Expectation Over Sensation: How Unblinding Undermines Placebo Effects in Pain Research

A mild nasal burning sensation enhanced the perceived efficacy of a sham painkiller, but only while participants believed they might be receiving fentanyl. Once half the volunteers were informed the spray contained no active drug, the identical burning sensation ceased to provide pain relief in the subsequent session.

This pattern emerged from the second phase of a randomized trial conducted by neuroscientists at University Medical Center Hamburg-Eppendorf, published in Brain in May 2024. The findings highlight a potential vulnerability in standard drug testing: when a medication’s side effects reveal to patients that they have received the active treatment, conventional placebo comparisons may inflate estimates of the drug’s true efficacy.

The Twist in Session Two

In the initial session, the 77 analyzed volunteers were told each nasal spray carried a 50% chance of containing fentanyl. In reality, neither spray contained the opioid. One spray delivered a minute dose of capsaicin—the compound responsible for chili pepper heat—while the other was entirely inert.

Before the second session, conducted the same day inside an MRI scanner, participants were randomly divided. One group of 42 was told the procedure would simply be repeated. The other 35 were fully debriefed: no fentanyl was present, and the pain differences they experienced stemmed solely from variations in the heat applied to their skin.

The physical sensation remained unchanged across groups. Side-effect ratings averaged 2.71 after the capsaicin spray and 0.05 after the inert spray, with no significant difference between the informed and uninformed cohorts.

What shifted was the meaning attributed to that sensation. Among volunteers who maintained the belief they might receive fentanyl, pain ratings on a 0-to-100 scale averaged 38.7 after the burning spray versus 43.0 after the inert spray. Among those who had been told the truth, pain scores were virtually identical: 45.0 after the burning spray and 43.5 after the inert spray—effectively no benefit.

This between-group difference was statistically significant, albeit modest. The sensation alone did not reduce pain; the belief attached to it did.

Brain imaging aligned with this interpretation. Only in the group that sustained the expectation of possible fentanyl did the burning spray correlate with enhanced connectivity between the rostral anterior cingulate cortex and the periaqueductal gray—regions implicated in endogenous pain modulation. The researchers interpret this as expectation engaging the body’s intrinsic pain-control systems, though imaging alone cannot establish causality.

A Hidden Vulnerability in Placebo-Controlled Trials

Standard drug trials compare an active medication against an inert placebo designed to appear identical. The underlying logic assumes that nonspecific factors—expectations, hope, conditioning—are equivalent in both arms, so any additional benefit in the drug group reflects pharmacology alone.

Side effects can violate this assumption. If a drug produces noticeable effects such as dry mouth, drowsiness, or tingling while the placebo does not, participants may correctly deduce their assignment. Researchers term this “functional unblinding.”

The Hamburg team argues their results demonstrate how such deduction can translate into exaggerated treatment effects. If side effects amplify expectations exclusively in the active-drug group, a portion of the observed benefit may stem from belief rather than biochemistry. The authors propose active placebos—inert interventions designed to mimic a drug’s side effects—as one corrective. Lead author Lieven Schenk emphasized to Swissinfo that when a drug causes side effects, researchers should “make sure that the placebo also causes them.”

Regulators Have Already Confronted Unblinding

The concern extends beyond theory. In June 2024, an FDA advisory committee voted 9–2 that Lykos Therapeutics had failed to demonstrate the efficacy of MDMA-assisted therapy for PTSD, and 10–1 that its benefits did not outweigh its risks. Among the cited issues was participants’ frequent ability to discern whether they had received the active drug, as reported by BioSpace.

The FDA rejected the application on August 9, 2024, requesting an additional Phase 3 trial. Reviewers also flagged other concerns, including selection bias and safety data, making functional unblinding one factor among several rather than the sole determinant. An expert quoted by BioSpace noted that functional unblinding occurs in virtually all psychiatric drug trials.

Active Placebos Are Not a Simple Solution

Empirically testing whether active placebos alter trial outcomes has proven difficult. A Cochrane review of 21 trials employing both active and standard placebos, encompassing 1,462 participants, found no clear overall difference in estimated drug effects—a result the authors deemed imprecise. Among the five trials at low risk of bias, the difference was larger, though the authors cautioned this association may be spurious.

The reviewers nevertheless advised that researchers carefully consider placebo selection when a drug has conspicuous side effects and outcomes are patient-reported, a scenario common in pain research.

The Hamburg study carries its own limitations. Participants were healthy young adults, pain was experimentally induced via heat in a laboratory setting, and the side effect was a mild, transient burn. At a follow-up session approximately one week later, the burning spray again reduced pain in both groups, and the between-group difference was no longer statistically significant. The authors speculate that debriefed volunteers may have reconstituted their expectations, though this was not directly measured. Roughly a quarter of the 104 enrolled participants were excluded from the analysis, and the team has made its data publicly available.

For patients, the study does not imply that side effects indicate a medication is working—or failing—in either direction. Anyone with concerns about a medication’s effects or side effects should consult their prescribing clinician before making any changes.

Key Questions Answered

What changed when volunteers learned the truth?

In the subsequent session, the burning spray no longer reduced pain. Volunteers who still believed they might receive fentanyl reported lower pain after the burning spray, whereas those who had been debriefed did not.

Did the burning sensation itself change?

No. Both groups rated the side effect similarly, indicating that belief—not the sensation alone—drove the analgesic effect.

What is functional unblinding?

Functional unblinding occurs when trial participants can deduce whether they received the active drug or a placebo, often due to noticeable side effects, potentially biasing the results.

What is an active placebo?

An active placebo is an inert treatment designed to mimic the side effects of the investigational drug so that participants cannot easily discern their group assignment.

Should patients interpret side effects as a sign a drug is working?

No. This laboratory study does not support that conclusion, and anyone with concerns about a medication should discuss them with their clinician.

Source link