[Occupational mortality in the population employed at 1991 Census].

Background: Despite limitations and problems connected to occupational surveillance systems based on mortality data, mortality from specific causes continues to be a crucial indicator for evaluating the differences in health among various occupations.

Objectives: To evaluate the potential of a surveillance system of occupational mortality based on census and mortality data obtained from ISTAT (Italian Central Statistics Institute).

Methods: By means of record-linkage between Census data and death records from ISTAT, occupational mortality was assessed during the twelve month period following the 1991 census, limited to subjects aged 18-64 years for whom occupational information was available. The study population consisted of deceased subjects, 19,527 of whom were men and 3,547 were women. A cross-sectional model was used to evaluate odds ratios for cause-specific mortality. The risk estimates were then compared to the results obtained in a previous analysis on Census data gathered in the 1981 census.

Results: Among the significantly increased risks observed both in 1981 and 1991, of particular interest are the excesses of mortality from liver cirrhosis in caretakers, janitors, and cleaning staff; and the excess of breast cancer in teachers. Mortality from cirrhosis in masons, porters, and waiters, and from violent causes in transport workers and bar and restaurant workers, was instead significantly in excess only in 1991.

Conclusions: The data on mortality by occupation does not permit inferences about occupational risks, though they can be useful to highlight socio-economic differences in premature death in the employed segment of the population and to draw attention to possible trends over time.