Empirical validation of competing definitions of schizophrenia: a poly-diagnostic study of cognitive impairment in non-affective psychosis

Abstract

Background: The empirical validation of diagnostic criteria for schizophrenia remains a controversial issue within psychiatry and allied sciences. Most diagnostic criteria are still influenced to a large extent by historical and consensus-based perspectives.

Methods: A poly-diagnostic approach including a set of 23 operationalized diagnostic criteria were administered to probands with non-affective psychosis (n=169). In addition, participants completed a neuropsychological battery during a stable phase of the illness. Attentional, verbal and visual memory and executive functions were assessed. The control group was composed of 26 demographically matched healthy subjects. Analysis of variance was conducted taking neuropsychological performance as response variables and the 23 binary diagnostic systems as explanatory variables.

Results: Four out of the 23 operationalized diagnostic systems for schizophrenia (Feighner, French, Kraepelin and Langfeldt criteria) demonstrated high empirical validity for memory and executive functions scores (medium to moderate effect sizes). These 4 systems resemble classic nosological approaches based upon the 'outcome principle' concept, which suggests that schizophrenia leads to deterioration. However, diagnostic effectiveness of neuropsychological tests for the 23 operationalized diagnostic systems of schizophrenia was low (likelihood ratio <2).

Conclusions: Neuropsychological functioning provides empirical validation to operationalized definitions of schizophrenia which are mainly based upon deterioration. It is suggested that some inconsistency of neurobiological studies in schizophrenia might result for using solely current consensus-based diagnostic systems. The implementation of poly-diagnostic strategies could contribute to improve the validity of the schizophrenia construct.