A structure-based expert model of the ICD-10 mental disorders

Area of Science:

• Psychiatry and Mental Health Classification
• Psychometrics and Data Analysis
• Clinical Psychology

Background:

• Current diagnostic systems like the International Classification of Diseases, 10th Revision (ICD-10) and the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) are primarily categorical.
• There is a growing consensus among researchers for classification systems to incorporate both categorical and dimensional aspects for a more comprehensive understanding of mental disorders.

Purpose of the Study:

• To present a novel method for classifying mental disorders that integrates both categorical and dimensional perspectives.
• To explore the structure of ICD-10 mental disorders from the viewpoint of experienced therapists.

Main Methods:

• A group of 20 therapists rated the similarity of various ICD-10 mental disorders based on their clinical experience.
• Nonmetric multidimensional scaling (NMDS) was applied to individual and mean similarity ratings.
• Cluster analysis was performed on the processed data to identify groupings of disorders.

Main Results:

• The resulting cognitive map and cluster analysis demonstrated good concordance with the established ICD-10 F groups from a categorical standpoint.
• Analogies were observed between the study's findings and existing critiques of current classification systems.
• Two primary dimensions, 'cognition' and 'psychosis', emerged from the dimensional analysis, showing significant correlations in the mean map.

Conclusions:

• The developed method successfully integrates dimensional and categorical aspects of mental disorder classification.
• This approach provides a therapist-centric model for understanding the structure of mental disorders within the ICD-10 framework.
• The findings support the development of more nuanced and expert-driven classification systems.