Printer Friendly

Dietary cadmium exposure and risk of breast, endometrial, and ovarian cancer in the women's health initiative.


Cadmium, a carcinogenic heavy metal, is released into the environment as a result of industrial and agricultural activities (Jarup and Akesson 2009). Tobacco, grains, and some vegetables can take up cadmium from soil and concentrate it above soil levels (Alloway et al. 1990; Hellstrom et al. 2007; Pappas et al. 2006; Peralta-Videa et al. 2009). Therefore, chronic low-level, nonoccupational exposure to cadmium through tobacco smoke and contaminated foods is common. Cadmium inhaled in cigarette smoke is readily absorbed by lung tissue (Agency for Toxic Substances and Disease Registry 2012). Less than 5% of cadmium ingested in food is absorbed, but low iron stores may increase absorption and may partly explain why women are consistently observed to have higher average urine and blood cadmium concentrations than men (Akesson et al. 2002; Berglund et al. 1994; Vahter et al. 1996). Thus, cadmium exposure may be especially relevant to women's health (Vahter et al. 2007).

Multiple mechanisms potentially link cadmium to cancer, including oxidative stress and inflammation (Lag et al. 2010; Liu et al. 2009), interference with DNA repair (Asmuss et al. 2000; Giaginis et al. 2006), and alterations of DNA methylation (Takiguchi et al. 2003). More relevant to hormone-related cancers, perhaps, is evidence that cadmium may act on estrogenic signaling pathways (Liu et al. 2008; Stoica et al. 2000), resulting in proliferation of breast cancer cells in vitro (Garcia-Morales et al. 1994), and inducing uterus and mammary gland weight increase in rats (Johnson et al. 2003). Long-term treatment with low concentrations of cadmium can malignantly transform breast cells in vitro, although the effect appears to be independent of estrogen receptor-[alpha] (Benbrahim-Tallaa et al. 2009).

Epidemiologically, occupational studies support a link between cadmium and lung cancer (Stayner et al. 1992; Thun et al. 1985), but have largely not addressed hormone-driven cancers in women. Three nonoccupational case-control studies have observed consistent associations between urinary cadmium and breast cancer risk (Gallagher et al. 2010; McElroy et al. 2006; Nagata et al. 2013). Prospective studies in Sweden showed an association between estimated dietary cadmium and endometrial cancer (Akesson et al. 2008) and postmenopausal breast cancer (Julin et al. 2012), but not ovarian cancer (Julin et al. 2011). In contrast, similar studies from the United States (Adams et al. 2012a) and Japan (Sawada et al. 2012) did not observe an association of dietary cadmium with postmenopausal breast cancer risk or risk of any cancer, respectively.

In this report we describe our prospective study of dietary cadmium intake and risk of breast, endometrial, and ovarian cancer in the Women's Health Initiative (WHI).


Study population. We selected study participants from the WHI, a large longitudinal study of postmenopausal women, 50-79 years of age, comprising observational study (OS) and randomized clinical trial (CT) arms. Details of the study design and recruitment have been extensively described (Anderson et al. 2003; Hays et al. 2003; Women's Health Initiative Study Group 1998). WHI recruited participants between 1 October 1993 and 31 December 1998 at 40 clinical centers across the United States. A total of 161,808 women enrolled in the WHI.

All participants provided written informed consent. Human subjects review committees at all participating sites approved WHI study protocols. The analyses presented here were reviewed and approved by the Fred Hutchinson Cancer Research Center Institutional Review Board as an ancillary study to WHI, and complied with all applicable U.S. regulations.

Exposure and covariate assessment. All women completed questionnaires at baseline screening and enrollment; these included detailed information on demographic characteristics, dietary habits, reproductive history [use of postmenopausal hormones for [greater than or equal to] 3 months (estrogen or estrogen plus progesterone; pills or patches)], medical history, lifestyle (tobacco use, alcohol use, dietary supplement use), and physical activity (Anderson et al. 2003). Anthropometric measurements were taken at baseline clinic visits using a standardized protocol, and body mass index (BMI; kilograms per meter squared) was calculated as weight divided by height squared.

To assess usual diet, all participants completed a baseline food frequency questionnaire (FFQ), specifically designed for WHI and previously described in detail (Kristal et al. 1997; Patterson et al. 1999). These FFQs captured usual intakes during the prior 3 months of 122 food and beverage line items, comprising 302 individual food and beverage components.

To estimate dietary cadmium intake, we adapted methodology commonly used for dietary micronutrient estimates (Schakel et al. 1997) and described previously for dietary cadmium (Adams et al. 2012a). We used measurements of the cadmium content of foods determined analytically by the U.S. Food and Drug Administration (FDA) as part of the Total Diet Study (TDS) (Egan et al. 2002, 2007; FDA 2013). Briefly, market baskets of 285 or 290 foods were typically purchased each year (1991-2008) from three locations in each of four regions of the United States. These foods were sent to a central laboratory for preparation according to predetermined recipes, and analysis for content of a number of contaminants including cadmium (Egan et al. 2002). Cadmium was determined with graphite furnace atomic absorption spectroscopy; the limit of detection (LOD) varied among food items and ranged from 0.001 to 0.007 mg/kg (Egan et al. 2002).

The arithmetic mean of cadmium content (milligrams per 100 g prepared weight) of all available samples of each food was assigned as the cadmium content for that food. Thus, we averaged over year and region of collection to estimate each food's cadmium content. To investigate whether this approach ignored important regional or secular trends in the cadmium content in foods, we examined measurements from the 20 foods with the highest reported cadmium concentrations. Regional variation was < 20% of the overall mean for 13 of the 20 top cadmium-containing foods; maximum variation of < 40% was observed for lettuces. No region with systematically higher or lower cadmium values was identified. Year-to-year variation in the 20 foods with highest mean cadmium concentration reached 50% of the mean cadmium. Qualitatively, however, trends with year of measurement were observed for only three of these 20 foods (peanut butter, decreasing; raisin bran cereal, decreasing; and egg noodles, increasing). Because of limited variation by region and year of measurement, we opted to include all available data from the FDA TDS for each food in order to obtain the best estimate of mean cadmium content of food typically consumed in the United States. We did not use baby foods (n = 57) in our calculation of dietary cadmium.

We assigned values of zero to individual cadmium measurements for food items below the LOD. For four foods, all measured cadmium values were below the LOD: tap water, olive/safflower oil, martini/palmarosa oil, and vegetable oil. For 127 foods, one or more cadmium measurements fell below the LOD, resulting in an overall mean cadmium content less than the LOD. These foods were primarily meats, fruits and fruit juices, dairy products, and beverages. Overall mean cadmium concentrations values (i.e., averaged over collection years and regions) below the LOD were retained in analysis.

We matched each of 302 food and beverage components comprising the 120 FFQ line items on the WHI FFQ to one of the foods analyzed by the FDA, based on the food names. To allow inclusion of participants at the Hawaii clinical center, 27 additional component foods specific to the Hawaii FFQ were matched. For component foods for which no obviously similar food was analyzed by the FDA, we relied on food "mapping" created by the FDA for the TDS (FDA 2013). In summary, 154 component foods were direct matches based on the food names; 122 component foods were close but differed in either unspecified details (e.g., "summer squash" and "squash") or preparation (e.g., "raw onion" and "cooked onion"); 26 component foods were matched using the TDS mapping file (e.g., "lentils" and "white beans"). Therefore, we attributed a cadmium content to every component food.

The FFQ analytic program calculated average annual servings of each FFQ line item, adjusted to sex-specific portion sizes, and estimated nutrient intakes based on the University of Minnesota Nutrition Data System for Research (NDSR, version 2007; http://www. Nutrient and cadmium calculations were performed by the Nutrition Assessment Shared Resource of the Fred Hutchinson Cancer Research Center (Seattle, WA).

Urinary cadmium and creatinine. Spot urine samples were collected at baseline from a subset of WHI participants (Anderson et al. 2003). The cadmium concentration in a subset (n = 1,050) of urine samples was measured at the Trace Elements Research Laboratory at the Wisconsin State Laboratory of Hygiene (Madison, WI), using sector field inductively coupled plasma mass spectrometry (SF-ICP MS; Thermo-Finnegan, Element 2) as described (Adams et al. 2011; Cheung et al. 2012). Values below the limit of quantification (3.5 ng/L) were assigned values of 2.5 ng/L. Urine creatinine was measured on a Molecular Devices Spectra Max M5e plate reader using BioAssay Systems QuantiChrom Creatinine Assay Kit configured for 96-well plate assays, following manufacturer's instructions. A modified Jaffe chemistry was employed to quantify picrate-creatinine spectrophotometrically at 510 nm. Samples were run in duplicate; a median coefficient of variation (CV) of 2.7% was observed. The method limit of quantification was 5 pM (0.06 mg/dL).

Exclusions and missing data. For these analyses we excluded women with incomplete or invalid (total energy < 600 or > 5,000 kcal/day) FFQ data (n = 4,624) or without follow-up information for cancer diagnosis (n = 650). We also excluded women with a previous diagnosis of a cancer of interest (breast, n = 5,545; endometrial, n = 1,005; and ovarian, n = 802) from analysis of that specific cancer. For analyses of endometrial cancer risk, we excluded women with hysterectomy before enrollment (n = 32,500). For analyses of ovarian cancer risk, we excluded women reporting bilateral oophorectomy before enrollment (n = 28,668). We included women with missing information on a given variable as a separate category for adjustment; 10.4% of participants had missing information on one or more variables. A total of 155,069 women were included in one or more analyses (breast cancer, n = 150,889; endometrial cancer, n = 91,643; ovarian cancer, n = 125,569).

Follow-up for cancer and censoring. Participants updated medical history annually (OS) or semiannually (CT) through a mailed self-administered or telephone-administered questionnaire. Breast, endometrial, and ovarian cancers reported by participants were adjudicated by WHI Clinical Coordinating Center (WHI-CCC) staff and physician review of medical records (Anderson et al. 2003; Curb et al. 2003). Separate analyses were conducted for each cancer of interest (breast, endometrial, and ovarian). In each analysis, women were followed until the earliest of incidence of the cancer of interest, death, or final contact. The original WHI study period ended on 31 March 2005; subsequent additional active follow-up continued through 2010. Follow-up for this report ended August 2009. For breast cancer analyses, women were censored at incidence of in situ breast cancer (n = 1,571) and were therefore not included as outcomes but did contribute time-at-risk before in situ diagnosis. For endometrial cancer analyses, women were censored at hysterectomy (n = 5,872). Hysterectomy was reported on annual or semiannual updates and adjudicated by WHI-CCC for hormone CT participants; analysis of OS participants relied on self-reported hysterectomy information. No updated information on oophorectomy was collected during the study, to enable censoring in analysis of ovarian cancer risk. Death was ascertained through clinical center follow-up of family reports and routine checks with the National Death Index.

Statistical analyses. Multivariable Cox proportional hazards regression was applied to estimate adjusted cancer hazard ratios (HRs) with 95% CIs by quintile of dietary cadmium, adjusted for total energy intake by the residual method (Willett and Stampfer 1986). The mean dietary cadmium intake (10.9 pg/day) was added to calculate "energy-adjusted dietary cadmium." Trends were examined by assigning to each quintile the ordinal value of that quintile and treating it as a continuous variable; p-trend is from a Wald test of this coefficient compared with zero in the fully adjusted model.

We selected confounders, each measured at baseline, based on knowledge of risk factors for breast, endometrial, and ovarian cancer, and sources of cadmium exposure. Multivariable models were stratified for enrollment age in bands (50-54, 55-59, 60-69, 70-79 years), and on WHI component (OS or CT), and adjusted for age (years) (in addition to stratification by age band), race/ethnicity (non-Hispanic white, other), education (high school diploma or less; some college or postsecondary education; college degree or more), BMI (< 25, 25-29.9, [greater than or equal to] 30.0 kg/[m.sup.2]), alcohol consumption (drinks/ week: none, < 1, 1-6.9, [greater than or equal to] 7), combined estrogen plus progesterone hormone therapy (never, past user, current user), unopposed estrogen hormone therapy (never, past user, current user), age at first birth (nulliparous, < 2, 20-29, [greater than or equal to] 30 years), age at menarche (< 12, 12, 13, > 13 years), age at menopause ([less than or equal to] 42, 43-47, 48-49, 50-52, [greater than or equal to] 53 years), physical activity (metabolic equivalent hours per week, quartiles), and cigarette smoking history (never, former, current). Breast cancer analyses were additionally adjusted for mammography in the 2 years before baseline (yes/no). In further analyses, we additionally adjusted for daily medium-sized servings of vegetables (< 1.5, 1.5-2.9, [greater than or equal to] 3), and daily medium-sized servings of grains (quartiles). To investigate alternatives to vegetable servings and grains, we performed additional analyses adjusted for intake of zinc and iron from diet and supplements (quartiles) along with servings of vegetables and servings of grains; or replaced adjustment for servings of vegetables and grains with computed grams of fiber and carbohydrates consumed daily (quartiles). Finally we examined adjustment for pack-years of smoking, in addition to smoking status.

To improve comparability with previous cohorts that have evaluated dietary cadmium and cancer incidence, we applied the same methods for each outcome of interest to selected subgroups of women: never-users of hormone therapy (at enrollment); women with BMI between 18.5 and 25 kg/[m.sup.2]; never-smokers; women without diabetes (at enrollment); women in the lowest quartile of zinc intake, iron intake, or servings of grains; women who consumed < 1.5 servings of vegetables/day; and women in the OS. Tests for the statistical significance of interactions, though, are not reported.

In further analyses, we estimated associations between dietary cadmium and breast cancer cases classified according to estrogen receptor status (ER+ or ER-). For analysis restricted to ER+ breast cancer (n = 5,161 cases), women were censored at incidence of ER-, borderline, or unclassified breast cancer; conversely, for analyses specific to ER-breast cancer (n = 948 cases), women were censored at incidence of ER+, borderline, or unclassified breast cancer.

For this report we estimated the partial Spearman correlation coefficient between creatinine-normalized urine cadmium concentrations and energy-adjusted dietary cadmium estimates, adjusted for age, for 565 never-smokers.

All analyses were completed in Stata Statistical Software, release 12 (StataCorp LP, College Station, TX, USA).


Estimated dietary cadmium ranged from 0.02 to 59.4 [micro]g/day (mean, 10.9 [micro]g/day; median, 10.3 [micro]g/day), and was higher among women reporting higher levels of vegetable and grain consumption, consistent with foods documented to be high in cadmium, or higher energy intake (Table 1). On average, the major sources of dietary cadmium were vegetables including potatoes (42% of dietary cadmium); grains including bread, pasta, and rice (29%); seafood (2.2%); fruit (3.8%); and meat, poultry, and dairy (3.8%). Dietary cadmium intake varied with many participant characteristics including age, BMI, race/ethnicity, education, smoking history, physical activity, and alcohol consumption. In comparison, estimated dietary cadmium varied only slightly with reproductive history, use of hormone therapy, and mammography utilization before study enrollment.

Results adjusted only for total energy, age, and WHI study component did not suggest statistically significant dose-response trends in associations between dietary cadmium and any of the three cancers (Table 2). Further adjustment for smoking, BMI, demographics, physical activity, and reproductive history largely left these results unchanged, as did further adjustment for servings of vegetables and servings of grains (Table 2). Notably, the FFQ-derived total energy intake and servings of grains and vegetables were correlated with dietary cadmium ([R.sup.2] of 0.4-0.5 for each in univariate analysis). Addition of zinc and iron intake (milligrams per day) to the model did not substantially change results; nor did substitution of daily total fiber (grams) and carbohydrates (grams) for vegetable and grain servings (data not shown). Not adjusting for total dietary energy intake assessed from the FFQ left the interpretation of results substantially unchanged, although some HRs for individual quintiles of dietary cadmium were significantly > 1 for endometrial and ovarian cancer in some but not all models (see Supplemental Material, Table S1). Adjustment for pack-years of smoking did not materially change results (data not shown).

Associations among the following subgroups of women were generally consistent with those estimated for the cohort as a whole: women with BMI 18.5-25 kg/[m.sup.2]; women without diabetes at enrollment; women consuming < 2.5 medium servings of fruits and vegetables and/or < 2.9 medium servings of grains per day (the lowest quartiles); women with zinc intake < 9.0 mg/day (the lowest quartile) from both diet and supplements; women with iron intake < 10.5 mg/day (the lowest quartile) from both diet and supplements; never-users of hormone therapy before enrollment; never-smokers; women with no history of any cancer before enrollment; and participants in the OS only (data not shown). Associations of ER+ and ER-breast cancer subtypes with dietary cadmium intake were similar to overall results for breast cancer (data not shown).

Mean creatinine-corrected urinary cadmium was 0.49 pg cadmium/g creatinine. The Spearman rank partial correlation coefficient (p) between energy-adjusted dietary cadmium and creatinine-corrected urinary cadmium, adjusted for age, was 0.085 (p = 0.007 for test of null hypothesis that p = 0).


Because of the apparent action of cadmium as an endocrine disruptor, or "metallohormone" (Byrne et al. 2009), we investigated the relation between dietary exposure to this potential environmental carcinogen and three hormone-driven cancers. We did not find evidence for an association of dietary cadmium with any of these cancers. Although HRs for breast and endometrial cancer were different from 1 for some quintiles of dietary cadmium, associations based on linear trends were not apparent or tested to be statistically significant. Thus, overall, we interpret our results to provide little evidence of associations between estimated dietary cadmium and risk of breast, endometrial, or ovarian cancer within this large cohort of postmenopausal women.

However, total energy intake and consumption of vegetables and grains, estimated from the FFQ were correlated with dietary cadmium. Thus, we considered whether adjusting for these variables might attenuate the relationship between dietary cadmium and cancer risk through "overadjustment." Without these adjustments, the endometrial and ovarian cancer HR estimates for the highest versus lowest quintile of dietary cadmium were > 1 and statistically significant in some models. No statistically significant dose-response trends were observed with increasing dietary cadmium exposure, however. Thus, although HR point estimates changed noticeably, overall, our interpretation of the results is largely unchanged.

We took advantage of the large size of the WHI to investigate whether the association between cadmium and cancer in selected subgroups of women was different than associations estimated from the cohort as a whole. We focused on three areas: whether the association between dietary cadmium and hormone-related cancer risk varied with BMI and hormone therapy (Akesson et al. 2008; Byrne et al. 2009); varied with dietary components such as fruits and vegetables, grains, zinc, or iron that could modulate uptake of dietary cadmium or mitigate the effects of cadmium (Beyersmann and Hartwig 2008; Klaassen et al. 2009; Tallkvist et al. 2001); or varied with tobacco use, a source of cadmium, that could mask an association with dietary cadmium (Akesson et al. 2008; McElroy et al. 2007; Richter et al. 2009). Furthermore, we conducted separate analyses restricted to ER+ or ER-breast tumors. We found no evidence supporting an association of cadmium with cancer risk in any subgroup examined.

In contrast to our results, prospective studies of postmenopausal women in the Swedish Mammography Cohort observed positive associations between dietary cadmium and risk of postmenopausal breast cancer and endometrial cancer, but not ovarian cancer (Akesson et al. 2008; Julin et al. 2011, 2012). Interestingly, the association of dietary cadmium with breast and endometrial cancers reported from the Swedish Mammography Cohort was strengthened by adjustment for intake of vegetables and grains (Akesson et al. 2008; Julin et al. 2012), opposite to our findings. Another study using a dietary cadmium database and FFQ similar to those used for this report, but in a different U.S. population, also observed no association with breast cancer risk (Adams et al. 2012a); and no statistically significant association between dietary cadmium and risk of all cancer, breast cancer, or endometrial cancer was observed in the Japan Public Health Center-based Prospective Study (Sawada et al. 2012). These studies used a methodology similar to that of the present study to estimate dietary intake of cadmium; the average estimated intake of dietary cadmium was similar between the U.S. and Swedish studies (10-15 [micro]g/day), but substantially higher in Japan (27 [micro]g/day). However, differences in the variation in the cadmium content of foods may partially explain differences in results; dietary cadmium estimates may be more accurate or precise in some populations than in others. Three retrospective case-control studies reported a positive association between cadmium exposure and risk of breast cancer (Gallagher et al. 2010; McElroy et al. 2006; Nagata et al. 2013), in contrast to our results. These studies assessed cadmium exposure through measurement of urinary cadmium, believed to be an objective marker of cadmium absorption over decades (Lauwerys et al. 1994; Nordberg and Kjellstrom 1979), which may explain the discrepant results in comparison to our study. On the other hand, the results from the retrospective case-control studies may be subject to biases that are not present in our prospective study; for example, treatment for breast cancer may influence urinary cadmium, as has been suggested for lead (McElroy et al. 2008). For each of these case-control studies, cases received some treatment before urine sample collection (Gallagher et al. 2010; McElroy et al. 2006; Nagata et al. 2013).

Cadmium is classified as a carcinogen by the World Health Organization (International Agency for Research on Cancer 1993) primarily on the basis of occupational studies. Nonoccupational exposure to cadmium occurs predominantly through tobacco smoke and food (Jarup and Akesson 2009), and the association between environmental cadmium exposure and various cancers has recently received increasing attention. Prospective epidemiological studies have reported associations between cadmium exposure and higher cancer mortality, including endometrial cancer mortality, but not breast or ovarian cancer mortality (Adams et al. 2012b; Menke et al. 2009; Nawrot et al. 2006).

It is possible that our method of exposure assessment resulted in misclassification that may have biased estimated associations toward the null. As described, our methodology was patterned on nutritional epidemiological studies of micronutrients and cancer risk that use a FFQ. Although the FFQ we used in this study was validated for intake of many micronutrients by comparison to daily food records (Patterson et al. 1999), those results may not extend to cadmium. Even if FFQ responses accurately captured food intake, variation in the cadmium content of food items was another potentially important source of measurement error because the amount of cadmium taken up by plants depends on agricultural conditions and crop varietals (Alloway et al. 1990; Arao and Ae 2003; Cataldo et al. 1981; Peralta-Videa et al. 2009). Last, the FFQ measured usual diet close to baseline, which may not be representative of lifetime exposure.

We compared our estimates of dietary cadmium to urinary cadmium concentrations, corrected for creatinine, in a sample of never-smokers in the WHI, and observed a small but statistically significant correlation (p = 0.085). Although urinary cadmium has been used to measure low-level environmental exposure in many epidemiological studies, recent reports have suggested that urinary cadmium may not accurately reflect cadmium accumulation in the kidney resulting from long-term, low-level exposure; therefore, urinary cadmium may reflect primarily recent exposure rather than cumulative exposure (Chaumont et al. 2012, 2013). Thus, the correlation between urinary and dietary cadmium may be lacking because these methods assessed cadmium exposure over different periods of time. Overall, measurement error in assessment of dietary cadmium would be nondifferential in this prospective cohort study and could have introduced substantial bias toward a finding of no association (Freedman et al. 2011; Kipnis et al. 2003).

Finally, although we did not assess occupational exposure to cadmium in the WHI, a previous study of the U.S. adult population suggests that elevated cadmium exposure occurs mainly in automotive and electrical repair, mining, metalworking, and similar jobs working directly with metals (Yassin and Martonik 2004). Because the participants in our study were women > 50 years of age, occupational exposure seems unlikely to have been important.

Despite these limitations, our study has substantial strengths, including its prospective design and the large size of the cohort. Our study included triple the number of cases of breast and endometrial cancer risk and nearly double the number of ovarian cancer cases of the largest prior studies from the Swedish Mammography Cohort (Akesson et al. 2008; Julin et al. 2011, 2012). Furthermore, WHI data on covariates are highly detailed, including only a small percentage of missing information on variables we considered in our analyses. Follow-up of participants through the established WHI-CCC and vital statistics minimized attrition from the cohort through loss to follow-up. Thus, selection resulting from missing data within the cohort, or differential attrition, is unlikely to have substantially biased our results.


The results of our study did not support the hypothesis that cadmium contamination of food, measured with an assessment of usual diet during the 3 months before baseline, is a risk factor for postmenopausal breast, endometrial, or ovarian cancer, but misclassification of exposure may have attenuated an association. In future prospective studies, alternative assessments of cadmium exposure, such as urinary cadmium concentration, should be tested in relation to risk of hormonal cancers.

Appendix: Short List of WHI Investigators

Program Office (National Heart, Blood, and Lung Institute, Bethesda, Maryland): Jacques Rossouw, Shari Ludlam, Dale Burwen, Joan McGowan, Leslie Ford, and Nancy Geller.

Clinical Coordinating Center (Fred Hutchinson Cancer Research Center, Seattle, WA): Garnet Anderson, Ross Prentice, Andrea LaCroix, and Charles Kooperberg.

Investigators and Academic Centers: JoAnn E. Manson (Brigham and Women's Hospital, Harvard Medical School, Boston, MA), Barbara V. Howard (MedStar Health Research Institute/Howard University, Washington, DC), Marcia L. Stefanick (Stanford Prevention Research Center, Stanford, CA), Rebecca Jackson (The Ohio State University, Columbus, OH), Cynthia A. Thomson (University of Arizona, Tucson/Phoenix, AZ), Jean Wactawski-Wende (University at Buffalo, Buffalo, NY), Marian Limacher (University of Florida, Gainesville/ Jacksonville, FL), Robert Wallace (University of Iowa, Iowa City/Davenport, IA), Lewis Kuller (University of Pittsburgh, Pittsburgh, PA), and Sally Shumaker (Wake Forest University School of Medicine, Winston-Salem, NC).


Adams SV, Newcomb PA, Shafer MM, Atkinson C, Bowles EJ, Newton KM, et al. 2011. Sources of cadmium exposure among healthy premenopausal women. Sci Total Environ 409(9):1632-1637.

Adams SV, Newcomb PA, White E. 2012a. Dietary cadmium and risk of invasive postmenopausal breast cancer in the VITAL cohort. Cancer Causes Control 23(6):845-854.

Adams SV, Passarelli MN, Newcomb PA. 2012b. Cadmium exposure and cancer mortality in the Third National Health and Nutrition Examination Survey cohort. Occup Environ Med 69(2):153-156.

Agency for Toxicological Substances and Disease Registry. 2012. Toxicological Profile for Cadmium. Available: http://www. [accessed 1 July 2013]. Akesson A, Berglund M, Schutz A, Bjellerup P, Bremme K, Vahter M. 2002. Cadmium exposure in pregnancy and lactation in relation to iron status. Am J Public Health 92(2):284-287.

Akesson A, Julin B, Wolk A. 2008. Long-term dietary cadmium intake and postmenopausal endometrial cancer incidence: a population-based prospective cohort study. Cancer Res 68(15):6435-6441.

Alloway BJ, Jackson AP, Morgan H. 1990. The accumulation of cadmium by vegetables grown on soils contaminated from a variety of sources. Sci Total Environ 91:223-236.

Anderson GL, Manson J, Wallace R, Lund B, Hall D, Davis S, et al. 2003. Implementation of the Women's Health Initiative study design. Ann Epidemiol 13(9 Suppl):S5-17.

Arao T, Ae N. 2003. Genotypic variations in cadmium levels of rice grain. Soil Sci Plant Nutr 49(4):473-479.

Asmuss M, Mullenders LH, Hartwig A. 2000. Interference by toxic metal compounds with isolated zinc finger DNA repair proteins. Toxicol Lett 112-113:227-231.

Benbrahim-Tallaa L, Tokar EJ, Diwan BA, Dill AL, Coppin JF, Waalkes MP. 2009. Cadmium malignantly transforms normal human breast epithelial cells into a basal-like phenotype. Environ Health Perspect 117:1847-1852.

Berglund M, Akesson A, Nermell B, Vahter M. 1994. Intestinal absorption of dietary cadmium in women depends on body iron stores and fiber intake. Environ Health Perspect 102:1058-1066.

Beyersmann D, Hartwig A. 2008. Carcinogenic metal compounds: recent insight into molecular and cellular mechanisms. Arch Toxicol 82(8):493-512.

Byrne C, Divekar SD, Storchan GB, Parodi DA, Martin MB. 2009. Cadmium--a metallohormone? Toxicol Appl Pharmacol 238(3):266-271.

Cataldo DA, Garland TR, Wildung RE. 1981. Cadmium distribution and chemical fate in soybean plants. Plant Physiol 68(4):835-839.

Chaumont A, Nickmilder M, Dumont X, Lundh T, Skerfving S, Bernard A. 2012. Associations between proteins and heavy metals in urine at low environmental exposures: evidence of reverse causality. Toxicol Lett 210(3):345-352.

Chaumont A, Voisin C, Deumer G, Haufroid V, Annesi-Maesano I, Roels H, et al. 2013. Associations of urinary cadmium with age and urinary proteins: further evidence of physiological variations unrelated to metal accumulation and toxicity. Environ Health Perspect 121:1047-1053; doi:10.1289/ehp.1306607.

Cheung K, Shafer MM, Schauer JJ, Sioutas C. 2012. Diurnal trends in oxidative potential of coarse particulate matter in the Los Angeles Basin and their relation to sources and chemical composition. Environ Sci Technol 46(7):3779-3787.

Curb JD, McTiernan A, Heckbert SR, Kooperberg C, Stanford J, Nevitt M, et al. 2003. Outcomes ascertainment and adjudication methods in the Women's Health Initiative. Ann Epidemiol 13(9 Suppl):S122-S128.

Egan SK, Bolger PM, Carrington CD. 2007. Update of US FDA's Total Diet Study food list and diets. J Expo Sci Environ Epidemiol 17(6):573-582.

Egan SK, Tao SS, Pennington JA, Bolger PM. 2002. US Food and Drug Administration's Total Diet Study: intake of nutritional and toxic elements, 1991-96. Food Addit Contam 19(2):103-125.

FDA (U.S. Food and Drug Administration). 2013. Total Diet Study. Available: totaldietstudy/ucm2006799.htm [accessed 1 August 2013].

Freedman LS, Schatzkin A, Midthune D, Kipnis V. 2011. Dealing with dietary measurement error in nutritional cohort studies. J Natl Cancer Inst 103(14):1086-1092.

Gallagher CM, Chen JJ, Kovach JS. 2010. Environmental cadmium and breast cancer risk. Aging (Albany NY) 2(11):804-814.

Garcia-Morales P, Saceda M, Kenney N, Kim N, Salomon DS, Gottardis MM, et al. 1994. Effect of cadmium on estrogen receptor levels and estrogen-induced responses in human breast cancer cells. J Biol Chem 269(24):16896-16901.

Giaginis C, Gatzidou E, Theocharis S. 2006. DNA repair systems as targets of cadmium toxicity. Toxicol Appl Pharmacol 213(3):282-290.

Hays J, Hunt JR, Hubbell FA, Anderson GL, Limacher M, Allen C, et al. 2003. The Women's Health Initiative recruitment methods and results. Ann Epidemiol 13(9 suppl):S18-S77.

Hellstrom L, Persson B, Brudin L, Grawe KP, Oborn I, Jarup L. 2007. Cadmium exposure pathways in a population living near a battery plant. Sci Total Environ 373(2-3):447-455.

International Agency for Research on Cancer. 1993. Beryllium, cadmium, mercury, and exposures in the glass manufacturing industry. IARC Monogr Eval Carcinog Risk Hum 58:1-444.

Jarup L, Akesson A. 2009. Current status of cadmium as an environmental health problem. Toxicol Appl Pharmacol 238(3):201-208.

Johnson MD, Kenney N, Stoica A, Hilakivi-Clarke L, Singh B, Chepko G, et al. 2003. Cadmium mimics the in vivo effects of estrogen in the uterus and mammary gland. Nat Med 9(8):1081-1084.

Julin B, Wolk A, Akesson A. 2011. Dietary cadmium exposure and risk of epithelial ovarian cancer in a prospective cohort of Swedish women. Br J Cancer 105(3):441-444.

Julin B, Wolk A, Bergkvist L, Bottai M, Akesson A. 2012. Dietary cadmium exposure and risk of postmenopausal breast cancer: a population-based prospective cohort study. Cancer Res 72(6):1459-1466.

Kipnis V, Subar AF, Midthune D, Freedman LS, Ballard-Barbash R, Troiano RP, et al. 2003. Structure of dietary measurement error: results of the OPEN biomarker study. Am J Epidemiol 158(1):14-21.

Klaassen CD, Liu J, Diwan BA. 2009. Metallothionein protection of cadmium toxicity. Toxicol Appl Pharmacol 238(3):215-220.

Kristal AR, Feng Z, Coates RJ, Oberman A, George V. 1997. Associations of race/ethnicity, education, and dietary intervention with the validity and reliability of a food frequency questionnaire: the Women's Health Trial Feasibility Study in Minority Populations. Am J Epidemiol 146(10):856-869.

Lag M, Rodionov D, Ovrevik J, Bakke O, Schwarze PE, Refsnes M. 2010. Cadmium-induced inflammatory responses in cells relevant for lung toxicity: expression and release of cytokines in fibroblasts, epithelial cells and macrophages. Toxicol Lett 193(3):252-260.

Lauwerys RR, Bernard AM, Roels HA, Buchet JP. 1994. Cadmium: exposure markers as predictors of nephrotoxic effects. Clin Chem 40(7 pt 2):1391-1394.

Liu J, Qu W, Kadiiska MB. 2009. Role of oxidative stress in cadmium toxicity and carcinogenesis. Toxicol Appl Pharmacol 238(3):209-214.

Liu Z, Yu X, Shaikh ZA. 2008. Rapid activation of ERK1/2 and AKT in human breast cancer cells by cadmium. Toxicol Appl Pharmacol 228(3):286-294.

McElroy JA, Shafer MM, Gangnon RE, Crouch LA, Newcomb PA. 2008. Urinary lead exposure and breast cancer risk in a population-based case-control study. Cancer Epidemiol Biomarkers Prev 17(9):2311-2317.

McElroy JA, Shafer MM, Trentham-Dietz A, Hampton JM, Newcomb PA. 2006. Cadmium exposure and breast cancer risk. J Natl Cancer Inst 98(12):869-873.

McElroy JA, Shafer MM, Trentham-Dietz A, Hampton JM, Newcomb PA. 2007. Urinary cadmium levels and tobacco smoke exposure in women age 20-69 years in the United States. J Toxicol Environ Health A 70(20):1779-1782.

Menke A, Muntner P, Silbergeld EK, Platz EA, Guallar E. 2009. Cadmium levels in urine and mortality among U.S. adults. Environ Health Perspect 117:190-196; doi:10.1289/ehp.11236.

Nagata C, Nagao y, Nakamura K, Wada K, Tamai y, Tsuji M, et al. 2013. Cadmium exposure and the risk of breast cancer in Japanese women. Breast Cancer Res Treat 138(1):235-239.

Nawrot T, Plusquin M, Hogervorst J, Roels HA, Celis H, Thijs L, et al. 2006. Environmental exposure to cadmium and risk of cancer: a prospective population-based study. Lancet Oncol 7(2):119-126.

Nordberg GF, Kjellstrom T. 1979. Metabolic model for cadmium in man. Environ Health Perspect 28:211-217.

Pappas RS, Polzin GM, Zhang L, Watson CH, Paschal DC, Ashley DL. 2006. Cadmium, lead, and thallium in mainstream tobacco smoke particulate. Food Chem Toxicol 44(5):714-723.

Patterson RE, Kristal AR, Tinker LF, Carter RA, Bolton MP, Agurs-Collins T. 1999. Measurement characteristics of the Women's Health Initiative food frequency questionnaire. Ann Epidemiol 9(3):178-187.

Peralta-Videa JR, Lopez ML, Narayan M, Saupe G, Gardea-Torresdey J. 2009. The biochemistry of environmental heavy metal uptake by plants: implications for the food chain. Int J Biochem Cell Biol 41(8-9):1665-1677.

Richter PA, Bishop EE, Wang J, Swahn MH. 2009. Tobacco smoke exposure and levels of urinary metals in the U.S. youth and adult population: the National Health and Nutrition Examination Survey (NHANES) 1999-2004. Int J Environ Res Public Health 6(7):1930-1946.

Sawada N, Iwasaki M, Inoue M, Takachi R, Sasazuki S, Yamaji T, et al. 2012. Long-term dietary cadmium intake and cancer incidence. Epidemiology 23(3):368-376.

Schakel S, Buzzard IM, Gebhardt S. 1997. Procedures for estimating nutrient values for food composition databases. J Food Compos Anal 10(2):102-114.

Stayner L, Smith R, Thun M, Schnorr T, Lemen R. 1992. A dose-response analysis and quantitative assessment of lung cancer risk and occupational cadmium exposure. Ann Epidemiol 2(3):177-194.

Stoica A, Katzenellenbogen BS, Martin MB. 2000. Activation of estrogen receptor-alpha by the heavy metal cadmium. Mol Endocrinol 14(4):545-553.

Takiguchi M, Achanzar WE, Qu W, Li G, Waalkes MP. 2003. Effects of cadmium on DNA-(Cytosine-5) methyltransferase activity and DNA methylation status during cadmium-induced cellular transformation. Exp Cell Res 286(2):355-365.

Tallkvist J, Bowlus CL, Lonnerdal B. 2001. DMT1 gene expression and cadmium absorption in human absorptive enterocytes. Toxicol Lett 122(2):171-177.

Thun MJ, Schnorr TM, Smith AB, Halperin WE, Lemen RA. 1985. Mortality among a cohort of U.S. cadmium production workers--an update. J Natl Cancer Inst 74(2):325-333.

Vahter M, Akesson A, Liden C, Ceccatelli S, Berglund M. 2007. Gender differences in the disposition and toxicity of metals. Environ Res 104(1):85-95.

Vahter M, Berglund M, Nermell B, Akesson A. 1996. Bioavailability of cadmium from shellfish and mixed diet in women. Toxicol Appl Pharmacol 136(2):332-341.

Willett W, Stampfer MJ. 1986. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol 124(1):17-27.

Women's Health Initiative Study Group. 1998. Design of the Women's Health Initiative clinical trial and observational study. Control Clin Trials 19(1):61-109.

Yassin AS, Martonik JF. 2004. Urinary cadmium levels in the U.S. working population, 1988-1994. J Occup Environ Hyg 1(5):324-333.

Scott V. Adams, (1) Sabah M. Quraishi, (1,2) Martin M. Shafer, (3,4) Michael N. Passarelli, (1,2) Emily P. Freney, (1) Rowan T. Chlebowski, (5) Juhua Luo, (6) Jaymie R. Meliker, (7) Lina Mu, (8) Marian L. Neuhouser, (1,2) and Polly A. Newcomb (1,2)

(1) Fred Hutchinson Cancer Research Center, Program in Cancer Prevention, Public Health Sciences Division, Seattle, Washington, USA; (2) Department of Epidemiology, University of Washington School of Public Health, Seattle, Washington, USA; (3) Environmental Chemistry and Technology; and (4) Wisconsin State Laboratory of Hygiene, University of Wisconsin-Madison, Madison, Wisconsin, USA; (5) David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, California, USA; (6) Department of Epidemiology and Biostatistics, Indiana University School of Public Health, Bloomington, Indiana, USA; (7) Department of Preventive Medicine, Program in Public Health, Stony Brook University School of Medicine, Stony Brook, New York, USA; (8) Department of Social and Preventive Medicine, School of Public Health and Health Professions, State University of New York at Buffalo, Buffalo, New York, USA

Address correspondence to S.V. Adams, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N, M4-B402, Seattle, Washington 98109 USA. Telephone: (206) 667-6427. E-mail:

Supplemental Material is available online (http://

Investigators gratefully acknowledge the staff of the Women's Health Initiative (WHI) Clinical Coordinating Center, the staff of the Nutritional Assessment Shared Resource at Fred Hutchinson Cancer Research Center, and E. Meier.

This research was supported by National Institutes of Health (NIH)/National Institute of Environmental Health Sciences (NIEHS) grant R01ES019667. S.M.Q. was supported by Biostatistics, Epidemiologic and Bioinformatic Training in Environmental Health Training Grant NIEHS T32ES015459. The WHI program is funded by the National Heart, Lung, and Blood Institute, NIH, through contracts HHSN268201100046C, HHSN268201100001C, HHSN268201100002C, HHSN268201100003C, HHSN268201100004C.

The authors declare they have no actual or potential competing financial interests.

Received: 7 May 2013; Accepted: 12 March 2014; Advance Publication: 14 March 2014; Final Publication: 1 June 2014.
Table 1. Selected baseline characteristics of participants in
one or more analyses (total n = 155,069), by quintile of
estimated total (not adjusted for energy) dietary cadmium
exposure [n (%)].

Characteristic                     Quintile 1     Quintile 2
                                   < 7.10         7.10-9.24
                                   [micro]g/day   [micro]g/day

Total                              31,013 (100)   31,014 (100)
Age (years)
  50-54                             3,965 (13)     3,867 (12)
  55-59                             5,858 (19)     5,976 (19)
  60-69                            13,645 (44)    14,143 (46)
  70-79                             7,545 (24)     7,028 (23)
WHI study component
  Clinical trial                   13,212 (43)    13,311 (43)
  Observational study              17,801 (57)    17,703 (57)
Total energy intake
  (kcal/day) (a)
  600-1,187                        21,082 (68)     9,839 (32)
  1,188-1,539                       7,308 (24)    11,988 (39)
  1,540-1,969                       2,198 (7)      7,281 (23)
  1,970-5,000                         425 (1)      1,906 (6)
BMI (kg/m.sup.2]) (b)
  < 25                             10,896 (35)    11,138 (36)
  25-29.9                          10,744 (35)    10,848 (35)
  [greater than or equal to] 30     9,110 (29)     8,757 (28)
Cigarette smoking history (b)
  Never                            15,843 (51)    15,585 (50)
  Former                           11,648 (38)    12,734 (41)
  Current                           3,098 (10)     2,283 (7)
Alcohol consumption
  (drinks/week) (b)
  None                             10,878 (35)     9,097 (29)
  < 1                              10,456 (34)    10,362 (33)
  1-6.9                             6,479 (21)     7,878 (25)
  [greater than or equal to] 7      2,931 (9)      3,508 (11)
Non-Hispanic white (b)             23,469 (76)    25,851 (83)
Education (b)
  High school or less              13,072 (42)    10,694 (34)
  Some college                      8,560 (28)     8,722 (28)
  College degree                    9,121 (29)    11,379 (37)
Mammography (b, c)                 24,094 (78)    25,144 (81)
Age at first birth (years) (b)
  Nulliparous                       6,744 (22)     6,179 (20)
  < 20                              4,908 (16)     4,006 (13)
  20-29                            16,753 (54)    18,129 (58)
  [greater than or equal to] 30     2,003 (6)      2,209 (7)
Age at menopause (years) (b)
  [less than or equal to] 42        6,324 (20)     5,565 (18)
  43-47                             5,832 (19)     5,822 (19)
  48-49                             2,771 (9)      2,898 (9)
  50-52                             7,887 (25)     8,363 (27)
  > 52                              6,089 (20)     6,520 (21)
Age at menarche (years) (b)
  < 12                              6,311 (20)     6,606 (21)
  12                                7,872 (25)     7,995 (26)
  13                                8,879 (29)     9,045 (29)
  > 13                              7,803 (25)     7,253 (23)
Unopposed E use (b,d)
  Never                            20,009 (65)    19,970 (64)
  Past                              4,164 (13)     4,026 (13)
  Current                           6,819 (22)     6,991 (23)
E + P use (b, d)
  Never                            23,818 (77)    22,918 (74)
  Past                              2,434 (8)      2,535 (8)
  Current                           4,749 (15)     5,553 (18)
Physical activity
  (MET-hr/week) (b)
  < 2.25                           10,502 (34)     8,069 (26)
  2.25-8.32                         8,162 (26)     8,234 (27)
  8.33-17.74                        5,990 (19)     7,153 (23)
  [greater than or equal            4,675 (15)     5,991 (19)
    to] 17.75
Servings vegetables (e)
  < 1.5                            23,319 (75)    16,935 (55)
  1.5-2.9                           7,043 (23)    12,685 (41)
  [greater than or equal to] 3        651 (2)      1,394 (4)
Servings grains (e)
  < 2.87                           19,850 (64)     9,795 (32)
  2.87-4.10                         8,005 (26)    11,659 (38)
  4.11-5.73                         2,587 (8)      7,196 (23)
  > 5.73                              571 (2)      2,364 (8)

Characteristic                     Quintile 3     Quintile 4
                                   9.24-11.35     11.35-14.21
                                   [micro]g/day   [micro]g/day

Total                              31,014 (100)   31,014 (100)
Age (years)
  50-54                             4,074 (13)     4,093 (13)
  55-59                             5,986 (19)     6,384 (21)
  60-69                            14,196 (46)    14,061 (45)
  70-79                             6,758 (22)     6,476 (21)
WHI study component
  Clinical trial                   13,338 (43)    13,361 (43)
  Observational study              17,676 (57)    17,653 (57)
Total energy intake
  (kcal/day) (a)
  600-1,187                         4,749 (15)     2,239 (7)
  1,188-1,539                      10,064 (33)     6,413 (21)
  1,540-1,969                      10,932 (35)    11,219 (36)
  1,970-5,000                       5,269 (17)    11,143 (36)
BMI (kg/m.sup.2]) (b)
  < 25                             11,064 (36)    10,902 (35)
  25-29.9                          10,914 (35)    10,621 (34)
  [greater than or equal to] 30     8,796 (28)     9,216 (30)
Cigarette smoking history*
  Never                            15,700 (51)    15,401 (50)
  Former                           13,036 (42)    13,553 (44)
  Current                           1,891 (6)      1,707 (6)
Alcohol consumption
  (drinks/week) (b)
  None                              8,262 (27)     8,044 (26)
  < 1                              10,296 (33)     9,841 (32)
  1-6.9                             8,399 (27)     8,863 (29)
  [greater than or equal to] 7      3,896 (13)     4,110 (13)
Non-Hispanic white (b)             26,832 (87)    26,934 (87)
Education (b)
  High school or less               9,372 (30)     8,569 (28)
  Some college                      8,584 (28)     8,445 (27)
  College degree                   12,837 (41)    13,789 (44)
Mammography (b, c)                 25,534 (82)    25,589 (83)
Age at first birth (years) (b)
  Nulliparous                       5,988 (19)     5,827 (19)
  < 20                              3,600 (12)     3,515 (11)
  20-29                            18,610 (60)    18,879 (61)
  [greater than or equal to] 30     2,369 (8)      2,352 (8)
Age at menopause (years) (b)
  [less than or equal to] 42        5,263 (17)     5,093 (16)
  43-47                             5,780 (19)     5,790 (19)
  48-49                             2,983 (10)     2,938 (9)
  50-52                             8,616 (28)     8,501 (27)
  > 52                              6,697 (22)     7,061 (23)
Age at menarche (years) (b)
  < 12                              6,678 (22)     6,919 (22)
  12                                8,191 (26)     8,192 (26)
  13                                9,041 (29)     8,980 (29)
  > 13                              6,987 (23)     6,821 (22)
Unopposed E use (b,d)
  Never                            19,809 (64)    20,026 (65)
  Past                              4,018 (13)     3,735 (12)
  Current                           7,161 (23)     7,241 (23)
E + P use (b, d)
  Never                            22,552 (73)    22,372 (72)
  Past                              2,786 (9)      2,745 (9)
  Current                           5,664 (18)     5,890 (19)
Physical activity
  (MET-hr/week) (b)
  < 2.25                            6,673 (22)     5,632 (18)
  2.25-8.32                         7,866 (25)     7,534 (24)
  8.33-17.74                        7,727 (25)     7,962 (26)
  [greater than or equal            7,248 (23)     8,527 (27)
    to] 17.75
Servings vegetables (e)
  < 1.5                             9,071 (29)     3,149 (10)
  1.5-2.9                          18,522 (60)    19,142 (62)
  [greater than or equal to] 3      3,421 (11)     8,723 (28)
Servings grains (e)
  < 2.87                            5,206 (17)     2,704 (9)
  2.87-4.10                         9,880 (32)     6,448 (21)
  4.11-5.73                        10,565 (34)    11,239 (36)
  > 5.73                            5,363 (17)    10,623 (34)

Characteristic                     Quintile 5
                                   > 14.21

Total                              31,014 (100)
Age (years)
  50-54                             4,667 (15)
  55-59                             6,518 (21)
  60-69                            13,690 (44)
  70-79                             6,139 (20)
WHI study component
  Clinical trial                   13,433 (43)
  Observational study              17,581 (57)
Total energy intake
  (kcal/day) (a)
  600-1,187                           858 (3)
  1,188-1,539                       2,994 (10)
  1,540-1,969                       7,137 (23)
  1,970-5,000                      20,025 (65)
BMI (kg/m.sup.2]) (b)
  < 25                             10,057 (32)
  25-29.9                          10,296 (33)
  [greater than or equal to] 30    10,374 (33)
Cigarette smoking history*
  Never                            15,311 (49)
  Former                           13,703 (44)
  Current                           1,581 (5)
Alcohol consumption
  (drinks/week) (b)
  None                              8,540 (28)
  < 1                               9,775 (32)
  1-6.9                             8,542 (28)
  [greater than or equal to] 7      3,990 (13)
Non-Hispanic white (b)             26,129 (84)
Education (b)
  High school or less               7,976 (26)
  Some college                      8,415 (27)
  College degree                   14,392 (46)
Mammography (b, c)                 25,438 (82)
Age at first birth (years) (b)
  Nulliparous                       6,113 (20)
  < 20                              3,459 (11)
  20-29                            18,487 (60)
  [greater than or equal to] 30     2,512 (8)
Age at menopause (years) (b)
  [less than or equal to] 42        5,123 (17)
  43-47                             5,630 (18)
  48-49                             2,851 (9)
  50-52                             8,715 (28)
  > 52                              6,921 (22)
Age at menarche (years) (b)
  < 12                              7,397 (24)
  12                                8,116 (26)
  13                                8,836 (28)
  > 13                              6,549 (21)
Unopposed E use (b,d)
  Never                            20,135 (65)
  Past                              3,750 (12)
  Current                           7,106 (23)
E + P use (b, d)
  Never                            22,407 (72)
  Past                              2,832 (9)
  Current                           5,764 (19)
Physical activity
  (MET-hr/week) (b)
  < 2.25                            4,432 (14)
  2.25-8.32                         6,422 (21)
  8.33-17.74                        8,145 (26)
  [greater than or equal           10,865 (35)
    to] 17.75
Servings vegetables (e)
  < 1.5                               612 (2)
  1.5-2.9                           9,120 (29)
  [greater than or equal to] 3     21,282 (69)
Servings grains (e)
  < 2.87                            1,192 (4)
  2.87-4.10                         2,793 (9)
  4.11-5.73                         7,159 (23)
  > 5.73                           19,870 (64)

Abbreviations: E, estrogen; E + P, estrogen and progesterone
postmenopausal hormone therapy; MET, metabolic equivalent.
(a) Participants with < 600 kcal/day or > 5,000 kcal/day were
excluded. (b) Numbers and percentages do not sum to total due
to missing information. (c) Within 2 years before enrollment.
(d) As of enrollment, including pills and patches. (e) Daily
medium-sized servings.

Table 2. Adjusted HRs (95% CIs) for breast, endometrial, and ovarian
cancer associated with energy-adjusted dietary cadmium exposure.

Outcome and             n      Cases   Model 1 HR (95% CI)   p-Trend

Breast cancer        150,889   6,658
Quintile dietary
  1                  30,171    1,198        Reference
  2                  30,185    1,378    0.96 (0.89, 1.04)
  3                  30,132    1,338    0.98 (0.91, 1.06)
  4                  30,202    1,416    1.00 (0.93, 1.08)
  5                  30,199    1,328    0.96 (0.89, 1.03)     0.63
Endometrial cancer   91,643    1,198
Quintile dietary
  1                  17,589     193         Reference
  2                  18,257     247     0.92 (0.77, 1.11)
  3                  18,423     231     0.96 (0.80, 1.15)
  4                  18,747     238     0.91 (0.76, 1.09)
  5                  18,627     289     0.89 (0.74, 1.06)     0.20
Ovarian cancer       125,569    735
Quintile dietary
  1                  25,056     123         Reference
  2                  25,091     153     1.15 (0.90, 1.46)
  3                  25,077     157     1.03 (0.80, 1.31)
  4                  25,222     138     1.42 (1.13, 1.78)
  5                  25,123     164     1.05 (0.82, 1.33)     0.22

Outcome and          Model 2 HR (95% CI)   p-Trend

Breast cancer
Quintile dietary
  1                       Reference
  2                   0.93 (0.86, 1.00)
  3                   0.94 (0.87, 1.02)
  4                   0.96 (0.89, 1.04)
  5                   0.93 (0.86, 1.00)     0.20
Endometrial cancer
Quintile dietary
  1                       Reference
  2                   0.89 (0.74, 1.07)
  3                   0.92 (0.77, 1.11)
  4                   0.87 (0.72, 1.04)
  5                   0.86 (0.72, 1.03)     0.12
Ovarian cancer
Quintile dietary
  1                       Reference
  2                   1.12 (0.88, 1.43)
  3                   1.00 (0.78, 1.28)
  4                   1.36 (1.08, 1.72)
  5                   1.01 (0.79, 1.29)     0.37

Outcome and          Model 3 HR (95% CI)   p-Trend

Breast cancer
Quintile dietary
  1                       Reference
  2                   0.92 (0.85, 1.00)
  3                   0.93 (0.86, 1.02)
  4                   0.94 (0.86, 1.03)
  5                   0.90 (0.81, 1.00)     0.12
Endometrial cancer
Quintile dietary
  1                       Reference
  2                   0.89 (0.73, 1.07)
  3                   0.92 (0.75, 1.12)
  4                   0.86 (0.69, 1.07)
  5                  0.86 (0.67, 1.1 1)     0.27
Ovarian cancer
Quintile dietary
  1                       Reference
  2                   1.05 (0.81, 1.34)
  3                   0.88 (0.67, 1.15)
  4                   1.12 (0.85, 1.48)
  5                   0.75 (0.54, 1.03)     0.22

Model 1: adjusted for total energy intake (residual method),
age and study component (observational, clinical trial).
Model 2: additional adjustment for body mass index, smoking,
alcohol consumption, race/ethnicity, education, physical
activity, age at first birth, age at menarche, age at menopause,
unopposed estrogen use, and estrogen and progesterone use.
For breast cancer only: also adjusted for mammography 2 years
before baseline. Model 3: additional adjustment for daily
vegetable servings and daily grain servings. p-Trend: Wald
test of ordinal variable for quintile of dietary cadmium.
COPYRIGHT 2014 National Institute of Environmental Health Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Research
Author:Adams, Scott V.; Quraishi, Sabah M.; Shafer, Martin M.; Passarelli, Michael N.; Freney, Emily P.; Ch
Publication:Environmental Health Perspectives
Article Type:Report
Geographic Code:1USA
Date:Jun 1, 2014
Previous Article:Assessment of the risk of medium-term internal contamination in Minamisoma city, Fukushima, Japan, after the Fukushima Dai-ichi nuclear accident.
Next Article:Cellular mechanism of the nonmonotonic dose response of bisphenol a in rat cardiac myocytes.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters