[Income, percent of women living in rural areas, parity, and breast cancer mortality in Spain, 1975-1991].Med Clin (Barc) 1997; 108(2):41-4MC
The aim of the present study was to analyze breast cancer mortality by provinces in Spain during the period 1975-91, and to assess the relationship with the geographical distribution of income level, percent of women living in rural areas and average parity of women in each province.
SUBJECTS AND METHODS
Data were obtained from national statistical sources. Standardized mortality ratios (SMR) for breast cancer were estimated by provinces for the periods 1975-1980, 1981-1986 and 1987-1991, and for the whole period 1975-1991. Poisson regression analysis was used to explore the association between breast cancer mortality and the above mentioned variables. Provinces were categorized according to the quintile distribution of independent variables, and ecological relative risks were estimated for each category.
Higher SMR were observed in island provinces (Canary and Balearic island), Catalonia, Basque Country, Navarre and the provinces of Saragosa, Seville and Valencia. Lowest SMR were observed in the inner provinces of Spain and the east part of Andalusian region. This pattern has remained very similar along the study period: income level showed a positive association with mortality from breast cancer. On the contrary, percent of women living in rural areas and parity were negatively associated to breast cancer mortality. The relative risk estimated for each child of parity adjusted by the other factors was 0.92 (95% confidence interval: 0.89-0.94).
The highest mortality from breast cancer in Spain has been observed in those provinces with the highest income level, the lowest percent of women living in rural areas and the lowest parity. These findings at the ecological (provinces) level are in concordance with results from other studies at the individual level, and further supports the hypothesis that for the etiology of breast cancer, environmental factors could play a dominant role.