Do We Need a DSM-V?

Below is an editorial (with the above title) taken from the newsletter of the Society of Biological Psychiatry and written by its editor Stephen M. Strakowski, MD, who is also Vice President of Research at the University of Cincinnati Health System and the Stanley and Mickey Kaplan Professor and Chairman of its Department of Psychiatry.

Dr Strakowski writes, “Recently, psychiatric diagnosis is on our minds with DSM-V committees in full swing. In my view, a diagnosis is used to isolate a ‘case’ from the general population for three primary reasons: 1) to define a treatment that will alleviate suffering; 2) to predict outcome (prognosis); and 3) to facilitate research into illness etiology. With these considerations in mind, do we need a DSM-V?

The Diagnostic and Statistical Manual: Mental Disease (DSM-I) originated in 1952. DSM-II expanded this volume and was published in 1968. In both publications, descriptions of psychiatric conditions were dominated by psychodynamic and psychoanalytic conceptualizations of most psychiatric conditions. Because these DSM-I and II categorizations were difficult to operationalize, they were of limited value for medical research; moreover they did not gel with the International Disease Classification (ICD), and frankly did not sufficiently address the three reasons for making a diagnosis in the first place.

Consequently, DSM-III (1980) was developed to address these shortcomings. DSM-III reconceptualized psychiatric disorders into a medical-research model, revolutionizing psychiatric nomenclature and invigorating neuroscience approaches to psychiatric research. Many of these diagnoses were useful for defining treatment and predicting outcome, although discoveries of etiologies have remained elusive. DSM-III was revised in 1987 (DSM-III-R) and again in 1994 (DSM-IV, and DSM-IV-TR).

Significant changes in diagnostic criteria occurred with each revision, largely to make the DSM more ‘user-friendly’ for clinicians and address concerns raised about being both overly narrow and too vague when “pathologizing” behavior. However, few studies were performed to determine whether changes in diagnostic criteria improved: 1) our ability to predict treatment response; 2) our ability to predict outcome; or 3) our ability to define etiologies.

As we face DSM-V then, has sufficient new information arisen to warrant changes in diagnostic criteria yet again? In short, the answer is: No.

So what harm is there in prematurely altering diagnostic criteria? Indeed, from a clinical viewpoint, such changes might open treatment options for people who might otherwise not receive them and may therefore benefit. Are we not trying to do exactly this?

Unfortunately, in the absence of treatment data justifying a criteria change, people are just as likely to be exposed to ineffective treatments and unnecessary side-effects. Prognostic data take years to accumulate, typically longer than between each revision of DSM. Consequently, another revision will simply restart and delay this incomplete process. Whether proposed changes will facilitate studies into the etiologies of the newly defined syndromes is not known, but previous revisions have not exhibited such successes in the past. Finally, with each revision, the number of diagnosable conditions increases. With each increase, psychiatry is criticized for ‘creating’ diagnoses to: 1) increase revenue to clinicians; 2) partner with big pharma to expand the mental health market; or 3) simply raise money for the DSM publishers. Consequently, in the absence of research demonstrating that new definitions meaningfully advance the utility of our diagnoses, our credibility with the public and our medical colleagues is challenged with each DSM revision. Only when we first accumulate research supporting changes in our diagnostic systems will we meet these challenges effectively. We are not at this point now.”

Thanks to Dr Strakowski and to the Society of Biological Psychiatry for their permission to reprint this edifying editorial. More information can be found in the October newsletter at http://www.sobp.org/files/public/October%202011%20Newsletter%20Opt.pdf.