Is psychedelic research biased?: 5 Questions for psychiatrist Josh Woolley

Woolley discusses addressing biases in psychedelic research.

According to Josh Woolley, oxytocin was his gateway drug into studying psychedelics. After receiving an M.D. and a Ph.D. in neuroscience from the University of California, San Francisco, Woolley joined its faculty in the early 2010s and began studying oxytocin-enhanced psychotherapy. When oxytocin, often called the “love hormone,” is administered via a nasal spray, it has been shown to increase bonding in group therapy sessions and may help control addiction cravings. Woolley’s work investigated how oxytocin affects people being treated for opioid dependency and how it influences social cognition in people with schizophrenia.

Then, in 2015, Woolley attended a party at the home of venture capitalist and philanthropist George Sarlo, who told his guests about his experience as a Holocaust survivor — and how an ayahuasca trip conjured visions of his late father and opened the door to confronting his painful past. Woolley was intrigued, and began searching for ways to start studies using psychedelics at UCSF.

The first psychedelics study Woolley worked on was published in The Lancet’s EClinicalMedicine in 2020, and looked at whether psilocybin could be used to treat demoralization in long-term AIDS survivors. As Woolley’s psychedelics work gained momentum, he became the director of UCSF’s Translational Psychedelic Research Program. The group is investigating psychedelics’ effectiveness in treating a variety of health issues, from depression and bipolar disorder to Parkinson's disease and chronic back pain. The Microdose spoke with Woolley about addressing biases in psychedelic research. 

Expectancy and placebo effects are a hot topic of conversation among psychedelics researchers at the moment. What do we know about placebo and expectancy effects in other areas of psychiatry and medicine?

When you expect certain things, those things are more likely; when you expect to get better, you engage the neural circuits involved in that happening. For instance, in pain studies, if you tell participants you are going to give them fentanyl and you don't, the pain goes way down, but only if the person believes you. The pain relief that you're getting is because you're releasing endogenous opioids. You can actually block those opioids with naltrexone and you won’t see that placebo effect. [Editor’s note: Naltrexone is a medication that blocks opioid receptors.] The same thing happens in depression: if you think you're supposed to get better, you engage neural and hormonal circuits involved in relieving depression.

But if the person administering fentanyl knows it's not fentanyl, you get less of an effect in the patient, even though you don't tell them; somehow, they pick up on it. You see the same thing in the other direction: if you tell a patient that something you are giving them is going to make pain worse and they believe you, the pain will be worse. That’s the nocebo effect.

One argument I’ve heard is that effect sizes for psychedelic trials are so big that it can't be placebo. But you can have gigantic placebo effects — for instance, you see that with sham surgery. If you put people in an operating room and they cut you open but don't do anything, you see very large placebo effects. That worries me in my work — whenever we have patients who have dramatic improvements, I think it’s a real effect, but maybe it's all expectation. I know first hand that the hype and hope is really high. We have participants battering down our door to get into studies saying, “I know this is going to work for me.”

How are you and other researchers working with psychedelic substances managing those effects in your study designs?

We can try to better mask the conditions with active placebos, like niacin — but those active placebos aren’t very strong, and not powerful enough to produce a subjective experience. I’m not sure they’re really going to trick anybody. Having a personally meaningful experience is likely therapeutic, too, so how do you control for that? Or can you get something else to induce a personally meaningful experience? It’s a philosophically complex question. Related to that, psychotherapy researchers think there is no placebo for psychotherapy because all of it is placebo — therapy is a manipulation of hope and expectation. The classic double blind placebo controlled trial is not super appropriate for psychotherapy in general. 

In our studies, we’re using what we call “the cone of silence,” or incomplete disclosure, for certain aspects of studies, where participants and staff don’t know details of the methods. It’s tricky in considering the ethics of doing that, but as a field, we can’t keep doing the same trials over and over again, where everyone knows the methods and can guess what might happen. 

Currently, most trials don't even ask what people think they got — the placebo or active drug — so we don't know how big the placebo effects could be. We need to start asking participants  that question.But let’s say it turns out it's impossible to fully mask in a study whether someone got a psychedelic or not — it's certainly possible that we may never be able to confuse people with psychedelic experience versus something else. 

What do we do then? 

Some people are nihilistic and say that we’ll never know if psychedelics really work. But I don’t think that’s a reasonable position. There are all sorts of things we can’t do clinical trials on that we believe in, like that cigarettes cause cancer or that education is good for you. So how did we figure those things out? We just used messier, longer processes — natural experiments, epidemiology, pre-clinical work with animals. If we use a combination of randomized controlled trials and different research approaches we might be able to understand those effects more holistically. Basically, we try to control variables in trials as best as we can, but also be open to the possibility that we may never fully succeed, and accept that.

You recently published a paper in Psychedelic Medicine surveying psychedelic clinical trial facilitators about their psychedelics use. Eighty eight percent of the therapists you surveyed had personal experience with at least one psychedelic, psilocybin being most common. Do you think facilitators’ or researchers’ personal experience with psychedelics matter?

Personally, I have not done psychedelics. But it’s a complicated issue. On the one hand, there’s always been this fantasy that there’s a divide between researchers or clinicians and patients. The truth is that we’re all patients sometimes. There are researchers who have become interested in mental health because they were exposed to mental illness through personal experience or through the experiences of family members and friends. Yet for the most part it has not been safe to disclose that, which makes it seem like all the researchers and clinicians are totally mentally healthy. It’s the same thing for drug use: more than 10% of Americans say they’ve used psilocybin, and yet, in my medical school class, no one said that they used psychedelics. People are coming out of the closet with it now, but it’s like: do you drink alcohol, another drug which is, in a lot of ways, more dangerous? People aren’t really asking researchers who study alcohol, “Do you really like alcohol?” or asking nicotine and tobacco researchers how much they smoke. I think stigma and shame sometimes underlies these questions.

But on the other hand, psychedelics advocates argue that these substances are powerful and change your mind and personality in fundamental ways. So perhaps it is the case that researchers or clinicians who have used psychedelics might actually be different in some way. In that study, those facilitators disclosed they’ve used a lot of psychedelics, and really believed in them working. That makes sense: Cognitive Behavioral Therapy therapists probably believe in CBT. Everyone comes from an intersection of identities and has their own positionality, and we have to be able to talk about these things transparently. 

Transparency about researchers’ biases has also come up in discussions of research funding. Most psychedelic research is funded by non-profits or companies, and I know the group you lead at UCSF, the Translational Psychedelic Research Program, has been funded by groups including Usona, MAPS, and the Heffter Research Institute. How do you think about best practices for managing conflicts of interest?

One of my mentors has always said that government money is the “cleanest” money — it's the most rigorous review system that we have. But for the most part, the government is not really funding psychedelic research. The National Institutes of Health has mostly gotten out of the drug development game, and now see their job as understanding the mechanisms of mental illness. The cost to bring a new drug to market is very expensive, over $100 million — and that cost has fallen to industry.

Personally, I always go through my university for these grants, and they have committees that review relationships I have. I disclose them any time I give a talk, and I try to acknowledge them and be transparent about it. I also don’t work with just one company or philanthropist. And I’ll note that none of this actually increases my salary, and I don’t own equity in any psychedelic companies.

To some extent, we’re all biased; if you build your career on a particular thing, we have biases towards that thing. Though I’ll say that if I could prove beyond a doubt that psychedelics didn't work, that’d probably be pretty good for my career too. 

I’m not sure what the right answer is, but I think it's dangerous when we’re not transparent about our conflicts of interest, or when we try to manage them in secret.

This interview has been edited and condensed for clarity and length.