If you want to know the moment when the era of consumer-driven health care took off, look no further than April Fool's Day, 2005. With TV cameras rolling and flash bulbs popping, Mark B. McClellan, MD, administrator of the Centers for Medicare & Medicaid Services (CMS), announced to a gathering of health care journalists in Durham, N.C. the government's opening gambit in the latest scheme to improve the quality of American medicine.

It's Hospital Compare, a report on how well over 4,200 hospitals measure up on 17 different standards of good medical practices. Hospital Compare, says McClellan, "gives consumers and health professionals quality of care information to help them make more informed decisions about their health care, while providing stronger rewards and support for high-quality, efficient care."

Behind this initiative lies the belief that medicine can function like a true market, one where consumers can drive quality. Right now, a dearth of outcomes data leaves payers and consumers in the position of being unable to distinguish between high quality and poor quality providers, and for the most part, hospitals that do a good job of caring for patients are paid no better than those that do a lousy job. CMS is hoping that if patients know enough about the care being delivered, they will choose higher quality providers -- and thus encourage the poor quality hospitals to get their acts together and improve. As McClellan put it, "Consumers are a powerful force for change."

Maybe so, but what McClellan did not say at the press conference was perhaps even more interesting than what he did. For instance, there was no mention of the fact that CMS has wanted hospitals to submit data on outcomes for at least two decades, and that hospitals have long resisted requests for such information, fearing that they will be penalized for taking on more complex cases, which can be expected to have worse outcomes. "My hospital takes care of sicker patients is the argument," says Chas Rhodes, a hospital industry analyst for the Advisory Board Company, a consulting firm in Washington, D.C. A few hospitals simply refused to participate in Hospital Compare, while those that did were given a small bonus on their Medicare payments.

McClellan also did not point out some glaring omissions in the new data. While the 17 measures on the CMS report card represent a good start toward improving the quality of care, they all track the underuse of proven treatments. For example, hospitals reported how consistently patients with an acute myocardial infarction receive a thrombolytic agent within 30 minutes of arrival and a prescription for a beta-blocker at discharge. Participating hospitals also released data on whether they offered smoking cessation counseling to patients with heart failure and assessed pneumoccal vaccination status of patients with pneumonia.

What's missing from this hospital report card, of course, is information about overutilization, the vast amount of unnecessary or redundant care that drives up costs, but fails to contribute to patients' health. Overutilization has been widely documented, yet many hospital administrators and physicians still argue that it plays only a minimal role in the poor quality and high cost of American health care. CMS and private insurers have an uphill battle ahead if they hope to tackle overutilization, which accounts for an estimated 30 percent of the nation's total health care bill, or more than $500 billion out of the $1.7 trillion spent last year.

Geography Lessons

For health plans, the fact that patients often receive unnecessary care probably doesn't seem like front-page news. Yet it's worth examining the scope of the problem and some of the economic and cultural forces that drive it. In a series of landmark papers dating back more than two decades, John Wennberg, MD, and his colleagues at Dartmouth's Center for Evaluative Clinical Sciences have documented wide and persistent variations in usage and quality health care, as well as spending, in different regions of the country. These geographic variations represent a natural experiment, which has opened a window on the extent of overutilization and the factors that contribute to it. Using Medicare patient records, Wennberg's group has shown that per capita spending on Medicare recipients can vary by as much as 300 percent among the more than 300 Medicare hospital catchment regions of the country, with no better outcomes in the high utilization regions.

Take the areas surrounding the cities of Miami and Minneapolis, for example, which represent the high and low ends of Medicare spending in many categories. Total lifetime Medicare outlays for a typical 65 year-old in Miami amount to $50,000 more than for a 65 year-old in Minneapolis. Each year, Medicare shells out nearly twice as much for a citizen in Miami as it does for the Minneapolitan -- about $7,847 in Miami, versus $3,663 -- and Miami and Minneapolis are by no means the only regions delivering disparate levels of medical services. The region around Palm Springs, in California, sits at the high end, while Richmond, Va., is low, along with Mason City, Iowa, and Honolulu.

Critics of Wennberg's work have long argued that these variations among regions represent differences in underlying illness of the resident populations. Wennberg's group has addressed this criticism in two ways, first by examining the records of Medicare recipients in the last six months of life, when it's safe to assume that everybody is quite ill. In a recent paper in the BMJ (formerly the Britisih Medical Journal) Wennberg and his colleagues looked at the rates of utilization at 77 individual hospitals around the country for Medicare patients in the last six months of life. The number of days dying Medicare recipients spent in the hospital ranged from an average of 9.4 at the low end, at Hannepin County Medical Center, in Minneapolis, to 27.1 days at the high end, at New York University's hospital. The average number of physician visits per patient at Stanford University Hospital was 22, while the average patient at NYU saw a physician a whopping 76.2 times. That's a physician visit every 2.5 days during a patient's last six months of life.

Looking at hospital catchment regions, underlying illness fails again to account for the variation. The region around Provo, Utah, for instance, one of the healthiest in the country, ought to get 14 percent fewer Medicare dollars than the national average, because its citizens are less likely to smoke, be obese or suffer from strokes, heart attacks and diabetes. Instead, Medicare spends seven percent more on Provo recipients than the national average. Elderly people in the region around Richmond, Va., by contrast, tend to be sicker than most elderly Americans and would receive 11 percent more than the national average if underlying illness dictated how much Medicare spends. Instead, they see 21 percent less. Nor are those regional differences explained by variations in the cost of care. Provo physicians are not charging significantly more for office visits or lumpectomies than physicians in Richmond, and their patients aren't getting more expensive artificial hips.

Rather, the variation across regions -- about 41 percent of it -- correlates with hospital resources and the number of physicians. In other words, health care in the United States turns the most fundamental of economic principles on its head, so that the amount of care delivered is driven more by the medical resource supply than by medical demand. At the clinical level, that means patients get more diagnostic tests in regions where there are more MRI and CT machines; they receive more angioplasties and stents where there are more catheterization labs; and they see more specialists where more specialists set up practice. During the last six months of life, a Medicare beneficiary in the region around Miami sees on average 25 specialists in a year versus two in the region around Mason City, Iowa, largely because Miami is home to significantly more specialists.

Regions with an overabundance of medical resources are not delivering higher-quality care, they are simply providing more of it, and wasting Medicare dollars in the process. As Harvard health care economist David Cutler puts it, "There's a lot of evidence out there that using more of what's good is not of incredibly high value. It's like going to a really nice restaurant and overeating. The food was worth it, but it would have been even better if you didn't overeat." The evidence is in the outcomes. Medicare recipients in high-cost regions suffer as many disabilities as citizens of low-cost regions; they are not more satisfied with their care; and rates of underuse of proven treatments, like beta blockers on discharge after an acute myocardial infarction, are no lower.

Nor does the population in high-use regions live any longer -- although with all that invasive medical treatment they receive, including days in the intensive care unit at the end of life, it might seem like it to some patients. In fact, according to a recent study published in the Annuals of Internal Medicine, mortality in high-cost regions appears to be about two to five percent higher than in the lowest cost regions of the country. The most likely explanation for this is that elderly people who live in high-cost regions spend more time in hospitals than citizens in low-cost regions, and hospitals are risky places, where patients are exposed to the possibility of medical errors, drug interactions, and life-threatening infections.

Upside-Down Economics

What's driving all this overutilization? Many physicians have stated patient demand and defensive medicine are to blame. They're only partly right. Physicians order more tests in regions of the country where malpractice suits are more common, and patients do ask for tests and prescription drugs, particularly those that are most heavily advertised. Patients can also push up rates of overuse by self-referring to specialists. When they don't get what they want, say physicians, patients will often go to another doctor, and there isn't enough time during an office visit to explain to them why they don't need antibiotics for a viral infection, and how an MRI is not really going to help in the diagnosis of Alzheimer's.

But patient demand and malpractice worries are clearly not the whole story when it comes to overutilization. First, there is the 41percent of geographic variation that correlates with supply of resources. Differences in malpractice law, on the other hand, can account for only about 14 percent of the variation, a finding that suggests that malpractice is not as important a driver of overutilization as physicians would like to believe. Then there is the question of why the physician, the person who spent several years in training, would order a test or perform a surgery simply because patients think they need it.

A better place to look for the source of overutilization is fee-for-service reimbursement. Medicine is the ultimate piece-work industry; physicians and hospitals are paid for what they deliver, not for keeping patients as healthy as possible, a system that creates a host of perverse incentives. For instance, cutting costs by slashing reimbursement rates can rein in health care inflation for a time, but eventually providers find ways to maintain their incomes, generally by delivering more services.

One former managed care administrator has a telling story about a group of radiologists in Northern California who maintained their income in the face of falling reimbursement rates from Medicare. It was the early 1980s, and CMS had given notice it would cut reimbursement for X-rays. The average income for each radiologist in the practice, about $400,000 a year, would go down by $50,000 when the new rates went into full effect. So the physicians decided that a certain number of films would go back to the referring physician with the note, "Normal, but suggest retake within 30 days to be sure."

Of course, physicians are not always so calculating when it comes to finding ways to maintain their incomes in the face of falling reimbursement rates. Indeed, a great deal of overutilization, it turns out, results in areas with high concentrations of physicians -- and the majority of variation between regions comprises what the Wennberg group calls "supply-sensitive" care. While big-ticket surgical procedures, like carotid endarterectomy and back surgery, are overused and misused, says Elliott Fisher, MD, a professor of medicine at Dartmouth and co-author of the study in the Annals of Internal Medicine, "most of this stuff is minor procedures, office visits, diagnostic tests. If there is only one doctor in an area, he or she tells patients to come back for my next available appointment, which is in six months. If there are two doctors in the area, and the doctor says come back for my first available appointment, that's in three months."

Elliott's observation seems to be borne out by the most recent estimates of 2004 Medicare costs. In March, CMS actuaries revised their estimate of 2004 increases from 12 percent upwards to 15 percent, nearly a third of which stemmed from growth in the number and intensity of physician office visits. A quarter was a boost in minor procedures, and a fifth came from greater use of imaging services.

In a recent paper in Health Affairs, Barbara Starfield, MD, of the Johns Hopkins Bloomberg School of Public Health, looked at the effect of specialist supply on health outcomes. There is considerable evidence that quality of care is better when a hospital performs at least a certain number of procedures per year, and the same appears to be true for individual surgeons. Patients of high-volume surgeons have lower death rates for heart bypass surgery, carotid endarterectomy, and five other cardiovascular procedures when compared with surgeons who have low volume. But as more surgeons move to a particular area, there are fewer patients available per surgeon -- unless they perform more non-indicated procedures.

Finally, overutilization is driven by a patchy overabundance of hospital resources, particularly new technology that offers high profit margins to hospitals, which are constantly struggling to stay ahead of the uncompensated care they deliver to the uninsured and underinsured. Think of it as the Willie Sutton School of Hospital Management: Hospitals invest in new technologies because that's where the money is. "Hospitals say, if we can build it, we are going to bill it," says Stan Borg, MD, chief medical officer for Blue Cross Blue Shield of Illinois. All hospitals, including not-for-profits, are in an arms race of sorts, where new technology not only brings in higher margins, it also helps attract well-insured patients and retain specialists.

This competitive environment pits the true medical needs of the surrounding population against the need for margin when it comes time for a hospital to decide whether or not to invest in a new technology. As an example, Borg points to the rise in the number of bariatric surgery centers. Hospitals say the building boom is a response to projected increases in need; Americans are getting fatter, and their health is suffering as a result. That's not the only reason, says Borg: "Bariatric surgery is a popular surgery, and hospitals are going into the business for revenue enhancement."

When multiple hospitals in a particular region decide to open a bariatric surgery center, the likely result is overuse, because each individual hospital in town must do whatever it takes to find enough paying patients in order to recoup its investment. Hospitals run ads on TV and billboards, touting their bariatric centers and other high-margin departments, while surgeons use other methods of keeping their volumes high. "We had members telling us there were weight requirements, and that what they understood from their interviews with their surgeons was, you need to go home and gain weight because you don't qualify for surgery yet," adds Borg.

Conscious Consumers

Not surprisingly, Americans undergo many high-tech -- and high-profit -- procedures far more often than their counterparts in western European countries, where health care resources are limited by state budgets. For instance, Americans were given 388.1 angioplasties per 100,000 people in 1999, double the per capita rate in Belgium, the next highest rate among countries in the Organization for Economic Cooperation and Development, more than four times the rate in Canada, and eight times the rate in the United Kingdom. Yet there's not much to suggest that the huge volume of angioplasty performed in this country has led to significant health benefits; the longevity at age 65 and health outcomes in Western Europe are no lower than in the United States.

At the most fundamental level, overutilization everywhere stems from the pervasive lack of evidence for most medical practices. One of the fathers of evidence-based medicine, David Eddy, MD, estimates that more than three quarters of health care that's delivered has no real scientific basis. That means physicians perform procedures, prescribe drugs, call patients in for follow-up visits and send them off for hospitalization and diagnostic tests on the basis of solid evidence only a quarter of the time. Even when evidence exists, physicians are not always aware of it, or they may not know how to apply it to an individual patient, lending a degree of uncertainty to most medical decisions, either because the evidence for what works and what doesn't simply does not exist, or because it has not been gathered, analyzed, and then disseminated in a way that physicians can use.

Given all the forces driving the American health care system toward overutilization, and the entrenched political interests of physicians, hospitals, drug companies and medical device manufacturers, it's understandable that policy makers and many private health plans have decided to start small and tackle patient demand first. Call it "cost sharing," or "consumer-driven medicine," the idea behind health savings accounts, tiered benefits, high deductibles and high co-payments is the same: get consumers to feel at least a little of the sting of health care costs. If patients have a more direct financial stake in their health care decisions, goes the thinking, they will become more prudent buyers of medical services, choosing prevention over discretionary care, and investing the time and effort that's needed to mange chronic illnesses.

To make more prudent decisions, consumers need more information. Health plans have begun to couple cost-sharing plans with disease management services, aimed at educating the 20 percent of the workforce that generates 80 percent of health care costs. Blue Cross Blue Shield of Illinois, for instance, offers members a personalized, interactive Web-based source of information about their own health conditions. Members can take a health risk assessment and receive reminders for physician appointments, and they have free access to registered nurses, who can help them manage such chronic conditions as asthma and diabetes, or answer questions about pregnancy.

Other health plans, like CIGNA, provide members with Web-based information about their conditions and ways to monitor their own health. CIGNA's online service offers best practices-based information for about 35 different ailments, and sets up graphs members can fill in, to help them track medication schedules and such health indicators as blood pressure, lipid levels, and blood glucose levels.

Targeting high-volume users of emergency departments (ED) and specialists is allowing other companies to tackle overutilization more directly. Two years ago, AmeriHealth Mercy Health Plan, a Philadelphia Medicaid managed care organization that serves about 80,000 Medicaid recipients in 19 counties, launched a program aimed at identifying its "frequent flyers," members who go to the ED four or more times a year and rack up charges that often toped $1,000 a visit for even minor complaints. One member visited the emergency department 56 times in a year, not once for a true emergency.

AmeriHealth Mercy contacted its frequent flyers, either in person or by phone, referring those with chronic conditions to case managers. For some members, reducing the number of ED visits was as simple as teaching a family how an asthmatic child should use an inhaler. AmeriHealth Mercy helped other members find a primary care physician. "A lot of folks just didn't know they needed to have a primary care doctor," says Sherry Knowlton, senior vice president and general manager at AmeriHealth Mercy. The savings to AmeriHealth Mercy thus far is about $350,000 per year.

Programs like AmeriHealth Mercy's will undoubtedly improve care for many Americans, but how big a bite they will take out of the high cost of American health care, and the enormous amount of excess care it delivers, is debatable. Borg acknowledges that getting patients more involved in their own care is only a first step. "It's more than a nibble," he says, "but instead of these being surgical approaches, it's whacking at things with a butter knife." Still, what health care observers should know from witnessing the result of Medicare price controls in the 1980s and the HMO era of the 1990s, is that health care providers have myriad ways of maintaining their incomes.

It's not a stretch to predict that if cost-sharing plans result in significantly lowered revenue streams for hospitals and physicians, providers will likely respond by increasing their volume, just as they did in the face of lowered reimbursements in the 80s and 90s. If the United States hopes to rein in health care inflation and reduce overutilization, it's going to take more than getting patients on the Web and asking them to share more of the costs.