Background: Precarious employment has been associated with poor health, but the potential mechanisms are unclear. We examined the relationships between precarious employment and health, and investigated psychosocial working conditions as potential mediators. Methods: A cross-sectional population-based survey was conducted in South Australia in 2009 (N=1,016 employed). SF-12 measures of mental and physical health were modeled using logistic regression in relation to employment arrangement, controlling for socio-demographics, years in job and psychosocial working conditions. Results: There was no association between casual full-time or part-time employment and poor mental health in multivariate analyses. Conversely, there was a significant association between casual full-time employment and poor physical health (compared to permanent full-time workers, OR=3.14, 95% CI 1.26-7.85). The association with physical health was unaffected by adjustment for psychosocial working conditions. Conclusions: Casual full-time employment was strongly associated with poor physical health but not with poor mental health. This association was not mediated by the psychosocial working conditions measured in this study, but may be related to other (unmeasured) working conditions.