Phenotypic dissection of mood disorder

The pathogenesis of affective disorders is not clearly understood and the diagnostic validity of psychiatric disorders remains unclear. The aim of this thesis was to identify sub-phenotypes in affective disorders that may be biologically validated by future molecular genetic studies. The dataset comprised over 1000 subjects meeting diagnostic criteria (DSM-IV) for bipolar I disorder (BPI) or major recurrent depression (MDDR) who were previously recruited as part of ongoing molecular genetic studies. Subjects had been assessed in detail using Schedules for Clinical Assessment in Neuropsychiatry (SCAN) and case-note reviews. I undertook hypotheses testing and exploratory analyses in this dataset using a range of univariate and multivariate statistical tests. I found that a depressive episode at illness onset in BPI subjects was associated with a more depressive course of illness. I also found clinical characteristics of depression that were associated with a bipolar, rather than unipolar, course of illness. Using the HCL-32, I identified a substantial number of MDDR subjects (17%) who reported bipolar symptoms at a level similar to that reported by BPI subjects. I found significant differences in the clinical course of illness of MDDR and BPI subjects according to the lifetime presence of recurrent panic attacks, as well as clinical characteristics that appeared to be associated with the presence of panic attacks only in BPI subjects. In the unipolar sample, I found that within subjects psychotic episodes tended to be more severe than non-psychotic episodes. However, between subjects there was wide variation in severity in both those that did, and did not, experience psychotic episodes. In MDDR subjects, I found that episodes of postpartum depression clustered in families (p=0.015). I found no significant evidence for the familiality of reporting of life events in the MDDR sample. These studies identify sub-phenotypes that may be of use in future genetic studies