Long-term mortality of oil refinery workers. II. Comparison of the experience of active, terminated and retired workers.

In occupational epidemiology a retrospective cohort study normally includes active, terminated, and retired employees and the mortality results may vary considerably if any of the three groups is excluded from the study. From a large refinery cohort of 12,526 white male workers followed between January, 1937 and January, 1978, the mortality experience of three groups (the active, terminated and retired) has been examined; detailed results, along with the merits and problems of studying these groups separately, are presented. The standardized mortality ratios (SMRs) for all causes are 0.68, 1.04 and 0.89 for the active, the terminated and the retired, respectively, and for all cancer, 0.85, 0.98 and 1.05. Significantly decreased SMRs are seen for most of the causes among the active and may be attributed to the "healthy worker effect." Exclusive study of active workers, although it may yield certain useful information, particularly on diseases of the young and those with short latency periods, is primarily a study of the healthy worker effect. Many favorable effects of the active worker will be encountered. The retirees as a whole experienced no significant excess mortality for any causes, although examination of a subgroup, the early retirees, did reveal a significant excess of deaths from diseases of the nervous system and sense organs. The retired may appear to be an ideal group for study because they usually have worked for an extended period of time, they may have experienced long-term occupational exposure and they have lived long enough to develop diseases with long latency periods; however, serious problems arise from studying only the retirees and these are discussed. The terminated group contributed 41% of the person-years, 49% of the total number of individuals and 38% of the deaths and is far too important to be omitted. Contrary to previous reports, the terminated did not demonstrate a significantly adverse mortality experience when compared with the general population, although they did not show the healthy worker effect that was seen among the active.