Rearrange the order of the following paragraphs to make a coherent and effective discussion section. Discussion A. The strength of the present analysis is the use of valid and reliable psychometric measurements for data collection, especially the measurement of anxiety and depression. Furthermore, the large sample size and the recruitment from a general population in selected urban areas strengthen the analysis. Additionally, SES was available comprehensively from all participants, with only a few missing values. B. The analysis is limited, however, because of pooling information from two temporally different surveys. Socioeconomic characteristics were collected from Phase 1 and all other data were taken from Phase 2, conducted 2 years later from the same participants. SES was not resampled in Phase 2 because the data were already available and it is assumed that SES is stable over 2 years. Therefore, we do not think that hereby, the results were influenced strongly. Additionally, the QUEBEB study was not designed for the present analysis and therefore other sleep-influencing characteristics, such as noise exposure, individual physical activity, nutrition or substance abuse, were not included. Finally, mainly urban populations were selected. Therefore, the results focus on urban populations but, nevertheless, allow an interpretation of sleep quality and its dependencies in general. C. The present analysis examined the association of SES with sleep quality adjusted for a wide range of socio-demographical, physical and psychological characteristics among a sample of German adults living in urban areas. The regression analysis demonstrated that higher SES predicted good sleep quality. Other previous studies that have found higher socioeconomic factors to be associated with better sleep quality support the presented research. In addition, Soltani et al. and Mezick et al. showed associations between low socioeconomic factors and high PSQI scores (i.e. low sleep quality). D. A strong association between mental health and sleep is also reported by several studies; an association between insomnia symptoms and poor mental health was observed in a Japanese study. Another analysis showed that 77% of study participants with current depression and 45% of participants with a history of depression reported poor sleep quality on the PSQI in comparison with 15% of control subjects. This result explains the overlapping effect of sleep quality and anxiety/depression. In contrast, a study of twins and siblings reported that the correlations between sleep problems measured by the PSQI and symptoms of anxiety and depression were only moderate and that other factors took a more important role. E. The association in the present analysis persisted but attenuated after adjusting for other factors, especially for mental and health status. The adjusted OR for SES on sleep quality decreases, if named factors were included in the binary logistic regression model. The strongest confounding effect is caused by anxiety and depression as additional variables in the regression model: anxiety and depression, as well as physical diseases, are more frequent in persons with a low SES than in those with a medium or high SES. Simple and multiple regression analyses showed a partial mediating effect for anxiety and depression in the relationship between SES and sleep quality. Therefore, mental disturbances, but also health status, are important factors in explaining sleep quality. F. Regarding physical health, poor sleep efficiency increases and longer sleep latency even almost doubles the risk of mortality in healthy elders. Similarly, Furihata et al. found an association between poor sleep quality and physical health. Furthermore, with increasing comorbidities, sleep quality decreases and the presence of depression rises. G. In addition, the logistic regression identified further determinants affecting sleep. According to the literature, a lower age and the male gender were factors associated with good sleep quality. We could also show that married people and people living in more suburban areas have better sleep quality than single or widowed persons or persons living in urban areas. H. For measuring SES, the Winkler Index was chosen. The index takes into account the different school graduations of the former Federal Republic of Germany and the German Democratic Republic. It is also aimed at addressing the complexity of the training system and all relevant German occupational titles. The formation of an index is constituted by the complexity and multidimensionality of social inequality structure, which is represented inadequately by single indicators. 1._______ 2._______ 3. _______ 4._______ 5._______ 6._______ 7. _______ 8._______