A rich literature has studied the effect of job insecurity on health. However the causal link between these two variables remains unclear. We study the relationship between perceived job insecurity and health using longitudinal data on around 30 thousand older workers from 20 European countries covering a period of 14 years. The unprecedented size and nature of the dataset compared to previous studies on job insecurity and health allows us to apply different estimation methods and compare the results obtained. We do so using a wide range of health outcomes that include objective and subjective measures. Using pooled OLS, we estimate a strong association between job insecurity and health outcomes. A fixed effect estimator yields precisely estimated zeros with the exception of a few mental conditions. Additionally, we test the robustness of an IV strategy that uses an index for employment protection legislation (EPL) as an instrument for job insecurity. We conclude that the direct causal link between job insecurity and health for older workers is in any case rather weak and discuss several reasons for our findings.