Introduction: Breast cancer (BC) is the most prevalent and the leading cause of death from cancer in women worldwide. The low percentage of cases related to reproductive history risk factors and to genetics suggests that the environment may play a role in its etiology. Pesticide exposure has been hypothesized to have an important effect. Objective: This ecological study tests whether BC incidence in Costa Rica is related to pesticide environmental exposure (PEE), controlling for access to health care, fertility, age at first pregnancy, and socioeconomic status. Methods: A Poisson regression model was run. Spatial analysis techniques to test for spatial autocorrelation, and geographically weighted regressions were used. Results: PEE had a statistically significant direct association with BC for women 45 yrs+. The corresponding incidence rate ratio for PEE was 1.29. This means that after controlling for other risk factors, moving a district to the next decile of PEE was associated with 29% increase in BC incidence for women 45+. PEE was significant in some rural and agricultural areas of the country, after controlling for other risk factors. Conclusions: There seems to be an actual relation between breast cancer and pesticides. Since it is a preventable risk factor, this is an important public health issue to be debated. Paying more attention to health consequences that derivate from environmental exposure would imply a shift toward the application of the precautionary principle. Conclusions about causality can not be drawn from an ecologic approach, like the one taken in this study.Introduction: Breast cancer (BC) is the most prevalent and the leading cause of death from cancer in women worldwide. The low percentage of cases related to reproductive history risk factors and to genetics suggests that the environment may play a role in its etiology. Pesticide exposure has been hypothesized to have an important effect. Objective: This ecological study tests whether BC incidence in Costa Rica is related to pesticide environmental exposure (PEE), controlling for access to health care, fertility, age at first pregnancy, and socioeconomic status. Methods: A Poisson regression model was run. Spatial analysis techniques to test for spatial autocorrelation, and geographically weighted regressions were used. Results: PEE had a statistically significant direct association with BC for women 45 yrs+. The corresponding incidence rate ratio for PEE was 1.29. This means that after controlling for other risk factors, moving a district to the next decile of PEE was associated with 29% increase in BC incidence for women 45+. PEE was significant in some rural and agricultural areas of the country, after controlling for other risk factors. Conclusions: There seems to be an actual relation between breast cancer and pesticides. Since it is a preventable risk factor, this is an important public health issue to be debated. Paying more attention to health consequences that derivate from environmental exposure would imply a shift toward the application of the precautionary principle. Conclusions about causality can not be drawn from an ecologic approach, like the one taken in this study.