Emotional demands and exhaustion: cross-sectional and longitudinal associations in a cohort of Danish public sector employees Abstract Purpose To examine cross-sectional and longitudinal associations between content-related emotional demands at work and exhaustion, and to investigate if these associations were modified by other psychosocial work characteristics. Methods In 2007, 4489 Danish public service employees participated in the PRISME study by completing postal questionnaires, and 3224 participated in the follow-up in 2009. Content-related emotional demands were measured by a scale (scored 1 to 5) based on five work-content-related items, and exhaustion was measured with the general exhaustion scale from the Copenhagen Burnout Inventory (CBI) (scored 1 to 5). The cross-sectional and longitudinal associations with exhaustion were analysed in the same model and adjusted for effects of potential confounders. Effect modifications were examined separately for self-reported emotional enrichment, meaningful work, job control, social support at work and quantitative demands. Results Exhaustion increased with increasing emotional demands, both cross-sectionally and longitudinally. However, although statistically significant, the effect was small. In the longitudinal analysis, a one unit increase in emotional demands was associated with a 0.03 unit (95% CI: 0.01–0.06) increase in exhaustion. We found statistically significant effect modification for three of six potentially modifying work characteristics. The effect of emotional demands on exhaustion was lower for participants with high levels of emotional enrichment (cross-sectionally and longitudinally), high levels of meaningful work (longitudinally), and higher for high levels of quantitative demands (cross-sectionally). Conclusions Increasing content-related emotional demands were associated with increasing levels of exhaustion, cross-sectionally and longitudinally. This effect was reduced if the work was experienced as emotionally enriching and meaningful.