Food and Drug Administration Advisory Committees (1992)

The Food and Drug Administration is responsible, among other things, for ensuring the safety and effectiveness of drugs, biologics, and medical devices. Its scientific and regulatory responsibilities in these areas arise from different historical periods, derive from different statutory bases, affect different industries, and are embedded in different organizations and processes. Its responsibilities encompass investigational drugs, biologics, and devices; the evaluation and regulation of new products; postmarketing surveillance of some products; licensing of establishments; oversight of manufacturing processes; product labeling and advertising of prescription drugs and restricted devices; and other functions.

Public advisory committees are used widely throughout the federal government for a wide array of purposes.^1,2 They are subject to the provisions of the Federal Advisory Committee Act (FACA). As used within the FDA, such committees may be ad hoc or standing; they are further classified as policy advisory committees and technical advisory committees. The former advise on ''broad and general matters''; the latter deal with "specific technical or scientific issues, which may relate to regulatory decisions before FDA (21 CFR 14.1(b)(2), 1991). In this report, we are primarily concerned with standing technical advisory committees to the FDA that deal with drugs, biologics, and medical devices. Only brief consideration is given to policy advisory committees.

The Code of Federal Regulations (CFR) sets forth the FDA's definition of the primary characteristics of an advisory committee:

The FDA uses technical advisory committees of outside scientific experts to advise it on the approvability of specific products and on the scientific and clinical policy issues it confronts regarding product development and evaluation. The agency also uses these committees to legitimate the soundness of its analysis of a given product, as a public forum for discussion of controversial issues, and, on occasion, as an "appeals court" for disputed agency decisions.^*

This chapter recounts the history of FDA advisory committees as it has evolved along somewhat different pathways for drugs, biologics, and medical devices. Variations are due partly to the differences in regulatory responsibilities in these areas and partly to the administrative entities and their cultures. At the end of the chapter, a brief section contrasts FDA advisory committees with the study sections of the National Institutes of Health.

The FDA initiated the use of advisory committees in the 1960s and 1970s for the evaluation of drugs. It extended their use in the early 1970s to the review of biologics soon after the Division of Biological Sciences of the National Institutes of Health was transferred to the FDA as the Bureau of Biologics. Finally, following the Cooper report of 1970, FDA in the early 1970s began to use such committees to classify medical devices, a step that Congress later mandated in the Medical Device Amendments of 1976 for both classification and product evaluation purposes.

The FDA's use of agency-chartered advisory committees for drug evaluation has evolved over the three decades since the 1962^** drug amendments to the Food, Drug, and Cosmetic Act. Those amendments required FDA to assess all new drugs for effectiveness, in addition to safety (as required by the 1938 amendments), and to reassess for effectiveness

nearly 4,000 prescription drugs that had been introduced to the market between 1938 and 1962—before proof of effectiveness was required.

The FDA responded by seeking external advice from the National Academy of Sciences-National Research Council (NAS-NRC) on previously marketed prescription drugs, establishing its own review committees for over-the-counter drugs, and extending such committees to new prescription drugs. The evolution of this use of outside scientific experts, recounted by Cooper for the 1960s, is briefly summarized here and then discussed at greater length below for the NAS-NRC Drug Efficacy Study and for the OTC review.

The Thalidomide controversy brought the teratogenic, mutagenic, and carcinogenic effects of drugs to public and scientific consciousness and provided a powerful stimulus for enacting the 1962 drug amendments. In partial response, Commissioner George Larrick established an Advisory Committee on Teratology. He also established an ad hoc committee to review the new drug application for Enovid, the first oral contraceptive. ^*

Dr. Joseph Sadusk, Jr., director of the Bureau of Medicine (predecessor to CDER) under Larrick, created a Medical Advisory Board and a series of standing and ad hoc advisory committees. His justification for the latter was to broaden the flow of communication to the bureau director beyond the immediate, full-time staff, there being no way in Sadusk's judgment for the FDA to acquire all of the staff expertise needed for drug evaluation. He hoped that advisory committees would upgrade the quality of inputs to the evaluation process. Sadusk visualized open advisory committee discussions, with industry scientific and medical personnel present, along with a representative of the American Medical Association's Council on Drugs, and executive sessions involving only committee members and FDA staff.

Dr. James Goddard, who succeeded Larrick as Commissioner in 1966, suspended the development of these ad hoc advisory committees and reviewed their use. Subsequently, Goddard and Dr. Herbert Ley, initially director of the Bureau of Medicine and then Commissioner, reactivated a program for standing advisory committees in 1967. In August 1967, the agency established a number of committees for drugs, with staff support of one medical officer and one executive secretary provided by the Division of Research and Liaison. One particularly active committee was the Obstetrics and Gynecology Advisory Committee, chaired by Dr. Louis Hellman. The

Anti-Infective Agents Advisory Committee, chaired by Dr. Calvin Kunin, was also active.

Cooper, analyzing the 1960s experience with advisory committees, observed that the dynamic of committee deliberations depended on the personality of the committee chairman, the advance preparation of agenda materials by staff, the significance of the items that came before the committee, and the extent to which FDA staff genuinely wished to obtain independent inputs. These same questions remain salient today, as this report indicates.

In 1969, a new Republican administration took office and placed FDA Advisory committee operations on hold while it reviewed the agency's resources and assessment capabilities. It permitted the Obstetrics and Gynecology Advisory Committee to continue its work, and the committee consequently issued its second report on oral contraceptives in August. In December 1969, Dr. Charles Edwards became Commissioner and, within a relatively short time, reactivated standing advisory committees in the following program areas: anti-infective agents, biometrics and epidemiological methodology, cardiovascular and renal disorders, dermatology, endocrinology and metabolism, food standards, methadone maintenance, neuropharmacology, obstetrics and gynecology, psychomimetic agents (jointly with the National Institute of Mental Health), radioactive pharmaceuticals, and respiratory and analgesic drugs.

This periodic use of advisory committees in the 1960s eventually led to the present prescription drugs advisory committee system, which is currently administered by the Center for Drug Evaluation and Research (CDER). Two other events helped form the system: the Drug Efficacy Study of the mid-1960s and the review of over-the-counter (OTC) drugs in the 1970s. These developments are reviewed below. Importantly, not one of these efforts was required by the Food, Drug, and Cosmetic Act; each was initiated by the FDA as a way to implement key portions of the statute.

The Drug Efficacy Study, conducted by the NAS-NRC at the request of the FDA, began in mid-1966 and concluded in 1969. The study came about because the 1962 drug amendments required, among other things, that drugs approved between 1938 and 1962 on the basis of safety alone be reviewed for their effectiveness as well. Commissioner Goddard, in a March 31, 1966, memorandum to Dr. Keith Cannan, director of the NAS-NRC Division of Medical Science, wrote:

Goddard indicated that the FDA had estimated that there were 3,000 marketed products for which applications had been filed in the 1938–1962 period and perhaps another 1,000 that were being marketed without applications. These products involved only a fraction of this number of chemical moieties; the agency estimated their sum as between 300 and 400, which could be grouped into about 60 categories of therapeutic effect. "These categories," Goddard wrote, "could be combined into ten or twelve groups, each group being appropriate for consideration by a different panel of experts."⁴ Thus, the Drug Efficacy Study began with the judgment by the FDA that expert outside advice was required. Goddard saw no conflict between the NAS-NRC effort and existing FDA advisory groups. The latter, he wrote, "are not equipped to undertake a task of this magnitude and cannot be expected to alter their other activities to the extent that would be required.''⁵

What is also interesting in retrospect is that Goddard also wrote that no problems were anticipated with conflict of interest:

The NAS-NRC responded favorably in April, and the two parties signed a contract in mid-June. Indicating the agency's eagerness to get under way, a May 1 FDA press release announced the study:

The NAS-NRC proposed that a Policy Advisory Committee be established to develop guidelines for the review panels, of which there were to be approximately 30.⁷ The committee consisted of 29 members and the chairmen of an initial 27 evaluation panels.⁸ This body met in July with medical and pharmacy professionals, the pharmaceutical industry, and the FDA, and generated procedural guidelines for submission of data on drugs under review. The 27 panels (later increased to 30) were staffed by 10 Public Health Service physicians assigned to the NAS-NRC effort by the FDA. Each panel had 6 members.

The Policy Committee, in consultation with the appropriate chairman, assigned drugs to a panel for review. Each panel was asked to designate a reviewed drug as effective, probably effective, possibly effective, or ineffective. The factual basis for these determinations was to be information submitted by sponsoring firms, the medical literature, information supplied from FDA files, and the experience and judgment of panel members.

According to a 1968 report, 237 firms submitted a total of 3,637 drug preparations for review;⁹ a later report put the figure at 2,824.¹⁰ According to the latter, most were prescription drugs, but about 15 percent were over-the-counter products; two-thirds were single-entity drugs, the rest were combinations. The panels completed their work in 1968, and in 1969 the NAS-NRC submitted reports to the FDA on more than 2,800 drugs. Each panel, it was estimated, reviewed approximately 150 drugs.

Although the reports of the Drug Efficacy Study were only advisory, in the sense that FDA retained both the authority to disagree and the responsibility for all implementing decisions, they were often decisive in the agency's decision making. Implementation, however, required that the FDA formally accept the study's recommendations; if it decided to withdraw approval for a drug, it was obliged to announce its plan to do so and afford the sponsor an opportunity to respond.

How was the NAS-NRC Drug Efficacy Study received? In retrospect, its reception appears to have been mixed. Apparently, the study earned some respect from the pharmaceutical industry, in part because the expeditious review contrasted with current industry concerns about the length of FDA reviews of new drug applications. However, because the study recommended the withdrawal of some products from the market, litigation resulted. This generated advice to the FDA that the recommendations of the advisory panels not be accorded undue weight.¹¹ For example, Warren Whyte, senior attorney for Abbott Laboratories, wrote:

In a similar vein, in 1971, Rodney Munsey, then associate general counsel of the Pharmaceutical Manufacturers Association, offered the following judgment:

Although the Drug Efficacy Study dealt primarily with prescription drugs, the NAS-NRC panels also considered 420 over-the-counter drugs (out of a total of 3,500 drugs reviewed).¹⁵ In 1972, the FDA faced the mammoth problem of reviewing all of the OTC drugs that had been marketed between 1938 and the enactment of the 1962 Drug Amendments. Its solution was to establish an OTC review system. This review focused on ingredients, not finished individual products, and on assessing those that were "generally

recognized as safe and effective" (GRAS/E) and that were not misbranded. The regulatory product of this process was a series of monographs consisting of approved active ingredients, labeling, and other general requirements.

A regulation promulgated in May 1972 described the administration of this OTC review system, which included four phases.¹⁶ The critical first phase was the review of OTC ingredients by expert panels responsible for specific product classes (e.g., analgesics). This was followed by publication of a proposed monograph on the basis of the panel recommendations; publication of a tentative final monograph, based on the agency response to comments on the proposed monograph; and the issuance of a final monograph.

To launch the first phase, the FDA established 17 panels of expert advisers to review the literature, data, and studies that applied to the labeling and active ingredients for 27 categories of OTC drugs. Its intent was to give credibility to the OTC review by having outside experts bring their independent judgments to bear on the safety and efficacy of every active ingredient in the OTC marketplace. A related objective in establishing the panels was to handle a heavy workload, for which the FDA staff were limited in number and scientific competence.¹⁷ Indeed, in this instance, the basic role of the FDA staff was to administer the OTC panel process.

Each panel had three types of members: seven scientific voting members, one nonvoting consumer member, and one nonvoting industry representative. The scientific voting members included physicians, pharmacologists, and toxicologists from both active medical practice and academia; for the most part, however, these individuals were subspecialty academic physicians, depending on the drug category. Approximately 185 experts served on these panels, assisted by another 74 experts. All served as special government employees.¹⁸ The OTC Drug Division managed all of these panels; in essence, this was its major function.

Baumgartner's history of the OTC review concluded that the FDA grossly underestimated the size and complexity of the panel phase, which lasted 10 years. The panels reviewed, by therapeutic category, 722 individual active ingredients that had 1,454 active uses in the hundreds of thousands of marketed OTC drug products; in the process, they evaluated more than 14,000 volumes of submitted data and other scientific materials. Collectively, the panels met more than 513 times on more than 1,050 calendar days, and their deliberations spanned an average of 4.5 years each.

This system, which was heavily influenced by Peter Barton Hutt, FDA Chief Counsel at the time that it was established, articulated some general principles of FDA advisory committees. Such committees should include nonvoting industry and consumer representatives, for example, to increase the likelihood that the results of the review would be accepted in both

quarters. The participation of the former helped to avoid surprising the industry, to maintain contact with it, to detect problems early, and to minimize opposition.

The OTC review advisory committees operated under the Federal Advisory Committee Act (adopted in 1972), federal conflict-of-interest statutes, and, before the review was finished, the Government in the Sunshine Act. This environment of openness and public scrutiny was very different from that of the Drug Efficacy Study and much more characteristic of our present period.

In the early 1970s, following the episodic efforts of the prior decade, an advisory committee system evolved for prescription drugs. Its purpose was to secure expert advice on the evaluation and approval of new therapeutic products. A number of controversial cases influenced the design of this system. One concerned the drug Promalin,^* for which the Bureau of Drugs established an ad hoc committee to review the data about its approval. The expert committee heard presentations from industry, the American Academy of Allergy, and the FDA staff.

Dr. Charles Edwards, then Commissioner of Food and Drugs, sat through this entire meeting. Afterward, he instructed the FDA staff to create similar advisory committees for all drug areas. His main aim was to ground the regulatory process in the mainstream of science; he also hoped to generate some protective cover for agency decisions and to establish a sound review process.¹⁹

The purpose of the committee system that was established was to advise the FDA on drug approval decisions and on guidelines for drug development in order to bring credibility to approval decisions and to enhance their quality. Advisory committees served both as consulting groups and as open forums for discussing controversial issues, a function that the agency considered particularly important.

The FDA used internal memoranda to create these prescription drug advisory committees administratively. Subsequently, the agency promulgated general regulations governing the formation and operation of advisory committees, which are now codified in (21 CFR 14). In general, from one to three committees advised each of the six to eight FDA prescription drug products divisions. The advisory committee system that emerged during this

period fulfilled the continuing function of reviewing new drugs, unlike the backlog-clearing task of the OTC review.

Management of these committees has not always been without controversy, as the Fountain Committee report of 1976 (see Chapter 4) makes clear. Nevertheless, in these early years, the FDA did learn the importance of asking precise questions, of formulating fixed agendas, and of reaching closure on what the committee actually thought—usually by soliciting a vote. Committee members later felt that they had been well or badly used as a function of how well the FDA performed these tasks.²⁰

The FDA created a system of standing, rather than ad hoc, committees so that committee members would see the fruits of their labor.^* Terms for advisory committee members were four years but were often shortened by such factors as slow appointments and early departures. Committee members were primarily academic physicians, although it soon became clear that other expertise was also needed. As a result, most committees have had a statistician, some have an epidemiologist, and occasionally a committee has a toxicologist member.

In the late 1970s, the FDA added consumer representatives to its prescription drug advisory committees. This arrangement did not work well, because such representatives were frequently at a scientific disadvantage in discussions with other committee members. Subsequently, the agency began to appoint technically qualified, consumer-nominated members to drug advisory committees, which provided the FDA with qualified experts with personal ties to the consumer community. (Chapter 5 discusses this topic at greater length.)

In the Drug Efficacy Study, the FDA turned to the NAS-NRC to obtain access to independent expertise through an organization whose modus operandi was expert committees. In doing so, it sought to clear a large backlog of work—for which it lacked adequate staff—in a short period of time. It learned in implementing that study's recommendations the practical limits of time and resources, and the consequent inability of expert advisory panels to review all relevant data. They agency learned as well that it was not relieved of the responsibility to decide issues. Most important, however, the

study established the pattern within the FDA of seeking outside expert advice on major decisions.

Several noteworthy differences between the Drug Efficacy Study and current FDA practices can be noted. For example, having charged the NAS-NRC, the FDA turned the review task over to the NAS-NRC; it had no role in selecting panel members, and the composition of panel membership originally was not disclosed. The panels met in secret, and the frequency and location of their meetings were not known. Indeed, the FDA at first was not willing to disclose the study reports. In addition, the federal conflict-of-interest requirements were not applicable, and the NAS's own procedures were not very stringent by today's standards.

The FDA set up the OTC review panels to bring independent expert judgments to decisions about the safety and efficacy of over-the-counter drugs. A key objective of the review was to generate outcomes that the agency could enforce. A secondary justification for the action was that the panels were needed to take over a workload that was too great for the FDA staff to handle and that it was not competent to handle.

The prescription drug review process evolved to assist the FDA in its continuing work of reviewing new drugs. As such, it departed from the large-scale, one-time efforts of the Drug Efficacy Study and the OTC drug review. The drug advisory committees were intended to ensure the scientific soundness of the agency's regulatory decisions, to establish the credibility of the process, and to provide a way to air controversial issues.

In each of these efforts, the FDA initiated the use of advisory committees. The Drug Efficacy Study suggested a model, and the OTC review moved advisory committees under the direct management of the FDA. Prescription drug review activities incorporated advisory committees into the continuing operations of the agency in its review of new drugs.

The federal government regulated biological products intended for human use even before it began to regulate drugs and medical devices. In the United States, biologics regulation dates to the Virus, Serum, and Toxins Act of 1902 (later expanded and consolidated in the Public Health Service Act). Drug regulation began some four years later with passage of the Food and Drug Act in 1906; it was extended substantially in 1938 and then again in 1962. Comprehensive device regulation did not receive explicit statutory underpinning until 1976.

The biological products covered by the 1902 act were those that had come into general use by that time—that is, viral vaccines, bacterial vaccines, antitoxins and toxoids. (It is of some historical interest that the Biologics

Act also included arsphenamine and its derivatives and other trivalent arsenic compounds, because at that time there was no existing separate statutory provision for the regulation of drugs.) Later, as blood and blood fractions and allergenic extracts came into medical use, they, too, were regulated as biologics. Thus, Section 351 of the Public Health Service Act, as currently amended, defines biological products as ''any virus, therapeutic serum, toxin, antitoxin, vaccine, blood, blood component or derivative, allergenic product, or analogous product, or arsphenamine or its derivatives (or any other trivalent organic arsenic compound), applicable to the prevention, treatment, or cure of diseases or injuries of man."^*

The main use of early biologics was to control communicable diseases, a public health purpose logically pursued by the government. Indeed, the development and regulation of biological products are embedded in a public health rationale and public institutional context. As a result, in the early years of this century, national governments, states, and sometimes even cities were involved in the production and use of biologics through public laboratories. No equivalent of today's industrial capability in biologics, biotechnology, or drugs then existed; one public health task was to create such a capability. These public health origins of biological products give a distinctive cast to this area of regulation.

In the United States, responsibility for regulating biologics was initially assigned to the U.S. Public Health Service and to the laboratories that later came to be the National Institute of Health (NIH). In the post-World War II National Institutes of Health, biologics regulation was originally part of the Microbiological Institute; the NIH created a separate Division of Biologic Standards (DBS) in 1956.

The NIH years of the biologics program left several legacies. First, the present-day Center for Biologics Evaluation and Research (CBER) is a regulatory unit deeply embedded in a medical science research organization. Its professionals engage in both the regulation of biological products—inspecting manufacturers and sampling vaccine production lots—and scientific research dealing with biological products, often in collaboration with the laboratories of other public health agencies. The aims of this research are to improve existing products and, in some cases, to explore the feasibility of developing new ones. Other tasks include the development of new tests for determining safety and efficacy and the production of laboratory reference standards.

On occasion, the FDA's biologics organization has also conducted clinical trials of either experimental or licensed products (sometimes in collaboration with other NIH components or the Centers for Disease

Control). Its research and regulatory decisions are tied to similar programs in the biologics control laboratories of other nations through the World Health Organization (WHO) and its Committee on Biological Standards, as well as through the various WHO working groups that report to the Committee on Biological Standards.

This extensive involvement with the public health community and the scientific research community created a climate favoring the seeking of scientific consensus in regulatory decision making. For example, the lengthy development of a major new product, such as a vaccine, includes the organization of numerous national and international scientific workshops and conferences. The biologics regulatory organization typically was the organizer of these conferences, often with joint sponsorship by interested components of other public health agencies here and abroad. Although these conferences did not provide advice on specific regulatory decisions, they contributed data and discussions that often influenced the development of requirements for the product and, in the process, generated international consensus about standards. ^*

During the years in which the biologics regulatory program was a component of the NIH, ad hoc advisory committees were sometimes formed to deal with matters of high public visibility, such as a major new product that was being considered for approval or an important problem that occurred with an existing marketed product. The membership of these committees generally included experts from government as well as the academic community. Although the work of a committee might extend over a period of months or years, its charge was limited to advising on the particular product or problem under review. Committee members would routinely participate in any pertinent workshops or conferences and be familiar with laboratory programs of the DBS. Thus, they acquired great understanding of the issues before the committee. Ad hoc committee meetings were closed to the general public. Initially, these committees reported to the Surgeon General of the Public Health Service; later, they reported to the Director of the NIH.

Historically, manufacturers of biological products have submitted license applications as "rolling submissions"; that is, they submitted portions of the application during the entire development process, with supplements added as additional studies are completed. Consequently, unlike most new drug applications, an application for a biologic typically was reviewed in its parts long before it was formally submitted in complete form. Thus, any ad hoc committee would usually meet on several occasions to review a product (or

group of similar products) under development before all of the studies needed for licensure had been completed. By the time of licensure, the committee would have had ample opportunity to become familiar with all research data relating to the product and would have participated in the development of the regulatory requirements for it. The Surgeon General's ad hoc live poliovirus vaccine advisory committee, for example, had many meetings and participated in a number of conferences over an extended time before the licensing decision arose. During this time, both the committee and the agency's scientific staff had formed close working relationships with the research personnel of the potential manufacturers of the vaccine, as well as with other scientists in the national and international public health communities involved in poliovirus vaccine research.

In this same period, licensing applications for lesser biological products usually received no outside attention and were reviewed and acted on solely by the staff of the biologics regulatory organization. Thus, the ad hoc committees had substantial involvement with certain high-visibility products, but they provided no comprehensive oversight of the overall regulatory program or of biologics development generally.

The other use of advisory committees by the biologics program that developed at the NIH and has been extended to the FDA was in the review of intramural research. (No comparable function exists for advisory committees of the other two FDA centers, the centers for Drug Evaluation and Research and for Devices and Radiological Health.) Just as each NIH institute had one or more standing committees for the review of intramural research, the DBS had its "board of scientific counselors," experts who reviewed the quality and appropriateness of intramural biologics research as well as the qualifications of individual scientists and made recommendations to the Director of the DBS. The DBS specifically charged the review committee not to review regulatory matters. The current CBER advisory committees continue this tradition.

In the early 1970s, the Division of Biologic Standards at the NIH came under fire from Senator Abraham Ribicoff's Subcommittee on Government Operations. The DBS and the NIH management did not fare well in the ensuing controversy. This embarrassed the Nixon administration, which inspired the administrative transfer of the biologics regulatory function from the NIH to the FDA. Thus, in the summer of 1972, the DBS was transferred from the NIH to the FDA, and renamed the Bureau of Biologics. The intention of the move was to strengthen the division's regulatory role and dilute the public health emphasis. The FDA itself had become a component of the Public Health Service in the 1960s. Yet despite its transfer to the FDA, many of the CBER's present policies, procedures, and practices stem from its years within the NIH. For example, CBER headquarters and its

principal laboratories continue to be located on the NIH campus, and it has maintained extensive research contacts with the NIH.

The transfer was followed by a number of management changes in the biologics organization and a decision to reexamine the efficacy of all existing licensed biological products. At that time, there were no standing committees for biologics regulatory decisions, and the ad hoc committees that were involved had a narrow focus. To carry out its regulatory functions, the bureau created a process similar to the comprehensive OTC drug review and formed six standing committees to review the principal categories of biological products: major vaccines, bacterial vaccines, blood products, and products for which the science base was substantially less, such as allergenic extracts.

Each committee had approximately seven scientifically qualified voting members who were expert in the particular area under review. In addition, the bureau chose a nonvoting ''consumer" representative and a nonvoting "manufacturing" representative for each of the panels. These committees had FDA staff support and were responsible for reviewing licensed products and making recommendations to the Commissioner about those that did and did not meet contemporary standards of efficacy. (Safety was considered indirectly because it bore a relation to efficacy.) The process, which was quite similar to the OTC review, began in late 1972 and early 1973 and took several years.

The scale and scope of the review were substantial in both administrative and logistical terms. As the committees began their work, it became apparent that these same experts could be helpful to the agency in other ways: giving ongoing advice about new products (assuming the role of earlier ad hoc committees in this regard); advising on general problems that occurred with both marketed and experimental products; and reviewing intramural research similarly to the function of the NIH DBS Board of Scientific Counselors. As the agency completed its one-time comprehensive reviews of existing products, it reduced the number of these original committees and rechartered them to provide continuing advice on all of the organization's regulatory and research programs.

In 1982, the biologics bureau was consolidated with the Bureau of Drugs into the Center for Drugs and Biologics. This arrangement lasted five years; the FDA separated the two units in 1987 into the Center for Biologics Evaluation and Research and the Center for Drug Evaluation and Research.

The use of advisory committees by the Center for Devices and Radiological Health (CDRH) differs from that of the CDER and CBER in

one critical aspect: it is required by statute. The Radiation Control Act of 1968 mandated the establishment of the Technical Electronic Products Radiation Safety Standards Committee (see the later discussion), which was the first acknowledgment by Congress that advisory committees could be useful in the administration of food and drug law. It was the Medical Device Amendments of 1976, however, that required the extensive use of advisory committees as an integral aspect of the device regulation authorized by that act.

In an October 30, 1969, message to Congress on protecting the interests of consumers, President Richard M. Nixon called for "certain minimum standards" for medical devices and declared that "the government should be given additional authority to require premarketing clearance in certain cases [of medical devices]." In response, the Secretary of Health, Education, and Welfare (HEW) appointed a committee to study the regulation of medical devices and to recommend a legislative program to implement the President's message.

The Study Group on Medical Devices of the Department of HEW issued its report, Medical Devices: A Legislative Plan, in September 1970. ²¹ This committee and its report were known as the Cooper Committee and Cooper report, respectively, after the chairman, Dr. Theodore Cooper, then Director of the National Heart and Lung Institute of the NIH. Among the committee members were Dr. Charles Edwards, Commissioner of the FDA, and Dr. Mark Novitch, then Special Assistant for Pharmaceutical Affairs, Office of the Assistant Secretary for Health and Scientific Affairs, and later Deputy Commissioner and Acting Commissioner of the FDA.

The committee's charge was to recommend procedures for establishing standards for certain medical devices and for the review and regulation of other devices before marketing. It found three issues central to a sound legislative proposal: (1) an immediate and systematic review of all devices "available and in use" in order to group them in one of three categories—those that should be exempt from standards and premarketing review, those for which standards should be established to ensure "safety and reliability," and those requiring premarketing review; (2) delineation of an acceptable plan "for assuring expert scientific review of the safety and effectiveness of medical devices at the clinical application phase'' and before marketing; and (3) defining the government's role in standard setting and enforcement.²² Regarding classification, the report recommended that "appropriate scientific, health, and engineering experts" be organized to conduct the initial review of existing devices and to advise on their

classification. For premarketing review, it recommended the establishment of "standing permanent advisory scientific review panels ... to assist in the review of data on new medical devices and to advise the Secretary of their safety and effectiveness." In addition, the report recommended that the Secretary be granted authority to certify existing standards or establish new ones and to audit manufacturers for compliance. It noted that prior experience with the initial and long-term safety and effectiveness of medical devices was often limited, that device development was dynamic, that hazards arose from use as well as from design and manufacture, and that existing data were quite inadequate.

Although various private organizations had tried to provide manufacturers with standards in specific areas, these efforts were poorly financed and coordinated, were not comprehensive, and lacked the force of law. The FDA was the only government agency with authority to regulate all medical devices, but its authority was limited to preventing misbranding and adulteration. Although the courts had recently upheld the agency's efforts to regulate certain devices through drug premarketing controls, the scope of this authority had not been clearly defined. Manufacturer concerns for product liability further underlined the need for "a system of device regulation."

Consultants to the Cooper Committee agreed that "the public deserves more protection," but they regarded drug regulation as an inappropriate model that was likely to inhibit innovation if applied to devices. However, the approaches favored by the consultants ranged from self-regulation to insistence on Good Manufacturing Practices to premarketing notification to premarketing approval, and did not reflect consensus.

The report noted "that a system of 'peer group' review of scientific data would induce confidence within the medical device community that decisions related to devices and standards were soundly based." The basic logic regarding the use of outside experts, whether for classification, standard setting, or premarketing review, was spelled out as follows:

To summarize, the Cooper report recommended that three categories of devices be established—those "so well recognized as safe and effective" that neither standards nor premarketing approval was needed, those that could be regulated by standards, and those "new and unproven critical devices that are at the leading edge of technological innovation and biomedical explorations" and that required premarketing review. For the latter, it recommended that the Secretary be authorized to establish standing advisory panels ''generally patterned along clinical sub-specialty lines, composed of appropriate experts from the physical and biological sciences, engineering, medicine, and dentistry qualified to evaluate medical devices." These panels should be permanent, meet regularly, have appropriate regulations governing conflicts of interest, and advise the Secretary about the acceptability of each device for clinical application or marketing. The report also recommended that a Medical Devices Advisory Council should be available to the Secretary for policy issues related to medical devices and should include experts in device development and representatives of manufacturers, users, and patients.

On the basis of the Cooper report's recommendation, the FDA initiated an inventory of medical devices and began classifying medical devices that were already on the market.²⁴ Two pilot panels—one for orthopedics and another for cardiovascular devices—met in November 1971 to develop a system for classifying devices, and by late 1972, the FDA had indicated its intention to classify all devices over the next 12 to 18 months with the aid of outside expert panels.²⁵ The agency also announced that it was initiating efforts to develop device standards.

In the period following the Cooper report, a broad consensus developed regarding the need for increased legislative authority over medical devices. The FDA authority to regulate drugs as devices was cumbersome, time-consuming, and inadequate. The consensus was stimulated in part by manufacturers seeking to alleviate their concerns about product liability.^26*

The legislative process moved slowly, culminating in new legislation six years after the Cooper report. In early 1976, the House Committee on Interstate and Foreign Commerce issued its report on pending device legislation. It reviewed the contents and recommendations of the Cooper report in detail. In addition, it approvingly noted the several references of the Cooper report to "peer review" groups for review of scientific data about devices, consisting of representatives from industry, the federal government, the academic community, and other concerned organizations, including consumers.

The House Commerce Committee, drawing on the Cooper report recommendations, "and in an effort to afford the [FDA] the best possible scientific advice and expose the agency's decisions to public scrutiny," drafted a bill that relied heavily "on the proceedings of experts, with ultimate authority vested in the Secretary." This proposed legislation provided for classification panels, which were to use the prior FDA classification panel efforts and review them "in light of the statutory classification criteria." It also provided for establishing advisory committees for product approval evaluation.

The Medical Device Amendments, which were enacted in late 1976, were heavily influenced by the Cooper report. They required the creation of advisory panels or committees for two purposes. The first was the classification of medical devices (Section 513). Following adoption of the amendments, the FDA revisited the classification process in accordance with the act. The new classification advisory panels had the benefit of prior efforts, which were a useful point of departure.

The second purpose was the evaluation of medical devices regulated by risk tier. Class I devices required no standards or premarketing review. Class II devices were to be regulated by performance standards (Section 514). Class III devices required premarketing approval (Section 515). Given that no performance standards have been issued by the FDA since the 1976 amendments, it is chiefly the premarketing approval process of Class III devices that concerns the IOM committee.

For product evaluation, the amendments called for establishment of permanent advisory committees. To these advisory committees the FDA was to appoint "persons qualified in the subject matter to be referred to the committee and of appropriately diversified professional backgrounds"; the agency was also mandated to appoint a chairman and provide necessary clerical support. The performance standards section called for the appointment of nonvoting consumer and industry members; this provision was omitted from the premarketing approval section, but the practice was

adopted for the product review committees nevertheless and is the basis for current policy.^*

Although the Safe Medical Devices Act of 1990 modified the original 1976 legislation in certain respects—giving the agency greater discretion in the use of advisory committees—it did not change the basic mandate to use such committees. For example, the 1976 amendments required the FDA to bring all pre-market approvals (PMAs) to advisory committees; now it has discretion to do so after "the fourth device of a kind" has been approved.

In addition to the requirements of the medical device amendments, the Center for Devices and Radiological Health administers the 1968 Radiation Control for Health and Safety Act, which is a section of the Public Health Service Act. Congress enacted this legislation to protect the public against the dangers of electronic product radiation through the development of performance standards for controlling the emission of such radiation. Implementation of this law required the creation of the Technical Electronic Product Radiation Safety Standards Committee (TEPRSSC).

In 1990, the CDRH rechartered its advisory committees into a single Medical Device Advisory Committee with a number of panels. Concurrently, the agency implemented the combination products requirements of the Safe Medical Devices Act by issuing regulations on product jurisdiction (which encompassed combination products) and negotiating three intercenter agreements on this subject. The CDRH rechartering of its device advisory committees converged with these product jurisdiction efforts and led to a rechartering of CDER and CBER advisory committees as well. These developments are reviewed in greater detail in Chapter 3.

It is worth contrasting FDA advisory committees with the study sections of the National Institutes of Health, if only because academic medical scientists are very familiar with the latter and often extrapolate these experiences to the operations of the former. The functions of the two sorts of committees, and the contexts in which they operate, are quite different. The pertinent distinctions are these:

• The 21 original NIH study sections were announced in late 1946; they had clear policy purposes and relatively simple administrative needs related to the review of extramural research grant proposals by academic scientists and the awarding of grants.²⁷ As discussed earlier, FDA advisory committees have a more recent history, are used for multiple purposes, and lack the academic science constituency of the NIH study section system.

• An NIH study section is responsible for reviewing a number of research grant proposals in their entirety. In contrast, the FDA review of a new drug application (NDA), a product license application (PLA), or a pre-market approval (PMA), involves a larger, more complicated application. Moreover, such a review is the legal responsibility of the FDA and, as a practical matter, cannot feasibly be performed by an advisory committee.

• A research proposal may take several days to review; an evaluation of a product application (NDA, PLA, or PMA) requires months or even years.

• A study section reviews only documents and hears no presentations, but an FDA advisory committee hears sponsor and FDA presentations and sometimes other experts.

• FDA medical reviewers develop detailed knowledge of a given application and often firm judgments regarding it, based on their own views of what constitutes good science and good medicine. (FDA executive secretaries in drugs and biologics are unlikely to develop an independent view of the merits of a proposal; in devices, they often are also proposal reviewers). By contrast, although NIH executive secretaries may have a sense for good science, they are not formally responsible for more than the efficient performance of the study section.

• FDA professional staff incur few costs if they hold up an application but confront great pressures to reach the "right" decision. In contrast, it is expected that all research proposals received by the NIH by a certain date will be reviewed at a specified time and that the study section will determine what is "right" in terms of scientific merit, technical feasibility, budget, and priority.

• Although medical scientists care intensely about study section reviews, society is largely uninterested in which scientist gets what research dollars. By contrast, society is not indifferent to the recommendation of an FDA advisory committee on the public availability of a medical technology (as evidenced by the press and television coverage of such meetings).

Thus, as will be clear throughout this report, FDA advisory committees differ in important policy, administrative, and operational respects from the more familiar NIH study sections.

A number of themes can be identified in the historical development of the use of advisory committees by the FDA. Among the most prominent are the following:

• FDA leadership took the initiative in introducing advisory committees for the review of drugs and biologics; these committees were not imposed on the agency by the Congress. The device amendments, which do require such committees for evaluation of medical devices, constitute congressional recognition of their importance for the agency.

• In the case of the Drug Efficacy Study, the OTC drug review, and the biologics review, these committees fulfilled major workload functions—clearing a backlog of work for which the agency lacked adequate personnel—and provided independent expertise. Over time, they came to play a central role in the assessment of new products and technologies.

• The use of advisory committees in the biologics program involved the review of intramural research programs and personnel, in addition to product evaluation. In addition, in vaccine development, the committees were used throughout the product development cycle and participation included scientists from other units of the Public Health Service as well.

• For medical devices, the anticipated uses of advisory committees were product classification, standard setting, and review of new products. The initial classification panels did their work and were not continued; the function is now fulfilled by product review panels. The standard-setting function did not develop as anticipated and committees, save the Technical Electronic Product Radiation Safety Standards Committee, were not used for this purpose. The primary purpose of device advisory committees today is the review of new products.

• In the FDA setting, advisory committees play a supportive role to agency professionals. They do exercise great influence in agency decisions, but their role as advisers is more literally so than is the case for NIH study sections.