Medical Device Link.

Originally Published MX September/October 2002

GOVERNMENTAL & LEGAL AFFAIRS

The FDA Approval and Medicare Coverage Processes

Part I: Current Interconnections

Bradley Merrill Thompson

For U.S. medical technology manufacturers, negotiating current government processes remains a duplicative and delay-prone exercise.

These days, when medical device executives and investors look at the business plan for a new medical technology, one of the first things they examine is its so-called policy risk—meaning the full range of governmental policies that could affect the viability of the technology. Two of the key elements of such policy risk are the uncertainties related to securing FDA approval and obtaining Medicare coverage. As executives and investors study these elements, they often wonder whether there is some connection between the two. If the technology is good enough to attain FDA approval, will it be good enough to deserve Medicare coverage?

Together, the administrative processes for FDA approval and Medicare coverage exert significant influence over the speed with which a product moves from its development phase to regulatory approval and reimbursement. Both processes include requirements that manufacturers must take into account during the development phase, and both can potentially drive up the cost of development. Indeed, the degree of certainty and predictability with which a product is likely to negotiate the FDA approval and Medicare coverage processes represents a major criterion that device executives and investors apply when evaluating potential projects.¹

The relationship between these two processes is a timely topic, as efforts to improve the current system are now under way on several fronts. In Congress, work is being directed toward enacting legislation that would ensure greater coordination between FDA and the Centers for Medicare and Medicaid Services (CMS).² In turn, such activity on Capitol Hill has stimulated interest by the Department of Health and Human Services (HHS), where the secretary's advisory committee on regulatory reform is examining this issue. At a June meeting of the committee attended by HHS secretary Tommy Thompson and representatives of FDA and CMS, the committee examined and debated the linkages between FDA approval and Medicare coverage. It is likely that the advisory committee will adopt recommendations on the subject this fall. Moreover, the agencies themselves are also discussing administrative changes that could result in improved linkages between their efforts. With so much activity being devoted to the relationships between FDA and CMS, it seems inevitable that there will be some significant changes over the coming months.

Coordination v. Harmonization. This article also examines the relationships between the FDA approval and the Medicare coverage processes, but from a different viewpoint. Rather than examining methods for better coordination of FDA and CMS activities, this article will specifically address the possibilities for harmonization of the FDA and CMS processes. The difference is important. Coordination implies two agencies working together toward a common purpose. Harmonization, in this context, suggests two agencies with separate missions whose decisions and actions are nevertheless consistent with one another.

As a goal for the processes used by FDA and CMS, the notion of coordination has certain pitfalls. In the view of many device manufacturers, coordination between the two agencies could easily turn into competition, with each agency's regulators trying to identify more deficiencies than the other. To the extent that such regulatory one-upmanship might delay governmental decision making—or even elevate requirements so that beneficial technologies would not be approved at all—it would be to the detriment of patients needing access to new technologies.

By contrast, using a harmonized approach would virtually eliminate the possibility of FDA and CMS regulators competing with one another. Moreover, harmonization offers opportunities for reducing duplication of efforts between the two agencies, achieving greater consistency in agency decisions, and minimizing the time required for decision making.

This two-part article explores these and other issues related to harmonization of the FDA approval and CMS coverage processes. This installment discusses the current interconnections between the two processes, and explains why many device manufacturers favor harmonization as an approach to improving the present system. The second installment will explore a proposed solution along those lines, and will discuss the consequences of the proposal. Also included will be some recommended strategies that manufacturers can use to manage their current relationships with the two agencies to produce harmony.

Current Processes

The FDA approval and Medicare coverage processes have each been detailed elsewhere, and this article does not attempt to repeat those explanations.^3–5 Nevertheless, it is important to define precisely the processes that are analyzed here.

FDA. There are several ways to obtain marketing authority for a medical device. The most widely used method is the premarket notification (510(k)) route, which requires the manufacturer to demonstrate that its medical device is substantially equivalent to those already lawfully on the market. Among other things, the definition of substantial equivalence includes a demonstration that the new device is as safe and effective as devices already on the market.⁶

In the case of novel devices for which there are no prior examples, the manufacturer must pursue a premarket approval (PMA). Obtaining a PMA requires proving that the device is safe and effective in a more abstract sense, as opposed to comparing it with other products that have already been marketed.

FDA also offers a number of less widely used routes to market, including those that involve such submissions as a product development protocol or a petition for reclassification. There are also regulatory categories for products that are under investigation, including some special categories for in vitro diagnostic devices. However, this article focuses principally on the 510(k) and PMA routes.

FDA is also responsible for protecting the consumer from economic fraud. Such fraud can result from, among other things, misrepresenting what a device will do—even if such misrepresentation does not mean the device is unsafe or ineffective.

CMS. With regard to Medicare coverage, CMS offers at least three processes leading to a coverage determination. The best-known process—but also the least used—is the national coverage determination (NCD) process. CMS reserves its use of this process for a very limited number of technologies that raise novel and difficult issues that merit a uniform national policy. The process is initiated when a stakeholder requests an NCD—usually after FDA has approved the device. CMS then determines whether the request includes enough information to be acted upon and, if it does, conducts internal deliberations to evaluate whether a positive NCD is warranted.⁷ During the process, CMS may seek a technology assessment from any one of a variety of outside bodies, or may seek the advice of the Medicare Coverage Advisory Committee (MCAC), which is composed of experts in technology assessment.

Separate from the NCD process is the local coverage process, through which local Medicare contractors such as the Part B carriers or the Part A fiscal intermediaries evaluate new technologies. These local Medicare contractors are typically insurance companies performing the service of administering Medicare for a local area. Local contractors are supposed to accept suggestions for local coverage decisions (LCDs) and evaluate those requests through a deliberative process that may involve consultation with local stakeholders.⁸

Someplace in between the NCDs and the LCDs are regional coverage determinations. While a local contractor may have responsibility for a specific state or other local area, regional contractors are responsible for much larger portions of the country. In the case of durable medical equipment, for example, there are four regional carriers that are each responsible for approximately one-fourth of the country. Unlike local Medicare contractors, whose determinations may reflect a wide variety of local coverage policies, regional contractors are supposed to act in concert with one another. As a result, the regional coverage process produces what is in effect a uniform national decision—but without using the process required by law for national coverage policymaking.

Interconnections between FDA Approval and Medicare Coverage

Area of Comparison	Similarities
Purpose	Evaluation of new medical devices to determine if they work.
Criteria	Clinical focus. Examine similar body of scientific evidence. Assess safety and effectiveness, including evidence of clinical utility in certain cases.
Agency	Governmental processes. Subject to certain statutory requirements, including appeals. Use advisory committees for input in certain cases.
Character of decision	Not simply yes or no decisions. Can involve conditions and limitations. Affect patient access to medical technologies.
Table I. Basic similarities between the FDA approval and Medicare coverage processes.

There is no direct connection between the two statutes that underlie the FDA approval and Medicare coverage processes. Consequently, the two processes are typically treated as independent and sequential—not least by the agencies that administer them. Nevertheless, there are some basic similarities between the processes and criteria that these two governmental agencies utilize (see Table I).

The Processes. In formal terms, the process is simply the mechanical flow of a technology assessment as it proceeds through the agency on its way to a decision. From this point of view, the FDA approval and Medicare coverage processes have a number of elements in common.

However, a side-by-side comparison of such similarities obscures the important fact that the two processes are often thought of as sequential. This attitude arose, in part, because CMS has historically preferred to wait until a product has received FDA approval before spending any appreciable amount of time evaluating the technology.⁷ CMS has thus avoided examining technologies that did not ultimately earn FDA approval, and has also been able to accommodate instances in which FDA approval modified the terms of the original product approval submission. By adopting such a sequential approach, CMS also sought to conserve scarce resources available for technology reviews.

In recent years, however, CMS's attitude toward early dialogue with technology developers has begun to change. By opening its doors to technology discussions in advance of FDA approval, CMS hopes to improve its reviewers' readiness to assess novel technologies. It also hopes to encourage manufacturers to develop and submit evidence with the characteristics most likely to be useful to CMS reviewers.⁵

While still sequential for the most part, the two processes can and sometimes do overlap. Moreover, because of some recent public disconnects between the two agencies, FDA and CMS are trying to communicate more often regarding pending coverage decisions.⁵

The Criteria. The factors that FDA and CMS take into account when making their decisions constitute the criteria for FDA approval and Medicare coverage. When the processes of the two agencies are compared, it is usually these decision criteria that get the most attention. These criteria are set by statute, and are expressed in different terms (see Table II).

Area of Comparison	FDA Clearance or Approval	Medicare Coverage
Statutory purposes	Public health protection. Protection against fraud.	Insurance.
Criteria (statutory)	Is device safe and effective? Is device substantially equivalent to a predicate device?	Does device fit a benefit category? Is use of the device reasonable and necessary?
Criteria (practical)	Is device safe and effective? Does device have clinical utility? Is device labeling truthful?	Is device safe and effective? Does device have clinical utility? Is it worth paying for?
Scope	Single product from one manufacturer. Products, including devices, drugs, and biologics.	A product class. Items and services, including procedures.
Population protected	All United States citizens.	Medicare beneficiaries.
Agency culture	Law enforcer.	Insurer. Law enforcer against fraud.
Process structures	Initiated by manufacturer of the product. Confidential until approval or disclosure.	Initiated by interested person (e.g., beneficiary). Trade secrets can be protected. Process is not confidential.
Impact of positive decision	Manufacturer can lawfully sell the product in the United States.	End-user can seek payment of reimbursement from Medicare.
Table II. Basic differences between the FDA product approval process and the Medicare coverage process.

On the FDA side, the primary criterion for novel products is to be safe and effective. The primary criterion for not-so-novel products is to be as safe and as effective as predicate devices lawfully on the market. To learn how FDA interprets the meaning of safe and effective, a manufacturer would first turn to the definitions included in the relevant FDA regulations, and then to the myriad FDA guidance documents where those definitions are applied to very specific technologies.⁹

FDA guidance documents also discuss the concept of clinical utility—whether a device actually does anything clinically useful.^10–11 This concept is most often raised in the context of novel devices that require a PMA, or as a way of ensuring that the measurements taken by a diagnostic device offer a clinically meaningful result.

On the CMS side, the first criterion applied by the agency is whether an item or service falls into a covered benefit category. If it does, the next criterion is whether the item or service is reasonable and necessary—a consideration that CMS argues is quite different from that used by FDA. To figure out exactly how CMS interprets the phrase reasonable and necessary, however, is a challenge. CMS has spent more than 20 years wrestling with how to define this key criterion in a regulation, but so far has not been able to do so.¹²

In the interim—skipping over the step of defining the terms of its key criterion—CMS has begun to release memorandums that explain how the agency reached its conclusions about specific technologies. Theoretically, one could read all of these decision memos and extract from them the general principles that the agency follows. Those who have attempted this process, however, have met with little success. Because each CMS decision seems to have characteristics that make it unique, it is difficult to identify general guiding principles even from the entire body of decision memoranda.

A review of CMS decisions to date suggests, among other things, that CMS is more impressed by technologies that earn PMAs than by those cleared through the 510(k) process. In addition, CMS applies special rules to devices that FDA considers investigational. For instance, IVD devices that are classified as investigational research only (IUO) or research use only (RUO) products—and are therefore not approved by FDA—can encounter difficulty in obtaining a favorable coverage determination from CMS.

Although many outsiders assume that economics play a role in CMS's definition of its reasonable and necessary criterion, CMS says that, at least currently, this is not the case. CMS has on occasion proposed the addition of economic considerations to its coverage determination process, but the resulting controversy has each time forced the agency to retract the proposal. Instead, CMS can bring certain economic considerations into play when the agency determines how much it will pay for a new medical technology.

Comparing Agency Criteria

Considering the differences between the statutory criteria for the FDA approval and Medicare coverage processes, it can seem difficult to compare the ways that the two agencies apply those criteria to the evaluation of particular technologies. The task can be made easier, however, by stepping back from comparing the way the processes work in specific instances. A clearer picture can be obtained by starting with an understanding of the purposes that the statutory criteria are meant to serve.

Legislative History. According to the legislative history of the Federal Food, Drug, and Cosmetic Act (FD&C Act), the criterion of safety and effectiveness was put in place in order to protect the public health of all Americans. By contrast, the legislative history of the Social Security Act reveals that the criterion of reasonable and necessary was meant to ensure that the Medicare program would only reimburse for those items and services that it should pay for. Those purposes are very different, and as a result one can readily understand how the clinical end points of interest to FDA and CMS would also differ.

Even so, the FDA and CMS criteria still have many points in common. For instance, CMS often states that FDA approval is a prerequisite to Medicare coverage of a product (except for some investigational devices). In other words, for both FDA approval and Medicare coverage, a device must be safe and effective.

By requiring FDA approval, however, CMS does not mean that FDA must have approved the device for the specific use in question. CMS has been known to grant Medicare coverage for off-label uses of a technology that is lawfully on the market.¹³ CMS uses this approach to emphasize that its interests go beyond the safety and effectiveness that FDA assesses.

Ultimately, looking only at the legislative history of the regulations that give rise to the FDA approval and CMS coverage processes has its own set of limitations. The resulting analysis is inevitably very superficial and potentially misses a lot of practical overlap between the two processes. A more useful analysis might be developed by comparing specifics of the FDA and CMS processes, but the lack of adequate definitions for key elements of the Medicare coverage process inhibits such efforts.

Agency Practice. The next-best approach is to review how the decisions are made in practice. A review of published coverage decision memos suggests that CMS uses the Medicare coverage process—and in particular the test of what is reasonable and necessary—for three purposes.

First, the coverage process seeks to determine whether the item or service at issue falls within a covered benefit category.

Second, the coverage process acts as a brake to prevent overutilization. In this sense, the intent of the process is to prevent the Medicare program from covering—and paying for—too much of a good thing. While this purpose is economically motivated, the decisions developed through the coverage process are not based on economic factors. Instead, they are based on an assessment of clinical data to determine the level of utilization appropriate for producing a satisfactory clinical result.

Third, CMS uses the coverage process to assess the safety and effectiveness of new technologies in order to protect the health of beneficiaries. In spite of the fact that the products in question have already received FDA approval or clearance, CMS often examines devices with these issues in mind. The agency's decision memorandums frequently cite concerns over the safety or effectiveness of a product as a basis for noncoverage.¹⁴ In some of the published decisions, it is clear that CMS simply disagreed with FDA's determinations about the basic safety and efficacy of devices cleared through the 510(k) process.

Why does this happen? It may happen simply because the coverage staff at CMS is populated by physicians and others who care deeply about people. They spend much of their time studying clinical evidence about new medical technologies, and in the course of that study they develop strong personal opinions. It is natural and laudable that they should then follow their consciences to protect the health of Medicare beneficiaries.

CMS reviewers have almost unlimited discretionary power to effect their personal views on new technologies as part of the Medicare coverage process. Permitting individual reviewers such free rein can cause significant problems for novel technologies on which opinions are mixed. Even when FDA has previously approved a device, for instance, nothing in the CMS regulations requires its reviewers to abide by that assessment. When even well-meaning scientists at FDA and CMS disagree about the value of a new technology, what should happen to that technology? The policy implications of this dilemma will be discussed in the second installment of this article.

While it is true that the criteria applied by FDA and CMS differ in many respects, they are also connected in many ways. Conceptually, FDA's criterion of safety and effectiveness is a subset of CMS's criterion of reasonable and necessary. And in practical terms, the two sets of criteria overlap by examining many of the same issues. As a result, if policymakers want to streamline the time it takes for patients to obtain access to new medical technology, the two assessment processes must be viewed together and the overall time compressed as much as possible.

The Impact of the Overlap

There are at least three problems with permitting the current overlap in governmental technology assessments to continue.

Delays. Most important, the overlap causes unnecessary delays in providing patient access to important new therapies. The fact that the Medicare coverage process frequently revisits basic public health issues is partly responsible for such delayed coverage decisions, which in turn delay access to high-quality care.

A study by AdvaMed (Washington, DC), an industry trade association, found that the average length of the review process for 26 national coverage decisions issued from the beginning of 1999 through the first half of 2001 was 175 days—even though none of the decisions involved an external technology assessment or referral to an expert advisory panel. This average exceeds the 90-day time frame within which Medicare has said it would generally complete decisions of this type. For decisions that did involve external assessments or expert-panel reviews, the average time was 306 days.

Moreover, none of these calculations included time for the prerequisite FDA review of these same technologies or the subsequent time required to implement the Medicare decisions. Although CMS data about national coverage determinations issued in 2001 indicate that there has been some reduction in the average time for review, there is no doubt that the coverage process adds considerable time to the overall goal of patient access.¹⁵

Wasted Resources. Having two federal agencies examining the safety and effectiveness of devices leads to wasted governmental resources. In almost every coverage decision, CMS spends a considerable amount of time reviewing much the same scientific material already studied by FDA, to learn how the technology works and to reach conclusions regarding the effectiveness of the technology. In these days of substantial belt-tightening, when government resources to review breakthrough medical technologies are already stretched thin, such duplication of effort is hard to accept.

Inconsistencies. Having two separate agencies assess the intrinsic safety and effectiveness of new medical technologies produces inconsistencies, which one government should not permit. It's not hard to understand why two sets of scientists might view complex medical evidence differently. But ultimately, all of the stakeholders in the healthcare system would benefit from having a federal government that speaks with a single voice on such issues.

The delays inherent in the current system, resulting both from overlapping processes and from the inconsistency of decisions rendered by those processes, remain a major concern. Such bureaucratic obstructions ultimately lead to unnecessary and unacceptable delays in providing patients access to important new therapies.

Conclusion

It has been clear for many years that FDA approval is necessary but not sufficient to earn Medicare coverage. As a payer, Medicare looks beyond FDA's determination of safety and effectiveness. How far beyond depends in part on the regulatory pathway that a new medical device took through FDA.

Recognizing that there are differences in the FDA approval and Medicare coverage processes, there are also great similarities. The two processes overlap tremendously, and that overlap can lead to needless delays, wasted resources, and inconsistent decisions. It is hoped that policy makers will explore ways to harmonize the two processes to avoid the consequences of that overlap.

The next installment of this article (November/December 2002) will explore some of the policy options for achieving harmonization. Until the two processes are harmonized, it is up to manufacturers to ensure the appropriate coordination by managing communications with the two agencies. That topic, too, will be explored in greater detail in the next installment.

---

References

1. Venture Capital and Medical Technology: The Impacts of Public Policy on Capital Flows and Innovation (McLean, VA: KPMG Consulting, 2001).

2. Medicare Modernization and Prescription Drug Act of 2002, HR 4954, sec. 842 (Engrossed as passed by House; went to the Senate floor July 15, 2002).

3. BM Thompson, FDA's Regulation of Medical Devices (Englewood, CO: Interpharm Press, 1995).

4. BM Thompson, Guide to Medicare Coverage Decision-Making and Appeals (Washington, DC: ABA Publishing, 2002).

5. J Whyte, "Covering New Ground," MX (November/December 2001): 68–71.

6. 21 USC 360c(i).

7. Health Care Financing Administration, "Medicare Program; Procedures for Making National Coverage Decisions," Federal Register, 64 FR:22619–22625 (April 27, 1999).

8. Medicare Program Integrity Manual, chap. 13 (Bethesda, MD: Centers for Medicare and Medicaid Services, 2002).

9. Code of Federal Regulations, 21 CFR 860.7.

10. Guidance on Clinical Utility and Premarket Approval, PMA Memorandum P91-1 (Rockville, MD: Office of Device Evaluation, Center for Devices and Radiological Health, FDA, 1991).

11. Guidance for the Submission of a Premarket Notification for Magnetic Resonance Devices (Rockville, MD: Office of Device Evaluation, Center for Devices and Radiological Health, FDA, 1998).

12. Health Care Financing Administration, "Medicare Program; Criteria for Making Coverage Decisions," Notice of intent to publish a proposed rule, Federal Register, 65 FR:31124–31129 (May 16, 2000).

13. "Coverage Policy Decision Memorandum on Electrostimulation for Wounds," CAG-00068N (Bethesda, MD: Centers for Medicare and Medicaid Services, July 23, 2002).

14. "Coverage Policy Decision Memorandum on Electrodiagnostic Sensory Nerve Conduction Threshold," CAG-000106N (Bethesda, MD: Centers for Medicare and Medicaid Services, February 14, 2002).

15. Tommy G. Thompson, "Report to Congress on National Coverage Determinations," (Washington, DC: Department of Health and Human Services, 2002).

Bradley Merrill Thompson, Esq., is a partner and chair of the drug and medical device law practice at the law firm of Baker & Daniels (Indianapolis). He is also chairman of the board of Aventor, a global, multidisciplinary consulting firm that helps medical technology companies and investors navigate regulatory and reimbursement requirements worldwide.

Copyright ©2002 MX