For the past four years, I have been deconstructing my views of my profession. My focus has been primarily on two areas: the efficacy and safety of the drugs I prescribe and so-called “alternative” approaches (in this I include many things such as Open Dialogue, Hearing Voices groups, and Intentional Peer Support to name a few). I have shared much of this in the blogs I wrote during this time. I am also interested in how we can improve and reform the public mental health system since this is not only where I work but where most people seek services and help. I have wondered where – if anywhere – psychiatrists fit in to a reformed system. I choose to post on some websites that are filled with deep criticism of my profession. I take it in and ponder in an Hamletian way if it is ethical to continue to practice given the serious problems in my field. I have come to an idea about how we can use our medical expertise to contribute something meaningful while acknowledging the value of non-medical thinking and remaining honest about the profound limitations of our knowledge. What follows is an attempt to articulate these ideas.
This discussion is divided into three parts. It begins with a review of Joanna Moncrieff’s ideas about a drug-centered vs. disease-centered approach to psychoactive compounds. In part II, I discuss these in the context of two classes of drugs: the neuroleptics and the psychostimulants. In the final section, I discuss need-adapted approaches and conclude with the proposition that the integration of a drug-centered pharmacology with a need-adapted approach might reclaim psychiatry – a slow psychiatry – as a more humble, humane, and honest endeavor.
Reflections on the Medicalization of Mental Health Care
Psychiatry in 2015 is the field of medicine that specializes in prescribing psychoactive drugs to people who we have identified as having psychiatric disorders. This is not to say that is all we do, and practice patterns vary. But at least where I work — within the community mental health care system in the U.S. — psychiatrists constitute a small but expensive proportion of the overall work force and one of our core functions consists of categorizing symptom clusters as specific disorders and recommending drugs to treat those putative conditions.
Although this may have begun as a well-intentioned effort to reduce suffering, it has run into a variety of problems. There has been an ever-expanding tendency to characterize a vast array of human behavior as medical disorders. It is this definition of problems-as-medical-conditions that brings them under the purview of the medical specialty of psychiatry. (Hugh Middleton addresses this in his recent post.) While characterizing a problem in medical terms is not synonymous with stating the problem requires a drug treatment, there has clearly been an increasing use of psychoactive drugs to treat an increasing array of problems in the past few decades. The beginning part of this post focuses on the drugs, but the latter part will address why a medical approach to understanding many kinds of problems – as a general principle — is often unhelpful, regardless of whether or not drugs are prescribed.
Regardless of one’s particular inclination towards using drugs to relieve mental suffering, it seems likely that people have and will seek out drugs to alter mental state and mood. I have argued, and continue to believe, that it is a good idea to have a field of medicine that specializes in understanding how to best use these drugs. But I think the current model of psychiatry has failed us.
The Drug-Centered Approach
Joanna Moncrieff has made the distinction between a drug-centered and disease-centered approach1 to thinking about psychoactive substances. In a disease-centered approach, one assumes that drugs correct abnormal brain chemistry that is responsible for the condition being treated, they are considered medical treatments, and the beneficial effects of drugs are derived from their effect on a presumed disease process. A drug-centered approach posits that the drugs create a state different from the one that existed before the drug was administered but do not correct a specific abnormality or defect; drugs are considered psychoactive substances that alter the expression of psychiatric problems through the superimposition of drug-induced effects.
In a disease-centered approach, one thinks about main effects — those directed at a specific disease state and side effects — versus unwanted effects that are incidental annoyances. In a drug-centered approach, one understands that most drugs have broad effects that may be useful in some contexts. In a disease-centered approach, one is more likely to consider long-term poor outcomes as a consequence of the natural course of an underlying disease state, and when symptoms recur after drugs are stopped, to consider that a recurrence of illness. In a drug-centered approach, one is more likely to recognize that drugs can have long-term impacts, and when the drugs are stopped, one understands that some form of withdrawal is expected.
Many psychoactive drugs were first introduced in the 1950s and 1960s and at that time they were thought about and studied in a drug-centered way. Drugs were classified broadly as tranquilizers and stimulants. Drugs were marketed for a wide variety of indications. Combination drugs were popular. In that era, stimulants, for example, were marketed to treat depression as well as over-eating. A popular drug called Dexamyl contained a barbiturate and a stimulant. It was marketed to treat depression, anxiety, overweight, and fatigue among housewives.
This began to change in the 1960s. In 1962 in the US, an amendment was passed to the Food and Drug Act in response to the discovery that the drug thalidomide caused severe birth defects. Going forward, drug makers were required to demonstrate that a drug approved for the market was not only safe but also effective in treating a specific condition. During that same era, recreational drug use became more common in the US and, as concern about this grew, laws were passed in the 1970s to restrict the prescription use of stimulants. At the same time, psychiatrists had a need to distinguish their concept of drugs as medicine from recreational users’ concept of drugs as, well, recreation.
Coincident with these forces were struggles within psychiatry. The psychoanalysts led many major academic psychiatry departments from the 1950s through the 1970s, but a growing group of psychiatrists wished to restore psychiatry to what they considered a more scientific and medical approach. They included those from the Washington University School of Medicine who were the leaders in restoring fidelity to psychiatric diagnosis. Known as the neo-Kraepelinians, they were the driving force behind the emergence of the modern diagnostic manual, the DSM-III, which was published in 1980.
The neo-Kraepelinians wanted an approach to diagnosis that would allow researchers to use modern methodologies to finally figure out the underlying etiologies and pathophysiologies of psychiatric disorders. This required a consistent and systematic approach to classification so that people who were given the label of schizophrenia, for example, were more likely to bear relevant similarities to one another, despite where and by whom their diagnoses were assigned.
The 1962 Food and Drug Act2 made this system critical for the ongoing drug development that proliferated over the next few decades because diagnostic categories – disease targets — were required by the law. The publication of the DSM III, which addressed the aspirations of both the neo-Kraepelinians and the drug companies, was the final stage in the transformation of psychiatry’s approach to understanding psychoactive drugs from a drug-centered to a disease-centered one.
But to this day, we have a big problem. Despite the promise heralded by the Washington University group and their many adherents, the etiologies of these disorders remain murky at best; there is no clear pathophysiology for any of the disorders in the DSM.
There is now overwhelming evidence of the damage caused by the diseased-centered approach to psychopharmacotherapy. In many ways, one can consider Anatomy of an Epidemic a treatise on these harms. We promoted these drugs based on their short-term effects. When the drugs are discontinued, we have come to consider all consequent problems to be re-emergences of putative underlying disease processes. This has led to the recommendation that many of the people who began taking these drugs remain on them indefinitely. We failed to reckon fully with the consequences of drug discontinuation. Some of the emergent problems people experienced on these drugs were considered evidence of the presence of an additional disorder and this led to an increased use of polypharmacy.
In this section, I apply the construct of the drug-centered versus disease-centered approach to psychopharmacotherapy to two classes of drugs: the neuroleptics and the psychostimulants.
I chose the neuroleptics in part because reconsidering their use has been a focus of mine. However, I have also found that when I discuss this with my colleagues, this is the class of compounds that they have the hardest time thinking of in a drug-centered way. Most psychiatrists begin their careers on inpatient units and we have all observed people who come in to the hospital extremely preoccupied by delusional thoughts and voices who, after taking these drugs for a few days, are no longer hearing voices or are not as bothered by delusional beliefs. Thus, the drugs appear to have specific anti-psychotic effects.
How can we reconcile these observations with a drug-centered approach?
Laborit, the French physician who first suggested that chlorpromazine might be of benefit to the people housed in France’s mental hospitals, noted that the drugs induced indifference. In the 2009 edition of the American Psychiatric Publishing Textbook of Psychopharmacology, edited by Alan Schatzberg and Charles Nemeroff, the authors write that in normal volunteers, “neuroleptics induce feelings of dysphoria, paralysis of volition, and fatigue” (emphasis mine). This indifference might be helpful at times when a person is psychotic and deeply troubled by intrusive and disturbing thoughts and voices. However, long-term use of these drugs would be expected to be associated with apathy and impaired function.
This is a matter of perspective similar to the parlor trick of looking at the picture in which some see an old woman’s face and others see a young one. Both images are there but our context might lead us to see one image more readily than another. So if we are trained in a disease-centered model and we note that our patients seem less bothered by voices, we might believe that the drug targeted some pathophysiology specific to this experience. We can believe that any apathy the person experiences is the “untreated” part of this condition that we have been told includes apathy as a core symptom, and when the person stops the drug and the voices recur, we can easily believe that to be a recurrence of the disease that that the drug had once treated effectively.
But if we take the perspective of a drug-centered approach, we understand that we are exploiting the state of indifference induced by the drug to provide what may be temporary relief for a person so troubled by harassing voices. We understand that we need to be careful about causing apathy with the drug and that when the drug is stopped a person might experience effects of its withdrawal that could include the same problems we were intending to suppress.
It is only in the context of a disease-centered approach that recent data suggesting impaired functional outcomes in those who remain on these drugs long-term is surprising. With a drug-centered approach, this outcome is predictable, or at least more understandable.
The stimulant story
I suspect that it is easier for psychiatrists to think about psychostimulants in a drug-centered way. They were prescribed in this manner for decades. They were given to overweight people to help them lose weight, they were given to depressed people to help improve their moods, and, as noted above, they were prescribed to housewives – sometimes in the form of Dexamyl in which they were combined with a barbiturate – to give a general boost to help them deal with the drudgery of daily life. As early as the 1930’s, they were reported to help calm children.
But they were also widely abused and with the beginning of the “drug wars” in the 1970’s, the use of these drugs in the US was restricted. By that time, many new antidepressants were available and stimulants were used primarily in children who were diagnosed with ADHD. So it was legislation and marketing that led to the changing indications and not any alteration in the evidence of their efficacy. A prominent textbook from the 1990’s3 still talked about their efficacy in the treatment of depression.
Over the subsequent decade, their indications began once again to expand. First of all, there was a broadening of the definition of ADHD. Whereas the DSM IV required “clinically significant impairment” before age 7, the DSM 5 the criteria only require that “the symptoms interfere with, or reduce the quality of, social, academic, or occupational functioning” and that symptoms be present before age 12. In addition, adults only need to have 5 rather than the 6 symptoms required of children to meet the current diagnostic criteria.
The stimulants are old drugs. Amphetamines were first synthesized in the 1880’s and methylphenidate in the 1940’s. But in 2007, a new drug gained FDA approval – lisdexamfetamine, marketed as Vyvanse. This is not exactly a new drug but a new preparation. These drugs are popular and widely abused; lisdexamfetamine is just harder to crush and snort or inject than other psychostimulants on the market. But what happened over the next few years was an expansion in the indications for stimulants, specifically lisdexamfetamine:
How does a drug-centered approach help us to understand this?
In a disease-centered approach, these drugs are treatments for specific disorders: ADHD, BED, MDD. These drugs are thought to relieve the suffering of people impaired by these disorders. But this implication of specificity is more of a pseudospecificity of action; after all, people who get some relief from these drugs are probably doing so via the same mechanism of action regardless of diagnosis. It is not likely that they act in one way in people who are depressed and another in those who overeat. It is just that some of the effects are considered of particular benefit to some people. Put another way, an overeater will experience an alteration of mood and a depressed person may eat less while on these drugs, but those are not the reasons why such people derive benefits from them. With a drug-centered approach it would be acknowledged that the drugs affect cognitive function in everyone, they suppress appetite in everyone, and they affect mood in everyone. It is the requirements of the disease-centered approach – the illusion of specificity – that results in this circuitous route we have traversed over the past 60 years from broad to narrow and more recently to ever broadening indications. It is also the impacts of the disease-centered approach that results in our conceptualization of many human experiences as medical disorders.
There are more serious and dangerous impacts of the disease-centered approach. In our current system, in order to get effective treatments to the market, we only need to test drugs over brief periods to evaluate their impacts on target syndromes. As we expand the marketing to categories that may come close to capturing many for what are conceptualized as chronic conditions, we might take a moment to think about what it means for people to be on these drugs for years on end or what it means to stop them. With a drug-centered paradigm, these concerns are more likely to come to our attention.
In addition, in a disease-centered approach, if psychosis were to emerge during the course of pharmacotherapy with a psychostimulant for, let’s say, ADHD, one might conceptualize that as the emergence of a psychotic condition and justify adding another drug to target this newly diagnosed condition. In a disease-centered paradigm, one might be inclined to diagnose ADHD in a person being treated with sedating drugs as another “co-morbid” condition. In a drug-centered approach, one is more likely first to stop the stimulant when psychosis emerges and to reduce the sedating drugs when inattention is observed.
Part III. Need-Adapted Approaches
In the first two sections, I drew the distinction between disease-centered and drug-centered approaches to psychopharmacotherapy. I strongly favor the latter as more honest and cautious. But a drug does not just end up in a person’s body. There is a process – a meeting, a consultation, a discussion of effects – that has to occur before a person swallows a pill, and this is where the need-adapted approach offers a path for psychiatrists to remain humble, honest, and humane.
The need-adapted treatment model was developed in Finland in 1970s and 1980s. As described by Yrjo Alanen in his book, Schizophrenia, Its Origins and Need-Adapted Treatment4, multiple theoretical frameworks were considered helpful but not definitive. This led the clinicians to approach a person not with the goal of applying a fixed theoretical framework but with an openness to using all models as needed. When they brought in the family, they found that this alone was often enough to resolve the crisis. This was the forerunner to Open Dialogue that evolved in Tornio, Finland. Over time, a broader array of approaches have evolved in Scandinavia and northern Europe. Tom Anderson and colleagues were simultaneously working on reflecting therapies. Carina Håkansson started the Family Care Foundation. Shared among them is a deep appreciation of the value and importance of social networks in helping to develop understandings of human problems. Diagnosis – and the diagnostic process – is held lightly in these models. The uncertainty many of us find inherent in this work is acknowledged. “Treatment” proceeds from individual/network needs and it remains flexible. The psychotherapeutic attitude is considered at least as important as the technical aspects of the treatment. In keeping with the value placed upon relationships, there is also a recognition of the value of psychological continuity, i.e., to the extent possible the team involved remains constant.
This contrasts with the more traditional medical approach in which there is a focus on the individual who is presumed to be experiencing some sort of psychopathology that the experts will characterize through the evaluative process. Families are a source of further history and support but often are not considered intrinsic to the recovery process. The treatments that are offered are based on this evaluative process whereby a diagnosis is made and treatment recommendations are based on that diagnosis. Treatments are considered in a more technical way and it is often assumed that they work independent of the relationship.
What has been interesting to me is the overlap in values that have been emphasized in other so-called “alternative” approaches. In the past 25 years, the recovery movement has grown in the US. If one goes to SAMSHA, one can find a set of recovery principles that include:
When I read these, I see an important overlap between the values of need-adapted approaches and the recovery movement. And part of that is embodied in the construction of the sentence – there is an emphasis on values — how one is with a person and his network. I find other overlapping values when I talk to those who are connected to the work of the Hearing Voices Network. It was these resonating themes that I noted in my post about the ISPS conference last spring.
Integration of Drug-centered and Need-adapted Approaches
This blog started with an explanation of a drug-centered paradigm for thinking about psychoactive drugs, followed by a discussion of need-adapted models. The connection may not seem apparent so let me try to make it more explicit. A drug-centered approach acknowledges that we do not fully understand the causes of peoples’ troubles. We understand more about drug action although our knowledge is certainly incomplete on that subject as well. A needs-adapted approach provides a framework in which we can talk about these drugs, acknowledge the many uncertainties, and support a person in deciding whether to take them. It acknowledges that this is likely to be an ongoing process that may be revisited time and again. It allows for the person’s own values and understanding of the problem to be both recognized and respected and it offers the space for many views to be heard. It allows that what psychiatrists label “symptoms” might not be the most important focus for a person. It gives space for a person to identify what is most important to him and places the discussion of drug treatment or any treatment for that matter within that context. It allows for a physician to be on the team but not necessarily as the leader. There may be discussion of drugs, the brain, what the physician has observed in others in similar situations, and whether there are relevant studies, but it does not require that the physician is the only expert or authority. If there is discussion of brain function and even dysfunction, this in no way precludes a person finding meaning in the experience. It allows for a frank discussion of what psychiatric diagnosis is (a classification system) and is not (a deep understanding of the nature of the problem). And it accepts that all of this occurs in the context of a relationship – usually multiple relationships — that will exert their influences on this process.
I have a longstanding interest in the problems and perils of industrial agriculture and I have often thought there were similarities between that field and the topic under discussion here. Industrial agriculture has valued production and profit above all else. Along the way, we have damaged our environment, our health and our culture. The Slow Food movement arose as a grass roots attempt to recapture our food and the culture attached to food. A Slow Medicine movement has now emerged and David Healy has written about this in an elegant blog. In an earlier blog, I suggested that psychiatrists have a relatively small part to play in the lives of people who struggle to navigate in this world. I think some colleagues who are generally sympathetic to my views were put off. They thought I went too far in reducing the scope of psychiatry. Was I supporting something along the lines of the 15-minute “med check”?
The simple answer to that question is no. In fact, what I think we need is Slow Psychiatry.
While I contend that psychiatry – medicine – can step aside with most people who experience emotional distress, when physicians are involved, these encounters will take time. To reduce emotional distress into small parcels of time and then parse the variety of human experience into rapidly determined and poorly validated diagnoses makes no sense. In addition, it is likely to foster a climate in which we continue to do harm.
I look forward to reading and reflecting on your thoughts and critiques.
1. Moncrieff, Joanna. The Bitterest Pills. London: Palgrave, 2013.
2. Healy, David, The Creation of Psychopharmacology. Cambridge, MA: Harvard University Press, 2002.
Note: Much of the history of drug development is from Dr. Healy’s book.
3. Schatzberg, Alan and Nemeroff, Charles, eds. Textbook of Psychopharmacology. Washington, DC: The American Psychiatric Publishing, 2009.
4. Alanen, Yrjo O. Schizophrenia, Its Origins and Need-Adapted Treatment. London: Karnac, 1997.
Named to “Best Doctors in America,” Dr. Sandra Steingard is Medical Director at HowardCenter, a community mental health center where she has worked for the past 17 years. She is also clinical Associate Professor of Psychiatry at the College of Medicine of the University of Vermont. For more than 20 years, her clinical practice has primarily included patients who have experienced psychotic states. Dr. Steingard is Board Secretary for the Foundation for Excellence in Mental Health Care.