Overdiagnosis, Mental Disorders and the DSM-5

Is the DSM-5 — the book professionals and researchers use to diagnose mental disorders — leading us to a society that embraces “over-diagnosis”? Or was this trend of creating “fad” diagnoses started long before the DSM-5 revision process — perhaps even starting with the DSM-IV before it?

Allen Frances, who oversaw the DSM-IV revision process and has been an outspoken critic of the DSM-5, suggests melodramatically that “normality is an endangered species,” due in part to “fad diagnoses” and an “epidemic” of over-diagnosing, ominously suggesting in his opening paragraph that the “DSM5 threatens to provoke several more [epidemics].”

First, when a person starts throwing around a term such as “over diagnosing,” my first question is, “How would we know we’re ‘over diagnosing’ a condition, versus gaining a better understanding of a disorder and its prevalence within modern society?” How can we determine what is being accurately, better and more frequently diagnosed today, versus a disorder that is being “over diagnosed” — that is, being diagnosed when it shouldn’t be due to marketing, education or some other factor.

We could look at attention deficit disorder (also known as attention deficit hyperactivity disorder, or ADHD). The National Institutes of Health convened a panel in 1998 to examine the validity of attention deficit disorder and its treatments, out of concern for the rising amount of children being diagnosed with attention deficit disorder. However, they barely mention overdiagnosis as a concern for ADHD in their consensus statement. They do point out one of the primary problems is inconsistent diagnosing, which I agree represents a real, ongoing concern across the spectrum of mental disorders.

Research into this question has produced mixed results, showing that on one hand, we are indeed over-diagnosing even common, serious mental disorders like bipolar disorder, but we’re also missing a lot of people who have the disorder and have never been diagnosed — again, inconsistent diagnosing. Bipolar disorder should be fairly accurately diagnosed because its diagnostic criteria are clear and overlap with only a few other disorders. One such study that examined whether we are “over diagnosing” bipolar disorder was conducted on 700 subjects in Rhode Island (Zimmerman et al, 2008). They found that less than half the patients who self-reported as being diagnosed with bipolar disorder actually had it, but that over 30 percent of patients who claimed never to have been diagnosed with bipolar disorder actually did have the disorder.

What this kind of study perhaps best demonstrates is the deeply flawed nature of our current diagnostic system based upon the categories set forth by the DSM-III, expanded upon in the DSM-IV, and now being further expanded upon in the DSM5. It is not simply a black and white issue of “over diagnosis.” It is a subtle, complex problem that requires subtle, complex solutions (not a machete taken to pare down sheer numbers of diagnoses). It shows, to me anyway, that perhaps the criteria are fine — the quality, reliable implementation of those criteria continue to leave a lot to be desired.

But diagnoses are not a finite numbers game. We don’t stop adding to the ICD-10 just because there are already thousands of diseases and medical conditions listed. We add to it as the medical knowledge and research supports the addition of new medical classifications and diagnoses. The same is true for the DSM process — hopefully the final revision of DSM5 won’t have added dozens of new disorders because the workgroup believed in a “fad” diagnosis. Rather, they add them because the research base and consensus of experts agrees it’s time to recognize the problem behavior as a real concern worthy of clinical attention and further research.

Who is Dr. Frances to say whether “binge eating disorder” is “real” or not? Has he replicated the work of the DSM5 eating disorders workgroup to arrive at that conclusion? Or is he just picking some diagnoses he feels are “fads” and makes it so? I wouldn’t dream of second-guessing a panel of experts in an area, unless I also spent some significant time reading up on the literature and arriving at my own conclusions through the same type of study and discussion the workgroups use.

The article goes on to list the possible reasons that over-diagnosis takes place, but the list basically boils down to two things — more marketing and more education. Nowhere on his list does he mention the most likely cause of ‘over diagnosis’ — the general unreliability of diagnoses in everyday, real clinical practice, especially by non-mental health professionals. For instance, he’s concerned that websites setup to help people better understand a mental health concern (such as ours?) may lead to people self-overdiagnosing. Self overdiagnosing? I think Dr. Frances just coined a new term (and perhaps a new phenomenon unto itself)!

Outside of this strange vortex, I call such websites and support communities “education” and “self-help.” The research literature is full of studies demonstrating that these websites help people better understand issues and get emotional support and direct, immediate help for them. Could some people use them to inaccurately diagnose themselves? Certainly. But is it a problem of epidemic proportions? I’ve seen no evidence to suggest it is.

Education is key to reaching out to people to help address the decades worth of mis-information and stigma surrounding mental health concerns. Do we just turn off the spigots and lock up the knowledge again in inaccessible books where only the elite and “properly trained” professional has access to it (as psychiatry has traditionally done with the DSM-III-R and even the DSM-IV)? Or do we keep the doors and windows of knowledge wide open and invite as many people as we can into to take a look around and better understand the serious emotional or life issues they are dealing with?

Last, if the DSM itself is partially to blame for over-diagnosis — e.g., because the diagnostic criteria are set too low, as Dr. Frances suggests — then I reiterate my previous suggestion: perhaps the usefulness of the DSM itself has passed. Perhaps it’s time for a more nuanced, psychologically-based diagnostic system to adopted by mental health professionals, one that doesn’t medicalize issues and turn every emotional concern into a problem that has to be labeled and medicated.

I think that the problems of over- and under-diagnosis of mental disorders should be addressed, but I see them as an entirely separate (and more complex) issue from the current revision of the DSM-5 and using the quantity of mental disorders as some sort of gauge to address the quality of diagnosis. Because I believe it’s the quality of our diagnoses — the ability to accurately translate diagnostic criteria to symptoms presented by real people — that most affects “over diagnosis,” not marketing or patient education.

Would we be looking to blame Merriam Webster for all of the trash romance novels that exist? Or do we blame the authors who put the words together to create the novels? Do we blame the DSM for poor diagnoses, or do we blame the professionals (many of whom are not even mental health professionals) who make the poor diagnoses in every day practice?

Read the full article: Normality Is an Endangered Species: Psychiatric Fads and Overdiagnosis