In April, the U.S. Department of Health and Human Services (HHS) announced five new payment models for primary care, known as the CMS Primary Cares Initiative.
The five models are:
• Primary Care First
• Primary Care First – High Need Populations
• Direct Contracting – Global
• Direct Contracting – Professional
• Direct Contracting – Geographic
R. Shawn Martin, senior vice president, advocacy, practice advancement and policy at the American Academy of Family Physicians (AAFP), says the organization has partnered with CMS over the past few years to help develop proposed alternative payment models that include a strong foundation in primary care, and helped contribute to the development of this initiative.
“We have seen positive development and implementation of these new models,” he says. “While we still have some questions with respect to how the Direct Contracting models will work, we are very pleased to see the construct of the Primary Care First path strongly resembles core elements of the AAFP’s Advanced Primary Care Alternative Payment Model, which was presented to CMS in 2017.”
Chris Dawe, senior vice president, Medicare partnerships at Evolent Health and former White House policy advisor for healthcare, says the company is pleased to see the administration pulling out all the stops to showcase their commitment not just to healthcare value but the need for providers to assume meaningful financial risk for the patients they serve. Still, he understands there are more details that need to be worked out.
“We are still awaiting details on the benchmark methodology—how the Direct Contracting Entity (DCE) will be measured on costs—and the payment methodology,” he says. “We know the benchmark methodology will involve some combination of historical expenditures and risk adjusted Medicare Advantage expenditures, but the details matter.”
For example, how will regional variation be factored in? Will the payment rate allow for sufficient support of administrative costs and enough flexibility that DCEs can offer attractive quality incentives to the participating providers and preferred providers they have agreements with?
“We are also curious how robust the beneficiary incentives will be to encourage beneficiaries to seek care from high-value providers, and there are still details outstanding on how models overlap or interact with one another,” Dawe says. “Once those details are published, we will be able to assess the model and make decisions.”
Norman Chenven, MD, founder of Austin Regional Clinic, a 340-physician multispecialty medical group and vice chairman of the Council of Accountable Physician Practices, is excited about what he’s seeing in these new payment system proposals.
“There appears to be an effort to engage physicians and systems from a variety of settings including the spectrum from the rural primary care MD to the specialist in a large vertically integrated group,” he says. “We’ve learned a lot in the past decade about accountable care. MDs will be solicited to share their thoughts during the commentary phase and should help shape the outcomes for the better.”
He explains the CMS Center for Medicare and Medicaid Innovation (CMMI) was given broad authority to test new payment and delivery models that aim to lower costs while maintaining or improving quality. The Center is not required to go through the federal rulemaking process wherein the agency releases a proposed regulation for public comment and then a final regulation some time later.
CMMI’s process is to develop models with input from stakeholders, including physicians. This input comes in a variety of forms: private meetings and multi-stakeholder roundtable discussions, and/or soliciting input somewhat more formally in writing.
Pros and Cons
Martin says the general Primary Care First model will include a risk-adjusted, population-based payment, plus a flat visit fee for every face-to-face encounter a patient has with a primary care physician. It also provides performance-based payments up to 50 percent of revenue for practices, along with a small downside risk of 10 percent.
The general Primary Care First model will include a risk-adjusted, population-based payment, plus a flat visit fee for every face-to-face encounter a patient has with a primary care physician, Martin says. Physicians can face payment penalties as high as 10 percent and bonuses of up to 50 percent depending on their performance.
“One of the fears we always have and a caution we raise is that groups that have more sophisticated infrastructure tend to be better positioned because they have more resources to apply vs. a smaller practice,” he says. “This model offers a lot of opportunity to help practices move away from fee for service and in turn focus on patient-centered comprehensive and quality care. Offering this flexibility to physicians and their practices will be key in helping to drive innovation and forward-thinking care delivery in the primary care space.”
Furthermore, capping the risk at 10 percent is particularly important and appropriate, he says since many primary care practices have limited revenue and this mirrors the model that the AAFP developed.
What Evolent Health appreciated about this model is that CMS is, for the first time, incorporating some downside risk (10 percent), whereas CPC+ only featured potential payment bonuses, not penalties.
“The idea of the model is to provide upfront care management payments to physicians and an opportunity to earn performance-based payments to support care redesign and care coordination,” says Ashley Ridlon, vice president of health policy at Evolent Health. “To-date the CPC+ model has not produced net savings for Medicare but CMMI chose to create a similar modelwhich has similar eligibility and care requirements but with both upside performance-based payment potential (up to 50 percent of primary care revenue) and downside risk (10 pecent of revenue).”
She adds that while the model helps practices dip their toes in the water of performance-based risk, she fears that the levers aren’t strong enough to drive meaningful value relative to total-cost-of-care models with more two-sided risk.
“And while the model may benefit truly small practices in building capacity, we also worry that the continued proliferation of lower-risk options will make it difficult for Direct Contracting Entities to recruit primary care providers to their more advanced models,” Ridlon says.
Ripley Hollister, MD, family medicine specialist with The Physicians Foundation, says since the CMS Center for Medicare and Medicaid Innovation does not have a physician as part of its leadership structure, these models were proposed by government officials or representatives of large health systems who do not understand how such proposals will impact the average small physician practice.
“Setting aside the over simplified and flawed assumption that physicians are more concerned about their revenue than their patients, there are many concerns with this type of reimbursement model,” he says. “The CMS value is measured by any number of metrics, often requiring data collection and benchmarking costs that are beyond the means of many independent practices. In other instances, we may have one case that becomes a complex case, an outlier that consumes an enormous amount of resources that skews our entire practice’s metrics.”
Need for New Models
The Primary Care First–High Need Populations model focuses on providing primary care or care coordination for Seriously Ill Population (SIP) patients.
“Inside both the Primary Care First and Direct Contracting paths, physicians will be able to identify high-need populations and then be eligible to receive higher payments for more intense services, especially for homebased care,” Martin says. “This is a very positive and innovative program for caring for patients who are homebound. The big question is ultimately who the eligible SIP populations will be.”
The initial concern of this model for Hollister is that the “transformation” does not work if a small practice wants to take care of complex patients.
“High need populations are at higher risk for hospitalizations or remaining unhealthy, which would decrease the chances for performance-based payment,” he says. “We need more details to determine if it would be possible for an independent practice to mitigate the risks associated with high needs populations.”
The Direct Contracting Models
The Direct Contracting — Global option builds on the successful Next Generation ACO model and offers a full risk of total costs for patients.
“Besides offering full-risk, one pro of this option is the availability of multiple options for beneficiary alignment,” Dawe says. “Prospective, claims-based alignment can be helpful for providers to get the population scale they need to be successful in managing risk, and voluntary alignment options will help other entities new to Medicare fee-for-service build an aligned beneficiary population.”
Additionally, CMS is exploring allowing the same benefit enhancements and waivers that Next Gen ACOs have enjoyed (for example, the 3-Day SNF Rule Waiver, Post-Discharge Home Visits Rule Waiver), plus perhaps a couple of additional ones to help beneficiaries access services in their homes.
“We understand CMS intends to use a prospective blend of historical spending and adjusted Medicare Advantage regional expenditures to develop the benchmark against which the Direct Contracting Entities will be measured, which sounds promising, but with limited details, it is difficult to assess at this point,” Dawe says.
While the global package seems to mimic the typical Medicare Advantage arrangement, Chenven says there would be a need to invest in staff and systems in order to have the proper infrastructure to successfully manage the program.
“Medical groups not proficient or comfortable with MA arrangements would probably need an external source of capital and expertise to support the PCPs in the program, because risk sharing can initially seem complex,” he says. “Doing these programs right the first time is important so those of us who have developed successful programs should reach out to support those who are just getting involved.”
There are also many questions about how CMS will address overlap of this model and other CMMI models and the Medicare Shared Savings Program, and Dawe worries about the presence of so many models fragmenting markets and making it harder for providers to achieve the scale they need to meaningfully impact population health and costs.
An example of the fragmentation is physicians participating in the Next Generation Accountable Care Organization (Next Gen ACO) model—the precursor to Direct Contracting—were prohibited from also participating in the CPC+ model, and vice versa. So when Next Gen ACO leaders operating in the 18 CPC+ regions wanted to recruit physicians, they often had a hard time convincing the doctors to join the full, two-sided risk ACO when there was a much easier, zero-downside risk model available to them.
“In a sense, the presence of multiple low-risk options in the market has a dampening effect on the overall move toward higher value in Medicare,” he says. “To be successful, ACOs need good networks of high-value providers and a large enough beneficiary population to spread risk.”
When there is overlap of models, he says CMS has to figure out how to manage the payments across multiple models so there is not duplication, e.g., paying twice for the same services.
“CMS did allow overlap of CPC+ and one-sided ACO models. The way they handled overlap in those models is to say that the care payments doctors receive through CPC+ must be counted toward medical expense in the ACO,” Ridlon says. “So receiving these payments may provide some upfront cash flow, but ultimately makes it harder for the practice to achieve savings compared to benchmark expenditures. Neither scenario is helpful, in our view.”
David Belk, MD, an internist in Alameda, Calif., says both the Direct Contracting – Professional and Direct Contracting – Geographic models are confusing.
“The more confusing a payment model is for anything, the less likely it will save money because people will always find ways to use inherent confusion to game a system to their advantage,” he says.
Chenven says that there’s not enough detail yet regarding the contracting entities’ organizational structures or their anticipated responsibilities to weigh in on the plusses or minuses for these.
“My instinct is that only larger integrated delivery systems will be in a position to participate in the geographic model and that the professional direct contracting will appeal to smaller and local entities,” he says.
HHS’ new models are also expected to help bring about more value-based transformation in primary care settings around the country.
Medicare is an influence payer, even beyond the Medicare market. Therefore, Martin says these models give primary care more opportunity to deliver comprehensive, continuous, coordinated and connected patient-centered care, instead of sporadic, episodic care, signaling that primary care is transforming toward value-base care along with other parts of the industry.
“Primary care can now move away from fee for service and that’s a very good thing,” he says. “However, this is just an agitator for practices to begin that shift. But for physicians and practices that are ready to transform, these models will help them do that much more.”
Ridlon says Direct Contracting will offer a more advanced payment model to move the needle on value in Medicare and CMS estimates the models will shift a quarter of the country into outcomes-based payment.
Chenven says the goals articulated by CMS are to entice/engage more PCPs to get involved, and move a significant percentage of traditional Medicare patients into this program to get away from the wasteful fee-for-service arrangement.
“To transform our healthcare system to deliver better, more efficient care, we must change to a true value-based system—not only in payment but in scalable clinical and operational methods as well,” he says. “All players must be aligned in this seismic shift—including hospitals, primary care, and specialists.”