The mind is a powerful medicine. Given an ineffective treatment, patients can experience real health improvements by simply believing that the treatment works—the placebo effect. But this blissful delusion has a dark side: when a harmless placebo becomes effective, it becomes harmful, too, causing side-effects seen in actual therapies.
In a new study exploring this mysterious “nocebo effect,” researchers pinpoint regions of the brain that seem to be behind phantom injuries. They also assess factors—framing and price—that can increase the potency of the effect. These may be critical to designing and assessing clinical practices and trial results, they argue.
Specifically, researchers gave patients a sham anti-itch cream for eczema (atopic dermatitis) and told them it increases sensitivity to pain as a side effect—which is a side effect of real medicines, but the phony cream shouldn’t have any side effects. Nevertheless, patients not only reported more pain, but the amount of pain they reported depended on the cream’s price and packaging. The cream caused more pain in patients when they were told it had a hefty price tag and came in a brand-name-looking box, compared with when they thought it was a cheap cream that came in a generic-looking box. The researchers, led by neuroscientist Alexandra Tinnermann of University Medical Center Hamburg-Eppendorf, published the results recently in Science.
Patients reported no heightened pain when using a control cream, even though the same benign cream was used for all three types: the expensive, cheap, and control. The only differences were the prices, packaging, and the patients’ expectations. The researchers speculate that patients expected the expensive, brand-name-looking drug to simply be more effective than the one that looked like a cheap knock-off. Thus it would be more potent and have stronger side effects.
In an accompanying editorial, pain and placebo expert Luana Colloca says the findings show that nocebo effects may skew clinical trial data and patient’s adherence to drugs.
Given these effects, Dr. Colloca urges:
We should consider how to avoid them in clinical trials and practices—for example, by tailoring patient-clinician communication to balance truthful information about adverse events with expectancies of outcome improvement, exploring patients’ treatment beliefs and negative therapeutic history, and paying attention to framing (i.e., treatment description) and contextual effects (i.e., price).
For the study, Tinnermann and colleagues recruited 49 healthy participants and told them the trial was comparing the pain-sensitizing side effects of two creams, one cheap and one expensive. Twenty-five participants got the expensive cream and the remaining 24 tested the cheap one. As a reference, the researchers also included a “control cream” that would supposedly have no side effects. In reality, the whole study used only one type of fatty cream, which contained no active ingredient.
Next, the researchers primed the participants for a “nocebo effect” so they could test how the cream’s price and packaging altered said effect. To do this, the researchers used a tricky pain test. Each participant rubbed some “control cream” on one patch of their left forearm and some “test cream” (expensive or cheap) on a different patch of the same forearm. After 30 minutes of sinking-in time, the researchers wiped off the creams and applied a small device that would deliver a mildly painful flash of heat. The researchers told the participants that the device would deliver the same painful flash to each patch of treated skin.
Over the skin treated with the control cream, the device delivered a mild blast of heat that would register as a “30” on a pain sensitivity scale calibrated for each participant. But, on the test cream-treated skin patch, the device delivered a more painful blast set to register as a “70.”
This little lie fed into the participant’s expectations that the test creams (cheap or expensive) would boost pain sensitivity, while the control cream would not.
Next, the researchers repeated the pain testing 16 times with each participant over the course of a few days. But, for these tests, the researchers delivered heat blasts that all should have registered as “50” on the calibrated pain scales, regardless of cream treatment. While all of this was going on, the participants were in an MRI machine so researchers could monitor their brain and spinal cord activity.
As expected, the participants displayed a nocebo effect, collectively reporting pain levels higher than 50 on skin treated with test creams, but around 50 for skin treated with control cream.
Moreover, the participants who tested the supposedly expensive cream reported more pain than those who thought they were testing a cheap cream. And the expensive cream seemed to become more harmful over the 16 trials. In early runs, the cheap-cream testers registered around 55 on the pain scale, while the expensive cream testers landed around 60. By the end of the 16 trials, the cheap testers’ pain levels stayed about the same, but the expensive testers’ pain jumped to a 70 on the scale.
When the researchers looked at the MRI data, they found that there was more activity in spinal cord regions of the expensive-cream testers. This suggested that the testers weren’t just imagining and reporting more pain—they were actually feeling it.
The researchers also pinpointed areas of the brain that seemed to be involved in the nocebo effect overall. These are the rostral anterior cingulate cortex and the periaqueductal gray, which are involved in higher-level functioning and pain, respectively.
Together, the data suggest that the cream’s price didn’t just alter pain report but activity in the body’s brain circuitry. It “modulated coupling between prefrontal areas, brainstem, and spinal cord,” Tinnermann and her colleagues concluded. And this might give researchers hints on how to tap into and alter early pain processing.
Science, 2017. DOI: 10.1126/science.aan1221 (About DOIs).