High risk occupations may differ from other employees by socioeconomic characteristics, which may obscure association of occupation and mortality. A considerable part of occupational differences may reflect more general socioeconomic differences. The main aim is to examine, which are the occupations with the highest all-cause mortality, and whether these differences could be explained by family type, education, income and former unemployment.

Materials and methods

We used longitudinal individual level register based data from the registers of Statistics Finland. Study population consisted of employees aged 30-64 at the end of the year 2000. The follow-up period was 2001-2015. We used Cox proportional hazard regression models to estimate the association between occupation and mortality.

Results

The occupations with increased risk of death were mostly manual workers.

Controlling for family type had only minor effect on excess mortality. Adjusting for education diminished the differences more, especially among men (30-50%). There were large differences, how much income explained. The effect of adjusting for unemployment varied largely between occupations.

On average, the adjustment of all sociodemographic characteristics explained 36-98% of excess mortality in ten occupations with the highest mortality among men and 11-61% among women. The percent was high eg. among cleaners and in construction industry.

Conclusion

In some occupations controlling for the variables explained all excess mortality. In many occupations, the association between occupation and mortality could be explained partly by sociodemographic characteristics. Some occupations were high risk independently of family type, education, income and unemployment. High mortality cannot be explained by a single factor, reasons behind high mortality vary between occupations.