Research consortium develops evidence-based diagnostic model for mental illness

BUFFALO, N.Y. — A consortium of psychologists and psychiatrists has developed a new, evidence-based alternative to the mental health field’s long-established diagnostic tools for the classification, treatment, and research of mental disorders, according to a University at Buffalo psychologist who is one of the co-authors of a paper that explains the groundbreaking approach.

The Hierarchical Taxonomy of Psychopathology (HiTOP) addresses what the authors say are limitations to the reliability and validity of traditional models like the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), the American Psychiatric Association’s (APA) authoritative handbook used by clinicians and researchers around the world to diagnose and treat mental disorders.

“HiTOP is the first attempt by any group of individuals to put forth a classification and diagnostic system that has the features we’ve described,” says Leonard Simms, an associate professor in UB’s Department of Psychology and one of the 40 researchers who worked with team leaders Roman Kotov of Stony Brook University, Robert Krueger of University of Minnesota, and David Watson of University of Notre Dame on the study that appears in the latest issue of the Journal of Abnormal Psychology.

Simms, an expert in the description and classification of mental disorders, says the potentially paradigm-shifting model has the capacity to advance research efforts and improve clinical outcomes related to the causes and treatments of mental disorders.

HiTOP’s guiding spirit is correcting the shortcomings of DSM-5 and other similar classification schemes, like the World Health Organization’s International Classification of Diseases (ICD), by changing the way mental disorders are classified and diagnosed.

HiTOP uses a diagnostic approach that is dimensional and hierarchical. Traditional systems, like DSM-5, are categorical.

Categorical systems associate each disorder with a set of symptoms. Clinicians make a diagnosis of a disorder only when patients present an established minimum number of those symptoms. For instance, Major Depressive Disorder is associated with nine symptoms. At least five of those symptoms must be present for a patient to receive a diagnosis of Major Depressive Disorder.

“That’s an arbitrary classification,” says Simms. “Somebody with four symptoms of depression could be experiencing as much if not more impairment than someone who meets the five criteria. Yet five gets the diagnosis and four does not. You see this throughout DSM-5.”

“I use a word like ‘arbitrary’ because in many cases the threshold in the diagnostic manual is usually half the number of symptoms. There is no evidence brought to bear on that threshold.”

Forcing people into categories means losing critical information because of distinctions between symptoms and impairment.

“That distinction creates a false negative,” says Simms. “A patient can have one symptom of depression and still be impaired.”

By eliminating arbitrary boundaries that separate either having a disorder or not having a disorder, researchers and clinicians can make more meaningful decisions.

Simms says statistical analysis shows that shades of gray, or dimensions, are more meaningful than categories.

“There are a variety of statistical techniques that have been in use over the last 25 years that allow us to determine whether underlying symptoms are better described as a categorical or dimensional phenomenon, with the vast majority of that evidence favoring a dimensional approach to psychiatric classification,” he says.

HiTOP’s hierarchical component is based on analysis of symptom similarities.  Any group of symptoms might be very close to others.

“There are various ways to talk about depression or anxiety,” says Simms. “Statistics provide researchers with evidence-based ways of combining those symptoms or not. DSM-5 has more disorders than we need. It’s not always clear how one disorder differs from another.”

The core issue for the authors is that traditional systems have been shaped by considerations other than empirical evidence.

“A lot of this is inertia,” says Simms. “We’ve had categories for mental disorders for decades, and that inertia has been an impediment to making changes in the way we think about the mental disorders.”

It comes down to a system characterized by past practice.

“Imagine a physician saying, ‘the research says we should do an MRI on your knee, but my training was in the 1970s, so we’re going to take an X-ray and that’s going to have to be good enough.’ The same thing applies here. Many current clinicians are not being influenced by the evidence.”

Simms says a system, like HiTOP, that’s based on solid evidence is an advance all by itself.

“A diagnostic system that places people into these messy categories that aren’t necessarily distinct from one another creates a lot of noise in the research world,” he says. “We can make further advances in research into the causes and treatments of these disorders if we have an evidence-based system with known patterns of correlation among these symptoms.

“If we have a system that’s cleaned up this way, not only would the research be stronger in terms of the causes and treatments of these disorders, but it presumably would lean toward better connections with different treatment modules that would be useful clinically.”

The HiTOP classification system remains a work in progress, but several parts of the model are ready for clinical and research applications, according to Simms.