Solving The Problem Of Questionable Diagnoses Grandfathered Into DSM

Charles Moser, PhD, MD, has forwarded an interesting suggestion to solve the problem of weak diagnoses that have received a free ride through previous revisions of DSM. His is a middle way intended to steer between the contrasting risks of continuing questionable diagnoses and the risks of eliminating them.

Dr Moser writes:

"Recently, Dr Frances discussed the problem of whether to keep some of the older diagnoses in the DSM that would not have been adopted under the more stringent empirical standards now in use. He argued to keep these diagnoses in the DSM-IV because “…when in doubt, stand pat and do least harm.” I would suggest that standing pat can actually do a lot of harm. The problem is not only the inconvenience to those who use the DSM, but also to those people who are labeled inappropriately and to Psychiatry’s scientific reputation.

Psychiatry, the APA, and the DSM editors should be dedicated to straightforward presentation of the facts (or lack thereof), exploring the problems with these diagnoses while adhering to the highest level of accuracy and completeness in reporting the data on which these decisions are made. Unfortunately, the DSM editors have not acknowledged any of the problems inherent in listing these grandfathered diagnoses in the DSM. Instead they keep suggesting that the DSM is a scientific document supported by an extensive empirical foundation. In fact, for some diagnoses (eg, the Paraphilias) there is minimal science to support inclusion, some studies refuting that they are even mental disorders, and serious risks attached to misuse.

All diagnoses should be subjected to a risk/benefit analysis and ideally the risk/benefit to the “patient” should take precedence over the risk/benefit to Psychiatry. For many of these grandfathered diagnoses, there is no clear benefit and many people so labeled have been harmed. The APA and the DSM editors should not need to be reminded that a guiding principle of all Medicine is “First, do no harm.” Ethically, the editors should explain the benefit and announce plans to limit the harm, or just remove these diagnoses. At the very least, the DSM editors need to minimize the potential harm from these diagnoses by acknowledging their problems and limitations.

There is a middle ground between continuing the status quo and complete removal. The DSM editors could acknowledge clearly which diagnoses do not meet the stringent criteria, either by creating and applying objective criteria or by creating a level of evidence system similar to those used evaluate medical interventions (for example, see http://www.uspreventiveservicestaskforce.org/uspstf/grades.htm). Alternatively, they could drop the pretense that the DSM is a scientific document based on an extensive empirical foundation.

I am sure there will be unforeseen consequences if either of these suggestions were adopted, but not all unforeseen consequences are negative. Any problem created could promote further research, something the DSM editors say they want but previous DSM editions have failed to spur. If the DSM is to live up to its purported reputation as a scientific document, the editors should submit their proposed draft for independent peer review. If the DSM is characterized as a consensus document, the editors should demonstrate consensus beyond the small group of DSM workgroup members and APA officers.

I really do not understand the editors’ reluctance to admit these diagnoses are problematic and openly discuss them in the DSM. I also do not understand why the APA membership has not demanded adherence to a more rigorous scientific standard."

My response:

Dr Moser offers an excellent idea, although one that has little chance of being accepted in the current climate and would be difficult to implement even under the best of circumstances.

An evidence-based rating of the scientific support, utility, and risks for all diagnoses in DSM-5 would constitute an invaluable consumers' guide and provide appropriate cautions for the use of the least supported and most potentially harmful categories. The credibility and usefulness of DSM-5 would be greatly enhanced were the ratings to be done by an independent group expert in the methods of evidence based medicine. The exercise would also go far in dispelling the misleading notions that DSM is some sort of "Bible of Psychiatry" and that all diagnoses are created equal and deserve equal belief and fealty.

But there are problems large and small. The immediate block to the implementation of Dr Moser's suggestion is the reluctance of the DSM-5 to allow open and independent review of any aspect of its work. After many delays, the DSM-5 Field Trials have finally begun, but we still don't know the wording of the criteria sets being tested. A "scientific review" of new proposals is also being done, but for no earthly reason it is meant to be kept strictly confidential. So the chances APA would now allow an independent and public review of every diagnosis in DSM-5 are below nil. This closed shop mentality calls into question the credibility of DSM 5 and the legitimacy of APA as custodian of the diagnostic system.

But even were everyone on board, rating the value of all the DSM criteria would still be a tough and controversial job. The best studied diagnosis is probably Major Depressive Disorder--but even MDD has controversial features, particularly the short 2 week duration required for milder depressions and the recommended subtypes. One could not meaningfully provide just one overall rating for MDD; many of the independent judgments would have to be scored separately.  

And consensus would be difficult to reach because there is much room for honest differences of opinion.

Even more challenging to an evidence-based approach are conceptual questions that have no right scientific answers. Antisocial personality disorder is the clearest cut and best studied of all the personality disorders, but one can have an interesting debate whether it is best considered a mental disorder at all-- many would say it is little more than a cluster of undesirable behaviors that have no place in a psychiatric nomenclature.

Dr Moser's suggestion would have its greatest ease of application and clearest utility in the example he provides from his own area of expertise--the Paraphilias. While this category has some clinical utility and a long historical tradition, its current dreadful misuse in the legal system sanctions a clear abridgment of constitutional rights and an abuse of psychiatry. It would be a great service to make explicit the remarkably weak research foundation for the Paraphilias and the great potential for their misapplication in court proceedings.
 
Dr Moser's suggestion will likely fall on deaf ears and would be tough to accomplish-- but it is clearly the right way to go.