A latent variable approach to understanding the relationships between sexual dysfunctions and depressive and anxiety disorders

Interest in the underlying structure of mental disorders has grown in recent years. Studies concur that the high rates of comorbidity between sexual dysfunctions and depressive and anxiety disorders are related to increased chronicity and severity, and poorer quality of life. However, the direction and nature of the relationships between the disorders remains unclear,and sexual dysfunctions have not been considered in the formulation of a nosological metastructure. Accurate and empirically based models of these relationships are important precursors to effective diagnosis and treatment, but have not been considered in the literature to date. Thus, the aim of this thesis was to empirically evaluate theory-driven models of these relationships as a first step towards understanding their aetiology, course, and the most effective treatment response for clinicians. The relationships between depression, anxiety, and sexual problems were investigated using large samples and advanced statistical techniques,including structural equation modelling, latent profile analyses, and factor mixture analysis (FMA). Five separate studies were conducted: The first examined a dimensional model of the relationships, which extended the established internalising spectrum to include sexual problems. The second investigated the measurement capabilities of popular sexual function measures -to evaluate their pertinence to latent structure models- and found them lacking for the measurement of sexual desire in particular. The third compared a dimensional model with a categorical framework model to determine which conceptualisation was more appropriate for the relationships between the disorders, and found a dimensional model provided the best fit for the majority of men and women. The fourth allowed for simultaneous dimensional and categorical components in a structural model by using FMA. The models of best fit implied that that the relationships between the disorders should be recognised in our nosological systems, but separate diagnostic groups can also be maintained. The final study was a preliminary examination of the interrelationships between the latent variables over time, and found relationships consistent with a shared underlying factor that drives change in the disorders over time. Taken together, these studies advance our understanding of the relationships between these highly comorbid disorders that can have a severe impact on quality of life. Understanding these relationships is a prerequisite for effective diagnosis and treatment, and is crucial to having valid nosological systems. This body of work contributes to the literature by (i) extending our knowledge about the nature of the relationships between depression, anxiety, and sexual problems; (ii) developing preliminary structural models of these relationships that are suited to the nature of psychopathology and to our nosological systems; and (iii) by providing specific recommendations about ways these models can be incorporated into our nosology.