High rates of placebo responses are consistently reported across medical conditions, notably mood disorders, Parkinson disease, and pain, but also schizophrenia, substance use disorders, and surgical procedures. Placebo response rates in antidepressant trials average 31% to 45% compared with approximately 50% responses to antidepressants, and they have increased over the last 30 years. The failure of antidepressant responses to separate from placebo has contributed to the reduction or discontinuation of research on new treatments for depression and other neuropsychiatric illnesses, hindering the development of novel neuropsychiatric treatments.

So begins a new paper considering the relationship between placebo and depression treatment.

This week’s Reading: “Association Between Placebo-Activated Neural Systems and Antidepressant Responses Neurochemistry of Placebo Effects in Major Depression” by Dr. Marta Peciña et al., which was just published “online first” by JAMA Psychiatry.

Dr. Marta Peciña

Here’s a quick summary: this is a big paper in a big journal that seeks to better understand the placebo effect and antidepressants, and taps neuroimaging to do so. There is, however, a catch: the number of patients involved is small.

You can find the paper here:


The placebo effect has drawn more interest of late. In this month’s Psychopharmacology, for example, a clever paper finds that post-SRI trial patients respond less well to future medication trials, and that they also respond less to placebo.

If the placebo effect is of ongoing interest, no wonder – patients on placebo respond almost as well to the dummy medications as to true antidepressants, something vexing and frustrating to us clinicians. And there are so many unanswered questions: for instance, why is the placebo effect stronger in some patients than others?

And that’s why I’m so excited about the renewed energy of study in this area. The Peciña et al. paper focuses on placebo and neuroimaging.

Here’s what they did:

· 35 people were selected with a DSM-5 diagnosis of depression, recruited by advertisement.

· During the first phase, patients were randomized to one of two placebo treatments: an “active oral placebo” (2 pills a day) for a week, having been told that this was a fast-acting antidepressant, or an “inactive oral placebo,” having been told it was an inactive control.

· Patients had a 3-day washout period, and then each group was given the other placebo treatment (active to inactive placebo, and vice versa). PET scanning was then done but after a placebo IV was given to the people who had just completed the active placebo treatment.

· Depressive symptoms were measured with a 16-item, self-reported scale (QIDS-SR16) before and after placebo. People who responded to the placebo could then be identified as responders. Researchers also administered another scale (PIDS) as patients received their IV placebo.

· Patients were then invited to participate in a 10-week open-label antidepressant trial (citalopram, unless they had a history of non-response to this medication, in which case another agent was chosen). Further QIDS-SR16 testing was done.

Here’s what they found:

· Of the 35 people selected, 23 were women and the median age was 35.

· 35 people started the open-label antidepressant trial, but 10 dropped out – meaning that only 25 people finished the full study.

· The patients moved through different phases, results were reported accordingly.

· After the placebo phase, not surprisingly, people reported a reduction in symptoms – both oral “active” placebo and IV, though not with the non-active placebo. (A QIDS-SR16 score drop of 1.75.)

· When antidepressants were introduced, the responder group (those who had responded to the oral placebo medications) showed larger reductions in QIDS-SR16 scores during antidepressant treatment compared with non-responders. This effect took time: present 4 weeks after starting the antidepressants. And it was even more pronounced over time: by weeks 8 and 10, the mean QIDS-SR16 score was roughly twice as high among placebo nonresponders compared with placebo responders. (!)

· Remission was higher in the responders: 60% vs. 20% for non-responders. (!!)

· As for neuroimaging, “the capacity to activate the MOR system during placebo administration was associated with greater reductions in QIDS-SR16 score over the 10-week trial.”

· The authors did a regression model that included “objectively measured placebo-induced opioid release in the sgACC, NAc, THA, and AMYG [subgenual anterior cingulate cortex, nucleus accumbens,
thalamus, amygdala] as regressors accounted for 43% of the variance in the response to open-label antidepressant treatment.” Responsiveness to placebo “predicted 46% of the variance in the response to 10 weeks of antidepressant treatment, while the combination of both the clinical and the opioid release measures predicted 57% of the variance in the response to 10 weeks of antidepressant treatment.” (!!)

· Below: Figure 4 from the paper; the regression analysis is in the top right corner.

JAMA Psychiatry

The authors write:

To our knowledge, the present study is the first direct demonstration of the role of a specific neurotransmitter system, namely MOR-mediated neurotransmission, in the formation of placebo effects in MDD and provides an explanation of the variability in antidepressant treatment responses.

The authors conclude that there are implications to drug development and depression treatment:

In clinical trials, this evidence could help inform decisions regarding patient stratification and drug-specific or nondrug-specific effects. In clinical practice, placebo responsiveness could potentially indicate the likelihood of responsiveness to enhanced patient-clinician interactions or psychosocial or cognitive approaches.

A few thoughts:

1. This is a thoughtful and well designed study. Tying neuroimaging to the study of placebo and antidepressant effects is, frankly, cool.

2. That said, it should be stated again: only 25 people (of a total of 35) actually completed the open-label antidepressant phase.

3. It’s interesting but is it important? The authors clearly think so: they go so far as to suggest implications for drug development. One co-author told Time: “The findings show that the response to medications, in this case antidepressants in major depression, is a complex process that also engages internal mechanisms of resiliency as well as the potential effects of medications. Examining the source of individual variations in the capacity to engage those resiliency mechanisms provides new targets for drug development to enhance those mechanisms.”

4. Another day, another study showing something interesting about the placebo effect. Forget about drug development, are there implications for how we treat patients beyond just the medications? Can we tickle the nucleus accumbens or amygdala better? Like all good studies, this paper ends up raising more questions than it answers.

Further Reading

The Psychopharmacology paper on declining placebo effect and antidepressants can be found here:


Reading of the Week. Every week I pick a reading — often an article or a paper — from the world of Psychiatry.