Background. Breast cancer is the most frequent cancer in women and the incidence has increased over time. Our objectives were to study: (1) the socioeconomic differences in breast cancer incidence and mortality in Denmark, (2) how different socioeconomic groups have contributed to the increasing incidence, (3) whether the diverging trend between breast cancer incidence and mortality reflects different socioeconomic distributions of breast cancer cases and breast cancer deaths, and (4) to compare measures of socioeconomic status based on own and spouses' occupation, respectively. We addressed these questions by studying the socioeconomic distribution of breast cancer incidence and breast cancer mortality in Danish women during the last 25 years. Methods. In all 1 402 225 women in Denmark were individually followed up for death, emigration, and incident breast cancer in 1970-1995. Of the 1 402 225 women included in the study, 730 549 were economically active in 1970, and 480 379 women were both married and economically active. Socioeconomic status was assessed based on the occupation in 1970. Results. For all women classified by their own socioeconomic group, the standardized incidence (SIR) and the standardized mortality ratios (SMR) were highest in academics (SIR = 1.39, SMR = 1.29), and lowest in women in agriculture (SIR = 0.77, SMR = 0.75). For married, economically active women classified by their own socioeconomic group the SIR and SMR were highest in academics (SIR = 1.40, SMR = 1.44) and lowest in women in agriculture (SIR = 0.76, SMR = 0.76). Classified by their husbands' socioeconomic group, the SIR and SMR were highest in women married to academics (SIR = 1.21, SMR = 1.16) and lowest in women married to men in agriculture (SIR = 0.79, SMR = 0.79). From 1970 to 1995, the risk of developing breast cancer increased by 38% in women aged 50-64. All social groups contributed to this increase, the increase being 45% in unskilled workers, and 26% in academics. Conclusion. During the last quarter of the 20th century academics had the highest risk of breast cancer in Denmark. The size of the social gradient in breast cancer occurrence depended on the measure used. The time trends in social distribution will result in breast cancer becoming more frequent.