The Death Toll of Delay

U.S. regulators’ overly cautious approach to risk during the pandemic has been deadly from the start.

A nurse holding a vaccine needle
Anthony Behar / Sipa / AP

About the author: Conor Friedersdorf is a California-based staff writer at The Atlantic, where he focuses on politics and national affairs. He is the founding editor of The Best of Journalism, a newsletter devoted to exceptional nonfiction.

This morning, the FDA granted full approval to the Pfizer vaccine for use in people 16 and older. Although “the vaccine approval was the fastest in the agency’s history,” as The Washington Post noted, serious side effects have proved extremely rare. Nevertheless, anti-vaccine activists—and the politicians and pundits pandering to them—have criticized the accelerated approval process as rushed. Yet the problem with the FDA is precisely the opposite: The agency too often fails to recognize the danger of being too cautious.

The same tendency is evident in the FDA’s review of whether to authorize the COVID-19 vaccines for preteens. Late last month, the government asked Pfizer and Moderna to expand the number of children ages 5 to 11 in their clinical trials and to provide follow-up data for a longer period. The companies complied. In early August, the American Academy of Pediatrics sent a letter to the FDA expressing concern.

“While we appreciate this prudent step to gather more safety data,” the pediatricians group wrote, “we urge FDA to carefully consider the impact of this decision on the timeline for authorizing a vaccine for this age group. In our view, the rise of the Delta variant changes the risk-benefit analysis for authorizing vaccines in children. The FDA should strongly consider authorizing these vaccines for children ages 5–11 years based on data from the initial enrolled cohort, which are already available.”

The group expressed additional worry about kids younger than 5. Noting that more than 340 million jabs had already been administered to Americans, the pediatricians argued that “there is no biological plausibility for serious adverse immunological or inflammatory events to occur” more than two months after getting vaccinated, and that demanding six months of follow-up data before approval for children under 5 “will significantly hinder the ability to reduce the spread of the hyper infectious COVID-19 Delta variant among this age group.” The subtext: Do you understand that the coronavirus is quickly spreading now?

Each time a new weapon emerges in the fight against a deadly disease, the bureaucrats charged with judging its safety and efficacy face a trade-off: Proceed too quickly and a dangerous or useless drug or medical device might be granted approval; proceed too slowly and the sick and dying might be denied a lifesaving intervention. Like Odysseus and his crew, an effective bureaucracy must be as concerned about Scylla as Charybdis, knowing that death is the consequence of tacking too far in either direction.

But public-health officials operate under incentives that distort their judgment. If they approve a drug or medical device that hurts even a tiny number of people, or that proves ineffective, the harms are clear, and those responsible may be pilloried in the press as negligent or corrupt. When the bureaucracy moves too slowly, the harms are harder to see. As the economist Alex Tabarrok put it in 2015, people still die, but “the bodies are buried in an invisible graveyard.”

The coronavirus pandemic has illuminated the costs of a sclerotic public-health agency’s inaction like never before.

“In my many years of writing about the FDA,” Tabarrok declared after seeing the AAP’s letter, “I can’t recall a single instance in which a major medical organization told the FDA to use a smaller trial and speed up the process because FDA delay was endangering the safety of their patients. Wow. The invisible graveyard is invisible no more.” The regulatory system urgently needs reforms. Among them: When judgment calls divide medical experts because they disagree on how to make thorny trade-offs, Americans should retain more freedom to make their own choices—and not just about vaccines.

The FDA’s current approach to drug approval was shaped by a tragedy in Europe, where the drug thalidomide, meant to treat morning sickness, caused birth defects and deaths in perhaps tens of thousands of babies in the 1950s. Largely in response, Congress passed legislation in 1962 that raised barriers to bringing new drugs to market. Thereafter, the FDA was charged with ensuring not only that a new drug was safe, but also that it was effective. And Congress eliminated the time constraint that had forced the FDA to move expeditiously on new drugs. By about 1980, new drugs needed an average of 10 years to gain approval.

In the 1980 manifesto Free to Choose, the Nobel Prize–winning economist Milton Friedman and his wife, Rose Friedman, showed how FDA regulation had done more to slow the introduction of valuable drugs than to prevent the distribution of harmful and ineffective ones. They wrote:

The effect on the rate of innovation of new drugs is dramatic: the number of “new chemical entities” introduced each year has fallen by more than 50 percent since 1962. Equally important, it now takes much longer for a new drug to be approved and, partly as a result, the cost of developing a new drug has been multiplied manyfold. According to one estimate for the 1950s and early 1960s, it then cost about half a million dollars and took about twenty-five months to develop a new drug and bring it to market. Allowing for inflation since then would raise the cost to a little over $1 million.

That cost had ballooned to $54 million by 1978—a hundredfold increase during a period when prices roughly doubled. Pharmaceutical development suffered in other ways too, the Friedmans noted:

As a result, drug companies can no longer afford to develop new drugs in the United States for patients with rare diseases. Increasingly, they must rely on drugs with high volume sales. The United States, long a leader in the development of new drugs, is rapidly taking a back seat. And we cannot even benefit fully from developments abroad because the FDA typically does not accept evidence from abroad as proof of effectiveness.

Of course, definitionally, the large majority of people don’t have a rare disease, and few relatively healthy people sit around thinking about the absence of new drugs that could improve their life.

The AIDS epidemic made the problem far more visible. In the late 1980s, when HIV was ravaging the gay community and was fatal for nearly all of those infected, public-health bureaucrats initially resisted activists’ requests to try new experimental drugs. In response, members of the advocacy group ACT UP converged on the FDA headquarters and demanded a chance to save their own lives. “One group were wearing lab coats that were stained with bloody hands,” David Barr, an ACT UP founder, recounted in a 2019 NPR interview. “Other people brought tombstones that they made and lied down in front of the building and held up the tombstones: ‘Dead from FDA red tape.’”

That epidemic ultimately yielded reforms that eased some patients’ access to drugs still under review. Another change was spurred by the anthrax attacks that followed 9/11. The Project BioShield Act of 2004 was meant “to provide protections and countermeasures against chemical, radiological, or nuclear agents.” Among other things, the act “allowed FDA to authorize formally unapproved products for emergency use against a threat to public health and safety (subject to a declaration of emergency by [the Department of Health and Human Services]),” according to a Harvard Law School analysis. “The record indicates that Congress was focused on the threat of bioterror specifically, not on preparing for a naturally-occurring pandemic.”

The “emergency-use authorization” created by that law is the legal basis on which Americans have been permitted to get the Pfizer, Moderna, and Johnson & Johnson COVID-19 vaccines. Tabarrok estimates that the speed and flexibility it permitted likely saved 100,000 lives.

Yet the federal government’s response to COVID-19 has been too slow in other ways. Early restrictions on which COVID-19 tests doctors could use were arguably the worst error. “The FDA’s guidance documents led to decreased availability of testing, particularly in the early stages of the pandemic, which contributed to the catastrophic course of the COVID-19 pandemic in the United States,” a study published in The Yale Law Journal concluded. Its authors, the law professors Barbara J. Evans and Ellen Wright Clayton, said that the FDA had interfered in the oversight of tests that it lacked clear authority over—a move they described as “possibly the deadliest regulatory overreach in U.S. history.”

Another significant failure was the lengthy delay in approving an over-the-counter, at-home COVID-19 test. “The F.D.A. needs to catch up to the science,” the Harvard epidemiologist Michael Mina told Ezra Klein in a January interview, complaining about the restrictions. “They are inadvertently killing people by not following the science.”

Those failures are noteworthy in part because, while ineffective tests could cause harm, the possible negative consequences of acting too quickly were not as dire as, say, potential birth defects. The danger of letting a new virus spread undetected was obvious. And still the FDA erred on the side of restriction.

Of course, approving a vaccine for children is a thornier matter: Injecting a substance into one’s body carries risks that taking a test does not; COVID-19 appears to pose less immediate risk to the young than to adults; and children have more years of life ahead, raising the costs of any downside to mRNA technology that manifests only years later. The American Academy of Pediatrics considered all this and still urged the FDA to speed up its review:

Based on scientific data currently available on COVID-19 vaccines, as well as on 70 years of vaccinology knowledge in the pediatric population, the Academy believes that clinical trials in these children can be safely conducted with a two-month safety follow-up for participants. Assuming that the two-month safety data does not raise any new safety concerns and that immunogenicity data are supportive of use, we believe that this is sufficient for authorization in this and any other age group.

Waiting for an additional four months of data, the group pointed out, will give the virus more time to spread. The academy’s claim is not that the Pfizer and Moderna vaccines carry no short- or long-term risks. The organization acknowledges that extremely rare cases of a heart inflammation called myocarditis have been reported among adolescents and young adults. But it also notes that such cases are most likely to occur within four weeks and not after six weeks. The pediatricians argue that six months of data are highly unlikely to tell us anything about those risks that we don’t know after two months. If your concern is that an mRNA vaccine could have an unforeseen downside that doesn’t manifest for 10 years, delaying the vaccine another four months isn’t going to catch that. Meanwhile, society would incur the cost of delay without the benefit of assurance that there is no long-term risk.

A convincing rebuttal to the AAP would articulate why the FDA thinks that four more months of data may realistically yield something that alters the decision to approve mRNA vaccines for the young. Of course, the truth is that there is neither an objectively correct period of time to track safety data for a vaccine nor anything like a consensus among experts about the best process given our novel circumstances. That subjectivity strengthens the case for considering a value that is implicated in these matters but that the AAP neglects to mention: the value of personal liberty, in this case the freedom to make choices about risks to our own body.

On crucial occasions during the pandemic, America would have fared better had regulators given Americans more individual agency over how to proceed. If allowed to use whichever COVID-19 tests they judged useful, medical professionals at the beginning of the pandemic could have greatly expanded their ability to track infections. If permitted to participate in human-challenge trials early on, potential volunteers such as the ones I interviewed in April 2020 could have accelerated vaccine approval in adults, moving up the point when jabs started going into arms. If permitted to purchase cheap COVID-19 tests without a prescription, consumers a bit later in the pandemic would have been able to check their own status more frequently, to voluntarily quarantine, and to avoid some spread. Had all that additional liberty been granted, many buried in “the invisible graveyard” would be alive today.

On the matter of vaccine approval for children, there’s no certain way to know if the process that the FDA appears to be pursuing or the alternative that the AAP has urged would prove superior. But uncertainty only strengthens the case for allowing families to reach their own conclusions. A parent homeschooling a 10-year-old on a farm in rural Idaho might be content to wait for six months of safety data, or two years for that matter.

But a single parent with a 9-year-old living in a crowded Florida motel full of anti-vaxxers and anti-maskers might see a more urgent need for vaccination. A pediatrician might agree, as might the 9-year-old herself. Under the status quo, though, none of their opinions matters, no matter how anxious that makes them.

If the FDA were merely in the business of studying safety and efficacy, and if its decisions were merely Good Housekeeping seals of approval for drugs and medical devices, instances when it proceeds too slowly would be less troubling. But because, under current law, the FDA is also in the prohibition business––because the public is effectively barred from using a new drug or medical device until the FDA signs off on it––every instance in which it proceeds too slowly transgresses against Americans’ rights to life, liberty, and making decisions about our own bodies, in consultation with doctors as we see fit.

During this pandemic alone, those transgressions have almost certainly cost many more lives than the thalidomide debacle, delaying both testing and vaccinations for months amid widespread death. Going forward, as new variants emerge and spread, and in preparation for future pandemics, the FDA should be reformed, retaining its capacity to judge a drug’s safety and efficacy, to prevent the advertisement of drugs that have not yet met its standards, and perhaps even to forbid drugs and devices that no reasonable person would consider safe.

But it should no longer have the power to deny medical professionals or the public access to drugs and tests that many experts believe would save lives by the thousands. The FDA will never have optimal incentives to move drugs through the review process quickly. But more liberty can provide a check on the agency’s failures.