Evidence From Humans
Print this page
Pesticide use and breast cancer risk among farmers' wives in the agricultural health study
Engel, L. S., Hill, D. A., Hoppin, J. A., Lubin, J. H., Lynch, C. F., Pierce, J., Samanic, C., Sandler, D. P., Blair, A., Alavanja, M. C. American Journal of Epidemiology. 2005. 161:2, 121-35.
Topic area
Environmental pollutant - Pesticide, DDT, 2,4,5-TP, dieldrin, captan, 2,4,5-
Study design
Prospective cohort
Funding agency
National Cancer Institute NIEHS US EPA
Study Participants
Menopausal Status
The menopausal status of women included in this study is listed here.
Pre menopausal
Post menopausal
Number in Cohort
Cohort: 30,454
Cohort participation rate
Greater than 70%
Participant selection: Inclusion and exclusion criteria
Criteria used to select participants in the study.
Women were eligible if they were married to one of the 43,475 male private pesticide applicators participating in the Agricultural Health Study between 1993 and 1997. Female licensed pesticide applicators were not included in the analysis, because of relatively small numbers (1,347, 15 cases), and because their exposure profile is different from the other female participants. Women residing out of state at the time of cohort enrollment were excluded. Women diagnosed with breast cancer prior to enrollment, and/or incident in-situ breast cancer were excluded from this analysis.
Exposures investigated
Exposure was assessed by self-report questionnaire. Female participants' direct exposure was assessed by questions asking about ever/never use of 50 pesticides, as well as the number of years and frequency of mixing or applying ANY pesticides. Questions a
How exposure was measured
Questionnaire, self-administered
Statistical Analysis
Ethnic groups with separate analysis
If this study provided a separate analysis by ethnic or racial group, the groups are listed here.
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, state, BMI, age at menarche, parity, age at first birth, menopausal status, age at menopause, family history of breast cancer, physical activity, smoking, alcohol consumption, fruit and vegetable consumption, education.
Genetic characterization included
If the study analyzed relationships between environmental factors and inherited genetic variations, this field will be marked “Yes.” “No”, if not.
Description of major analysis
Poisson regression was used to estimate rate ratios and 95% confidence intervals. Analyses were stratified by state (Iowa and North Carolina) to test for consistency, and by menopausal status.
Strength of associations reported
Husbands' use of 2,4,5-TP was significantly and consistently associated with breast cancer, both across states and in a dose-response fashion.
Each of 2,4,5-T, dieldrin, and captan was more weakly associated with breast cancer- either consistently associated between states, or with an exposure-response trend.
Husbands' use of all organochlorines was associated with a relative risk of 1.3 (95% CI = 0.9-2.0).
Husbands' use of DDT was inconsistently associated with breast cancer, but husbands' use of the organochlorines Aldrin, Chlordane, Dieldrin, Heptachlor, and Lindane were significantly associated with increased breast cancer. However, except for Dieldrin, these associations were seen in Iowa but not in North Carolina, and not with a dose-response relationship.
Results Comments
This large study concluded that the chemicals most strongly associated with breast cancer (i.e. consistently between states, and showing a dose-response relationship) were 2,4,5-T, dieldrin, 2,4,5-TP, and captan. However, the relative risk for the use of many organochlorine pesticides by the husband were also significantly elevated, although only for postmenopausal women, and in Iowa. In addition, these relationships were seen in spite of both direct and indirect exposure measures being classified as ever/never, which means that many people with low exposures (and potentially low cancer risk) would be grouped with the 'ever exposed', decreasing the relative risk. Also, for indirect exposure, we only have information about pesticides applied by the husband, even though others may also have applied pesticides in the vicinity (increasing potential exposure misclassification). Furthermore, female pesticide applicators were excluded even though they would be expected to have the highest exposures and thus the highest cancer risk.
Author address
Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA. engell@mskcc.org
Reviewers Comments
Substantial efforts were made (in analyses reported in other publications) to characterize the self-reported exposures whose associations with breast cancer were examined in this study.
Privacy notice   |   Copyright statement