The negative health effects of working long hours are well-known in Japan, where “karoshi” or death due to overwork is an official cause of death. However, European research is lagging behind. This study aims to investigate the association between working hours and mortality among Belgian male workers with a lag time of 10 years.

The total male working population in Belgium is selected from an anonymous record linkage between the 1991 Belgian Census, the Population register (1991-2011) and death-certificate data (2001-2011). Approximately 1.5 million men between 30 and 60 years were employed at the time of the census. A total of 134,320 men died between 1/10/2001 and 31/12/2011. Incidence Rate Ratios (IRR) and 95% confidence intervals (CI) were calculated using (stratified) Poisson regression models. Self-reported information about working hours was derived from the 1991 census. Overall, 21% of the 1991 male workforce worked longer than 40 hours a week. Covariates include age in years and occupation in 1991.

Men working less than a full-time workweek of 35-40 hours experienced increased all-cause mortality risks after adjusting for age and occupation. Men working less than 20 hours a week had an IRR of 1.16 (CI 1.11-1.21). For working between 21 and 34 hours a week, we found an IRR of 1.10 (CL 1.06-1.14).

Working more than 40 hours a week was only associated with increased all-cause mortality risks for specific occupations, including labourers in mining, construction, manufacturing and transport. Preliminary results indicate that these workers experienced increased respiratory mortality risks. Examination of cause-specific mortality also shows elevated suicide risks for corporate managers and managers of small enterprises working more than 40 hours a week compared to their respective colleagues working 35-40 hours a week.

The relation between working hours and mortality risks should be assessed within the specific occupational context.