OCM Participants Consider Costs of 2-Sided Risk

Barbara McAneny, MD

After 3 years of struggling to meet targets for improved care quality and cost efficiency, many of the 175 practices that have remained in the Oncology Care Model (OCM) of the Centers for Medicare & Medicaid Services (CMS) have a decision to make: Assume a share of the downside risk for failure to meet financial targets or exit the OCM by December 3.

OCM critics and proponents agree that the voluntary program has improved many aspects of how participating practices provide cancer care. But they say that aspects of the program remain challenging and that good results by CMS’ standards are difficult to achieve, partly because the system is so complex that practices can’t tell if they are doing well or not. For this reason, many are leery of taking on downside risk, also referred to as 2-sided risk.

CMS recently issued practices their performance results from the program’s fourth period, which spanned January 2 to December 31, 2018.¹ Notably, it was the last report that OCM participants would get before deciding whether to make the leap to 2-sided risk.

One of those participants is Barbara McAneny, MD, a managing partner of New Mexico Oncology Hematology Consultants, Ltd, which has locations in Albuquerque and Gallup. “I think that [CMS] has the flaw of not understanding that a couple of adverse years with a 2-sided risk plan could put little practices like mine out of business,” McAneny said. “If we take 2-sided risk and we spend more than Medicare’s target [for spending] that year, we could probably pay it back using cash flow, but…a second year [of underperformance] would shut us down.”

New Mexico Oncology Hematology Consultants fell short of CMS’ targets in a previous performance period because 2 patients were admitted to a specialized nursing facility for treatment for costly diabetic leg ulcers, for which the practice was held responsible by CMS, McAneny said. The 2 admissions cost $100,000 apiece. “Those 2 patients meant we missed the target even though we had nothing to do with managing those conditions,” McAneny said.

Oncologists and industry observers said that smaller practices are less able to absorb the impact if CMS doesn’t pay them for all oncology care costs. “We don’t have a big pot of reserves—we just have whatever we earn in a given year,” said McAneny, who is president of the American Medical Association.

Practices that have earned payment incentives, called performance based payments (PBPs), for meeting improvement targets do not have to assume 2-sided risk. They can remain in 1-sided risk, which means that they enjoy a share of upside gains from cost savings but not downside risk.

McAneny said her practice will probably be in that latter group: “We are far enough under the target price that I think we will be able to remain in 1-sided risk and not have to stop the program.”

It's more likely that the larger practices will be able to take on the risk from a 2-sided arrangement, said Bo Gamble, director of strategic practice initiatives for the Community Oncology Alliance (COA). “The practices that are looking at 2-sided risk have the resources to do it, and some are also reviewing insurance in case they do have poor performance. I don’t anticipate that many small or medium practices will take 2-sided risk—it’s just too risky,” he said.

He agreed with McAneny that “[some] practices are working off such a small advantage of financial margin that a [poor performance period] could potentially close their doors.”

Although they relate to practice events that happened a year ago, the latest round of performance results from CMS will give practices useful insight into whether they are ready for 2-sided risk, said Alyssa Dahl, director of informatics for DataGen, a Rensselaer, New York—based consulting business that is working with about 20 oncology practices enrolled in the OCM.

“I think a lot of practices were anxious to see their initial performance period [PP] 4 results; it’s something that practices were eagerly awaiting, and they’re going to be spending a lot of time delving into what the results mean and deciding whether or not they want to make a risk decision to go into 2-sided risk for the upcoming performance period that starts in January,” Dahl said.

Integra Connect of West Palm Beach, Florida, is a health consulting company that works with 14 practices and 964 practitioners participating in the OCM. Charles Saunders, MD, chief executive officer of the company, said a higher incidence of cancer diagnoses that necessitated higher-cost treatments caused several of Integra’s practices to perform worse in the fourth performance period than in the preceding period. “We saw a positive trend between periods 1, 2, and 3 where [ there was] consistent improvement,” Saunders said. “We saw that trend reverse a little bit in period 4, where there was worse-than-expected performance across the board.”

Saunders said part of this shift was due to costlier care episodes, specifically for cases of multiple myeloma and lung cancer, 2 diseases that he calls “financial losers from an OCM standpoint.”

“The case mix was particularly [detrimental for cancer] centers that have a high percentage of tertiary referrals because those patients tend to be sicker than others, and CMS doesn’t fully adjust for the severity of illness in cancer,” Saunders said.

CMS failed to consider stage of cancer and its impact on the costs of care when it formulated its target cost of care, causing it to be “unrealistically low and unachievable for most practices,” according to Saunders.

He said also that prices of cancer agents—specifically, for Medicare Part D drugs—rose since the OCM began and contributed to plunges in performance among Integra Connect’s practices.

“The cost of cancer agents [novel or otherwise] probably increased more rapidly than CMS accounted for because, compared with the historical periods that predated the OCM, the cost of drugs wasn’t rising as rapidly as it actually did during performance periods 1 through 3, so that [led to] a target price that was too low,” Saunders said.

The problem with financial targets is that they can put a practice in an uncomfortable position, according to McAneny.

“When we looked at the drugs that caused us to miss the target [in a past period], a lot of them were new biologics for cancer care, which put us in a dilemma,” McAneny said. “I’m not going to give a patient an inferior treatment so I [can] hit a financial target, but we’re put in a situation where we know that if we treat the patient the best way we should, we’re not going to hit the target.”

Integra Connect’s practices are divided on whether to continue their involvement in the model or withdraw.

“About half of the practices that have not received a PBP are seriously considering dropping out of the program, and about half that did not receive the bonus are opting in because they have confidence in the programs they’ve put in place and have worked on over past performance periods and [feel] that [these programs] are likely to gain traction,” Saunders said. He noted that Integra Connect is now working with the practices that are entertaining 2-sided risk to model the impact of stop-loss insurance and determine whether it will be beneficial.

Although 2-sided risk can be cause for concern for smaller practices without an established record of cost-mitigation success in the OCM to date, it can be enticing for those who have achieved ≥1 PBP in the past. Practices that assume 2-sided risk can potentially earn more from CMS than those who stick with 1-sided risk. “Practices that have been consistent high performers could maximize their savings [in 2-sided risk] if they can maintain that level of performance,” Dahl said. Yet, as Gamble notes, some practices are unclear about why they have performed well or poorly under the model. The data are complicated to parse: “Two-sided risk could provide a financial benefit to your team if you have a track record of doing well, but the caveat is that most teams don’t realize why they’re doing well or not doing well,” he said.

“Those that do well can sometimes point to specifics, such as low utilization or increased hospice, but they can’t specifically say 80% of [the success] was due to this and 20% was due to this. They can’t take it down to that granular level; it’s all speculation, primarily because the factors that determine their success are changing. There are so many unknowns,” Gamble said.

For all its problems, the OCM seems to have improved the quality of care, although the financial health of practices participating has not fared as well, according to a recent Genentech survey, in which 79% of 43 OCM participants said that care quality improved since they began participating in the program.² Nearly half (47%) said their involvement in the model caused them to experience higher costs compared with 19% who did not. Results showed that OCM participation rose from 15% to 21% between 2017 and 2018.

Two-Sided Risk

For practices that remain in the OCM and accept the 2-sided risk arrangement, the risk plan that a practice chooses and how well the practice can align with CMS’ target costs of care are important factors.

For practices that take on 2-sided risk, CMS makes it easier to meet targets for costs of care. CMS first establishes the acceptable cost of care, which it then lowers using a “discount.” The deeper the discount, the tougher it is for practices to align their costs of care with what CMS is willing to pay.

Practices that stay with 1-sided risk must accept a 4% discount, but practices that accept 2-sided risk work with more favorable discounts of 2.75% or 2.5%, based on the type of model they choose.1

Under the original 2-sided risk plan, practices are responsible for expenses that exceed the target costs. The alternative 2-sided risk plan allows a “safe zone.” If the sum of a practice’s expenditures lands somewhere in this zone, the practice will not earn PBPs, but it will not be required to pay back costs as practices in the original 2-sided-risk plan would.¹

Research from Avalere indicated that most OCM participants would fare better under the alternative 2-sided risk model than the original. Findings showed that roughly half of practices in the alternative risk arrangement would owe CMS money versus approximately 70% of practices in the original 2-sided risk plan.³

The OCM will conclude in June 2021, but CMS is already looking ahead to its next-generation bundled payment system.¹ In November 2019, CMS proposed a successor to the OCM, the Oncology Care First model. Developed through the Center for Medicare & Medicaid Innovation (CMMI), this volunteer, 5-year model would begin in January 2021 and be modified based on practices’ experience in the OCM.⁴

CMMI initiated a public comment period through November 25, 2019, to field feedback from potential stakeholders and OCM-participating practices. Although the call for remarks aligns with the goal to design a practice-centric, value-based care follow-up to the OCM, CMMI has been criticized for its timing. Notably, the COA said the amount of time that CMMI allotted for practice response was insufficient because many practices were busy trying to decide whether to continue with the OCM.⁵

References

1. The Physicians Advocacy Institute’s Medicare quality payment program (QPP) physicians education initiative: Oncology Care Model (OCM). Physician’s Advocacy Institute website. physiciansadvocacyinstitute.org/Portals/0/assets/docs/Advanced-APM-Pathway/Oncology%20Care%20Model%20Overview.pdf?ver=2019-05-10-093131-197.
2. Genentech. The 2019 Genentech oncology trend report: 11th edition. genentech-forum.com/content/dam/gene/managedcare/forum/pdfs/Oncology-Trends/Genentech_2019_Oncology_Trend_Report_Digital_version.pdf. Published 2019. Accessed November 11, 2019.
3. Kane R, Bekele B, Grady L, Adelsberg M, Whittam D. More than half of all OCM providers could owe CMS money if required to join in 2-sided risk model. Avalere website. avalere.com/press-releases/more-than-half-of-all-ocm-providers-could-owe-cms-money-if-required-to-join-in-2-sided-risk-model. Published May 21, 2019. Accessed September 23, 2019.
4. Oncology Care First model: informal request for information. CMS website. bit.ly/2Ca1Sjd. Accessed November 11, 2019.
5. Community needs more time to assess and comment on CMMI’s proposed Oncology Care First Model. Community Oncology Alliance website. communityoncology.org/community-needs-more-time-to-assess-and-comment-on-cmmis-proposed-oncology-care-first-model/. Published November 5, 2019. Accessed November 14, 2019.