Could a simple test save Medicare hundreds of millions?

Study looks at cost-effectiveness of testing heart patients to help decide who needs implanted defibrillators

ATLANTA -- Next week, the Medicare agency will announce whether it will cover the cost of a $400 heart test that assesses a person's risk of dying suddenly from a heart condition. Today, a study led by University of Michigan researchers suggests that the test could actually save Medicare hundreds of millions of dollars in the long run.

The researchers used a sophisticated computer model to calculate the potential impact of using the test, called microvolt T-wave alternans (MTWA), to help determine which patients would benefit most from implanted devices that automatically re-start a stopped heart.

Those devices, called implantable cardioverter defibrillators, or ICDs, have been shown to save lives that would otherwise have been lost to sudden cardiac death, which kills 300,000 Americans each year.

One year ago, Medicare expanded its coverage of ICDs to include many more heart patients; an estimated 500,000 people over age 65 are now candidates. Covering the cost of ICDs for patients who have the same characteristics as participants in the MADIT-2 ICD trial that led to Medicare approval could add $3 billion annually to the program's budget. ICDs cost $35,000 including implantation, require periodic battery replacement, and pose a device failure risk.

In a presentation at the American College of Cardiology Scientific Session, the U-M team suggests that the MTWA test could spare a significant fraction of those costs, if doctors focused on providing ICDs only to patients with an abnormal or inconclusive MTWA test result. Previous studies have suggested that patients with a negative, or normal, MTWA test result have a far lower risk of sudden cardiac death than others.

This risk-stratification of ICD candidates who meet MADIT-2 criteria, say the U-M researchers, could potentially save Medicare $690 million a year -- even after the cost of the MTWA test and the care of patients who don't receive ICDs are included.

"ICDs have been shown in several studies to be cost-effective, which means the cost is considered acceptable given the benefit to patients," says lead author and U-M cardiology fellow Paul Chan, M.D., M.Sc. "But a very expensive device can be cost-effective and still not affordable to society, if the condition it treats is highly prevalent. Our study demonstrates the potential impact of using additional factors to aid decisions about the use of expensive devices."

The MTWA test, he notes, is gaining acceptance among cardiologists but will likely become more common if Medicare decides to cover it.

Microvolt T-wave alternans are small variations in the electrical impulses in the heart. They can't be detected on the conventional heart-rhythm test called the electrocardiogram (ECG), but MTWA test equipment can detect them via special sensors placed on the patient's body during a brief exercise session. Two companies make equipment that are used to perform MTWA tests, but Chan and his colleagues have no association with those companies, nor with companies that make ICDs.

At the ACC meeting, Chan will receive the Parmley Prize for his previous research on the use of the MTWA test as a way to predict sudden cardiac death in patients with heart failure. In a study he presented last fall at the American Heart Association Scientific Sessions that will soon be published in the Journal of the American College of Cardiology, he showed that ICD-eligible patients who met MADIT-2 trial criteria for primary prevention of sudden cardiac death, and who had positive or inconclusive MTWA results, were more than twice as likely to die than similar patients with negative MTWA results. Other MTWA studies have suggested a larger difference, but did not control for other patient characteristics. Chan's study also showed that about one-third of patients who meet Medicare criteria for an ICD have normal MTWA test results.

For the new study being presented at the ACC meeting, which has been accepted for publication in JACC, Chan and his colleagues used a computer model called a Markov analysis to simulate the costs, benefits, and health outcomes of 65-year-old individuals who meet the MADIT-2 trial criteria for Medicare coverage of an ICD. They looked at three scenarios: all such patients get an ICD, only those with positive or inconclusive MTWA results receive an ICD, or no patients get an ICD.

The model used a range of assumptions about what percentage of patients would suffer sudden cardiac problems and long-term complications with or without an ICD, and how many ICD patients would suffer complications related to their ICDs, It also included assumptions about what proportion of patients would die depending on their treatment or MTWA status. All of these assumptions were based on data from previous ICD trials, or best case/worst case scenarios. Costs for care were also included, such as MTWA test costs, ICD costs, and the cost of medical care for those who stayed healthy or suffered a heart problem with or without an ICD.

To account for the uncertainties inherent in each of the clinical variables, the researchers performed various levels of sensitivity analyses, including an analysis using 10,000 simulations in which all model variables were randomly sampled. This latter analysis, known as a Monte Carlo simulation, allowed them to examine the distribution of possible cost-effectiveness ratios for every quality-adjusted life year (QALY) saved, and is a standard way to assess the cost-effectiveness of a medical intervention. It allows researchers to assess the cost of saving a year of a person's life, adjusted for the quality of that life and the life-saving power of a particular course of action.

In general, the cost per QALY for giving ICDs only to risk-stratified patients who had been tested with MTWA was about $50,000, compared with medical therapy. "This is well within the usual range considered cost-effective for new medical technologies," says Chan. However, compared with MTWA testing followed by ICD use only in those with a positive or inconclusive MTWA result, the all-ICD option would cost $88,700 per QALY gained -- under best-case scenarios.

Chan cautions that the new results are not meant to guide doctors in decisions about whether or not to suggest that particular primary prevention patients receive ICDs. But he hopes the data may be useful in decisions about how to spend resources for this type of care -- and that further research will examine how often doctors should conduct MTWA tests in patients who don't receive an ICD, to help them monitor for changes in sudden cardiac death risk over time that might warrant an ICD.

In addition to Chan, the authors of the ACC presentation are: Kenneth Stein, M.D., of Weill Cornell Medical Center; Theodore Chow, M.D. of the Ohio Heart and Vascular Center; A. Mark Fendrick, M.D., and Sandeep Vijan, M.D., M.Sc., of the U-M Division of General Medicine; and J. Thomas Bigger, M.D., of Columbia University.


Last reviewed: By John M. Grohol, Psy.D. on 21 Feb 2009
    Published on All rights reserved.



A man can get discouraged many times, but he is not a failure until he begins to blame somebody else and stops trying.
~ John Burroughs