•This study examines the association of patterns of precarious employment with self-rated health among US workers.•Latent class analysis identifies four distinct classes of precarious employment experienced by workers.•Logistic regression analysis shows that there are differences in perceived health across precarious employment classes. Although the prevalence of precarious employment has been on the rise due to structural changes in the global labor market, there is still lack of a clear understanding of whether precarious employment is a social determinant of health. Data from the 2006–2010–2014 General Social Survey (N=5,411) were used to examine the relationships between patterns of precarious employment and perceived health among US workers. Based on a multidimensional construct of precarious employment, latent class analysis identified four differential patterns of precarious employment experienced by workers: (1) the most precarious group, (2) low precarious with middle income group, (3) low precarious with high income group, and (4) mixed precarious group. I then conducted a multinomial logistic regression and found that socio-demographic characteristics, such as gender, race/ethnicity, and education, were significantly associated with precarious employment class membership. Finally, a logistic regression analysis showed that there were significant differences in perceived health status across precarious employment classes, controlling for individual background characteristics. Study findings highlight the heterogeneity and various patterns of precarious employment experiences and indicate a need for the use of a multidimensional construct of employment precariousness in determining its health impact on the working population.