On (and off) labels

Asks Density Duck:

What do you think about off-label usage of drugs?

Broadly speaking, not a whole lot.  It’s a very, very common practice, and for the most part not problematic.

Anyone who works in pharmaceutical development should feel free to correct any misstatements I make in the comments, but let me begin with how medications are approved by the FDA.  (The FDA’s website is surprisingly unhelpful when it comes to medications for humans, but the process for approval of veterinary medications is roughly the same and is presented clearly here.)  Medications are approved after a series of clinical trials, first on healthy volunteers, then on patient groups in different settings/contexts.  If the drug proves sufficiently safe and effective, it is approved for use in the general population.

However, this process is very indication-specific.  Efficacy is assessed with regard to particular symptoms or conditions.  The medication may have numerous therapeutic effects, but if the pharmaceutical company wants to have the medication labeled as an approved treatment for all of them then the studies must demonstrate efficacy for every separate indication.  This amounts to separate trials (or a very complicated single trial).  Clinical trials are expensive, and unless there’s a clear financial benefit for having numerous clinical indications on the label, the drug company’s interests will be in getting their medication on the market sooner rather than having more expansive approved usage.  (Again, if this is in some way inaccurate or incomplete, I welcome corrections in the comments.)

Often after a medication has gone on the market, it will receive further study (sometimes sponsored by the drug company, but often not) as a possible treatment for other conditions.  Sometimes the drug company will submit an application for an additional label indication, but often sufficient evidence will accumulate that a medication has other uses that it has kind of a de facto clinical indication even if it’s not on the label.  No matter how much information there may be about additional benefits or uses of a drug, the FDA will only broaden the label if the drug company submits an application do so.  Why bother if most providers are already using that way anyhow?

By and large I think this is a good thing.  Without getting inadvisably specific with regard to my own clinical practices, I can think of several medications that I use commonly for indications that aren’t on the label.  Almost always they’re for related or clinically similar conditions, and there’s sufficient evidence to support the alternate use.  Obviously the more potentially toxic the medication, the more likely providers will be to stick to the clinical indications.  (Not a lot of off-label use of vinblastine or halothane, I’m guessing.)  Assuming a basic degree of common sense on the part of most providers, I think we should be able to use medications in ways that the FDA hasn’t specifically approved.

However, there is one area where I believe a certain class of medications are egregiously over-prescribed.  Excuse me as I trundle up onto my soapbox.  (It’s my blog, and I’ll rant if I want to.)  I am deeply disturbed by reports of antipsychotic medications being over-prescribed, particularly for children and the elderly.  These medications are heavily-marketed, expensive and are laden with side effects.  (The Better Half knows to fast-forward whenever an ad pops up urging viewers to “ask their doctor” about Abilify, because I have a tendency to scream at the TV.)  From an NPR report:

Last week, FDA staff members released reports saying the three drugs [Seroquel, Zyprexa and Geodon] do help kids, but that there are serious side effects. They can cause sedation, heavy weight gain and other problems that can lead to heart disease and diabetes.

“If these children are starting these medications at very young ages, they’re likely to be on them for many years,” says Dr. William Cooper, a professor of pediatrics at Vanderbilt University School of Medicine. “So the adverse effects, such as weight gain, increased lipids, increased blood sugar are likely to have potentially long-term and important side effects for these children.”

Cooper says bad outcomes may be worse in kids than they are in adults, and there needs to be more study of the side effects in children.

After seeing a lot of kids coming to his clinic who had been prescribed these expensive and heavily marketed medications designed for adults, Cooper did a national survey. He found that the number of prescriptions to children had increased five times over a recent seven-year period.

Over a similar period, another study found, the number of children diagnosed with bipolar disorder had increased 40 times. Still, Cooper says that doesn’t explain all of the increase in the use of antipsychotics among children and adolescents.

Cooper says he was surprised that, in most cases in his survey, doctors weren’t prescribing the drugs for serious mental illness. “Only one-third of the use was for bipolar or schizophrenia,” says Cooper.

Instead, nearly 30 percent were prescribed to kids diagnosed with attention deficit hyperactivity disorder; nearly 15 percent went to children and adolescents with no psychiatric diagnosis at all. [emphasis mine]

That last bit is grossly inappropriate.  These medications have serious adverse effects on children’s weight, cholesterol levels and blood sugar.  It is very, very difficult to help patients taking them lose weight, and I’ve seen dramatic weight gains for many patients on them (even as appropriately prescribed by child psychiatrists.)  In my opinion, providers who are not psychiatrists or specialists in behavioral pediatrics have no business writing prescriptions for these medications (with possible exceptions for rare, judicious use by generalists practicing in underserved areas).

In a similar vein, from the New York Times:

Nearly one in seven elderly nursing home residents, nearly all of them with dementia, are given powerful atypical antipsychotic drugs even though the medicines increase the risks of death and are not approved for such treatments, a government audit found.


“Government, taxpayers, nursing home residents as well as their families and caregivers should be outraged and seek solutions,” Daniel R. Levinson, inspector general of the Department of Health and Human Services, wrote in announcing the audit results.

Mr. Levinson noted that such drugs — which include Risperdal, Zyprexa, Seroquel, Abilify and Geodon — are “potentially lethal” to many of the patients getting them and that some drug manufacturers illegally marketed their medicines for these uses “putting profits before safety.”

This is nigh unto malpractice, if not barreling right into it.  Antipsychotics are NOT simple sedatives, even though they sure as hell sedate patients.  Using them to treat agitated patients with dementia at the risk of shortening their lives is grossly unacceptable.  This article made my blood boil.

For the most part, I believe quite strongly that physicians and other prescribers should have the privilege of writing prescriptions for medications based upon their own clinical judgment.  I believe most of us understand the risks of doing so, and most of us are prudent enough to know when off-label use is appropriate and when it isn’t.  There appears to be a troubling tendency to over-prescribe antipsychotic medications (and, as a topic for a later post, psychiatric medications in general) by far too many providers, but for the most part I think the medical community can operate safely outside the FDA’s imprimatur.

Russell Saunders

Russell Saunders is the ridiculously flimsy pseudonym of a pediatrician in New England. He has a husband, three sons, daughter, cat and dog, though not in that order. He enjoys reading, running and cooking. He can be contacted at blindeddoc using his Gmail account. Twitter types can follow him @russellsaunder1.


  1. And, relevant to this post:


    “FDA also advises manufacturers of products labeled as RUO or IUO that if they discover that a clinical laboratory to which it sells its IUO or RUO device is using these IUO or RUO labeled products for a non-investigational diagnostic use, “it should halt sales for such use or comply with FDA regulations for IVD products, including premarket review requirements.” ”

    Interesting. A trial balloon for coming down hard on off-label use of other things as well?

    • I am deeply skeptical that the FDA has the authority, the resources or the motivation to police whether most drugs are used on- or off-label. It would involve accessing myriad confidential medical records, first of all. Secondly, I do not believe the FDA has the authority to dictate how providers treat patients, by and large. I take a certain calculated risk by prescribing medications off-label, but not of being cited by the FDA. If there were to be an adverse outcome, theoretically I have less cover in a malpractice suit if I use a medication in a manner other than how I’m “supposed to.”

      I’m sure the FDA pays closer attention to how some medications are being used than others, but I don’t think off-label use of medications is sufficiently problematic to warrant action on their part, even if it IS within their ambit.

Comments are closed.