In recent years, some welcome FDA reform has occurred, particularly with the PDUFA of 1992 and the 1997 Modernization Act. The significant component of these reforms is the partial shifting of FDA funding from general appropriation to “user fees” for FDA services. FDA funding is now tied to NDAs, so the FDA has managed to approve drugs more rapidly because such promptness induces more NDAs and more “user fee” revenues for the FDA.
We believe, however, that the health and well-being of Americans and others worldwide would be best served by more significant reductions in FDA powers. In these matters, we favor adult freedom and hence the repeal of all forms of premarket approval. We believe that nongovernmental parties and tort law should generate and administer all of the rules, standards, institutions, and practices that make up drug affairs. We describe how such a system would work in the section below called The Sensible Alternative.
Freedom may be the North Star, but in the American system decontrol comes slowly and gradually, when it comes at all. Drug affairs are now fettered by numerous restrictions. Relaxing or removing any of the significant restrictions would help. The following subsections suggest reforms that fall short of the full freedom of what we consider the sensible alternative but nevertheless would substantially improve on current policy.
- Improve Consumer Information and Control
- Create a Field of Drug-Certifying Bodies
- Implement International Reciprocity
- Drop the Proof-of-Efficacy Requirement
- Make FDA “Approval” Voluntary
Traditionally, patients have been expected to be patient while doctors and other elites make treatment decisions on their behalf. FDA paternalism, therefore, has been part and parcel of the paternalism of American medicine. The latter paternalism has a number of social and historical roots but is also tied up with the professionalization and regulation of American medicine (see, e.g., Starr 1982). In recent years, as competition in the market for health services has increased, consumers have become more informed about health matters and more willing to question their doctors and participate closely in their treatments. Yet consumers remain uninformed in many respects, and, unfortunately, the FDA has tended to resist the trend toward greater consumer information and control. The FDA, for example, tried for decades to suppress the use and sale of vitamins and other dietary supplements such as St. John’s Wort (for more, see the History section).
In the tradition of American medical paternalism, drug information has tended to flow hierarchically from the FDA to the physician and then (perhaps) to the consumer. Physicians have supported this hierarchical information structure because it places them in control of the highly profitable information gateway (Benham and Benham 1975). The FDA has, for the most part, supported or acquiesced to this hierarchy. It has used its power over labeling, for example, to create the class of prescription-only drugs (see the History section). Prescription-only drugs did not need to carry any consumer labeling except for the warning “Caution: to be used by or on the prescription of . . . ” (3 Federal Register 3168, Dec. 28, 1938). Thus, product labeling evolved so that it became common for consumers to be given potentially dangerous drugs with no written information other than the doctor’s instructions.
Even today, consumers typically do not receive a copy of the product label with their prescription. If they are lucky, their pharmacist may give them a computer-generated synopsis of warnings and contraindications. Consumers can and should ask for product labels with their prescriptions. The FDA should also pay more attention to designing labels that can be easily read and understood. At present, product labels are difficult to read, badly organized, and poorly formatted. Gieringer (1984) argues that “the organization of warnings and risk analysis would seem to deserve a sophisticated effort at risk analysis . . . this could be done through some form of dedicated labeling agency” (121).
Aside: Questioning prescription-only restrictions.
Restrictions on direct-to-consumer advertising is another example of the ways in which the FDA has supported the hierarchical transmission of information. Drug advertising on television has only recently been allowed, a good fifty years after the invention of that medium.
A minor reform in the direction of improving the transmission of information would be to return control over drug promotion and advertising to the FTC (which had it until 1962). The FDA often acts unreasonably in regulating promotional literature and advertising. Drug companies fear the FDA and refrain from challenging its edicts because it has a stranglehold on the whole industry, and no company wants to get on its bad side (Volokh 1995). Separating NDA approval from advertising approval would increase the willingness of firms to stand up to unreasonableness. In addition, since at least the 1980s, the FTC has been more respectful than the FDA of the consumer’s ability to filter and grade information from advertising claims. The FTC, for example, was instrumental in allowing food manufacturers to make substantiated health claims regarding food, with notable improvements in the health of consumer food products as a result (Ippolito and Mathios 1990).
A large improvement would be to allow “Not FDA Approved” claims. Under a split-label regime, the product label would consist of a part for FDA-approved health and nutrition claims and a part for “Not FDA Approved” claims. Although retaining FDA certification for those who want assurance from the FDA, the split-label approach increases the amount of information available to the consumer. Recall that many patients are prescribed drugs for off-label uses. By definition, FDA labeling contains no information specific to off-label uses, and thus many patients are not well served by the FDA label. (In addition, the well-known Physician’s Desk Reference [PDR] is only a compilation of FDA labeling; the U.S. Pharmacopoeia Drug Information (USPDI) and its consumer version, Complete Drug Reference, are better sources of information than the PDR). A split-label approach would mean that all of a drug’s common uses were described on the label. Manufacturers of aspirin, for example, would be allowed to state its uses as a prophylactic for first heart attacks on the “Not FDA Approved” portion of the label. The same idea could also be applied to advertising claims.
A split-label approach now exists, following the 1994 passage of the DSHEA, for claims made on behalf of vitamins, minerals, and other dietary supplements. Dietary supplements can now make “nutritional support” claims but not “drug” claims so long as they carry the warning, “These statements have not been evaluated by the Food and Drug Administration.” The split-label approach would extend the idea of a split label to all drugs and medical devices.
In September 2003 the FDA further liberalized by beginning to accept applications for health claim labels on food items based on scientific evidence provided to the FDA using a grading scale of A-B-C-D, where “A” represents “significant scientific agreement” and “D” represents “little scientific evidence supporting this claim.” The scheme potentially allows qualified health claims that have not met the FDA’s efficacy requirements. Allowing ranked health claims is a liberalization, although how effective the system works in practice will depend on the FDA—it may be better to refer health claim adjudication to an independent panel of experts.
As of 2008 the FDA may deny seriously ill patients—those facing morbidity or death from their illnesses—the use of unapproved drugs. This FDA intervention has resulted in incidents where physicians were unable to provide experimental or unapproved treatments despite the terminal nature of their patients’ illnesses. Members of the U.S. Senate have proposed the “Access, Compassion, Care, and Ethics for Seriously Ill Patients Act” or “ACCESS” bill to Congress that would amend the Federal Food, Drug, and Cosmetic Act to provide seriously ill patients and their physicians the freedom to choose these riskier treatment options and to protect drug companies from any subsequent lawsuits.
The bill would provide this through a number of reforms to the FDA, though it would retain pre-market approval authority over both the drugs used in the program and the applications for use of the approved investigational drugs. Reforms included in the bill include:
- It would declare that the use of investigational drugs or combinations of drugs for treatment is the responsibility of the physician and patient rather than the FDA. Drug companies would have to apply for their drugs to be categorized under the “Compassionate Investigational Access” program, but then would be able to provide them to seriously ill patients.
- It would require the FDA to include at least two patient representatives on the FDA committees responsible for evaluating applications related to drug uses relevant to the Act. This is intended to increase the chances of these drugs being approved for use in the program by the committees.
- Drug companies providing drugs through this program would be immune to prosecution for any claims resulting from their use by the patients or physicians.
- The Medicare program would be expanded to provide coverage for any drugs approved under the Compassionate Investigational Access program.
- The FDA would be required to give consideration to non-statistical measures such as clinical evaluations and case history reports, data from animal and computer models, comparison with historical data, and evaluations of the adverse effect of delaying the availability of an investigational drug to even a small subpopulation of seriously ill patients.
The bill includes a number of quarter measures that, while an improvement over current FDA policies, would not substantially reduce drug lag or drug loss. The key to reducing drug lag and drug loss is to reduce the costs of researching and developing new drugs by limiting FDA regulation.
Henry Miller recently developed an intermediate reform proposal in which drug development and application would be overseen by nongovernmental drug-certifying bodies. They would compete with one another for hire by companies developing a new drug. The hired drug-certifying body would oversee investigation, help develop the NDA, and then make an initial decision on the application—that is, decide whether to certify the drug. The European agencies would also be permitted to serve as drug-certifying bodies. The company and its certifying body would then go together to the FDA for final approval of the new drug. The FDA, therefore, would retain final authority, but would rely on a set of trusted drug-certifying bodies, which would compete to get it right, do it quickly, and keep fees low. Unlike the FDA, the certifying bodies would be liable to litigation for negligence. Under such a regime, according to Miller (2000), the FDA “becomes primarily a certifier of certifiers, rather than a certifier of products” (106).
Once the FDA receives the NDA with the certifying body’s recommendation for approval, it will have ninety days to respond. The FDA would not be able to extend the ninety-day review period, and failure to act will constitute automatic approval at the close of the ninety-day period (a feature reminiscent of the pre-1962 FDA). If the FDA rejects the application, it must provide an explanation that rebuts the presumption given to the recommendation (Miller 2000, 94). Rejection would be open to appeal to “an independent standing advisory committee of experts, located organizationally in the Department of Health and Human Services (DHHS), where the FDA is located, at the level of the office of the secretary (therefore, not reporting to the FDA commissioner). [The committee would] advise the secretary of DHHS, who, it is expected, [would] accept their advice and instruct the FDA commissioner to implement the committee’s decision” (100).
Miller thoughtfully seeks to design a new set of institutions to sustain in the United States a cooperativeness that is more natural to the countries of western Europe. Miller’s plan might vastly improve the situation in the United States. But Miller’s plan gives the FDA final authority over new drugs and devices as well as hegemony over the setting of drug quality and safety standards and the recognition of drug-certifying bodies.
In addition, Miller is explicit in seeking reforms that would lower type 2 errors without increasing type 1 errors. Although all such reforms should be pursued, there is little reason to believe that the current level of type 1 error is best. Indeed, an understanding of FDA incentives suggests that at present type 1 errors are too high relative to type 2 errors, and some trade-off is thus warranted.
If the United States and, say, Great Britain had drug-approval reciprocity, then drugs approved in Britain would gain immediate approval in the United States, and drugs approved in the United States would gain immediate approval in Great Britain. Some countries such as Australia and New Zealand already take into account U.S. approvals when making their own approval decisions. The U.S. government should establish reciprocity with countries that have a proven record of approving safe drugs—including most west European countries, Canada, Japan, and Australia. Such an arrangement would reduce delay and eliminate duplication and wasted resources. By relieving itself of having to review drugs already approved in partner countries, the FDA could review and investigate NDAs more quickly and thoroughly.
EU countries established a limited reciprocity system in 1983. Over time, however, the European system has evolved in a somewhat different direction. The reciprocity principle has two virtues, competition and elimination of duplication. Unfortunately, the EU has begun to subvert the first virtue (Hansen 2000). Today many products must be reviewed by the European Agency for the Evaluation of Medicinal Products (EMEA), headquartered in London. EMEA approval opens the market to all EU countries. Still, the EMEA regime is better than the FDA. For most drugs, the EMEA is not the sole source of authorization (hence, there is competition among the EU drug agencies); it contracts most of its reviews to outside experts; it takes a less adversarial posture; and it places more confidence in the sponsors ’ summary reports (rather than reanalyzing the raw data) (Miller 2000). Nevertheless, there is little guarantee that the system will maintain its current performance, and, besides, being better than the FDA is nothing to brag about. If the EU centralizes authority, the vestiges of competitive governance might give way to monopoly governance.
International reciprocity would eliminate the FDA’s monopoly on drug approval. Under such a system, U.S. drug companies could submit the NDA equivalent to the authorities in partner countries and thereby gain approval in the United States. Thus, the FDA would have to compete for business. It would have to shape up or lose out on the fees that come with NDAs.
Prior to 1962, the FDA screened drugs only for safety. NDAs did not have to demonstrate efficacy. As we have seen, the pre-1962 years were a time of great pharmaceutical advance. Also, we know from the current practice of off-label prescriptions that a panoply of nongovernmental institutions research, mediate, and certify efficacy of drugs, quite apart from FDA efficacy evaluation. Hence, there is strong evidence that private enterprise and tort law takes care of efficacy and that the costs, delays, and drug loss from FDA efficacy requirements are unredeemed. A splendid reform, therefore, would be to drop the proof-of-efficacy requirements in drug approval and return to the pre-1962 principle of requiring only proof of safety. That simple reform would greatly expand the range of drugs developed (in particular for rare diseases), increase the speed with which they get to market, and significantly reduce costs and drug prices.
FDA drug testing and certification might be made optional. Drug companies would be welcome to submit applications for voluntary certification by the FDA. If doctors and patients were assured of quality and safety apart from FDA certification (as they are now with respect to off-label drug uses), they would be free to purchase the drug clearly labeled “Not FDA Approved.” On the other hand, if FDA certification is as valuable as the FDA claims, patients and doctors would avoid non-FDA-certified drugs and rely on FDA certification. Under this plan, the FDA would become a genuinely voluntary institution, much like Underwriters’ Laboratories.
Under such a regime, the image of the FDA would change. Today, when a drug company applies for FDA “approval,” there is an ambiguity about whether the company actually sees value in the FDA’s seal of approval or merely seeks legal permission. Under the proposed regime, the meaning of FDA approval would be unambiguous. Approval would be sought because the company believes the FDA seal of approval to be valuable. The relationship between the company and the agency would be one of integrity, appreciation, and cooperation because the agency would not exercise coercive power over the company.
Voluntary certification would make the FDA one among a set of many certification agencies, a set that might include governmental drug agencies in other countries, the American Medical Association, the Mayo Clinic, the U.S. Pharmacopoeia, and private for-profit or nonprofit firms such as Underwriters’ Laboratories, which would specialize in testing drugs and medical devices. In the section below, The Sensible Alternative, we discuss in greater detail how a voluntary system would provide safe and efficacious drugs.