CMS Comprehensive Care for Joint Replacement Payment Model.pdf

8/20/2019 CMS Comprehensive Care for Joint Replacement Payment Model.pdf

1/120

Vol. 80 Tuesday,No. 134 July 14, 2015

Part III

Department of Health and Human Services

Centers for Medicare & Medicaid Services

42 CFR Part 510Medicare Program; Comprehensive Care for Joint Replacement PaymentModel for Acute Care Hospitals Furnishing Lower Extremity JointReplacement Services; Proposed Rule

VerDate Sep2014 21:01 Jul 13, 2015 Jkt 235001 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\14JYP3.SGM 14JYP3


2/120

41198 Federal Register / Vol. 80, No. 134 / Tuesday, July 14, 2015/ Proposed Rules

DEPARTMENT OF HEALTH ANDHUMAN SERVICES

Centers for Medicare & MedicaidServices

42 CFR Part 510

[CMS–5516–P]

RIN 0938–AS64

Medicare Program; ComprehensiveCare for Joint Replacement PaymentModel for Acute Care HospitalsFurnishing Lower Extremity JointReplacement Services

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

ACTION: Proposed rule.

SUMMARY: This proposed rule proposesto implement a new Medicare Part Aand B payment model under section1115A of the Social Security Act, called

the Comprehensive Care for JointReplacement (CCJR) model, in whichacute care hospitals in certain selectedgeographic areas will receiveretrospective bundled payments forepisodes of care for lower extremityjoint replacement or reattachment of alower extremity. All related care within90 days of hospital discharge from thejoint replacement procedures will beincluded in the episode of care. We

believe this model will further our goalsin improving the efficiency and qualityof care for Medicare beneficiaries forthese common medical procedures.

DATES: Comment period: To be assuredconsideration, comments on thisproposed rule must be received at oneof the addresses provided in theADDRESSES section no later than 5 p.m.EDT on September 8, 2015.

ADDRESSES: In commenting, please referto file code CMS–5516–P. Because ofstaff and resource limitations, we cannotaccept comments by facsimile (FAX)transmission.

You may submit comments in one offour ways (no duplicates, please):

1. Electronically. You may (and weencourage you to) submit electronic

comments on this regulation to http:// www.regulations.gov. Follow theinstructions under the ‘‘submit acomment’’ tab.

2. By regular mail. You may mailwritten comments to the followingaddress ONLY: Centers for Medicare &Medicaid Services, Department ofHealth and Human Services, Attention:CMS–5516–P, P.O. Box 8013, Baltimore,MD 21244–1850.

Please allow sufficient time for mailedcomments to be received before theclose of the comment period.

3. By express or overnight mail. Youmay send written comments via expressor overnight mail to the followingaddress ONLY: Centers for Medicare &Medicaid Services, Department ofHealth and Human Services, Attention:CMS–5516–P, Mail Stop C4–26–05,7500 Security Boulevard, Baltimore, MD21244–1850.

4. By hand or courier. If you prefer,you may deliver (by hand or courier)your written comments before the closeof the comment period to either of thefollowing addresses:

a. For delivery in Washington, DC—

Centers for Medicare & MedicaidServices, Department of Health andHuman Services, Room 445–G, HubertH. Humphrey Building, 200Independence Avenue SW.,Washington, DC 20201.

(Because access to the interior of theHubert H. Humphrey Building is notreadily available to persons without

Federal Government identification,commenters are encouraged to leavetheir comments in the CMS drop slotslocated in the main lobby of the

building. A stamp-in clock is availablefor persons wishing to retain a proof offiling by stamping in and retaining anextra copy of the comments being filed.)

b. For delivery in Baltimore, MD—

Centers for Medicare & MedicaidServices, Department of Health andHuman Services, 7500 SecurityBoulevard, Baltimore, MD 21244–1850.

If you intend to deliver yourcomments to the Baltimore address,please call the telephone number (410)786–7195 in advance to schedule yourarrival with one of our staff members.

Comments mailed to the addressesindicated as appropriate for hand orcourier delivery may be delayed andreceived after the comment period.

For information on viewing publiccomments, we refer readers to the

beginning of the SUPPLEMENTARY INFORMATION section.

FOR FURTHER INFORMATION CONTACT:Claire Schreiber, Claire.Schreiber@

cms.hhs.gov, 410–786–8939Gabriel Scott, Gabriel.Scott@

cms.hhs.gov, 410–786–3928

SUPPLEMENTARY INFORMATION:Inspection of Public Comments: All

comments received before the close ofthe comment period are available forviewing by the public, including anypersonally identifiable or confidential

business information that is included ina comment. We post all commentsreceived before the close of thecomment period on the following Web

site as soon as possible after they have been received: http:// www.regulations.gov. Follow the searchinstructions on that Web site to viewpublic comments.

Comments received timely will also be available for public inspection,generally beginning approximately 3weeks after publication of the rule, at

the headquarters of the Centers forMedicare & Medicaid Services, 7500Security Boulevard, Baltimore, MD21244, on Monday through Friday ofeach week from 8:30 a.m. to 4:00 p.m.EDT. To schedule an appointment toview public comments, phone 1–800–743–3951.

Electronic Access

This Federal Register document isalso available from the Federal Registeronline database through Federal DigitalSystem (FDsys), a service of the U.S.Government Printing Office. This

database can be accessed via theinternet at http://www.gpo.gov/fdsys/.

Alphabetical List of Acronyms

Because of the many terms to whichwe refer by acronym, abbreviation, orshort form in this proposed rule, we arelisting the acronyms, abbreviations andshort forms used and theircorresponding terms in alphabeticalorder.

mSA Micropolitan Statistical AreaACO Accountable Care OrganizationASPE Assistant Secretary for Planning and

EvaluationBPCI Bundled Payments for Care

ImprovementCBSA Core-Based Statistical AreaCMS Centers for Medicare & Medicaid

ServicesCPT Current Procedural TerminologyCCJR Comprehensive Care for Joint

ReplacementCSA Combined Statistical AreaDME Durable Medical EquipmentFFS Fee-for-serviceHCAHPS Hospital Consumer Assessment of

Healthcare Providers and SystemsHHA Home health agencyHOPD Hospital outpatient departmentHHPPS Home Health Prospective Payment

SystemHIQR Hospital Inpatient Quality Reporting

HRRP Hospital Readmissions ReductionsProgram

HRR Hospital Referral RegionHVBP Hospital Value Based Purchasing

ProgramICD–9–CM International Classification of

Diseases, 9th Revision, ClinicalModification

IPPS Inpatient Prospective Payment SystemIPF Inpatient psychiatric facilityIRF Inpatient rehabilitation facilityLEJR Lower extremity joint replacementLOS Length of stayLTCH Long term care hospitalLUPA Low Utilization Payment Adjustment


http://www.regulations.gov/http://www.regulations.gov/mailto:[email protected]:[email protected]:[email protected]:[email protected]://www.regulations.gov/http://www.regulations.gov/http://www.gpo.gov/fdsys/http://www.gpo.gov/fdsys/mailto:[email protected]:[email protected]://www.regulations.gov/http://www.regulations.gov/http://www.regulations.gov/http://www.regulations.gov/mailto:[email protected]:[email protected]


3/120

41199Federal Register / Vol. 80, No. 134 / Tuesday, July 14, 2015/ Proposed Rules

MAC Medicare Administrative ContractorMCC Major complications or comorbiditiesMSA Metropolitan Statistical AreaMS–DRG Medical Severity Diagnosis-

Related GroupMP MalpracticeNPP Nonphysician PractitionerNPRA Net Payment Reconciliation AmountOPPS Outpatient Prospective Payment

System

PAC Post-acute careSNF Skilled nursing facilityTHA Total hip arthroplastyTKA Total knee arthroplasty

Table of Contents

I. Executive SummaryA. PurposeB. Summary of the Major Provisions1. Model Overview: LEJR Episodes of Care2. Model Scope3. Payment4. Similar Previous and Concurrent Models5. Overlap With Ongoing CMS Efforts6. Quality Measures and Reporting

Requirements7. Data Sharing Process

8. Beneficiary ProtectionsC. Summary of Economic EffectsII. BackgroundIII. Provisions of the Proposed Rule

A. Proposed Definition of the EpisodeInitiator and Selected Geographic Areas

1. Background2. Proposed Definition of Episode Initiator3. Financial Responsibility for the Episode

of Care4. Proposed Geographic Unit of Selection

and Exclusion of Selected Hospitalsa. Overview and Options for Geographic

Area Selection b. MSA Selection Methodology(1) Exclusion of Certain MSAs(2) Proposed Selection Strata(a) MSA Average Wage-adjusted Historic

LEJR Episode Payments(b) MSA Population Size(c) Analysis of Strata(3) Factors Considered but Not Used in

Creating Proposed Strata(4) Sample Size Calculations and the

Number of Selected MSAs(5) Method of Selecting MSAsB. Episode Definition for the

Comprehensive Care for JointReplacement (CCJR) Model

1. Background2. Clinical Dimension of Episodes of Carea. Definition of the Clinical Conditions

Included in the Episode b. Definition of Related Services Included

in the Episode

3. Duration of Episodes of Carea. Beginning the Episode and Beneficiary

Care Inclusion Criteria b. Middle of the Episodec. End of the EpisodeC. Proposed Methodology for Setting

Episode Prices and Paying ModelParticipants Under the CCJR Model

1. Background2. Performance Years, Retrospective

Episode Payment, and Two-Sided RiskModel

a. Performance Period b. Proposed Retrospective Payment

Methodology

c. Proposed Two-Sided Risk Model3. Adjustments to Payments Included in

Episodea. Proposed Treatment of Special Payment

Provisions Under Existing MedicarePayment Systems

b. Proposed Treatment of Payment forServices That Extend Beyond theEpisode

c. Proposed Pricing Adjustment for High

Payment Episodes4. Proposed Episode Price SettingMethodology

a. Overview b. Proposed Pricing Features(1) Different Target Prices for Episodes

Anchored by MS–DRG 469 vs. MS–DRG470

(2) Three Years of Historical Data(3) Proposed Trending of Historical Data to

the Most Recent Year of the Three(4) Update Historical Episode Payments for

Ongoing Payment System Updates(a) Proposed Inpatient Acute Services

Update Factor(b) Proposed Physician Services Update

Factor(c) Proposed IRF Services Update Factor(d) Proposed SNF Services Update Factor(e) Proposed HHA Services Update Factor(f) Proposed Other Services Update Factor(5) Blend Hospital-Specific and Regional

Historical Data(6) Define Regions as U.S. Census Divisions(7) Normalize for Provider-Specific Wage

Adjustment Variations(8) Proposed Combination of CCJR

Episodes Anchored by MS–DRGs 469and 470

(9) Discount Factorc. Proposed Approach To Combine Pricing

Features5. Proposed Use of Quality Performance in

the Payment Methodologya. Background

b. Proposed Implementation of QualityMeasures for Reconciliation PaymentEligibility

(1) General Selection of Proposed QualityMeasures

(2) Proposal To Adjust the PaymentMethodology for Voluntary Submissionof Data for Patient-Reported OutcomeMeasure

(3) Measure Risk-Adjustment andCalculations

(4) Applicable Time Period(5) Criteria for Applicable Hospitals and

Performance Scoring(a) Identification of Applicable Hospitals

for the CCJR Model(b) Methodology to Determine Performance

on the Quality Measures(c) Proposed Methodology To Link Quality

and Payment(i) Background(ii) Alternatives Considered To Link

Quality and Payment(iii) Proposal To Link Quality and Payment

through Thresholds for ReconciliationPayment Eligibility

6. Proposed Process for Reconciliationa. Net Payment Reconciliation Amount

b. Payment Reconciliation7. Proposed Adjustments for Overlaps With

Other Innovation Center Models andCMS Programs

a. Overview b. CCJR Beneficiary Overlap With BPCI

Episodesc. Accounting for CCJR Reconciliation

Payments and Recoupments in OtherModels and Programs

d. Accounting for Per Beneficiary PerMonth (PBPM) Payments in the EpisodeDefinition

e. Accounting for Overlap With Shared

Savings Programs and Total Cost of CareModels

8. Proposals To Limit or Adjust HospitalFinancial Responsibility

a. Overview b. Proposed Limit on Raw NPRA

Contribution to Repayment Amounts andReconciliation Payments

(1) Proposed Limit on Raw NPRAContribution to Repayment Amounts

(2) Proposed Limit on Raw NPRAContribution to Reconciliation Payments

c. Proposed Policies for Certain Hospitalsto Further Limit RepaymentResponsibility

d. Proposed Hospital Responsibility forIncreased Post-Episode Payments

9. Proposed Appeal Procedures forReconciliationa. Payment Processes

b. Calculation Errorc. Dispute Resolution(1) Limitations on Review(2) Matters Subject to Dispute Resolution.(3) Dispute Resolution Process.10. Proposed Financial Arrangements,

Beneficiary Incentives, and ProposedProgram Rule Waivers and Amendments

a. Financial Arrangements and BeneficiaryIncentives

(1) Financial Arrangements PermittedUnder the CCJR Model

(a) CCJR Sharing ArrangementRequirements.

(b) Participation AgreementsRequirements.

(c) Gainsharing Payment and AlignmentPayment Conditions and Restrictions.

(d) Documentation and Maintenance ofRecords

(2) Beneficiary Incentives Permitted Underthe CCJR Model

(3) Compliance with Fraud and AbuseLaws

11. Proposed Waivers of Medicare ProgramRules

a. Overview b. Post-Discharge Home Visitsc. Billing and Payment for Telehealth

Servicesd. SNF 3-Day Rule

e. Waivers of Medicare Program Rules ToAllow Reconciliation Payment orRecoupment Actions Resulting From theNet Payment Reconciliation Amount12.Proposed Enforcement Mechanisms

D. Quality Measures and Display of QualityMetrics Used in the CCJR Model

1. Backgrounda. Purpose of Quality Measures in the CCJR

Model b. Public Display of Quality Measures in

the CCJR Model2. Proposed Quality Measures for

Performance Year 1 (CY 2016) andSubsequent Years



4/120


1 In this proposed rule, we use the term LEJR torefer to all procedures within the MedicareSeverity-Diagnosis Related Groups (MS–DRGs) wepropose to select for the model, includingreattachment of a lower extremity, as described insection III.B. of this proposed rule.

2Sylvia Mathews Burwell, HHS Secretary,Progress Towards Achieving Better Care, SmarterSpending, Healthier People, http://www.hhs.gov/ blog/2015/01/26/progress-towards-better-care-smarter-spending-healthier-people.html (Jan 26,2015).

a. Hospital-Level Risk-StandardizedComplication Rate (RSCR) FollowingElective Primary Total Hip Arthroplasty(THA) and/or Total Knee Arthroplasty(TKA) (NQF #1550)

(1) Background(2) Data Sources(3) Cohort(4) Inclusion and Exclusion Criteria(5) Risk-Adjustment

(6) Calculating the Risk-StandardizedComplication Rate and PerformancePeriod

b. Hospital-Level 30-day, All-Cause Risk-Standardized Readmission Rate (RSRR)Following Elective Primary Total HipArthroplasty (THA) and/or Total KneeArthroplasty (TKA) (NQF #1551)

(1) Background(2) Data Sources(3) Cohort(4) Inclusion and Exclusion Criteria(5) Risk-Adjustment(6) Calculating the Risk-Standardized

Readmission Rate and PerformancePeriod

c. Hospital Consumer Assessment of

Healthcare Providers and Systems(HCAHPS) Survey

(1) Background(2) Data Sources(3) Cohort(4) Inclusion and Exclusion Criteria(5) Case-Mix-Adjustment(6) HCAHPS Scoring(7) Performance periodd. Applicable Time Period3. Possible New Outcomes for Future

Measuresa. Hospital-Level Performance Measure(s)

of Patient-Reported Outcomes FollowingElective Primary Total Hip and/or TotalKnee Arthroplasty

(1) Background

(2) Data Sources(3) Cohort(4) Inclusion and Exclusion Criteria(5) Outcome(6) Risk-Adjustment (if Applicable)(7) Calculating the Risk-Standardized Rate(8) Performance Period(9) Requirements for Successful

Submission of THA/TKA Voluntary Data b. Measure that Captures Shared Decision-

Making Related to Elective Primary TotalHip and/or Total Knee Arthroplasty

c. Future Measures Around Care Planningd. Future Considerations for Use of

Electronic Health Records4. Form, Manner and Timing of Quality

Measure Data Submission

5. Proposed Display of Quality Measuresand Availability of Information for thePublic From the CCJR Model

E. Data Sharing1. Overview2. Beneficiary Claims Data3. Aggregate Regional Data4. Timing and Period of Baseline Data5. Frequency and Period of Claims Data

Updates for Sharing Beneficiary-Identifiable Claims Data During thePerformance Period

6. Legal Permission to Share Beneficiary-Identifiable Data

F. Monitoring and Beneficiary Protection

1. Introduction and Summary2. Beneficiary Choice and Beneficiary

Notification3. Monitoring for Access to Care4. Monitoring for Quality of Care5. Monitoring for Delayed CareG. Coordination With Other AgenciesIV. Evaluation ApproachA. BackgroundB. Design and Evaluation Methods

C. Data Collection MethodsD. Key Evaluation Research QuestionsE. Evaluation Period and Anticipated

ReportsV. Collection of Information RequirementsVI. Response to CommentsVII. Regulatory Impact AnalysisA. Statement of NeedB. Overall ImpactC. Anticipated Effects1. Overall Magnitude of the Model and its

Effects on the Market2. Effects on the Medicare Programa. Assumptions and Uncertainties

b. Analysesc. Further Consideration3. Effects on Beneficiaries4. Effects on Small Entities5. Effects on Small Rural Hospitals6. Unfunded MandatesD. AlternativesE. Accounting StatementF. ConclusionRegulations Text

I. Executive Summary

A. Purpose

The purpose of this proposed rule isto propose the creation and testing of anew payment model called theComprehensive Care for JointReplacement (CCJR) Model under theauthority of the Center for Medicare and

Medicaid Innovation (Innovation Centeror CMMI). Section 1115A of the SocialSecurity Act (the Act) authorizes theInnovation Center to test innovativepayment and service delivery models toreduce program expenditures whilepreserving or enhancing the quality ofcare furnished to Medicare, Medicaid,and Children’s Health InsuranceProgram beneficiaries. The intent of theCCJR model is to promote quality andfinancial accountability for episodes ofcare surrounding a lower-extremity jointreplacement (LEJR) or reattachment of alower extremity procedure.1 CCJR will

test whether bundled payments to acutecare hospitals for LEJR episodes of carewill reduce Medicare expenditureswhile preserving or enhancing thequality of care for Medicare

beneficiaries. We anticipate the CCJRmodel being proposed would benefitMedicare beneficiaries by improving the

coordination and transition of care,improving the coordination of items andservices paid for through Medicare Fee-For-Service (FFS), encouraging moreprovider investment in infrastructureand redesigned care processes for higherquality and more efficient servicedelivery, and incentivizing higher valuecare across the inpatient and post-acute

care spectrum spanning the episode ofcare. We propose to test CCJR for a 5year performance period, beginning

January 1, 2016, and ending December31, 2020. Under FFS, Medicare makesseparate payments to providers andsuppliers for the items and servicesfurnished to a beneficiary over thecourse of treatment (an episode of care).With the amount of paymentsdependent on the volume of servicesdelivered, providers may not haveincentives to invest in qualityimprovement and care coordinationactivities. As a result, care may be

fragmented, unnecessary, or duplicative.We have previously used ourstatutory authority under section 1115Aof the Act to test bundled paymentmodels such as the Bundled Paymentsfor Care Improvement (BPCI) initiative.Bundled payments for multiple servicesin an episode of care hold participatingorganizations financially accountablefor an episode of care. They also allowparticipants to receive payment in part

based on the reduction in expendituresfor Medicare arising from their careredesign efforts.

We believe the CCJR model beingproposed would further the mission of

the Innovation Center and theSecretary’s goal of increasingly payingfor value and outcomes, rather than forvolume,2 because it would promote thealignment of financial and otherincentives for all health care providerscaring for a beneficiary during an LEJRepisode. In the proposed CCJR model,the acute care hospital that is the site ofsurgery would be held accountable forspending during the episode of care.Participant hospitals would be affordedthe opportunity to earn performance-

based payments by appropriatelyreducing expenditures and meeting

certain quality metrics. They would alsogain access to data and educationalresources to better understand post-acute care and associated spending.Payment approaches that rewardproviders that assume financial andperformance accountability for aparticular episode of care create


http://www.hhs.gov/blog/2015/01/26/progress-towards-better-care-smarter-spending-healthier-people.htmlhttp://www.hhs.gov/blog/2015/01/26/progress-towards-better-care-smarter-spending-healthier-people.htmlhttp://www.hhs.gov/blog/2015/01/26/progress-towards-better-care-smarter-spending-healthier-people.htmlhttp://www.hhs.gov/blog/2015/01/26/progress-towards-better-care-smarter-spending-healthier-people.htmlhttp://www.hhs.gov/blog/2015/01/26/progress-towards-better-care-smarter-spending-healthier-people.htmlhttp://www.hhs.gov/blog/2015/01/26/progress-towards-better-care-smarter-spending-healthier-people.html


5/120


3For example, Total Hip Arthroplasty and TotalKnee Arthroplasty procedures are very high volumeLEJR procedures that together represent the largestpayments for procedures under Medicare. Suter L,Grady JL, Lin Z et al.: 2013 Measure Updates andSpecifications: Elective Primary Total HipArthroplasty (THA) And/Or Total KneeArthroplasty (TKA) All-Cause Unplanned 30-DayRisk-Standardized Readmission Measure (Version2.0). 2013. http://www.cms.gov/Medicare/Quality- Initiatives-Patient-Assessment-Instruments/ HospitalQualityInits/Measure-Methodology.html; Bozic KJ, Rubash HE, Sculco TP, Berry DJ., Ananalysis of Medicare payment policy for total jointarthroplasty. J Arthroplasty. Sep 2008; 23(6 Suppl1):133–138.

incentives for the implementation andcoordination of care redesign betweenhospitals and other providers.

The proposed model would requirethe participation of hospitals inmultiple geographic areas that might nototherwise participate in the testing of

bundled payments for episodes of carefor LEJR procedures. Other episode-

based, bundled payment models beingtested by Centers for Medicare &Medicaid Services (CMS), such as theBPCI initiative, are voluntary in nature.Interested participants must apply tosuch models to participate. To date, wehave not tested an episode paymentmodel with bundled payments in whichproviders are required to participate. Werecognize that realizing the fullpotential of new payment models willrequire the engagement of an even

broader set of providers than haveparticipated to date, providers who mayonly be reached when new payment

models are applied to an entire class ofproviders of a service. As such, we areinterested in testing and evaluating theimpact of a bundled payment approachfor LEJR procedures in a variety ofcircumstances, especially among thosehospitals that may not otherwiseparticipate in such a test.

This proposed model would allowCMS to gain experience with making

bundled payments to hospitals whohave a variety of historic utilizationpatterns; different roles within theirlocal markets; various volumes ofservices; different levels of access tofinancial, community, or other

resources; and various levels ofpopulation and health provider densityincluding local variations in theavailability and use of differentcategories of post-acute care providers.We believe that by requiring theparticipation of a large number ofhospitals with diverse characteristics,the proposed model would result in arobust data set for evaluation of this

bundled payment approach, and wouldstimulate the rapid development of newevidence-based knowledge. Testing themodel in this manner would also allowus to learn more about patterns of

inefficient utilization of health careservices and how to incentivize theimprovement of quality for commonLEJR procedure episodes. This learningpotentially could inform futureMedicare payment policy.

Within this proposed rule we proposea model focused on episodes of care forLEJR procedures. We chose LEJRepisodes for the proposed model

because as discussed in depth in sectionIII.C. of this proposed rule, these arehigh-expenditure, high utilizationprocedures commonly furnished to

Medicare beneficiaries,3 wheresignificant variation in spending forprocedures is currently observed. Thehigh volume of episodes and variationin spending for LEJR procedures createa significant opportunity to test andevaluate the proposed model thatspecifically focuses on a defined set ofprocedures. Moreover, there is

substantial regional variation in post-acute care referral patterns and theintensity of post-acute care provided forLEJR patients, thus resulting insignificant variation in post-acute careexpenditures across LEJR episodesinitiated at different hospitals. Theproposed model would enable hospitalsto consider the most appropriate post-acute care for their LEJR patients. Theproposed model additionally wouldoffer hospitals the opportunity to betterunderstand their own processes withregard to LEJR, as well as the processesof post-acute providers. Finally, while

many LEJR procedures are planned, theproposed model would provide a usefulopportunity to identify efficiencies bothfor when providers can plan for LEJRprocedures and for when the proceduremust be performed urgently.

We note that we seek public commenton the proposals contained in thisproposed rule, and also on anyalternatives considered as well.

B. Summary of the Major Provisions

1. Model Overview: LEJR Episodes ofCare

LEJR procedures are currently paid

under the Inpatient ProspectivePayment System (IPPS) through one oftwo Medicare Severity-DiagnosisRelated Groups (MS–DRGs): MS–DRG469 (Major joint replacement orreattachment of lower extremity withMajor Complications or Comorbidities(MCC)) or MS–DRG 470 (Major jointreplacement or reattachment of lowerextremity without MCC). Under theproposed model, as described further insection III.B of this proposed rule,episodes would begin with admission toan acute care hospital for an LEJRprocedure that is assigned to MS–DRG

469 or 470 upon beneficiary dischargeand paid under the IPPS and would end90 days after the date of discharge fromthe acute care hospital. This episode ofcare definition offers operationalsimplicity for providers and CMS. Theepisode would include the LEJRprocedure, inpatient stay, and all relatedcare covered under Medicare Parts A

and B within the 90 days afterdischarge, including hospital care, post-acute care, and physician services.

2. Model Scope

We propose that participant hospitalswould be the episode initiators and bearfinancial risk under the proposed CCJRmodel. In comparison to other healthcare facilities, hospitals are more likelyto have resources that would allow themto appropriately coordinate and managecare throughout the episode, andhospital staff members are alreadyinvolved in hospital discharge planningand post-acute care recommendationsfor recovery, key dimensions of highquality and efficient care for theepisode. We propose to require allhospitals paid under the IPPS andphysically located in selectedgeographic areas to participate in theCCJR model, with limited exceptions.Eligible beneficiaries who receive careat these hospitals will automatically beincluded in the model. We propose toselect geographic areas through astratified random samplingmethodology within strata based on thefollowing criteria: Historical wageadjusted episode payments and

population size. Our proposedgeographic area selection process isdetailed further in section III.A of thisproposed rule.

3. Payment

We propose to test the CCJR model for5 performance years. During theseperformance years we propose tocontinue paying hospitals and otherproviders according to the usualMedicare FFS payment systems.However, after the completion of aperformance year, the Medicare claimspayments for services furnished to the

beneficiary during the episode, based onclaims data, would be combined tocalculate an actual episode payment.The actual episode payment is definedas the sum of related Medicare claimspayments for items and servicesfurnished to a beneficiary during a CCJRepisode. The actual episode paymentwould then be reconciled against anestablished CCJR target price, withconsideration of additional paymentadjustments based on qualityperformance and post-episode spending.The amount of this calculation, if


http://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Measure-Methodology.htmlhttp://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Measure-Methodology.htmlhttp://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Measure-Methodology.htmlhttp://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Measure-Methodology.htmlhttp://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Measure-Methodology.htmlhttp://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Measure-Methodology.html


6/120


positive, would be paid to theparticipant hospital. This paymentwould be called a reconciliationpayment. If negative, we would requirerepayment from the participant hospital.We propose Medicare would requirerepayment of the difference between theactual episode payments and the CCJRtarget price from a participant hospital

if the CCJR target price is exceeded.We propose to make reconciliation

payments to participant hospitals thatachieve quality outcomes and costefficiencies relative to the establishedCCJR target prices in all performanceyears of the model. We also propose tophase in the requirement thatparticipant hospitals whose actualepisode payments exceed the applicableCCJR target price pay the difference

back to Medicare beginning inperformance year 2. Under thisproposal, Medicare would not requirerepayment from hospitals for

performance year 1 for actual episodepayments that exceed their target pricein performance year 1.

We also propose to limit how mucha hospital can gain or lose based on itsactual episode payments relative totarget prices. We also proposeadditional policies to further limit therisk of high payment cases for allparticipant hospitals and for specialcategories of participant hospitals asdescribed in section III.C. of thisproposed rule.

4. Similar Previous and ConcurrentModels

This proposed model is informed byother models and demonstrationscurrently and previously conducted byCMS and would explore additionalways to enhance coordination of careand improve the quality of servicesthrough bundled payments.

We recently announced the OncologyCare Model (OCM), a new voluntarypayment model for physician practicesadministering chemotherapy. UnderOCM, practices will enter into paymentarrangements that include financial andperformance accountability for episodesof care surrounding chemotherapy

administration to cancer patients. Weplan to coordinate with other payers toalign with OCM in order to facilitateenhanced services and care atparticipating practices. Moreinformation on the OCM can be foundon the Innovation Center’s Web site at:http://innovation.cms.gov/initiatives/ Oncology-Care/.

Medicare tested innovativeapproaches to paying for orthopedicservices in the Medicare Acute CareEpisode (ACE) demonstration, a priordemonstration, and is currently testing

additional approaches under BPCI. Bothof these models have also informed thedesign of the CCJR model.

Under the authority of section 1866Cof the Act, we conducted a 3-yeardemonstration, the Medicare Acute CareEpisode (ACE) Demonstration. Thedemonstration used a prospective globalpayment for a single episode of care as

an alternative approach to payment forservice delivery under traditionalMedicare FFS. The episode of care wasdefined as a combination of Part A andPart B services furnished to MedicareFFS beneficiaries during an inpatienthospital stay for any one of a specifiedset of cardiac and orthopedic MS–DRGs.The MS–DRGs tested included 469 and470, those proposed for inclusion in theCCJR model. The discounted bundledpayments generated an average grosssavings to Medicare of $585 per episodefor a total of $7.3 million across allepisodes (12,501 episodes) or 3.1

percent of the total expected costs forthese episodes. After accounting forincreased post-acute care costs that wereobserved at two sites, Medicare savedapproximately $4 million, or 1.72percent of the total expected Medicarespending. More information on the ACEDemonstration can be found on theInnovation Center’s Web site at: http:// innovation.cms.gov/initiatives/ACE/.

We are currently testing the BPCIinitiative. The BPCI initiative iscomprised of four related paymentmodels, which link payments formultiple services that Medicare

beneficiaries receive during an episode

of care into a bundled payment. Underthe initiative, entities enter intopayment arrangements with CMS thatinclude financial and performanceaccountability for episodes of care.Episodes of care under the BPCIinitiative begin with either—(1) aninpatient hospital stay or (2) post-acutecare services following a qualifyinginpatient hospital stay. The BPCIinitiative is evaluating the effects ofepisode-based payment approaches onpatient experience of care, outcomes,and cost of care for Medicare FFS

beneficiaries. Each of the four models

tests LEJR episodes of care. While finalevaluation results for the models withinthe BPCI initiative are not yet available,we believe that CMS’ experiences withBPCI support the design of the CCJRmodel. Under section 1115A(c) of theAct, the Secretary may, taking intoconsideration an evaluation conductedunder section 1115A(b)(4) of the Act,‘‘through rulemaking, expand (includingimplementation on a nationwide basis)the duration and the scope of a modelthat is being tested under’’ theInnovation Center’s authority. CCJR is

not an expansion of BPCI, and BPCI may be expanded in the future. CMSpublished a discussion item solicitingpublic comment on a potential futureexpansion of one or more of the modelswithin BPCI in the CY2016 IPPS rule, 80FR 24414 through 24418. CCJR wouldnot be not an expansion or modificationof BPCI; nor does it reflect comments

received in response to the NPRM forthe 2016 IPPS Rule. CCJR is a uniquemodel that tests a broader, differentgroup of hospitals than BPCI. It isnecessary to provide CMS withinformation about testing bundledpayments to hospitals that are requiredto participate in an alternative paymentmodel. For a discussion of why we arerequiring hospitals to participate in theCCJR model, see section III.A of thisproposed rule.

The CCJR model’s design wasinformed to a large degree by ourexperience with BPCI Model 2. BPCI’s

Model 2 is a voluntary episode paymentmodel in which a qualifying acute carehospitalization initiates a 30, 60 or 90day episode of care. The episode of careincludes the inpatient stay in an acutecare hospital and all related servicescovered under Medicare Parts A and Bduring the episode, including post-acutecare services. More information on BPCIModel 2 can be found on the InnovationCenter’s Web site at: http:// innovation.cms.gov/initiatives/BPCI-Model-2/.

Further information of why elementsof the OCM, the ACE Demonstration,and BPCI Model 2 were incorporated

into the design of the CCJR model isdiscussed later in this proposed rule.

5. Overlap With Ongoing CMS Efforts

We propose to exclude fromparticipation in CCJR certain hospitalsparticipating in the risk-bearing phase ofBPCI Models 2 and 4 for LEJR episodes,as well as acute care hospitalsparticipating in BPCI Model 1. Wepropose not to exclude beneficiaries inCCJR model episodes from beingincluded in other Innovation Centermodels or CMS programs, such as theMedicare Shared Savings Program, as

detailed later in this proposed rule. Wepropose to account for overlap, that is,where CCJR beneficiaries are alsoincluded in other models and programsto ensure the financial policies of CCJRare maintained and results andspending reductions are attributed tothe correct model or program.

6. Quality Measures and ReportingRequirements

We are proposing to adopt threehospital-level quality of care measuresfor the CCJR model. Those measures


http://innovation.cms.gov/initiatives/Oncology-Care/http://innovation.cms.gov/initiatives/Oncology-Care/http://innovation.cms.gov/initiatives/ACE/http://innovation.cms.gov/initiatives/ACE/http://innovation.cms.gov/initiatives/BPCI-Model-2/http://innovation.cms.gov/initiatives/BPCI-Model-2/http://innovation.cms.gov/initiatives/BPCI-Model-2/http://innovation.cms.gov/initiatives/ACE/http://innovation.cms.gov/initiatives/ACE/http://innovation.cms.gov/initiatives/BPCI-Model-2/http://innovation.cms.gov/initiatives/BPCI-Model-2/http://innovation.cms.gov/initiatives/BPCI-Model-2/http://innovation.cms.gov/initiatives/Oncology-Care/http://innovation.cms.gov/initiatives/Oncology-Care/


7/120


include a complication measure,readmission measure, and a patientexperience survey measure. We proposeto use these measures to test the successof the model in achieving its goalsunder section 1115A of the Act and tomonitor for beneficiary safety. Weintend to publicly report thisinformation on the Hospital Compare

Web site. Additionally, we areproposing and requesting publicfeedback on possible voluntarysubmission of data to support thedevelopment of a hospital-level measureof patient-reported outcomes followingan elective Primary Total Hip (THA) orTotal Knee Arthroplasty (TKA).

7. Data Sharing Process

We propose to share data withparticipant hospitals upon requestthroughout the performance period ofthe CCJR model to the extent permitted

by the HIPAA Privacy Rule and otherapplicable law. We propose to shareupon request both raw claims-level dataand claims summary data by serviceline with participants. This approachwould allow participant hospitalswithout prior experience analyzingclaims to use summary data to receiveuseful information, while allowingthose participant hospitals who preferraw claims-level data the opportunity toanalyze claims. We propose to providehospitals with up to 3 years ofretrospective claims data upon requestthat will be used to develop their targetprice, as described in section III.C ofthis proposed rule. In accordance with

the HIPAA Privacy Rule, we wouldlimit the content of this data set to theminimum data necessary for theparticipant hospital to conduct qualityassessment and improvement activitiesand effectively coordinate care of itspatient population.

8. Beneficiary Protections

Under the CCJR model, beneficiariesretain the right to obtain health servicesfrom any individual or organizationqualified to participate in the Medicareprogram. Under the CCJR model,eligible beneficiaries who receive

services from a participant hospitalwould not have the option to opt out ofinclusion in the model. We propose torequire participant hospitals to supply

beneficiaries with written informationregarding the design and implications ofthis model as well as their rights underMedicare, including their right to usetheir provider of choice. We will alsomake a robust effort to reach out to

beneficiaries and their advocates to helpthem understand the CCJR model.

We also propose to use our existingauthority, if necessary, to audit

participant hospitals if claims analysisindicates an inappropriate change indelivered services. Beneficiaryprotections are discussed in greaterdepth in section III.E. of this proposedrule.

9. Financial Arrangements and ProgramPolicy Waivers

We propose to hold participanthospitals financially responsible forCCJR LEJR episodes as participants inthe model as discussed in sectionIII.C.10.a. of this proposed rule.Specifically, only these hospitalparticipants would be directly subject tothe requirements of this proposed rulefor the CCJR model. Participanthospitals would be responsible forensuring that other providers andsuppliers collaborating with the hospitalon LEJR episode care redesign are incompliance with the terms andconditions of the model.

Several of the proposed Medicareprogram policy waivers outline theconditions under which skilled nursingfacilities (SNFs) and physicians couldfurnish and bill for certain servicesfurnished to CCJR beneficiaries wherecurrent Medicare programs rules wouldnot permit such billing. We draw theattention of SNFs and physicians tothese proposals that are included insection III.C.10.b.(5). of this proposedrule.

C. Summary of Economic Effects

As shown in our impact analysis, weexpect the proposed model to result in

savings to Medicare of $153 millionover the 5 years of the model. Morespecifically, in performance year 1 ofthe model, we estimate a Medicare costof approximately $23 million, as wehave proposed that hospitals will not besubject to downside risk in the first yearof the model. As we introduce downsiderisk beginning in performance year 2 ofthe model, we estimate Medicaresavings of approximately $29 million. Inperformance year 3 of the model, weestimate Medicare savings of $43million. In performance years 4 and 5 ofthe model, as we have proposed to move

from target episode pricing that is basedon a hospital’s experience to targetpricing based on regional experience,we estimate Medicare savings of $50million and $53 million, respectively.

Additionally, hospitals must meet orexceed specific thresholds onperformance on certain quality of caremeasures in order to be eligible for areconciliation payment and as theperformance threshold increases inperformance years 4 through 5, weestimate additional savings. As a result,we estimate the net savings to Medicare

to be $153 million over the 5 years ofthe model. We anticipate there would bea broader focus on care coordinationand quality improvement for LEJRepisodes among hospitals and otherproviders within the Medicare programthat would lead to both increasedefficiency in the provision of care andimproved quality of the care provided to

beneficiaries.We note that under section

1115A(b)(3)(B) of the Act, the Secretaryis required to terminate or modify amodel unless certain findings can bemade with respect to savings andquality after the model has begun. Ifduring the course of testing the modelit is determined that termination ormodification is necessary, such actionswould be undertaken throughrulemaking.

II. Background

This proposed rule proposes theimplementation of a new innovativehealth care payment model under theauthority of section 1115A of the Act.Under the model, called the CCJRmodel, acute care hospitals in certainselected geographic areas will receive

bundled payments for episodes of carewhere the diagnosis at dischargeincludes a lower extremity jointreplacement or reattachment of a lowerextremity that was furnished by thehospital. We are proposing that the

bundled payment will be paidretrospectively through a reconciliationprocess; hospitals and other providersand suppliers will continue to submit

claims and receive payment via theusual Medicare FFS payment systems.All related care covered under MedicarePart A and Part B within 90 days afterthe date of hospital discharge from thejoint replacement procedure will beincluded in the episode of care. We

believe this model will further our goalsof improving the efficiency and qualityof care for Medicare beneficiaries forthese common medical procedures.

III. Provisions of the Proposed Rule

A. Proposed Definition of the EpisodeInitiator and Selected Geographic Areas

1. BackgroundThe CCJR model is different from

BPCI because it would requireparticipation of all hospitals (withlimited exceptions) throughout selectedgeographic areas, which would result ina model that includes varying hospitaltypes. However, a discussion of BPCI isrelevant because its design informs andsupports the proposed CCJR model. TheBPCI model is voluntary, and under thatmodel we pay a bundled payment for anepisode of care only to entities that have



8/120


elected to participate in the model. Weare interested in testing and evaluatingthe impact of an episode paymentapproach for LEJRs in a variety of othercircumstances, including among thosehospitals that have not chosen tovoluntarily participate because we havenot tested bundled payments for thesehospitals previously. This would allow

CMS and participants to gain experiencetesting and evaluating episode-basedpayment for LEJR procedures furnished

by hospitals with a variety of historicutilization patterns; roles within theirlocal markets; volume of servicesprovided; access to financial,community, or other resources; andpopulation and health care providerdensity. Most importantly, participationof hospitals in selected geographic areaswill allow CMS to test bundledpayments without introducing selection

bias such as the selection bias inherentin the BPCI model due to self-selected

participation.2. Proposed Definition of EpisodeInitiator

In BPCI Model 2, LEJR episodeinitiators are either acute care hospitalswhere the LEJR procedure is performedor physician group practices whosephysician members are the admitting oroperating physician for the hospitalstay. Thus, under BPCI, it is possiblethat only some Medicare cases thatcould potentially be included in anLEJR episode at a specific hospital areactually being tested in BPCI. Forexample, if the hospital itself is not

participating as an episode initiatorunder BPCI, yet some physicians whoadmit patients to the hospital aremembers of physician group practicesparticipating in BPCI, not all of thehospital’s possible LEJR episodes aretested and paid under BPCI.

Under the proposed CCJR model, asdescribed further in section III.B of thisproposed rule, episodes would beginwith admission to an acute care hospitalfor an LEJR procedure that is paid underthe IPPS through Medical SeverityDiagnosis-Related Group (MS–DRG) 469(Major joint replacement or

reattachment of lower extremity withMCC) or 470 (Major joint replacement orreattachment of lower extremity withoutMCC) and end 90 days after the date ofdischarge from the hospital. For theCCJR model, we propose that hospitalswould be the only episode initiators.For purposes of CCJR, the term‘‘hospital’’ means a hospital as definedin section 1886(d)(1)(B) of the Act. Thisstatutory definition of hospital includesonly acute care hospitals paid under theIPPS. Under this proposal, all acute carehospitals in Maryland would be

excluded from CCJR. The state ofMaryland entered into an agreementwith CMS, effective January 1, 2014, toparticipate in CMS’ new Maryland All-Payer Model. In order to implement theMaryland All-Payer Model, CMS waivedcertain requirements of the Act, and thecorresponding implementingregulations, as set forth in the agreement

between CMS and Maryland.Specifically, under the Maryland All-Payer Model, Maryland acute carehospitals are not paid under the IPPS orOPPS but rather are paid under rates set

by the state. Following the model’sperformance period, Maryland willtransition to a new model thatincorporates the full spectrum of care,not just hospital services. As such, withrespect to Maryland hospitals, CMSintends to test and develop newpayment and delivery approaches thatcan incorporate non-hospital services ina manner that accounts for Maryland’s

unique hospital rate setting system andpermit Maryland to develop its ownstrategy to incentivize higher qualityand more efficient care across clinicalsituations within and beyond hospitals,including but not limited to LEJRepisodes of care. We are proposing thatpayments to Maryland hospitals would

be excluded in the regional pricingcalculations as described in sectionIII.C.4 of this proposed rule. We seekcomment on this proposal and whetherthere are potential approaches forincluding Maryland acute care hospitalsin CCJR. In addition, we seek comment

on whether Maryland hospitals should be included in CCJR in the future uponany termination of the Maryland All-Payer Model.

We propose to designate IPPShospitals as the episode initiators toensure that all Medicare FFS LEJRservices furnished by participanthospitals in selected geographic areas to

beneficiaries who do not meet theexclusion criteria specified in sectionIII.B.3 of this proposed rule and are notBPCI episodes that we are proposing toexclude as outlined in this section andalso in section III.C.7 of this proposed

rule are included in the CCJR model. Weare proposing certain exceptions to theinclusion of hospitals in the CCJRModel, as discussed in section III.C. ofthis proposed rule. Given that ourproposal to initiate the LEJR episode

begins with an admission to a hospitalpaid under the IPPS that results in adischarge assigned to MS–DRG 469 or470, we believe that utilizing thehospital as the episode initiator is astraightforward approach for this model

because the hospital furnishes the LEJRprocedure. In addition, we are

interested in testing a broad model in anumber of hospitals under the CCJRmodel in order to examine results froma more generalized payment model.Thus, we believe it is important that, ina model where hospital participation isnot voluntary, all Medicare FFS LEJRepisodes that begin at the participanthospital in a selected geographic area

are included in the model for beneficiaries that do not meet theexclusion criteria specified in sectionIII.B.3 of this proposed rule and are notBPCI episodes that we are proposing toexclude as outlined in this section andalso in section III.C.7 of this proposedrule. This is best achieved if the hospitalis the episode initiator. Finally, asdescribed in the following sections thatpresent our proposed approach togeographic area selection, thisgeographic area selection approachrelies upon our definition of hospitals asthe entities that initiate episodes. We

seek comment on our proposal to definethe episode initiator as the hospitalunder CCJR.

3. Financial Responsibility for theEpisode of Care

BPCI Model 2 participants that haveentered into agreements with CMS to

bear financial responsibility for anepisode of care include acute carehospitals paid under the IPPS, healthsystems, physician-hospitalorganizations, physician grouppractices, and non-provider businessentities that act as conveners bycoordinating multiple health care

providers’ participation in the model.Thus, our evaluation of BPCI Model 2will yield information about how resultsfor LEJR episodes may differ based ondifferences in which party bearsfinancial responsibility for the episodeof care.

For the CCJR model, we propose tomake hospitals financially responsiblefor the episode of care for severalreasons. We recognize that ideally all ofthe providers involved in thecontinuum of care for Medicare

beneficiaries in a 90-day post-dischargeLEJR episode would work together to

determine the best structure formanaging the LEJR episode, develop anefficient process that leads to highquality care, track information acrossthe episode about quality and Medicareexpenditures, and align financialincentives using a variety of approaches,including gainsharing. However,

because the proposed CCJR model istesting a more generalizable model byincluding hospitals that might notparticipate in a voluntary model andincludes episodes initiated at a widevariety of hospitals, we believe it is



9/120


most appropriate to identify a singletype of provider to bear financialresponsibility for making repayment toCMS under the model.

Hospitals play a central role incoordinating episode-related care andensuring smooth transitions for

beneficiaries undergoing LEJRprocedures. Moreover, the episode

always begins with an acute carehospital stay, IPPS payments for LEJRscomprise about 50 percent of Medicarepayments for a 90-day episode, and the

beneficiary’s recovery from surgery begins during the hospital stay. Mosthospitals already have someinfrastructure related to healthinformation technology, patient andfamily education, and care managementand discharge planning. This includespost-acute care (PAC) coordinationinfrastructure and resources such ascase managers, which hospitals can

build upon to achieve efficiencies under

this episode payment model. Manyhospitals also have recently heightenedtheir focus on aligning their efforts withthose of community providers toprovide an improved continuum of caredue to the incentives under other CMSmodels and programs, includingAccountable Care Organization (ACO)initiatives such as the Medicare SharedSavings Program (MSSP), and theHospital Readmissions ReductionProgram (HRRP), establishing a base foraugmenting these efforts under the CCJRmodel.

In view of our proposal that hospitals be the episode initiators under this

model, we believe that hospitals aremore likely than other providers to havean adequate number of episode cases tojustify an investment in episodemanagement for this model. We also

believe that hospitals are most likely tohave access to resources that wouldallow them to appropriately manage andcoordinate care throughout the LEJRepisode. Finally, the hospital staff isalready involved in discharge planningand placement recommendations forMedicare beneficiaries, and moreefficient PAC service delivery providessubstantial opportunities for improving

quality and reducing costs under CCJR.We considered requiring treatingphysicians (orthopedic surgeons orothers) or their associated physiciangroup practices, if applicable, to befinancially responsible for the episodeof care under the CCJR Model. Weexpect that every Medicare beneficiarydischarged with a diagnosis groupedunder MS–DRG 469 or 470 would havean operating physician and an admittingphysician for the hospital stay.However, the services of providers otherthan the hospital where the acute care

hospital stay for the LEJR procedure(hereinafter ‘‘the anchorhospitalization’’) occurs would notnecessarily be furnished in every LEJRepisode. For example, that physicians ofdifferent specialties play varying rolesin managing patients during an acutecare hospitalization for a surgicalprocedure and during the recovery

period, depending on the hospital andcommunity practice patterns and theclinical condition of the beneficiary andcould not be assumed to be included inevery LEJR episode. This variabilitywould make requiring a particularphysician or physician group practice to

be financially responsible for a givenepisode very challenging.

If we were to assign financialresponsibility to the operatingphysician, it is likely that there would

be significant variation in the number ofrelevant episodes that could be assignedto an individual person. Where the

physician was included in a physiciangroup practice, episodes could beaggregated to this group level but thiswould not be possible for all cases andwould likely still have low volumeconcerns. We believe that the smallsample sizes accruing to individualphysician and physician group practiceswould make systematic care redesigninefficient and more burdensome, giventhat we are proposing to test allepisodes occurring at hospitals selectedfor participation for beneficiaries that donot meet the exclusion criteria specifiedin section III.B.3 of this proposed ruleand are not BPCI episodes that we are

proposing to exclude as outlined in thissection and also in section III.C.7 of thisproposed rule.

Finally, we note that although theBPCI initiative includes the possibilityof a physician group practice as a typeof initiating participant, the physiciangroups electing to participate in BPCIhave done so because their practicestructure supports care redesign andother infrastructure necessary to bearfinancial responsibility for episodes andis not necessarily representative of thetypical group practice. In addition, mostof the physician group practices in BPCI

are not bearing financial responsibility, but are participating in BPCI as partnerswith convener organizations (discussedlater in this section), which enter intoagreements with CMS, on behalf ofhealth care providers such as physiciangroup practices, through which theyaccept financial responsibility for theepisode of care. The infrastructurenecessary to accept financialresponsibility for episodes is not presentacross all physician group practices, andthus we do not believe it would beappropriate to designate physician

group practices to bear the financialresponsibility for making repayments toCMS under the proposed CCJR model.We seek comment on our proposal torequire the hospital to bear the financialresponsibility for the episodes of careunder CCJR.

We are proposing that hospitals will bear the financial responsibility for LEJR

episodes of care under CCJR. However, because there are LEJR episodescurrently being tested in BPCI Model 1,2, 3 or 4, we believe that participationin CCJR should not be required if itwould disrupt testing of LEJR episodesalready underway in BPCI models.Therefore, we are proposing that IPPShospitals located in an area selected forthe model that are active Model 1 BPCIparticipant hospitals as of July 1, 2015or episode initiators for LEJR episodesin the risk-bearing phase of Model 2 or4 of BPCI as of July 1, 2015, would beexcluded from participating in CCJR

during the time that their qualifyingepisodes are included in one of the BPCImodels. Likewise, we are proposing thatif the participant hospital is not anepisode initiator for LEJR episodesunder BPCI Model 2, then LEJRepisodes initiated by other providers orsuppliers under BPCI Model 2 or 3(where the surgery takes place at theparticipant hospital) would be excludedfrom CCJR. Otherwise qualifying LEJRepisodes (that is, those that are not partof a Model 3 BPCI LEJR episode or aModel 2 physician group practice-initiated LEJR episode) at theparticipant hospital would be included

in CCJR.While we propose that the participant

hospital be financially responsible forthe episode of care under CCJR, we also

believe that effective care redesign forLEJR episodes requires meaningfulcollaboration among acute carehospitals, PAC providers, physicians,and other providers and supplierswithin communities to achieve thehighest value care for Medicare

beneficiaries. We believe it may beessential for key providers to be alignedand engaged, financially and otherwise,with the hospitals, with the potential to

share financial responsibility with thosehospitals. We note that all relationships between and among providers andsuppliers must comply with all relevantlaws and regulations, including thefraud and abuse laws and all Medicarepayment and coverage requirementsunless otherwise specified further laterin this section and in section III.C.10 ofthis proposed rule. Depending on ahospital’s current degree of clinicalintegration, new and differentcontractual relationships amonghospitals and other health care



10/120


4As stated in the FY 2014 IPPS/LTCH PPSproposed rule (78 FR 27552) and final rule (78 FR50586), on February 28, 2013, OMB issued OMBBulletin No. 13–01, which established reviseddelineations for MSAs, mSAs, and CSAs, andprovided guidance on the use of the delineations ofthese statistical areas. A copy of this bulletin may be obtained at http://www.whitehouse.gov/sites/ default/files/omb/bulletins/2013/b-13-01.pdf.

providers may be important, althoughnot necessarily required, for CCJR modelsuccess in a community. Weacknowledge that financial incentivesfor other providers may be importantaspects of the model in order forhospitals to partner with these providersand incentivize certain strategies toimprove episode efficiency.

In the BPCI initiative, participantshave entered a variety of relationshipswith entities above the hospital level.Some of these relationships are oneswhere the financial risk is borne by theentity other than the hospital, such as aparent organization (known as awardeeconveners) and others have managerialor other responsibility relationshipswith other organizations (known asfacilitator conveners) but financialresponsibility remains with the episodeinitiator . We acknowledge theimportant role that conveners play inthe BPCI initiative with regard to

providing infrastructure support tohospitals and other entities initiatingepisodes in BPCI. The convenerrelationship (where another entityassumes financial responsibility) maytake numerous forms, includingcontractual (such as a separate for-profitcompany that agrees to take on ahospital’s financial risk in the hopes ofachieving financial gain through bettermanagement of the episodes) andthrough ownership (such as when riskis borne at a corporate level within ahospital chain).

However, we are proposing that for

the CCJR model, we would hold onlythe participant hospitals financiallyresponsible for the episode of care. Thisis consistent with the goal of evaluatingthe impact of bundled payment and careredesign across a broad spectrum ofhospitals with varying levels ofinfrastructure and experience inentering into risk-based reimbursementarrangements. If conveners wereincluded as participants in CCJR, wemay not gain the knowledge of how avariety of hospitals can succeed inrelationship with CMS in which they

bear financial risk for the episode of

care. We acknowledge that CCJRhospitals may wish to enter intorelationships with other entities in orderto manage the episode of care ordistribute risk. We do not intend torestrict the ability of hospitals to enterinto administrative or risk sharingarrangements related to this model. Werefer readers to section III.C.10 of thisproposed rule for further discussion ofmodel design elements that may outlinefinancial arrangements betweenparticipant hospitals and otherproviders and suppliers.

4. Proposed Geographic Unit ofSelection and Exclusion of SelectedHospitals

In determining which hospitals toinclude in the CCJR model, weconsidered whether the model should

be limited to hospitals where a highvolume of LEJRs are performed, whichwould result in a more narrow test onthe effects of an episode-based payment,or whether to include all hospitals inparticular geographic areas, whichwould result in testing the effects of anepisode-based payment approach more

broadly across an accountable carecommunity seeking to coordinate carelongitudinally across settings. Selectingcertain hospitals where a high volumeof LEJRs are performed may allow forfewer hospitals to be selected as modelparticipants, but still result in asufficient number of CCJR episodes toevaluate the success of the model.However, there would be more potential

for behavioral changes that couldinclude patient shifting and steering

between hospitals in a given geographicarea that could impact the test.Additionally, this approach wouldprovide less information on testingepisode payments for LEJR proceduresacross a wide variety of hospitals withdifferent characteristics. Selectinggeographic areas and including all IPPShospitals in those areas not otherwiseexcluded due to BPCI overlap aspreviously described and in sectionIII.C.7 of this proposed rule as modelparticipants would help to minimize the

risk of participant hospitals shiftinghigher cost cases out of the CCJR model.Moreover, in selecting geographic areaswe could choose certain characteristics,stratify geographic areas according tothese characteristics, and randomlyselect geographic areas from within eachstratum. Such a stratified randomsampling method based on geographicarea would allow us to observe theexperiences of hospitals with variouscharacteristics, such as variations insize, profit status, and episodeutilization patterns, and examinewhether these characteristics impact theeffect of the model on patient outcomesand Medicare expenditures withinepisodes of care. Stratification wouldalso substantially reduce the extent towhich the selected hospitals will differfrom non-selected hospitals on thecharacteristics used for stratification,which would improve the statisticalpower of the subsequent modelevaluation, improving our ability toreach conclusions about the model’seffects on episode costs and the qualityof patient care. Therefore, given theauthority in section 1115A(a)(5) of the

Act, which allows the Secretary to electto limit testing of a model to certaingeographic areas, we propose to use astratified random sampling method toselect geographic areas and require allhospitals paid under the IPPS in thoseareas to participate in the CCJR modeland be financially responsible for thecost of the episode, with certain

exceptions as previously discussed andin sections III.B.3 and III.C.7 of thisproposed rule.

a. Overview and Options for GeographicArea Selection

In determining the geographic unit forthe geographic area selection for thismodel, we considered using a stratifiedrandom sampling methodology to select(1) certain counties based on their Core-Based Statistical Area (CBSA) status, (2)certain zip codes based on theirHospital Referral Regions (HRR) statusor (3) certain states. We address each

geographic unit in turn.We considered selecting certaincounties based on their CBSA status.The general concept of a CBSA is thatof a core area containing a substantialpopulation nucleus, together withadjacent communities having a highdegree of economic and socialintegration within that core. Countiesare designated as part of a CBSA whenthe county or counties or equivalententities are associated with at least onecore (urbanized area or urban cluster) ofat least 10,000 in population, plusadjacent counties having a high degreeof social and economic integration withthe core as measured throughcommuting ties with the countiesassociated with the core. There are 929CBSAs currently used for geographicwage adjustment purposes acrossMedicare payment systems.4 The 929CBSAs include 388 MetropolitanStatistical Areas (MSAs), which have anurban core population of at least 50,000,and the 541 Micropolitan StatisticalAreas (mSA), which have an urban corepopulation of at least 10,000 but lessthan 50,000. CBSAs may be furthercombined into a Combined StatisticalArea (CSA) which consists of two or

more adjacent CBSAs (MSAs or mSAs or both) with substantial employmentinterchange. Counties not classified as aCBSA are typically categorized andexamined at a state level.


http://www.whitehouse.gov/sites/default/files/omb/bulletins/2013/b-13-01.pdfhttp://www.whitehouse.gov/sites/default/files/omb/bulletins/2013/b-13-01.pdfhttp://www.whitehouse.gov/sites/default/files/omb/bulletins/2013/b-13-01.pdfhttp://www.whitehouse.gov/sites/default/files/omb/bulletins/2013/b-13-01.pdf


11/120


5The Dartmouth Atlas of Healthcare, http:// www.dartmouthatlas.org/data/region/ . Accessed onApril 9, 2015.

The choice of a geographical unit based on CBSA status could meanselection of a CBSA, an MSA, or a CSA.We propose basing the selection on anMSA, which we will discuss later inthis section.

In determining which geographicareas will be potentially subject toselection, we focused on MSAs, which

is a subcategory within CBSAcharacterized by counties associatedwith an urban core population of at least50,000. It is our intention at this timethat counties not in an MSA would not

be subject to the selection process.These counties not subject to selectionwould include the mSA counties and thecounties without a core urban area of atleast 10,000. These areas are largelyrural areas and have a limited numberof qualifying LEJR cases. Relatively fewof these areas would be able to qualifyfor inclusion based on the minimumnumber of LEJR episodes in year

requirement discussed later in thissection.We considered, but ultimately

decided against, using CSA designationinstead of MSAs as a potential unit ofselection. Under this scenario, wewould look at how OMB classifiescounties. We would first assess whethera county has been identified as

belonging to a CSA, a unit whichconsists of adjacent MSAs or mSAs or

both. If the county was not in a CSA, wewould determine if it was in an MSAthat is not part of a larger CSA. Countiesnot associated with a CSA or an MSAwould be unclassified and excluded

from selection. These unclassified areaswould include the counties in a statethat were either not a CBSA (no corearea of at least 10,000) or associatedwith a mSA (core area of between 10,000and 50,000) but unaffiliated with a CSA.

Whether to select on the basis of CSA/MSAs or just on MSAs was influenced

by a number of factors including anassessment with respect to theanticipated degree to which LEJRpatients would be willing to travel fortheir initial hospitalization, the extent towhich surgeons are expected to haveadmitting privileges in multiple

hospitals located in different MSAs andconsiderations related to the degree towhich we desire to include hospitalswithin mSAs that are part of a largerCSA. It was believed that theanticipated risk for patient shifting andsteering between MSAs within a CSAwas not severe enough to warrantselecting CSAs. However, for these samereasons, we believe that selectingcomplete MSAs is preferable toselecting metropolitan divisions ofMSAs for inclusion in the CCJR model.We use the metropolitan divisions to set

wage indices for its prospectivepayment systems. Of the 388 MSAs,there are 11 MSAs that contain multiplemetropolitan divisions. For example,the Boston-Cambridge-Newton, MA–NHMSA is divided into the followingmetropolitan divisions:

• Boston, MA.• Cambridge-Newton-Framingham,

MA.• Rockingham County-Strafford

County, NH.The Seattle-Tacoma-Bellevue, WA MSAis divided into the followingmetropolitan divisions:

• Seattle-Bellevue-Everett, WA.• Tacoma-Lakewood, WA.We propose selecting entire MSAs

rather than sub-divisions within anMSA.

We next considered selecting hospitalreferral regions (HRRs). HRRs representregional health care markets for tertiarymedical care. There are 306 HRRs with

at least one city where both majorcardiovascular surgical procedures andneurosurgery are performed. HRRs aredefined by determining where themajority of patients were referred formajor cardiovascular surgicalprocedures and for neurosurgery.5 Compared to MSAs, HRRs are classified

based on where the majority of beneficiaries within a zip code receivetheir hospital services for selectedtertiary types of care. The resultingHRRs represent the degree to whichpeople travel for tertiary care thatgenerally requires the services of amajor referral center and not the size ofthe referral network for more routineservices, such as knee and hiparthroplasty procedures. In addition,

because HRRs are defined based onreferrals for cardiovascular surgicalprocedures and neurosurgery, they maynot reflect referrals for orthopedicprocedures. Therefore, we believe thatMSAs as a geographic unit arepreferable over HRRs for this model.

We also considered selecting states forthe CCJR model. However, weconcluded that MSAs as a geographicunit are preferable over states for theCCJR model. As mentioned in section

III.A.4.b of this proposed rule, weanticipate that hospitals that wouldotherwise be required to participate inthe CCJR model would be excludedfrom the model because their relevantLEJR episodes are already being testedin BPCI. If we were to select states as thegeographic unit, there is a potential thatan entire state would need to beexcluded because a large proportion of

hospitals in that state are episodeinitiators of LEJR episodes in BPCI. Incontrast, if we excluded a specific MSAdue to BPCI participation, as discussedin the next section, we could still selectanother MSA within that same state.Likewise, if we chose states as thegeographic unit, we wouldautomatically include hospitals in all

rural areas within the state selected. IfMSAs are selected for the geographicunit, we anticipate that fewer smallrural hospitals would be included in themodel. Using a unit of selection smallerthan a state would allow for a moredeliberate choice about the extent ofinclusion of rural or small populationareas. Selecting states rather than MSAswould also greatly reduce the number ofindependent geographic areas subject toselection under the model, which woulddecrease the statistical power of themodel evaluation. Finally, MSAsstraddle state lines where providers and

Medicare beneficiaries can easily crossthese boundaries for health care.Choosing states as the geographic unitwould potentially divide a hospitalmarket and set up a greater potential forpatient shifting and steering to differenthospitals under the model. The decisionthat the MSA-level analysis was moreanalytically appropriate was based onthe specifics of this model and notmeant to imply that other levels ofselection would not be appropriate in adifferent model such as the proposedHome Health Value Based Purchasing(HHVBP) model.

For the reasons previously discussed,we propose to require participation inthe CCJR model of all hospitals, withlimited exceptions as previouslydiscussed in section III.A.2.of thisproposed rule, paid under the IPPS thatare physically located in a county in anMSA selected through a stratifiedrandom sampling methodology,outlined in section III.A.3.b in thisproposed rule, to test and evaluate theeffects of an episode-based paymentapproach for an LEJR episodes. Wepropose to determine that a hospital islocated in an area selected if thehospital is physically located within the

boundary of any of the counties in thatMSA as of the date the selection ismade. Although MSAs are revisedperiodically, with additional countiesadded or removed from certain MSAs,we propose to maintain the same cohortof selected hospitals throughout the 5year performance period of the modelwith limited exceptions as describedlater in this section. Thus, we proposenot to add hospitals to the model if afterthe start of the model new counties areadded to one of the selected MSAs or


http://www.dartmouthatlas.org/data/region/http://www.dartmouthatlas.org/data/region/http://www.dartmouthatlas.org/data/region/http://www.dartmouthatlas.org/data/region/http://www.dartmouthatlas.org/data/region/


12/120


remove hospitals from the model ifcounties are removed from one of theselected MSAs. We believe that thisapproach will best maintain theconsistency of the participants in themodel, which is crucial for our abilityto evaluate the results of the model.However, we retain the possibility ofadding a hospital that is opened or

incorporated within one of the selectedcounties after the selection is made andduring the period of performance. (Seesection III.C.of this proposed rule fordiscussion of how target prices will bedetermined for such hospitals.)Although we considered includinghospitals in a given MSA based onwhether the hospitals were classifiedinto the MSA for IPPS wage indexpurposes, this process would be morecomplicated, and we could not find anycompelling reasons favoring thisapproach. For example, we assignhospitals to metro divisions of MSAs

when those divisions exist. See ourprevious discussion of this issue. Inaddition, there is the IPPS process ofgeographic reclassification by which ahospital’s wage index value orstandardized payment amount is basedon a county other than the one wherethe hospital is located. For the purposeof this model, it is simpler and morestraightforward to use the hospital’sphysical location as the basis ofassignment to a geographic unit. Thisdecision would have no impact on ahospital’s payment under the IPPS. Weseek comment on our proposal toinclude participant hospitals for the

CCJR model based on the physicallocation of the hospital in one of thecounties included in a selected MSA.

b. MSA Selection Methodology

We propose to select the MSAs toinclude in the CCJR model by stratifyingall of the MSAs nationwide according tocertain characteristics.

(1) Exclusion of Certain MSAs

Prior to assigning an MSA to aselection stratum, we examined whetherthe MSA met specific proposedexclusion criteria. MSAs were evaluated

sequentially using the following 4exclusion criteria: First, MSAs in whichfewer than 400 LEJR episodes(determined as we propose to determineepisodes included in this model, asdiscussed in section III.B.2) occurredfrom July 1, 2013 through June 30, 2014were removed from possible selection.The use of the 400 LEJR cases in a yearwas based on a simple one-sided powercalculation to assess the number ofepisodes that would be needed to detecta 5 percent reduction in episodeexpenditures. Accordingly, cases in

hospitals paid under either the criticalaccess hospital (CAH) methodology orthe Maryland All-Payer Model are notincluded in the count of eligibleepisodes. This criterion removed 156MSAs from possible selection.

Second, MSAs were removed frompossible selection if there were fewerthan 400 non-BPCI LEJR episodes in the

MSA in the reference year. For thepurposes of this exclusion, the numberof BPCI episodes was estimated as thenumber of potentially eligible casesduring the reference year that occurredin acute care hospitals participating inBPCI Model 1, or in phase 2 of BPCIModels 2 or 4 as of July 1, 2015 and thenumber of LEJRs in 2013 and 2014associated with these hospitals wasexamined. This criterion removed anadditional 24 MSAs from potentialselection.

Third, MSAs were also excluded frompossible selection if the MSA wasdominated by BPCI Models 1, 2, 3, or 4episodes to such a degree that it wouldimpair the ability of participants ineither the CCJR model or the BPCImodels to succeed in the objectives ofthe initiative or impair the ability to setaccurate and fair prices. We anticipatethat some degree of overlap in the twoprograms will be mutually helpful for

both models. There are two steps to thisexclusion. First, we looked at thenumber of LEJR episodes at BPCI Model1, 2 or 4 initiating hospitals and second,the number of LEJR episodes amongBPCI Model 3 SNF and HHA episodeinitiators. We set the first cut off for this

exclusion if, within an MSA, more than50 percent of otherwise qualifyingproposed CCJR episodes were in Phase2 of BPCI Model 2 or 4 with hospitalinitiators. We set the second cut off forBPCI Model 3, based on if either SNF orHHA BPCI Model 3 initiating providersaccounted for more than 50 percent ofLEJR referrals to that provider type, theMSA would be eliminated from thepossibility of selection. As a result ofthis third criterion, 4 additional MSAswere removed from possible selection.No MSAs were excluded based onSkilled Nursing Facility (SNF) or Health

Home Agency (HHA) participation inModel 3.Finally, MSAs were removed if, after

applying the previous 3 criteria theyremained eligible for selection, but morethan 50 percent of estimated eligibleepisodes during the reference year werenot paid under the IPPS system. Pleaserefer to the appenda for this proposedrule for the status of each MSA based onthese exclusion criteria, available athttp://innovation.cms.gov/initiatives/ ccjr/. After applying these fourexclusions, 196 MSAs remained to be

stratified for purposes of our proposedselection methodology.

(2) Proposed Selection Strata

Numerous variables were consideredas potential strata for classifying MSAsincluded in the model. However, ourproposal is intended to give priority totransparency and understandability of

the strata. We propose creating selectionstrata based on the following twodimensions: MSA average wage-adjusted historic LEJR episodepayments and MSA population size.

(a) MSA Average Wage-adjustedHistoric LEJR Episode Payments.

We were interested in being able toclassify and divide MSAs according totheir typical patterns of care associatedwith LEJR episodes. As astraightforward measure of LEJRpatterns of care, we selected the meanMSA episode payment, as defined inthis proposed rule. MSAs vary in theiraverage episode payments. The averageepisode payments in an area may varyfor a variety of reasons including—1) inresponse to the MS–DRG mix and thusthe presence of complicating conditions;2) readmission rates; 3) practice patternsassociated with type of PAC provider(s)treating beneficiaries; 4) variations ofpayments within those PAC providers,and 5) the presence of any outlierpayments.

The measure of both mean episodepayments and median episodepayments within the MSA wasconsidered. We propose to stratify by

mean because it would provide moreinformation on the variation in episodepayments at the high end of the rangeof payments. We are interested in thelower payment areas for the purpose ofinforming decisions about potentialfuture model expansion. However, theCCJR model is expected to have thegreatest impact in areas with higheraverage episode payments.

The average episode payments usedin this analysis were calculated basedon the proposed episode definition forCCJR using Medicare claims accessedthrough the Chronic Conditions

Warehouse for 3 years with admissiondates from July 1, 2011 through June 30,2014. Episode payments were wage-adjusted using the FY 2014 hospitalwage index contained in the FY 2014IPPS Final Rule, downloaded at http://www.cms.gov/Medicare/Medicare-Fee-

for-Service-Payment/ AcuteInpatientPPS/FY-2014-IPPS-Final-Rule-Home-Page-Items/FY-2014-IPPS-Final-Rule-CMS-1599-F-Data-Files.html.The adjusted payment was calculated bydividing the unadjusted payment by afactor equal to the sum of 0.3 plus the


http://innovation.cms.gov/initiatives/ccjr/http://innovation.cms.gov/initiatives/ccjr/http://innovation.cms.gov/initiatives/ccjr/http://innovation.cms.gov/initiatives/ccjr/


13/120


multiplicative product of 0.7 and thewage index value of the hospital wherethe LEJR was performed. Episodes in thedatabase with IPPS payments less than$4,000 for the DRG 469 or 470 case weredeleted as indicating that the hospitaldid not receive full payment for theLEJR procedure. We also truncated theepisode payment at the 99.9th

percentile of the distribution ($135,000)to limit the impact of extreme outliers.

(b) MSA Population Size

The second dimension proposed forthe CCJR selection strata is the numberof persons in the MSA. In deciding how

best to incorporate the dimensions ofurban density and availability ofmedical resources, a variety of measureswere considered, including overallpopulation in the included counties,overall population in the core area ofthe MSA, population over the age of 65in the MSA, the number of hospital bedsand the number of Medicare FFS LEJRprocedures in a year. The reason wedecided to include this dimension inthe strata definition is that these factorsare believed to be associated with theavailability of resources and variations

in practice and referral patterns by thesize of the healthcare market. Whenexamined, these alternative measureswere all very highly correlated with oneanother, which allowed the use of oneof these measures to be able tosubstitute for the others in the definitionof the stratum. From these alternativeapproaches, we choose to use MSA

population.In operationalizing this measure,

MSAs were classified according to their2010 census population.

(c) Analysis of Strata

The two proposed domains, MSApopulation and MSA historic LEJRepisode spending, were examined usinga K-Means factor analysis. The purposeof this factor analysis was to inform theprocess of which cut points mostmeaningfully classify MSAs. Factoranalysis attempts to identify and isolatethe underlying factors that explain thedata using a matrix of associations.Factor analysis is an interdependencetechnique. Essentially, variables

Documents

CMS Comprehensive Care for Joint Replacement Payment Model.pdf