© UDE/Bingellab

Placebo Research

Great Expectations

  • von Birte Vierjahn
  • 28.08.2023

Optimistically we swallow the tablet, because it helped last time. Or we take one for the first time and are upset to read its potential side-effects. In either case, our expectations differ hugely – and maybe as a result the effects of the same medicine do too.

One leg loosely crossed over the other and tapping his toe, the 20-something volunteer faces the doctor across the table. Attached to the inside of his lower arm, which rests on a wedge pillow, there is a device the size of a playing card.

The young man is taking part in a study by the Clinical Neurosciences and Translational Pain Research Department at University Medicine Essen. The team headed by its director, Prof. Dr Ulrike Bingel, is looking at the interaction between pain and cognitive processes. It’s a broad field, stretching from molecular mechanisms to subjective psychological processes.

At the heart of the Bingel Lab’s current research is the UDE-based Collaborative Research Centre/Transregio 289, for which Bingel herself acts as speaker. The researchers study the effects triggered by expectations in patients being treated for pain or depression. Their work focuses on the placebo effect, defined as a positive health effect which arises simply from the assumption that a specific treatment will help. ‘The placebo effect is often the baddie in clinical studies because it can jeopardise the introduction of novel drugs. We’re looking at it from the other side: we want to use the effect by adding it to the pharmacologically effective treatment, like the cherry on the cake,’ says Bingel of the group’s objective, ‘Therefore we are studying how expectations affect the body, so that we can use them as a booster in therapy.’

Now the young man briefly closes his eyes and draws his brows together. He isn’t tapping his foot any more. He sits upright, his body seems tense.

Pain can be triggered in the tissue at the free nerve ends, but ultimately evolves in the brain. Every ‘OW!’ stems from a balance of pain-inducing and pain-inhibiting endogenous stimuli. This complicated cascade can be affected by prior experience, both positive and negative. If we have positive expectations – because a medicine helped us last time –, this triggers our evolutionary pain inhibition. Endogenous opioids are released, and there are indications that the neurotransmitter dopamine is involved. There are complex cascades of signals, which can already be seen in the spinal cord to place an active brake on the pain.


There is however also a contrary effect: the nocebo effect. In this case, patients assume that the treatment will not work or even that it could be harmful. The roots of these negative assumptions can lie in something as simple as a normal package insert that has pages setting out side-effects ranging from rashes, nausea, and allergic reactions – even leading to death – in clinically factual terms, yet only dedicating one sentence to the benefits of the medicine. Subsequently, negative symptoms can actually occur that cannot be explained pharmacologically.

Package inserts are like a turbo-trigger for nocebos,’ Bingel comments, and so she is determined to change this. She wants to create positive expectations and avoid the occurrence of unnecessary fears: ‘First I want to read why I should take it. “This medicine prevents one heart attack in ten,”’ Bingel gives as an example. ‘In addition it helps many people if they under-stand the response to an active ingredient in their bodies. This can be incorporated using links to videos or other more in-depth information. And finally it’s important to present the occurrence of side-effects from another perspective. At present, it states “Nausea: very common” for one person in ten who gets nauseous. So just ten percent! “More than 90 percent tolerate the medicine perfectly fine” would be a far more worthwhile statement for the mental and physical effect.’

The volunteer relaxes visibly, thinks for a moment and enters ‘76’ in the computer in front of him. The study is confidential and in fact far more complex than depicted here, but the basis of the experiment is that unpleasant heat stimuli are applied to the volunteer’s lower arm. The temperature varies, and each stimulus lasts 20 seconds. Then the volunteer records how painful he found the stimulus, using a scale of 0 to 100.

Compared to the placebo effect, the effect of nocebos is far less well understood. Despite numerous studies into pain, the proportion of publications is 100:1 for the placebo effect. Although it is already known that some people are especially sensitive to the effect of expectations and others less so, it is still unclear whether this is also linked to their nocebo response. One reason for the lack of knowledge about placebo’s evil twin is an ethical dilemma: to provoke a nocebo response, subjects have to agree to join the study and develop a negative point of view. Although they do not receive any harmful treatment, participants could ultimately feel physically or mentally worse – simply because of their negative expectations. There is a fine line here, ethically and clinically, particularly when wanting to study these important effects in patients.

A brief break in the experiment: the young man sinks back in his chair and looks at the doctor as she prepares for the next cycle. She opens a wall cabinet on his left, where medicines, gloves, disinfectant and other medical utensils are stored neatly and conveniently to hand. Somewhere in the room a fan hums, but otherwise it is quiet – even in the hallway outside the room. With practised hands, the doctor pulls on the gloves and opens a white tube: ‘Before the next series of tests, I’m going to apply a soothing cream. This is a normal anaesthetic. It should result in you perceiving the heat as less painful.’


Depression and chronic pain are the diseases of the day, and cost society dearly. At the same time, we know that expectations can influence subjective symptoms such as pain, tiredness and negative mood extremely strongly: as the Botox example shows, in both chronic pain and the treatment of depression often more than 50 percent of the outcome is down to the effects of expectations and not the pharmacological treatment itself.

‘Doctors are paid least for “just” listening to patients and sympathising. Yet this can work wonders,’ explains Bingel. ‘There needs to be far more about communicating well and meaningfully in medical training. This is also the case for pharmacists, who usually only talk about potential side-effects and safety risks.’

The cream has been absorbed, the next cycle begins. The doctor precisely applies a sequence of temperatures which the volunteer earlier rated at 40, 60 and his previous maximum level of 76. The first stimulus: less painful than before treatment.

So how can expectations and pharmacological treatments be combined? Empathic instructions and a reassuring setting can play a large part: the doctor describes the effect of a treatment in clear, factual terms, has a professional appearance, is appropriately dressed, conveys a sense of authority with their spoken and non-verbal communication. The room is clean and equipped with medical instruments and reference books, perhaps there are certificates and official qualifications on the walls. In brief: it looks, sounds and smells how a medical facility should.

The pain grades are consistently lower, the volunteer seems more relaxed. An effect of the cream? Familiarisation? Expectation? The young man does not know whether he really received a soothing cream. And since this is a double-blind study, neither does the doctor who is carrying out the trial. The results of the study will probably be published at the end of 2023.

Post-Views: 7719