Abbreviated New Drug Application (ANDA)
The ANDA was established by the Waxman-Hatch Act. Under an ANDA, proof of bioequivalence is enough to satisfy FDA safety and efficacy requirements for a generic drug. In addition, the manufacturer of the generic must certify that the original drug is either (a) not patented, (b) has an expired patent, (c) has a patent expiry date before which the generic will not be sold, or (d) has an invalid patent or a patent that the generic does not infringe. If the manufacturer claims the original patent is invalid or not infringed, there is a thirty-month litigation period during which the FDA cannot approve the generic.
A-B-C-D Health Claim Categories refer to a scoring method by which health claims manufacturers can be measured based on the amount of supporting evidence. Since September 2003 the FDA has accepted applications for health claims from food items using a form of the A-B-C-D system where “A” represents “significant scientific agreement” and “D” represents “little scientific evidence supporting this claim.” Proponents believe this has allowed food producers to advertise benefits that have not met the FDAs proof-of-efficacy requirements. One option for labeling reform is that the FDA could accept similar qualified claims of health benefits from drugs.
As of 2008 the FDA has currently approved 12 health claims, while allowing 16 qualified health claims. The number of petitioned claims is uncertain. As of late 2007, members of Congress were calling for a halt of qualified health claims until a GAO study could be completed and had prompted the FDA to conduct a re-evaluation of a number of pre-existing qualified health claims, but no such halt has taken place.
In 2006 the European Union passed regulation electing to implement health benefit claims on food without a report card but with specific requirements for various claims that will be assessed by the national authority of a member state. It has since provided additional guidance on implementing the regulation.
The implementation of the food health claims categories, in response to a court decision from Pearson v. Shalala requiring the allowance of qualified health claims, has liberalized dietary supplement food sales by enabling producers to advertise and differentiate their products without meeting the FDA criteria for significant scientific agreement. However, the FDA still determines the grade received by a health claim, and therefore retains a form of pre-market approval of these qualified health claims.
Research has suggested that the distinctions as used in the food health claim labels have proven unclear to consumers in testing (IFIC, 2005, Derby and Levy, 2005, Teratanavat, 2005). The most recent GAO study notes that some organizations such as the American Medical Association, have called for the elimination of qualified health claims on the grounds that consumers cannot distinguish between the four existing categories.
After the IND has been filed, a drug that can be used by patients who are suffering from life-threatening or seriously debilitating conditions, for which no other drug treatment exists, can qualify for accelerated development. Approval can be based on surrogate endpoints or on an FDA determination that the drug can be used safely if distribution is limited.
Aspirin has been called the wonder drug of the twentieth century. It is used extensively in the treatment of mild to moderate pain, fever, and inflammation, and has also been discovered to be useful in the management and prevention of heart attacks and strokes (which are off-label uses) and perhaps also of colon cancer and other maladies. For many years, the FDA prohibited manufacturers from advertising or promoting the use of aspirin for heart attack victims, a decision that was responsible for thousands of needless deaths (see the section Advertising Restrictions). The FDA continues to prevent manufacturers from advertising and promoting some of aspirin’s many uses.
Luckily for Americans, aspirin was available in the United States long before the modern FDA (aspirin was known to Hippocrates, although that knowledge was lost and not rediscovered until the late eighteenth century). Given the FDA’s dangerously risk-averse attitude and its reliance on animal studies, many people have speculated that aspirin would not be approved by the FDA if it were discovered today(see, for example, Wardell and Lasagna 1975, 137–39). Although aspirin is in general a very safe drug, it can cause stomach bleeding and is responsible for some seven thousand deaths and seventy-six thousand hospitalizations per year in the United States (Grinspoon 1997). In addition, aspirin, along with a number of other familiar drugs such as penicillin, produces very negative effects in some laboratory animals. All of this leads us to the question, How many wonder drugs have we lost because of the FDA’s deadly overcaution?
Consumers want quality and safety in their drugs and devices. Moreover, consumers, prior to purchase and use, want assurance of quality and safety. Society has three broad approaches to quality and safety assurance. The first is voluntary practices and institutions—such as reputation, knowers, and middlemen—which assure quality and safety because it is almost always unprofitable to harm or cheat the customer. The second is tort remedy, by which consumers who are harmed or cheated may litigate under the tort doctrines of fraud, negligence, breach of contract, and so on. The third is governmentally imposed restrictions on voluntary exchange, whereby government attempts to determine the quality and safety of goods and services, and prohibits exchange until it has given permission. The FDA represents the third approach, which is the only approach that entails coercion against innocent parties. The theme of this Web page is that the best way to meet the demand for assurance is the combination of voluntary means and the tort system. We argue that the third approach, government restriction, does not achieve anything beyond what would be achieved by the other two approaches, yet prevents people from consummating many important and legitimate activities and exchanges. We argue, essentially, that the demand for assurance, like the demand for clothing, is best met by the free-enterprise system working within a sensible tort system.
Krauss (1996) explains: “[In the early 1990s, the FDA] ordered hospitals to stop using specialized baby ventilators, which are irreplaceable in saving sick infants because they provide uniquely tiny breaths of air, because hospitals refused to ‘blind-test’ them and thereby condemn[ed] fifty percent of air-deprived infants. Dr. Martin Kessler of Georgetown University Medical Center estimates that scores of babies died as a result. Subsequent to the FDA decision, protest from doctors who pointed to named infants’ deaths were aired on the [ABC] television show ‘20/20’ [“So Safe You Could Die—Overregulation by the FDA,” broadcast 12 August 1994]. Only then did the agency again allow use of the ventilators” (468–69).
This incident (see also Some Remarks about Medical Devices) illustrate what it takes to pry permission from the FDA stranglehold. The special circumstances that made the public-information campaign successful in this case are, tragically, extremely rare. Usually, doctors are not able to identify the victims of the FDA’s withholding of new therapies.
Biological drugs (or biologics) such as insulin, penicillin, blood and blood products, vaccines, derivatives of natural substances, and extracts of living cells are grown or cultured in separate batches. Just as with beer or wine, the quality can vary considerably by batch depending on small differences in inputs. Thus, in addition to obtaining marketing approval, a biologics manufacturer previously also had to have its production methods and facilities FDA licensed. Moreover, every batch of biologics had to be FDA tested. Recent advances in biotechnology, however, have diminished the variation and made production more like that of nonbiological (or chemically synthesized) drugs. In 1995, the FDA announced simplified rules on “well-characterized” biologics, dropping manufacturing-facility licensing and batch certification in such cases. Today many biologics are treated in the same fashion as nonbiological drugs. The FDA’s rules on biologics were codified in the 1997 Modernization Act.
In 1992, journalists highlighted claims by women that their silicone breast implants caused connective tissue diseases. Without clinical tests or clear scientific evidence, the FDA banned most implants and crusaded against the industry, leading to a $3.2 billion court settlement that drove manufacturer Dow Corning into bankruptcy. Studies by the University of Michigan, the University of Maryland, the Mayo Clinic, the Harvard Medical School, the American Medical Association, and others, however, have found no causal link between implants and connective tissue diseases. An independent panel of experts commissioned by the British government concluded that “Silicone gel breast implants are not associated with any greater health risk than other surgical implants.” And “In particular, there is no evidence of an association with an abnormal immune response or typical or atypical connective tissue diseases or syndromes.” The British study (Silicone Gel Breast Implants: The Report of the Independent Review Group [London: Department of Health, 1998]) references previous reviews and the voluminous scientific literature. Angell (1997) is an accessible yet scientifically accurate account of the breast implants controversy.
The brief summary is a technical document written for physicians and other health care professionals that the FDA requires to accompany an advertisement for a drug. It is typically a page in length, written in small, often illegible print and is so technical that it is virtually incomprehensible to consumers. The brief summary is not required if the advertisement mentions the name of the drug but not the use or if the remedy is mentioned without the name of the drug. Because the brief summary typically requires an additional page in a printed ad or additional airtime in a TV or radio ad, many pharmaceutical companies have less incentive to inform consumers of useful information.
The DSHEA rejected the FDA’s attempt to regulate vitamins, minerals, oils, fibers, and other dietary supplements as drugs. Under the DSHEA, manufacturers can make statements of nutritional support but not drug claims. Manufacturer’s of St. John’s Wort, for example, may claim that St. John’s Wort “promotes healthy emotional balance and well-being,” but they cannot say St. John’s Wort “is useful in the treatment of depression.” The distinction is mostly one for lawyers, not consumers, considering that many consumers do take St. John’s Wort for depression. (Such consumers are in fact justified in doing so: a number of studies indicate that not only is St. John’s Wort effective at relieving mild cases of depression (e.g., Woelk 2000), but does so with fewer side effects than many antidepressive pharmaceuticals. In addition, St. John’s Wort is considerably cheaper than pharmaceuticals and does not require a prescription.) As another example, a manufacturer of cranberry juice might claim that “cranberry juice helps to maintain a healthy urinary tract” but could not maintain that cranberry juice is “useful in preventing a urinary tract infection (UTI)” despite the fact that millions of women do drink cranberry juice for that reason (although there is some scientific evidence that cranberry juice can prevent UTIs, it is, as of 2001, very preliminary).
The DSHEA also allowed manufacturers and retailers of dietary supplements to disseminate scientific information on the value and use of supplements.
The 1962 amendments authorized the FDA to test old as well as new drugs for efficacy. The FDA lacked the personnel to do these tests, so the National Academy of Sciences appointed a special committee to perform efficacy investigations on some four thousand pharmaceuticals. The committee, the Drug Efficacy Study Implementation (DESI), was composed of 160 physicians, who relied on their pharmacological knowledge, literature reviews, clinical experience, and intuition. The appointment of this group was somewhat paradoxical: instead of drawing on the knowledge of all doctors, which included knowledge about specific patients and what treatments had and had not worked on those patients, 160 doctors were empowered to decide for the nation in the absence of much relevant information.
Europe’s more efficient regulatory structure typically allows new drugs to be approved sooner in Europe than in the United States. The difference between the date of European adoption and that of U.S. adoption is known as drug lag. Drug lag has been important in the debate about the FDA because it is measurable and because U.S. legislators do not like to think that Europe does things better than the United States. Drug lag, however, is far from an ideal measure of the costs of the FDA. Note that drug lag will be reduced the more inefficient the European system becomes, but this is scant comfort for patients in the United States who die from lack of access to new drugs. Moreover, it is far from obvious that Europe has the best system of drug approval, even if it is faster than the system in the United States. More generally, as a result of the FDA’s long review times and extensive regulations, new drugs take longer to reach the marketplace than they would in a voluntary certification system. Because a voluntary certification system doesn’t presently exist, we cannot measure drug lag with respect to such a system, but such lag is almost certainly considerably longer than drug lag with respect to Europe.
The FDA does more than delay the introduction of new drugs; it also reduces the total number of new drugs created. This is known as drug loss. Drug loss is difficult to measure, although Peltzman’s (1973) results (see also the discussion in the text) suggest that it is very extensive.
The Durham-Humphrey Amendment codified the distinction between prescription and OTC drugs. A drug was to be an OTC drug unless (1) it was habit forming, (2) had potentially harmful effects or potentially harmful effects when used by a lay person, or (3) was limited to prescription use by a New Drug Application. Prior to this amendment, manufacturers decided whether a drug was to be sold OTC or by prescription only, but the company could be sued for misbranding if the FDA believed that the drug could not be properly labeled for consumer use.
More information on this amendment can be found in the History section.
Elixir Sulfanilamide is a sulfa drug (antibiotic) released by Massengil in1937 in liquid form without prior toxicity testing of its solvent. The solvent diethylene glycol, used today as automotive antifreeze, caused the death of 107 people, mostly children. The chemist who created the elixir committed suicide. The “Elixir Sulfanilamide tragedy” prompted the passage of the Food, Drug, and Cosmetic Act of 1938. Although the deaths of 107 people were tragic, many more people have died because of FDA regulation. Deaths caused by drugs are seen, however, whereas deaths caused by a lack of drugs are unseen.
The old term ethical drugs signified drugs advertised only to doctors. The expression refers to the original 1847 code of ethics of the AMA, which deemed advertising directly to the public to be unethical. Over time, the term came to mean legal drugs.
The 1997 Modernization Act moved in the direction of “reform”; thus, most of its requirements were in the direction of reducing FDA bureaucracy and speeding drugs and devices to patients. The act was, however, at best a modest reform package. Specifically, the Modernization Act reauthorized user fees for another five years, codified rules for fast-track approval, codified the rule that only one adequate and well-controlled clinical study and confirmatory evidence could be the basis of approval, and codified restrictive FDA policies on dissemination of information regarding off-label uses of drugs. (Washington Legal Foundation v. Friedman found such restrictions unconstitutional and further opened up the ability of firms to disseminate off-label information.) Concerning medical devices, the Modernization Act exempted most Class I and some Class II devices from premarket approval, and it increased physician authority to use investigational devices. Finally, in a variety of clauses, the FDA was required to provide manufacturers with better and more timely information concerning its procedures.
Set up in 1915 to restrain monopoly, the FTC monitored drug advertising from its inception until 1962, when the regulation of advertising of prescription drugs matter was transferred to the FDA. The FTC still regulates the advertising of OTC drugs, with a few exceptions, notably aspirin. On food products, the FDA and the FTC have split responsibilities, with the FDA regulating food labeling and the FTC regulating food advertising.
Manufacturers of a new medical device need not file a premarket approval form if the device is “substantially similar” to an already approved device that did not require a premarket approval form. Instead manufacturers may file the simpler form 510(k). The 510(k) was supposed to be a premarket notification that did not require approval. In the years following 1976, the FDA began to require that more and more information be included with the 510(k) notification. Now the industry considers the 510(k) to be the first in a series of premarket approval hurdles, rather than a notification. With the passage of the SMDA in 1990, the 510(k) officially became a premarket approval regulation.
Passed after the Elixir Sulfanilamide tragedy, the Food, Drug, and Cosmetic Act required pharmaceutical manufacturers to submit a New Drug Application to prove the drugs’ safety. The act also provided the basis for the distinction between prescription and nonprescription drugs that was further developed in the Durham-Humphrey Amendment.
After a drug goes off-patent, any manufacturer who finds it profitable may produce an equivalent drug and sell it under the drug’s chemical or “generic” name. The original version is then called the brand-name drug, and the new competitive versions generic drugs. The Waxman-Hatch Act made it much easier for generic drugs to compete against their originals.
The various Good Practices are lengthy FDA rules and regulations governing manufacturing, laboratory, and clinical facilities. Most of the rules are obvious and include such things as, “Adequate washing facilities shall be provided, including hot and cold water, soap or detergent, air driers or single-service towels, and clean toilet facilities easily accessible to working areas,” and “Personnel shall practice good sanitation and health habits.” Many rules require the keeping of extensive written records. The current GMPs can be found here. Although the rules and accompanying inspections are not without utility, manufacturers rarely challenge the FDA, in part because they fear retaliation through arbitrary and capricious interpretation of the various standards and regulations (see, for example, the cases discussed by Volokh 1995).
The private sector also offers quality-control certification through such services as ISO 9000 Certification. The ISO 9000 standards are a series of standards for quality management that are produced by the Swiss organization International Organization for Standardization. The standards involve repeated audits and inspections.
IRBs review, approve, and monitor research involving human subjects in order to evaluate the ethical acceptability and the scientific validity of any such studies. An IRB is formally designated by an institution in which research takes places, such as a hospital or university. FDA regulations on IRB membership are quite strict. The IRB must be composed of at least five members including at least one scientific member, one nonscientific member, at least one person not affiliated with the research institution, no members with conflicts of interests, both genders if at all possible, and so forth. Research cannot begin until the IRB approves.
Medical devices must receive FDA approval before being marketed (since the 1976 Medical Device Amendments to the Food, Drug, and Cosmetic Act). Similar to an IND, an IDE allows companies to sell and use a limited number of devices for investigation purposes and clinical trials. The IDE exempts the device not only from the premarket approval regulation but also from a host of other reporting and recording regulations.
Before testing a new drug on human subjects, the company must file an IND. Prior to filing an IND, the sponsor develops information on the chemistry of the drug so that it can be produced in batches of known strength and purity. In addition, the sponsor must conduct a number of animal studies to produce information on the pharmacology and toxicology of the drug. Information, for example, must be produced on the absorption, distribution, metabolism, and excretion properties of the drug. Finally, detailed protocols for testing on human subjects must be submitted. In addition, since 1971, the FDA has required that all proposed clinical studies be reviewed by an institutional review board. Technically, unless otherwise notified, the sponsor can begin clinical studies within thirty days (if the IRB approves). The FDA can, however, terminate an IND at any time; thus in practice the FDA must approve the IND proposal. The IND stage of drug approval is broken into three phases. The FDA exerts considerable control over all phases of the clinical trial process, and at any stage the FDA can and often does request additional clinical trials and changes in trial protocols.
This phase consists of short-term clinical tests of the drug on twenty to eighty healthy volunteers to determine basic pharmacological and toxicological information in humans especially as regards safety. The FDA can stop clinical testing if they deem the drug unsafe.
This phase consists of small-scale, longer-term tests for efficacy and safety. Typically the drug is tested in one hundred to three hundred patients. In phase II trials, dosage levels are experimented with to find optimal dosage levels, and further information on safety is collected.
Large-scale testing for safety and effectiveness is conducted in phase III. Typically one thousand to three thousand patient volunteers are used in this phase. The primary information the FDA will use to decide whether the drug satisfies its (often arbitrary) benefit-risk relationship is developed in phase III trials. The trials are tightly controlled, may involve a large number of patients, and can take several months to several years for completion.
Upton Sinclair’s famous muckraking novel about the Chicago meatpacking industry contained outlandish, graphic images of Durham’s Pure Leaf Lard being made out of the remains of workmen who had fallen into the cooking vats. (See Libecap and Pasour for critical examinations of the issues raised in this book.) When this 1906 novel upset the public, regulation advocates used the public’s reaction to secure passage of the Pure Food and Drug Act of 1906 (as well as the Meat Inspection Act of 1906).
Also known as the 1962 Amendments to the Food, Drug, and Cosmetic Act of 1938, the Kefauver-Harris bill required pharmaceutical firms to wait for FDA approval prior to marketing their product. By law, the FDA is supposed to review an NDA within 180 days, but no penalties for failure exist, and the FDA has never come close to meeting this requirement. Additionally, the 1962 Amendments required the firm to show that a drug is efficacious in addition to safe for all labeled uses. The 1962 Amendments also brought clinical research and development under the authority of the FDA. Drug sponsors henceforth had to file an Investigational New Drug Application and obtain FDA approval for all investigational studies involving humans. Laboratories and clinics engaged in research also became subject to Good Laboratory Practice and Good Clinical Practice regulations that required extensive paperwork.
More information can be found in the History section.
An organization that knows more than the consumer about a seller’s reputation or about the quality and safety of the seller’s products is called a knower organization. The term usually refers to private, voluntary organizations. Knower organizations, when paid by the seller, often inspect quality and safety, and grant a seal of approval or certification mark, just as Underwriters’ Laboratories does for electronic safety, Orthodox Union does for kosher foods, Moody’s does for securities, medical schools do for newly graduated practitioners, and the American Dental Association does for dental products. Alternatively, knower organizations may investigate quality and safety, and sell their reports directly to consumers, as do Consumer Reports, credit bureaus, and doctors and pharmacists who recommend which drugs to take. Medical groups, hospital affiliations, and medical degrees are types of seals of approval that assure consumers that they can trust the doctor who recommends a drug. In the drug field, the growth and development of knowers and knower organizations are severely stunted by FDA command of the drug and device industries. Nonetheless, such organizations would surely expand and mature if people were freer in making their own drug choices.
These amendments greatly increased the FDA’s control over medical devices. The amendments divided medical devices into three classes. Class I devices are subject to general control requirements (certain records and reports provided to the FDA by the manufacturer); Class II devices are subject to general control and specific performance-based standards; and Class III devices are subject to a premarket approval process that is much like the approval process for new drugs. In addition, the amendments gave the FDA power to ban medical devices, to require manufacturers to inform consumers of potential hazards, and to force manufacturers to give refunds.
See the History section for more information.
This law requires hospitals, ambulance services, surgical facilities, nursing homes, and outpatient facilities to file a report to the FDA if a device is “associated with” any death, injury, or illness. With respect to sick individuals, however, it’s often difficult to say whether a device is “associated with” an adverse event in any meaningful way. Medical device reports have thus involved extensive paperwork, a large fraction of which the FDA never reads (Higgs 1995c). Adverse events reports involving devices and patient death must be sent directly to the FDA, whereas reports of events that have caused injury or illness are sent to the manufacturer.
A retail drug store or pharmacy is an example of a middleman. The middleman purchases goods from suppliers and then sells to consumers. Middlemen are very important in providing assurance of quality and safety because they have repeated dealings with customers and wish to induce the customer to continue buying from them. A pharmacy that sold someone an unsafe drug not only would be subject to suit, but would probably lose that person’s business and perhaps the business of those who learned of the mishap. Seeking to build and preserve a good reputation, pharmacies have a strong incentive to exclude unsafe or ineffective drugs from their shelves. They have strong incentives to know which drugs are safe and which dangerous. Another form of middleman is the pharmaceutical company. The company purchases the inventions and discoveries of researchers, develops them into brand-name products, and then sells them to the public. To preserve the reputation of the brand name and to avoid lawsuits, the company thinks carefully before putting a new drug on the market. Another example of a middleman is a hospital or clinic that purchases supplies and equipment, and employs staff. Consumers have repeat dealings with the hospital, and the hospital has repeat dealings with its suppliers and staff; in this way, the middleman creates a bridge of reputation and trust from consumers to the suppliers and the staff.
Published by the U.S. Pharmacopoeia (USP), a private nonprofit organization, the National Formulary is the official compendium of standards for drugs, excipients, dietary supplements, and vitamins and minerals. The USPNF defines standards of strength, quality, purity, identity, packaging, labeling, and storage, and describes and defines the appropriate tests, assays, and analytical methods that are used to measure strength, purity, and so forth. The USP also publishes the USPDI, a compendium of drug information that is also officially recognized in the United States and in many other countries worldwide.
In the early stages of research, a new substance is called a new chemical entity, a chemical entity that has never been used by humans and has been tested only on animals.
Under the Food, Drug, and Cosmetic Act of 1938, the NDA was submitted to the FDA enumerating the uses of the drug and providing evidence of its safety. If the FDA found no reason to object, the NDA was automatically approved within sixty days. But since 1962 the FDA evaluates proof of efficacy as well as proof of safety, and the company must wait for FDA approval no matter how long that takes. On average, the NDA review process lasts for two years (as of the mid-1990s, some evidence suggests times have shortened). The NDA review is handled by the FDA’s Center for Drug Evaluation and Research (CDER). Once the CDER deems the NDA fileable, the medical, biopharmaceutical, pharmacology, statistical, chemistry, and microbiology departments of the CDER review it. The length of an NDA application can reach one hundred thousand pages of material. If the departments pass the NDA, an advisory committee meets. If the advisory committee is satisfied with all of the findings, a labeling review takes place. Once the labeling review is complete, the NDA is approved, and the drug is ready to be marketed.
The NLEA required food manufacturers to include nutritional labeling on most food products. The NLEA also codified the FDA’s authority to allow health claims on foods and dietary supplements. Although the intent of the NLEA was to increase the amount of information consumers received by broadening the health claims allowed on foods and dietary supplements, the FDA officials took an aggressive stance and announced that they planned to regulate supplements as drugs. The resulting backlash led to the passing of the Dietary Supplement Health and Education Act (DSHEA) of 1994.
Many drugs that have gained FDA approval have uses other than those for which the drug was officially evaluated and approved. Once a drug has been approved for some use, however, it can be legally prescribed for any use. Approved uses are known as on-label uses, whereas other uses are off-label uses. Manufacturers may disseminate information, either in literature to doctors or in advertisements to the public, about on-label uses, subject to restrictions such as the brief summary. But disseminating information about off-label uses was prohibited between 1962 (the Kefauver Amendments) and 1997 (the Modernization Act). Off-label uses are often very important and common. Amoxicillin and tetracycline are today routinely used to treat stomach ulcers following Barry Marshall’s revolutionary discovery that ulcers are caused by Helicobacter pylori, but they are not approved for such uses (as of the late 1990s). Aspirin is also routinely used to prevent heart attack, even though it is not an approved use (as of 2000). Restriction on advertising and information dissemination have limited the information available to doctors and have thus sometimes prevented patients from receiving needed medication. Many of the FDA’s restrictions on dissemination of information related to off-label prescriptions were declared unconstitutional in Washington Legal Foundation v. Friedman D.D.C. See also the discussion in the text and Tabarrok (2000).
The market for some drugs is small because the disease treated is rare or because only a small number of people do not respond satisfactorily to the existing therapies. Nevertheless, because there are many diseases that affect only a small number of people, the total number of Americans with an orphan disease has been estimated to be twenty to thirty million (Meyers 1991). Because the costs of getting through the FDA process are the same whether the intended market for the drug is fifty thousand patients or five million patients, drug loss has been especially prevalent for orphan diseases. The 1983 Orphan Drug Act, amended in 1984, recognized this problem, but rather than reducing review times or directly lowering the burden of FDA-required trials, it created more government programs to try to counter the effect of the FDA. In particular, the ODA provided for grants to defray the costs of testing products for rare diseases and gave sponsors of orphan products tax credits on their development costs. Most importantly, the FDA agreed that once it approved one sponsor’s drug, it would not approve any other similar drug for the same indication for a period of seven years. In effect, sponsors were given monopoly rights. Thus, the number of drugs for orphan indications was increased but at the cost of higher-priced drugs. A notable example of how the ODA has been abused is that the AIDS drug AZT was granted orphan status and has since earned many billions in sales revenue.
A drug that is available to the consumer without a prescription. As recently as 1940, all nonnarcotic drugs were available over the counter.
Initially designed to help AIDS patients, parallel tracking makes drugs showing promising results in phase III of the IND process available to patients whose condition prevents them from participating in controlled clinical trials. Parallel tracking is similar to the treatment IND, a program started several years earlier.
Patent medicine is a term from the pre-1938 era signifying not that the drug was patented, but that it was advertised directly to the public and that its ingredients were not being fully disclosed. Patent medicines often contained alcohol, opiates, or cocaine, providing relief rather than cure.
Pearson v. Shalala forced the FDA to relinquish some control over medical statements on dietary supplements. Congress authorized the FDA to pre-approve health claims on foods [in the Nutrition Labeling and Education Act (NLEA)] but the FDA created an undefined and apparently very difficult to pass standard for approving such health claims, and after 10 years had approved only 10 such claims (Steinborn and Todd 1999).
Durk Pearson and Sandy Shaw, supplement manufacturers, wanted to make four health claims on their dietary supplement labels. The proposed claims were “(1) Consumption of antioxidant vitamins may reduce the risk of certain kinds of cancers. (2) Consumption of fiber may reduce the risk of colorectal cancer. (3) Consumption of omega-3 fatty acids may reduce the risk of coronary heart disease. And (4) .8 mg of folic acid in a dietary supplement is more effective in reducing the risk of neural tube defects than a lower amount in foods in common form.”
Although there is scientific information supporting the claims and although such claims are widely heard in the mainstream health-media, the FDA refused to allow any of them. Pearson and Shaw sued in federal district court based on the First Amendment and the FDA’s failure to define their standard for accepting a health claim. After losing initially, the D.C. Court of Appeals ruled that dietary supplements may be labeled with medical claims as long as they bear a disclaimer that such claims have not received FDA approval.
A placebo is a pharmacologically inert capsule, injection, or procedure. The placebo effect is the improvement in well being that is due to medical attention coupled with the use of mere placebos.
The placebo effect must be distinguished from “regression to the mean,” the statistical likelihood that a sick person will get better absent any treatment. The placebo effect is best defined as the improvement in well being in a group treated with a placebo compared to a similar (ideally, randomly assigned) group that is not treated at all (Hróbjartsson and Gotzsche, 2001).
Though controversial, there is evidence that a significant placebo effect exists, especially when treating the symptoms of pain, stress, depression, and anxiety. Scientific evidence also supports the notion that a patient’s attitude and spirits affect her recovery. Brown and Severs provide a good description of how the placebo effect may be tied to such factors:
“A sense of control is restored when a patient decides to seek medical assistance. Furthermore, the symbols and rituals of healing—the doctor’s office, the stethoscope, a thorough physical examination, and a treatment prescription—offer reassurance. So, too, an explanation of the illness and the prognosis, when favorable, reduces fear and uncertainty . . . Plausible treatment mobilizes a patient’s hope and creates expectation of improvement” (Brown and Severs 1999, 33).
Brown and Severs provide a sample statement that a doctor might make to a patient when prescribing a placebo found to improve health—a statement that is truthful and nonmisleading yet might accommodate an improvement in attitude and spirits.
FDA efficacy requirements demand that a drug be proven “effective beyond a placebo.” Sobel (2002) argues that this standard unnecessarily reduces the development of and public access to placebos that would alleviate suffering.
Early in the twentieth century, Harvey Wiley, chief of the Bureau of Chemistry, was intent on showing the dangers of an uncontrolled food industry. He recruited a volunteer group of young men and had them ingest large quantities of foods with additives and preservatives, such as formaldehyde and boric acid. The consequent ailments of the Poison Squad (1902–6) were well publicized and helped to pave the way for the Pure Food and Drugs Act of 1906.
The first stage in drug development, preclinical research, involves synthesis and purification testing in the lab and animal testing. Thousands of compounds are tested in preclinical research before a handful are chosen to enter the second stage, which requires filing of an IND.
Premarket approval means that regulators must authorize a product, process, batch, or facility before it is allowed to serve the market. That which is not specifically permitted is forbidden. Aaron Wildavsky (1988) has distinguished the hubristic anticipatory approach of premarket approval from the humble resilience approach of the freedom of contract plus postmarket inspection and recourse (through tort, legal, or recall procedures). Wildavsky argues that the resilience approach allows for flexibility, differentiation, experimentation, and entrepreneurship. Beginning with the 1902 Biologics Act, however, government control has consistently enacted the anticipation approach by expanding premarket approval requirements. Today, drugs and medical devices must obtain premarket approval by the FDA before the product is allowed on the market. Devices categorized as Class III require premarket approval. According to the 1976 Amendments, all new devices were automatically placed in Class III even if they were low-risk devices. Since the FDA Modernization Act of 1997, new low-risk devices can go through an abbreviated process.
Drugs that consumers may purchase only if they have a doctor’s prescription, in contrast to OTC drugs, are called prescription drugs. Prescription requirements induce consumers to call on doctors; hence, they raise doctors’ income. Prescription requirements may also be viewed in relation to drug prohibition. Prescription laws give doctors the privilege of authorizing the sale of FDA-approved narcotics that are substitutes for illegal “recreational” drugs. To prevent doctors from becoming authorized “drug dealers,” the DEA and other authorities monitor and enforce against doctors who “overprescribe” narcotics (known as “script doctors”).
See also the subsection Prescription-Only Requirements.
Traditionally, the FDA, like other bureaucracies, obtained its funding from general tax revenues according to congressional appropriations. In addition to having biased incentives, the FDA has at times been slow to approve new drugs because it lacked adequate resources to hire enough competent investigators to examine NDAs. Thus, many people died as NDAs sat unexamined on reviewers’ desks. In 1992, Congress authorized the FDA to charge drug companies a user fee. The term fee is dubious because fee usually means payment for services rendered, but is it appropriate to call the FDA’s deliberation over whether to grant permission a “service”? The “user fee” is rather like ransom received for “the service” of releasing a kidnapped child.
In 2006 the FDA collected over $300 million in so-called user fees, which constituted 58 percent of its total drug and biological review funding (with the rest coming from FDA appropriations). The FDA must report these figures every year in a PDUFA financial report; the link is to the 2006 report. Initially authorized for five years, the fee act was extended for another five years in the FDA Modernization Act of 1997. Subsequently PDUFA III and PDUFA IV followed in five-year intervals.
User fees have reduced the FDA’s average review times and increased access to new drugs (Philipson et al. 2005 and Berndt et al. 2005) but total time to bring a drug to market was not changed appreciably because of increases in the clinical development time (GAO 2002, Kaitin and DiMasi 2000). The FDA must provide performance figures once a year in a PDUFA performance report; the link is to the 2006 report.
More information is available in the History section.
Part of the 1962 Amendments, the proof-of-efficacy standard requires firms to produce evidence demonstrating that their product is efficacious for claimed uses. The proof-of-efficacy standard creates the distinction between on-label and off-label uses. The FDA reasons that because off-label uses have not passed proof of efficacy, the company is banned from disseminating information about such uses.
This act banned the adulteration and mislabeling of food and drugs. It required that products specify the quantity of certain substances (alcohol, morphine, opium, cocaine, heroin, alpha- or beta-eucaine, chloroform, cannabis indicia, chloral hydrate, and acetanilide). It declared the U.S. Pharmacopoeia and the National Formulary to be the official documents determining standards. If a drug differed from these standards, the difference had to be stated on the package. The act did not address advertisements. Regarding therapeutic claims for a drug, in 1911 the Supreme Court ruled that the act did not prohibit claims that were false but not fraudulent. Along similar lines, the Sherley Amendment of 1912 decisively banned fraudulent claims. In other words, drug sellers could make therapeutic claims, even false claims, as long as they could plausibly show that they believed their own claims.
Rare diseases, as defined by the Orphan Drug Act, are those diseases or conditions affecting less than two hundred thousand persons in the United States at the time of designation. Or they may be diseases or conditions that affect more than two hundred thousand persons but for which the costs of developing a drug cannot be recouped within seven years from sales in the United States.
If the United States and, say, Britain had drug approval reciprocity, then drugs approved in Britain would automatically and immediately gain approval in the United States as well. The logic of such a proposal suggests that the U.S. government ought to establish reciprocity with countries that have a proven record of approving safe drugs. Such an arrangement would eliminate the delay with which drugs approved abroad become available to Americans. The FDA opposes the proposal, presumably because the reform would introduce competition. American drug companies would apply to, say, the British authority for approval because it is more efficient and reasonable in reviewing applications.
In any industry, trade, or profession, a seller‘s trading partners, associates, and customers develop opinions of his trustworthiness. They develop a sense of whether the seller’s products and services live up to the quality and safety that he promises. A good reputation is one of the most important keys to success because a good reputation will bring satisfied customers calling again and will bring others who hear of the seller’s good reputation. If the seller sells an unsafe product, he not only will pay tort penalties, but will lose reputation and business. Reputation is generated by word of mouth and by other informal means, but also by various knower organizations that evaluate, rate, and report on the seller’s quality and safety.
Passed in 1990, the SMDA has greatly broadened FDA authority over the medical devices industry. The act requires Medical Device Reports; not filing the reports can result in fines. The SMDA also formally changed the 510(k) procedure, which was originally intended to be a notification procedure, to a premarket approval procedure. The SMDA also permitted the assessment of substantial civil penalties for violating the Food, Drug, and Cosmetic Act relating to devices.
More information is available in the History section.
When a knower organization evaluates a product or service, it often grants a seal of approval such as a certification mark, a degree, or a rating, which helps to determine the seller’s reputation and provides assurance to consumers.
The split-label proposal is a reform proposal that would reduce the FDA suppression of information. The product label would consist of a part for FDA-approved health and nutrition claims and a part for non-FDA-approved claims. (A larger stride toward medical freedom would be to permit the marketing of products not approved by the FDA, provided that the label clearly indicates that the product is not FDA approved.) Although retaining FDA certification for those who want assurance from the FDA, the split-label approach would increase the amount of information available to the consumer and evoke market mechanisms, such as knowers and middlemen, to provide nongovernmental assurance of quality and safety.
The off-label uses of a drug may become additional on-label uses if the company submits a Supplemental New Drug Application and the FDA approves the application. SNDAs may take years to process and can be expensive. When a drug is off-patent or if the off-label use is for only a small population, there is little incentive (except that it is easier to advertise on-label uses) for a firm to obtain an SNDA.
It’s often very expensive or time-consuming to measure the effect of a drug on an ultimate goal such as mortality. Drugs to reduce cholesterol, for example, are intended ultimately to reduce the number of heart attacks and thus to lengthen life expectancy. It could take twenty or more years to test this hypothesis adequately, however. A surrogate endpoint, such as a reduction in cholesterol counts, is a more easily measured endpoint. A drug may be approved based on clinical trials showing a positive surrogate endpoint if there is evidence from other studies that the surrogate endpoint accurately predicts an ultimate benefit. (We know, for example, that men with high cholesterol are at greater risk of a heart attack, but this is not the same as knowing that a reduction in cholesterol will reduce heart attacks, although it is suggestive.) Postmarketing studies can continue to track the effectiveness of drugs that were approved using surrogate endpoint methodology.
The use of surrogate endpoints is controversial because a positive surrogate endpoint does not necessarily predict a positive ultimate endpoint. Encainide and flecainide were widely prescribed because they prevented premature beats of the heart on the theory that such prevention would reduce heart attacks. The Cardiac Arryhythmia Suppression Trial later showed that not only was this claim false but that encainide and flecainide could actually increase the number of heart attacks. See Moore (1995) and the brief discussion in Tabarrok (2000).
The sedative thalidomide was released in Europe in 1957 and taken by pregnant women to relieve morning sickness. Tragically, the drug caused severe birth defects in more than ten thousand children. When the horrible side effects were discovered, the drug was still pending approval by the FDA, which held up the drug for reasons unrelated to its danger to fetuses. Hence, the old, pre-1962 FDA was adequate in screening out thalidomide. Nonetheless, supporters of FDA power used the thalidomide tragedy to secure the 1962 Amendments, which vastly enhanced FDA power. Since that time, thalidomide has been a terrifying word. Yet as early as 1965 the drug was discovered to be effective in treating leprosy and in places other than the United States has long been the standard treatment for that disease. Thalidomide has also been used to treat other diseases, including lupus, some cancers, and Kaposi’s sarcoma. Yet only in 1998 did the FDA finally permit Americans to use thalidomide. Special rules require that doctors and their patients register with the drug manufacturer and the FDA before thalidomide can be prescribed; women who take the drug must agree to use two forms of contraception and to submit to biweekly pregnancy tests.
This application allows drugs that are at the end of phase III clinical testing to be made available to patients who are suffering from a serious or immediately life-threatening condition when there is no other treatment available. An “immediately life-threatening” condition is defined as one that will result in death within a few months. Instituted in the late 1980s in response to the AIDS crisis, this practice is similar in effect to parallel tracking.
Even after extensive testing, the safety and effectiveness of a new drug are always somewhat uncertain. The FDA can thus never be certain that a new drug, device, use, or claim will be a net good or a net bad for society. Type 1 errors occur when the FDA approves a drug that ends up being a net bad for society. Type 2 errors occur when the FDA rejects a drug that would be a net good for society. Type 1 errors produce identifiable victims. Such errors are very visible and often result in major media attention, public concern, and congressional action against the FDA. Hence, the FDA has a strong incentive to avoid type 1 errors. The FDA makes sure that type 1 errors are rare by increasing the length of the drug approval process and by requiring more extensive clinical trials. Such overcaution is deadly, however, because it leads to more type 2 errors: not permitting a drug, device, use, or claim that would be a net good to society. Type 2 errors are less visible because patients, doctors, and journalists are usually unaware that a drug has been delayed or suppressed and that it would have saved individuals. The victims of type 2 errors, the people who would have lived if the FDA had not delayed the legal use of a new drug, are just as real as those who die because the FDA approved a bad drug, but they are known only in a statistical sense and are much less salient. Intelligent drug policy should aim to minimize the harm associated with both types of error. Although AIDS and cancer patients have sometimes been vocal in protesting FDA drug suppression, for the most part there is no informed, organized constituency to represent those who suffer as the consequences of type 2 errors. Thus, instead of minimizing total harm, the FDA focuses excessively on avoiding type 1 errors, resulting in many type 2 victims, invisible to the public eye but no less real.
For further discussion, see FDA Incentives.
Underwriters Laboratories Inc. (UL) is a private, not-for-profit, product safety testing and certification organization. Founded in 1894, UL tests more than eighteen thousand different products for more than fifty thousand customers. It operates testing centers and has customers throughout the world. On a voluntary basis, manufacturers submit products to UL for testing and safety certification. There are no laws specifying that a UL mark must be used. (In the United States, many municipalities have laws, codes, or regulations that require a product to be tested by a nationally recognized testing laboratory before it can be sold in the area, but not necessarily by UL, so UL does not have a monopoly.)
UL successfully ensures high-quality standards in the fields of electrical products, fire suppression devices, automotive equipment, and much more. UL even certifies the electrical and mechanical aspects of medical devices. The success of UL in these fields suggests that UL and similar organizations might also ensure safe and high-quality drugs (Campbell 1999).
Published by U.S. Pharmacopoeia (USP), a private nonprofit organization, the USP-DI is a compendium of drug uses, covering both on-label and off-label uses. Using panels of expert physicians who evaluate the literature and clinical practice, the USP-DI presents information on which drugs are recommended for which uses, warnings, contraindications, dosages, etc. To keep up with best practices, it is updated regularly. The USP-DI is the best known of several such compendia. The USP also publishes the National Formulary.
The Complete Drug Reference, published jointly with Consumer Reports, is the consumer version of the USP-DI. The Complete Drug Reference contains more information than the Physicians Desk Reference (PDR), which is useful but compiles product label information only and thus does not deal with off-label usages.
Prior to this ruling, the FDA had maintained that manufacturers could not disseminate information about off-label uses to physicians, even photocopies of peer-reviewed journal articles, except under strict conditions (mainly that the manufacturer had to be in the process of submitting an SNDA for the off-label use). In WLF v. Friedman, the court ruled that many of the FDA’s restrictions violated the commercial free-speech rights of manufacturers. The FDA’s policies had in the meantime been codified by Congress in Section 401 of the FDA Modernization Act of 1997 (which did not become effective until late 1998). This aspect of the 1997 statute was subsequently struck down as unconstitutional by the same district court. On appeal, in Washington Legal Foundation v. Henney 202 F.3d 331 (D.C. Cir. 2000), the FDA backed down from its earlier position and reinterpreted section 401 of the Modernization Act in a way consistent with the district court’s ruling in WLF v. Friedman. Hence, the court decisions have broadened manufacturers’ freedom to disseminate information about off-label uses, though such freedom remains restricted.
The 1984 Drug Price Act extended patent terms to account for the time it took a drug to receive FDA approval, and it reduced the barriers to entry for generic drug producers by allowing them to assert safety and efficacy based on information in the original NDA and on a proof that the original drug and generic drug are bioequivalent. The submission procedure for a generic drug is called an Abbreviated New Drug Application (ANDA).
More information is available in the History section.