Telling people about the side effects of a drug can make them sick

The phrase "First, do no harm" doesn't actually appear in the text of the Hippocratic Oath, yet it's frequently considered a fundamental principle of medicine.

At the same time, we believe firmly in the principle of informed consent, the idea that people should know the risks of a drug or medical procedure.

But sometimes, trying to follow one of these principles violates the other.

Specifically, if you tell people that they are likely to experience a number of side effects, those people are significantly more likely to experience those side effects, according to Ted Kaptchuk, a professor of medicine at Harvard Medical School and the director of the Program in Placebo Studies at Beth Israel Deaconess Medical Center.

"Almost anything can be a side effect," says Kaptchuk. And whatever side effects you tell a person are associated with a drug — dizziness, depression, dry mouth, drowsiness, insomnia, gastrointestinal trouble, foggy mind — "that's what people get," he says.

This effect is called a "nocebo" effect, a negative placebo. Instead of getting healthier because people expect to get healthier, they get sicker because they expect to get sicker.

And contrary to what some people may think about placebo or nocebo effects, these effects are very much "real" things that have an impact on people's health, affect their lives, and can be costly.

"Nocebo effects are very, very important," says Kaptchuk.

Darren Weaver

How belief can harm

"Information can be self-fulfilling," writes Rebecca Wells, along with Kaptchuk, in a paper titled "To tell the truth, the whole truth, may do patients harm: The problem of the nocebo effect for informed consent," published in The American Journal of Bioethics.

As that paper explains, there's some pretty clear research demonstrating ways that telling people about an effect could create it.

In one study involving aspirin the paper describes, some of the 555 patients involved were told they might experience gastrointestinal trouble as a result of the medication. That group was six times more likely to withdraw from the study because of gastrointestinal trouble.

In another study of 96 people testing beta-blockers, 3.1% of people who didn't know what drug they were taking experienced erectile disfunction. From the group that was told the name of the drug, 15.6% had similar troubles. The third group of patients in that study was told that erectile disfunction was a side effect of the drug they were on —from that group, a whopping 31.2% reported having those troubles. Without knowing the risk, a small number of people experienced that side effect. But with knowledge, people were ten times as likely to have a problem.

The mind is a powerful thing that can affect the body.

People told that something is going to hurt experience much more pain than people told they'll feel a cooling sensation or that they will be too numb to feel pain. If someone asks, "will it hurt," and the answer is (convincingly) "no," then the answer is no. But if they believe the answer to be "yes," they'll feel pain. As the paper explains, studies of the brain show that people experiencing this nocebo effect show brain activity indicating pain and show decreased dopamine and natural opioid activity — their brain physically is making them feel that hurt.

"There is no absolute 'truth' about a medication's side effect profile independent of what the physician says or does not say," the authors write.

Darren Weaver

So what should people be told?

Just because there's no "absolute" truth about side effects doesn't mean that all side effects are changed by the nocebo effect.

"It's only things the brain can modulate," says Kaptchuk. "The brain can make you feel more pain or less pain, more depressed or less, more or less nausea." Some drugs have side effects that exist independent of neuromodulation — potential liver damage from acetaminophen or increased ulcer risk from ibuprofen. Those things aren't changed by placebos or nocebos, meaning that people should always be told about those risks.

It's what to tell patients about those other slightly more vague side effects — which some people will experience anyway, though not nearly as many if they aren't aware of it — that is more complicated.

In the paper, Wells and Kaptchuk argue that one potential solution is "contextualized informed consent." Tell people about definitive side effects, but depending on their individual risks (some people are probably more susceptible to nocebos and placebos than others), they may not need all the details. In some cases, a doctor might tell a patient to call if they experience anything unexpected, though even this is likely to increase the chance of unwanted side effects.

Still, it's complicated. Ethically, doctors don't want to be accused of withholding information from patients — something that might open them up to costly malpractice lawsuits. And yet at the same time, they don't want to make people sicker, even if that's something that Kaptchuk says is unfortunately a side effect of the way that medicine works.

"A placebo [or nocebo] is about predicting the future at some level," he says.

NOW WATCH: NASA is bringing back one of its most experimental and dangerous programs