Gary Greenberg on the Placebo Effect

Episode Summary

Executive Summary: Russ Roberts and Gary Greenberg examine the placebo effect as a serious, under-theorized force in medicine, psychotherapy, and healing. They argue that expectation, empathy, ritual, and confidence can materially affect outcomes, especially for chronic and stress-related conditions, while also warning about ethical risks, victim-blaming, and the limits of current clinical-trial methods and biomedical models.

Main Topics: What the placebo effect is—and why it’s hard to define (Priority: 5/5): Greenberg explains that placebo is usually defined either as any medication effect not due to the drug, or as everything in a clinical trial not attributable to the active medicine; both definitions reveal how placebo is bound up with trial design rather than a simple standalone phenomenon. Clinical trials, deception, and hidden assumptions (Priority: 5/5): The conversation explores double-blind placebo-controlled trials, why subjects are told they may get placebo, and how trial structure itself may alter outcomes. Roberts and Greenberg note that trials may not cleanly isolate drug effects if placebo and drug effects interact. Empathy, therapeutic alliance, and healing rituals (Priority: 5/5): Greenberg argues that warmth, attention, empathy, and ritual may activate bodily healing processes, potentially via brain concordance and mirror-neuron-related mechanisms. The quality of the relationship between healer and patient may matter as much as the treatment itself. Placebo effects in chronic pain and difficult-to-treat conditions (Priority: 4/5): They discuss conditions such as chronic pain, irritable bowel syndrome, chronic fatigue syndrome, and back pain, where modern medicine often struggles and where placebo responses appear especially strong. Mindfulness and expectation can help, but may also invite blame if they fail. Genetics and biological variability in placebo response (Priority: 4/5): Greenberg highlights emerging evidence that placebo responsiveness varies with genetic markers, including dopamine-related pathways. This suggests some people may be biologically more responsive to expectation effects than others. Limits of the current biomedical model (Priority: 5/5): The discussion challenges the “magic bullet” model of medicine, arguing that many outcomes likely reflect complex systems behavior rather than simple molecule-on-molecule causation. The hosts suggest current methods may miss important healing mechanisms. Implications for psychotherapy, psychiatry, and confidence (Priority: 4/5): Greenberg reflects on psychotherapy as a likely paradigm case for placebo-like healing effects, where care and relationship matter but are hard to isolate. They also discuss how psychiatric labels and confidence can shape outcomes, for better or worse.

Key Arguments: Placebo is not just a trick; it may reflect real biological and psychological processes embedded in healing encounters and trial contexts. Even double-blind trials do not cleanly separate drug effects from placebo effects, because expectation and the meaning of treatment can influence outcomes. Empathy, trust, and ritual may trigger physiological changes that support healing, particularly in conditions with strong stress or pain components. The placebo effect may be strongest where modern medicine is weakest, suggesting that complex conditions need a broader model than simple pharmacology. Some placebo responses may be genetically mediated, especially through dopamine-related pathways, implying meaningful individual differences in responsiveness. Open-label placebo studies suggest deception may not be necessary in all cases; explanation and supportive interaction can still produce benefit. A major risk of leveraging placebo-like mechanisms is moral harm: patients may be blamed for not healing if they are told to think positively or be more mindful. Psychotherapy may work partly through a “secret radio frequency” of care and attention that current scientific tools struggle to measure. The increasing strength of placebo effects in trials may reflect rising expectations and improved trial care rather than better interventions. Medical science has often assumed additivity—drug effect plus placebo effect—but that assumption may be wrong if these forces interact or interfere.

Data Points: Episode date: December 17, 2018 - EconTalk episode featuring Gary Greenberg First appearance on EconTalk: September 2010 - Roberts notes Greenberg’s prior appearance Years of archives available: Back to 2006 - Host mentions the podcast archive history Clinical-trial design: Double-blind, placebo-controlled - Described as the standard method where neither subject nor experimenter knows assignment Chance of placebo in trial: 50% - Roberts notes participants are told they may have a 50-50 chance of getting placebo Back-pain study result: No difference between vertebroplasty and sham procedure - Discussed as an example where cement injection and saline/sham produced similar outcomes Study size: 40,000 participants - Greenberg cites a large longitudinal study linking a genomic region to placebo response Study duration: 10 years - The 40,000-participant study followed people over time Historical threshold: Prior to about 1860 - Greenberg argues most medical treatment before then largely relied on placebo-like effects Placebo response conditions: Chronic stress-related conditions - Greenberg says these show the strongest placebo effects Prevention trial examples: Vitamin E or aspirin vs placebo - Used to illustrate genetic links to placebo-related prevention outcomes Genetic mechanism: Dopamine-related pathways - Greenberg says placebo effects may be modulated through dopamine Research setting effect: Patients are treated like royalty - Roberts describes the intensive attention and punctuality in clinical trials as a possible placebo enhancer

Pivotal Quotes: "What if the placebo effect isn't a trick?" — Gary Greenberg (article title referenced by Russ Roberts): Central framing of the episode and Greenberg’s New York Times Magazine piece "the ritual is very, very important to the outcome." — Gary Greenberg: On why medical and therapeutic processes may matter beyond the active treatment itself "there's stuff going on here that I don't really fully understand." — Gary Greenberg: On psychotherapy and the limits of current scientific understanding of healing

Implications: Medicine should take expectation, relationship, and ritual seriously, especially for chronic conditions. But using placebo-like effects raises ethical risks and may require new research methods that capture interactive, not merely additive, healing mechanisms.