This paper analyses the causal effects of temporary employment on work-related stress and mental health before (2006/07) and during the economic crisis (2011/12) and examines whether the economic recession worsened these two health outcomes. To control for selection bias, propensity scores (PS) are computed separately for men and women using microdata from two cross-sectional surveys, considering temporary (treatment group) versus permanent employment (control group). Next, we use difference-in-differences estimators stratifying by age, education level, and regional unemployment differences using PS as weights. Our results indicate that a male salaried worker with a temporary labour contract tends to have lower levels of work-related stress in the pre-crisis period, but not for women. The stratification analysis shows lower work-related stress levels among older male adults, workers with a high education level, and employees in regions with high unemployment rates. The economic crisis is responsible for increasing stress only among older temporary workers and male university graduates, without affecting women. We also see evidence of a positive link between temporary employment and poor mental health in both periods, although only for men. We neither find significant impacts for our sample of men or women, nor for most of our population subgroups with the exception male workers with a university degree.