WORK STRESS IN THE CONTEXT OF SHIFTWORK: A LONGITUDINAL INVESTIGATION

Abstract

Although prior research has documented the harmful effects of shiftwork on health and well-being, the relationship between shiftwork and other workplace stressors remains unclear. The current study addresses this gap by using longitudinal data from three cycles of the National Population Health Survey to explore 1) the direct effects of shiftwork and JDCSM variables (job demands, job control, social support, and mastery) on health and well-being, 2) whether shiftwork moderates the effects of JDCSM variables, and 3) whether the effects of shiftwork are mediated by these variables. In addition, “reverse” causal relationships between four measures of strain and workplace stressors are investigated. The results suggest that shiftwork is associated with a reduction in job satisfaction over two cycles. However, there was no evidence of a relationship between shiftwork and changes in chronic health conditions, psychological distress, or health-related quality of life. Consistent with the JDCSM model, increased job demands predicted more distress and less job satisfaction, while greater mastery predicted less distress and more job satisfaction. In addition, job control and social support predicted improved health-related quality of life. The associations were more consistent with the thesis’s predictions across two cycles than across three. There was only weak support for the hypotheses that shiftwork moderates the effects of JDCSM variables. However, there was evidence that changes in job control and social support mediate the effects of shiftwork on job satisfaction. There was also support for reverse causal effects between various outcomes and workplace stressors. However, the prediction that shiftwork moderates these reverse effects received only weak and inconsistent support. Exploratory analyses found specific JDCSM and strain variables to predict an individual’s continuation in shiftwork over time. The theoretical and practical implications of these findings are discussed.