Mental health conditions: five genetic factors reshape psychiatry

Depression, autism, schizophrenia and addictions: five genetic factors account for most of the predisposition to these mental health conditions.

Psychiatry’s bible is the DSM (Diagnostic and Statistical Manual of Mental Disorders). Its first edition, DSM-I, was published in 1952 in the aftermath of the Second World War. Like the Vidal drug reference in medicine, it is the definitive reference manual used by psychiatrists and mental health professionals to identify and classify psychological disorders. It has undergone major changes-eight in total when revised editions are included-and DSM-5-TR (Text Revision) is now the governing version.

How the DSM transformed psychiatric diagnosis

When DSM-III was published in 1980, psychiatry made its most dramatic shift, moving away from a psychoanalytic understanding of mental disorders and towards fixed diagnostic categories. A particular combination of symptoms was henceforth matched to a specific illness and to treatment intended to cure or alleviate it, in the same manner as the rest of medicine distinguishes pneumonia from an ulcer. That framework remains in place today, even though modern psychiatry fully recognises that the human brain cannot be so readily placed into neatly labelled boxes.

That view began to take shape 20 years ago, particularly as genetics and brain imaging advanced. Numerous studies had shown that psychiatric disorders share common biological roots, yet no research had conclusively demonstrated this proposition across such a broad set of conditions until this study was published in Nature on 10 December 2025. Its findings indicate that five genomic factors-a profile or cluster of DNA variations shared by several people-are enough to account for most of the genetic inheritance common to fourteen mental health conditions.

The major breaking down of mental illness boundaries

To reach this conclusion, Andrew Grotzinger, a behavioural genetics researcher at the University of Colorado Boulder, and his team compared the DNA of more than one million patients with that of five million healthy people, allowing correlations to emerge. This enabled them to identify the five main genomic factors.

The first cluster covers disorders with a compulsive component: anorexia nervosa, Gilles de la Tourette syndrome and obsessive-compulsive disorder (OCD). The second brings together so-called internalising disorders, including depression, anxiety and post-traumatic stress disorder (PTSD). Addictive behaviours form the third group, while neurodevelopmental disorders, such as autism and attention deficit hyperactivity disorder (ADHD), make up the fourth. This may explain why the latter two conditions so frequently occur together.

Schizophrenia and bipolar disorder share genetic factors

The fifth group is arguably the hardest for psychiatry to accept: schizophrenia and bipolar disorder share 70% of the genetic variations that predispose people to either condition. Yet separating these two disorders has been one of the discipline’s oldest foundations since the late 19th century, following the work of German psychiatrist Emil Kraepelin. This divide prevented clinicians from assigning both diagnoses to the same patient and, consequently, from considering that they might arise from shared biological foundations.

A century-old diagnostic divide shattered

The DSM had already drawn considerable criticism for its arbitrary nature and for encouraging doctors towards overmedication, but this finding is undoubtedly one of the most significant blows dealt to it. A century of clinical practice had kept the genetic relationship between schizophrenia and bipolar disorder at arm’s length, turning their separation into an untouchable dogma; genetics has now demonstrated its lack of consistency.

The validity of the categorical model imposed since DSM-III, already heavily disputed since the 1990s, is now contradicted by the evidence that is hardest to dismiss. Although the manual will not disappear any time soon, as it remains practitioners’ common language and the framework of healthcare systems, it will undoubtedly lose some of the lustre of its authoritative status. While this study does not invalidate its use in any way, it strikes at its most sensitive claim-one which none of its seven revisions had managed to challenge: that its compartmentalisation of disorders corresponded to patients’ biological and genetic reality.