Episodic care forms a payment methodology of increasing relevance to neurointerventional specialists and other providers. Episodic care payment models are currently recognized in both payment paths described by the Medicare Access and Children's Health Insurance Program (CHIP) Reauthorization Act (MACRA): the Merit-Based Incentive Payment System and Advanced Alternative Payment Models. Understanding the cost of care, as well as how such costs are shaped in the context of episodic care, will be critical to success in both of these paths.
Statistics from Altmetric.com
The Affordable Care Act of 2010 sought to rationalize the delivery of healthcare in the USA.1 ,2 Among a wide range of reforms, it created the Center for Medicare and Medicaid Innovation (CMMI), which set the stage for piloting new payment models focused on providing value over volume. From its outset, the CMMI studied two payment initiatives potentially impacting neurointerventionalists: the bundled payments for care initiative (BPCI) and accountable care organizations. Both of these approaches have been explored in detail in this journal.3–5 In early 2015, Secretary of the Department of Health and Human Services Sylvia Burwell articulated the Obama administration's vision6 by restating the federal goal to transition away from fee-for-service (FFS) medicine. This aim, which manifested in the administration's policies, was further studied in this journal, given its future criticality.7
The Medicare Access and CHIP Reauthorization Act (MACRA) became law in April 2015.8 As outlined previously, it pushed forward the agenda of transforming the FFS volume-based US healthcare delivery system into one that is value-driven.9 ,10 An important component of MACRA was expansion of episode groups, representing sets of all rendered services relating to a given clinical condition or diagnosis. The promotion of episode groups in MACRA further empowers regulators to evaluate and attribute the ‘cost’ of providing care to Medicare beneficiaries in a more granular fashion. Over the summer of 2016 the Center for Medicare & Medicaid Services (CMS) put out a call for nominations for a MACRA Episode Based Resource Use Measures Clinical Committee,11 with an ostensible goal of expanding cost performance measures for the Merit-Based Incentive Payment System (MIPS).12 Two of the authors of this manuscript (JAH and GNN) serve as representatives to that committee. To put this into context for JNIS readers, some initial episodes that the committee proposed reviewing directly related to neurointerventionalists. While that list is confidential, we can share the publicly developed episodes, which include one for ischemic stroke.13
The vast majority of neurointerventional (NI) specialists will, in all likelihood, interact with the Quality Performance Program, (QPP) the program established by CMS for MACRA implementation through MIPS. The Society of NI Surgery is making specific efforts to develop the Neurovascular Quality Initiative (NVQI) into a Qualified Clinical Data Registry (QCDR). The QCDR approach will address three of the four components (quality, improvement categories, and advancing care information) of the MIPS final score that determines payment adjustments, but does not directly address one category (cost).14
Episodic payment models (EPMs) seek to encourage care coordination, improve efficiency, eliminate waste, and lower cost by centralizing care around a clinical condition. The underlying concept of EPM is that a single ‘bundled’ payment is provided for all the work and services rendered in association with a single condition or procedure. Episodic care thus requires a clear understanding of resource use as models of this single payment approach reward those with lower costs. The episode groups being developed for measuring cost in MIPS have a direct parallel with episodic payments and, in fact, episode groups may in time evolve into episodic payments.
The CMMI was formed with the specific goal and expectation that it would develop and test novel delivery systems and payment models for improving healthcare quality while lowering costs.15 Of particular note for NI specialists are its exploration and testing of the BPCI approach. However, other early CMMI initiatives including the Oncology Care Model (OCM) and Comprehensive Care for Joint Replacement (CJR) offer clear lessons for neurointerventionalists.
As reviewed previously, BPCI includes four distinct models for participating entities.3 The first three models are based on retrospective payment history whereas the fourth involves a lump sum prospective payment to providers for the entire episode of care.16 The voluntary program was designed to be attractive to healthcare systems and to entice interested entities to apply for participation. For example, in BPCI, the vast majority of CMS savings are returned directly to the participants. Forty-eight disease-related groups (DRG), including ischemic stroke, are eligible for participation. Unfortunately, while 1468 healthcare entities are participating in this initiative, it seems that few are modeling around ischemic stroke.17 Specifically, during the initiative's first year, 15 (mostly teaching) hospitals participated in at least one non-surgical neurovascular episode, the majority of which involved patients hospitalized under an ischemic stroke DRG. Unfortunately for proponents of this model, significant changes in standardized allowed payments were not significantly different between BPCI and comparator (reference) episodes for this group of stroke patients.18 Bearing that in mind, for elective conditions such as unruptured aneurysms, it would be relatively straightforward to model and thus design an episode, allowing for differences in materials cost. Since the bundle is formed around a clear clinical condition familiar to and managed by NI specialists, the specialty could attractively work with its healthcare system to manage costs for the condition. The increasing granularity of episodic payment methodology may eventually allow their application in a far more complex group of NI centric conditions. An alternative perspective would reflect the reality that CMS might prefer to focus its energies on high impact measures and broaden the associated episodes rather than seeking new conditions or procedures. This type of approach would have the potential benefit of incrementally encouraging elements that CMS considers important in any future state —for example, populations-based payments and care coordination.
The aforementioned CJR and OCM initiatives were noted above to be potentially instructive for neurointerventionalists and thus bear further investigation. The CJR is a bundled payment for lower extremity joint replacement. Whereas traditional BPCI is voluntary, CJR participation is required for all eligible hospitals within predefined geographic areas. The rationale from CMS is clear: by combining all Part A and B payments within a 90-day global fee, impacted hospitals will be positioned to manage their total expenditures. CJR provides further insight for NI providers as it includes all facility and imaging payments (eg, rehabilitation, recovery, and follow-up that occur within the 90-day window). OCM is a multipayer initiative that extends beyond Medicare. NI specialists with limited Medicare exposure should take note, as the model's application for other insurers aside from Medicare will likely become more common for other episode payment models in the future. Moreover, participating oncology practices must offer their patients enhanced services including what could loosely be considered care coordination; services that NI currently provide although are not typically reimbursed in FFS methodology.
The MACRA proposed rule
CMS released a 962-page proposed rule specific for MACRA in April 2016, renaming the entire approach as the ‘Quality Payment Program’.19 A somewhat unexpected development in this proposed rule was the very low number of recognized Advanced Alternative Payment Models (APMs) at the time of its publication.15 Lucrative bonuses are available for clinicians participating in Advanced APMs.15 However, MACRA incorporates strict statutory requirements for Advanced APMs, including;
Use of certified electronic health record technology
Submission of quality measures that are comparable to those in the MIPS quality performance category
With some exceptions, bearing of more than nominal financial risk for monetary losses.
These requirements effectively excluded from being deemed an Alternative APM episodic payment approaches such as BPCI, OCM and CJR, as well as many other accountable care approaches that had been originally designed by CMS as APMs. These various value-based initiatives were subsequently rapidly modified to meet the criteria for Advanced APMs, thereby better reflecting the spirit of MACRA.20
The episodic payment model advances
CMS invited comments to its proposed MACRA rule. Many organizations and entities provided feedback including the American College of Radiology21 and the American Society of Neuroradiology.22 Both organizations had direct NI involvement in the creation of their response letters. Then, on 25 July 2016, CMS released the Advancing Care Coordination through EPM notice of rule-making.23 For reasons delineated below, we believe that this expansion will be a watershed moment for NI. In the rule, CMS proposed to create three new DRG-based EPMs that would fall under the domain of the CMMI. The EPMs will include the initial event and 90-day global care for acute myocardial infarction, coronary artery bypass graft surgery, and surgical hip/femur fracture treatment. The proposal also indicated that incentive payments would be made to hospitals for cardiac rehabilitation, thereby creating another episode.
Neurointerventionalists would do well to note the far more complex nature of the underlying clinical circumstances in these new episodes than those associated with lower extremity joint replacements. CMS is intentionally targeting conditions initiated in the hospital setting, thereby making the facility responsible for saving money and improving care. Many opportunities exist for hospitals to achieve such aims, including taking advantage of the relationship between inpatient and post-acute settings. These hospital-based EPMs will likely create downstream pressure on impacted providers to participate in episodic care. As in the CJR model and as opposed to traditional BPCI models, participation will be mandatory for hospitals and systems in predefined geographic regions. Importantly, this proposed rule allows far greater participation for EPMs in Advanced APM models through the creation of two distinct tracks, the details of which are beyond the scope of this review.16 Of note, despite the broadening opportunity for systems to participate in Advanced APMs, the Bundled Payment for Care Initiative that is perhaps most associated with NI care is not currently eligible for the benefits associated with Advanced APMs. Providers must thus be familiar with episodic payment, given the impact of such payments on both the MIPS final score as well as eligibility for Advanced APMs.
Enter the PTAC
APMs have suffered from the failure of specialists to participate in a robust manner despite the financial benefits of doing so. This perception was in part propagated by the initial exclusion of numerous episodic payments from meeting MACRA criteria for Advanced APMs. Physician-focused payment models (PFPMs) are designed to address this challenge. These are APMs that address a new issue or specialty as compared with existing largely primary care-focused models. Importantly, by design, these proposals are designed around physician services. The PFPM Technical Advisory Committee (PTAC) is a powerful group established by the MACRA responsible for reviewing proposed new PFPMs and to make recommendations to CMS regarding their approval.
Despite the impact of the EPM expansion by CMS in July 2016 in allowing a larger number of initiatives to qualify as Advanced APMs under MACRA, the large majority of specialists are likely to remain unable to voluntarily participate in these initiatives. CMS currently defines PFPMs as an APM that includes Medicare as a payer, a physician group practice or individual physician (rather than a facility) as the APM entity, and targets the quality and costs of physician services, for example by addressing physician behavior or decision-making.24 The PFPM is likely to provide the path by which NI specialists (ie, a new issue or specialty) will be able to move past MIPS and attain Advanced APM participation. It is important to remember that, even if one obtains PTAC endorsement and CMS acceptance of a PFPM equivalent of an advanced APM, the participant will still need to meet Medicare threshold requirements for QP or partial QP status.
As defined in the statute, the PTAC is composed of up to 11 members who are considered expert in aspects of care delivery that pertain to this type of alternative approach.8 The PTAC currently includes seven physicians with only two specialists, neither of which is specifically involved in the care of cerebrovascular patients. The PTAC will begin to accept proposals very shortly (before the end of 2016) and make initial recommendations to CMS in the spring of 2017. Physician payments for PFPMs are expected to begin in 2018. The PTAC is on target to meet these timelines.
Neurointerventionalists should note that there is no requirement for the PTAC to involve a content expert in a highly specialized area, although it may choose to consult with an impartial outside person with domain expertise. The optional nature of seeking domain expertise may concern NI specialists, given the absence of cerebrovascular representation amongst the two specialists on the PTAC. Finally, it is important to note that, like many committees associated with the legislature (eg, Medicare Payment Advisory Commission (MEDPAC)) or CMS (eg, the Specialty Society Relative Value Update Committee (RUC)), the PTAC is limited to an advisory role; its recommendations are not binding.
NI specialists, like the vast majority of healthcare providers, are currently paid for professional services using the FFS system. For all of its potential flaws, it is straightforward conceptually. A practitioner provides a service; that service has a defined value; and, in some form or another, that service is reimbursed.25 ,26 Multiple elements involved in the treatment of typical cerebrovascular patients, exampled by care coordination, do not have specific reimbursement associated with them.
Episodic care, by definition, provides a methodology for organizing activities around a diagnosis or procedure. It can reasonably be hoped that a bundled payment around a condition would lead to an optimization of care with potential diminution of cost. The providers and the facility would, in essence, regulate/manage themselves to make sure that resource utilization is warranted. Care would presumably be coordinated to, for example, limit expenses around procedures, use step-down units in lieu of intensive care, and facilitate transfer to the lowest cost site for treatment as early as possible. One can imagine meaningful discussions that would result among providers in their own space27 ,28 or between specialists with potentially different viewpoints.29
MIPS include four components that contribute to a final performance score that determines bonus payments or penalties. Qualified clinical data registries address three of the four components. The fourth component, cost, is typically difficult to get at and, at the current time, based entirely on billed claims. In order for NI specialists to obtain maximal reimbursement, a rationalization and clear understanding of underlying resource use and cost is important. Active participation in the MACRA Episode Based Resource Use Measures Clinical Committee will help achieve that goal. The clinical episodes that are currently being defined for determining cost in MIPS could translate to bundles for guiding payments in the future. An easy-to-understand current method of providing episodic care is the BPCI. At this time and at least until 2018, BPCI participants will continue to get paid via the MIPS system.
CMS continues to develop opportunities for greater participation in Advanced APMs. Updated CMS rules have retrofitted existing and new episodic approaches to a novel alternative payment architecture. The PFPM and its technical advisory committee, the PTAC, likely provide the best opportunity for specialist physicians to develop Advanced APMs that reflect the activity in their respective specialties. As specialty-specific Advanced APMs are proposed and potentially employed by NI physicians in the future, they will likely to some extent surround episodes of care. In this situation, understanding cost will be pivotal to achieving success.
The MACRA put forward two approaches to delivering healthcare in the USA: MIPS and Advanced APMs. To view MIPS as the linear descendent of FFS would be an incomplete perspective. By including cost as a critical component of a provider's final score, MIPS introduced an element of value-based purchasing to CMS payment methodology. NI was an area of interest for CMS in episodic care dating back to the launch of BPCI, as evidenced by the inclusion of ischemic stroke among the included disease states. Participants in that bundled payment (BPCI) approach will continue to be compensated by MIPS.
Updated rules have moved types of episodic care into the realm of Advanced APMs. The physician-focused payment approach holds promise for NI specialists as a method of participating in this methodology that removes the onerous reporting requirements of MIPS and provides a pathway towards better reimbursement. PFPM-based APMs of interest to NI will feature elements of pathology- and procedure-specific focus built on an episodic framework. Understanding cost is key to success, whether through MIPS or Advanced APMs.
Contributors JAH drafted the original manuscript. All authors reviewed the draft, provided meaningful edits, and contributed to the final version.
Competing interests JAH consults for Medtronic. ABR is supported by a Research Grant from the Harvey L Neiman Health Policy Institute.
Provenance and peer review Not commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.