American Journal of Epidemiology Copyright« 1992 by The Johns Hopkins University School of Hygiene and Put*; Health All rights reserved

Vol. 136, Mo. 10 Printed in U.S.A.

Calcium Intake and Blood Pressure in Elderly Women

Joel A. Simon, u Warren S. Browner,1'2 Jo Law Tao,2 and Stephen B. Hulley2

The authors studied the relative importance of calcium intake as a correlate of Wood pressure in a cross-sectional analysis of 6,517 nonWack women aged >65 years from Portland, Oregon; Minneapolis, Minnesota; Baltimore, Maryland; and the Monongahela Valley near Pittsburgh, Pennsylvania, who enrolled in the Study of Osteoporotic Fractures in 1986. Dietary calcium intake was measured by using a modified food frequency questionnaire. There were weak associations between total calcium intake and Wood pressure, such that each 1,000 mg/day increase in calcium intake was associated with a 1.5 mmHg decrease in systdic blood pressure (95% confidence interval -2.2 to -0.8, p < 0.0001) and a 0.5 mmHg decrease in diastolic pressure (95% confidence interval 0.8 to -0.20, p < 0.004). After adjustment for confounding variables, total calcium intake continued to have a weak inverse association only with diastolic Wood pressure. The multivariate models revealed several stronger independent risk factors for blood pressure, including age, body mass index, alcohol consumption, and level of education (all p < 0.01). The small size of the inverse association between calcium intake and blood pressure limits its Wotogical importance in this population of elderly women. Am J Epidemiol 1992;136:1241 - 7 . blood pressure; calcium; women

Dietary calcium, while less well established than dietary sodium as a determinant of blood pressure, has nevertheless been a topic of considerable research (1). Many observational studies have suggested that increased dietary calcium may be associated with a modest lowering of blood pressure (2-16), although others have not found this association (17-20). Even in the studies that have found an association between calcium

Received for publication January 17, 1992, and in final form May 11, 1992. 1 General Internal Medicine Section, Department of Medicine, Veterans Affairs Medical Center, San Francisco, CA. 2 Division of Clinical Epidemiology, Department of Epidemiology and Biostarjstics, University of California, San Francisco, CA. Reprint requests to Dr. Joel A. Simon, General Internal Medians Section (111A1), Veterans Affairs Medical Center, 4150 Clement Street, San Francisco, CA 94121. The Study of Osteoporotic Fractures is supported by research grants AG05407 and AG5394 from the National Institute on Aging and grant AR35582-4 from the National Institutes of Arthritis and Musculoskeletal Diseases

intake and blood pressure, the relation has not been observed uniformly for systolic and diastolic blood pressures or across race and sex categories. In a recent review (21), it was suggested that certain subgroups might reduce their risk of hypertension by ingesting substantially higher amounts of calcium than the 800 mg/day currently recommended for providing adequate bone mineralization in adults (22). These groups included elderly women, blacks, alcoholics, diabetics, and pregnant women. A safe and effective nonpharmacologic intervention that reduced the need for antihypertensive medication would certainly be welcome, even if it were restricted to selected populations. There is a particular need for better information on the relation between calcium intake and blood pressure in elderly women since this population is at high risk for strokes and other complications of hypertension. The Study of Osteoporotic Fractures is a prospective cohort study of the predictors of

1241

1242

Simon etal.

bone mineral density and fractures in a cohort of elderly women. In this report, we examine the relation between calcium intake and blood pressure in this population. MATERIALS AND METHODS

The participants in the Study of Osteoporotic Fractures have been described previously (23). Briefly, 9,704 nonblack women age 65 years or older were enrolled from population-based listings in Portland, Oregon; Minneapolis, Minnesota; Baltimore, Maryland; and the Monongahela Valley near Pittsburgh, Pennsylvania. In Oregon and Minnesota, women were recruited from health maintenance organizations. Additional participants in Minnesota were obtained by using jury selection, Hypertension Detection and Follow-up Study, and Systolic Hypertension in the Elderly study lists. In Maryland, women were selected from lists of licensed drivers and holders of identification cards. The data we collected at baseline included calcium intake, blood pressure, activity level, level of education, body mass index, and use of tobacco, caffeine, and alcohol. We ascertained dietary calcium intake by an interviewer-administered food frequency questionnaire that focused on calcium-rich foods. A validation study comparing a 34item food frequency questionnaire and an 18-item food frequency questionnaire with 7-day food records resulted in the selection of the 34-item instrument (r = 0.76) as the most accurate for estimating calcium intake in this age group (24). A complete description of the foods included in the 34-item food frequency questionnaire can be found elsewhere (24). The instrument was not designed to estimate the consumption of sodium, potassium, or magnesium. We specifically asked about the use of calcium supplements. For women who reported the use of calcium supplements at least once a week, we multiplied the milligrams per pill by the number of pills taken daily to estimate calcium ingested in this manner. Dietary and supplemental calcium intakes, including calcium from antacid sources such as Turns

(SmithKline Beecham Corp., Pittsburgh, Pennsylvania), were added to yield total calcium intake. We determined caffeine consumption by multiplying the reported number of cups of coffee consumed by 95 mg, cups of tea by 55 mg, and cola drinks by 45 mg and summing these three parameters. We asked about alcohol (beer, wine, and liquor) use in ounces per week and converted to grams per week (I ounce = 23.34 g), and we asked about number of cigarettes smoked per day. Recreational and sports activities, as well as routine climbing of stairs and blocks walked outside the home, were added to yield kilocalories burned per week. We used the highest number of years of education achieved by either the study participant or her spouse as an indicator of socioeconomic status. Subjects were weighed by using a standard balance beam scale after they had removed shoes, outdoor clothing, and heavy sweaters. We measured height in centimeters with a wall-mounted portable stadiometer, and we calculated body mass index (weight (kg)/ height (m)2) for each subject. We measured blood pressure in the supine position using pediatric, regular adult, and large adult cuffs, as appropriate, and a Standard Baum sphygmomanometer (Baum Co., New York, New York). A single determination of the first and the fifth phase Korotkoff sounds was recorded. Hypertensive subjects were defined by either systolic pressure of 140 mmHg or greater or diastolic pressure of 90 mmHg or greater. We excluded 3,095 women who reported diuretic use and 65 women who reported total calcium intakes greater than 3,500 mg/ day, since these intakes were thought to be of questionable validity. Sixty-three of the 65 women with intakes of greater than 3,500 mg/day reported consumption of calcium supplements. The systolic and diastolic pressures (mean ± standard deviation) for the women with total calcium intakes of greater than 3,500 mg/day were 142 ± 19 and 76 ± 11, respectively. We also excluded 27 subjects whose blood pressure levels were missing, leaving 6,517 available for analysis.

Calcium and Blood Pressure

1243

TABLE 1. Baseline characteristics of 6,517 study subjects from Portland, Oregon; Minneapolis, Minnesota; Baltimore, Maryland; and the Monongahela Valley near Pittsburgh, Pennsylvania, who were enrolled in the Study of OsteoporotJc Fractures, 1986* Measurement

Mean±SDt

Range

Median

Age (years) Education level (years):): Alcohol Intake (g/week) Caffeine Intake (mg/day) Cigarettes/day Body mass index (kg/m2) Systolic btood pressure (mmHg) Diastolic blood pressure (mmHg) Energy expenditure above basal (kcal/weekft Total calcium intake (mg/ day)§ Calcium supplement intake (mg/day)

71 ± 5 14 ± 3 41 ± 59.1 228±117 16 ± 10 26±5

65-99 3-19 1.5-458 23-600 1-60 13-60

70 13 17.5 238 15 25

140 ±19

86-240

138

76 ± 9

8-118

76

0-28,433

1,688 ±1,170

1,198

1,132 ±657

45-3,500

1,020

771 ±484

10-3,000

615

• The means, ranges, and medians reported for alcohol, caffeine, tobacco, and calcium supplement use Include only those women reporting current consumption (n - 3,729, 4,819, 723, and 3,054, respectively), f SD, standard deviation. t For education level and energy expencSture, n - 6,516 and 6,477, respectively. § Includes dietary and supplemental calcium intakes.

2). Tests for linear trend revealed that total We used general linear regression models calcium intake was significantly related to to examine the associations between predicmean systolic and diastolic pressures (p < tor variables and blood pressure. Predictor 0.0001 and p < 0.01, respectively). variables included daily calcium intake, age, body mass index (kg/m2), years of education In univariate analyses, total calcium in(using the higher of the subject's or her take as a continuous variable was inversely spouse's educational level), total energy exrelated to both systolic and diastolic prespended above basal rates, weekly alcohol sures: Each 1,000 g/day increase in calcium intake, and daily tobacco use and caffeine intake was associated with a 1.5 mmHg deintake. We also examined the independent crease in systolic pressure {p < 0.000 l)(table relation between calcium supplement intake 3) and a 0.5 mmHg decrease in diastolic and blood pressure using linear regression pressure (p < 0.004) (table 4). However, in models. We compared mean total calcium intake between normotensive and hypertensive subjects within 5-year age groups using TABLE 2. Mean blood pressure per 500 mg/day t tests. We considered two-sided p values of increment In total calcium intake from the Study of less than 0.05 to be statistically significant. OsteoporotJc Fractures, 1986 RESULTS

Most study participants were nonobese, well-educated women who did not smoke (table 1). Mean total calcium intake was 1,132 mg/day, and ranged widely from 45 to the upper cutoff of 3,500 mg/day. Mean blood pressures were 141/77 mmHg and 134/74 mmHg in women who consumed less than 500 and greater than 3,000 mg/day of calcium, respectively (table

Blood pressure (mmHg)

Total calcium intake (mg/day)

No of subjects

£500 501-1,000 1,001-1,500 1,501-2,000 2,001-2,500 2,501-3,000 3,001-3,500

1,151 2,035 1,602 1,013 469 167 80

Systolic* (mean± SD)f

Diastole* (mean ± SO)

141 ±19 140±19 140±19 139± 18 139 ±19 139 ±17 134±20

77±9 77±9 76 ±9 76 ±9 76 ±9 77±9 74 ±8

• Tests for Bnear trend yielded p < 0.0001 and p < 0.01 for systolic and diastole blood pressures, respectively, f SD, standard deviation.

1244

Simon eta).

TABLE 3. Relation of total calcium Intake and other covariates to systolic blood pressure from the Study of OsteoporotJc Fractures, 1986 Variable Unlvariate model Total calcium intake (1,000 mg/ day) Muttjvariate model Total calcium intake (1,000 mg/ day) Age (years) Body mass index (kg/m2) Education (years) Energy expended (1,000 kcal/ week) Ateohol intake (100 g/week) Caffeine intake (100 mg/day) Cigarettes/day

Stopef

95% confidence tntervarf

p value

-0.05

-1.5

-2.2 to -0.80

0.0001

-0.05 0.21 0.09 -0.08

-0.58 0.84 0.39 -0.51

-1.27 0.75 0.29 -0.67

to to to to

0.12 0.93 0.49 -0.35

0.10 0.0001 0.0001 0.0001

0.02 0.05 0.04 0.03

0.29 1.71 0.51 0.09

0.02 0.77 0.19 0.02

to to to to

0.56 2.64 0.83 0.17

0.04 0.0004 0.002 0.02

Correlation coefficient*

• Partial correlation coefficients in the case of the muttivariate model. R' - 0.07 for muttivariate model. t Indicates change (and 95% confidence Intervals) In systolic Wood pressure (in mmHg) per Indteated unit of each variable.

TABLE 4. Relation of total calcium intake and other covariates to diastolic Wood pressure from the Study of Osteoporotic Fractures, 1986 Variable Unlvariate model Total calcium intake (1,000 mg/ day) MultJvariate model Total calcium Intake (1,000 mg/ day) Age (years) Body mass index (kg/m2) Education (years) Energy expended (1,000 kcal/ week) Alcohol Intake (100 g/week) Caffeine intake (100 mg/day) Cigarettes/day

Correlation coefficient*

Stopet

95% confidence Intervalt

p value

-0.04

-0.49

-0.83 to -0.16

0.004

-0.04 -0.08 0.14 -0.03

-0.35 -0.11 0.28 -0.12

-0.69 -0.15 0.23 -0.19

to -0.01 to-0.06 to 0.33 to-0.04

0.04 0.0001 0.0001 0.003

0.01 0.05 0.02 0.02

0.09 0.86 0.12 0.03

-0.05 0.40 -0.04 -0.01

to to to to

0.21 0.0003 0.13 0.18

0.22 1.33 0.28 0.06

• Partial corretaticin coefficients In the case of the muttivariate model, fl* •= 0.03 for muttivariate model. f Indicates change (and 95% confidence Intervals) In diastole blood pressure (in mmHg) per Indicated unit of each variable.

the multivariate models, the strongest relation observed was between increasing age and systolic pressure: Each 10-year increase in age was associated with a 8.4 mmHg increase in systolic pressure (p < 0.0001). Body mass index had a consistent relation with both systolic and diastolic pressures: Each unit increment was associated with an increase of 0.4 mmHg and 0.3 mmHg in systolic and diastolic pressures, respectively. Alcohol was also positively correlated with blood pressure: Each 120 g (approximately 12 drinks) consumed per week was associ-

ated with an approximate 2.0 mmHg increase in systolic pressure and a 1.0 mmHg increase in diastolic pressure. As education level increased, systolic and diastolic pressures decreased modestly. Because other investigators have found that alcohol modifies the effect of dietary calcium on blood pressure (25, 26), we also analyzed our data for this interaction by including alcohol x total calcium terms in the regression models. We found no evidence for such an interaction. Adjustment for potential confounders such as age and body mass index reduced

Calcium and Blood Pressure

the magnitude of the associations between total calcium intake and blood pressure: Each 1,000 mg/day increase in calcium intake was associated with decrease of 0.6 mmHg (p = 0.10) in systolic pressure and a decrease of 0.4 mmHg (p = 0.04) in diastolic pressure. We examined separately women who used calcium supplements and those who did not. A total of 3,054 women reported calcium supplement use, ingesting 10-3,000 mg/day in this manner. Univariate analyses failed to demonstrate an association between amount of supplemental calcium consumed and blood pressure. This held true even after adjustment for dietary calcium intake. Among women who did not use calcium supplements, an association was found only in the univariate analysis of dietary calcium and systolic pressure: Each 1,000 mg/day of dietary calcium intake was associated with a decrease of 2.3 mmHg (p< 0.01). However, after adjustment for age, body mass index, and other variables, this relation was no longer statistically significant (p = 0.08). There were no statistically significant differences in mean total calcium intake between normotensive and hypertensive subjects in any of the 5-year age categories examined. DISCUSSION

Our results indicate that total calcium intake may be weakly related to systolic and diastolic pressures. The size of the association was small, a 1.5 mmHg decrease in systolic pressure for each 1,000 mg/day increase in calcium intake and a 0.5 mmHg decrease in diastolic pressure for each 1,000 mg/day increase in calcium intake. This effect was even smaller after adjustment for other variables, especially age. Two analyses of elderly women in the first and second National Health and Nutrition Examination Surveys found dietary calcium to be inversely related to blood pressure (14, 27). These two studies, like ours, also found other variables, such as age and body mass index, to be more strongly predictive of blood pressure than was dietary calcium.

1245

Most observational studies that have included elderly women, however, have been unable to demonstrate an association between dietary calcium and blood pressure (8-10, 18, 19). The failure of these studies to observe an association could be due to the smaller size of some of these studies, to problems with the accurate measurement of dietary calcium, or to the very small magnitude of the association. Clinical trials (28-40) have suggested that blood pressure reduction resulting from increased calcium intake, if observed at all, is generally of small magnitude and may be restricted to certain population subgroups. Some clinical trials have found that only hypertensive individuals benefit from calcium supplementation (31, 34, 35, 40). Others have suggested that only particular subgroups of hypertensives, such as those with high parathyroid hormone levels or low serum total calcium (30, 36) or those with above average renal calcium excretion, might benefit (36, 37). We were unable to find any difference in calcium intake between normotensive and hypertensive elderly women, but we were unable to further categorize hypertensive women on the basis of calcium excretion or parathyroid hormone levels. Experimental studies of calcium supplementation in middle-aged and elderly women have yielded inconsistent results. One very small, uncontrolled study that reported a blood pressure reduction with calcium supplementation included only six elderly women and lasted only 2 weeks (38). Other studies that included mostly middleaged women have demonstrated inconsistent blood pressure responses to calcium supplementation (31, 35, 37). The failure to observe a consistent effect from calcium supplementation may be the result of the impaired intestinal absorption of calcium that may develop in postmenopausal women (1). While our study benefited from a very large sample size and from a well-validated dietary questionnaire that focused on calcium consumption, there were some limitations. All questionnaire measurements of diet have imperfect precision and accuracy,

1246

Simon etal.

and our dietary instrument did not measure other dietary components that might be related to blood pressure, such as sodium, potassium, and magnesium. We did not ascertain the use of antihypertensive medications other than diuretics; thus, it is possible that we misclassified the blood pressure in women taking such medications. However, if there were substantial misclassification of treated hypertensive subjects, this would also be expected to affect the associations between other predictors and blood pressure. Yet age, body mass index, alcohol intake, and education level all remained strong predictors of blood pressure. Moreover, the magnitude of the effect we observed, for example, an 8 mmHg increase in systolic pressure per 10 years of age, agrees with that observed in other studies of elderly women (41, 42). Although we measured blood pressure in the supine and standing positions, we analyzed only the supine measurement because of the high prevalence of orthostasis among the elderly. While it is possible that blood pressure taken in the standing or seated positions might have yielded different results, others have documented an inverse association between supine blood pressure and calcium intake (5, 15, 35). Further, we were limited by the availability of only a single blood pressure determination for each study subject. The elderly, who differ from other populations because of their high prevalence of systolic hypertension, are important because of their increased risk of suffering the consequences of hypertension, such as heart failure and stroke (43). An effective nonpharmacologic treatment for hypertension would be particularly useful among the elderly, who often are unable to tolerate the effects of blood pressure medication. The results of our study, however, indicate that total calcium intake is not a major determinant of blood pressure in elderly white women. At least in this population, other blood pressure predictors such as age, body mass index, and alcohol are far more important. Nevertheless, since many elderly women have low calcium intakes and since

increased dietary calcium may be beneficial for other disorders such as osteoporosis (1), the promotion of increased calcium intake among elderly women may still represent a prudent course of action even in the absence of a meaningful effect on blood pressure.

REFERENCES 1. Committee on Diet and Health, Food and Nutrition Board, Commission on Life Sciences, National Research Council. Diet and health: implications for reducing chronic disease risk. Washington, DC: National Academy Press, 1989:347-66. 2. He J, Tell GS, Tang Y-C, et al. Relation of electrolytes to blood pressure in men. The Yi people study. Hypertension 1991 ;17:378-85. 3. Joffres MR, Reed DM, Yano K. Relationship of magnesium intake and other dietary factors to blood pressure: the Honolulu Heart Study. Am J Clin Nutr 1987;45:469-75. 4. Reed D, McGee D, Yano K, et al. Diet, blood pressure, and multicollinearity. Hypertension 1985; 7:405-10. 5. Kromhout D, Bosschieter EB, Coulander CDL. Potassium, calcium, alcohol intake and blood pressure: the Zutphen Study. Am J Clin Nutr 1985; 41:1299-1304. 6. Garcia-Palmieri MR, Costas R, Cruz-Vidal M, et al. Milk consumption, calcium intake and decreased hypertension in Puerto Rico: Puerto Rico Heart Health Program Study. Hypertension 1984; 6:322-8. 7. McCarron DA, Morris CD, Henry HJ, et al. Blood pressure and nutrient intake in the United States. Science 1984,224:1392-8. 8. Ackley S, Barrett-Connor E, Suarez L. Dairy products, calcium, and blood pressure. Am J Clin Nutr 1983,38:457-61. 9. Kesteloot H, Joossens JV. Relationship of dietary sodium, potassium, calcium, and magnesium with blood pressure: Belgian interuniversity research on nutrition and health. Hypertension 1988; 12: 594-9. 10. Gruchow HW, Sobocinski KA, Barboriak JJ. Alcohol, nutrient intake, and hypertension in US adults. JAMA 1985;253:1567-70. 11. Iso H, Terao A, Kitamura A, et al. Calcium intake and blood pressure in seven Japanese populations. Am J Epidemiol 1991 ;133:776-83. 12. Yamamoto ME, Kuller LH. Does dietary calcium influence blood pressure? Evidence from the Three Area Stroke Mortality Study. (Abstract 464). Circulation 1985;72(Suppl. III):III-116. 13. Harlan WR, Hull AL, Schmouder RL, et al. Blood pressure and nutrition in adults. The National Health and Nutrition Examination Survey. Am J Epidemiol 1984; 120:17-28. 14. Harlan WR, Hull AL, Schmouder RL, et al. High blood pressure in older Americans: the First National Health and Nutrition Examination Survey.

Calcium and Blood Pressure Hypertension l984;6:802-9. 15. Kok FJ, Vandenbroucke C, van der Heide-Wessel C, et al. Dietary sodium, calcium, and potassium, and blood pressure. Am J Epidemiol 1986;123: 1043-8. 16. Liebman M, Chopin LF, Carter E, et al. Factors related to blood pressure in a biracial adolescent female population. Hypertension 1986;8:843-50. 17. Feinleib M, Lenfant C, Miller SA. Hypertension and calcium. Science 1984^226:384-6. 18. Sempos C, Cooper R, Kovar MG, et al. Dietary calcium and blood pressure in National Health and Nutrition Examination Surveys I and II. Hypertension 1986;8:1067-74. 19. Sowers MR, Wallace RB, Lemke JH. The association on intakes of vitamin D and calcium with blood pressure among women. Am J Clin Nutr 1985,42:135-42. 20. Hancock RL, Tylavsky FA, Moore R, et al. Hereditary and environmental influences on blood pressure values of premenopausal women and their college-age daughters. J Am Coll Nutr 1991; 10: 376-82. 21. McCarron DA, Morris CD, Young E, et al. Dietary calcium and blood pressure: modifying factors in specific populations. Am J Clin Nutr 1991 ;54: 215S-19S. 22. Subcommittee on the Tenth Edition of the RDAs. Food and Nutrition Board. Commission on Life Sciences. National Research Council. Recommended dietary allowances. Washington, DC: National Academy Press, 1989. 23. Cummings SR, Black DM, Nevitt MC, et al. Appendicular bone density and age predict hip fracture in women. JAMA 1990,263:665-8. 24. Cummings SR, Block G, McHenry K, et al. Evaluation of two food frequency methods of measuring dietary calcium intake. Am J Epidemiol 1987; 126:796-802. 25. Criqui MH, Langer RD, Reed DM. Dietary alcohol, calcium, and potassium: independent and combined effects on blood pressure. Circulation 1989,80:609-14. 26. Hamet P, Mongeau E, Lambert J, et al. Interactions among calcium, sodium, and alcohol intake as determinants of blood pressure. Hypertension 1991;17(Suppl. 1):I-150-4. 27. Harlan WR, Landis R, Schmouder RL, et al. Blood lead and blood pressure: relationship in the adolescent and adult US population. JAMA 1985;253: 530-4. 28. Belizan JM, Villar J, Pineda O, et al. Preliminary evidence of the effect of calcium supplementation on blood pressure in normal pregnant women. Am J Obstet Gynecol 1983; 146:175-80.

1247

29. Belizan JM, Villar J, Pineda O, et al. Reduction of blood pressure with calcium supplementation in young adults. JAMA 1983;249:1161-5. 30. Groebbe DE, Hofman A. Effect of calcium supplementation on diastolic blood pressure in young people with mild hypertension. Lancet 1986;2: 703-6. 31. Johnson NE, Smith EL, Freudenheim JL. Effects on blood pressure of calcium supplementation of women. Am J Clin Nutr 1985;42:12-17. 32. Lopez-Jaramillo P, Naravaez M, Felix C, et al. Dietary calcium supplementation and prevention of pregnancy hypertension. (Letter). Lancet 1990; 1:293. 33. Lyle RM, Melby CL, Hyner GC, et al. Blood pressure and metabolic effects of calcium supplementation in normotensive white and black men. JAMA 1987;257:1772-6. 34. McCarron DA, Henry HJ, Morris CD. Randomized placebo-controlled trial of oral Ca2+ in human hypertension. (Abstract). Clin Res 1984;32:37A. 35. McCarron DA, Morris CD. Blood pressure response to oral calcium in persons with mild to moderate hypertension: a randomized, doubleblind, placebo-controlled, cross-over trial. Ann Intern Med 1985;1O3:825-31. 36. Resnick LM, Nicholson JP, Laragh JH. Outpatient therapy of essential hypertension with dietary calcium supplementation. (Abstract). J Am Coll Cardiol 1984;3:616. 37. Strazzullo P, Siani A, Guglielmi S, et al. Controlled trial of long-term oral calcium supplementation in essential hypertension. Hypertension 1986;8: 1084-8. 38. Tabuchi Y, Ogihara T, Hashizume K, et al. Hypotensive effect of long-term oral calcium supplementation in elderly patients with essential hypertension. J Clin Hypertens 1986;2:254-62. 39. Zoccali C, Mallamaci F, Delfino D, et al. Doubleblind randomized, crossover trial of calcium supplementation in essential hypertension. J Hypertens 1988;6(suppl. 6):S451-5. 40. Van Beresteyn ECH, Schaafsma G, de Waard H. Oral calcium and blood pressure: a controlled intervention trial. Am J Clin Nutr 1986,44:883-8. 41. Joossens JV, Kesteloot H. Trends in systolic blood pressure, 24-hr sodium excretion, and stroke mortality in the elderly in Belgium. Am J Med 1991; 90(Suppl3A):5S-llS. 42. Foster TA, Hale WE, Srinivasan SR, et al. Levels of selected cardiovascular risk factors in a sample of geriatric participants-The Dunedin Program. J Gerontol 1987,42:241-5. 43. Kaplan NM, Lieberman E. Clinical hypertension. 5th ed. Baltimore, MD: Williams & Wilkins, 1990.