Mortality and work conditions: a retrospective follow-up assessment of the effects of work conditions on the mortality of male employees in the manufacturing industry.

A retrospective follow-up study was conducted to investigate the effects of work conditions on mortality. Mortality rates among male employees aged 40-60 years in the Japanese steel industry were calculated for separate work condition factors, such as job site location (parent company or subsidiary company), job category (white-collar or blue-collar) and work schedule (day work or shift work). Odds ratios were calculated to evaluate the independent effects of work condition factors on mortality using a logistic regression method. Death certificates and demographic statistics, 1991-1995, from the annual personnel reports of the employees which had been compiled and preserved by the company, were utilized. There were 19,642 employees in April, 1991, and 171 deaths and 76,761.7 person-years were observed. No work condition factors were associated with the mortality rates. Selection bias was speculated from the findings. Since there is a thorough medical checkup system at this company, the employees identified with serious health problems would probably already have been excluded as potential candidates for transferring to subsidiaries or for work schedule changes. These findings and the interpretation were consistent with those of other reports which have been conducted in Japanese industrial fields. Mortality was not an appropriate indicator for assessing health risks in the occupational environment, but it reflects the real status of health care management. Through careful interpretation it has been suggested that there is a possibility that past health care work conditions activity can be accessed.