Verified by 
Depression
Peter D. Kramer on Depression, Placebo, and Psychotherapy
An interview with the legendary Brown University psychiatrist.
Posted March 1, 2021 Reviewed by Jessica Schrader
Peter D. Kramer, M.D., is one of America's best-known psychiatrists. He won international acclaim in the early 1990s upon publication of his bestselling book Listening to Prozac, and has since authored six other books relating to psychiatry, psychopharmacology, and psychotherapy. He received his M.D. from Harvard University and spent his career in the psychiatry department at Brown University, where he is currently Clinical Professor Emeritus of Psychiatry and Human Behavior. I thank him for his gracious willingness to be interviewed here.
Ruffalo: Your 1993 book Listening to Prozac has become a classic in psychiatry and represents one of the most influential books ever written for a general audience in the history of the field. Can you tell us a little bit about what the atmosphere was like in psychiatry in the early 1990s, and what motivated you to write the book?
Kramer: Thank you. Most of my books originate in my clinical practice. When the first SSRIs became available in the late 1980s, I prescribed them to patients and was struck by two responses. Some patients declared themselves "better than well"—often, more comfortable socially than they had been before the depressive episode. Others made observations about the self. Traits that disappeared, like obsessions, they judged now to be biological happenstance, not essential parts of their psychic makeup.
I wrote about those responses, first for colleagues and then, in Listening to Prozac, for general readers. I intended the book as a work of cultural commentary. What did it mean that patients were willing to allow a technology to define them? If we had access to what I called "cosmetic psychopharmacology"—if we could prescribe to move people from a normal but less desired or socially rewarded temperament to a better favored one—how ought the medical profession to respond? I was especially interested in the congruence between antidepressant effects, like increased assertiveness, and demands that the culture placed on women.
The early reviews attended to those issues. In philosophy, the book stimulated interest in the topic that medical ethicists call enhancement. Carl Elliott, a philosopher who trained in medicine, titled his book on enhancement Better Than Well. Because the public was eager for news about depression, Listening to Prozac was often read as a discussion of the treatment of mood disorders. Some psychiatrists took the book that way as well.
Back then, antidepressants were largely reserved for use in major depression. Treatment duration was short, a matter of months, and focused on acute episodes. If an episode ended, and a patient remained socially inhibited, pessimistic, and plagued by low self-esteem, the medication had done its job. For various reasons—but I think the book was one—the scope of prescribing widened and the standard for successful responses rose. In the book, I warned about that shift, referring to "diagnostic bracket creep," the tendency for diagnosis to shift to reflect the potentials of new technology.
In part because SSRIs were well-tolerated, but perhaps also because of recognition of their effects on temperament, the scope of psychopharmacology soon became more congruent with the scope of psychotherapy.
Ruffalo: In recent years, there has been a great deal of debate surrounding antidepressants and the placebo effect. What are your thoughts on this debate, and are antidepressants any more effective than placebo?
Kramer: I wrote a recent book on this topic, Ordinarily Well: The Case for Antidepressants. In it, I conclude that in the treatment of depression, antidepressants work about as well as most treatments in medicine for chronic and recurrent illness.
The evidence accords so well with the clinical experience—patients do better—that the debate itself calls for explanation. I attribute it partly to professional rivalry—psychologists and psychotherapists versus medical-model psychiatrists—partly to stigma, and partly to the romanticization of depression that I explored in an earlier book, Against Depression.
SSRIs may work yet better for dysthymia—chronic lower-level depression—than they do for major depression. Interestingly, SSRIs' advantage over placebo appears to be most marked in the mitigation of neuroticism—fragility in the face of stress. In other words, subsequent research largely confirmed my speculation about certain patients' enjoying personality effects, sometimes independent of effects on core depressive symptoms.
The professional literature on placebo can be confusing or misleading. Some of it comes from professionals who appear to enter the discussion with advance hostility toward psychiatry. For a counterbalance, I suggest my piece in the LA Review of Books, "Why Are We So Eager to Hear 'Placebo' Speak?"
To point to one little-appreciated problem: For safety, and to ensure "retention" in the study, depressed patients in the control arm of medication trials receive extensive social, emotional, and financial support. Often, the research subjects are poor and isolated, with little structure in their day. In a study, they tend to improve via what we might call psychotherapy effects.
In practice, then, trials contrast a non-specific psychotherapy (the control condition) to that psychotherapy plus medication (the active arm). But we know that psychotherapy and medication effects are not additive. The benefit of the two together is never the sum of the benefits from each. When we subtract out the psychotherapy effect from the combined effect, the remainder is less than the effect of medication alone. The subtraction understates the inherent efficacy of antidepressants, the benefits that they would confer if given in clinical practice by, say, an internist who offers no special support.
If you have not encountered this paradox before, I walk through it (and others) in slow motion in Ordinarily Well.
What is hard to find in the research literature is evidence for the efficacy of what I call the classic placebo effect, improvement via hopeful expectations tied to pill-taking. It's unclear that the classic placebo effect does much in depression. As a clinician, when your patient improves on antidepressants, you can be fairly sure that the contributors are medication efficacy, the support or psychotherapy that you provide, natural remission, and regression to the mean. The contribution of pill placebo effects may be minimal.
Some of the most convincing evidence for antidepressants comes from research that stands apart from the often-problematic candidate drug trials conducted by drug companies to gain FDA approval. Studies that use unambiguous entry criteria are especially impressive. If you look at trials that admit consecutive hospitalized stroke patients, giving them drug or placebo, SSRIs turn out to be highly protective.
As a side note: The framing of the question—"Do antidepressants work?"—has changed. It used to mean: If I have a patient a few weeks into his second or third episode of uncomplicated major depression, will medicine shorten the duration? Studies on that question are no longer feasible, precisely because our current antidepressants work. Patients with non-chronic depression without serious comorbidities get treated by non-psychiatrists and do so well that it's not clear that it would be ethical to suggest that they enter a trial of an untested drug.
In our larger studies, the typical patient is two years into the seventh episode of depression that has not responded to extensive treatment. Amazingly, standard courses of antidepressants still seem to help.
Anyway, the question that we once would have asked—how simpler cases fare—has been answered. I doubt that this result will ever be challenged by contradicting evidence. Only a new and especially promising medication would merit testing on patients who are likely to respond to effective medications we know well, through extensive clinical experience.
Ruffalo: The British Psychological Society's report on depression caused controversy in 2020 for its description of depression as an "experience," not a disease. I wrote a brief, critical response to it here on Psychology Today. Having spent much of your career studying depression and its treatment, how do you conceptualize depression? Is it a disease, an experience, or something else?
Kramer: I dedicated Against Depression to that question. There are many ways to conceptualize mood disorders, but the medical model stands up well. What do you call an experience that is chronic or recurrent, that is associated with adverse changes in your brain, bones, endocrine glands, heart, and blood elements, and that is likely to shorten your life? Even a culture that had no concept of mental illness might well call that condition a disease. Depression checks all those boxes—and of course, we do recognize mental illnesses.
In Against Depression, I trace the history of resistance to the medical model. I associate it with the romanticization of melancholy, stigma against mental illness, the anti-psychiatry movement, and conflicts among the mental health professions. In Ordinarily Well, I look specifically at the claim that antidepressants are placebos with side effects. It arose in the 1950s when antidepressants were among the best-studied drugs in all of medicine. The two forms of resistance to medical-model approaches have had similar subsequent courses.
Ruffalo: Many psychiatry residents I teach have a great interest in psychotherapy and would like to incorporate it into their future practices. As a psychiatrist who values and has practiced psychotherapy, do you have any advice for psychiatry trainees looking to practice as psychotherapists in this current climate of managed care?
Kramer: It is hard for one generation to advise the next. I can say that as I finished residency training, my mentors told me that I would be unable to make a living practicing fee-for-service, psychotherapy-centered, outpatient psychiatry. Their mentors had likely told them the same.
My practice filled immediately, and I continued for almost 40 years without ever having an open slot. Each new patient was an exception, an extra case that a Dean or close colleague insisted I see. In later years, some of the pressure on the practice came from colleagues' and prospective patients' awareness of my writing, but I would say that when I tried to refer out to psychotherapist-psychiatrists I respected, generally their rosters were full as well.
I do think that we need to discuss the morality of the full-fee private practice. When I started, I treated well-insured working-class patients, and later I managed to see patients at a reduced fee. Some jobs still promise that combination. The positions protect time for psychotherapy and allow treatment of less well-off patients. In this regard, sub-specialization—for example, in substance abuse treatment or hospice or hospital liaison work—may make a difference.
Ruffalo: Thank you very much for your time and thoughtful responses. I am certain that your work will continue to inspire future generations of psychiatrists and psychotherapists.
