The episodes of care payment model is designed to put the patient at the center of care delivery, with their care coordinated among all providers for more comprehensive care to achieve the best possible outcomes. Unlike other disease areas, oncology care is not as clear cut; for instance, there is tremendous variation in care even at diagnosis, and care has no set start and end date. Nonetheless, the benefits of the episodes of care model are far-reaching and are worth the effort of overcoming design and implementation barriers.
The episodes of care model, also often referred to as bundled payments or episode-based payments, is a model that is increasingly being used to address care for individuals who require care from specialists (ie, in addition to primary care practitioners). It is a model that puts the patient at the center of care delivery, seeking to optimize care and costs of care by reducing unnecessary variation across the full continuum of care. The ultimate goal is to ensure that all providers who touch the patient during the course of their care are coordinated, communicating and collaborating to create a comprehensive and holistic plan for that individual that, ideally, will lead to the best outcome and the most optimal use of limited resources.
In recent years, many payers began testing the episode of care model with procedural episodes (eg, hip replacement, knee replacement surgery), because they are fairly straightforward from a construct design perspective. An “episode” is triggered by the surgery and typically includes the preoperative preparation and risk assessment (ie, labs, radiology before surgery), the surgery itself (the episode trigger), and the post-surgical care (eg, physical therapy, home care, wound care, additional hospitalizations, etc). It is easy to define what the trigger is (ie, the surgery), what clinical intervention or analysis will mark Day 1 of the episode (eg, some number of days prior to surgery to ensure capture of the pre-surgical workup), and when to end the episode (eg, some number of days after surgery to ensure capture of the post-surgical care). While there may be some variations between exactly how many days before and after the surgery are included in the episode, based on stakeholder perspectives and goals, generally, the model is easily defined. Typically, once patients have the surgery and recover, the episode is completed.
Episodes of Care for Oncology and Other Chronic Diseases
Cancer and other chronic conditions are different; care is not quite as neatly defined. In most cases, they do not share a common initial starting point for the episode nor a universal fixed endpoint. In particular, in the case of cancer, there is tremendous variation in the initial diagnosis of the disease. Patients with breast cancer, for example, may be diagnosed at various stages of disease; the cancer may be adjuvant or metastatic. The patient’s HER2 receptor status further differentiates the patient’s disease state and likely response to treatment. In addition, there is no standard beginning and end as there is in a procedural episode. This does not mean, however, that we cannot build a very effective episode of care model for individuals with cancer. It requires a bit more precision, but it can be done effectively following the same criteria as in procedural episodes (Box 1).
The goal is to capture patients who are clinically alike and who would be expected to have similar outcomes. In other words, you want to take a snapshot of the individuals in the group at about the same point in their disease state to ensure comparable and meaningful measurement. This can be very difficult for a number of reasons. First, payers do not have enough information on claims alone to be able to identify the patient’s current disease state (ie, claims data does not include stage of disease, size of tumor, HER2 receptor status, etc). Just effectively stratifying patients with cancer diagnoses requires an immediate collaboration between payer and clinicians and a close examination of medical records as well as claims.
Second, because the stratification of this population requires more precision than those that require only claims data, it is possible, and often likely that you will wind up with a very small population of patients tied to a particular physician group practice, which makes it statistically difficult to study and therefore create effective clinical and other pathway protocols. It also makes it difficult to get the attention of a payer or even a provider to engage in this kind of model if the number of patients is not significant. Stratifying based on all of the breast cancer patients connected to a payer and extrapolating from that, or creating virtual episodes, are two ways to address this particular issue.
Additionally, patients with cancer often do not complete their treatment or their disease remains active, after the first episode period. So again, unlike individuals going through procedural episodes, patients in oncology episodes will have one episode end and then immediately roll over into a follow-up episode. Moreover, because year 1 of the disease and treatment are often radically different from the follow-up years, those episodes and their goals need to be defined differently and with different outcomes measures.
Further complicating oncology episodes is the fact that new drugs and technology are coming to market very quickly, sometimes with insufficient information about side effects and impact to patient outcomes. These innovative treatments also often come at significant cost. It is critical for those building episodes around cancer and other chronic conditions to create enough flexibility to address these issues and continually refine the model.
The Benefits Outweigh the Challenges
While building episodes for cancer and other chronic conditions is more complicated than for the more typical procedural episodes, the benefits to these patients cannot be overstated. Let us consider the current patient experience in cancer. Patients managing a diagnosis of cancer are often terrified, in pain, and left—as an unfortunate result of the fee-for-service (FFS) payment methodology—to navigate the complexities of the health care spectrum on their own. They are also often treated to duplicative and unnecessary, sometimes painful and/or harmful services, as a result, again, of the disjointed care that has grown out of the FFS payment model.
There are a number of payers, both governmental and private, who have begun to implement value-based care models for individuals with oncologic diagnoses. The CMS Oncology Care Model, which is a 5-year pilot, launched in 2016. The program has financial and performance accountability metrics for episodes of care surrounding chemotherapy administration.1 The goal of the program is to provide more coordinated, higher quality care at the same or lower cost to CMS.1
Cigna has partnered with Memorial Sloan Kettering Cancer Centers (MSK) to better coordinate care for patients with cancer. Patients are assigned a care coordinator from MSK to help navigate the most appropriate care and also a care coordinator from Cigna to help navigate and optimize patient benefits.2
CVS’ Transformation Oncology Care model, which has been adopted by Aetna, uses genomic testing results to expedite the therapy prescribing process and match patients to appropriate clinical trials.3 And Horizon BCBS of New Jersey employs both an Episode of Care and Specialty Care Medical Home model to engage oncology practices and to improve patient outcomes, experiences, and cost of care.
Subsequent articles in this column series will explore in-depth other existing oncology payment models and newly created models. Articles will detail the model structures, targeted patient populations, and how payers are working with providers to execute such models.
Conclusion
The goal of episodes of care or any value-based care model is to ensure that the patient is at the center of everything, with their care coordinated among all providers, to ensure that they receive the comprehensive care that is required to achieve the best possible outcomes. There may not be a more appropriate group of patients as cancer patients, those whose chronic disease spans multiple body systems, specialties, and caregivers.
References
1. Centers for Medicare & Medicaid Services. Oncology care model. Innovation CMS. Updated February 4, 2020. Accessed March 3, 2020. https://innovation.cms.gov/initiatives/oncology-care/
2. Cigna collaborates with Memorial Sloan Kettering to improve care for people undergoing cancer treatment. News release. Cigna Corporation. July 24, 2019. Accessed March 3, 2020. https://www.cigna.com/newsroom/news-releases/2019/cigna-collaborates-with-memorial-sloan-kettering-to-improve-care-for-people-undergoing-cancer-treatment
3. CVS Health launches precision medicine program for payers. J Clin Pathways.
December 20, 2019. Accessed March 3, 2020. https://www.journalofclinicalpathways.com/index.php/news/cvs-health-launches-precision-medicine-program-payers
