Episodic Bundled Payment Models

Christopher Doria Skeehan, MD, FAAOS

James D. Slover, MD, MS, FAAOS

Ran Schwarzkopf, MD, MSc, FAAOS

Dr. Slover or an immediate family member is a member of a speakers’ bureau or has made paid presentations on behalf of Pacira; serves as a paid consultant to or is an employee of Horizon Pharma; has received research or institutional support from Biomet and Smith & Nephew; and serves as a board member, owner, officer, or committee member of American Academy of Orthopaedic Surgeons, AJRR Hip Society Steering Comm Member, American Association of Hip and Knee Surgeons, Hip Society, and PCORI Advisor Board Shared Decsion Making. Dr. Schwarzkopf or an immediate family member has received royalties from Smith & Nephew; serves as a paid consultant to or is an employee of Intelijoint, Smith & Nephew, and Zimmer; has stock or stock options held in Gauss surgical, Intelijoint, and PSI; has received research or institutional support from Smith & Nephew; and serves as a board member, owner, officer, or committee member of American Academy of Orthopaedic Surgeons and American Association of Hip and Knee Surgeons. Neither Dr. Skeehan nor any immediate family member has received anything of value from or has stock or stock options held in a commercial company or institution related directly or indirectly to the subject of this chapter.

INTRODUCTION

Currently, total joint arthroplasty (TJA) is one of the most successful surgical procedures performed, generating 2.07 and 1.85 lifetime quality-adjusted life years per patient for total hip arthroplasty (THA) and total knee arthroplasty (TKA), respectively.¹ Although extremely effective at improving the lives of patients, the annual combined cost of these procedures accounts for more than $7 billion from the Centers for Medicare & Medicaid Services (CMS). This represents 1% to 2% of the entire CMS annual payouts to physicians and hospitals and is the single greatest procedural cost to the agency.² In 2014, approximately 1 million primary TKAs and THAs were performed in the United States, with nearly 2 million annual procedures anticipated by 2030.³,⁴ Although orthopaedic surgeons’ Part B payments represent only 12% of the total reimbursement for THA and TKA, surgeons control approximately 90% of the cost of care through clinical decision making and orders.⁵ In an attempt to encourage surgeons and hospitals to become better financial stewards of patient care, the US Congress empowered CMS with new legislation to create novel alternative payment models (APMs) for the delivery of TJA.

THE DEVELOPMENT OF BUNDLED CARE IN THE UNITED STATES

In an effort to contain the cost of surgical care, CMS developed the concept of APMs through the Affordable Care Act of 2010. In 2015, the Sustained Growth Rate was repealed and replaced by the Medicare and CHIP Reauthorization Act
(MACRA) which replaced the Sustained Growth Rate with Quality Payment Programs. Two pathways within the framework of Quality Payment Programs were the Merit-Based Incentive Payment System (MIPS) and APMs.

MIPS consolidated three already existing value-based fee-for-service programs into one. The program created four performance categories that would generate a final score that was used to adjust hospital or clinician reimbursement. MIPS was developed to be budget neutral, so both negative and positive adjustments to reimbursement were limited to 4% of the original rate. In contrast, APMs are a payment approach that gives added incentive payments to provide high-value care. APMs can provide structure that can apply a specific clinical condition, a care episode, or a population.⁶

What is Episodic Bundled Care?

Episodic bundled care is an APM where a triggered episode of care (EOC) is the basis for the treatment of a patient condition. All costs of care during the episode are tallied and any savings below a predetermined target price are shared with clinicians. Additionally, any expenses over the target price are assessed as penalties to the clinician. Therefore, the clinician or organization assumes risk, with responsibility for both the cost and outcomes of the care delivered during the EOC. Spending below the preset target price can establish eligibility for incentive payments, while spending above the target price will result in the clinician or organization being found financially accountable for the difference. APMs include the Bundled Payments for Care Improvement (BPCI), the Comprehensive Care for Joint Replacement (CJR), and the Advanced Bundled Payments for Care Improvement (BPCI Advanced).

Bundled Payments for Care Improvement

The CMS BPCI was established in 2011. This was a voluntary episodic bundled care program created “to improve patient care through payment innovation that fosters improved coordination and quality through a patient-centered approach.”⁷ It was considered an introduction to value-based payment systems in which hospitals or clinicians could pick the type of procedure or medical treatment to be included in the bundle. BPCI provided a flat target price that included all services rendered 72 hours prior to admission, any acute care services, and any services during the 90-day postacute period. This postadmission period included all postacute care services, such as rehabilitation and readmission costs that were related to the initial Medicare severity diagnosis-related group (MS-DRG) that triggered the episode. Most readmissions were considered related, with some studies demonstrating 92% of all readmissions were considered related.⁸ A few separate unrelated procedures were excluded from the bundle. As part of the program, CMS also audited charges that occurred between 90 and 120 days to ensure services were not being delayed to avoid the bundle.

Participants in BPCI could choose from one of four models for reimbursement:

Model 1: Retrospective payments for acute care hospital stays only, where a hospital was paid a discounted amount based on the inpatient prospective
payment system rates for specific MS-DRGs while physicians were paid separately under the Medicare physician fee schedule (PFS).
Model 2/3: A retrospective bundled payment arrangement where expenditures were quarterly reconciled against an EOC’s target price.
- Model 2: Acute and postacute care
- Model 3: Postacute care only
Model 4: A prospective bundled payment that encompassed all services performed by the hospital and clinicians during an EOC that lasted the entire patient stay.⁷

Model 2 was the most popular model utilized for lower extremity joint replacement (LEJR). BPCI for LEJR was initiated in two phases. Phase 1 commenced on January 1, 2013, which allowed participants to develop patient care pathways and cost savings measures prior to assuming full financial risk in phase 2. Phase 2 commenced on July 1, 2015, with phase 1 ending on September 30, 2015. Phase 2 was extended and completed on September 20, 2018.

Comprehensive Care for Joint Replacement

On April 1, 2016, the next iteration of bundled care, the CJR commenced. This was an involuntary program for those hospitals already enrolled in BPCI as well as other hospitals in 67 different metropolitan service areas. The program was proposed to end on December 31, 2020, but was extended for an additional 3 years in June 2020. On December 1, 2017, CMS excluded involuntary enrollment for rural hospitals and hospitals in 33 of the 67 metropolitan service areas, allowing clinicians to opt out of the program. Unlike BPCI, CJR included only MS-DRG 469 and 470 (Major Joint Replacement or Reattachment of Lower Extremity with or without major complications or comorbidities) with incorporation of ICD10 diagnosis codes for hip fracture into the target price. Similar to BPCI, participating hospitals were financially accountable for the quality and cost of the care rendered during the bundled period. Payments were made to the hospital following the standard fee-for-service model. Target prices for years 1 through 3 were established based on the hospital’s historic standardized spending and regional historical standardized spending. Target prices incorporated a 3% discount to create initial savings. Years 4 and 5 target prices were entirely based on regional pricing. CMS would perform an annual audit of the total expenditures for services rendered under Parts A and B, compare those against the target price, and based on the cost award the hospital additional payments or penalize them for overpayment. Quality metrics were used to alter the 3% discount rate and increase the savings for hospitals that performed well to ensure quality was maintained. This was accomplished by using a CMS CJR composite quality score. Those hospitals with higher quality scores would see a reduction in the discount rate from 3% to 1.5%. These quality bonuses were provided if the score was in the top 70th percentile in years 1 to 3 and top 60th percentile years 4 to 5 against the regional standard. Fifty percent of the quality score represented the NQF#1550 for hospital-level risk standardized complication rate following elective THA/TKA, 40% the NQF#0166
HCAHPS (Hospital Consumer Assessment of Healthcare Providers and Systems) survey, and 10% the voluntary THA/TKA patient-reported outcomes (PROs). In 2015, stakeholders recognized Hip dysfunction and Osteoarthritis Outcome Score for Joint Replacement/Knee dysfunction and Osteoarthritis Outcome Score for Joint Replacement and PROMIS Global Health or the Veterans RAND 12-Item Health Survey as acceptable PRO scores for reporting.⁵ From 2016 to 2020, the target price for an uncomplicated LEJR without femoral neck fracture in the CJR program decreased by 2.9% to 6.6%, depending on the region. Although this decrease in target price demonstrates that increases in efficiency can result in decreases in the regional cost per EOC for LEJR, continued decreases in target prices each year are quickly approaching a floor where participating institutions are no longer able to earn a net income from participation in bundled payments programs.

CJR did not initially include outpatient LEJR in bundles. The removal of TKA and THA from the inpatient-only list in 2018 and 2020 allowed clinicians to bill CMS using the outpatient prospective payment system instead of the inpatient prospective payment system. This avoided triggering a bundle with an anchor MS-DRG 469/470 hospitalization. With healthier, younger patients having procedures being performed as outpatients, this meant that those patients entering into bundles would be more infirm, medically complex, surgically complex, and/or older. Recently CMS released proposed rule CMS-5529-P, which extends CJR for an additional 3 years beyond the earlier end date of December 31, 2020. In addition to extending the program, the new rule will establish a blended rate MS-DRG 470 without hip fracture combining both outpatient and inpatient TKA/THA.

Although most patients who were enrolled in BPCI and CJR underwent LEJR for primary or secondary osteoarthritis, bundles could also include LEJR for patients with hip fractures, musculoskeletal tumors, and revision LEJR. Althausen and Mead⁹ demonstrated that hip fractures were more costly to treat in the bundle and therefore represented a greater financial liability in the bundled payment model for participating hospitals. Lott et al¹⁰,¹¹ demonstrated that despite decreases in length of stay, readmission, cost of care per episode, and increased discharges to home, hip fracture patients still had a total cost of care while utilizing an established BPCI pathway of $49,993 per EOC. CJR accounted for this by separating the fractures from the nonfractures and creating a different target price, allowing institutions to successfully treat these patients using the bundled payment system. Gammal et al¹² reviewed an administrative claims database within their hospital system for the oncologic and nononcologic cost of an EOC. They found the average cost of an oncologic EOC necessitating THA was $43,771 versus $23,779 for a nononcologic EOC. THA performed for oncologic reasons was also found to be an independent risk factor for greater EOC costs. Although not commonplace, some institutions opted to include revision LEJR in their bundled care arrangement. Courtney et al¹³ examined the appropriateness of bundles for revision LEJR. Although revision LEJR performed in a bundle did demonstrate a reduced length of stay compared to the prebundled cohort, the LEJR revision bundled group received less reimbursement from CMS for the index hospitalization with equivalent costs of care.

BPCI Advanced

The third CMS iteration of episodic bundled care in the United States is BPCI Advanced. This program started accepting its first cohorts on October 1, 2018, and its second cohorts on January 1, 2020. CMS made BPCI Advanced a voluntary program. Participation would exempt physicians from MIPS, but those hospitals already enrolled in CJR were not allowed to switch to BPCI Advanced. Enrollment criteria included an electronic medical record system, payments based on MIPS comparable quality measures, and entities willing to bear more than nominal financial risk. Similar to CJR, BPCI Advanced was a single retrospective bundled payment system with a 90-day EOC duration. Target pricing was calculated from the regional cost average, then a 3% cost savings discount was applied by CMS to set the price. Instead of just including MS-DRG 469/470 as does CJR, BPCI Advanced included 31 inpatient and 4 outpatient clinical episodes starting at model year 3. The clinical episodes included were back and neck except spinal fusion, double joint replacement of the lower extremity, fractures of the femur and hip or pelvis, hip and femur procedures except major joint, lower extremity/humerus procedure except hip, foot, femur, major joint replacement of the lower extremity, major joint replacement of the upper extremity, and spinal fusion. The program exempted hospitals from the reporting requirements associated with MIPS and qualified them for incentive payments. Payments would ultimately be tied to quality metrics. For years 1 through 3, those included all-cause hospital readmissions, hospital-level complication rates, hospital 30-day all-cause mortality rates, advanced care plans, and some perioperative care metrics, including selection of prophylactic antibiotic, excess days in acute care, and CMS patient safety indicators. During year 4, alternative quality measures would be available to use instead of the administrative quality measures, including five claims-based and registry-based measures as well as all-cause readmission and advanced care plan. Preliminary target prices were provided prior to each model year.

OUTCOMES OF EPISODIC BUNDLED CARE FOR HIP AND KNEE REPLACEMENT

The Bundled Payments Care Initiative and the follow-up program CJR were successful in achieving their goals of containing the cost of TJA, seeing significant reductions in the cost per EOC while maintaining or improving outcomes¹³,¹⁴,¹⁵,¹⁶ and improving hospital efficiency. In a systematic review by Agarwal et al,¹⁵ the authors noted that the bundled care programs were effective at maintaining quality while reducing cost of LEJR EOCs, but not other conditions. Most notably, Dummit et al¹⁷ who reported that those hospitals performing in BPCI relative to control hospitals reduced their EOC cost for TJA by $3,286 versus $2,119 in the control hospitals. After the first 2 years of CJR, Barnett et al¹⁸ found a 3.1% decrease in the cost to perform LEJR relative to those performed immediately prior to the implementation of the program.

Early participants in the program developed strategies to reduce costs to less than the initial target episode price set by CMS, were better positioned to retain
financial solvency when performing LEJRs on the Medicare population in future iterations of the program and in other bundled payment plans. For example, at a single, large urban, academic, tertiary care hospital, reforms were targeted at reducing readmissions, reducing the length of stay, reducing implant/supply/drug costs, reducing the time spent in the operating room, changing the discharge disposition of patients to more cost-efficient alternatives, and decreasing the use of unnecessary testing or consultations.⁸ These goals were achieved by developing standardized clinical pathways that directed the care of more than 90% of patients with LEJR, with exclusion from the pathways directed by set criteria rather than physician preference. Clinical care coordinators were hired to oversee patient involvement and facilitate adherence to the pathways, facilitate care at transition points, and provide a communication pathway from patients to clinicians.⁸ At the same institution, those interventions resulted in the average length of stay decreasing from 4.27 to 3.58 days, discharges to inpatient facilities decreased from 65% in 2012 to 44% in 2013, and readmissions decreased from 17% in 2011 to 11% in 2013.¹⁹ Furthermore, from 2013 to 2016, during that same institution’s involvement in BPCI, the length of stay decreased from 3.58 to 2.96 days. Discharges to an inpatient rehabilitation facility decreased from 44% to 28%, the 90-day all-cause readmission rate decreased from 13% to 8% and the cost per EOC decreased by 20%, demonstrating continued improvement is possible.²⁰ As a result, in February 2020, CMS issued proposed rule CMS-5529-P, which extended the program for an additional 3 years. It also proposed reforms to the process of target price reconciliation, risk adjustment for hip fractures by proposing the creation of a new MS-DRG for LEJR for hip fractures, and the inclusion of outpatient LEJR into bundles.²¹

FACTORS THAT CAN INFLUENCE PATIENT OUTCOMES IN EPISODIC BUNDLED CARE

One of the challenges for success in bundled payment models has been the lack of risk adjustment of target payments for the diversity of patients that enter into bundles. BPCI and CJR provide two risk-adjusted categories, LEJR with and without major comorbidities. CJR added additional risk adjustment by including a third pathway for hip fractures that was also implemented in BPCI Advanced.

Only gold members can continue reading. Log In or Register to continue