Validity of current somatoform disorder diagnoses: perspectives for classification in DSM-V and ICD-11

Abstract

Background: The impending revisions of DSM-IV and ICD-10 provide an excellent opportunity to improve the diagnostic accuracy of the current somatoform disorder classification. To prepare for these revisions, this study systematically investigates the validity of the current classification of somatoform disorders.

Methods: We searched Medline, Psycinfo and reference lists to investigate convergent, divergent, criterion and predictive validity of the current somatoform disorder classification.

Results: Substantial associations of somatoform disorders with functional impairment and elevated health care costs give evidence for the clinical and societal importance of somatoform disorders and for the convergent validity of the current operationalization. The specificity of the current somatoform disorder classification, i.e. their divergent validity, is demonstrated by the fact that functional somatic syndromes and their consequences are only partially explained by association with anxiety and depression. However, the imprecision of the diagnostic criteria, which are not based on positive criteria but on the exclusion of organic disease, largely limits the criterion validity of the current classification systems. Finally, studies investigating the predictive potential of somatoform disorders are lacking, and to date predictive validity has to be considered as low.

Conclusions: The insufficient criterion and predictive validity of the present somatoform classification underlines the need to revise the diagnostic criteria. However, an abolishment of the whole category of somatoform disorders would ignore the substantial convergent and divergent validity of the current classification and would exclude patients with somatoform symptoms from the current health care system. A careful revision of the current somatoform disorder diagnoses, based on positive criteria of psychological, biological and social features, has the potential to substantially improve the reproducibility and clinical utility of the existing classification system.