Comments on an article by Flanders et al, entitled Confounding by Time Since Hire in Internal Comparisons of Cumulative Exposure in Occupational Cohort Studies (Epidemiology, Vol 4, pages 336-341, 1993; see NIOSHTIC record NIOSH-00226660) were presented. The authors argue that, in contrast to suggestions by Flanders et al, time since hire should not be treated as a positive confounder in analyzing exposure mortality trends in occupational cohort studies. Arguments in support of their position included: mortality does not increase with time since hire for actively employed workers; and the lessening of the healthy worker effect over time for the cohort as a whole is a function of increasing numbers of workers leaving work, some of whom leave due to health problems. Data from ten large occupational studies in which exposure was judged to have caused no increase in mortality were used to illustrate purported flaws in the approach of Flanders et al. The authors conclude that problems arise when considering time since hire as a positive confounder; nonetheless, it is important to consider potential interactions with time since hire for diseases requiring a certain latency. For response by Flanders et al see NIOSHTIC record NIOSH-00232578.