DSM-5 in the Homestretch-1. Integrating the Coding Systems

With DSM-5 scheduled for publication a little more than a year from now, we may safely assume that barring unannounced surprises from, say, the APA Scientific Review Committee, what we will see on the DSM-5 Web site is what we will get. With that in mind, it’s time to review what we will indeed get. But before moving to significant changes in the major disorder categories, we should remind ourselves where DSM-5 fits into the larger picture of coding mental illnesses.

There are, in case you have forgotten, two classificatory systems of mental disorders-the International Classification of Diseases (ICD), produced by the World Health Organization (WHO), and the Diagnostic and Statistical Manual (DSM), produced by the American Psychiatric Association. How are they related? It is a question that has confused me, and I assume, some of my psychiatric colleagues as well as others-other mental health professionals, and still others. For an answer to this question, I asked Michael First, MD, Editor of DSM-IV-TR, Consultant on the WHO ICD-11 revision, and someone who knows much more about these matters than most of us. Here is his response to what I put as a question that could lend itself to a quick answer.

Dr First:I wish I could give you a quick answer, but as with any question about the coding, I need to give the background first.

This is how it works. The only official coding system in the United States is the ICD, produced by the WHO. The US has a treaty obligation to report health statistics using the ICD system. The US is still using a "clinical modification" of the WHO's ICD-9 system (released in 1978) called ICD-9-CM. The diagnostic codes that appear in the DSM-IV are all legal ICD-9-CM codes. Clinicians fulfill their obligations to use the ICD-9-CM coding system by using the DSM-IV. When DSM-IV created its few new disorders, which were obviously not in the ICD-9-CM, the APA could either assign an existing ICD-9-CM code (which might be already used for another DSM-IV disorder, in which case multiple disorders would share the same code), or else request that the US Government to add a new code to the ICD-9-CM system (such a provision exists in the ICD-9-CM system to accommodate newly discovered or new subtypes of diseases), which is a year-long process.

When assigning an existing code, we would pick the code that was closest to the phenomenology of the new disorder. For example, ICD-9-CM had a code for "depressive neurosis," which was then assigned to the DSM-III disorder "dysthymia," because that ICD-9-CM concept came closest to the DSM-III concept. More often, for a category that was not really reflected in the ICD-9-CM at all, we would assign an "other" code, which are available throughout the ICD-9-CM precisely to accommodate the addition of new disorders. So for Bipolar II Disorder, which we added to DSM-IV, we picked the code 296.89, which corresponded to "Other Bipolar Disorder" in ICD-9-CM.

For DSM-5, the same general rule applies. For new diagnoses such as Attentuated Psychosis Syndrome, APA would need to look at the existing system and either find some category that corresponded most closely to this condition (although it is not clear to me that there is any such condition already in ICD-9-CM), or else assign some "other" code; perhaps the code for "Other Psychotic Disorder," although even that would be a stretch since, by definition, these individuals are not psychotic. Even though the newly assigned code may not up conceptually match very well to the existing ICD-9-CM entity, from a practical perspective it doesn't matter-insurance companies seem to accept and pay for most ICD-9-CM codes no matter what they are, unless they have been specifically excluded by the insurance company. For example, perhaps an insurance company would want to refuse payment for treatment of personality disorders; they could exclude any of the codes for personality disorders.

Further complicating matters is the recent news that implementation of ICD-10-CM is being delayed from October 2013 to God knows when. Originally, DSM-5 was planning to include ICD-10-CM codes to be compatible with the requirement that clinicians use ICD-10-CM codes starting in October 2013 (5 months after publication of DSM-5). Now that there will be a delay, DSM-5 will need to use ICD-9-CM codes (ie, those in the DSM-IV), instead.

So welcome to the Byzantine process of psychiatric coding. Further implications of the above are that (1) it’s not legitimate to use ICD-10 codes in the US, although they are used in the rest of the world, and (2) you may use any ICD-9-CM code, even if it’s not in the DSM-IV.

For those of you who are interested in diagnostic coding parlor games, this situation leaves you with two such games. First, take your guess as to what ICD-9 code will be dragooned to cover DSM-5 diagnoses such as Attenuated Psychosis Syndrome and Disruptive Mood Dysregulation Disorder (as you saw above, Michael First already stated his guess for APS). And second, take a tour through that 1978 document, ICD-9-CM, and see whether there are any diagnoses left out of DSM-III and IV that you would like to add to your personal arsenal.