O8D.5\u2005Physical activity as a moderator of effort-reward imbalance over time – results of the german lida-cohort study

Abstract

Background Work-related stress is widespread in modern societies and an important risk factor for common health impairments. The model of effort-reward imbalance (=ERI), one of the best-investigated occupational stress models, is associated with musculoskeletal and coronary heart diseases as well as depressive disorders. A common assumption is that physical activity (PA) antagonizes the harmful effects of distress on health. Yet, the evidence of the effect of PA on work-related stress, especially on ERI, is limited so far. The aim of this investigation was to analyze the effect of PA on ERI over time. Methods Work-related stress was measured by ERI and physical activity by a single question asking for the weekly frequency of physical activity leading to sweating or getting out of breath over two study waves (t1, t2). Multiple linear regression analysis adjusting for sociodemographic factors (age, sex, education) was performed including interaction testing. Results Work-related stress at t1 was a significant predictor for work-related stress at t2. Physical activity at t2 - itself not significantly associated with work-related stress - had a significant moderating effect on work-related stress over two study waves. The higher the frequency of PA was, the lower the stress level at t2 in comparison to t1. This interaction of physical activity with ERI was also observable after adjustment for sociodemographic factors. Discussion We found a moderating effect of physical activity on ERI over time in agreement with former studies using other measures of work-related stress. A dose-response relationship of physical activity and ERI over two study waves was observable in our study. Regarding the detrimental effect of chronical stress on health, PA might be beneficial by mitigating the health hazards of ERI and their long-term consequences (e.g., loss of work force). Longitudinal studies over more than two waves have to prove our findings.