Original Article

Regular Yogurt Intake and Risk of Cardiovascular Disease

Among Hypertensive Adults
Justin R. Buendia,1 Yanping Li,2 Frank B. Hu,2 Howard J. Cabral,3 M. Loring Bradlee,1
Paula A. Quatromoni,4 Martha R. Singer,1 Gary C. Curhan,5 and Lynn L. Moore1

BACKGROUND RESULTS
High blood pressure (HBP) is a major cardiovascular disease (CVD) risk Yogurt intake was inversely associated with CVD risk (myocardial

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

factor. Clinical trials including Dietary Approaches to Stop Hypertension infarction and stroke) among hypertensive participants (P <0.01 in
(DASH) have demonstrated beneficial effects of dairy consumption on both cohorts). Among participants consuming ≥2 servings/week of
risks of HBP and CVD. Yogurt, a fermented dairy product, may indepen- yogurt, NHS women had a 17% (95% CI: 0.74–0.92) lower risk while
dently be related to CVD risk. HPFS men experienced a 21% (95% CI: 0.66–0.96) lower CVD risk com-
pared to those who consumed <1 serving/month. Regular yogurt
OBJECTIVE consumers with higher DASH diet scores had 16% (95% CI: 0.73–0.96)
To evaluate the association between yogurt consumption and CVD risk and 30% (95% CI: 0.57–0.85) CVD risk reductions in the 2 cohorts,
among hypertensive individuals in 2 large cohorts and to determine respectively.
whether the association differs among those whose eating pattern
more closely resembles the DASH diet.
CONCLUSION
METHODS Hypertensive men and women who consumed ≥2 servings/week of
Overall, 55,898 female Nurses’ Health Study (NHS) and 18,232 male yogurt, especially in the context of a healthy diet, were at lower risk for
Health Professionals Follow-Up Study (HPFS) participants with preva- developing CVD.
lent HBP were included. Cumulative average estimates of yogurt intake
from validated food frequency questionnaires were related to verified Keywords: blood pressure; dairy; diet; dietary approaches to stop hyper-
self-reported CVD outcomes using Cox proportional hazards models. tension; hypertension; yogurt.
Hazard ratios and 95% confidence intervals (CI) were adjusted for CVD
risk factors, medications, and diet covariates. doi:10.1093/ajh/hpx220

High blood pressure (HBP) not only affects about 1 billion which participants in the combination diet group (fruits
people worldwide but there is compelling evidence that it is a and vegetables plus low-fat dairy) experienced the greatest
major cause of cardiovascular morbidity and overall mortal- BP-lowering effect compared to both the control Western
ity.1 HBP is linked with an increased risk of cardiovascular diet and the fruits and vegetables only groups.8 Later
disease (CVD) through a variety of mechanisms including studies demonstrated that the DASH dietary pattern was
arterial stiffness2 and direct effects on vascular endothelial associated with a reduced risk of both fatal and nonfatal
function.3 CVD.9,10
Higher dairy consumption has been associated with Yogurt intake in the United States has increased substan-
beneficial effects on CVD-related comorbidities such as tially in the past decade.11 Early studies suggested that regular
hypertension,4,5 type 2 diabetes, insulin resistance,6,7 and consumption of fermented dairy products such as yogurt
dyslipidemia.5 One of the earliest studies to show a benefi- was associated with a lower risk of atherosclerotic vascular
cial effect of dairy intake on risk of HBP was the Dietary disease12 and a reduction in arterial stiffness13 in hyperten-
Approaches to Stop Hypertension (DASH) clinical trial, in sive subjects. A meta-analysis of 13 randomized controlled

1Department of Medicine, Preventive Medicine and Epidemiology,

Correspondence: Lynn L. Moore (llmoore@bu.edu).
Boston University School of Medicine, Boston, Massachusetts, USA;
Initially submitted December 4, 2017; accepted for publication 2Department of Nutrition, Harvard T. H. Chan School of Public Health,

January 3, 2018; online publication February 15, 2018. Boston, Massachusetts, USA; 3Department of Biostatistics, Boston
University School of Public Health, Boston, Massachusetts, USA;
4Department of Health Sciences/Programs in Nutrition, Sargent

College of Health and Rehabilitation Sciences, Boston University,

Boston, Massachusetts, USA; 5Channing Division of Network Medicine,
Department of Medicine, Brigham and Women’s Hospital and Harvard
Medical School , Boston, Massachusetts, USA.
© The Author(s) 2018. Published by Oxford University Press on
behalf of American Journal of Hypertension, Ltd. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com

American Journal of Hypertension 31(5) May 2018 557

Buendia et al.

trials found that probiotic fermented milk (vs. placebo) was that is made from milk and retains its calcium content.17
associated with a 3.98 mm Hg reduction in systolic blood Finally, a DASH diet score was calculated for each par-
pressure among hypertensive adults and a 2.09 mm Hg re- ticipant.9 While yogurt is a part of the usual DASH score,
duction among normotensives,14 raising the possibility of a it was excluded from calculation of the score in these
greater CVD benefit of yogurt consumption among hyper- analyses.
tensives than nonhypertensives.15
While yogurt is known to be a nutrient-rich food and a Assessment of HBP
source of probiotic bacteria, there are few large, longitudinal
studies of yogurt’s specific health effects rather than overall A participant who reported a HBP diagnosis on the
dairy consumption. To our knowledge, there is no published enrollment questionnaire in 1980 (NHS) or 1986 (HPFS)
evidence on long-term yogurt intake and CVD risk among was considered to have prevalent HBP. Those without HBP
individuals with prevalent HBP. Our goal was to examine the at enrollment were asked on each subsequent biennial ques-
association between yogurt intake and risk of CVD among tionnaire for any new diagnoses of HBP. Participants were
those with prevalent HBP and to determine whether this

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

considered to have prevalent HBP at first report and through-
association differs among those whose diet more closely out the rest of the study. The validity of self-reported HBP
resembles a heart-healthy dietary pattern. has been examined in both the NHS and HPFS cohorts.18,19
In the NHS, 77% of 51 cases of self-reported HBP in the val-
METHODS idation study had either systolic blood pressure or diastolic
blood pressure levels >160 or 95 mm Hg via standard BP
For the current analyses, participants included those measurement.18 In the HPFS, a medical record review of a
with prevalent HBP from the Nurses’ Health Study (NHS, random sample of 100 participants reporting a diagnosis of
N = 121,700, ages 30–55 at enrollment in 1980) and the HBP in 1988 confirmed the diagnosis in all participants.20
Health Professionals Follow-Up Study (HPFS, N = 51,529, Further, self-reported HBP among participants in the valid-
ages 40–75 at enrollment in 1986). The date of the first self- ation study was highly predictive of subsequent cardiovas-
reported HBP diagnosis served as the baseline for these cular events.19
analyses. Hence, the baseline visit differed depending on
when HBP was first reported. Participants without HBP
were excluded, as were those with missing data on dairy CVD outcome ascertainment
intake, missing, or implausible energy intake (<500 or
≥3,500 kilocalories/day for NHS and <800 or ≥4,200 kilocal- The assessment and diagnosis of CVD in these cohorts
ories/day for HPFS), or prevalent cancer, diabetes, or CVD has been previously described.21 For the current analyses,
(including revascularization, angina, myocardial infarction the primary outcomes include major CHD (nonfatal and
(MI), or stroke) at or before first diagnosis of HBP, leaving fatal MI) and stroke (nonfatal and fatal) cases. As a second-
55,898 and 17,984 NHS and HPFS participants, respect- ary outcome, cases of revascularization are included.
ively, for the current analyses. Analyses were approved by On each biennial questionnaire, participants reported
the Institutional Review Board of Boston University School any interim physician-diagnosed events including MI,
of Medicine. stroke, and revascularization. Permission was requested to
access medical records to confirm all reported new diagno-
ses. World Health Organization criteria (diagnostic symp-
Dairy and yogurt intake assessment toms, electrocardiographic findings, or elevated cardiac
enzyme concentrations) were used to confirm nonfatal MI.22
In the NHS, participants were asked to complete a mailed Nonfatal stroke was confirmed using computed tomog-
61-item semiquantitative food frequency questionnaire raphy, magnetic resonance imaging results or the National
(FFQ) in 1980 to report usual dietary intake in the preceding Survey of Stroke criteria (i.e., medical record confirmation
year. They were asked how often, on average, they consumed of a neurological deficit with sudden or rapid onset that per-
servings of each listed food item from the questionnaire sisted more than 24 hours or until death).23 Fatal MI and
using 9 possible response categories, ranging from “never or stroke were validated via physician review of death or med-
less than once a month” to “6 or more a day.” An expanded ical records. CVD events in which no medical records were
131-item FFQ was sent in 1984, 1986, and every 4 years available but were deemed probable were included in these
thereafter until 2006. The same version of the FFQ was used analyses, as was previously shown.21
from 1986 through 2006 for HPFS participants.16
The following foods were included in the estimation
of total dairy intake in our analyses: milk (skim, low- Covariates assessment
fat, reduced-fat, and whole), yogurt, ice cream, sher-
bet, frozen yogurt, cottage/ricotta cheese, and other A large number of potential confounding variables were
cheese. Servings of dairy foods were defined using explored including sociodemographic factors, family his-
MyPyramid servings from the United States Department tory, and other diet and lifestyle factors. Questionnaires
of Agriculture (USDA).17 Cream cheese and cream were were used to collect data on smoking status, physical activity
excluded from the dairy variable we created as they did (estimated as metabolic equivalent tasks [METs] per week),
not meet the MyPyramid definition of a dairy food—one other lifestyle factors, and family history of HBP and MI.

558 American Journal of Hypertension 31(5) May 2018

 Long-Term Yogurt Intake and Cardiovascular Disease

Body mass index was calculated as the self-reported weight or elevated cholesterol.24,25 In these cases, cumulative aver-
(in kg) divided by the height squared (in meters). age dietary intakes were carried forward to the exam prior
to the occurrence of one of these events. We compared these
results with our primary analyses in which cumulative aver-
Statistical analyses
age intake was calculated up to the stop censoring date and
Yogurt intake for each participant was estimated as the found that the hazard ratios (HRs) were virtually identical.
cumulative average intake starting at the time of the first Thus, our primary analyses are presented in the results of
HBP diagnosis and stopping at the time of first occurrence this manuscript.
of one of the following censoring events: date of CVD diag- HRs and 95% confidence intervals (CI) were calculated
nosis, date of death, loss to follow-up, or end of follow-up using Cox proportional hazards models to adjust for poten-
(30 June 2010 for NHS and 31 January 2010 for HPFS). tial confounding factors. HRs for the risk of MI, stroke, and
The calculation of cumulative average intakes has been revascularization associated with average yogurt intake
previous described in detail.24 This methodology is used to (categorized as <once/month, once/month to <once/week,
once/week to <twice/week, and ≥twice/week) were first esti-

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

minimize exposure measurement error and is intended to
reflect changes in yogurt intake over time. The DASH diet mated. Since MI and stroke were the primary outcomes of
score was also computed from cumulative average of each interested, we estimated HRs for this combined outcome
score component. Due to concerns about changes in self- as well as for total CVD (MI, stroke, or revascularization).
reported dietary intakes following certain diagnoses, sec- Finally, risk of each CVD outcome was estimated accord-
ondary analyses were conducted in which dietary intakes ing to combined categories of yogurt intake and DASH diet
were not updated following after exams at which partici- scores. For these latter analyses, both yogurt intake and the
pants reported the occurrence of angina, revascularization, DASH diet score were dichotomized (lower vs. higher) and

Table 1. Baseline characteristics of participants according to usual yogurt intake

NHS (N = 55,898) HPFS (N = 18,232)

Yogurt intake categories (1C s/d)

1/month– 1/month–
<1/month <1/week 1–<2/week ≥2/week <1/month <1/week 1–<2/week ≥2/week

n 21,474 10,565 11,827 12,032 9,920 3,369 2,751 2,192

Age2, year 58.7 (10.1) 58.5 (9.8) 58.4 (9.8) 58.5 (9.9) 60.9 (10.0) 58.9 (9.9) 59.3 (10.1) 60.6 (10.2)
Activity, MET-hours/ 13.0 (18.5) 14.5 (19.0) 15.8 (22.0) 18.3 (22.4) 25.7 (35.3) 28.4 (33.5) 30.9 (34.5) 34.4 (42.5)
week
BMI, kg/m2 26.8 (5.3) 27.1 (5.3) 27.0 (5.2) 27.0 (5.3) 24.6 (7.8) 24.7 (7.9) 24.7 (7.7) 24.2 (7.9)
Calories, kcals/day 1,598 (526) 1,629 (518) 1,702 (522) 1,805 (542) 1,918 (599) 1,970 (603) 2,019 (596) 2,135 (630)
Alcohol, g/day 6.8 (12.2) 6.2 (11.1) 5.9 (10.2) 5.8 (9.5) 13.9 (17.7) 11.7 (15.1) 11.3 (14.0) 10.9 (14.1)
DASH diet score 22.4 (4.6) 23.8 (4.5) 24.6 (4.4) 25.7 (4.4) 22.6 (4.9) 24.5 (4.8) 25.2 (4.8) 26.7 (4.6)
Total fiber, g/day 17.2 (5.1) 18.3 (5.3) 18.7 (5.3) 19.4 (5.4) 20.8 (7.3) 22.5 (6.8) 23.3 (7.2) 24.2 (7.0)
Fruits/vegetables, 4.4 (1.8) 4.8 (1.9) 5.1 (1.9) 5.5 (2.0) 5.1 (2.4) 5.8 (2.6) 6.1 (2.6) 6.7 (2.9)
s/day
Red and processed 1.2 (0.93) 1.1 (0.86) 1.1 (0.83) 0.93 (0.76) 1.2 (0.84) 1.0 (0.79) 0.98 (0.75) 0.88 (0.73)
meats, s/day
Total dairy, s/day 1.3 (1.0) 1.5 (1.0) 1.7 (1.1) 2.1 (1.2) 1.3 (1.0) 1.4 (1.0) 1.5 (0.99) 2.0 (1.2)
Yogurt, s/day 0.00 (0.02) 0.06 (0.06) 0.14 (0.16) 0.45 (0.44) 0.00 (0.01) 0.06 (0.06) 0.15 (0.15) 0.48 (0.40)
Current smoker, % 23.2 15.2 12.2 11.1 10.0 5.1 4.3 3.5
Antihypertensive 44.3 46.0 47.6 49.0 55.2 52.3 51.4 48.8
medication use, %
HBP family history, % 48.2 52.8 55.4 53.3 48.0 49.5 53.0 49.5
MI family history, % 21.2 20.5 19.7 20.3 34.0 32.7 32.6 35.1
Current 23.7 28.0 29.7 30.9 N/A N/A N/A N/A
postmenopausal
hormone use, %

Values are mean (SD) for continuous variables and % for categorical variables. Abbreviations: BMI, body mass index; DASH, Dietary
Approaches to Stop Hypertension; HBP, high blood pressure; HPFS, Health Professionals Follow-Up Study; MET, metabolic equivalent task;
MI, myocardial infarction; NHS, Nurses’ Health Study.

American Journal of Hypertension 31(5) May 2018 559

Buendia et al.

then cross-classified, yielding 4 mutually exclusive exposure status, cigarette pack years, alcohol intake, postmenopau-
categories: (i) low yogurt + low DASH diet score (refer- sal hormone use (in NHS), aspirin and multivitamin use,
ence group); (ii) low yogurt + high DASH diet score; (iii) body mass index (baseline and updated every 2 years), and
high yogurt + low DASH diet score; (iv) high yogurt + high cumulative average intakes of the following dietary factors:
DASH diet score. The dichotomous cutpoints used were total energy intake, carbohydrates, total fat and fat subtypes
selected using sensitivity analyses to optimize analytic power (saturated, monounsaturated, polyunsaturated, omega-3,
and to remain consistent with FFQ and usual yogurt serving trans fatty acids), protein (total, animal, and plant), whole
sizes: yogurt (<2 vs. ≥2 servings/week in the NHS; <1 vs. ≥1 grains, fiber (total, cereal), nuts, fruits and vegetables,
serving/week in the HPFS) and DASH diet score (<25 vs. sugar-sweetened beverages, potatoes, beans, red and pro-
≥25). Twenty-five was chosen as the cutpoint for the DASH cessed meats, sodium, potassium, calcium, magnesium,
diet score as it was around the median in both cohorts (24). and vitamin E. Only covariates that changed the HRs by
The following potential confounders were explored in >10% were retained in the final models. These included:
the proportional hazards models: age, race, family histo- age, race, physical activity, MI family history, antihyperten-
ries of HBP, diabetes, and MI, antihypertensive medica- sive medication use, and intakes of total energy, total fiber,

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

tion use, hypercholesterolemia, physical activity, smoking trans fats, and other dairy foods (milk and cheese). Fixed

Table 2. Yogurt intake and subsequent risk of major CHD, stroke, and CABG in 2 cohorts

Risk of major CHD

NHS HPFS

Age-adjusted HR Age-adjusted HR
Yogurt intakea Cases IR (95% CI) Multivariable HR (95% CI)b Cases IR (95% CI) Multivariable HR (95% CI)b

<1/month 961 303 1.00 1.00 705 634 1.00 1.00

1/month–<1/week 334 228 0.79 (0.70–0.90) 0.85 (0.75–0.97) 197 516 0.91 (0.77–1.06) 0.94 (0.80–1.10)
1–<2/week 252 199 0.69 (0.60–0.79) 0.75 (0.65–0.87) 134 530 0.91 (0.75–1.09) 0.93 (0.77–1.12)
≥2/week 217 177 0.64 (0.55–0.75) 0.70 (0.60–0.82) 96 481 0.80 (0.64–0.99) 0.81 (0.65–1.02)
P for linear trendc <0.001 <0.001 0.03 0.06

NHS HPFS

Risk of stroke
<1/month 848 268 1.00 1.00 349 313 1.00 1.00
1/month–<1/week 337 230 0.90 (0.79–1.02) 0.98 (0.86–1.11) 86 224 0.81 (0.64–1.03) 0.85 (0.67–1.08)
1–<2/week 302 239 0.87 (0.76–0.99) 0.97 (0.84–1.11) 40 158 0.54 (0.39–0.76) 0.58 (0.41–0.81)
≥2/week 262 214 0.83 (0.72–0.95) 0.94 (0.81–1.09) 42 210 0.71 (0.51–0.98) 0.75 (0.54–1.05)
P for linear trendc <0.01 0.44 <0.01 0.04

Meta-analysis

Risk of major CHD Risk of stroke

<1/month 1,666 389 1.00 1.00 1,197 280 1.00 1.00
1/month–<1/week 531 287 0.83 (0.76–0.92) 0.88 (0.80–0.98) 423 229 0.88 (0.79–0.98) 0.95 (0.85–1.06)
1–<2/week 386 254 0.76 (0.68–0.85) 0.81 (0.72–0.91) 342 225 0.82 (0.72–0.92) 0.90 (0.79–1.02)
≥2/week 313 219 0.69 (0.61–0.78) 0.74 (0.65–0.84) 304 213 0.81 (0.71–0.92) 0.91 (0.79–1.04)
P for linear trendc <0.001 <0.001 <0.001 0.12
I2 69.0 30.2 63.7 69.4

Abbreviations: BMI, body mass index; BP, blood pressure; CABG, coronary artery bypass grafting; CHD, Coronary Heart Disease; CI, con-
fidence interval; CVD, cardiovascular disease; HBP, high blood pressure; HPFS, Health Professionals Follow-Up Study; HR, hazard ratio; IR,
incidence rate per 100,000 person-years; MI, myocardial infarction; NHS, Nurses’ Health Study.
aCumulative average of yogurt intake was calculated from first report of HBP up to the first of the following events: CVD diagnosis, lost to

follow up, death, end of study.

bAdjusted for age, race, smoking (defined as never, past, current with 1–14 cigs/day, current with 15–24 c/day, 25+ c/day), family history of

MI, physical activity (continuous from baseline), BMI, BP-lowering medication use, and intakes of total energy, alcohol, trans fatty acids, fiber,
milk, and cheese.
cLinear trend across yogurt intake categories was quantified with a Wald test for linear trend by assigning the median value to each category

and modeling it as a continuous variable.

560 American Journal of Hypertension 31(5) May 2018

 Long-Term Yogurt Intake and Cardiovascular Disease

effects meta-analyses were used to pool estimates from both separately (Supplementary Table 1), there were 3,300 and
cohorts. All analyses were performed with SAS software 2,148 cases in NHS and HPFS, respectively. Higher yog-
(version 9.4; SAS Institute, Cary, NC). urt intake in women was associated with a 16% lower risk
of undergoing revascularization (P trend <0.01) while there
was no significant association observed among HPFS men
RESULTS
(Supplementary Table 1). In separate analyses (not shown)
The baseline characteristics of NHS and HPFS partici- restricted to cases of ischemic stroke only, results were simi-
pants with prevalent HBP are shown according to yogurt lar to those of total stroke presented in Table 2 but the power
intake in Table 1. Those with the highest yogurt intakes of these analyses was very low given the small numbers of
(≥2 servings/week) tended to be more physically active, ischemic strokes in HPFS.
drank less alcohol, and were less likely to smoke. Higher We examined our primary outcome (incident major
yogurt intake was also associated with a healthier diet as CHD or stroke) first (Table 3) before investigating the sec-
indicated by a higher DASH diet score, higher fiber and ondary outcome including revascularization procedures
(Supplementary Table 1). In both cohorts, participants con-

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

fruit and vegetable intakes, as well as lower intakes of red
and processed meats. Both studies had similar distribu- suming 1–<2 cups or ≥2 servings/week of yogurt had an
tions of the DASH diet scores (mean/median around approximately 20% lower risks of major CHD or stroke dur-
23/24). ing the follow-up period (P for trend <0.01 in both cohorts).
In Table 2, there were 1,764 cases of incident MI (major When revascularization was added to the total CVD out-
CHD) and 1,749 strokes in the NHS. In the HPFS, there come variable, the risk estimates were attenuated for both
were 1,132 reported MIs and 517 strokes. Higher intakes men and women, but remained significant for both cohorts
of yogurt (≥2 servings/week) were associated with a 30% (P for trend <0.01 for both).
reduction (95% CI: 0.60–0.82) in risk of MI among NHS Table 4 shows the independent and combined associations
women (P trend <0.001) and a 19% reduction (95% CI: of higher yogurt intake and diet quality (as measured by the
0.65–1.02) in HPFS men (P trend = 0.06). Estimates for DASH diet score) with major CHD and stroke risk. Among
stroke risk were weaker than those for MI among women. women in the NHS cohort, a higher DASH score alone was
When revascularization procedures (CABG) was examined unassociated with risk of major CHD or stroke. For each

Table 3. Risk of incident major CHD or stroke according to usual yogurt intake

Risk of incident major CHD or stroke

NHS HPFS

Age-adjusted HR Multivariable HR (95% Age-adjusted HR

Yogurt intakea Cases IR (95% CI) CI)b Cases IR (95% CI) Multivariable HR (95% CI)b

<1/month 1,793 568 1.00 1.00 1,026 928 1.00 1.00

1/month–<1/week 661 452 0.84 (0.77–0.92) 0.91 (0.83–0.99) 279 733 0.89 (0.77–1.01) 0.92 (0.80–1.05)
1–<2/week 542 429 0.77 (0.70–0.85) 0.85 (0.77–0.94) 170 675 0.79 (0.67–0.93) 0.82 (0.69–0.96)
≥2/week 479 392 0.74 (0.67–0.82) 0.83 (0.74–0.92) 135 679 0.77 (0.64–0.92) 0.79 (0.66–0.96)
P for linear trendc <0.001 <0.001 <0.01 <0.01

Meta-analysis

<1/month 2,819 662 1.00 1.00

1/month–<1/week 940 510 0.85 (0.79–0.92) 0.91 (0.84–0.98)
1–<2/week 712 470 0.77 (0.71–0.84) 0.84 (0.77–0.91)
≥2/week 614 432 0.75 (0.68–0.82) 0.82 (0.75–0.90)
P for linear trendc <0.001 <0.001
I2 0.0 0.0

Abbreviations: BMI, body mass index; BP, blood pressure; CHD, coronary heart disease; CI, confidence interval; CVD, cardiovascular dis-
ease; HBP, high blood pressure; HPFS, Health Professionals Follow-Up Study; HR, hazard ratio; IR, incidence rate per 100,000 person-years;
MI, myocardial infarction; NHS, Nurses’ Health Study.
aCumulative average of yogurt intake was calculated from first report of HBP up to the first of the following events: CVD diagnosis, lost to

follow up, death, end of study.

bAdjusted for age, race, smoking (defined as never, past, current with 1–14 cigs/day, current with 15–24 c/day, 25+ c/day), family history of

MI, physical activity (continuous from baseline), BMI, BP-lowering medication use, and intakes of total energy, alcohol, trans fatty acids, fiber,
milk, and cheese.
cLinear trend across yogurt intake categories was quantified with a Wald test for linear trend by assigning the median value to each category

and modeling it as a continuous variable.

American Journal of Hypertension 31(5) May 2018 561

Buendia et al.

Table 4. Associations between independent and combined yogurt intake and DASH diet scores with risk of CVD

Risk of major CHD

NHS HPFS

Yogurt/DASH diet Age-adjusted HR Multivariable HR Age-adjusted HR Multivariable HR

scorea Cases IR (95% CI) (95% CI)b Cases IR (95% CI) (95% CI)b

Low/low 963 263 1.00 1.00 595 591 1.00 1.00

Low/high 584 261 0.84 (0.76–0.94) 0.96 (0.86–1.07) 379 606 0.85 (0.74–0.97) 0.90 (0.78–1.04)
High/low 90 174 0.74 (0.60–0.92) 0.78 (0.62–0.97) 69 551 1.01 (0.78–1.29) 1.01 (0.79–1.30)
High/high 127 179 0.65 (0.54–0.79) 0.75 (0.61–0.91) 89 471 0.70 (0.56–0.87) 0.73 (0.58–0.93)

Risk of stroke

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

NHS HPFS

Low/low 907 248 1.00 1.00 268 266 1.00 1.00

Low/high 580 259 0.87 (0.79–0.97) 1.01 (0.90–1.13) 187 298 0.90 (0.74–1.09) 0.93 (0.76–1.15)
High/low 114 221 0.95 (0.78–1.16) 1.02 (0.84–1.25) 26 207 0.85 (0.56–1.27) 0.86 (0.57–1.30)
High/high 148 208 0.76 (0.64–0.90) 0.91 (0.76–1.10) 36 190 0.61 (0.43–0.87) 0.63 (0.44–0.91)

Risk of major CHD or stroke

NHS HPFS

Low/low 1,854 508 1.00 1.00 841 840 1.00 1.00

Low/high 1,142 512 0.85 (0.79–0.91) 0.97 (0.89–1.05) 554 890 0.87 (0.78–0.97) 0.91 (0.81–1.03)
High/low 202 393 0.85 (0.73–0.98) 0.90 (0.77–1.04) 93 746 0.96 (0.77–1.19) 0.97 (0.78–1.21)
High/high 277 391 0.72 (0.63–0.81) 0.84 (0.73–0.96) 122 649 0.67 (0.55–0.81) 0.70 (0.57–0.85)

Meta-analysis

Risk of major CHD Risk of stroke

Low/low 1,558 333 1.00 1.00 1,175 251 1.00 1.00

Low/high 963 336 0.85 (0.78–0.92) 0.94 (0.86–1.02) 767 268 0.88 (0.80–0.96) 0.99 (0.90–1.09)
High/low 159 248 0.84 (0.72–1.00) 0.87 (0.74–1.03) 140 218 0.93 (0.78–1.11) 0.99 (0.83–1.18)
High/high 216 240 0.67 (0.58–0.77) 0.74 (0.64–0.86) 184 204 0.73 (0.62–0.85) 0.85 (0.72–1.00)

Risk of major CHD or stroke

Low/low 2,695 579 1.00 1.00

Low/high 1,696 595 0.85 (0.80–0.91) 0.95 (0.89–1.02)
High/low 295 462 0.88 (0.78–0.99) 0.92 (0.81–1.04)
High/high 399 445 0.70 (0.63–0.78) 0.79 (0.71–0.89)

Abbreviations: BMI, body mass index; BP, blood pressure; CHD, coronary heart disease; CI, confidence interval; CVD, cardiovascular dis-
ease; DASH, Dietary Approaches to Stop Hypertension; HBP, high blood pressure; HPFS, Health Professionals Follow-Up Study; HR, hazard
ratio; IR, incidence rate per 100,000 person-years; NHS, Nurses’ Health Study.
aCumulative average of yogurt intake was calculated from first report of HBP up to the first of the following events: CVD diagnosis, lost to

follow up, death, end of study. Yogurt cutpoints are <2/week (low), ≥2/week (high) in NHS; <1/week (low), and ≥1/week (high) in HPFS. DASH
diet score cutpoints are <25 (low) and ≥25 (high) in both cohorts.
bAdjusted for age, race, smoking (defined as never, past, current with 1–14 cigs/day, current with 15–24 c/day, 25+ c/day), family history of

MI, physical activity (continuous from baseline), BMI, BP-lowering medication use, and intakes of total energy, alcohol, trans fatty acids, fiber,
milk, and cheese.

outcome, the greatest risk reduction occurred among those CHD and stroke, respectively. For the combined CVD outcome
jointly exposed to higher yogurt consumption and higher with major CHD and stroke, the lowest risk was again found
DASH diet score. Among HPFS participants, those with higher among those with higher yogurt intakes and a higher DASH
yogurt intakes and a higher DASH score had 27% (95% CI: diet score. These HRs were attenuated when revascularization
0.58–0.93) and 37% (95% CI: 0.44, 0.91) lower risks of major was added to the CVD outcome (Supplemental Table 2).

562 American Journal of Hypertension 31(5) May 2018

 Long-Term Yogurt Intake and Cardiovascular Disease

DISCUSSION These tripeptides may play a role in platelet aggregation and

clot formation via their inhibition of angiotensin-convert-
Higher yogurt consumption among hypertensive women ing enzyme in the renin–angiotensin system,32 one of the
and men in the NHS and HPFS cohorts was associated with major BP regulators in the body. Angiotensin-converting
a lower risk of MI and stroke and, among women, with a enzyme inhibition leads to lower concentrations of angio-
lower likelihood of undergoing a revascularization proced- tensin II, a potent vasoconstrictor32 that has also been shown
ure. The strongest and most consistent results were asso- to increase intracellular calcium in vascular smooth muscle
ciated with the lower risk of MI in both men and women. cells and pH in platelets, leading to increased platelet aggre-
Further, the associations between yogurt consumption and gation.33 Platelets from hypertensive patients produce more
CVD risk were stronger among participants with higher reactive oxygen species, thereby reducing the bioavailabil-
DASH diet scores. ity of nitric oxide,34 an important vasodilator and inhibitor
The independent effect of the original DASH diet interven- of platelet aggregation.35,36 In spontaneously hypertensive
tion8 was also observed in other trials including the Exercise rats, fermented dairy-derived peptides have been shown to
and Nutrition interventions for CardiOvasculaR hEalth improve vascular response in vitro possibly by the stimula-

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

(ENCORE) study of overweight, middle-aged hypertensive tion of nitric oxide release.37
men and women.26 That study found that compared to those This is the first large prospective study to show a reduced
consuming a typical American diet, those randomized to the risk of MI and stroke associated with regular yogurt con-
DASH diet intervention had lower BP, brachial artery flow- sumption among hypertensive adults. This study has several
mediated dilation, baroreflex sensitivity, and left ventricular important strengths including its prospective design and the
mass after a 4-month intervention period.26 use of updated repeated assessments of diet for estimating
A number of studies including the numerous DASH tri- long-term intakes and repeated measures of potential con-
als have found dairy intake to be associated with reduced founding variables. This study also had several limitations.
risks of both HBP and CVD.8–10,26 Several prospective stud- Since yogurt is highly correlated with an overall healthy diet
ies among those without HBP have found an inverse associ- and lifestyle, it is possible that some potential confounders
ation between yogurt intake and markers of CVD.27,28 In a associated with a healthy lifestyle were not measured or were
5-year prospective study, Australian women over the age of imperfectly measured, leading to uncontrolled or residual
70 who consumed >100 g/day of yogurt had significantly less confounding. Additionally, yogurt is a complex dairy prod-
common carotid artery intima–media thickness than those uct and we were not able to study specific types of yogurt such
who consumed less yogurt.27 A case–control study of Italian as those with different added probiotics. While few studies
adults found a 45% lower odds of acute MI among those who examined food sources of probiotics, some have shown that
consumed ≥7 cups of yogurt per week compared to nonyo- different probiotic combinations have unique effects on BP
gurt consumers.28 and CVD risk factors.14 Future studies are needed to assess
In early clinical trials, the BP-lowering effect of dairy different types of yogurt. Finally, the small number of stroke
was largely attributed to its calcium content. However, cases in HPFS yielded inadequate statistical power to exam-
it has been shown by others that factors in dairy other ine different types of stroke.
than or in addition to dietary calcium may explain the
beneficial effects on BP and CVD risk.29 Several small CONCLUSIONS
randomized controlled feeding trials have examined the
effects of fermented milk products on CVD risk factors Our results suggest that higher long-term yogurt intake
among hypertensive individuals.13,30,31 For example, a is associated with lower CVD risk among hypertensive men
randomized, placebo-controlled, double-blind trial with and women. Higher yogurt intake in combination with
89 hypertensive subjects tested the impact of adding the an overall heart-healthy diet as measured by the DASH
milk tripeptides, isoleucine-proline-proline (IPP), and diet score was associated with greater reductions in CVD
valine-proline-proline (VPP) to Lactobacillus helveticus- risk among hypertensive men and women. These findings
fermented milk. The investigators added 1.2 mg of IPP endorse that incorporation of yogurt into a healthy diet pat-
and 1.3 mg of VPP per 100 g of milk for 12 weeks and then tern for individuals with hypertension to aid in the preven-
increased the dose to 5.8 mg IPP and 6.6 mg VPP per 100 g tion of incident CVD.
for an additional 12 weeks. This protocol led to reductions
in arterial stiffness in the intervention group compared
with control subjects who were given Lactobacillus helve- SUPPLEMENTARY MATERIAL
ticus-fermented milk without IPP or VPP supplementa- Supplementary materials are available at American Journal
tion.30 Hirota et al. found that these same fermented milk of Hypertension online.
tripeptides improved vascular endothelial function inde-
pendent of their BP-lowering effects in a randomized trial
of hypertensive men.31
Taken together, these studies suggest that fermented dairy
products may lower CVD risk through effects on vascular ACKNOWLEDGMENTS
stiffness as well as through direct effects on BP. Consequently, The Nurses’ Health Study and Health Professionals Follow-up
yogurt consumption may be associated with lower CVD risk Study cohorts are supported by grants UM1 CA186107, UM1
by improving BP control among hypertensive individuals. CA176726, and UM1 CA167552 from the National Institutes

American Journal of Hypertension 31(5) May 2018 563

Buendia et al.

of Health. The current analyses were supported by small grants to vascular mortality: a meta-analysis of individual data for one million
from the National Dairy Council, the General Mills Bell Institute adults in 61 prospective studies. Lancet 2002; 360:1903–1913.
16. Willett WC, Sampson L, Stampfer MJ, Rosner B, Bain C, Witschi J,
for Health and Nutrition, and the Boston Nutrition and Obesity Hennekens CH, Speizer FE. Reproducibility and validity of a semi-
Research Center grant P30DK046200. The Boston Nutrition quantitative food frequency questionnaire. Am J Epidemiol 1985;
Obesity Research Center is administratively based at Boston 122:51–65.
Medical Center and is funded by the National Institutes of 17. Bowman SA, Friday JE, Moshfegh AJ. MyPyramid Equivalents
Database, 2.0 for USDA Survey Foods, 2003–2004: Documentation and
Health (NIH/NIDDK) grant P30DK046200. User Guide. Beltsville, MD: Food Surveys Research Group Beltsville
Human Nutrition Research Center, Agricultural Research Service, US
Department of Agriculture, 2008.
DISCLOSURE 18. Colditz GA, Martin P, Stampfer MJ, Willett WC, Sampson L, Rosner
B, Hennekens CH, Speizer FE. Validation of questionnaire information
The authors declared no conflict of interest. on risk factors and disease outcomes in a prospective cohort study of
women. Am J Epidemiol 1986; 123:894–900.
19. Ascherio A, Rimm EB, Giovannucci EL, Colditz GA, Rosner B, Willett

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

WC, Sacks F, Stampfer MJ. A prospective study of nutritional factors
and hypertension among US men. Circulation 1992; 86:1475–1484.
REFERENCES 20. Ascherio A, Hennekens C, Willett WC, Sacks F, Rosner B, Manson
J, Witteman J, Stampfer MJ. Prospective study of nutritional factors,
1. Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, blood pressure, and hypertension among US women. Hypertension
de Ferranti S, Després JP, Fullerton HJ, Howard VJ, Huffman MD, Judd 1996; 27:1065–1072.
SE, Kissela BM, Lackland DT, Lichtman JH, Lisabeth LD, Liu S, Mackey 21. Sun Q, Shi L, Rimm EB, Giovannucci EL, Hu FB, Manson JE, Rexrode
RH, Matchar DB, McGuire DK, Mohler ER 3rd, Moy CS, Muntner P, KM. Vitamin D intake and risk of cardiovascular disease in US men
Mussolino ME, Nasir K, Neumar RW, Nichol G, Palaniappan L, Pandey and women. Am J Clin Nutr 2011; 94:534–542.
DK, Reeves MJ, Rodriguez CJ, Sorlie PD, Stein J, Towfighi A, Turan 22. Mendis S, Thygesen K, Kuulasmaa K, Giampaoli S, Mähönen M, Ngu
TN, Virani SS, Willey JZ, Woo D, Yeh RW, Turner MB; American Heart Blackett K, Lisheng L; Writing Group on behalf of the Participating
Association Statistics Committee and Stroke Statistics Subcommittee. Experts of the WHO Consultation for Revision of WHO Definition of
Heart disease and stroke statistics–2015 update: a report from the Myocardial Infarction. World Health Organization definition of myo-
American Heart Association. Circulation 2015; 131:e29–322. cardial infarction: 2008-09 revision. Int J Epidemiol 2011; 40:139–146.
2. Rabkin SW, Waheed A, Poulter RS, Wood D. Myocardial perfusion 23. Robins M, Weinfeld FD. The National Survey of Stroke. Study design
pressure in patients with hypertension and coronary artery disease: and methodology. Stroke 1981; 12:I7–11.
implications for DBP targets in hypertension management. J Hypertens 24. Hu FB, Stampfer MJ, Rimm E, Ascherio A, Rosner BA, Spiegelman D,
2013; 31:975–982. Willett WC. Dietary fat and coronary heart disease: a comparison of
3. Lytsy P, Lind L, Sundström J. Endothelial function and risk of hyperten- approaches for adjusting for total energy intake and modeling repeated
sion and blood pressure progression: the prospective investigation of dietary measurements. Am J Epidemiol 1999; 149:531–540.
the vasculature in Uppsala seniors. J Hypertens 2013; 31:936–939. 25. Joshipura KJ, Hu FB, Manson JE, Stampfer MJ, Rimm EB, Speizer FE,
4. Elwood PC, Pickering JE, Fehily AM. Milk and dairy consumption, dia- Colditz G, Ascherio A, Rosner B, Spiegelman D, Willett WC. The effect
betes and the metabolic syndrome: the Caerphilly prospective study. J of fruit and vegetable intake on risk for coronary heart disease. Ann
Epidemiol Community Health 2007; 61:695–698. Intern Med 2001; 134:1106–1114.
5. van Meijl LE, Vrolix R, Mensink RP. Dairy product consumption and 26. Blumenthal JA, Babyak MA, Hinderliter A, Watkins LL, Craighead
the metabolic syndrome. Nutr Res Rev 2008; 21:148–157. L, Lin PH, Caccia C, Johnson J, Waugh R, Sherwood A. Effects of the
6. Pereira MA, Jacobs DR Jr, Van Horn L, Slattery ML, Kartashov AI, DASH diet alone and in combination with exercise and weight loss
Ludwig DS. Dairy consumption, obesity, and the insulin resist- on blood pressure and cardiovascular biomarkers in men and women
ance syndrome in young adults: the CARDIA Study. JAMA 2002; with high blood pressure: the ENCORE study. Arch Intern Med 2010;
287:2081–2089. 170:126–135.
7. Elwood PC, Pickering JE, Givens DI, Gallacher JE. The consumption of 27. Ivey KL, Lewis JR, Hodgson JM, Zhu K, Dhaliwal SS, Thompson PL,
milk and dairy foods and the incidence of vascular disease and diabetes: Prince RL. Association between yogurt, milk, and cheese consumption
an overview of the evidence. Lipids 2010; 45:925–939. and common carotid artery intima-media thickness and cardiovascular
8. Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks disease risk factors in elderly women. Am J Clin Nutr 2011; 94:234–239.
FM, Bray GA, Vogt TM, Cutler JA, Windhauser MM, Lin PH, 28. Tavani A, Gallus S, Negri E, La Vecchia C. Milk, dairy products, and cor-
Karanja N. A clinical trial of the effects of dietary patterns on blood onary heart disease. J Epidemiol Community Health 2002; 56:471–472.
pressure. DASH Collaborative Research Group. N Engl J Med 1997; 29. Ruidavets JB, Bongard V, Simon C, Dallongeville J, Ducimetière P,
336:1117–1124. Arveiler D, Amouyel P, Bingham A, Ferrières J. Independent contribu-
9. Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu tion of dairy products and calcium intake to blood pressure variations
FB. Adherence to a DASH-style diet and risk of coronary heart disease at a population level. J Hypertens 2006; 24:671–681.
and stroke in women. Arch Intern Med 2008; 168:713–720. 30. Jauhiainen T, Rönnback M, Vapaatalo H, Wuolle K, Kautiainen H,
10. Folsom AR, Parker ED, Harnack LJ. Degree of concordance with DASH Groop PH, Korpela R. Long-term intervention with Lactobacillus hel-
diet guidelines and incidence of hypertension and fatal cardiovascular veticus fermented milk reduces augmentation index in hypertensive
disease. Ame J Hypertens 2007; 20:225–232. subjects. Eur J Clin Nutr 2010; 64:424–431.
11. Nielsen SJ, Ogden C. Trends in yogurt consumption, US adults, 1999– 31. Hirota T, Ohki K, Kawagishi R, Kajimoto Y, Mizuno S, Nakamura Y,
2012. The FASEB Journal 2015; 29(Suppl. 1):587.517. Kitakaze M. Casein hydrolysate containing the antihypertensive tripep-
12. Mann GV. Studies of a surfactant and cholesteremia in the Maasai. Am tides Val-Pro-Pro and Ile-Pro-Pro improves vascular endothelial func-
J Clin Nutr 1974; 27:464–469. tion independent of blood pressure-lowering effects: contribution of
13. Jauhiainen T, Rönnback M, Vapaatalo H, Wuolle K, Kautiainen H, the inhibitory action of angiotensin-converting enzyme. Hypertens Res
Korpela R. Lactobacillus helveticus fermented milk reduces arterial 2007; 30:489–496.
stiffness in hypertensive subjects. Int Dairy J 2007; 17:1209–1211. 32. Erdmann K, Cheung BW, Schröder H. The possible roles of food-
14. Dong JY, Szeto IM, Makinen K, Gao Q, Wang J, Qin LQ, Zhao Y. Effect derived bioactive peptides in reducing the risk of cardiovascular dis-
of probiotic fermented milk on blood pressure: a meta-analysis of ran- ease. J Nutr Biochem 2008; 19:643–654.
domised controlled trials. Br J Nutr 2013; 110:1188–1194. 33. Touyz RM, Schiffrin EL. Effects of angiotensin II and endothelin-1 on
15. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R; Prospective platelet aggregation and cytosolic pH and free Ca2+ concentrations in
Studies Collaboration. Age-specific relevance of usual blood pressure essential hypertension. Hypertension 1993; 22:853–862.

564 American Journal of Hypertension 31(5) May 2018

 Long-Term Yogurt Intake and Cardiovascular Disease

34. Camilletti A, Moretti N, Giacchetti G, Faloia E, Martarelli D, Mantero 36. Wilkinson IB, Qasem A, McEniery CM, Webb DJ, Avolio AP, Cockcroft
F, Mazzanti L. Decreased nitric oxide levels and increased calcium JR. Nitric oxide regulates local arterial distensibility in vivo. Circulation
content in platelets of hypertensive patients. Am J Hypertens 2001; 2002; 105:213–217.
14:382–386. 37. Sipola M, Finckenberg P, Korpela R, Vapaatalo H, Nurminen ML. Effect
35. Azuma H, Ishikawa M, Sekizaki S. Endothelium-dependent inhibition of long-term intake of milk products on blood pressure in hypertensive
of platelet aggregation. Br J Pharmacol 1986; 88:411–415. rats. J Dairy Res 2002; 69:103–111.

Downloaded from https://academic.oup.com/ajh/article/31/5/557/4818397 by guest on 27 January 2024

American Journal of Hypertension 31(5) May 2018 565