The trouble with new drugs

Robert H. Shmerling, MD

Senior Faculty Editor, Harvard Health Publishing

When a drug is approved by the FDA, it may seem like it’s only a matter of time before some unexpected side effects are discovered. Perhaps it seems that way because it’s true! According to a study of all drugs approved between 2001 and 2010, the FDA announced alerts, warnings, or recalls on about one-third of them in the years after their approval.

Some of the side effects were minor and easily managed. For example, there might be a warning to avoid taking a new medication at the same time as another medication. But sometimes the “side effect” is death. And that’s the case with a new warning about the gout drug febuxostat.

An ancient disease, new treatments, new concerns

Gout can cause joint pain that’s excruciating and debilitating, but it is highly treatable and even preventable. In recent years, a number of new drugs have been approved, including febuxostat. Although it was approved in 2009, some of the early studies raised concerns that people who took febuxostat might have a higher rate of cardiovascular events such as heart attacks. As a result, the FDA required the drug’s maker to perform additional research after its approval.

The additional research confirmed that the safety concerns were justified: when compared with people taking allopurinol, an older gout drug, researchers observed a small but significant increase in the rates of cardiovascular death, and death from all causes, among people taking febuxostat. This led the FDA to issue a “black box warning” — the strongest action short of a recall — about the risks of taking this drug and how allopurinol should be tried first.

If you’re taking febuxostat now, talk to your doctor about this new development and whether you should continue it, stop it, or switch to another treatment.

What’s a black box warning, anyway?

Depending on the level of concern, the FDA has several ways to communicate new warnings regarding the safety of an approved drug:

  • Medication guides. These are printed reports routinely provided to patients by pharmacists for certain medications, intended to reduce serious side effects.
  • Prescription drug labeling. This is the package insert that provides detailed information about an individual drug. These often contain a section for patients using nonmedical terminology, though most of the content is in “medicalese” and intended for health professionals.
  • Drug Safety Communications. These include Public Health Advisories and other updated drug safety alerts.
  • Boxed warning. Often called a black box warning, this information appears at the beginning of the package insert and is intended to call attention to serious or life-threatening side effects. That’s what was added to the package insert for febuxostat.
  • Drug recalls. Although they may come at the request of the FDA, most drug recalls come from drug makers as a voluntary action. They may be permanent or temporary, depending on the problem.
  • Drug withdrawals. Even more rarely, evidence comes to light that convinces the FDA that the risk of taking a drug is simply too high when compared to its potential benefits. In these cases, the drug’s approval is essentially reversed, and the FDA requires the drug be taken off the market.

(You can find out more about how the FDA monitors the drug approval and post-approval process here.)

Why does this happen so often?

As I was beginning to write this post, yet another alert was released by the FDA. This time it was for tofacitinib (Xeljanz), a medication approved in 2012 to treat rheumatoid arthritis (RA). A higher than expected rate of blood clots in the lungs and death was found among the drug’s users. Some important caveats are worth noting, though: the trial was required by the FDA; it enrolled people with RA who were 50 and older and had at least one cardiovascular risk factor; the dose was 20 mg/day, roughly twice the currently approved dose for RA. (However, this dose is approved for people with ulcerative colitis.)

Why aren’t more of these problems detected before drugs are approved?

There are a number of reasons, but the biggest one has to do with the difference between clinical trials required prior to a drug’s approval and real world usage. A trial may exclude people with heart disease, or who are elderly, or may include only men or only women — but once the drug is approved, men and women, people with heart disease, and older individuals may start taking the medication. New problems may emerge in this expanded group of people that weren’t detected before.

In addition, there’s a numbers issue. Clinical trials may include hundreds or even thousands of study subjects, but to detect a small risk it may take tens of thousands of individuals, or even more. For these reasons, post-marketing surveillance plays a major role in ensuring the safety of medications approved by the FDA.

While the high rates of alerts and warnings may make it seem like the FDA is doing a lousy job of making sure drugs are safe before approving them, there’s another way to look at this. The high rate of alerts and warnings could mean that the post-approval monitoring is working to detect previously underappreciated problems and to publicize them.

Here’s where you come in

If you develop a side effect from a medication you’re taking, let your doctor know. You may need to stop taking the drug to avoid more trouble. But you can also report your experience to the FDA’s MedWatch Safety Information and Adverse Event Reporting Program (here or at 800-332-1088). You could be helping to identify an important and previously unrecognized problem with the medication.

While new drugs can truly be lifesaving, sometimes the latest breakthrough drug turns out to be worse than older treatments or the disease itself. Despite the years of research required to get a drug approved, important risks can be missed. We have to rely on post-market surveillance to detect these risks. You can help.

Follow me on Twitter @RobShmerling


  1. Bob

    I sympathize with the difficulties that new drugs are presented with and the inherent unknowns they create and then must somehow be managed. Nevertheless the problem does not go away and will get predictably worse as more and more drugs are introduced and adopted by MD’s. The number of reported issues and categories can be assumed to be understated as no adequate reporting is required by the patient and a reluctance to take drugs off the market unless there is solid evidence or a safer drug. I would assume that a person taking the drug has not sat down and read all the known side-affects and possible ones that that can occur due to poly pharmacy which now makes the possible side-affects multiplied again. But who is to be responsible for this as I think the family doctor is over worked and under pressure already and the extra paper work and surveillance to keep track and monitor historical data of each patient would be a monumental task. Yet perhaps some responsibility should be transferred to the patient to record and report on-line themselves for each drug and any red flags then sent to alert the family doctor. I can not see any body else doing it given the scope and size of the problem. But in doing this there may be some privacy issues and trust issues so I think it must be a volunteer system. I am sure clever programmers could create such a system or else you’d better come up with a non drug solution that has passed all the clinical trials as required by the FDA. Good luck there. But this article was a fair analysis of the problem but by no means a solution nor meant to be.

  2. azure

    In other words, everyone who takes a new drug is a guinea pig. Particularily w/regards to long term effects as drug trials last a relatively short time.

    Reporting side effects is a great idea, an even better idea is for health care providers to ASK if the person’s experiencing side effects from a new medication, if so, what are they (i.e, listen to the patient’s answer for as long as is necessary to elicit a full answer) and finally, actually take note of & report what the patient has said. Not many health care providers I’ve been to do so. In reality, when I once reported what to me was a significant side effect (sudden onset of a severe headache in a young /late 20’s person who rarely experienced headaches, then they were usually sinus headaches, which this was not) I was told, oh, just keep taking X medication, it’ll probably go away.
    My response was, no, what’s going to happen is that I’ll be given a prescription for a different medication. I was well enough to see the MD (not the provider who told me to “just keep taking the medication”), I was prescribed a different med & I had no more headaches.
    In addition, some cancer medications are known to have unpleasant side effects, and most MDs will tell their patients to keep taking the medication anyway, for as long as the person can tolerate the sx, if the med appears to be doing what it’s supposed to do. Seems to be particularly true both of chemo, and “living with cancer” meds like Ibrance & similar meds. Or that seems to be the experience of the people I’ve known who unfortunately have to take those meds (and their insurance pays for the bloated medication cost).

    My opinion is the high rate of warnings, etc. on new meds means that the FDA is too easily swayed by the corporations it’s supposed to regulate. It approves too many medications w/questionable trial structures and questionable results.

Commenting has been closed for this post.