Previous research indicates an increase in the prevalence of depression around adolescence, especially for females. Research suggests depressogenic cognitions play an essential role in the development of depression and may mediate the relation between risk factors and depression. Research has also shown the family environment, negative life events, and maternal depression are all related to the development of depressogenic cognitions. Additionally, few studies have tested models of depression while measuring both anxiety and depressive symptoms despite the high rates of comorbidity between the two disorders. The current study used path analytic techniques to integrate correlates of depression while accounting for comorbid anxiety symptoms in comprehensive model of depression for early adolescent girls. Participants included 203 girls, aged 9-14, along with their mothers. Participants completed self-report measures of the family environment, cognitive triad, and negative life events. Mothers of participants completed a self-report measure of psychopathology. Participants also completed a semi-structured diagnostic interview, which served as the measure for symptoms of depression and anxiety. Results supported previous literature finding a more depressogenic cognitive triad was significantly associated with higher depressive severity. Family environments, characterized by more cohesive and less conflictual family relationships, more communication, and higher engagement in social/recreational activities, were significantly associated with a more positive cognitive triad. Additionally, more negative life events were significantly associated with a more depressogenic cognitive triad. Both family social/recreational activities and negative life events had significant indirect effects on depression. Results indicated a strong relation between anxiety and depression, with anxiety having a significant positive direct effect on depression. The pathways from maternal depression and anxiety to the cognitive triad, anxiety symptoms to the cognitive triad, as well as family environment variables, maternal depression and anxiety and negative life events to anxiety symptoms were not found to be significant. Results from an exploratory analysis suggest anxiety may moderate the relation between the cognitive triad and depression. Implications of these results, limitations, and recommendations for future research are provided. / text

Previous research has suggested that depressive disorders are common in youth and are associated with many negative outcomes. As a result, understanding how to treat depression effectively is very important. It is unclear; however, what factors predict treatment success or failure for depressed youth. Researchers are starting to investigate whether comorbid anxiety is a possible moderator of treatment outcome for youth with depression. Studies of the relation between comorbid anxiety and treatment outcome have produced mixed findings and have almost exclusively focused on older depressed adolescents. There is also limited research exploring whether parent intervention moderates the effect of comorbid anxiety on treatment outcome in depressed youth. This study focused on investigating the relation between comorbid anxiety and treatment outcome in a sample of 84 depressed female early adolescents who received either group cognitive behavioral therapy (CBT) or group CBT plus a parent intervention. The addition of parent intervention was explored as a moderator of the relation between anxiety and treatment outcome. Treatment outcome was measured by changes in depression severity and global functioning during treatment. The depression severity and global functioning scores of depressed girls with comorbid anxiety were also compared to depressed girls without comorbid anxiety prior to treatment to determine whether the first group of girls entered treatment with a different level of psychopathology. Participants and their primary caregivers were administered a semi-structured diagnostic interview which was used as a measure of depression severity, global functioning, anxiety severity, and to determine whether participants met diagnostic criteria for depressive and anxiety diagnoses. The results of this study suggested that depressed youth with comorbid anxiety or higher anxiety severity started out treatment with higher depression severity and lower functioning. Results also suggested that comorbid anxiety was not related to negative treatment outcome and that youth with comorbid anxiety actually experienced larger reductions in depression severity over the course of treatment than youth without comorbid anxiety. Parent intervention did not significantly moderate the effect of comorbid anxiety on treatment outcome. The study’s limitations, implications of the results, and recommendations for future research were discussed. / text

My thesis aims at identifying both genetic and environmental causes of major depressive disorder (MDD), using a large case-control study: 6,000 Chinese women with recurrent MDD and 6,000 controls. One of the major challenges for conducting genetic research on MDD is disease heterogeneity. The first question addressed is how different MDD is from highly comorbid anxiety disorders. I examine how anxiety disorders predict clinical features of depression and the degree of heterogeneity in their predictive pattern. The second question addressed is whether clinically defined MDD is a single disorder, or whether it consists of multiple subtypes. Results are then compared with and interpreted in the context of Western studies. Furthermore, latent class analysis and factor analysis results are also used in association analysis to explore more genetically homogeneous subtypes. Genetic data were derived using a novel strategy, low pass whole genome sequence analysis. Using genotypes imputed from the sequence data, I show that a cluster of single nucleotide polymorphisms (SNPs) is significantly associated with a binary disease phenotype including only cases with = 4 episodes of MDD, suggesting that recurrence might be an indication of genetic predisposition. The third issue examined is the contribution of rare variants to disease susceptibility. Again using sparse sequence data, I identified exonic sequence variants and performed gene-based analysis by comparing the number of variants between cases and controls in every gene. Furthermore I performed gene enrichment test by combining P values of SNP association tests at different minor allele frequency ranges. Overall, I did not find convincing evidence that rare variants aggregately contribute to disease susceptibility. However, the gene-based analysis resulted in an unexpected finding: cases have an excess of variants in all thirteen-protein coding mitochondrial genes, which was due to copy number differences in the mitochondrial genome. Both human phenotypic data as well as mice experimental data show that the increase in the mitochondrial copy number in cases is due to chronic stress.