Dietary Intake of Calcium in Relation to Mortality From Cardiovascular Disease
The JACC Study
Background and Purpose— No prospective studies have examined the association between calcium intake and the risk of cardiovascular disease in Japanese populations with a low mean calcium intake.
Methods— Between 1988 and 1990, 110 792 Japanese subjects (46 465 men and 64 327 women) 40 to 79 years of age without a history of stroke, coronary heart disease, or cancer, completed a lifestyle questionnaire including food intake frequency under the Japan Collaborative Cohort (JACC) Study for Evaluation of Cancer Risk Sponsored by Monbusho. By the end of 1999, after 515 029 person years of follow-up, 566 deaths from stroke (101 subarachnoid hemorrhages, 140 intraparenchymal hemorrhages, and 273 ischemic strokes) and 234 deaths from coronary heart disease had been documented.
Results— The intake of total calcium tended to be inversely associated with mortality from total stroke but not from coronary heart disease or total cardiovascular disease for men and women. The inverse association with dairy calcium intake was apparent for total stroke, both hemorrhagic and ischemic. The multivariate relative risk for men with highest versus lowest quintiles of dairy calcium intake was 0.53 (95% CI, 0.34 to 0.81) for total stroke, 0.46 (0.23 to 0.91) for hemorrhagic stroke, and 0.53 (0.29 to 0.99) for ischemic stroke; corresponding relative risks for women were 0.57 (0.38 to 0.86), 0.51 (0.28 to 0.94), and 0.50 (0.27 to 0.95).
Conclusions— Dietary calcium intake from dairy products was associated with reduced mortality from stroke for Japanese men and women.
Findings from studies assessing the relationship between dietary intake of milk, dairy products, and calcium with the risk of cardiovascular disease have been inconsistent. Five prospective studies have examined the association between milk or calcium intake with the risk of mortality or incidence of coronary heart disease; however, none of these studies demonstrated any significant association.1–5 Two prospective studies have shown that dietary intake of milk or calcium reduces the risk of ischemic stroke;1,2 however, 2 other studies failed to demonstrate this association.3,4 These studies were undertaken in Western countries where the intake of calcium is greater than that in Eastern countries.
The mean calcium intake among the Japanese is far lower than that of whites in Western countries, largely because of a lower intake of milk and dairy products.6,7 According to the national nutrition surveys, the mean intake of calcium in Japan was 559 mg per day for men and 535 mg per day for women in 2002,6 whereas corresponding values in United States were 966 mg per day and 765 mg per day between 1999 and 2000.7 The purpose of the present study was to determine the relationship between calcium intake and mortality from cardiovascular disease in a large prospective study of Japanese men and women.
The Japan Collaborative Cohort (JACC) Study for Evaluation of Cancer Risks, sponsored by the Ministry of Education, Sport, and Science, was conducted from 1988 to 1990. A total of 110 792 subjects (46 465 men and 64 327 women) 40 to 79 years of age completed self-administered questionnaires about their lifestyles and medical histories of previous cardiovascular disease and cancer. They were enrolled from 45 communities across Japan, mostly when they underwent municipal health screening examinations according to the Health Law for the Aged.8,9 The sampling methods and protocols of the JACC Study have been described previously.8 Subjects with a previous history of stroke, coronary heart disease, or cancer were excluded from analysis. Therefore, 41 782 men and 52 901 women were followed; 21 068 men and 32 319 women provided valid responses to questions regarding milk, cheese, and yogurt consumption and were enrolled in the present study.
For mortality surveillance in each community, investigators conducted a systematic review of death certificates, all of which were forwarded to the public health center in the area of residency. Mortality data were sent centrally to the Ministry of Health and Welfare, and the underlying cause of death coded according to the International Classification of Disease, 9th Revision from 1988 to 1994, and 10th Revision from 1995 to 1999, for the National Vital Statistics. In Japan, registration of death is required by the Family Registration Law and is believed to be followed across Japan. Therefore, all deaths that occurred in the cohort were ascertained by death certificates from a public health center, except for subjects who died after they had moved from their original community, in which case the subject was treated as a censored case. The follow-up was conducted at the end of 1999. The average follow-up period for the participants was 9.6 years. The present study was approved by the ethics committees of the Nagoya University School of Medicine and the University of Tsukuba.
Intake of Milk, Yogurt, and Cheese and Calculation of Calcium Intake
Each participant was asked to record the frequency of the intake of 35 foods including milk, yogurt, and cheese. Five responses were possible for each food item: “rarely,” “1 to 2 days per month,” “1 to 2 days per week,” “3 to 4 days per week,” and “almost every day”; the consumption of each food was calculated by multiplying the frequency score of consumption of each food 0, 0.38, 1.5, 3.5, and 7, respectively. Each portion size was estimated from the validation study conducted in 85 individuals from the baseline participants. For example, 121 mg of calcium for 1 serving of milk, 118 mg for 1 serving of yogurt, and 107 mg for 1 serving of cheese. Based on Japan Food Table version 4, the average daily intake of nutrients and total energy was calculated by multiplying the frequency of consumption of each item by its nutrient content and energy per serving and totaling the nutrient intake for all food items.10
We obtained the nutritional data of valid calcium intake for 23 117 men and 35 609 women. The reproducibility and validity of this dietary questionnaire was reported previously.10 Briefly, the Spearman rank correlation coefficients between first and second questionnaires, conducted 1 year apart, were 0.76 in total calcium intake, 0.53 in dairy calcium intake, and 0.76 in nondairy calcium intake. The coefficients between the average of 2 questionnaires and 4 3-day dietary records were 0.44 in total calcium intake, 0.58 in dairy calcium intake, and 0.17 in nondairy calcium intake.
Statistical analysis was based on sex-specific mortality rates of stroke during the follow-up period from 1989 to 1999. For each participant, the person years of follow-up were calculated from the date of filling out the baseline questionnaire to death, moving out of the community, or the end of 1999, whichever was first. The sex-specific relative risk of mortality from stroke was defined as the death rate among participants according to the quintile of calcium intake: <315, 315 to 407, 408 to 490, 491 to 589, and ≥590 mg per day for total calcium, <7, 7 to 37, 38 to 120, 121 to 127, and ≥128 mg per day for dairy calcium, and <262, 262 to 335, 336 to 403, 404 to 484, and ≥485 mg per day for nondairy calcium, in men, and in women, <331, 331 to 421, 422 to 501, 502 to 595, and ≥596 mg per day for total calcium, <12, 12 to 64, 65 to 122, 123 to 143, and ≥144 mg per day for dairy calcium, and <269, 269 to 336, 337 to 399, 400 to 473, and ≥474 mg per day for nondairy calcium.
Age-adjusted means, proportions of selected cardiovascular risk factors, and the calcium intake were presented according to the quintile of calcium intake. Statistical testing was not conducted because of the large sample size. The age-adjusted and multivariate-adjusted relative risks and their 95% CIs were calculated after adjustment for age and potential confounding factors by using the Cox proportional hazard model. These confounding variables included body mass index (sex-specific quintiles), smoking status (never, ex-smokers, and current smokers of 1 to 19 and ≥20 cigarettes per day), alcohol intake category (never, ex-drinkers, and current drinkers of ethanol at 1 to 22, 23 to 45, 46 to 68, and ≥69 g per day), history of hypertension (yes or no), history of diabetes (yes or no), energy intake (sex-specific quintiles), and potassium intake (sex-specific quintiles). Test for a linear trend across the calcium intake quintiles were conducted by linear regression using a median variable of calcium intake in each calcium intake quintile.
The significance of the interactions of smoking status (yes or no) or alcohol status (yes or no) with dairy calcium intake (dummy variables for sex-specific quintiles) was tested using an interaction term of 2 variables in multivariate models.
Cause-specific mortality was calculated according to the International Classification of Disease, 9th and 10th revisions, which define total stroke (ICD-9 codes 430 to 438 and ICD-10 codes I60 to I69), and further subgroups total stroke into subarachnoid hemorrhage (ICD-9 code 430 and ICD-10 code I60), intraparenchymal hemorrhage (ICD-9 code 431 and ICD-10 code I61), and ischemic stroke (ICD-9 codes 433 to 434 and ICD-10 codes I63 and I693).
Among 21 068 men and 32 319 women followed up for 9.6 years, there were 284 deaths from stroke (37 subarachnoid hemorrhage, 76 intraparenchymal hemorrhage, and 146 ischemic stroke) in men and 282 in women (n=64, 64, and 127, respectively).
Table 1 shows sex-specific selected cardiovascular risk factors, total energy, and potassium intake according to quintiles of total, dairy, and nondairy calcium intake. Compared with men and women in the lowest quintile, those in the highest quintile smoked less, drank less, had higher intakes of total energy and potassium, and were less likely to have a history of hypertension.
Table 2⇓ shows age-adjusted and multivariate relative risks (95% CI) of stroke, stroke subtypes, coronary heart disease, and cardiovascular disease according to quintiles of total calcium intake. For men and women, total calcium intake tended to be inversely associated with mortality from total and ischemic strokes: the associations were of statistical significance for total stroke in women and of borderline significance for ischemic stroke in women but did not reach statistical significance for either end point in men. When men and women were combined, adjusting for sex, the multivariate relative risk in the highest versus lowest quintiles of total calcium intake was 0.86 (0.56 to 1.31) for total stroke (P for trend=0.003) and 0.74 (0.41 to 1.36) for ischemic stroke (P for trend=0.03). Total calcium intake was not associated with mortality from coronary heart disease or total cardiovascular disease.
Table 3⇓ shows multivariate relative risks according to quintiles of dairy calcium intake. For dairy calcium, there was an inverse association between dairy calcium intake and mortality from total stroke, either hemorrhagic or ischemic, and total cardiovascular disease but not coronary heart disease. The multivariate relative risk (95% CI) in the highest versus lowest quintiles of dairy calcium intake for men was 0.53 (0.34 to 0.81) for total stroke and 0.46 (0.23 to 0.91) for hemorrhagic stroke, 0.53 (0.29 to 0.99) for ischemic stroke, and 0.73 (0.55 to 0.95) for total cardiovascular disease. The respective relative risk for women was 0.57 (0.38 to 0.86), 0.51 (0.28 to 0.94), 0.50 (0.27 to 0.95), and 0.77 (0.58 to 1.03). These associations did not differ significantly when stratified by smoking and drinking status (data not shown in the table). For example, the multivariate relative risk of total stroke for men and women combined in the highest versus lowest quintiles of dairy calcium intake was 0.55 (0.32 to 0.94) among current smokers and 0.63 (0.44 to 0.92) among nonsmokers (P for interaction=0.31), 0.37 (0.23 to 0.59) among current drinkers, and 0.76 (0.52 to 1.11) among nondrinkers (P for interaction=0.84). There was no association between nondairy calcium intake and mortality from cardiovascular disease (data not shown in the table).
In this large prospective study of middle-aged Japanese men and women, we observed lower risks of mortality from total stroke, either hemorrhagic or ischemic, with increased intake of dairy calcium.
Two cohort studies have shown an inverse relationship between total calcium intake, specifically dairy calcium intake, and the risk of ischemic stroke.1,2 The Honolulu Heart Program of 3150 men 55 to 68 years of age reported that the multivariate relative risk of thromboembolic stroke in a 22-year follow-up was 1.8 (1.1 to 2.9) for the lowest versus highest quintiles of total calcium intake and 1.5 (1.0 to 2.2) for those of dairy calcium intake.1 The Nurses’ Health Study of 85 764 women 34 to 59 years of age showed that the multivariate relative risk of ischemic stroke was 0.69 (0.50 to 0.95) for the highest versus lowest quintiles of total calcium intake.2 A cohort study of 265 070 Japanese >40 years of age also showed lower mortality from cerebral hemorrhage and ischemic stroke for dairy milk intake ≥4× per week compared with less than once per week; the multivariate relative risk was 0.74 (0.68 to 0.80) and 0.85 (0.77 to 0.92), respectively.11 However, this study did not examine the reproducibility or validity of dietary assessments, nor did it calculate calcium intake.
The mechanisms responsible for the inverse association between dairy calcium and risk of stroke merit some discussion. A well-designed clinical trial demonstrated that diets rich in low-fat milk and dairy products reduced blood pressure levels.12 Calcium in milk and dairy products can be absorbed more efficiently than that in nondairy products because phosphorylated serine and threonine residues, abundant in milk and dairy products, make calcium ion soluble during digestion.13 In addition, calcium has a hypotensive effect, particularly in individuals with a high sodium intake. Calcium also has antiplatelet aggregation effects.14,15 In addition to these beneficial effects of calcium, whey peptides in milk and dairy products may have a hypotensive effect through inhibition of the angiotensin-converting enzyme.16
In the present study, calcium intake was not associated with mortality from coronary heart disease in either sex. Our finding is consistent with the results of previous studies that milk intake was not associated with mortality or incidence of coronary heart disease among middle-aged men.3,4 However, a recent prospective study of postmenopausal women showed a 33% lower mortality of coronary heart disease associated with the highest quintile of calcium intake.17
Some limitations of the present study warrant discussion. First, there would be the residual confounding on the association between calcium intake and cardiovascular disease. For example, histories of hypertension and diabetes were self-reported. Thus, it is possible that we did not adjust for these confounding variables fully. Also, current smokers or drinkers had a lower calcium intake than nonsmokers or nondrinkers. However, we found similar inverse associations between calcium intake and mortality for smokers and nonsmokers and for drinkers and nondrinkers.
Second, we used only 3 foods, namely milk, yogurt, and cheese, not other dairy foods and calcium supplements. However, validation and reliability assessments showed that the validity and reliability of dairy calcium intake was very good (r=0.55; P<0.0001). During the baseline survey, we did not ask for information on calcium supplements in the questionnaire; however, the use of calcium supplements was not common in Japan in the 1990s.
In conclusion, the present study showed that calcium intake from milk and dairy products was associated with a reduced risk of mortality from total stroke, either hemorrhagic or ischemic, among Japanese men and women. Clinical trials are necessary to confirm this finding.
The JACC Study was supported by grants-in-aid for scientific research from the Ministry of Education, Science, Sports and Culture of Japan (Monbusho; 61010076, 62010074, 63010074, 1010068, 2151065, 3151064, 4151063, 5151069, 6279102 and 11181101). The authors sincerely express their appreciation to Dr Kunio Aoki, professor emeritus, Nagoya University School of Medicine, and the former chairman of the JACC Study, and to Dr Haruo Sugano, the former director, Cancer Institute, Tokyo, who greatly contributed to the initiation of the JACC Study.
- Received August 4, 2005.
- Revision received October 19, 2005.
- Accepted October 31, 2005.
Abbott RD, Curb JD, Rodriguez BL, Sharp DS, Burchfiel CM, Yano K. Effect of dietary calcium and milk consumption on risk of thromboembolic stroke in older middle-aged men. Stroke. 1996; 27: 813–818.
Iso H, Stampfer MJ, Manson JE, Rexrode K, Hennekens CH, Colditz GA, Speizer FE, Willett WC. Prospective study of calcium, potassium, and magnesium intake and risk of stroke in women. Stroke. 1999; 30: 1772–1779.
Ness AR, Smith GD, Hart C. Milk, coronary heart disease and mortality. J Epidemiol Community Health. 2001; 55: 379–382.
Van der Vijver LP, van der Waal MA, Weterings KG, Dekker JM, Schouten EG, Kok FJ. Calcium intake and 28-year cardiovascular and coronary heart disease mortality in Dutch civil servants. Int J Epidemiol. 1992; 21: 36–39.
Ministry of Health and Welfare. Kokumin Eiyo no Genjyo (The National Nutrition Survey 2002) Tokyo: Daiichi Shuppan, 2004. [In Japanese]
Iso H, Date C, Yamamoto A, Toyoshima H, Tanabe N, Kikuchi S, Kondo T, Watanabe Y, Wada Y, Ishibashi T, Suzuki H, Koizumi A, Inaba Y, Tamakoshi A, Ohno Y. Perceived mental stress and mortality from cardiovascular disease among Japanese men and women. The Japan Collaborative Cohort Study for Evaluation of Cancer Risk Sponsored by Monbusho (J Am Coll Cardiol Study). Circulation. 2002; 106: 1229–1236.
Date C, Fukui M, Yamamoto A, Wakai K, Ozeki A, Motohashi Y, Adachi C, Okamoto N, Kurosawa M, Tokudome Y, Kurisu Y, Watanabe Y, Ozasa K, Nakagawa S, Tokui N, Yoshimura T, Tamakoshi A; J Am Coll Cardiol Study Group. Reproducibility and validity of a self-administered food frequency questionnaire used in J Am Coll Cardiol Study. J Epidemiol. 2005; (suppl 1): S9–23.
Renaud S, Morazain R, Godsey F, Dumont E, Symington IS, Gillanders EM, O’Brien JR. Platelet functions in relation to diet and serum lipids in British farmers. Br Heart J. 1981; 46: 562–570.
Bostick RM, Kushi LH, Wu Y, Meyer KA, Sellers TA, Folsom AR. Relation of calcium, vitamin D, and dairy food intake to ischemic heart disease mortality among postmenopausal women. Am J Epidemiol. 1999; 149: 151–161.