Objective Introducing the job embedded scale to explore the dropout intention of medical workers at tertiary hospitals.Methods Correlation analysis, analysis of variance and regression analysis were used to analyze with a job embedded scale, the dropout intention of medical workers of a Beijing hospital.Results There was a significant negative correlation between job embedding and dropout intention of such people(r=-0.223).The predictive ability of on-the-job embeddedness is better than off-the-job embeddedness.Fit-organizational, sacrifice-organizational, fit-community and sacrifice-community were significantly negatively correlated with dropout intention.There are differences in the relationship between job embedding and dropout intention of medical workers of different individual attributes.Conclusion Job embedded scale can effectively predict the dropout intention of medical workers, offering a usable tool to predict the dropout of these people.