Aims: To analyse the association between sickness absence and sex segregation of occupation and of work site, respectively, and to analyse work environmental factors associated with high sickness absence. Methods: The study group consisted of 1075 women employed as nurses, assistant nurses, medical secretaries, or metal workers who answered a questionnaire comprising 218 questions on women's health and living conditions. Sickness absence was collected from employers' and social insurance registers. Results: Women working in the male dominated occupation had in general higher sickness absence compared to those working in female dominated occupations. However, metal workers at female dominated work sites had 2.98 (95% CI 2.17 to 3.79) sick-leave spells per woman and year compared to 1.70 (95% CI 1.29 to 2.10) among those working with almost only men. In spite of a better physical work environment, female metal workers at a female dominated work site had a higher sickness absence than other women, which probably could be explained by the worse psychosocial work environment. Working with more women also had a positive association to increased frequency of sick-leave spells in a multivariate analysis including several known indicators of increased sick-leave. Conclusions: There was an association between sickness absence and sex segregation, in different directions at the occupational and work site level. The mechanism behind this needs to be more closely understood regarding selection in and out of an occupation and a certain work site.