Environment and Breast Cancer: Science Review

Evidence From Humans
Print this page
Investigating the role of transportation models in epidemiologic studies of traffic related air pollution and health effects
Shekarrizfard, M., Valois, M. F., Goldberg, M. S., Crouse, D., Ross, N., Parent, M. E., Yasmin, S., Hatzopoulou, M. Environ Res. 2015. 140, 282-91.
Topic area
Environmental pollutant - Air pollution Vehicle exhaust
Study design
Hospital-based case-control
Funding agency
Natural Sciences and Engineering Research Council
Study Participants
Menopausal Status
The menopausal status of women included in this study is listed here.
Analysis restricted to postmenopausal status
Number of Controls
Controls: 415
Participant selection: Inclusion and exclusion criteria
Criteria used to select participants in the study.
The study population -- the same as that reported on in Crouse et al., 2010 & Labreche et al., 2010 -- consisted of postmenopausal women, 50-75 years of age at time of diagnosis, who in 1996 and 1997 were residents of the Island of Montreal and Nun's Island. Cases had primary, histologically confirmed, invasive breast cancer, and were identified through 18 hospitals in the region that treated breast cancer. Controls were matched to cases by hospital and age, and they had 1 of 32 other selected sites of incident, histologically confirmed cancers. Liver, intrahepatic bile duct, pancreas, lung, bronchus and trachea, brain and central nervous system, leukemias and lymphomas were excluded because of their possible association with occupational exposures.
Exposure Investigated
Exposures investigated
Ambient NO2 concentrations from a land use regression (LUR) model and NOx emissions from a traffic assignment model, estimated at geocoded self-reported residence at diagnosis/interview.
How exposure was measured
Questionnaire GIS/geographic Environmental sample
Exposure assessment comment
LUR model was the same as that developed by Crouse et al., 2010, except that it was based on air monitoring data (integrated 2-week sampling) from a particular period -- August 2006. The traffic assignment model used 2008 traffic data (daily diaries from the Origin-Destination survey) and a vehicle allocation algorithm. The output from this model were then combined with emission factors (season-, road type-, speed-, and vehicle-specific) to estimate daily NOx emissions. Traffic assignment model was compared to actual traffic counts at several major intersections and bridges in the region and found to perform well. Authors note that while only current estimates of exposure were used, the analysis by Crouse et al., 2010 suggested spatial distributions do not vary dramatically over a 10-20 year period.
Breast cancer outcome investigated
Primary incident breast cancer
Confounders considered
Other breast cancer risk factors, such as family history, age at first birth, and hormone replacement therapy use, that were taken into account in the study.
Age at diagnosis, first degree family history, education, family income, ethnicity, drinking and smoking status, body mass index, immigration status, age at oophorectomy, age at menarche, oral contraception use, duration of HRT use, total duration of brea
Genetic characterization included
If the study analyzed relationships between environmental factors and inherited genetic variations, this field will be marked “Yes.” “No”, if not.
Strength of associations reported
LUR model August NO2 concentrations (per 2.7 ppb increase): aOR 1.14 (95% CI 0.94-1.38)
Transportation model, summer NOx emissions (per 481g increase): aOR 1.04 (95% CI 0.98-1.10)
Results Comments
Correlation between LUR (NO2) and transportation model (NOx) estimates were good near roads, less so moving away from roads. Authors note this may be a limitation of the transportation model, which does not include other sources of air pollution.
Author address
Department of Civil Engineering and Applied Mechanics, McGill University, 817 Sherbrooke St. W., Room 492, Montreal, Quebec, Canada H3A 2K6. Electronic address: maryam.shekarrizfard@mail.mcgill.ca. Department of Medicine, McGill University, Division of Cl
Reviewers Comments
The purpose of this study was to compare traffic-related pollution exposure estimates and effect estimates for breast and prostate cancer produced by the LUR vs transportation model. LURs generally depend on air monitoring data; the transportation model may offer a good alternative, though authors note that this model may tend to overestimate exposure near roadways (because of lack of accounting for dispersion) and underestimate exposure elsewhere.