Breast cancer incidence and exposure to environmental chemicals in 82 counties in Mississippi.

Objective: Breast cancer is a major public health problem in the United States and in most industrialized countries. Environmental exposure to several chemicals has been implicated in the cause of breast cancer. However, data are not consistent about the role of the environment in breast cancer incidence. To assess environmental risks for breast cancer, patterns of breast cancer incidence in relation to environmental chemicals in Mississippi counties were mapped.

Methods: This article presents an analysis of age-adjusted incidence rates of female breast cancer and environmental chemical emissions by county in the state of Mississippi. The incidence data were obtained from the State Department of Health, and emissions data sources included the U.S. Environmental Protection Agency's Air-Data and Toxics Release Inventory Program.

Results: Counties having a significantly higher incidence of breast cancer compared with the state's rate were identified. Also identified were counties with higher levels of chemical emissions. The incidence of breast cancer in 1998 was significantly associated with the amount of ammonia (r = 0.268, P = 0.015), minimum emissions (r = 0.233, P = 0.035), and maximum emissions (r = 0.237, P = 0.032) of the facilities in the county. A linear dose-response relationship was observed between increased amounts of chemical emissions and breast cancer incidence.

Conclusions: This study found a significant link between breast cancer incidence and maximum emissions of environmental chemicals. However, further research using individual-level data is needed.

**********

Breast cancer is the most common form of cancer among women in the United States. A National Cancer Institute report estimates that approximately one in eight women in the U.S. (approximately 13.3%) will develop breast cancer during her lifetime. (1) The incidence of invasive breast cancer, particularly lobular carcinoma, has been increasing steadily for the past two decades. (2) Important risk factors for female breast cancer include early age at onset of menarche (younger than 12 years), late age at onset of menopause (55 years or older), first full-term pregnancy after age 30, nulliparity, history of premenopausal breast cancer for mother or a sister, personal history of breast cancer or of benign proliferative breast disease, obesity, urban residence, lack of breast-feeding, oral contraceptive use, and inherited mutation in breast cancer-susceptible genes, such as BRCA1 and BRCA2. (3,4)

Several studies have indicated that exposure to exogenous chemicals in the environment is associated with a higher incidence of breast cancer. (5) Those chemicals include pesticides, polychlorinated biphenyls, radon, radium, and electromagnetic fields. (6-10) Use of diethylstilbestrol (11) and hormone replacement therapy (12) were found to increase female breast cancer incidence. Several known carcinogens found in tobacco (eg, polycyclic hydrocarbons, aromatic amines, and N-nitrosamines) may induce breast tumors. (13) Recent studies suggest that both active and passive smoking may increase the risk of breast cancer. (14,15) However, several studies have disputed the role of some environmental chemicals in the causation of breast cancer. (16-19) Because of the controversies in the existing literature, we undertook a study to demonstrate the relationship between environmental chemicals and breast cancer incidence in the 82 counties in Mississippi by using geospatial information system (GIS) mapping.

Materials and Methods

The study included Mississippi county-level data on invasive breast cancer incidence in women from 1996 through 1998 obtained from the Mississippi State Department of Health. The age-adjusted incidence rates were calculated per 100,000 population using the 2000 population as the denominator. Subjects were assigned to the different counties according to their residence address. The study also included data on environmental chemical emissions. The emissions data sets were developed from county-wide levels of emissions for certain chemicals, as reported by the U.S. Environmental Protection Agency (EPA) AirData and the EPA Toxics Release Inventory Program. Other sources for environmental chemicals included the Mississippi Department of Environmental Quality, the U.S. Agency for Toxic Substances and Disease Registry, the American Cancer Society, and the Sierra Club. The chemicals causing air pollution were classified into two groups: 1) criteria air pollutants (the EPA has set health-based standards for them) and 2) total core chemicals (the EPA listed more than 600 chemicals as Toxics Release Inventory 1995 core chemicals). Criteria air pollutants included carbon monoxide (CO), nitrogen oxides, sulfur dioxide (S[O.sub.2]), particulate matters (P[M.sub.10] and P[M.sub.2.5]), ozone precursors in the form of volatile organic compounds (VOCs), and ammonia (N[H.sub.3]). Emissions data found on the AirData website also included fields labeled "minimum," "maximum," and "total" air pollutant emissions. The minimum and maximum emissions fields referred to the air pollutant emissions for a single source. For example, if a single facility emitted 10 tons of S[O.sub.2], 20 tons of particulate matter, and 30 tons of CO, then minimum emissions = 10 tons, maximum emissions = 30 tons, and total emissions = 60 tons. All data were aggregated by county. No individual-level data were used for GIS mapping. The study was approved by the Institutional Review Board of the University of Southern Mississippi.

Data Analysis

Data were analyzed using SPSS for Windows, version 11.0 (SPSS, Inc., Chicago, IL), and EpiInfo, version 2002 (Centers for Disease Control and Prevention, Atlanta, GA). Trends of breast cancer incidence were calculated by race and by year. Because of skewed distribution, median rates were used for further comparison. Statistical significance was calculated by a comparison of the incidence rate of breast cancer for each county with the state's rate. Percent difference in incidence of each county from the state's rate was calculated by using the following formula:

Percent difference in county incidence = [[County incidence-State incidence]/[County incidence]] X 100.

Counties were classified into four categories according to the percent difference in incidence: no increase from the state's rate, less than 20% increase, 20 to 39% increase, and 40% or more increase. To demonstrate the high-risk areas, counties were mapped according to the breast cancer incidence categories using ArcView GIS 3.2 (ESRI, Redlands, CA). Pearson's correlation was used to analyze 1) bivariate correlation between breast cancer incidence and independent variables, including CO, nitrogen oxides, S[O.sub.2], P[M.sub.10], P[M.sub.2.5], VOCs, and N[H.sub.3]; 2) minimum, maximum, and total emissions; and 3) total amount of core chemicals. To examine a dose-response relationship, the variables found statistically significant at bivariate analysis (eg, minimum and maximum emissions) were further grouped into multiple-dose categories. One-way analysis of variance and an appropriate post hoc test were performed to find which doses of the independent variables (minimum and maximum emissions) significantly related to breast cancer incidence. A value of P [less than or equal to] 0.05 (two-tailed) was considered statistically significant.

Results

The incidence rate of female breast cancer in the 82 counties in Mississippi decreased from 61.2 per 100,000 in 1996 to 58.4 per 100,000 in 1998. Whites had a higher incidence rate than nonwhites; however, the difference was not statistically significant (61.4 per 100,000 versus 52.3 per 100,000 in 1998; P = 0.40). Further analysis was restricted to the 1998 incidence data, because those were the latest completed records available for the state.

High-Risk Counties by Breast Cancer Incidence

Counties were categorized according to the percent difference in incidence of breast cancer from the state's rate (Fig. 1). Six counties had incidence rates 40% or higher than the state's rate: Yazoo (117 per 100,000), Copiah (114 per 100,000), George (106 per 100,000), Forrest (105 per 100,000), Stone (99 per 100,000), and Hinds (97 per 100,000). The other counties that had significantly higher rates than the state's rate included Lee, Noxubee, Jefferson, Jones, Perry, Scott, Chichasaw, Madison, Yalobusha, Clay, Tishomingo, and Warren, in order of incidence frequency. Counties with very low rates were Coahoma, De Soto, Quitman, Tallahatchie, and Tate. No reports were available from Issaquena and Tunica.

[FIGURE 1 OMITTED]

Relation Between Breast Cancer Incidence and Environmental Chemicals

Incidence of breast cancer correlated significantly with air pollutants including N[H.sub.3] (r = 0.268, P = 0.015), minimum emission (r = 0.233, P = 0.035), and maximum emission (r = 0.237, P = 0.032). Figure 2 shows the correlation between maximum emissions and breast cancer incidence.

Maximum emissions of a facility ranged from 2,091 to 105,510 tons, with a median of 11,984 tons. The top 10 counties, in order of the amount of maximum emissions, included Harrison, Hinds, Jackson, Forrest, Rankin, Jones, Lauderdale, Perry, De Soto, and Scott, all having 60% or higher amounts of chemical emissions than the state's median amount of emissions. Table 1 shows the major sources of air pollution in those counties.

To examine a dose-response relationship, maximum emissions were classified into three dose categories: 1) 2,091 to 11,984 tons, which corresponds to no increase from the state's median amount of emissions; 2) 11,985 to 22,523 tons, which is a less than 50% increase; and 3) greater than 22,523 tons, which is a greater than or equal to 50% increase. The incidence of breast cancer increased with increasing doses of maximum emissions, as suggested by the rates of 51 per 100,000, 59 per 100,000, and 71 per 100,000 in dose categories 1, 2, and 3, respectively. One-way analysis of variance and Tukey honest significant difference test showed that breast cancer incidence increased significantly in dose category 3 compared with dose category 1 (P = 0.027). A similar direct association was observed between breast cancer incidence and increasing doses of minimum emissions.

[FIGURE 2 OMITTED]

Discussion

This study identified several counties in Mississippi that had a higher incidence of female breast cancer compared with the state's median rate in 1998. The study also showed a dose-response relationship between breast cancer incidence and the amount of environmental chemical emissions in the state. However, no single chemical agent was incriminated for the increased breast cancer incidence, mainly because the available data at the county level included aggregated amounts of chemicals from multiple sources of chemical plants. Sources that produced the most emissions included inorganic pigments, electrical services, natural gas transmission, sawmills and planing mills, paper mills, petroleum refining, general medicine and surgical hospitals, and ship building and repairing, among others. These data may be helpful in developing measures against exposure to the offending agents in high-risk counties.

To our knowledge, this is the first report to confirm the relationship between environmental chemicals and breast cancer incidence in Mississippi. However, a recent report showed an association between pesticide exposure and increased risk of breast cancer mortality in Mississippi. (20) In that study, the total number of acres planted during 1997 to 2000 was used as a proxy measure for pesticide exposure. Three statistical economic areas including Greenville, Corinth, and Yazoo counties were identified as the areas of greatest vulnerability to increased pesticide exposure and higher risk of breast cancer. Several studies documented evidence in favor of the relationship between breast cancer and environmental exposure to chemicals, including organochlorines, polychlorinated biphenyls, and other agents. (5) Our study was consistent with another study that showed a significantly elevated risk of breast cancer incidence among women in Nassau and Suffolk counties in New York who were ever exposed to chemical facilities. (21) However, the association between breast cancer and chemical exposure is not consistent in the literature. In Montreal, Quebec, the incidence of stomach cancer and cervical cancer increased, but the incidence of breast cancer decreased in women living near a solid waste landfill site that generated volatile organic compounds. (22)

One of the limitations of our study is that it did not provide individual-level data. It is likely that many of the counties would be heterogeneous for exposure and for other personal characteristics that would affect the risk of breast cancer. In addition, one could imagine that disadvantaged persons would be living in higher pollution areas. For example, one could hypothesize that these areas would consist of women who had lower levels of education and possibly more children at younger ages, factors that tend to decrease the incidence of breast cancer. This may result in an underestimate of the true association. However, most of the limitations to this study are intrinsic to these types of epidemiologic studies. Because of the nature of the study, it was difficult to control for other factors that may influence the outcome of breast cancer incidence.

Conclusion

This study identified a significant association between maximum emissions of air pollutants and breast cancer incidence in Mississippi. Further research is needed to measure the association more precisely by using individual-level data.

Table 1. Sources of pollution in the top 10 polluted counties in
Mississippi

            Maximum emissions
County         (tons/yr)       Major source of pollution (a)

Harrison       105,510         Inorganic pigments; electric services;
                                 national security; hoists, cranes, and
                                 monorails; wood partitions and
                                 fixtures; petroleum bulk stations and
                                 terminals; ship building and repairing;
                                 plastic pipe production; ready-mixed
                                 concrete
Hinds          104,236         Electric services; natural gas
                                 transmission; general medicine and
                                 surgical hospitals; brick and
                                 structural clay tile; folding
                                 paperboard boxes; minerals; ground or
                                 treated brick; paints and allied
                                 products; fabricated metal products;
                                 ready-mixed concrete
Jackson         68,659         Paper mills; petroleum refining; electric
                                 services; ship building and repairing;
                                 phosphatic fertilizers
Forrest         52,878         Sawmills and planing mills, general;
                                 crude petroleum and natural gas;
                                 electric services; natural gas
                                 transmission; national security
Rankin          44,660         Blast furnaces and steel mills; natural
                                 gas transmission; products of purchased
                                 glass; industrial inorganic chemicals;
                                 miscellaneous metal work
Jones           44,370         Electric services; pulp mills; farm
                                 machinery and equipment; reconstituted
                                 wood products; petroleum refining;
                                 transformers
Lauderdale      39,268         Electric services; sawmills and planing
                                 mills, general; manufacturing
                                 industries; bread, cake, and related
                                 products; paper-coated and laminated
                                 packaging; petroleum bulk stations and
                                 terminals
Perry           36,241         Sawmills and planing mills, general;
                                 natural gas transmission; wood products
De Soto         33,445         Natural gas transmission; sawmills and
                                 planing mills, general; paving mixtures
                                 and blocks; aluminum extruded products;
                                 metal cans; boat building and
                                 repairing; metal doors, sash, and trim;
                                 fabricated metal products;
                                 transportation equipment; corrugated
                                 and solid fiber box production; iron
                                 and steel forgings
Scott           29,905         Sawmills and planing mills, general;
                                 refuse systems

(a) Based on the Standard Industrial Classification of the primary
activity performed or product produced by a source.

Acknowledgments

We gratefully acknowledge Michelle Richard, director of the cancer registry; Dick Johnson, Department of Public Health Statistics, Mississippi State Department of Health; and Tom Dessent, Internet support contractor for U.S. EPA Office of Air and Radiation, for their valuable contributions in providing the data.

Accepted August 4, 2003.

Copyright [c] 2004 by The Southern Medical Association

0038-4348/04/9703-0259

References

1. National Cancer Institute. Cancer facts: Lifetime probability of breast cancer in American women. Available at: http://www.nci.nih.gov/fact/5_6.htm. Accessed January 14, 2004.

2. Li CI, Anderson BO, Daling JR, Moe RE. Trends in incidence rates of invasive lobular and ductal breast carcinoma. JAMA 2003;289:1421-1424.

3. Hulka BS, Moorman PG. Breast cancer: Hormones and other risk factors. Maturitas 2001;38:103-116.

4. American Cancer Society. Breast Cancer Facts and Figures 2001-2002. Available at: http://www.cancer.org/docroot/STT/content/STT_1x_Breast_Cancer_Facts_and_Figures_2001-2002.asp. Accessed September 24, 2003.

5. Mitra AK, Faruque FS, Avis AL. Breast cancer and environmental risks. J Environ Health In press.

6. Hoffman W. Organochlorine compounds: Risk of non-Hodgkin's lymphoma and breast cancer? Arch Environ Health 1996;51:189-192.

7. Demers A, Ayotte P, Brisson J, et al. Plasma concentrations of polychlorinated biphenyls and the risk of breast cancer: A congener-specific analysis. Am J Epidemiol 2002;155:629-635.

8. Soto J, Quindos LS, Cos S, et al. Influence of low doses of radiation due to [.sup.222]Rn on proliferation of fibroblasts and MCF-7 human breast cancer cells in vitro. Sci Total Environ 1996;181:181-185.

9. Stebbings JH. Health risks from radium in workplaces: An unfinished story. Occup Med 2001;16:259-270.

10. Verkasalo PK, Pukkala E, Kaprio J, et al. Magnetic fields of high voltage power lines and risk of cancer in Finnish adults: Nationwide cohort study. BMJ 1996;313:1047-1051.

11. Laitman CJ. DES exposure and the aging woman: Mothers and daughters. Curr Womens Health Rep 2002;2:390-393.

12. Rossouw JE, Anderson GL, Prentice RL, et al; Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321-333.

13. Terry PD, Rohan TE. Cigarette smoking and the risk of breast cancer in women: A review of the literature. Cancer Epidemiol Biomarkers Prev 2002;11:953-971.

14. Terry PD, Miller AB, Rohan TE. Cigarette smoking and breast cancer risk: A long latency period? Int J Cancer 2002;100:723-728.

15. Kropp S, Chang-Claude J. Active and passive smoking and risk of breast cancer by age 50 years among German women. Am J Epidemiol 2002;156:616-626.

16. Stellman SD, Djordjevic MV, Britton JA, et al. Breast cancer risk in relation to adipose concentrations of organochlorine pesticides and polychlorinated biphenyls in Long Island, New York. Cancer Epidemiol Biomarkers Prev 2000;9:1241-1249.

17. van't Veer P, Lobbezoo IE, Martin-Moreno JM, et al. DDT (dicophane) and postmenopausal breast cancer in Europe: Case-control study. BMJ 1997;315:81-85.

18. Zheng T, Holford TR, Mayne ST, et al. [beta]-benzene hexachloride in breast adipose tissue and risk of breast carcinoma. Cancer 1999;85:2212-2218.

19. Zheng T, Holford TR, Mayne ST, et al. Risk of female breast cancer associated with serum polychlorinated biphenyls and 1,1-dichloro-2,2'-bis(p-chlorophenyl)ethylene. Cancer Epidemiol Biomarkers Prev 2000;9:167-174.

20. Abdalla MH, Gutierrez-Mohamed ML, Farah IO. Association of pesticide exposure and risk of breast cancer mortality in Mississippi. Biomed Sci Instrum 2003;39:397-401.

21. Lewis-Michl EL, Melius JM, Kallenbach LR, et al. Breast cancer risk and residence near industry or traffic in Nassau and Suffolk Counties, Long Island, New York. Arch Environ Health 1996;51:255-265.

22. Goldberg MS, al-Homsi N, Goulet L, et al. Incidence of cancer among persons living near a municipal solid waste landfill site in Montreal, Quebec. Arch Environ Health 1995;50:416-424.

RELATED ARTICLE: Key Points

* One in eight American women who live to age 85 will develop breast cancer at some time during her life.

* Exposure to environmental chemicals has been linked to higher breast cancer incidence, but there are controversies regarding the association.

* Counties in Mississippi that had a higher incidence of breast cancer included Yazoo, Copiah, George, Forrest, Stone, and Hinds.

* The study found a significant association between environmental-chemical exposure and female breast cancer incidence.

* A dose-response relationship was observed between maximum emissions of air pollutants and breast cancer incidence.

Amal K. Mitra, MD, MPH, DRPH, and Fazlay S. Faruque, PHD, PG

From Epidemiology and Biostatistics, Center for Community Health, University of Southern Mississippi, Hattiesburg; and Geographic Information Systems, School of Nursing, University of Mississippi Medical Center, Jackson, MS.

Supported by the 2002 Faculty Summer Research Grants of the University of Southern Mississippi.

Reprint requests to Amal K. Mitra, MD, MPH, DrPH, Epidemiology and Biostatistics, Center for Community Health, University of Southern Mississippi, Hattiesburg, MS 39406-5122. Email: [e-mail]