Questions about cost effectiveness are paramount as payers slowly integrate new diagnostic tools. Evidence to help them make careful coverage determinations is often incomplete.
Abstract
Formulary decisions might be considered a snap when compared to decisions about coverage of a genetic test. The presence – or lack – of credible data to demonstrate cost-effectiveness is usually the touchstone.
Not long ago, a technology proving cost-effectiveness over five years “was a strong recommendation to payers,” says Frank Papatheofanis, MD, MPH, PhD, of Aequitas. Today, because of enrollee turnover, he says, “we’re down to a three-year window.”
PHOTOGRAPH BY ROBERT BURROUGHS
What does it mean when major health insurers say they prefer not to comment on the topic of genetic testing?
America’s Health Insurance Plans President and CEO Karen Ignani may have unwittingly provided an answer in the news release that accompanied the trade association’s Board of Directors Statement on Genetic Testing and the Use of Genetic Information, back in 2002.
“The possibility of genetic-based research and treatment to treat and cure a variety of conditions is vast,” said Ignani, “but it is equally clear that when it comes to understanding the benefits and risks [of] this emerging science, questions far outweigh answers.”
Aetna Chairman and CEO John W. Rowe, MD, a recognized leader in encouraging genetic testing with proven clinical utility, led the board committee that developed the AHIP (then the American Association of Health Plans) document.
Almost three years later, payers’ questions about genetic testing still far outweigh the answers, especially questions about cost-effectiveness.
“The value proposition in genomics is that diagnostic tests will help determine which patients will benefit from a drug or other intervention and potentially make that drug or intervention available to them earlier in their treatment,” says Aetna Senior Medical Director Joanne Armstrong, MD. “You may actually get some cost savings by avoiding ineffective therapies and adverse effects, but — for most of these tests — that value proposition has not been demonstrated.”
Some tests — Armstrong mentions HER2/neu testing for receptivity to trastuzumab (Herceptin) therapy, and genotyping for HIV and hepatitis C therapy as examples — are covered by various benefit plans, but most of the 824 clinical genetic tests currently available are not.
SAW IT COMING
None of this will come as a surprise if you caught “Personalized Medicine” when it was released by Forrester Research in 2000. Predicting that “Science will produce a torrent of products in need of payers,” principal author Michael Barrett nailed many of today’s trends, such as health care cost shifting, consumer self-direction, genomics, and the concern about discrimination based on genetic information.
Here’s his take, five years ago, on how health insurers would deal with baby boomers who pay out of pocket for genetic tests that predict their disease susceptibilities — and then try to protect themselves from future medical costs by buying more health insurance.
As insurers experiment with consumer-directed plans, they will move toward the individual market, only to encounter the trap of adverse selection. Barrett believes that, after a period of turmoil around equity issues, insurers will fashion hybrids — combining voucherlike shopping and group-risk pooling — with the groups arising variously from employment, residence, or affinity ties.
We may not be there yet, but adverse selection is undoubtedly on payers’ minds when it comes to genetic testing. For now, their reluctance to pay for testing, and in some cases even to talk about it, seems to have to do with economics.
“No one wants to be in a position of denying a potentially useful treatment,” says Scott Ramsey, MD, PhD, an economist, internist, and professor at the University of Washington Medical Center. “On the other hand, health plans don’t have the data to know whether they should cover these tests.”
Ramsey, also the director of Outcomes Research at the Fred Hutchinson Cancer Research Center in Seattle, explains that it’s not just a matter of demonstrating that a test works in the lab (its analytical validity), or that it accurately predicts the presence or absence of a clinical condition or predisposition (its clinical validity), or that it provides information that is useful for prevention or treatment (its clinical utility). The point is, acting on the test result changes the outcome for a patient.
“Most tests go to market without high-quality clinical trials to tell where their value is.”
— Scott Ramsey, MD, PhD, Univ. of Wash.
“For example, there’s evidence that you can identify women at high risk of developing breast cancer, but that doesn’t necessarily mean [such a test] has saved lives,” Ramsey continues. “To demonstrate that, you would have to randomize women into groups that get the test and don’t get the test, and follow them forward and look at outcomes. Those studies are very hard to do, and to my knowledge, no one has done them for any genetic test.”
CEAS AND HTAS
Even tests with proven analytical validity, clinical validity, and clinical utility often must be shown to be more cost-effective than existing diagnostic tests (or, in Ramsey’s example, the group that does not get the test). Cost-effectiveness analyses (CEAs) compare an intervention, such as a genetic test, to one or more alternatives in terms of quality-adjusted life years.
“Cost-effectiveness is not just about up-front costs,” cautions Frank Papatheofanis, MD, MPH, PhD, president of Aequitas, a research and analysis firm that helps biotech, pharma, and medical device manufacturers make the case for payer coverage. “The acquisition cost associated with a technology may appear to be very high — but the value of that technology over. a period of time may confer advantages over alternatives with lower up-front costs.”
Health technology assessments (HTAs), on the other hand, focus only on effectiveness and don’t consider cost. Several organizations, including the Blue Cross and Blue Shield Association’s Technology Evaluation Center (TEC), ECRI, and Hayes Inc., provide both HTAs and CEAs to the Centers for Medicare and Medicaid Services and private payers. Like a CEA, an HTA compares a candidate technology to current or emerging alternatives.
The Blues’ TEC, for example, evaluates drugs, devices, procedures, and biological products using five criteria (see box). Meeting these criteria does not guarantee coverage, but they are important considerations for coverage decisions.
BCBS TECHNOLOGY EVALUATION CRITERIA.
|
SOURCE: BLUE CROSS BLUE SHIELD ASSOCIATION
Provided to all Blue Cross and Blue Shield member plans, TEC assessments are researched and drafted by a staff of physicians and doctorate-level scientists who are supported by professionals in nursing, pharmacology, statistical methods, and library science. They are based on clinical and scientific evidence and evaluate whether a technology improves health outcomes, such as length of life, quality of life, and ability to function. TEC is also one of 13 Evidence-Based Practice Centers for the U.S. Agency for Healthcare Research and Quality.
PUBLISH OR PERISH
In the real world, test developers don’t wait in line for CEAs and HTAs, at least not initially. Developers typically collaborate with independent academics and researchers in launching studies to demonstrate the value of their genetic tests.
“You discover the genes, you find lots of mutations, and you try to determine the risk of an individual with one or more mutations for developing cancer and what can be done about it,” explains Gregory Critchfield, MD, president of Myriad Genetic Laboratories, which has developed and marketed predictive genetic tests for breast, ovarian, pancreatic, colorectal, uterine, and skin cancer. “Studies have to be built all along this process to tell you that interventions will make a difference. Once you have solid data, professional societies generally will issue opinions and insurers will be in a position to say, ‘We will reimburse.’”
An important link in this evidentiary chain is the publication of studies in peer-reviewed journals. For Myriad’s BRACAnalysis breast cancer test, the key was a 2000 article in JAMA.1 For Exact Sciences’ PreGen-Plus, a test that detects acquired mutations associated with colorectal cancer, the breakthrough article was in the New England Journal of Medicine last December.2
“The fact that we had done the study and had presented the data at a medical meeting was almost irrelevant to the payers,” says Exact Sciences President and CEO Don Hardison. “They wanted to see it in a peer-reviewed journal.”
Typically, test developers contract with one or more reference labs, which in turn negotiate a rate with payers for a genetic test, based on the CPT codes a test involves. But that’s only the first of a two-part process, says Hardison. It’s more difficult — but potentially more lucrative — to include a genetic test in a screening program. This requires that a payer make a policy decision to cover a test after a favorable technology review, a task Blue Shield of California is facing regarding Pre-Gen-Plus, Hardison says, following such a decision this past February by the California Technology Assessment Forum that the test meets all of Blue Shield’s national criteria.
WHY PAYERS MAY BAIL
So, has the advent of genetic testing resulted in consumers enjoying the benefits of genetic testing, labs expanding their menus of services, and payers saving downstream medical costs? Not quite. Third-party payers may have as many dis-incentives as incentives for getting on the genetic testing bandwagon.
Predictive and diagnostic genetic testing could result in dramatic treatment-cost reductions by identifying enrollees who would benefit from preventive strategies and earlier treatment. But would such an investment make economic sense when the average plan’s membership turnover is 20 percent a year?
“The enrollee may not be around long enough for the health plan to save the money it would have spent,” notes Papatheofanis. “Three or four years ago, if a technology could prove cost-effectiveness over five years, that was a pretty strong recommendation to payers. Because of enrollee turnover, we’re now down to a three-year window.”
Even if diagnostic testing and preventive services are part of a member’s benefit package, and even if a specific genetic test is covered, the plan may not necessarily consider that test medically necessary, given a particular member’s circumstances. This was noted in a draft report last February of the Secretary’s Advisory Committee on Genetics, Health, and Society,3 part of the National Institutes of Health and established in 2002 to study issues raised by genetic testing and make recommendations.
Payment rates are negotiated, but the added expense of genetic tests may be another disincentive for health plans.
“They’re expensive,” acknowledges Papatheofanis. “A lot of innovation goes into genetic test development. The materials are expensive. The intellectual property that goes into them is expensive. Developers are going to try to recoup that investment, so the early rollout of these tests will come with a higher price tag than a serum potassium test that’s been around for 30 years.”
The success of Myriad Genetic Laboratories and Exact Sciences Corp. in providing the kind of data payers want to see is not typical in the genetic testing industry, and for good reason. These studies take lots of patients, time, and money. Hardison says Exact Sciences spent $15 million on a study that included 5,500 patients in 80 centers.
“It’s the same for all medical tests, not just genetic tests,” says Ramsey. “Most go to market without high-quality clinical trials to tell where their value is. It’s not like the drug model, which is evaluated through prospective, randomized trials.”
It’s a situation that allows some payers to stall when it comes to making coverage decisions about genetic tests, says Rick Carlson, JD, clinical professor of policy programs at the University of Washington School of Public Health, in Seattle, where he is involved in pharmaceutical cost-effectiveness studies. Carlson has consulted for health insurers and large hospital systems on strategic planning and product development. Recently, he also provided public comment on genetic discrimination issues before the Secretary’s Advisory Committee.
“I understand how they make business decisions,” says Carlson, “and my practical experience of 30 years is that there are serious companies within the carrier community and companies that follow the leaders. The serious companies are asking for data because they genuinely want to make a rational decision about the issue. The others are more inclined to sit back and try to avoid the cost by waiting for somebody else to do the work.”4
The way Carlson sees it, many payers will not cover genetic tests because many such tests are not supported by cost-effectiveness data. That will push many developers further into the retail market, which will gain momentum from consumer-directed health plans and high-deductible health savings accounts. Not only is a small but growing percentage of consumers already paying out of pocket for genetic testing, but the popularity of HSAs appears to be increasing steadily.
“Take Alzheimer’s,” says Carlson, “if you knew that you had a 5 percent greater likelihood of getting it than not, that information has utility to you in terms of organizing your life. Plans don’t see data the same way the consumer does. So, I think we’ll see greater consumer involvement in genetic testing because plans will have a hard time figuring out what to do with it.
“A creative solution, frankly, is to dump predictive and informational genetic tests into the consumer-directed product line, where members can spend their tax-favored money while plans avoid liability.”
A LOSING PROPOSITION?
An important coverage-and-payment issue is the use of CPT codes for calculating genetic test reimbursement. Right now, payment for a lab test is the sum of the dollar values of the CPT codes for each procedure. Test developers and laboratories argue that this method is archaic, and in many cases neither covers actual costs nor reflects the value of a test.
Medicare, usually an industry bellwether for coverage and reimbursement issues, may not be an appropriate model for private insurers when it comes to genetic testing because of limitations that typically don’t apply to most private payers, such as Medicare’s lack of a benefit category for preventive care. Further, the Medicare Modernization Act froze lab fees — the schedule of which was last revised more than 20 years ago — until 2009.
“A creative solution is to dump predictive and informational genetic tests into the consumer-directed product line, where members can spend their tax-favored money while plans avoid liability,” says Rick Carlson, JD, of the University of Washington.
PHOTOGRAPH BY RICK DAHMS
Interest-group standoffs threaten to inhibit the development of this remarkable technology in its infancy; or perhaps they will simply divert the flow of new genetic testing technologies into a retail environment, as Carlson and others suggest. Many Americans, however, will not be able to afford retail genetic services. That’s why work of the Secretary’s Advisory Committee — which, among other things, assesses how genetic tests are integrated into health care and public health and serves as a public forum for discussion of emerging scientific, ethical, legal, and social issues raised by genetic tests — may be an important milestone in the future of genetic testing and services.
Whether or not the committee has an effect, the cost of genetic tests and therapies is likely to create an opportunity for government intervention to assure access for those without deep pockets. But Uncle Sam will be at least as tight with a buck as private payers, so we may be back to where we started.
“The science is very young,” says Ramsey. “When the genome was sequenced, this was going to be big stuff in terms of making people live longer and better. A lot of capital flowed to these companies, and now we’re seeing the early results of that research and that funding.
“Many such tests hold promise, but I don’t know that there’s enough capital for the very large and long-term studies needed to establish their value in clinical practice. At some point, the NIH is going to have to step in and fund a big study. Until then, you will have a lot of companies advocating products, a lot of insurers nervous about paying for them, and not a lot of data to help make the decision.”
Footnotes
Schrag D, Kuntz KM, Garber JE, Weeks JC. Life expectancy gains from cancer prevention strategies for women with breast cancer and BRCA1 or BRCA2 mutations. JAMA. 2000;283:617–624.
Imperiale TF, Ransohoff DF, Itzkowitz SH, et al., for the Colorectal Cancer Study Group. Fecal DNA versus fecal occult blood for colorectal-cancer screening in an average-risk population. N Engl J Med. 2004;351:2704–2714.
Aetna, mentioned at the outset of this article, was an early adopter of coverage for BRCA for members and others whose test results could benefit the member. According to Armstrong, this position represents a rational and cost-effective way to provide risk information to members.