Frances Kelsey, the FDA Reviewer Who Kept Thalidomide Off the US Market

On August 7, 1962, in a ceremony at the White House, President John F. Kennedy pinned a medal on a 48-year-old government employee whose job was reading paperwork. Frances Oldham Kelsey had been a medical officer at the Food and Drug Administration for barely two years. Her achievement, as the public had learned only weeks earlier, was that she had said no, repeatedly, for more than a year, to a pharmaceutical company that wanted to sell a sedative in the United States. The drug was thalidomide. In West Germany, Britain, Canada, Australia and dozens of other countries, it had been marketed as a safe remedy for sleeplessness and morning sickness, and by 1962 it was clear that it had caused thousands of babies to be born with severe deformities, many with shortened or missing limbs, and many others to die. The United States had been largely spared. The President's Award for Distinguished Federal Civilian Service went to the reviewer who had kept asking for more data.

In short: Frances Kelsey, a new FDA reviewer, refused to approve thalidomide in 1960 and 1961 because the manufacturer's safety evidence was thin. When the drug's catastrophic effects on babies became known abroad, her stand helped push Congress to pass the 1962 Kefauver-Harris Amendments, which required drugmakers to prove their products work as well as that they are safe. Her story is about what a well-staffed, skeptical regulator can prevent.

The application that landed on her desk

Kelsey joined the FDA in August 1960. She was a Canadian-born pharmacologist with a doctorate from the University of Chicago and a medical degree, and she had taught and done research for years before coming to Washington. Within weeks she was assigned a new drug application from the William S. Merrell Company of Cincinnati, which wanted to sell thalidomide in the United States under the brand name Kevadon. The application was considered routine. The drug was already widely sold abroad. The company expected approval quickly.

Under the law at the time, a new drug application took effect automatically after 60 days unless the FDA found it inadequate and said so. That put the burden on the reviewer. To stop a drug, Kelsey had to keep finding reasons, in writing, why the application was not yet good enough.

She found them. The company's data on how the drug was absorbed and how it behaved in the body was incomplete. Its animal studies were not persuasive to her. Then, in early 1961, she read a letter in the British Medical Journal from a physician reporting peripheral neuritis, nerve damage causing tingling and numbness in the hands and feet, in patients who had taken thalidomide over long periods. To Kelsey, a drug that could damage nerves in adults raised an obvious question: what might it do to a developing fetus? She asked Merrell for evidence of safety in pregnancy. The company did not have it.

Pressure from the other side of the table

Merrell pushed back hard. Company representatives contacted the FDA many times, by phone, by letter and in person, and went over Kelsey's head to her superiors. They called her demands unreasonable and her objections bureaucratic. They pointed out, correctly, that the drug was in widespread use in other countries without obvious problems in adults.

It helps to sit with how that pressure looked from inside the agency. Kelsey was new. She was a woman in a mostly male profession at a time when that alone could invite condescension. The drug was approved in many wealthy countries. Her supervisors, to their credit, backed her. She kept returning the application.

Meanwhile, in late 1961, physicians in West Germany and Australia were independently connecting a wave of severe birth defects to mothers who had taken thalidomide in early pregnancy. The German pediatrician Widukind Lenz and the Australian obstetrician William McBride raised the alarm, and the drug was withdrawn from the German market in November 1961. Merrell withdrew its American application in early 1962.

The loophole and the babies born here anyway

The United States was spared the worst, but not entirely. Under the rules of the time, companies could distribute unapproved drugs to physicians for "investigational" use with very few controls. Merrell had sent thalidomide samples to more than a thousand American doctors, who gave them to many thousands of patients, often without telling them the drug was experimental and without systematic records. When the danger became known, the FDA had to track down the supply. A small number of babies in the United States were born with thalidomide-related injuries, and the true count of exposures was hard to establish because record-keeping had been so loose.

One American case became especially well known. In 1962 Sherri Finkbine, an Arizona woman who hosted a children's television program, learned that a sedative her husband had brought back from Europe contained thalidomide, which she had taken early in her pregnancy. Unable to obtain an abortion in Arizona, she traveled to Sweden to end the pregnancy. Her case drew national attention and became an early moment in the American debate over abortion law.

How a near miss became a law

The public learned of Kelsey's role in July 1962, when The Washington Post published a front-page story about the FDA reviewer who had kept a dangerous drug off the market. The timing was crucial, because a drug bill was already sitting in Congress.

Senator Estes Kefauver of Tennessee had been investigating the drug industry since 1959, holding hearings on drug prices, marketing and the weak evidence behind many products' claims. His bill had stalled under industry opposition. Thalidomide changed the politics overnight. Within months, Congress passed the Kefauver-Harris Drug Amendments, named for Kefauver and Representative Oren Harris of Arkansas, and Kennedy signed them on October 10, 1962.

The new law did several things at once:

• Proof of efficacy. For the first time, drugmakers had to show that a drug actually works for its intended use, with "substantial evidence" from adequate and well-controlled investigations. Before 1962, the law required evidence of safety only.
• Affirmative approval. The FDA now had to approve a drug before it could be marketed, rather than letting applications take effect automatically.
• Oversight of trials. Clinical investigations became subject to FDA regulation, and the law required that patients be told when they were receiving experimental drugs and give consent, with limited exceptions.
• Advertising and manufacturing. The FDA gained authority over prescription drug advertising and stronger rules for manufacturing practices.

Notice what the law did not do. It did not single out thalidomide, and it did not punish Merrell. It changed the default. Before October 1962, a drug reached the market unless the government found a reason to stop it in time. After, a drug stayed off the market until its maker proved its case. Kelsey had been doing by stubbornness what the new law made routine.

The FDA then began a large review of drugs approved between 1938 and 1962 to check whether they worked, a project that took years and led to many products being withdrawn or relabeled.

What "proof that it works" means in practice

The phrase "adequate and well-controlled investigations" sounds like legal boilerplate. In practice it changed how medicine is done.

Before 1962, a company could support its claims for a drug with testimonials from doctors who had tried it on patients and liked the results. Those impressions are notoriously unreliable. Patients often improve on their own; doctors who believe in a treatment tend to notice successes and forget failures; and people given any pill frequently report feeling better, the placebo effect. A drug could be sold for years on the strength of enthusiasm rather than evidence.

The 1962 standard pushed the industry toward controlled trials, in which a drug is compared with a placebo or an existing treatment, patients are typically assigned to groups at random, and, where possible, neither patients nor researchers know who is getting what until the end. The FDA later spelled out what an adequate trial needs in its regulations. Randomized controlled trials had existed before; a famous British study of the antibiotic streptomycin for tuberculosis had been published in 1948. But the American law made them the expected route to market for the world's largest drug buyer.

The review of older drugs that followed showed why it mattered. The FDA asked the National Academy of Sciences to evaluate the effectiveness claims of thousands of drug products already on the market. Many claims could not be supported, and the agency eventually removed hundreds of products or required changes to their labels.

The debate over that standard has never fully settled. In recent years the FDA has used accelerated approval, which allows a drug for a serious condition to reach patients based on an early marker of benefit, with confirmatory trials required afterward. The pathway has brought important drugs to patients faster, especially in cancer and HIV. It has also produced controversies, most visibly the 2021 approval of the Alzheimer's drug Aduhelm over the objections of the agency's own outside advisory committee, several of whose members resigned. Each such fight is, in a sense, a replay of the question Kelsey faced: how much evidence is enough before people take a drug?

There is no permanent answer. There is only a process, staffed by people, that has to make the call one application at a time.

The earlier disaster that trained her

There is a detail in Kelsey's career that makes her stand less surprising. In 1937, as a young graduate student in Chicago, she had worked on research into Elixir Sulfanilamide, a liquid antibiotic product that a Tennessee company had dissolved in diethylene glycol, a toxic solvent. More than a hundred people died, many of them children. That disaster led to the Federal Food, Drug, and Cosmetic Act of 1938, which first required companies to show a drug's safety before selling it.

So the reviewer who stopped thalidomide had seen, a quarter century earlier, what happens when no one checks. She is also the subject of a story she told herself about how she got that Chicago job: her first name, Frances, was taken for the masculine spelling Francis, and the professor who hired her assumed she was a man. She said later that she was not sure she would have been hired otherwise.

Kelsey worked at the FDA for 45 years, much of that time on the protection of people in clinical trials. She died in 2015 at the age of 101 in Ontario, shortly after receiving the Order of Canada.

The argument about regulators, then and now

The thalidomide story is often told as a simple fable: brave reviewer, villainous company. A fairer telling has to engage with the argument on the other side, because it has never gone away.

Critics of the FDA, including many economists, argue that the 1962 law made drug development slower and more expensive, and that the cost is invisible. A drug delayed or never developed harms patients who would have benefited from it, and those patients do not appear in newspaper photos. Some studies after 1962 found a drop in the rate of new drugs introduced. The argument is serious, and it has shaped reforms such as accelerated approval pathways and user fees that fund faster reviews.

Here is my answer. The 1962 law's core demand, that a drug be shown to work before people pay for it and take its risks, is not bureaucratic excess. It is the minimum a patient should expect. Delays are a real cost, and the right response is a well-funded agency that reviews quickly and carefully, not one that reviews less. Kelsey's skepticism was valuable precisely because she had the time and backing to apply it. A reviewer buried under applications, or frightened for her job, cannot do what she did.

That is why the story belongs on our topic page about a government being gutted. In 2025 the Department of Health and Human Services announced deep staff reductions, including thousands of positions at the FDA, as part of the administration's broader cuts to the federal workforce. Supporters said the agency could do more with less. Former officials and outside experts warned that losing experienced reviewers, inspectors and support staff would slow reviews and weaken oversight at the same time. Whatever the final effect, the thalidomide episode is a reminder that the value of a regulator lies in what does not happen, which makes it easy to cut and hard to measure.

What you can do

Weigh in when the FDA writes rules. Proposed federal regulations on drugs, devices and food safety regularly open for public comment, and they appear on our public comment page, sorted by closing date and grouped by issue, with practical guidance on writing a comment the agency must consider.

Ask your members of Congress how they will fund the agencies that check what we take. Appropriations for the FDA and the rest of HHS come up every year. Find your representatives and look at their latest votes before you call.

Follow the bigger fight over the federal workforce. Our topic page on health care for all and the one on government cuts collect reporting on how staffing changes affect patients.

Report a problem. If you or a family member has a serious reaction to a drug, your doctor can report it, and patients can report directly to the FDA's MedWatch program. This is general information; ask your doctor or pharmacist about any medication question.

Frances Kelsey's medal was for something that did not happen. Generations of American families never knew her name because her work succeeded. Protecting the next reviewer who says "not yet" is how we honor it.