How a Fake Side Effect Made a Placebo Painkiller More Effective

Researchers in Germany told healthy volunteers that a nasal spray might contain fentanyl, one of the most powerful painkillers in medicine. None of the sprays did. But one version carried a trace of capsaicin, the compound that makes chili peppers hot, and it left a mild burning sensation in the nose.

In a controlled study conducted by German researchers, healthy volunteers were informed that a nasal spray they were testing might contain fentanyl, one of the most potent pain relievers in clinical medicine. In reality, none of the administered sprays contained the opioid. However, one version of the spray included a trace amount of capsaicin—the active compound responsible for the heat of chili peppers—which induced a mild, localized burning sensation in the nasal passage.

This simulated side effect significantly influenced the participants’ pain perception. When subjected to painful thermal stimuli on their skin, volunteers reported lower pain levels following the burning spray compared to a completely inert spray that produced no physical sensation.

The randomized trial, led by Lieven Schenk, Tahmine Fadai and Christian Büchel at University Medical Center Hamburg-Eppendorf, was published in Brain in May 2024. Its central idea, that people may read a side effect as proof a treatment is working, earned the team the 2024 Ig Nobel Prize in Medicine.

The randomized trial, led by Lieven Schenk, Tahmine Fadai, and Christian Büchel at the University Medical Center Hamburg-Eppendorf, was published in the journal Brain in May 2024. The study’s central premise—that patients may interpret a side effect as evidence of a treatment’s efficacy—earned the research team the 2024 Ig Nobel Prize in Medicine.

A Chili Compound Dressed Up as a Drug Side Effect

The setup relied on carefully planned deception. Participants were told they were testing a fentanyl nasal spray, described as a powerful painkiller used for cancer pain, and that each spray had a 50% chance of containing the drug. They were also told that a burning feeling in the nose was a possible fentanyl side effect.

The experimental setup relied on a carefully orchestrated protocol of deception. Participants were told they were testing a fentanyl nasal spray, described as a potent analgesic used for severe cancer pain, with a stated 50% probability that any given spray contained the active opioid. Crucially, they were informed that a burning sensation in the nose was a known potential side effect of fentanyl administration.

In reality, no spray contained fentanyl. One contained a tiny dose of capsaicin, 0.15 micrograms per puff, while another contained nothing active. A device called a thermode delivered controlled heat to the skin, and participants rated the pain on a 0-to-100 scale.

In reality, none of the sprays contained fentanyl. One formulation contained a minute dose of capsaicin—0.15 micrograms per puff—while the other was entirely inert. To measure pain thresholds, a specialized thermode device delivered controlled thermal stimuli to the participants’ skin, who then rated their discomfort on a standardized scale from 0 to 100.

The burning spray worked as a convincing fake side effect. Volunteers rated side effects at 2.74 after it, compared with 0.13 after the inert spray. Nearly 95% believed the burning spray contained fentanyl, compared with about 17% for the spray that caused no sensation.

The capsaicin-induced burning sensation served as a highly convincing simulated side effect. Volunteers rated the severity of side effects at 2.74 after using the burning spray, compared to a mere 0.13 following the inert spray. Furthermore, nearly 95% of participants believed the burning spray contained fentanyl, compared to only about 17% for the odorless, sensation-free spray.

Pain ratings followed. In the first session, average pain was 26.2 after the burning spray versus 30.2 after the inert one, a statistically significant difference.

Pain tolerance ratings mirrored this psychological expectation. During the initial testing session, average pain scores were 26.2 following the burning spray, compared to 30.2 after the inert spray—a statistically significant difference that highlighted the power of sensory cues.

Of 104 people enrolled, 77 were included in the final analysis. The others were excluded for reasons that included a spray prepared with too little capsaicin, a technical problem, very low pain ratings, and failing a check on whether the deception had worked. The trial was registered in advance, and an earlier version of the paper was posted as a preprint in November 2023, before peer review.

Of the 104 individuals enrolled in the study, 77 were included in the final analysis. The remaining participants were excluded due to factors such as a formulation error involving insufficient capsaicin, technical difficulties, exceptionally low baseline pain ratings, or failure to pass a post-study verification of the deception protocol. The trial was registered in advance, and an earlier draft of the manuscript was posted as a preprint in November 2023, prior to peer review.

The effect did not vanish quickly. At a follow-up session about a week later, the burning spray still produced lower pain ratings than the plain spray. The International Association for the Study of Pain later selected the study as one of its featured papers of the week.

The psychological analgesic effect proved to be relatively persistent. At a follow-up session conducted approximately one week later, the burning spray continued to yield lower pain ratings than the plain, inert spray. The International Association for the Study of Pain (IASP) subsequently selected the study as one of its featured papers of the week.

The Belief That Strong Drugs Come with Side Effects

The burn did not appear to reduce pain directly. When the researchers analyzed the pathway statistically, the side effect raised people’s expectations of relief, and those expectations drove the drop in pain.

The physical burn itself did not appear to exert a direct analgesic effect. Instead, statistical pathway analysis revealed that the presence of the side effect elevated the participants’ expectations of pain relief, and it was these positive expectations that directly drove the reduction in reported pain.

How much the burn shifted expectations depended on a belief participants brought with them: that stronger treatments tend to cause more side effects. For people who held that view more firmly, a stinging nose looked more like evidence of a real drug.

The degree to which the burn shifted expectations was moderated by a pre-existing belief held by the participants: the notion that stronger therapeutic treatments tend to produce more pronounced side effects. For individuals who held this belief more firmly, a stinging nose served as compelling subjective evidence of a genuine pharmacological agent.

One possible alternative explanation is that one painful sensation dampened another, a process scientists call conditioned pain modulation. The authors argue that is unlikely here, because the capsaicin dose was very small and the burning felt the same across groups, though that conclusion rests on their reasoning rather than a direct test.

One potential alternative explanation is that one painful sensation simply dampened another, a physiological phenomenon known as conditioned pain modulation. The authors argue against this possibility, noting that the capsaicin dose was extremely small and the subjective intensity of the burn was consistent across groups. However, this conclusion relies on logical inference rather than a direct experimental test.

Brain Scans Point to the Body’s Own Pain Brakes

A second session, held the same day, took place inside an MRI scanner. Among volunteers who still believed the sprays might contain fentanyl, the burning spray was linked to reduced activity in the rostral anterior cingulate cortex, a region involved in processing pain and expectations.

A second experimental session, conducted on the same day inside an MRI scanner, provided neurobiological insights. Among volunteers who maintained their belief that the sprays might contain fentanyl, the burning spray was associated with reduced neural activity in the rostral anterior cingulate cortex (rACC), a brain region critical to processing both pain and cognitive expectations.

It was also linked to stronger communication between that region and the periaqueductal gray, a brainstem area that helps the body turn down pain signals. Together, these structures form part of what scientists call the descending pain modulatory system.

This fake side effect was also linked to enhanced functional communication between the rACC and the periaqueductal gray (PAG), a key brainstem region involved in downregulating incoming pain signals. Together, these structures form the core of what neuroscientists refer to as the descending pain modulatory system.

The researchers interpret this as a sign that expectation recruited the brain’s built-in pain control circuitry. Brain imaging of this kind shows associations, not proof of mechanism. The team has made its study data publicly available for other scientists to examine.

The researchers interpret these findings as evidence that positive treatment expectations can actively recruit the brain’s endogenous pain-control circuitry. While neuroimaging data of this nature demonstrate associations rather than absolute causal mechanisms, the research team has made its raw dataset publicly available to allow independent scrutiny by other scientists.

An Ig Nobel Winner with a Serious Point

The Ig Nobel Prizes, awarded each year to research that seems funny at first glance but raises real questions, honored the work at a ceremony at MIT on September 12, 2024. The prize citation credited the team with showing that fake medicine with painful side effects worked better than fake medicine without them.

The Ig Nobel Prizes, awarded annually to research that initially seems humorous but ultimately prompts serious scientific inquiry, honored the team at a ceremony held at MIT on September 12, 2024. The prize citation recognized the researchers for demonstrating that a simulated medicine accompanied by a painful side effect outperformed a completely inert fake treatment.

Schenk told Swissinfo he was “certainly very surprised” by the award. He explained that side effects can signal to people that they are receiving a real drug, which can strengthen their response.

Schenk told Swissinfo that he was “certainly very surprised” by the award. He explained that perceived side effects can serve as a behavioral signal to individuals that they are receiving an authentic, active drug, thereby reinforcing and strengthening their therapeutic response.

The study has clear limits. Participants were young, healthy adults with an average age of about 24; the pain came from heat in a lab; and the side effect was a brief, mild burn rather than nausea, fatigue, or other effects patients commonly experience. About a quarter of enrolled volunteers were excluded from the analysis.

The study has several clear limitations. The participant pool consisted of young, healthy adults with an average age of approximately 24 years. The pain tested was acute, laboratory-induced thermal pain, and the simulated side effect was a brief, mild burn rather than the chronic nausea, fatigue, or systemic discomfort typically experienced by patients. Additionally, approximately a quarter of the enrolled volunteers were excluded from the final analysis.

The findings do not mean a treatment that causes unpleasant side effects is necessarily more effective, or that side effects should be sought out. They suggest that how side effects are explained to patients may shape how they experience a treatment. Anyone with questions about side effects from a medication should speak with the prescribing clinician rather than stopping or changing it on their own.

These findings do not imply that treatments causing unpleasant side effects are inherently more effective, nor do they suggest that side effects should be deliberately sought. Instead, they suggest that the clinical context and the explanation provided to patients regarding side effects can significantly shape their overall treatment experience. Anyone with questions about medication side effects should consult their prescribing clinician rather than discontinuing or modifying their regimen independently.

Key Questions Answered

Did any of the sprays contain fentanyl?

No. One spray contained a very small dose of capsaicin to create a mild burning sensation that participants were told could be a fentanyl side effect. The other contained nothing active.

How much did the burning spray reduce pain?

In the first session, average heat-pain ratings were 26.2 out of 100 after the burning spray and 30.2 after the spray that caused no sensation.

Why would a side effect make a treatment seem to work better?

The analysis suggests the burn raised expectations of relief, especially among people who believe stronger drugs cause more side effects, and those expectations lowered pain.

Does this apply to real medications and patients?

That has not been established. The study used healthy young adults, lab-induced heat pain, and a mild sensation, so results may differ for patients with clinical pain and real side effects.

What is the Ig Nobel Prize?

An annual research award that seems amusing at first but prompts serious thought. This study won the 2024 Ig Nobel Prize in Medicine.

Source link