NIMH Won't Follow Psychiatry 'Bible' Anymore

The fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5)—slated for release this month—has lost a major customer before even going to print. Thomas Insel, director of the National Institute of Mental Health (NIMH), declared last week on his blog that the institution will no longer use the manual to guide its research. Instead, NIMH is working on a long-term plan to develop new diagnostic criteria and treatments based on genetic, physiologic, and cognitive data rather than symptoms alone.

Insel's pronouncement is the most recent hit in a long barrage of criticism that has rained down upon the latest DSM revision process since it began over a decade ago. "While DSM has been described as a 'Bible' for the field," he wrote, "it is, at best, a dictionary, creating a set of labels and defining each." Although the manual's strength has been to standardize these labels, he wrote, "[t]he weakness is its lack of validity," and "[p]atients with mental disorders deserve better."

Although Insel's blog was reported as a "bombshell," and "potentially seismic," NIMH's decision to scrap the DSM criteria has been public for several years, says Bruce Cuthbert, director of NIMH's Division of Adult Translational Research and Treatment Development. In 2010, the agency began to steer researchers away from the traditional categories of DSM by posting new guidance for grant proposals in five broad areas. Rather than grouping disorders such as schizophrenia and depression by symptom, the new categories focus on basic neural circuits and cognitive functions, such as those for reward, arousal, and attachment.

Helena Kraemer, a biostatistician at Stanford University in Palo Alto, California, who was responsible for field trials of diagnostic categories proposed for DSM-5, says that Insel is right that the NIMH's new program, called Research Domain Criteria (RDoC) is "the direction we have to go." However, she says, "he's wrong in saying that DSM-5 is to be set aside." When it comes to validity, there now is no gold standard, she says. "The DSM is a series of successive approximations." Kraemer's vision is that future versions of the manual will not have to wait 10 to 15 years for revision, but incorporate new scientific data from RDoC as it emerges. She says that a meeting is scheduled in June to discuss the possibility of converting the DSM into an electronic document that could incorporate those changes. "Everybody I've talked to about it thinks that's a good idea."

Frank Farley, a psychologist at Temple University in Philadelphia, Pennsylvania, and former president of the American Psychological Association (APA), isn't convinced that the whole process doesn't need to start from scratch, however. The measures of agreement between experts for several of the disorders in the new DSM-5 were "terrible," he says. "What it suggests is that we need to go back to the drawing board." In 2011, Farley and colleagues circulated a petition for APA to submit the new revisions to independent review. Although 14,000 professionals and more than 50 organizations signed on, he says, "Nothing happened. We got a 'Thanks, but no thanks' letter back."

Both RDoC and DSM are necessary, says William Carpenter, a psychiatrist at the University of Maryland School of Medicine in Baltimore. Carpenter chairs the psychosis working group for the new DSM-5 manual and is one of three external advisers to RDoC. On a practical level, researchers and physicians need DSM to help characterize and treat patients in the field, he says. "If you don't, you just have 'Mental health I, II, III, IV and V.' "

On the other hand, Carpenter says, drug development for psychiatric disorders "has been stalemated for decades" due to our lack of understanding of the biological roots of psychiatric disease. "What I would hope for our field, is that clinically we get into the habit of deconstructing these syndromes into the specific pathologies that the patients have," such as hallucinations and impaired emotional processing. Once we understand the neural circuitry and neurobiology that cause such symptoms, he adds, "hopefully this will help drag drug companies into trying to make novel discoveries, instead of me-too drugs that they've lived off of for all these years."

Implementing RDoC will present some practical challenges, Carpenter acknowledges. "This does shift the paradigm." Rather than excluding all study subjects who do not fit a DSM diagnosis, such as major depression, for example, the new approach might include a range of participants with different diagnoses who all demonstrate anhedonia, the impaired ability to experience pleasure, and might look for underlying brain abnormalities that they share in common. "I bet that the rough spots are overcome pretty quickly," Carpenter says, "but of course we have to see how well that actually works out."

Cuthbert emphasizes that the new system is a framework for research, not a diagnostic manual, and that it has not yet been tested. "It's a platform to get people moving in the right direction," he says. In the meantime, the DSM "has been and continues to be very useful in psychiatry," he says. For the sake of patients, he says, "it's important to communicate that we do have good treatments for mental disorders."