Objective The aim of this study was to test the extent to which physical activity performed during work and leisure is associated with systemic inflammation. Methods Data regarding job history and high-sensitivity C reactive protein (hs-CRP) levels, as well as potential confounders, came from the Copenhagen Aging and Midlife Biobank. The participants' self-reported job history was combined with a job exposure matrix to give a more valid assessment of cumulated occupational physical activity compared with conventional self-reported activity. Occupational physical activity was measured as cumulative ton-years (lifting 1000 kg each day for a year). Current leisure time physical activity was self-reported into four different categories. We analysed the association between occupational physical activity, current leisure time physical activity and hs-CRP level in a multivariable linear regression model with adjustment for age, sex, smoking history, number of chronic diseases, body mass index and alcohol. Results In unadjusted analysis, higher occupational physical activity was associated with increased hs-CRP levels, while higher leisure time physical activity was associated with lower hs-CRP levels. In adjusted analysis, lower leisure time physical activity resulted in 12% higher hs-CRP levels while higher occupational physical activities showed a 6% increase in hs-CRP. When we analysed occupational and leisure time physical activity as continuous variables, only leisure time physical activity affected hs-CRP. Conclusion This study indicates that the relationship between physical activity and hs-CRP depends on the setting of physical activity, with lower hs-CRP related to leisure time physical activity and higher hs-CRP related to occupational physical activity. The results suggest that systemic inflammation may explain the physical activity paradox.