Necessary but Not Sufficient: Amending the Medical Device Law

Originally Published MDDI July 2002

DEVICE REGULATION

Kshitij Mohan

There is an old saying about laws and sausages: If you like them, you should not watch them being made. One may or may not agree with this dubious wisdom, but one thing is clear—we expect our law-producing factories to roll them out at rates similar to those of sausage makers.

This need to constantly create new laws sometimes generates laws that are of little use, and others that end up having unintended consequences (consider the laws to simplify the tax code). The medical device industry has had its share of both types.

SMDA. The Safe Medical Devices Act of 1990 (SMDA), encouraged by the first Bush administration, was originally intended to further reduce industry's regulatory burden. Prior to this act, administrative processes had helped eliminate backlogs and applications were being reviewed within statutory time frames. But, worried that these remedies might be misinterpreted as going beyond the legislative mandate, the administration supported SMDA.

Some strange slips between the pen and print resulted in several more stringent requirements for the industry. Mandatory adverse reporting was extended from manufacturers to hospitals and distributors, and the 510(k) process morphed from a notification scheme to an approval process "by order." Requirements were put in place for tracking certain devices throughout their useful life.

SMDA included the codification of the so-called "Mohan Memo"— a 1986 guidance memo that defined the decision criteria for determining substantial equivalence for devices cleared through the 510(k) process— even though the scheme in that memo did not need to be codified; it had worked fine for almost four years without legal challenges or controversy. The net effect of SMDA was that it either contributed to or was unable to prevent the astronomical increases in both the review times and device application backlogs that occurred shortly after its adoption.

SMDA is not solely to blame for the dark ages of medical device regulation in the early 1990s. Also responsible were a new administration and a generic-drug scandal that sent several industry and FDA officials to jail.

The generic-drug approval process was, by congressional design, an abbreviated process that allowed some flexible decision making, not unlike the 510(k) process for devices. There was an unspoken suspicion that if one was susceptible to corruption, the other might be as well. But various audits and investigations of the device approval processes failed to reveal any irregularities. Nevertheless, that era at FDA was defined by a strong suspicion of all industry. Reviewers were forbidden to talk with industry for fear that they might be compromised. It appeared to be the end of the open dialogue that had existed between industry and FDA.

FDAMA. Another attempt at "fixing" medical device law came with the passage of the FDA Modernization Act (FDAMA) in 1997. While FDAMA may have been of limited effectiveness, it was not harmful and was at least a step in the right direction. The worst it could do was have industry and FDA wonder what Congress meant by "least burdensome," or to ask why one needed to pass a law for FDA to have meetings with industry.

FDAMA did endorse the trend in FDA to exempt several low-risk Class I and II devices from 510(k) requirements; overall, however, it was more of a congressional call for FDA to go forth and do good. To some extent FDAMA was a Washington win-win. It sounded good ("modernization" and "least burdensome" sound great) and was noncontroversial—consensus defined as the lowest common denominator. There was only one problem: as a sausage, it needed more meat.

H.R. 3580. H.R. 3580, the new proposals introduced last December, offers another opportunity at revising the law to fix some of the gaps left over from the previous legislation. It is time we get the product right so that patients and all the members of the medical technology community can benefit. As William Graham, legendary Baxter CEO, used to say, "We are lucky to be in an industry in which you can do well by doing good."

There are many omissions in the current proposals (see sidebar). They need to be reformulated to accomplish what they intend. The spirit of harmony and consensus that the bill proponents seek may itself be a problem. It seems designed to have a relatively easy passage. But legislation intended to drive meaningful change need not have the smoothest ride; it may require strong leadership, however, to convince enough legislators to vote for it.

There are a number of issues this bill does not address. This is due to a variety of reasons: some issues are not addressed because the different trade groups involved in drafting the legislation don't have consensus; others because it would broaden the debate to areas that are too controversial, such as relief from product liability litigation.

Unfortunately, the issues that are probably the most important for the community of patients, FDA, and industry fall into this category of "hot potatoes." If this community became inclined to meet these issues head on, however, it would propose legislation that would address the problem areas.

CDRH RESOURCES AND USER FEES

The President's FY 2003 budget proposals include a 6% increase for CDRH, which is entirely offset by federal worker pay raises already approved by Congress. This is in contrast to increases of 25% for CDER and 19% for CBER. The low priority administrations have put on CDRH is evident from the fact that between 1995 and 2002, CDRH has experienced a budgetary erosion of 8%—while CDER and CBER have had small growth of 22% and 8%, respectively. Only the drug center has been able to keep pace with inflation over that time period. If the budget for devices had merely kept pace with inflation, CDRH would have over $35 million more in FY 2002.

Over the last several years, there have also been attempts by administrations to impose user fees, which have been successfully fought off by industry, particularly the smaller firms and their advocates.

In any case, this is an unacceptable situation that must be addressed. I submit the following rationale:

It is imperative that the medical technology industry sit down with FDA, congressional, and OMB staff to develop a proposal for user fees that would not only be acceptable, but attractive to the smaller firms that feel the most threatened. Fortunately, AdvaMed now supports user fees, though it had previously been opposed to or ambivalent about them. There is still apprehension among some small companies, however. Given the importance of small companies in the innovation of new devices, a construct acceptable to all is imperative.

ENHANCING QUALITY

FDA's belief in quality and quality systems, beneficial though it may be for the regulated industry, needs to manifest itself within the agency as well. In the area of quality, FDA must provide leadership by example, not merely pass judgment. New legislation should authorize and enable the agency to establish an FDA quality leadership program for its own organizations. As with any good quality program, FDA's quality leadership program should include the following:

CONSTRUCTIVE ENFORCEMENT

Depending on the leadership of the agency or of a subunit—or even on the bent of a reviewer or investigator—FDA swings back and forth between its two identities: one, a public health agency that uses laws and enforcement as a tool; the other, primarily a law enforcement agency with responsibility for laws and regulations in the public health area. FDA must assert its role primarily as a public health agency; enforcement is an important and powerful tool for the agency, but it needs to be used judiciously and thoughtfully.

FDA has a variety of enforcement tools—from warning letters, recalls, and seizures to civil penalties and criminal prosecutions. The effectiveness of each varies, but any can be severe on the economic status of a company. As a public health agency, FDA should be only as severe as is needed so that the impact of its action on a company's behavior remains high, while the destructive impact on the company's economic or organizational health is low.

As one-time speaker of the House Sam Rayburn once said, "any[one] can burn down a barn; it takes a skilled carpenter to build one." As a public health agency, FDA must preserve the health and integrity of an industry that can provide safe and effective products to patients in a timely manner.

In addition to regulations and enforcement, FDA also has other tools available that it can use more aggressively—training and education programs, communication and information dissemination, and research and technology programs, to name a few.

Civil penalties and criminal prosecutions of individuals may be appropriate in individual cases of egregiously violative or criminal behavior. The use of any of these penalties, however—especially when perceived as an effort to make an example of one company—often has unintended adverse consequences.

When a company is distracted, damaged, or destroyed, in addition to the negative impact on its employees and investors, there is also a negative impact on its customers—in this case, the patients. Many times this negative impact cannot be totally avoided, but in almost every instance it can be minimized.

Often, the negative impact forces the firms and venture capitalists to become more risk averse; this results in a reluctance to invest in innovation, and can slow down technology development.

But the unnecessarily destructive impact of enforcement actions can be minimized by using the least-destructive tool in a particular situation. While making an example may get everyone's attention, it also deters the growth of a vibrant, innovative industry that treats as a business imperative its compliance with value-enhancing, rational regulations.

In any overhaul of the FDA law, Congress should articulate guidelines along the lines of the federal sentencing guidelines. These requirements should define the elements of a strong environment of constructive enforcement within the agency. They should include the following:

PREEMPTION OF STATE LAW

In the Medical Device Amendments of 1976, Congress provided the medical device industry with special protection from multiple and diverse requirements in different states by making the amendments preemptive to any state law.

In addition to this preemption and to underscore it, Congress also provided a mechanism for exemption: it allowed state or local governments to apply for a preemption exemption from any specific state law, under certain conditions.

The courts have been contradictory on the issue of preemption, however. In the 1996 case of Medtronic v. Lohr, the Supreme Court ruled that the agency requirements under the 510(k) provisions were not specific enough concerning an individual product, and thus would not trigger a preemption of state laws. In addition, the court stated that there was no congressional intent specific to the preemption of state torts claims as evidenced in the legislative history of the Medical Device Amendments.

In subsequent cases related to the PMA application process, lower courts have held contradictory views. In general, courts are reluctant to use the constitutional privilege of the supremacy of federal law to tread on state laws unless there is clarity of congressional intent, specificity with respect to a particular product, or clear contradiction or obstruction of a federal law (or regulation) by a state law. Also coming into play here is the so-called Chevron deference. It says that if there is a lack of clarity in congressional intent, the courts give deference to an agency's own interpretation of its statute—particularly in the case of regulatory agencies that deal with complex technical products and activities.

Curiously, FDA has worked hard to limit the preemptive scope of its own statute. Even though it was not a party in specific cases, FDA has filed briefs in almost all the cases on this issue and finally developed regulations to define that limited scope. Whether this enthusiasm for state rights was due to constitutional concerns or to an ideological sympathy to the plaintiff's bar is unclear.

It is time for Congress to clarify its intent on two counts. First, it must state whether it regards the FDA requirements for PMAs, 510(k)s, quality systems, etc., to be specific requirements for medical devices. If it does, they would be applicable to specific products through a system of classification regulations, guidelines, and precedents related to those specific products, and should trigger preemption. And second, Congress must state whether or not it intends for the dederal medical device laws to preempt booth product-specific state laws and general state laws, such as those governing state tort claims.

Such clarification would accomplish some important things. First, it would provide the kind of harmony between the state and federal requirements that was anticipated in the original Medical Device Amendments; if Congress had not intended to provide preemption, why would it have put the special language specifically in the Medical Device Amendments in the first place? After all, a special provision is not part of the boilerplate for all legislation. Second, it would add greater weight to FDA regulations and provide a strong incentive to industry for voluntary compliance, because such compliance would provide relief from unnecessary and debilitating litigation.

ORPHAN DEVICES AND COMBINATION PRODUCTS

The Orphan Drugs Act, because of the effective nurturing and advocacy of the Office of Orphan Products, provides a fast route to market for drugs and biologics that treat rare disorders. It also provides attractive incentives to developers of such drugs and biologics, including the following:

The Orphan Drug Act also provides grants and contracts for device development, but does not provide market exclusivity. Under SMDA, however, there was a provision for a humanitarian device exemption, as a counterpart of the open protocols or treatment IND for drugs.

The rationale for providing incentives for orphan medical devices is not significantly different than that for drugs. Legislative change, therefore, is needed to provide parity for medical devices. Following are my proposals for the legislation:

CONCLUSION

The strength of FDA's regulatory system is its risk- and evidence-based regulations. To remain strong and relevant to new technologies and innovation, the system must grow with the growth in new technologies. At the same time, however, it should be pruned where it spends resources for more mundane products or where it is not adding significant value either in the establishment or demonstration of safety and effectiveness for devices.

There is no evidence that the exemption in the last decade of a large number of devices from the 510(k) process has adversely affected the public health. It is now time to further diminish the unnecessary regulatory burdens in order to refocus FDA on the needs of the new century.

Just as bitterness is not a measure of a drug's efficacy (as FDA would agree), so it is that excessive burdens though regulations are not necessary to ensure the safety and effectiveness of medical devices. On the other hand, the lack of new technologies to treat diseases and conditions has a predictably negative effect on public health.

The current proposals before Congress are a small step in the right direction, but an even stronger effort is needed.

Kshitij Mohan is a former head of FDA's Office of Device Evaluation and has served as the top science and technology officer at Baxter International and Boston Scientific Corp. He is currently chief regulatory and technology strategist for King & Spalding in Washington, DC.

Copyright ©2002 Medical Device & Diagnostic Industry