The Nocebo Effect: How We Think Ourselves Sick, According To Psychiatrists

A few months ago, I (Colloca) was asked by a reporter to comment on the role of the nocebo phenomenon in Havana Syndrome. The reporter was referring to a set of symptoms experienced mostly by government officials and military personnel that first occurred at the U.S. Embassy in Havana.

"Is this a nocebo effect?" the reporter asked. I explained that I had never heard about the disease but wanted to learn about it. I did some quick research, and the syndrome reminded me of other mass psychogenic illnesses (see Chapter 12), whereby people in a group may feel sick as a result of thinking that they were exposed to something dangerous—even though there is no real noxa, or harmful agent.

Nocebo effects are adverse outcomes due to negative expectations. The clearest example of nocebo effects come from placebo treatment in clinical trials. Up to 19% of adults and 26% of older adults report adverse effects when they are given placebos in clinical trials. A quarter of those given a placebo in clinical trials discontinue their participation because of adverse effects. This discontinuation can negatively impact clinical trial enrollment and the ability to retain participants in clinical trials.

We are now beginning to understand some of the mechanisms—psychological and biological—that give rise to nocebo effects. Studies in both laboratory and clinical settings, some of which are described in other chapters, document the important role of information and expectations in generating nocebo effects. For example, asthmatic patients who were given a medication called a bronchoconstrictor, which narrows certain airways in the lungs, but who were told that the treatment they received was a bronchodilator (a medication that widens those airways) showed a widening of the airways. The opposite is also true: Patients with asthma showed a narrowing of the airways when the bronchodilator they were given was described to them as a bronchoconstrictor.

Along these lines, another paradoxical nocebo response applies to muscle responses. Participants who were told that they had been given a muscle stimulant (a medication that increases muscle tone) experienced muscle tension even though in reality they had received a muscle relaxant (a medication that decreases muscle tone). 

Nocebo effects can also affect Parkinson's disease, a condition that causes, among other symptoms, bradykinesia, an extreme slowness of reflexes and movements. Often when medications do not work to reduce these symptoms, Parkinson's patients undergo a neurosurgical procedure called deep brain stimulation involving the placement of electrodes connected to a neurostimulator, which can be turned on and off to deliver electrical impulses. 

In the hospital, relieving pain after surgery is critical. In a landmark study, pain treatments were delivered into the bloodstream through an automatic pump, but the patient was unaware of the timing of the infusion. 

Patients underwent thoracotomy, a surgical procedure, in order to remove lung cancer. In the postoperative period, levels of pain and anxiety peak. Pain is controlled with opioids and non-opioid painkillers. When morphine, an opioid, was interrupted openly (that is, patients were told about the interruption), pain increased substantially. On the contrary, when the interruption of morphine was undisclosed (hidden), the level of clinical pain remained consistently low, as if the opioid had continued being pumped into the bloodstream.

A revolutionary discovery in nocebo mechanisms came with the advent of brain imaging techniques that can depict changes in the brain associated with nocebo effects and shed light on their neural signature. Using the open-hidden procedure described above, a pioneering study indicated that the effects of a strong narcotic such as remifentanil can be completely blocked by the suggestion that the infusion of the drug has been stopped.

 In one study, remifentanil was continuously infused in the participants' veins, but the participants were told that the drug was discontinued. As expected, participants experienced increased pain (hyperalgesia) after being told the drug had been discontinued. Brain activity was measured with functional magnetic resonance imaging (fMRI). When participants experienced nocebo hyperalgesia, brain activity increased in a part of the brain called the hippocampus, which is involved in learning and memory. 

Nocebo effects also depend on the order in which treatments are delivered.

Luana Colloca and Fabrizio Benedetti conducted a study in which participants were assigned to one of two groups. Group 1 received a treatment presented as effective, and Group 2 received the same treatment, not presented as effective or ineffective but taken after an ineffective one. The groups differed in the degree of pain reduction (49.3% versus 9.7%, respectively). Similarly, Simon Kessner and colleagues found that starting a new medication after an unsuccessful medication created nocebo pain increases. The brain processes therapeutic failure with an activation of the posterior insular cortices, areas of the brain related to feeling pain.

Interestingly, the price of a medication described as inducing pain is also associated with nocebo effects and brain activation. Alexandra Tinnermann and colleagues showed that marketing a cream (in reality a sham cream) as an expensive one elicited higher nocebo hyperalgesia than a control cream that was marked as less expensive—that is, the expensive medication increased pain. This effect was mirrored by an increase of activity in brain areas that are hubs for the overall experience of pain, suggesting that nocebo effects have a critical role in processes that signal underlying pain sensation.

Nocebo effects aggravate not only pain but also other symptoms, such as itchiness and shortness of breath. Itchiness can be increased by negative expectation, and this worsening is paralleled by more communication between the insula and the periaqueductal gray—two regions of the brain involved in altering the sensation of pain.

At the molecular level, nocebo effects have been linked to the release of a hormone called cholecystokinin (CCK). CCK acts on the brain to increase anxiety. Cholecystokinin also plays a role in temperature regulation. CCK increases during times of heightened anxiety and appears to have a role in nocebo responses.

In an early study that explored the effect of suggestions of hyperalgesia, the researchers measured two hormones that rise during stress, adrenocorticotropic hormone and cortisol. The suggestions of pain increased not just the pain itself but also adrenocorticotropic hormone and cortisol. Interestingly, when participants were given a drug called proglumide, which blocks the effects of CCK, pain was also blocked. This result indicates that CCK is a critical component of the nocebo effect. In general, this research points to a relationship between nocebo effects and how anxiety and stress are regulated.