Defining mental disorders, DSM-5 style

By Jonathan Raskin

What is a mental disorder? This is a question the American Psychiatric Association contemplated while preparing the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). DSM-5 factors include:

  1. A behavioral or psychological syndrome or pattern that occurs in an individual
  2. Reflects an underlying psychobiological dysfunction
  3. The consequences of which are clinically significant distress (e.g., a painful symptom) or disability (i.e., impairment in one or more important areas of functioning)
  4. Must not be merely an expected response to common stressors and losses (ex. the loss of a loved one) or a culturally sanctioned response to a particular event (ex. trance states in religious rituals)
  5. Primarily a result of social deviance or conflicts with society

To many people, this sounds pretty good from a first read. But how readily does this definition allow us to truly distinguish what is or isn’t a disorder? What are the underlying assumptions that this definition implies? Let’s consider the first two criteria, which when combined hold that a mental disorder is “a behavioral or psychological syndrome or pattern that occurs in an individual that reflects an underlying psychobiological dysfunction.”

The DSM has long claimed to be “atheoretical” about the causes of mental disorders. This makes sense if you think about all the constituencies the DSM has to please. Mental health professionals have many different (often conflicting) ideas about what causes people to experience problems in living. They also often disagree on how best to alleviate such problems. Should they rely on medication, psychoanalysis, behavioral conditioning, rational argument, extended family discussions, sociopolitical consciousness raising, or any number of other possible intervention strategies to help those they serve? In order to avoid alienating any particular constituency of mental health professionals, the DSM has strategically adopted an atheoretical stance on the etiology (causes) of mental disorders. At the same time, the DSM hews to a medical model by organizing mental disorders into discrete categories, just as medicine does with diseases. That is, DSM is a medical-model manual that is nonetheless atheoretical about the causes of the disorders it catalogs. This may be confusing but important to keep in mind.

Trying to be atheoretical about causes makes defining disorders difficult. This is readily apparent in the DSM-5’s proposed definition, which says that a mental disorder is “a behavioral or psychological syndrome or pattern that occurs in an individual.” What does this mean? To start with, it means that disorders are inside people. They are things people “have.” Can I have a psychological syndrome or pattern inside me? Even though we can’t observe it directly, the idea that our “psychology” is inside us seems commonsensical to most people. But what about our behavior? Is it also inside us? No. Behavior is something people do. It is observable, not inside us. Thus, to say that behavior is something that occurs in an individual doesn’t quite hold together theoretically. At the very least, it might irritate died-in-the-wool behaviorists, who discourage us from relying on abstract mental concepts to explain behavior.

However, things get hairy when we shift to the second definitional criterion, which holds that these “behavioral or psychological syndromes or patterns” reflect an “underlying psychobiological dysfunction.” First, this marks a clear shift away from the aforementioned “atheoretical” position that has been a hallmark of the DSM for the past 30-plus years. Second, in claiming that mental disorders are psychobiological, the DSM’s reach clearly exceeds its grasp. Let me take these two points one at a time.

The first point concerns the move away from an atheoretical stance on the causes of disorders. As already noted, the DSM has long sought to keep the peace among professionals of varying theoretical orientations by remaining mute when it comes to specifying the causes of mental disorders. It has prided itself on sticking to descriptions of disorders and avoiding speculation about causes. Discovering etiology, according to past DSMs, is best left to researchers. Given this longstanding commitment to an atheoretical position on etiology, the prospect of changing the definition of mental disorders to one that explicitly defines disorders as “psychobiological dysfunctions” is big news indeed because doing so is overtly theoretical. Psychobiology conceptualizes human psychology as something that can be reduced to and explained exclusively in biological terms. As such, the proposed new definition of mental disorders contends that all DSM disorders have biological causes. The goal of being atheoretical goes out the window if DSM explicitly defines mental disorders as biological.

The second point is that by moving so explicitly in a biological direction, DSM’s reach exceeds its grasp. Moving in an openly biological direction might make sense, but only if the DSM restricts itself to disorders where the underlying biological causes are known. In other words, if the DSM plans to shift from an atheoretical to a psychobiological stance, it should probably have pretty clear evidence that the disorders it contains can be diagnosed using biological tests or markers. Yet, this is not the case. DSM-5 will carry on the longtime DSM tradition of diagnosing mental disorders using behavioral criteria alone. These criteria will continue to take the form of lists of behaviors. Biological indicators will not be used to diagnose mental disorders because we simply don’t have the ability to do that at the moment. This is why DSM’s reach exceeds its grasp. It wants to define mental disorders as having underlying psychobiological dysfunctions, but very few of the disorders it contains can be diagnosed biologically. DSM makes diagnoses based on what people do, not tests of biological functioning.

Even in cases where people take drugs for a disorder and feel better, we cannot say with certainty that an underlying biological cause is being remedied. Improved mood notwithstanding, whether we have cured something remains the subject of speculation because, when it comes down to it, we just don’t know enough to say with certainty what the underlying biological cause of any given DSM disorder actually is in the first place. We must keep in mind that changing a person’s behavior (whether through drugs or other methods) doesn’t necessarily mean we have corrected a psychobiological dysfunction. One might smoke a joint and feel more relaxed, but this doesn’t mean marijuana cures anxiety disorders or that the person was suffering from a marijuana deficiency. It just means that drugs can alter experience. To feel confident that a drug cures an underlying disorder, we need to know what the biological etiology of the disorder is and how the drug “fixes” that etiology. Even if we believe drugs can be helpful, their effectiveness doesn’t necessarily resolve the etiological uncertainty about what mental disorders are.

All the confusion about defining “mental disorder” makes sense when one thinks about the term more carefully. Mental has to do with mind, and disorder is often (though not always) a euphemism for disease. Therefore, another way to think about mental disorder is as some kind of “mind disease.” Of course, as Thomas Szasz pointed out more than 50 years ago, minds—unlike brains—are not biological, and in a literal sense, cannot be afflicted by diseases. Most of the disorders listed in DSM, therefore, fall into two likely categories: (a) everyday problems in living that warrant professional attention but are not diseases, and (b) suspected brain diseases whose etiologies may one day be uncovered but currently remain unknown.

When it comes to mental disorder, the psychobiological definition being considered for DSM-5 is not only overtly theoretical but also poorly drawn. The things we presently call mental disorders have not been convincingly explained in psychobiological terms, even if the authors of DSM-5 would like to insist otherwise.


Are you a Saybrook University alumna or faculty member who is interested in submitting a blog post for potential publication on Unbound? You may do so by submitting your finished piece or pitch on the right-hand side panel of UNBOUND pages.