The placebo effect has been part of medical conversation since the late 1700s, when an English physician named Dr. Alexander Sutherland coined the term to describe a doctor with an especially reassuring bedside manner. Today, the concept plays a central role in how we test new drugs: to prove a medication actually works, researchers have to compare it against a “dummy” pill or injection—something inert, with no active ingredient. Neither the patient nor the doctor is told which one the patient is getting, a setup called a double-blind, placebo-controlled trial. This precaution exists because up to 40% of patients feel real, measurable benefits from a completely inactive treatment, simply because they believe it’s the real thing.

When Belief Changes the Body

What’s fascinating—and a little unsettling—is that over three centuries of medicine, we’ve learned that believing a treatment will help can cause the body to respond as if that treatment (proven or not) actually did. Studies have shown that a plain saltwater injection can ease pain, and a sugar pill can lower blood pressure.

I saw this firsthand during my medical residency. At the time, there was a common (and, I now think, troubling) practice: if a patient came in reporting pain and staff suspected they were exaggerating to get narcotics, we’d give them a shot of saline instead. If their pain went away, we took that as proof they hadn’t really been in pain—they were just drug-seeking. Looking back after more than 40 years in medicine, I’m no longer confident that conclusion was right, and I’ve come to have real concerns about the ethics of that approach.

I’m not alone in thinking about this. A 2008 survey of nearly 700 physicians found that roughly half of internists and rheumatologists reported regularly prescribing placebo treatments, and most—62%—considered the practice ethically acceptable (1). In a separate, well-known study, 165 patients with knee arthritis underwent one of two real surgical procedures—or a fake surgery, where surgeons made incisions and went through the motions without actually treating anything. The results were no different across the groups (2). And in another striking experiment, Dr. Robert Ader showed that simply giving people a flavored drink could suppress their immune system if their body had been conditioned to associate that taste with an immune-suppressing drug (3).

Rethinking What’s Actually Happening

Because of the ethical discomfort around deceiving patients with fake treatments, researchers have proposed better ways to describe what’s really going on. One idea is the “meaning response”—the notion that healing is driven not just by a treatment’s chemistry, but by the meaning, context, and ritual surrounding it: the doctor’s confidence, the setting, the expectation of getting better. This same idea helps explain the flip side, called the nocebo effect, where a patient given an inactive substance but warned it might cause side effects actually experiences those side effects (4).

Where Cannabis Comes In

We’ve long known that the body’s own endorphin system—the same one responsible for a “runner’s high”—plays a role in placebo pain relief. But a Swiss study looking at what happens in the brain during placebo pain relief found something else: the endocannabinoid system, or ECS, was involved too (5).

The ECS is a body-wide network of receptors and signaling molecules—related to, but distinct from, the compounds in cannabis—that helps regulate things like pain, mood, appetite, and immune response. It’s part of how the body maintains balance, or homeostasis.

A 2017 study put this connection to the test directly. Twenty participants were told they’d be given a cannabis lollipop containing a high dose of THC (the main psychoactive compound in cannabis). In reality, they received an inactive version. Remarkably, many still experienced effects similar to genuine THC intoxication—the first study to show that people’s expectations alone could produce a placebo “high” (6).

This lines up with something experienced practitioners in the cannabis and psychedelics field have long emphasized: set and setting—a person’s mindset going in, and the environment they’re in—can shape their response to a substance as much as the substance itself, alongside simple conditioning: your body learning to expect a certain effect (7).

A Physician’s Perspective

Long before I specialized in cannabis medicine, I noticed something in the emergency room that I couldn’t quite prove, but strongly suspected: keeping a calm, warm, humorous (when appropriate) presence with patients—even in the middle of treating serious, life-threatening conditions—seemed to genuinely improve how they did.

Years later, working as a cannabis physician and seeing the sheer range of conditions that cannabis seemed to help, I started paying closer attention to the ECS’s central role in keeping the body in balance. Every organ system has cannabinoid receptors woven through it, tied into all sorts of regulatory processes. Watching patients respond to different cannabinoids—the well-known ones like THC and CBD, and lesser-known ones too—I began to wonder: could the ECS also be involved in the placebo effect itself, beyond just pain relief?

The Open Question

Could there be a placebo-like benefit, driven by the ECS, in other conditions—Parkinson’s disease (there are notably high concentrations of cannabinoid receptors in the brain regions Parkinson’s affects), mood disorders, addiction, autoimmune illness, and more?

Here’s where things stand: we know the ECS matters for all of these conditions in its own right. There’s solid, independent evidence that ECS activity affects mood and anxiety, nausea, motor control in Parkinson’s, and cravings related to addiction. What we don’t yet have is the kind of study that would prove expectation and belief specifically recruit the ECS in these other conditions, the same way it appears to during pain relief—the equivalent of blocking a cannabinoid receptor and watching a placebo effect disappear. That experiment hasn’t been done.

But there’s no obvious reason it couldn’t work that way. If the brain can tap into endocannabinoid signaling in response to expectation and conditioning for pain, it’s reasonable to think it might do the same in other places where those same receptors show up and matter.

If that turns out to be true, it raises a compelling possibility: could doctors and patients learn to intentionally engage this system—this naturally occurring, ethical mind-body interaction—to support healing across a wide range of conditions, without needing to introduce any outside substance at all?