Skip to main content

Rethinking How We Diagnose Psychosis

A new study opens a door to more biologically based categories of major mental illness

Credit:

©iStock.com

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


If you are unfortunate enough to develop acute chest pain this winter you will probably be assessed by a clinician who will order a battery of tests to determine if your symptoms result from pneumonia, bronchitis, heart disease, or something else. These tests not only can yield a precise diagnosis, they ensure you will receive the appropriate treatment for your specific illness.

If you are unfortunate enough to have a psychotic episode this winter, the process of arriving at a diagnosis will be quite different. In fact, there are not many choices. Most people with a psychotic disorder are labeled as having either schizophrenia or bipolar disorder. The distinction has been in textbooks for a century: schizophrenia (originally dementia praecox) is associated with delusions, hallucinations, an absence of affect, and a chronic course; bipolar (originally manic depressive disorder) can also involve delusions and hallucinations, and ,typically, dramatic swings in mood and a fluctuating course. But outside of textbooks, in the real world of the emergency room or clinic, these distinctions are less clear as many patients do not neatly fit the formal descriptions. Sadly, there are no blood tests or scans to distinguish schizophrenia from bipolar disorder.

While clinicians have become very skilled at assessing symptoms and signs, the absence of diagnostic laboratory tools or biomarkers poses a serious problem in psychiatry. Do all people with a label of schizophrenia have the same disorder? What about the large number of people who appear to have aspects of both schizophrenia and bipolar? Are these disorders, diagnosed exclusively by signs and symptoms, identifying distinct biological entities or could there be many different illnesses with a continuum of psychotic signs and symptoms? These questions are not merely academic.  As with chest pain, getting a precise diagnosis is important for selecting the best treatment. 


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Moving psychiatry into a new era of biologically based diagnosis has been a long sought goal—and a priority at the National Institutes of Mental Health (NIMH), where I served as director for 13 years until recently. Now a study published online in the American Journal of Psychiatry raises fresh hope. A study led by neuroscientist Brett Clementz of the University of Georgia, psychiatrist Carol Tamminga of the University of Texas, and colleagues at Yale and Harvard found distinct “biotypes” of psychosis that can be identified with quantitative biomarkers. In this study, from the Bipolar-Schizophrenia Network on Intermediate Phenotypes (B-SNIP) consortium, 711 people with a diagnosis of schizophrenia, bipolar disorder, or schizo-affective disorder (a hybrid of schizophrenia and bipolar disorder) were assessed with what the investigators call a brain-based panel of cognitive tests, studies of eye movements, a test of cognitive control, and electro-encephalogram. In addition, each subject had a brain imaging scan. 

Ignoring the clinical diagnosis, researchers pooled the data and analyzed it with unbiased, criterion-free statistical methods to look for what they called biotypes. Perhaps it is not surprising that the computer analysis from a large population with three diagnostic categories would find three clusters or biotypes.  But the three biotypes have very little relationship to the three diagnostic categories. In fact, people with schizophrenia and bipolar and schizo-affective disorders were distributed across the three biotypes.  And the biotypes did not differ simply by symptom severity or the presence of mania-related symptoms.

Is there any reason to think that these biotypes are more valid than a clinical diagnosis based on symptoms? A few observations suggest that the B-SNIP investigators may be on to something. First, some of the biotype differences were also found in first-degree family members, for whom data was also collected, suggesting a genetic basis for the new categories. Second, biotypes differed in social functioning—with people in Biotype 1 showing more serious functional impairment relative to the other biotypes. Third, the brain imaging studies (not used in defining the biotypes) showed clear differences in regional gray matter, especially in frontal, cingulate, temporal, and parietal cortex. While none of these observations proves that the biotypes are more valid than clinical diagnosis, these findings together encourage a fresh approach to the diagnosis of psychotic disorders. It will be important to know whether genomic variation, functional brain measures, or other behavioral measures can refine or further validate these biotypes.

Precision medicine has become a buzzword to describe the diagnostic revolution in cancer and other diseases. The concept is simply that the tools of modern biology including genetics and imaging can deconstruct current diagnostic categories yielding more precise disease groups that can be matched to more personalized treatments. For mental disorders, where laboratory tests have not been used in the clinic, precision medicine could be a disruptive innovation, revealing that many of the current diagnostic categories are imprecise and biologically heterogeneous. We are seeing evidence of this in autism and attention deficit hyperactivity disorder (ADHD) as well as depression and anxiety disorders. 

The NIMH has proposed a new approach to the diagnosis of mental disorders that calls for the incorporation of biological, cognitive, behavioral and social data in addition to observed symptoms and signs. This new approach, known as the Research Domain Criteria or RDoC, has been controversial both because it represents a major break with traditional psychiatry and because some doubt that it is viable.  The data from the B-SNIP study provides early evidence for the RDoC approach.

Of course, the real test of the B-SNIP biotypes—or any new method of diagnosis—is whether they will be effective in targeting treatments for patients and predicting outcomes. That will require studies of treatment response. Usually in the U.S., people with schizophrenia receive antipsychotic medication and sometimes psychosocial supports; bipolar disorder is treated with mood stabilizers, antipsychotic medications, sometimes antidepressants, and sometimes psychosocial supports. In truth, treatment remains empirical, with few guides to know which person will respond to a given therapy. One of the hopes of precision medicine for psychiatry is that the use of biomarkers will provide more predictability so that patients and clinicians can make more informed and precise treatment decisions. Psychotic disorders are among the most disabling conditions in all of medicine—with enormous costs in dollars and suffering for patients and their families. New approaches to diagnosis and treatment are long overdue.

Thomas Insel is a neuroscientist and psychiatrist who served as Director of the National Institute of Mental Health from 2002-2015. He has accepted a role at Verily, a new life sciences company within Alphabet (formerly known as Google).

 

Editor's Note: This post was adapted for the Scientific American MIND March/April 2016 Perspectives.

Thomas R. Insel is a psychiatrist and neuroscientist who served as director of the National Institute of Mental Health from 2002 to 2015. He is an adviser to Compass Pathways, as well as to several digital mental health companies, and author of Healing: Our Path from Mental Illness to Mental Health (Penguin, 2022).

More by Thomas R. Insel