A Third of New Drugs Hit the Market With Side Effects No One Knew About

Image: Derek Gavey/Flickr
Image: Derek Gavey/Flickr

Let’s say you’ve got some pretty severe arthritis pain. Your doctor prescribes you the same anti-inflammatory they’ve prescribed everyone else, and it works! This new drug has given you new life! But then, you start hearing disturbing news reports—the same drug seems to be causing an increase in the rates of heart attacks and strokes. What do you do? How do you weigh the risks and the benefits?


This happened (albeit with more foul play) with the popular, since-withdrawn anti-inflammatory drug Vioxx. And Vioxx might not be alone. According to a new study, roughly a third of drugs approved by the United States Food and Drug Administration (FDA) seem to have some sort of negative side effect that didn’t reveal itself during smaller, pre-market studies. But the study’s authors didn’t seem surprised, or even upset about the number. After all, a drug coming to market is the biggest clinical trial of all.

“New drugs are called new drugs for a reason: because they’re new,” Dr. Nicholas Downing, the study’s first author from Boston’s Brigham and Women’s Hospital, told Gizmodo. “Unless you want to test drugs in huge populations, there’s going to be uncertainty at the time when the regulator says they’re going to approve a drug.”

The researchers looked at the public information on all 222 new FDA drug approvals from 2001 to 2010. They found that 71 of them, or 32 percent, had some adverse effect not caught during the clinical trials. That stuff is supposed to gets caught when drugs go through three phases of testing, which let researchers identify the side effects and ensure the drug actually cures some ill. The new analysis, published in the Journal of the American Medical Association today, found that the rates of undiscovered adverse effects were higher in drugs taken from biological sources, psychiatric drugs, and those that received an accelerated and near-deadline approval.

These numbers didn’t surprise anyone I spoke to—prior studies of drugs approved by the Europeans Medicine Agencies (EMA) found similar safety event rates. “The fact that they could identify types of drugs more likely than others, like biologics or psychiatric drugs” was important, and should possibly lead to increased scrutiny once these drugs hit the market, Dr. Jean-David Zeitoun, first author of the EMA study, told Gizmodo. And “the fact that they found some kind of link between the deadline and the [safety] events is important.”

That last discovery perked the metaphorical ears of one FDA watchdog and a former professor of mine, journalist Charles Seife (who was not involved with the study but is working with its principle investigator on another project). “The researchers are careful to craft their findings in neutral language: these are the characteristics that appear to be associated with safety issues after the drugs and biologics get to market,” he told me in an email. “But to me, the implication is that when the FDA is under pressure to move faster, safety seems to suffer.” Of course, there could be other interpretations—he thought the study’s sample size was limited, for one.


Seife also thought that anyone hoping the FDA would approve drugs faster (like our Apprentice Host-In-Chief, for example) should take caution, as did Harvard Medical School Professor Jerry Avorn, who told Gizmodo in an email: “It’s another vital piece of data making it clear that FDA should not be in a hurry to approve drugs more quickly, or with less data, as the administration and congress seem to be advocating.”


Neither Zeitoun nor Downing thought there was evidence of foul play. Instead, they pointed to a balance of getting drugs approved in time to help people versus studying them long enough to determine all the side effects.


None of the drug companies I reached out to for comment responded. The only thing that Downing (and others) requested was more data on safety and more surveillance of drugs after they’re on the market. They really just wanted this data out there so you, the consumer, can understand that drug approval is far from a perfect science.

“The learning process doesn’t stop at approval, and we need to use high quality clinical and claims data to learn about drugs in the years after,” said Downing. “Sometimes we may learn that [drugs] are more or less effective or more or less safe.”


The FDA would not comment on this specific study, but did tell Gizmodo that they do post-market monitoring, and are reviewing the paper’s findings. Additionally, the agency pointed out that some adverse drug events may occur that people don’t report. You can get adverse event information from this link by filing a Freedom of Information Act request, if you’re worried.

But most importantly, just realize that taking medication is always a balance between risks and benefits.


“Every individual taking prescription medicines has to understand that no medication, even those over the counter or those sold as food supplements, is without risk and no drugs are assessed for long term use,” director of research at the UCLA Center for Health Policy Research Nadereh Pourat told Gizmodo in an email.

She continued: “When a doctor tells you that there are no risks associated with long term use of drugs as simple as ibuprofen, this simply means there are no studies that can tell you the risks or that the benefits may outweigh the risks.”



Former Gizmodo physics writer and founder of Birdmodo, now a science communicator specializing in quantum computing and birds


Dont use drugs that have not been studied! Opioids is a prime example.

I was given “Oxichodone” (full 5 inch bottle) after heart surgery. Horrible addictive drug that was never tested and look at our society. Terrible addicted.

Dont you hear the disclaimers at the end of commercials on television where your head will fall off and you will die. People are so gullible....Jesus!

Why are people like sheep... they simply do what ever a doctor and drug company tells them...... So willing to harm themselves at the drop of a pill.

We have brains and we dont use them... its a very inadequate evaluation of your own personal life and its future to say yes to a doctor ... who wont even remember your name.

time to get smarter ...people