Depression symptoms mediate the association between workplace stress and interleukin 6 in women, but not men: The Whitehall II study.

Workplace stress and depression are positively related with inflammation, and each other. Low-grade inflammation and concurrent high levels of workplace stress or depression has been related with future morbidity. The potential pathway between constructs however, remains elusive. For the first time, this study explored the concurrent relationship between workplace stress, depressive symptomology and low-grade inflammation, and considered the role of gender in these relationships. Data from the Whitehall II cohort study ( ​= ​2528, M ​= ​57.01, 23.7% females) provided measures of workplace stress (job demand-control; JDC), depressive symptomology (Centre for Epidemiological Studies Depression scale; CES-D) and circulating inflammatory markers, interleukin-6 (IL-6) and C-reactive protein (CRP) collected on the same day from a single time point. Females had higher workplace stress, depressive symptoms and lower serum IL-6 concentrations. For males, higher workplace stress was associated with higher depressive symptoms. For females, higher depressive symptoms were related with elevated IL-6 levels, and both higher workplace stress and IL-6 levels were associated with higher depressive symptoms. Higher depressive symptoms were related with higher CRP levels in men only. Higher depressive symptoms statistically mediated the relationship between higher workplace stress and IL-6 levels in females only,  ​= ​0.016, CI [0.002, 0.039]. Females in this large cohort had higher levels of job strain, depression and lower IL-6 concentrations than males. In females, higher depressive symptoms were associated with higher serum IL-6 levels and workplace stress was not. Considered together, these findings suggest that low job control may be more apparent in females than males, but it is primarily negative affect that drives the positive relationship between work stress and serum IL-6 concentrations in females. Replicating the current design with a suitably proximal follow-up is required to determine if the associations identified are causal.