Brain scan sheds light on Medicare's technology decisions

The development of a new medical technology can take years, but so, too, can the process of persuading Medicare to pay for it.

The list of services the nation's largest payer of health care covers is constantly changing to account for advances in drugs, devices, screenings and other types of physician care for patients. Medicare is guided by its enabling legislation to apply a “reasonable and necessary” standard when adding to its benefit package, but interpreting and applying those criteria has been challenging — and at times controversial.

Medicare officials can be influenced by nonclinical factors — including politics — when trying to determine the value of a service that might be used by thousands of physicians to treat millions of patients. Evidence regarding the projected cost of covering new technology is not supposed to be part of the discussion, but a favorable determination for a service can translate into hundreds of millions of dollars in additional program spending over the long term, and the program's stewards are sensitive to those dollar figures.

The process should include more of an incentive to approve technology that will be cost effective and improve quality in Medicare, argued Harold Miller, executive director of the Center for Healthcare Quality and Payment Reform in Pittsburgh. But there has been resistance to such an approach.

In 1989, Medicare proposed adding the words “cost effective” to coverage criteria deliberations, but the blowback from industry members and other stakeholders led the agency to withdraw the idea, said Peter J. Neumann, ScD, director of the Center for the Evaluation of Value and Risk in Health at the Institute for Clinical Research and Health Policy Studies at Tufts Medical Center in Boston. At least in their public deliberations, Medicare officials are directed to discuss only whether a new service will improve patient outcomes.

But despite efforts to prohibit discussions of cost, the issue inevitably will come up, as it did during a Jan. 30 meeting of the Medicare Evidence Development & Coverage Advisory Committee.

The committee gathered testimony and asked expert witnesses questions about using beta amyloid positron emission tomography scans to help diagnose dementia and neurodegenerative disease. When the discussion veered toward how much the Centers for Medicare & Medicaid Services would be expected to spend on the service if it were covered, an official interjected. CMS, as a matter of policy, does not consider cost during a coverage determination process unless it's connected to such other relevant factors as preventing admissions to the hospital, the official reminded the panel.

The projected up-front costs of a service under consideration often come up at hearings even though coverage decisions officially are made based on the downstream effects of ordering the service and its overall value to Medicare as an insurer, to physicians and to patients, said Steven Pearson, MD, president of the Institute for Clinical and Economic Review in Boston and a former vice chair of the Medicare Evidence Development & Coverage Advisory Committee. ICER and Dr. Pearson presented a white paper on Alzheimer's diagnostics tests to inform insurance policy during the most recent MEDCAC hearing.

“If it cost 10 cents, there would be less of an emphasis or scrutiny of it,” Dr. Pearson said. The PET scan used to show amyloid is estimated to cost $3,000 per screening, according to MEDCAC testimony.

Measuring confidence in a technology

MEDCAC flies largely under the radar considering the role the advisory panel plays in the health care system. A few times a year it holds public meetings at CMS headquarters in Baltimore that are broadcast on the Internet to review medical studies, technology assessments and other sources of information on the effectiveness and appropriateness of a major proposed new Medicare service. The committee provides an open forum to gather testimony from stakeholders and make a recommendation to the agency based on evidence, Dr. Neumann said.

Ultimately, CMS will decide whether to approve national coverage of the service and can disregard the panel's advice.

During the amyloid imaging hearing, panel members and witnesses discussed how a positive or negative Alzheimer's screen using the technology might change a physician's behavior or treatment choices for the patient. The panel indicated that it saw little evidence to change patient management.

On a scale of one to five, voting MEDCAC members gave PET imaging of brain beta amyloid a 2.14 score, showing that they had low to intermediate confidence in evidence that the test would improve outcomes. They also had little doubt that the conclusion applied to the general Medicare population.

The MEDCAC process can be challenging for advocates of a new technology trying to obtain a favorable opinion for national coverage, said Dr. Neumann, who has studied and written about the Medicare coverage decision-making process. The program allows individual Medicare administrative contractors to make their own coverage decisions at the local level. That might seem preferable to some drug and device makers, but that leads to confusing coverage variation throughout the country.

A national review that would ensure coverage continuity takes a long time and can be contentious, Dr. Neumann said. A request for coverage that doesn't require a technical assessment or a MEDCAC hearing is decided within six months. Other determinations can take nine months, with additional time tacked on for public comment before a final decision is reached.

At this article's deadline, the Medicare agency had not decided whether to cover the amyloid imaging service.

The Jan. 30 MEDCAC meeting was not the first time the issue of using PET imaging to diagnose Alzheimer's disease has come up before the committee. The panel's predecessor, the Medicare Coverage Advisory Committee, in 2002 reviewed using fluorodeoxyglucose PET scans for evaluating patients with suspected dementia. At the time, the Medicare agency issued a noncoverage determination, but it changed course under heavy political pressure from Washington.

A 2004 Washington Post article, “A Tale of Politics: PET Scans' Change in Medicare Coverage,” detailed how the late Sen. Ted Stevens (R, Alaska), a legendary appropriator in the upper chamber of Congress, became involved in the debate to ensure adequate payment for the PET service. CMS decided to cover the test, but only narrowly.

Making cases for and against coverage

The Food and Drug Administration's April 2012 approval of Amyvid (florbetapir F 18), a drug used in PET imaging of the brain to show evidence of cognitive decline, helped push forward the latest coverage discussion of the diagnostic service.

Despite the MEDCAC low-confidence vote, the testimony from medical experts was overwhelmingly positive, said Wei-Li Shao, senior director of Eli Lilly and Co.'s Alzheimer's business division, in a statement. Eli Lilly, which makes the radioactive diagnostic agent Amyvid, will continue pushing for the imaging coverage.

“When making a final coverage decision, we encourage the Centers for Medicare & Medicaid Services to heavily consider the real-world medical experience presented today, and the appropriate use criteria recently released by the Society of Nuclear Medicine and Molecular Imaging and the Alzheimer's Assn. supporting the use of these imaging agents,” Shao said.

The organizations assembled experts to review research to produce evidence-based recommendations on appropriate use. There is a role for PET scans in the fight against Alzheimer's, said Maria Carrillo, PhD, senior director of medical and scientific relations at the Alzheimer's Assn. The panel released its appropriate use criteria to be published in The Journal of the Alzheimer's Association and The Journal of Nuclear Medicine on Jan. 28.

“The basic premise is they have to be done in concert with a clinical diagnostic workup by a dementia specialist,” Carrillo said. The test itself does not replace the clinical evaluation, in which there is a cognitive complaint and a confirmed impairment, she later added.

MEDCAC's vote was unfortunate, Carrillo said. The panelists sought more information on how the tests might change a physician's treatment decision and improve patient outcomes. Because the test is relatively new, information on improved outcomes will not be understood fully for several years, she said.

Neurologist David Kuhlmann, MD, from Bothwell Regional Health Center in Sedalia, Mo., said during his presentation that studies have not asked whether patients would do better as a result of the PET scan. Dr. Kuhlmann argued against Medicare coverage.

A wrong coverage decision can be harmful. Patients in their mid-50s, for example, might be screened and started on anti-amyloid therapies even though 30% of adults with normal cognition would trigger positive amyloid findings, he said. Some of those placed on those medicines incorrectly might end up continuing treatment for the rest of their lives.

“If they remain cognitively normal, the doctor will tell them it is working and they should continue therapy, even though there may not be a reason why their cognition remains normal,” Dr. Kuhlmann said.

Proponents of coverage, on the other hand, argued that there are several patient benefits to obtaining a timely and definitive diagnosis — something MEDCAC did not understand, Carrillo said. For instance, an early diagnosis allows a patient and family to begin preparing for the onset of Alzheimer's. Financial planning would allow more families to ensure that loved ones are cared for, and more Alzheimer's patients could join clinical trials early on, she said.

“Early and timely diagnosis is important,” Carrillo said. “It's not just about medications.”

Since the MEDCAC vote in January, the Alzheimer's Assn. said it has met face to face with CMS officials to emphasize these points. At the end of the decision-making process, they hope such pressure is enough that Medicare will be persuaded to pay for the diagnostic service.