Geospatial distribution of pesticide use and disparities in breast cancer mortality in Mississippi women.

Abstract

Breast Cancer is the most common form of cancer among women in the United States. Established risk factors include advancing age, early menarche, late menopause, positive first relative, late age at first birth and socioeconomic status. Mississippi has a combination of risk factors making it suitable for studying the pathways of breast cancer etiology. Data for this study consisted of secondary analyses of the Mississippi age-adjusted breast cancer mortality aggregated by two periods (1970-1994 & 1996-1999) and total number of acres of planted crops (as a proxy for pesticide exposure) for 1997-2001. Descriptive statistics, Pearson correlations as well as geospatial analysis (GIS) by State Economic Area (SEA) were used as methods to analyze the data. We found significant correlation between pesticide exposure and risk of breast cancer mortality in three SEAs: Yazoo, Vicksburg and Corinth. Statistically significant linear associations were found between level of pesticide exposure (acres planted) and breast cancer mortality rate in Mississippi women per SEA for the first period of study 1970-1994. Significant associations (p < 0.05) using Pearson correlation analysis (r) were found for Greenville SEA with both white (r-0.882) and black females for cotton (r=0.4187) and the rice crop (r=0.678). In contrast, Hattiesburg SEA showed significant associations only for black females with the soy crop (r=0.720) and wheat (r=0.570). Yazoo SEA showed correlations for black females and the soy crop (r=0.507) as well as white females with the catfish crop (r=0.592). Columbus SEA, however, showed associations that are significant for white females with the soy crop (r=0.792) and black females with the catfish crop (r=0.720). A significant association was found for Corinth SEA for black females with total plants (r=0.667). Correlations for Greenville and Vicksburg SEAs were not significant for total plants with both black and white females. Our findings may well relate to the pesticide bioaccumulation hypothesis, where over time, pesticides will make their way into the soil, food chain and finally human fatty tissues. Data collected on income and health care levels showed a variable distribution and skewed disparities towards low income and black women. We conclude that there are moderate statistically significant associations between the number of acres planted, type of crops, race and the mortality rates for breast cancer in Mississippi women.

Introduction and Background

The Mississippi River basin contains the largest and most intensively farmed region in the nation. In order to increase yields from crops, large amounts of pesticides are used to protect against weeds, insects, and other pests. The major categories of pesticides are herbicides, insecticides, and fungicides. It is estimated that about two-thirds of all pesticides used for agriculture in the United States are applied to cropland and pasture land in the Mississippi River Basin (Gianessi and Puffer 1991; Phillips et al, 1997; Lee, et al 2007). The intense use of pesticides is of concern because of potential adverse effects on the quality and use of water resources. A study by Mitra et al, 2004 linked environmental risk factors to breast cancer and another study by Mitra and Faruque, 2004 found a significant link between breast cancer incidence and the maximum emissions of environmental chemicals in 82 counties in Mississippi. The most immediate concerns are for aquatic life and for the 18 million people in the basin who rely on surface-water sources for drinking water (Abdalla et. al, 2003).

According to the U.S. Geological Survey (USGS, 1991), the Mississippi River Basin contains about sixty-five percent of the total harvested cropland in the nation, producing about eighty percent of the corn and soybeans, and much of the cotton, rice, sorghum, and wheat (U.S. Department of Agriculture. 1985; U.S. Department of Commerce, 1993). Many of the pesticides used in the basin are present in the Mississippi River and its tributaries. A regional study of the upper Midwest showed that large amounts of herbicides are flushed into streams during storm runoff in late spring and summer each year following cropland application drained down to the state of Mississippi (Thurman et al. 1992). Storm runoff produces high concentrations of many herbicides in streams across the upper Midwest from Nebraska to Ohio for periods of several weeks to several months. Concentration of herbicides in some small streams may exceed 100 micrograms per liter ([micro]g/L) for short periods of time. Flow from these streams, in turn, transports significant amounts of herbicides into large rivers such as the Missouri, Ohio, and Mississippi, and eventually to the Gulf of Mexico (Pereira and Rostad 1990; Goolsby et al 1991; Duell et al, 2001; Pereira and Hostettler 1993).

There have been recent concerns that long-term exposure to certain persistent organochlorine compounds may increase breast cancer risk by affecting estrogen metabolism. There are several equivocal risk factors where evidence is conflicting: hormone replacement therapy, pesticide exposure and ultra violet radiation, oral contraceptives prior to first pregnancy, smoking, ionizing radiation and benign breast disease. Only about 30-40 percent of breast cancer risk can be explained by the established risk factors and the roles of environmental risk factors in breast cancer have long been suspected in influencing breast cancer risk (Ames, 1979; Krieger, 1989; Valentagas and Daling, 1994; Aspelin, 1997; Anderson et al, 2002; Boarders and Verbeek, 1997; and Mitra et. al, 2004).

Xenoestrogenic pesticides currently in wide use in agriculture are the chlorinated hydrocarbons Endosulfan and Methoxy-chlor. They are much less persistent than DDT and are not stored in body fat. Other pesticides besides the chlorinated hydrocarbons, however, may increase the risk of breast and other cancers in women. Atrazine, a persistent herbicide that is the most widely used in the US, and an extensive groundwater contaminant, causes mammary cancer in rats and is linked to ovarian cancer in agricultural areas (Donna et. al, 1989; Theo-Colborn et al, 1993; Charlier et al, 2003; USATS, 1994; Davis et al, 1997).

Human breast milk is contaminated with xenoestrogenic and carcinogenic pesticide residues worldwide and that pesticide contamination of breast milk has been found even in remote villages in the third world (Rogan et al, 1986). The preliminary findings of a possible link between pesticides and breast cancer are already having an impact on the pesticide debate. The involvement of women as well as concerned citizens and scientists in the pesticides issue, even if only peripherally, can support efforts towards pesticide reform. In breast cancer research, pesticides are only one of several environmental exposures that are being considered as risk factors for the disease (SCDPR, 1985; Rogan et al, 1986; LWVEF, 1989; Brody and Rudel, 2003; PAN, 2007).

According to the bioaccumulation hypotheses (Theo-Colborn et al, 1993; Reddetzke and Applegate, 1993; Schafer et al, 2004), substances remain in the environment (soil, water, etc.) long after their use. As a result, the food chain can become biologically affected through hormonal and molecular transformations. Furthermore, the water topology in Mississippi serves as both a reservoir and a carrier of previously used chlorinated insecticides. These insecticides are primarily transported on sediment particles and persist in the food chain due to long soil half-life. In comparison to other areas of the country, residents of Mississippi were being at greater risk of exposure to bioaccumulation of pesticides over very long periods of time. This environment is brought about by at least three factors: (1) the heavy cropland concentrated in areas drained by the Mississippi River Basin; (2) the majority of crops in the nation are harvested in Mississippi river basin; and (3) two-thirds of all agricultural (crop and pasture) use of pesticides in the nation are applied to the Mississippi river basin. The endogenous hormonal effects caused by pesticides, accumulation of metabolites in water bodies, and their inhabitant fish, may contribute to increased development of breast cancer in MS. The high proportion of water bodies in the state and the high consumption of fish and high saturated fat food preparation, may add to carcinogenic promotion and the development of breast cancer (Solly and Shanks, 1974; Roddetzke et al, 1993; Wolf, 1993; Marshall, E. 1993; Schafer et al, 2004, PAN, 2007).

Pesticides may have adverse effects on aquatic life, however, the U.S. Environmental Protection Agency (IARC, 1979; USEPA, 1987; IPCS, 1991) has established maximum standard contaminant levels for aquatic life for only very few current-generation pesticides. As a result of increasing pest resistance, combining pesticides, increasing applications, substitution of more expensive, toxic, or ecologically hazardous pesticides has occurred more frequently. In addition to the problem of pesticide resistance, millions of dollars worth of crops have been lost as a result of improper pesticide application (Benbrook et al. 1996). Consequently, more and more farmers and other pesticide users are seeking to better target their use of pesticides and implement pesticide-use-reduction and alternative strategies. Most pesticides including chlorinated hydrocarbons are animal carcinogens and may increase the risk of breast and other cancers in women (Donna et al, 1989, Hunter and Kelsey, 1993; PAN, 2007).

The economic value of the pesticide market is quite significant. In 1995, almost one billion pounds of conventional active pesticide ingredients were used in the United States, amounting to approximately 4.6 pounds of pesticides per person. About three-fourths of the total annual pesticide use is for agriculture (Aspelin et al. 1997). Most of the pesticides used in the Mississippi River Basin are herbicides used for weed control (Gianessi and Puffer 1990). More than 40 pesticides and pesticide degradation products were detected during 1987--92 in water samples collected from the Mississippi River or from the mouths of large rivers that flow directly into the Mississippi (Goolsby and Battaglinl993). Many pesticides are highly persistent in stream water and in reservoirs, but very little specific information is available on pesticide half-lives in natural water bodies (Duell et al, 2001, Abdalla, 2003). The information that is lacking about pesticides include the level of toxicity, health effects resulting from specific exposure or combination of exposures, and complete test data on the 270 varieties of pesticides

Breast cancer is the most prevalent cancer in women in the US where estimates are that 40,730 will die from the disease in the year 2008 and a total of 271,530 will be diagnosed with the disease (ACS, 2008). During the period of 1950 to 1989, the incidence of breast cancer increased by 53% and since 1940, mortality rates have been increasing by about 1% a year in industrialized countries (Davis et al. 1997). The cause is unknown for 70% of the cases and the remaining 30% have known risk factors, the most significant are lifetime exposures to female hormones (estrogens), which play a critical role in the development of breast cancer. According to the American Cancer Society (ACS 2008) breast cancer case estimates were at 1,630 with 440 deaths in Mississippi for the year 2008.

Since there wasn't a clear understanding of the role of agricultural occupational exposure to pesticides and breast cancer mortality, the objective of this study was to analyze the association of pesticide exposure and breast cancer mortality in Mississippi. Specifically, to determine if there are associations between pesticide exposure and risk of breast cancer mortality, and whether or not the associations persist over time, after controlling for socioeconomic status and access to early detection of breast cancer. In this regard, we will test four hypotheses: 1) there are linear associations between level of pesticide exposure (number of acres planted; proxy) and female's breast cancer mortality rates in Mississippi; 2) the association between level of pesticide exposure and breast cancer mortality rate/level will persist after controlling for race; black women will be more likely to have significant correlations between exposure and breast cancer mortality; 3) the association between level of pesticide exposure and breast cancer mortality rate will vary by type of crop planted; and 4) there is a persistent association over time between level of pesticide exposure and female's breast cancer mortality in Mississippi.

Material and Methods

Population and Sample Selection:

The samples for the analysis consisted of women who died from breast cancer between 1970-1994 and 1996-1999 in Mississippi. The first set of data was obtained from the National Cancer Institute Web site, aggregated by State Economic Area (SEA), and the second set of data was obtained from the Mississippi Cancer Registry, aggregated by county. The statistical power for both samples differed; the first sample consisted of only nine units (SEA), whereas the second sample consisted of the 82 counties in the state. The State of Mississippi is divided into nine State Economic Areas (SEA): Jackson, Greenville, Yazoo, Vicksburg, Corinth, Columbus, Meridian, Hattiesburg, and Biloxi.

Initial analysis using these nine categories indicated that two SEAs did not produce findings due to less than five counties per SEA having planted crops (Jackson and Biloxi). A decision was made to collapse these two SEAs into other larger SEAs in order to obtain significant findings across all units. It was felt that in doing this, the groupings were not biased as the water topology; economic areas and concentration of planted crops did not vary. The only essential difference or bias introduced is that the Jackson SEA, which is essentially Hinds County, contains the great majority of primary care providers (n=447). However, the final analysis was not controlled by this measure. Statistical correlations (data not shown) indicated no significance between the numbers of primary care providers and the breast cancer mortality rate in a given SEA. Given the relatively small sample sizes of the unit of analysis (modified SEA), A significance level of .05 was defined as statistically significant, and borderline significance at the 0.05 < p < 0.10 was also noted in the findings.

Statistical analysis to test the hypothesis included descriptive tabulations with accompanying graphs, correlation analysis by race, and controlling by SEA. The independent variables for both time periods were those representing the pesticide exposure (total acres of planted crops, and number of acres planted for each type of crop). The dependent variables for both time periods were the female breast cancer mortality rate (per 1000,000 age-adjusted to the 1970 U.S. Census population). Both independent and dependent variables were continuous measures with a normal distribution curve. Therefore, the statistical analysis for a normally distributed continuous variable, Pearson correlation, was used for the main findings. The analyses were controlled by three nominal measures: race (white, black), SEA (7 geographic areas), and type of crop planted.

Data files were constructed by downloading data from websites into Excel files, or data was manually entered directly into excel files. Accuracy of data retrieval and entry was achieved by repeated visual inspection. Analyses were performed using the Statistical Analysis Software (SAS, version 8e). Data integrity was preserved by calculating and comparing averages and standard deviations in both software packages, SAS and Microsoft Excel. In addition, the Geographic Information System (GIS) software was used to allow visualization of information in a manner that reveals relationships, patterns, and trends not visible with other systems.

Geographical Information System GIS:

GIS is an information system that is designed to work with data referenced by spatial or geographic coordinates GIS is a database system with specific capabilities for spatially referenced data, as well as a set of operations for analysis of the data. (Dunn, 1987; Selvin, 1988; Star and Estes 1990; Marshall, 1991, Haslett, 1992; and Walter, 1993).

Spatial Analytic Techniques:

Spatial variation in health related data is well known, and its study is a fundamental aspect of epidemiology. Representation and identification of spatial patterns play an important role in the formulation of public health policies. The spatial analytic techniques used here are limited to those involving a graphic exploratory analysis of data. These spatial techniques include point, line, area, surface and contour patterns.

Area Patterns:

The first stage of data analysis is to describe the available data sets through tables or one-dimensional graphics, such as the histogram. For spatial analysis, the obvious option is to present data on maps, with the variable of interest divided into classes or categories, and plotted using colors or hachure within each geographic unit, known as a choropleth map (Marshall, 1991). A literature search on spatial analysis revealed that among seventy-six papers, fifty-four percent used the choropleth map as an analytical tool (Walter et al,. 1993).

The most common maps use pre-specified classes of health events, or the mean and standard deviation of their distribution. The maps are usually represented using administrative boundaries such as counties, municipalities, health districts, and so on, where data are usually collected. Major variables used for area pattern analyses are incidence rates, mortality rates, and standardized mortality ratio (SMR). The latter is most common in health atlases (Walter et al,. 1993). At times, area pattern analysis uses statistical significance rather than raw data.

Area pattern analysis also uses empirical Bays estimates of the relative risk (Marshall,. 1993). When two or more health-related variables are available in each area unit, multivariate analysis, can synthesize the information. Measures of coherency between two variables were used as the variable of interest to explain the spatial pattern of disease or event occurrence.

Surface and Contour Patterns:

Data of epidemiological or public health interest often occur as spatial information during each of several time epochs. The analytical techniques described previously require the pooling of information in administrative areas with well defined geographic boundaries (e.g., counties, municipalities, and health districts), and the presentation of the spatial process with maps constrained to them. These maps are often unable to capture health problems at the locality or sub-county level. As well, epidemiological variables do not necessarily recognize political boundaries. To overcome the limitation of administrative regions for mapping, surface, and contour pattern analysis presents an alternative by representing the distribution of the health event. The advantage of this spatial analytical technique is that the variable under study is treated as a continuous process throughout the region.

Surface and contour analysis assumes that a health event is a continuous process observed at a set of geographic points, known as sampling points. Using the x and y coordinates of these sampling points, with an associate z value corresponding to the health event, the estimated spatial relative risk is depicted as a three-dimensional map or surface. The contour map, known as an iso-line or iso-pleths, is the projection of the surface in a plane, and corresponds to constant z values of the defined surface.

Although these techniques may overcome the limitations of political boundaries and help in the representation of spatial processes collected as point data, they are often not used by health researchers. This may be due to the very fact that they lose the geopolitical information known to the researcher. One possibility, which is already available in some G1S packages, is the capacity of overlaying the geographic map of the region with one of these analytical techniques.

Study Strengths and Limitations:

This type of study uses an ecological study design. (in this case, this is a correlation ecological study). The strengths of this type of study are that large areas, such an entire state, can be studied, assessing relationships over long periods of time and it is not labor-intensive as archival data can be easily used. Thus, it is very cost-effective to carry out this type of study. However, casual and temporal association cannot be determined. For instance the concentration of pesticides cannot be determined exactly by region. Also, the onset of the accumulation of pesticides in specific communities and penetration into human systems cannot be determined. Furthermore, other endogenous and exogenous factors, such as diet, immunity, exposure to synthetic estrogens and other carcinogens that may contribute to incidence of breast cancer are unknown and beyond the scope of this study. Individual level data are needed to be able to determine some of these additional exposures in relation to breast cancer. Finally, a prospective design would be needed to be able to determine a causal relationship between exposure to pesticides and breast cancer events in general

RESULTS AND DISCUSSION

The objective of this study was to determine if the bioaccumulation hypothesis could be applied to study the long-term health effects of harmful environmental agents. Specifically, to determine if there are associations between pesticide exposure and risk of breast cancer mortality and whether or not the associations persist over time after controlling for socioeconomic status and access to early detection of breast cancer.

Data presented here include graphic representations and descriptions of the water topology and crops planted using GIS software. It also included a description of the patterns of breast cancer mortality in the U.S. and in Mississippi using geospatial maps and data tabulations. A geospatial map showing the distribution of primary care providers in the state is also presented to emphasize the relationship between access to early detection of breast cancer and subsequent reduction of breast cancer disparity/mortality. Results on four tested hypotheses were also included.

Breast Cancer Patterns in the U.S:

In reference to the national GIS maps from the NCI web site for the graphic representation of the spectrum of breast cancer mortality for 1970-1994 (data not shown). The published breast cancer maps illustrate the national distribution of breast cancer mortality for black and white women. This pattern shows that white women die from breast cancer in all areas of the country with the greatest mortality rates in the northeastern part of the country. In comparison, a pattern for black women is concentrated in the southeast regions of the country with the highest mortality rates for this group. This finding is consistent with national incidence patterns of breast cancer, in which case, white women have much higher incidence of this cancer compared to black women. However, due to lack of early detection, black and other minority women tend to get diagnosed at later stages of the disease and therefore, to have poorer survival rates (www.nci.gov).

Table 1. Breast Cancer Incidence in the State of Mississippi.

Table 1: Mean Age-Adjusted Breast Cancer Incidence Rates and Standard
Deviations Mississippi, 1970-1994 & 1996-1999 (N = 82)

Year/Race           Mean   S.D.  Minimum  Maximum

1970-1994  Whites   20.8   4.0    6.0      34.7
           Blacks   22.7   6.5    9.0      45.4
1996-1999  Whites  113.9  40.9    0       200.0
           Blacks  101.5  48.0    0       285.1

Table 1 illustrates the increase in lifetime prevalence of breast cancer in Mississippi women, with the rate of breast cancer mortality more than five-fold in 1996-1999 compared to 1970-1994 (110.6 vs. 20.8 for white women and 107.6 vs. 22.7 for black women). The rate for black women in both time periods had the highest rates (maximum values) across both time periods (285.1 vs. 45.4 during 1996-1999 and 200.0 vs. 34.7 during 1970-1994). It should be noted that statistics such as mortality and incidence rates are more stable when aggregated over several years. While the time period of 1970-1994 brings great statistical stability, it is not comparable to the fewer years of data in the period of 1996-1999. However, any bias that this difference may carry is not felt to interfere with the correlation analysis as the mathematical formula involved in calculating the Pearson correlation uses the slope of the change in one axis over the other, and not the magnitude of the difference between the two.

The age-adjusted incidence rate for Mississippi women for two periods of time is shown in table 1. For 1970-1994 period, there was a slightly higher rate difference between black and white women (22.7 vs. 20.8). During the period of 1996-1999, the white women had a higher incidence than black women, but it could be generally noticed that the rate was almost five fold higher compared to the first time period. This points out to a trend for increased incidence with time for generally unknown reasons.

State Economic Areas (SEA):

Figure 1 shows the Mississippi State Economic Areas (SEA). The state is divided into nine areas. For purposes of gaining statistical power, the Jackson SEA was combined with the Yazoo SEA. and the Hattiesburg SEA with the Biloxi SEA. This yielded seven areas in the analysis. The map also shows the eighty-two county boundaries and labels in the State.

[FIGURE 1 OMITTED]

  Figure 2 shows the State's breast cancer mortality rates/SEA
  for both white and black women during 1950-1969 & 1970-1994 and
  1970-1994 respectively. The data showed rate differences within
  different SEA

[FIGURE 2 OMITTED]

Topographic predisposition of Mississippi to pesticide contamination:

Water Topology of Mississippi:

Figure 3 show the water topology and the Mississippi major lakes, rivers and catfish farms. The figure shows the feasibility of these scattered water bodies in contribution to the transport of pesticides (planted acres as a proxy of source) from one place to another across the state through water runoff. As can be seen on the map, the catfish farms are concentrated particularly in the Mississippi delta region of the state. The accumulation and transport of pesticides across the water topology and the introduction of pesticide metabolites into the food chain, especially from catfish farms, poses a great risk for long-term health effects.

[FIGURE 3 OMITTED]

Cropland in Mississippi

  Figure 4 shows the total number of planted acres by SEA;
  Greenville, Yazoo, Jackson, Vicksburg and Columbus SEAs showed
  the highest number of planted land.

[FIGURE 4 OMITTED]

Figure 5 is a GIS overlay map showing the total planted acres in Mississippi for the year 1997-2001 and the age-adjusted breast cancer mortality rate for each of the years 1996-1999. The SEAs with the highest number of acres of crops planted include Greenville, and parts of Yazoo, Vicksburg and Meridian. Scott, Benton, Lee, Holmes and Stone counties have the highest number of planted crops in the state. The overlay shows county mortality ratios and planted acreage as depicted in a pie chart for comparison.

[FIGURE 5 OMITTED]

Table 2 shows the analysis of different crops planted in the state, showing the highest percentages for rice followed by cotton and catfish. The table present descriptive data for the independent, dependent and control variables studied. The table also shows the percent distribution by type of crop, mean planted crops in acres, standard deviations, and range (minimum, maximum). In the state of Mississippi, rice (6,231 acres), cotton (960 acres) and catfish (800 acres) account for the majority of the distribution of acres of crops planted, out of a total of 7,321 acres planted between the years 1997-2001. The dispersion (S.D. and range) of acres among the SEAs varies greatly with several areas not having at least one of the crops planted and some having several types of crops planted in the SEA.

Table 3 Shows the Pearson correlation and p-value analysis for crop correlations with SEA and race. There are combinations of different correlations with regards to the three parameters.

Table 2: Percent Distribution and Mean Number of Planted Crops (Acres),
Standard Deviations, and Range Mississippi. 1997-2001 (N = 7,321)

Type of Crop  % N = 7,321    Mean      S.D.    Minimum   Maximum

Corn               1.9       139.1     46.6       0        137.7
Cotton            13.1       959.1    361.8       0        960.0
Rice              71.5     5,334.5  2,083.0       0      6,231.0
Soy                0.5        36.6     12.4       0         33.1
Sorghum            1.4       102.5     29.0       0        101.1
Wheat              0.7        51.2     22.4.0     0         52.5
Catfish           10.9       798.0     88.3       0        800.0

Table 3: Association of Breast Cancer mortality and Planted Acres

SEA          Independent  Dependent  Pearson (r)  P-value

Greenville   T Plants     Blacks       0.573       0.066

Greenville   Catfish      Blacks       0.648       0.031

Yazoo        Rice         Whites-94    0.674       0.030

Vicksburg    T Plants     Whites-69    0.607       0.062

Corinth      T Plants     Blacks       0.667       0.049

Greenville   Cotton       Blacks       0.419       0.050
             Rice         Blacks       0.678       0.035
             Rice         Whites       0.888       0.049

Hattiesburg  Soy          Blacks       0.720       0.050

Yazoo        Wheat        Blacks       0.570       0.030
             Soy          Blacks       0.506       0.072
             Catfish      Whites       0.592       0.044

Columbus     Catfish      Blacks       0.720       0.050
             Soy          Whites       0.792       0.033

Health Care and Income Disparities:

Distribution of Primary Care Providers in Mississippi:

The impact of access to health care and to early detection of cancer is generally linked to access to a regular primary care provider. As seen in Figure 6, the distribution of primary care providers in Mississippi is concentrated in the metropolitan areas of Hinds county (state capital) and along the Gulf Coast. Most of the other areas, especially the Mississippi delta where most black Mississippians live and work in the agriculture fields, have very few providers with some counties having less than five providers for ten to forty thousand persons Table 4 provides some useful information on the economic and access indicators for the people in Mississippi. The average household income, percent of people living below the poverty line and the number of primary care providers are well below national averages.

[FIGURE 6 OMITTED]

Figure 6 shows the distribution of health care providers within the 82 counties in MS. There is a clear disparity within counties and hence SEAs regarding the level of health care especially within rural Mississippi.

Table 4: Mean Income, Percent Poverty and Primary Care Providers by
County Mississippi, 1995 (N=82)

Factor                       Mean    S.D.   Minimum  Maximum

Income ($)                  23,573  5,011   14,382   41,241
Poverty (% below)           24.1     7.9     8.8       44.9
Primary Care Providers (#)  20.0    51.0     0.0      447.0

Findings from Hypothesis Testing

Hypothesis 1:

There is a linear association between level of pesticide exposure (number of acres planted) and breast cancer mortality rate. The findings indicate that this hypothesis is supported (overall trend in Table 3 and fig 5). There is a linear association between level of pesticide exposure and breast cancer mortality rate (although not statistically significant across the board (SEA, race and type of crop). In both time periods that were studied (1970-1994 and 1996-1999), there were both positive and negative but linear associations (p<. 05) for rice (r = 88819, r = .67844, r = 0.63370) soy (r =0.72034, 0.79209), and a lower Pearson Correlation for cotton (r = 0.41878). Both time periods had positive linear associations, while only the period 1970-1994 had some unexplained negative linear associations (data not shown), which may possibility be an indication of spurious (by chance) events?

Hypothesis 2:

The association between level of pesticide exposure and breast cancer mortality rate persists after controlling for race; black women will be more likely to have significant correlations between exposure and cancer. The findings in support these hypotheses were shown only for the second time period. As shown in Table3 and figure 4 and 5, the association between acres of crops planted by type of crop and breast cancer mortality rate is linearly and significantly associated but differs by race and type of crop planted, and by SEA.

For the years 1970-1994, there are moderate statistically significant correlations between the breast cancer mortality rate for the black and white females and the use of pesticides in the State of Mississippi. The association varies by SEA, race and type of crop. Black females are at increased risk of breast cancer mortality rate in areas with planted acres for catfish, wheat, soybeans and rice. White women are at risk in areas planted with rice and catfish; all of which are food crops and require massive exposures during their cultivation and harvest, in contrast to cotton which is not a food crop and mostly dried in its cultivation. The association persisted after controlling for race only in the period 70-94; the association was significant for the rice crop for both white (r =0.88818) and black women (r =0.67844). Black women were more likely to have significant correlation with the total number of planted acres of soy (r =0.72034) for 70-94, as well as for cotton (r =0.41878) and rice (r =0.6780), and wheat (r=o.5700). White women were having significant correlation with the soy crop (r = 0.79203) for 70-94, Catfish (r=0.59200 and rice (r = 0.88820). White women have a significant association only for total plants (r = 0.60709) in 1996-1999. Variations within SEAs existed (table 3).

Hypothesis 3: The association between level of pesticide exposure and breast cancer mortality rate will vary by type of crop planted.

This hypothesis was not supported. The associations varied by type of crop planted in addition to variation by race, as shown in Table 3 and figures 4 and 5 which were discussed in findings for hypothesis 2. The association between level of pesticide exposure and breast cancer did vary by type of crop planted, only for wheat, rice, soy and catfish crops. There was a persistent association per time period (70-94 to 96-99) between level of pesticide exposure and breast cancer mortality for rice and soy crops. In 1970-1994 the association for both white and black women living in Greenville SEA for the rice crop was statistically significant (table 3).

Hypothesis 4:

There is a persistent association over time between level of pesticide exposure and breast cancer mortality. This hypothesis is supported. There is a persistent association over time, in 1970-1994 and 1996-1999 data (Table 3 and figures 4 and 5). Table 3 shows the correlation coefficients (r) between crops, planted acres and age adjusted breast cancer mortality rate by State Economic Area for the period 1970-1994. There is a high correlation between the rice crop and the white female breast cancer mortality rate in Greenville SEA and a moderate correlation with the black females for the same area. There was a positive correlation between the soy crop and the black females in the Hattiesburg/Biloxi SEA. A positive correlation was found between the black females and wheat in Yazoo SEA, and finally a positive correlation between the soy crop and white females in Columbus SEA.

Table 3 also shows similar findings for the period 1996-1999. There is a positive correlation between the rice crop and the black females in Greenville SEA, a positive correlation between the soy crop and the white females in Columbus. SEA and finally, there is a positive correlation between black females and the soy crop in Yazoo SEA.

DISCUSSION

An ecological study design model was used to analyze secondary data to establish correlations and test four hypotheses related to pesticide exposure and the risk of breast cancer mortality in the state of Mississippi. Findings from the study showed moderate correlations between total planted acres as a proxy of pesticide use, as well, positive correlations using Pearson analysis model (SAS software) were established for different crop types, and race for pesticide use/exposure and the risk of mortality from breast cancer in Mississippi women. Disparities in income levels and health care provision (providers within the 82 counties ranged between 0-447 and there were 24% of the population who have income levels below the poverty line; table 4 and Figure 6) were also established from data analysis. Studies supporting the role of environmental exposure to chemicals and the risk of cancer were published by Mitra et al, 2004. Mitra and Faruque, 2004 did a similar study in 82 counties from the state of Mississippi relating chemical emissions to breast cancer incidence, where they have found statistically significant correlations in support of our findings. Even though emissions offer direct exposure potential, their volatility and dilution by ambient air, will generally reduce the effective concentrations that may be involved unlike our study where the persistence of pesticides in soil, plants, food and water make a stronger argument for occupational/residential exposure and related consequences (Abdalla et al, 2003, PAN, 2007)..

The Mississippi River basin contains the largest and most intensively farmed region in the nation. The intense use of pesticides is of concern because of potential adverse effects on the quality and use of water resources (Gianessi and Puffer 1991). A number of studies have described the ability of technical pesticides (eg.chlorodane) and its metabolites to disrupt endocrine pathways. Disturbance of the endocrine system may occur through changes in the activity of liver microsomal enzymes that are important in the metabolism and degradation of ovarian and thyroid hormones. Endocrine disruption may also occur at the level of target tissues including the breast (Marshall, 1993; Anderson, 2002; and Fan et al, 2007).

Many chemicals are essential for life and are beneficial, while exposure to other chemicals can be harmful and affect our health. Some chemicals need to be activated by enzymes in the body to become cancer-causing chemicals (Carcinogens). A number of chemicals also pose no cancer risk, while others may act as beneficial agents. It's impossible to make generalizations about environmental chemicals. Much of the concern about whether pesticides affect breast cancer risk stems from observations of higher rates of cancer in male workers with high exposure to pesticides. However some scientists have found higher cancer rates in farmers exposed to certain pesticides (Flower et al, 2004; Fan et al, 2007; Lee et al, 2007).

Many chemicals have to be activated in the body to become carcinogens. Some people have differences in certain genes that control these activation pathways. This is an example of a gene-environment interaction. More research needs to be done to identify important gene-environment interactions. This will help identify groups of women who may have a higher breast cancer risk if they are exposed to certain chemicals (LWVEF, 1989; IPCS, 1991, and Pereira et al, 1997).

If a higher level of the active form of the carcinogen exists, this may put individuals at greater risk for developing certain cancers, including breast cancer. For example, women with high body levels of environmental chemicals called polychlorinated biphenyls (PCBs) usually do not have a higher risk of breast cancer. However, in one study breast cancer risk was higher in a group of women who had both high level of PCBs and variations in the activation of a gene called CYPIAI (Moysich et al. 1999). Chlordane is extremely persistent in the environment (Phillips et al. 1997). Chlordane is one of the chemicals that had been most widely used for termite control. Adipose tissue level of chlordane rose in the week following ingestion of a solution of technical chlordane, indicating tissues redistribution of this chemical.

We are exposed to thousands of naturally occurring and synthetic chemicals over our lifetime. Many chemicals are essential for life and are beneficial, while exposure to other chemicals can be harmful and affect our health. There are many ways our bodies can be exposed to chemicals. This includes exposure in air, food and beverage, and through skin contact. Fetuses can be exposed to chemicals that cross the placenta during pregnancy. Some environmental contaminants can pass from a mother's body to an infant through breast milk. Certain chemicals can be stored in the fat of fish or animals, becoming more concentrated as they pass up the food chain (Theo-Calborn et al, 1993; PAN, 2007).

Factors consistently associated with a higher breast cancer risk are called established risk factors. Established risk factors include progressing age, starting menstrual periods earlier in life, late menopause, having a relative (mother or sister) with breast cancer, and past exposure to breast disease. Breast cancer rates vary widely in different parts of the world. Breast cancer rates are much lower in Japan, China, Africa and India (IARC, 1979; Globecan 2000). It is not clear why there are geographical differences in breast cancer rates. Studies of breast cancer rates of Japanese women who migrate to the US suggest an environmental influence on the risk of breast cancer. Within one or two generations, the breast cancer rate of descendants of migrant Japanese women increased and became similar to the higher breast cancer rate in western women (Shimizu 1991). Results of studies on twins in Scandinavia also suggest that a woman's environment plays a significant role in determining her breast cancer risk, and that environmental factors play a major role in determining her breast cancer risk, and that environmental factors lay a major role in determining the risk of breast cancer (Lichtenstein 2000).

There are very few studies that have evaluated whether female farmers have a higher risk of breast cancer (Blair 1995). The results of this small study suggest that breast cancer risk may be increased in some female farmers with high exposure to pesticides. Early reports suggested that women with high levels of DDE in their blood or fat had a higher risk of breast cancer. However, the majority of the more recent, well-controlled studies have looked at breast cancer risk in white women living in North America and Europe. For dieldrin, and other organochlorine pesticides, there are too few studies in women to make conclusion of whether or not body levels are associated with breast cancer risk (Snedeker 2001). For example, preliminary research suggests that occupational exposure to the environmental estrogen 4-octylphenol is associated with a higher risk of breast cancer (Aschengrau et al. 1998). EPA is in the process of validating screening tests for more than 865 pesticide active ingredients and about 150 high volume industrial chemicals for endocrine-disrupting effects. Exposure to cancer-causing chemicals when the breast is developing may affect breast cancer risk later in life. Studies have shown that breast development may affect breast cancer risk later in life. During pregnancy breast cells undergo changes making them more mature (BCERF, 1998).

There are many types of exposures on the farm that may affect cancer risk, including exposure to pesticides, solvents, fuel exhaust, and toxins from molds that from in stored crops. However some scientists have found higher cancer rates in farmers exposed to certain pesticides.

In a study of North Carolina female farmers, overall breast cancer rates were lower in women who lived or worked on a farm compared to women who did not work or live on a farm. It has been suggested that these female farmers/farm residents may have lifestyles or risk factors that could have reduced their risk of breast cancer (earlier age of first pregnancy, higher number of pregnancies, less likely to smoke or drink alcohol, higher level of exercise). However in this study, one group of females on farms who did not wear protective clothing or gloves when applying pesticides had two-fold higher risk of breast cancer compared to women who took proper precautions. The results of this small study suggest that breast cancer risk may be increased in some female farmers with high exposure to pesticides. This study illustrates the importance of reducing exposures to pesticide in the workplace (Duell et al 2000).

Breast cancer takes many years to develop- often up to thirty or more years-because of the many changes that must occur before a normal cell becomes a cancerous cell that divides out of control. Scientists are concerned that some environmental chemicals can either mimic the effect of hormones or growth factors, or affect how fast the body makes or breaks down these hormones. Through these actions, an environmental chemical could affect the delicate balance that controls cell division. More than half of all breast tumors depend on estrogen for growth. Chemicals that mimic the effect of estrogen may play a role in supporting the growth of estrogen-dependent breast tumors. For example, preliminary research suggests that occupational exposure to the environmental estrogen 4-octylphenol is associated with a higher risk of breast cancer (Aschengrau et al. 1998).

In addition to concerns about how environmental estrogens may affect breast cancer risk, there is also evidence that these"xenoestrogen" can affect reproduction in wildlife and possibly in humans (Fan et al. 2007). Because of these concerns, the US Congress passed the Food Quality Protection Act in 1996. This legislation mandates that all pesticide active ingredients be tested for their estrogen mimicking and other hormone disrupting effects. Childhood and adolescence are critical periods of breast development. Exposure to cancer-causing chemicals when the breast is developing may affect breast cancer risk later in life. Studies have also shown that breast development may affect breast cancer risk later in life. Studies have shown that the developing mammary glands (breast tissue) of young rats and mice have bud-like structures composed of rapidly dividing cells. These dividing immature breast cells are more susceptible to the damaging effect of cancer-causing chemicals. During pregnancy breast cells undergo changes making them more mature. Mature breast cells appear to be more resistant to the effects of carcinogens, and can more easily repair damage caused by cancer - causing chemicals (BCERF. 1998).

Health disparities research is at the forefront of the health care and public health agenda. There are various studies that support the socioeconomic and race disparities within minority, black and white women with regards to breast cancer and special reference to pesticides and farmland setting as risk factors (Krieger, 1989; Worthing and Hance, 1991; Whitten, 1993; Moormier, 1996; Newman and Alfonso, 1997; Simon and Severson, 1997; Lannin et al, 1998; McCarthy et al, 1998, Hill, 2002; and Lee et al, 2007),. Our findings support the need to address the health disparity problem related to breast cancer in minority women within cultivated lands in Mississippi.

CONCLUSION

For the years 1970 and 1994, there are moderate statistically significant correlations between the breast cancer mortality rate for the black and white females and the use of pesticides in the State of Mississippi. The association varies by SEA, race and type of crop. Black females are at increased risk of breast cancer mortality in areas with planted acres of catfish, wheat, soybeans and rice. For the years 1996-1999 there are moderate statistically significant correlations between the breast cancer mortality rate for the black and white females and the use of pesticides in the State of Mississippi. Black women are at increased risk of breast cancer mortality in areas with planted acres of rice, soybeans, wheat and catfish and White females are at risk in areas planted with rice, catfish and soybean crops

There is chemical persistence in the environment and great potential for continuous human exposure. There are several studies that are currently being conducted to address the research needs to determine whether environmental contaminants, including organochlorine pesticides and other chemicals increase the risk of breast cancer among women. More studies are needed to examine if there is an association between the levels of organochlorine residues in serum and increased risk of breast cancer among African American females.

There are lots of available areas to test and examine the relation between pesticide use and breast cancer mortality. The State of Mississippi local agencies need to address this issue and advance the research agenda in this area to estimate this risk of breast cancer and its impact on the State population.

As a result of the new mammography and early detection activities that has been added to different communities in Mississippi, the mortality rate is expected to decrease over time, due to diagnosis at earlier stages and longer survival rates. However, from the results shown in this study, there is evidence of a persistent association between pesticide exposure and breast cancer mortality that may be explained by the bioaccumulation hypothesis where over time, pesticides will make their way into the soil, food chain and finally human fatty tissues and the threat may not be limited to farm exposures.

ACKNOWLEDGEMENTS

We would like to thank Dr. Mark G. Hardy, Dean, College of Science Engineering and Technology, Dr. Paul B. Tchounwou, Associate Dean, College of Science engineering and Technology and Dr. Gregorio Begonia, Interim chair, Department of Biology, Jackson State University for cordial support.

LITERATURE CITED

Abdalla, M.H., Gutierrez-Mohamed, M.L., and Farah, I.O. Association of Pesticide Exposure and Risk of Breast Cancer Mortality in Mississippi. Biomed Sci Instrum. Vol. 39:397- 401, 2003 (ISA Vol. 437, paper #2003-068 / 0067-8856/2003).

American Cancer Society 2008, Cancer facts and figures. [http://www.acs.org].

Ames, B.N 1979. Identification of Environmental chemicals causing mutations and cancer. Science 204, 283-292.

Anderson, H.R., Vinggaard, A.M., Rasmussen, T.H., Gjermandsen, I.M., and Bonefeld- Jorgensen, E.C. 2002. Effects of Currently Used Pesticides in Assays for Estrogenicity, Androgenicity and Aromatase Activity in Vitro. Toxicol Appl. Pharmacol. 197:1-12 Aschengrau,

A., Rogers, S., and Ozonoff, D. Perchloroethylene-Contaminated Drinking Water and the Risk of Breast Cancer: Additional Results from Cape Cod, Massachusetts, USA Journal article by; Environmental Health Perspectives, Vol. 111, 2003

Aspelin. A.L.., 1997, 2-3. (Washington, D.C.: U.S. Environmental Protection Agency, Office of Prevention, Pesticides and Toxic Substances, Biological and Economic Analysis Division.

BCERF, 1998. Cornell University Program on Breast Cancer and Environmental Risk Factors in New York State (BCERF)* Critical Evaluation # 8

Benbrook, C. 1996. Pest Management at the Crossroads, Consumers Union.

Blair and Zaham. 1995. Environ. Health prospect. Prospect vol. 103[supply. 8], pp 205-8.

Brody, J.G., and Rudel, R.A. 2003. Environmental Pollutants and Breast Cancer. Environ.Health Perspect: 111: 1007-1019.

Broeders, MJM, Verbeek ALM. 1997. Breast cancer epidemiology and risk factors. 41:QJ Nucl Med: 179-88.

Charlier, C., Albert, A, Herman, B., Hamoir, E., Gaspard, U., Meurisse, M., and Plomteux, G. 2003. Breast Cancer and Serum organochlorine residues. Occup. Environ. Med.60:348-351.

Gianesse, L.P., and Puffer, C.A., 1991, Herbicide use in the United States - national summary report (rev. ed.): Washington D.C., Resources for the Future, 128 p.

Hunter, D.J. and Kelsey,. K.T. 1993. Pesticide residues and breast cancer: The Harvest of a Silent Spring? Journal of the National Cancer Institute Vol. 85: 598-9.

Davis DL, Axelord D Osborne M Telang N Bradlow HI Sittner E. 1997. Avoidable cases of breast cancer: The known, unknown, and the suspected. Ann N Y Acad Sci.

Donna, A. Crosignani P. Robutti F. Betta P. G. Bocca R. Mariani N. Ferrario F. Fissi R. and Berrino F. 1989. Triazine herbicides and ovarian epithelial neoplasms. Scandinavian Journal of Work, Environment and Health 15, 47-53.

Duell, E. J. Millikan R. C. Savitz D. A. Newman B. Smith J. C. Schell M. J. and Sandler D. P.A,. 2001. Population based case-control study of farming and breast cancer in North Carolina. Environ Health Prespect, vol 109 [Suppl.1:35-47,.

Dunn, R. 1987. Statistical mapping. The American Statistician 41: 153-6.Marshall, E. 1993. "Search for a Killer: Focus Shifts From Fats to Hormones. Science Vol. 259. 618-21.

Fan, W.Q., Yanase, T., Moringaga, H., Gondo S. Okabe, T., Nomua, M., Komatsu, T.Morohashi, K-I., Hayes, T.B., Takayanagi, R., and Nawata, H. 2007. Atrazine-Induced Aromatase Expression is SF-1-Dependent: Implications for Endocrine Disruption in Wildlife and Reproductive Cancers in Humans. Environ. Health Perspect doi:10.1289/ehp.9758.

Flower, K.B., Hoppin, J.A., Lynch, C.F., Blair, A., Knott, C., Shore, D.L., and Sandler, D.P. 2004. Cancer Risk and Parental Pesticide Application in Children of Agricultural Health Study Participants. Environ. Health Perspect. 112: 631-635.

Globocan 2000.Estimating the world cancer burden: Int J Cancer. 2001 Oct 15; 94(2): 153-6.

Goolsby DA, Coupe RC, Markovchick DJ. 1991. Distribution of selected herbicides and nitrate in the Mississippi River and its major tributaries, April through June 1991. US Geological Survey Water-Resources Investigations Report 91-4163, 35.

Haslett, J. 1992. Dynamic graphics for exploring spatial data with application to locating global and local anomalies. The American Statistician, 45: 234-43.

Hill, M. 2002. Breast cancer among Mississippi women: a preliminary report of excess mortality in African Americans. Journal od the Mississippi Academy of Sciences Vol. 47 No 3: p. 234.

IARC. 1979. Hexachlorocyclohexane (Technical Lindane and Lindane). In IRAC Monographs on the evaluation of chemicals to humans. World Health Organization: 195-223.

IPCS. 1991. In environment Health Criteria 124: United Nations Environmental program, International Labor Organization, and the World Health Organization. 208.

Krieger N. 1989. Exposure, susceptibility, and breast cancer risk: A hypothesis regarding exogenous carcinogens, breast tissue development, and social gradients, including black/white differences in breast cancer incidence. Breast Cancer Res Treat_13: 205-23.

Lannin DR, Mathews HF Mitchell J Swanson MS Swanson HF Edwards MS. 1998. Influence of socioeconomic and cultural factors on racial differences in late stage presentation of breast cancer. JAMA. 279: 1801-7.

LWVEF 1989: League of Women Voters Education Fund, America's Growing Dilemma. Washington, D.C., League of Women Voters Education Fund, America's Growing Dilemma. P111.

Lee, W.J., Sandler, D.P., Blair, A., Samanic, C., Cross, A.J., and Alavanja, M.C. 2007. Pestcide use and colorectal Cancer risk in the Agricultural Health Study. Int. J. Cancer 121 (2): 339-346.

Marshal, R. J. 1991. A review of methods for the statistical analysis of spatial patterns of disease. Statist. Soc 154(3), 421-41.

Wolff, M.S. 1993. "Blood Levels of Organochlorine Residues and Risk of Breast Cancer_Journal of the National cancer Institute, vol. 85: 648-52.

McCarthy AP, Burns RB Coughlin SS. 1998. Mammography use help to explain differences in breast cancer stage at diagnosis between older black and white women. Ann Intern. Med._ 128: 729-36.

Mitra, A.K., Faruque, F., and Avis, A.L. 2004. Breast Cancer and Environmental Risks: Where is the link?. Journal of Environmental Health, vol. 66:7: 24-32.

Mitra, A.K., and Faruque, F.S. 2004. Breast Cancer Incidence and Exposure to Environmental Chemicals in 82 Counties in Mississippi. Southern Medical Journal: 97: 259-263.

Moormier J. 1996. Breast Cancer in black women. Ann Intern Med 124: 897-905.

Kirsten B. Moysich (2), Peter G. Shields, Jo L. Freudenheim, Enrique F. Schisterman, John E. Vena, Paul Kostyniak, Hebe Greizerstein, James R. Marshall, Saxon Graham and Christine B. Ambrosone, 1999. Polychlorinated Biphenyls, Cytochrome P4501A1 Polymorphism, and Postmenopausal Breast Cancer Risk. Cancer Epidemiology Biomarkers & Prevention Vol. 8, 41-44.

Newman LA, Alfonso AE. 1997. Age-related differences in breast cancer stage diagnosis between black and white patients in an urban community hospital.. J Surg OncolPesticide Action Network International (PAN). 2007. A position on Synthetic Pesticides.A PAN International Position Paper- Working Group I.

Pereira, W.E., and Hostettler, F.D., 1993, Nonpoint source contamination of the Mississippi River and its tributaries by herbicides: Environmental Science and Technology, v. 27, no. 8, p. 1542-1552, doi: 10.1021/es00045a008.

Pereira, W.E., and Rostad, C.E., 1990, Occurrence, distributions, and transport of herbicides and their degradation products in the lower Mississippi River and its tributaries: Environmental Science and Technology, v. 24, no. 9, p. 1400-1406, doi: 10. 1021/es00079a015.

Pereira, W.E., Rostad, C.E., and Leiker, T.J., 1992, Synthetic organic agrochemicals in the lower Mississippi River and its major tributaries--Distribution, transport and fate: Journal of Contaminant Hydrology, v. 9, p. 175-188, doi: 10.1016/0169-7722(92)90057-L.

Pereira, W.E., Rostad, C.E., and Leiker, T.J., 1990, Distribution of agrochemicals in the lower Mississippi River and its tributaries: The Science of the Total Environment, v. 97/98, p. 41-53, doi: 10.1016/0048-9697(90)90229-N.

Phillips P.J., Ria-Murray K. Hollister H. M. and Falnaray E. A. 1997. Distribution of DDT, Chlordane and total PCBs in bed sediments in the Hudson River Basin. NY Earth Science Environ 3: 26 47.

Reddetzke, K. A and Applegate H. G. 1993. Organochlorine pesticides in adipose tissue of persons from EL Paso, Texas. J Environ. Health 56: 25-7.

Rogan WJ, Gladen BC McKinner JD., 1986. Polychlorinated biphenyls (PCBs) and Dichlorodiphenyl Dichloroethene (DDE) in human milk: effects of maternal factors and previous lactation., Am J Pub Health 76: 172-7.

Schafer, K.S., Reeves, M.., Spitzer, S., and Kegley, S.E. 2004. Chemical trespass, pesticides in our bodies and corporate accountability. Pesticide Action Network North America.

Selvin, S. 1988. Transformations of maps to investigate clusters of disease. Soc. Sci. Med. 26, 215-21.

Shimizu, H. Ross R. K. Bernstein L. Yatani R. Henderson B. E. and Mack T. M. 1991. cancers of the prostate and breast among Japanese and white immigrants in Los Angeles coutty., British Journal of Cancer 63: 963-6.

Simon MS, Severson RK. 1997. Racial differences in breast cancer survival: theinteraction of socioeconomic status and tumor biology. Am J Obbtest Gynecology.

Snedeker, S.M., 2001. Pesticides and Breast Cancer Risk: A Review of DDT, DDE, and Dieldri Environmental Health Perspectives, 109(Suppl 1): 35-47.

Solly, S. R. B. and Shanks V. 1974. Polychlorinated biphenyls and organochlorine pesticides in human fat in New Zealand. N.Z J Sci 17, 535-44. SCDPR 1985: State of California, Dept of Pesticide Regulation. SB-950.

Swanson GM. 1994; Lin CS Survival pattern among younger women with breast cancer: the effect of age, race, stage, and treatment. . J Natl Cancer Inst Monogr 16: 69-77.

Theo-Colborn, Frederick, S. Vom Saal and Ana M. Soto. 1993. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ. Health Perspectives, 10:378-84.

Thurman, M., Goolsby, D., Meyer, M.T., Mills, M.S., Pomes, M.L., Kolpin, D.W. (1992). A reconnaissance study of herbicides and their metabolites in surface water of the Midwestern, U.S, using immunoassay and GC-MS. Environmental Science and Technology, 26, 2440-2447.

USGS, 1991. Pesticides in surface waters of the Santee river basin and coastal drainages, North and South Carolina

USATS 1994: U.S Agency for Toxic Substance and Disease Registry.

USEPA. 1987. United States Environmental Protection Agency, Springfield, VA National Technical Information Service.

USDA, 1985. Wetlands Converted to Agricultural Use before December 23, 1985.

U.S. Department of Commerce. 1993. Census of Agriculture. U.S. Gov't Printing Office, Washington, D.C.

Velentagas P, Daling JR. 1994. Risk factors for breast cancer in younger women.. J Natl Cancer Inst Monogr 416: 15-22.

Walter, S. D. 1993. Visual and statistical assessment of spatial clustering in mapped Data. Stat. Med 12, 1275-91.

Whitten, P.L.1993. Chemical revolution to sexual revolution: historical changes in human reproductive development," in Theo Colborn and Coralie Clement.pp. 311-34.

Worthing, C and Hance R. J. 1991. In the pesticide Manual, A World Compendium. C Worthing and Hance, Eds. The British Crop Protection Crop Protection Council: pp 95.

Ibrahim O. Farah * (1), Mohamed H. Abdalla (1), and Elgenaid I. Hamadain (2), (1) Jackson State University, Jackson, MS 39217 and (2) University of Mississippi Medical Center, Jackson, MS 39216

Corresponding Author: Ibrahim O. Farah [e-mail]