Skip to main content

Transforming the understanding
and treatment of mental illnesses.

Celebrating 75 Years! Learn More >>

 Archived Content

The National Institute of Mental Health archives materials that are over 4 years old and no longer being updated. The content on this page is provided for historical reference purposes only and may not reflect current knowledge or information.

New Processing Technique Helps Researchers Use Electronic Health Records to Study Biological Contributors to Mental Illnesses

Symptom Dimensions Based on NIMH Research Domain Criteria Definitions Linked to Genes Relevant to Psychopathology

Science Update

Researchers have found a way to scan electronic health records (EHRs) that helps identify associations between broad dimensions of behavioral function and genes relevant to mental disorders. Use of the technique opens an enormous source of data to researchers who are interested in taking a dimensional approach to the study of mental illnesses instead of using traditional diagnostic categories.  The study , funded in part by the National Institute of Mental Health (NIMH), was published online February 26, 2018 in the journal Biological Psychiatry.

As medicine has entered the digital age, the use of electronic systems for managing health data has skyrocketed. These electronic health records provide a trove of information for researchers who want to understand factors that contribute to health and illness.

“Electronic health records are like very large, very messy studies,” said lead author Thomas McCoy Jr., M.D., of Massachusetts General Hospital and Harvard Medical School. “They contain a wealth of information about a huge number of people but can require special methods to utilize this information.”

Dr. McCoy and coauthors applied a new natural language processing method (described in a companion paper  also published in Biological Psychiatry) to the narrative discharge summaries of individuals who were genotyped as part of the Partners HealthCare Biobank initiative and also hospitalized between 2010 and 2015. The researchers employed the processing method to extract symptom dimensions based on the NIMH Research Domain Criteria (RDoC) from the EHRs — even though the EHRs were recorded without considering RDoC dimensions.

“RDoC is a research framework for psychopathology that is based upon dimensions of human behavior and functioning that reflect contemporary knowledge about major systems of emotion, cognition, motivation, and social behavior,” said Bruce Cuthbert, Ph.D., Director of the NIMH RDoC Unit. “The purpose of the RDoC project is to support research that explores the range of these behavioral or cognitive domains— from normal to abnormal—in order to understand the developmental processes and environmental influences that can lead to increasingly dysregulated functioning and the onset of symptoms of various types.”

Dr. McCoy and colleagues examined the genomes of participants to determine whether any common genetic variations were associated with each of the five RDoC-based symptom dimensions extracted from the EHRs. They found:

  • One genetic locus (a fixed position on a chromosome) associated with the arousal dimension. This locus spanned a genetic area that research suggests is important in primate neocortical evolution and development;
  • Two loci associated with the social dimension; and
  • One locus associated with the cognitive dimension and thought to be related to immune response in Alzheimer’s disease.

The authors note that neither of the loci associated with the arousal and cognitive dimensions were previously associated with mental disorders.

“We found that a simple method can extract RDoC-informed symptom dimensions from health records, and that the scores related to these dimensions correlate with expert ratings, clinical outcomes, and genetic variation,” said Dr. McCoy. “This method is a step toward enabling researchers with a health record data set to study transdiagnostic symptom domains.”

References

McCoy, T., Castro, V., Hart, K., Pellegrini, A., Yu, S., Cai, T., & Perlis, R. (in press). Genome-wide association study of dimensional psychopathology using electronic health records . Biological Psychiatry

McCoy. T., Yu, S., Hart, K., Castro, V., Brown, H., James N. Rosenquist, … Perlis, R. (in press). High throughput phenotyping for dimensional psychopathology in electronic health records.  Biological Psychiatry

Grants

MH106933-04 , MH106577-01A1 , HG00738-01 , MH106577-01A1