Buy the numbers

Data on how successfully, not how often, hospitals treat their patients will drive dollars under health-care reform.
By Edward Martin

 

In late winter, guests in white rockers sip coffee on the veranda of the clubhouse at Pinehurst Resort, overlooking a lawn that looks like it was clipped by a barber. Visitors stroll streets lined with boutiques and art galleries. Outside of town, wind whispers through longleaf pines: It’s a good day for golf. In the heart of the village, daffodils are blooming early this year on the campus of FirstHealth Moore Regional, centerpiece of a three-hospital network that covers 15 counties.

One hundred twenty miles to the west, many businesses are boarded up and cars park behind chain-link fences in downtown Gastonia. Sidewalks are empty, unlike when this was the heart of Tar Heel textiles and busy merchants were updating storefronts with the now-weathered aluminum facades popular in the 1960s. Ten minutes away in the northeast suburbs, CaroMont Health’s Gaston Memorial Hospital, with flowing curves and sparkling white exterior, belies Gaston County’s fading industrial heritage.

Despite their dissimilar settings, the hospitals could be twins. With fewer than 450 beds each, they are medium in size but among the state’s best in various ratings. And both are at the forefront of a revolution in how Tar Heels and their employers pay health-care bills that totaled $60.2 billion in 2010. Value-based healthcare is coming, which means hospitals will increasingly be paid for how successfully they treat patients, not merely for how often they treat them.

In the complex world of medicine, it is a stunningly simple but radical reversal of a compensation model that traditionally rewarded hospitals and physicians for poor results and penalized them for excellence. If, for instance, unsterilized equipment infected a patient, the hospital and surgeons were paid for the operation — then paid again to cure the consequences. That’s one reason the state’s per-capita hospital bill grew to $2,280 in 2009, up an average of 5% a year since 1991.

For the last year, CaroMont and Blue Cross and Blue Shield of North Carolina, the state’s largest insurer, have replaced gimpy knees for a fixed price that includes preliminary work-ups in doctor’s offices, surgery and rehabilitation after the patient goes home. FirstHealth will ask Medicare this spring to let it replace hips and knees and perform heart-bypass surgery on the same basis. For competitive reasons, neither will discuss bundle charges, but North Carolina hospitals typically charge $26,000 to nearly $60,000 for joint-replacement surgery alone, depending on the joint and hospital location. Bypass surgery frequently tops $70,000.

“On one level, you can ask why the heck we weren’t doing this all along,” says Kevin Schulman, a Duke University physician who also teaches in its business school. “Why weren’t people holding hospitals accountable for the quality of care provided at their sites? What Medicare is pursuing now, along with a variety of private payers, is a step away from just writing checks to hospitals irrespective of whether the service was indicated or provided in the best possible manner — and then writing another check to fix it if there’s a problem.”

Bundled-care plans are value-based. If FirstHealth’s bypass patient recovers better than expected, the hospital and physicians will receive Medicare’s fixed price, yet to be negotiated. If the patient falters or has to return to the hospital, perhaps with an infection because doctors and nurses failed to give him antibiotics dictated by surgical protocol, FirstHealth and its physicians could be forced to eat tens of thousands of dollars in extra care. Quality with a carrot and stick? “Yes,” FirstHealth CEO David Kilarski says. “Tying reimbursement to quality certainly ups the stakes for hospitals. These are things we’ve prided ourselves on doing better than a lot of surrounding institutions, so I’d like to think we’re doing it to provide better patient care and not just because it’s going to increasingly determine our reimbursement. But no doubt, the stakes have been increased.”

Only gradually will patients become aware of the trend. Not so for insurers, hospital administrators and employers, who bear the brunt of health-care costs. At least eight provisions, most embedded in the Affordable Care Act of 2010, link compensation to quality. They began in 2011 with rules that prohibit Medicare from paying extra to treat hospital-acquired infections. More will be phased in through 2015, including across-the-board reductions in payments for patients readmitted within 30 days, additional bundled-care pilot programs and steeper cuts in Medicare and Medicaid payments to treat hospital-acquired conditions and medical accidents. Medicare accounts for 60% or more of many hospital budgets, and private-sector insurers such as Blue Cross typically follow its lead on pricing and policies.

Value-based care is already affecting hospital bottom lines. Insurers and employers are guiding workers to standout hospitals and physicians, bypassing those with less-sterling records. Self-insured Salisbury-based Food Lion LLC, which operates more than 1,100 supermarkets nationwide, provides injured and sick workers information on hospitals and specialists that, based on data filtered from health-care payouts for its 73,000 employees, get better and cheaper results.

However, Food Lion also illustrates a paradox of value-based hospital compensation. Though medicine is quintessentially personal, pay-for-performance relies on huge stores of data that harbor silent stories of medical success and failure. Nowhere is that more obvious than on the outskirts of Chapel Hill, where inside the glass walls of Blue Cross and Blue Shield’s trapezoidal headquarters, analysts are mining records of more than 3.7 million customers who submitted $11.5 billion in claims in 2010. What emerges is a list of more than 30 Tar Heel hospitals that have outstanding success for procedures ranging from bariatric surgery for weight control to treatment of exotic cancers.

“Improving quality of care for our members when they’re in the hospital is better for them — fewer poor outcomes and complications — and who wants to have to be readmitted? And it ultimately will reduce costs for the health-care system as a whole,” says Blue Cross Vice President Eugenie Komives, a physician who is the insurer’s senior medical director. For several years, Blue Cross has negotiated contracts that financially reward hospitals for better results in treating heart attacks and heart failure, pneumonia and other potential killers and for cutting readmission rates and controlling infections.

Underlying the quality movement is a grim reality: Medical bills were the leading cause of more than 25,000 personal bankruptcies in North Carolina in 2010. “Payment reform is absolutely necessary,” says Adam Searing, director of the Raleigh-based North Carolina Health Access Coalition, a nonprofit consumer group. “We simply must get a handle on costs and improve quality, and things such as bundling payments and linking payments to outcomes are some of the most promising strategies to achieve that end.” Behind the promise, though, could lay pain for hospitals.


In Winston-Salem, a city with one foot in the past and the other in the future, researchers grow replacements for diseased human organs, while health-care ironies sprout at every turn. Laboratories in a city built on cigarette manufacturing probe nicotine as a possible treatment for Alzheimer’s and other neurological ailments. Medicine is the new tobacco, and Wake Forest Baptist Medical Center is widely considered one of the nation’s best hospitals.

But not by Medicare. As usually happens when hospital rankings are revealed, the federal insurance program for people older than 65 or permanently disabled ignited a storm last fall by ranking Baptist below par in four critical measures. These included heart-attack death rates from 2007 through mid-2010, along with other criteria such as complications after surgery. In contrast, Moses H. Cone Memorial Hospital in Greensboro — Cone Health’s flagship is highly regarded but lacks Baptist’s national reputation — was zinged only for having a slightly worse-than-average pneumonia death rate. Most hospitals in the state fell in the average range. Such scores are more than ammunition for advertising campaigns — they’re quality measures that will increasingly determine billions of dollars in hospital revenue. Are they fair?


“We don’t question if they’re fair or not,” says surgeon Russell Howerton, Baptist’s chief medical officer. He, like others here, stoutly defends the hospital’s quality of care. “They are what they are, which is, useful but not perfect. If those are the rules of the game, those are the rules we’ll play by, and we’ll seek to optimize the outcome for our patients.” Like medical directors at FirstHealth and other hospitals around the state, Howerton says such ratings are useful yardsticks for improvement. But his sense of resignation goes only so far. “Our chief financial officer will tell you that quality has a financial return of its own, as well as benefiting the patient. He wants to make sure we’re not financially disadvantaged by measuring tools that by virtue of the outcome of only one patient could have a large financial outcome for us.” Some Medicare scores that dropped Baptist below national norms turned on minute statistical differences.


At the heart of most ratings, including the methodology Business North Carolina used in its rankings this year, is Medicare data that compares hospitals. Our scores are based on 12 weighted criteria that not only examine patient outcomes but patient satisfaction and how hospitals are rated by other organizations. Some of those include data collected by the Research Triangle Institute’s health-care research arm and opinion surveys of hundreds of the nation’s top doctors and surgeons. The enormous volumes of information gathered by Medicare and closely related Medicaid, which mostly insures patients who can’t afford to pay, allows analysts to spot trends and outline deficiencies.

Based on Medicare data, 30.1% of Baptist’s heart-failure patients were readmitted to the hospital in less than a month, compared with a national average of 24.7%. That drew a rating of “worse than the U.S. rate,” while smaller Northern Hospital of Surry County in Mount Airy, at 23.2%, was deemed “no different than U.S. rate.” Does that mean Northern is a better hospital than Baptist, one of a quartet of top teaching and research hospitals in the state — Duke University Medical Center, UNC Hospitals and East Carolina University’s Brody School of Medicine are the others — where cardiologists from around the world send some of their most difficult cases? Not even Northern’s boosters make that argument.

Its rating is based on 239 patients, fewer than half of Baptist’s 786. At some small hospitals, data such as heart-attack death rates are based on even smaller numbers, sometimes fewer than 50 cases. Those are levels at which luck — good or bad — can statistically swing a hospital from above to below average. Other, harder-to-define factors can be involved, too.

One is unintentional screening, which happens when small hospitals with few specialists and little sophisticated equipment send high-risk patients directly to larger hospitals from their emergency rooms rather than admitting them. Patients don’t show up in a hospital’s data unless they’re admitted. The victim of a serious stroke who arrives at a rural hospital might be sent to a medical center while a patient with a transient ischemic attack, a treatable mini-stroke with a likely good outcome, would be admitted. “The tool CMC — Centers for Medicare & Medicaid Services — devised is obligated to measure every facility from small to large,” Howerton says. “It does a good job at trying to measure every kind of facility, but when applied to certain kinds, it’s not as precise as it could be. It just happens that the experience of our organization is not one well measured by that tool.”

In February, Medicare released a study that supports his premise. It says higher complication rates at teaching hospitals probably stem from more-complex illnesses and injuries. Such skepticism is not universally shared. At least not by a doctor who in the 1990s collaborated with Medicare on developing its ratings. That’s Duke University’s Schulman, who says the database, called Hospital Compare, is adjusted for risk. “As long as I’ve been involved in medicine — 20-some years — people have been arguing that their patient populations are sicker on average than others’.” In fact, he adds, because large hospitals are judged on larger numbers of patients, scores typically show less volatility. One or two bad outcomes, for example, wouldn’t have a devastating impact.

Others contend that this year’s ratings, including those by BNC, show that good small hospitals are often overlooked. Nearly half of those that made the list had fewer than 200 beds, and two — McDowell Hospital in Marion and Presbyterian Hospital Huntersville — had fewer than 100. “Small organizations have fewer people, which may give them a more-focused capacity to deliver the best patient care,” says Jeff Spade, executive director of the North Carolina Hospital Association’s Center for Rural Health Innovation and Performance. “They may have only a few nursing units, not the 10s or 20s at big hospitals, and they may have 40 or 50 doctors on the medical staff, not 400 or 500.”

While such hospitals might lack the money for a top magnetic-resonance scanner — about $3 million, including the suite in which it’s housed — they may partially offset that with tighter teams. An operating-room nurse might be less reluctant to point out a problem to a surgeon he or she knows. Quality care extends to the relationships between patients and their doctors and nurses, which can show up in patient-satisfaction scores that Medicare and other insurers are increasingly tapping to determine how much hospitals are paid.

With only 54 beds, Angel Medical Center in Franklin, deep in the mountains of western North Carolina, ties Duke Medical Center, with nearly 1,000 beds, and UNC Hospitals, with almost 700, for highest patient satisfaction. And if pay-for-performance care brings simplicity to the complex financial world of medicine, improving quality is not always about complex technology. Emphasis on hand-sanitizing by doctors and nurses before and after contact with each patient has made dramatic improvements in controlling infections in the last two decades. Counting scalpels and surgical sponges before and after operations has reduced accidents in which they are left in patients. Many improvements were adopted after a 1999 report by the Institute of Medicine that concluded that as many as 100,000 patients a year died because of medical errors.

“We’ve known that preoperative, one-time antibiotics within an hour of surgery is the right thing to do since I was a resident 30 years ago,” says the medical director of one of the state’s largest hospitals. But about 15 years ago, when medical organizations started measuring so- called “processes”— such as starting heart-attack victims on preventive aspirin regimens — that Medicare uses in grading hospitals, he was shocked. “We found out, whoa! Doctors were only doing this 80-some percent of the time. You would not take a flight that landed safely only 80% of the time.”
He draws his own conclusion. “That’s not an indication we’re bad people or bad organizations. But it is an indication that if we apply the tools they’ve tested in other organizations, we can move from very good to near perfect.”


Regenerative knee disease has taken its toll. Now 60, retired from a company that makes nuclear reactors, he has come to this Triangle hospital from Virginia because of its reputation as one of the nation’s best. A nurse carefully writes “wrong limb” on his knee with an indelible marker, a practice adopted by orthopedic surgeons to prevent mistakes. In his preoperative haze, he looks down in bed at his knee and is jolted alert. The nurse has written “wrong limb” on the wrong knee. Value-based care is unfolding in an arena of infinite opportunities for error with the vagaries of human anatomy and illness — difficult to reduce to scores and ranks. Few think it will be painless, but many think it will be worth it. “Well, you know,” Schulman says, “they charge an awful lot of money to put you in bed overnight, and you should get something for it.”

Administrators fret that hospitals will be expected to take on greater responsibilities, including aspects of health care over which they exercise little control. Health-care reform, for example, establishes hospital Accountable Care Organizations, which might be pressured to take on such community-health threats as diabetes and smoking. Similar to bundled care, these measures would be overseen by hospitals but depend on physicians, clinics and others that might not even be affiliated with the hospital. “We’re leaning on hospitals to be accountable for what happens within their walls,” Schulman says. “But one cloud in this is that ACOs say the hospital is now in charge of providing continuity of services, even when patients don’t even go to the hospital. That’s a huge leap in terms of the scope of responsibility for hospital management.”

Searing, the health-care consumer advocate, says there’s also concern that patients will be bypassed as value-based care takes hold. “That includes things like releasing data on how patients are doing in the new payment models.” Some fear that hospitals and physicians, already under pressure from slim Medicare margins, will cut corners — such as gambling on releasing patients prematurely. In a profession equal parts art and science, dramatic change is coming.

In the ’90s, when best practices and formal process control took hold, doctors grumbled about being forced to practice what they called cookbook medicine. Those kinds of measurements are what Medicare and private insurers increasingly will rely on to judge quality and compensation. Hospitalist Daniel Barnes, until recently interim chief medical officer and now president of FirstHealth Physician Group, says doctors gradually changed their culture to accept process measures. “They’re evidence-based and have been demonstrated to improve patient care, so they’ve become more accepted.” At FirstHealth, the same culture shift is already under way.

“We’ve had a focus on readmissions and a number of these quality indicators now for a number of years. We’ve tried to incorporate them into our program and into reimbursement for physicians so quality is really incentivized.” Hospitals will be increasingly applying the same principle with value-based care and hoping for similar results for the same reason Barnes says it’s working at FirstHealth. “It’s also the right thing to do.”

To purchase the issue with the complete Top Hospitals listing, click here.