Abstract
Background
Adherence to several diet quality scores including the Alternative Healthy Eating Index (AHEI), Alternative Mediterranean diet score (AMED), and Dietary Approach to Stop Hypertension (DASH) has been associated with lower risk of cardiovascular disease (CVD), but little is known about how changes in these scores over time influence subsequent CVD risk.
Methods and Results
We analyzed the association between 4-year changes in three diet quality scores (AHEI, AMED, and DASH) and subsequent CVD risk among 29,343 men in the Health Professionals Follow-up Study and 51,195 women in the Nurses’ Health Study (1986–2010). During 1,394,702 person-years of follow up, we documented 11,793 CVD cases. Compared with participants whose diet quality remained relatively stable in each 4-year period, those with the greatest improvement in diet quality scores had a 7%–8% lower CVD risk in the subsequent 4-year period (pooled hazard ratio, 0.92 [95% confidence interval (CI): 0.87–0.99] for AHEI; 0.93 [95% CI: 0.85–1.02] for AMED; and 0.93 [95% CI: 0.87–0.99] for DASH; all P-trend<0.05). In the long term, increasing the diet scores from baseline to the first 4-year follow up was associated with lower CVD risk during the next 20 years (7% [95% CI: 1% to 12%] for AHEI and 9% [95% CI: 3% to 14%] for AMED). A decrease in diet quality scores was associated with significantly elevated risk of CVD in subsequent time periods.
Conclusions
Improving adherence to diet quality scores over time is associated with significantly lower CVD risk both in the short term and the long term.
Keywords: Diet, change, cardiovascular disease, longitudinal cohort study
Organizations such as the American College of Cardiology and the American Heart Association as well as The 2015 Dietary Guidelines Advisory Committee Report support and emphasize studying overall dietary patterns as a primary approach when investigating how dietary factors may influence cardiovascular outcomes 1,2. Because cardiovascular disease (CVD) is the primary cause of death in the US 3, studies assessing diet quality as a whole may provide valuable information on how to prevent the burden of CVD through better dietary habits2. One approach is to calculate a score based on adherence to dietary patterns or to recommendations. The Alternative Healthy Eating Index-2010 (AHEI-2010) is one a priori-defined diet score based on recommendations for food and nutrient consumption with current scientific evidence of beneficial health effects 4. Two other commonly studied scores are the Alternative Mediterranean diet (AMED) score that comprises foods that are characteristic of the Mediterranean pattern 5, and the Dietary Approach to Stop Hypertension (DASH) score that was developed from the DASH dietary recommendations aiming to reduce blood pressure 6.
Although adherence to several diet quality scores including the AHEI, the AMED, and the DASH has been associated with lower CVD incidence and mortality 4, 5, 7–13, most previous studies have had only one-time measurement in diet 7, 10, 14–16 and thus have been unable to evaluate changes in diet quality and subsequent CVD risk. Therefore, using two large prospective cohorts with data on repeated measures of diet, lifestyle, medical conditions, and other chronic diseases collected every two to four years over a 24-year period, we evaluated the association between 4-year changes in three diet quality scores (AHEI, AMED, and DASH) and the subsequent risk of CVD (in the subsequent 4-year follow up). In addition, this study examined long-term effects of diet quality scores on CVD using changes in the three diet quality scores from baseline to the first 4-year follow up to predict subsequent 20-year risk of CVD.
Methods
Study Population
The Health Professional Follow-up Study (HPFS), an ongoing cohort, was established in 1986 and consists of 51,529 US male health professionals aged 40 to 75 years from all 50 US states. The Nurses’ Health Study (NHS), also an ongoing cohort, began in 1976 and consists of 121,700 registered female nurses aged 30 to 55 years from 11 US states. Participants from both cohorts responded to validated questionnaires inquiring about detailed medical history, lifestyle, and other health information 17 at baseline and every two years thereafter to update information on lifestyle practices, risk factors, and chronic disease occurrence 18. Detailed description of the two cohorts has been reported elsewhere 19. Using 1986 as baseline for both cohorts in the current investigation, detailed information about diet and lifestyle was assessed with follow up until 2010.
We excluded men and women who had a baseline history of CVD (myocardial infarction, angina, stroke, transient ischemic attack, and coronary revascularization) or cancer because the diagnoses of these conditions might have changed diet. We also excluded participants who left more than 10 blank food items on the baseline food frequency questionnaire (FFQ), had missing information at baseline on the three diet quality scores and other lifestyle covariates, and were out of the predefined limits of energy intake levels (<800 or >4200 kcal/day for men and <500 or >3500 kcal/day for women). After exclusions, we followed up 29,343 men in the HPFS and 50,195 women in the NHS (1986–2010).
The study protocol was approved by the institutional review boards of Brigham and Women’s Hospital and Harvard Chan School of Public Health. All participants gave informed consent.
Assessment of diet quality
In the HPFS and the NHS, dietary information was collected using a validated, 131-item semi-quantitative FFQ. Every four years, FFQs were sent to the participants to update information on diet. Participants were asked how often, on average, they consumed each food of a standard portion size in the past year. The frequency responses ranged from “never or less than once per month” to “six or more times per day”. The reproducibility and validity of the FFQs have been described in detail elsewhere 4, 20, 21, showing good correlation between nutrients assessed by the FFQ and multiple weeks of food records 21. We calculated the three diet quality scores from the FFQ (Supplemental Table 1).
The AHEI-2010 was based on a comprehensive review of foods and nutrients that have been associated consistently with lower risk of chronic disease in clinical and epidemiologic investigations. 4 The score emphasized higher intake of vegetables, fruit, whole grains, nuts and legumes, long-chain n-3 fats, polyunsaturated fatty acids (PUFAs), and lower intake of sugar-sweetened beverages and fruit juice, red and processed meat, trans fat, sodium, and alcohol. All components were scored from 0 (unhealthy) to 10 (healthiest), and the total score ranged from 0 (non-adherence) to 110 (perfect adherence).
The AMED score was modified and adapted to the Mediterranean diet scale designed by Trichopoulou et al 22. The components included vegetables, fruits, whole grains, nuts, legumes, fish, red and processed meat, alcohol consumption, and monounsaturated-to-saturated fat ratios. For these components, participants consuming above the median intake received 1 point while those below the median intake, received 0 points; red and processed meat consumption below the median received 1 point; alcohol intake between 5–15 g/day (women) and 10–25 g/day (men) was assigned 1 point. The total score ranged from 0 to 9, with a higher score representing closer resemblance to the Mediterranean diet.
The DASH score was based on food and nutrients emphasized or minimized from the DASH diet and focused on 8 components 6. For each component, we classified participants into quintiles according to intake ranking (ranging from 1 to 5; 5 being the best score for higher intake of fruits, vegetables, nuts and legumes, low-fat dairy products, and whole grains; 5 being the best score of lower intake in sodium, sweetened beverages, and red and processed meat). Total score ranged from 8 to 40 points.
Assessment of cardiovascular disease
We included incident cases of CVD defined as coronary heart disease (CHD, including fatal and non-fatal myocardial infarction and coronary artery bypass surgery) and stroke. When a participant reported an incident event on each biennial questionnaire, permission was requested to examine medical records, reviewed by study investigators blinded to the participant’s risk factor status. For each event, the month and year of diagnosis was recorded as the diagnosis date. Myocardial infarction was defined according to the World Health Organization criteria and cardiac-specific troponin levels23. Stroke was confirmed using the National Survey of Stroke criteria, requiring a constellation of neurologic deficits, sudden or rapid onset, and duration of at least 24 hours or until death 24. The information on coronary artery bypass surgery was based on unconfirmed self-reports, which have high validity in these cohorts 25. When medical records were not available, interviews or letters confirmed CHD and stroke events that were designated as “probable”. Deaths were identified from the state vital statistics records and the National Death Index, or reported by families and the postal system.6
Assessment of covariates
Updated biennial information on lifestyle and CVD risk factors were assessed including age, weight, smoking status, aspirin use, menopausal status, postmenopausal hormone therapy and oral contraceptives use, physical activity, and newly diagnosed chronic disease, including hypertension, hypercholesterolemia, and diabetes. Physician-diagnosed hypertension, diabetes and high cholesterol were ascertained biennially. Height and parental history of CVD prior to age 60 years were ascertained for women in 1976 and for men in 1986. Height and weight were used to calculate body mass index (BMI) (kg/m2). Every four years, alcohol intake was updated on the FFQs.
Statistical methods
We used Cox proportional hazards models with time-varying covariates and age as the underlying time scale to assess the association between the updated 4-year changes in the three diet quality scores and the subsequent risk of CVD, including separate models for CHD and stroke. The models incorporating diet scores that were updated every 4 years provided an estimate of the increased risk of a person having a CVD event in the subsequent 4 years given that they were event-free prior to that 4 year period. Changes in the three diet quality scores were divided into quintiles to calculate the hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) for CVD. Person-years were calculated from the date of return of the baseline questionnaire to the date of diagnosis of CVD, death, loss of follow up or the end of follow up (January 31, 2010 for HPFS and June 30, 2010 for NHS), whichever came first. We adjusted for the following potential confounders: age, race, family history of CVD, menopausal status and postmenopausal hormone use, aspirin use, and change in smoking status (never to never, never to current, past to current, current to past, and current to current) during each 4-year period. We also adjusted for baseline and 4-year changes in physical activity and energy intake. For each individual diet quality score as exposure, we also adjusted for baseline diet quality score. In addition, baseline and 4-year time changes in alcohol intake were included in the model when analyzing the DASH score because this component was not included in the score definition. A test for linear trend across quintiles of change was performed by assigning a median value to each quintile, producing a single continuous variable used to model P-trend. Additionally, a 20-percentile increase in each score was calculated from the median value of each quintile In an additional model, to assess if hypertension, hypercholesterolemia, diabetes, BMI, and weight change were potential mediators or confounders, we further included updated 2-year history of hypertension, hypercholesterolemia, diabetes, baseline BMI (<23, 23–24.9, 25–29.9, 30–34.9, ≥35), and changes in body weight in each 4-year period as time-varying covariates.. To examine long-term association of diet quality scores on CVD, changes in the three diet quality scores were analyzed from baseline to the first 4-year follow up (1986–1990) to predict subsequent 20-year risk of CVD (1990–2010).
In addition, a sensitivity analysis was conducted assessing the AHEI and the AMED without the alcohol component and including both baseline and changes in alcohol intake as covariates. We tested for potential effect modification by current smoking status, presence of hypertension, hypercholesterolemia or diabetes, baseline diet quality scores, physical activity, baseline BMI, and changes in weight.
In each cohort, all analyses were performed separately to achieve better control of confounding. For the primary analyses, to obtain overall estimates for both men and women and to increase statistical power, HRs and CIs were calculated from the multivariable adjusted models in both cohorts combined using an inverse, variance-weighted meta-analysis with random-effects model, accounting for heterogeneity between studies. All statistical tests were 2-sided and performed using SAS version 9.4 for UNIX (SAS Institute, Cary, NC, USA).
Results
During 1,394,702 person-years of follow up, we documented 11,793 incident CVD cases (6228 in HPFS and 5565 in NHS). In both cohorts, participants with the largest decrease in diet quality scores tended to have greater weight gain over a 4-year period and higher diet scores at baseline compared with those with relatively stable diet (Table 1). Participants with the greatest increase in diet scores initially tended to have lower total energy intake, decreased consumption of alcohol, lower diet scores and increased physical activity over a 4-year period. The updated and baseline changes in the three diet quality scores were significantly correlated (Supplemental Table 2).
Table 1.
Age-adjusted characteristics based on baseline 4-year changes in the three, diet quality scores (quintiles AHEI, AMED, and DASH)1
| AHEI (score range 0–110) |
AMED (score range 0–9) |
DASH (score range 8–40) |
|||||||
|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||
| Q1 (Largest decrease) |
Q3 (Relatively no change) |
Q5 (Largest increase) |
Q1 (Largest decrease) |
Q3 (Relatively no change) |
Q5 (Largest increase) |
Q1 (Largest decrease) |
Q3 (Relatively no change) |
Q5 (Largest increase) |
|
| Health Professional Follow-up Study | |||||||||
| No. of participants | 5868 | 5869 | 5868 | 5945 | 7074 | 5176 | 5008 | 6297 | 5339 |
| Initial diet score | 59.2 (11.2) | 52.0 (10.8) | 47.0 (10.4) | 5.55 (1.60) | 4.24 (1.88) | 2.98 (1.59) | 27.5 (4.58) | 23.7(4.85) | 20.5(4.41) |
| Changes in diet score | −11.2 (4.1) | 0.76 (1.20) | 13.3 (4.39) | −2.58 (0.81) | 0 (0.0) | 2.55 (0.80) | −5.55 (1.65) | 0.49 (0.5) | 5.76 (1.85) |
| Age, years | 56.9(9.53) | 56.8(9.37) | 57.3(9.36) | 56.8(9.48) | 56.9(9.36) | 56.7(9.23) | 56.6(9.58) | 56.8(9.37) | 57.3(9.29) |
| Initial BMI, Kg/m2 | 25.4(3.19) | 25.3(3.11) | 25.3(3.09) | 25.4 (3.15) | 25.3(3.07) | 25.4(3.00) | 25.2(2.99) | 25.3(3.05) | 25.6 (3.12) |
| Weight change, Kg | 2.34(8.05) | 1.32(7.70) | −0.24(8.30) | 1.80 (8.02) | 1.34(7.74) | 0.53(8.13) | 2.47 (7.79) | 1.44 (7.54) | −0.14 (8.67) |
| Initial physical activity, MET- h/wk | 22.8 (32.5) | 21.1 (28.3) | 21.2(29.3) | 22.7 (31.9) | 22.1 (30.2) | 20.5 (28.9) | 24.0(32.9) | 21.3(27.4) | 19.1(2.6.5) |
| Changes in physical activity, MET-h/week | 1.52(3.29) | 1.69 (3.34) | 1.77 (3.51) | 1.48 (3.34) | 1.68 (3.38) | 1.79 (3.29) | 1.46 (3.27) | 1.65 (3.29) | 1.82 (3.39) |
| Initial alcohol intake, g/d | 10.3 (12.2) | 11.6(15.4) | 13.1 (17.8) | 11.3(14.0) | 11.1(15.4) | 12.2(16.1) | 11.0(14.7) | 11.6(15.4) | 11.6(15.7) |
| Changes in alcohol intake, g/d | 0.34(9.30) | −1.08(8.38) | −3.33 (11.2) | −1.40(9.04) | −1.10(9.13) | −1.49(10.1) | −1.11(9.20) | −1.11(9.32) | −1.72(9.81) |
| White race, % | 96 | 96 | 95 | 95 | 96 | 95 | 96 | 96 | 94 |
| Current smoker, % | 9 | 9 | 9 | 9 | 9 | 10 | 8 | 9 | 11 |
| Hypertension, % | 17 | 17 | 17 | 16 | 17 | 16 | 16 | 16 | 18 |
| High cholesterol, % | 16 | 20 | 25 | 17 | 19 | 26 | 16 | 19 | 27 |
| Diabetes, % | 3 | 2 | 4 | 3 | 3 | 3 | 3 | 3 | 4 |
| Parental history of MI before 65y, % | 32 | 31 | 32 | 31 | 32 | 32 | 32 | 32 | 32 |
| Total energy intake, Kcal/d | 2012(607) | 2018(610) | 1958(583) | 2062(592) | 2010(605) | 1899(558) | 2039(586) | 2011(609) | 1952(583) |
| Change in total energy intake, kcal/d | −53(536) | −80.0(495) | −113 (514) | −259(498) | −69(497) | 120 (500) | −127 (530) | −65.1(491) | −47.1(523) |
|
| |||||||||
| Nurses’ Health Study | |||||||||
| No. of participants | 10039 | 10039 | 10039 | 9576 | 11673 | 9742 | 9265 | 10377 | 8824 |
| Initial diet score | 57.6(10.7) | 50.2(10.2) | 44.8(9.44) | 5.39(1.49) | 3.99(1.76) | 2.72(1.58) | 27.2(4.35) | 24.1(4.68) | 20.4(4.20) |
| Changes in diet score | −10.3 (4.06) | 1.72 (1.23) | 14.58 (4.63) | −2.57 (0.80) | 0.0 (0.0) | 2.55 (0.79) | −5.68 (1.69) | −0.51 (0.50) | 5.81 (1.88) |
| Age, years | 56.1(7.12) | 55.9(7.09) | 56.0(7.00) | 56.1(7.10) | 55.9(7.14) | 55.7(7.00) | 56.1(7.11) | 56.0(7.11) | 55.6(6.97) |
| Initial BMI, Kg/m2 | 25.1 (4.68) | 25.1(4.63) | 25.3(4.85) | 25.2(4.56) | 25.1(4.63) | 25.2(4.79) | 25.1(4.60) | 24.9(4.54) | 25.6(4.95) |
| Weight change, Kg | 3.96(10.8) | 2.54(10.2) | 0.48(11.3) | 2.96(10.6) | 2.51(10.5) | 1.63(11.0) | 3.98(10.7) | 2.58(9.92) | 0.25(11.6) |
| Initial physical activity, MET- h/wk | 14.7(20.8) | 14.1(20.7) | 14.1(20.4) | 14.8(21.6) | 14.3(20.6) | 13.3(18.4) | 15.5(20.1) | 14.4(21.3) | 12.4(16.2) |
| Changes in physical activity, MET-h/week | 0.07(1.59) | 0.15(1.55) | 0.18(1.59) | 0.11(1.58) | 0.14(1.55) | 0.21(1.54) | 0.04(1.62) | 0.11(1.55) | 0.23(1.53) |
| Initial alcohol intake, g/d | 5.78(8.95) | 5.88(10.3) | 7.84(13.2) | 6.20(9.84) | 6.16(10.8) | 6.61(11.3) | 5.92(10.1) | 6.56(10.9) | 5.94(10.6) |
| Changes in alcohol intake, g/d | −0.44(6.83) | −0.72(5.59) | −2.59(9.20) | −1.17(6.78) | −1.01(6.76) | −1.24(7.42) | −0.91(6.84) | −1.07(6.99) | −1.40(7.26) |
| White race, % | 99 | 98 | 97 | 98 | 98 | 98 | 99 | 98 | 97 |
| Current smoker, % | 19 | 19 | 22 | 18 | 20 | 21 | 19 | 20 | 22 |
| Hypertension, % | 24 | 23 | 24 | 23 | 24 | 24 | 25 | 22 | 25 |
| High cholesterol, % | 32 | 35 | 38 | 32 | 35 | 38 | 32 | 33 | 40 |
| Diabetes, % | 3 | 3 | 4 | 3 | 3 | 3 | 3 | 3 | 4 |
| Parental history of MI before 65y, % | 20 | 20 | 20 | 20 | 20 | 20 | 21 | 20 | 20 |
| Total energy intake, Kcal/d | 1807(527) | 1776(514) | 1723(507) | 1839(509) | 1774(523) | 1683(491) | 1811(500) | 1775(526) | 1712(506) |
| Change in total energy intake, kcal/d | −6(471) | −13(433) | −34(463) | −176(447) | −20(432) | 142(450) | −77(463) | −16(429) | 33(469) |
| Menopausal status and postmenopausal hormone use, % | |||||||||
| Premenopausal | 30 | 31 | 30 | 29 | 31 | 32 | 30 | 30 | 32 |
| Postmenopausal+never users | 28 | 27 | 29 | 27 | 27 | 28 | 28 | 28 | 28 |
| Postmenopausal+past users | 26 | 25 | 26 | 27 | 26 | 25 | 26 | 26 | 25 |
| Postmenopausal+current users | 15 | 14 | 14 | 14 | 14 | 13 | 15 | 14 | 14 |
| Missing information | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 | 2 |
Values are means (SD) or percentages and are standardized to the age distribution of the study population. We presented the results based on quintiles of change in any of the three dietary quality scores: Q1 (representing the largest decrease in diet), Q3 (reference category considered as relatively no change) and Q5 (representing the largest increase in diet).
BMI, body mass index; MET, metabolic equivalent of task.
The HR and 95% CI for CVD risk based on 4-year changes in diet quality scores are presented in Table 2. Compared with individuals whose diet quality remained relatively stable (no change) in each 4-year period, those with the greatest improvement (11%–22%) in diet quality had a 7%–8% lower risk of CVD in the subsequent 4-year period (pooled HR [95% CI]: AHEI, 0.92 [0.87–0.99]; AMED, 0.93 [0.85–1.02]; DASH, 0.93 [0.87–0.99]; all P-trend<0.05]. In addition, a 20-percentile increase in diet scores was associated with 3%–9% lower risk of CVD (pooled HR [95% CI]: AHEI, 0.91 [0.86–0.97]; AMED, 0.97 [0.94–0.99]; and DASH, 0.94 [0.90–0.98]) (Table 2). We explored the association with CHD and stroke, separately. For CHD, individuals with the highest quintile of change in the AHEI had a 12% lower CHD risk (95% CI: 4%–19%).
Table 2.
Hazard Ratios and 95% Confidence Intervals for incidence CVD based on quintiles of updated 4-year changes in diet quality scores
| Q1 | Q2 | Q3 | Q4 | Q5 | P- trend | HR equivalent to 20-percentile of increase* | |
|---|---|---|---|---|---|---|---|
| HPFS | |||||||
|
| |||||||
| AHEI | |||||||
| Median change (%) | −8.11 (−7.37) | −1.82 (−1.66) | 0 (0) | 4.43 (4.03) | 11.63 (10.57) | ||
| Cases/person-years | 1452/104850 | 1053/84414 | 1086/87586 | 1331/106700 | 1306/105642 | ||
| Multivariable model 1 | 1.06 (0.96–1.16) | 1.01 (0.92–1.11) | 1 [Reference] | 1.01 (0.93–1.11) | 0.98 (0.90–1.08) | 0.10 | 0.93(0.85,1.01) |
| Multivariable model 2 | 1.05 (0.96–1.15) | 1.01 (0.92–1.11) | 1 [Reference] | 1.01 (0.92–1.11) | 0.97 (0.88–1.06) | 0.049 | 0.92 (0.84–0.99) |
| AMED | |||||||
| Median change (%) | −2 (−22.22) | −1 (−11.11) | 0 (0) | 1 (11.11) | 2 (22.22) | ||
| Cases/person-years | 1433/109474 | 652/56809 | 2192/159682 | 808/69253 | 1143/93972 | ||
| Multivariable model 1 | 0.99 (0.92–1.07) | 0.93 (0.84–1.02) | 1 [Reference] | 0.93 (0.85–1.02) | 0.94 (0.86–1.01) | 0.26 | 0.98 (0.94–1.02) |
| Multivariable model 2 | 0.99 (0.92–1.07) | 0.93 (0.85–1.02) | 1 [Reference] | 0.92 (0.85–1.01) | 0.92 (0.85–0.99) | 0.02 | 0.97 (0.93–0.99) |
| DASH | |||||||
| Median change (%) | −4 (−12.5) | −1 (−3.13) | 0 (0) | 2 (6.25) | 4 (12.5) | ||
| Cases/person-years | 1472/104940 | 998/90505 | 1546/113874 | 978/83240 | 1234/96633 | ||
| Multivariable model 1 | 1.04 (0.95–1.13) | 0.95 (0.86–1.04) | 1 [Reference] | 0.94 (0.85–1.03) | 0.99 (0.90–1.08) | 0.13 | 0.95 (0.89–1.01) |
| Multivariable model 2 | 1.04 (0.95–1.13) | 0.95 (0.87–1.04) | 1 [Reference] | 0.93 (0.85–1.02) | 0.96 (0.88–1.05) | 0.03 | 0.93 (0.88–0.99) |
|
| |||||||
| NHS | |||||||
| AHEI | |||||||
| Median change (%) | −8.96 (−7.86) | −2.52 (−2.29) | 0.36 (0.31) | 4.99 (4.53) | 12.15 (11.04) | ||
| Cases/person-years | 1238/189147 | 990/164467 | 1131/168579 | 1106/191772 | 1100/191128 | ||
| Multivariable model 1 | 0.98 (0.90–1.07) | 0.96 (0.88–1.05) | 1 [Reference] | 0.91 (0.83–0.99) | 0.92 (0.84–1.01) | 0.08 | 0.93 (0.85–1.01) |
| Multivariable model 2 | 0.98 (0.90–1.07) | 0.97 (0.88–1.06) | 1 [Reference] | 0.91 (0.83–0.99) | 0.91 (0.83–0.99) | 0.03 | 0.91 (0.84–0.99) |
| AMED | |||||||
| Median change (%) | −2 (−22.22) | −1 (−11.11) | 0 (0) | 1 (11.11) | 2 (22.22) | ||
| Cases/person-years | 1201/190520 | 892/134161 | 1628/250486 | 783/142070 | 1061/187857 | ||
| Multivariable model 1 | 1.01 (0.93–1.09) | 1.05 (0.97–1.15) | 1 [Reference] | 1.00 (0.91–1.09) | 0.99 (0.91–1.08) | 0.51 | 0.99(0.95–1.03) |
| Multivariable model 2 | 1.01 (0.93–1.10) | 1.06 (0.97–1.15) | 1 [Reference] | 0.99 (0.91–1.08) | 0.98 (0.90–1.06) | 0.25 | 0.98(0.94–1.02) |
| DASH | |||||||
| Median change (%) | −4 (−12.5) | −2 (−6.25) | 0 (0) | 2 (6.25) | 5 (15.63) | ||
| Cases/person-years | 1272/195557 | 986/164612 | 1076/162690 | 1116/196422 | 1115/185812 | ||
| Multivariable model 1 | 0.99 (0.91–1.08) | 0.95 (0.87–1.04) | 1 [Reference] | 0.96 (0.88–1.05) | 0.92 (0.84–1.01) | 0.15 | 0.96 (0.90–1.02) |
| Multivariable model 2 | 1.01 (0.93–1.11) | 0.97 (0.89–1.07) | 1 [Reference] | 0.97 (0.89–1.07) | 0.91 (0.83–1.00) | 0.04 | 0.94 (0.89–0.99) |
|
| |||||||
| Pooled analysis | |||||||
| AHEI | |||||||
| Multivariable model 1 | 1.01 (0.94–1.07) | 0.97 (0.91–1.03) | 1 [Reference] | 0.95 (0.86–1.05) | 0.94 (0.88–1.00) | 0.01 | 0.93 (0.87–0.99) |
| Multivariable model 2 | 1.01 (0.94–1.07) | 0.97 (0.91–1.04) | 1 [Reference] | 0.95 (0.87–1.05) | 0.92 (0.87–0.99) | 0.003 | 0.91 (0.86–0.97) |
| AMED | |||||||
| Multivariable model 1 | 1.00 (0.95– 1.06) | 0.98 (0.85–1.13) | 1 [Reference] | 0.96 (0.90–1.03) | 0.95 (0.87–1.03) | 0.07 | 0.98 (0.95–1.00) |
| Multivariable model 2 | 1.01 (0.95– 1.07) | 0.99 (0.86–1.14) | 1 [Reference] | 0.96 (0.90–1.02) | 0.93 (0.85–1.02) | 0.01 | 0.97 (0.94–0.99) |
| DASH | |||||||
| Multivariable model 1 | 1.01 (0.95–1.08) | 0.94 (0.88–1.00) | 1 [Reference] | 0.95 (0.89–1.01) | 0.95 (0.89–1.01) | 0.01 | 0.95 (0.92–1.00) |
| Multivariable model 2 | 1.02 (0.96–1.09) | 0.95 (0.89–1.02) | 1 [Reference] | 0.96 (0.90–1.02) | 0.93 (0.87–0.99) | 0.003 | 0.94 (0.90–0.98) |
Model 1: adjusted for age, initial diet quality score (quintiles), race (white nonwhite), family history of MI, aspirin use, initial body mass index (<23, 23–24.9, 25–29.9,30–34.9, and >=35 kg/m2), weight change (quintiles) during the 4-year period, and simultaneous changes in other lifestyle factors: smoking status (never to never, never to current, pasta to past, past to current, current to past, current to current or missing indicator) and initial and changes (all in quintiles) in physical activity and total energy intake. Model 2: Model 1+ type 2 diabetes (yes or no), hypertension (yes or no), and hypercholesterolemia (yes or no). For DASH additionally adjusted for by change an initial alcohol intake (in quintiles).
The 20-percentile increase in each score was calculated from the median value of each quintile.
There was a significant inverse trend in CHD risk across quintiles in the DASH score over 4-year periods (P-trend=0.023) (Supplemental Table 3). No significant association was found for the association between the AMED and CHD risk; however, for stroke risk, there was a significant inverse association across quintiles over 4-year periods (P-trend=0.026). For the DASH score, participants in the highest quintile presented a 15% lower stroke risk (95% CI: 3%–26%). A 20-percentile increase in the AHEI and the DASH score was significantly associated with a 6% and 9% lower CHD risk, respectively, and a 20-percentile increase in the AMED and the DASH scores was significantly associated with a 7% and 9% lower stroke risk, respectively (Supplemental Table 3). The associations remained after removing weight change from the multivariable model.
To examine the long-term association for changes in diet quality scores and CVD risk, we used changes in diet quality scores from baseline to the first 4-year follow-up to predict risk of CVD in the subsequent 20-year follow up (Table 3). In pooled multivariable analyses, increasing the diet scores (12%–22%) from baseline to the first 4-year follow-up was associated with lower risk of CVD during the next 20-year follow up (AHEI, 7% [95% CI: 1%–12%] and AMED, 9% [95% CI: 3%–14%]). Although an increase in the DASH score was not associated with lower CVD risk, a decrease (16%) was associated with an 8% (95% CI: 2%–15%) higher CVD risk over the next 20 years. Consistent results were found for CHD risk, while no significant association was found for stroke risk (Supplemental Table 4). Examining changes in the three diet scores in longer period (1986–2002), an increase in CVD risk from 2002 to 2012 was observed with a decrease in the three diet quality scores (AHEI, 14% [95% CI: 2%–27%], AMED, 15% [95% CI: 4%–26%], and DASH, 16% [95% CI: 4%–26%]) (Supplemental Table 5). Additionally, the overall Kaplan Meier estimate of the cumulative risk of CVD and the crude incidence each 4-year period and at 20-year of follow-up was provided in the Supplemental figure 1.
Table 3.
Pooled Hazard Ratios and 95% Confidence Intervals for incidence CVD during 20-year follow up based on baseline 4-year changes (1986–1990) in diet quality scores
| Q1 | Q2 | Q3 | Q4 | Q5 | P-trend | HR equivalent to 20-percentile of increase* | |
|---|---|---|---|---|---|---|---|
| AHEI | |||||||
| Multivariable model 1 | 1.03 (0.97–1.09) | 1.05 (0.99–1.11) | 1 [Reference] | 0.99 (0.90–1.09) | 0.96 (0.90–1.02) | 0.008 | 0.92 (0.87–0.98) |
| Multivariable model 2 | 1.03 (0.96–1.11) | 1.05 (0.99–1.11) | 1 [Reference] | 0.98 (0.88–1.10) | 0.93 (0.88–0.99) | 0.0003 | 0.90 (0.85–0.95) |
| AMED | |||||||
| Multivariable model 1 | 1.00 (0.94–1.06) | 0.94 (0.88–1.01) | 1 [Reference] | 0.97 (0.92–1.03) | 0.93 (0.88–0.99) | 0.26 | 0.98 (0.96–1.01) |
| Multivariable model 2 | 1.01 (0.95–1.07) | 0.95 (0.89–1.00) | 1 [Reference] | 0.97 (0.92–1.03) | 0.91 (0.86–0.97) | 0.04 | 0.97 (0.94–0.99) |
| DASH | |||||||
| Multivariable model 1 | 1.07 (1.01–1.14) | 0.98 (0.93–1.04) | 1 [Reference] | 1.00 (0.95–1.06) | 1.03 (0.97–1.10) | 0.75 | 0.99 (0.91–1.07) |
| Multivariable model 2 | 1.08 (1.02–1.15) | 0.99 (0.94–1.05) | 1 [Reference] | 1.00 (0.94–1.05) | 0.99 (0.93–1.05) | 0.21 | 0.96 (0.89–1.03) |
Data are based on 24 years of follow –up (1986–2010). The exposure was changes in diet quality scores in the first 4 years period (1986–1990) and the outcome was the incidence of CVD in the subsequent follow-up years (1990–2010). Model 1 and 2 were adjusted for the some covariates as in Table 2.
The 20-percentile increase in each score was calculated from the median value of each quintile.
The joint analysis using initial and subsequent 4-year diet quality scores by tertiles, showed that compared with participants who had the lowest adherence to diet scores at baseline and four years later, those with the poorest adherence at baseline but made the largest improvements to diet quality in the following period had a significant 15% lower CVD risk for both the AHEI (95%CI: 2%–25%) and the AMED (95% CI: 4%–25%). Having the highest adherence at baseline and maintaining this adherence over four years later was associated with 13%–16% lower CVD risk (pooled HR [95% CI]: AHEI, 16% [11%–21%]; AMED, 13%[7%–18%]; DASH, 13%[7%–18%]) (Supplemental Figure 2).
After multiple testing corrections, no significant effect modification was found for current smoking status, presence of hypertension, hypercholesterolemia or diabetes, baseline diet quality scores, physical activity, baseline BMI, and changes in weight.
In sensitivity analyses, after removing the alcohol component from the AHEI and the AMED scores, the association between the highest quintile of change in the diet scores and CVD risk was attenuated; however, there was a significant inverse linear trend in CVD risk across quintiles of change (Supplemental Table 6).
Discussion
In two large prospective cohorts of men and women, we found that an increase in diet quality as assessed by three scores over 4-year periods was associated with lower CVD risk in the following four years, independent of baseline diet quality score or changes in other lifestyle factors. This short-term association persisted in the long term for the AHEI and the AMED. In addition, a decrease in the DASH score was associated with higher CVD risk in the long term. The CVD risk was more pronounced for a reduction in any of the three diet quality scores in longer-term than in short-term follow up.
Although the three diet quality scores analyzed in this study were developed for slightly different purposes, and vary in the food and/or nutrient components and optimal cutoffs, they all capture essential elements of a high-quality diet. This is evidenced by the strong correlations between the scores observed in our study, as well as the consistent magnitudes of association with CVD. Existing evidence supports the association between the adherence to these three diet quality scores and lower risk of chronic diseases, including CVD 4–6. The AHEI-2010 has been associated with major chronic disease risk 7, 26, 27, including an inverse association with CVD risk 4. Further observational studies have shown that a higher AHEI score was associated with a reduction of 18%–26% for CVD death in different populations 8–10, 14. As for the AMED, the results from previous meta-analyses of prospective studies have reported an inverse association between the Mediterranean diet and CVD risk 11, 12 – for each two additional points in the AMED, there was a 13% (95%CI: 10%–15%) significant reduction in cardiovascular events 12. Analyzing the DASH and CVD incidence, another meta-analysis for six cohort studies with a total of 260,011 adults showed that higher adherence to a DASH-style diet was associated with a 20-percentile decrease in risk of CVD (95% CI: 24%–36%) 28. The findings in our current study extend the previous results and also demonstrate that changes in diet quality scores are associated with both short- and long-term CVD risk.
To translate our results to the general population and better compare the scores, we reported that a 20-percentile change in any of the three diet quality scores every four years was associated with a significantly lower CVD risk in the following four years (9% for the AHEI, 3% for the AMED, and 6% for the DASH). As an example, a person increasing 22 points out of 110 for the AHEI-2010 score (e.g., increasing nuts and legumes up to 1 s/day and not consuming sugar sweetened beverages and fruit juice daily) over a 4-year period could lower the CVD risk by 9% in the subsequent four years independently of the initial diet quality score and other lifestyles. The slight variations in the strength of the association between the three scores and CVD risk may be due to the differences in the scoring system. When evaluating changes over time, the wider categorizations such as those used to define the AMED and the DASH may narrow the range of possible scores and likely limit the detection of differences among groups. In contrast, the continuous scale of the AHEI has the advantage of increasing the sensitivity and therefore the ability to detect changes over time.
When we studied CHD separately, the subsequent 4-year CHD risk had similar significant association as CVD with 20-percentile changes in diet quality scores for the AHEI and the DASH, but not for the AMED. Although the three diet scores share common healthy components, the AHEI and the DASH have specific variables more relevant to Western diets (e.g., sugar-sweetened beverages, red and processed meat, and sodium intake) than the AMED, which could explain the different results. With regard to stroke, we found an association between the 20-percentile change in the AMED and the DASH scores, but not with the AHEI. The DASH diet was first tested in randomized control trials to reduce blood pressure 29, a major risk factor for stroke 30. Moderate alcohol consumption that has been consistently associated with reduced risk of CHD 31 was included in the AHEI and the AMED, but not in the DASH because the association for blood pressure was not as clear 32. In sensitivity analysis, removing the alcohol component from the AHEI and the AMED attenuated the association to null in the AHEI for the highest quintile of change for CVD and CHD risk. However, for stroke, the association did not change. These results were consistent with our previous findings that moderate alcohol intake was an important contributor to cardiovascular mortality 8.
Our study showed that the association between increasing diet quality and lower CVD risk in the short term (4 years) persists in the long term. Using the first 4-year changes diet quality in the analysis, we found that increasing the AHEI and the AMED scores were associated with significant lower CVD risk at 7% and 9% in the long term, respectively. On the contrary, the decrease in the DASH score was associated with an 8% increase in CVD risk. The increase of CVD was greater (14–16%) with a decrease in diet quality in longer period (16 years) suggesting that this association is more pronounced in longer-term than short-term follow up. The results with CVD risk were consistently seen in CHD risk, but not in stroke. Although Fung et al 6 found an inverse association between the DASH and risk of stroke, other studies could not replicate this association 15, 16, 28, suggesting that the DASH diet may not offer long-term benefit for stroke risk 16. In addition, it is possible that DASH is harder to capture with a FFQ since sodium is more difficult to measure.
Joint analysis of initial and subsequent 4-year diet scores confirmed that maintaining consistently high adherence to diet quality scores was related to lower CVD risk as compared with a consistently low adherence to diet scores. In addition, our results underscore that middle age participants with low adherence to diet quality scores in the beginning can still reduce their CVD risk by improving their diet quality.
The strengths of this study include a prospective population-based design, a large sample size, a long follow up, and repeated validated dietary and lifestyle data. The consistent results from these longitudinal cohorts combined with randomized control trials data (e.g., the PREDIMED Study 13 and the Lyon Diet Heart Study 33) support diet quality improvements as an important strategy to prevent CVD. Finally, looking at short- and long-term diet changes, our analysis allows for a more appropriate and practical evaluation of how dynamic and realistic changes in diet quality are associated with CVD risk.
Limitations need to be considered in our study. First, because dietary information was self-reported, measurement error and misclassification was inevitable. This error is likely to be substantially greater than in our typical prospective analyses because in the assessment of change both the baseline and follow-up questionnaire contribute to error, whereas in our usual prospective analysis error is dampened by using the average of repeated questionnaires34. Due to the prospective design, misclassification and measurement error was most likely non-differential, but attenuation of associations toward the null is likely large. Because AMED or the DASH score assignment depends on the population relative intake, when applied to different populations, individuals with the same score may have different intake of each component. Second, our study population, consisting of white nurses and health professionals, could limit the generalizability of the results. Third, residual confounding is always a concern in observational studies; however, we were able to adjust for measured time-dependent confounders. Finally, despite being able to adjust for diabetes, hypertension, and hypercholesterolemia, other potential clinical conditions could modify diet and confound the associations.
Conclusion
In this large study of US men and women, improving adherence to diet quality scores over time is associated with significantly lower risk of CVD, both in the short term and the long term. The increase in CVD with a reduction in diet quality is more pronounced in longer-term than short-term follow up. Our results provide further evidence that modest improvement in diet quality over time confers benefits for CVD prevention.
Supplementary Material
Acknowledgments
Funding Sources: The study was supported by research grants UM1 CA186107, UM1 CA167552, HL60712, P01 CA87969, P01 CA055075, R01 HL034594, R01 HL088521, R01 HL35464 from the National Institutes of Health. Sotos-Prieto M was supported by a research fellowship from Fundación Alfonso Martín Escudero (FAME), Spain.
Footnotes
Disclosures: None.
References
- 1.Eckel RH, Jakicic JM, Ard JD, de Jesus JM, Houston Miller N, Hubbard VS, Lee IM, Lichtenstein AH, Loria CM, Millen BE, Nonas CA, Sacks FM, Smith SC, Svetkey LP, Wadden TA, Yanovski SZ. 2013 aha/acc guideline on lifestyle management to reduce cardiovascular risk. J Am Coll Cardiol. 2014;63:2960–2984. doi: 10.1016/j.jacc.2013.11.003. [DOI] [PubMed] [Google Scholar]
- 2.Scientific report of the 2015 dietary guidelines advisory committee (advisory report) 2015 Feb; [Google Scholar]
- 3.Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, Dai S, Ford ES, Fox CS, Franco S, Fullerton HJ, Gillespie C, Hailpern SM, Heit JA, Howard VJ, Huffman MD, Judd SE, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Mackey RH, Magid DJ, Marcus GM, Marelli A, Matchar DB, McGuire DK, Mohler ER, 3rd, Moy CS, Mussolino ME, Neumar RW, Nichol G, Pandey DK, Paynter NP, Reeves MJ, Sorlie PD, Stein J, Towfighi A, Turan TN, Virani SS, Wong ND, Woo D, Turner MB. Heart disease and stroke statistics--2014 update: A report from the american heart association. Circulation. 2014;129:e28–e292. doi: 10.1161/01.cir.0000441139.02102.80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chiuve SE, Fung TT, Rimm EB, Hu FB, McCullough ML, Wang M, Stampfer MJ, Willett WC. Alternative dietary indices both strongly predict risk of chronic disease. J Nutr. 2012;142:1009–1018. doi: 10.3945/jn.111.157222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Fung TT, Rexrode KM, Mantzoros CS, Manson JE, Willett WC, Hu FB. Mediterranean diet and incidence of and mortality from coronary heart disease and stroke in women. Circulation. 2009;119:1093–1100. doi: 10.1161/CIRCULATIONAHA.108.816736. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu FB. Adherence to a dash-style diet and risk of coronary heart disease and stroke in women. Arch Intern Med. 2008;168:713–720. doi: 10.1001/archinte.168.7.713. [DOI] [PubMed] [Google Scholar]
- 7.Akbaraly TN, Ferrie JE, Berr C, Brunner EJ, Head J, Marmot MG, Singh-Manoux A, Ritchie K, Shipley MJ, Kivimaki M. Alternative healthy eating index and mortality over 18 y of follow-up: Results from the whitehall ii cohort. Am J Clin Nutr. 2011;94:247–253. doi: 10.3945/ajcn.111.013128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Li S, Chiuve SE, Flint A, Pai JK, Forman JP, Hu FB, Willett WC, Mukamal KJ, Rimm EB. Better diet quality and decreased mortality among myocardial infarction survivors. JAMA Intern Med. 2013;173:1808. doi: 10.1001/jamainternmed.2013.9768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Mursu J, Steffen LM, Meyer KA, Duprez D, Jacobs DR. Diet quality indexes and mortality in postmenopausal women: The iowa women’s health study. Am J Clin Nutr. 2013;98:444–453. doi: 10.3945/ajcn.112.055681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Reedy J, Krebs-Smith SM, Miller PE, Liese AD, Kahle LL, Park Y, Subar AF. Higher diet quality is associated with decreased risk of all-cause, cardiovascular disease, and cancer mortality among older adults. J Nutr. 2014;144:881–889. doi: 10.3945/jn.113.189407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Sofi F, Abbate R, Gensini GF, Casini A. Accruing evidence on benefits of adherence to the mediterranean diet on health: An updated systematic review and meta-analysis1. Am J Clin Nutr. 2010;92:1189–1196. doi: 10.3945/ajcn.2010.29673. [DOI] [PubMed] [Google Scholar]
- 12.Martinez-Gonzalez MA, Bes-Rastrollo M. Dietary patterns, mediterranean diet, and cardiovascular disease. Curr Opin Lipidol. 2014;25:20–26. doi: 10.1097/MOL.0000000000000044. [DOI] [PubMed] [Google Scholar]
- 13.Estruch R, Ros E, Salas-Salvado J, Covas MI, Corella D, Aros F, Gomez-Gracia E, Ruiz-Gutierrez V, Fiol M, Lapetra J, Lamuela-Raventos RM, Serra-Majem L, Pinto X, Basora J, Munoz MA, Sorli JV, Martinez JA, Martinez-Gonzalez MA. Primary prevention of cardiovascular disease with a mediterranean diet. N Engl J Med. 2013;368:1279–1290. doi: 10.1056/NEJMc1806491. [DOI] [PubMed] [Google Scholar]
- 14.George SM, Ballard-Barbash R, Manson JE, Reedy J, Shikany JM, Subar AF, Tinker LF, Vitolins M, Neuhouser ML. Comparing indices of diet quality with chronic disease mortality risk in postmenopausal women in the women’s health initiative observational study: Evidence to inform national dietary guidance. Am J Epidemiol. 2014;180:616–625. doi: 10.1093/aje/kwu173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Agnoli C, Krogh V, Grioni S, Sieri S, Palli D, Masala G, Sacerdote C, Vineis P, Tumino R, Frasca G, Pala V, Berrino F, Chiodini P, Mattiello A, Panico S. A priori-defined dietary patterns are associated with reduced risk of stroke in a large italian cohort. J Nutr. 2011;141:1552–1558. doi: 10.3945/jn.111.140061. [DOI] [PubMed] [Google Scholar]
- 16.Folsom A, Parker E, Harnack L. Degree of concordance with dash diet guidelines and incidence of hypertension and fatal cardiovascular disease. Am J Hypertens. 2007;20:225–232. doi: 10.1016/j.amjhyper.2006.09.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Colditz GA, Manson JE, Hankinson SE. The nurses’ health study: 20-year contribution to the understanding of health among women. J Womens Health. 1997;6:49–62. doi: 10.1089/jwh.1997.6.49. [DOI] [PubMed] [Google Scholar]
- 18.Rimm EB, Giovannucci EL, Willett WC, Colditz GA, Ascherio A, Rosner B, Stampfer MJ. Prospective study of alcohol consumption and risk of coronary disease in men. Lancet. 1991;338:464–468. doi: 10.1016/0140-6736(91)90542-w. [DOI] [PubMed] [Google Scholar]
- 19.Mozaffarian D, Hao T, Rimm EB, Willett WC, Hu FB. Changes in diet and lifestyle and long-term weight gain in women and men. N Engl J Med. 2011;364:2392–2404. doi: 10.1056/NEJMoa1014296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.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. doi: 10.1093/oxfordjournals.aje.a114086. [DOI] [PubMed] [Google Scholar]
- 21.Salvini S, Hunter DJ, Sampson L, Stampfer MJ, Colditz GA, Rosner B, Willett WC. Food-based validation of a dietary questionnaire: The effects of week-to-week variation in food consumption. Int J Epidemiol. 1989;18:858–867. doi: 10.1093/ije/18.4.858. [DOI] [PubMed] [Google Scholar]
- 22.Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a mediterranean diet and survival in a greek population. N Engl J Med. 2003;348:2599–2608. doi: 10.1056/NEJMoa025039. [DOI] [PubMed] [Google Scholar]
- 23.Rose GA. Cardiovascular survey methods. Geneva Albany, N.Y: World Health Organization; WHO Publications Centre distributor; 1982. [Google Scholar]
- 24.Walker AE, Robins M, Weinfeld FD. The national survey of stroke. Clinical findings. Stroke. 1981;12:I13–44. [PubMed] [Google Scholar]
- 25.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. doi: 10.1093/oxfordjournals.aje.a114319. [DOI] [PubMed] [Google Scholar]
- 26.McCullough ML, Feskanich D, Stampfer MJ, Giovannucci EL, Rimm EB, Hu FB, Spiegelman D, Hunter DJ, Colditz GA, Willett WC. Diet quality and major chronic disease risk in men and women: Moving toward improved dietary guidance. Am J Clin Nutr. 2002;76:1261–1271. doi: 10.1093/ajcn/76.6.1261. [DOI] [PubMed] [Google Scholar]
- 27.Fung TT, McCullough ML, Newby PK, Manson JE, Meigs JB, Rifai N, Willett WC, Hu FB. Diet-quality scores and plasma concentrations of markers of inflammation and endothelial dysfunction. Am J Clin Nutr. 2005;82:163–173. doi: 10.1093/ajcn.82.1.163. [DOI] [PubMed] [Google Scholar]
- 28.Salehi-Abargouei A, Maghsoudi Z, Shirani F, Azadbakht L. Effects of dietary approaches to stop hypertension (dash)-style diet on fatal or nonfatal cardiovascular diseases--incidence: A systematic review and meta-analysis on observational prospective studies. Nutrition. 2013;29:611–618. doi: 10.1016/j.nut.2012.12.018. [DOI] [PubMed] [Google Scholar]
- 29.Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM, Bray GA, Vogt TM, Cutler JA, Windhauser MM, Lin PH, Karanja N. A clinical trial of the effects of dietary patterns on blood pressure. Dash collaborative research group. N Engl J Med. 1997;336:1117–1124. doi: 10.1056/NEJM199704173361601. [DOI] [PubMed] [Google Scholar]
- 30.Marsh JD, Keyrouz SG. Stroke prevention and treatment. J Am Coll Cardiol. 2010;56:683–691. doi: 10.1016/j.jacc.2009.12.072. [DOI] [PubMed] [Google Scholar]
- 31.Lichtenstein AH, Appel LJ, Brands M, Carnethon M, Daniels S, Franch HA, Franklin B, Kris-Etherton P, Harris WS, Howard B, Karanja N, Lefevre M, Rudel L, Sacks F, Van Horn L, Winston M, Wylie-Rosett J. Diet and lifestyle recommendations revision 2006: A scientific statement from the american heart association nutrition committee. Circulation. 2006;114:82–96. doi: 10.1161/CIRCULATIONAHA.106.176158. [DOI] [PubMed] [Google Scholar]
- 32.Williams B. The year in hypertension. J Am Coll Cardiol. 2008;51:1803–1817. doi: 10.1016/j.jacc.2008.03.010. [DOI] [PubMed] [Google Scholar]
- 33.de Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J, Mamelle N. Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: Final report of the lyon diet heart study. Circulation. 1999;99:779–785. doi: 10.1161/01.cir.99.6.779. [DOI] [PubMed] [Google Scholar]
- 34.Hu FB, Stampfer MJ, Rimm E, Ascherio A, Rosner BA, Spiegelman D, Willett WC. Dietary fat and coronary heart disease: A comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Am J Epidemiol. 1999;149:531–540. doi: 10.1093/oxfordjournals.aje.a009849. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.