The Problem With Labeling: Too Many Side Effects
16 CommentsBy Ed Silverman // May 25th, 2011 // 7:03 am
In response to regulatory demands and mounting litigation, drugmakers have puffed up their product labeling to the point where the verbiage occupies numerous pages and contains a laundry list of possible side effects. Yet all these warnings have a down side, because they may undermine patient adherence and overwhelm docs as they try to find suitable treatments, according to a new study.
In fact, the study found that the list of potential side effects averaged 70 per drug, and the 200 most widely prescribed meds averaged 105 possible adverse events, leading to what the researchers labeled (pun intended) overwarning. The study, which analyzed 5,602 labels and was published in The Archives of Internal Medicine, even found one label listing 525 possible side effects.
“The idea that labeling and side effects are out of control is nothing new. The problem is a disconnect – there is no correlation between safety and number of side effects on a label,” Jon Duke, a co-author and an assistant professor of medicine at Indiana University, tells us. “What we were trying to do is quantify and put some kind of baseline measurement on this information overload…We should note that, having a lot of side effects on a label doesn’t in anyway apply toxicity, but it does often suggest labeling itself may be difficult for patients and doctors to utilize.”
In fact, Duke notes the same issue arose five years ago, when the FDA issued new guidelines that attempted to make it easier to digest labeling (see here). But since then, he says there has been no way to measure whether the effort succeeded and so the researchers developed a software program to extract adverse event data from labels.
To test their hypothesis, they examined labels formatted in accordance with the 2006 guidelines, which contained more side effects than other labeling. Was this due to new labeling formats associated with new drugs? To rule out the possibility, they repeated the comparison looking only at drugs approved since 1980 and again found a significantly higher number of side effects in the newest labeling format than in older label formats - 113 versus 72 (read the abstract here).
They also examined different classes of meds to determine if certain types of drugs were more likely to have a large number of side effects on the labeling. There were more side effects listed on labels for antidepressants, antivirals and newer drugs for Restless Legs Syndrome and Parkinson’s disease. In general, medications typically used by psychiatrists and neurologists had the most complex labels, while drugs used by dermatologists and ophthalmologists had the least, they wrote.
As a practical matter, though, Duke acknowledges that the propensity toward listing numerous side effects is unlikely to change, especially given the ongoing surge in product-liability litigation.
“There’s no question the driving factor for pumping up labels are medical-legal concerns,” says Duke, who is a practicing physician himself. “As long as something is put in a label to cover failure-to-warn concerns, then no, it won’t change. So the likelihood of seeing the amount of data poured into labels is low. But I want to at least float the notion that there are some adherence issues by creating these monster labels. And there may be a win for drugmakers – they can improve adherence with more clear and more focused labels that are less threatening in a sense.
“So I am advocating two things. First, try to come up with a way to reduce number of warnings and focus on meaningful events. Ultimately, I’m sure medical and legal exploration will be a part of that. What I’m trying to drive is making smarter drug labels, dynamic drug labels that can use patient specific information to create a personalized safety profile. Integrate electronic medical and health records. A patient has a certain history, so they may have a higher risk of certain events. The technology exists. This is where we could be headed.”
pill bottle thx to bartificial on flickr
SteveM
The obligation to clearly inform the patient of drug side and after-effects has become systemically perverse.
Physicians minimize time spent with patient to maximize patient volume. That means they off load medication counseling the pharmacists. And the pharmacists off load patient counseling to the PI or tell customers to “talk to your doctor”.
The average patient meanwhile is not equipped to parse through a dense PI to ascertain what the real risks of a a drug are.
The psycho-pharmacologist 15 minute “med check” model is especially disturbing.
Salient point
FDA, which reviews every word of every label before approval, plays a larger role in how PI’s are constructed than indicated in this post, which seems to place the onus for this development on legal departments.
If a side effect occurs in a pivotal trial, regardless of potential causation, it will be somewhere in the label. If a side effect is reported in PMS that did not occur in the pivotal trials, or seems to occur much more frequently, this will also be noted in the label.
If FDA & drug companies took the leeway to pick & choose what to report, there would be massive outrage here & on other sites about how nefarious they are.
Odd
Interesting that the other day there is an article about a side effect not being on the labeling for an OTC med and then this piece about too much information. Just goes to show that sometimes you can never be right…
industry insider
Some of the required reporting of side effects makes no sense and is actually detrimental, IMHO. For example, in the Cymbalta fibromyalgia trials there was no increased risk of suicide as compared to the depression trials. However, in the Cymbalta DTC ads for fibromyalgia the company is required to mention the suicide risk. This is probably more lawyer driven than FDA driven as I suspect is the case with the explosion in AE reporting.
Can anyone explain to me how this makes any sense?
Betsy
One never knows exactly what drives the promotion/advertising folks to determine what to say in ads, but, vis a vis Cymbalta, and the suicide warning, the suicide RATE for people with fibromyalgia is very high, so that may figure in to their legal concerns and to a connection ANY drug my have in that area.
I find myself listening to TV ads for meds, particularly my favorite one which I shan’t mention here (but you can guess what this fibromyalgia activist has in mind), and saying to the TV, after all the possible bad side effects are mentioned, “so don’t take it.”
And some side effects are NEVER mentioned despite reporting of it, like one of mine when I nearly lost my tongue from taking a famous anti-depressant and another fibromyalgic developed serious throat problems from the same thing. Did anyone care? Nope.
Jim
It may be all well and good to worry over TMI on a label, but the critical thing is to put the worst side effects there. These are often NOT on the label. Take suicide, for example. On SSRIs, the suicide warning implies that the risk of suicide/homicide stops at age 22 or whatever, which is blatantly wrong - but thanks to industry lobbying and FDA compliance, the label is silent on the all-too real risk for older adults. And for atypical antipsychotics, the suicide warning is for the condition itself as opposed to the drug - this is just wrong. Check out Seroquel and Zyprexa clinical trial results and Adverse Events reporting to learn the truth of the matter. This critical public health issue remains untold, at least on warning labels - but not in headlines.
SteveM
Re: industry insider & Betsy - Cymbalta
Cymbalta is one of the most perverse examples of failure to inform. It’s side effect and discontinuation profiles are atrocious.
Extending perverse to pathological, Eli-Lilly is offering a refund “Promise” on their Cymbalta web site:
http://www.cymbalta.com/promise/cymbaltapromise.jsp
Only Lilly does not tell the patient in the advert that no clinically proven discontinuation protocol for Cymbalta has been published. Low dose tapering formulations are not available from Lilly. And that it could take months to taper off the drug by breaking capsules apart to count beads, (which runs counter to the published product use instructions) or paying a compounding pharmacist to formulate low dose mixtures.
In the end, the discontinuation cost to the patient could be substantially greater than the 60 days out of pocket refund the patient would get back from the “Promise”. To say nothing of the serious after effects Cymbalta discontinuation can cause.
Salmon
Labeling regarding AEs is a difficult issue.
Of course neuro and psych drugs will have the most AEs as they tend to effect organ systems all over the body whereas other classes of meds don’t necessarily do so.
First regarding Salient’s statement that “If a side effect occurs in a pivotal trial, regardless of potential causation, it will be somewhere in the label.”
This is not true. For the past few years FDA has been using a rule that the AE must occur at a rate of 5% of double placebo.
Before that companies would negotiate so if they had something significant occurring at 0.9% they would propose a cut off of 1% if it was at 0.4% the cutoff would be proposed or if at 1.9% they would propose a 2% cutoff.
Then there’s a issue of serious life threatening toxicities that occur at much lower rates and toxicities that are different in different populations or with different duration of use or dose.
Listing AEs is useful if an individual patient comes in and you need to know if it could be due to the drug, but otherwise with regards to counseling, common, distressing, and life threatening AEs need to be communicated and labeled as to such to guide counseling. The total number of AEs should not be an issue when these others are considered.
John Sam Campana, RPH
The side effect profile for any drug is stupid. I’m sure it will keep some patients from taking their medication the way it should be taken. IF YOU WERE TO DO A PLACEBO STUDY AND WERE TO LIST THE SIDE EFFECTS FROM THE STUDY, YOU WOULD FIND THE SAME OLD SIDE EFFECTS. Patients would complain about dry mouth, constipation, being tired, headache, nausea, depression, dry eyes, gas, etc, etc.
Salient point
Salmon-Yes you’re right. I’m dating myself. They used to contain an all encompassing list (Cymablta, for instance, has one in section 6.11) but not anymore.
I wonder if an economic model could be devised to assign a “cost” to a given adverse event & these events could be stratified based on a combination of percentage incidence & this “cost.” If, say, a stroke costs $10K to treat & a bout of pneumonia costs $500, you would need 2% incidence of pneumonia to equal 0.1% incidence of stroke. (These figures are all make-believe, of course.)
An AE would be listed in the label of its “incidence cost” were equal to 1% of the median expenditure on health care for whatever population you’re discussing–from all adults to a specific group if the adverse event is more common for it. A “red letter” event is one where the “incidence cost” is 5% instead of 1%.
Seems like this would help account for severity since, in our economic model, severity means higher cost.
Lana Keeton
How about pharmaceutical companies just accept responsibility for the people they harm? What a novel concept…doing the right thing instead of depending on percentages and words to get out of liability for destroying lives?
industry insider
The ultimate irony here is that many in the medical community, including myself believe that fibromyalgia is not a real disease in the first place. It is just another example of what my industry has unfortunately promulgated over the past two decades as “disease mongering” when we ran out of real disease to try to cure.
Fibromyalgia, IMHO, is an example of what psychiatrists classify in the DSM-IV as a somatoform disorder, pain subtype. If you think about it, it makes perfect sense that a psychotropic drug such as Cymbalta would be effective in what lay persons call a “psychosomatic disorder”.
Betsy
That any intelligent lay or medical person in 2011 would consider Fibromyalgia a somatoform disorder is as shocking as what Big Pharma does to patients. Over 10% of the general pop. has FM, and if you want to check the reality of it, read the newly published Breaking Thru the Fibro Fog by Kevin P. White, MD, Phd. FM, a CNS disorder, can be excruciatingly painful, and I’m one patient who was born with it but has led an active, productive life in spite of it.
Why is there so much suicide? Because families and friends haven’t believed in FM, so they send the patient to a doctor who also doesn’t believe in it, and the patient is left feeling hopeless and facing a life of pain (with other unpleasant symptoms).
Moreover, there are many doctors who won’t see patients for something else unrelated to FM simply because they have FM. I am one of those patients, and I’ve dealt with over 25,000
patients who have it and have heard their very ugly tales.
Thank goodness, more doctors and more humans DO believe in FM. Unfortunately, this doesn’t reduce the pain, but it’s just nicer.
And, by the way, Cymbalta is not particularly effective in PWFM, ‘though Big Pharma would have you believe it is. And why should it be effective since this is a CNS/ANS disease and is not qualified for a spot in the DSM.
You are welcome to discuss this with I. Jon Russell, MD in San Antonio, or Kevin White, or Harvey Moldofsky, MD, the famous sleep specialist in Toronto or any number of doctors who DO know and treat FM or read a paper or two about it or Dr. White’s book. Informative people might save lives.
industry insider
Moldofsky is on the pharma payroll. care to reconsider?
Betsy
Every doctor on earth is on the Pharma payroll, but if you know medicine, you know that Harvey Moldofsky is one of the most important sleep specialists on earth, and if he has taken a bit of money from Pharma, it is really only a bit. He’s a very decent and honest doctor and has done more to understand sleep than any doctor on earth.
No-one, but NO-ONE has anything bad to say about Harvey Moldofsky.
And, by the way, I have taken every anti-depressant manufactured, because doctors thought I suffered from depression. Not one of them, including the very modern ones, (SNRI’s, etc.)works at all. Guess I’m not suffering from a somatoform disorder. But I sure as heck have FM.
A very, very well known doctor who helped write the 1990 ACR criteria, finished his role in this effort and went out and started dissing FM, saying PWFM were simply trying to get insurance reimbursements & disability, etc.,(he is said to be on the payroll of insurance companies, and embarked on a well known public campaign against FM. All the well known doctors in the FM community know his M.O. and are disgusted by it. And the details of this person’s campaign were told to me at a medical conference by a close friend and colleague of his that didn’t mind being quoted (which I’ve never done, nonetheless). I guess he was trying to tell me something I’d share. I didn’t need to. The world knows about him.
I won’t continue this discussion any more. I’m into helping to save lives not destroy them.
FDA Blogger
Here is an article that describes the FDA’s guidance for the labeling of adverse reactions on prescription drugs. As both the article I linked and this article suggest, the FDA does not adequately handle the information overload we see on medicine bottles today.
goo.gl/HjZAD