Dietary Calcium and Magnesium Intake and Mortality: A Prospective Study of Men

American Journal of Epidemiology ª The Author 2010. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org. Vol. 171, No. 7 DOI: 10.1093/aje/kwp467 Advance Access publication: February 19, 2010 Original Contribution Dietary Calcium and Magnesium Intake and Mortality: A Prospective Study of Men Joanna Kaluza, Nicola Orsini, Emily B. Levitan, Anna Brzozowska, Wojciech Roszkowski, and Alicja Wolk* * Correspondence to Prof. Alicja Wolk, Division of Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, PO Box 210, SE-171 77, Stockholm, Sweden (e-mail: alicja.wolk@ki.se). Initially submitted August 4, 2009; accepted for publication December 22, 2009. The authors examined the association of dietary calcium and magnesium intake with all-cause, cardiovascular disease (CVD), and cancer mortality among 23,366 Swedish men, aged 45 79 years, who did not use dietary supplements. Cox proportional hazards regression models were used to estimate the multivariate hazard ratios and 95% confidence intervals of mortality. From baseline 1998 through December 2007, 2,358 deaths from all causes were recorded in the Swedish population registry; through December 2006, 819 CVD and 738 cancer deaths were recorded in the Swedish cause-of-death registry. Dietary calcium was associated with a statistically significant lower rate of all-cause mortality (hazard ratio () ¼ 0.75, 95% confidence interval (CI): 0.63, 0.88; P trend < 0.001) and a nonsignificantly lower rate of CVD ( ¼ 0.77, : 0.58, 1.01; P trend ¼ 0.064) but not cancer mortality ( ¼ 0.87, : 0.65, 1.17; P trend ¼ 0.362) when the highest intake tertile (mean ¼ 1,953 mg/ day; standard deviation (SD), 334) was compared with the lowest (990 mg/day; SD, 187). Dietary magnesium intake (means of tertiles ranged from 387 mg/day (SD, 31) to 523 mg/day (SD, 38) was not associated with all-cause, CVD, or cancer mortality. This population-based, prospective study of men with relatively high intakes of dietary calcium and magnesium showed that intake of calcium above that recommended daily may reduce all-cause mortality. calcium; dietary supplements; eating; magnesium; men; mortality Abbreviations: CVD, cardiovascular disease; FFQ, food frequency questionnaire. Although the role of calcium and magnesium in homeostasis has been examined extensively, the optimal intake of the nutrients and balance between them is not clear. Calcium and magnesium may antagonize each other in terms of absorption and metabolism, particularly when intake of one is significantly higher than intake of the other (1, 2). Associations between intakes of these 2 micronutrients and incidence of major chronic diseasessuchashypertension, type 2 diabetes, cardiovascular disease (CVD), and specific cancers have been investigated in many epidemiologic studies (2 12). However, the association of calcium intake with mortality has been addressed in few investigations, with no consistent inverse associations observed with CVD mortality (13 15) and an adverse association observed with prostate cancer mortality (16, 17); to our knowledge, there are no studies of dietary magnesium intake and mortality. Most studies were limited to only drinking water as a source of calcium and magnesium when evaluating the association with CVD (18 21) or cancer (22 24) mortality. Therefore, we conducted a prospective study of dietary intakes of calcium and magnesium in relation to all-cause mortality as an ultimate outcome, taking into account both potentially protective and potentially harmful health effects of the 2 micronutrients. We analyzed data from 23,366 participants in the population-based Cohort of Swedish Men, which we followed for 10 years. We also examined whether the observed associations of each of the micronutrients with all-cause mortality depend on the level of the other one. Furthermore, we examined the associations of dietary calcium and magnesium more specifically with CVD and cancer mortality. 801

802 Kaluza et al. MATERIALS AND METHODS Study population The Cohort of Swedish Men was established in central Sweden (Västmanland and Örebro Counties) in 1997 1998. A questionnaire was sent to all men (n ¼ 100,303) aged 45 79 years living in the study area, and 48,645 men answered. For the analysis, we excluded those who returned a blank questionnaire (n ¼ 92); those who died before January 1, 1998 (n ¼ 55); those with a previous cancer diagnosis other than nonmelanoma skin cancer (n ¼ 2,592); those with a history of CVD (n ¼ 5,069) or diabetes (n ¼ 3,204); those with extreme energy intake estimates (i.e., 3 standard deviations from the mean value for log-transformed energy); and those who reported use of any types of dietary supplements (n ¼ 14,267). Men who consumed dietary supplements were excluded because there was not sufficient information on supplement composition to estimate calcium and magnesium intake from these sources. After these exclusions, the study cohort comprised 23,366 men. There were no differences in mortality rates between users (12%) and nonusers (11%) of dietary supplements during 10 years of follow-up of the cohort. Assessment of diet and other exposures Food frequency questionnaires (FFQs) were used to assess dietary intake during the previous year before cohort entry. The FFQ contained questions on 96 food items, which included the most commonly consumed foods in the study population as well as foods with high concentrations of some nutrients and/or bioactive substances. Frequency of consumption was reported according to 8 predefined categories: never/seldom, 1 3 times/month, 1 2 times/week, 3 4 times/week, 5 6 times/week, once/day, 2 times/day, and 3 times/day. Intakes of calcium, magnesium, and other nutrients were calculated by multiplying the frequency of consumption of each food item by the nutrient content of appropriate age-specific portion sizes from the Swedish National Food Administration Database (25). Validity of the FFQ has been evaluated among 248 men by using 14 24-hour recall telephone interviews spread over a whole year (26). The Spearman correlation coefficients (r) between energy-adjusted calcium and magnesium intake assessed by the FFQ and by 24-hour recalls were 0.77 and 0.73, respectively. Sociodemographic data, waist and hip circumferences, total physical activity, self-perceived health status, smoking history, and alcohol drinking habits were collected by using a selfadministered questionnaire. The questionnaire also included information about hypertension and hypercholesterolemia. Follow-up of the cohort and ascertainment of cases Dates of death or migration from the study area were obtained from the Swedish Death and Population Registers at Statistics Sweden through December 31, 2007. Detailed information on causes of death was available from the Swedish Register of Death Causes at the National Board of Health and Welfare through December 31, 2006, because there is up to a 1-year delay in completion of these register data. from cardiovascular (codes I00 I79) and cancer (codes C00 C97) causes were classified according to the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision. Completeness of the registers is nearly 100%. Statistical analysis Cox proportional hazards regression models were used to estimate the hazard ratios and 95% confidence intervals of all-cause, CVD, and cancer mortality among men. Survival time was calculated as the time from study entry until death; migration; December 31, 2007 (for analysis of all-cause mortality); or December 31, 2006 (for cause-specific mortality analysis), whichever was earlier. Multivariate hazard ratio estimates were adjusted for age (continuous variable); marital status (single, married, divorced or widowed); education (less than high school, high school or university); total physical activity (40.9 metabolic equivalent task hours 3 hours/day (median), >40.9 metabolic equivalent task hours 3 hours/day) (27); selfperceived health status (very good, good, average, bad, or very bad); waist-to-hip ratio (<0.94 (median), 0.94); smoking status (never, former, or current); alcohol consumption (5 g ethanol/day, >5 g ethanol/day); and intakes of dietary fiber (quartiles), saturated fatty acids (quartiles), vitamin D (quartiles), and phosphorus (quartiles). Intakes of nutrients were adjusted for energy intake of 2,600 kcal/day by using the residual method (28). Multivariate hazard ratio estimates for tertiles of intake of calcium and magnesium were mutually adjusted. Including 4 categories (quartiles) for waist-to hip ratio, alcohol consumption, and total physical activity did not materially alter the results. We checked whether the proportional hazards assumption was reasonable in the multivariate models. Scaled Schoenfeld s residuals were regressed against survival time. There was no evidence of departure from the assumption. To calculate P values for trend, the median values of tertiles of calcium and magnesium intake were used as a continuous variable. Using the likelihood ratio test, we tested statistical interactions between calcium and magnesium intake in predicting mortality. All statistical analyses were performed with Stata, version 10, software (StataCorp, College Station, Texas). All reported P values are 2-sided; P values 0.05 were considered statistically significant. RESULTS The baseline characteristics of the study population, by tertiles of energy-adjusted calcium and magnesium intake, are shown in Table 1. Mean dietary calcium intake increased 2-fold between the lowest tertile (990 mg/day (standard deviation, 187)) and the highest tertile (1,953 mg/day (standard deviation, 334)). The main sources of this nutrient in the diet were milk and milk products (68.5%) and cereal products (8.4%). Compared with men in the lowest tertile of calcium intake, those in the highest tertile had higher intakes of protein, fat, and phosphorus and lower intakes of alcohol,

Calcium and Magnesium Intake and Mortality 803 Table 1. Age-standardized Baseline Characteristics a of a Cohort of 23,366 Swedish Men Aged 45 79 Years by Tertiles of Energy-adjusted Dietary Calcium and Magnesium Intake, 1998 Characteristic <1,230 (n 5 7,786) Calcium Intake, mg/day b 1,230 1,598 (n 5 7,788) carbohydrates, and dietary fiber. Intake of magnesium was generally high in the study population and ranged from 387 mg/day (standard deviation, 31) in the lowest tertile to 523 mg/day (standard deviation, 38) in the highest tertile. Only 5% of men had a magnesium intake of less than 314 mg/day. Thirty-three percent of magnesium was obtained from bread and cereal products, 17.5% from milk and dairy products, and 11.9% from vegetables. Among men in the lowest tertile of magnesium intake, dietary fiber, calcium, and phosphorus intake was lower and fat intake was higher than among men in the highest tertile. Men in the lowest tertile also had a lower educational level, and more of them were smokers. The Pearson s correlation between calcium and magnesium intake was 0.18. During an average of 10 years of follow-up (224,206 person-years, 1998 2007), 2,358 deaths from all causes were registered. From baseline 1998 through December 2006 (203,036 person-years), 819 CVD deaths and 738 cancer deaths were registered. 1,599 (n 5 7,792) <426 (n 5 7,786) Magnesium Intake, mg/day b 426 480 (n 5 7,788) 481 (n 5 7,792) Calcium, mg/day 990 (187) 1,408 (105) 1,953 (334) 1,357 (428) 1,456 (431) 1,541 (489) Magnesium, mg/day 443 (65) 454 (60) 466 (62) 387 (31) 453 (15) 523 (38) Age, years 57.6 (9.0) 58.1 (9.0) 58.5 (9.2) 58.6 (9.5) 58.5 (9.2) 57.0 (8.5) Waist-to-hip ratio 0.94 (0.08) 0.94 (0.07) 0.94 (0.07) 0.94 (0.08) 0.94 (0.07) 0.94 (0.07) Total physical activity, MET 3 41.6 (5.0) 41.7 (4.9) 41.9 (5.0) 41.6 (5.1) 41.6 (4.8) 41.9 (5.0) hours/day University education >12 years 17.2 17.6 15.9 14.3 17.7 18.8 Married 83.0 84.8 83.6 81.9 85.7 83.9 Health status very good or good c 79.0 82.4 81.6 77.6 81.7 83.9 Smoking Never 35.5 37.5 36.0 35.6 37.1 36.3 Current 26.3 24.0 26.6 28.6 24.3 23.9 Energy from diet, kcal/day 2,608 (791) 2,715 (742) 2,673 (760) 2,655 (793) 2,677 (744) 2,663 (761) Alcohol consumption, g/day 13.1 (11.5) 10.7 (9.3) 8.4 (7.7) 10.0 (9.8) 11.1 (9.7) 11.2 (9.9) Adjusted daily intake of energy Total protein, g 91 (13) 101 (9) 113 (12) 97 (14) 102 (13) 106 (14) Total fat, g 86 (15) 91 (13) 95 (15) 97 (15) 91 (13) 83 (13) Saturated 36.9 (7.8) 41.6 (7.7) 46.1 (9.8) 45.4 (9.3) 41.9 (8.2) 37.2 (8.4) Monounsaturated 30.3 (5.4) 30.8 (4.6) 30.8 (4.8) 33.2 (4.7) 30.9 (4.1) 27.7 (4.4) Polyunsaturated 12.5 (2.7) 11.9 (2.1) 10.8 (2.0) 11.8 (2.5) 11.8 (2.3) 11.7 (2.3) Total carbohydrates, g 339 (39) 323 (33) 307 (37) 314 (41) 321 (36) 335 (36) Dietary fiber, g 31.2 (8.5) 30.2 (7.4) 27.6 (7.4) 23.9 (5.4) 29.4 (5.6) 35.7 (7.6) Vitamin D, lg 6.2 (2.3) 6.3 (2.1) 6.7 (2.4) 6.2 (2.2) 6.5 (2.2) 6.6 (2.6) Phosphorus, mg 1,754 (223) 2,016 (178) 2,350 (265) 1,840 (284) 2,044 (269) 2,238 (314) Abbreviation: MET, metabolic equivalent task hours. a Values are expressed as mean (standard deviation) or percentage. b Energy adjusted to 2,600 kcal/day by using the residual method (28). The 5th 95th percentiles were 798 2,258 mg/day for calcium and 355 561 mg/day for magnesium. c Self-reported. We observed inverse associations between energyadjusted calcium intake and risk of all-cause and CVD mortality (Table 2). Compared with men in the lowest tertile of calcium intake, those in the highest tertile had a 25% (95% confidence interval: 0.63, 0.88; P trend < 0.001) lower risk of all-cause mortality and a 23% (95% confidence interval: 0.58, 1.01; P trend ¼ 0.064) lower risk of CVD mortality, although the latter association was not statistically significant at the 0.05 level. Further adjustment of CVD mortality for hypertension and high cholesterol did not change the risk estimates. We found no association between calcium intake and cancer mortality. In multivariate analyses, we did not observe associations between magnesium intake and all-cause, CVD, or cancer mortality. Exclusion of the first 2 years of follow-up did not substantially change the observed risk estimates. Men in the highest, compared with the lowest, tertile of dietary calcium intake had a 24% lower mortality rate from all causes

804 Kaluza et al. Table 2. Age- and Multivariate-adjusted Hazard Ratios and 95% Confidence Intervals a for All-cause, CVD, and Cancer Mortality Among 23,366 Swedish Men by Tertiles of Energy-adjusted Dietary Calcium and Magnesium Intake, 1998 2007 Calcium Intake, mg/day b <1,230 (n 5 7,786) 1,230 1,598 (n 5 7,788) 1,599 (n 5 7,792) All-cause mortality deaths 783 762 813 Age adjusted 1.00 0.91 0.83, 1.01 0.93 0.84, 1.02 0.169 Multivariate adjusted 1.00 0.89 0.80, 1.00 0.75 0.63, 0.88 <0.001 CVD mortality deaths 257 265 297 Age adjusted 1.00 0.96 0.81, 1.14 1.01 0.85, 1.19 0.889 Multivariate adjusted 1.00 0.91 0.75, 1.10 0.77 0.58, 1.01 0.064 Cancer mortality deaths 232 250 256 Age adjusted 1.00 1.01 0.85, 1.21 0.99 0.83, 1.18 0.887 Multivariate adjusted 1.00 1.00 0.82, 1.22 0.87 0.65, 1.17 0.362 Magnesium Intake, mg/day b <426 (n 5 7,786) 426 480 (n 5 7,788) 481 (n 5 7,792) All-cause mortality deaths 889 831 638 Age adjusted 1.00 0.97 0.88, 1.07 0.91 0.82, 1.01 0.067 Multivariate adjusted 1.00 1.09 0.97, 1.22 1.06 0.91, 1.24 0.385 CVD mortality deaths 307 289 223 Age adjusted 1.00 0.98 0.83, 1.15 0.95 0.80, 1.13 0.559 Multivariate adjusted 1.00 1.15 0.95, 1.38 1.25 0.96, 1.61 0.086 Cancer mortality deaths 260 259 219 Age adjusted 1.00 1.01 0.85, 1.21 1.00 0.84, 1.21 0.969 Multivariate adjusted 1.00 1.05 0.86, 1.28 0.98 0.75, 1.28 0.878 Abbreviations: CI, confidence interval; CVD, cardiovascular disease;, hazard ratio. a Multivariate hazard ratios and 95% confidence intervals adjusted for age, marital status, education, self-reported health status, smoking status, physical activity, waist-to-hip ratio, alcohol consumption, energy-adjusted dietary fiber, saturated fatty acid, vitamin D, and phosphorus intake. Hazard ratios for energy-adjusted calcium and magnesium intake are mutually adjusted. b Energy adjusted to 2,600 kcal/day by using the residual method (28). The 5th 95th percentiles were 798 2,258 mg/day for calcium and 355 561 mg/day for magnesium. P trend (hazard ratio ¼ 0.76, 95% confidence interval: 0.64, 0.90; P trend ¼ 0.002) and a 28% lower risk of CVD mortality (hazard ratio ¼ 0.82, 95% confidence interval: 0.61, 1.11; P trend ¼ 0.209), which remained not statistically significant. We next cross-classified subjects by tertiles of calcium and magnesium intake to examine a potential interaction between intake of energy-adjusted calcium and magnesium in predicting mortality (Table 3). However, we found no evidence of statistically significant interaction for mortality from all causes (P ¼ 0.148) or for CVD (P ¼ 0.478) and cancer (P ¼ 0.768). DISCUSSION In this population-based, prospective cohort study of men not taking dietary supplements, we found that dietary calcium intake was inversely associated with mortality from all causes. Men in the highest, compared with the lowest, tertile of calcium intake had a statistically significant 25% lower rate of mortality from all causes. Cardiovascular mortality was decreased by 23% when we compared the highest with the lowest tertiles of calcium intake, although this reduction was not statistically significant. We did not observe an

Calcium and Magnesium Intake and Mortality 805 Table 3. Multivariate-adjusted Hazard Ratios and 95% Confidence Intervals a for All-cause Mortality Among 23,366 Swedish Men by Energy-adjusted Dietary Intake of Calcium and Magnesium, 1998 2007 Calcium Intake, mg/day a Magnesium Intake, mg/day b <426 426 480 481 <1,230 395 1.00 236 1.07 0.90, 1.27 152 0.95 0.76, 1.17 1,230 1,598 272 0.85 0.72, 0.99 307 0.99 0.84, 1.17 183 0.86 0.69, 1.07 1,599 222 0.70 0.57, 0.86 288 0.74 0.60, 0.91 303 0.85 0.67, 1.07 Abbreviations: CI, confidence interval; CVD, cardiovascular disease;, hazard ratio. a Multivariate hazard ratios and 95% confidence intervals adjusted for age, marital status, education, self-reported health status, smoking status, physical activity, waist-to-hip ratio, alcohol consumption, energy-adjusted dietary fiber, saturated fatty acid, vitamin D, and phosphorus intake. The interaction between calcium or magnesium intake was not statistically significant (P ¼ 0.148). b Energy-adjusted to 2,600 kcal/day by using the residual method (28). The 5th 95th percentiles were 798 2,258 mg/day for calcium and 355 561 mg/day for magnesium. association between calcium intake and cancer mortality. Dietary magnesium intake was not associated with allcause, cardiovascular, or cancer mortality. Our findings for calcium are in line with results from the Women s Health Initiative randomized trial, in which calcium plus vitamin D supplementation was associated with a nonsignificant decrease in all-cause mortality (14). They are also in line with results from a prospective cohort study of 34,486 postmenopausal US women, in which the highest versus the lowest quartiles of dietary and total (diet þ supplements) calcium intake were associated with statistically significant 37% and 33% lower mortality from ischemic heart disease, respectively (13). Similarly, in an ecologic study, a high level of calcium in drinking water (94 146 mg/l vs. 9 53 mg/l) was associated with a statistically significant lower risk of noncerebrovascular (10%) and cerebrovascular (14%) causes of death among elderly people from the southwest of France (29). Dairy products as a source of calcium were considered in a prospective study of a Japanese population (46,465 men and 64,327 women aged 40 79 years) with a generally low dietary calcium intake (15). The risk of mortality for men in the highest versus the lowest quintile of dairy calcium intake was 47% lower for total stroke, 54% lower for hemorrhagic stroke, and 47% lower for ischemic stroke, and the associations were statistically significant; corresponding risk estimates for women were 43%, 49%, and 50% lower. However, there was an internal inconsistency in that study because total calcium intake was nonsignificantly associated with stroke mortality. Adjustment for hypertension and high cholesterol did not change the estimates of risk of CVD mortality associated with calcium intake. This finding may be because both blood pressure and cholesterol levels have a graded, continuous association with CVD mortality and our study assessed them as only dichotomous variables or because the pathways are not the most important in cardiovascular death. An alternate explanation that cannot be ruled out is confounding by factors associated with dietary calcium intake that were not measured in this study. The beneficial effect of calcium intake on mortality could be explained by a positive influence of calcium on blood pressure, serum cholesterol level, and blood glucose. Results of a meta-analysis of randomized controlled trials indicated a statistically significant decrease in systolic blood pressure with calcium supplementation (30). Calcium supplement intake (mean dose ¼ 1,200 mg/day) was associated with reduced systolic blood pressure, and the effect was greater for people with a relatively low dietary calcium intake (800 mg/day). It was also reported that a highcalcium diet was associated with significantly reduced total cholesterol and low density lipoprotein cholesterol and did not change high density lipoprotein cholesterol in serum (31). During a weight-loss intervention among women, a significantly greater decrease in low density lipoprotein cholesterol was observed in the group with calcium and vitamin D supplementation than in the placebo group (32). Dietary calcium has also been associated with lower risk of diabetes mellitus, which increases the risk of all-cause and CVD mortality (33). An inverse association between magnesium concentration in drinking water and CVD mortality was observed for women from southern Sweden (19) and in a French population (29). To the best of our knowledge, there are no studies of dietary and total magnesium intake and mortality. A dietary deficiency of magnesium and abnormalities in magnesium metabolism are often associated with an imbalance of calcium, sodium, and potassium (34). Magnesium deficiency plays an important role in heart diseases, such as ischemic heart disease, congestive heart failure, sudden cardiac death, atherosclerosis, and a number of cardiac arrhythmias (34). We did not observe an association between dietary magnesium intake and mortality in the present study. However, note that, compared with other populations, ours had, in general, a high magnesium intake (mean ¼ 450 mg/day, 5th percentile ¼ 355 mg/day) and a high calcium intake (mean ¼ 1,400 mg/day, 5th percentile ¼ 798 mg/day). For example, in the US National Health and Nutrition Examination Survey, mean intake of magnesium was less than 300 mg/day among middle-aged and older men who did not use supplements (35). Among 51,529 US men aged 40 75 years included in the Health Professionals Follow-up Study, the

806 Kaluza et al. median of the lowest quintile of magnesium intake (diet and supplements) was 243 mg/day and in the highest quintile was 452 mg/day; the medians of calcium intake were 500 mg/day and 1,400 mg/day, respectively (36). In other words, it seems that there were no magnesium-deficient participants in our study population, which could explain the overall lack of association with magnesium. Further studies are needed in other populations with lower dietary magnesium intakes to address this issue. Strengths of our study include the population-based, prospective design and detailed information on diet. The prospective design and practically complete cohort follow-up minimized the possibility of recall bias and bias caused by loss to follow-up. The extensive data on possible risk factors for mortality allowed comprehensive adjustment of confounders. Although the FFQ-based estimates of dietary calcium and magnesium intake had relatively high validity, some misclassification of calcium and magnesium intake is inevitable. Measurement error in assessing dietary intake in a prospective setting leading to nondifferential misclassification would have likely attenuated rather than exaggerated the true associations. Furthermore, it is not known whether participants diet during the baseline year reflected their diet during the biologically relevant period. Moreover, we did not have data about calcium and magnesium intake from drinking water, which could be a relevant source of these minerals, and men who consumed these nutrients in dietary supplements were excluded because of lack of information on the composition of the supplements. In Sweden, dietary supplement use is associated with higher education, lower body weight, more physical activity, and worse subjective health (37). Exclusion of supplement users could result in a loss of generalizability if they were different from the analytical cohort in terms of occurrence of death. However, we found that users and nonusers of dietary supplements had very similar mortality rates. We cannot rule out the possibility that some unmeasured confounders or residual confounding accounted for the observed associations. In conclusion, the findings from this prospective cohort study of men support an inverse association between dietary calcium intake and all-cause mortality. The inverse association between dietary calcium intake and CVD mortality was not statistically significant, and there was no apparent relation between magnesium and all-cause, CVD, or cancer mortality. However, these results have to be considered in the context of the study population with relatively high intakes of dietary calcium and magnesium. Future studies should examine intake of calcium and magnesium from diet, from drinking water, and from dietary supplements in relation to all-cause and specific causes of mortality, especially in populations in which potential deficiency of these minerals is suspected. ACKNOWLEDGMENTS Author affiliations: Division of Nutritional Epidemiology, Karolinska Institutet, Stockholm, Sweden (Joanna Kaluza, Nicola Orsini); Department of Human Nutrition, Warsaw University of Life Sciences-SGGW, Warsaw, Poland (Joanna Kaluza, Anna Brzozowska, Wojciech Roszkowski); and Department of Epidemiology, University of Alabama at Birmingham, Birmingham, Alabama (Emily B. Levitan). This study was supported by the Swedish Council for Working Life and Social Research (FAS), the Swedish Research Council/Longitudinal studies, the Swedish Cancer Society, and the Swedish Foundation for International Cooperation in Research and Higher Education (STINT). Conflict of interest: none declared. REFERENCES 1. Solomons NW. Physiological interactions of minerals. In: Bodwell CE, Erdman JW Jr, eds. Nutrient Interactions: IFT Basic Symposium Series. New York, NY: Marcel Dekker, Inc; 1988:115 141. 2. Weaver CM. Calcium. In: Bowman BA, Russell RM, eds. Present Knowledge in Nutrition. Washington, DC: ILSI Press; 2001:273 280. 3. Chacko SA, Song Y, Nathan L, et al. Relations of dietary magnesium intake to biomarkers of inflammation and endothelial dysfunction in an ethnically diverse cohort of postmenopausal women. Diabetes Care. 2009;33(2):304 310. 4. Cho E, Smith-Warner SA, Spiegelman D, et al. Dairy foods, calcium, and colorectal cancer: a pooled analysis of 10 cohort studies. J Natl Cancer Inst. 2004;96(13):1015 1022. 5. Iso H, Stampfer MJ, Manson JE, et al. Prospective study of calcium, potassium, and magnesium intake and risk of stroke in women. Stroke. 1999;30(9):1772 1779. 6. Larsson SC, Bergkvist L, Rutegård J, et al. Calcium and dairy food intakes are inversely associated with colorectal cancer risk in the Cohort of Swedish Men. Am J Clin Nutr. 2006; 83(3):667 673; quiz 728 729. 7. Larsson SC, Bergkvist L, Wolk A. Magnesium intake in relation to risk of colorectal cancer in women. JAMA. 2005; 293(1):86 89. 8. Song Y, Li TY, van Dam RM, et al. Magnesium intake and plasma concentrations of markers of systemic inflammation and endothelial dysfunction in women. Am J Clin Nutr. 2007;85(4):1068 1074. 9. Song Y, Manson JE, Buring JE, et al. Dietary magnesium intake in relation to plasma insulin levels and risk of type 2 diabetes in women. Diabetes Care. 2004;27(1):59 65. 10. Song Y, Manson JE, Cook NR, et al. Dietary magnesium intake and risk of cardiovascular disease among women. Am J Cardiol. 2005;96(8):1135 1141. 11. Song Y, Sesso HD, Manson JE, et al. Dietary magnesium intake and risk of incident hypertension among middle-aged and older US women in a 10-year follow-up study. Am J Cardiol. 2006;98(12):1616 1621. 12. Strain JJ, Cashman KD. Minerals and trace elements. In: Gibney MJ, Vorster HH, Kok FJ, eds. Introduction to Human Nutrition. Berlin, Germany: Blackwell Sciences Press; 2002:178 185. 13. Bostick RM, Kushi LH, Wu Y, et al. Relation of calcium, vitamin D, and dairy food intake to ischemic heart disease mortality among postmenopausal women. Am J Epidemiol. 1999;149(2):151 161. 14. LaCroix AZ, Kotchen J, Anderson G, et al. Calcium plus vitamin D supplementation and mortality in postmenopausal

Calcium and Magnesium Intake and Mortality 807 women: the Women s Health Initiative calcium-vitamin D randomized controlled trial. J Gerontol A Biol Sci Med Sci. 2009;64(5):559 567. 15. Umesawa M, Iso H, Date C, et al. Dietary intake of calcium in relation to mortality from cardiovascular disease: the JACC Study. Stroke. 2006;37(1):20 26. 16. Giovannucci E, Liu Y, Platz EA, et al. Risk factors for prostate cancer incidence and progression in the Health Professionals Follow-up Study. Int J Cancer. 2007;121(7):1571 1578. 17. Skinner HG, Schwartz GG. A prospective study of total and ionized serum calcium and fatal prostate cancer. Cancer Epidemiol Biomarkers Prev. 2009;18(2):575 578. 18. Rubenowitz E, Axelsson G, Rylander R. Magnesium in drinking water and death from acute myocardial infarction. Am J Epidemiol. 1996;143(5):456 462. 19. Rubenowitz E, Axelsson G, Rylander R. Magnesium and calcium in drinking water and death from acute myocardial infarction in women. Epidemiology. 1999;10(1):31 36. 20. Yang CY. Calcium and magnesium in drinking water and risk of death from cerebrovascular disease. Stroke. 1998;29(2):411 414. 21. Yang CY, Chang CC, Tsai SS, et al. Calcium and magnesium in drinking water and risk of death from acute myocardial infarction in Taiwan. Environ Res. 2006;101(3):407 411. 22. Chiu HF, Chang CC, Yang CY. Magnesium and calcium in drinking water and risk of death from ovarian cancer. Magnes Res. 2004;17(1):28 34. 23. Yang CY, Chiu HF, Cheng MF, et al. Calcium and magnesium in drinking water and the risk of death from breast cancer. J Toxicol Environ Health A. 2000;60(4):231 241. 24. Yang CY, Chiu HF, Tsai SS, et al. Magnesium and calcium in drinking water and the risk of death from esophageal cancer. Magnes Res. 2002;15(3-4):215 222. 25. Bergström L, Kylberg E, Hagman U, et al. The food consumption database KOST: the National Food Administration s Information System for nutritive values of food. VårFöda. 1991;43:439 447. 26. Messerer M, Johansson SE, Wolk A. The validity of questionnaire-based micronutrient intake estimates is increased by including dietary supplement use in Swedish men. J Nutr. 2004;134(7):1800 1805. 27. Norman A, Bellocco R, Bergström A, et al. Validity and reproducibility of self-reported total physical activity differences by relative weight. Int J Obes Relat Metab Disord. 2001;25(5):682 688. 28. Willett W, Stampfer MJ. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol. 1986;124(1):17 27. 29. Marque S, Jacqmin-Gadda H, Dartigues JF, et al. Cardiovascular mortality and calcium and magnesium in drinking water: an ecological study in elderly people. Eur J Epidemiol. 2003; 18(4):305 309. 30. van Mierlo LA, Arends LR, Streppel MT, et al. Blood pressure response to calcium supplementation: a meta-analysis of randomized controlled trials. J Hum Hypertens. 2006;20(8): 571 580. 31. Denke MA, Fox MM, Schulte MC. Short-term dietary calcium fortification increases fecal saturated fat content and reduces serum lipids in men. J Nutr. 1993;123(6):1047 1053. 32. Major GC, Alarie F, Doré J, et al. Supplementation with calcium þ vitamin D enhances the beneficial effect of weight loss on plasma lipid and lipoprotein concentrations. Am J Clin Nutr. 2007;85(1):54 59. 33. van Dam RM, Hu FB, Rosenberg L, et al. Dietary calcium and magnesium, major food sources, and risk of type 2 diabetes in U.S. black women. Diabetes Care. 2006;29(10):2238 2243. 34. Chakraborti S, Chakraborti T, Mandal M, et al. Protective role of magnesium in cardiovascular diseases: a review. Mol Cell Biochem. 2002;238(1-2):163 179. 35. Ford ES, Mokdad AH. Dietary magnesium intake in a national sample of US adults. J Nutr. 2003;133(9):2879 2882. 36. Ascherio A, Rimm EB, Hernán MA, et al. Intake of potassium, magnesium, calcium, and fiber and risk of stroke among US men. Circulation. 1998;98(12):1198 1204. 37. Messerer M, Johansson SE, Wolk A. Sociodemographic and health behaviour factors among dietary supplement and natural remedy users. Eur J Clin Nutr. 2001;55(12):1104 1110.