Thoughts on the DSM-V

I’ve been reading about the debate over the upcoming DSM-V, the new version of psychiatry’s reference manual, and wanted to express my own opinions. I think that a utilitarian criterion should be used to evaluate not only the DSM, but psychiatry as a whole, including the prescribing (and overprescribing) of medications. By this I mean: does this diagnostic manual or treatment properly diagnose and treat people with psychiatric disorders? Does it falsely diagnose and improperly treat people by diagnosing them with disorders that they don’t have?

I’ll make an assertion that people with full-blown (out-of-control) schizophrenia, bipolar disorder, and OCD are better off medicated. To be medicated, they must first be diagnosed, so an instrument that aids in correct diagnosis is useful. The first two editions of the DSM came out when psychiatry was still heavily influenced by psychoanalytic theory. As a consequence, they had vague, unreliable categories (e.g. neurotic vs. psychotic), and contained spurious speculation on etiology. Since psychoanalytic treatment never helped patients with serious disorders, these early editions of the DSM were pretty much useless.

Robert Spitzer led the development effort of the major revision of the DSM, known as the DSM-III, which came out in 1980. (Interestingly, Spitzer is a critic of the upcoming DSM-V.) The DSM-III largely abandoned the psychoanalytic orientation of its predecessors. It abandoned any attempt to classify by etiology. It created the symptom-based, multiaxial categorical system which, with some changes and extensions, was incorporated into the DSM-IV in 1994. The DSM-III helped psychiatrists make more reliable and valid diagnoses than past editions. It must be remembered, however, that without any knowledge of etiology, and without any valid diagnostic tests, there would always be less reliability and validity than with physical disorders.

The symptom-based model of the DSM-III and DSM-IV sets up fairly steep criteria by which to diagnose someone. For example, click here for the DSM-IV criteria for a major depressive episode. Of course, if you go to a PCP and tell him that you’re depressed, in less than a minute he’ll write you a prescription for an antidepressant, but the DSM states what criteria this PCP should be using. It also states the criteria by which patients need to be evaluated to be diagnosed in research studies. This is important because in order for drugs to be approved, they must first go through clinical trials on patients diagnosed with the disorder that these drugs are supposed to be treating. The fact that once they’re approved, they can be prescribed off-label for other disorders is a separate issue that I won’t take up here.

Enough about the history. Let’s ask the question: with the current DSM-IV, are people who should be diagnosed and medicated actually diagnosed and medicated? I think the answer is yes. If Jim has bipolar disorder, and has some means (insurance or self-pay) to see a doctor and to purchase drugs, he’ll be diagnosed, and get the drugs he needs. It’s possible that Jim doesn’t want to see a doctor or take drugs, but that’s a separate issue. No diagnostic manual can “treat” that problem (assuming that it is a problem).

The scary thing about some of the proposed revisions to the DSM-V is that they will expand the number of diagnosable disorders. This will in turn expand the use of drugs to people who don’t need them. I don’t think that this is a good thing. There are already too many people taking drugs who don’t fit the diagnostic criteria of the disorder for which the drugs were originally approved. An example of this is the unconscionable use of drugs approved for adults in children, sometimes very young children.

What psychiatry desperately needs is a more scientific basis. It needs better diagnostic tests, and a better understanding of etiology. These will in turn lead to better treatments. In the 15 years since the DSM-IV came out, while there has been some great research going on, there hasn’t been a major breakthrough. So the DSM-V is not going to help things, and will probably hurt things. Hopefully by the time the DSM-VI is published, there will be some breakthroughs. We can only hope.