Upcoming CMS Payment Models And the Importance of AI

Carol McCall
January 6, 2021

As an industry, healthcare is undergoing a transformation.  Its long-standing business model — which rewards the number of services performed as opposed to the outcomes those services achieve — is in sharp decline.  Fee-for-service medicine is giving way to value-based care that incentivizes and rewards personalized care, prevention, and the management of chronic illness.

To thrive in this environment, healthcare organizations are adding skills, capabilities and technologies to enhance their ability to deliver value. If your organization hasn’t started yet, you need to pick up the pace because as 2020 draws to a close, more changes are right around the corner.

Value-Based Payments: Here to Stay

In early 2021, the Centers for Medicare & Medicaid Services (CMS) will take another bold step to underline their commitment to healthcare’s transformation and will implement several new payment models that demonstrate their “strong commitment to advancing value-based care” that can “lower costs and improve health outcomes.”

The CMS Primary Cares Initiative will expand reimbursement for primary care.  It is designed to support greater attention to patients with costly, chronic conditions and to enable providers to spend more time interacting with high-risk populations.  The new primary care payment models are predicated on maximizing the value and quality of care patients receive throughout the healthcare system and CMS considers them to be potential game-changers.  They have also set aggressive goals.  CMS is targeting the program to cover 11 million Medicare beneficiaries and, when introducing the models, said their goal was to “dismantle” fee-for-service payments.

 

The Primary Cares Initiative is not CMS’s only effort.  CMS has continued to roll out other new payment models, including those for cancer patients and Chronic Kidney Disease which, despite delays to deployments caused by COVID-19, are also set to commence or expand in early 2021.  Moreover, CMS recently issued guidance to state Medicaid directors to accelerate the adoption of value-based healthcare in an attempt to both realize its advantages more quickly and better respond to unforeseen challenges, such as the COVID-19 pandemic.

Population health experts and medical practitioners have lauded the benefits of value-based care for over a decade, and while the transformation has sometimes seemed slow, the widespread adoption of these payment models is finally becoming a reality.

Value-Based Payments: Opportunity or Threat?  

Whether this new reality becomes an opportunity or a threat is not completely clear.   This is partly because the transformation to value-based care (VBC) is disrupting nearly every aspect of how care is organized, delivered, measured, and reimbursed.  The process is reshaping the industry and its key segments, including where profit pools lie and who gets them, and is creating opportunities for new entrants and start-ups with more compelling offerings.

 

VBC is more than simply focusing on high cost patients or certain conditions.  It redefines care as systematically addressing the preventive and chronic care needs of everyone (including the social and environmental determinants that affect health).  Success will demand having the ability to prioritize outcomes, systematically and proactively identify at-risk patients, foster health promotion, and exchange and share information with the broader health system.

Value-based payments are vital because they reward organizations that are able to achieve better outcomes.  By themselves, however, such payment models are not enough.  They must be coupled with the AI and analytic technologies that enable organizations to achieve better outcomes with a high degree of certainty.  The industry’s traditional rules-based methods are simply inadequate to the task; they are incapable of prioritizing opportunities, of learning directly from data, or offering insights into what actions may be beneficial.  Absent those kinds of capabilities, systematically producing better outcomes is next to impossible and value-based payments become more threat than opportunity.

 

AI in Practice: Examining Chronic Kidney Disease (CKD)

To better understand why integrating AI-based models with value-based payment mechanisms is so critical, it helps to look at a specific example.  Consider one of new models CMS specifically introduced to target Chronic Kidney Disease (CKD).

 

CKD is a complex, clinically dynamic, and progressive condition.  It is increasingly common yet suffers from low awareness, inadequate treatment, fragmented care, poor outcomes, and high healthcare costs.  In fact,

  • 15% of US adults—37 million people—are estimated to have CKD
  • Most (9 in 10) adults with CKD are unaware of their condition—especially in the early stages, and 1 in 2 people with extremely low kidney function (but not on dialysis) don’t know they have CKD
  • People with CKD experience an excess of complications and adverse events which commonly become life threatening.  An example is hyperkalemia, which is detected in more than 35% of CKD patients, which has a mortality rate of 25% (versus 10% for CKD patients without hyperkalemia).

Successfully managing CKD requires close surveillance and continuous monitoring so that clinicians are able to balance its clinical, medical, and psychological effects while avoiding its excess of adverse events.

 

Until now, CKD has traditionally been overlooked in value-based care programs.  This is partially due to artificial boundaries that have traditionally segmented the management of kidney disease.  It has also been challenging for payers to invest in long-term outcomes when beneficiaries leave before the economic impact of investments of better outcomes have been achieved.

   

CMS new payment models aim to address this.  Healthcare organizations now have an opportunity to invest in programs that cover a larger spectrum of the disease and create comprehensive care delivery and health management initiatives capable of generating better health outcomes.

AI-based models are ideally suited to enable such efforts.  Data from EHRs, labs tests, and medical and pharmacy claims can be used in models designed to address CKD’s specific challenges.  Predictive models can support a comprehensive approach to care and empower organizations to reimagine their CKD programs with broad care management goals while addressing specific population needs.  AI can enable comprehensive CKD care programs to:

  • Promote early diagnosis (e.g. identify at-risk diabetics and suspect CKD patients)
  • Slow CKD progression (e.g. identify patients at-risk of rapid progression)
  • Anticipate adverse events (e.g. hospital-acquired complications, nephrotoxic medications, anemia, or hyperkalemia)

Programs can be designed to achieve these goals but without the capabilities that AI provides, they will be unable to systematically improve outcomes and will not succeed in a value-based care paradigm.

For a more detailed analysis about using AI to improve CKD outcomes, you can download our white paper, The Importance of AI in the Management of Chronic Kidney Disease.  

For more information about the complementary technologies and competencies required for a value-based healthcare system, you can download our white paper, Precision Health Intelligence.

Register and get updated on events and news from ClosedLoop

We add new resources regularly. Enter your email address to get them directly in your inbox.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.