Nassir Ghaemi, MD


June 01, 2012

But DSM-IV Has Limitations, Too

Also at this year's APA meeting, Steven Hyman, a psychiatrist and neurobiology researcher who is former head of the National Institute of Mental Health, gave a plenary lecture on DSM-5 that was refreshingly honest in its appreciation of the limitations that the DSM-IV has placed on research. Rewinding to DSM-III, from the 1980s, he made the point that although that edition was a major advance, it is now out of date, and that DSM-IV, which merely continued the basic DSM-III structure, needs major changes. "The DSM-III was a brilliant document that could not have foreseen the science. It's time to move on scientifically," said Hyman.

Hyman noted that DSM-III actually hinders science. Researchers have difficulty getting funding from the National Institutes of Health or publishing papers that go outside DSM criteria: "For example, it was very hard to get a grant to test the hypothesis that maybe the apparent comorbidity of multiple anxiety disorder and mood disorders was just that there was a single underlying process or single disorder that got expressed with different symptom complexes in different times in life."

There was a name for that condition -- neurotic depression -- and Sir Martin Roth, the great British psychopathologist, warned repeatedly in the 1970s and 1980s that it would be a mistake for DSM-III to remove it. DSM-III made that mistake, and the field has since acted like it would be a sin to study the matter any further.

There are many examples of this ilk in DSM-III and DSM-IV. Some who are upset with proposed changes in DSM-5 are diagnostic conservatives who seem to think that all our questions were answered in 1980 and 1994.

Dr. Hyman has been influential in designing the new Research Domain Criteria (RDOC), an attempt to create a DSM for research that begins with biological, rather than clinical, terms. I agree with the need for a DSM for research, but I don't think our biological knowledge is advanced enough yet -- despite all the advances that have been made -- to build a diagnostic system from them, even for research purposes.

I think we should have a new DSM just for research: a system of Research Diagnostic Criteria (RDC), like what was created in the 1970s that led to DSM-III to begin with. I've started that process with my colleagues in the world of bipolar disorder research. We will publish a new RDC for bipolar disorder within the coming year -- before DSM-5, I hope. If we do so, I hope that colleagues in other specialties in psychiatry will produce similar RDCs.

With these new publications, psychiatry may then be in a position for real advance. We will then have 3 nosologies, all complementary to each other and able to improve the others:

  1. DSM-5: a nosology based on a mix of research, economic concerns, social preferences, and professional consensus that is used for basic practice, insurance reimbursement, and short-term consensus.

  2. RDOC: a nosology based solely on biological research that is used for research.

  3. RDC: a nosology based solely on clinical research that is used for research.

In summary, DSM-5 is on its way, and May 2013 is as good a date as any for its publication. In some places, it will be a much-needed advance over the now-outdated DSM-IV. But in other places, it keeps old categories that are not as well proven as they should be, and it even adds a few new categories that are mainly based on professional, economic, and social concerns rather than on sufficient scientific evidence.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.