DSM-5 enters the diagnostic fray

To a cacophony of boos, so-whats and even a few cheers, the American Psychiatric Association released the fifth edition of its Diagnostic and Statistical Manual of Mental Disorders, DSM-5, on May 18 at its annual meeting in San Francisco.

Controversy always flares when psychiatrists redefine which forms of human suffering will count as real and reimbursable by medical insurance. This time, though, the stakes are raised by competing efforts to classify mental disorders.

The World Health Organization plans to release a new version of its own system for identifying mental ailments in 2015 as part of the 11th edition of the International Classification of Diseases. It’s not clear how much the ICD will mirror DSM-5.

Some differences have already emerged. Clinicians working on the international classification report in the May 11 Lancet that they plan to pare down the number and types of symptoms needed to diagnose post-traumatic stress disorder, or PTSD, and add a severe form of the condition triggered by long-lasting or frequent harrowing events. These departures from DSM-5 would make it easier for mental health workers to help victims of conflict and natural disasters in poor, non-Western countries, say psychologist Andreas Maercker of the University of Zurich and his colleagues. 

Meanwhile, the National Institute of Mental Health in Rockville, Md., has launched the Research Domain Criteria, or RDoC, a 10-year effort to define mental disorders based on behavioral and brain measures. DSM’s approach, by contrast, relies on rulings by groups of psychiatrists about which symptoms characterize particular disorders. The approach has yielded imprecise diagnostic labels that advance neither treatment nor research, argued psychiatrist and NIMH director Thomas Insel in an April 29 blog post.

Insel’s statement raised hackles at the psychiatric association. On May 13, Insel and American Psychiatric Association president Jeffrey Lieberman together released a conciliatory statement declaring that DSM-5 and RDoC complement each other on the path to better diagnoses for mental disorders.

RDoC will fund research that examines how lots of factors — fear, attention, parenting styles and neighborhood qualities, to name a few — interact to produce symptoms that may or may not jibe with DSM-5 categories, said the RDoC project’s director, psychologist Bruce Cuthbert, at the Association for Psychological Science annual meeting in Washington, D.C., on May 23. Not everyone with, say, autism spectrum disorder or PTSD has the same underlying problems, he says.

“RDoC is about understanding the biology and the psychology of mental illness,” Cuthbert says. “DSM-5 is sloppy on both accounts.”

DSM-5 has also gotten hammered — especially by psychiatrist Allen Frances, chair of the task force that produced the previous DSM — for allegedly turning some common forms of distress into medical conditions, encouraging physicians to prescribe unneeded psychoactive medications.

Normal grief will become an illness in DSM-5, Frances says. Mourning and eventual acceptance of a loss will be replaced, in his view, by “pills and superficial medical ritual.”

As with any highly upsetting event, a loved one’s death triggers major depression in some people, responds University of Pittsburgh psychiatrist David Kupfer, who chaired the psychiatry association’s DSM-5 task force. The new manual makes clear that in grief, painful feelings come in waves and self-esteem is preserved; in depression, dark moods endure and include feelings of worthlessness and self-loathing. Clinicians should be able to separate grief from depression, Kupfer says.

When the process of developing DSM-5 started 14 years ago, those involved were optimistic that biological markers of mental disorders were just around the corner. To their disappointment, scientific validation of DSM-5 categories, from schizophrenia to major depression, remains a distant goal.

“DSM-5 isn’t perfect, but it is the best we can do with the science available,” Kupfer says.

Field trials leading up to the final version of DSM-5 assessed the extent to which 279 clinicians trained in the new manual and prompted by a computerized checklist agreed on diagnoses for nearly 2,000 patients at seven adult and four child psychiatric hospitals. Results were mixed.

Good agreement existed about which patients qualified for conditions such as PTSD, the most severe form of bipolar disorder, borderline personality disorder, autism spectrum disorder and attention-deficit hyperactivity disorder. But discord reigned when the clinicians tried to determine which patients met DSM-5 criteria for major depression, generalized anxiety disorder, antisocial personality disorder and several other ailments.

Many psychiatric disorders include symptoms of depression and anxiety that can complicate diagnostic and treatment decisions. Fuzzy boundaries separating many mental ailments mean that, as with previous manuals, most people deemed to have one DSM-5 disorder will also have one or more additional disorders, Kupfer says.

Kupfer and his colleagues faced intense scrutiny for revising how autism and related conditions are diagnosed. A tightened autism definition in DSM-5 has raised fears among advocacy groups that some children with this condition will go undiagnosed and be denied special school services. DSM-5 folds four previously separate categories on the autism spectrum, including Asperger syndrome, into an umbrella term, autism spectrum disorder. Under DSM-5, individuals with this diagnosis get rated on the severity of their social problems, the restrictiveness of their interests, and the extent to which they engage in repetitive behaviors. Language difficulties can coexist with these symptoms.

No one knows how, or whether, these changes will affect autism rates.

---

Back Story | Changing notions of autism

As autism rates have climbed, physicians and parents have taken an increasing interest in how the DSM defines the condition. Before 1980 the book didn’t offer separate criteria for autism, but mentioned it within the entry for childhood schizophrenia.

DSM-I
The final sentence in the entry for “Schizophrenic reaction, childhood type,” dictates that “psychotic reactions in children, manifesting primarily autism, will be classified here.”

DSM-II
The second edition of the manual takes a similar approach to that of the first. Autism appears in the entry for “Schizophrenia, childhood type,” which notes that “the condition may be manifested by autistic, atypical and withdrawn behavior.”

DSM-III
Autism is broken out as its own disorder in an entry headed “Infantile Autism.” There are six criteria for the diagnosis, including onset before 30 months of age, pervasive lack of responsiveness to other people and gross deficits in language development.

DSM-III-R
Now called autistic disorder, the requirements for diagnosis are much more complex. The manual specifies 16 characteristics grouped into three categories. At least eight of the 16 are required for a diagnosis, with at least two from one category and one from each of the other two.

DSM-IV and DSM-IV-TR
These editions require six or more characteristics from three categories, with at least two from the first and one from each of the second two, for an autism diagnosis. The criteria primarily focus on impairments in social interaction, communication and behavior.

DSM-5
Previous editions identified a number of developmental disorders similar to autism, including Asperger syndrome, childhood disintegrative disorder and pervasive developmental disorder not otherwise specified. This edition largely combines those categories into a single entity labeled autism spectrum disorder, though a separate diagnosis called social communication disorder also exists.