Since March, St. Charles Health System has faced a new financial incentive to keep Oregon Health Plan patients out of its hospitals.
Before then, the hospitals were reimbursed by the state Medicaid program only if a patient walked through the doors and sought treatment. This is the standard financial model known as “fee for service.”
Patients who are insured under Oregon’s version of Medicaid are still covered if they have to go to the hospital. But under a new contract, St. Charles now receives a monthly payment for each Central Oregon member of the Oregon Health Plan — and will profit the most if the patients don’t have to be admitted.
That shift is part of a big experiment in health care, to encourage coordinated care that emphasizes prevention. Gov. John Kitzhaber developed this model for Medicaid, in an effort to accomplish the so-called triple aim: better health care, better health and lower costs. Securing the third leg of that tripod — lower costs — involves what is likely the most revolutionary change of all. It also involves a generous definition of “lower,” as the state is cutting the rate of growth in per-patient spending, not the total dollar amount.
Instead of paying health care providers for each service they perform, the new system offers a fixed per-member, per-month sum. The monthly payments are augmented by possible performance bonuses.
Several primary care clinics have switched to this payment system. But for hospital services, Central Oregon is the first community to put its money where the governor’s mouth is.
This could someday become the new model for all health care in Oregon. Earlier this year, the state Public Employees’ Benefits Board, which insures more than 130,000 public employees and their families, announced plans to switch its insurance plans to more closely resemble the one used for the Oregon Health Plan.
Regardless of how you get your health insurance — whether through Medicare or a private insurance company such as Moda — the story of Medicaid in Oregon could be a glimpse into your own future.
Costly treatment, prevented
Lisa Severa’s experience illustrates the state’s overall strategy for saving money, as a comprehensive checkup this spring helped prevent what likely would have been costly treatments later.
Severa, 45, was uninsured from the fall of 2008 until the start of this year, when her name was added to the rolls of the Oregon Health Plan. That’s when the Affordable Care Act raised the Medicaid income requirements, to cover adults who earn up to about $16,100 a year for a single person, or $32,900 per year for a family of four.
Severa had been a patient at Mosaic Medical, a nonprofit chain of clinics throughout Central Oregon, since moving from Vancouver, Washington, to Crooked River Ranch in 2012. But until she became an Oregon Health Plan member, each visit to Mosaic cost Severa $24. If she didn’t have the money, she was sent a bill — for $30.
Decades earlier, Severa had been diagnosed with irritable bowel syndrome, a chronic disease causing constipation, diarrhea, cramping and bloating. It’s diagnosed by ruling out other possible causes. But Severa had never undergone a colonoscopy, so she wondered if there was some other explanation hidden inside her large intestine. The procedure “had been on my to-do list for eight years or so,” she said.
Out of pocket, a colonoscopy costs up to several thousand dollars. For Severa, it was covered by the Oregon Health Plan.
A doctor at Bend Memorial Clinic performed the procedure in April. When Severa woke up, she learned two polyps had been removed from her colon and sent to a lab for biopsy.
For nearly a week, she worried. She Googled “colon cancer symptoms” and became convinced she was sick. Finally, her phone rang. The polyps were precancerous, which means she’ll simply need another colonoscopy in three years. If they hadn’t been removed, they could have turned into cancer — and cost upward of $10,000 in treatment, not to mention lost time at Severa’s jobs as caregiver for her wife’s grandfather and a neighbor.
Severa felt doubly grateful for the procedure when, two days after her colonoscopy, her father told her he’d had colon cancer.
“I guess he just didn’t want me to worry about it,” she said with a laugh. “The things we do to protect our kids.”
The ‘hotspotting’ strategy
In 2011, facing a $2 billion hole in Oregon’s budget, Kitzhaber struck a deal with the federal Centers for Medicare & Medicaid Services. CMS agreed to give the state $1.95 billion over the next five years. In return, the Oregon Health Plan’s budget would grow 2 percent slower than other Medicaid programs across the country, saving an estimated $11 billion in 10 years.
Cutting health care costs isn’t a new goal. But most savings are achieved by dropping enrollment or slashing coverage. Kitzhaber’s agreement, on the other hand, pledged to improve health care quality, hitching funding to a set of performance measurements sometimes called metrics.
Today, Oregon is about two years into its five-year experiment. So far, 16 coordinated care organizations, also called CCOs, have popped up around the state, each tasked with managing the Oregon Health Plan in their surrounding areas. Central Oregon’s CCO is PacificSource Community Solutions. Each CCO gets a pool of money based on the number of Oregon Health Plan members it serves, and greater freedom in how that money is spent.
One of the state’s primary strategies is to shift care away from emergency departments and into primary care clinics.
While making his pitch for CCOs, Kitzhaber repeatedly touted the anecdote of an elderly woman with heart disease who lives in an apartment without air conditioning. A heat wave sends her into heart failure. The state covers tens of thousands of dollars in medical bills, he pointed out, but won’t pay for a $200 air conditioner that could have prevented her health crisis in the first place.
Deborah Cohen is part of a team of Oregon Health & Science University researchers studying CCOs.
“We hear a lot of anecdotes about the air conditioner story or the one very expensive (patient) kept out of the emergency room,” she said. “They are nice to create more energy around an idea, but unless it’s happening consistently and unless the cost of health care is in some real way changing, you do worry a little bit about how successful these CCOs are going to be.”
A 2010 pilot study in Central Oregon focused on 144 of the prior year’s most frequent visitors to area emergency rooms. Health workers met and devised an individual, coordinated plan for each patient who agreed to participate. In six months, the patients’ ER visits dropped 49 percent, and the average per-patient costs dropped 62 percent.
This strategy, called “hotspotting,” has been adopted by most, if not all, of the state’s CCOs.
But John McConnell, a health care economist at OHSU, isn’t convinced it’s a smart move. McConnell said patients who make frequent trips to the emergency room tend to be in the throes of a short-term medical crisis. With or without intervention, he said, they usually revert to a normal rate of ER visitation within two years.
“There’s a lot of intuition behind (hotspotting, but) not a lot of empirical evidence yet that it’s going to work,” he said.
Even if it does, McConnell said, those savings might not add up to much.
“You look at the total amount spent on Medicaid — and what’s happening in the emergency department, per se, is small,” he said.
Emergency room costs are about 2 to 9 percent of total health care spending, with the exact figure dependent on the accounting method.
However, hospitalizations — overnight stays, generally speaking — seem to make up a much greater portion of Medicaid spending, McConnell said, though the exact amount isn’t known.
“The emergency room is responsible for a big percentage of people being admitted to the hospital,” he said. “You can find safe, high-quality pathways out of the emergency room and into an outpatient clinic (that) could generate cost savings. But that’s slightly different than making sure no one gets to the emergency room in the first place.”
In the first year of the CCO experiment, the rate of growth in Medicaid spending — not the amount of Medicaid spending, but the rate of increase — dropped 1 percent, which is about what state officials had anticipated. Officials admit that curbing spending is likely to get harder as the years go on and clinics have already made simple, obvious improvements.
To make a bigger dent in costs, officials said, it’s going to take not only new ways of caring for patients but new ways of paying for that care.
Michael Bailit, a health care economics consultant from Massachusetts, said the problem with “fee for service” health care is it encourages care that doesn’t necessarily benefit health. Instead, it offers a financial incentive to perform the most treatments, especially those with a high profit margin.
“It’s inherently inflationary,” he said. “And it doesn’t reward desired behaviors.”
The vast majority of health care in the U.S. is paid using a “fee for service” model, though others are gaining popularity. A survey of Blue Cross insurers released this month found that $1 out of every $5 those insurers spent on health care last year was tied to performance measurements.
The reason for moving away from fee for service is not just to reduce costs, Bailit said, but to align the financial incentives with the goal of high-quality care.
The CCOs, which often act like insurance companies, must negotiate contracts with health care providers in their regions that serve Oregon Health Plan patients. Instead of paying per treatment or office visit, a CCO may give a clinic a lump sum of money per month per Oregon Health Plan patient. It may offer financial bonuses for excellent performance. Each CCO has flexibility, and some may try different models with different providers. That means Oregon could soon be home to dozens of experimental payment methods.
Dr. Bruce Goldberg, a former family practice physician, was director of the Oregon Health Authority until late 2013, when Kitzhaber tapped him to run the state’s beleaguered health insurance exchange, Cover Oregon. Earlier this year, he resigned in disgrace as the state website, despite nearly $250 million in costs, was scrapped in favor of the national exchange. The debacle is now under investigation by the FBI.
In health care, Goldberg said in an interview before he resigned, budgets were traditionally set by actuaries who estimated prices based on trends from previous years. Annual increases of 6, 8 or even 12 percent were common.
“The same system that rewarded inefficiency also penalized efficiency,” Goldberg said. “Because if you saved money this year, you would get even less money next year.”
Now, he said, savings remain with the CCO and its providers.
In Central Oregon, the CCO gets about $245 per month for each Oregon Health Plan patient, though the exact per-patient amount depends on factors such as age. All of that money goes into what’s often called the “global budget,” to be spent according to the CCO’s contracts, and guidance from its board of directors, the Central Oregon Health Council.
From the global budget, PacificSource takes an administrative fee.
The primary care physician assigned to — or, in some cases, selected by — each patient receives a monthly payment, usually about $20, again depending on factors such as age.
Since March, when St. Charles signed its new contract with PacificSource, the hospital system has received a monthly payment for each patient enrolled, too, to cover inpatient and outpatient hospital services for all Oregon Health Plan patients in the region.
The remaining money in the global budget is used to pay specialists, who are reimbursed for actual treatments provided, using a traditional “fee for service” payment system.
Robin Henderson, chief behavioral health officer and director of government relations for St. Charles Health System, said moving from a “fee for service” model to a global-payment model requires thoughtful decision-making.
“Everything that you touch in this space is interconnected,” she said.
Until this year’s Medicaid expansion flooded local primary care clinics, Central Oregon had reduced the rate of emergency room visits among Oregon Health Plan patients. In 2013, members in Central Oregon made about 49.9 visits to emergency rooms per 1,000 member months, down from 61.6 in 2011, the year before Medicaid reform started.
Henderson said the downward trend was good news for the patients who managed to avoid the emergency room, and good news for the CCO’s budget. She said, however, officials didn’t want emergency room visits drop off too steeply, too quickly.
“We have to be very thoughtful, to still keep the basic infrastructure in place so that … when your kid falls off a bike, there is a trauma surgeon at St. Charles,” she said.
McConnell said it’s too soon to know if moving away from the “fee for service” model will actually save money.
He and his team of researchers are using billing and performance-quality data to compare the expenditures of each CCO with estimated equivalent costs under the old “fee for service” model, to identify exactly where the savings — if any — came from. By fall, he said, he expects to better understand how much money is being saved, how it’s being saved and whether the state’s estimates for future savings are realistic.
Michael Cannon, the Cato Institute’s director of health policy studies, is skeptical.
For one, the Oregon Health Plan ballooned following the Jan. 1 expansion under the Affordable Care Act, also called Obamacare. More than 971,000 Oregonians are now covered, including more than 357,500 who, like Severa, received coverage since Jan. 1.
Even if the state continues to chip away at the growth of per-patient costs, the massive spurt in membership dwarfs those savings, he said.
Cannon said even if Obamacare hadn’t prompted Oregon Health Plan membership to swell, he believes enrollment would have surged because of plans to pay primary care clinics per-member, per-month.
“There’s always a large chunk of the Medicaid-eligible population that is not enrolled because they’re healthy,” Cannon said. “Now there’s a huge incentive to find these people and enroll them because the provider will get the full … amount for a very low-risk patient.”
Pay for performance
One of the biggest changes to the state’s new Medicaid system is the idea of paying for health outcomes.
In 2013, 98 percent of Oregon’s Medicaid budget was claims-based. The remaining 2 percent was set aside and awarded according to performance, once data from the year was collected and verified.
Each quarter, the state produces a report card for each CCO, with 33 grading categories, or metrics. Of those, 17 are tied to financial incentives. (See how Central Oregon’s CCO performed in 2013 on the facing page.)
Most of the scores, however, are gleaned from the old “fee for service” claims system. In other words, providers who treat Oregon Health Plan patients — even at clinics receiving a per-patient, per-month payment — still have to submit claims for each procedure or exam they perform, so the state can calculate its report card.
So the new health care system is inherently contradictory: Even as it changes the ways people get paid, its quality measurements — and the financial incentives tied to them — are based on the old “fee for service” system.
“That’s the reality that we have,” said Tina Edlund, director of the Oregon Health Authority until she was recently being tapped to help switch Cover Oregon to the federal exchange, Healthcare.gov. “We still collect claims and we still collect billing data. The idea of paying for outcomes is really still in its infancy. We’re not there, and it’s going to take some time to get there.”
Last month, the U.S. Government Accountability Office published a report that identified a gap in the Centers for Medicare & Medicaid Services’ ability to accurately audit Medicaid payments in states switching to managed care programs, such as Oregon.
The report said that a failure to beef up CMS’ auditing system “will leave a growing portion of federal Medicaid dollars vulnerable to improper payments.”
Studies repeatedly show wasteful spending in health care. A 2012 report by the Institute of Medicine found that roughly 30 percent of health care spending is wasted on unnecessary or poorly delivered services. Other studies put the estimate closer to 40 percent. Officials say if they can capture just a fraction of that, it will add up to huge savings.
“What a lot of patients don’t understand is that more isn’t better. In fact, sometimes more is more harmful to a patient,” said Bailit, the consultant.
One of the state’s metrics for evaluating the success of the new Medicaid system focuses on an often-overlooked example: the rate of children tested for strep throat before being prescribed an antibiotic. Antibiotics are effective in treating strep throat, but not viruses. The overuse of antibiotics can lead to drug-resistant superbugs, which pose a growing public health threat.
Dr. Knute Buehler, an orthopedic surgeon at The Center for Orthopedic & Neurosurgical Care & Research, is also a Republican running for Bend’s seat in the Oregon House of Representatives. Buehler said he believes the CCO experiment could work. But additional reforms could help its chances. Studies have shown, for example, that doctors sometimes order extra tests or images to avoid liability in case of a malpractice lawsuit. “Defensive medicine” can run up the cost of care.
“If you’re really trying to wring costs out of the system … a big part of that is medical malpractice reform,” Buehler said.
Twenty states have laws to cap the noneconomic damages for patients who win medical malpractice suits, but Oregon is not one of them. To Buehler’s mind, this contributes to Oregon’s access problem.
“It drives up malpractice insurance rates. And that’s a factor when people finish medical school and decide where to set up shop,” he said.
New math for clinics
Each CCO is overseen by a board of directors that includes patients as well as representatives of the various hospitals, clinics and agencies in a region.
A global budget means these players have to come together to divvy up the pie. That’s a tall order among entities used to competing. Providers are wary of losing money, especially as Medicaid is long known for offering the lowest reimbursement rates.
In May, the Central Oregon Health Council, which oversees the local CCO, reviewed two pilot projects it funded last year. One added a pediatric nurse care coordinator to Central Oregon Pediatric Associates, to work with the 100 patients with the most complex medical cases and highest rates of hospitalization. The program cost $78,000 and saved an estimated $1.3 million to $2.6 million.
A second pilot project developed a multidisciplinary pain clinic for patients with chronic, unmanaged pain.
Rebecca Babcock, a 29-year-old single mother, enrolled in the eight-week clinic and said it changed her life. After more than a year of being bedridden, she is now a full-time psychology student. Last year, she went to the emergency room six times and urgent care four times, she said. This year, she hasn’t gone once.
The program’s organizers had not yet completed their financial analysis but said the pretreatment population spent an average of almost $15,000 per year on health care.
During their board meeting, COHC members were enthusiastic about continuing both projects but struggled over how to divide their costs.
“The outcomes sound fabulous,” said Karen Shepard, chief financial officer for St. Charles Health System. “But I’m having a hard time weighing what the cost savings to (St. Charles) is.”
Mike Shirtcliff, of Advantage Dental, said, “If you’re gonna help me save a million dollars, then I’ll write you a check to do it. But it sounds like we don’t know how much it saves (each of) us yet. We don’t have that data.”
The interaction illustrated the difficulty of getting stakeholders to agree on a shared pool of finances.
“You have to monetize health, and that’s hard to do,” said Buehler, who also serves on St. Charles’ board of directors.
Some providers are skeptical that achieving “lower costs” within the Oregon Health Plan will translate into anything but even lower reimbursement rates for them.
Earlier this year, Central Oregon Family Medicine, in Redmond, announced it would no longer accept Oregon Health Plan patients. A physician at the clinic said the recent Medicaid expansion meant hundreds of patients had switched from commercial insurance to the Oregon Health Plan, and the clinic was suddenly losing money.
Nationwide, the percentage of primary care physicians who accept Medicaid has declined in recent years, despite a 2012 increase in reimbursement rates. According to a survey by health care consulting firm Merritt Hawkins, 50.6 percent of primary care physicians accepted Medicaid in 2013, down from 65.4 percent in 2009.
Dr. Dan Murphy is a primary care physician who co-owned a Redmond medical practice, Cascade Medical Clinic, for 15 years. Several years ago, he learned about coordinated, team-based care and wanted to offer it at his clinic. But as a small, independent practice, he couldn’t figure out how to make the finances pencil out. Three years ago, he sold his practice to St. Charles. The hospital system now operates primary care clinics throughout Central Oregon that specialize in this model of coordinated care.
Experts say the kind of integration encouraged by CCOs favors large practices that can afford to employ staff — nurse care coordinators, community health workers and behavioral health specialists, for example — who provide care that isn’t reimbursed under the “fee for service” model.
Murphy said his annual salary hasn’t really changed since the clinic switched ownership and payment models.
When a CCO succeeds in saving money, the savings are unlikely to end up in the pockets of doctors, nurses and therapists.
To Bailit, that’s a problem. He said it’s important to commit to a payment method all the way down to the individual clinician. If a CCO uses bonuses to promote coordination within a physician group, for example, but savings don’t trickle down to the individual doctor, the incentive disappears as soon as the exam room door closes.
Joel “Charley” Jones is an Oregon Health Plan patient on the southern Oregon coast who shattered his heel in late 2012. After a series of surgeries to rebuild the foot, his surgical wounds became infected and doctors said his lower leg would have to be amputated. A nurse care coordinator employed by the region’s CCO, Western Oregon Advanced Health, helped Jones, 45, avoid amputation.
An amputee’s lifetime costs related to the missing limb average $509,000, according to a 2007 study by the Center for Injury Research and Policy at Johns Hopkins University. So saving Jones’ foot led to hundreds of thousands of dollars in savings for the state, at least in theory.
A medical assistant for the surgeon who treated Jones, Dr. Ryan Pederson of South Coast Orthopaedic Associates, said she couldn’t disclose the exact agreements the clinic has with different insurers. But Shanna Toney, a medical assistant for Pederson, said, generally speaking, private insurance reimburses 45 to 50 percent of the bill amount, Medicare 33 percent and the Oregon Health Plan 21 percent.
In total, Jones’ charges surpassed $14,000, according to Toney, but the clinic was only paid $3,300 by the Oregon Health Plan.
In the throes of contract negotiations, CCO officials from across the state were wary of speaking to The Bulletin about what their new finances will look like.
In the last couple of months, Cohen, from OHSU, said, more information has started to trickle in.
“What we’re getting is sort of the 5,000-foot view: What are the real principles and approaches that are informing how they think about their budgets? Not who gets the money,” she said.
Because funding for mental health organizations has been separate from the rest of the Oregon Health Plan budget for almost two decades, Cohen said pooling those finances has been tricky.
“Of all the system-level changes, integration of physical and behavioral health services is one of the hardest ones to make,” she said. “This is like years and years and years of (financial) silos that have been set up.”
Integration is also one of the ways to reap the greatest cost savings, she said, and make the biggest improvements in patient health.
In Central Oregon, officials said there have been technical obstacles to integrating care for behavioral and physical health. Privacy laws made it tough for medical providers to access information on patients’ mental health. And there are separate systems of billing codes for mental health and physical health.
Despite the intended flexibility within the state, Medicaid is largely funded by the federal government. And it turns out the state’s bills to the federal government are still restricted.
“They still don’t have the (billing) codes they can use to report they bought that air conditioner,” said Dennis McCarty, professor in the Department of Public Health and Preventive medicine at OHSU, referring to Kitzhaber’s campaign trail story. McCarty is studying the impacts of health care reform on addiction treatment and mental health.
“There are anecdotes emerging that (CCOs) are doing creative things by being flexible, but CMS isn’t as flexible as you would hope they would be,” he said.
McCarty said this will likely become more of a problem as CCOs look for more innovative ways to bend the cost curve. Right now, there is still plenty of low-hanging fruit, he said.
Changes to ‘charity care’
Because Oregon has fewer uninsured patients since Medicaid was expanded this year, some clinics, such as Mosaic Medical, report spending less on unreimbursed care, also called charity care.
Shepard said that’s not true at St. Charles, however. First, some patients are still uninsured.
Also, the hospital usually loses money on Medicaid patients, she said, so instead of covering the uninsured, what St. Charles calls its “community benefit” budget is filling the gap between what it’s paid to care for the Medicaid population and the actual cost.
Maryclair Jorgensen, director of health plan administration, payer relations and contracting at St. Charles, said the hospital is seeing more privately insured patients who can’t afford their high deductibles. So the hospital is using more of its charity care budget on privately insured patients, too.
Last year, Medicaid accounted for 14 percent of the hospital’s income. This year, with so many new Medicaid enrollees, that has swelled to 19 percent, Shepard said.
So far, according to Shepard, emergency room visits are not down in 2014. She attributed the steady flow of hospital visits to the dearth of primary care providers accepting Oregon Health Plan patients.
“There’s a backlog. … It will get caught up, but so far (this year) we haven’t had any lower volume in the emergency room,” Shepard said.
In most communities, said Cohen, of OHSU, hospitals are viewed as “cash cows.” They provide the most complex care for the most complicated conditions — at the highest cost. But St. Charles, through its new contract with PacificSource, is banking on coordinated care keeping more patients out of the hospital.
Jorgensen said St. Charles will withhold about $5 million of its Medicaid dollars this year, and redistribute it to primary care providers in the region, depending on their performance on certain prevention-related metrics.
Shepard conceded that some of the hospital’s profits — or losses — will depend on the region’s primary care providers. She said the hospital wanted to offer a real incentive. And there are many costs — including some performance metrics — that the hospital can control. To help cut down on readmissions, for example, St. Charles is working to improve its system for notifying primary care providers that a patient has been admitted to the hospital.
Currently, primary care providers have to log into a secure website each day, to see if any of their patients have been hospitalized and then schedule any follow-ups. The hospital is working to automatically notify each patient’s primary care provider as soon as he or she is admitted. This kind of change will affect all patients at the hospital, not just Oregon Health Plan members.
Kitzhaber, who is a former emergency room physician, has said he wants to reform all of health care, but Medicaid offers a way of easing into a new system.
Before this year, Medicaid patients made up, on average, 10 percent of a typical Oregon doctor’s patient load. The state figured providers would be more willing to experiment when only a small portion of their business is involved.
Lawsuits have already been filed against at least two CCOs, in Salem and Portland, by providers upset about their Medicaid reimbursement rates. One suit was dropped and the other is pending.
State officials acknowledge some of the current CCOs may fail. Providers could walk away from contract negotiations, for example. But officials said they expect other CCOs will form to replace them.
The new system means the state is experimenting with its most vulnerable population: patients who have the fewest resources to get help if shifting away from “fee for service” turns out to mean “fee for no service.”
Some are skeptical the state’s cost-cutting measures can be expanded to other populations.
“We are trying to create a lot of initiatives on the lowest payer in the system,” said Dr. J. Mark Maddox, who was medical director for PacificSource until April, when he accepted a job in California. “Most providers balance the number of government providers (such as Medicaid) with private providers (which pay more). So if you change a little on the Medicaid side, they can make it work. When you start to … roll in other populations, it becomes very challenging.”
St. Charles officials decided to accept the risk of Central Oregon’s new Medicaid agreement in part because the contract involved the lowest reimbursement rate among insurers.
“So, generally, it’s a loss anyway,” Shepard said. “And it’s the right thing to do.”