|Home | About | Journals | Submit | Contact Us | Français|
Obesity is associated with increased risks of several cancers including, colon, lower esophagus, kidney, female breast, and endometrium. Some studies have associated pesticides with higher risks of cancer in agricultural populations. The interaction between obesity and pesticide use on cancer risk has not been well studied. Using data from the Agricultural Health Study we examined the association between body mass index (BMI) and the risk of cancer at 17 sites, and the interaction between BMI and pesticide use. Pesticide applicators (primarly farmers), and their spouses residing in Iowa and North Carolina were enrolled between 1993 and 1997 and followed through 2005. This analysis included 39,628 men and 28,319 women who provided information on pesticide use, height and weight data, and were cancer-free at enrollment. Of all subjects, 64% were overweight or obese, and 4,432 incident cancers were diagnosed during the follow-up period. We found positive associations between BMI (continuous) and colon cancer among men (Hazard Ratio (HR) 1.05, 95% confidence interval (CI) 1.02–1.09) and breast cancer among postmenopausal women (HR 1.03, 95% CI 1.01–1.06), as well as an inverse association with lung cancer among men who were ever smokers (HR 0.92, 95% CI 0.88–0.96). Men who ever used carbofuran (HR=1.10, 95% CI 1.04–1.17), metolachlor (HR=1.09, 95% CI 1.04–1.15), and alachlor (HR=1.08, 95% CI 1.03–1.13) had significant positive associations between BMI and colon cancer, but non-users did not. Men who ever smoked and used carbofuran had a positive, although not significant, association between BMI and lung cancer, while users of carbofuran had a significant inverse association. These findings, which suggest that certain pesticides may modify the association between BMI and colon and lung cancer risk, should be further evaluated in other populations.
The prevalence of obesity in the U.S. has doubled since the 1980’s. Currently, one-third of U.S. adults are obese and another third are overweight (Ogden, 2007). Obesity has been associated with higher risks of several cancers, including colon, lower esophagus, kidney, gallbladder, female breast, and endometrium (Calle, 2004; Semanic, 2004; Hjartaker, 2008;). The association between obesity and cancer is likely linked to several related factors including diabetes, hypertension, hyperlipidemia, low physical activity, and a high-fat diet (Hjartaker, 2008), and involve several possible biological mechanisms such as increased availability of growth and sex hormones, increased insulin, inflammation, altered immunity, cell proliferation, and angiogenesis (Calle, 2004; Hyartaker, 2008; Bianchini, 2002).
Excess cancer rates, particularly prostate, lymphohematopoetic, brain, stomach and soft-tissue tumors have been reported among agricultural workers and pesticide applicators and manufacturers around the world (Blair 1992; Keller-Byrne, 1997; Khuder, 1997; Khuder, 1999; Alavanja, 2007). Also, several epidemiological studies, have reported associations between specific pesticides and increased cancer risks (Alavanja, 2004; Alavanja Bonner 2005). It has been suggested that lipophilic pesticides, such as organochlorines, may increase cancer risk via their persistent accumulation in adipose tissue, slow release into the blood and peripheral tissues, and eventual impact on the endocrine system (IARC, 1991; Kelce, 1995; Soto, 1995). Because of the accumulation of lipophilic pesticides in adipose tissue, there is speculation that obese persons may have a greater capacity to accumulate and have a higher body burden of these contaminants (Schildkraut, 1999; Pelletier, 2002). Non-lipophilic pesticides have been linked to cancer via pathways closely related to obesity such as hormones and oxidative stress (Diamanti-Kandarakis, 2009; Slotkin, 2009). Furthermore, laboratory studies have reported that toxicants, including pesticides, may contribute to obesity (Baillie-Hamilton, 2002; Heindel, 2003; Irigaray, 2006). Together, these factors support the possibility of a joint effect of pesticides and obesity on cancer risk, but this has not been well studied in humans. In this analysis, we examine the association between body mass index (BMI) and the risk of cancer, as well as the interaction between BMI and pesticide use, in a population of U.S. farmers, pesticide applicators and spouses.
Details of the AHS have been previously reported (Alavanja, 1996). Briefly, the AHS cohort is composed of over 89,000 participants, including licensed private pesticide applicators and their spouses residing in Iowa and North Carolina, and commercial applicators residing in Iowa. In total, 57,310, or 82%, of applicators seeking pesticide licensing in each state, were enrolled in the study between December 13, 1993 and December 31, 1997. In addition, spouses of enrolled private applicators were asked to participate, and a total of 32,346, or 75%, chose to participate. The study protocol was approved by all appropriate institutional review boards.
At enrollment, applicators completed a self-administered questionnaire collecting information on demographics, lifestyle factors, medical histories, ever/never use of 50 pesticides, and detailed use information (number of days and years pesticides were applied, use of personal protective equipment, and application methods) on 22 of the 50 pesticides. A self-administered take-home questionnaire was given to all enrolled applicators to collect additional information, including weight, height, and detailed use on the other 28 pesticides. The take-home questionnaire was returned by 44% of the enrolled applicators. No meaningful differences were found between applicators who completed and did not complete the take-home questionnaire (Tarone, 1997). For spouses, information on demographics, lifestyle factors, medical histories, and ever/never use of the same 50 pesticides was obtained from a self-administered questionnaire (81%) or telephone interview (19%), and height and weight data was only collected from the self-administered questionnaire. Of all participants, 43% were missing height and weight data. Thus, to supplement this missing data, two additional sources, the 5-year follow-up phone interview and driver’s license data from 1985, were used. In Iowa, drivers’ licenses contain height and weight information, while in North Carolina they contain only height. Approximately 70% of the cohort completed the 5-year follow-up phone interview and provided information on current height and weight. After using these additional sources, we had complete height and weight data on 67,947 participants (76% of total cohort). Of these subjects, 76% had height and weight from the take-home questionnaire, and the remaining 24% came from supplemental sources. Thus, it is possible that some cancers cases could have weight reported after their cancer diagnosis if they were diagnosed within the first 5 years after enrollment. Subjects with implausible BMI values were removed (<11 or >70 kg/m2). To have the most current menopause status of female participants, we used data from the 5-year follow-up phone interview. Questionnaires are available on the AHS website (www.aghealth.org/questionnaires.html).
Cancer cases were identified using population-based state cancer registries. Only incident cancer cases diagnosed between enrollment and December 31, 2005 (median follow-up of over 10 years) were included, and participants with any type of cancer diagnosed prior to enrollment were excluded from the analysis. Vital status was obtained from the state death registries and the National Death Index. Participants who moved out of North Carolina or Iowa (based on IRS record) were not followed for cancer occurrence after moving and follow-up times were censored at date of exit from the state.
For this study, we included participants with weight and height data, which totaled 38,190 private applicators, 2,325 commercial applicators, and 27,432 spouses. BMI was calcuated as weight (kg) divided by the square of height (m2), and grouped as follows: underweight (<18.5 kg/m2), normal (18.5–24.9 kg/m2), overweight (25–29.9 kg/m2), obese I (30.0–34.9 kg/m2) and obese II/III (≥35 kg/m2) (WHO, 1998). The chi-square test was used to assess statistically significant differences between BMI groups and selected characteristics.
Associations between BMI and cancer were evaluated separately by sex due to their differences in risk factors for cancer. There were a sufficient number of cases (at least 10 for each sex) to examine 17 cancer sites. Hazard rate ratios (HR) and 95% confidence intervals (95% CI) were calculated using a Cox Proportional-Hazard regression model using age at enrollment and age at exit (minimum of: cancer diagnosis, death, or end of follow-up (December 31, 2005)) as the start and end of follow-up, respectively. A backward elimination procedure was conducted in order to determine the most parsimonious model for each of the 17 cancers evaluated. Initially, models included the following potential confounding factors: race (white, other), state of residence (Iowa, North Carolina), education (high school or lower, greater than high school), smoking status (never, former, current), alcohol intake in past 12 months (no, yes), hypertension (no, yes), diabetes (no, yes), family history of cancer (no, yes), meat intake (≤3 per month, 1–2 per week, ≥3 per week), fruit intake (<1 per day, 1–2 per day, ≥3 per day), vegetable intake (<1 per day, 1–2 per day, ≥3 per day), vitamin supplement (no, yes-not regularly, yes-regularly), and leisure-time exercise (0, <1, 1–2, 3–5, >5 hours per week), as well as menopause status (pre, post) and parity (0,1,2,3,4,>4) for women. Iteratively, the factor with the highest p-value above 0.05 was removed from the model until only those factors that contributed 10% or more change in the HR remained. BMI values were generally normally distributed, thus we also examined associations between BMI and cancer using a linear test of trend with a continuous BMI variable.
For the interaction between BMI and pesticide use, we examined pesticides used by at least 10% of male participants (36 pesticides: 16 insecticides, 17 herbicides, 2 fungicides, 1 fumigant), and 10% of female participants (5 pesticides: 3 insecticides, 2 herbicides). We also examined use of any organochlorine insecticide (aldrin, chlordane, DDT, dieldrin, heptachlor, lindane, toxaphene) and any organophosphate insecticide (chlorpyrifos, coumaphos, diazinon, dichlorvos, fonofos, malathion, parathion, phorate, terbufos, trichlorofon). The cancer sites examined were those with at least 100 cases per sex (bladder, colon, lung, melanoma, non-Hodgkin lymphoma, prostate, and rectal for men; and breast and colon for women). Tests of interaction between ever/never pesticide use and continuous BMI were conducted using a cross-product interaction term in the Cox regression model. We also calculated HRs and 95% CIs for cancer risk in relation to BMI (<25, 25–29.9, and ≥30 kg/m2) stratified by ever/never use of each pesticide using a Cox regression model with the appropriate adjustments for each cancer. Associations with less than 5 exposed subjects per BMI category were not evaluated. SAS version 9.1 (Cary, NC) and the AHS data release PIREL0612 were used to conduct all analyses.
Of the 67,947 subjects, 0.8% were underweight, 34.8 were normal, 43% were overweight, 16.7% were moderately obese (obese I) and 4.7% were highly obese (obese II/III) (Table 1). Compared to normal weight subjects, underweight subjects were more likely to be women, young (<40 years) or old (≥70 years), current smokers, never alcohol drinkers, non-hypertensives, high fruit consumers (≥3 servings/day), and do no or little exercise. In contrast, compared to normal weight subjects, overweight or obese subjects were more likely to be men, middle-aged (40–69), lower educated (high school or less), former smokers, hypertensive, diabetic, low fruit and vegetable consumers, not take vitamin supplements, and do no or little exercise.
We found a statisitcally significant positive linear trend for BMI and colon cancer among men (HR 1.05, 95% CI 1.02–1.09; p-trend=0.005) adjusting for race, education, and family history of colon cancer (Table 2). Overweight men had a 1.26-fold (95% CI=0.86–1.86), obese I men a 1.88-fold (95% CI =1.23–2.91), and obese II/III men a 2.03-fold 95% CI=1.05–3.93) risk compared to normal weight men. For rectal cancer, obese II/III men had a 3.21-fold (95% CI=1.34–7.71) risk compared to normal weight men adjusting for meat consumption, but the test of trend was not statistically significant. We also observed increased, but not statistically significant, risks of stomach and bladder cancers with higher BMI. In contrast, the risks of lung and oral cancers decreased with higher BMI. Stratifying by smoking status, we found a statistically significant test of trend for lung cancer among ever smokers (HR=0.93, 95% CI=0.88–0.97, p-trend=0.001) adjusting for state, race, vegetable consumption, exercise, and pack-years of cigarette smoking. There were only 21 men with lung cancer who never smoked, thus we could not completely calculate risk estimates for this group or the test of interaction between smoking status and BMI. In order to rule out the possibility of weight loss due to preclinical disease, we restricted the cases of lung cancer among ever smokers to those diagnosed 2 or 5 years after enrollment and found that the inverse association with BMI remained (2 years: HR=0.94, 95% CI=0.89–0.99, p-trend=0.01; 5 years: HR=0.93; 95% CI=0.88–0.99, p-trend=0.02).
Among women, we found statistically significant positive linear trend for BMI and all cancer sites combined (HR=1.02, 95% CI=1.01–1.04; p-trend=0.001) adjusting for smoking status, hypertension, taking vitamin supplements, and parity; and specifically for postmenopausal breast cancer (HR=1.03, 95% CI =1.01–1.06; p-trend=0.02) adjusting for diabetes, vitamin supplement, parity, and family history of breast cancer. For breast cancer, overweight postmenopausal women had a 1.22-fold (95% CI 0.93–1.60), obese I a 1.62-fold (95% CI 1.17–2.24), and obese II/III a 1.07-fold (95% CI 0.61–1.88) risk compared to normal weight postmenopausal women. To explore the null association for breast cancer in obese II/III women, we compared the frequency of the factors in Table 1 between the obese I and II postmenopausal women using the chi-square test, and found that only alcohol consumption was significantly different (p=0.02), with 50% of the obese II versus 26% of the obese I women ever drinking alcohol. Stratifying by alcohol consumption, the linear trend for BMI and breast cancer among postmenopausal women was statistically significant among those who never drank alcohol (HR=1.05, 95% CI=1.02–1.08, p-trend=0.003), but not among those who drank alcohol (HR=1.00, 95% CI=0.95–1.04, p-trend=0.85) (p-interaction=0.01). In addition to breast cancer, women with higher BMI had increased, but not statistically significant, risks of pancreatic and kidney cancers, and melanoma. BMI was not associated with colon cancer in women (HR=0.99, 95% CI 0.95–1.04). Also, the risk of lung cancer decreased with increasing BMI, but the test of trend was not statistically significant for all women, ever smokers, or those diagnosed 2 or 5 years after enrollment.
For the interaction between pesticide use and BMI on colon cancer risk in men, we evaluated 22 pesticides that were used by at least 10% of the male study population and for which there were at least 5 cancer cases in each BMI group, as well as any organochlorine or organophophate insecticide (Table 3). Of these 22 pesticides, carbofuran, metolachlor and alachlor had statistically significant (p≤0.05) modifying effects. Specifically, BMI was significantly associated with colon cancer among male users of carbofuran (HR=1.10, 95% CI 1.04–1.17, p-trend=0.002), but not among non-users of carbofuran (HR=1.02, 95% CI 0.97–1.06, p-trend=0.52). Similarly, BMI was significantly associated with colon cancer among male users of metolachlor (HR=1.09, 95% CI 1.04–1.15, p-trend=0.001) and alachlor (HR=1.08, 95% CI 1.03–1.13, p=trend=0.002), but not among non-users of each of these herbicides (No metolachlor: HR=1.01, 95% CI 0.96–1.06, p-trend=0.70; No alachlor: HR=1.01, 95% CI 0.95–1.06, p-trend=0.87). To explore possible confounding by use of multiple pesticides, we examined the correlation between never/ever use of carbofuran, metolachlor and alachlor among male participants and found that they were not highly correlated (metolachlor and alachlor r=0.31, metolachlor and carbofuran=0.18, and alachlor and carbofuran=0.26). We also examined whether each of these pesticides still had statistically significant modifying effects when we further adjusted for use of the other two pesticides, and found that the test of interaction was still statistically significant (<0.05). In addition to these three pesticides, we also found that men who used DDT, fonofos, malathion, carbaryl, permethrin, atrazine, cyanazine, trifluralin, EPTC, imazethpyr, and glyphosate had significant associations between BMI and colon cancer, but non-users of these pesticides did not, and the tests of interaction were not statistically significant. Furthermore, we also observed that men who did not use terbufos, chlorpyrifos, metribuzin, pendimethalin, chlorimuron-ethyl, and methyl bromide had significant associations between BMI and colon cancer, but users of these pesticides did not, and tests of interaction were not statistically significant.
We evaluated the interaction between pesticide use and BMI on postmenopausal breast cancer for four pesticides used by at least 10% of the women in the study population and for which there were at least 5 cancer cases in each BMI group, as well as any organochlorine or organophosphate insecticide. There was no evidence of a significant interaction between pesticide use and BMI on breast cancer risk among postmenopausal women; however, women who did not use malathion, diazinon, carbaryl or any organochlorine insecticide or any organophosphate insecticide had significant associations between BMI and breast cancer, but users did not. The significant associations among non-users of these pesticides remained statistically significant among women who did not drink alcohol, but not among women who drank alcohol.
We also examined the interactions between pesticide use and BMI on the risks of lung, bladder, prostate, and rectal cancers, and NHL and melanoma in men, as well as colon cancer in women. For lung cancer among male ever smokers, we found that the inverse effect of BMI was statistically significant among non-users of carbofuran (HR=0.87, 95% CI=0.82–0.92, p-trend=<0.0001), but not among users (HR=1.01, 95% CI=0.93–1.09; p-trend=0.77) (p-interaction=0.02). We did not find any other statistically significant modifying effects on any of the other associations examined.
In our AHS population of U.S. farmers and their spouses and commercial pesticide applicators, 64% were overweight or obese, which is comparable to the frequency reported in the non-Hispanic white U.S. population, but our population had a higher frequency of overweight (43%) and lower frequency of obesity (21%) than the U.S. population, where 33% are overweight and 31% are obese (Ogeden, 2006). The frequency of overweight and obesity in our population was similar to the frequency reported in an Austrian farming population where 43% were overweight and 15% obese (Dorner, 2004). The mean BMI in our study population (27.6 kg/m2) is very close to the mean BMI reported in New York dairy farmers (26.9 kg/m2)(Jenkins, 2004).
Consistent with previous studies, we found a significant positive association between BMI and colon cancer among men (Samanic, 2006; Rapp, 2005; Samanic, 2004; Moore, 2004). This association was not significantly modified by diet (meat, fruit, vegetable consumption) or leisure-time exercise, which is also in agreement with a previous study of male health professionals that showed an association between BMI and colon cancer independent of physical activity (Giovanuunci, 1995). Obesity has been suggested to contribute to colon cancer via increased leptin that is linked to growth of colonic epithelial cells (Hardwick, 2001), and increased insulin-like growth factor (IGF) that promotes tumor growth (Calle, 2004). We found no association between BMI and colon cancer among women, which has also been previously observed and attributed to waist size being a stronger predictor of colon cancer than BMI in women (Rapp, 2005; Moore, 2004).
We found that men who used metolachlor, alachlor, and carbofuran had significant positive associations between BMI and colon cancer, but non-users did not, with statistically significant interactions. The modifying effects of each of these pesticides remained statistically significant when adjusting for reported use of the other two pesticides. These findings suggest that pesticides somehow modify the carcinogenic effect of obesity, although the current understanding of the effect of pesticides in humans is limited. Alachlor and metolachlor are chloroacetanilide herbicides that have a chorinated amide chemical structure (Kidd). Carbofuran is a carbamate insecticide that inhibits cholinesterase and has been shown to generate reactive oxygen species and induce lipid peroxidation (Kamboj et al., 2006). Based on U.S. EPA guidelines, alachlor is classified as a likely human carcinogen at high doses, but a non-likely carcinogen at low doses, metolachlor is as possible carcinogen, and carbofuran a non-likely carcinogen (Kegly, 2009). None of these three pesticides have been linked to colon cancer in the AHS (Rusiecki, 2006; Lee 2004; Bonner, 2005) or reported in other populations. The effect of these pesticides on obesity has not been reported. We found that users of metolocahlor, alachlor, and carbofuran had slightly higher and statistically significant mean BMIs than non-users, adjusting for age, sex, and education, providing some additional evidence that the metabolism of these pesticides may be linked to obesity-related factors.
We also observed that men who used 11 other pesticides [insecticides: carbaryl (carbamate), fonofos and malathion (organophosphates), DDT (organochlorine), permethrin (pyrethroid)]; herbicides: atrazine and cyanazine (triazine), trifluralin (dinitroanaline), EPTC (thiocarbamate), imazethypyr (imidazolinone), glyphosate (phosphinic)] had positive associations between BMI and colon cancer, but non-users did not, and the interaction tests were not statistically significant. Of these, atrazine, cyanazine, trifluralin, and imazethpyr have an aromatic amine chemical structure (Kidd). Previous studies in the AHS reported that use of two aromatic amine pesticides, imazethypyr (Koutros) and trifluarlin (Kang) were linked to increased colon cancer risk. Aromatic amine compounds, mostly those formed from meats cooked at high temperatures, have been linked to colon cancer (Butler, 2003; Nowell, 2002). It is important to note that we also observed that non-users of six pesticides [insecticide: terbufos and chlorpyrifos (organophosphate); herbicides: metribuzin (triazinone), pendimethalin (dinitroaniline), chlormurion-ethyl (urea); fumigant: methyl bromide] had significant associations between BMI and colon cancer, but users did not (interaction tests were not statistically significant). Of these, pendimethalin and chlormurion-ethylare also aromatic amine compounds (Kidd). The occurrence of significant associations among users and non-users of pesticides in the same chemical group may indicate that these are chance findings.
Our finding of an inverse association between BMI and lung cancer in men, particularly in ever smokers, is consistent with previous studies (Renehan, 2008). For example, a meta-analysis of five prospective studies of BMI and lung cancer reported a statistically significant summary risk ratio of 0.76 (95% CI 0.67–0.85) among smokers, but not among non-smokers (RR=0.91, 95% CI 0.76–1.10) (Renehan, 2008). This could be due to uncontrolled confounding from smoking because smokers have lower BMIs and a higher risk for lung cancer (Canoy, 2005). In the AHS, underweight subjects were more likely to be current smokers, while overweight/obese subjects were more likely to be former smokers compared to normal weight subjects. Also, current smoking was associated with a significant 4-fold risk of lung cancer among men in our study population. Leanness due to preclinical disease did not appear to be a confounding factor in our study or other previous studies (Kark, 1995; Olson, 2002; Kanashiki, 2005). Biological mechanisms explaining the relationship between leanness and lung cancer risk among smokers is not well understood, but it has been suggested that leanness may serve as a marker of vulnerability to the effects of smoking either through acquired or inherited factors (Sidney, 1987; Shields, 1991).
The slight positive association between BMI and lung cancer among male ever smokers who used carbofuran, but the statistically significant inverse association among non-users of carbofuran, suggests that carbofuran use may interact with obesity to increase the risk of lung cancer. This observation corroborates previous findings from the AHS that found elevated risks of lung cancer with reported use of carbofuran (Alavanja, 2004; Bonner, 2005).
The positive association between BMI and breast cancer among postmenopausal women is in agreement with previous studies (Carmichael 2004, Kuriyama 2005). The main source of estrogen in postmenopausal women is from androstenedione in adipocytes; therefore, heavier postmenopausal women tend to produce more estrogen, which is linked to higher cell proliferation and DNA damage (LaGuardia 2001). In pre-menopausal women, regular anovulatory cycle reduces exposure to estrogen and decreases their breast cancer risk (Carmichael 2004, Potischman, 1996). It is unclear why the positive association between BMI and breast cancer was apparent in those who did not drink, since alcohol consumption is considered a risk factor for postmenopausal breast cancer due to the decrease in the clearance of estradiol (Wayne, 2008; Duffy, 2008). In our population, ever alcohol drinking was associated with an excess risk of postmenopausal breast cancer in women with a BMI <25 kg/m2 (HR=1.25, 95% CI=0.85–1.86), but not in women with a BMI ≥25 kg/m2 (HR=0.87, 95% CI=0.63–1.19). The interaction between BMI and alcohol consumption was also found in a case-control study that reported that one drink per day was associated with estrogen receptor positive tumors in women with a BMI < 25 kg/m2, but not in women with a BMI ≥ 25 kg/m2 (Terry, 2006). A few other studies have suggested that alcohol intake and BMI may only be risk factors for certain histologic types of breast cancer (Suzuki, 2008; Li 2006), which we were not able to evaluate in our study.
The positive associations between BMI and postmenopausal breast cancer were present among women who did not use malathion, diazinon, carbaryl or any organophosphate or organochlorine insecticide, but not in women who used these pesticides. Since this modifying effect was not specific to a particular pesticide, this finding appears to suggest some uncontrolled difference between women who applied and did not apply pesticides. A previous study of breast cancer in the AHS found that women who never applied pesticides, including organophospahtes or organochlorines, had higher risks of breast cancer than women who applied pesticides (Engel, 2005). This previous finding together with our observations is suggestive of possible protective effect of working on the farm for breast cancer.
This study has several important strengths. Due to the prospective study design, pesticide use and other self-reported data was obtained prior to cancer diagnosis, thus any recall bias should be non-differential and unlikely to create false-positive associations. Self-reported pesticide use among farmers in the AHS has been shown to be reasonably reliable (Blair, 2002; Hoppin 2002). Detailed demographic, lifestyle, and medical history data allowed us to adjust for several important risk factors. Misclassification of cancer diagnoses is unlikely given the linkage to population-based state cancer registries.
There also are limitations to this study. Although BMI is a well-defined indicator of obesity, it has been shown to overestimate obesity in athletes (Ode, 2007). This potentially could have occurred in our population given the physical activity of farmers, but the total prevalence of overweight and obesity was in agreement with that reported in the U.S. population. Due to the use of additional data sources to supplement missing height and weight, it is possible that some cancer cases had weight reported after their cancer diagnosis if they were diagnosed within the first 5 years after enrollment. This potentially could have resulted in lower weight values due to the effects of disease in cancer cases; however for lung cancer risk, we excluded cases diagnosed within the first 2 and 5 years of enrollment, and found the associations were not altered. Since we examined the interactions of several pesticides, it is possible that some of our results could have occurred by chance alone. Also, it is not clear whether pesticide use resulted in obesity, or if heavier people used certain pesticides, since pesticide use was reported as ever use and BMI was recorded at or near enrollment. Furthermore, since the mechanisms by which pesticides may interact with BMI on cancer risk in humans are unclear, our findings should be considered hypothesis generating for future studies.
In conclusion, in this U.S. population of farmers, pesticide applicators and their spouses, we confirmed previous findings that BMI is positively associated with colon cancer in men and with breast cancer in postmenopausal women, as well as inversely associated with lung cancer among men who ever smoked. This is the first large prospective cohort study with the capacity to examine possible modifying effects of several pesticides on the association between BMI and several cancer sites. The findings from this analysis, which suggest that certain pesticides may modify the association between BMI and colon and lung cancer risk, should be further evaluated in other populations.
References—should either alphabetize the reference list or use a nric system in the text