Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 207
A History of Medical-Device Legislation and Regulation in the United States T he committee was charged to review the 510(k) clearance process and to evaluate whether it protects patients optimally and promotes innovation in support of the public health. The Food and Drug Ad- ministration (FDA) strives to find an appropriate balance between promot- ing and protecting the public health.1 To aid in understanding the current 510(k) system and how it evolved to its present state, this appendix outlines pertinent elements in the history of device regulation as they are related to Class II devices generally and to the 510(k) clearance process in particular. The 510(k) system is a type of gatekeeping for managing the transfer of new technology from the laboratory to the bedside. Entering the commer- cial market is a single point in the life cycle of a product. The need for and rigor of premarket review are directly affected by the availability of other tools to protect patients after marketing commences. Effective postmarket- ing2 controls can reduce the burdens of premarket review and accelerate its speed. Thus, this appendix will examine device regulation beyond the 510(k) process itself. Of particular relevance is the safety surveillance system by which the FDA can identify and act on risks that emerge after a device enters the market. A robust and effective system might provide metrics for assessing 1Federal Food, Drug, and Cosmetic Act, § 1003(b) (the most current version will hereinafter be referred to as FFDCA), 21 USC § 393(b) (2006). 2The appendix will use the word postmarketing to refer to the situation after a product enters commercial distribution. The FFDCA uses postmarket specifically to describe surveillance studies that FDA can order under Section 522. 21 USC § 360l (2006). To avoid confusion, postmarket will be used only with reference to that provision of the law. 207
OCR for page 208
208 MEDICAL DEVICES AND THE PUBLIC’S HEALTH the nature and extent of harm that is associated with products cleared or approved by the FDA and therefore indirectly reflecting on the quality of the FDA review processes. In contrast, if safety surveillance is weak and of limited scope, that injuries are not identified can yield little confidence in the review process—absence of evidence of harm is not evidence of the absence of harm. Therefore, this appendix will review the history of adverse-event monitoring of medical devices. Consideration of the 510(k) clearance process largely involves Class II devices. Since 1997, most Class I devices have been exempted from 510(k) review. Most Class III devices are subject to FDA review through the pre- market approval (PMA) process, not the 510(k) pathway, although 25 types of Class III devices are still eligible to enter the market by the 510(k) clearance process. As recommended in Chapter 7, the committee urges the FDA to cease reliance on the 510(k) process for Class III devices as soon as possible. Once that step is completed, the 510(k) process will apply only to Class II devices—although a few are exempt from 510(k) requirements— and some Class I devices. In considering the gatekeeping function related to the introduction of new medical products to the market, the Federal Food, Drug, and Cosmetic Act (FFDCA) has adopted a variety of approaches: no government preclear- ance review whatsoever, a limited preclearance review, or extensive govern- ment review. For most drug products, Congress chose detailed review and affirmative approval by the FDA before commercialization, which has led to long and expensive processes of drug development. For medical devices, however, Congress has adopted all three approaches: exemption from any premarket review by the FDA before marketing of a product; limited pre- market review through a notification of intent to market, commonly called the 510(k) submission; and thorough review and affirmative approval before launching of a product (through the PMA). After entering the market, products continue to be subject to a variety of legal requirements and prohibitions, which are enforced primarily by the FDA. Congress has given the FDA diverse powers in a patchwork fashion, usually in response to demonstrated or alleged deficiencies in current sys- tems.3 In medical-device terminology, the requirements are called general and special controls and apply to devices according to the degree of potential 3The laws governing medical devices have evolved separately from those applicable to drugs even though many of the authorities appropriate for one system might seem equally appropriate for the other. The nonparallel development reflects the fact that Congress often approached new issues that emerged in one field of medical products without considering whether the issues might also exist in the other. Thus, for example, with respect to medical devices Congress has given the FDA authority to order the recall of products that violate the law, to require notification of voluntary product removals, and to impose civil money penalties for violations. The FDA has none of those authorities over drugs to the same degree. Conversely, the agency has much more extensive authority over labeling, advertising, and marketing of prescription
OCR for page 209
209 APPENDIX A risk presented but generally without regard to the extent of review before market entry. As with premarket review, postmarketing protections of health can affect innovation, depending on the rigidity and zeal with which the regulation with which regulations are enforced. The history of federal regulation of medical devices reflects several important forces. First, the FDA has legal constraints on its discretion in exercising its powers. To guarantee that the will of Congress would not be subverted by the executive branch, Congress set forth detailed instructions on how the FDA should carry out the law. The original Medical Device Amendments of 1976 more than doubled the length of the FFDCA as it stood at that time. Later amendments added far more text than they removed. Congressional micromanagement can be found in many parts of the current law. To ensure presidential control over executive branch activities, the FDA has been required to submit proposed major regulations and other policies to the Office of the Secretary of Health and Human Services and to the White House Office of Management and Budget (OMB) for review and ap- proval before publication, and there are procedural requirements for public participation to safeguard interests of patients, manufacturers, and others. Federal courts review the FDA’s interpretations of law, its regulations, and its actions. Judicial rulings can profoundly influence implementation of the medical-device statute. Second, the FDA does not operate in a political vacuum but is subject to scrutiny as it carries out its mandates. Regulatory policies and philosophies change from one president to the next. Congressional committees that have jurisdiction over the FDA can attempt to drive the agency toward policies and actions that they prefer through oversight hearings, committee inves- tigations, and reviews by the General Accounting Office (GAO, now the Government Accountability Office), formerly by the Office of Technology Assessment (no longer in existence), and by the Department of Health and Human Services inspector general. Third, the modern history of the FDA shows a disparity between con- gressional mandates, under the various legislative enactments, and congres- sional appropriations of persistently insufficient resources (FDA Science Board, 2007, 9-10). The appropriation process is subject to its own set of pressures, such as tax and fiscal policy, competing demands for scarce fed- eral revenues, and changing agendas in Congress and the executive branch. Thus, funding for the agency often bears little relation to the mandates and missions assigned by Congress. The most elegant and detailed schemes en- visioned when a statute is written can become unworkable when resources drugs than of prescription Class II devices. The disjointed evolution of two regulatory schemes governing medical technologies can lead to public misunderstanding and confusion over the FDA’s powers and responsibilities.
OCR for page 210
210 MEDICAL DEVICES AND THE PUBLIC’S HEALTH are inadequate. Necessary improvisation and tough choices at the FDA between competing priorities can move the regulatory regime in unforeseen directions. The last force affecting the history of medical-device regulation is the inherent difficulty of resolving some core issues, which can cause them to arise again and again as different solutions prove unsatisfactory. That those issues have recurred throughout the 35 years since the medical-device statute’s enactment in 1976 demonstrates how difficult they are to resolve. Those issues include • hether the 510(k) mechanism as originally enacted in 1976 or W even as modified in 1990 or 1997 can even theoretically provide determinations about the safety and effectiveness of each product reviewed. Simply stated, must a finding that a new device is substan- tially equivalent to another lawfully marketed product (the predicate device) necessarily be a determination that the new device is safe and effective? • hether the FDA can reasonably ensure the quality and consistency W of 510(k) clearance decisions. Manufacturers desire predictability of the regulatory process to manage what they maintain is the inherent uncertainty and risk in innovation. Moreover, because the FFDCA re- quires the FDA to accept any marketed device as a predicate, flawed 510(k) decisions become embedded and must be perpetuated. • hether the FDA has appropriately used its authority to establish W performance standards or special controls over Class II devices. What does it mean for public health if the “reasonable assurance of the safety and effectiveness” of a Class II device cannot, by definition, be ensured by general controls alone but the agency does not impose any additional controls? Is the adoption of special controls mandatory or discretionary? • hether the procedural requirements imposed by law properly bal- W ance the interests of private parties with the need for efficient and ef- fective government actions. The preliminary recommendations of the Center for Devices and Radiological Health (CDRH) 510(k) Working Group asserted that “CDRH’s postmarket tools, while valuable, have important limitations” (FDA, 2010, p. 4). When powers given to the FDA are so encumbered by “due process” demands that they become unworkable and are not used, no one is served. • hether lack of safety or effectiveness of marketed products can be W promptly identified and addressed. Ultimately, the regulatory scheme to safeguard public health can be considered reliable only if society can be confident that consumers are being protected. The most im- portant element in gaining that knowledge is surveillance of adverse
OCR for page 211
211 APPENDIX A medical experiences. Without a robust surveillance network to detect injuries and product ineffectiveness, there is no basis for judging the system. The absence of data on a problem does not demonstrate that a problem does not exist. • hether the appropriate balance has been achieved between premar- W ket clearance (the gatekeeping function) and postmarketing controls. Premarket clearance may add little value to protecting the public (while inhibiting innovation); in 1997, Congress eliminated 510(k) notification requirements for most Class I devices. But the prelimi- nary recommendations of the CDRH 510(k) Working Group stated that “CDRH’s postmarket tools, while valuable, have important limitations and are not sufficient to serve as a substitute for a high- quality premarket review” (FDA, 2010, p. 4). The Working Group seems to have concluded that inadequate or ineffective postmarketing controls justify increased rigor in the preclearance process to protect consumers. This appendix seeks to provide a chronologic inventory of medical- device legislation and external studies of its implementation. It is regrettably sprawling, shifting back and forth over seemingly disparate events and de- scribing laws as originally enacted, even though they have been superseded by amendments. Readers are encouraged to keep in mind the six issues just identified so that they can understand the relevance of the materials presented here and to accept our apologies for the somewhat disjointed structure adopted. Although the FDA has undertaken many internal reviews and efforts to improve the process, this appendix draws most of its infor- mation from congressional materials. The focus here is on what Congress has perceived and has done concerning the goals and implementation of the FDA’s regulation of medical devices. DEVICE REGULATION UNDER THE 1938 FEDERAL FOOD, DRUG, AND COSMETIC ACT The Pure Food and Drug Act of 1906 did not cover medical devices. As Congress approached overhaul of the statute in the 1930s, it added that class of product to FDA jurisdiction. For all intents and purposes, however, government authority and industry legal requirements in the draft legislation were identical for both drugs and devices. Until the brink of enactment, the bills imposed no premarket review of either but subjected them to prohibi- tions against being “adulterated” or “misbranded.” Those terms were legal words of art and covered a variety of specific mandates or prohibitions. For example, a drug or device would be “adulterated” if it had been prepared under insanitary conditions whereby the product may have been rendered
OCR for page 212
212 MEDICAL DEVICES AND THE PUBLIC’S HEALTH injurious to health. A medical product would be “misbranded” if its label failed to identify the name and address of the manufacturer or distributor.4 The draft statute further provided that once the FDA discovered a viola- tion of the requirements, it could seize the noncompliant product, seek an injunction against its manufacturer or distributor to prevent future viola- tions, or recommend to the Department of Justice criminal prosecution of the persons responsible.5 Those three options were the only enforcement tools in the bill. Final legislative action was precipitated by the elixir sulfanilamide disas- ter, which involved a product that put an effective drug (sulfa) into solution with a toxic solvent (diethylene glycol) that caused upward of 100 deaths in the winter of 1937–1938 (Carpenter, 2010, pp. 85–108). In response, Con- gress added a distinct provision applicable only to “new drugs”: premarket review of a new drug application (NDA) for safety. The manufacturer was not required to obtain FDA approval but only to provide the agency 60 days in which to make a safety assessment. If the government failed to find any safety issues within the 60-day window, the drug could enter the market.6 In that form, the FFDCA became law in 1938. Over the next several decades, the FDA directed much of its enforce- ment efforts to devices, using the adulteration and misbranding provisions of the FFDCA to deal with grossly hazardous or fraudulent products. By the early 1960s, however, the agency’s attention was drawn to new technolo- gies that were being introduced without premarket clinical testing, quality control, or patient consent.7 In 1962, Congress amended the new drug provisions of the FFDCA to require affirmative FDA approval of an NDA before marketing could begin and to require that a new drug be shown to be effective as well as safe.8 The broadened premarket authority for “new drugs” encouraged the FDA to consider applying these powers to medical devices. The opportunity to do so was supplied by the overlap in the definitions of drug and device in the original 1938 Act. As written, the terms had similar scope and were differentiated only by a drug’s being an “article” whereas a device was an “instrument, apparatus, . . . [or] contrivance” that was either recognized in an official compendium, such as the US Pharmacopeia; was “intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease”; 4FFDCA, Pub. L. No. 75-717, §§ 501-02, 52 Stat. 1040, 1049-50 (1938). 5FFDCA §§ 301-04, 52 Stat. at 1042-45. 6FFDCA § 505, 52 Stat. at 1052-53. 7H.R. Rep. No. 94-853, at 7-8 (1976) (House Report on the Medical Device Amendments of 1976). 8Drug Amendments of 1962, Pub. L. No. 87-781 (codified as amended at 21 USC § 355(b) (1)(a)(A)).
OCR for page 213
213 APPENDIX A or was “intended to affect the structure or any function of the body.”9 In two important legal cases, the FDA persuaded federal courts that it had statutory authority to declare that products that were generally thought of as medical devices were drugs and new drugs requiring agency approval of an NDA before marketing. One product consisted of a disposable applica- tor, a nylon ligature loop, and a nylon locking disk used to tie off severed blood vessels during surgery; the disk and at least some of the nylon thread remained in the patient’s body.10 The other product was an antibiotic- sensitivity disk.11 Ultimately, the FDA classified a number of products as drugs before the Medical Device Amendments were enacted, including con- tact lenses, injectable silicone, pregnancy-test kits, and bone cement (Hutt et al., 2007, p. 977). Only devices classified as drugs were subject to any premarket review by the FDA before 1976. DEVELOPMENT AND ENACTMENT OF THE MEDICAL DEVICE AMENDMENTS OF 1976 Presidents Kennedy, Johnson, and Nixon all recognized “the need for more comprehensive authority to regulate medical devices.”12 At the direc- tion of the secretary of the Department of Health, Education, and Welfare (now the Department of Health and Human Services), Theodore Cooper, director of the National Heart and Lung Institute (now the National Heart, Lung, and Blood Institute) chaired a study group to consider how to ap- proach new legislation; the group issued its report in September 1970. Among its recommendations were better differentiation of devices and drugs in federal law with the adoption of a new and distinct regulatory regime for medical devices, preclearance review of some devices with the extent of premarket review depending on the novelty and potential hazards of the devices, and expansion of the FDA’s nonpreclearance authority over medical devices. The last category included mandatory records and reports, adherence to good manufacturing practices, manufacturer registration, and FDA inspection of factories.13 Over the next few years, several high-profile public-health problems that involved medical devices were observed. Among the most publicized was the Dalkon Shield, an intrauterine contraceptive device (IUD) that was introduced into the market in late 1970. By 1975, at least 16 deaths, 25 miscarriages, numerous cases of pelvic perforation and pelvic infection, re- 9FFDCA § 201(g)-(h), 52 Stat. at 1041 (1938). 10AMP, Inc. v. Gardner, 389 F.2d 825 (2d Cir. 1968), cert. denied, 393 U.S. 825 (1968). 11U.S. v. An Article of Drug . . . Bacto-Unidisk, 394 U.S. 784 (1969). 12H.R. Rep. No. 94-853, at 9. 13H.R. Rep. No. 94-853, at 9-10; U.S. Dep’T of HeaLTH, eDUc., & WeLfaRe, MeDicaL DeviceS: a LegiSLaTive pLaN passim (1970) (Cooper Committee Report).
OCR for page 214
214 MEDICAL DEVICES AND THE PUBLIC’S HEALTH moval of the IUD for medical reasons, and pregnancies due to IUD failure had been reported. Marketing of the device ceased by 1976. Other examples cited by the House of Representatives committee responsible for the 1976 legislation included pacemaker failures and dangerous eye infections after implantation of intraocular lenses.14 In 1974 and 1975, the Senate passed comprehensive legislation for the regulation of medical devices.15 The House of Representatives, however, did not move with its own bill until March 1976.16 A conference commit- tee reconciled the differences between the two bills.17 The Medical Device Amendments of 1976 (MDA) passed both houses of Congress and were signed into law by President Ford on May 28.18 As discussed below, the medical-device law has since been amended many times, most significantly in 1990, 1992, 1997, 2002, 2004, 2005, and 2007. None of those amendments, however, has altered the fundamental regulatory regime established in 1976. BASIC STRUCTURE OF THE MEDICAL DEVICE AMENDMENTS Although the new statute provided for premarket review of some prod- ucts, protection of public health rested predominantly on rules enforced by the FDA after a product entered the market. In that respect, the MDA departed from the approach taken for drugs in 1962, which put extraor- dinary emphasis on the FDA premarket approval process as the primary means of protecting consumers. For devices, the “gatekeeper” function of premarket approval provided an additional layer of protection but only for a small array of products (Class III), and it did not replace or render un- necessary the variety of other requirements applicable to all other devices. As originally enacted, the 510(k) clearance process was not intended to be a major preclearance mechanism. Distinguishing Devices from Drugs To establish a distinctive regulatory scheme for medical devices, Con- gress first had to address the overlapping definitions of drug and device in the 1938 law. As noted above, federal courts found the terms sufficiently congruent to uphold the FDA’s determination that several products thought 14H.R. Rep. No. 94-853, at 8 (1976). 15S.Rep. No. 94-33, at 1 (1975) (accompanying S. 510). 16H.R. Rep. No. 94-853, at 1 (accompanying H.R. 11124). 17H.R. Rep. No. 94-1090, at 1 (accompanying S. 510). 18Medical Device Amendments of 1976, Pub. L. No. 94-295, 90 Stat. 539 (1976) (hereinafter 1976 MDA).
OCR for page 215
215 APPENDIX A of as devices could nevertheless be regulated as drugs. The MDA provided as follows The term “device” . . . means an instrument, apparatus, implement, ma- chine, contrivance, implant, in vitro reagent, or other similar or related article, including any component, part, or accessory, which is— 1. recognized in the official National Formulary, or the United States Pharmacopeia, or any supplement to them, 2. intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease, in man or other animals, or 3. intended to affect the structure of any function of the body of man or other animals, and which does not achieve any of its principal intended purposes through chemical action within or on the body of man or other animals and which is not dependent upon being metabolized for the achievement of any of its principal intended purposes.19 General Controls The MDA established requirements for industry and authorities for the FDA that would apply to all medical devices. Those provisions are usually described as the general controls and include the following:20 • pplying to all devices the “adulteration” provisions of the 1938 A Act, including— o prohibiting potential or actual contamination of the product;21 and o requiring adherence to good manufacturing practice regulations promulgated by the FDA.22 • pplying to all devices the “misbranding” provisions of the 1938 A Act, including— o prohibiting any false or misleading statements in labeling,23 19FFDCA § 201(h), 90 Stat. 575 (1976). The definition was changed in 1992 to substitute “its primary intended purposes” for “any of its principal intended purposes.” 106 Stat. 240 (1992). For current definition see FFDCA § 201 (h), 21 USC § 321(h) (2006). 20This summary necessarily abbreviates the statutory language. It is intended to give an overview, not provide a detailed description of each provision of the law. 21FFDCA § 501(a)(1), (a)(2)(A), (3), (4)(A), 21 USC § 351(a)(1), (a)(2)(A), (3), (4)(A) (2006). 22FFDCA §§ 501(h), 520(f), 21 USC §§ 351(h), 360j(f) (2006). 23FFDCA § 502(a), 21 USC § 352(a) (2006).
OCR for page 216
216 MEDICAL DEVICES AND THE PUBLIC’S HEALTH o requiring disclosure of the name and address of the manufacturer,24 and o mandating “adequate directions for use” and “adequate” warn- ings against unsafe use in labeling for the product.25 • A uthorizing the FDA to restrict the sale, distribution, or use of a device (a) only on the authorization of a licensed practitioner or (b) upon such other conditions as the FDA may prescribe, if—because of the potential for harm or the collateral measures necessary to its use—the FDA determines that there cannot otherwise be reasonable assurance of the safety and effectiveness of the device (devices so designated are called “restricted devices”).26 rohibiting false or misleading advertising of a “restricted device.”27 • P • M andating disclosures in all advertising for “restricted devices” of a brief statement of intended uses and relevant warnings.28 • R equiring registration with the FDA by all manufacturers of medical devices.29 • M andating listing with the FDA by a registered firm of all devices it currently markets.30 • R equiring notification to the FDA of any new device proposed to be marketed by a registered firm, at least 90 days prior to introduction for commercial distribution [this is the provision that gave rise to the 510(k) process that is the focus of the committee’s study].31 • A uthorizing the FDA to inspect factories in which medical devices are manufactured and to inspect records relating to “restricted devices” in those facilities.32 • A uthorizing the FDA to ban a device from sale or use if it presents substantial deception or an unreasonable and substantial risk of ill- ness or injury.33 • A uthorizing the FDA to order notification to physicians and others (including patients and other end users) if three conditions are all met: (a) a device presents an unreasonable risk of substantial harm to the public health, (b) notice is necessary to eliminate the risk, and (c) no other more practicable means is available to eliminate this risk.34 24FFDCA § 502(b), 21 USC § 352(b) (2006). 25FFDCA § 502(f), 21 USC § 352(f) (2006). 26FFDCA § 520(e), 21 USC § 352(e) (2006). 27FFDCA § 502(q), 21 USC § 352(q) (2006). 28FFDCA § 502(r), 21 USC § 352(r) (2006). 29FFDCA § 510(c), 21 USC § 352(c) (2006). 30FFDCA § 510(j)(1), 21 USC § 360(j)(1) (2006). 31FFDCA § 510(k), 21 USC § 360(k) (2006). 32FFDCA § 704(a), 21 USC § 374(a) (2006) (2006). 33FFDCA § 516, 21 USC § 360f (2006). 34FFDCA § 518(a), 21 USC § 360h(a) (2006).
OCR for page 217
217 APPENDIX A • uthorizing the FDA to order a manufacturer to repair a device, A replace the device, and/or refund the purchase price of the device, if four conditions are all met: (a) the device presents an unreasonable risk of substantial harm to the public health, (b) the device was not properly designed or manufactured to the state of the art when de- signed or made, (c) the unreasonable risk was not caused by someone other than the manufacturer to exercise due care with the installa- tion, maintenance, or use of the device, and (d) notification by itself would not be sufficient to eliminate the unreasonable risk, so that repair, replacement, and/or refund by the manufacturer is necessary to eliminate this risk.35 • uthorizing the FDA to require manufacturers to maintain records A and submit reports (such as reports of adverse medical events associ- ated with medical devices) to assure that a device is not adulterated or misbranded and to otherwise assure its safety and effectiveness36 (the statute presumed, however, that Class I devices would not be subject to reporting in ordinary circumstances).37 • uthorizing the FDA to order detention of a device for up to 20 days, A if the agency believes it is adulterated or misbranded, to permit the preparation and filing of a court action for seizure of the product.38 It must be emphasized that the foregoing requirements and authorities were established in 1976 almost always without regard to the assignment of a medical device into Class I, Class II, or Class III (the classes are defined below). The law has since been amended to exempt many Class I products from some general controls or to limit the application of general controls to subsets of Class II or III products that pose higher risks. But the original MDA rarely made those distinctions. In cases in which authority was con- ditioned on an FDA finding of a higher degree of risk, the finding was to be made without regard to the prior classification of the device. Device Classification and Premarket Review The 1976 act adopted a three-tier system to determine the need for ad- ditional regulatory controls in the form of premarket review to protect and promote public health. The tiers correspond to the perceived risks posed by the devices. Thus, the law represented a sharp break with the approach 35FFDCA § 518(b), 21 USC § 360h(b) (2006). 36FFDCA § 519(a), 21 USC § 360i(a) (2006). 37FFDCA § 519(a)(8), 21 USC § 360i(a)(8) (2006) (redesignated from 519(a)(5) by the Medical Device Amendments of 1992, Pub. L. No. 102-300, § 5(a)(1), 106 Stat. 238, 239 (1992) (hereinafter 1992 MDA). 38FFDCA § 304(g), 21 USC § 334(g) (2006).
OCR for page 268
268 MEDICAL DEVICES AND THE PUBLIC’S HEALTH Limitation of Food and Drug Administration Authority to Withhold 510(k) Clearance Because of Failure to Comply with Other FFDCA Requirements Congress understood that the FDA had undertaken a policy of withhold- ing clearance of a 510(k) submission when an inspection of the manufac- turer revealed GMP violations. To countermand that use of FDA authority, the 1997 law required the agency to act on the 510(k) submission without regard to the manufacturer’s failure to comply with any other provision of the FFDCA “unrelated to a substantial equivalence decision.” To be clear, Congress went on to say that “including a finding that the facility in which the device is manufactured is not in compliance with good manufacturing requirements . . . (other than a finding that there is a substantial likelihood that the failure to comply . . . will potentially present a serious risk to hu- man health).”207 Restriction of Scope of Review to Proposed Indications for Use In response to complaints that the agency was denying 510(k) clearances because of potential “off-label” uses, Congress specified that the FDA was to make determinations solely on the indications for use proposed by the manufacturer in the labeling with the 510(k) submission.208 The new law did, however, permit the agency to require warnings in labeling with respect to off-label uses if there is a reasonable likelihood that the device will be used for an intended use not identified in the proposed labeling and that use could cause harm. That special authority could be exercised only by the director of the Office of Device Evaluation, not by any subordinate managers. Restriction of Scope of Review to Least Burdensome Means of Demonstrating Equivalence The SMDA of 1990 had codified the FDA’s approach to determining substantial equivalence, including authorizing the FDA to request addi- tional information in the 510(k) submission to confirm equivalence.209 The FDA had already, on its own initiative, revised the substantial-equivalence standard to limit requests for data only when the differences presented new 207FFDCA § 513(f)(5), 21 USC § 360c(f)(5) (2006) (added by FDAMA § 206(b), 111 Stat. at 2339). 208FFDCA § 513(i)(1)(E), 21 USC § 360e(i)(1)(E) (2006) (added by FDAMA § 205(b), 111 Stat. at 2337-38). 209FFDCA § 513(i)(1)(A)(ii), 21 USC § 360c(i)(1)(A)(ii) (2006) (“the information submitted that the device is substantially equivalent to such other device, including clinical data if required by [FDA], demonstrates that the device is as safe and effective as such other device which is currently being sold”).
OCR for page 269
269 APPENDIX A types of questions or new risks that could not be mitigated (IOM, 2010, p. 11). In 1997, Congress restricted the agency further, specifying that the FDA could request information only if it were necessary to make the substantial- equivalence determination. Moreover, in making such a request, the FDA has to “consider the least burdensome means of demonstrating substantial equivalence and request information accordingly.”210 Device Classification (and Reduction in Scope of PMA Review Regarding Effectiveness) to Consider Substitution of Postmarket Controls The 1997 law directed the FDA to “consider the extent to which reli- ance on postmarket controls may expedite the classification of devices” into Class I or Class II instead of Class III.211 Congress ordered that “in making a determination of a reasonable assurance of effectiveness” in reviewing a PMA, the FDA must “consider whether the extent of data that otherwise would be required for approval . . . with respect to effectiveness can be reduced through reliance on post- market controls.”212 The legislation also provided for collaborative determinations by the FDA and a PMA applicant of the type of scientific evidence needed to ap- prove a particular PMA. The law created a process for meetings, internal appeals of scientific disagreements between a company and the FDA re- viewers, and binding written FDA commitments on the evidence needed.213 Congress was again interested in reducing unnecessary clinical testing of medical devices. It provided214: Any clinical data, including one or more well-controlled investigations, specified in writing by [FDA] for demonstrating a reasonable assurance of device effectiveness shall be specified as [a] result of a determination by [FDA] that such data are necessary to establish device effectiveness. [FDA] shall consider, in consultation with the applicant, the least burdensome appropriate means of evaluating device effectiveness that would have a reasonable likelihood of resulting in approval. 210FFDCA § 513(i)(1)(D), 21 USC § 360c(i)(1)(D) (2006) (added by FDAMA § 205(b), 111 Stat. at 2337). 211FFDCA § 513(i)(C), 21 USC § 360c(i)(C) (2006) (added by FDAMA § 205(b), 111 Stat. at 2337). 212FFDCA § 513(a)(3)(C), 21 USC § 360c(a)(3)(C) (2006) (added by FDAMA § 205(a), 111 Stat. at 2336). 213FFDCA § 513(a)(3)(D), 21 USC § 360c(a)(3)(D) (2006) (added by FDAMA § 205(a), 111 Stat. at 2336-37); FFDCA § 562, 21 USC § 360bbb-1 (2006) (added by FDAMA § 404, 111 Stat. at 2368). 214FFDCA § 513(a)(3)(D)(ii), 21 USC § 360c(a)(3)(D)(ii) (2006) (added by FDAMA § 205(a), 111 Stat. at 2337).
OCR for page 270
270 MEDICAL DEVICES AND THE PUBLIC’S HEALTH Finally, the new law clarified that establishing effectiveness would require only one clinical investigation.215 The legislative history of these provisions contains the following explanation, which underscores the nega- tive congressional reaction to the Temple report of 1993, which had recom- mended greater consideration of randomized controlled studies to support medical-device safety and effectiveness216: The Committee believes that [these amendments] are necessary to and consistent with improving communication between the FDA and regu- lated persons, increasing regulatory efficiency, and decreasing the length of product review and approval. In particular, the Committee is aware of examples where the FDA has requested inappropriate types of clinical testing for certain breakthrough devices and is concerned about instances in which the agency has required sponsors to conduct unnecessary ran- domized clinical studies to demonstrate device effectiveness. Although randomized clinical testing may be the best means of demonstrating de- vice effectiveness for some products, the Committee is informed that such testing is often unnecessary to demonstrate effectiveness for many devices. Alternative Procedures for Setting Performance Standards for Class II Devices The 1997 law provided that the FDA could recognize an appropriate performance standard established by a nationally or internationally recog- nized organization engaged in standard development. In lieu of an extended notice-and-comment rule-making, the FDA could simply issue a notice of recognition.217 In addition, Congress provided a simplified process for ap- plicants to certify compliance with the standard.218 Priority PMA Review for Important New Devices Congress directed the FDA to provide review priority for medical de- vices that represented breakthrough technologies, for which no approved alternatives existed, that offered substantial advantages over existing ap- 215FFDCA § 513(a)(3)(A), 21 USC § 360c(a)(3)(A) (2006) (revised by FDAMA § 217, 111 Stat. at 2350). 216H.R. Rep. No. 105-307, at 24 (1997). 217FFDCA § 514(c)(1)(A), 21 USC § 360d(c)(1)(A) (2006) (added by FDAMA § 204(a), 111 Stat. at 2335-36). 218FFDCA § 514(c)(1)(B), 21 USC § 360d(c)(1)(B) (2006) (added by FDAMA § 204(a), 111 Stat. at 2335-36).
OCR for page 271
271 APPENDIX A proved alternatives, or whose availability would be in the best interest of patients.219 Certainty of Review Timeframes Congress reinforced the requirement that 510(k) submissions be acted on within 90 days of filing.220 Modifications of Postmarketing Controls For the first time, Congress substantially aligned postmarket controls to device classifications in 1997. Perhaps unwittingly, that restructuring may have discouraged the FDA from down-classifying device types from Class II to Class I. Doing so would have sharply reduced the potential regulatory tools available to the agency in the future if a problem with the down- classified type of device emerged. Limiting Device Tracking to Class II and Class III Devices The 1997 law limited the FDA’s authority to require device tracking to Class II or Class III devices whose failure would be reasonably likely to have serious adverse health consequences or that are intended to be implanted for more than 1 year or are life-sustaining or life-supporting and used outside a device-user facility.221 Limiting Postmarket Surveillance Requirements to Class II and Class III Devices Congress also restricted the FDA’s power to order postmarket surveil- lance for devices to those in Class II or Class III whose failure would be reasonably likely to have serious adverse health consequences or that are intended to be implanted for more than 1 year or are life-sustaining or life- supporting and used outside a device-user facility.222 219FFDCA § 515(d)(5), 21 USC § 360e(d)(5) (2006) (added by FDAMA § 202, 111 Stat. at 2334). 220FFDCA § 510(n), 21 USC § 360(n) (2006) (added by FDAMA § 209(a), 111 Stat. at 2341). 221FFDCA § 519(e), 21 USC § 360i(e) (2006) (revised by FDAMA § 211, 111 Stat. at 2345-46). 222FFDCA § 522, 21 USC § 360l (2006) (revised by § 212, 111 Stat. at 2346).
OCR for page 272
272 MEDICAL DEVICES AND THE PUBLIC’S HEALTH Changes in Requirements for Reporting Adverse Events Associated with Medical Devices The 1997 law eliminated mandatory reporting by distributors of medi- cal devices.223 It also eased reporting requirements for device-user facilities and directed the FDA to establish a sentinel reporting system, relying on a representative sample of all device-user facilities, to collect information on device-related deaths and serious illness or injuries.224 1997–2009 DEVELOPMENTS The Medical Device User Fee and Modernization Act of 2002 The Medical Device User Fee and Modernization Act had two major components but did not make any fundamental change in the general regula- tory scheme created by the prior enactments. The first established a “user- fee” program whereby medical-device manufacturers paid fees for some applications, reports, and supplements sent to the FDA for evaluation and imposed performance goals for the FDA in reviewing these applications. The user-fee program was to operate for 5 years. “The fees are based on the rela- tive level of effort required for reviews.”225 It is interesting to observe, there- fore, that Congress fixed the effort required for a 510(k) review at 1.75% of that for a PMA review.226 That ratio is not substantially different from the one described by the FDA in 1987, when the agency said that the average PMA review took 1,200 hours but the average 510(k) review took only 20 hours (1.67%).227 Congress also authorized specific federal appropriations for the FDA to conduct safety surveillance of marketed medical devices. The other part of the legislation consisted of targeted changes to reduce regulatory burdens and agency workload. For example, Congress • reated a system whereby people other than federal employees could C be accredited to conduct inspections of manufacturers of Class II and Class III devices.228 • ermitted labeling for prescription devices intended for use in health- P 223FFDCA § 519(a), 21 USC § 360i(a) (2006) (revised by FDAMA § 213(a), 111 Stat. at 2346-47). 224FFDCA § 519(b), 21 USC § 360i(b) (2006) (revised by FDAMA § 213(c), 111 Stat. at 2347-48). 225H.R. Rep. No. 107-728, pt. 1, at 24 (2002). 226Id. at 24 table l.1. 227S. Rep. No. 101-513, at 15; Medical Devices and Drug Issues: Hearings Before the Subcomm. on Health and the Env’t of the H. Comm. on Energy and Commerce, 100th Cong. 384 (1987). 228FFDCA § 704(f)–(g), 21 USC § 374(f) (2006) (added by the Medical Device User Fee and
OCR for page 273
273 APPENDIX A care facilities to be provided solely by electronic means (for example, over the Internet).229 • irected that registration of device manufacturers and listing of mar- D keted devices be converted to electronic filings as soon as feasible.230 • equired the FDA to undertake “modular review” of discrete por- R tions of a PMA before the filing of a complete PMA so that questions on these portions could be resolved early and the number of issues that might delay final approval could be reduced.231 The Medical Devices Technical Corrections Act of 2004 The Medical Devices Technical Corrections Act of 2004 made a series of minor corrections in elements of the 2002 law but did not modify the substance of that act or the overall system for regulating medical devices.232 The Medical Device User Fee Stabilization Act of 2005 The Medical Device User Fee Stabilization Act of 2005 was intended to address gaps in expected revenues in the 2002 user-fee legislation.233 The Food and Drug Administration Amendments Act of 2007 The most recent device-related legislation renewed the user-fee system but did not alter the regulatory scheme in any meaningful way.234 January 2009 Government Accountability Office Report The 2007 amendments directed GAO to examine which premarket review process—510(k) or PMA—the FDA used to review selected types of devices from FY 2003 through FY 2007.235 It found that the agency had reviewed 13,199 submissions for Class I and Class II devices through the Modernization Act of 2002, Pub. L. No. 107-250, § 201, 116 Stat. 1588, 1602-09 (2002) [hereinafter, MDUFMA]). 229FFDCA § 502(f), 21 USC § 352(f) (2006) (added by MDUFMA § 206, 116 Stat. at 1613). 230FFDCA § 510(p), 21 USC § 360(p) (2006) (added by MDUFMA § 207, 116 Stat. at 1613). 231FFDCA § 515(c)(3), 21 USC § 360e(c)(3) (2006) (added by MDUFMA § 209, 116 Stat. at 1613). 232Medical Devices Technical Corrections Act, Pub. L. No. 108-214, 118 Stat. 572 (2004). 233Medical Device User Fee Stabilization Act of 2005, Pub. L. No. 109-43, 119 Stat. 439 (2005). 234Medical Device User Fee Amendments of 2007, Title II of the Food and Drug Administration Amendments Act, Pub. L. No. 110-85, 121 Stat. 823, 842 (2007). 235Food and Drug Administration Amendments Act of 2007, Pub. L. No. 110-85 § 225, 121 Stat. 823, 854 (2007).
OCR for page 274
274 MEDICAL DEVICES AND THE PUBLIC’S HEALTH 510(k) clearance process and cleared 90% for marketing. In addition, 342 submissions under 510(k) were for Class III devices, of which 67% were cleared. Finally, 217 original PMA applications and 784 supplemental PMA applications for Class III devices had been reviewed, of which 78% were approved. As of October 2008, 20 types of Class III devices could still be cleared through the 510(k) clearance process (GAO, 2009b, pp. 6, 17).236 GAO observed that, relative to the PMA process, the 510(k) review system is generally less stringent, faster, and less expensive. With respect to strin- gency, GAO said: For most 510(k) submissions, clinical data are not required and substan- tial equivalence will normally be determined based on comparative device descriptions, including performance data. In contrast, in order to meet the PMA approval requirement of providing reasonable assurance that a new device is safe and effective, most original PMAs and some PMA supple- ments require clinical data. In addition, other aspects of FDA’s premarket review are [also] less stringent. . . . For example, FDA generally does not in- spect manufacturing establishments as part of the 510(k) premarket review process—the 510(k) review process focuses primarily on the end product of the manufacturing process rather than the manufacturing process itself. In contrast, the agency does inspect manufacturing establishment as part of its review of original PMA submissions [Footnotes omitted]. (GAO, 2009b, pp. 14–15) The estimated average cost to the FDA for reviewing a 510(k) in FY 2005 was about $18,200, compared with $870,000 for a PMA (GAO, 2009b, pp. 14–15). That is a ratio of roughly 2.09%, not far from the 1987 ratio of 1.67% based on the estimated numbers of staff-hours to review a 510(k) (20) and a PMA (1,200). As mentioned before, the FDA does not believe that the 1987 estimates are reliable today (Desjardins, 2011). October 2009 Department of Health and Human Services Inspector General Report The HHS inspector general conducted a review of medical-device adverse-event reporting in 2003–2007 to assess the extent of reporting, the extent of compliance with reporting requirements by manufacturers and user facilities, and the use of these reports by the FDA to identify and ad- dress safety concerns (OIG, 2009). Among the findings, the FDA received twice as many reports in 2007 as in 2003; most manufacturer reports were on time, but many of the user-facility reports and those from manufacturers that were supposed to be filed in 5 days because of their seriousness were 236The FDA reports that as of 2009, 26 types of Class III devices could still be cleared through the 510(k) clearance process (FDA, 2011).
OCR for page 275
275 APPENDIX A late; and CDRH was not using the reports in a systematic manner to identify and address safety issues (OIG, 2009, ii). On the last point, the inspector general made a number of specific observations, which echoed the GAO report of 1997 (OIG, 2009, pp. ii–iii): • nalysts did not adequately document information regarding their A reviews, so it was difficult to trace a response to an individual event. • he FDA could not link postmarket surveillance activities to particu- T lar events. • he agency lacked a system to document when adverse-event reports T resulted in onsite inspections. • here was a serious delay between receipt of reports and when they T are substantively reviewed in the FDA. • he FDA rarely acted when manufacturers or user facilities submit- T ted late reports. • he agency made little use of annual reports from user facilities. T The inspector general recommended that the FDA develop a protocol for reviewing adverse-event reports, including documentation requirements, provisions for timely review of high-priority reports, and follow-up with late reporters (OIG, 2009, pp. iii–iv). CLOSING OBSERVATIONS This chronologic inventory has attempted to identify materials bearing on six issues relevant to the consideration of the effectiveness of the 510(k) system: • hether a 510(k) determination that a new device is substantially W equivalent to a predicate device is a finding that the new device is safe and effective. • hether 510(k) decisions made over the past 35 years are of suf- W ficient quality and reliability to serve as predicates for future 510(k) actions. • hether each type of Class II device must be subject to special con- W trols to provide reasonable assurance of the safety and effectiveness of devices of that type. • hether procedural requirements have unduly interfered with the W FDA’s ability to execute its duties. • hether the postmarketing system for detecting device-related ad- W verse medical events or failures of effectiveness can be relied on in
OCR for page 276
276 MEDICAL DEVICES AND THE PUBLIC’S HEALTH drawing conclusions about the overall effectiveness of the regulatory scheme. • hether the appropriate balance has been found between premarket W clearance and postmarketing controls. Those issues have repeatedly been identified and discussed by congres- sional bodies, by HHS, by the courts, and by the FDA itself. The commit- tee’s conclusions and recommendations on those questions are set forth in Chapter 7 of the present report. As a general observation, however, it is clear that the consistent lack of adequate resources can undo much of what Congress envisions when it enacts a regulatory law. The experience of the FDA in implementing the medical-device law is an excellent example. Congress set forth an elaborate regulatory scheme with classification of all devices, PMA for novel and high- risk devices, performance standards and special controls for medium-risk devices, a preclearance mechanism for medium-risk and low-risk devices, surveillance for adverse events related to devices in use, and numerous pow- ers by which the FDA could act to address problems with marketed devices. Congress never, however, provided the funds to perform all those tasks in strict accordance with the statute and within the timeframes that Congress desired. The agency later concluded that reviewing PMAs and reclassifying Class III devices were too expensive, as would establishing performance standards have been. The FDA evolved the 510(k) mechanism into an al- ternative to those options. The system that the FDA operates today reflects cumulative adminis- trative interpretations of the laws enacted by Congress in 1976, 1990, and 1997, informed by internal and external criticisms, advanced by technologic processes, and adjusted to the political and budget priorities of 6 presiden- tial administrations and 18 Congresses. But it is not necessarily the system that would be designed if Congress started over today in recognition of the likelihood of the FDA’s resources for the foreseeable future. REFERENCES Carpenter, D. 2010. Reputation and power: Organizational image and pharmaceutical regula- tion at the FDA (Princeton studies in American politics: Historical, international, and comparative perspectives). First edition. Princeton, NJ: Princeton University Press. Cooper, R. 1987. Clinical data under section 510(k). Food, Drug, Cosmetic Law Journal 42(2):192-202. Desjardins, P. R. 2011. FDA response to information inquiry from the committee on the public health effectiveness of the 510(k) clearance process: 510(k) and PMA FTEs. Silver Spring, MD, 01/11/2011. FDA (Food and Drug Administration). 1993. The Temple report: Final report of the commit- tee for clinical review. Based on a review of selected medical device applications. Silver Spring, MD: Food and Drug Administration.
OCR for page 277
OCR for page 278
277 APPENDIX A ———. 2010. CDRH preliminary internal evaluations—Volume I: 510(k) working group pre- liminary report and recommendations. Silver Spring, MD: Food and Drug Administration. ———. 2011. 515 program initiative: The 515 program initiative (aka reclassification of pre- amendments class III 510(k)s).
278 MEDICAL DEVICES AND THE PUBLIC’S HEALTH OTA (Office of Technology Assessment). 1984. Federal policies and the medical devices indus- try. Washington, DC: Congress of the United States, Office of Technology Assessment. Vodra, W. W. 1981. Litigating an “old drug”/”new drug” case: Victory lies outside the court- room. Food, Drug, Cosmetic Law Journal 36:191.
OCR for page 278