Top experts from Penn LDI and the Healthcare Transformation Institute (HTI) discuss the first decade of alternative payment model (APM) experiments and their recommendations for the next 10 years of APM development.

In “Redefining Health Care: Creating Value-Based Competition on Results, “the 2006 book widely credited with adding “value-based care” to the lexicon, Michael Porter of Harvard and Elizabeth Teisberg of the University of Virginia posited the puzzle of health care this way:  “Why is competition failing to lower prices or increase quality as it does in all other business sectors? There was not just too much care, but also too little care, and the wrong care. There were not just skewed regulatory and private-sector incentives, but also a fundamental misalignment between the nature of competition and value for the patient.”

Fifteen years later, the debate around how to realign the U.S. health care’s payment system around that value for the patient rather than volume for the provider, rages on.

Looking at the latest trends and developments in this field of research was the March 12 Penn LDI “Value-Based Payment: False Hope or Future of Health Care” panel discussion co-sponsored by the Healthcare Transformation Institute at the University of Pennsylvania and the Center for Health Incentives & Behavioral Economics (CHIBE).

During this next decade, the number one priority needs to be seeing gains against health equity through the innovation agenda of the federal government and the Medicare program. We need every APM model launched in the next decade to have some directly-focused quality metrics, or incentive programs or bonuses aimed at closing equity gaps.

Amol Navathe

The four panelists—all top experts in the field of health care payment models—were Ezekiel Emanuel, MD, PhD, Co-Director of the Penn Healthcare Transformation Institute and former Obama White House special advisor for health care; Amol Navathe, MD, PhD, Co-Director of the Penn Healthcare Transformation Institute and a Commissioner of the Medicare Payment Advisory Commission (MedPAC); Mai Pham, MD, MPH, President of the Institute for Exceptional Care and former Chief Innovation Officer at the Center for Medicare and Medicaid Services (CMS); and Rachel Werner, LDI Executive Director and Professor of Medicine and Health Care Management at the Perelman School of Medicine and the Wharton School. All four are LDI Senior Fellows.

The four are also co-authors of the just-released white paper, “The Future of Value-Based Payment: A Road Map to 2030,” and policy recommendations produced with input from the LDI-HTI Next Decade of Payment Innovation Working Group in collaboration with 13 other top national experts on health care payment reform.

The first decade

Like that white paper, the seminar looked back from 2020 to 2010 when passage of the Affordable Care Act created the Center for Medicare and Medicaid Innovation (CMMI) to fund and test a wide range of experimental alternative payment models (APMs). These included Accountable Care Organizations (ACOs), bundled or episode-based reimbursements, a number of transformational primary care models, and other innovative programs. The goal was to gather evidence on how best to transition U.S. health care away from fee-for-service system.

The general consensus after 10 years is that these experiments weren’t very successful in meeting either their fiscal or care quality goals, but they did provide useful insights that now inform work on future APM strategies. 

Rachel Werner, MD, PhD
Rachel Werner, MD, PhD

“Despite these limitations of the effect of APMs so far,” said Werner as she opened the seminar, “there are a number of lessons we can learn about how to improve them. There are a number of ways we can think about why they’ve been less effective than we had hoped. One of these is that the APM models were built primarily on a fee-for-service chassis. Fee for service provides very strong incentives to maximize volume, and APMs alone may have not had strong enough incentives to overcome those fee-for-service-based incentives. That’s in part because APMs don’t often hold providers to clear budgets and don’t often transfer meaningful risks to providers.”

“Other factors,” Werner continued, “include program designs that have been quite complex for participants, conflicting incentives across the decade’s many different programs, and the fact that often the participation in these programs is voluntary rather than mandatory.”

Werner emphasized that the seminar—like the new white paper—was focused on what the federal government should be doing over the next decade to achieve a more meaningful impact across health care with value-based payment.

Mandatory rather than voluntary

All four panelists agreed strongly on the paramount importance of future APM programs being mandatory rather than voluntary.

Mia Pham, MD, MPH
Mia Pham, MD, MPH

“I think all of our thinking on this comes from a decade of experience and our realization of the different archetypes among providers,” said panelist Pham, one of the top founding managers of the CMMI APM efforts. “You have true believers who will do value-based care no matter what happens, unless they really go off a financial cliff. You have what I would consider empiricists who are very willing to be first at trying it, but only as long as it works out for them financially. As soon as it doesn’t, they’re out of there. You have followers who wait for those leaders to go first and then follow them whether they stay in or get out. And then you have the group we would consider laggards. These are the resisters who say ‘not until you make me.'”

“Our concern,” Pham continued, “is that to date, payers have not deployed the tools that they really need to to get the third and fourth groups into the game. It is very important to make the payment models as appealing as fiscally possible to the leader groups. But there are laggards who are not going to change unless there is a real reason for them to change. But we sense there has been an undue skittishness about deploying mandatory models that payers need to get over.”

Asked what it would take to achieve APM-based health care by 2030, Emanuel said, “First of all, I think all providers have to be in predominantly paid by an APM through the government and private payers.” He pointed out that federal and state governments control about 45% of health care payments and could potentially “harmonize the same structure of a value-based payment model through” Medicare, Medicaid, the Federal Employees Health Benefits (FEHB) Program, and the Defense Department Health Agency Tricare program. He also noted that the ACA exchange subsidies might be leveraged to require exchange payers to adopt the same APM model.

Ezekiel Emanuel, MD, PhD
Ezekiel Emanuel, MD, PhD

Changing clinical practice

“What we really need,” Emanuel continued, “is to understand that for a physician practice or a hospital to actually change how they care for patients in a way that puts greater focus on the quality of care and the total cost of care requires investment on their part. Investment of resources, energy, and the time of some of their leadership to manage that implementation. There are a lot of kinks that inevitably arise when you’re changing the workflow around patients. That may require hiring care managers or behavioral health specialists or whatever else is actually necessary to effectuate the change.”

“Groups are going to be resistant and reluctant to do that because of just the time,” Emanuel continued. “But if you have a mandatory model, it takes people over that potential energy barrier, as it were, because they’re required by Medicare, and Medicare is too big a portion of their billing to ignore. But then, they might even be required by Medicaid and some of their commercial business as well. Making it mandatory is extremely important to get them on the ‘escalator’ to a new kind of reimbursement. It also changes the APM psychology from ‘this is a demonstration or an experiment’ to ‘this is an inevitability and I better get on that train.’ And I think those two psychological things are really important to address.”

Another barrier to more enthusiastic APM participation among providers was the relatively short length of time of past CMMI experiments, said panelist Navathe.

Amol Navathe
Amole Navathe, MD, PhD

“One of the things that came out of our group’s conversations with national experts,” he said, “is that part of changing the psychology of the physician group is having longer programs. We need APM programs that commit to doing this for five or seven years. Providers need to know that if they actually invest in the infrastructure, invest in the fees, invest in the changing workflows of a program, that it won’t all just go away in a couple years.”

“If we’re going to invest in population management strategies,” Navathe continued, “then we need to know this is a pillar of the program designs going forward and it’s not going to go away. We might make tweaks to models overall, but those investments over five and seven years are actually going to bear out going forward.”

Health equity focus

He also pointed out that another shortcoming of the last 10 years of APM demonstrations and experiments has been underscored by the current COVID-19 pandemic: That the first decade of CMMI APM efforts had no real focus on health equities.

“The first decade was a flurry of experimentation,” Navathe said. “And while I think the federal government and CMMI contractors put real effort toward monitoring some of the impacts on vulnerable populations that face outsized disadvantages, we haven’t really seen to date an expressed effort to design models oriented directly toward addressing health equity issues.”

“For example, I think we’ve seen in some of these programs, like accountable care organizations, that you end up getting non-systematic participation differences,” Navathe said. “In areas that have higher proportions of low-income populations, health systems and provider groups are less likely to join those ACOs. We’ve also recently done a study looking at the mandatory bundled payment program and we again got systematic differences. Populations that had greater social risk were less likely to be able to access providers participating in bundled payment programs.”

“So, while I think that monitoring is great, and clearly critical, I don’t think it gets us fully there in this situation. During this next decade, the number one priority needs to be seeing gains against health equity through the innovation agenda of the federal government and the Medicare program. We need every APM model launched in the next decade to have some directly-focused quality metrics, or incentive programs, or bonuses aimed at closing equity gaps. That means measuring how those models are producing access and outcomes for disadvantaged populations over time.’

“Lastly,” Navathe concluded, “We need to have greater participation. I would love to see every clinician and every beneficiary aligned to some value-based model by 2030. That should be the benchmark we try to hit.”