Environment and Breast Cancer: Science Review
About
A follow-up study of mortality among women in the North American synthetic rubber industry
Sathiakumar, N., Delzell, E. J Occup Environ Med. 2009. 51:11, 1314-25.
Topic area
Environmental pollutant - Occupation Styrene
Environmental pollutant - Occupation Styrene
Study design
Retrospective cohort
Retrospective cohort
Funding agency
International Institute of Synthetic Rubber Produc
International Institute of Synthetic Rubber Produc
Study Participants
Menopausal Status
The menopausal status of women included in this study is listed here.
No analysis based on menopausal status
Number in Cohort
Cohort: 4,863 women
Cohort: 4,863 women
Country where study was conducted
USA Canada
USA Canada
Cohort participation rate
Not applicable. This study was records-based.
Not applicable. This study was records-based.
Participant selection: Inclusion and exclusion criteria
Criteria used to select participants in the study.
The cohort is comprised of 4,863 female workers hired before January 1, 1991 and employed for at least 1 day at eight styrene butadiene rubber plants located in Kentucky, Ontario, Texas, and Louisiana. Inclusion in the cohort was based on availability of personnel records with work history information. Vital status was followed through 2002 using plant records, SSA, NDI, Cambridge Statistical Research Associates, Centers for Medicare and Medicaid services, Canadian Mortality Data Base and annual tax files, beginning in 1943, 1950, 1960, and 1965 (depending on the plant). Cause of death was determined by nosologists' review of death certificates for deaths before 1979, and from NDI Plus from 1979 on.
Exposure Investigated
Exposures investigated
Exposure estimates to butadiene and styrene were derived from job groups, tasks, and average exposure intensities compiled in job exposure matrices. Job groups were based on personnel records of job changes, duration, work area, title, and pay status. An
Exposure estimates to butadiene and styrene were derived from job groups, tasks, and average exposure intensities compiled in job exposure matrices. Job groups were based on personnel records of job changes, duration, work area, title, and pay status. An
How exposure was measured
Job history Environmental sample Other: Job exposure matrix (JEM)
Job history Environmental sample Other: Job exposure matrix (JEM)
Exposure assessment comment
About 34% of women at the plant worked in exposed jobs. The median duration of employment was 1.6 years, with the large majority having worked between 0 and 5 years.
About 34% of women at the plant worked in exposed jobs. The median duration of employment was 1.6 years, with the large majority having worked between 0 and 5 years.
Breast cancer outcome investigated
Mortality
Mortality
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, race, calendar period
Genetic characterization included
If the study analyzed relationships between environmental factors and inherited genetic variations, this field will be marked Yes. No, if not.
No
Strength of associations reported
External comparison:
Breast cancer mortality rates at the plants compared to the surrounding population, defined as countries within a 50 mile radius
of the U.S. plants, and the province of Ontario as comparison to Ontario's plant.
SMR 97 (95% CI 76-122)
Internal cohort comparisons:
1,3-butadiene exposure (Ref = 0 ppm-yrs)
> 0 to 19.9 ppm-years (8 cases): RR 0.9 (95% CI 0.4 β2.0)
>19.9 ppm-years (9 cases): RR 1.9 (95% CI 0.8-4.4)
Styrene exposure (Ref = 0 ppm-yrs)
> 0 to 5.5 ppm-yrs (8 cases): RR 0.8 (95% CI 0.4 β1.7)
>5.5 ppm-yrs (9 cases): RR 1.5 (95% CI 0.6-3.4)
External comparison:
Breast cancer mortality rates at the plants compared to the surrounding population, defined as countries within a 50 mile radius
of the U.S. plants, and the province of Ontario as comparison to Ontario's plant.
SMR 97 (95% CI 76-122)
Internal cohort comparisons:
1,3-butadiene exposure (Ref = 0 ppm-yrs)
> 0 to 19.9 ppm-years (8 cases): RR 0.9 (95% CI 0.4 β2.0)
>19.9 ppm-years (9 cases): RR 1.9 (95% CI 0.8-4.4)
Styrene exposure (Ref = 0 ppm-yrs)
> 0 to 5.5 ppm-yrs (8 cases): RR 0.8 (95% CI 0.4 β1.7)
>5.5 ppm-yrs (9 cases): RR 1.5 (95% CI 0.6-3.4)
Results Comments
Because styrene and 1,3 butadiene are correlated in this industry, the authors did not attempt to model independent effects of these chemicals (i.e. considering both chemicals in one model with mutual adjustment).
Because styrene and 1,3 butadiene are correlated in this industry, the authors did not attempt to model independent effects of these chemicals (i.e. considering both chemicals in one model with mutual adjustment).
Author address
Department of Epidemiology, School of Public Health, The University of Alabama at Birmingham, Birmingham, Ala, USA. nalini@uab.edu
Department of Epidemiology, School of Public Health, The University of Alabama at Birmingham, Birmingham, Ala, USA. nalini@uab.edu
Reviewers Comments
Mortality is not a sensitive measure of breast cancer risk. Further, the lack of control for breast cancer risk factors is a concern for this analysis because the comparison groups may not be appropriate. Specifically, the SMR compares to the general population, which could bias the estimate downward due to a Healthy Worker Effect. For the internal comparisons, the comparison group consists of women who worked in administration or 'residual' operations. Differences in breast cancer risk factors in administrators and exposed plant workers could bias estimates.
Mortality is not a sensitive measure of breast cancer risk. Further, the lack of control for breast cancer risk factors is a concern for this analysis because the comparison groups may not be appropriate. Specifically, the SMR compares to the general population, which could bias the estimate downward due to a Healthy Worker Effect. For the internal comparisons, the comparison group consists of women who worked in administration or 'residual' operations. Differences in breast cancer risk factors in administrators and exposed plant workers could bias estimates.