Opinion: A Tale of Two ACOs

Joel C. Cantor | February 7, 2012 | Health Care, Opinion
New Jersey and the federal government have very different takes on ACOs, but both are tasked with driving down the cost of healthcare

It is the hottest new acronym in healthcare – ACO – or accountable care organization. A centerpiece of the Patient Protection and Affordable Care Act, the ACO concept is one of the few significant pieces of the reform law that is designed to truly bend the cost curve. According to the non-partisan Congressional Budget Office, the Affordable Care Act will reduce the federal deficit even as it extends coverage to over 30 million uninsured. But it achieves this largely by cutting rate increases for some Medicare providers and Advantage plans and raising new revenue.

But trimming rate increases and finding new revenue sources is not the same as stemming unsustainable cost growth. That will take broader changes in the way healthcare is paid for and delivered. Reimbursement systems that reward delivering the ever-growing volume of services without regard to their value to patients must be changed. That’s where ACOs come in.

The Medicare ACO program will take shape this year. Today in New Jersey and around the nation, hospitals, physician groups, and other care providers are scrambling to respond to complex new rules for creating these ACOs. But there is another kind of ACO under development right here in New Jersey. Taking a look under the hood of these two very different ACO models can help us understand whether and how they might contribute to a better, and more affordable, healthcare system.

The federal ACO program, also known as the Medicare Shared Savings Program, is a national effort to improve care and lower costs for Medicare beneficiaries.

The Garden State ACO was enacted by the legislature and signed into law by Gov. Chris Christie in August. Called the Medicaid ACO Demonstration Project, it applies to the state-run healthcare program for low-income people and will run for three years. Whether it is continued beyond three years depends on if it cuts costs and reaches quality benchmarks. (In full disclosure, the New Jersey ACO law calls on the Center for State Health Policy to help implement and evaluate the program).

The two ACO models have some features in common but also important differences. Both encourage groups of healthcare providers to collaborate to reduce wasteful services while maintaining or improving quality of care, patient satisfaction, and outcomes. And both programs will return a portion of the savings that the ACOs achieve as a reward for reaching those goals.

Both models also require a great deal of up-front investment in health information technology and care management infrastructure, and neither program offers startup resources or guarantees reimbursement for these investments. In both cases, providers must find venture capital to get started, and they are at risk of never recouping these costs. This is a level of entrepreneurism unfamiliar to many healthcare providers.

This is where the major similarities between the ACO models end.

One of the most important differences between the programs is in how Medicare or Medicaid beneficiaries will be enrolled in an ACO. The Medicare model has a complex set of rules that assigns beneficiaries based on where they have received primary care before the ACO was formed. ACOs must include panels of primary care practices whose assigned patients then “count” toward calculation of potential savings.

In spite of these restrictions, however, Medicare ACOs are largely free to pick and choose among primary care practices. Consequently, they will have a large measure of control over the patients for which they will be responsible.

Their incentives are clear — avoid the most difficult patients. The trouble is that a very small number of the sickest patients account for a very large share of the costs. For instance, just the top 1 percent of spenders generates more than one in five dollars in healthcare spending. These same high-cost patients offer the greatest opportunities for savings through improved care management, but ACOs have a strong incentive to steer clear of them. To make matters worse, Medicare will hold its ACOs harmless for the top 1 percent most-costly patients, further reducing their incentives to achieve better care at lower cost for the highest-cost patients.

The New Jersey Medicaid ACO model avoids these perverse incentives. ACOs are to be held accountable for improving care and reducing costs for the entire population of Medicaid beneficiaries in a defined community. That motivates ACOs to focus on the top 1 percent — and even the top 0.1 percent — of high-cost patients. This is precisely what the Camden Coalition of Healthcare Providers, the inspiration of the New Jersey ACO law, does. The CCHP has created a new lexicon in going after “hot spots” of “super-users”. This is a paradigm shift for American healthcare.

Another important difference between the two ACO models is the nature of the health services each must reduce in order to generate savings. For Medicare ACOs to succeed, they must redirect patients from high-profit, specialty care and high-tech procedures to low-cost, low-tech but highly effective preventive and primary care.

Medicare reimburses specialty care and high-tech procedures well, driving service patterns that maximize the use of these services. This pattern is more evident in New Jersey than elsewhere. According to the Dartmouth Atlas of Healthcare, New Jersey ranks first in the number of different physicians Medicare patients see in the last six months of life, averaging more than 11 doctors. In short, successful Medicare ACOs will have to steer patients away from specialist physicians and hospitals that profit from these services but add little to better outcomes. At the same time, many ACOs will have to tap these same deep pockets to finance their startup operations, raising questions about whether this model can adequately disrupt the status quo.

In contrast, as Camden Coalition founder Dr. Jeffrey Brenner point out, attracting super-user Medicaid patients is not part of the business plan of any hospital. Indeed, these patients frustrate emergency department staff and are typically money losers for hospitals because of Medicaid’s low reimbursement rates. Thus, there is much more alignment of incentives between Medicaid ACOs and their local hospitals than is the case in the Medicare model.

One other important difference between these models is how the revenue returned from reduced spending may be used by the ACOs. It would not be a great overstatement to say that the Medicare Shared Savings Program is akin to bounty hunting. Again, in spite of safeguards, there is no guarantee in the Medicare model that savings will be reinvested in care improvement. The New Jersey Medicaid ACO law is explicit on this point: ACOs must be nonprofit collaboratives of local providers that reinvest every dollar of shared savings to make care better for local Medicaid beneficiaries.

As a nation, we are putting a lot of eggs in the ACO basket. Realigning the perverse, cost-increasing incentives of our largely fee-for-service healthcare system is imperative. Both the Medicare and Medicaid ACO models seek to do that. Undoubtedly, both models will require course corrections as they are rolled out, and both federal Medicare and New Jersey Medicaid leaders have signaled an eagerness to adjust as experience is gained. Let’s all hope that they succeed!