September 20th, 2010 by gelenberg Leave a reply »

Most psychiatrists are aware of the turmoil swirling around the next iteration of the venerable DSM: DSM-V. The process and many proposed changes are under scrutiny and sometimes attack. I have two lower-key concerns, which apply to DSM-IV and probably will be relevant to the next generation.

The shift from DSM-II to DSM-III gave us the current multi-axial format, based largely on the systematic descriptive approach used  at Washington University in St. Louis. DSM-III brought superior reliability: independent assessors were more likely to agree on a diagnosis.

But many senior clinicians have lamented the “checklist interview” this approach has fostered. We observe residents rush from question to question, dutifully recording the spoken answer, without taking in the nuance, tone of voice, facial expression, and body posture of the person sitting opposite. The follow-up question that a patient may be begging to have asked goes unaddressed. Surely, we need to collect and record data. But we can do so without losing the art of good interviewing. Psychiatrists still need to “listen with the third ear.”

And if reliability with DSM-III and IV are so high, how do we account for patients with long strings of incompatible diagnoses? Sit at the door on any psychiatric unit, and those conflicting diagnoses in patients’ histories are the rule, not the exception. In my experience, they reflect the sloppy use of the DSM rules. Sure, you and I are likely to agree on a diagnosis in most axes—provided that we both play by the same rules.  Like it or not, the DSM is our dictionary, the common language that allows communications. And like any language, it requires precise use per agreed-upon definitions. When we do less, any “signal” gets lost in the “noise.”


Comments are closed.