The Placebo Effect in Therapy (Part 2/2)

Psychotherapy doesn't make much sense, and that's okay.

“Doctors look for cure, but patients still want care.” – Howard Spiro¹

Doctoring by Faith

When I was little and got sick, my dad would come into my room and put a trembling hand on my head. After a few seconds, he’d pull his hand back like a plunger and slap it on his own forehead, saying “I’ve got it now.” Sure enough, a few days later, he was sick and I was not.

What was “really” happening, of course, is that sicknesses are contagious and time-limited, but I didn’t know that at the time. And that’s not the whole story, either. One of the most consistent, bewildering scientific findings is that belief, hope, and expectation play an active role in healing. The only part of my dad’s act that was “real” was the extent to which I believed it, yet somehow that was as real as real medicine. This is known as the placebo effect.

The literature on placebo is fascinating and ultimately confusing. A number of factors weaken or strengthen a placebo’s power. Moderators include the form of the medicine (pill or injection? big pill or small pill?) and the associated ritual (taken how many times a day? shot where?), as well as the personalities of the patient and prescriber. For example, in George Bernard Shaw’s play, The Doctor’s Dilemma, Dr. Sir Ralph Bloomfield Bonington possesses such unwavering positive energy that patients get better so as not to disappoint him.

The cultural salience of the intervention is important, too. Howard Brody, referring to Kleinman and Sung’s Why Do Indigenous Practitioners Successfully Heal, writes of Taiwanese healing practices that “the patient who submits to [the healers’] care and experiences the proper rituals must be healed; to go on believing that one is sick afterward is tantamount to declaring that one of no longer a part of the Taiwanese culture.”² Similarly, in the U.S., the placebo effect is likely stronger when medication is prescribed by a psychiatrist compared to a nurse practitioner or primary care physician. In these examples, cultural expectations are doing the heavy lifting.

Basically, anything that influences belief, hope, or expectation can influence the placebo effect. We typically think of this only from the patient’s side, but it applies to the healer as well. Healing power depends in large part on the confidence to heal. As Gary Snyder writes in The Practice of the Wild about Álvar Núñez Cabeza de Vaca: “Once he had made his way back to Mexico and was again a civilized Spaniard he found he had lost his power of healing…for as he said, there were ‘real doctors’ in the city, and he began to doubt his powers.”³

These modifiers to placebo are confusing for the same reason that placebo is confusing: what the hell is happening? What is the process by which hope, faith, belief, or expectation is converted into the active healing currency of, say, white blood cells?⁴ If placebo could be put into a pill, it would be one of the most powerful drugs in the world—and the funny thing is that it is put into every pill to some extent. And not only pills, but remedies of many kinds…

I hope, therefore I heal.

The way I see it, psychotherapy is no different than surreal caves, psychiatric medication, religion, Reiki, or whatever the hell this is. They are all platforms for belief, in addition to whatever else they might be.

The fact that psychotherapy is a platform for belief helps to explain why different versions nevertheless produce similar results. Therapeutic methods are akin to religious sects in that arguments about the “right” sect are pointless because the basic premise of a God just isn’t true. The healing value of religion isn’t in getting the “facts” right; it’s in sharing something to believe in. Ditto with therapy.

As Steinbeck writes in East of Eden, “It took a smart man to know where the difference lay between the sects, but anyone could see what they had in common.”⁵ The same could be said not only for the full list of therapeutic approaches, but indeed for everything that proves therapeutic to humans.

Sleight of Hand

Is magic fake or real?

On the one hand it’s clearly fake because what the magician claims to be doing—conjuring a rabbit out of thin air, cutting someone in half—isn’t happening. But people are being awed.⁶ They do experience the trick as if it were magic. Nobody would go to magic shows otherwise.

Therapy, too, is “fake” in that what the therapist claims or thinks is happening, usually is not. What the therapist considers the “agent of change” is not the primary one. Clients aren’t getting better, by and large, by making the unconscious conscious, improving communication skills, or correcting maladaptive patterns in thought and behavior.⁷ All that stuff is the magical equivalent of “abracadabra” and other misdirection. It keeps the client focused on something; it distracts them while the real work is being done.

Note that the distraction must be realistic, believable. You couldn’t have “ketchup therapy” in which the therapist and client squirt ketchup on each other because there would be no reason to believe that could work. But something that is the functional equivalent of ketchup therapy might do the trick as long as its purported mechanism makes more cultural and/or intuitive sense. And, of course, if enough so-called experts endorsed ketchup therapy over several decades, it would probably catch on.⁸

So, something like psychoanalysis, as noted in Harrington (1999),

…might be completely wrong, according to external, scientific standards. But it nevertheless might work very effectively if both patient and psychoanalyst are committed to the story it tells about human beings. In fact, the most true story that something like science may be able to develop of human behavior at the same time may not be the most subjectively persuasive story for either patient or therapist.⁹

Isn’t that the rub, then?—that in many instances, truth and healing are opposed to one another? If a client believes that their father came to them in a dream and delivered an important message, do I reply: “That’s nonsense”? Do I say: “Of course he did”? Or do I say nothing?

Sometimes the tension between truth and story resolves easily. An oncologist friend once told me that when a patient needs medication and the options are equivalent, she nevertheless promotes one of them. “Your options are A and B, but I think A will be better.” Not only is this morally acceptable in my book, but I think it would be poor practice not to leverage placebo in this instance because there’s no cost to it. I do the same when providing therapist referrals. “Here are two options. I’d start with Miranda first. If she’s not available, though, or turns out not to be a good fit, Gary is great, too.”

Other times, the tension between truth and a healing narrative is more difficult, and I don’t wrestle with it only as a therapist—I wrestle it with a human, too, one who would be better off with a little less skepticism and a little more faith.

I think this tension, by the way, accounts for the current equilibrium in the field between story and science. We at Living Fossils have been beating the drum for over two years now that clinical psychology ought to take a more scientific—specifically, evolutionary—approach. Indeed, clinical psychology itself wants nothing more than to be a science of “evidence-based” interventions. But this shift has been slow in developing because it would invalidate many healing narratives that both therapists and clients have come to rely upon. And how pointless would it be to invalidate these stories if we didn’t have anything complete with which to replace them? Even the evolutionary view is ultimately silent when it comes to the kind of problems most clients bring to therapy.

The Real Work

Meanwhile, what is the “real work” being done while the therapist misdirects their client’s attention with fancy notions of transference, resistance, core beliefs, birth order, systems of interaction, and so forth? A relationship, of course. One built on care and compassion, and dedicated to making sense of—and finding meaning in—the client’s life.

More specifically, the real work is being done at the level of common factors, or elements that all psychotherapies share. These include the therapeutic alliance, the placebo effect (i.e., expectation of improvement), time dedicated to solving problems, the benefits of an outside perspective, and so on. While therapists drone on about correcting core beliefs, then, or interpret everything in a client’s life through the lens of birth order, clients are benefitting from the fact that someone cares enough to take their problems so seriously.

As Thoreau says, “The greatest compliment that was ever paid me was when one asked me what I thought, and attended to my answer.”

Therapeutic orientations are important, but not for the reasons most adherents think. Belief in a chosen method is itself a common factor, amplifying the placebo effect for both the therapist and client. A therapeutic orientation, as the name suggests, helps therapists make sense of and believe in their work. It frames an otherwise overwhelming display of suffering, and comes with a supportive community of like-minded individuals, too. The therapist, meanwhile, is the main vehicle through which a client comes to make sense of their suffering and believe in their recovery. That is why therapeutic approaches are important, not because they contain accurate information. They’re two-lane placebo highways.

That said, if therapists focus too much on the specifics of their chosen method, they might be distracted from forming authentic, healing relationships with each unique client. A common factors approach would de-emphasize many of the “distinctions without a difference” that the field has long fought over, recognizing them instead as analogous to the magician’s misdirection. As “mere” ingredients of the placebo effect.

Now, you might be wondering if “just” a relationship—with nothing else to it, such as an empirical understanding of the brain—can make that much of a difference. You mean to a highly social species, one whose survival has in large part depended on its ability to leverage the group? Of course it can.

Specifically, I often think of my work as that of a diplomat. Broad strokes, people usually come in because of a social rupture. We might say that they are out of harmony with their social world. In establishing a bond with me, clients subconsciously hope to re-establish harmony with their partner, job, parent, society-at-large, what have you. My role is to broker the reconciliation. To bring the client back into the fold, as it were, and if necessary, help explain their absence. Then, when the client is firmly back in the hands of the group, I take my leave.

In the end, therapy is mostly a meaningful—albeit proxy—social relationship. That’s all it is, and yet it can be everything.

The Implications of Placebo

Much of life is bait-and-switch. For example, a person falls in love and wants to have sex with the person they love. Before they know it, they’re staying up all night changing diapers and definitely not having sex.¹⁰

However, as that example illustrates, not all bait-and-switches are bad. I find therapy to be an especially lovely one: the client comes in suffering, with the belief that I have the magic potion. By the time they realize I don’t, a wonderful—dare I say magical?—thing has occurred. We’ve formed a bond, albeit one that will end. But our relationship has developed as an aside, in the background. I have indeed remained a Wizard of Oz during most of the act, concealing the fact that I am another error-prone, answer-less muggle. I deceived them with my diploma on the wall and the use of big, meaningless words—to their benefit.

So, therapy is fake—is placebo—in its stated means of operation, but real in its delivered effect. A dialectical behavior therapist, for example, might think the client is improving because they are finally doing their homework. Yet it’s more likely that the client is improving because someone cares enough about them to assign homework and insist that it gets done. Most of the stuff that psychotherapists have long hung their hat upon—delivering brilliant interpretations, creating self-aware and compliant clients, uncovering the root cause of all subsequent illness—is irrelevant outside of its ability to legitimize the therapy, to make it seem like serious business with a good chance of success. Such content becomes the foreground, allowing the actual process of healing—via common factors such as the relationship and placebo—to operate smoothly in the background.

But the elusive mechanisms of healing by common factors helps to explain why we do not typically identify them as central—why clients will often say that therapy has been helpful and then balk at the natural follow-up of “why?” The reality is that they don’t know; and most times, neither does the therapist.

After all, the placebo effect doesn’t make much sense. How exactly would a temporary relationship with a stranger heal? In the midst of such confusion, clients will fall back on answers that aren’t true but at least sound more plausible. Thus I hear “tools and techniques,” “self-awareness,” “insight,” and even “homework” as the causes of therapeutic success. Never mind that I hate the phrase “tools and techniques,” or that most clients forget that they were assigned homework in the first place. Long after therapy is over, it won’t be what the homework was about that matters, only that I cared enough to give it.

Here are some implications for the field if everything I’ve said is true:

1. The main value of therapeutic training is signaling value, which strengthens the placebo effect. The patient’s knowledge that their therapist has a Ph.D. contributes more to their healing than whatever their therapist happened to learn (or unlearn) in those 5-7 years.

2. If therapeutic training were designed to be technically helpful, it would focus on how to leverage common factors. For example: how to build strong relationships, listen carefully, facilitate open conversation, and use placebo. Training should also focus on what is knowable about mental health, much of which comes from an evolutionary perspective. For example, that emotions measure and motivate, or that emotions can misfire on the principle “better safe than sorry.”

3. Many alternatives to therapy exist. Most of these alternatives are less costly. If we want to give everyone access to mental health, these low-cost, widely-accessible alternatives should be front and center. Yes, I’m talking about diet, sleep, exercise, and social connection.¹¹

4. The field should stop wasting its time comparing therapies. Someone ought to knock CBT off its high horse specifically.¹²

End of Series

I’ve been working on this piece for well over a year. I had most of it worked out, but I kept deliberating on the following question:if I say that therapy is placebo, am I cutting out its legs?

A positive placebo response is most likely to occur in a patient when the meaning attached to that illness experience by the patient is altered in a positive direction. “Meaning,” in turn, consists of at least three general components—providing an understandable and satisfying explanation of the illness; demonstrating care and concern; and holding out an enhanced promise of mastery or control over the symptoms.¹³

If all I can provide is the second—demonstrating care and concern—then aren’t I leaving a lot of healing on the table? The last thing I wanted to do was undermine something that many people have found helpful, including my clients.

But then, in the course of my research, I learned an amazing thing: the placebo effect still works if you tell people about it. I don’t really have anything to say about this, other than it makes sense because nothing about the placebo does.

Separately, though, the conclusions I arrived at in this article have improved my practice. For example, I no longer concern myself as much with the stuff that doesn’t matter—even the question of whether my client has changed—and instead focus on listening better and caring more. It’s actually been liberating to realize that the correct interpretation of my clients’ problems isn’t all that important, nor is their degree of insight. Case conceptualization and self-awareness can help, to be sure, but are not the end-all, be-all.

There are a few areas where I’m still uncertain though. I sometimes leverage placebo in what feels like inauthentic ways, and that gives me pause. For example, for clients getting over heartbreak, I’ll sometimes feign more confidence than I feel that they will get over it and be okay. Sometimes, they won’t. As George R.R. Martin says, “Some old wounds never truly heal, and bleed again at the slightest word.”

Or perhaps a client is enthusiastic about a new direction in life, but I don’t think it will work out. Do I keep my mouth shut? I try not to race ahead of them, of course, but I also tend not to rain on their parade. This all but guarantees that when the inevitable downfall comes, I’ll be saddled with not a little guilt.

I’ve also had clients that I simply can’t bring myself to like or respect, and I often wonder if I should refer them elsewhere, and what I should say if I do.

But, for the most part, I have benefitted from leaning into the non-sense of my work. Placebo makes no sense; it bucks the logic of scientific scrutiny for now. Yet smart people in particular—or at least people who think they are smart—seem allergic to doing things that work for unknown reasons. (Remember, I know this because my tagline is that I work best with people who think they are too smart for therapy.) Yet sometimes it makes sense to do something because it feels good, even if it makes no sense. Even after all these years of providing therapy, I cannot put my finger on anything more predictive of success than the client’s initial belief that therapy will work. To a large degree, I find myself as effective as my client thinks I am.

As Howard Spiro says, “The placebo effect and the benefit it brings should make us all once again regard physicians as potentially active healing agents, however idiosyncratic and differing their personalities.”¹⁴ To that we could add: “however idiosyncratic and differing their approaches.”

In the end, being a therapist is a difficult job—a wonderful job—but not yet much of a scientific one. And that’s okay.

---

References

Harrington, A. (Ed.). (1999). The placebo effect: An interdisciplinary exploration (Vol. 8). Harvard University Press.

1. as quoted in Harrington (1999), p. 39.

2. as quoted in Harrington (1999), p. 83, emphasis original.

3. p. 24

4. There is, as yet, no mechanistic explanation for placebo, as there is for many other treatments, from casts to morphine.

5. p. 215.

6. Speaking of, check out Rob’s post on what awe might be measuring and motivating.

7. If these were legitimate agents of change, then presumably they would produce varying effects. They don’t.

8. See what I did there? :p

9. p. 230.

10. Keep in mind that this bait and switch was much more prevalent in the past when people had far less knowledge of and control over reproduction.

11. This is the conclusion Tom Insel came to in Healing after serving as director of the National Institute of Mental Health, in which he oversaw billions of research dollars into genetic testing. There may be promise to that line of research, but people are suffering now.

12. As the saying goes, if you want something done right, do it yourself—so stay tuned on this.

13. Howard Brody, as quoted in Harrington (1999), p.79.

14. as quoted in Harrington (1999), p. 47.