| dc.description.abstract | Background: The extant literature indicates that the distinction between affective and non-affective psychotic symptoms is arbitrary. Emerging evidence indicates that psychosis may be best represented as a multidimensional construct comprising a General dimension that captures the covariance shared across all affective and non-affective symptoms, and five specific dimensions that capture the unique shared variance within positive, negative, depression, manic, and disorganized symptoms. In addition, sustained and repeated childhood trauma is a risk factor for psychosis and Complex PTSD (CPTSD). However, there is limited research on (1) the latent structure of psychosis in general population samples, (2) the reliability and replicability of these dimensions, (3) the associations between these dimensions and established risk factors of psychotic illness, (4) what psychosocial variables might mediate the association between childhood interpersonal trauma and these dimensions of psychosis, and (5) if psychosis and CPTSD are related constructs.
Objectives: (1) To determine the appropriate latent structure of psychosis in the general population; (2) To assess the reliability and replicability of each dimension of the best fitting model of the latent structure of psychosis; (3) To assess the association between each dimension of psychosis and a range of established risk factors for psychotic illness; (4) To assess if the association between childhood interpersonal trauma and each dimension of psychosis is mediated by multiple psychosocial variables; (5) To determine if CPTSD and psychosis are related constructs.
Methods: Empirical assessments were conducted in three phases. Research objectives 1-3 were assessed in phase one and were based on a nationally representative sample of the adult population of the United States (N = 36, 309). Confirmatory factor analysis and confirmatory bifactor modelling were used to test a series of alternative models of the latent structure of psychosis. Bifactor strength indices were used to assess the reliability and replicability of each dimension of the most appropriate measurement model of psychosis. Finally, structural equation modelling (SEM) was used to test the associations between environmental, developmental, demographic, social, and psychological variables and each dimension of psychosis. Research objective 4 was assessed in phase two and was based on the same sample as used in phase one. SEM was used to test if the associations between childhood interpersonal trauma and each dimension of psychosis was mediated via multiple psychosocial variables. The findings from phases one and two were used to inform the final phase of the study which was based on a sample of former residents of institutional care facilities (N = 45). The fifth objective used independent samples and chi-square tests to examine the association between psychosis and CPTSD.
Results: The latent structure of psychosis was best represented by a bifactor model which included a General dimension (e.g. affective + non-affective symptoms) and four specific dimensions reflecting positive, negative, manic, and disorganized symptoms. The General and Negative dimensions demonstrated adequate reliability and replicability. The General dimension was correlated with several established risk factors for psychotic illness including childhood trauma, diminished social support, history of attempted suicide, younger age, being male, ethnic majority status, and lower socio-economic status. Multiple mediating effects were identified in the association between childhood interpersonal trauma and the General dimension of psychosis. Posttraumatic stress disorder (PTSD), diminished social support, and history of attempted suicide mediated the association between childhood interpersonal trauma and the General dimension of psychosis. There were also a number of mediating effects from childhood interpersonal trauma and the specific dimensions. Finally, psychosis and CPTSD were found not to be associated with one another in the sample of former residents of institutional care facilities.
Conclusion: These findings represent an important addition to the existing psychosis literature by addressing several gaps in knowledge. Results indicate that psychosis is best understood as a dimensional construct in the general population, and there exists a general vulnerability to affective and non-affective symptoms. The internal reliability, replicability, and external validity of the General dimension of psychosis was supported. In addition, the association between childhood interpersonal trauma and this General dimension of psychosis was found to be influenced by social support networks, PTSD symptomology, and a history of attempted suicide. These finding provide clinicians with evidence for how best to conceptualise psychotic symptoms, and targets for treatment in highly vulnerable groups. Finally, the independent nature of psychosis and CPTSD in former residents of institutional care facilities suggests that while both forms of psychological distress are common in this cohort, distinct clinical interventions may be required to address these problems. | en |
