Why the FDA Has an Incentive to Delay the Introduction of New Drugs

FDA drug reviewers are immune from legal liability, but they may be reproached and humiliated by congressional hearings, television exposés, and newspaper condemnations. Why don’t congressional representatives, television reporters, and newspaper columnists bring the deadly consequences of FDA delay to light? Answer: When the side effects of a new drug cause little Tommy of 236 Elm Street, Saginaw, Michigan, to become gravely ill, television reporters show the poor lad languishing in a hospital bed, and viewers respond emotionally. When little Tommy dies, reporters interview the grieving parents. Blame falls on the drug company, on the FDA officials who approved an unsafe drug, and maybe on the doctor. FDA reviewers are anxious to avoid such censure, which might damage their careers and reputations.

The consequences of error in the other direction, however, are not symmetric. If little Tommy suffered from a disease that would be cured by a drug not yet allowed by the FDA, it is unlikely that Tommy’s parents or doctors would even be aware of that fact. If they heard about the not-yet-allowed drug and inquired into its availability, the FDA may simply say that it “must hold the unproved drug until safety questions and risks to the public health are resolved.” No one who could counter such claims would be in a position to do so. Thus, the bad consequences of disallowing the drug would not be identifiable and would not revisit the FDA. In consequence, FDA officials are much less concerned about such consequences. Only in rare cases in which suffering patients have been well organized and vocal, in particular in AIDS cases, have FDA officials taken much heat for withholding approval. Even then the heat is not extreme because the FDA officials can always claim that they are simply doing their job in delaying the approval of experimental drugs. Researchers have long noted that in anxiously seeking to avoid risk of approving an unsafe drug, FDA officials often fall deeply into the inverse error: disallowing valuable drugs. Figure 3 shows the two types of error and indicates the reason for systematic bias toward type 2 errors.

Figure 3
Two Types of Error in FDA Approval Decision

Thus, in addition to the error of approving a bad drug, there is the error of withholding a good drug. It would be foolish to minimize the first type without minding the effects on the incidence of the second type. Attention to the two types of error reminds us that FDA policy can be very harmful overall even if reducing the FDA’s power would increase type 1 errors. Reducing FDA delay in approving drugs, for example, could result in perhaps hundreds of additional mortalities, but if the benefit is thousand of lives saved (fewer type 2 errors), then consumers are on net better off. This analysis shows the danger of comparing two situations based solely on the type 1 error. Willman (2000), for example, has excoriated the FDA for speeding up the approval of new drugs in the 1990s, but he looks only at type 1 errors and not at all at type 2 errors. The current level of type 1 errors should not be made sacrosanct. It is possible to have too few type 1 errors when this means a high rate of type 2 error.

It should go without saying that any reform that reduces one type of error without increasing the other is to be applauded. The FDA reforms of the 1990s, in particular with regard to user fees, have reduced the number of type 2 errors without increasing the number of type 1 errors. Although more drugs have been withdrawn in recent years, it is solely because more drugs have been accepted; the rate of drug withdrawals has not increased. Sadly, Willman (2000) misses even this elementary point. See FDA Reform for a further discussion of reforms.

Type 2 errors should be not only acknowledged but perhaps regarded as even more pernicious than type 1. Economists Friedrich Hayek, Armen Alchian, and Israel Kirzner have stressed that the strength and flexibility of an economic system lies in its propensities to correct its own errors. In the medical system, an error involving a bad drug tends to be corrected: it leads to bad outcomes, suffering, complaint, loss of reputation, legal penalties, and serious medical review. An error involving a good drug, on the other hand, does not lead to any specific or readily identifiable problem. Even after years of delaying important new drugs, there may be little public pressure to get the FDA to grant permission. This type of error is not inherently self-correcting. The difference in propensity for error correction gives a good reason for concerned parties, in managing the trade-offs between the two types of error, to be especially vigilant against policies and procedures that will introduce type 2 errors.

One of the most active current critics of the FDA is physician and Hoover Institution fellow Henry I. Miller. Miller’s most significant qualification, however, is that he worked in the FDA from 1979 to 1994 serving its regulatory apparatus. Speaking from personal experience, he described (in Miller 2000) the incentives inside the FDA:

In the early 1980s, when I headed the team at the FDA that was reviewing the NDA for recombinant human insulin, . . . we were ready to recommend approval a mere four months after the application was submitted (at a time when the average time for NDA review was more than two and a half years). With quintessential bureaucratic reasoning, my supervisor refused to sign off on the approval—even though he agreed that the data provided compelling evidence of the drug’s safety and effectiveness. “If anything goes wrong,” he argued, “think how bad it will look that we approved the drug so quickly.” (41)

Miller explains that the anecdote is representative of FDA behavior:

A classic type 1 error . . . was the FDA’s approval in 1976 of the swine flu vaccine. Although the vaccine was effective at preventing influenza[,] it had a major side effect that was unknown at the time of approval: temporary paralysis from Guillain-Barré Syndrome in a small number of patients. This kind of mistake is highly visible and has immediate consequences—the media pounces, the public denounces, and Congress pronounces. Both the developers of the product and the regulators who allowed it to be marketed are excoriated and punished in modern-day pillories: congressional hearings, television news magzines, and newspaper editorials. Because a regulatory official’s career might be damaged irreparably by his good faith but mistaken approval of a high-profile product, decisions are often made defensively—in other words, to avoid type 1 errors at any cost. . . . This predilection, or pressure, toward commission of type 2 errors is not the way the system is supposed to operate. Lifetime tenure for civil servants—which itself extracts costs—is supposed to free regulators to consider only the public interest as they render decisions. Instead, it has given us the worst of both worlds. (42–43)

Miller is not the first whistle-blower, and virtually all economists who study the FDA are critical (see Quotes Section). When enlightened criticism hits at the exterior walls of the FDA, how do those inside respond? Typically they do not respond at all.

American citizens have been led to think that agencies such as the FDA protect and benefit them. The official interpretation is propagated by America’s most powerful cultural institutions: the major media, the schools and colleges, and the government itself. To dissent publicly and seriously from the official interpretation is to confuse and upset the Americans who passively accept it. To dissent is also to court the enmity of the agency itself. Drug companies, although in the best position to know the sins of the FDA, are especially not inclined to vex or challenge their masters. FDA officials enjoy a respected “consumer protection” status and operate in a world of informational imperviousness and legal immunity (NDA reviewers are immune from legal liability for negligence).

Vincent Kleinfeld (1970), a Washington, D.C., attorney who worked on drug approval affairs, wrote, “the courts in this area of the law tend to equate the Food and Drug Administration with God, motherhood, and country.” He describes how the situation goes down in court when someone challenges the FDA for disallowing a drug:

If the Food and Drug Administration yells “hazard” or “danger,” you are not going to get one judge out of a hundred to hold against the Food and Drug Administration; this I can testify to from very bitter experience since I have been in private practice and with the government. When you are before a judge, it makes practically no difference whether the government is right or wrong. The government attorney looks up gravely at the judge and says, “Your Honor, the Food and Drug Administration . . .”—there is a pause right there—“takes the position that this product is dangerous; it may cause death either directly or because it keeps the patient away from the knife, the X-ray machine, radium. If Your Honor wishes to take the grave responsibility of substituting your judgment for that of the Food and Drug Administration . . .”—another pause. That is it. (180, 191)

Thus, the official interpretation of the FDA remains hegemonic.

When response does occasionally become necessary, Miller (2000) explains the consequences: "boxes on the organizational chart are arranged and rearranged, added and eliminated; names of entities are changed (and then changed back); and various pilot programs come and go. FDA managers avidly craft and meet new performance milestones, but there is little impact on the bottom line of timely patients’ access to new therapies.” He notes that more than one hundred studies and inquiries have documented the negative impacts of the existing system. “The reports of these studies, however, just seem to get filed away and forgotten” (49, 45–46).