DSM Disorders Disappear in Statistical Clustering of Psychiatric Symptoms
7 hours ago
- #Psychopathology
- #DSM-5
- #HiTOP
- The study 'Reconstructing Psychopathology' by Forbes et al. presents a data-driven reorganization of DSM-5 symptoms, challenging traditional psychiatric classifications.
- Using a large online survey with 14.8K participants, symptoms were assessed randomly to avoid bias, resulting in 680 items analyzed via statistical clustering methods.
- The final hierarchical structure identified 8 spectra (e.g., Externalizing, Internalizing) and 27 subfactors, diverging from DSM disorders like Major Depressive Disorder (MDD).
- Classic DSM disorders such as MDD, GAD, and PTSD did not emerge as distinct clusters, instead dissolving into smaller, homogeneous syndromes or merging into higher-order spectra.
- The findings suggest that DSM criteria index heterogeneous symptom subsets rather than fixed disorders, highlighting limitations in current diagnostic frameworks.
- Limitations include reliance on self-reports, decontextualized symptoms, and a 12-month assessment window, calling for future multi-method and diverse sample studies.