PurposeWe aim to extend current knowledge on associations between stressful work and sickness absence, first, by studying associations between ERI and sickness absence among full-time employees from various occupations, and second, by investigating if associations vary by age.MethodsWe use data from four waves of the German socio-economic panel (GSOEP), collected among men and women between 2006 and 2012, with 9418 observations. Stressful work is measured with a short form of the ERI questionnaire. We investigate an imbalance between effort and reward (ER ratio) as well as the two main components (“high effort” and “low reward”). Sickness absence is measured by self-reported number of sickness days (assessed the following year). After descriptive analyses, we estimate a series of multivariable regressions, including tests for interactions between age and work stress.ResultsEach of the three indicators of stressful work is related to higher number of sickness days, with except of “high effort” in case of men. Findings remain significant after adjusting for social position (income, education and occupational class) and health. In addition, for both men and women, associations were slightly higher among older workers, though interactions did not reach statistical significance.ConclusionOur findings support that stressful work is linked to sickness absence across a wide spectrum of jobs with varying incomes and educational levels, and also that associations are slightly more pronounced among older workers.Electronic supplementary materialThe online version of this article (10.1007/s00420-018-1298-3) contains supplementary material, which is available to authorized users.