Feature
The Medical Arms Race
By Jack Gordon
How much high-tech medical equipment does Minnesota really need? And how much can we afford before the system collapses under the expense?
Last December, Medica announced that beginning in March, it will require physicians to contact a third-party radiology consultant prior to ordering high-tech imaging such as CT, MRI, MRA, or PET scans. HealthPartners, Blue Cross and Blue Shield of Minnesota, and the Minnesota Department of Human Services have similar plans in the works for prior consultations or authorizations for imaging.
High-tech imaging is the current focus of insurance companies seeking to curb spending associated with expensive medical technology. It’s also a prominent example of what a growing number of critics of the current health care system describe as a “medical arms race” in action. The complaint coming from a number of quarters is that the proliferation of high-priced technology is driving up health care costs to crippling levels with insufficient evidence of corresponding benefits to patients—and no regard at all for cost-effectiveness.
According to the National Coalition on Health Care in Washington, D.C., total U.S. health care spending reached $1.9 trillion in 2004, representing 16 percent of the gross domestic product. Spending is expected to reach $4 trillion—20 percent of GDP—by 2015. Between 2000 and 2005, employer-sponsored health insurance premiums increased 73 percent, compared with cumulative inflation of 14 percent and cumulative wage growth of 15 percent during the same period.
The Minnesota Department of Health reports total health care spending of $25.6 billion in the state in 2004, up 5.4 percent from 2003, compared with a national one-year increase of 7.9 percent.
Regarding technology’s contribution to those escalating costs, concern has focused especially on imaging systems. According to a 2006 report from HealthLeaders-InterStudy of Nashville, which compiles information for the managed-care industry, radiology costs “have quietly risen to be 10 percent of the health care dollar and are growing at a rate of 18 to 20 percent a year.” That is approximately double the growth rate of prescription-drug costs, which receive far more attention, the report says.
Runaway Driver
When it comes to adding new high-tech equipment and expensive services to the health care system, critics say many forces act as feet on the accelerator. These include patient demand and physicians’ sincere desire to provide the best possible care, as well as economic incentives. But there is no corresponding foot on the brake.
In Minnesota, those critics include physicians, as well as insurers, pundits, and policymakers. One question they raise is whether expensive, high-tech tests and procedures are always used appropriately for individual patients. Others are more fundamental to the workings of the health care system as a whole: How many children’s hospitals, heart centers, bariatric centers, robotic surgery systems, PET scanners, MRI scanners, and so on does a given population area need to serve its genuine health care requirements, as opposed to the maximum convenience of patients or the economic interests of physicians and hospitals?
Former U.S. Sen. David Durenberger, chair of the National Institute of Health Policy (NIHP) at the University of St. Thomas in Minneapolis, is one of the more vocal critics. Last July, the NIHP launched a two-year series of conferences intended to come to grips with what it calls the “medical arms race syndrome.”
As an example, he points to the CyberKnife, a radiosurgery system that enables treatment of intracranial and extracranial tumors previously considered inoperable. According to news reports, a CyberKnife lists for more than $3.5 million. In 2003, St. Joseph’s Hospital in St. Paul was the only center in the Upper Midwest that had one. Now there is another CyberKnife at Miller-Dwan Hospital in Duluth. “Where did all the extra patients come from?” Durenberger asks. “Was there a long waiting line at St. Joseph’s? I suspect not.”
Another case in point: the da Vinci robot, a $1 million-plus system used for laparoscopic surgeries such as gallbladder removals, prostatectomies, and bariatric surgery. In 2004, three Minnesota hospitals (the University of Minnesota Medical Center, HealthEast St. John’s, and Mayo Clinic in Rochester) had da Vinci systems. By 2006, the robots were in six hospitals, including facilities in Duluth and St. Cloud, according to manufacturer Intuitive Surgical Systems of Sunnyvale, California. Surgeons report a number of benefits these systems provide, such as an improved view of the surgical field and the elimination of tremor in movement. But a study of the 202 initial robotic abdominal surgery cases performed at Johns Hopkins University Hospital in Baltimore found that “clinical data demonstrating improved [patient] outcomes are lacking.”
The concern is that the proliferation of CyberKnifes and da Vincis and free-standing $2-million MRI scanners and $1-million CT scanners may be driven less by patient need than by market share or providers’ desire to use the latest technology to attract specialists who do procedures that pay well: gastroenterologists, cardiologists, and neurosurgeons, for example. A claim on the Web site of Sunnyvale, California-based Accuray, CyberKnife’s maker, is that buyers say the device “attracts new patients and increases their practice volumes.”
Considered on its own, each new addition to the technology arsenal is wonderful, Durenberger says. Whether it’s a robot or a drug-eluding stent or a new atrial fibrillation stroke preventer, “it’s all incredibly useful,” he says. “There’s not a thing you can think of that isn’t good.”
But the issue of cost-effectiveness is nowhere addressed, he says. If you asked how many CyberKnifes or da Vincis or PET scanners a given market ought to have in order to provide high-quality and reasonably convenient patient care without driving up costs unnecessarily, Durenberger says he couldn’t tell you and neither could anyone else. The right questions haven’t been asked, and the right information isn’t available.
Why Market Forces Don’t Cut It
Elsewhere in the U.S. economy, the question “How much is too much?” is answered more or less automatically by the free market. According to those advancing the arms race theory, however, market forces don’t operate in health care the way they do in other industries. “In any other industry,” Durenberger says, “when someone invents something new, market mechanisms determine whether it catches on and to what extent.” For instance, if someone invents a video game, “there is a market that tells you this is something that X number of people want and are willing to pay X dollars for. That analogy applies practically everywhere except in health care.”
In health care, the checks and balances inherent in ordinary market systems do not operate “to temper our enthusiasm for novelty and innovation,” Durenberger says. There are two reasons. First, consumers—that is, patients—don’t know what treatments they need; they rely on physicians to tell them. Second, the consumer doesn’t pay directly for those treatments; private or public insurers do. “So someone else decides what we need, and someone else pays for it,” he says. “That’s true for our personal health, and it’s true of [health care] decisions made for us as a community.”
Because of those peculiarities in the system, it is a myth that free-market competition among providers works to reduce prices and improve quality in health care, as it does in other industries, says Allen Horn, M.D., president of CentraCare Clinic in St. Cloud, a not-for-profit multispecialty group.
A supporter of efforts to grapple with the arms-race issue, Horn says that if competition among medical providers reduced prices, one would expect to see higher hospital charges in St. Cloud, which has one hospital, than in the Twin Cities, where a number of facilities compete. Yet as he points out, a browse through the Minnesota Hospital Association’s PriceCheck Web site reveals that charges for most procedures are higher in the Cities than in St. Cloud and other less-competitive regions of the state. “If you built two more hospitals in St. Cloud to make the market more competitive,” Horn insists, “prices would go up, not down.”
Horn allows that competition does improve customer service, which “used to be godawful,” since that is the primary criteria by which consumers actually rate their own medical care: “If you ask someone, ‘Is your physician any good?’ they’ll say, ‘Yes, he listens to me, he explains things to me.’” As for the actual quality of care, however, “patients just have to assume that doctors know what they’re doing,” Horn says. And, of course, money is no object if the patient’s health is at stake and the insurer to whom the patient has been paying those high premiums is picking up the tab.
When competition takes the form of “you have an MRI scanner, so I need one too,” Horn says, “there is really no data to support the conclusion that this improves the quality of care that patients receive.”
Driving Blind
In its April report “Developing Informed Decisions: Seeking Market Reforms to Advise Medical Facility Expansion,” the Citizens League framed the dilemma this way: “Minnesota is in the midst of a wave of medical facility investment, yet we lack the basic information to make good decisions about the expansion of medical facilities. We don’t have a functioning market to do it for us, and there is no process in place to inform decisions or to make needed changes.”
As an example, the report cites the state Legislature’s 2006 decision to allow North Memorial Health Care and Fairview Health Services to build a new 300-bed hospital in Maple Grove: “For the first time since Minnesota established a hospital-bed moratorium in 1984, there was competition to build a new hospital facility. At one point as many as 40 lobbyists were working at the Legislature in some capacity on this issue. Several providers of medical care reportedly spent millions of dollars in an effort to convince the Legislature that they should be the provider allowed to build a hospital in Maple Grove.… There was no process to guide the Legislature or the providers in responding to this competitive situation.”
The point is not that it is a bad idea to build a new hospital in Maple Grove, says Bob DeBoer, director of policy development for the Citizens League. Rather, the League’s concern is that neither the Legislature nor anyone else knows whether it is a good idea from a cost-effectiveness standpoint. The information isn’t there.
Data from the Minnesota Department of Health, which keeps track of capital expenditures for projects totaling more than $1 million, show that from 1993 to 2006, providers in the state purchased 173 new MRI scanners, 171 CT scanners, and 16 PET scanners. During that same period, the number of outpatient surgery centers increased from 10 to 46. But the Citizens League says that what’s lacking are the answers to fundamental questions such as What medical services are currently available in all medical facilities? and What is the capacity and use of existing medical facilities?
The current system operates under “supplier-induced demand,” DeBoer says. Facilities, equipment, and services are added because suppliers think they will make money. But because the money comes from third-party payers, the system gets “no meaningful signals from the consumer side. The doctor is the proxy for the patient, and insurance companies are proxies for payment.”
Supplier-induced demand “doesn’t mean there is a nefarious plot to raise costs,” DeBoer says. But physicians and hospitals “naturally will compete for high-reimbursement services.” And once those services and facilities are in place, he says, “the tendency will be to use them. If someone invests in high-quality, high-cost equipment, they’ll want to use it even if the benefit is marginal.” Couple the physician’s natural desire to do everything possible for the patient with the need to make expensive equipment pay for itself, and it becomes difficult to sort out quality-of-care issues from monetary considerations.
“Especially as treatment patterns change from inpatient to outpatient,” DeBoer says, “we don’t have a good sense of what we really need, or what we have overcapacity in. We just know that overcapacity leads to higher costs.”
On the national scene, not all critiques of supplier-induced demand are that delicate. A December story in the New York Times singled out intensity modulated radiation therapy (IMRT), a treatment for prostate cancer that “stands out for its profit potential,” and raised pointed questions about how often it is selected precisely because it is expensive. IMRT “can mean reimbursement of $47,000 or more a patient,” which is “many times the fees that urologists make” on prostate treatments such as surgery and radioactive seed implants, the Times reported. This “may help explain why urologists have started buying multimillion-dollar IMRT equipment and software, and why many more are investigating it as a way to increase their incomes.”
A Complicit Few or Many?
Questions about an inherent conflict of interest arise especially when physicians own their own expensive equipment. In such cases, orders for high-tech radiology scans, for instance, amount to self-referrals. A recent report on high-tech imaging by Blue Cross and Blue Shield of Tennessee cites national studies showing that physicians who own their own radiology equipment are two to seven times more likely to order an imaging test.
Medica’s new requirement for prior consultations for high-tech imaging includes exemptions for tests ordered in emergency, urgent-care, or inpatient hospital situations. The primary target appears to be outpatient services. Even then, Medica’s chief medical officer, Charles Fazio, M.D., stresses that the requirement is for a consultation, not a pre-authorization, and says that the decision will remain in the physician’s hands.
Why single out radiology rather than, say, robotic surgery systems? Because high-tech scanners have become so ubiquitous: Witness the Minnesota Department of Health’s count of 173 new MRIs and 171 CT scanners added to the system from 1993 to 2006. Fazio says that Medica now pays for about 170,000 high-tech imaging studies in Minnesota per year. If 10 or 20 percent of those scans are either unnecessary or inappropriate for the clinical situation, “we’re talking $10 or $15 million a year” for Medica alone. For the state as a whole, he says, inappropriate imaging might cost $50 million or $60 million a year.
“We have an extraordinary number of CT, MRI, and PET scanners in this community,” Fazio says. “Two years ago, our figures for high-tech imaging showed that the number of tests being done were 15 to 20 percent beyond what evidence-based [uses] would suggest we need. That leads us to conclude that the [plain] availability of a machine leads to more tests.”
This may be especially true if the machine is owned by the physician ordering the test. Medica “made no friends” with its consultation announcement, Fazio admits. Indeed, physician reaction was a combination of anger and concern, according to the Minnesota Medical Association. But from physicians who are sympathetic in principle, he says, “the feedback we’ve gotten most often is from doctors who don’t own their own machines telling us to focus on preventing self-referrals.” Many physicians are convinced that is the real problem, Fazio says. But Medica thinks it’s only part of it. Data from Minnesota is lacking, but studies elsewhere have shown that practitioners who own their machines order 20 to 30 percent more images than practitioners who don’t own machines, he says.
But Fazio says a 2006 pilot test of the consultation program indicated a different problem. Although Medica’s findings may be skewed by the fact that physicians who participated in the pilot were volunteers, the data doesn’t point to a small number of bad actors. “There’s no Pareto Principle [80 percent of consequences stem from 20 percent of causes],” he says. “It’s not that a few people are making 80 percent of the non-evidence–based decisions.” Instead, the problem comes from physicians who order the right test 80 or 90 percent of the time but have trouble making a good call in one or two out of 10 cases.
Elusive Remedies
If a medical arms race is, indeed, creating overcapacity and generating unreasonable costs, insurers are likely to respond with more demands for consultations and pre-authorizations. What of other ideas for how to apply the brakes?
Durenberger says flatly that relying on “government” is not the way to solve the problem. “We have to find other solutions,” he insists. When asked what those solutions might be, Durenberger admits he doesn’t know. “You’re asking me to answer in 20 minutes what we’ll take two years to study,” he replies.
Horn has reservations about government intervention but says he sees no other entity that could meaningfully address the issue.
For Minnesota, specifically, the Citizens League recommends that the state establish a “permanent, quasi-public body to act as a consumer voice in medical care decision-making.” That body’s first task would be to “oversee the gathering of statewide information” about the current capacity and use of medical facilities.
Meanwhile, the Legislature has ordered the Minnesota Department of Health to report by February 15 on the need for a new process for approving the construction of medical facilities or the addition of certain services at existing ones.
Whatever the answer may be, Durenberger says, “We’re having more done to us than needs to be done.… With every investment in medical technology, people have to make offsetting decisions someplace else. And at some point, people just can’t afford whatever it is.” MM
Jack Gordon is a freelance writer in Eden Prairie.