Awais Aftab, MD

Conversations in Critical Psychiatry is an interview series that explores critical and philosophical perspectives in psychiatry and engages with prominent commentators within and outside the profession who have made meaningful criticisms of the status quo.

 is the Theodore J. Castele Professor of the History of Medicine and Associate Director of the Program in Medicine, Society, and Culture at Case Western Reserve University, Cleveland, Ohio. He holds secondary appointments in the Departments of Bioethics and Psychiatry. He earned degrees in history from Wesleyan (BA), Stanford (MA), and Johns Hopkins University (PhD); he also studied psychiatric epidemiology at Columbia University. He is the author of 

Imperial Bedlam: Institutions of Madness and Colonialism in Southwest Nigeria

 (University of California Press; 1999); 

Electroconvulsive Therapy in America: The Anatomy of a Medical Controversy

 (Polity Books; 2020). His articles have appeared in journals including the

 Journal of the History of Medicine and Allied Sciences, Harvard Review of Psychiatry, Bulletin of the History of Medicine

Dr Sadowsky is a leading historian of psychiatry, and one I am proud to call him my teacher. As a psychiatry resident, I attended his lectures on history of psychiatry as part of our residency didactics and benefitted greatly from them. Sadowsky was 

 in 2017 by Greg Eghigian, PhD, about his book, 

, an interview that is well worth revisiting. The present interview is about his new book 

. I admire Sadowsky for his ability to tackle complex and controversial issues with scholarly rigor and fair-minded appraisal, and I am delighted to have him participate in this series.

 Before I jump into my questions, I want to give you an opportunity to introduce your book to the readers on your own terms. What are the issues that occupy you? I suppose some folks on hearing about 

 may reflexively think, “What… another book about depression? Don’t we have enough of those already?” Perhaps in your response you can also touch on why a new history of depression was needed in your opinion.

 First, not many of the books about depression are histories. Take 

The Noonday Demon: An Atlas of Depression

 I think it is great, but it only has a single chapter on history, and it is a brief account. With a few exceptions, most of the history books are focused on antidepressants, and many other treatments have important histories—psychodynamic psychotherapy, cognitive behavioral therapy (CBT),electroconvulsive therapy (ECT), and the many approaches that preceded the 20th century, for example.

A couple of long, detailed histories of depression barely mention ECT. This is weird. I know opinions about ECT are divided—

 is also a book about controversy—but whatever you think about ECT, many psychiatrists consider it among the best treatments psychiatry has for serious affective disorder. Imagine for a moment a historian of medicine in another specialty writing a long history of a particular ailment and skipping briefly over a treatment many doctors say is among the most effective! Previous histories of depression have also been a bit weak on psychoanalysis, which is a strong interest of mine.

Second, little existing historiography of depression focused on patient experiences and self-representations. This is not an easy area to research, because a lot of depression treatment takes place in private office practice, a confidential setting, and little documentation is publicly available. I would have liked to do more, but I hope I pushed the field a little in that direction. But I want to add immediately that some historical and other critical studies of psychiatry see patient voices as only a way to criticize psychiatry. It is important to document patient complaints, but some critics of psychiatry go to tortuous lengths to deny any therapeutic benefits. Honoring patient voices also means seeing that many feel helped by treatment.

I also wanted to draw attention to the politics of inequality. Like so many other illnesses, depression hits different populations to different extents. Solomon did draw attention to this, but most of the historical work on depression restricts inquiry to the gender ratio, which is an important axis, but not the only one. Class, race, and LGBT status, for example, also matter for depression, and I hope future historians will develop this further.

Finally, I wanted to make the story more global, less focused entirely on the West and especially the United States, though I do give the US a lot of attention. Anthropologists and global mental health workers have looked at the wider world, but historians really have not, not for depression. I would like to have gone further here, too, and I hope others will.

 What I love about your book is that it does not settle for easy answers. It embraces the uncertainty, ambiguity, contradictions, and flux inherent in the very subject matter. One of the difficulties with studying depression is it is a fluid entity with variations across both time and space. Is depression the same today as it was in the 17th century when Robert Burton wrote 

? Is depression the same thing in the United States and Nigeria? These are obviously thorny issues. Can you elaborate on the approach you take to address this? How can we tell a coherent history of something if we are having a hard time pinning down what that “something” is?

 The various illness categories—or, as it may be in some contexts, conceptions that are not illness categories—are already included in a long running discussion of the comparability of diagnostic categories and sickness experience across time and space. The discussion is a story itself.

Most of the illness states I compare have a core feature, which is excessive sadness or blue mood, in whatever that particular context considers excessive. But some definitions of depressive illness consider this a common, but not necessary, feature for diagnosis. As you know, major depressive disorder (MDD) in DSM-5 does not require depressed mood as a necessary criterion, provided there is loss of interest and 4 other symptoms. I draw (lightly) on Ludwig Wittgenstein, PhD’s concept of “family resemblances.”

 Things do not need precise overlaps to have similarities that justify comparison. For all the imperfections of the DSM, I think if you look at the menu of symptoms for MDD, they do form a constellation of related signs of suffering that have been clumped together in disparate contexts.

Over-emphasizing cultural differences has its own risks. I show in chapter 1 that the idea that African societies had little or no depression had roots in some frankly racist ideas. Universalizing labels can be a form of cultural imperialism, but over-insistence on difference forecloses possibilities for meaningful comparisons. And some who think depression is a purely modern or Western illness might be surprised at how many cultures, across time and space, have had some conception of excessive sadness as a malady.

Chiara Thumiger, PhD, a great historian of madness in antiquity with whom I am collaborating on a new project, set forth 4 principles for comparisons across time and space

1) The human mind is biological, a part of a shared evolutionary heritage, so you can expect some measure of universality;

2) the mind and mental life are not confined to the brain, but involve other parts of the body;

3) mind and brain are situated in and shaped by culture, so universality will always be limited; and;

4) every individual has an irreducible quality, uniqueness. I think these are great principles to keep in mind.

 It is apparent from your book that one cannot even begin to do justice to the history of depression if one is not willing to tackle the anthropology, literature, phenomenology, philosophy, science, and sociology of depression. Despite the demands, you really do quite a remarkable job of drawing on these diverse fields to piece together the history. In the course of writing this book, did you suffer from a sort of identity crisis at any point, along the lines of, “Am I even writing history anymore?”

 Making sense of anything for me starts with chronology and making rough periodizations. But, as you say, depression demands multiple perspectives. I did sometimes think I was straying from my disciplinary home. But who cares? My goal was to illuminate my subject, not to show loyalty to my discipline.

I have known some academics who are oddly certain that their own discipline provides the ultimate answers. Not me. I consider history a good lens among many, and I am still learning much about the others. I look forward to you and others helping me learn more philosophy.

 You are clearly dissatisfied with a lot of popular critiques of depression. You write, “One thing this book is not is a long lament on the over-diagnosis of depression, and the turning of life’s normal suffering into a medical problem.” Tell us more about your dissatisfaction with this line of criticism.

 Many individuals look at the high numbers of individuals with a depression diagnosis compared to previous times and take it as self-evident that we are over-diagnosing. It is not. Rising diagnosis rates is a valid reason to be concerned about over-diagnosis, but it is not proof. In 1950, much of psychiatry was concerned that we were under-diagnosing depression, so from that perspective, we may be finding a lot that we had been missing. We should at least consider the upside: More diagnosis means that more individuals who could use professional help are getting it.

As for medicating the normal pain of living, we should keep in mind that getting treatment for depression, whether in psychotherapy or with antidepressants, does not protect anyone from suffering, though one hopes it is reduced. Life remains hard. And many patients with depression welcome sadness as a relief from feelings of deadness or numbness inside.

I do offer some defense of antidepressants against their most severe critics in the book. But I also think we are probably over-reliant on them. I think the benefits of psychotherapy, particularly long-term insight-oriented therapy, have become underrated in the antidepressant era.

Still, why some advocates of psychotherapy are so adamantly and categorically opposed to any medication use, or to any suggestion of a biological component to depression, is a mystery to me. I have read many hardline critiques of biological and pharmaceutical psychiatry that are focused on whether the evidence for efficacy is robust, or questioning the philosophical underpinnings of the enterprise, or worried about adverse effects, or indicting the pharmaceutical industry. The thoughtful challenges have reinforced my view that we are over-reliant on drugs, at least for some populations. Other critiques have struck me as dogma masquerading as critique. But even the smartest of them have not convinced me that antidepressants are worthless, a claim several important voices have made.

The biology of depression is incompletely known, and much of it may remain permanently elusive. But the idea that depression is purely psychological is, I think, implausible. But so is the view that depression is purely biological. The dichotomy itself is the problem. Mark Solms, PhD, says that mind and brain are 2 ways of looking at the same thing. I find this more helpful. I do not understand how something could affect either the mind or the brain independently of the other. That is not a coherent idea.

I am getting away from your question a little, but it is an important point. Proponents of a life history emphasis in etiology sometimes take it as self-evident that if trauma or neglect is the cause, the better solution must be psychotherapeutic, as opposed to somatic. Psychotherapy might be the better treatment, but it is not self-evident just from the etiology. And the reverse: individuals who lean towards more organic theories of etiology—genes, brain chemistry—sometimes take it as self-evident that the better solution must be something more physical, like a drug or ECT. Again, this could be true, but it is far from self-evident. In any event, I think monocausal approaches to etiology are misguided.

 You write at one point, “The growing interest in depression led more people, both professional and lay, to learn to label problems as depression, which was a growing idiom of distress before the advent of antidepressants . . . The medications, though, provided new incentives for multiple actors—pharmaceutical companies, doctors, patients, and patient’s families—to identify cases of depression” (page 98).

I suspect you would have been writing a very different history of depression if antidepressant medications had not been developed or if their use had continued to be severely restricted due to tolerability. A curious thing though is that anxiety disorders are more common in the general community than depressive disorders, and antidepressants such as selective serotonin reuptake inhibitors (SSRIs) are not just antidepressants but also antianxiety medications. We could perhaps imagine an alternative history in which anxiety was the idiom of distress, SSRIs had been marketed as antianxiety rather than antidepressant, and you would perhaps be writing a book about the empire of anxiety, and the continuous historical links between melancholia and anxiety! Is it merely an accident of history that we ended up in the age of depression rather than the age of anxiety?

Cross-cultural study has shown that the separation of anxiety and depression is a convention of Western psychiatry (which is rapidly becoming a global or cosmopolitan psychiatry, with both good and bad effects). Many healing systems around the world assume anxiety and depression are aspects of a single thing. And even in the West, the separation is somewhat new, a Kraepelinian innovation.

You present an interesting counter-history. I am not sure the trajectory has been solely drug-driven, though. When the monoamine oxidase inhibitors and tricyclics were developed, psychiatrists were looking for depression treatments, and had been for decades. That is part of why these drugs came to be called antidepressants.

One curious thing though is that important anxiety medications and antidepressants were introduced around the same time, and yet from a broad cultural point of view, we had first an age of anxiety, the immediate postwar period, followed by an age of depression, starting sometime in the 1970s and becoming more visible in the 1980s. In what I consider the most speculative portion of my book, I try to understand that as possibly representing wider cultural moods, perhaps rooted in different forms of capitalism. As speculative as that section is, I included it because I thought it was worth thinking about.

 One thing I had not quite appreciated before reading your book was that the psychoanalysts were quite sympathetic to the notion of gene-environment interactions when it came to depression (the genetic factors being described in the language of their day as a constitutional factor or inborn tendency). It is rather unfortunate that we have this caricature of psychoanalysts in contemporary consciousness as psychological reductionists and dismissive of biology.

 I have long thought that psychoanalytic hostility to biology was a caricature created by the critics of psychoanalysis—I wrote an article about it in 2006

—but I was surprised by how right I was in the case of depression. The psychoanalysts who denied any role of biology in causing or treating depressive illness were defying the views of the most prominent theorists, including Freud; Karl Abraham, PhD; Otto Fenichel, MD; Edith Jacobson, MD; John Bowlby; MD; Silvano Arieti, MD; Julia Kristeva, PhD; Nancy McWilliams, PhD, ABPP; and Otto Kernberg, MD, FAPA. I am naming names here because they are so prominent, and they wrote such widely read books.

 You do not spend much time on the creation of MDD as a diagnosis in DSM-III (and the surrounding politics of it). I wonder how different future course of events would have been in the absence of such a diagnostic category. If you look at DSM-II, depression was fragmented across many different sections under many different diagnoses. We had diagnoses in DSM-II such as “involutional melancholia,” “manic-depressive illness, depressed type,” “psychotic depressive reaction,” “depressive neurosis,” and “neurasthenic neurosis.” I suspect that this DSM-II organization of depression would not have been particularly conducive to the process of empire-building!

 That is a great question, and I could have given this change more attention in my discussion of the DSM. But I wonder if the creation of the category (if not the precise name) of “Major Depressive Disorder” was already in process before DSM-III, reflecting the growing research interest in depressive illness in the years preceding the revision of the manual.

 As a psychiatrist, when I look back at the evolution of depression, a pair of developments in recent decades appear to me to be notable, although their historical significance is probably yet to be seen. The first is the notion of treatment-resistant depression (TRD). There has been increasing recognition that a majority of patients either do not experience resolution of their depression with standard antidepressants, or if do, they do not stay well over time. This has led to more aggressive pharmacotherapy, such as augmenting with atypical antipsychotics, and the recently approved medication esketamine. The second is that bipolar disorder has been steadily encroaching on the territory of depression, such that even in the absence of mania or hypomania, many psychiatrists classify the presentation as being on the bipolar spectrum. Just as the development of antidepressants led to an increase in the diagnoses of depression, development of medications for bipolar depression have led to an increase in the diagnoses of bipolar disorder. Any preliminary thoughts of these developments as a historian?

 It is of course awful that some depressions are so hard to treat. I am not a clinician, but I wonder if many of those that seem resistant to aggressive medication treatment might respond to intensive psychotherapy, which can be expensive, or to ECT, which has risks of adverse effects (risks which are, in my opinion, generally greater than the risks of antidepressants, though I think in some cases, the risks may be worth it). As a historian though, I would ask if the whole category of TRD does not represent some progress, since it implies the other kind. A 19th century alienist could not do much about any severe depressions at all. Many of my colleagues in history of medicine, and especially history of psychiatry, get nervous at any hint of a progress narrative, but I do not mind saying I think we have a better repertoire of treatments for depressive illness now than we did in 1850. All the treatments have downsides, but that is true of medical treatments in general.

I am aware of the rise of bipolar diagnosis, but I have not followed it closely. From a practical point of view, of course, what matters about a label is matching the individual in pain to the best treatment possible, and that is a clinical question. My historian’s take is: The safest prediction is that changes in diagnostic fashions are going to keep happening, and they will not necessarily be driven by genuine scientific advance. The whole enterprise is not worthless, but we should not assume the present trend represents the final or best word. On this point, I do worry about progress narratives.

 Your admiration for Adolf Meyer, MD’s, approach is obvious in your book. At one point you write, “The product they [the patients] want resembles the one Meyer was selling, though it is hard to find in the current market” (p 163).

 The assessment of some of your colleagues has been rather harsh in comparison. Edward Shorter, PhD, in 

 described him Meyer as (unfairly I think) a “second-rate thinker . . . [who was unable] to disentangle schools that were absolutely incompatible, and ended up embracing whatever new came along.”

, is also rather unkind in his assessment, saying that Meyer’s success had “little to do with nosology or therapeutics and everything to do with marketing.”

 Perhaps you would like to set the record straight here regarding Meyer?

First of all, a confession: Meyer bores me. I do not find him an exciting thinker. Second, some of the criticism of Meyer has merit. Greenberg has a point Meyer probably did not advance diagnosis or therapeutics much, at least not in a systematic way that could be passed down. To be fair, though, much progress in psychiatric therapeutics has been somewhat accidental. ECT therapy was based on a hypothesis—an alleged antagonism between schizophrenia and epilepsy—that is no longer considered likely, and schizophrenia is not now the main indication for ECT anyway. A lot of psychopharmaceutical development came from chance observations of the effects of drugs that were being tried for other purposes.

The reason I want to give Meyer some credit is his rejection of reductionism. He wanted clinicians to see patients as biological organisms, with a particular life history and psychology, and a social context. This should be what we all want now. If Meyer was such an effective marketer, why has there been so much reductionism?

 In a very memorable passage of your most recent book, you compare history with psychoanalysis: “One theory about how psychotherapy works is that patients are locked into repeatedly telling the same story about themselves – about their loneliness, or their victimization, for example. The therapy helps patients see that they do not have to repeat the same story. They can tell new stories of their lives. History can have a similar role . . . We do not have to live the same story, time after time” (page 133).

 Later in the epilogue you again refer to the study of history as a means of guarding against compulsive repetition. Can you summarize here what history tells us we can do differently going forward when it comes to our understanding and treatment of depression?

 I will name 2 things. First, we do not need to keep having the same arguments about whether depression is fundamentally biological, or really the result of trauma in life, or the result of social forces. I fear that we will, because of the tenacity with which some individuals hold to the insistence that we must choose. It is my hope that by showing historically how insistence on a single approach leads to dead ends, we might be liberated from this. In the epilogue, I used the image of a quarreling couple, stuck in the same fights for decades on end. A good marriage counselor might be able to remind each partner that the other has things to offer.

I also hope we might be liberated from the cycle of hype and disappointment about new treatments. Most of the major treatments for depression over the last century (psychoanalysis, ECT, CBT, and antidepressants) have been overemphasized by some zealous proponents. Then disappointment (dare I say, depression?) comes when the harms or limitations of the treatments become clear. In turn, we rush to condemn the old treatment as worthless and latch on to something new, which is thought to lack the flaws of the earlier ones. But the old treatments still had something to offer, and the new ones prove to have flaws, though we may not see them right away.

New treatments are going to come, as they should. When they do, let us try not to overdo the hype, the radical rejection of the old ones, or the disappointment in the new that follows. In addition to helping patients overcome repetitions, psychotherapy can help patients to see individuals in and things in their complexity. Sometimes we, as professionals or as whole societies, can act like an individual searching for love, and finding partners we initially idealize, only to come to despise them when they inevitably disappoint us. We would do better to tolerate ambiguity. I hope 

The opinions expressed in the interviews are those of the participants and do not necessarily reflect the opinions of 

is a psychiatrist in Cleveland, Ohio, and clinical assistant professor of Psychiatry at Case Western Reserve University. He is a member of the executive council of Association for the Advancement of Philosophy and Psychiatry and has been actively involved in initiatives to educate psychiatrists and trainees on the intersection of philosophy and psychiatry. He is also a member of the 

Dr Aftab and Dr Sadowsky have no relevant financial disclosures or conflicts of interest.

A History of the Mind and Mental Health in Classical Greek Medical Thought

5. Sadowsky J. Beyond the metaphor of the pendulum: electroconvulsive therapy, psychoanalysis, and the styles of American psychiatry. 

A History of Psychiatry: From the Era of the Asylum to the Age of Prozac

Manufacturing Depression: The Secret History of a Modern Disease

Articles

The story of depression, through time and around the world. 

A People’s History of Depression: Jonathan Sadowsky, PhD

Kenneth S. Kendler, MD, is the Banks Distinguished Professor of Psychiatry and professor of human genetics at Medical College of Virginia, 

. Kendler received his medical and psychiatric training at Stanford and Yale University, respectively. Since 1983, he has been engaged in researching the genetics of psychiatric and substance use disorders. He has utilized methods ranging from family studies to large-sample population-based twin and adoption studies to molecular genetic studies, including linkage, genome-wide association studies, and whole genome sequence aimed at identifying specific genes that influence the vulnerability to a variety of psychiatric disorders. Since 1996, he has served as director of the Virginia Institute of Psychiatric and Behavioral Genetics. Kendler has received numerous national and international awards and is an editor of the journal Psychological Medicine. He played an active role in developing DSM-III-R, DSM-IV, and DSM-5 (for which he chaired the Scientific Review Committee), and he is currently the vice-chair of the 

American Psychiatric Association DSM Steering Committee

. He has published more than 1200 articles. Dr Kendler is one of the highest-cited researchers in psychiatry: according to Google Scholar, his academic work has been cited nearly 170,000 times, and his H-index is 196.

 is internationally renowned and highly respected for his work in psychiatric genetics, he has been further catapulted into fame over the past 2 decades through his widely read commentaries on the philosophical foundations of psychiatry. I recently had the pleasure of reading the book 

Toward a Philosophical Approach to Psychiatry: The Writings of Kenneth Kendle

The book is a selection of 21 of his most important philosophical and historical papers published throughout his career, addressing topics such as the classification and nature of mental disorders, mind-body relationship, causality and explanation in psychiatry, and historical studies in psychiatric nosology. My admiration of Kendler is no secret at this point.

 I really like your description of how psychiatric nosology sits in a historically contingent developmental arc. How is this 

 relevant to ongoing nosological debates in psychiatry, and how has ignorance of this history impeded our efforts at making progress?

With respect to psychiatric nosology, there is quite a bit of truth to 2 worn maxims: (1) If you don’t know where you have been it is hard to see where you are going, and (2) if you don’t know your history, you’re are at high risk to repeat your prior mistakes. So, I think history can provide a context and a background for what nosology can do and where it has taken wrong roads in the past.

 I read that you collaborated with a translator to obtain English translations of 

 previously untranslated works. Will these translations be published or see the day of light in some manner?

They are all sitting on my hard drive, and that of the translator Ms Astrid Klee, but there is a lot more there than just Kraepelin. A very small percentage of the relevant German psychiatric literature of the late 19th and early 20th centuries have been translated. I have thought about trying to set up a website to make these widely available. That will take time, energy, and a bit of resources. If any of these readers want to help, be in touch.

 You have argued for a scientific pluralism where there are multiple explanatory perspectives available to us to understand psychiatric disorders, and one perspective cannot be reduced to another perspective. You describe your pluralism as empirical and hard-nosed, by which you mean that risk factors must earn their place at the table. It is a little unfortunate that many discussions of pluralism in psychiatry tend to get stuck in debating the merits and demerits of the 

. For the most part, you have managed to stay away from those controversies and have successfully charted a course for pluralism independent of any baggage that the biopsychosocial model brings. Is it time for our field to abandon the biopsychosocial model?

Its core idea was on target, but its implementation was so non-specific as to blunt any rigor it might have once had and its ultimate utility. I would not mourn its passing, but it did some important historical work.

 DSM-5 defines mental disorder as a syndrome that, among other things, “reflects a dysfunction in the psychological, biological, or developmental processes underlying mental functioning.” What is the intended meaning of the term dysfunction? Did the members of the DSM committee have any clarity or consensus regarding what it means?

The hope that we could have a crisp set of inclusion and exclusion criteria for what constitutes a psychiatric dysfunction is a wonderful idea but (in my view, not for lack of trying) impossible. There are too many social and conceptual nuances. So, although there is a general idea of what dysfunction means, operationalizing in the way that would give substantial sharp precision to the definition has not proved possible.

 Could you articulate what that general idea of dysfunction is? I am also interested in whether dysfunction necessarily implies that the locus of the problem is primarily inside the individual and not in interpersonal relations and social context? And, if there is such an implication, does this relegate interpersonal and social causal risk factors to a more secondary status?

 The general idea of dysfunction is commonsensical—that the relevant psychobiological system is not doing what it is supposed to do. Examples might include providing your higher centers with an approximately veridical sense of the world around you, keeping levels of anxiety roughly appropriate to the real dangers being confronted, producing mood states approximately congruent to the environmental situation, etc. The DSMs have traditionally seen disorders as existing within individuals and, for example, avoided providing 

 for dysfunctional marriages or families. So, in that sense, the underlying disturbance is seen to exist within individuals. I do not see that definition having much of anything to do with the causes. Environmental experiences like severe childhood sexual abuse can clearly cause dysfunctional mood-modulation systems as well as a high genetic vulnerability.

 To what extent is the dappled distribution of causal risk factors of psychiatric disorders a result of the heterogeneity of constructs? For example, a discussion of the distribution of causal risk factors of chronic fever disorder is not likely to be very meaningful. Furthermore, it is true for major depressive disorderas a category that no single causal risk factor has dominant explanatory power, but it may be the case for specific individuals with depression that there are causal risk factors of large explanatory power. What are your thoughts?

I don’t buy it. Most psychiatric disorders are, I think, multifactorial all the way down. I also think that for a few affected individuals they do suffer from disorders largely as a result of one major cause. I sometimes call your model the mental handicap model. Imagine you were a physician in the 

 caring for what would then have been called idiots or imbeciles. If you could apply modern diagnostic methods to that group, a fair proportion would have a range of specific causes: Down syndrome, fragile X, a host of autosomal recessive disorders impacting on all kinds of brain-relevant genes that, when mutated, produce widespread central nervous system dysfunction, a range of small deletions, and a whole bunch would be nonspecific multifactorial. That is, a lot of your cases (but far from all) would have largely monocausal conditions. This has been long postulated for psychiatric illness. It has, in my judgment, been largely a pipe dream. I think that lesson applies more broadly.

 Your philosophical writings are published and featured prominently in leading American psychiatric journals, which do not usually publish articles related to philosophy of psychiatry. I have wondered if one of the reasons mainstream journals are so receptive to your philosophical work—aside from the undisputed academic quality and rigor—is that your conclusions do not threaten or destabilize the status quo in a way that conclusions of some of the other philosophical commentators do. It was interesting to me that one of your more controversial articles—on the dopamine hypothesis of schizophrenia—was also the one that had a difficult time getting accepted. Why are leading American psychiatry journals publishing so few philosophical articles and so afraid of controversy?

I have made a deliberate attempt over the past 15-plus years to try to crack open the leading psychiatric journals for articles with a primary philosophical and/or historical content.

I agree with you that my ability to do that has been, in part, a result of the fact that I have achieved some standing in empirical areas of psychiatric research. To be vernacular, I have accumulated some street cred. I also think the papers were written in a way the audience could understand and that spoke to their concerns. I did not consciously edit the papers to make them less controversial. I still think that nonscholarly issues impacted on the problems we had with the dopamine hypothesis paper—that was perhaps a special case. I do not want to blow up psychiatry. I believe quite deeply in our clinical and research mission, but we surely can think more clearly about a number of issues in our clinical work, nosology and research.

 Some critics hold the view that if a classification can be misused, it will be misused. To what extent should the concern for misuse constrain classification decisions? Do the creators of DSM have a responsibility to make academic as well as public educational efforts to reduce the ways in which the diagnostic manual is misunderstood and misused?

 Our primary responsibility on DSM is to our patients and the research community that we assist. But possible misuse does, appropriately, arise in nosologic debates. For example, it played a key role in the opposition to the late-luteal-phase dysphoric debate in DSM-IV—that the diagnoses would be used in ways prejudicial to women. So, it would be unrealistic for DSM to ignore completely the possible misuse of the document. But, if the chips are down, I think serving our patients and research is the more important mission. It is impossible to control all the possible misuses of DSM and, if you took that concern too far, it would be paralyzing.

You have argued persuasively that the conditions we call psychiatric disorders are not merely constructed by individuals and cultures, and that these conditions have some basis in the objective reality, but what do you say about the construct of mental disorder itself? To what extent are concepts of health and disease, and characterizations of disorder grounded in objective reality?

 As an aggregate concept, I think psychiatric illness (or mental illness) is a real thing out in the world. Going from that to specific disorders, be it 

or narcissistic personality disorder, that is harder, and the role of social construction or historical accident becomes greater. A metaphor (not my own) is to imagine rewinding the tape of history, say, to 20 centuries BCE and re-running it 100 times until society and medicine developed enough to establish something like what we call psychiatry. I would bet that some construct like insanity/schizophrenia would be there almost all the time. I would not say that for a number of more specific disorders in our manual.

You have stated that psychiatric disorders are multicausal, similar to how coronary artery disease, hypertension, and type 2 diabetes are multicausal in medicine. Multifactorial disorders can still have final common pathways onto which those risk factors converge, and those final common pathways provide a lot of explanatory power.

For example, multiple causal risk factors for diabetes converge onto insulin production or insulin resistance. Do you imagine that various causal risk factors for psychiatric disorders also converge onto final common pathways? If it turns out that psychiatric disorders are more like homeostatic property clusters, then it may very well be the case that there may be no final common pathway. In that situation would it still be fair to say that psychiatric disorders are multicausal in the same way as 

 Great question and topical. I have been involved with a research team trying to determine the biological coherence of the signals emerging from genome-wide association studies. This is a question rather closely related to the one you pose. I guess my main answer is “I hope so.” As you note, that is the case for lots of other complex disorders. To be a bit more precise, I think it is realistic that our “big” syndromes (eg, schizophrenia, 

, depression, anxiety disorders) reflect broad syndromes with some meaningful subtypes within them for which different therapies might have different success. It is awfully optimistic to think that all of them go through one tight final common pathway at which an intervention, properly designed, could be nearly curative for all cases. But similarly, the idea that there are hundreds of kinds of each disorder, each with its own needed therapy, is both very pessimistic and likely unrealistic.

You may not remember during the 2016 American Psychiatric Association Annual Meeting, I approached you nervously after a session to express my admiration of your philosophical work. I was a second-year resident at that time. You gave me advice that was something along the lines of “It is very important that you read a lot and that you read very widely.” That often comes to my mind because every day I discover that there is so much more to learn. I am impressed by how you have managed to apply ideas from other areas of philosophy of science to psychiatry. Given that you have limited time, how do you prioritize and decide what you will read?

There is never enough time to read everything of interest; I often think of the metaphor of trying to drink a waterfall. I am rather disciplined in what I read. I am good at skimming, and if the first 20 pages don’t look good, then I bail. The older I have gotten, the less patient I am with philosophical books written for other professional philosophers, filled with philosophy-speak. If they are not interested in communicating with me, I have to be very, very interested in what they are saying to soldier on.

I read 5 or 6 books at once. I am now reading about the 

 of genetics and cannot get enough of it. I read very little fiction. But I do keep up other lines of reading, usually late at night, that help keep me rounded, or listen to Audible when I bike or commute. Recently, I have been reading more history of psychiatry than philosophy. I have lots of bookshelves at my home and work office, but I am running out of space.

This may be an unfair question: If posterity could remember you predominantly for either your research work in psychiatric genetics or your philosophical work in psychiatry, what would you prefer it to be?

I have to laugh—yes unfair—Sophie’s choice that I cannot answer. I hope I am remembered, if at all, as someone who has tried to weave together empirical and conceptual work. It has been a wonderful career, and the older I get, the harder it is to pull these 2 parts of what I do apart. My identity is with the whole of it.

Conversations in Critical Psychiatry is an interview series that explores critical and philosophical perspectives in psychiatry and engages with prominent commentators within and outside the profession who have made meaningful criticisms of the status quo. The opinions expressed in the interviews are those of the participants and do not necessarily reflect the opinions of 

Interviews published in this series can be accessed at www.psychiatrictimes.com/series/critical-conversations-in-psychiatry.

 is a psychiatrist in Cleveland, Ohio, and clinical assistant professor of psychiatry at Case Western Reserve University. He has been actively involved in initiatives to educate psychiatrists and trainees on the intersection of philosophy and psychiatry. He is also a member of the 

Weaving Conceptual and Empirical Work in Psychiatry: Kenneth S. Kendler, MD

, was published online ahead of print on May 26, 2020. -Ed

An interview with Kenneth S. Kendler, MD, vice-chair of the American Psychiatric Association DSM Steering Committee, author of more than 1200 articles, and one of the highest-cited researchers in psychiatry. 

A Psychiatrist Weaving Conceptual and Empirical Work

 is a health sciences clinical professor in the Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine at UCLA. His clinical work focuses on the treatment of individuals with severe mental disorders, including schizophrenia. His research covers clinical trials in schizophrenia and early intervention for young individuals at high risk for psychosis. In 2005, he received Young Investigator Awards from both the International Congress of Schizophrenia Research and the American College of Neuropsychopharmacology. He has authored over 100 journal articles, abstracts, and book chapters. He also writes the 

, exploring the grey area between psychopathology and normality and the psychological underpinnings of everyday life.

My interest in highlighting critical perspectives extends to the way in which these critical perspectives are being received by thoughtful, well-meaning individuals from the mainstream psychiatric community. Joseph Pierre, MD is such a psychiatrist. He has engaged with the critical perspectives in his scholarly writings as well as in his online interactions. This interview focuses on the debates surrounding the DSM and psychiatry generally, delusion-like beliefs, social-epistemological conceptions of delusions, engaging with service users, and controversies relating to long-term outcomes with the use of antipsychotics. On most of these issues, Dr Pierre’s views are closer to prevailing psychiatric wisdom, but he also incorporates aspects of critical thought in productive ways that show the potential for common ground. My hope is that this interview will be informative for psychiatrists on both critical and conventional ends of the spectrum. For the critically minded, it highlights why many criticisms fail to persuade someone like Dr Pierre, and for the conventionally minded, it highlights what psychiatry stands to gain by engaging with the criticisms.

 In your contributions to the pluralogue on essential questions in 

 you have entered the conceptual debates surrounding definitions of mental disorder and the nature of psychiatric diagnoses. I am curious to know how your understanding of these issues has evolved over the course of your career. What has been the shape and form of your own critical engagement with the field, and what have you learned from it?

 My interest in critical psychiatry started at least a decade ago, sparked in large part by Allen Frances, MD’s series of critiques of the forthcoming 

. He highlighted the potential unintended consequences of premature revisions in the absence of a true paradigm shift. I was particularly interested in, and concerned about, the issue of diagnostic expansion, and how dimensional perspectives were going to be included in 

, which focuses on how we define psychotic disorders like schizophrenia, as well as individual 

 In an attempt to work through these issues in my own head, I wrote a series of papers about the borders of mental illness in DSM.

 I also participated in several collaborations with Dr Frances and others like Ronald W. Pies, MD, with whom I had established a collegial kinship around this topic.

In addition to reviewing the academic literature for my writing, I also read books like Richard Bentall, PhD’s 

. In doing so, I began to see some flaws with approaches that were critical of the DSM. For example, the pre-

 criticism that each successive edition of the DSM resulted in the further proliferation of diagnoses could be explained by the development of finer distinctions of psychopathology, rather than necessarily reflecting more individuals being classified as mentally ill. Alternative definitional approaches, such as Jerome Wakefield, PhD’s harmful dysfunction model, did not seem to make it any easier to define mental illness operationally. My chapter in 

 was the culmination of my attempts to reconcile criticisms of the DSM with clinical pragmatism. I also advocated a broader mental health-mental illness continuum within psychiatry. Although I started my journey with a critical eye towards 

, and have never felt that a new edition was justified, I nonetheless ended up becoming something of an apologist for the DSM as a good enough guide for clinical work, for which I have yet to see a better alternative.

That said, in my writing about the how we define mental disorders, I call for greater appreciation of contextual utility, whereby psychiatric diagnosis serves disparate purposes in different settings. DSM is a reasonable guide for clinical work and interventional research, but it is not as well-suited to other domains, such as etiological research or forensic work where it can be counter-productively confining. The biggest problem with the DSM’s contextual utility may be that it is primarily used not so much as a clinical guide but as a billing manual. It incentivizes over-diagnosis and over-estimates of severity based on reimbursement issues that demand unwarranted diagnostic confidence.

 Are there erroneous ways of thinking about psychiatry that you observe to be common among your psychiatric colleagues and trainees these days, and which you would like to correct?

 Although I can be a vocal defender of psychiatry against calls to abolish the DSM, psychiatric medications, or even our profession as we know it, I do see a variety of appropriate criticisms about the way psychiatry is sometimes practiced. These criticisms deserve greater attention.

As someone who did their residency training during the decade of the brain, at the height of optimism that neuroscientific discovery would result in both thorough understanding of mental illness as well as its cure, I have no doubt 20 years later that this was a vastly over-optimistic ambition. There is also little question that robotic 15-minute med checks that focus on symptom checklists and medication side effects represent bad psychiatry. In my work as an educator, I teach and advocate for a biopsychosocial perspective that views mental disorders as arising from interactions between environment and individual biology, similar to how we understand cardiovascular disease, diabetes, or even broken bones within the so-called medical model. Within this framework, psychotherapy and social interventions (eg, treating homelessness) are integrated with pharmacotherapy and sometimes used in lieu of somatic interventions. The systemic issues that relegate psychiatrists to mere prescribers (which you discussed with 

, PhD, DFAPA), and have severely curtailed the duration of inpatient stays, has done our patients, and our profession, a disservice.

Over-prescription at the healthier end of the mental health-mental illness continuum is a valid concern about psychiatric care as it is practiced in some settings. I often see anecdotal evidence of treatment moving too readily to prescription, polypharmacy, and high-dose medication at the expense of watchful waiting, supportive care, psychotherapy, social interventions, treatment of substance use disorders, careful diagnostic re-assessment, and de-prescribing. That is partly due to psychiatrists over-identifying with their role as psychopharmacologists, but it is also driven by patients desperate for a pharmacologic quick fix. SSRIs and second-generation antipsychotics have thankfully reduced the risk-benefit ratio of psychiatric medications, but this has resulted in misperceptions that they are happy pills to be doled out like candy, often by primary care providers. These are important critical issues that deserve more attention in our profession. At the same time, however, they should not be used to argue against pharmacotherapy in toto, especially at the more severe end of the mental health-mental illness continuum.



 You have a long-standing interest in delusion-like beliefs (DLBs), conspiracy theories, and overvalued ideas, including the challenges these grey areas present in forensic settings. An important recognition, as you put it, is that “the formation and maintenance of beliefs and DLBs do not occur exclusively within the vacuum of an individual's mind, but are influenced by social forces.”

 Can you expand on your work in this area and some of the main conclusions that you have arrived at?

Yes, psychosis has been a central interest of mine as a clinician and former early career researcher, especially with regard to the phenomenology of delusions and hallucinations and the grey area between pathology and normalcy. The DSM may be a good rough guide for diagnosing and treating disorders, but it falls far short of capturing the heterogeneity of individual symptoms or experiences like voice hearing and delusion-like beliefs. It does little to help clinicians decide what is or is not pathological. I do not see that as DSM’s fault though—clinical expertise and research discoveries are needed to move us forward in that regard.

In my academic work, I have been curious about how understanding pathological phenomena like delusions might contribute to a better understanding of more normative but potentially problematic phenomena, like political, religious, and false beliefs held with extreme conviction. The idea that a kind of subclinical psychopathology, including cognitive biases, underlies not only dysfunctional but also normal human behavior is the premise of my blog 

.

Back in 2001, I wrote a paper about distinguishing between religious and delusional beliefs in which I noted, “even the most bizarre delusional idea might find a match in an average episode of the X-Files.”

 In recent years, while focusing on conspiracy theories, I have paid more attention to how historically unprecedented access to information, misinformation, and deliberate disinformation shapes our beliefs and how we, as psychiatrists, sort out whether a belief is a delusion in that context. These days, the vexing part of that challenge is that both pathological delusions and normative delusion-like beliefs are typically drawn from pre-existing cultural ideas and memes with fringe ideas increasingly shared in the internet era. With delusion-like beliefs such as conspiracy theories, the evidence to support them is usually found while searching for answers online. Misinformation is so easy to find not only because individuals believe and propagate it as a form of social affiliation, but also because peddling disinformation can be profitable, as well as a tool of political propaganda. In that sense, delusion-like beliefs, not unlike like psychiatric disorders, are very much a byproduct of interactions between individuals and their social environments.

 Similar to the way you emphasize how DLBs are formed and maintained by social forces, there is an emerging social epistemological conception of delusion that sees delusions as characterized not merely by individual epistemic deficits but also social epistemic deficits.

 Has your engagement with the conceptual literature on DLBs had any influence on your understanding of delusions?

 My recent work on conspiracy theories is based on a similar argument, that existing research has focused too much on individual psychology and that common belief in conspiracy theories can be better understood within a socio-epistemic model, whereby mistrust of authoritative sources of information leaves individuals vulnerable to misinformation in the form of conspiratorial counter-narratives.

 I would agree that such social perspectives have been given short shrift in understanding delusions proper as Kengo Miyazono, PhD et al

 and more recently Vaughn Bell, PhD and colleagues

 have argued, but overreliance on social theories fails to adequately account for delusions. For example, Miyazono et al hypothesize that “testimonial isolation” and “testimonial discounting” (being sheltered from or otherwise ignoring alternative explanations) can explain delusion formation and maintenance in schizophrenia, but a quick look at the world around us reveals that neither is specific to psychotic disorders. Most of us are shielded from views that contradict our own within social echo chambers and online filter bubbles and we routinely discount the dissenting testimonies of others (shouting “fake news”) when they are encountered.

In my experience, the key to distinguishing between pathological delusions and normative delusion-like beliefs lies within exploring their evidential justifications. We do not hold beliefs only due to epistemic trust in the testimony of those who tell us or teach us things, as Miyazono et al suggest, but also based on folk intuitions and subjective experience. Phenomenologically, subjective experience, and in particular anomalous experience, figures more heavily into clinical delusions than it does in delusion-like beliefs. Conspiracy theories, for example, are out there, waiting to be found and shared, whereas delusions have been traditionally regarded as unshareable

echoing what Jaspers wrote about delusions lying beyond intersubjective understanding. The unshareability of delusions often hinges on their idiosyncratic and self-referential content, based on anomalous subjective experiences that are prioritized over other types of evidence. In that sense, I see individual psychopathology as indeed central to understanding delusions in a way that is not as relevant to delusion-like beliefs.

As I suggested earlier however, the internet has thrown a fascinating wrench into efforts to distinguish delusions from delusion-like beliefs. Bell et al

 suggest that unlike shared delusion-like beliefs, idiosyncratic delusions are socially maladaptive, but it is now possible for subjective anomalous experiences and their delusional misinterpretations to be validated and reinforced within social spaces online. This can provide a false normalization of psychotic symptoms, whether we are talking about voice-hearing or paranoia in the form of so-called “gangstalking.” This might be de-stigmatizing within those online spaces, but can also steer individuals away from the potential benefits of psychiatric treatment. DSM-5 all but did away with 

or shared psychotic disorder, leaving little guidance for clinicians grappling with seemingly delusional beliefs that are supported by testimonial validation in the internet era.

As with mental disorders more broadly, what we really need to understand delusions and delusion-like beliefs are truly integrative models that do not point to individual biology or social/environmental exposure as a reductionistic either/or. Psychiatric disorders are not merely normal responses to difficult life circumstances and trauma, just as they are not inevitable byproducts of genetics or chemical imbalances. We should also bear in mind that while experiences like delusions and delusion-like beliefs can be modeled as existing on a continuum, they may yet be important categorical distinctions, not only between delusions and delusion-like beliefs, but also between different types of delusions.

 As you are well aware, there is considerable anger directed at psychiatry from consumers/survivors/ex-patients and the self-described “harmed patient” community. I personally was not aware of the intensity and magnitude of this until I became active on social media. I am interested in your thoughts on this phenomenon and how it is changing the nature of public psychiatric discourse. This also taps into the larger issue of service-user involvement in psychiatry. As 

, PhD, emphasized in her recent interview with me,

 there is an immense need for diverse service users to be invited to the table as epistemic agents and interlocutors. I see that as an admirable and essential goal, although it is not without its challenges when it comes to implementation. The historical attitude of the field toward such an undertaking has been one of reluctance, perhaps because mental health consumerism has been viewed by many with some degree of suspicion and as an extension of the antipsychiatry movement.

 What is your opinion on the involvement of service users in psychiatry as epistemic agents and how should the field navigate the associated challenges?

 Yes, the internet has profoundly affected social discourse and even epistemology itself by providing a platform for previously unheard voices to be heard. Some have reasonably argued that this democratization of knowledge has led to the death of expertise, such that encountering dissenting opinions is par for the course within online spaces whether we are experts or not. During my career, there is always been a picket line of protesters at American Psychiatric Association meetings, but now that picket line shows up on my Twitter feed. Navigating such interactions and deciding how to engage, or not, is fraught with challenges, especially given that we are often talking about patients or ex-patients outside of the doctor-patient relationship

Generally speaking, there is potential for dialogue around improving psychiatric care, with medication side effects like akathisia or phenomena like benzodiazepine and antidepressant withdrawal in mind, but accounts of more idiopathic experiences (eg, withdrawal symptoms associated with being bed-ridden for years on end) and the use of polemical language (eg, “prescribed harm,” “brain damage,” “torture”) can provoke defensiveness on the part of clinicians and render common ground elusive. And unfortunately, online group processes often foster fractious discord, splitting, and projective identification rather than facilitating the kind of face validation that such patients are often looking for.

Like so much online discourse these days, there is also a dialogical impasse created by competing epistemologies. Self-identified “harmed psychiatric patients” are typically speaking from individual experience, whereas clinicians see things based on clinical experience with our own patients as well as the published literature. The idea of validating service users as epistemic agents is not novel, however; it is a fundamental part of our clinical work with patients, albeit with an eye out for potential cognitive distortions and false beliefs based on subjective experience. I maintain that many who identify as “harmed patients” would be best served by finding an empathic clinician who can listen to their stories and work with them in a therapeutic setting, but I understand that some have given up on that possibility and see it as anathema.

My work on conspiracy theories, and especially following the anti-vaccine movement, has led me to appreciate that offering consumers a seat at the table is a necessity for authoritative institutions (such as science, medicine, and psychiatry) to mend the kind of mistrust that can underlie belief in misinformation. But how to best listen to, learn from, and dialogue with patients outside of clinical spaces—and just what table we are referring to—is far from clear. As with so many polarized debates these days, a good start might be to steer clear of polemics in favor of finding middle ground, but that is often not what individuals want and is better achieved offline. We should also avoid the pitfall of framing the dialogue as between an “us” and a “them,” or as necessarily a dichotomy between subjective lived experience vs objective clinical experience and evidence, bearing in mind that psychiatrists (and other mental health professionals) and patients are not mutually exclusive groups, as Dr Jones exemplifies. Historically, lying on the couch was a core part of psychiatric training and many of us have done so as real patients as well.

 The view that individuals with schizophrenia who stay on antipsychotic medications do worse over the long-term than those who discontinue them has generated a lot of controversy, and you have also commented on this issue.

 I do not think there is any conclusive evidence that this is the case and I certainly am well-aware of the therapeutic potential of antipsychotic medications and the immense relief they bring to many, but I also do not think that the issue is quite that clear-cut.

Your view seems to be that the only thing that can really resolve this is a randomized-controlled trial (RCT) of antipsychotics in the long-term treatment of schizophrenia. Using RCTs to look at outcomes at, say, the 5-year mark (let alone 10 years or 15 years) is extremely challenging. The expected dropouts alone would probably render the results inconclusive, and the difficulty of recruitment would bring up limitations of generalizability. So I am not confident that an RCT of such duration, even if it is somehow conducted, will inevitably satisfy critics on either side. Furthermore, I want to push back against the idea that only RCTs can settle questions such as this. Henrik Ohlsson, PhD, and Kenneth Kendler, MD, made a similar point in a different context (psychiatric epidemiology) when they wrote, “The view that causation can be definitively resolved only with RCTs and that no other method can provide potentially useful inferences is simplistic . . .We need to avoid the extremes of overzealous causal claims and the cynical view that potential causal information is unattainable when RCTs are infeasible.”

 Obviously, observational studies always have inherent limitations, but it does not mean that they cannot offer us meaningful information with a certain degree of confidence.

Secondly, one of your arguments against the observational studies, such as by Martin Harrow, PhD (and there are at least 8 observational studies which have looked at this)

 is that the unmedicated patients who did better likely had schizophrenia of lesser severity or better prognosis to begin with. But that is not really consistent with the data from Harrow’s study, in which the results remain unchanged even when data is analyzed separately for good-prognosis and poor-prognosis patients.

 Again, I do not want to give the impression that these data are conclusive or are not subject to limitations or confounding, but I also think it is about time we start taking observational data a bit more seriously and move beyond the position that in the absence of RCTs the only reasonable option for us is to stick to our guns (as a reminder, the causal association between cigarette smoking and lung cancer was established based on non-RCT data). At the very least we can openly acknowledge uncertainty both in communication with the public as well as with our individual patients, that when it comes to long-term outcomes (3 years or more) of antipsychotic treatment in schizophrenia, we do not quite know with certainty whether antipsychotics on average improve them or worsen them, so that our patients and their families can make informed decisions taking into account this uncertainty. Your thoughts?

 RCTs are often the best way to establish causal relationships between clinical interventions and associated events, but they are obviously not the only way. For example, we do not have RCTs to prove that cannabis causes schizophrenia, but we do have longitudinal cohort studies that support the hypothesis. Note that the idea that cannabis use over time, particularly at an early age, can trigger or increase the risk of developing a chronic psychotic disorder like schizophrenia matches observations that cannabinoids can also cause psychotic symptoms more acutely.

With antipsychotics and psychosis, RCTs have resulted in consensus that antipsychotics can reduce psychotic symptoms in the short term. There is also support for antipsychotic effectiveness in relapse prevention from both longer-term RCTs as well as discontinuation studies.

 And anecdotally, in my clinical practice as an inpatient psychiatrist, psychotic relapses are almost always preceded by discontinuation of antipsychotic therapy (and/or illicit drug use) in the weeks to months before rehospitalization. That should be the starting point for making sense of non-randomized longitudinal studies in schizophrenia with the seemingly paradoxical finding that those who continue on antipsychotic treatment have a worse course of illness than those that do not. While it is theoretically possible that antipsychotics could cause that worsening (eg according to the unproven “dopamine super-sensitivity psychosis” hypothesis), especially when we consider that side effects like extrapyramidal toxicity can result in poorer outcomes, it makes more sense to suspect that this is a disease severity effect within a naturalistic, non-randomized study design. If we imagined similar long-term studies of medications like anti-hypertensives or sulfonylureas, I do not think we would be surprised if illness morbidity was greater among those who had to remain on medication due to illness severity, without invoking a medication toxicity effect.

, Robert Whitaker quotes Martin Harrow as explaining the paradoxical finding of his 2007 study as follows: “It’s not that those who went off medications did better, but rather it was those who did better [initially] who then went off medications.” Subsequently however, both Dr. Harrow

 blog), have noted that that explanation does not seem to match estimates of prognosis at baseline. But baseline estimates of prognosis are not a great proxy for whether patients were actually doing better over time and do not explain why antipsychotics were deprescribed or discontinued. In fact, Harrow’s 2017 study

 demonstrated that better work functioning was associated with not being prescribed antipsychotic medication in their 20-year follow-up study. While we are still left with a potential chicken and egg debate about the arrow of causality, this could be interpreted as validation of Harrow’s original disease severity explanation: that those who were functioning better were able to stop and stay off medications.

 commentary was one of many responses to a review article about the risk-benefit of long-term antipsychotic treatment in schizophrenia, which collectively demonstrate that psychiatry recognizes different explanatory hypotheses and that there is room for healthy academic debate on this issue (as well as many others, such as antidepressant effectiveness). In the spirit of seeking middle ground, we should acknowledge the take home message of the longitudinal studies in schizophrenia by Harrow and others—that lower doses of antipsychotic medication that minimize side effects are usually best and that some individuals diagnosed under the umbrella of schizophrenia may be able to stop and stay off antipsychotic medications. In my practice, I often deprescribe antipsychotics and revise chart diagnoses of schizophrenia. What we need to clarify based on well-designed clinical trials is when that is most appropriate and when it is not.

It is said that tolerating uncertainty and ambiguity is a requisite for practicing psychiatry. The challenge with acknowledging areas of uncertainty within a public sphere is that doing so can be weaponized to reinforce a post-truth view of the world in which everything is debatable, and any counter-position is just as valid. Although I like to think of myself as a middle of the road kind of individual, it is important to keep in mind that the truth does not always lie in the middle of a debate, whether we are talking about climate change, vaccines, or antipsychotic medications.



 There seems to be an implicit hint in your response that individuals diagnosed with schizophrenia who are able to stop and stay off antipsychotic medications may not really have had schizophrenia to begin with. Is this something you wanted to imply or perhaps I am misreading? While misdiagnosis certainly happens, I think it is commonly accepted–for instance, by psychiatrists such as E. Fuller Torrey, MD

–that a subset of patients with schizophrenia (a minority but could be as much as 20%) are able to recover and stay well off antipsychotic medications. Would you agree?

 The conundrum of schizophrenia is that even Bleuler conceptualized the term to account for a heterogeneous group of superficially similar disorders. So, it makes sense, almost tautologically, that those who benefit from or need long-term medication and those that can safely stop and stay off them (something that has been reported as far back as the 1950s) might at the very least have different types of schizophrenia. In retrospect, that might also mean that a patient did not have schizophrenia in the first place (eg, perhaps they had a substance-induced psychosis or voice-hearing experiences inappropriately labeled as a psychotic disorder) or it might reflect a variant of spontaneous recovery (eg, schizophreniform disorder or simple schizophrenia). It goes back to thinking of DSM diagnoses within a particle and wave model: both as a continuum but also with meaningful categorical distinctions within a spectrum. Personally, I think both conceptualizations are useful for understanding schizophrenia.

 I definitely share your clinical experience that psychotic relapses in individuals with schizophrenia are very commonly preceded by discontinuation of antipsychotic therapy. However, some psychiatrists are beginning to wonder to what extent such relapses may be precipitated by abrupt rather than gradual discontinuation. An interesting on-going RCT investigating this question is the RADAR trial by Joanna Moncrieff, MD, et al.

 In this RCT individuals with schizophrenia and related disorders (who have had more than 1 episode) are randomized to either a clinically supervised, gradual reduction of antipsychotic medication, leading to discontinuation when possible, or to continued maintenance treatment. Investigators will look at social functioning and relapses, with the primary outcome being social functioning at 24 months. Now this trial will not end until 2022, so we will not get to see the results any time soon, but hypothetically speaking, if the trial were to demonstrate that a large number of individuals are able to gradually discontinue antipsychotic medication with superior social functioning at 24 months—would that have any impact on your confidence that the results of long-term observational studies are adequately and sufficiently explained by disease severity confounding?

 I would certainly welcome a well-designed RCT that might help to resolve any debate there is about the association between poorer outcome and antipsychotic treatment in naturalistic studies. However, the more important question to answer is how to predict who can and cannot safely discontinue antipsychotic medications. Inclusion criteria that allow entities like “other non-affective psychosis” as proposed in Moncrieff et al

 mean that the study population may end up being farther away from the true schizophrenia side of the psychotic spectrum, which has presented a problem in interpreting recent studies of first-episode psychosis.

 What we really need are clinical trials that explore the risk-benefit of antipsychotic medications and non-antipsychotic interventions within more narrowly-defined psychotic disorder subtypes.

With regard to antipsychotic discontinuation and the explanatory value of super-sensitivity psychosis, it is important to note that psychotic exacerbations often do not come immediately as a rebound effect—they often take weeks to months. This challenges the notion that relapse is best attributed to the effects of acute antipsychotic withdrawal.

 In a meta-analysis of placebo-controlled RCTs for relapse prevention in schizophrenia, antipsychotic medications were favored over placebo regardless of abrupt discontinuation or gradual tapering.

 Note also that in individual studies like that of Gitlin and colleagues,

 where depot antipsychotic medication was discontinued (amounting to a slow taper), the vast majority of patients still experienced a psychotic relapse months later. Anecdotally, I once discontinued antipsychotic medication in a patient who I did not think had schizophrenia and I am sorry to say that after several months of a slow taper, I was proven wrong. I still regret that clinical error.

 In conversations with you (and with some of the other respected colleagues, such as 

How can the best insights of mainstream and critical psychiatry be brought together? 

Awais Aftab, MD

Articles

A People’s History of Depression: Jonathan Sadowsky, PhD

A Psychiatrist Weaving Conceptual and Empirical Work

There and Back Again: Joseph Pierre, MD

Integration, Common Principles, and the Future of Psychotherapy: Marvin R. Goldfried, PhD

The Bureaucratic Takeover of American Psychiatry: George Dawson, MD, DFAPA

Sense-Making and the Enactive Turn in Psychiatry: Sanneke de Haan, PhD

3E Approach to Psychopathology: Kristopher Nielsen, PhD

Phenomenology, Power, Polarization, and the Discourse on Psychosis: Nev Jones, PhD

Managing Pandemic Anxiety

Psychiatry’s Intellectual Crisis: Giovanni Fava, MD