Gesundheit am Arbeitsplatz: Meta-Analysen zum Zusammenhang von Arbeitsmerkmalen und Depression

On the one hand depressive disorders lead to substantial impairment in the life of the people concerned. On the other hand they lead to tremendous costs for businesses and society. The Global Burden of Disease Study (WHO, 2008) predicts that depressive disorders will be the second-leading cause of disease burden in 2020 (Murray & Lopez, 1997) and the leading cause by 2030 (WHO, 2008). While there has been a growing number on primary research on depression in the last two decades, reviews are still rare. A few reviews exist that analyze the relationships of specific work characteristics and depression (Tennant, 2001; Bonde, 2008; Netterstrøm et al., 2008; Siegrist, 2008; van der Doef & Maes, 1999). A first meta-analysis analyzes job characteristics and their relationship to the wider conception of common-mental-disorders (Stansfeld & Candy, 2006). A comprehensive meta-analysis that relates job characteristics at the workplace to depression is not present. This present work evaluates if specific work characteristics exist that possess relationships to depression and might be acting as promoters for developing a depression. This was done against the background of potential prevention at the workplace. The aim was to conduct meta-analyses that analyze the relationship of job characteristics and depression. To do so, present primary studies were integrated with meta-analytical techniques. Furthermore the concepts were evaluated in order to gain theoretical information. Two publications were composed. As being the first meta-analysis in this area of research, publication 1 investigated the relationship of role stress and depression. This incorporated two variables, role ambiguity and role conflict according to Kahn et al. (1964). Results showed positive relationships for both variables, showing slightly higher correlation coefficients for role conflict. When correcting the results for the influence of the other construct, coefficients remained in considerable amount. Further, the independence of the constructs was meta-analytically assessed. Results revealed a common component of the variables, which could arise from the basic concept. Despite this assumption, theoretical considerations support the emergence from the work environment. A statistically significant moderator was found that showed a moderating influence of the geographic region where the data was conducted. These findings agree with primary research of Hofstede (1994), who identified six dimensions that vary in cultural groups. Abstract 127 The second publication includes two research questions that differ in their content-related and methodological approach. First, the relationship of the variables included in the job-demand-control-model (Karasek, 1979; Karasek & Theorell, 1990) and depression was meta-analytically investigated. Second, the appropriateness of the job-demands-measurement was assessed. The meta-analytical results showed small to moderate, but significant, findings. The highest relationship was found for iso-strain (high job demands / low control/ isolated working environments). Detailed Analysis showed that the relationship of iso-strain and depression is the highest in female samples. Statistical variables that moderated the relationship were extracted. In conclusion the second research question have shown that previous measurements need to be revised in order to adequately assess todays working environments. Thus, they supplied practical implications. It became apparent that the answer pattern of employees has not changed in the last two decades. Nevertheless, the relationship of job-demands and depression increased. The results suggest that previous measurements do not represent the increasing concentration of the work environment (i.e. rising qualitative and quantitative job-demands that have to be fulfilled by less employees).