CMS Keeps Innovation Under Lock and Key

Gary Kirsh, MD

Several organizations are developing alternative payment models (APMs) they hope CMS will adopt as ways to encourage oncologists to deliver care at a lower cost while improving its quality. If approved, these proposals could become customized alternatives to the agency’s Oncology Care Model (OCM), which is being piloted by 190 practices, and the Merit-based Incentive Payment System (MIPS), the default Medicare reimbursement system for most clinicians.

However, as physicians, practice administrators, and advocates navigate the alphabet soup of federal payment reform, they continue to harbor serious doubts about the prospects for alternative models proposed by medical organizations. They note that CMS expects participants in advanced APMs, unlike those in MIPS models, to quickly take on heavy financial risk for meeting cost-saving and quality goals, and they say it is unclear how serious the agency is about approving physician-proposed payment arrangements. They also note the added uncertainty of a new administration and Republican efforts to repeal the Affordable Care Act, which had a major role in advancing the government’s promotion of value-driven care. The departure of experienced staff from CMS’s Center for Medicare & Medicaid Innovation (CMMI) could complicate the implementation of MIPS and APMs, and CMMI has been targeted for elimination by some politicians. At the same time, new CMS leadership might end up being more welcoming toward physicians’ reimbursement proposals.

“CMS has laid out a process, but we don't know how fast it’ll be to go through that process. We don't know what the regulatory burdens will be, whether the process will be efficient or inefficient, whether CMS will listen or won’t,” said Gary Kirsh, MD, immediate past president of the Large Urology Group Practice Association (LUGPA). “And at the end of the day, even if the process isn’t onerous, what will be their threshold for approving any of this? We have some ideas about what will be good, but we don't know if they’ll be interested in these things.”

LUGPA is developing one APM for localized prostate cancer and a second for sepsis after prostate biopsy. The sepsis model will include what Kirsh calls a “warranty” guaranteeing a lower cost for the bundle of services provided to each patient. Organizations that are submitting APMs include the American Society of Clinical Oncology (ASCO), the Community Oncology Alliance (COA), the American Society for Radiation Oncology (ASTRO), and the Society of Gynecologic Oncology (SGO).

The two-track system of MIPS and advanced APMs was created by the Medicare Access and CHIP Reauthorization Act (MACRA) of 2015 and codified in a final rule issued in October. Medicare providers may opt out and face a 4% payment cut in 2019; participate fully or partially in MIPS, which consolidated existing payment adjustment programs and added new elements; or join an advanced APM that balances a 5% bonus against sizeable potential penalties for underperformance. MIPS includes a number of nonadvanced APMs, some of which may be used in advanced form. Advanced APMs involve higher financial risk for providers than regular APMs, but the financial rewards may be higher, too.

MACRA calls for CMS to accept outside proposals for physician-focused payment models. The proposals will then be evaluated by a technical advisory committee, which can recommend models to the agency for testing as APMs or advanced APMs. Currently, the only oncology APM is the OCM, which supplements traditional fee-for-service reimbursement for chemotherapy care with monthly and performance-based payments. Medical organizations are eager to try out alternative models designed to spur efficiencies in other types of oncology treatments (eg, by testing clinical pathways).

“This has great appeal to us. Our ability to have large groups of urologists and allied professionals with business and clinical infrastructure get doctors to ‘swim in lanes’ is our sweet spot,” Kirsh said. “When we're able to do that, we’re going to deliver better care for the patient, and we’re going to get ourselves off fee-for-service. Fee-for-service is not our friend. It's a handcuff on our entrepreneurism and our creativity.”

Kirsh said the urology APMs would serve not just as payment models, but also as studies, allowing oncologists to test different prevention protocols or treatment methods. Those that prove most effective at the least cost could become standard methods. The proposed models could also be accepted by commercial payers even if CMS passes on them, he said.

Kirsh and others agree that the two-sided risk required in advanced APMs, with the real possibility of financial losses for physicians who do not meet a model’s benchmarks, is essential to promoting value-driven care. However, organizations like ASCO argue that requiring such substantial risk levels early on runs counter to the intent of MACRA. They say Congress invited payment proposals and created the Physician-Focused Payment Model Technical Advisory Committee (PTAC) to encourage testing of APMs. They contend that CMS has made it too financially risky to engage in bold experiments.

“CMS thinks the benefit to being in an APM is so high—this 5% bonus and being exempt from MIPS—that they really want to make sure that you have skin in the game if you're going to do this,” said Blase Polite, MD, chair of ASCO's Government Relations Committee.

“But we think the intent of Congress was, ‘We really want to encourage people to come up with APMs, and let’s test some things out and give some people credit for that,’” he said. “Congress wasn't looking to set a huge bar for APMs. They wanted a lot of APMs out there. We don't want CMS being the only arbiters and developers of APMs, and that's basically how the current system was set up.”

ASCO has been working on an episode-based payment model for a few years and will send CMS a version called the Patient-Centered Oncology Payment. This medical home-inspired model resembles the OCM, but differs in a number of ways, focusing more on rewarding physicians for reducing patient hospitalizations, among other elements.

Under MACRA, PTAC passes APM recommendations on to CMS but the agency is not required to test any of the models. SGO and other groups have urged CMS to make a rule saying it will accept PTAC recommendations.

For failure to meet performance goals under advanced APMs, CMS has proposed potential penalties of 4% of the total cost of care for each patient, including drug spend. This could represent a big chunk of a doctor’s Medicare revenues, amounting to 20% in some cases. Polite, whose practice is enrolled in the OCM and who is in regular contact with other participating practices, said that none of them are currently planning to join the two-sided risk version of the model. “This really puts you at huge, huge financial risk, to the point where you’d be almost crazy to do it as an oncologist,” he said.

Deepak Kapoor, MD, LUGPA’s Health Policy committee chair, said he favors using urology APMs in nonadvanced form with MIPS, which would exempt practices from MIPS’ most challenging performance categories while avoiding excessive risk.

A number of observers have told CMS that the structure of payment adjustments required in advanced APMs is too complex and will deter participation, especially by small and medium-sized practices. Polite said it’s still unclear how CMS will handle risk adjustment, which could change quality scores and payments based on patient comorbidities, cancer site, and other factors.

Even if CMS does prove interested in adopting physician-proposed APMs, it could take years before they are available. An agency spokesman said it takes CMS 18 months to develop its own APMs, and while physician proposals that meet its criteria may need less time, once they undergo PTAC review they will still go through the regular developmental process for APMs, including design changes, public announcement, and a request for applications.

The agency’s implementation of its sole existing oncology APM, the OCM, has already had some stumbles, said Robert“Bo”Gamble, COA’s director of strategic practice initiatives. He said CMS forced some physicians to either stop working with access hospitals in underprivileged areas or drop out of the program, and it didn’t thoroughly explain crucial performance reports to practices before the OCM launched. For patient surveys, practices are required to use a new paper document the agency developed rather than a widely used electronic survey, he said.

CMS is aware of physicians’ concerns and has taken steps to address them, but not to the extent that Polite and others would like. The agency softened the general MACRA compliance requirements, saying practices need only comply minimally this year to avoid penalties in 2019. It also put of finalizing the advanced APM risk requirements for a year and said it is committed to designing models with “stop-loss” amounts that would sufficiently limit potential penalties.

Stephen S. Grubbs, MD, ASCO’s vice president of clinical a airs, said he believes regulators want to see new risk models succeed. “I can tell you, from the conversations I've had with folks at CMS, they don't want to see practices going out of business because they took risk and got into trouble. That's not their goal here at all,” he said.

Grubbs said CMS might be able to ease advanced APM adoption by using risk corridors, much like the Affordable Care Act does, to subsidize practices that experience losses. Polite said that Rep Tom Price (R-GA), the expected next chief of the Department of Health and Human Services, has criticized mandatory models and may want CMS to ease the conditions for advanced APMs rather than corral oncologists into the OCM.

Others proposed APMs include ASTRO’s shared-savings model, called Radiation Oncology Total Cost of Care, which would apply to breast, lung, prostate, upper and lower gastrointestinal, gynecological, head and neck, skin, and central nervous cancers. SGO has been working on an APM for the surgical management of endometrial cancer and plans to create additional models. Gamble said COA is working on an OCM variant that would provide a uniform set of clinical pathways for use by multiple payers, both public and private.