When Josephine Taylor was hospitalized for a seizure in 2014 at St. Joseph Mercy Ann Arbor, she noticed something new–for the first time, her primary care doctor’s office got involved in her care, even though a different doctor admitted her to the hospital.

The difference didn’t stop there.

“If the [primary-care doctor’s] nurse care manager doesn’t hear from me in a week or so, she calls me,” says Taylor, of Pittsfield Township. “It’s more one-on-one follow-up, and I really like it.”

It turns out that Taylor spotted changes prompted by the Patient Protection and Affordable Care Act, aka Obamacare. She’s one of an estimated fifty million Americans whose health care is paid for on a fee-for-service basis by Medicare, the government’s insurance for the elderly and disabled. Under the ACA, Medicare is using its financial clout to drive greater efficiency and better care for the people it covers.

They’re making some progress. Taylor’s doctor and other health care professionals in the Physician Organization of Michigan saved Medicare $27 million in 2014. POM received a share of the savings–$12 million.

“We’re not able as a group to tell you the precise interventions that led to success,” says Steven Bernstein, a U-M physician who’s on POM’s board of managers. But Bernstein cites several possible contributors to the savings, including more nurse care managers like the one who keeps in touch with Taylor; more at-home rehab for joint replacement patients; getting seniors in for an annual wellness visit; and more people creating advance directives for end-of-life care.

It took years for Ann Arbor health professionals to achieve that success.

Carrots and sticks

POM is an “accountable care organization,” a group health care provider created by the Affordable Care Act. Since 2012, more than 400 ACOs have been created to serve 7.9 million Medicare beneficiaries.

POM alone encompasses more than 6,300 health-care professionals, including the U-M’s Faculty Group Practice; Integrated Healthcare Associates, the big medical group linked to St. Joe’s; and the independent Huron Valley Physicians Association.

ACOs are “accountable” because Medicare gives them incentives to reduce costs while maintaining quality. If an ACO reaches a savings target while meeting Medicare’s quality standards, it gets to share in the money saved, like POM did last year.

Getting there, though, has been harder than expected. Neither the U-M faculty nor IHA signed up with POM initially. They started out, in 2012, with their own “Pioneer” ACO. An elite group of early adopters, the Pioneers keep a bigger share of Medicare’s savings if they meet the agency’s goals–but also face financial penalties if those targets aren’t met.

They weren’t. Medicare aimed to cut costs by 2 percent that year, but the Ann Arbor-area Pioneer saved just 0.3 percent.

Dr. David Spahlinger, executive director of the U-M faculty practice, explains there’s not much savings that can be made in the care of 80 percent of Medicare patients.

“There are still savings, but it’s harder,” he says.

The real savings possibilities, Spahlinger says, are in Medicare patients who have five or more chronic medical conditions and are hospitalized repeatedly. The challenge is to manage their complicated conditions outside the hospital, so they make fewer emergency department visits and need less inpatient care.

Adding to that challenge, from 2006 to 2010, the U-M doctors participated in demonstration projects designed to save money on patient care. That success may have left them with fewer opportunities for savings going forward. After all, there’s only so much money to cut before reductions affect the quality of care.

“It’s harder over time,” says David Introcaso, vice president at the American Association of Accountable Care Organizations. “I think that’s why savings year 1 [2012] over year 2 [2013] dropped 60 percent for Pioneers.” To date, thirteen of the thirty-two Pioneers have dropped out.

The Dartmouth-Hitchcock health system in New Hampshire has considered dropping out of the Pioneer system, too, according to its June editorial in Modern Healthcare magazine. It points out that health systems whose costs were already lower before Obamacare will find it harder to meet Medicare’s savings goals. That may be the case with the U-M’s doctors as well.

While many Pioneer ACOs saved Medicare some money, few met the targets necessary for them to share in the savings. “Pioneer ACOs had a small collective impact on slowing total Medicare spending growth in the first year, but most Pioneer ACOs saw [spending] growth similar to their local markets,” according to a 2013 report prepared for the Centers for Medicare & Medicaid Services.

Since the Pioneers are penalized when costs rise, why are some groups still willing to take the risk? Introcaso offers a possible motive: “The industry knows we’re moving to pay-for-value, so the sooner you move your learning curve the better, to be more competitive in your market,” he says. “Part of the problem is that nobody really knows what explains success or failure.”

Impeding the success of Pioneer and non-Pioneer ACOs alike is an inherent systematic flaw.

Unlike a health maintenance organization, where a primary care doctor controls referrals to other caregivers, patients in an ACO are free to seek care wherever they like. Nevertheless, all their medical costs are attributed to the ACO.

Unlike HMOs, many patients don’t even know they’re in an ACO. While it is required to notify beneficiaries they belong to it, POM has heard from patients who think the notification letter is a scam, or they confuse it with other health care mailings, according to a letter it filed with Washington in response to proposed ACO changes.

Even doctors don’t know at the time of treatment which of their Medicare patients are in an ACO.

A medical monopoly?

POM saved Medicare $4.7 million in 2013, its first year. It didn’t share in those savings because it missed the 2 percent goal, but at least it wasn’t penalized for falling short.

Avoiding that risk may explain why St. Joe’s and the U-M abandoned their own Pioneer ACO for POM. Though few patients realize it, the group now includes almost every doctor who treats adults in the Ann Arbor area.

Does that raise legitimate concerns about a monopoly? Introcaso points out that Washington granted ACOs several waivers from laws against kickbacks, self-referrals, and eliminating competition. The exemptions were needed, he says, to accomplish the law’s goals: “If you’re going to ask providers to give more integrated care, that leads to consolidation.”

Martin Gaynor, an economics and health policy professor at Carnegie Mellon University in Pittsburgh who has studied hospitals merging and buying physician practices, is a little more circumspect.

“Well, if indeed there is a monopoly, then yes,” it’s concerning, he says. “Monopolies are good for the monopoly and nobody else.”

Marianne Udow-Phillips, director of U-M’s Center for Healthcare Research & Transformation, says the motives for consolidation matter: “I really think it’s quite important to distinguish what’s happening with POM ACO and the mergers and acquisitions of hospitals, because [ACOs] are not financially linked to the reimbursement structure that’s established with the payers.

“The reason they’re doing what they’re doing … is because they can share data and coordinate the care of patients.”

Including many caregivers helps POM save money and provide better care, says Dr. Mary Durfee, IHA’s executive vice president and chief medical officer.

“Making the transitions from one care setting or provider to another needs to be as consistent and as seamless as possible,” she says. “Working together through POM … has helped us to build a collaborative relationship that benefits the community.”

POM’s growth created challenges. Leigh Ann Bowers, who retired this year as the group’s administrative director, says that when the IHA and the HVPA doctors joined in 2014, they brought with them some sicker patients, who are more expensive to treat. Bowers believes that’s why POM’s cost to treat a Medicare beneficiary increased from an average of $8,789 in 2013 to $9,000 in 2014.

Spahlinger, who chairs POM’s board of managers, declines to break out local results for Ann Arbor.

“We measure to improve, not judge,” he says.

Closing the revolving door

Among ACOs, only the Pioneers are penalized for missing cost targets. But all hospitals are penalized for overspending. Medicare reduces its payments when patients return to the hospital for the same problem within thirty days of discharge.

Based on percentages alone, St. Joe’s, with half of its patients insured by Medicare, has more at stake than U-M, whose Medicare patient population is 29 percent.

In 2012, Medicare started penalizing hospitals with high readmission rates for patients admitted for heart attack, heart failure, and pneumonia. The penalty is broad: all of a hospital’s Medicare payments are reduced, not just those for the heart and pneumonia patients.

While both hospitals’ penalty rates for readmissions were on an initial downward trend, St. Joe’s trajectory is starting to move upward.

In 2013, U-M’s readmission penalty was a 0.65 percent reduction in its total Medicare inpatient payments. Last year, U-M’s penalty was reduced to 0.26 percent; this year the rate fell to 0.22 percent.

At St. Joe’s in 2013, the readmission penalty was 0.11 percent; in 2014, it dropped to 0.06 percent; this year it will increase to 0.41 percent. Either the hospital is admitting sicker patients, or it has work to do to reduce readmissions–or both.

Though hospitals are financially responsible for readmissions, most of the care that can prevent them happens outside the hospital. That’s why St. Joe’s and its parent, Trinity, have been busily buying up physician practices. And despite the hospital’s rising readmission penalty, Durfee believes the coordinated effort is beginning to pay off. She says that between January 2013 and April 2014, IHA reduced readmissions among its higher-risk patients by eight percentage points, from 25 percent to 17 percent.

“The Medicare penalties for readmissions have been a stimulant that drives hospitals, emergency rooms, primary care and specialty care providers, and other members of the care team to work together to provide a better transition from the inpatient to the outpatient setting,” says Durfee. “This is good for patients in that it supports them to manage the transition from the hospital to the comfort and familiarity of their homes.”

Durfee’s national colleagues may be very interested in her results. While published studies about reducing readmissions exist, there’s a “paucity” of high-quality trials to evaluate interventions, according to a 2011 review paper from Northwestern University that examined thirty-six years’ worth of such efforts.

The Northwestern authors concluded that post-discharge telephone calls were part of the most successful bundled interventions, but they couldn’t identify any single intervention or group of efforts that reliably worked in cutting readmissions.

Avoiding Medicare’s penalties is not about to get any easier for either Ann Arbor hospital. The agency is adding penalties for high readmission rates for hip and knee replacement patients in certain geographic locations, and it may not be long before the new categories are applied universally. Washington plans to add chronic obstructive pulmonary disease to that penalty list in 2016. More procedures can be added to the penalty list in the future.

Both hospitals do have reason to be hopeful, however. They already have fairly favorable readmission rates for hip and knee replacements, at or slightly below the national average of 5.4 percent.

POM is also having success with just the kind of coordinated services that the Affordable Care Act envisioned. According to Bowers, its hip and knee replacement patients who got an early start on post-surgical therapy and then continued on the aggressive therapy after they moved to a skilled nursing facility went home in an average of seventeen days, while those who did not averaged twenty-nine days.

Of course, some factors influencing readmissions are beyond any hospital’s control. Hospitals who have more patients who are poor–which often translates to more chronic and mental illnesses, weak social support, and less access to primary care and healthy food–say they’re at an unfair disadvantage.

There’s also the issue of people making end-of-life decisions, like entering hospice, instead of seeking an ever-more-elusive cure for what ails them.

Another solution to curbing revolving-door admissions may be to employ more palliative care specialists, doctors who are trained to provide comfort measures, to see the sickest patients.

Justin DeWitte, president of Residential Hospice, Troy, says hospitals in Southeast Michigan refer 2 percent or fewer of their discharged patients to hospice, when best practices say the referral rate should be closer to 6 percent.

“Many of those who should be in hospice are not educated about hospice, haven’t been told the reality of their disease state, and haven’t been empowered to make their own decision,” says DeWitte.

Spahlinger agrees with DeWitte, and says doctors have a moral duty to talk with patients about realistic expectations from medical care. He cites a 2012 New England Journal of Medicine article that found a majority of patients with metastatic lung or colorectal cancer didn’t understand that chemotherapy wasn’t likely to cure their illness. In reality, it only offered them a little more time to live.

The instinct to seek a cure “is hard to change,” Spahlinger says. “It requires a longer conversation.”

When patients want to continue aggressive therapy, Medicare will continue to pay for it. But starting in 2016, the agency has proposed paying doctors for end-of-life conversations as well.