The predictive validity and outcome of ICD-10 and DSM-5 short-lived psychotic disorders: a review and meta-analysis

The ICD-10 Classification of Mental and Behavioural Disorders introduced the category of 'acute and transient psychotic disorders' (ATPDs) encompassing polymorphic, schizophrenic and predominantly delusional subtypes, and the forthcoming ICD-11 revision has restricted it to polymorphic psychotic disorder, while the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) listed 'brief psychotic disorder' (BPD). To assess the predictive validity and outcome of ATPDs and BPD, relevant papers in English, French and German were searched in PubMed and Web of Science. Where possible meta-analysis of prognostic validators (diagnostic stability, course, outcome and response to treatment) was conducted. Fifty studies published between January 1993 and July 2019 were found. The clinical and functional outcome of ATPDs proved better than in schizophrenia and related disorders, but mortality risk is high, particularly suicide, and treatment trials provide little evidence. Meta-analysis of 25 studies (13,507 cases) revealed that 55% (95% CI 49-62) do not change diagnosis, 25% (95% CI 20-31) converted into schizophrenia and related disorders, and 12% (95% CI 7-16) into affective disorders on average over 6.3 years. Subgroup meta-analysis estimated prospective consistency of polymorphic psychotic disorder (55%; 95% CI 52-58) significantly greater than that of the ATPD subtypes with schizophrenic (OR 1.7; 95% CI 1.4-2.0) and predominantly delusional (OR 1.3; 95% CI 1.1-1.5) symptoms. Moreover, the diagnostic stability of BPD (13 studies; 294 cases) was 45% (95% CI 32-50) over a mean 4.2 years. Although these findings indicate that short-lived psychotic disorders have little predictive validity, significant differences among the ATPD subtypes support the revised ICD-11 ATPD category.