The DSM: How Psychiatrists Redefine 'Disordered'

The attempt to catalog all the ways that a person can go crazy dates at least to 1840, when the Census included a question on "idiocy/insanity." From those two simple categories, there are now more than 300 separate disorders; they are listed in a 943-page book called the Diagnostic and Statistical Manual of Mental Disorders, or DSM for short. The book is important because doctors, insurers and researchers all over the world use it as a reference, a dictionary of everything humanity considers to be mentally unbalanced.


Here is the first comprehensive look at what might go into the book's latest version, the DSM-5. Currently, the DSM is disjointed and disorganized — at times well researched and at times anachronistic. The present version, the DSM-IV-TR (the TR stands for "text revision"), was published in 2000. It begins with "mild mental retardation" moves on to common illnesses like depression and odd ones like dyspareunia (painful sexual intercourse not due to a medical condition) and ends with the vague "personality disorder not otherwise specified." The rhyme and reason behind the DSM have always been murky; the book, like our brains, is a huge, complicated beast.


The American Psychiatric Association (APA), which publishes the DSM, has long wanted the fifth version to be a more rational, understandable document, but that's not proving to be easy. Publication has been delayed at least twice, and the association now doesn't expect to produce DSM-5 until 2013, 14 years after research on it began. One reason is that there are so many stakeholders: patients, shrinks, HMOs, academics. Patients want their illnesses covered; shrinks need to get paid, and academics want definitions to be consistent with research — research that is itself uneven. Sometimes, DSM changes can be made on the basis of long-term, peer-reviewed studies. But other times, such gold-standard research data is lacking, and changes must be made on the basis of consensus among clinicians. The process is fraught and confusing, even for those in the middle of it.


Still, the launch of dsm5.org — where suggested changes to the DSM were posted Feb. 10 for public comment — is a major step. Here are five ways the APA is proposing to address major criticisms of older versions of the book:


1. Contain the definition of a mental illness within sensible borders. 
A major problem with earlier versions was mission creep: In 1980, the APA published DSM-III, which radically expanded what clinicians could define as disordered. One example: depression. The pre-1980 definition had described "depressive neurosis" as "an excessive reaction of depression due to an internal conflict or to an identifiable event such as the loss of a love object." The much longer 1980 definition (which carried on into DSM-IV and DSM-IV-TR, with slight modifications) omitted the requirement that symptoms be "excessive" in proportion to cause. In fact, the revised manual said nothing about causes and listed symptoms instead.


To be diagnosed with major depressive disorder today, you need have only five symptoms for two weeks, which can include such common problems as depressed mood, weight gain, insomnia, fatigue and indecisiveness. The current DSM does make an exception for bereavement: if you recently lost a loved one, such symptoms are not considered disordered. But the manual doesn't make exceptions for other things that make us sad — divorce, financial stress, a life-threatening illness.


The proposed revisions would change that and once again take into account severity of symptoms. The new definition of all mental disorders would include the proviso that they "must not be merely an expectable response to common stressors and losses."


2. Define mental disorders along a continuum rather than as binary possibilities. 
When he spoke at a New York City DSM conference last year, Harvard provost Dr. Steven Hyman, a former director of the National Institute of Mental Health, argued that most mental disorders cannot be seen as discrete all-or-nothing illnesses like leukemia (which you either have or don't). Rather, he said, they should be seen as "continuous with normal," less like leukemia and more like hypertension. Hyman seems to have won the battle here — in particular, social-interaction disorders like autism and Asperger's will now be defined along a single spectrum (autism spectrum disorders), rather than as separate conditions. The proposed change has brought controversy: many high-functioning people with Asperger's disorder would rather not see themselves in the same category as those whose autism is so severe that they cannot dress themselves.


3. Address the problem of including certain minor addiction disorders (caffeine intoxication) but excluding others (compulsive gambling). 
These are relatively infrequent diagnoses, but they seem highly capricious. Isn't compulsive gambling a sign of a bigger problem? Isn't caffeine intoxication usually an accident? That's one reason the whole category of "substance-related disorders" has chipped away at the authority of the DSM. The new DSM would rationalize the system. There are no plans to change the diagnostic criteria of "caffeine intoxication" (essentially, drinking so much coffee or Red Bull that you go nuts, at least temporarily), but the APA is considering whether "non-substance addictions" like compulsive gambling, shopping and eating are related to traditional substance abuse — and, if so, how. Also, it has proposed re-titling the category of substance-related disorders to "Addiction and Related Disorders." No decisions have been made, but this research process is promising and long overdue.


4. Overhaul the strange grouping of personality disorders. 
Currently, personality disorders include everything from the debilitating, often deadly illness known as borderline personality disorder to the dated, rather sexist "illness" known as histrionic personality disorder, a symptom of which is that the sufferer "consistently uses physical appearance to draw attention to self." Who doesn't do that?


In the DSM-5, histrionic personality disorder would be eliminated, and personality disorders would be reduced to five key types: antisocial, avoidant, borderline, obsessive-compulsive, and schizotypal (a word for so eccentric that you don't get along with others).


5. Rethink the definitions of sexual and gender identity disorders. 
Today, heterosexual men can be diagnosed with a supposed disorder called "transvestic fetishism" if they meet only two criteria: they have sexual fantasies about cross-dressing, and those fantasies cause "impairment in social, occupational, or other important areas." What's more, the DSM considers aversion to sex a sex disorder, even though the condition has less to do with low sex drive than outsized feelings of fear and avoidance — more like a phobia.


The DSM-5 proposes to update this category by including "hypersexual disorder." Although the name sounds like something Han Solo might have had, the proposed criteria make sense: sexual fantasies take up so much time that they become repetitive, debilitating and harmful to normal functioning. Also, "sexual avoidance disorder" would be dropped and "transvestic fetishism" would become "transvestic disorder," although the diagnostic criteria themselves would not change: the DSM still seems to have a problem with cross-dressing.


Overall, the DSM-5 is shaping up to be a much better reference than its predecessor. There will be months of negotiations — anyone can register at dsm5.org to comment, and consumer groups, day-to-day therapists, research psychologists and many others will have a say. But give the APA one thing: it seems to acknowledge, finally, that it is not the sole arbiter of what makes a person crazy.