Brain Imaging Studies May Distort Picture of Mental Disorders

January 6, 2017

Advances in brain imaging techniques have made it possible to look for the underlying neurobiological causes of psychiatric disorders, but no matter how detailed the images are, such studies may not paint an accurate picture of the people who live with these disorders.

According to a study from Columbia psychiatrist Mark Olfson, MD, MPH, brain imaging studies are often narrowly focused on the disorder of interest to the exclusion of those with common comorbid psychiatric conditions.

“Participants in psychiatric brain imaging studies are carefully chosen to have a given disorder without other co-occurring conditions,” said Dr. Olfson. “As a result, brain imaging studies leave out most of the people who are representative of those with the target disease.”

Dr. Olfson and colleagues compared eligibility criteria from 112 brain imaging studies of people with nicotine dependence, alcohol dependence, drug use disorder, major depressive disorder, or post-traumatic stress disorder with information gathered by the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) on patients with these disorders.

The researchers found that more than 86 percent of NESARC respondents with drug use disorder would have been barred from participating in brain imaging studies of the condition. The majority of patients with nicotine dependence (roughly 64 percent), alcohol dependence (nearly 60 percent), and PTSD (nearly 70 percent) also would have been ineligible. In contrast, only 18 percent of patients with major depression would have been excluded from such brain imaging studies.

“Imposing narrow inclusion criteria makes it easier to see what’s happening in the brains of people with a particular psychiatric disorder,” said Dr. Olfson. “Yet people with psychiatric conditions are often much more complex. They frequently have several concurrent conditions with different causes and effects. The highly selective nature of brain imaging studies raises questions about the extent to which their results apply to the average people with these conditions and whether larger studies should be designed to account for many of the co-existing conditions we commonly see in practice.”

Mark Olfson is professor of psychiatry at Columbia University Medical Center (CUMC).


The paper, titled “Generalizability of Neuroimaging Studies in 5 Common Psychiatric Disorders Based on the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC),” was published Dec. 28, 2016, in the Journal of Clinical Psychiatry.


Additional contributors were Carlos Blanco, MD, PhD (National Institute on Drug Abuse, Rockville, MD); Melanie M. Wall, PhD (CUMC); Martin A. Lindquist, PhD (Johns Hopkins Bloomberg School of Public Health, Baltimore, MD); Jorge Mario Rodriguez-Fernandez, MD (CUMC); Silvia Franco, MD (CUMC); and Shuai Wang, PhD (CUMC).


The research was supported by NIH grants DA019606MH076051, and MH082773; Agency for Healthcare Research and Quality grant HS02112; and the New York State Psychiatric Institute (Drs. Blanco, Olfson, and Wall).


The authors declare no conflicts of interest.



Press Release