Does the DSM need to be re-written?

I recently came across an interesting paper that looked at the likelihood that a person, once diagnosed with one mental disorder, would be diagnosed with another. (Exploring Comorbidity Within Mental Disorders Among a Danish National Population, by Oleguer Plana-Ripoll.)

This was a remarkable study in two ways. First, it had a sample size of 5,940,778, followed up for 83.9 million person-years–basically, the entire population of Denmark over 15 years. (Big Data indeed.)

Second, it found that for virtually every disorder, one diagnoses increased your chances of being diagnosed with a second disorder. (“Comorbid” is a fancy word for “two diseases or conditions occurring together,” not “dying at the same time.”) Some diseases were particularly likely to co-occur–in particular, people diagnosed with “mood disorders” had a 30% chance of also being diagnosed with “neurotic disorders” during the 15 years covered by the study.

Mood disorders includes bipolar, depression, and SAD;

Neurotic disorders include anxieties, phobias, and OCD.

Those chances were considerably higher for people diagnosed at younger ages, and decreased significantly for the elderly–those diagnosed with mood disorders before the age of 20 had a +40% chance of also being diagnosed with a neurotic disorder, while those diagnosed after 80 had only a 5% chance.

I don’t find this terribly surprising, since I know someone with at least five different psychological diagnoses, (nor is it surprising that many people with “intellectual disabilities” also have “developmental disorders”) but it’s interesting just how pervasive comorbidity is across conditions that are ostensibly separate diseases.

This suggests to me that either many people are being mis-diagnosed (perhaps diagnosis itself is very difficult,) or what look like separate disorders are often actually one, single disorder. While it is certainly possible, of course, for someone to have both a phobia of snakes and seasonal affective disorder, the person I know with five diagnoses most likely has only one “true” disorder that has just been diagnosed and treated differently by different clinicians. It seems likely that some people’s depression also manifests itself as deep-rooted anxiety or phobias, for example.

While this is a bit of a blow for many psychiatric diagnoses, (and I am quite certain that many diagnostic categories will need a fair amount of revision before all is said and done,) autism recently got a validity boost–How brain scans can diagnose Autism with 97% accuracy.

The title is overselling it, but it’s interesting anyway:

Lead study author Marcel Just, PhD, professor of psychology and director of the Center for Cognitive Brain Imaging at Carnegie Mellon University, and his team performed fMRI scans on 17 young adults with high-functioning autism and 17 people without autism while they thought about a range of different social interactions, like “hug,” “humiliate,” “kick” and “adore.” The researchers used machine-learning techniques to measure the activation in 135 tiny pieces of the brain, each the size of a peppercorn, and analyzed how the activation levels formed a pattern. …

So great was the difference between the two groups that the researchers could identify whether a brain was autistic or neurotypical in 33 out of 34 of the participants—that’s 97% accuracy—just by looking at a certain fMRI activation pattern. “There was an area associated with the representation of self that did not activate in people with autism,” Just says. “When they thought about hugging or adoring or persuading or hating, they thought about it like somebody watching a play or reading a dictionary definition. They didn’t think of it as it applied to them.” This suggests that in autism, the representation of the self is altered, which researchers have known for many years, Just says.

N=34 is not quite as impressive as N=Denmark, but it’s a good start.