| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 1996;27:1065-1072.)
© 1996 American Heart Association, Inc.
Articles |
Prospective Study of Nutritional Factors, Blood Pressure, and Hypertension Among US Women
From the Channing Laboratory (C.H., W.W., F.S., B.R., M.S.) and Division of Preventive Medicine (C.H., J.M.), Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Mass; Department of Nutrition (A.A., W.W., F.S.) and Department of Epidemiology (A.A., W.W., M.S.), Harvard School of Public Health, Boston, Mass; and Department of Epidemiology and Biostatistics, Erasmus University School of Medicine, Rotterdam, Netherlands (J.W.).
Correspondence to Alberto Ascherio, MD, Department of Nutrition, Harvard School of Public Health, 665 Huntington Ave, Boston, MA 02115.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAppendix 1References |
|---|
Abstract We examined prospectively the relation of nutritional factors with hypertension and blood pressure levels among 41 541 predominantly white US female nurses, aged 38 to 63 years, who completed a detailed semiquantitative food frequency questionnaire in 1984 and were without diagnosed hypertension, cancer, or cardiovascular disease. During 4 years of follow-up, from 1984 to 1988, 2526 women reported a diagnosis of hypertension. Age, relative weight, and alcohol consumption were the strongest predictors for the development of hypertension. Dietary calcium, magnesium, potassium, and fiber were not significantly associated with risk of hypertension, after adjusting for age, body mass index, alcohol, and energy intake. Among women who did not report hypertension during the follow-up period, calcium, magnesium, potassium, and fiber were each significantly inversely associated with self-reported systolic and diastolic pressures, after adjusting for age, body mass index, alcohol consumption, and energy intake. When the four nutrients were added simultaneously to the regression model, only fiber and magnesium intakes retained significant inverse associations with systolic and diastolic pressures. In analyses of food groups, intakes of fruit and vegetables were inversely associated with systolic and diastolic pressures, and intakes of cereals and meat were directly associated with systolic pressure. These results support hypotheses that age, body weight, and alcohol consumption are strong determinants of risk of hypertension in middle-aged women. They are compatible with the possibilities that magnesium and fiber as well as a diet richer in fruits and vegetables may reduce blood pressure levels.
Key Words: calcium • epidemiology • magnesium • nutrition • potassium • fiber
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
A role for diet in determining changes in blood pressure (BP) level is suggested by international comparisons1 and studies of migrants2 3 4 5 and vegetarians.6 7 8 9 Alcohol intake10 11 and obesity12 13 14 increase BP, but these factors do not entirely explain observed differences among populations,3 9 suggesting etiologic roles of other aspects of diet. In a recent international study, large variations in salt intake had only a minor relation with BP level,15 but reanalyses of observational16 17 and experimental18 data have suggested a stronger association of salt intake with BP. The role of other specific nutrients is also unclear.
In 1980, information on dietary intake over the preceding year was collected from a large cohort of US women. Dietary intake was reassessed in 1984 with a more refined instrument. The association of the diet reported in 1980 with the development of hypertension during the first 4 years of follow-up was described in a previous report.19 In the present study, we analyzed the association between diet reported in 1984 and development of hypertension between 1984 and 1988 and self-reported BP in 1986 and 1988. Our primary hypothesis was that calcium, magnesium, potassium, and dietary fiber intakes are inversely associated with risk of hypertension and BP levels.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
Nurses' Health Study
The Nurses' Health Study is a prospective cohort study of major diseases among a large group of US female nurses. In 1976, 121 700 women (98% white) aged 30 to 55 years completed mailed questionnaires regarding risk factors for cardiovascular diseases and cancer as well as other major health conditions. Follow-up questionnaires are sent biennially to update the information.20
Dietary Assessment
Semiquantitative food frequency questionnaires were
included in the 1980 (61 food items) and 1984 (126 food items)
follow-up periods. Extensive data on the reproducibility and
validity of the food frequency questionnaires have been published
elsewhere.21 22 23 In a validation study conducted in 1986
among 200 cohort participants, the correlations between intakes
estimated by the food frequency questionnaire and two 1-week diet
records were .62 for calcium, .76 for magnesium, .61 for potassium,
and .60 for fiber after deattenuation for within-person variation
in the diet records. For each food, a standard portion size was
specified and participants were asked to estimate average intake over
the previous year. Intake scores for different nutrients were computed
by multiplying the reported frequency of each food by the nutrient
content of the specified portion. The food composition database used
for calculation of nutrient values is based primarily on US Department
of Agriculture publications and is continually supplemented by other
published sources and personal communications from laboratories and
manufacturers. Calculations included specific information for brand and
types of vitamin and mineral supplements and breakfast cereals.
Calculation of sodium intake also included the average number of shakes
of salt added to food.
Diagnosis of Hypertension
BP status was defined by self-reported responses to the
questionnaires. In 1976, participants were asked whether they had ever
had a diagnosis of high BP (excluding during pregnancy). On subsequent
biennial questionnaires, we inquired whether subjects had been newly
diagnosed as having high BP and, if so, the date of diagnosis. The
validity of self-reported diagnosis of hypertension was documented
by review of the medical records of a sample of nurses who reported
a diagnosis of high BP on the 1982 questionnaire.24
Briefly, recorded BP was greater than 160/95 mm Hg in 39 (77%) of
the 51 records reviewed, and greater than 140/90 mm Hg in all of
them. In 1978, participants were asked to report the number of
physician examinations during the previous year, and in 1988, whether
they had had a BP measurement during the previous 2 years. In 1986 and
1988, participants were also asked to report their usual
systolic and diastolic BPs, choosing among the
following categories: In 1986, less than 75, 75 to 84, 85 to 89, 90 to
94, 95 to 104, 105+ mm Hg for diastolic; less than 120, 120
to 139, 140 to 149, 150 to 159, 160 to 169, 170+ mm Hg for
systolic; in 1988, less than 65, 65 to 74, 75 to 84, 85 to 89,
90 to 94, 95 to 104, 105+ mm Hg for diastolic; less than
115, 115 to 124, 125 to 134, 135 to 144, 145 to 154, 155 to 164, 165 to
174, 175+ mm Hg for systolic.
Population for Analysis
A total of 98 462 women returned the 1980 dietary
questionnaire. We excluded those who reported one or more of the
following diagnoses on the 1980 or previous questionnaires: high BP,
myocardial infarction, coronary artery surgery, stroke, angina
pectoris, diabetes mellitus, and all cancers except nonmelanoma skin
cancer. Women were also excluded if they indicated on the 1980 or
previous questionnaires that they currently used antihypertensive
medication, were on a special diet (1980 only), or had been pregnant
for at least 6 months within the previous 2 years. After exclusion of
these 38 792 women, the baseline population in 1980 consisted of
59 670 women. Of these, 48 311 women returned the 1984 dietary
questionnaire and 41 930 met the inclusion criteria at the date of
return of the 1984 questionnaire. Follow-up questionnaires were
sent in 1986 and 1988 to all study participants. The 389 (0.9%)
nonrespondents to both follow-up questionnaires were excluded,
leaving 41 541 women for analysis. During the period from the
return of the 1984 questionnaire through June 1, 1988, these women
contributed 149 294 years of follow-up. Analyses with BP
level as the dependent variable were restricted to women who
reported their systolic (n=28 971) or diastolic
(n=28 757) BP in both the 1986 and 1988 questionnaires.
Statistical Analysis
Two outcomes were considered in the study: self-report of
diagnosed hypertension during the 4 years of follow-up, and the
average of self-reported BP in 1986 and 1988 among women who did
not report a diagnosis of hypertension.
The analysis of hypertension was based on incidence rates, with person-months of follow-up used as the denominator. For each participant, person-months were allocated according to the 1984 exposure variables, until death or hypertension was reached, or until June 1, 1988. Women who reported a diagnosis of cardiovascular disease, diabetes, or cancer in 1986 were excluded from the 1986-88 follow-up. We calculated the relative risk associated with nutrient intakes, defined as the incidence rate of hypertension among women in each category of nutrient intake divided by the corresponding rate among women in the lowest category of intake. Age- and body mass index–specific rates were calculated and used for computation of age- and body mass index–adjusted relative risks with 95% confidence intervals. To adjust for several risk factors simultaneously, we used proportional hazards models.25
The association between nutrient intakes and self-reported BP among women who did not report a diagnosis of hypertension was analyzed with multiple linear regression models. To control for confounding, models included age, body mass index, and alcohol consumption; a quadratic term for age was included to account for the nonlinearity of the age-BP association. Nutrients were added to the models as continuous variables; however, analyses with indicator variables corresponding to categories of intake were also performed.
Body mass index was calculated as weight (kilograms) divided by the square of height (meters). Nutrients were adjusted for total calories as described elsewhere21 by regressing the nutrient intake on total energy intake. For each calorie-adjusted nutrient, we specified five levels of intake. Cutoff points were chosen a priori to obtain equally spaced categories, each containing a comparable number of subjects. In addition to analyses by nutrient, we also grouped foods into the following categories: sweets, meats, fish, cereals, fruits, vegetables, high-fat dairy, and low-fat dairy (see Appendix). We also assessed nutrient intakes from these specific food groups. Intakes of groups of food or individual food items were entered in regression models as continuous variables or as quintiles.
All probability values are two-tailed. Results were defined as statistically significant for a value of P<.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
During the 4 years of follow-up between 1984 and 1988, 2526 women reported a new diagnosis of hypertension. The incidence rate of hypertension increased from 6.9 per thousand person-years for women aged 35 to 39 years to 23.1 per thousand person-years for women aged 60 years or more (Figure
). Body mass index was
strongly associated with an increasing risk of hypertension, and
alcohol consumption was associated with risk of hypertension at intakes
higher than 20 g/d (Table 1). Therefore, we controlled
for age, body mass index, and alcohol consumption in all further
analyses. No significant association was found between intake
of sodium and risk of hypertension. The relative risk comparing the top
(median sodium intake, 2.8 g/d) with the bottom (median sodium intake,
1.2 g/d) quintile of sodium intake was 0.93 (95% confidence interval,
0.82 to 1.06). Also, no significant associations were observed between
intakes of calcium, magnesium, potassium, and fiber and incidence of
hypertension (Table 2). These results were not
materially changed by excluding from the analysis women who
reported not having had a BP examination in 1988 or women with unknown
BP in either 1986 or 1988. We also examined the association between
nutrients and hypertension within strata of other variables.
Comparing the top with the bottom category of nutrient intakes, no
significant associations for calcium, magnesium, potassium, and fiber
were seen within strata of age (<50, 
50 years), body mass index
(<23, 23 to 28, 29 kg/m2), alcohol consumption (0, 0.1
to 19, 20 g/d), menopause (before versus after menopause), and
hormone use (never, past, current).
|
|
|
In analyses of women who did not report hypertension during the
follow-up period, age, body mass index, and alcohol intake were
strongly associated with BP level (Table 3); the model
containing these variables will be subsequently referred to as the
basic model. When added to the basic model, sodium intake was not
significantly associated with systolic BP (regression
coefficient for 1 g/d=-0.03, SE=0.14) or diastolic BP
(regression coefficient for 1 g/d=-0.07, SE=0.20). Calcium,
magnesium, potassium, and fiber—when added one at a time to the
basic model—were each significantly inversely associated with
systolic and diastolic BPs. When the four nutrients
were added simultaneously, fiber and magnesium retained
significant inverse associations with both systolic and
diastolic BPs, and calcium and potassium had no significant
relation with either systolic or diastolic BP
(Table 3). These associations were not materially changed by adjusting
for waist-to-hip ratio, physical activity, smoking, caffeine
intake, menopausal status, and use of hormones. Also, the inverse
associations between intakes of magnesium and fiber and BP remained
significant after adjusting for intake of vitamin C and other nutrients
(one at a time). Similar results were obtained within strata of body
mass index, alcohol intake, and age. When the four nutrients were
included simultaneously in regression models, only fiber
and magnesium were independently associated with BP in most of the
strata. The only exception was women younger than 50 years, among whom
calcium, fiber, and magnesium each retained an inverse association with
diastolic BP.
|
To compare the magnitude of the effect of dietary variables with
factors known to influence BP, we also fitted regression models in
which age, body mass index, and alcohol intake were entered as
categorical variables. Average systolic BP was 7.6 mm Hg
higher and average diastolic BP was 3.0 mm Hg higher among
women aged 60 years or older compared with women aged 35 to 39 years.
Analogous differences were 9.4/5.4 mm Hg
(systolic/diastolic) comparing obese women (body
mass index 32 kg/m2) with lean women (body mass index
<23 kg/m2) and 2.4/1.2 mm Hg comparing drinkers of more
than 30 g/d of alcohol with nondrinkers. When indicator variables
for each category of nutrient intake were added to the model (one
nutrient at a time), average BP was 1.3/1.0 mm Hg lower in women with
magnesium intake greater than or equal to 0.35 g/d compared with women
with magnesium intake less than 0.2 g/d, and 1.1/1.1 mm Hg lower
comparing women with a fiber intake higher than 25 g/d with women with
a fiber intake less than 10 g/d (Table 4).
|
Considered as separate nutrients and simultaneously
included in the basic models, intakes of fiber from fruits, vegetables,
and cereals were each inversely associated with diastolic
BP; intakes of fiber from fruits and vegetables were also inversely
associated with systolic BP (Table 5). We also
included in regression models intake of magnesium separately from
fruits, cereals, vegetables, dairy products, and other sources.
Magnesium from fruits and other sources, but not from cereals,
vegetables, or dairy products, was inversely associated with BP
(Table 5).
|
When the food groups defined a priori were simultaneously
entered in regression models, intakes of fruits and vegetables were
inversely associated with both systolic and
diastolic BPs, whereas intakes of meat and cereals were
directly associated with systolic but not diastolic
BP (Table 6). Fruit retained a significant inverse
association with systolic as well as diastolic BP,
and vegetables with systolic but not diastolic BP
after controlling simultaneously for intakes of magnesium
and fiber; in the same model, both nutrients were significantly and
inversely associated with BP. Intakes of fish and dairy products
were not significantly associated with BP.
|
To further explore the associations between foods and BP, we added to
the basic models all the items that compose the groups found to be
significantly associated with BP. Separate models were fitted for each
food group, with all items composing the group included
simultaneously in the model. Among fruits, intakes of
apples, oranges, prunes, and grapes were significantly inversely
associated with systolic or diastolic BP; none of
the fruit items was directly associated with BP. Among vegetables,
intakes of carrots, alfalfa, mushrooms, raw spinach, tofu, and celery
were each inversely associated with BP; only intake of eggplants was
directly associated with BP (Table 7). However,
individual items within the cereal and meat food groups had opposite
associations with BP (direct: white bread, potatoes, processed meats,
hamburger, and beef; inverse: rice, dark bread, and chicken) (Table 7).
Adding all individual fruits, vegetables, cereals, meats, and sweets
simultaneously to the basic regression models (including
total energy intake), intakes of grapes, prunes, oranges, raw spinach,
mushrooms, rice, chicken, dark bread, and jam were inversely associated
with either systolic or diastolic BP, and eggplant,
white bread, potatoes, hamburger, cookies, and sweet rolls were
directly associated with either systolic or
diastolic BP. These associations remained significant after
adjusting for dietary fiber and magnesium intake.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
In this large cohort of women, age, body weight, and alcohol intake are significant predictors of hypertension and BP. Furthermore, we observed inverse relations of calcium, dietary fiber, potassium, and magnesium with reported BP but not with the incidence of hypertension. These associations remained after adjusting for age, body mass index, alcohol consumption, waist-to-hip ratio, physical activity, smoking, caffeine intake, menopausal status, and use of hormones. Adjustment for fiber and magnesium intakes eliminated the observed inverse associations of dietary potassium and calcium with both systolic and diastolic BPs.
The validity of BP values and self-report of hypertension in this study is supported by the results of a substudy and by confirmation of established associations of age, alcohol intake, and relative weight with incidence of hypertension and BP levels. Also, after 4 years of follow-up, the relative risk of stroke after adjusting for age and smoking was 4.3 (95% confidence interval, 1.7 to 11.0) among women who in 1986 reported a systolic BP of 140 to 149 mm Hg compared with women who reported a systolic BP of 120 mm Hg or lower, and 2.8 (95% confidence interval, 1.2 to 6.2) among women who reported a diastolic BP of 85 to 89 mm Hg compared with women who reported a diastolic BP of 75 mm Hg or lower (our unpublished data, 1992). While a direct measurement of BP is more objective than self-reported values, the validity of a single measurement is limited because of the intraindividual variability in BP.26 Hypertension is usually asymptomatic, and its detection depends on behavioral factors that may be associated with diet. Although the lower frequency of BP measurements among women with low intakes of magnesium, potassium, and fiber suggested that detection bias may have contributed to the lack of association between the nutritional variables and hypertension, such a contribution must have been small. Nevertheless, we based our main conclusions on analyses with BP as the dependent variable; BP measurements are less likely to be biased and are statistically more powerful. Knowledge of borderline hypertension might have caused women to change their diets, but the direction of such change would have been more likely to attenuate relations with nutrients than to create them. For this reason, and because of random error in the measurement of nutrient intakes, the strength of the observed associations probably underestimates the true effects of diet. However, our results do not exclude a positive association between sodium intake and risk of hypertension, which may have been obscured by error in the measurement of sodium intake or by a reduced sodium intake in response to high BP values at baseline. Also, a positive association of sodium intake with BP may exist within a lower range of sodium intake than that observed in our study population.
We found no independent association of calcium intake with BP in this large cohort of women who reported a wide range of intake. Although our results do not exclude the possibility that BP in some individuals may be reduced by increasing calcium intake, we observed no association within strata defined by age, body mass index, or alcohol intake. However, an inverse association between calcium intake and risk of hypertension was observed during the first 4 years of follow-up in this same cohort19 as well as in several observational studies.27 We have considered several possible causes for this discrepancy. The effect of calcium may be modified by other factors not yet identified, or its effect may be present only at very low levels of intake that are uncommon in the current diet of healthy middle-aged white US women. Also, we cannot exclude the possibility that women with higher BP increased their calcium intake between 1980 and 1984, possibly in response to the publication of the first studies that showed an inverse association between calcium intake and BP. However, this explanation seems unlikely because a diagnosis of hypertension between 1980 and 1984 was not associated with changes in calcium intake during the same period (data not shown). To address the hypothesis that the different results between the two follow-up periods in the Nurses' Health Study were due to the inclusion of calcium from supplements in the 1984 but not in the 1980 questionnaire, we estimated the association between dietary calcium alone in 1984 and risk of hypertension or BP levels. Results were similar to those reported for total calcium intake. Also, it is unlikely that the difference is due to the inclusion in the 1984 questionnaire of sources of calcium that may adversely affect BP. Foods contributing significantly to calcium intake were the same in both questionnaires, and average dietary calcium was similar in 1980 (720 mg/d) and 1984 (706 mg/d). Results of a recent large randomized trial in individuals with high normal diastolic BP have not supported the hypothesis that calcium intake reduces BP levels.28 Previous smaller trials had provided conflicting evidence.29
A low magnesium intake was found to be the dietary factor more strongly associated with high BP in the analyses of data from the Honolulu Heart Study30 and from the first 4 years of follow-up of the Nurses' Health Study.19 Most randomized, double-blind, placebo-controlled supplementation trials conducted have failed to show an effect of magnesium on BP.27 These studies, because of small size or short duration, were unlikely to detect a short-term modest effect or long-term inhibition of magnesium intake on the rise of BP. Also, participants in these studies were hypertensive patients or pregnant women, who may respond differently to magnesium supplementation than normotensive nonpregnant subjects. These limitations, however, do not apply to phase I of the Trials of Hypertension Prevention.28 In that trial, no difference in BP was observed among 227 subjects randomized to 6 months of magnesium supplementation (360 mg/d) and 234 who received placebo. However, at 6 months, the urinary excretion of magnesium in 24 hours was only 39 mg higher in the supplemented than in the placebo group, raising the possibility that absorption of supplemental magnesium, or compliance, may have been low. Also, during the trial, BP in the placebo group decreased significantly (2.67 mm Hg systolic and 2.95 mm Hg diastolic), probably because of regression to the mean. The lower bounds of the 95% confidence interval for the difference between the magnesium and placebo group were -0.98 and -1.47 mm Hg for systolic and diastolic BP, respectively. Thus, the results of this trial are compatible with a small effect of magnesium on BP levels. The inverse association between magnesium and BP in our study suggests that intake of this mineral may play a role in BP regulation.
An inverse association between dietary fiber intake and BP has previously been observed. In the 4 years of follow-up between 1980 and 1984,19 the relative risk for developing hypertension in women was 0.76 for fiber intake greater than 25 g/d compared with intake less than 10 g/d (P=.002). The association was weakened after adjustment for intakes of calcium, magnesium, and potassium but still compatible with a protective effect (95% confidence interval, 0.71 to 1.05). Other cross-sectional studies have reported inverse associations of BP with dietary fiber.30 31 32 33 Also, a BP-reducing effect of an increased fiber intake has been observed in a nonrandomized intervention study.34
A significant treatment-related reduction in BP (-11 mm Hg systolic, P=.02; -3 mm Hg diastolic, P=.04) was obtained in a randomized double-blind study comparing tablets providing 7 g/d of fiber from grain, citrus fruit, and vegetables with a placebo.35 However, other randomized trials did not show an effect of fiber on BP in healthy normotensive volunteers,36 37 38 hypercholesterolemic subjects,39 40 41 42 or overweight individuals who received fiber tablets as a supplement to a weight-reducing diet.43 44 Specific fiber components used as supplements in these trials included soy cotyledon fiber; wheat, rice, or oat fiber; guar gum; and instant oats; baseline fiber intake ranged from 13 to 19.8 g/d among the six trials in which it was reported.
Overall results from dietary trials do not support a BP-reducing effect of short-term fiber supplementation. The combined evidence from observational studies and randomized trials is consistent if fiber intake affects BP only after prolonged administration45 or only in subjects with a lower baseline intake than that of subjects recruited in clinical trials. Alternatively, some other dietary component highly correlated with fiber, or possibly a specific fiber type, may have a BP-lowering action. Although an inverse association between vitamin C intake and BP level has been reported previously,46 vitamin C intake did not explain the association between fiber intake and BP in our data. Also, supplements may not fully reproduce some of the physiological effects of a high-fiber diet that may be related to BP regulation, including a reduction in overfeeding and a lower postprandial peak of plasma insulin concentration.47 48
The observation that intakes of food groups and individual food items were significantly associated with BP, and the persistence of these associations after adjusting for fiber and magnesium intakes, suggests that other dietary factors may contribute to the observed associations of diet with BP. Although we cannot conclusively identify specific nutrients that will affect BP levels, our results support the hypothesis that dietary factors contribute to the regulation of BP and suggest that a diet richer in fruit, vegetables, dark bread, and rice may have a BP-lowering effect. Whether magnesium and fiber reduce BP levels requires further investigations. These should include further observational studies as well as randomized trials, ideally among subjects with low baseline intake of these nutrients, and basic research on mechanisms by which magnesium and fiber may regulate BP.
|
Acknowledgments |
|---|
This work was supported by research grants HL-35464, HL-34594, and CA40935 from the National Institutes of Health. The authors thank Frank Speizer, MD, principal investigator of the Nurses' Health Study. We also thank the participants in the study for their continuing cooperation and Karen Corsano, Mary Johnson, Laura Sampson, Ichiro Kawachi, Eric Rimm, and Ed Giovannucci for their expert advice.
|
Appendix 1 |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
Foods were grouped as follows: sweets (chocolate, candies, cookies, brownies, donuts, cakes, pies, sweet rolls, and jam); meats (processed meats, bacon, hot dogs, hamburgers, beef, pork, lamb, and chicken); fish (tuna, dark fish, other fish, and shrimp); cereals (cold cereal, cooked oats, cooked cereals, white bread, dark bread, English muffins, muffins, brown rice, white rice, pasta, other grains, pancakes, potatoes, and crackers); fruits (grapes, avocados, bananas, cantaloupes, watermelons, apples, apple juice, oranges, orange juice, grapefruits, grapefruit juice, other fruit juice, strawberries, blueberries, and peaches); vegetables (string beans, broccoli, sauerkraut, coleslaw, cabbage, cauliflower, brussels sprouts, carrots, corn, peas, beans, lentils, alfalfa sprouts, celery, mushrooms, yellow squash, yams, spinach, kale, iceberg lettuce, romaine lettuce, green pepper, garlic, tomatoes, tomato juice, tomato sauce, red chili sauce, tofu, and soybeans); high-fat dairy (milk, cream, ice cream, cream cheese, other cheese, butter, sour cream); and low-fat dairy (skim milk, sherbet, cottage cheese, yogurt).
Received December 7, 1995; first decision January 29, 1996; accepted January 29, 1996.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1References |
|---|
- Epstein FH, Eckoff RD. The epidemiology of high blood pressure: geographic distribution and etiological factors. In: Stamler J, Stamler R, Pullman TM, eds. The Epidemiology of Hypertension. New York, NY: Grune & Stratton; 1967:155-166.
- Cassel J. Studies of hypertension in migrants. In: Paul O, ed. Epidemiology and Control of Hypertension. Miami, Fla: Symposium Specialists; 1975:41-58.
- He J, Tell GS, Tang YC, Mo PS, He GQ. Effect of migration on blood pressure: the Yi people study. Epidemiology. 1991;2:88-97. [Medline] [Order article via Infotrieve]
- Poulter NR, Khaw KT, Hopwood BEC, Mugambi M, Peart WS, Rose G, Sever PS. The Kenyan Luo migration study: observations on the initiation of a rise in blood pressure. Br Med J. 1990;300:967-972.
- Elfor J, Phillips A, Thomson AG, Shaper AG. Migration and geographic variations in blood pressure in Britain. Br Med J. 1990;300:291-294.
-
Armstrong B, van Merwyk AJ, Coates H. Blood
pressure in Seventh-Day Adventist vegetarians. Am J
Epidemiol. 1977;105:444-449.
[Abstract/Free Full Text] - Rouse IL, Armstrong BK, Beilin LJ. The relationship of blood pressure to diet and lifestyle in two religious populations. J Hypertens. 1983;1:65-71. [Medline] [Order article via Infotrieve]
-
Sacks FM, Rosner B, Kass EH. Blood pressure in
vegetarians. Am J Epidemiol. 1974;100:390-398.
[Abstract/Free Full Text] -
Sacks FM, Kass EH. Low blood pressure in
vegetarians: effects of specific foods and nutrients. Am
J Clin Nutr. 1988;48:795-800.
[Abstract/Free Full Text] -
Jackson R, Stewart A, Beaglehole R, Scragg R.
Alcohol consumption and blood pressure. Am J
Epidemiol. 1985;122:1037-1044.
[Abstract/Free Full Text] -
Puddey IB, Beilin LJ, Vandongen R, Rouse IL, Rogers
P. Evidence for a direct effect of alcohol consumption on blood
pressure in normotensive men: a randomized controlled trial.
Hypertension. 1985;7:707-713.
[Abstract/Free Full Text] - Kannel WB, Brand N, Skinner J, Dawber TR, McNamara PM. The relation of adiposity to blood pressure and development of hypertension. Ann Intern Med. 1967;67:48-59.
- Dyer AR, Stamler J, Shekelle RB, Schoenberger JA, Stamler R, Shekelle S, Berkson DM, Paul O, Lepper MH, Lingberg HA. Relative weight and blood pressure in four Chicago epidemiologic studies. J Chronic Dis. 1982;35:897-908. [Medline] [Order article via Infotrieve]
- Stamler J, Farinaro E, Mojonnier LM, Hall Y, Moss D, Stamler R. Prevention and control of hypertension by nutritional-hygienic means. JAMA. 1980;243:1819-1823. [Abstract]
- INTERSALT Cooperative Research Group. INTERSALT: An international study of electrolyte excretion and blood pressure: results for 24-hour urinary sodium and potassium excretion. Br Med J. 1988;297:319-328.
- Law MR, Frost CD, Wald NJ. By how much does dietary salt reduction lower blood pressure? I: Analysis of observational data among populations. Br Med J. 1991;302:810-815.
- Frost CD, Law MR, Wald NJ. By how much does dietary salt reduction lower blood pressure? II: Analysis of observational data within populations. Br Med J. 1991;302:815-818.
- Law MR, Frost CD, Wald NJ. By how much does dietary salt reduction lower blood pressure? III: Analysis of data from trials of salt reduction. Br Med J. 1991;302:819-824.
-
Witteman JC, Willett WC, Stampfer MJ, Colditz GA, Sacks
FM, Speizer FE, Rosner B, Hennekens CH. A prospective study of
nutritional factors and hypertension among US women.
Circulation. 1989;80:1320-1327.
[Abstract/Free Full Text] - Willett WC, Stampfer MJ, Colditz GA, Rosner BA, Hennekens CH, Speizer FE. Dietary fat and the risk of breast cancer. N Engl J Med. 1987;316:22-28. [Abstract]
-
Willett WC, Sampson L, Stampfer MJ, Rosner B, Bain C,
Witschi J, Hennekens CH, Speizer FE. Reproducibility and
validity of a semiquantitative food frequency questionnaire.
Am J Epidemiol. 1985;122:51-65.
[Abstract/Free Full Text] - Willett WC, Reynolds RO, Cottrell-Hoehner S, Sampson L, Browne ML. Comparison of a semiquantitative food-frequency questionnaire with a one-year diet record. J Am Diet Assoc. 1987;87:43-47. [Medline] [Order article via Infotrieve]
- Willett WC. Nutritional Epidemiology. New York, NY: Oxford University Press; 1990.
-
Colditz GA, Martin P, Stampfer MJ, Willett WC, Sampson
L, Rosner B, Hennekens CH, Speizer FE. Validation of
questionnaire information on risk factors and disease outcomes in a
prospective cohort study of women. Am J Epidemiol. 1986;123:894-900.
[Abstract/Free Full Text] - Cox DR. Regression models and life-tables. J R Stat Soc B. 1972;34:187-220.
-
Rosner B, Polk BF. Predictive values of routine
blood pressure measurements in screening for hypertension.
Am J Epidemiol. 1983;117:429-442.
[Abstract/Free Full Text] - Witteman JCM, Grobbee DE. Calcium and magnesium in hypertension: current evidence. Magnes Bull. 1990;3:87-97.
- The Trials of Hypertension Prevention Collaborative Research Group. The effects of nonpharmacologic interventions on blood pressure of persons with high normal levels. JAMA. 1992;267:1213-1220. [Abstract]
- Grobbee DE, Waal-Manning HJ. The role of calcium supplementation in the treatment of hypertension: current evidence. Drugs. 1990;39:7-18. [Medline] [Order article via Infotrieve]
-
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;45:469-475.
[Abstract/Free Full Text] -
Fehily AM, Milbank JE, Yarnell JW, Hayes TM, Kubiki AJ,
Eastham RD. Dietary determinants of lipoproteins, total
cholesterol, viscosity, fibrinogen, and blood
pressure. Am J Clin Nutr. 1982;36:890-896.
[Abstract/Free Full Text] -
Ascherio A, Stampfer MJ, Colditz GA, Willett WC,
McKinlay J. Nutrient intakes and blood pressure in normotensive
males. Int J Epidemiol. 1991;20:886-891.
[Abstract/Free Full Text] -
Jenner DA, English DR, Vandongen R, Beilin LJ,
Armstrong BK, Miller MR, Dunbar D. Diet and blood pressure in
9-year-old children. Am J Clin Nutr. 1988;47:1052-1059.
[Abstract/Free Full Text] - Wright A, Burstyn PG, Gibney MJ. Dietary fiber and blood pressure. Br Med J. 1979;2:1541-1543.
- Schlamowitz P, Halberg T, Warnoe O, Wilstrup F, Ryttig K. Treatment of mild to moderate hypertension with dietary fibre. Lancet. 1987;2:622-623. [Medline] [Order article via Infotrieve]
- Margetts BM, Beilin LJ, Vandongen R, Armstrong B. A randomized controlled trial of the effects of dietary fibre on blood pressure. Clin Sci. 1987;72:343-350. [Medline] [Order article via Infotrieve]
-
Brussaard JH, van Raaij JM, Stasse-Wolthuis M, Katan
MB, Hautvast JG. Blood pressure and diet in normotensive
volunteers: absence of an effect of dietary fiber, protein, or
fat. Am J Clin Nutr. 1981;34:2023-2029.
[Abstract/Free Full Text] - Swain JF, Rouse IL, Courley CB, Sacks FM. Comparison of the effects of oat bran and low-fiber wheat on serum lipoprotein levels and blood pressure. N Engl J Med. 1990;322:193-195. [Medline] [Order article via Infotrieve]
- Lo GS, Cole TG. Soy cotyledon fiber products reduce plasma lipids. Atherosclerosis. 1990;82:59-67. [Medline] [Order article via Infotrieve]
-
Kestin M, Moss R, Clifton PM, Nestel PJ.
Comparative effects of three cereal brans on plasma lipids, blood
pressure, and glucose metabolism in mildly
hypercholesterolemic men. Am J Clin
Nutr. 1990;52:661-666.
[Abstract/Free Full Text] - Tuomilehto J, Silvasti M, Aro A, Koistinen A, Karttunen P, Gref CG, Ehnholm C, Uusitupa M. Long term treatment of severe hypercholesterolaemia with guar gum. Atherosclerosis. 1988;72:157-162. [Medline] [Order article via Infotrieve]
-
Van Horn L, Moag-Stahlber A, Liu K, Ballew C, Ruth K,
Hughes R, Stamler J. Effect on serum lipids of adding instant
oats to usual American diets. Am J Public Health. 1991;81:183-188.
[Abstract/Free Full Text] - Rigaud D, Ryttig KR, Angel LA, Apfelbaum M. Overweight treated with energy restriction and a dietary fiber supplement: a 6-month randomized, double blind, placebo-controlled trial. Int J Obes. 1990;14:763-769. [Medline] [Order article via Infotrieve]
- Rossner S, Anderson IL, Ryttig K. Effects of a dietary fibre supplement to a weight reduction programme on blood pressure: a randomized, double-blind, placebo-controlled trial. Acta Med Scand. 1988;223:353-357.[Medline] [Order article via Infotrieve]
- Hennekens CH, Buring JE. Observational evidence. Ann N Y Acad Sci. 1993;703:18-24. [Medline] [Order article via Infotrieve]
- Jacques FP. A cross-sectional study of vitamin C intake and blood pressure in the elderly. Int J Vitam Res. 1991;62:1-4.
- Young JB, Landsberg L. Diet-induced changes in sympathetic nervous system activity: possible implications for obesity and hypertension. J Chronic Dis. 1982;35:879-886. [Medline] [Order article via Infotrieve]
- Anderson JW. Fiber and health: an overview. Am J Gastroenterol. 1986;81:892-897.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  I. Tzoulaki, I. J Brown, Q. Chan, L. Van Horn, H. Ueshima, L. Zhao, J. Stamler, P. Elliott, and for the International Collaborative Research Group Relation of iron and red meat intake to blood pressure: cross sectional epidemiological study BMJ, July 15, 2008; 337(jul15_1): a258 - a258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Nakamura, C. Nagata, S. Oba, N. Takatsuka, and H. Shimizu Fruit and Vegetable Intake and Mortality from Cardiovascular Disease Are Inversely Associated in Japanese Women but Not in Men J. Nutr., June 1, 2008; 138(6): 1129 - 1134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Champagne Magnesium in Hypertension, Cardiovascular Disease, Metabolic Syndrome, and Other Conditions: A Review Nutr Clin Pract, April 1, 2008; 23(2): 142 - 151. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. D. Sesso, N. R. Cook, J. E. Buring, J. E. Manson, and J. M. Gaziano Alcohol Consumption and the Risk of Hypertension in Women and Men Hypertension, April 1, 2008; 51(4): 1080 - 1087. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Larsson, M. J. Virtanen, M. Mars, S. Mannisto, P. Pietinen, D. Albanes, and J. Virtamo Magnesium, Calcium, Potassium, and Sodium Intakes and Risk of Stroke in Male Smokers Arch Intern Med, March 10, 2008; 168(5): 459 - 465. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Touyz Transient receptor potential melastatin 6 and 7 channels, magnesium transport, and vascular biology: implications in hypertension Am J Physiol Heart Circ Physiol, March 1, 2008; 294(3): H1103 - H1118. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. L. Lutsey, L. M. Steffen, and J. Stevens Dietary Intake and the Development of the Metabolic Syndrome: The Atherosclerosis Risk in Communities Study Circulation, February 12, 2008; 117(6): 754 - 761. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. Lichtenstein, H. Rasmussen, W. W. Yu, S. R. Epstein, and R. M. Russell Modified MyPyramid for Older Adults J. Nutr., January 1, 2008; 138(1): 5 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kan, J. Stevens, G. Heiss, R. Klein, K. M Rose, and S. J London Dietary fiber intake and retinal vascular caliber in the Atherosclerosis Risk in Communities Study Am. J. Clinical Nutrition, December 1, 2007; 86(6): 1626 - 1632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Wang, J M. Gaziano, S. Liu, J. E Manson, J. E Buring, and H. D Sesso Whole- and refined-grain intakes and the risk of hypertension in women Am. J. Clinical Nutrition, August 1, 2007; 86(2): 472 - 479. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R Harriss, D. R English, J. Powles, G. G Giles, A. M Tonkin, A. M Hodge, L. Brazionis, and K. O'Dea Dietary patterns and cardiovascular mortality in the Melbourne Collaborative Cohort Study Am. J. Clinical Nutrition, July 1, 2007; 86(1): 221 - 229. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Dauchet, E. Kesse-Guyot, S. Czernichow, S. Bertrais, C. Estaquio, S. Peneau, A.-C. Vergnaud, S. Chat-Yung, K. Castetbon, V. Deschamps, et al. Dietary patterns and blood pressure change over 5-y follow-up in the SU.VI.MAX cohort Am. J. Clinical Nutrition, June 1, 2007; 85(6): 1650 - 1656. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Song, T. Y Li, R. M van Dam, J. E Manson, and F. B Hu Magnesium intake and plasma concentrations of markers of systemic inflammation and endothelial dysfunction in women Am. J. Clinical Nutrition, April 1, 2007; 85(4): 1068 - 1074. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bo, M. Durazzo, S. Guidi, M. Carello, C. Sacerdote, B. Silli, R. Rosato, M. Cassader, L. Gentile, and G. Pagano Dietary magnesium and fiber intakes and inflammatory and metabolic indicators in middle-aged subjects from a population-based cohort. Am. J. Clinical Nutrition, November 1, 2006; 84(5): 1062 - 1069. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Liu, H. K. Choi, E. Ford, Y. Song, A. Klevak, J. E. Buring, and J. E. Manson A Prospective Study of Dairy Intake and the Risk of Type 2 Diabetes in Women Diabetes Care, July 1, 2006; 29(7): 1579 - 1584. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. J. Appel, M. W. Brands, S. R. Daniels, N. Karanja, P. J. Elmer, and F. M. Sacks Dietary Approaches to Prevent and Treat Hypertension: A Scientific Statement From the American Heart Association Hypertension, February 1, 2006; 47(2): 296 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M Steffen, C. H Kroenke, X. Yu, M. A Pereira, M. L Slattery, L. Van H |