First, some background. DSM-V stands for Diagnostic and Statistical Manual--Fifth Edition. Of course, in psychiatry, numbers can be misleading. DSM-V will actually be the seventh revision, as DSM-III was revised once (DSM-III-R) and DSM-IV was also revised (DSM-IV-TR. TR stands for "Text Revision.") It's published (and jealously guarded) by the American Psychiatric Association.
When I started out in the 1970's, DSM-II was the manual we used, and it was awful. Definitions were vague, and you could diagnose the same person with anything from depressive neurosis to schizophrenia without much difficulty. DSM-II was generally denounced as irrelevant and unnecessary by behaviorists and humanists. At the time, diagnosis had little to do with treatment, so it really was unnecessary. In hospitals, physicians would argue endlessly about a patient's diagnosis, with no change in treatment after they had differentiated what kind of schizophrenic they were dealing with.
DSM-III changed all that, as diagnoses were made on the basis of behavior. Specific symptoms were identified and you had identify a certain number of symptoms in order to diagnose depression. This was a great advance in two areas. First, research could be focused. If you wanted to research a treatment for depression, there was now a clear operational definition for depression. That means that we could compare different studies without comparing apples and oranges.
Second, DSM-III gave us the ability to communicate what we were treating to third parties. It was published when insurance companies were beginning to pay for psychotherapy. Unfortunately, many therapists didn't know an adjustment disorder from major depression. Consequently, managed care used DSM-III to refuse to pay for many treatments. We had to learn to say, "Here are the symptoms that show the client is depressed and here is the progress I am making on them. Once we learned to do that, managed care stopped refusing psychotherapy and we now generally get as much time to treat clients as we need.
DSM-IV has been a small improvement over DSM-III, but it's still controversial. Now DSM-V is on the horizon and it's generating more heat. A recent article in the New York Times picked up on the controversy. There are a number of points of controversy.
First, there is no real understanding of the causes of mental illness. We're still not even sure if there is such a thing as a mental illness. Psychiatry has been moving toward a neurobiological model of mental illness, but is still far away. If you ask a psychiatrist what causes depression, you'll hear one of two answers. You might hear, "We don't know," which is the honest answer. However, more likely, you will hear some mumbling about serotonin and biochemical imbalances. Those guys are just blowing smoke. Just ask them to define a healthy biochemical balance; their response will be entertaining. Unfortuantely, more than half of the members of the task force writing DSM-V have drug company connections, so you know they will remain committed to a neurobiological model.
Second, diagnoses are often points on a continuum. Consider what is now called Attention Deficit/Hyperactivity Disorder (ADHD). Our ability to pay attention falls along a continuum from fleeting to intense. When we diagnose someone with ADHD we are saying that their attention on the average, falls below some imaginary point on that continuum. That imaginary point is also arbitrary. Do we abandon children who fall just to the normal side of that point? How about children who fall mostly to the average side of that point, but occasionally their attention and concentration crash and burn?
Third, the inclusion of a diagnosis in the manual has always been unsystematic and has many social implications. Why is repetitive handwashing considered obsessive compulsive disorder, while repetitive shopping is not? Is binge eating a disorder, or does the person just need to grow up and get some self control? In DSM-II, homosexuality was considered a diagnosable mental illness. In DSM-III, homosexuality was removed as a diagnosis. There is a straight line from deciding homosexuality is not a mental illness to deciding gays should have the right to marry.
Fourth, how open should the process be? The APA has had the members of the DSM-V task force sign conficentiality agreements. Dr. Robert Spitzer, a member of the APA task force on DSM-V has raised concerns about it. (More about the controversy can be found here.) Given the social implications for some diagnoses, there's a lot at stake here. But, it's difficult to balance the need for open, scholarly discorse, with the risk of being personally attacked by one pressure group or another.
Who gets treatment is affected by diagnosis. When I diagnose a child with autism, that child becomes eligible for a wide range of services that a child with plain old "mental retardation," would not get. Interest groups in a lot of areas would like to see their own diagnoses included, so more services become available. A battle is raging over whether or not transsexuality should be included in DSM-V as a mental illness. Some would like it to be normalized; a person should be able to cross dress if they want to. However, others would like to have insurance cover sexual reassignment surgery. You can't cover a procedure if it isn't treating a diagnosis.
DSM-V will almost certainly move closer to a neurobiological model and away from a psychosocial model. That will lead to less emphasis on psychosocial treatment. I don't hold that against psychiatry; they treat mental illness biologically. However, psychotherapy still has much to offer. Psychotherapy is still the best (or at least a competitive) treatment for personality disorder, post-traumatic stress disorder, and some depressive and anxiety disorders. Generally, psychotherapy reduces the risk of relapse. Yet, if we structure our diagnoses around neurobiolgical models, psychotherapy will be marginalized. Already, family and marital therapies are extremely difficult to justify using DSM-IV. Somehow, diagnoses should also be there to allow for psychosocial treatment and to encourage more research in that area.
Despite these problems, we still need a diagnostic manual. Psychiatric diagnoses, for all their arbitrariness, give us a way of organizing research and communicating therapeutic information. At some point research will give us a handle on the nature of mental illness, and a clear diagnostic system will emerge from that. In the meantime, this is the best we've got, and none of the critics have given us a good alternative to the writing of a new manual.