Context The aim of this study was to describe the factors associated with mortality by suicide among women at work with a particular focus on work-related factors.

Population and methods The study population was based on the data of the Swiss National Cohort (SNC) consisting in all Swiss residents recorded in the 1990 and/or the 2000 compulsory national censuses linked to emigration and mortality registers. Within the SNC we selected all working women aged between 18 and 65 at the official census dates.

Following work-related variables were available in both censuses: Socio-economic status, weekly hours of work, the sector of activity according to the Swiss NOGA scheme and the job title coded according to ISCO. The date of beginning of follow-up was the date of the relevant census. The date of end of follow-up was the earliest of the 65th anniversary, dec 31 st 2014 (the end of mortality follow-up), the date of loss of follow-up or dec 4th 2010 if no declared work in the 2000 census. The risk of suicide was modelled using negative binomial regression incorporating the number of person-years as offset.

Results The cohort comprised 1,843,619 women and 2524 deaths by suicide corresponding to 24.9 million person-years. The most significant non-occupational predictors of suicide were age, civil status, period, nationality and geographical regions. Part-time work and long working hours were associated with increased suicide rates. Health and social activities had the highest suicide risks and education the lowest. Overall the job title according to ISCO-1d was not significant. However the rank of suicide risk among women followed approximately the skill levels associated with these codes. Of interest is however the high RR among professionals.

Conclusion After accounting for the main non-occupational factors, the suicide risk varied significantly with some work-related factors that were specific for women.