Evidence From Humans
Print this page
A population-based case-control study of farming and breast cancer in North Carolina
Duell, E. J., Millikan, R. C., Savitz, D. A., Newman, B., Smith, J. C., Schell, M. J., Sandler, D. P. Epidemiology. 2000. 11:5, 523-31.
Topic area
Environmental pollutant - Pesticide
Study design
Population based case-control
Funding agency
Study Participants
Menopausal Status
The menopausal status of women included in this study is listed here.
Pre menopausal
Post menopausal
Number of Controls
Controls: 790
Participant selection: Inclusion and exclusion criteria
Criteria used to select participants in the study.
Participants of the Carolina Breast Cancer Study. Cases were diagnosed with breast cancer between May 1993 and May 1996 in 24 counties of North Carolina. Those who mentioned farming in the preliminary interview were asked to participate in the current study. Controls were selected from the North Carolina Division of Motor Vehicles lists for women 65 and younger, and from Health Care Financing Administration lists for women 65 and older.
Exposures investigated
Potential pesticide exposures were evaluated by self-report. Measures included duration of living/working on farm, direct application of pesticides, use of protective equipment, etc.
How exposure was measured
Questionnaire, in person
Early life exposures considered
Exposure during thelarche
Statistical Analysis
Ethnic groups with separate analysis
If this study provided a separate analysis by ethnic or racial group, the groups are listed here.
African Americans
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, age at menarche, parity-age at first pregnancy, lactation, current BMI, education, duration of smoking, alcohol consumption, first-degree family history of breast cancer, oral contraceptive use, duration of laundry (for pesticide users).
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
Multiple unconditional logistic regression was used to calculate ORs.
Strength of associations reported
No association was seen between ever working on a farm and breast cancer.
Women who worked by hand with crops had a reduced risk compared to women who never farmed.
Women who reported presence in a field during or shortly after pesticide application were at similar risk to women who had never farmed, but risk was elevated when the referent group was farming women who were not exposed (OR = 1.8, 95% CI: 1.1-2.8).
Similarly, the risk was non-significantly elevated for pesticide application to crops (OR = 1.3, 95% CI: 0.8-2.3) when referents were non-exposed farm women, but no risk increase was seen when the referents were non-farming women.
Women who reported they did not use exposure protection while applying pesticides had a higher risk (when compared to non pesticide-appliers; OR = 2.0, 95% CI: 1.0 - 4.3) than women who reported using exposure protection while applying pesticides (OR = 0.8, 95% CI: 0.4 - 1.8).

The ORs for 1-5, 6-10, 11-20, and >20 years of washing laundry for a pesticide user were 0.8 (0.5-1.3), 0.9 (0.6-1.5), 2.0 (1.2-3.4), and 1.7 (1.0-2.9).
Results Comments
Information about pesticide exposure was gathered by questionnaire, and no information was presented about which specific pesticides participants might have been exposed to. Farming women were more likely than non-farming women to be older and be post-menopausal, but also to have an older age at menarche, younger age at menopause, younger age at first full-term pregnancy, more children, a lower likelihood of being nulliparous, and have done vigorous chores at age 12, have higher BMI, be less educated, African American, nonsmokers, nondrinkers, and nonusers of oral contraceptives. On this basis we might expect farming women to be at decreased risk of breast cancer. Even though these factors were controlled for in the multivariate analysis, residual confounding could still exist. This could explain why elevated risks were not seen when comparing farm workers to non-farmworkers, but were often elevated when comparing farm workers who used to pesticides to farmworkers who didn't. Exposure during thelarche (breast formation; ages 9-16) was not associated with increased risk of breast cancer compared to exposure at other times. However, exposure during thelarche may also be correlated with exposure at other times, which would diminish the chances of seeing a difference.
Author address
Department of Epidemiology, School of Public Health, and the Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, 27599-7400, USA.
Privacy notice   |   Copyright statement