated its stay-based model using 2017 data to reflect more current costs and utilization. It then constructed episodes from individual PAC stays that are within seven days of each other. MedPAC focused on solo and pairs of stays to start because these stays make up a little over two-thirds of all PAC stays. As for patient characteristics, the episode model includes a patient's age, disability status, the primary reason to treat, comorbidities and risk score, medical complexity, cognitive status, and other disabilities such as difficulty swallowing. These factors were derived from claims and other administrative data.

Results Currently, aggregate PAC payments are on average 12 percent higher than costs for the 40 patient groups that MedPAC examined. MedPAC found that an episode-based model could align payments more closely with costs for most of the patient groups it examined. Payments would also be redistributed across clinical conditions in the episode-based model.

Specifically, payments would move from care episodes that include rehabilitation therapy “not predicted by patients' clinical characteristics” to episodes for medical complex care needs.

However, these positive results are observed only in the broad aggregate. MedPAC also found that its episode-based model would exacerbate existing payment-to-cost differentials, depending on the length of a stay. MedPAC attributes this to the fact that episode-based payments are based on the average costs across all episodes – short, medium and long. This resulted in payments for short episodes being more than double their costs (indicated by the green circles below) and, conversely, payments being about three-quarters of their costs for long episodes (see yellow circles). In other words, payments for short

18 AMRPA Magazine / April 2019

The analysis also examined home health stays and institutional PAC stays separately, recognizing that payment-to-cost differences reflect, to some extent, the clinical differences of patients in the “short” group versus those in the “long” group. Nonetheless, even after this case mix adjustment was incorporated, the shorter episodes still remained disproportionally profitable and both home health and institutional PAC providers would incur a loss for longer episodes.

MedPAC expressed reservations about proceeding with an episode-based model due to the cost-to-payment misalignment, stating that providers have been behaviorally responsive to payment system changes in the past. The staff concluded that an episode-based model could create unintended consequences, such as providers inappropriately cutting lengths of stay or prematurely discharging patients, avoiding patients likely to need extended care, stinting on care during the episode, and transfer or continue to treat patients based solely on financial considerations.

Next Steps The episodic model MedPAC designed would result in substantial overpayment for short stays and underpayments for long ones. As such, while commissioners viewed the episodebased model as one they would like to revisit in the future, they agreed that it was not feasible to proceed with at this time, primarily due to the administrative complexity of determining appropriate payments. Several commissioners recommended the staff develop a robust quality measure framework that could be coupled with a PAC value-based purchasing (VBP) program help protect against the perverse incentives in any PAC PPS model, be it a stay- or episode-based.

AMRPA anticipates MedPAC will address the episode-based PAC PPS model in its June Report to Congress.

Check out

Engage AMRPA!

A new online community forum enabling a private social network for AMRPA members to form communities, collaborate, manage industry profiles, and connect with peers and colleagues.

On Engage AMRPA, members are also able to:  View AMRPA Upcoming Events  Listen to AMRPA Member’s Only Calls  Read the latest AMRPA Comment Letters  And much more!

engage.amrpa.org

Come see what people are talking about!

AMRPA Magazine / April 2019 19

AMRPA Submits Comments on CMS’ 2020 Medicare Advantage and Part D Advance Notice Part II and Draft Call Letter Editor’s Note: On March 1, 2019, the American Medical Rehabilitation Providers Association (AMRPA) submitted comments to the Centers for Medicare and Medicaid Services’ (CMS) regarding proposed updates to the Medicare Advantage (MA) program affecting patient access to medical rehabilitation services. The complete letter is provided below and at www.amrpa.org.

March 1, 2019 The Honorable Seema Verma Administrator Centers for Medicare and Medicaid Services U.S. Department of Health and Human Services 7500 Security Boulevard Baltimore, MD 21244

Re: CMS-2018-0154 (2020 Medicare Advantage and Part D Advance Notice Part II and Draft Call Letter).

Dear Administrator Verma: On behalf of the American Medical Rehabilitation Providers Association (AMRPA), I am submitting this letter regarding the proposed updates to the Medicare Advantage (MA) and Part D programs through the 2020 Advance Notice and Draft Call Letter released by the Centers for Medicare and Medicaid Services (CMS).1 Our comments focus on concerns regarding MA enrollee access to medical rehabilitation services, particularly access to inpatient rehabilitation hospitals and units.

AMRPA members provide rehabilitation services across the spectrum of health care settings including inpatient rehabilitation hospitals and units (referred to by Medicare as inpatient rehabilitation facilities, or IRFs), hospital outpatient departments, and settings independent of the hospital, such as comprehensive outpatient rehabilitation facilities (CORFs), rehabilitation agencies, skilled nursing facilities (SNFs) and long term care hospitals (LTCHs). AMRPA members help patients maximize their health, functional skills, and independence, so they can participate in society by returning to home, work, or an active retirement. As part of furnishing care in the IRF setting, AMRPA members provide intensive, comprehensive, hospital-based, rehabilitation therapy programs coupled with complex medical management of the patient. Some of the therapy services provided include physical and occupational therapy, speech language pathology, and prosthetic/orthotic services, to name a few.

Summary of Comments AMRPA appreciates CMS’ efforts to administer the MA program in a transparent manner that affords stakeholders the opportunity to provide comments. With steady growth in managed care, the MA program now covers one-third of all Medicare beneficiaries.2 It continues to be important that the MA program be administered in a way that protects beneficiaries’ legal rights and guarantees their access to medically necessary health care.

MA plan enrollees are routinely denied access to post-acute care benefits to which they are entitled. CMS’ urgent attention to this matter is therefore required. AMRPA Centers for Medicare and Medicaid Services, 2020 Medicare Advantage and Part D Advance Notice Part II and Draft Call Letter (Jan. 2019) [hereinafter “Draft Call Letter”].

1

20 AMRPA Magazine / April 2019

recommends and requests CMS action to address barriers to access as outlined below: Require disclosure of Medicare post-acute care coverage rules so beneficiaries receive adequate information about potential options upon admission to, and especially at discharge from, a short-term acute care hospital; Restrict the use of proprietary decision tools, which are inconsistent with Medicare coverage policy and clinical decision-making; Ensure MA plan enrollees are reasonably able to appeal improper coverage denials; Require reporting of utilization, denial, and overturn rates for enrollee utilization of post-acute care; and Audit MA plan performance to ensure comparable access to inpatient hospital rehabilitation across MA and fee-forservice Medicare beneficiaries.

Background: Access Challenges As AMRPA has commented in response to prior call letters, significant administrative hurdles are embedded in the MA program that prevent patients from accessing the post-acute care they need. We have detailed information regarding the challenges MA enrollees routinely face in trying to access medical rehabilitation in AMRPA’s prior letters. We will refrain from restating the entirety of the information here, but we remain concerned—and increasingly so—that many MA plans continue to deny medically necessary inpatient rehabilitation care and are circumventing Medicare coverage rules in the process.

When Medicare beneficiaries are injured, become seriously ill, or require surgery, they often require medical rehabilitation to regain functional losses. The acute hospital stay is often just the first step toward recovery and returning to life in the community. Patients frequently require a course of hospital-based rehabilitation that is intensive, rehabilitation physician-directed and coordinated, and delivered by a multidisciplinary team.

IRFs strive to continue the healing process and deliver the medical and nursing care needed while also improving the quality of life for patients recovering from surgical procedures, strokes, spinal cord injuries, brain injuries, amputations, hip fractures, and many other conditions that decrease a person’s ability to function, live independently, and perform common

daily activities, such as walking, using a wheelchair, bathing, or eating. For example, a patient who sustains a stroke may be left with permanent neurological deficits and need to overcome or adapt to physical, language and cognitive impairments. Other post-acute care settings generally provide less intensive and less coordinated rehabilitation services without the nursing care levels and hours or physician availability of IRFs. Due to the uniquely intensive medical and rehabilitation services provided in an IRF, Medicare has incredibly rigorous screening criteria and other regulatory requirements to ensure that each and every patient admitted to an IRF is appropriate. The agency has developed detailed coverage regulations for Medicare IRF coverage.3 These coverage rules also apply to both Part A fee-forservice and Part C Medicare Advantage beneficiaries. Medicare regulations are clear that MA plans must provide “all Medicarecovered services.”4 These covered services include “all services that are covered by Part A,” which are the “basic benefits” available to MA enrollees.5 MA plans must comply with all Medicare coverage regulations and manuals.6 Medicare manuals are equally clear that an MA plan “must provide enrollees in that plan with all Original Medicare-covered services.”7 The relevant manual instructs that “[i]f the item or service is covered

by Original Medicare under Part A or Part B, including Part B prescription drugs, then it must be offered.”8 Therefore, MA plans must determine IRF coverage using the Part A regulations at 42 C.F.R. § 412.622 and other applicable guidance.

However, instead of following these Medicare IRF coverage criteria, many MA plans improperly apply private decision tools, such as Milliman and InterQual, to make coverage decisions that override clinical decision-making, both prospectively and retrospectively. The effect of this practice is to divert many enrollees who qualify for inpatient hospital rehabilitation to less appropriate, lower-acuity settings, such as nursing homes and homecare, inevitably decreasing their prospects for full recovery.

Even more frustrating, AMRPA members increasingly report that MA plans inform their enrollees that IRF care is not covered under their plan. This is why it is unsurprising that in its March 2017 Report to Congress, the Medicare Payment Advisory Commission (MedPAC) once again found that MA enrollees were admitted to IRFs at approximately one-third the rate of Medicare fee-forservice beneficiaries in 2015.9 To illustrate the tension between standard MA operating procedures and best clinical practices, the American Stroke Association (ASA) and American Heart Association (AHA) emphatically recommend that all stroke patients receive their immediate post-acute care in the IRF setting. The ASA/AHA

2 C.F.R. § 412.622(a). Among other requirements, to be covered in an IRF, the patient must need an interdisciplinary approach to care and be stable enough at 4 admission to participate in intensive rehabilitation. There must also be a “reasonable expectation” that the patient will need multidisciplinary therapy, intensive rehabilitation, and supervision by a rehabilitation physician. The requirement for multidisciplinary therapy must include physical or occupational therapy. Intensive rehabilitation is defined as three hours per day, five days per week (or 15 hours per week). The therapy must be reasonably likely to result in measurable, practical improvement to the patient’s functional capacity or adaptation to impairments. The rehabilitation physician must see the patient at least three times per week. Medicare coverage may not be denied based on treatment norms or rote “rules of thumb.”

4 42 C.F.R. § 422.10(c).

5 Id. § 422.101(a).

6 Id. § 422.101(b).

7 Medicare Managed Care Manual, ch. 4 § 10.2.

8 Id. § 10.3.

9 MEDICARE PAYMENT ADVISORY COMMISSION, REPORT TO THE CONGRESS: MEDICARE PAYMENT POLICY 298 (Mar. 2017) (finding that 2015 Medicare admissions to IRFs were 10.3 for every 1,000 FFS patients compared to 3.7 for every 1,000 MA patients).

3

AMRPA Magazine / April 2019 21

guidelines are based on years of clinical analysis, including the most comprehensive independent analysis ever undertaken in the field. However, many MA enrollees who suffer strokes are denied access to inpatient rehabilitation and redirected to nursing homes for their post-acute care. In one recent survey of our membership, AMRPA found that patients with a primary diagnosis of stroke constitute 30 percent of cases denied preadmission approval by MA plans. This practice directly contravenes evidence-based best practices and is purportedly based on the aforementioned decision support tools that MA plans refuse to divulge.

Further illustrating the deficiency in access to post-acute care provided by MA plans, independent researchers recently found that MA plan benefits are not designed to adequately meet enrollees’ post-acute and long-term care needs, and that post-acute provider networks and cost-sharing restrict access to needed care. The study’s authors identified these practices as driving a unidirectional flow of higher-cost enrollees from enrollment in MA back to traditional fee-for-service Medicare. 10 Improper Use of Non-Medicare Guidelines A number of problematic practices by managed care organizations are contributing to this worsening phenomenon.

Based on reports from AMRPA members, the rates of preadmission denials and retroactive claims denials have steadily risen as MA plans increasingly rely on proprietary guidelines such as Milliman and InterQual guidelines, defer to medical or clinical staff who lack rehabilitation expertise, and erect other administrative barriers that make appealing initial denials untenable for hospitalized patients, their caregivers, and the acute care hospitals forced to extend their stays until discharge plans are arranged.

These proprietary guidelines do not appear to mirror Medicare coverage but are nevertheless being used to deny patients access to medically necessary and clinically appropriate medical rehabilitation services. MA plans often refuse to share their placement assessments with providers, caregivers or others on the basis that the underlying decision tool is proprietary. This posture puts patients in an unwinnable situation and flaunts one of the underlying premises for having uniform Medicare basic benefits coverage policies that are available to all. AMRPA has sought to understand the Milliman product and through smallsample modeling it has become clear that virtually no patients are recommended for placement in the IRF setting, including those recovering from major strokes with paralysis and other debilitating injury and illness. Based on this modeling, 95 percent of reviewed cases qualifying for inpatient-level rehabilitation care were directed to a lower acuity setting, such as a nursing home or homecare.

To avoid such blatant disregard for Medicare requirements, AMRPA requests that the final Call Letter instruct MA plans to apply CMS’ coverage regulations governing IRFs. CMS must

ensure that MA plans are not designing benefits to discriminate against beneficiaries or discourage enrollment by inhibiting access to services or steering particular subsets of Medicare beneficiaries to specific coverage options.11 AMRPA believes that CMS should reiterate concerns that some MA plans may be disregarding anti-discrimination provisions with policies that impermissibly discourage particular services through measures such as excessive enrollee cost-sharing requirements.

However, AMRPA does not believe that monitoring cost-sharing practices is alone sufficient to combat discriminatory practices that both deny access and discourage enrollment among certain Medicare beneficiaries. For these reasons, AMRPA renews our request that CMS remedy this situation in the final Call Letter by explicitly instructing MA plans to: (1) refrain from using private decision tools to override clinical decisionmaking and subvert Medicare beneficiaries’ rights under the law; and instead (2) apply the existing Medicare coverage rules governing IRF care.

Disregard for MA Enrollee Appeal Rights Hospitalized MA enrollees are often precluded from exercising fundamental appeal rights in seeking clinically appropriate postacute care. In the final Call Letter, CMS has an opportunity to ensure patients’ basic appeal rights will be met.

MA enrollees are often completely unaware of their rights to the same benefits of those enrolled in traditional Medicare, as well as their right to appeal a denial of a preauthorization for services in a particular setting. The most vulnerable beneficiaries are often at the greatest risk of being denied access to medically necessary rehabilitation services without knowledge of the decisions being made behind the scenes, and may lack the social or financial supports necessary to appeal without guidance. Accompanying any preauthorization request, MA plans should be required to inform enrollees about their redetermination and appeal rights, including information about resources to help them navigate the process. There should also be a required disclosure of Medicare post-acute care coverage rules so Medicare beneficiaries receive adequate and explicit information about potential options upon admission to, and especially at discharge from, a short-term acute care hospital.

The operating procedures of MA plans erect numerous barriers, bureaucratic processes and delays, as well as unreasonable paperwork demands which restrict access to higher-acuity post-acute care settings, such as IRFs, and limit opportunities for timely redeterminations. MA plans frequently deny a referral to an IRF but decline to provide a copy of the denial notice to the patient or caregiver, thereby hindering the possibility of a successful appeal. MA plans are presently required to provide these notices upon request, but in light of the obvious access problems, CMS should instruct MA plans to provide denial information to the patient automatically, and to other health care providers whenever requested by the patient, a

omotazur Rahman et al., High-Cost Patients Had Substantial Rates of Leaving Medicare Advantage and Joining Traditional Medicare, 34(10) HEALTH AFF. 1675, 1679M 80 (Oct. 2015).

11 42 C.F.R. §422.100(f)(1)-(3).

10

22 AMRPA Magazine / April 2019

caregiver, or providers involved in delivering the patient’s acute or planned post-acute care. Additionally, patients should be held harmless when plans fail to comply with notice rules and requirements.

Further, managed care organizations often employ reviewers who lack relevant clinical experience to advise on referrals for medical rehabilitation. Based on AMRPA members’ experiences, it is rare for an MA plan’s medical reviewer to have any expertise or even baseline knowledge in medical rehabilitation, and thus most reviewers are often unable to understand the patient’s rehabilitation needs. In contrast, IRFs are required to employ a rehabilitation physician with specialized training in preadmission review to determine the appropriateness of a patient’s admission to an IRF, consistent with Medicare regulations.12 Despite this expertise, our members report that a substantial number of MA plans will only correspond with the referring physician from the acute care setting, who may be less qualified to make this determination, and also often refuse to correspond with the medical director of the referred-to setting, such as an IRF. To ensure patients are entitled to informed medical review, MA plans should be required to elevate an appeal to a clinician with relevant expertise within a reasonable amount of time, certainly within 24 hours, regardless of the day of the week. Further, CMS should direct MA plans to correspond with any clinician involved in the discharge planning process when making referral determinations and redeterminations.

MA plans often maintain unreasonably limited hours for considering preauthorization requests and redeterminations and stretch out their review processes over several days, essentially forcing hospitalized patients to be discharged to alternative settings. Current appeal processes permit MA plans to take up to 72 hours to render an initial decision or redetermination.

Moreover, AMRPA members report that if a determination period ends on a Friday, plans will often respond that no one is available to reconsider the determination until the following week. The aggregate effect of the high rate of initial denials, combined with administrative hurdles that slow the redetermination process, is that patients are stuck in the acute care setting, which is clinically inappropriate, introduces additional health risks to the patient and costs to the health care system. Additionally, MA plans often waive precertification requirements for subacute rehabilitation settings such as nursing homes. As a result, hospital personnel are pressured to discharge to these settings rather than wait days for MA plans to consider and reconsider referrals for inpatient rehabilitation. Over time, acute care providers and their discharge planning personnel become less willing to assist patients in obtaining the requisite approvals to access the appropriate level of rehabilitation care, especially when the administrative timeline needlessly prolongs the acute care stay as well as in light of increasing pressures to limit inpatient services Instead, discharge personnel increasingly make referrals only to post-acute care settings that they know will not be denied by the MA plan.

CMS should revisit these practices and use the Call Letter as an opportunity to eliminate unnecessary requirements that challenge beneficiaries’ ability to appeal. The agency must ensure 24/7 access by enforcing existing timelines and should work with plans to further expedite their processes to enable timely appeals. At a minimum, MA plans should be able to review and process post-acute care preauthorizations and redeterminations seven days a week and should never take more than 24 hours to respond. To that end, we request that all hospitalized patients needing a placement/admission determination be entitled to the emergency protocols with regard to medical review.

In addition to ensuring technical conformance with regulations, CMS must also do more to also ensure that hospitalized patients actually receive decisions that are timely enough to impact their future trajectory of care. Without this assurance, appeal rights are hollow.

To that end, AMRPA urges CMS take steps in the final Call Letter to ensure MA plan enrollees have a meaningful opportunity to appeal improper coverage denials.

Specifically, CMS should direct MA plans to: (1) provide Medicare post-acute care coverage rules so beneficiaries receive adequate information about potential upon discharge from a short-term acute care hospital; (2) automatically provide a copy of the denial notice to the patient or caregiver; (3) inform enrollees about their redetermination and appeal rights; (4) process preauthorizations and redeterminations within 24 hours for all hospitalized patients seeking authorization for post-acute care admissions; (5) enforce these timelines on weekends and holidays; (6) consult clinicians with relevant expertise for placement decisions; (7) be willing to communicate with the medical director of the referred-to setting; (8) increase the weight of the following measures capturing access: Plan Makes Timely Decisions about Appeals and Reviewing Appeals Decisions; and (9) make plan-level appeals and grievances publicly available so that Medicare beneficiaries are able to make informed choices about plan selection during open enrollment.

Inadequate Measurement of MA Performance There are currently insufficient incentives for MA plans to authorize referrals to inpatient medical rehabilitation settings, which can cost slightly more than non-hospital settings in the short term, but produce better outcomes and savings in the longterm. Notably, patients’ long-term survival and outcomes have been shown to vary significantly by post-acute setting, hence the AHA/ASA recommendation. The most robust study on this topic, performed by Dobson DaVanzo & Associates, found that Medicare beneficiaries admitted to IRFs for their immediate postacute care had significantly better outcomes across a range of quality indicators compared to highly matched beneficiaries who received their immediate post-acute care in a SNF.13 According to the findings, modestly higher spending on immediate post-acute

42 C.F.R. § 412.622(a)(4)(i).

See DOBSON DAVANZO & ASSOCIATES, ASSESSMENT OF PATIENT OUTCOMES OF REHABILITATIVE CARE PROVIDED IN INPATIENT REHABILITATION FACILITIES (IRFS) AND AFTER DISCHARGE (July 2014).

12 13

AMRPA Magazine / April 2019 23

care in the IRF setting was generally offset over the course of the two-year period. 14 MA plans are not currently held accountable for many relevant quality outcomes, such discharge back to the community, or longterm health outcomes, such as days in the community, and thus systematically fail to make an investment in enrollees’ long-term health. Since these quality indicators do not impact MA plans’ payment, decision-making too often ignores consideration of what is best for the patient. AMRPA would like to work with CMS and other stakeholders to rectify this overarching shortcoming of the MA program.

In the meantime, we encourage CMS to enhance transparency about MA utilization of post-acute care services. While individual medical providers experience inappropriate denials of patient referrals on a daily basis, CMS appears to lack robust data on the aggregate number (and proportion) of placement decisions, including the total number of patients being referred to different post-acute care settings. We encourage CMS to enhance reporting on the total numbers (and proportion) of denials which are successfully and unsuccessfully appealed and the number of retroactive/post-payment denials and appeals. CMS should promptly institute reporting requirements for MA plans to begin recording this baseline data in uniform data sets and be required to report this information to CMS on a quarterly basis. The agency should make this information public on an annual basis.

Additionally, AMRPA encourages CMS to make changes in the final Call Letter to audit plan performance along additional

14 15

Id. Draft Call Letter, at 110.

24 AMRPA Magazine / April 2019

dimensions including compliance with Medicare coverage rules and beneficiary protections. AMRPA appreciates that CMS includes several appeals measures in the Star Ratings program because these metrics are critical indicators of overall plan performance.15 However, CMS should do more than just require MA plans to submit data on appeals, but should audit the information to ensure its accuracy and conformance with regulatory requirements. *** AMRPA appreciates the opportunity to provide comments on the Draft Call Letter and is hopeful that many of these concerns can be addressed in the final. If you have any questions regarding our comments, please contact Carolyn Zollar, J.D., AMRPA’s Senior Counsel at (202) 223-1920 or czollar@amrpa.org, Kate Beller, J.D., AMRPA’s Executive Vice President for Government Relations and Policy Development at (202) 223-1920 or kbeller@amrpa.org, or AMRPA’s Washington Counsel Martha Kendrick, J.D., at (202) 8874215 or mkendrick@akingump.com.

Sincerely,

Richard Kathrins, Ph.D. Chair, AMRPA Board of Directors President and CEO, Bacharach Institute for Rehabilitation

2019 AMRPA Schedule of Events CONFERENCE DATES 2019 Regional Meeting in Grand Rapids, MI Friday, June 7, 2019: Mary Free Bed Rehabilitation Hospital 2019 Regional Meeting in Jacksonville, FL Friday, June 14, 2019: Brooks Rehabilitation Hospital 2019 Fall Educational Conference & Expo in San Diego, California Sunday, October 13, 2019: IRF Boot Camp October 14-16, 2019: Fall Conference & Expo AMRPA WEBINARS Thursday, April 11, 2019, 12:30 p.m.: Medical Rehabilitation Headlines and Highlights

by Carolyn Zollar, JD, AMRPA Senior Policy Counsel AMRPA MEMBERS ONLY CALLS Wednesday, April 17 at 1:00 p.m. ET Wednesday, June 19 at 1:00 p.m. ET Wednesday, August 21 at 1:00 p.m. ET Wednesday, October 23 at 1:00 p.m. ET Wednesday, December 18 at 1:00 p.m. ET eRehabData® WEBINARS: AVAILABLE TO eRehabData® SUBSCRIBERS ONLY Tuesday, April 2: Nursing and Therapy Documentation Tips Tuesday, May 7: Physician Documentation Tuesday, June 4: Managing Outcomes with eRehabData

Please visit www.amrpa.org for registration information.

AMRPA Magazine / April 2019 25

AMRPA Responds To RTI on the Development of a Unified Post-Acute Care Prospective Payment System Editor’s Note: On February 15, 2019, the American Medical Rehabilitation Providers Association (AMRPA) submitted comments to RTI International (RTI) and the Centers for Medicare and Medicaid Services (CMS) on the development of a unified post-acute care prospective payment system (unified PAC PPS).

The complete letter is provided below and is available at www.amrpa.org.

February 15, 2019 Melissa Morley, Ph.D. Senior Health Economist RTI International 3040 East Cornwallis Road P.O. Box 12194 Research Triangle Park, NC 27709 Anne Deutsch, Ph.D Senior Research Public Health Analyst RTI International 3040 East Cornwallis Road P.O. Box 12194 Research Triangle Park, NC 27709 Re: RTI/CMS Technical Expert Panel (TEP) on a Unified Post-Acute Care Prospective Payment System -- Questions for TEP Input Dear Drs. Morley and Deutsch: On behalf of the American Medical Rehabilitation Providers Association (AMRPA), we welcome the opportunity to offer feedback on the development of a unified post-acute care prospective payment system (unified PAC PPS) pursuant to the requirements of the Improving Medicare Post-Acute Care Transformation (IMPACT) Act of 2014. AMRPA strongly supports RTI and its governmental colleagues, the Department of Health and Human Services (HHS) Office of the Assistant Secretary for Planning and Evaluation (ASPE) and the Centers for Medicare and Medicaid Services (CMS), for involving the affected post-acute care (PAC) stakeholders in this initial effort to define what is or can be a unified PAC PPS. This paper reflects our preliminary but not exhaustive views on the questions posed to the technical expert panel (TEP) first convened on September 24, 2018. We look forward to continued dialogue.

AMRPA is the national trade association representing more than 625 freestanding inpatient rehabilitation hospitals and rehabilitation units of general hospitals (referred to as inpatient rehabilitation facilities (IRFs) by Medicare), outpatient rehabilitation service providers, and several long-term care hospitals (LTCHs) and skilled nursing facilities (SNFs). Inpatient rehabilitation hospitals and units (IRH/Us) provide hospital-level care, which is significantly different in intensity, capacity, and outcomes from care provided in non-hospital post-acute settings. AMRPA members help patients maximize their health, functional skills, independence, and participation in society so they can return to home, work, or an active retirement.

I. General Comments First, AMRPA appreciates RTI’s willingness to extend the deadline for these comments which has allowed us to produce more substantive written feedback with broader input from our member clinicians and administrators. We respectfully request that RTI/CMS grant TEP stakeholders a minimum 60-day comment period moving forward. Developing a unified PAC PPS is indeed monumental work and

26 AMRPA Magazine / April 2019

a minimum 60-day comment deadline would better enable stakeholders to provide RTI/CMS with comprehensive and insightful input.

Second, AMRPA encourages RTI and CMS to continue to involve PAC stakeholders frequently as the project progresses to discuss additional policy and structural issues and to discuss data analyses as they are developed. In that same vein, AMRPA strongly encourages CMS and its various contractors to be fully transparent in information sharing with stakeholders and make available their technical data and analysis. Doing so will lead to a greater understanding of the importance of various issues and the subsequent decisions made by the researchers. The inclusion of stakeholders is more likely to result in their support for a final proposal once designed. The RAND Corporation and CMS engaged the inpatient rehabilitation provider community actively and frequently when they were developing the inpatient rehabilitation facility prospective payment system (IRF PPS). And, AMRPA believes, as a result of that extensive involvement the final system was successful in its design and implementation.

Third, according to RTI, the focus of this initial TEP exercise is to establish foundational principles for developing a unified PAC PPS, and specifically to understand the beneficiaryspecific costs and setting-specific costs underpinning PAC today. To understand the true extent of these costs, RTI/CMS must recognize the value of hospital-level care provided in higher intensity PAC settings such as IRH/Us and LTCHs. As hospitals, IRH/Us are able to address the needs of higher-acuity and medically complex patients who require intensive, 24-hour-a-day, interdisciplinary care provided under the direct supervision of a rehabilitation physician. Operating under federal and state requirements for hospitals, as well as setting-specific regulations, layers additional costs to the care provided in IRH/Us. However, the value proposition is indisputable: IRH/Us deliver better, more durable outcomes and return patients to their home and communities sooner.1,2,3 The unified PAC PPS must not diminish or undervalue the critical role of hospital-level PAC in the continuum of care, and it must be developed to meet the needs of all persons who require PAC. Patient acuity level and PAC setting selection drives both differences in cost and outcomes. Under a unified PAC PPS, patients should receive care in the PAC setting that can address the full range of their clinical needs to achieve the best outcomes for their acuity level.

RTI/CMS must remain cognizant that the cost, intensity, and quality of clinical services – even when they may seem to be identical services being delivered across PAC – vary dramatically across settings due to three key discriminating factors in today’s environment:

1. Institutional PAC versus home health; 2. Hospital-based versus non-hospital PAC providers; and 3. Asymmetry in setting-specific regulations.

We discuss these in greater detail below.

Finally, a unified PAC PPS would mark a seismic shift in how Medicare pays for PAC and likely the regulatory and operational paradigm under which providers deliver care.

As CMS continues its work to implement the IMPACT Act, AMRPA urges the agency to conduct a demonstration or pilot to test a unified PAC PPS model for its feasibility and viability as a prototype payment system under the authority granted to the Center for Medicare and Medicaid Innovation (CMMI). Specifically, CMS should conduct this pilot before sending a report to Congress on the unified PAC PPS prototype and accompanying policy recommendations, as such demonstration pilots should inform both the prototype and accompanying policy recommendations.

II. Foundational Principles to Guide Post-Acute Care Payment Reform In 2009, AMRPA initiated discussions about a new type of PAC service delivery system and, separately, the impact of any national bundling project on the rehabilitation community. Both of those efforts were guided by several foundational principles. We firmly believe that as RTI progresses in developing the unified PAC PPS that it should reflect these principles. Specifically, we urge that any PAC payment reform be based on the guiding principles summarized below. They are:

A. Non Discrimination in Access and Service

• The payment system should assure that there is no discrimination in admission or treatment of vulnerable populations, particularly persons with disabilities.

B. Patients

• The payment systems needs to be patient-centered with a focus on restoring health and independence, enhancing function, and returning patients to their homes, schools, jobs and communities.

• Service delivery should be organized to optimize meeting the needs of patients.

• It must serve the needs of persons with disabilities and chronic conditions in particular, as well as those with acute health problems and assure full access to care. Hence, persons with functional loss must have access to medical rehabilitation and medically complex services that are:

o Focused on prevention of further medical complications; and

o Intended to improve health, outcomes, optimize functional ability, and activity and participation in society, not just survival.

Dobson Davanzo & Associates, Assessment of Patient Outcomes of Rehabilitative Care Provided in Inpatient Rehabilitation Facilities and After Discharge (July 2014).

Kramer AM, Steiner JF, Schlenker RE, et al: Outcomes and costs after hip fracture and stroke, A comparison of rehabilitation settings, JAMA 1997; 277:396-403.

3 American Heart Association/American Stroke Association, Guidelines for Adult Stroke Rehabilitation and Recovery. May 4, 2016.

1 2

AMRPA Magazine / April 2019 27

• Any model of change must provide patients with an adequate choice of providers, suppliers and services.

• It should maximize outcomes and patient satisfaction.

C. Providers

• Appropriate physician involvement, direction and oversight are key and essential to the delivery of medical rehabilitation and complex medical care.

• Providers should deliver services commensurate with the intensity of service needs of patients.

• Intensity of services received should be provided based on the best available clinical evidence and expert judgment.

• Medical staff organization should have mechanisms for credentialing that ensure appropriate physician involvement, direction and oversight in order to deliver complex medical and medical rehabilitation care.

• Providers should be able to receive reasonable payment for delivering high quality care.

• The model should be designed to maximize innovation and investment in staff infrastructure.

• Providers should be free of regulatory barriers in order to organize the delivery of services to patients in the most effective ways.

D. Payment

• Cost-effective and cost-efficient care should be promoted.

• Payment must thoroughly account for all payments, costs and resources which reflect the characteristics of patients served as well as costs not related to patient characteristics.

• The system should maximize administrative simplicity for providers and payers.

E. Quality

• High quality care should be enhanced, sought, delivered, fairly reimbursed and maximize patient/ family outcomes and satisfaction.

• Reimbursement and measures of success for providers should be risk adjusted to promote the care of those with the greatest need. Such measures must meet accepted measurement standards. Hence, care must be taken in the design to assure that any bias against caring for the hardest cases is removed and that there are no incentives to stint on care or game the payment.

• Risk adjustment will promote treatment for all who need services and not solely those categories of patients whose quality outcomes can be achieved at low cost and therefore who perform well on selected quality measures.

4

• Quality measures selected should promote positive outcomes, avoidance of adverse events and demonstrate effectiveness and efficiency of care.

F. Additional Criteria

• Should adjust adequately for environmental factors.

• Should encourage an economically rational organization of service capacity.

• Financial risk should be minimized.

III. Responses to RTI’s Questions

1. What clinical services should be considered in a unified PAC PPS given the differences in the packages of services included in the current PAC PPSs? For example, LTCH payments include nursing, therapy, comprehensive ancillaries, and an interrupted stay policy for acute hospitalizations. In contrast, HHA payments include nursing, therapy, aide, social work and supplies, but do not include comprehensive ancillaries or an interrupted stay policy for acute hospitalizations.

With regard to clinical services, the unified PAC PPS should include those services that are currently covered under the Medicare Part A fee-for-service benefit in institutional PAC settings. However, AMRPA recommends that certain high-cost services that are unrelated to the patient’s need for PAC (such as chemotherapy, dialysis, or infusion therapy) be carved out of the unified PAC PPS payment rate to better align incentives between provider costs and patient access.

We note that RTI is also examining Part B Medicare utilization and payments in addition to Part A.4 If CMS’ implementation of the IMPACT Act were to consider including other clinical services used by PAC patients, then Part B services such as physician services and outpatient rehabilitation should be included in RTI/CMS’ assessment of costs. PAC payment reform should be cognizant of outpatient rehabilitation use in particular, as many PAC patients utilize those services today and this component is likely to continue to grow in the future.

The clinical services covered within the current PAC PPSs vary, as RTI recognizes in this question, in large part because they reflect the service that can be delivered in institutional settings (IRH/U, LTCH, and SNF) versus in HHAs. Hence, achieving the longer-term goal of more uniform service coverage across PAC is complicated by the infrastructure realities of PAC settings today and specifically, their capabilities. In other words, Medicare’s expectations for covered services in a LTCH or IRH/U versus an HHA differ because institutional PAC is able to provide a broader range of services (while incurring the associated costs) and these services and costs are reflected in the settings’ Medicare Part A payments. See Table A below.

RTI International, Developing a Unified Post-Acute Care Payment System Exhibit Booklet. September 24, 2018. See Exhibits B.1 and B.2 where “Mean 90-payments” is defined as inclusive of all Medicare Parts A and B service use for 90 days following acute care hospital discharge.

28 AMRPA Magazine / April 2019

Table A: Services Covered in Today’s PAC Payment Systems Leading to Costs Included in Medicare Part A Payment Services

HHA

Therapy (PT, OT, SLP)

Yes

Nursing, aides

Yes

Medical social services

Yes

Other routine costs (room and board, etc.)

No

Non-therapy ancillary (drugs, supplies)

No

IRH/U

LTCH

SNF

Yes*

*Note: Although these clinical services are uniformly covered in Part A in institutional PAC settings, the cost to providers of providing these services differs significantly across institutional settings due to other, setting-specific factors.

When applicable, payment policies exist for certain outlier patients who require the provision of services beyond the established discharge-based payment rate. By way of example, these policies include the LTCH interrupted stay policy, and for IRH/Us, the high-cost outlier payment policy for extraordinarily costly cases, and the interrupted stay policy, in which services rendered within the interrupted days are billed separately to Medicare and not part of the IRH/U’s financial responsibility.5 Notably, the IRH/U interrupted stay policy is different from the homonymous one in the LTCH PPS, in which the LTCH is responsible for services provided in other settings during those days.6 This policy may be designed to satisfy the expectation that LTCHs, as hospitals treating a focused group of clinically complex individuals, would be able to address the vast majority of patients’ emergent needs, and hence the LTCH interrupted stay policy seeks to disincentivize the inappropriate shifting of care to other providers. The HHA PPS does not include many services that would otherwise be absorbed under Part A payments to IRH/ Us, LTCHs, and SNFs because home health agencies do not have the infrastructure to provide it, and hence HHA PPS payments are considerably lower in comparison.

But, in the longer run, defining the services that should be in a unified PAC PPS must have a more fundamental nascence than simply repurposing what currently is covered in the siloed payment systems. Speaking bluntly, adhering to the current definitions places too much credence on the prudence and rationality of these policies in the first place. In a genuine sense, we now enjoy an opportunity to examine the postacute continuum afresh, as a whole and across silos, to determine the services again that should be covered and then the concomitant payment levels needed to cover the cost of treating patients, agnostic of the treatment setting.

5 6

In the longer term, AMRPA recommends that RTI/ CMS conduct a comprehensive evaluation to define the medical, rehabilitation, and other ancillary service needs of PAC patients, and this evaluation should inform what clinical services are considered in a unified PAC PPS. The unified PAC PPS should also account for other appropriate services such as care coordination, medication reconciliation, discharge planning, transitional care services, and other patient-centered activities. Following that exercise, CMS can then turn to developing payment policies to address services to be included in the unified PAC PPS, which should include companion, special payment policies for unanticipated events such as acute hospital transfers. Of course, these companion policies would likely vary by the capability and service intensity of a given “setting” (however that term may be defined in a unified PAC PPS), for the reasons described above.

Ultimately, the unified PAC PPS must be developed with the overarching goal of meeting the needs of all persons who require PAC, from medically complex patients with acute health care needs, individuals with disabilities and chronic conditions, and those that have continuing maintenance needs. AMRPA thinks the unified PAC PPS presents a unique opportunity to achieve this goal of ensuring full access as well as preserving the solvency of the Medicare Trust Fund. With those goals in mind, the question becomes less: “What are the services the PAC PPS should cover?” and more: “How can a unified PAC payment system be designed so that it aligns incentives appropriately across patients, providers, and payers to meet patient needs in an efficient and effective manner?”

2. Is it necessary or desirable to differentiate between beneficiary-specific costs and setting-specific costs in the development of a unified PAC PPS to reduce over or under payment? Does the focus on beneficiary-specific costs address the goal of payment based on patient characteristics?

It is critical to clearly identify and distinguish what costs are beneficiary-specific versus those that are settingspecific, because the total and comprehensive cost of patient care is a combination of the two. Independent of patient case-mix, there are considerable differences in provider costs across PAC settings that must be addressed in order to develop appropriate payment levels. We discuss these issues in greater detail in the next question.

A focus on beneficiary-specific costs is a step in the right direction towards developing appropriate payment rates based on patient characteristics. However, beneficiary

42 CFR § 412.624 Id. § 412.531

AMRPA Magazine / April 2019 29

costs, particularly when that cost information is derived solely from Medicare’s administrative data, does not necessarily represent the optimal allocation of services or Medicare dollars based on patient characteristics (see: the historically misaligned and volume-based payment incentives in the SNF and HHA PPSs).

The IMPACT Act mandates that the unified PAC PPS be a payment system driven by patient care needs and not the setting of care. The challenge therein is identifying and attributing beneficiary-specific costs appropriately.

This step is complicated by various factors, including the vast heterogeneity yet occasional apparent overlap of PAC patients across the four settings, and the lack of routine cost data for providers. A stroke rehabilitation patient in an IRH/U might look clinically similar to a SNF stroke patient “on paper” (i.e., according to administrative data), but in actuality evidence suggests they are likely very different based on acuity, prognosis, functional ability, and other more granular clinical characteristics.

As another clinical example: A traumatic brain injury (TBI) patient in an IRH/U exhibits agitation and significant behavioral issues, and requires additional pharmacological intervention as well as additional staffing resources in the form of intensive nursing intervention and team management to monitor and de-escalate the patient overnight. IRH/Us, as hospitals, have around-the-clock nursing and physician coverage and thus are able to provide the services this patient needs. The same cannot be said if the patient were in a non-hospital PAC setting. For such a patient, could Medicare’s administrative data adequately distinguish between an IRH/U-TBI patient and a SNF-TBI patient in a way that recognizes the significant difference in provider costs? And for this patient, where do beneficiary-specific costs end and setting-specific costs begin?

Routine Costs In its work on a PAC PPS prototype, the Medicare Payment Advisory Commission (MedPAC) uses provider costs as one approach to calculate hypothetical payment levels for a unified PAC PPS. Provider costs, however, are not the same thing as beneficiary needs or costs. Secondly, existing provider cost data is limited as institutional PAC claims data do not include patient-level data of routine service use, only a flat daily room and board charge.7 Hence, the Post-Acute Care Payment Reform Demonstration (PAC PRD) project collected separate data on all direct patient care and support staff resources per patient stay; the data were then weighted to account for licensure and occupation.

To estimate the costs of stays accurately, CMS needs patient/stay-level data on routine costs, particularly nursing costs. While the cost structure differences between institutional PAC versus HHAs is clear, the unified PAC PPS design must also consider and account for differences in routine costs across institutional settings in order to establish appropriate payment levels. This granularity has been not factored into unified PAC PPS models presented to policymakers to date.8 While the differences may not be as large a differential as institutional versus home health care costs, significant considerations remain.

RTI/CMS should also consider intra-setting differences in costs. For example, a rehabilitation hospital that is a Spinal Cord Injury (SCI) Center of Excellence and provides more specialized care would have higher costs for treating SCI patients than the typical IRH/U. Experience with certain case types or conditions will allow some PAC providers to have higher levels of competencies, treatment protocols and better outcomes which should be considered as RTI/CMS look at patient volume and the severity levels within each condition.

Thanks to advances in medicine, many additional patients, including cancer patients and organ transplant patients, today benefit from the intensive rehabilitation and medical management services provided in IRH/ Us. These patients have much higher acuity due to their medically complex conditions. Patients who have had solid organ transplant, stem cell transplant or bone marrow transplants require more resources for care related to their conditions such as immunosuppression, viral loads, anti-rejection medications and medical fragility. Similarly, patients receiving chemotherapy and radiation during their rehabilitation require more resources due to the types of drugs, procedures, and other factors associated with their treatment such as nausea, vomiting, and fatigue. Certain chemotherapy drugs require administration by specialty staff and medication exposure precautions. The development of a unified PAC PPS will require RTI/CMS to consider all of these costs to pay providers fairly and accurately. To do so will require the consideration of the patient clinical characteristics and the setting-specific routine costs.

CMS must ensure that it can adequately disentangle true beneficiary costs from setting-specific costs in evaluating the total cost of care. Only then can the unified PAC PPS appropriately pay for care based on patient characteristics. How to achieve this should be a focus of future TEP discussions.

Urban Institute, Designing a Unified Prospective Payment System for Postacute Care: A report by staff from the Urban Institute for the Medicare Payment Advisory Commission, 8 (June 2016).

8 Medicare Payment Advisory Commission, Report to Congress: Medicare and the Health Care Delivery System, 99 (June 2018).

7

30 AMRPA Magazine / April 2019

3.

Are there setting-specific considerations that a unified PAC PPS should take into account?

a. Setting-specific costs (e.g., inpatient settings versus home health)?

As mentioned above, the cost, intensity, and quality of clinical services vary dramatically across settings due to three key discriminating factors in today’s PAC environment:

•

• •

H ospital-level PAC versus non-hospital PAC providers; A symmetry in setting-specific regulations; and Institutional PAC settings versus home health.

RTI/CMS must bear in mind that these factors influence both provider costs and the services covered under Part A payments and thereby the cost to the Medicare program for addressing beneficiary-related needs, i.e., beneficiary-specific costs.

Costs Due to Hospital CoPs and Setting-Specific Requirements and Standards It is well recognized that some PAC settings have significantly higher costs due to Medicare’s conditions of participation (CoPs) requirements for hospitals and Medicare’s setting-specific regulations. For IRH/Us, maintaining compliance with the hospital CoPs, coverage requirements, and classification requirements adds significant administrative and clinical staffing costs when compared with other PAC settings that may treat similar patient diagnoses. We illustrate the key differences in Table B: Contributors to SettingSpecific Costs and Table C: Differences in PAC Coverage Requirements Contribute to SettingSpecific Staffing and Administrative Costs.

Moreover, by the very nature of each setting, the patients treated in that setting, and the underlying regulatory requirements, RTI/CMS must recognize and account for the added setting-specific cost particularly applicable to operate IRH/Us and LTCHs. The cost to construct rehabilitation hospitals or LTCHs physical plants vary greatly from SNFs and HHAs, and must meet specific federal and state requirements to be licensed and operate as hospitals. These physical plants require additional overhead costs to maintain them in order to pass multiple state-specific surveys.

For IRH/Us and LTCHs, the hospital CoPs add to staffing costs (physician and nursing compensation) and numerous administrative or capital costs in maintaining compliance. For example, all 9

hospitals have costs related to maintaining the physical plant that not all SNFs or HHAs have to incur, such as equipment standards for backup generators, oxygen tank storage, among many other requirements. The average cost of a backup generator is $350,000-$400,000. By contrast, most HHAs lease a small office space. These cost differences are only a small subset of the actual cost differentials to operate in each setting and must be accounted for to adequately pay for care in each PAC setting.

Routine and ancillary costs vary greatly not only between HHAs and institutional settings, but across institutional settings as well. Due to state and federal requirements, each setting has a completely different compliment of staff types and level. IRH/U services differ significantly from other settings of PAC due to the close medical supervision and intensive, interdisciplinary approach to rehabilitation care. Rehabilitation hospitals must have a medical director who is a physician with special training in rehabilitation medicine, and an interdisciplinary treatment team comprised of a rehabilitation medicine physician, rehabilitation nurses, registered nurses (RNs), and certified physical, occupational, and speech language therapists. By contrast, utilization of RN personnel occurs less consistently within SNFs, where licensed nurse practitioners (LPNs) and nurse aides are utilized more frequently and physician oversight is comparatively far more limited.

Costs related to certain specialized patient care should also be considered. There are additional costs incurred to treat these specific patients. These costs include (but are not limited to) specialized therapy equipment, ventilators, costs to meet accreditation thresholds granted by agencies such as the Joint Commission, staffing mix and higher costs related to such services as dialysis, drugs and medical supplies. An example of the higher level of care needed by this group is exemplified by the American Stroke Association’s recommendation that stroke patients be treated at an IRH/U rather than a SNF due to their studies showing that the intensive level of therapy and hospital-level setting achieves a better outcome for this subset of patients.9

American Heart Association/American Stroke Association, Guidelines for Adult Stroke Rehabilitation and Recovery (May 2016).

AMRPA Magazine / April 2019 31

Table B: Contributors to Setting-Specific Costs IRH/Us

LTCHs

SNFs

HHAs

Medicare Classification Criteria

60% Rule

25-Day Length of Stay Rule; non-CCI site-neutral payments

None

None

Setting-Specific Coverage Criteria

Yes, see below.

Yes

Significantly less restrictive, see Table C below.

Significantly less restrictive, see Table C below.

Medicare Conditions of Participation (CoPs)

IRH/Us and LTCHs must comply with Hospital CoPs: • Physician must be on duty or on call 24/7 to provide close medical supervision • Physician must write admission orders • Nursing must be furnished by an RN • RN or LPN must be on duty 24/7

SNF CoPs • H ave arrangements for physician services in case of emergencies • N ursing may be furnished by LPN or RN • R N or LPN must be on duty at least 8 hrs/day

HHA CoPs • M ust have policies established by a group of professionals including at least one physician and one or more RNs • R N or physician can supervise overall provision of services

Administrative Resources to Comply with Hospital CoPs: • Compliance with state and hospital accreditation organizations (e.g., Joint Commission) • CoPs require hospitals to establish, review and revise policies and procedures; collect and report data; develop and issue patient notices; comply with specific building codes; establish committees to address quality, credentialing and other activities; document specific elements of patient care and services in specific ways; and make medical records available to patients • While not required by Medicare, many IRH/Us voluntarily acquire accreditation from the Commission on Accreditation of Rehabilitation Facilities (CARF) Accreditations

Spinal Cord Injury Rehabilitation, Brain Injury Rehabilitation, Parkinson's Disease Rehabilitation, Stroke Rehabilitation, Hip Fracture Rehabilitation, Advanced Inpatient Diabetes, Heart Failure

Quality Respiratory Care Hospital

Specialized Patient Populations

Brain Injury, Parkinson, Stroke,

Ventilator; Complex Wound Care

HomeCare Elite

Wound Care

N/A

Spinal Cord

Institutional PAC versus Home Health: Considerations for Assessing Setting and Beneficiary Costs Without a doubt, the unified PAC PPS needs to adjust for the cost structure differences between HHA and institutional settings as HHA costs do not include non-therapy ancillaries (NTAs) or routine costs such as room and board. This adjustment is well-recognized, as reflected in the MedPAC unified PAC PPS and PAC PRD work.

Because the Medicare HH benefit does not include NTAs, we further recommend that RTI/CMS expand its analysis of beneficiary costs in HHAs beyond Part A to include spending that may be showing up in other line items of Medicare’s balance book. There are beneficiary-specific costs and costs of treatment that are currently being absorbed in institutional PAC settings that are not necessarily reflected when looking at spending related strictly to the HH Part A benefit.

32 AMRPA Magazine / April 2019

To illustrate, Mrs. Jones is a hip fracture patient being discharged to PAC, and she is also receiving infusions for an unrelated reason. If Mrs. Jones were to go to an IRH/U, the cost of infusions is absorbed by the provider under the IRF PPS Part A payment (and those of other institutional PACs) and does not show up elsewhere as a Medicare cost. This is one reason why these types of patients can be so costly to IRH/Us. If Mrs. Jones were to go to a HHA, the infusions would be paid separately and not reflected in the HHA PPS payments. Hence the cost of treating Mrs. Jones must account for this difference in order to assess total Medicare beneficiary costs representatively. In other words, while it is understood that setting-specific costs differ, RTI/CMS must recognize that beneficiaryspecific costs, even for clinically similar patients, also differ significantly across PAC settings due to current asymmetries in Medicare coverage across silos. This is especially relevant for medically

Table C: Differences in PAC Coverage Requirements Contribute to Setting-Specific Staffing and Administrative Costs IRH/U Coverage Requirements • M edical record must document the reasonable expectation that patient will need and benefit from multidisciplinary therapy, intensive rehabilitation, and supervision by a physician with specialized training in rehabilitation • A licensed or certified clinician must conduct a pre-admission prior to admission to the IRH/U • C lose physician supervision with the following responsibilities required: o P ost-admission evaluation conducted and documented by a physician within 48 hours after admission o C omplete an individualized plan of care (IPOC) within the first 4 days after admission o P hysician visits 3x/week o L ead a multidisciplinary team meetings once a week, and documented by a physician in the plan of care • Weekly multidisciplinary team meetings with attendance attestation from RN with specialized training or experience in rehabilitation, social worker or case manager, licensed or certified therapists (PT, OT, SLP), with additional attendance from neuropsychologists, audiologists, prosthetist/ orthotists • I ntensity of therapy requirements regarding the amount of therapy (generally interpreted as the “3-Hour rule”), and type of therapy (IRH/U patients must need at least two disciplines, PT and OT, with care led by licensed therapists)

SNF Coverage Requirements

HHA Coverage Requirements

• Stay must be preceded by 3-day inpatient hospital stay • SNF residents must be under the care of a physician at least once every 30 days for the first 90 days and at least once every 60 days thereafter*

• P hysician face-to-face visit certification that patient is eligible and needs HH services • M ust be homebound and need intermittent skilled services less than 8 hours a day • P lan of care reviewed and signed by physician in consultation with HHA professional every 60 days

*After the initial visit, the required physician visit may alternate between physician, physician’s assistant, nurse practitioner, or a clinical nurse specialist

complex patients and more comorbid patients who require specialized treatments, as these are not furnished in HHAs (and some of which are not even furnished in all institutional PAC settings).

While the unified PAC PPS may be focused on Part A fee for service costs, RTI/CMS must account for the full array of Medicare expenditures or otherwise risk underpaying institutional settings. RTI/CMS could begin to do so by ensuring that its cost analysis encompasses other sources of Medicare payment, such as Part B and Part D, particularly as high-cost beneficiaries use more of these services than others.10 How accurately representative the “cost of a stay” is will depend on how many sources of Medicare spending are incorporated into the analysis. We further recommend that RTI/CMS also look at Medicaid spending, particularly as this is a factor in the SNF benefit after a certain time.

b. Regulatory burden, provider burden, beneficiary protection, or Medicare program impact?

As described above, the overall regulatory burden on providers is significant, most especially for hospital-level PAC settings. A survey by the American Hospital Association found that providers dedicate the largest proportion of resources to documenting CoPs adherence and coverage verification processes. 11 That said, AMRPA agrees that criteria such as minimum staffing levels may be necessary to ensure safety and quality to meet the needs of patients who require higher-intensity, hospital-level care.

With regard to beneficiary protections, the unified PAC PPS as a Medicare FFS program must preserve beneficiaries’ rights to free choice of provider. If there is a demonstration of the unified PAC PPS as we recommend, freedom of choice must be fully protected such that beneficiaries are not required

High-Cost Medicare Beneficiaries Spend More on Outpatient Care and Medications. See the Commonwealth Fund study, “More Than One-Quarter of High-Cost Medicare Patients Have Persistent High Costs Over Three Years” (Jan 2019).

11 American Hospital Association, Regulatory Overload: Assessing the Regulatory Burden on Health Systems, Hospitals, and Post-Acute Care Providers (Oct 2017).

10

AMRPA Magazine / April 2019 33

Medicare dollars are supposed to follow the patients in a unified PAC PPS, then for non-overlapping conditions payment is already following patients to a unique site of care.

to participate in any demonstration or particular type of service delivery system in order to receive services.

c. Other adjustments (e.g., outlier policies or geographic adjustments)?

Certain cost variations that are not within providers’ control should be accounted for in the unified PAC PPS system. These variants may be a function of governmental (state, local, or federal) requirements, geography or unique system delivery factors and need to be recognized. These adjustments include wage adjustment, teaching status adjustment, rural area location, and low-income patient (LIP) load, among others. One example of a state regulatory cost is a North Carolina requirement that IRH/Us must provide 4 or 4.5 hours of therapy for spinal cord injury and brain injury patients, regardless of payer source. As such, it costs North Carolina IRH/ Us 150 percent more to provide these patients’ therapy compared to an IRH/U just across the border in South Carolina.

In addition to facility-based adjustors, additional aspects of the unified PAC PPS should be outlier payment policies for extraordinary situations at the patient level, as discussed previously. The unified PAC PPS should not discriminate in any way against higher-needs patients or unintentionally limit patient access to necessary services. On the other end of the spectrum, unexpectedly short-stays should also be addressed in a companion payment policy so as not to overpay for inappropriate admissions.

4. In developing a unified PAC PPS, what options can be considered in the shorter versus longer term?

[From RTI’s September 24, 2018 presentation]

In the short term, CMS should gain more detailed information on provider costs across settings as there is very limited stay-level information available on routine costs. As noted above, while a sample of this information was collected in the PAC PRD, these data are not current nor are they not representative of today’s practices or patient case mix. Similarly, CMS must have a fuller understanding of which costs are truly beneficiaryspecific and based on patient needs, agnostic of setting.

CMS should then continue to align payments with costs to minimize the likelihood that legacy misaligned incentives are carried over into a new system and to ease any transition to a new payment system.

The TEP members suggested that as an initial test CMS focus on developing a unified PAC PPS for high-volume conditions where there may be overlap in the settings in which patients receive care. We think this approach would provide a good foray in the shorter term. If

34 AMRPA Magazine / April 2019

Over the longer term, CMS should consider revising the regulatory requirements to support a care and payment delivery system based on patient characteristics rather than the site of care in order to achieve a greater degree of equity and comparability in costs. In doing so, however, CMS must not inadvertently discourage higher-intensity and specialized care or create access issues for medically complex patients. While the postacute needs of some patients may be addressed in less intensive settings such as SNFs and HHAs, the unified PAC PPS must recognize that there always will be higher acuity patients who require hospital-level care to address their constellation of medical needs.

5. What period of time should be considered in a unified PAC PPS? Stay? Episode? Period of days?

Are there different considerations for inpatient versus home health settings?

In the near term, CMS could consider an approach based on a per-stay or per-discharge (however defined) period of time. However, in the longer run, CMS should consider basing the unified PAC PPS on an episodic framework. This approach would include all the PAC services provided to the patient (with carve-outs for high-cost services unrelated to the reason for PAC, as discussed above) and focus on a comprehensive look at patient needs with a “treatment in place” model. If done correctly, this approach would enable providers and CMS to optimize efficiencies and minimize shifting care costs to the next site of care.

In the longer term, CMS could also consider paying a staggered per diem rate, recognizing that the highest costs of care frequently occur during the front part of a patient’s stay and the lower costs towards the end of the stay. As RTI/CMS are well-aware, the cost of care for the PAC patient population ranges extensively and, even among similar diagnostic groups, can be highly variable.

When considering the period of time, there are of course differences in institutional PAC settings versus HHA. Operating costs and covered services aside, many HHA patients are community admissions and have care needs, such as chronic disease management, that are different from HHA patients admitted from short-term acute care hospitals (STACHs) or even from institutional PAC settings. More than 85 percent of sequential HHA stays of three episodes or more are community admissions.12 In other words, multi-episode home health “episodes” are likely very different in nature from institutional PAC stays. With the HHA payment system changes adopted in the CY 2019 final rule, CMS will be able to differentiate community versus

institutional admissions to home health. CMS could then evaluate the impact and consequences of not including community HHA admissions in a unified post-acute prospective payment system.

6. Analyses looking at a unified PAC PPS will include all PAC settings (LTCH, IRF, SNF, and HHA). Is it also valuable to examine different combinations of settings?

As we have detailed above, there are significant differences in costs and patient case-mix across PAC settings. These differences are most notable between institutional PAC settings versus home health, and between hospital-based PAC versus non-hospital PAC. As such, AMRPA strongly recommends that CMS examine multiple combinations of settings to better equalize the asymmetries in setting-specific costs.

This approach was also used in the PAC PRD. The combinations should model:

• Institutional PAC versus HHA;

• Hospital PAC versus lower-level institutional care;

• Institutional PAC + HHA Institutional Admissions model (exclude HHA community admissions); and

• Hospital PAC + SNF + HHA Institutional Admissions.

7. Should the development of a unified PAC PPS assume today’s utilization and transfer patterns?

Or should the development of a unified PAC PPS assume the provision of multiple levels of care within a single PAC setting (i.e., treatment in place)?

In the near term, RTI/CMS would need to use today’s utilization and transfer data for development purposes. However, beneficiary costs based on current administrative data do not necessarily represent the optimal allocation of services based on patient characteristics. CMS would also need to account for how systemic payment changes, such as recent ones to the SNF, HHA and IRH/U PPSs, impact the continuity of administrative data.

CMS, under the authority granted to CMMI, should pilot test any unified PAC PPS prototype and especially a “treatment in place” model, as it represents such a significant shift from current practices before moving forward with any national prototype or proposal thereof.

The transfer, utilization, and cost data gathered from this exercise would greatly inform the development of a unified PAC PPS.

a. What are the pros and cons of each of these approaches for providers and beneficiaries?

There would be benefits for both providers and beneficiaries, given that the regulatory framework of a unified PAC PPS would allow providers to treat in place. There is an opportunity to realize cost savings due to efficiencies of being able to treat in place. For patients, there is tremendous benefit in more coordinated care and minimized transitions and handoffs to a next site of care. Furthermore, patients and providers would both benefit from not needing to wade through the labyrinthine coverage requirements currently in place when patients move through the PAC continuum. b. What are the considerations for provider certification and other regulatory requirements?

A PAC provider should demonstrate certain capabilities, characteristics, and physical and service attributes based on the clinical needs of their patients. More complex PAC patients would have higher-intensity needs and would require access to hospital-level provider settings. In the shorter term, it would be more feasible for higher-intensity settings to have the flexibility to offer step-down care, rather than expecting SNFs or HHAs to develop the physical and staffing infrastructure needed to sustain higher acuity patients safely.

In the longer term, any provider should be able to care for patients if they can demonstrate (via certification or other requirements) that they have the ability to address specific patients’ needs. As MedPAC has recommended, these provisions could be a set of requirements specific to certain clinical conditions.13

8. What assessments or claims variables should be considered in the PAC PPS development? [From RTI’s Sept 24, 2018 presentation]

The data must include, at a minimum, attributes such as diagnosis, motor and cognitive function on admission and discharge, severity, age, sex, race, co-morbidities, length of stay, medical information available on admission and discharge, status prior to acute care admission, home status, socioeconomic data, infection rates and impairments, discharge destination, and outcomes expectations.

Although the goal of this work is to achieve equity and comparability across sites, CMS must not focus myopically on examining only the patient clinical variables that are standardized or present across

Medicare Payment Advisory Commission, Data Book: Health Care Spending and the Medicare Program, 96 (June 2018).

Medicare Payment Advisory Commission, Report to Congress: Medicare and the Health Care Delivery System, 92-94 (June 2016).

12 13

AMRPA Magazine / April 2019 35

settings – if anything, it is just as, if not more important to also isolate and identify those variables that are unique to one setting because these data can represent patients who are being addressed in only that particular setting. For example, CMS may want to develop a claims-based or assessment-based indicator to identify post-organ transplant rehabilitation patients, as these patients require specialized ancillaries and additional staffing resources in IRH/Us, as described above.

This adjustment is necessary for the development of equitable risk adjustment and case-mix weighting methodologies for the unified PAC PPS. Equally critical, RTI/CMS need to account for the significant differences in volume within the PAC settings and the impact this factor may have on any unified PAC PPS case-mix weights generated by regression modeling. Given the disproportionate volumes between IRH/U and LTCH discharges versus SNF and HHA discharges (see Table D), RTI/CMS must ensure that the contributions of hospital-based PAC settings to the cost analysis are not overrun in any regression model.

Table D: Medicare FFS Utilization, 201614

IRH/U

LTCH

Spending

$7.7 billion

$5.1 billion

$29.1 billion

SNF

$18.1 billion

HHA

Beneficiary Volume

391,000 stays for 350,000 beneficiaries

126,000 stays for 111,000 beneficiaries

2.3 million stays for 1.6 million beneficiaries

6.5 million episodes for 3.4 million beneficiaries

This sensitivity is especially critical when the administrative data being used neither capture nor reflect the complete nuances of the patient’s clinical needs. Again, a stroke patient in an IRH/U is not likely to be the same stroke patient in a SNF in terms of their systemic costs and beneficiary-specific needs. One way CMS can account for these discrepancies is by ensuring that there are necessary risk-adjustment variables in the model to stratify, for example, an IRF orthopedic patient from an HHA orthopedic patient, or an LTCH ventilator patient from a SNF ventilator patient. Above all, risk adjustment in the unified PAC PPS must take into account all chronic conditions/comorbidities and be tested to ensure that the costs of treating the sickest patients with the most complex conditions are adequately accounted for.

9. What quality measures will be important to monitor to ensure quality of care for beneficiaries under a unified PAC PPS?

The quality measures should be relevant to the population of patients, cover a broad scope of medical status and functional ability, be reliable and valid as measures, and be operationally feasible to use.

14

In addition, if such measures become performance measurements, they should assure that patients are receiving the level of services and care their conditions require and that there is no stinting on care. Any quality measure adopted should directly drive better outcomes for the patients and the quality measure has sufficient room for improvement across the setting(s) over time, i.e., not be “topped out.” At a minimum, the measures should include:

• Functional outcomes measures:

o Functional improvement

o Functional maintenance

• Resource use and efficiency measures:

o Discharge to the community

o Preventable hospital readmissions

o Emergency department visits

• Patient safety and adverse events:

o Healthcare-associated infections

o Mortality

o Complications

• Measures of patient-centeredness of care

o Quality of life

o Days in the community, or a patient-centered “efficiency measure”

Efficiency measures Most efficiency measures used today assess “efficiency” in purely fiscal terms and from the payer’s perspective (e.g., Medicare Spending Per Beneficiary). However, CMS should consider the “efficiency of spending” in tandem with value from the patient’s perspective. If a PAC provider can successfully transition a patient back to their home or community in 10 days and help the patient stay longer in the home or community without adverse outcomes, then that provider should be considered more efficient than one that keeps the patient in the facility for 25 days (assuming total Medicare spending is equal). CMS should explore and adopt measures that address a beneficiary-centered – not just Medicare Trust Fund-centered – definition of efficiency.

Additional Considerations If, under a unified PAC PPS, payments are designed to be beneficiary-based, then it follows that quality measurement could be designed that way as well. This would be a structural change from Medicare’s current setting-specific PAC quality reporting programs (QRPs): the IRF QRP, LTCH QRP, SNF QRP, and HHA QRP. In terms of comparability, RTI/ CMS could consider using a measure set by clinical condition instead of by setting. For example, CMS could develop a uniform “PAC Stroke Quality Measure Set” so that, agnostic of the setting of care, the resulting patient outcomes would be judged against a common yardstick. This approach also has the potential to simplify both measure development and performance evaluation, since the inclusion criteria would limit the measure to a specific clinical subset and

Medicare Payment Advisory Commission, Report to Congress: Medicare Payment Policy (Mar 2018).

36 AMRPA Magazine / April 2019

thereby inherently risk adjust for case-mix, in some regard.

The quality measures should also assume adjustments for social risk factors such as social support at home or in the community, socioeconomic status, and language among others.

It would be important to account for the differences in patient goals in a unified PAC quality framework. Currently that is mitigated in some part by the different settings – IRH/ Us accept Medicare patients on the expectation that they would benefit and functionally improve whereas many SNF residents receive rehabilitation for functional maintenance.

If a unified PAC PPS does not contain segregations along setting-specific delineations, then “goals for post-acute care” should be factored into the quality framework in order to produce more appropriate performance comparability.

Furthermore, patient goals would be a valuable data point to collect in the longer term since functional improvement versus maintenance would correlate closely with anticipated resource use.

Despite the desirability of using measures of functional improvement, RTI/CMS must recognize that the functional measures currently used in the PAC QRPs are not sensitive at both the most severely impaired level and at the highest level of ability, e.g., floor and ceiling effects. Current measures also do not capture critically important benefits in the quality of life domain for certain extreme patient conditions, such as patients with tetraplegia or severe brain injury, among others.

11. A unified PAC PPS puts an emphasis on patient needs rather than setting of care. What are the design elements of a unified PAC PPS that will help empower patients and their families to work with providers to make care decisions?

CMS should better educate beneficiaries about PAC. One suggestion is to encourage earlier discussions in STACHs about discharge planning and educating the patient’s family or caregiver about post-discharge needs. Many families and caregivers would benefit from being more aware of PAC needs and potential options earlier in the continuum. CMS should also consider revising hospital discharge planning requirements to facilitate more transparent dialogue between hospitals and patients about PAC, such as by specifically recommending a high-quality provider. Nonetheless patients should retain freedom of choice, as discussed above.

To the extent that CMS would give hospitals the ability to steer patients and incorporate risk or accountability in the unified PAC PPS, it will need to balance patient preferences versus provider recommendations. This issue is a challenge today in Medicare’s bundled payment models or ACOs, where oftentimes patients do not adhere to provider recommendations but the provider is still “on the hook” for post-discharge spending and outcomes. ***

10. Can these quality measures be used to inform value-based payments in a unified PAC PPS?

Quality measures, when developed and implemented appropriately, could be used eventually to inform value-based design features of a unified PAC PPS. Once data are collected over a period of years, the use of performance measures would allow development of incentive payment methods to reward those institutions that achieve better risk-adjusted medical and functional outcomes, which probably require longer lengths of stay.

Other concepts should also be considered for the longer term. Should the unified PAC PPS evolve to be episodic in nature, CMS should consider monitoring spending and outcomes on beneficiaries who continue to receive Medicare services after the episode concludes. If spending spikes after the PAC PPS episode has ended, it may be an indication of inappropriate care stinting during the episode. CMS already includes such a post-episode monitoring period in the Comprehensive Care for Joint Replacement (CJR) bundled payment program.

IV. Suggestions for Future TEP Discussions AMRPA appreciates the opportunity to continue dialogue with RTI/CMS over the course of this project. We have several suggestions at this juncture for topics to address in future TEP meetings, as described in the comments above, as well as recommendations for the TEP process. With regard to process, it would be helpful to stakeholders if the questions posed to the TEP were more focused and narrow in scope. Specifically, RTI/CMS should offer more structured parameters when asking TEP participants to consider shorter-term and longer-term options by, for example, outlining what regulatory requirements could change in the shorter versus longer-term. There is an extensively wide range of possibilities when stakeholders are asked to consider options that are conceivable in either the current PAC silos in the near future, or in a unified PAC PPS in the longer run.

Topics that are worth discussing in future TEP meetings include:

• How to better capture routine cost data;

• How recent PAC payment system changes (especially SNF, HHA and, to a lesser degree, anticipated IRF PPS changes) will impact the unified PAC PPS development work:

AMRPA Magazine / April 2019 37

o Will the recent changes in the HHA and SNF PPSs affect the incoming data RTI/CMS will examine? For example, the HHA PPS will begin to pay differently for community versus institutional admissions – does CMS plan to bifurcate its analysis of HHA data based on admission source? Will future analysis include the new ICD-10-CM data being collected under the SNF Patient Driven Payment Model?

• How and when would CMS’ work with RAND on crosssetting standardized patient assessment data elements (SPADEs) integrate with this work with RTI? • The value of functional status outcomes in PAC and how CMS intends to incorporate functional status data in the unified PAC PPS; and • The feasibility of using quality measures developed and stewarded by industry stakeholders in the unified PAC PPS.

V. Conclusion AMRPA remains a strong supporter of the principles and objectives of the IMPACT Act and remains committed to working with CMS, RTI, and other governmental entities to ensure the Act is implemented to achieve these objectives.

As RTI and CMS continue to establish the foundational principles for a unified PAC PPS, AMRPA urges the agency to be comprehensive in its analysis of beneficiary-specific and setting-specific costs, and to account for the systemic cost differences across PAC settings as we have discussed above. Ultimately, the unified PAC PPS must be designed to be recognize the evolving and diverse health care needs of Medicare beneficiaries and do so by reimbursing providers adequately for addressing those needs.

38 AMRPA Magazine / April 2019

AMRPA thanks RTI and CMS for their efforts towards developing a unified PAC PPS and we look forward to continuing to work together. If you have any questions regarding our comments, please contact Carolyn Zollar, J.D., AMRPA Executive Vice President for Policy Development and Government Relations (czollar@amrpa.org / (202) 860-1002), or Mimi Zhang, AMRPA Senior Policy and Research Analyst (mzhang@amrpa.org / (202) 860-1003).

Sincerely,

Richard Kathrins, Ph.D. Chair, AMRPA Board of Directors President and CEO Bacharach Institute for Rehabilitation

Suzanne Kauserud, FACHE, MBA, PT Chair, AMRPA Unified PAC PPS Workgroup Chair, AMRPA Quality Committee Vice President, Carolinas Rehabilitation

CC: Susanne Seagrave, Ph.D, CMS Susan Bogasky, ASPE

PAC Market Analysis Reports Find out where your institution stands with a Market Analysis of Medicare Post-Acute Care (PAC) Referral Patterns, Episode Spending, Performance Measures and Impact of Medicare Bundled Payment Models

Using the most recent two years of Medicare claims data, Dobson DaVanzo & Associates delivers inpatient rehabilitation providers with a general market-level analysis on their facility’s episode spending and key performance metrics across all Medicare discharges. Benchmark your facility against state and national inpatient rehabilitation providers and find out where you stand.

Dobson DaVanzo & Associates can also help you better understand how the Bundled Payment for Care Improvement (BPCI) initiative and the Comprehensive Care for Joint Replacement Payment Model (CJR) are impacting the markets.

Stay informed! Order your PAC report today.

AMRPA Members Receive Reports at Discounted Rates.

Visit https://amrpa.org/PAC-Market-Analysis-Reports for more information, or contact Rachel Koresky, AMRPA Operations Manager, at rkoresky@amrpa.org.

OIG Estimates $84 Million in Improper Payments for SNF Services Due to Violations of Three-Day Rule

Jonathan M. Gold, JD, AMRPA Regulatory and Government Relations Counsel

Highlight: »»

Audit Covered Calendar Years 2013-2015

»»

OIG Also Identified Systemic Flaws In Ability Recoup Improper Payments

In February, the U.S. Department of Health and Human Services Office of Inspector General (OIG) released a report on an audit it conducted on Medicare payments for skilled nursing facility (SNF) claims for calendar years 2013-2015. The focus of this review was whether SNF claims complied with the requirement that a SNF beneficiary must have spent at least three consecutive days as an inpatient in a hospital (not including the day of discharge or time spent as an outpatient or under observation status) within 30-days preceding the SNF admission. This rule is commonly referred to as the “three-day rule.”

To focus on potential violations of the three-day rule, OIG excluded claims that were preceded by a hospital stay of more than two days (not including the discharge day), claims for which there is an exception to the rule (such as for certain alternative payment models), claims under investigation by fraud contractors, and claims for which the service dates overlapped with dates of the inpatient hospital stay. After excluding these claims, OIG identified more than 22,000 claims totaling more than $134 million in calendar years 2013-2015 that did not appear to have a corresponding qualifying hospital stay.

OIG then randomly sampled 99 of the 22,000 claims it had identified through its exclusion criteria and reviewed the medical records and other information for those claims. It determined that 65 of the 99 claims were improperly paid for failing to meet the requirements of the three-day rule. These 65 claims totaled more than $484,000, which OIG extrapolated to determine that approximately $84 million in SNF services were improper for failing to meet the three-day rule in 2013-2015.

For 18 of the 65 improper claims, OIG was able to determine that the hospital provided incomplete, misleading or erroneous discharge information to the SNF, which may have led to the SNF falsely believing that the patient was eligible for services. In other instances, the OIG found that SNFs often combined non-inpatient time spent in the hospital when asserting the patient was eligible for services, including time spent in emergency rooms and in observation status. For the 34 remaining SNF claims that did not initially show a qualifying hospital stay, OIG found several reasons the claims were nonetheless appropriate. OIG said that some of the reasons these claims were not improper were due to hospital error in reporting the number of inpatient days, or because the hospital claim was submitted to another payer other than Medicare.

OIG said it was unable to determine who was at fault for some of the improper claims it identified, which it notes presents serious road blocks to recouping improper payments. Medicare can only recover improper payments from a provider when a provider is determined to be at fault. In order to prove the provider is at fault, Medicare may need to show that the SNF knew or should have known the

40 AMRPA Magazine / April 2019

patient did not have a qualifying hospital stay. However, OIG notes there is not a coordinated notification mechanism to require hospitals to provide SNFs with accurate information about the dates of the inpatient stay. Therefore, it is difficult to prove the SNF knew or should have known the correct dates of the patient’s stay, and determine the SNF was at fault.

As a result of these findings, OIG made a number of recommendations to Medicare. First it recommended that an edit be included in the Medicare Common Working File (CWF) to identify claims submitted that do not have a corresponding qualifying hospital stay. OIG also recommends Medicare require hospitals to provide written notification to beneficiaries including their total inpatient hospitals days, and that the SNF obtain a copy of that notification upon admission to ensure the patient is eligible.

Medicare responded to OIG’s recommendations by concurring that an edit should be included in the CWF, and it had implemented one recently. However, Medicare did not agree that there should be a coordinated notification mechanism and noted that hospitals already provide notice to outpatients on observation status, and that the discharge procedures for hospitals are dictated in the Medicare Conditions of Participation.

JOIN TODAY!

EDUCATION , COMMUNICATION, PARTICIPATION & OPERATIONAL ASSISTANCE

AMRPA: Working Together Access to Medical Rehabilitation AMRPA: Working Together to To Preserve Preserve Access To Medical Rehabilitation Maggie RamirezAMRPA · VP ofMember Membership Services · 347-573-3732 · mramirez@amrpa.org Anna Kruskop, Services Associate, akruskop@amrpa.org, 202-207-1120.

AMRPA Magazine / April 2019 41

Issue Brief Examines Medicare Advantage Market Concentration and Potential Competition

Highlight: »»

70 percent or more of enrollees were in highly concentrated markets from 2009–2017.

»»

Existing evidence suggests that the primary drivers of consumer choices are differences in the premiums, quality of care and benefits among Medicare Advantage plans.

The Commonwealth Fund’s latest Issue Brief, Market Concentration and Potential Competition in Medicare Advantage, examines the market structure of the Medicare Advantage (MA) program from 2009–2017, including the insurers offering plans in each county and the level of enrollment by county and plan.

Medicare Advantage, the private option to traditional Medicare (or fee-for-service [FFS]), now serves roughly 37 percent of beneficiaries through health care plans. Recent data has shown that many MA markets are served by just one or a small number of insurers.

Market

// Medicare Advantage, the private option to traditional Medicare (or fee-for-service [FFS]), now serves roughly 37 percent of beneficiaries through health care plans.

In 2012, 97 percent of county markets in the MA program were designated as highly concentrated according to the Federal Trade Commission (FTC) and the U.S. Department of Justice (DOJ), with a Hirschman-Herfindahl Index (HHI) of greater than 2,500. In 2016, the Medicare Payment Advisory Commission (MedPAC) observed that local markets for MA plans were becoming increasingly concentrated. Recently, courts have blocked mergers that would further erode competition within the MA market.

Key Findings MA markets are highly concentrated and have become more concentrated since 2009.

From 2009–2017, 70 percent or more of MA enrollees were in highly concentrated markets, dominated by two or three insurers.

Since the payment system used to reimburse MA plans relies on competition to spur efficiency and premiums that more closely reflect insurers’ actual costs, these developments suggest that taxpayers and beneficiaries will overpay.

42 AMRPA Magazine / April 2019

Data Methods The authors measured both actual and potential competitors for each county for each year. Actual competitors are those insurers that participate in MA in a specific county; potential competitors are the insurers participating in MA in a state but not in the county of interest. The data allowed the authors to compute concentration ratios and the HHI for each county and in each year. In some analyses the counties were categorized according to the HHI corresponding to the FTC/DOJ classifications of concentration: 1) not concentrated, HHI <1,501; 2) moderately concentrated, HHI=1,501–2,500; and 3) highly concentrated, HHI >2,500.

Implications of MA Market Concentration Even though 37 percent of all Medicare beneficiaries are enrolled in private plans, when compared with employer-based health insurance Medicare’s transition to managed care has been slow. Traditional Medicare is the last major mainstay of open-network, fee-for-service health insurance, although the fee-for-service component is beginning to change with the spread of accountable care organizations (ACOs).

Several factors may have driven greater concentration in MA markets since 2009. First, consolidation in the health insurance industry generally may have affected the MA market structure.

Concentration in provider markets also has been increasing, which has made price negotiations for health care services more difficult for insurers, especially smaller ones. Medicare policy changes over these years may have inadvertently limited the supply and market entry of MA insurers. When Medicare rules were changed to require all MA plans to create networks of providers, the effect of provider concentration was heightened and some health insurers were less willing to remain in and/ or enter MA markets. This effect may have been especially significant in rural areas. At the same time, there appears to be a substantial number of potential MA insurer entrants in most moderate to highly concentrated markets, yet there appears to have been little clear impact on market outcomes in terms of premiums and quality.

Since the payment system used to reimburse insurers selling in the MA market relies on competition to drive premiums toward insurers’ actual costs, these developments suggest that taxpayers and beneficiaries will overpay for MA products, compared with what they might have paid in markets with more robust competition.

Results In 2017 Medicare beneficiaries could choose from a relatively large number of private plans (roughly seven) by the standards of the private insurance market. The number of insurers declined from 2009 to 2011 then remained steady through 2017, averaging 2.5 in 2017. For comparison, in

2017, the average metropolitan area had two insurers competing in the health insurance marketplaces created by the Affordable Care Act (ACA).

Insurer concentration increased from 2009 to 2011 (the number of insurers selling MA plans fell from 4.5 to 2.9) then remained at about the same, high level of concentration.

The two-firm concentration ratio was already high in 2009 (81 percent); it rose to 91 percent by 2011 and remained through 2017. The average county-level HHI was 4,914 in 2009, rising to 6,360 in 2013, and declining slightly to 6,285 in 2017.

Notably, the number of potential competitors also fell over the same period. Nevertheless there are now more potential than actual competitors in each county.

Virtually all Medicare enrollees face MA markets that are moderately to highly concentrated. Among sparsely populated markets, which are largely rural, the mean HHI is 6,684 — indicating that they are highly concentrated.

This is in part because of the difficulty that managed care plans, such as HMOs and PPOs, have in establishing provider networks in rural areas where providers are scarce and provider markets are highly concentrated. In highly populated markets, the average HHI shows that they too are highly concentrated HHI = 3,774), but the index value is considerably lower than in sparsely populated markets.

While Medicare beneficiaries have a choice between FFS and MA, in assessing the competitive forces on MA plans they assume that the actual or potential competition from other MA plans matters most. While the presence of FFS likely affects the conduct of MA plans, existing evidence suggests that the primary drivers of consumer choices are differences in the premiums, quality of care, and benefits among MA plans.

Conclusion The authors concluded that the effect of Medicare Advantage market power on prices or quality of care needs to be assessed empirically. There is some, but limited, evidence on the exercise of MA market power. However, the authors state that further research is needed to understand how potential competitors affect the actions of existing competitors. It also will be important to understand the barriers to market entry for potential competitors, especially those that might be lowered to spur greater competition.

For the complete report see, Market Concentration and Potential Competition in Medicare Advantage, Commonwealth Fund, Issue Brief, February 14, 2019.

AMRPA Magazine / April 2019 43

CMS Office of the Actuary Releases 2018-2027 National Health Expenditure Projections

Highlight: »»

The average annual spending growth in Medicare is expected to exceed that of Medicaid and private health insurance.

National health spending growth is expected to average 5.5 percent annually from 2018-2027, reaching nearly $6 trillion by 2027, according to the latest projections from the Office of the Actuary within the Centers for Medicare and Medicaid Services (CMS).

Health expenditures are expected to outpace growth in the Gross Domestic Product (GDP) by 0.8 percent over the same period. The health share of GDP is projected to rise from 17.9 percent in 2017 to 19.4 percent by 2027.

The outlook for national health spending and enrollment over the next decade is expected to be largely driven by:

// Health expenditures are expected to outpace growth in the Gross Domestic Product (GDP) by 0.8 percent over the same period.

Key economic factors, such as growth in income and employment, and demographic factors such as the baby-boom generation continuing to age from private insurance into Medicare; and Increases in prices for medical goods and services (projected to grow 2.5 percent over 2018-2027 compared to 1.1 percent during the period of 2014-2017).

Similar to the findings in last year’s report, the latest report found that by 2027, federal, state and local governments are projected to finance 47 percent of national health spending, an increase of 2 percentage points from 45 percent in 2017.

44 AMRPA Magazine / April 2019

Projected health insurance enrollment and national health expenditures by sector and payer Medicare

Medicare spending growth is projected to average 7.4 percent over 2018-2027, the fastest rate among the major payers.

Medicaid

Average annual growth of 5.5 percent is projected for Medicaid spending for 2018-2027. Medicaid spending growth is projected to average 6.0 percent for 2020 through 2027 as the programâ&#x20AC;&#x2122;s spending patterns reflect an enrollment mix of beneficiaries who are older and with disabilities.

Health Insurance Enrollment

Net enrollment gains across all sources are generally expected to keep pace with population growth with the insured share of the population going from 90.9 percent in 2017 to 89.7 percent in 2027.

Private Health Insurance and Out-of-Pocket

For 2018-2027, private health insurance spending growth is projected to average 4.8 percent, slowest among the major payers, which is partly due to slow enrollment growth related to the baby-boomers transitioning from private coverage into Medicare. Out-of-pocket expenditures are also projected to grow at an average rate of 4.8 percent over 2018-2027 and to represent 9.8 percent of total spending by 2027 (down from 10.5 percent in 2017).

Prescription Drugs

Spending growth for prescription drugs is projected to generally accelerate over 2018-2027 (and average 5.6 percent) mostly as a result of faster utilization growth. Underlying faster growth in the utilization of prescription drugs, particularly over 2020-2027, are a number of factors including efforts on the part of employers and insurers to encourage better medication adherence among those with chronic conditions, changing pharmacotherapy guidelines, faster projected private health insurance spending growth in lagged response to higher income growth, and an expected influx of new and expensive innovative drugs into the market towards the latter stage of the period.

Hospital

Hospital spending growth is projected to average 5.6 percent for 2018-2027. This includes a projected acceleration in 2019, to 5.1 percent from 4.4 percent in 2018, reflecting the net result of faster expected growth in both Medicare (higher payment updates) and Medicaid (as a result of expansion in five states), but slower projected growth in private health insurance as enrollment declines slightly due to the repeal of the individual mandate.

Physician and Clinical Services

Physician and clinical services spending is projected to grow an average of 5.4 percent per year over 2018-2027. This includes faster growth in prices over 2020-2027 for physician and clinical services due to anticipated rising wage growth related to increased demand from the aging population.

The Office of the Actuaryâ&#x20AC;&#x2122;s report is available on CMSâ&#x20AC;&#x2122;s website.

AMRPA Magazine / April 2019 45

A M E R I C A N M E D I C A L R E H A B I L I T AT I O N P R O V I D E R S A S S O C I AT I O N

Announcing The AMRPA QRP Certification Program A New Era for Rehabilitation has Begun!

Starting October 1, 2019, CMS is eliminating FIM from the IRF-PAI and introducing CMGs based on the GG-H QRP questions. AMRPA is working on key aspects of this change, from policy review and critique to helping hospitals adjust their day-to-day operations.

As part of this work, AMRPA is happy to announce the AMRPA QRP Certification Program.

Available in early 2019, this system will allow AMRPA member hospitals, as well as eRehabData® subscribers, to electronically test staff on their expertise of the QRP data elements in the IRF-PAI. The tests will be administered through AMRPA’s own eRehabData® system.

Best of all, AMRPA member hospitals and eRehabData® subscribers will have unlimited use of this powerful resource, free of charge.

Together, we can make the transition from FIM to GG-H smooth and seamless.

To learn more, please visit eRehabData.com

46 AMRPA Magazine / April 2019

NCHS Releases Long-term Care Provider and Services Trends in the United States

Highlight: »»

»»

In 2016, about 65,600 paid, regulated long-term care services providers in five major sectors served more than 8.3 million people in the United States.

The number of older people with significant physical or cognitive disabilities is projected to increase from 6.3 million in 2015 to 15.7 million in 2065.

A report issued by the by the National Center for Health Statistics (NCHS) released results from the National Study of Long-term Care Providers (NSLTCP) that describe providers and services users in five major sectors of paid, regulated long-term care services in the United States. Services users include residents of nursing homes and residential care communities, patients of home health agencies (HHAs) and hospices, and participants of adult day services centers.

The report provides information on the supply, organizational characteristics, staffing and services offered by providers; and the demographic, health and functional composition, and adverse events among users of these services.

Long-term Care Services Long-term care services include assistance with activities of daily living (ADLs; dressing, bathing and toileting), instrumental activities of daily living (IADLs; medication management and housework) and health maintenance tasks. Long-term care services assist people in improving or maintaining an optimal level of physical functioning and quality of life, and can include help from other people and special equipment or assistive devices. The need for long-term care services is generally defined based on functional limitations (need for assistance with or supervision in ADLs and IADLs) regardless of cause, age of the person, where the person is receiving assistance, whether the assistance is human or mechanical, and whether the assistance is paid or unpaid.

Individuals may receive long-term care services in a variety of settings: In the community, such as at an adult day services center; In the home, for example from a home health agency, hospice or family and friends; In institutions, such as in a nursing home or skilled nursing facility; or In other residential settings, for instance, in an assisted living or similar residential care community.

Although people of all ages may need long-term care services, the risk of needing these services increases with age. The number of Americans over age 65 is projected to shift from 47.8 million in 2015 to over 87.9 million in 2050, representing an increase of 84 percent and comprising 22 percent of the population. The population aged 85 and over is projected to triple, from 6.3 million in 2015 to over 18.9 million in 2050, and will account for almost 5 percent of the U.S. population. This “oldest old” population tends to have the highest disability rate and highest need for long-term care services, and is also more

AMRPA Magazine / April 2019 47

likely to be widowed and without someone to provide assistance with daily activities. The number of older persons in the United States with significant physical or cognitive disabilities is projected to increase from 6.3 million in 2015 to 15.7 million in 2065.

The projections estimate that the number of older adults using paid long-term care services will grow considerably in the coming years, and a substantial share of paid long-term care services is publicly funded through programs such as Medicaid.

Methods Data sources include NCHS surveys of adult day services centers and residential care communities and administrative records from the Centers for Medicare and Medicaid Services (CMS) on home health agencies, hospices and nursing homes (covers 2015 and 2016 data years).

Key Findings In 2016, about 65,600 long-term care services providers in five major sectors served more than 8.3 million people in the United States.

Long-term care services were provided by 4,600 adult day services centers, 12,200 home health agencies, 4,300 hospices, 15,600 nursing homes and 28,900 assisted living and similar residential care communities.

In 2016, there were an estimated 286,300 current participants enrolled in adult day services centers, 1,347,600 current residents in nursing homes and 811,500 current residents living in residential care communities.

In 2015, about 4,455,700 patients were discharged from home health agencies, and 1,426,000 patients received services from hospices.

Sectors differed in ownership and chain status, and supply varied by region.

The majority of home health agencies, hospices, nursing homes and residential care communities were for-profit, while a minority of adult day services centers were for-profit. Similarly, the majority of nursing homes and residential care communities and a minority of adult day services centers were chain-affiliated.

The supply of residential care beds per 1,000 people aged 65 and over was higher in the Midwest and West than in the Northeast and the South, and the capacity of adult day services centers was higher in the West than in the other regions.

Almost 1.5 million nursing employee full-time equivalents (FTEs)—including registered nurses (RNs), licensed practical or vocational nurses (LPNs or LVNs), and aides—and about 35,000 social work employee FTEs worked in the five sectors.

The relative distribution of nursing and social work employee FTEs varied across sectors: the most common employee FTEs were aides in adult day services centers, nursing homes and residential care communities while RNs were the most common employee FTEs in home health agencies and hospices.

48 AMRPA Magazine / April 2019

Sectors differed in their average staffing levels for nursing, social work and activities employees.

Among the three sectors where nursing staff levels (RNs, LPNs or LVNs, and aides) could be examined, the average total nursing staff hours per resident or participant day was higher in nursing homes than in residential care communities and adult day services centers.

In contrast, the average social work staffing level was higher in adult day services centers than in nursing homes or residential care communities, and the average activities staffing level in adult day services centers was more than twice that of nursing homes and residential care communities.

Daily-use rates among individuals aged 65 and over per 1,000 people aged 65 and over varied by sector.

The highest daily-use rate was for nursing home residents, followed by residential care residents, and the lowest daily use rate was for adult day services center participants.

Long-term care services users varied by sector in their demographic and health characteristics and functional status.

Adult day services center participants tended to be younger and more racially and ethnically diverse than services users in other sectors. At least one-quarter of services users in each of the five sectors had Alzheimer’s disease or other dementias, arthritis, heart disease or hypertension. However, the prevalence of these and six other reported diagnosed chronic conditions varied widely between sectors.

Fewer adult day services center participants needed assistance with four of six activities of daily living (ADLs; bathing, dressing, toileting and walking or locomotion) than services users in other sectors.

Adverse events among long-term care services users varied by sector.

Compared with adult day participants and residential care residents, more home health patients had overnight hospital stays and emergency department visits.

More residential care residents had falls compared with adult day participants and nursing home residents.

Short-stay (less than 100 days) residents differed from long-stay (100 days or more) residents by age and sex, and in the prevalence of numerous diagnosed conditions, overnight hospital stays, and falls.

For the complete report, see Long-term Care Providers and Services Users in the United States, 2015–2016, Analytical and Epidemiological Studies, National Center for Health Statistics, Vital and Health Statistics, Series 3, Number 43, February 2019.

Reductions in Post-Acute Care Spending Under BPCI Were Driven by Shift from Higher-to LowerCost Discharge Settings

Highlight: »»

»»

Overall institutional PAC declined 4.4 percent in BPCI hospitals vs 2.1 percent in non-BPCI hospitals.

Hospitals that were more successful in reducing Medicare payments started with higher payments and higher use of institutional PAC settings.

A recent study published in the Journal of the American Geriatrics Society characterized patterns of change in post-acute care (PAC) under lower-extremity joint replacement (LEJR) episodes in Medicare's Bundled Payments for Care Improvement (BPCI) program in 2013 to 2015.

PAC use has declined significantly in alternative payment models such as bundles or accountable care organizations (ACOs) but the study authors say that little is known about whether this reduction is due to provider selection or consolidation, and whether particular patterns of changes in PAC are associated with success under the program.

The study included 264 hospitals participating in BPCI for LEJR and matched controls.

The measurements were use and duration of institutional PAC (proportion discharged to a skilled nursing facility (SNF), an inpatient rehabilitation facility (IRF), and a long-term care hospital (LTCH), dispersion of PAC (proportion of discharges to commonly used providers), and quality of PAC (Star Ratings, readmission rates, length of stay and nurse staffing); and part A Medicare payments.

The study found that BPCI participants decreased the use and duration of institutional PAC compared to controls: overall institutional PAC declined 4.4 percent in BPCI hospitals vs 2.1 percent in non-BPCI hospitals (difference = -2.2 percent; P = .033), and duration decreased by 1.6 days in BPCI hospitals compared to 0.0 days in non-BPCI hospitals (difference in differences = -1.5 days; P < .001). However, BPCI participants did not change their PAC referral patterns to reduce dispersion or refer patients to higherquality PAC providers.

The authors concluded that BPCI hospitals that were more successful in reducing Medicare payments started with higher payments and higher use of institutional PAC settings and demonstrated greater drops in use and duration of institutional PAC. However, but there were no changes in dispersion or referral to high-quality PAC providers.

For the study see, Post-Acute Care After Joint Replacement in Medicare's Bundled Payments for Care Improvement Initiative, Journal of the American Geriatrics Society, February 25, 2019.

AMRPA Magazine / April 2019 49

Compliance with CMS Three-Hour Rule Alone Does Not Have a Substantive Impact on TBI Outcomes, Study Finds A study published in the Archives of Physical Medicine and Rehabilitation (AAPM&R) sought to determine if patients' level of effort (LOE) in therapy sessions during traumatic brain injury (TBI) rehabilitation modifies the effect of compliance with the Centers for Medicare and Medicaid Servicesâ&#x20AC;&#x2122; Three-Hour Rule for inpatient rehabilitation hospitals and units (IRH/Us).

The study participants included 1,820 patients who received their first IRH/U admission for TBI in the United States and were enrolled for three- and ninemonth follow-up. The main outcome measures were Participation Assessment with Recombined Tools-Objective-17, FIMâ&#x201E;˘ Motor and Cognitive scores, Satisfaction with Life Scale and Patient Health Questionnaire-9.

The study found that when the full cohort was examined, no strong main effect of compliance with the Three-Hour Rule was identified and LOE did not modify the effect of compliance with the Three-Hour Rule. In contrast, LOE had a strong positive main effect on all outcomes, except depression. When the sample was stratified by level of disability, LOE modified the effect of compliance, particularly on the outcomes of participants with less severe disability. For these patients, receiving three hours of therapy for 50 percent+ of therapy days in the context of low effort resulted in poorer performance on select outcome measures at discharge and up to nine months post discharge compared to patients with <50 percent of three-hour therapy days.

The authors concluded that while LOE is an active factor in inpatient TBI rehabilitation, compliance with the Three-Hour Rule was not found to have a substantive impact on the outcomes. The results support matching time in therapy during acute TBI rehabilitation to patients' LOE in order to optimize long-term benefits on outcomes.

For the study abstract, see Impact of Level of Effort on the Effects of Compliance with the Three-Hour Rule, Archives of Physical Medicine and Rehabilitation, February 20, 2019.

50 AMRPA Magazine / April 2019

Latest Research Findings

Hospital Ownership of Post-Acute Care Providers and the Cost of Care A recent study published in Health Care Management Review examined the association between the episodic costs of care and hospital ownership of post-acute care (PAC) providers, including skilled nursing facilities (SNFs), home health agencies (HHAs) and inpatient rehabilitation hospitals and units (IRH/Us).

Hospitals are facing incentives to manage the total cost of care for episodes of illness, including the costs of inpatient care and the cost of care provided by physicians and PAC providers. PAC is an especially important component of the overall cost of care.

One strategy hospitals are employing to manage this cost is to own PAC providers. Prior work on the relationship between PAC ownership and cost has reached mixed conclusions.

a voluntary shared saving program (Bundled Payment for Care Improvement). The effects of PAC ownership vary by the type of PAC provider owned. According to the study authors, the results suggest that ownership of SNFs and HHAs may be a viable strategy for success in reimbursement programs that reward hospitals for reducing the total costs for episodes of care.

For the study abstract see, Hospital ownership of postacute care providers and the cost of care, Health Care Management Review, February 21, 2019.

The authors examined panel data on hospital ownership of PAC providers from the American Hospital Association for 2013-2015 and cost of care data from the Centers for Medicare & Medicaid Services' Value-Based Purchasing Program (VBPP). Using ordinary least squares, the study quantified the association between a hospital's PAC ownership choice (both ownership of any PAC provider and ownership of particular types of providers) and the episodic cost of care.

The study found that in 2015 80 percent of hospitals owned some type of PAC provider and ownership of SNFs and HHAs is associated with a lower episodic cost of care, whereas ownership of IRFs is associated with higher episodic costs of care. The effects of ownership do not differ for hospitals that participate in

AMRPA Magazine / April 2019 51

Social Engagement, Mild Cognitive Impairment and Falls Among Older Primary Care Patients A recent study published in the Archives of Physical Medicine and Rehabilitation examined the association between mild cognitive impairment (MCI) and falls among primary care patients, and investigated whether social engagement (SE) modifies these associations.

The study participants included 430 community-dwelling older adults at risk of mobility decline with a mean age of 76.6 (range 65 to 96). The main outcome measure was the number of falls in the past year reported at the baseline interview. MCI was identified using a cutoff of 1.5 SD below the age-adjusted mean on at least two of the standardized cognitive performance tests. SE (e.g., keeping in touch with friends and family, volunteering, and participating social activities, etc.) was assessed with the Late Life Function and Disability Instrument, and required a score above the median value 49.5 out of 100.

While MCI is associated with a greater risk for falls, higher levels of SE may play a protective role.

For the study abstract see, The Association Between Social Engagement, Mild Cognitive Impairment and Falls Among Older Primary Care Patients, Archives of Physical Medicine and Rehabilitation, February 27, 2019.

The study found that MCI was present among 42 percent of participants and 42 percent reported at least one fall in the preceding year. Using Generalized Estimating Equations, MCI was associated with a 77 percent greater rate of falls (p<0.05). There was a statistically significant interaction between SE and MCI on the rate of falls (p<0.01), such that at a high level of SE, MCI was not statistically associated with falls (p=0.83). In participants with lower levels of SE, MCI is associated with 1.3 greater rate of falls (p<0.01).

Post-Discharge Decolonization Helps Reduce Infection Risk Among MRSA Carriers Hospitalized patients who are colonized with methicillin-resistant Staphylococcus aureus (MRSA) are at high risk for infection after discharge, according to a recent study in the New England Journal of Medicine.

The authors conducted a multicenter, randomized, controlled trial of post-discharge hygiene education, as compared with education plus decolonization, in patients colonized with MRSA (carriers).

Decolonization involved chlorhexidine mouthwash, baths or showers with chlorhexidine, and nasal mupirocin for five days twice per month for six months. Participants were followed for one year.

The primary outcome was MRSA infection as defined according 52 AMRPA Magazine / April 2019

to Centers for Disease Control and Prevention (CDC) criteria.

Secondary outcomes included MRSA infection determined on the basis of clinical judgment, infection from any cause, and infectionrelated hospitalization. All analyses were performed with the use of proportional-hazards models in the per-protocol population (all participants who underwent randomization, met the inclusion criteria, and survived beyond the recruitment hospitalization) and as-treated population (participants stratified according to adherence).

The study found that in the per-protocol population, MRSA infection occurred in 98 of 1,063 participants (9.2 percent) in the education group and in 67 of 1,058 (6.3 percent) in the

decolonization group; 84.8 percent of the MRSA infections led to hospitalization. Infection from any cause occurred in 23.7 percent of the participants in the education group and 19.6 percent of those in the decolonization group; 85.8 percent of the infections led to hospitalization. The hazard of MRSA infection was significantly lower in the decolonization group than in the education group (hazard ratio, 0.70; 95 percent confidence interval [CI], 0.52 to 0.96; P=0.03; number needed to treat to prevent one infection, 30; 95 percent CI, 18 to 230); this lower hazard led to a lower risk of hospitalization due to MRSA infection (hazard ratio, 0.71; 95 percent CI, 0.51 to 0.99).

The authors concluded that post-discharge MRSA decolonization with chlorhexidine and mupirocin led to a 30 percent lower risk of MRSA infection than education alone.

For the study see, Decolonization to Reduce Post-discharge Infection Risk Among MRSA Carriers, New England Journal of Medicine, February, 2019.

The decolonization group had lower likelihoods of clinically judged infection from any cause (hazard ratio, 0.83; 95 percent CI, 0.70 to 0.99) and infection-related hospitalization (hazard ratio, 0.76; 95 percent CI, 0.62 to 0.93); treatment effects for secondary outcomes should be interpreted with caution owing to a lack of pre-specified adjustment for multiple comparisons. In as-treated analyses, participants in the decolonization group who adhered fully to the regimen had 44 percent fewer MRSA infections than the education group (hazard ratio, 0.56; 95 percent CI, 0.36 to 0.86) and had 40 percent fewer infections from any cause (hazard ratio, 0.60; 95 percent CI, 0.46 to 0.78). Side effects (all mild) occurred in 4.2 percent of the participants.

AMRPA Magazine / April 2019 53

REGIONAL MEETING SERIES

Now Announcing AMRPAâ&#x20AC;&#x2122;s Regional Meeting Series!

Registration now open!

For more information, visit the AMRPA website: amrpa.org/Education/Events/2019-Regional-Meeting-Series

ADVERTISING RATES:

ADVERTISING CONTACT:

Full page - $1500 Half page - $1000 Third page - $750 Ads may be B&W or full color.

Julia Scott Phone: +1-202-207-1110 Email: jscott@amrpa.org

ADVERTISE IN OUR MAGAZINE

DISCOUNTS FOR MEMBERS

AMRPA & eRehabDataÂŽ Have Your Back

We adapt to regulatory updates to provide real time outcomes reports to inpatient rehabilitation hospitals and units.

Visit eRehabData.com to learn more, or contact Sam Fleming at sam@erehabdata.com to receive a free demo. Without losing any historical data, our staff help you migrate to the only patient assessment system that is trusted, owned, used, and supported by the medical rehabilitation industry.

56 AMRPA Magazine / April 2019