< Back to previous page
Publication
Educational differences in breast cancer mortality in the Brussels-Capital Region
Book Contribution - Book Chapter Conference Contribution
Objective: The aim of this study is to investigate whether there is an inversion of the gradient in breast cancer mortality in the Brussels-Capital Region. Several studies have shown a shift from a positive to a negative relationship between education and breast cancer mortality over time. In this presentation, trends in inequalities in breast cancer mortality are addressed through a comparison of two longitudinal datasets. In addition, the study investigates whether the gradient persists after adjustment for reproductive history.
Methods: The study population consists of all women in the Brussels-Capital Region aged 50-79 in 1991 (N=155,481) and in 2001 (N=144,571). The first dataset consists of a linkage of the Belgian 1991 census with registration records of all deaths and emigrations between March 1991 and December 1995. The second dataset consists of a linkage of the Belgian 2001 census with registration records for the period October 2001-December 2004. Rate ratios are calculated through survival analyses with breast cancer mortality as dependent variable and educational attainment as independent variable. Reproductive factors are controlled for through the inclusion of number of children and age at first birth.
Results: During 1991-1995, tertiary educated women show a higher rate ratio compared to women with primary education. Control for number of children and for age at first birth reduces the excess mortality of tertiary educated women, but differences remain significant.
In the 2001-2004 period, the educational differences are no longer significant. Rate ratios have become smaller. Even more, inclusion of reproductive factors changes the direction of the gradient, women with secondary education having the highest mortality in breast cancer. Trends in breast cancer mortality between 1991-1995 and 2001-2004 differ significantly by educational attainment. Rates diminish considerably among women with tertiary education, slightly among women with primary education and increase among women with secondary education.
Conclusions: The results show that educational differences in breast cancer are changing over time. Highly educated women may have benefited more from screening programs and health interventions. In order to achieve further declines in breast cancer mortality, policy interventions should take this into account. Other hypotheses should be considered as well.
Methods: The study population consists of all women in the Brussels-Capital Region aged 50-79 in 1991 (N=155,481) and in 2001 (N=144,571). The first dataset consists of a linkage of the Belgian 1991 census with registration records of all deaths and emigrations between March 1991 and December 1995. The second dataset consists of a linkage of the Belgian 2001 census with registration records for the period October 2001-December 2004. Rate ratios are calculated through survival analyses with breast cancer mortality as dependent variable and educational attainment as independent variable. Reproductive factors are controlled for through the inclusion of number of children and age at first birth.
Results: During 1991-1995, tertiary educated women show a higher rate ratio compared to women with primary education. Control for number of children and for age at first birth reduces the excess mortality of tertiary educated women, but differences remain significant.
In the 2001-2004 period, the educational differences are no longer significant. Rate ratios have become smaller. Even more, inclusion of reproductive factors changes the direction of the gradient, women with secondary education having the highest mortality in breast cancer. Trends in breast cancer mortality between 1991-1995 and 2001-2004 differ significantly by educational attainment. Rates diminish considerably among women with tertiary education, slightly among women with primary education and increase among women with secondary education.
Conclusions: The results show that educational differences in breast cancer are changing over time. Highly educated women may have benefited more from screening programs and health interventions. In order to achieve further declines in breast cancer mortality, policy interventions should take this into account. Other hypotheses should be considered as well.
Book: Paper presented at the 5th Conference of Epidemiological Longitudinal Studies in Europe
Keywords:breast cancer mortality, educational differences, evolution of inequalities