A cohort study is the most important study design for analyzing industrial exposures. One of the sources of bias in industrial cohort studies is healthy worker effect (HWE). HWE consists of two selection processes: healthy worker hire and healthy worker survivor. In this article, methods of reducing HWE presented in the literature are compared and its limitations are discussed. The best method of reducing healthy worker hire effect is an internal reference group. But in the case of time related exposures (cumulative exposure) time from hire have to be controlled in the regression model. To minimize healthy worker survival effect internal reference group should be used and time from hire should be controlled.