New research has shown that, even after drugs are approved by the US Food and Drug Administration (FDA), nearly one in three of them go on to have safety issues.
That's not necessarily as bad as it sounds - the good news is that this shows the FDA is taking lifetime monitoring of its approved drugs seriously. But perhaps more concerning is the fact that so many drugs with safety issues are getting approved in the first place.
"The fact that so many new safety risks are being identified after FDA approval indicates that the FDA is taking its responsibility of ensuring the safety of new drugs throughout their lifetime seriously," said one of the researchers, Nicholas Downing from Brigham and Women's Hospital.
"However, these safety risks emerge, on average, four years after approval. This means that many patients are exposed to these medications before the risks become clear."
In particular, drugs used to treat mental health issues such as depression, and those that had been rushed through the approval process, were more likely to have safety issues down the line.
The study is particularly timely, with the government currently calling on the FDA to speed up the drug approval process.
"It shows that there is the potential for compromising patient safety when drug evaluation is persistently sped up," said lead researcher Joseph Ross from Yale University.
The research looked into 222 therapeutic drugs approved by the FDA between 2001 and 2010, and found that 32 percent of them - almost one third - were affected by a postmarket safety event.
To be clear, only three drugs were actually pulled from the market during the study period.
But 61 of them received boxed warnings - which are issued to doctors when new, life-threatening risks are identified about a drug - and 59 warranted FDA safety communications, which are put out when new, serious risks are found out.
On average, it took a median of 4.2 years after the drugs had been approved for these safety concerns came to light - which means that many people would have been unknowingly taking them in the meantime.
In particular, antipsychotics, a type of antidepressant known as selective serotonin reuptake inhibitors (SSRIs), as well as autoimmune, migraine, and erectile drugs were likely to attract new safety warnings.
So what's going wrong? To work out whether a drug is safe, the FDA makes drug companies go through three phases of clinical trials. After preclinical trials in the lab and animals, a drug then enters Phase I trials, where it's given to a small group of humans to work out whether its safe, correct dosage, and what types of side effects it might have.
Phase II trials are then conducted on another small group of humans to work out if the drug works or not. And finally, the drug is tested on a larger group to make sure it's effective in a clinical setting.
"If a review by FDA physicians and scientists shows the drug's benefits outweigh its known risks and the drug can be manufactured in a way that ensures a quality product, the drug is approved and can be marketed in the United States," the FDA explains on its site.
That sounds pretty thorough, but previous research by Ross's team revealed that the FDA approves drugs faster than its counterpart in Europe, and that most clinical trials involved fewer than 1,000 patients.
The majority of pivotal trials used to make approval decisions also lasted six months or less, which makes it difficult to identify long-term safety issues.
There's also the fact that drugs are sometimes rushed through the process if they're deemed urgent enough - and the latest study found that these drugs in particular are more likely to have higher rates of safety interventions.
One of the reasons for this, Ross told Sydney Lupkin from Kaiser Health News, is that drugs being rushed through often rely on something called 'surrogate endpoints' - which means researchers measure outcomes that are related to therapeutic benefit, such as changes in tumour size, instead of patient survival directly.
"This [finding on surrogate endpoints] has the greatest relationship to policy today," Ross told Lupkin. "In the 21st Century Cures Act, there's a push to have the FDA move to further support the use of surrogate markers … [but] they're more likely to have concerns in the post-market setting."
"This analysis highlights that there is residual uncertainty about the risks and benefits of new drugs at the time of approval, thereby demonstrating the need for all stakeholders engaged in the drug development process to commit to the generation of clinically useful information both before and after regulatory approval," said Downing.
The research has been published in the Journal of the American Medical Association.