Article
Author(s):
The Community Oncology Alliance (COA) has been working with its member practices and some payer partners who have successfully implemented innovative care delivery and payment models to develop a 2.0 version of CMS’ Oncology Care Model (OCM). A progress report was presented at COA’s Payer Exchange Summit held October 29-30 in Tyson’s Corner, Virginia.
The Community Oncology Alliance (COA) has been working with its member practices and some payer partners who have successfully implemented innovative care delivery and payment models to develop a 2.0 version of CMS’ Oncology Care Model (OCM). A progress report was presented at COA’s Payer Exchange Summit held October 29-30 in Tyson’s Corner, Virginia.
Moderated by Kavita Patel, MD, MS, co-founder, Tuple Health, and nonresident fellow, Brookings Institution, participants included Bo Gamble, director of strategic practice initiatives, COA; Bruce Gould, MD, president, COA, and medical director, Northwest Georgia Oncology Centers; and Lalan Wilfong, MD, vice president of quality programs, Texas Oncology.
Patel clarified that COA’s OCM 2.0 is not a CMS-developed model, rather “we are thinking of a more universal approach to develop models that can be used for Medicare and by commercial payers.”
Wilfong reiterated Patel’s explanation, saying that the goal of their collaboration was to develop a payment model template that can be used to frame a new payment system for cancer patients that can be used by Medicare, commercial plans, and by self-insured employers, irrespective of their type or size, whose ultimate mission is to provide cost-effective quality care.
He added that the focus of their meetings and discussions was what we liked or did not like about current payment models and learning experiences for practices that have not been through value-based models or have not negotiated contracts or looked at reconciliation reports.
Wilfong then highlighted some of the key features of OCM 2.0. It is vital, he explained, to maintain a consistent and frequent channel of communication among collaborators. Open, timely, and frequent communications and reporting are vital to fine-tune, implement, and monitor the model with a goal to meet at least on a quarterly basis. Trust among collaborators is vital for this model to work.
A few of the other considerations within the model include:
Wilfong said that the model should have a base of 5-7 base quality measures and additional measures can be included following a payer-provider agreement. These measures have to be patient-centered, he said, and can be used a multiplier when calculating payment.
Risk adjustments in this model would be based on grouping by primary ICD-10 diagnosis codes with the same first 3 digits. Finally, shared savings would be allocated to payer, provider, and the employer associated with the patient care team.
Following a preview of the universal payment model, Gould highlighted key features of CMS’ OCM model that providers appreciate. “We like that it’s open to all payers. Also, there are no boundaries on when an episode may begin for a patient,” he said. Currently, the episode begins when the patient receives treatment for the first time.
“The monthly enhanced oncology services payment is vital to help practices transition to cost-effective and efficient care and enhanced services,” he said.
Finally, winsorization, whereby the Center for Medicare & Medicaid Innovation makes adjustments when calculating performance-based payments (PBPs) so practices that are 5% or lower or 95% and higher, their costs of care are not included when calculating their PBPs.
However, OCM as it currently exists has several problems, Gould explained and listed them for the audience:
1. OCM is prescriptive. “We would recommend a more comprehensive model that follows Oncology Medical Home [OMH] accreditation, which has been developed by COA.” Key elements of a comprehensive OMH are:
2. Attribution remains a significant issue; Gould said, and it can be overcome by using G codes: practice can register patients, so it’s clear that the patient belongs to that specific practice.
3. Reporting and measures are a problem that have evolved from poor technology interfaces, too much data entry, and too many measures. A solution for this is better coordination with CMS and their contractors to develop seamless, user-friendly interfaces and harmonization of measures.
4. Performance-based payments are a problem, too, because they are based on claims data, not clinical data. “Baseline prices do not reflect true drug costs, and the solution lies in considering both clinical and claims data when calculating baseline prices and for risk adjustment,” Gould said. “It is important that the model keep up with real-time changes in indications and price of a drug. Also, a shared savings model might be a better calculator than a gains savings model. We like a lot of what we have experienced with OCM 1.0 and we’d like to see it improve,” Gould said.
Finally, Gamble spoke to the audience about drugs and their value proposition in OCM 2.0.
He said that COA has had several meetings with pharmaceutical manufacturers to understand their progress and approach to explain the value of their product. “They have a different set of challenges that are both regulatory and cultural in nature,” he said. Additionally, manufacturers are also wrestling with measures for ‘value’.
Gamble listed a few key considerations for drugs and value-based arrangements (VBAs) in COA’s universal payment model:
“COA wants to be a facilitator for these discussions with providers, for payers and the pharmaceutical industry,” Gamble said.