Are Warnings About Drug Side Effects Actually Making Us Sick?

By Guest Blogger | July 24, 2012 12:53 pm

Steve Silberman (@stevesilberman on Twitter) is a journalist whose articles and interviews have appeared in Wired, Nature, The New Yorker, and other national publications; have been featured on The Colbert Report; and have been nominated for National Magazine Awards and included in many anthologies. Steve is currently working on a book on autism and neurodiversity called NeuroTribes: Thinking Smarter About People Who Think Differently (Avery Books 2013). This post originally appeared on his blog, NeuroTribes.


Patient receiving a vaccinePhoto by Flickr user Noodles and Beef

Your doctor doesn’t like what’s going on with your blood pressure. You’ve been taking medication for it, but he wants to put you on a new drug, and you’re fine with that. Then he leans in close and says in his most reassuring, man-to-man voice, “I should tell you that a small number of my patients have experienced some minor sexual dysfunction on this drug. It’s nothing to be ashamed of, and the good news is that this side effect is totally reversible. If you have any ‘issues’ in the bedroom, don’t hesitate to call, and we’ll switch you to another type of drug called an ACE inhibitor.” OK, you say, you’ll keep that in mind.

Three months later, your spouse is on edge. She wants to know if there’s anything she can “do” (wink, wink) to reignite the spark in your marriage. She’s been checking out websites advertising romantic getaways. No, no, you reassure her, it’s not you! It’s that new drug the doctor put me on, and I hate it. When you finally make the call, your doctor switches you over to a widely prescribed ACE inhibitor called Ramipril.

“Now, Ramipril is just a great drug,” he tells you, “but a very few patients who react badly to it find they develop a persistent cough…” Your throat starts to itch even before you fetch the new prescription. Later in the week, you’re telling your buddy at the office that you “must have swallowed wrong” — for the second day in a row. When you type the words ACE inhibitor cough into Google, the text string auto-completes, because so many other people have run the same search, desperately sucking on herbal lozenges between breathless sips of water.

In other words, you’re doomed. Cough, cough!

Coughing man Photo by Flickr user Jeff and Colin

What’s going on here? Just medicine-as-usual in a world where valuable drugs have annoying side effects and conscientious health professionals do their best to protect their patients from unpleasant (and potentially litigious) surprises? Sure. But a provocative new report by Winfried Häuser, Ernil Hansen, and Paul Enck in the journal of the German Medical Association suggests that the side effects of some drugs, and the discomfort of certain medical procedures, may be inadvertently intensified by doctors and nurses trying to keep patients fully informed of the consequences of their medical care. The culprit behind this phenomenon is the nocebo effect.

You can think of the nocebo effect as the evil twin of the placebo effect — the body’s healing response to the act of taking a pill or receiving medical care, even if the pill itself is inert. The most familiar example of the placebo effect is what happens in trials of experimental drugs. One group of volunteers is randomly assigned to take the drug in question; another group is assigned to take placebo — a fake drug designed to look just like the real one. Neither the volunteers nor the staff know which group is which. If the drug group improves significantly more than the placebo group, the drug is judged to be effective. This kind of test — the double-blind, placebo-controlled trial — has been the gold standard of drug development in medicine for half a century.

In real life, gauging the effectiveness of a new medication is not quite that easy. In 2009, I wrote a widely-circulated article in Wired magazine about a mysterious increase in placebo effects in trials in recent years that is making it harder for Big Pharma to bring new drugs to market. I explored some of the reasons that might be happening in the article.

A placebo, you might say, is an ersatz drug that makes you feel better, while a nocebo is a fake drug that makes you feel worse. Of course, in both cases, it’s not the pill that’s doing the work; it’s your own body, responding to the social context in which you take the pill. If a skilled doctor with kindly bedside manner tells you that drug X will reduce the inflammation of a minor injury, it often will — even if the drug itself is nothing but a capsule full of lactose, milk sugar. One of the astonishing things we’ve discovered about the placebo effect in recent years is how wide a range of ailments can be ameliorated by it, at least temporarily — from chronic pain, to high blood pressure, to inflammation, to depression and anxiety, to sexual dysfunction, to the nausea and vomiting caused by chemotherapy. Perhaps unsurprisingly, it turns out that the nocebo effect is equally capable of making you feel more miserable, in a similarly broad range of ways.

One of the most interesting findings in the new report from Germany is about the underappreciated — and under-studied — role of nocebo effects in clinical trials.

If you tell a group of trial volunteers that they’re testing a new drug that may relieve the pain of migraines, a significant number of volunteers will experience pain relief after taking the drug — even if they’ve been randomly assigned to the placebo group and are receiving nothing but sugar pills. The placebo effect in action.

But here’s where it gets interesting. If you tell the volunteers that the side effects of this new medicine may include dry mouth, tingling in the hands and feet, and slight dizziness, some volunteers will experience precisely these side effects — in both groups. In fact, some volunteers who are taking nothing but sugar pills will be made so uncomfortable by these symptoms that they will choose to drop out of the trial early.

Pills for pain
Photo by Flickr user MadcowIV

The relevance of the nocebo effect is not limited to clinical trials. A particularly nasty demonstration of its power can be provided by patients who are being treated for Parkinson’s disease with small doses of electrical current delivered to certain structures in the brain. This technique, known as Deep Brain Stimulation (DBS), has been shown to be effective in controlling tremor and other motor issues in patients who are resistant to drugs. But if you tell a DBS patient that the current has been switched off — even though it remains on — their coordination and other motor functions will abruptly decline until you tell them that the juice has been switched on again. Classic nocebo.

The German researchers cite many other examples. In one study, 50 patients with chronic back pain were randomly divided into two groups before a leg flexion test. One group was informed that the test could lead to a slight increase in pain, while the other group was instructed that the test would have no effect. Guess which group reported more pain and was able to perform significantly fewer leg flexions?

Another example from the report: Patients undergoing chemotherapy for cancer treatment who expect these drugs to trigger intense nausea and vomiting suffer far more after receiving the drugs than patients who don’t. That’s not to say that this nausea is imaginary, purely psychological, or “all in their heads.” Like placebo effects, nocebo effects are physiological, not just mental. Researchers like Franklin Miller at the National Institutes of Health and Tor Wager at the University of Colorado are learning that nocebo effects are mediated by neurotransmitters like dopamine, endogenous opioids, and cholecystokinin. Telling a volunteer that the scented Vaseline you’re slathering on their arm is a pain-intensifying gel triggers a cascade of chemicals in the body that magnify sensations of discomfort.

Skull made of pills
Photo by Flickr user markkilner

How evil can the evil twin of the placebo effect get? Even setting aside anecdotal accounts of evil eyes and “voodoo death” (first reported by a doctor named William Cannon in 1942), nocebo effects can kill, or at least accelerate the progress of disease. In traditional Chinese astrology, some potentially fatal maladies like cancer are associated with certain birth years. A 1993 study of Chinese-Americans who came down with these diseases found that they were more likely to die faster if they were born in an “ill-fated” year associated with the disease — but only if they believed in traditional Chinese astrology. (In these tragic cases, a dose of skepticism could have at least staved off the inevitable for a little while.)

In recent years, the phrase “informed consent” has become a mantra in public health, which is good news for many reasons. But what if the lengthy litanies of side effects we’ve become accustomed to skimming over — on printouts at the pharmacy, in ads for the latest antidepressants — are self-fulfilling prophecies? What if our well-intentioned desire to prepare patients for the worst increases the probability that the worst will happen?

Even the specific words employed by a doctor can influence the course of a patient’s treatment. Some particularly dreadful word choices cited by the German researchers include doctors telling patients:

You must strictly avoid lifting heavy objects — you don’t want to end up paralyzed.

You are a high-risk patient.

We’re putting you to sleep now. It will soon be all over.

Though the German report focuses on clinical practice and bedside manner, it’s hard not to think about the broader implications of nocebo effects in daily life. If you’re ever been told twice in one afternoon that you “look a little worn out,” you know that it’s a miracle if you don’t feel utterly exhausted – if not physically ill — by the end of the day. Knowing about nocebo has made me much more careful when making observations about a friend’s physical condition.

Gluten allergy icon from Kids Aware

The German report also points out that nocebo effects deserve much more study. One promising avenue for research might be examining the role of negative messaging in large-scale social phenomena like the apparent increase in food allergies in recent years, or the surge in popularity of gluten-free diets. Having friends with celiac disease and autism, I’m fully aware that some people are truly gluten intolerant, and that banishing the dreaded protein from the menu can avert serious medical problems. But the new report makes me wonder if the flood of anti-gluten “awareness” messages on social networks isn’t lengthening the lines for pricey GF pasta, crackers, and doughnuts at my neighborhood Whole Foods Market.

The conflict between the need for informed consent in medicine and our bodies’ propensity to take social factors into consideration when calculating our own health status is not an easy problem. In recent years, the pendulum has swung so far in the direction of more-bad-news-is-better that it seems almost like heresy to suggest that some of this paternalistic advice may actually be making us sicker.

The new generation of placebo and nocebo research is teaching us that how we feel is highly dependent on the feedback we get from the people around us, particularly from trusted health professionals. The next time I get a headache, you’ll forgive me if I don’t re-read the label on the aspirin bottle warning me about the possibility of hives, facial swelling, asthma, shock, Reye’s Syndrome, nausea, vomiting, severe stomach bleeding, allergic reactions, and distressingly vague “changes in behavior.”

I’ll just take two and hope for the best.

  • Andrew

    Have studies been done about whether or not it is possible to remove the placebo / nocebo effect by describing it to patients? Like saying: “In some cases this pill can cause headaches, but often times the headaches are a result of patients expecting negative side-effects, not the pill itself. So just think positive.”

    If telling someone they will fall ill causes them to feel ill, then warning them that it probably won’t happen could prevent it, right?

  • Sharon

    I truly subscribe to this notion as I feel I’ve done this before…I read the side effects of a prescription and immediately after starting the course of treatment, I see the side effects popping up. One thing that has made a huge difference…awareness that we are doing this. Ignorance is bliss is my mantra with prescription meds. If there is anything very serious as a possible side effect, I ask my pharmacist/doctor to tell me me about it but I purposely ask them to forgo the description of each small side effect like nausea etc.. and miraculously, I no longer suffer from any of those side effects. Informed consent with limits.

  • Anthy Hellmers

    How about keeping it as simple as saying; please let us know how you are feeling when on this placebo. Do not give an example. If words lead to pre-conceived feelings then don’t make statements that could effect the results. Let the user express their symptoms good or bad in context to themselves. The user should keep notes and share them with their doctor. That is the only way a fair consensus of the placebo nocebo can be valued. I wonder how many drugs have made it to market due to mind manipulation?

  • Eva Beaton

    maybe this is why I seem to be the one in one hundred, five hundred, a thousand, that gets whatever reaction or problem. Although I first heard the phrase, “this is unusual it only happens to 1 in a 100” when I went back to the dentist, with an abscess that had developed in an air bubble under my upper lip, that had come up during a root canal. They had to clean the abscess without freezing (can’t freeze infected gums) and I bled like a stuck pig, but maybe it triggered off something in my brain that makes me the one! I have had this phrase said to me, several times since by drs, I didn’t even know that I could have been the one to have whatever happen, or to get whatever bug etc! But I’m that one!

    I definitely don’t read the papers I get detailing every last possible thing, I have constant nausea due to inner ear troubles anyway, do I need it increased because the drug I’m taking for my diabetes, high blood pressure, high heart rate, high cholesterol or my asthma might make me nauseated?? I don’t think so.

  • Kim

    I truly did have the blood pressure medicine cough described in the article! Since I’m prone to respiratory problems anyway, I went to the doctor complaining of a cold I couldn’t kick and found out that it was the medicine. So basically I suffered for a couple of weeks when changing the med quicker would have been better, had I but known to be aware of the symptom. Perhaps someone should develop a quick test to show who’s susceptible to the effect and who’s not?

  • Byron Winchell

    “Extraordinary Popular Delusions and the Madness of Crowds” a history of popular folly by Scottish journalist Charles Mackay, first published in 1841, modern version of.

    Honest, I take so many prescriptions that I avoid reading about new ones until a week or so has passed. I’m that suggestible. Yes, I listen to whatever the doctor says to avoid dropping dead straight away.


The Crux

A collection of bright and big ideas about timely and important science from a community of experts.

See More


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar