Examining the latent structure of ICD-11 depression symptoms

ICD-11 presents updated descriptions for single episode and recurrent depressive disorder. A depressive episode (DE) is integral to these disorders and contrasting descriptions of the structure of DE symptoms are presented in ICD-11 and the Clinical Descriptions and Diagnostic Requirements for ICD-11 (CDDR). The former presents symptoms as unidimensional while the latter suggests symptoms are multidimensional reflecting affective, cognitive-behavioural, and vegetative clusters. This study examined if the latent structure of the DE symptoms is better represented by a one- or three-factor model. Data were collected from general population samples from the United Kingdom (UK: N = 975) and Ukraine ( N = 2050). DE symptoms were measured using the International Depression Questionnaire, and the latent structure of the DE symptoms was assessed using confirmatory factor analysis. The three-factor model of DE symptoms outlined in the CDDR performed substantially better than the one-factor model and was a close fit to the UK and Ukrainian data. This model was invariant for sex in each nation. Prevalence estimates based on a three-factor model were similar to those based on the ICD-11 requirements, and similarly sized sex differences were observed. The multidimensional description of the DE symptoms outlined in the CDDR better represented data from two culturally distinct samples. Discrepancies in how DE symptoms are described in ICD-11 and the CDDR should be resolved to facilitate future research efforts. • ICD-11 depression symptoms were best explained by a multidimensional model. • Symptoms reflect affective, cognitive-behavioural, and neurovegetative clusters. • This model was a close approximation of sample UK and Ukraine sample data. • This model was operated equivalently for males and females in each sample. • Prevalence rates based on a three-factor model were similar to those based on the ICD-11 rules.