Squandering Medicare’s Money

San Francisco

MEDICARE has suddenly taken center stage in American politics, with Democrats now trying to score an advantage from the unpopularity of the Republican plan to overhaul the government health insurance program. Apart from the politics, though, Medicare’s financing challenges are worsening: this month, Medicare’s trustees projected that the insurance program would become insolvent by 2024, five years earlier than previously estimated.

Much has been said about the growing gap between the program’s spending and revenues — a gap that will widen as baby boomers retire — but little attention has been focused on a problem staring us in the face: Medicare spends a fortune each year on procedures that have no proven benefit and should not be covered. Examples abound:

• Medicare pays for routine screening colonoscopies in patients over 75 even though the United States Preventive Services Task Force, an independent panel of experts financed by the Department of Health and Human Services, advises against them (and against any colonoscopies for patients over 85), because it takes at least eight years to realize any benefits from the procedure. Moreover, colonoscopies carry risks of serious complications (like perforations) and often lead to further unnecessary procedures (like biopsies). In 2009, Medicare paid doctors more than $100 million for nearly 550,000 screening colonoscopies; around 40 percent were for patients over 75.

• The task force recommends against screening for prostate cancer in men 75 and older, and screening for cervical cancer in women 65 and older who have had a previous normal Pap smear, but Medicare spent more than $50 million in 2008 on such screenings, as well as additional money on unnecessary procedures that often follow.

• Two recent randomized trials found that patients receiving two popular procedures for vertebral fractures, kyphoplasty and vertebroplasty, experienced no more relief than those receiving a sham procedure. Besides being ineffective, these procedures carry considerable risks. Nevertheless, Medicare pays for 100,000 of these procedures a year, at a cost of around $1 billion.

• Multiple clinical trials have shown that cardiac stents are no more effective than drugs or lifestyle changes in preventing heart attacks or death. Although some studies have shown that stents provide short-term relief of chest pain, up to 30 percent of patients receiving stents have no chest pain to begin with, and thus derive no more benefit from this invasive procedure than from equally effective and far less expensive medicines. Risks associated with stent implantation, meanwhile, include exposure to radiation and to dyes that can damage the kidneys, and in rare cases, death from the stent itself. Yet one study estimated that Medicare spends $1.6 billion on drug-coated stents (the most common type of cardiac stents) annually.

• A recent study found that one-fifth of all implantable cardiac defibrillators were placed in patients who, according to clinical guidelines, will not benefit from them. But Medicare pays for them anyway, at a cost of $50,000 to $100,000 per device implantation.

The full extent of Medicare payments for procedures with no known benefit needs to be quantified. But the estimates are substantial. The chief actuary for Medicare estimates that 15 percent to 30 percent of health care expenditures are wasteful. Medicare spending exceeded $500 billion in 2010, suggesting that $75 billion to $150 billion could be cut without reducing needed services.

Why does Medicare spend so much for procedures and devices on patients who get no benefit and incur risks from them?

One reason is that Medicare’s reimbursement procedures are not sophisticated enough to track the appropriateness of the care provided. Medicare delegates its claims administration to private local contractors based on how quickly and cheaply they can process claims.

These contractors have few incentives to audit the taxpayer dollars they are paying out, and even if they wanted to, they would need information often not available on the claim form. For example, a claims administrator, processing a claim for a screening colonoscopy, does not know when the patient’s last colonoscopy was, or whether there was a new clinical reason for repeating it. While this information is available, finding it would require extra steps, and there are no incentives to do so.

Moreover, denying payment after a procedure is performed invites the wrath of both patient and physician. Medicare and private insurers are also keen to avoid situations that could be viewed as telling doctors how to practice medicine — even if such advice is in the patient’s best interest. The political sensitivity of limiting services based on age, for example, was illustrated by the uproar over the Preventive Services Task Force’s finding two years ago that women in their 40s do not benefit from routine mammography.

Another factor is the shocking chasm between Medicare coverage and clinical evidence. Our medical culture is such that if the choice is between doing a test and not doing one, it is considered better care to do the test. So while Medicare is obligated to follow the task force’s recommendations to cover new preventive services, it has no similar mandate to deny coverage for services for which the task force has found no benefit.

Changing the system would be relatively easy administratively, but would require a firm commitment to determining whether tests and procedures truly benefit patients before performing them. Unfortunately, in a political environment in which doctors providing end-of-life counseling are called death panels, and in which powerful constituencies seek to preserve an ever-increasing array of procedures and device sales, this solution remains hidden in plain view.

Of course, doctors, with the consent of their patients, should be free to provide whatever care they agree is appropriate. But when the procedure arising from that judgment, however well intentioned, is not supported by evidence, the nation’s taxpayers should have no obligation to pay for it.