Using big data of genetics, health claims, and brain imaging to challenge the categorical classification in mental illness

Abstract

Psychiatric disorders in first-degree relatives (FDRs) often differ from the index patient's diagnosis, suggesting that there is genetic contribution to psychiatric disorders in which related cases do not all map to the same diagnosis as the index case. Our aim is to look for psychiatric comorbidities across major mental illnesses using three approaches, genetics, clinical diagnosis, and brain imaging to address common associations and pathology among mental illnesses. Genome-wide association studies from the Psychiatric Genomics Consortium showed single gene polymorphisms are common across 5 major psychiatric disorders, including schizophrenia (SZ), bipolar disorder (BD), major depressive disorder (MDD), autism spectrum disorder (ASD), and attention deficit hyperactivity disorder (ADHD). Clinically, results of Taiwan's nationwide population studies showed that other major psychiatric disorders were more likely to coaggregate in families with an index case of an individual with a psychiatric disorder, compared to control families. Finally, resting functional connectivity (FC) magnetic resonance imaging (MRI) and whole-brain connectomic analysis of SZ, BD I, BD II, MDD, and healthy controls revealed that the four groups of patients shared similar patterns of abnormal neural substrate in the brain that differed from controls. In conclusion, using big data from genetics, administrative health claims, and brain imaging, we identified concordance, indicating dimensional coherence of genetic heritability, clinical mutual associations, and common neurobiological substrates across major psychiatric disorders. These results will challenge the current diagnostic classification system and possibly move psychiatry beyond descriptive syndromes towards a nosology informed by disease cause.