What Health Reform Can Learn From Cleveland Clinic

The Cleveland Clinic, where president Obama went in July to see high-quality, cost-efficient medicine in action, has miniaturized robotic tools that can repair a heart valve through an incision less than an inch long, a computer system that allows doctors to read patients' charts and write orders from anywhere in the world, and the last word in networked, interactive supply closets. Any time a nurse takes something from a shelf, it's recorded by a program that keeps a running count of 350 items in hundreds of locations, and can dispatch a self-guided robot cart to bring replacements from the warehouse. A century after Henry Ford began building cars on an assembly line, Cleveland Clinic has brought that technique to medicine, updated to reflect the latest Japanese-inspired thinking on "lean manufacturing" and "continuous-cycle improvement." Cleveland Clinic is a hospital trying to be a Toyota factory.

In his efforts to improve the efficiency of medical care, Cleveland Clinic president and CEO Dr. Delos M. Cosgrove, a former cardiac surgeon, has enlisted every tool of modern management, obsessively tracking metrics of performance from blood-bank usage to market share, even redesigning hospital gowns in an initiative to "improve the patient experience." He has expanded the system to nine community hospitals and 15 "family health centers," plus satellite facilities in Florida and Toronto and one under construction in Abu Dhabi. Riding roughshod when necessary over the prerogatives of the existing staffs, he has consolidated some services (such as cardiac surgery) in the 1,100-bed main hospital campus and distributed others (including primary care and routine obstetrics) to the communities.

He has even taken on the most intractable driver of American health-care costs: Americans. Having already banned the hiring of smokers (a dictate enforced by urine tests for nicotine), Cosgrove declared this year that if it weren't illegal under federal law, he would refuse to hire fat people as well. The resulting outcry led him to apologize for "hurtful" comments. But he has not backed down from his belief that obesity is a failure of willpower, which can be attacked by the same weapons used to combat smoking: public education, economic incentives, and sheer exhortation. "You have to be an optimist to be a cardiac surgeon," says Cosgrove. You also need a measure of self-assurance, bordering on arrogance, to take a beating heart into your own hands, and Cosgrove—who is imposingly tall and fit, with an unsettlingly intense, unblinking stare—gives the impression he wouldn't hesitate to snatch a potato chip from the hand of anyone who dared pick one up in his presence. It would have to be brought from home, though, because under his administration, potato chips have been banished from the clinic's vending machines.

As Cosgrove told a Senate hearing in June, the clinic's business practices offer a potential model for the American health-care industry as it strains to bend the ever-rising cost curve. The evidence was in the 2008 Dartmouth Atlas of Health Care, which reported that of the five medical centers ranked best by U.S. News in 2007, Cleveland Clinic provided the most cost-efficient care, measured by expenses incurred during the last two years of life—$31,252, nearly 50 percent below the most expensive. The clinic's distinctive feature is that in contrast to most other American hospitals, where doctors are essentially autonomous professionals, at the clinic physicians work on fixed salaries and yearly contracts. An outsider might describe this relationship as "employer-employee," although Cosgrove prefers a teamwork analogy; he calls Cleveland Clinic "the world's second-largest group practice" (after Mayo Clinic, which is organized similarly). This saves money in many small ways, such as on expenses for medical supplies and devices. "Because we're all on a team," says Dr. Joseph Sabik, chairman of thoracic and cardiovascular surgery, "instead of stocking 30 different heart valves, we can stock two or three, and unless there's a good medical reason to do otherwise, that's what we use." And it saves money in one large way, by divorcing doctors' income from the number of procedures they perform. That, in turn, reduces the incentive for unnecessary tests, whose cost to the economy was estimated at $210 billion a year in a recent report by PricewaterhouseCoopers.

That's an eye-catching number, even in the context of $2.2 trillion in overall health-care expenditures. Along with many other health-care experts, Cosgrove worries that the bills now making their way through Congress won't do enough to control expenses. "They set out to improve quality, access to health care, and cost control," Cosgrove said after the House bill passed. "I give them an A so far on access, a C on quality, but I'm not sure they're going to get a passing grade on costs." A number of proposals have been floated to improve cost efficiency by "bundling" reimbursements—paying hospitals a fixed amount to treat a given condition, rather than require itemized bills for each test and procedure—or rewarding hospitals for improving care and holding down expenses. Cosgrove believes even more fundamental reforms are needed, including better coordination among hospitals and doctors, and between hospitals—along the lines of what the clinic has already achieved. But a broad reorganization of American health care may be a bridge too far, at least for now.

The same study estimated that another $210 billion is wasted each year on medical paperwork. That, though, is one potential savings that has mostly eluded Cosgrove. At the clinic's patients' accounts office, rows of cubicles are piled high with file folders and printouts, testimony to its dealings with thousands of different health plans from hundreds of insurance companies all over the country. Thousands of times a day, clerks pick up the phone and get put on hold like anyone else who calls an insurance company. Industry estimates put the average cost of handling a phone call at $3, to each party. This is the hidden cost of competition; whatever else a government-run health-insurance system would accomplish, it would impose a uniform billing system on the current one, in which clinic's 2,000 doctors require 1,400 clerks to handle their billing.

Cleveland Clinic works so well as a business in part because it works so well as a hospital. Cardiac medicine and cardiac surgery are its crown jewels, a major reason that it attracted 39,309 patients from other states and 2,380 from 102 foreign countries last year. Many were transferred from other hospitals aboard the clinic's fleet of helicopters and jet transports, outfitted as flying intensive-care units and kept on 24-hour standby. They will pick up any patient not actively receiving CPR, including one who was flown to the clinic on heart-lung bypass. When they take off, they don't know and don't ask whether they're coming back with a homeless drug addict, whose care may run into hundreds of thousands of virtually unrecoverable dollars, or a senior partner in a law firm. Like most hospitals, the clinic counts on privately insured patients, who generally are profitable, to balance those on Medicare, which pays about 6 percent below cost, and Medicaid, whose payments average about 14 percent below. Unlike most hospitals, though, its reputation assures it a steady stream of wealthy patients, including foreigners, whose bills do not have to pass the needle's eye of an insurance-company claims examiner.

Dr. Steven Nissen, the chief of cardiovascular medicine, sees many transfer patients on his morning rounds. "Ninety-nine percent of the hospitals in the world would walk away from this guy," Nissen says of one patient, an elderly man in acute heart failure who was flown in the night before from a hospital in another state. "He's 79 years old, diabetic, with three bad valves. But this is what we do here. By our standards he's a reasonable surgical candidate."

As he talks, Nissen scans patient charts on a cart-mounted computer that can also call up X-rays and lab results as soon as they're posted. Cosgrove was an early and enthusiastic adapter of electronic medical record-keeping. Orders can be entered on the keyboard; the software's "decision support" function alerts the doctor to possible adverse drug interactions and will even suggest medications for specific conditions. Under development, according to Dr. C. Martin Harris, the clinic's chief information officer, is a "smart IV" system that will record in real time the actual administration of intravenous medications. But the most revolutionary feature of the clinic's record-keeping system is that patients can access their own charts, including test results, diagnoses, and procedures. (Doctors can hide some information—such as a cancer diagnosis—from patients until they've had a chance to explain it to them in person.) Nearly 200,000 patients are enrolled in this program, called MyChart. A pilot program in cooperation with Microsoft supplies home-monitoring equipment to outpatients with chronic conditions such as diabetes and congestive heart failure. The hardware measures weight, blood pressure, or glucose, and the software enters it on the patient's chart and automatically alerts the doctor by e-mail to abnormal results, which otherwise might be caught only at a bimonthly checkup or when the patient shows up in the emergency room. This is a microcosm of the kind of medicine—harnessing advanced technology to basic preventive procedures—that Cosgrove wants the whole country to adopt. But it raises the question of why the model that has worked at the clinic hasn't been adopted more widely. Medicine is a notoriously conservative profession. Its prime principle, "First, do no harm," was also the foundation of Edmund Burke's political philosophy, and of 16 physicians in the U.S. House and Senate, 11 are Republicans. There's a built-in resistance to changes such as the shift to "evidence-based medicine," which simply means being guided by the latest scientific data rather than "what I remember from med school" or "what worked with the last guy I saw with this." Most laypeople were probably surprised to learn that medicine didn't work this way already, or that the idea could be dismissed as a plot to undermine doctors' autonomy. Cleveland Clinic has also been in the forefront of measuring and publicizing its results; of a dozen leading hospitals contacted by NEWSWEEK recently to compare outcomes for cancer patients, the clinic was the only one that could provide its own data, compiled, analyzed, and easily accessible to patients and the public. That goes even for statistics—such as readmission rates for heart attack and heart failure—that display what the clinic delicately describes as "opportunity for improvement."

The clinic has also revolutionized its table of organization. Most hospitals are organized by traditional specialties, each with its own medical perspective, records, and business practices. You may have been seeing the same gynecologist for 20 years, but if you get referred for a hysterectomy, the surgeon's office treats you as if you dropped into the waiting room from another planet, and it's up to you to prove your health insurance is good on Earth. But in 2006 Cleveland Clinic abandoned the traditional departments in favor of 25 "institutes" organized by disease or organ system. This works well for patients, who don't care whether their back pain is cured by a rheumatologist, a neurologist, or an orthopedic surgeon. But, says Regina Herzlinger, an expert in health-care economics at Harvard Business School, it runs afoul of the dominant fee-for-service system of medical billing, which discourages cooperation across fields. When Duke University Medical Center set up a disease-management system for congestive heart failure, coordinating the efforts of cardiologists, primary-care doctors, pharmacists, and nurse practitioners, it drove down the cost of treatment by 40 percent in a single year, while reducing readmissions and improving outcomes. But that highlighted the central paradox of health-care economics: a patient's "cost" is the hospital's "revenue." The unintended result of the Duke experiment, says Herzlinger, was that the unit lost tens of millions of dollars a year. The chief beneficiaries were the insurance companies, which saved on reimbursements.

At Cleveland Clinic, by contrast, the institute system worked because all its doctors were already on salary. This eliminates the competition for patients between departments, and the incentive for doctors to perform additional tests and procedures. The system also "drives our quality up," Cosgrove says, because it frees doctors to concentrate on their practices, not the minutiae of running a small business. "Day after day for 30 years I did nothing but fix hearts," he boasts. "That's how you get good at something." Only a relatively few hospitals have adopted the salary model, including the Veterans Administration system, Mayo Clinic, and the cancer-specialty hospitals Fred Hutchinson and M.D. Anderson. It's a question of how you export the culture; Cosgrove himself admits that "teamwork doesn't come naturally to doctors." Running a hospital this way requires evaluating every doctor annually on a long list of criteria including infection and readmission rates, patient satisfaction, research, and even how much blood they used. Most doctors aren't used to being treated like…well, like ordinary employees. But Cosgrove thinks the world is moving in his direction; young medical-school graduates don't want to deal with the ever-increasing expense of setting up their own offices, and are willing to trade some independence in exchange for not being on call seven days a week in a solo or small-group practice.

It's no coincidence that the salary system has been adopted first by hospitals with world-class reputations. If you put doctors on a salary, it had better be a good one. Most of the doctors at Cleveland Clinic say they are being paid a little less than the maximum they could earn in a more traditional practice, but they would have to work a lot harder and longer for it elsewhere. According to its 2007 tax filing, the highest-salaried doctor at Cleveland Clinic, excluding those who were also officers or trustees, was $1.2 million. That's comparable to the highest compensation—$1.1 million—reported by Cleveland's other leading medical center, Case Western Reserve's University Hospitals. (For medical students considering a specialty, those were, respectively, a bariatric surgeon and a plastic surgeon. Cosgrove was paid about $1.5 million.) Cosgrove says he brought up the salary model at a White House meeting on health-care reform with policy adviser Nancy-Ann DeParle. "There were 10 other heads of hospitals there and every one of them said, 'Oh, no, we could never do that.' Then I asked them, 'Who would do it by themselves, if they could just make it happen?' And they all raised their hands." Admittedly, not everyone buys the argument that salaries reduce costs. Joseph White, an expert on health economics at Case Western Reserve University, is one skeptic, based less on the data than on his knowledge of human nature. "If they claim they're not interested in doing more reimbursed procedures, they're lying," he says. "Even if you [the doctor] don't put the money in your own pocket, someone is keeping track of it. Your boss can always tell you to do more."

Perhaps only someone coming from outside medicine can fully appreciate how irrational and inefficient most American hospitals are. Cosgrove—although he has been a doctor his entire working life—thinks even nonprofit hospitals, such as the clinic, should be run like businesses. McKinsey & Co., the management experts, and Hill Holliday, a marketing firm, were paid nearly $5.2 million by the clinic in 2007 for consulting services. Cosgrove established a department called Strategic Planning and Continuous Improvement with 50 employees, headed by Darryl Greene, a systems engineer with experience in appliance manufacturing. Here is how Greene and chief medical operations officer Dr. A. Marc Harrison approached the problem of managing patients on blood-thinning drugs. These patients have a variety of underlying conditions, but they all face the same problem, Greene says, "that you don't want them to bleed to death, and you don't want them to clot, either." So the clinic created a unit just to manage their anti-coagulant levels, but it quickly had more patients than it could handle. By analyzing the steps involved in a visit and assessing their contributions to patient care, a process Greene calls "mapping the value stream," his team cut the standard visit from 30 to 15 minutes. Doctors were giving the same introductory talk to every patient who came in, so they made it into a DVD instead. If you doubt how seriously the clinic takes these issues, ask to see the "enterprise business intelligence performance management dashboard," which displays real-time updates on waiting times in various departments around the hospital, updated every 30 minutes.

What Cosgrove hasn't been able to do, though, is rationalize the other side of the balance sheet, the billing procedures. Even Cleveland Clinic has almost no leverage over private insurers. Cosgrove has been urging them to move toward bundling their payments, reimbursing a fixed amount for a given diagnosis. "Insurance companies have as many people looking at our claims as we do," he says, "so I'm saying to them, can't we work something out? Let's say you give birth in the hospital. We'll figure out what the average cost is, and instead of having a tug of war over $50, the differences will cancel each other out. We're spending our time checking up on each other. It seems crazy." But perhaps not so crazy, from the insurers' point of view. Except for Medicare, which is a uniform system where the rules, if complex, are at least consistent and transparent, every state Medicaid system and every insurance company sets its own policies and procedures. If you do a catheterization into the three branches of the cerebral artery, is that one procedure or three? Lyman Sornberger, executive director of patient financial services, keeps asking the insurance companies for their rules so he can submit a "clean claim," but without much success. One reason, he suspects, is that it would "upset their assumptions." That is a polite way of saying that insurers count on rejecting a proportion of claims the first time they are submitted, delaying as long as possible the disbursement of actual cash.

Cosgrove has carefully avoided choosing political sides in the health-care-reform debate. But a visit to Cleveland Clinic makes it hard to avoid the conclusion that if you're looking for "waste" in the health-care system—defined as expenses that do not directly contribute to medical outcomes—a good place to look is the nation's cobbled-together system of competing private insurers. Nissen, who considers himself less bound by the need for circumspection, points out that "the overhead for private insurers is 29 percent. For Medicare, it's 3 percent. If what's left over is what you can spend on patients, I think 97 percent is a much better deal." A message comes in to the critical-care-transport ready room; a suspected stroke case needs airlift to the main hospital, stat. A crew scrambles to the helipad on the roof and is airborne within minutes, disappearing over the trees. In the tunnels below, the robot carts whir and beep. And in the billing office, the phones blink and blink and blink, a reminder of how some things never change.

Join the Discussion