Taking Accountability: DHMC Explores ACOs
This past January, Dartmouth-Hitchcock became one of Medicare's first accountable care organizations, or ACOs. "The new model really puts the onus on the providers, on hospitals, physicians to engage patients in a conversation about their health that extends beyond the examination room," said Elliott Fisher, a professor at the Geisel School of Medicine and The Dartmouth Institute. (Dartmouth Institute photograph)
Lebanon — Diane Flint’s 85-year-old patient ended up in the emergency room recently and didn’t want to tell doctors about the pain in her abdomen.
As it turned out, she needed her gall bladder removed. Flint has been trying ever since to get this tough, reserved Granite Stater to open up and recognize that she needs surgery.
“I’m trying to educate her on what could happen if she waits too long,” said Flint, a care coordinator with Dartmouth-Hitchcock in Manchester. “Being 85 years old, the situation is a lot more critical because of her age.”
The conversations Flint has with her patients are perhaps more personal than you’d expect to have with a medical provider, but they are essential to Flint’s job and to the experimental model of care that is being pioneered at Dartmouth-Hitchcock and other hospital networks around the country.
Three years ago, when she was working as a registered nurse, Flint might have spent 10 minutes or so with patients to triage a problem and send them onto another physician or nurse. Now, she’s focused on getting into “the nitty gritty” of patients’ lives, working not only with people who are sick, but also their family members, social workers and occasionally outside agencies that have nothing to do with health care — all in a comprehensive effort to keep patients well.
This past January, Dartmouth-Hitchcock became one of Medicare’s first “accountable care organizations,” or ACOs, a concept developed at Dartmouth that is one of the major elements of the massive 2010 federal health law known as the Affordable Care Act or “Obamacare."
For more on ACOs, see related story, "OneCare: Profiting Off Patients?"
The ACO model has been in a testing phase for years, but is finally being introduced on a broader scale around the country. Dartmouth-Hitchcock is one of 32 “Pioneer” ACOs launched in January, an initiative the federal government hopes will save $1.1 billion over five years. Dartmouth-Hitchcock also recently proposed a new ACO arrangement with Fletcher Allen Health Care in Burlington that would involve more than a third of all Medicare beneficiaries in Vermont. (See related story)
The central tenet of ACOs is that doctors and hospitals should be held accountable not only for treating patients once they get sick, but also for helping to keep them out of the hospital and doctor’s office to begin with.
By helping patients avoid expensive trips to the hospital, the ACOs are designed to save the health system money. In return, the providers get a reward: If they cut back on costs while still furnishing high quality care, then insurance companies and government programs like Medicare are willing to split the savings with them — and perhaps move away from the fee-for-service mechanism that has been blamed for so much of the rise in health care spending.
“The new model really puts the onus on the providers, on hospitals, physicians to engage patients” in a conversation about their health that extends beyond the examination room, said Elliott Fisher, a professor at the Geisel School of Medicine and The Dartmouth Institute. “The kinds of language that you’re hearing in delivery systems that are stepping up to this model are continuous healing relationships where the notion is to engage patients wherever they are.”
Fisher was a leading architect of accountable care organizations — he coined the term. Studies of ACO-like arrangements have shown mixed results in the savings achieved, but Fisher and his colleagues believe the model shows promise to improve efficiency and lower the cost of care, particularly for patients who find themselves sick enough to require the most expensive kind of care.
And, like the health reform law itself, ACOs have not been without controversy.
Anti-trust regulators have said they will be watching closely for signs of price-fixing as hospitals work more closely together.
There’s also been concern about the notion of providing “bonuses” for providers who lower patient costs. The ACO with Dartmouth-Hitchcock and Fletcher Allen, called OneCare Vermont, is structured as a for-profit, a term that hospital officials say is misleading but that has nevertheless alarmed a number of people, some of them health care providers.
“There is nothing wrong with the concept of coordination and consultation, but the moment you introduce a profit motive in health care, down the line, the American propensity for greed will take over,” said Ann Raynolds, a psychologist with practices in Quechee and Claremont.
Then there is the question about what role, if any, patients play. Despite the goal of providing them with more thoughtful, cost-effective care, many people have no idea whether they are in an ACO.
Around 90,000 Dartmouth-Hitchcock patients are involved in some kind of ACO arrangement, be it with Medicare or one of three commercial insurers — Anthem, Harvard Pilgrim or Cigna. The only notification Medicare patients in the Pioneer ACO received was a two-page letter sent out last December, and there is no requirement to inform people with commercial insurance.
Some hospital officials, including Dartmouth-Hitchcock CEO (and physician) Jim Weinstein, have said patients need to share some of the responsibility for keeping heath care costs in check.
Despite these concerns, most everyone agrees on one basic assumption — that the current system is unsustainable and hospitals need to step up and find ways of reducing the cost of care.
“I think that the most thoughtful analysts and economists realize that continuing on the current path of cost growth for health care is not going to be sustained,” Fisher said. “Change has to come. We need to move to these new models of accountable care.”
What Does It Mean For Patients?
Although ACOs can take different forms, the same basic principle applies to all: hospitals agree to take responsibility for the cost and quality of caring for a certain group of patients, no matter what.
In essence, it is an agreement between providers and payers (i.e. insurers) that gives hospitals incentives to find more coordinated, efficient ways of treating patients. Together, the hospitals and insurers (or, in the cases of self-insured plans, the employers) sit down and agree on goals for saving costs and maintaining good care for the patient. The quality benchmarks are based on nationally recognized standards by such groups such as the National Quality Forum.
In the “Pioneer ACO” with Medicare, for example, if Dartmouth-Hitchcock finds a way to hit all 33 benchmarks on quality — as measured by patient surveys and claims data — while lowering costs by a certain amount, then Medicare keeps half the savings and sends the rest to Dartmouth-Hitchcock.
There’s some risk involved, too. If Pioneer ACOs fail to meet either quality or cost goals, then there are no savings. They may even lose money if, in the end, patient costs rise.
Dartmouth-Hitchcock has similar arrangements with commercial insurers, each one with its own set of quality benchmarks and cost saving targets. There are some arrangements in which there’s no downside risk — meaning, Dartmouth-Hitchcock wouldn’t end up paying more if costs rise. But otherwise, these arrangements generally work the same way.
To many, ACOs sound a lot like “managed care,” where a network of doctors shared in the financial rewards and risk in caring for groups of patients. But there’s one critical difference — patients in an ACO aren’t restricted in where they can get care.
If patients see any changes, it should be in getting better and more efficient care, said Barbara Walters, executive medical director at Dartmouth-Hitchcock, who is heading up the Pioneer ACO.
“The patient might have no idea that this is going on,” she said. “The patient could just be out there, going about their merry way, getting their care, wherever it is they want to get it.”
Dartmouth-Hitchcock’s Pioneer ACO includes all its clinics throughout the state, including Lebanon, Manchester, Concord and Keene. But if someone who has been “attributed” — that is, included in the pool of patients for which Dartmouth-Hitchcock is accountable — decides to go to Boston Medical Center, then the cost of that treatment also gets counted against Dartmouth-Hitchcock’s targeted amount.
That’s an uncomfortable risk for health care providers, says Weinstein, Dartmouth-Hitchcock’s CEO.
“Wherever they are, if they go from Manchester to Boston to get a procedure, unfortunately, we’re responsible for that cost,” Weinstein said in an interview earlier this year. “I don’t like that. That’s a problem with ACO Pioneer, because patients have to have some part of the responsibility in managing health care.”
Not everyone agrees on this point, including Fisher at Dartmouth. He said health care providers need to take responsibility for providing the high quality of care that makes patients want to stick with them.
For too long, the health care system has taken a passive approach to treating patients, waiting for people to come to them after they get sick, Fisher said. If they don’t show up, then it’s seen as the patient’s problem, not the doctor’s. That has to change, Fisher said.
“The language of ‘noncompliant’ patients, ‘nonadherent’ patients is a language of blame,” Fisher said.
Fisher said he’s sympathetic to Weinstein’s concerns. He hears similar complaints from other providers. But there’s a way to address this issue without putting the onus on patients.
“I talk to other ACO leaders and they talk about the green grass strategy,” Fisher said. “Our grass has to be greener than everybody else’s grass. People want to stay with us.”
There are all sorts of reasons someone might seek care outside the ACO network. Perhaps Dartmouth-Hitchcock doesn’t offer a particular type of treatment or another provider is simply more convenient.
If Dartmouth-Hitchcock is going to make itself more attractive and encourage people to stay inside the network, then it needs data, Walters said.
“From a provider perspective, I need as complete data as possible to help manage my population and my individual patients,” she said. “Especially because Dartmouth-Hitchcock doesn’t have every single provider that provides every kind of care that you can think of. We have snowbirds, the people who go down south, I don’t know what’s happening when they’re living down south for three months or six months of the year unless I get back the data on what’s happening to them.”
What kind of data? Basically, the claims data that Medicare and other insurers collect that says what treatments these people are getting. If Walters sees that a patient has gone outside the system for cancer treatment, for example, then her staff can reach out to that person and find out why he or she didn’t go to Norris Cotton Cancer Center. And, if there’s some way of making Dartmouth-Hitchcock’s “grass greener” — such as offering weekend hours and providing better service close to home — that can become part of the plan.
But getting that information in a timely fashion isn’t always easy. Right now, Dartmouth-Hitchcock gets claims information from Medicare monthly, a frequency she said “we can work with.” There is one commercial insurer — Walters wouldn’t say which one — that won’t share claims data with Dartmouth-Hitchcock, but the other insurers do.
Results on how Dartmouth-Hitchcock’s Pioneer ACO has fared on overall quality and cost won’t be available until sometime next year, according to Medicare officials. But the experience in the commercial market, as well as a five-year project that preceded the rollout of ACOs offers some insight as to how it might go.
A Mixed Bag
Dartmouth-Hitchcock’s ACO with Medicare may be in its early stages, but commercial insurers have had similar arrangements in place for several years.
Cigna launched its ACO-like pilot program in 2008, and both Harvard Pilgrim and Anthem had similar arrangements with Dartmouth-Hitchcock in 2010.
Spokespersons for each company said the programs have been successful.
“We both agreed that there was a better approach than the fee-for-service mechanism that has been in place for a long time,” said Robert Noonan of Anthem Blue Cross Blue Shield of New Hampshire.
The Anthem program covers 15,000 patients, or “lives” in the jargon of health care insurance. If Dartmouth-Hitchcock achieves mutually agreed-upon targets for quality and cost, then it gets to keep more than half of the savings, Noonan said.
Dartmouth-Hitchcock hit all its goals for quality and cost-savings from the first year, Noonan said. He declined to give specific dollar amounts, as did the other commercial insurers. But all praised the model as one that made sense.
“The fee-for-service world…a lot of people don’t feel good using that system today,” said Beth Roberts, vice president for regional markets with Harvard Pilgrim. “They can feel the fragmentation.”
Studies of an earlier pilot project that laid the groundwork for ACOs show that success is far from certain.
In 2005, Dartmouth-Hitchcock was one of 10 health care networks across the country to participate in something called the Physician Group Practice Demonstration. As with ACOs, providers were offered financial bonuses if they could slow down spending while providing high-quality care.
The experiment lasted five years. Maintaining quality care was not an issue, but finding enough savings was, according to a report issued last year by the Centers for Medicare & Medicaid Services. By the final year, only four of the 10 sites had saved enough money to qualify for the bonus. Dartmouth-Hitchcock was not among them.
Carrie Colla, an assistant professor at The Dartmouth Institute, led a team including Fisher that took another look at the Physician Group Practice Demonstration results. That study, published in September, yielded a more promising conclusion — in finding that the experiment was, in fact, successful in reducing spending for so-called “dual eligibles,” or people who qualify for both Medicare, which covers the elderly, and Medicaid, the federal program insuring the poor.
Such patients are among the most expensive and difficult to manage patients in the U.S. health care system, in part because of their high rates of illness, low socioeconomic status and lack of social supports, the report said.
“Our results suggest that while some care management or coordination programs have failed to demonstrate savings, ACOs and similar shared-savings contracts have the potential to improve care for this high-cost group,” Colla’s report said.
About 60 percent of health care spending in the U.S. goes toward caring for the 5 percent of people who are most sick, Fisher said. If hospitals and other providers, through ACOs, can find ways of bringing down the costs for those people, then this experiment will be worthwhile.
“Sick people are where all the money’s getting spent in health care… So you have to find that 5 percent of the population and pay great attention to them,” he said.
Some people, including Federal Trade Commissioner and Randolph resident Julie Brill, have raised concerns about accountable care organizations.
In July, Brill told a group of Dartmouth-Hitchcock physicians and administrators that when competitors come together to coordinate care, “the concern becomes that more gets coordinated than just care.”
Her suggestion was that ACOs could lead to larger, more powerful entities that will raise prices to private payers, such as insurance companies and self-insured employers, because they’ll have more of a bargaining position.
“It’s a legitimate concern,” Fisher said. That said, he’s not overly concerned that it will play out that way, especially once ACOs establish themselves.
During the first two years of Medicare’s 32 Pioneer ACOs, health care providers will share in both savings and losses in the cost of patient care. By year three, however, the Pioneer ACOs that have shown savings will move to a “population-based payment model,” in which providers receive a fixed amount of money per patient each month.
Raising the prices under this model won’t help doctors save money, because they’re getting paid the same amount regardless of what they do. In fact, Fisher argues, they will have a powerful incentive to keep patients healthy so they don’t have to invest in costly treatment.
Fisher’s colleague Valerie Lewis, a sociologist, is leading efforts to study poorer people — from the uninsured to Medicaid beneficiaries — and see how they are affected, Fisher said. In an August report published in Health Affairs, Lewis warned that ACOs have the potential to exclude racial minorities, the homeless and others who are “socially disadvantaged,” as well as people who have serious health problems.
Flint’s experience as a care coordinator offers a window on this last question.
A middle-aged patient of hers checked into an emergency room in southern New Hampshire not long ago complaining of chest pain.
Flint is notified daily when one of her patients ends up hospitalized. When she got the alert about this particular man, she noticed he hadn’t been in to see his primary care doctor in nearly two years.
Flint called him and learned he’d lost his job, and thus, his health insurance through a Cigna accountable care program. Flint gave him some information about financial services offered through Dartmouth-Hitchcock.
He likely won’t continue as one of Flint’s patients, because his loss of insurance means he’ll no longer be a part of Cigna’s program. Nonetheless, Flint has helped him explore options for how he could keep getting care.
She told him that she’d follow up with him in a week, realizing that he could easily slip through the cracks in the health care system. Flint said she’s kept her word.
“Actually, I called him today,” she said.
Chris Fleisher can be reached at 603-727-3229 or firstname.lastname@example.org.