How Scientists Finally Learned to Measure the Placebo Effect

A new way to separate belief from real change—without control groups.

Posted January 27, 2026 | Reviewed by Devon Frye

Placebo effects are real, powerful, and deeply human, as I've discussed in my previous article . But a thorny question remains: How can we tell whether a depression treatment is producing genuine therapeutic change, or whether improvement comes entirely from belief?

To understand the challenge, imagine the following scenario.

The "Placebin" Thought Experiment

Suppose I launch a new pharmaceutical company and announce a revolutionary antidepressant called "Placebin." It’s powerful, safe, and nearly free of side effects. To prove it, we will give Placebin to one million depressed Americans—free of charge.

Weeks later, 350,000 people report dramatic improvement from the amazing new drug. They flood the talk shows with testimonials. There are dramatic mood improvements and no side effects at all. Friends and family join the chorus. The headlines write themselves.

There’s just one catch: Placebin is a completely inert pill.

Was their “recovery” real? Yes, it was. But the highly touted “Placebin” had nothing to do with it. So how do we know whether any antidepressant—or supplement, or therapy —is doing something real?

Why Traditional Placebo Research Falls Short

For decades, researchers assumed the only way to study placebo effects was through randomized controlled trials, dividing patients into drug and placebo groups. While valuable, this approach has serious limitations:

Here’s an example. When our team sought to evaluate a digital mental health app, we were stunned to learn that a modest controlled outcome study comparing our app to any popular antidepressant could cost $30–100 million, or more. For a small startup, this was beyond impossible—so we took a different path.

Using real-world data from 290 beta testers for our 2023 Feeling Good app, we applied an advanced statistical method—structural equation modeling—to measure the placebo effect precisely and directly, as a causal variable. 1

Instead of asking, “Who got the drug versus placebo?” we tried to ask far more revealing questions: What, exactly, did people expect would happen to their moods once they used the app, and how did those expectations shape what actually did happen? And how big were the causal effects of these placebo expectations, as compared with the “real” app effects?

To our knowledge, this had never been done before. I owe special thanks to my esteemed colleague, Dr. Irving Kirsch, a pioneer in placebo research and Associate Director of the Program in Placebo Studies at Harvard, for insights that made this breakthrough possible. 2

The results were striking: Users’ expectations weeks before starting the app—their predictions of how much they would improve—were astonishingly accurate. In addition, these expectations exerted a strong causal influence on their emotional improvement.

In other words, the placebo effect was real, and it was powerful. But this wasn’t the whole story.

The app’s primary goal is to help people challenge and reduce their belief in the distorted negative thoughts that trigger depression and anxiety—thoughts like “I’m worthless” or “I’ll always be depressed.” And sure enough, reductions in belief in these thoughts (cognitive change) produced far larger reductions in negative feelings than the placebo expectations.

In addition, both forces worked together, but cognitive change accounted for the lion’s share of improvement. There are two ways to compare the impact of any variable on changes in mood:

How much improvement does this variable actually cause, on average?

What is the explanatory power of this variable? In simple terms, how much of the change can be attributed to this variable? These two measures, as it turns out, can be radically different.

In Figure 1, you can compare the mean causal effects of placebo vs “true” app effects. As you can see, the “true” app effects (cognitive changes) caused twice as much improvement as the placebo effect (all measured from 0 to 100).

And, as you can see in Figure 2, the change in belief in negative thoughts explained nearly three times as much of the variation in the changes in negative feelings during Basic Training (measured as percent of variance).

And another unexpected finding was this: The dreaded placebo effects and true app effects were not at war, competing with one another, but were actually independent and additive. This means that app users got the best of both possible worlds.

This is the first time, to the best of my knowledge, that “placebo” effects and the true therapy effects have stood next to each other, shoulder to shoulder, for precise comparison and evaluation. And while the placebo effect was real and impressive, the effects of the cognitive change triggered by the app were much greater. But what was perhaps most exciting is that the “effect sizes” (calculated by Cohen’s d ) were vastly larger than previously reported for digital devices, especially in patients with moderate to severe depression. 3

The Paradox of Placebo

For years, placebos were treated as the enemy of “real” treatment. Our findings suggest the opposite. Placebo is not a nuisance to be eliminated—it’s an ally.

Belief and meaning amplify healing. But when belief is paired with tools that target the actual causes of emotional suffering, the results are far more powerful. Dr. Irving Kirsch once described the phenomenon this way: “The placebo effect is a window into the power of meaning and belief.”

Perhaps we’ve been asking the wrong question all along. The issue isn’t whether improvement is “real” or “just a placebo.” The real question is this: How can we harness belief ethically, transparently, and effectively to help people heal to the greatest extent possible, and as fast as possible?

The Real Antidepressant

Expectation, hope, and meaning almost instantly alter brain function. Our research has proven that learning to change how you think—especially how you relate to self-critical thoughts—produces genuine and lasting emotional change.

So, if your mood improved because you believed you could feel better, or because you learned to think differently, or because you used a tool that helped you rediscover your worth, then yes, your improvement was real. But the healing didn’t come from a pill. It came from you—and you may be the most powerful antidepressant of all.

Our latest study on the efficacy of TEAM CBT (T = Testing, E = Empathy, A = Paradoxical Agenda Setting, M = Methods) is currently under peer review.
Kirsch, I. (2009). The emperor’s new drugs: Exploding the antidepressant myth . Basic Books.
Linardon, J. et. al. Current evidence on the efficacy of mental health smartphone apps for symptoms of depression and anxiety: A meta-analysis of 176 randomized controlled trials. World Psychiatry , 23(1), 139–149 (2024).

Share this post Facebook Bluesky Linkedin Email

There was a problem adding your email address. Please try again.

By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy

David Burns, M.D. , is Adjunct Clinical Professor Emeritus of Psychiatry and Behavioral Sciences at the Stanford University School of Medicine. He is a renowned psychiatrist, award-winning researcher, and author of Feeling Good and Feeling Great ,

Get the help you need from a therapist near you–a FREE service from Psychology Today.

This article is part of the Bringwise Psychology Journal — daily insights on human behavior, mental health, and personal growth.