Jay Sultan: Poor Utilization Management Will Doom Many ACOs

Accountable care organizations (ACOs) are implementing population health programs nationwide. This trend has been building momentum since the Affordable Care Act became law in 2010.

As physicians and hospitals form ACOs, however, they may be missing an important element of population health: the management of health care risk, says Jay Sultan, the vice president for strategy at Edifecs, a health care IT company. Failure to manage population risk, he says, will cause many ACOs to fail, particularly those serving small numbers of patients.

“We have population health analysis but not population health management,” he says. “This situation is similar to what we had in the 1990s when providers took on capitated risk contracts and most failed miserably. At the time, those physicians and other providers didn’t know how to measure risk for enrolled members. Not knowing how to measure risk meant they didn’t know how to manage risk. That meant they didn’t know how to determine whether their care patterns were going to succeed or fail. And many of them failed.”

Some things have changed since the 1990s. Physicians operating under capitated risk plans 20 years ago lacked adequate information about the populations they were serving, and they didn’t have sophisticated electronic health record systems to help them manage those populations. At the time, health plans knew how to manage risk but had no experience with how to shift risk to providers. Therefore, it’s no surprise the providers failed, Sultan says. He fears the situation is similar today.

“Payers have been managing risk since the early 1960s,” he says. “Providers don’t know how to do it because they don’t have any experience with it. And they haven’t had any education about it either.

“To their credit, providers today are doing population health analysis. By that, I mean, physicians and hospitals involved in developing ACOs have identified which patients are in their panels and they can do some level of rudimentary analysis on those patients. But that’s about all they can do.

“Few can identify their sick patients or those who are sickest. Few can identify where they have gaps in care. If unaddressed, these weaknesses will become very costly in the coming months or years. By that, I mean, if you are an organization taking risk for the delivery of population health and you can’t identify your sickest patients or your gaps in care, you’re in trouble.”

One of the most important aspects of managing the health of a population is controlling the use of services. Health plans have a long and contentious history with utilization review and utilization management. These strategies can be clumsy and imprecise, but since the 1960s, health plans have learned from experience how to control utilization and costs.

“The problem for ACOs is that they have little or no experience with utilization management. In fact when they hear the term utilization review or utilization management, physicians and other providers think the discussion is about denial of services. They hate the concept so much they don’t even want to admit to themselves that management of utilization is necessary.”

In fact, of course, utilization management is not always about denial of services, but some denial certainly is necessary.

“The reality is that we cannot give unlimited health care to all people when spending is limited under some form of capitation, episode-of-care payment, shared savings, or shared risk arrangement.”

In May, researchers from Harvard Medical School published the results of a spending analysis that found that the Centers for Medicare & Medicaid Services (CMS) wasted $1.9 billion on low-value care delivered to patients in 2009.

One of the most famous studies (“The Quality of Health Care Delivered to Adults in the United States”) on whether providers deliver appropriate care was published in the New England Journal of Medicine in 2003. It found that only 55% of participants in the study received recommended care.

We could be back in the 1990s, with providers learning the hard way that they don’t know how to measure — and therefore manage — risk.

“The most recent data I’ve seen shows that about two thirds of health care being delivered today may not be evidence-based,” Sultan says.

“Consider that a hospital is doing population health management and a doctor wants to order an MRI for a patient. Both the hospital and the physician are operating under a capitated contract in which the hospital manages the risk. If that patient had an MRI just a week ago, the hospital’s EHR may have that information. Therefore, the doctor and the hospital both know that another MRI may not be necessary. But does the hospital have a method built into its workflow system that will prevent that physician from ordering another MRI?” Sultan asks. “You cannot claim to be managing something if there is nothing in the process capable of saying ‘no.’”

In many ACOs today, this scenario is common because utilization controls have not yet been established, he says. “The providers are generating information on utilization. That’s part of all EHRs today. But they are not using that information to alter what happens. Conversely, health plans have systems now or they are enhancing such systems to prevent needless testing or other services because payers know that utilization management works.

“We talk a lot about price transparency for patients, but we have not yet developed the far more important price and quality transparency systems for the physician, the one purchaser who needs to understand how much the colonoscopy will cost and what level of quality those providers will deliver.

“Someone will have to make choices about what gets delivered because there simply isn’t enough money in our economy to continue on the path we’re on today. That sounds simplistic, but that’s what successful health plans do: They control utilization and premiums so that they can collect more money than they pay out,” he says. “One challenge for providers is that existing financial models for providers seek to maximize utilization, at the same time when new models need providers to control utilization.”

Repeat of the 1990s

For these reasons, Sultan says that ACOs that lack adequate utilization management systems will fail. “It will be a repeat of what happened in the 1990s. The same market dynamics are playing out now. A bunch of providers took risk in the 1990s. Most of them failed and the remaining successful ones acquired the unsuccessful ones.”

Not all ACOs will fail, of course. “There will be some winners and typically a distinguishing characteristic will be size. The winners will mostly be larger ACOs and the larger ones will acquire the smaller ones because those ACOs that have too small a risk pool will be most vulnerable.

“And failure will be dramatic. Some hospitals will be shuttered or sold. That’s already happening today under mostly fee-for-service payment, but when the predominant form of payment becomes capitation, episode-of-care payment, shared risk, or shared savings, the number of failures will accelerate.”