REGULATIONS & STANDARDS
![]() |
Michele M. Schoonmaker, PhD, is director, government affairs at Cepheid (Sunnyvale, CA). Her responsibilities include developing reimbursement plans for molecular diagnostic products, and working with federal agencies and legislative offices on regulatory policies. She can be reached at michele.schoonmaker
@cepheid.com. |
Despite the number of molecular tests on the market, payers have developed few coverage policies for molecular diagnostics. Even where policies do exist, the payment rates are often suboptimal. Such suboptimal payment can be similar to a noncoverage decision in that laboratories cannot afford to use the test.
Reimbursement for molecular diagnostics depends on the following: a current procedural terminology (CPT) code to identify the test, evidence in the medical literature to support a particular technology platform or a biomarker’s clinical validity, the type of insurance that a patient carries, and establishing equitable payment rates to reflect added value. Measuring such added value is hindered by a lack of consensus regarding the appropriate use of cost data in setting payment rates. Even though several groups have concluded that molecular tests are undervalued and that the current payment systems need reform, their recommendations have been largely ignored.2
Coding for Molecular Tests
In most cases, before a payer can establish a reimbursement policy, a test must be identified by a specific CPT code.3 Although tests lacking CPT codes can still be reimbursed using unlisted codes, they have a greater chance of being denied payment or requiring additional documentation. To apply for a new CPT code, an IVD manufacturer must demonstrate that the test has regulatory approval (if required), is clinically significant, and is used in a wide geographic region. In addition, in most insurance policies, a CPT code needs to match a specific code from the International Classification of Diseases, ninth revision (ICD-9) in order for a claim to be paid.
While the CPT codes for molecular diagnostics are process oriented, the codes for the molecular identification of infectious organisms are specific for the particular organism. Nonetheless, many analysts believe that the current codes do not adequately describe most molecular testing, and that the process of obtaining new CPT codes is not fast or flexible enough for emerging technologies.
Coverage Policies
Reimbursement decision making involves determining two factors: is the test medically necessary (coverage), and if so, how much is it worth (payment). For most medical products, payers prefer evidence from large, randomized clinical trials to demonstrate medical necessity. Many payers will not cover a new experimental technology unless it is presented in the context of an approved clinical trial.
With the increasing reliance on positive health outcomes as a primary factor in deciding coverage, molecular testing is at a disadvantage, since the effect of molecular test results on a patient’s health may not be apparent for many years. Moreover, clinical trials measuring health outcomes are rarely conducted for new IVDs. Most premarket notification 510(k) applications include data from studies that are designed to demonstrate substantial equivalence to a predicate device and show comparable performance in clinical specimens. Such studies are not designed to compare test results to clinical outcomes, or establish the superiority of one test over another.
A molecular test can attain Medicare coverage if the Centers for Medicare & Medicaid Services (CMS) determine the following: the test is part of a covered benefit, is not specifically excluded, and is reasonable and necessary for the diagnosis of illness or injury. While Congress decides which benefit categories Medicare can pay for, CMS or its local contractors decide the coverage for specific services within the categories.4 Consequently, CMS has interpreted section 1862(a)(1)(A) of the Social Security Act as meaning that unless otherwise specified in the statute, most preventive services (e.g., genetic screening tests) are not covered. However, while CMS may make national coverage determinations (NCDs) for certain molecular technologies, coverage for most laboratory tests is determined locally.
A local coverage determination (LCD) is based on a review of current medical practices and clinical data. While some Medicare contractors will develop LCDs for many covered services, others will write an LCD only if the coverage should be limited in some way (e.g., frequency, patient population, scope of benefit). Such practices often result in geographic variations in coverage policies. Because of the variability in how and when local contractors may develop LCDs, a contractor may cover molecular diagnostics even if there is no formal policy.
Private insurance companies are reviewing a small number of genetic and molecular diagnostic tests for coverage. Large insurers, such as BlueCross BlueShield (Chicago), Kaiser Permanente (Oakland, CA), and Aetna U.S. Healthcare (Hartford, CT), have well-defined formal processes to evaluate new technologies.5,6 Even though the final decision-making process is not always open, coverage benefits are determined locally and may be negotiated by specific groups or employers with the local plan.
Some insurers are becoming more open to meeting with IVD manufacturers to discuss their data needs for determining coverage for new technologies. Manufacturers should assess the regulatory and reimbursement landscape to identify data gaps in the literature that they will use to support the clinical utility of their product. For example, if Medicare is going to be a significant payer, are there enough persons over the age of 65 represented in the clinical study? If similar devices using different technological methods are on the market, IVD manufacturers should also consider conducting a comparative benefit study. Such a study may demonstrate added benefits over the predicate devices or current practices to support coverage for the new test.
If reimbursement is expected to be a hurdle, IVD manufacturers should collect any information during a clinical study that they can use to support the payers’ requirements. Some insurers may even cover some of the costs associated with a clinical study in return for information that would help them make coverage decisions. If it is not feasible to collect the data desired by insurers in the premarket study, manufacturers can explore postmarket options. A new guidance document, “Coverage with Evidence Development,” outlines circumstances under which Medicare may cover a new product while additional evidence is gathered through postmarketing studies.7
Payment Rates
Once payers decide to cover a new test, they assign a payment rate. Medicare reimburses laboratory services provided by independent labs according to a national fee schedule. Based on laboratory charges from 1983, this fee schedule was established under the Deficit Reduction Act of 1984, and has rarely been updated. Critics have charged that the fee schedule has not kept pace with inflation. They also charge that no practical method exists for making adjustments to the fee schedule, particularly for molecular tests which were not widely available in 1983. Currently, payments for existing laboratory tests are frozen until 2009. While the inherent reasonableness provision of the Balanced Budget Act of 1997 gives CMS the authority to adjust payments that are too high or too low, examples of its use and guidelines on how such provisions are applied are not available.
Table I. (click to enlarge) Total number of claims for molecular diagnostic laboratory tests
submitted to all Medicare carriers, 2000–2004. |
Table II. (click to enlarge) Charges submitted to and paid by Medicare for CPT codes in molecular diagnostics and infectious disease, 2000–2004. The amount paid equals the number of tests paid by each contractor in each year (2000–2004) multiplied by the contractor’s clinical laboratory fee schedule amount for that year.
|
Table II shows the amounts that Medicare was billed and paid for the same CPT codes during the same five-year period. While Medicare reimbursed 46% of the charges submitted for molecular infectious-disease testing, only 9% of the charges for nucleic acid extraction were paid. Such low reimbursement is alarming since the rates in state Medicaid programs are bound by each state’s Medicare limit. Many Medicaid programs reimburse at rates even lower than the state’s Medicare allowance.
Several organizations, such as the Secretary’s Advisory Committee on Genetics, Health, and Society (Bethesda, MD), the Lewin Group (Falls Church, VA), and the Institute of Medicine (Washington, DC), have questioned the adequacy and accuracy of Medicare’s national fee schedule for lab services, and have argued that molecular testing is undervalued. They also believe that reimbursement for molecular tests needs to be adjusted to reflect their added value. Such adjustments are necessary to continue developing innovative IVD products and allow patients to reap the benefits from technologies developed as a result of the Human Genome Project.
Private insurance companies use various methods for setting their reimbursement rates. Some insurers negotiate rates with either employers or participating laboratories, which are then contractual. Others establish their rates as a percent of billed charges (e.g., cover 85% of laboratory charges) or a fee schedule amount which may be based on Medicare’s fees.
Unlike Medicare that publishes its fee schedule, IVD manufacturers are often unable to collect data on private payers’ reimbursement rates. Even with willing laboratory customers, it can be difficult to obtain reimbursement data, as many lab providers are prohibited from sharing such data as a condition of their contracts.
In 1989, Medicare proposed using cost-effectiveness in its coverage decision-making process.8 However, CMS could not reach a consensus on many variables, and the proposal was never finalized. Cost has been implicated again in more-recent attempts by CMS to establish formal criteria for reimbursement decision making.9 It is also difficult to assess the use of formal cost-effectiveness data by private payers, since many of their decision-making processes are not made available to the public.
Despite payer interest, cost-effectiveness methodologies are still not standardized. For IVD tests, there is the added difficulty of assessing benefits that accrue indirectly from a test, rather than directly from a therapeutic intervention. Though cost studies may be too expensive to conduct for every test, they may be useful to overcome difficult reimbursement hurdles, such as for new technology platforms or CPT codes. However, there are no established guidelines to determine how much data are necessary, how they should be appropriately shared, and how they would be used for reimbursement.
Improving Reimbursement for Molecular Tests
Despite the many advantages to labs and clinicians, molecular testing has been slow to replace older diagnostic technologies simply because it costs more. Although they are more expensive and not required by FDA, IVD manufacturers could consider conducting comparative benefits studies that incorporate cost analyses.
In addition, IVD manufacturers can do several things to improve their chances of obtaining reimbursement for their molecular tests. Especially for tests that use new technology platforms or detect new genetic markers, manufacturers should meet early with payers to evaluate data needs and plan for the collection of that data during clinical studies. During premarket studies, IVD manufacturers can collect cost data and negotiate for their proper consideration in the reimbursement decision-making process. When such data collection is not feasible during the premarket phase, manufacturers can develop alternative plans during postmarket studies.
Following FDA clearance or approval, IVD manufacturers can submit an application for a new CPT code. Once the molecular test is on the market, manufacturers should monitor coverage and payment experiences, and follow up immediately with payers when noncoverage or low payment is an issue. Finally, even though molecular diagnostics is the fastest-growing segment of the IVD industry, it is still relatively small. The government is facing increasing pressure to develop better regulatory and reimbursement policies for molecular testing. IVD manufacturers need to be actively involved in public debates to ensure that reasonable and equitable policies are developed, thereby enabling molecular diagnostics to reach its full potential in benefiting patient care.
References
1. The Lewin Group, “The Value of Diagnostics: Innovation, Adoption and Diffusion into Health Care,” AdvaMed Web site (Washington, DC: AdvaMed, 2005 [cited 29 November 2006]); available from Internet: www.advamed.org/publicdocs/thevalueofdiagnostics.pdf.
2. The Secretary’s Advisory Committee on Genetics, Health, and Society, “Coverage and Reimbursement of Genetic Tests and Services,” Office of Biotechnology Activities Web site (Bethesda, MD: NIH, 2006 [cited 13 July 2006]); available from Internet: www4.od.nih.gov/oba/sacghs/reports/CR_report.pdf.
3. “CPT (Current Procedural Terminology),” American Medical Association (AMA) Web site (Chicago: AMA, 2006 [cited 13 July 2006]); available from Internet: www.ama-assn.org/ama/pub/category/3113.html.
4. “Exclusions from Coverage and Medicare as a Secondary Payer,” U.S. Social Security Administration (SSA) Web site (Baltimore: SSA, 2006 [cited 4 May 2006]); available from Internet: www.ssa.gov/OP_Home/ssact/title18/1862.htm.
5. “Technology Evaluation Center Criteria,” BlueCross BlueShield Association Web site (Chicago: BlueCross BlueShield Association, 2006 [cited 3 May 2006]); available from Internet: www.bluecares.com/tec/teccriteria.html.
6. “Clinical Policy Bulletins,” Aetna Web site (Hartford, CT: Aetna, 2006 [cited 8 May 2006]); available from Internet: www.aetna.com/cpb/cpb_menu.html.
7. “National Coverage Determinations with Data Collection as a Condition of Coverage: Coverage with Evidence Development,” CMS, Centers for Medicare & Medicaid Services Web Site (Baltimore: CMS, 2006 [cited 13 July 2006]); available from Internet: www.cms.hhs.gov/mcd/ncpc_view_document.asp?id=8.
8. Federal Register, 54 FR:4302, January 30, 1989.
9. Federal Register, 65 FR:31124, May 16, 2000.




