Increase Font Size Decrease Font Size View as PDF Print

What is the relationship between adherence to dietary guidelines/recommendations or specific dietary patterns, assessed using an index or score, and risk of cardiovascular disease?

Conclusion

There is strong and consistent evidence that in healthy adults increased adherence to dietary patterns scoring high in fruits, vegetables, whole grains, nuts, legumes, unsaturated oils, low-fat dairy, poultry and fish; low in red and processed meat, high-fat dairy, and added sugars; and moderate in alcohol is associated with decreased risk of fatal and non-fatal cardiovascular diseases, including coronary heart disease and stroke.
 

Grade

I – Strong

 

Key Findings:

  • Three major categories of dietary pattern scores were identified related to cardiovascular disease (CVD) risk: dietary exposure based on adherence to (1) a Mediterranean dietary pattern, (2) dietary guidelines recommendations, or (3) a DASH diet.
  • The preponderance of the evidence from studies carried out in large, well-characterized prospective cohorts from the United States, Europe, Japan, and Australia showed that, in healthy adults, an increase in a Mediterranean diet score or dietary guidelines-related score was associated with decreased risk of fatal and non-fatal CVD, defined as coronary heart disease (CHD) and stroke, as well as decreased risk of CHD and stroke as individual clinical outcomes. Fewer studies assessed the association between adherence to a DASH diet and CVD, CHD, or stroke outcomes, using an index or score, and their findings were inconsistent.
  • Scores that were most frequently associated with decreased risk of CVD, CHD, or stroke, in categorical comparisons of adherence, were the original Mediterranean Diet Score (MDS), the Alternate Mediterranean Diet Score (aMed), the Healthy Eating Index (HEI)-2005, the Alternate HEI (AHEI) and updated AHEI-2010, the Recommended Food Score (RFS), and one of the DASH scores.
  • Positive food components of scores that were associated with decreased CVD risk were fruits, vegetables, whole grains, nuts, legumes, unsaturated fats, and fish. Alcohol was included as a positive component when consumed in moderation, but not in all scores. Red and processed meats were negative components in the Mediterranean scores, AHEI scores, and DASH; whereas, poultry was included as a positive component in the original AHEI and RFS scores. Total high-fat dairy was a negative component in the MDS, but dairy was a positive component when meeting recommended intakes for the HEI-2005 or as low-fat dairy in the RFS and DASH scores. Certain scores also included sugars or sugar-sweetened beverages as negative components.
  • Studies that assessed the association between individual food components of scores and CVD risk were consistent with the identified food components from comparisons across predictive scores.
  • A smaller number of studies examined intermediate, secondary outcomes and other individual clinical endpoints outcomes with mixed results.

Evidence Summary Overview

Description of the Evidence

A total of 55 studies met the inclusion criteria for this systematic review on dietary patterns and incident cardiovascular disease (CVD) outcomes. The body of evidence consisted primarily of prospective cohort studies; 52 studies were prospective cohort studies and the remaining studies were randomized controlled trials (RCTs). In terms of study quality, 46 of the studies were of positive quality and 9 were of neutral quality. The studies were carried out between the years 2000 and 2013. The sample sizes of the RCTs were from 187 to 7,447 subjects. The sample sizes of prospective cohort studies ranged from 373 to as many as 380,296 participants (3 studies <1,000, 16 studies >1,000, 29 studies >10,000, and 4 studies >100,000). RCT duration ranged from 3 months to 4.8 years and prospective cohort study follow-up times ranged from 2 to 40 years. Several of the prospective cohort studies were of long duration, with 7 studies with greater than 20 years follow-up. The majority of these studies measured hard clinical endpoint outcomes.
 
Population: The prospective cohort studies were primary prevention studies of general populations and most studies were conducted with healthy adults who were free of CVD, coronary heart disease (CHD), hypertension (HTN), or diabetes. However, a few studies included participants with chronic diseases such as CVD, HTN, or diabetes, within a larger cohort of otherwise healthy individuals (Akbaraly, 2011; Kant, 2000; Nilsson, 2012; Russell, 2012; Tognon, 2012). The three RCTs were conducted in adults with elevated chronic disease risk: two studies reported on the Prevencion con Dieta Mediterranea (PREDIMED) trial of older adults at increased CVD risks (Estruch, 2006 and 2013) and one study reported on men with metabolic syndrome (Jacobs, 2009). In prospective cohort studies with adult participants, age ranges spanned from 18 to 97 years. Two studies focused on young adults with an age range of 18 to 30 years from the Coronary Artery Risk Development in Young Adults Study (CARDIA) cohort (Steffen, 2005; Zamora, 2011). One study was conducted specifically in the elderly with an age range of 70 to 90 years (Knoops, 2004). Only one study examined children; a longitudinal growth and health study examined boys and girls with a mean age of 13 years at baseline (van der Laar, 2012). Several studies examined only women (Belin, 2011; Chiuve, 2011; Fitzgerald, 2012; Fung, 2008 and 2009; Levitan, 2009a; McCullough, 2000b; Michels and Wolk, 2002) and one study focused specifically on post-menopausal women (Belin, 2011). Other studies examined only men (Fidanza, 2004; Jacobs, 2009; Kaluza, 2009; Levitan, 2009b; McCullough, 2000a; Menotti, 2012; Sjögren, 2010). Some studies that examined men and women assessed health outcomes separately by gender (Buckland, 2011; Dauchet, 2011; Dilis, 2012; Drake, 2012; McCullough, 2002; Mitrou, 2007; Nilsson, 2012; Oba, 2009; Seymour, 2003; Tognon, 2012; von Ruesten, 2010). Only four studies specifically identified the race/ethnic subgroups of their cohort. Two studies with the CARDIA cohort examined equal numbers of Black and White young adults (Steffen, 2005; Zamora, 2010). One study examined Hispanic, Black, and White urban Americans (the Northern Manhattan Study) (Gardener, 2011) and one study examined Whites, Blacks, and South Asians in the United Kingdom (Whitehall II study) (Akbaraly, 2011).
 
Taken together, studies were conducted in the United States, Europe, Japan, and Australia and included many large, well-characterized cohorts. Several studies analyzed the same cohorts; however, all of the included studies presented unique data related to the association between dietary patterns and CVD.
  • Nineteen studies were conducted in the United States with the following cohorts: the Nurses’ Health Study (Chiuve, 2011 and 2012; Fung, 2008 and 2009; McCullough, 2000b and 20002), the Health Professionals Follow-Up Study (Chiuve, 2012; McCullough, 2000a and 2002), the Women’s Health Initiative (Belin, 2011; Fitzgerald, 2012), the CARDIA Study (Steffen, 2005; Zamora, 2011), the Iowa Women’s Health Study (Folsom, 2007), the Framingham Offspring and Spouse Study (Rumawas, 2009), the NIH AARP Diet and Health Study (Mitrou, 2007), the American Cancer Society Cancer Prevention Study II (Seymour, 2003), the Breast Cancer Detection and Demonstration Project (Kant, 2000), and the Northern Manhattan Study (Gardener, 2011).
  • Nine studies were conducted in Spain with the following cohorts: the Seguimiento Universidad de Navarra Study (Martínez-González, 2011 and 2012; Núñez-Córdoba, 2009; Toledo, 2010; Tortosa, 2007), EPIC-Spain (Buckland, 2009 and 2011), and the Prevencion con Dieta Mediterranea Study (Estruch, 2006 and 2013).
  • Eight studies were conducted in Sweden with the following cohorts: the Swedish Mammography Cohort (Levitan, 2009a; Michels and Wolk, 2002), the Cohort of Swedish Men (Kaluza, 2009; Levitan, 2009b), the Vasterbotten Intervention Program (Nilsson, 2012; Tognon, 2012), the Uppsala Longitudinal Study of Adult Men (Sjögren, 2010), and the Malmö Diet and Cancer Cohort (Drake, 2012).
  • Four studies were conducted in Greece with the following cohorts: EPIC-Greece (Dilis, 2012; Misirli, 2012; Trichopoulou, 2003) and the ATTICA study (Pangiotakos, 2008).
  • Two studies were conducted in Italy, EPIC-Italy (Agnoli, 2011) and the Italian Rural Areas of the Seven Countries Study (Menotti, 2012). Two studies were conducted in France, both with the SU.VI.MAX cohort (Dauchet, 2007; Kesse-Guyot, 2011). Two studies were conducted in Norway, the Oslo Diet and Exercise Study (Jacobs, 2009) and the Tromso Study (Hansen-Krone, 2012). Two studies were conducted in the Netherlands, EPIC-Netherlands (Hoevenaar-Blom, 2012) and the Amsterdam Growth and Health Longitudinal Study (van der Laar, 2012). And two studies were conducted in Japan, including the National Integrated Project for Prospective Observation of Non-Communicable Diseases and Trends in the Aged (Nakamura, 2009) and the Takayama Study (Oba, 2009).
  • The remaining countries were represented in only one study: Germany (EPIC-Potsdam; von Ruesten, 2010), Denmark (WHO MONICA Study; Osler, 2012), Portugal (EPIPorto; Camoes, 2010), the United Kingdom (Whitehall II; Akbaraly, 2011), and Australia (Blue Mountain Eye Study; Russell, 2012).
  • Lastly, two large multi-country studies were included in this body of evidence: the Seven Countries Study (United States, Finland, Italy, Greece, Japan, Yugoslavia, and Serbia) (Fidanza, 2004) and Healthy Ageing: a Longitudinal study in Europe (HALE), comprised of individuals from the Survey in Europe on Nutrition and the Elderly: a Concerned Action (SENECA) and the Finland, Italy, the Netherlands Elderly (FINE) studies. In combination, the latter analysis included 11 European countries (Knoops, 2004).
 
Dietary Exposure: Methodologically, diet exposure was assessed by adherence to a hypothesis-driven dietary pattern, defined using a numerical scoring system. Three major categories of a priori dietary patterns were identified: (1) dietary pattern exposure based on a Mediterranean diet, (2) dietary pattern exposure based on dietary guidelines recommendations, and (3) dietary pattern exposure based on a DASH diet. The most common method of assessment of dietary intake for determining dietary pattern scores was food frequency questionnaires (FFQs), commonly validated for foods in the respective locations of the population of study. Many prospective cohort studies assessed dietary intake only at baseline.
  • Twenty-three studies examined health outcomes related to adherence to a Mediterranean-style dietary pattern. Of these studies, 12 studies used the Mediterranean Diet Score (MDS) 1 of Trichopoulou (Agnoli, 2011; Dilis, 2012; Gardener, 2011; Hoevenaar-Blom, 2012; Martínez-González, 2011 and 2012; Misirli, 2012; Mitrou, 2007; Núñez-Córdoba, 2009; Sjögren, 2010; Trichopoulou, 2003; Tortosa 2007). Four studies used the alternate Mediterranean Diet Score (aMed) (Fung, 2009; Mitrou, 2007; van der Laar, 2012; Chiuve, 2011). Two RCTs used the authors’ Mediterranean diet score based on a version of the MDS that assessed a cardio-protective Mediterranean diet (Estruch, 2006 and 2013). Numerous other Mediterranean diet scores were represented in this body of evidence including the relative Mediterranean diet score (rMED) (Buckland, 2009 and 2011), the modified MDS (MMDS) (Tognon, 2012; Knoops, 2004), the Mediterranean Adequacy Index (MAI) (Fidanza, 2004; Menotti, 2012), the Mediterranean Style Diet Pattern Score (MSDPS) (Rumawas, 2009), the MedDietScore (Pangiotakos, 2008), and the Italian Mediterranean Index (Agnoli, 2011).
  • Seventeen studies examined health outcomes related to adherence to dietary guidelines recommendations, including the United States, French, German, Swedish, and Japanese dietary guidelines recommendations. Of these studies, the largest number of studies assessed adherence to a version of the Healthy Eating Index (HEI). Four studies used the alternate HEI (AHEI) or updated AHEI-2010 (Belin, 2011; Chiuve, 2012; McCullough, 2002; Akbaraly, 2011). Two studies used the HEI-2005 (Agnoli, 2011; Chiuve, 2012) and two studies used HEI-f, a version of the original HEI adapted to the authors’ FFQs (McCullough, 2000a and 2000b). Several studies used the Recommended Food Score (RFS) in different versions, also including a non-RFS in some studies (Kaluza, 2009 [and non-RFS]; Kant, 2000; McCullough, 2002; Michels and Wolk, 2002 [and non-RFS]).  Different versions of a Diet Quality Index (DQI) were used including the DQI-2005 (Zamora, 2011), the DQI of Patterson (Seymour, 2003), and DQI-Swedish Nutrition Recommendations (DQI-SNR) (Drake, 2012). Two studies used a Japanese score including the Japanese Food Guide Spinning Top (Oba, 2009) and the Reduced Salt Japanese Diet Score (Nakamura, 2009). Lastly, one study used the French Programme National Nutrition Santé Guideline Score (PNNS-GS) (Kesse-Guyot, 2011) and one study used the German Food Pyramid (von Reusten, 2010).
  • Eight studies examined health outcomes related to adherence to a DASH diet. Of these studies, five studies used the DASH score of Fung (Agnoli, 2011; Fitzgerald, 2012; Fung, 2008; Levitan, 2009a and 2009b) and two used the DASH score of Folsom (Folsom, 2007; Levitan, 2009a). One study (Dauchet, 2007) used two scores adapted from the DASH eating plan of Appel (Appel, 1997) and one study (Camoes, 2010) used the DASH goals of Sacks (Sacks, 1995 and 2001). Finally, one study (Levitan, 2009a) used DASH scores based on NHLBI food and nutrient recommendations.
  • Five studies examined health outcomes comparing the association with two or more dietary pattern scores, including comparisons across Mediterranean, dietary guidelines, DASH, and other customized scores (Agnoli, 2011 [DASH, MDS, Italian MDS, HEI]; Belin, 2011 [AHEI, DMI]; McCullough, 2002 [AHEI, RFS]; Sjögren, 2010 [HDI, MDS]; and Toledo, 2010 [DASH, UMMDS, etc.]). Toledo, in particular, assessed a large number of scores across Mediterranean diet and dietary guidelines-related scores. Three studies measured different versions of related scores (Chiuve, 2012 [HEI, AHEI]; Levitan, 2009a [DASH]; Mitrou, 2007 [MDS, aMed]).
  • Seven studies examined health outcomes associated with the authors’ own a priori diet scores: Hansen-Krone (2012), Jacobs (2009), Nilsson (2012), Osler (2002), Russell (2012), Sjögren (2010), and Steffen (2005).

Qualitative Synthesis of the Collected Evidence

Themes and Key Findings for Total Scores

Health Outcomes: The studies in this body of evidence were subdivided into two broad categories based on (I) endpoint clinical outcomes and (II) intermediate (CVD risk factor) outcomes. Within the first category, there were five designated subcategories: studies that examined (1) cardiovascular disease (CVD), (2) coronary heart disease (CHD), (3) stroke, (4) myocardial infarction (MI), and (5) heart failure (HF). Within the second category, there were two designated subcategories: studies that examined (1) hypertension (HTN) and blood pressure (BP) and (2) blood lipids.
 
Endpoint Clinical Outcomes: Cardiovascular diseases are disorders, both fatal and non-fatal, that affect the heart and circulatory system. The two major subclasses of CVD are CHD and stroke, and CHD includes MI as a major event. In addition, HF, another prevalent disorder, is also included within CVD. For this systematic review, endpoint clinical outcomes were subdivided into CVD, CHD, stroke, MI, and HF; ranging from more inclusive to less inclusive cardiovascular disorders. Overall, the outcomes were most commonly identified using the 9th or 10th edition of the International Classification of Diseases (ICD-9 or ICD-10).
 
CVD Incidence and Mortality:
The CVD category included studies that assessed a composite measure of CVD as the primary or secondary outcome of the study, and overall 29 studies were included (table 4-B-I-2). Typically, studies that measured CVD as a primary outcome did not include all diseases of the circulatory system (i.e., the entire spectrum of disorders listed in the ICD for this category). Total CVD most commonly included CHD and stroke, both fatal and nonfatal events. CHD and stroke events, in addition to CVD, were also assessed individually in many of these studies. This category also included studies with the primary objective of assessing total mortality but that also assessed CVD mortality as a secondary outcome. These studies commonly assessed CVD as “diseases of the circulatory system” and included the complete spectrum, or a large spectrum, of disorders listed in the ICD for this category.
 
Seventeen studies examined the association between adherence to a dietary pattern and CVD incidence and/or mortality as primary outcomes. The studies were all prospective cohort studies with one exception, an RCT (PREDIMED trial) that reported endpoint clinical outcomes (Estruch, 2013 [Med Diet]). The prospective cohort studies were typically large studies with well-characterized cohorts representative of national populations from the United States, Europe, and Japan. Two major categories of a priori dietary patterns were identified in these cohorts: (1) Mediterranean-style dietary patterns and (2) dietary guidelines-related dietary patterns. Additionally, two studies examined adherence to a DASH dietary pattern and three studies examined adherence to the authors’ own a priori dietary pattern score.
 
Out of 17 studies that assessed CVD incidence and mortality, 14 studies reported a favorable association between increased adherence to a dietary pattern and risk of CVD (Belin, 2011 [AHEI, DMI]; Chiuve, 2012 [HEI, AHEI]; Estruch, 2013 [Med Diet]; Fitzgerald, 2012 [DASH]; Fung, 2009 [aMed]; Gardener, 2011 [MDS]; Hoevenaar-Blom, 2012 [MDS]; Kesse-Guyot, 2011 [PNNS-GS]; Knoops, 2004 [MDS]; Martínez-González, 2011 [MDS]; McCullough, 2000a [HEI-f]; McCullough, 2002 [AHEI, RFS]; Nakamura, 2009 [Japanese Diet Score]; Pangiotakos, 2008 [MedDietScore]). These studies had large sample sizes with generally healthy adults and long follow-up times. The long follow-up periods allowed for sufficient numbers of incident events and assessment of long-term associations between a given dietary pattern and CVD risk. The large cohorts in these studies were relatively homogeneous in terms of ethnicity and SES, which had the advantage of reducing confounding. In the prospective cohort studies, with few exceptions, common confounders were adjusted for, including age, sex, total energy, physical activity, BMI, smoking, and medications or supplements. Overall, in studies that reported favorable associations between a dietary pattern (based on categories of adherence) and CVD incidence or mortality, the decrease in risk of CVD ranged from 22 to 59 percent for increased adherence to a Mediterranean dietary pattern and from 20 to 44 percent for increased adherence to a dietary guidelines-related pattern. The one RCT reported that a Mediterranean diet (plus olive oil or nuts) decreased risk of CVD in high risk subjects, with an approximate 30 percent decrease in risk of major CVD events for both interventions, and the trial (PREDIMED) was stopped early for meeting benefit requirements (Estruch, 2013 [Med Diet]). One study found an association between adherence to a German dietary guidelines-related score in men, but not in women (von Ruesten 2010 [German Food Pyramid Index]), and one study of women reported that although there was a 14 percent decreased risk of CVD, it was not significant (McCullough, 2000b [HEI-f]). Lastly, although one study on DASH cited above found a 12-percent decrease in incident CVD risk in women (Fitzgerald, 2012 [DASH]), a second study reported no association between a DASH dietary pattern and CVD mortality in women (Folsom, 2007 [DASH]).
 
Twelve studies measured CVD mortality as a secondary outcome after total mortality. Of these studies, five studies reported a favorable association between adherence to a dietary pattern and CVD mortality (Akbaraly, 2011 [AHEI]; Buckland, 2011 [rMED]; Kaluza, 2009 [RFS]; Martínez-González, 2012 [MDS]; Mitrou, 2007 [MDS, aMed]).  Four studies reported different CVD mortality risk outcomes based on gender. Three of these studies found a favorable association between adherence to either a dietary guidelines-related score (Oba, 2009 [Japanese Food Guide Score]; Seymour, 2003 [DQI]) or a Mediterranean diet score (Tognon, 2012 [MMDS]) and CVD mortality in women, but not men. However, one study found a favorable association in men, but not in women (Drake, 2012 [DQI-SNR]). Three studies reported no association between dietary pattern and CVD mortality (Nilsson, 2012 [Traditional Sami diet score]; Russell, 2012 [TDS]; Sjögren, 2010 [HDI, MDS]). 
 
Sub-analysis–Gender: Several studies examined only women; three of these found a favorable association between dietary pattern and CVD incidence or mortality in women (Belin, 2011 [AHEI, DMI]; Fitzgerald, 2012 [DASH]; Fung, 2009 [aMed]) (one study focused on post-menopausal women [Belin, 2011]). However, one study, although reporting a 14 percent decreased risk of CVD in women, found it was not significant (McCullough, 2000b [HEI-f]). Another study found no association in women (Folsom, 2007 [DASH]). Three studies examined only men and found either a favorable association (McCullough, 2000a [HEI-f]; Kaluza, 2009 [RFS]) or no association (Sjögren, 2010 [HDI, MDS]). Other studies examined men and women separately relative to CVD risk and found either a favorable association for both between adherence to a dietary pattern and CVD (McCullough, 2002 [AHEI, RFS]; Mitrou, 2007 [MDS, aMed]) or a gender difference with a favorable association for men, but not women (Drake, 2012 [DQI-SNR]; von Ruesten, 2010 [German Food Pyramid Index]) or a favorable association for women, but not men (Oba, 2009 [Japanese Food Guide Score]; Seymour, 2003 [DQI]; Tognon, 2012 [MMDS]). One study that assessed men and women separately found no association for either men or women (Nilsson, 2012 [Traditional Sami diet score]). Two studies assessed men and women together, as well as men and women separately, and found a favorable association in all cases between adherence to a dietary pattern and CVD mortality (Buckland, 2011 [rMED]) or incidence (Chiuve, 2012 [HEI, AHEI]). Taken together, although there are reported gender differences across this series of studies, the evidence is mixed and does not demonstrate a consistently more favorable association for men or women as individual subgroups. Many of the studies that found gender differences used country-specific dietary guidelines scores within their target populations, including the German Food Pyramid Index, the Japanese Food Guide Score, and the DQI adapted to the Swedish Nutrition Recommendations (DQI-SNR).
 
CHD Incidence and Mortality:
CHD was commonly defined as fatal and nonfatal CHD and nonfatal MI. This category included 16 studies that measured CHD incidence or mortality. Nine of these studies assessed diet exposure as adherence to a Mediterranean dietary pattern; whereas, fewer studies assessed other diet pattern score categories such as dietary guidelines-related scores (4), DASH scores (3), and an authors’ customized a priori diet score (1). Overall, the CHD studies included seven of the above studies that measured both CVD and CHD as independent outcomes, six studies that examined CHD incidence or mortality, three studies that measured CHD mortality as a secondary outcome after total mortality, and one study that examined sudden cardiac death (table 4-B-I-3). Overall, the CHD outcomes were most commonly identified by ICD-9 or ICD-10 codes.
 
Out of 13 studies that assessed CHD as a primary outcome, including non-fatal and fatal outcomes, 10 studies reported a favorable association between adherence to a dietary pattern and CHD risk (Buckland, 20009 [rMed]; Chiuve, 2011 [aMed] and 2012 [HEI, AHEI]; Dilis, 2012 [MDS]; Fidanza, 2004 [MAI]; Fung, 2008 [DASH] and 2009 [aMed]; Knoops, 2004 [MMDS]; Martínez-González, 2011 [MDS]; Menotti, 2012 [MAI]). Similar to the above studies that assessed CVD, these studies had large, relatively homogeneous populations of generally healthy adults, with long follow-up times. Overall, based on categories of adherence, the decreased risk of CHD ranged from 29 to 61 percent for increased adherence to a Mediterranean dietary pattern, from 24 to 31 percent for increased adherence to a dietary guidelines-related pattern, and from 14 to 27 percent for adherence to DASH. Common confounders that were adjusted for across these studies were age, sex, physical activity, BMI, and smoking. Of the above studies, one study that compared the HEI-2005 with the AHEI-2010 and adjusted each for respective models, found a favorable association between adherence to the AHEI, but not HEI, and risk of CHD with that further adjustment (Chiuve, 2012 [HEI, AHEI]). For CHD, the AHEI-2010 was more strongly associated with risk than the HEI-2005 (P-difference = 0.002). Three studies reported no association between adherence to a dietary pattern and risk of CHD (Fitzgerald, 2012 [DASH]; Folsom, 2007 [DASH]; Osler 2002 [Healthy Food Index]). In studies that examined CHD mortality secondary to total mortality, all three studies found a favorable association between dietary pattern adherence and CHD mortality, ranging from 33 to 53 percent decreased risk (Kant, 2000 [RFS]; Michels and Wolk, 2002 [RFS]; Trichopoulou, 2003 [MDS]).
 
Sub-analysis–Gender: Several studies examined only women. Five studies found a favorable association between dietary pattern and CHD incidence and mortality in women (Chiuve, 2011 [aMed]; Fung, 2008 [DASH]; Fung, 2009 [aMe]); Kant, 2000 [RFS]; Michels and Wolk, 2002 [RFS]); whereas, two studies found no association in women (Fitzgerald, 2012 [DASH]; Folsom, 2007 [DASH]). Two studies examined only men and found a favorable association between adherence to a Mediterranean diet and CHD risk in men (Fidanza, 2004 [MAI]; Menotti, 2012 [MAI]). Three studies assessed men and women together, as well as men and women separately, and found a favorable association in all cases between adherence to a dietary pattern and CHD mortality (Buckland, 2011 [rMED]; Dilis, 2012 [MDS]) or incidence (Chiuve, 2012 [HEI, AHEI]). Taken together, the studies that assessed CHD risk in men and women separately did not show gender differences, although two studies that only examined women found no association between a DASH dietary pattern and CHD. 
 
Stroke:
This category included 10 studies that examined stroke in addition to CVD or CHD (above), or stroke alone. In addition, three studies measured stroke mortality as a secondary outcome after total mortality. The studies were all prospective cohort studies with one exception, an RCT (PREDIMED) that reported endpoint clinical outcomes (Estruch, 2013 [Med Diet]). Stroke outcomes often included ischemic and hemorrhagic stroke, in some cases results were reported as total stroke, ischemic stroke, and hemorrhagic stroke (table 4-B-I-4). Out of 13 studies that assessed stroke, including non-fatal and fatal outcomes, 10 studies reported a favorable association between adherence to a dietary pattern and stroke risk (Agnoli, 2011 [DASH, MDS, Italian MDS, HEI]; Chiuve, 2012 [HEI, AHEI]; Estruch, 2013 [Med]; Fung, 2008 [DASH]; Fung, 2009 [aMed]; Hoevenaar-Blom, 2012 [MDS]; Kant, 2000 [RFS]; Michels and Wolk, 2002 [RFS]; Misirli, 2012 [MDS]; Nakamura, 2009 [Japanese Diet Score]). Similar to the above studies that assessed CVD, these studies had large sample sizes with generally healthy adults and long follow-up times with relatively homogeneous populations. Overall, the decreased risk of stroke ranged from 13 to 53 percent for increased adherence to a Mediterranean dietary pattern and from 14 to 60 percent for increased adherence to a dietary guidelines-related pattern. One study on the DASH diet showed a 17 percent decreased risk of stroke in women with increased adherence to the DASH score of Fung. Common confounders that were adjusted for across these studies were age, sex, total energy, physical activity, BMI, and smoking. The one RCT reported that a Mediterranean diet (plus olive oil or nuts) decreased risk of stroke in high risk subjects (Estruch, 2013 [Med Diet]). Two of the studies found differing results based on a comparison between two or more dietary pattern scores. One study found the AHEI-2010 was more predictive of stroke than HEI-2005 (Chiuve, 2012 [HEI, AHEI]). A second study compared four different scores and found that two Mediterranean diet scores (MDS and Italian MDS) and the DASH score of Fung were predictive of total stroke risk, but HEI-2005 was not (Agnoli, 2011 [DASH, MDS, Italian MDS, HEI]). Lastly, one study reported no association between dietary pattern adherence and risk of ischemic stroke (Gardener, 2011 [MDS]) and two studies reported no association with stroke death (Folsom, 2007 [DASH]; Tognon, 2012 [MMDS]).
 
Sub-analysis–Gender: Several studies examined only women; four of these found a favorable association between dietary pattern and stroke incidence or mortality in women (Fung, 2008 [DASH]; Fung, 2009 [aMed]; Kant, 2000 [RFS]; Michels and Wolk, 2002 [RFS]), but one study found no association (Folsom, 2007 [DASH]). None of the studies examined stroke risk only in men. One study assessed men and women separately and found no association between a Mediterranean diet and stroke mortality (Tognon, 2012 [MMDS]). One study assessed men and women together, as well as men and women separately, and found a favorable association in all cases between adherence to a dietary guidelines-related pattern and stroke incidence (Chiuve, 2012 [HEI, AHEI]). Taken together, these studies did not show gender differences relative to risk of stroke. 
 
Myocardial Infarction:
This category included four of the above studies that also examined fatal and non-fatal MI as an independent outcome and one study that measured MI mortality as a secondary outcome after total mortality (table 4-B-I-4). Two studies reported a favorable association between adherence to a Mediterranean diet and risk of MI (Gardener, 2011 [MDS]; Hoevenaar-Blom, 2012 [MDS]). One study reported a favorable association between adherence to a Mediterranean diet and risk of MI mortality in women, but not in men (Tognon, 2012 [MMDS]). Two studies found no association between a dietary pattern score and risk of MI (Hansen-Krone, 2012 [Smart Diet Score]) or acute MI death (Nakamura, 2009 [Japanese Diet Score]). 
 
Heart Failure:
This category included three studies that measured heart failure as the primary outcome (table 4-B-I-4). In two of these studies, heart failure included the ICD categories for heart failure and hypertensive heart failure and these studies found a favorable association between adherence to a DASH diet and risk of HF in men and women (Levitan, 2009a and 2009b [DASH]). The third study found a favorable association between the AHEI and the authors’ own Dietary Modification Index (DMI) and risk of HF in the Women’s Health Initiative study.

Intermediate Outcomes: Hypertension, Blood Pressure, and Blood Lipids
The intermediate outcomes and CVD risk factors included in this review were HTN, BP (systolic and diastolic), and blood lipids including LDL-C, HDL-Cl, the total-C: HDL-C and LDL-C: HDL-C ratios, and triglycerides. Relative to the large body of evidence for this question, only a few studies examined intermediate outcomes related to CVD risk.  
 
Hypertension and Blood Pressure:
Of the three studies that examined the association between adherence to a dietary pattern and HTN, two studies found no association (Folsom, 2007 [DASH]; Núñez-Córdoba, 2009 [MDS]) and one study found different results comparing multiple dietary pattern scores. Toledo et al. reported a higher DASH score was associated with decreased risk of HTN, but a higher updated, modified MDS (UMMDS) was associated with increased risk of HTN (Toledo, 2010 [DASH and UMMDS]).
 
Three studies that examined BP as a primary outcome found an association between increased DASH scores and decreased systolic and diastolic BP (SBP and DBP) over time (Dauchet, 2007 [DASH]; Steffen, 2005 [Food Index]; van der Laar, 2012 [aMed]). One of these studies also assessed men and women separately and found a favorable association for SBP in men, but not women (Dauchet, 2007 [DASH]). One study also assessed risk of elevated BP (EBP) and found an association between increased DASH score and decreased risk of EBP (Steffen, 2005 [Food Index]). An early report from a subsample of the PREDIMED trial, with subjects at elevated CVD risk, found the Mediterranean diet treatment groups (Med diet + olive oil or + nuts) had improved SBP and DBP measures (Estruch, 2006 [Med Diet]). On the other hand, a second RCT that examined men with metabolic syndrome found no effect of a diet that matched adherence to the authors’ a priori diet score and changes in SBP and DBP (Jacobs, 2009 [a priori diet score]).
 
Three studies examined BP as a component of metabolic syndrome. One study found an association between adherence to a dietary pattern and decreased SBP and DBP (Zamora, 2011 [DQI]). But two studies reported no association between adherence to a Mediterranean diet and BP (Rumawas, 2009 [MSDPS]; Tortosa, 2007 [MDS]).
 
Blood Lipids:
Of the three studies that examined blood lipids as primary outcomes, two of these studies were RCTs with populations at elevated risk. An early report from a subsample of the PREDIMED trial, with subjects at elevated CVD risk, found the Mediterranean diet treatment groups had favorable changes in blood HDL-C, the total-Cl: HLD-C ratio, and triglycerides (Estruch, 2006 ([Med Diet]). On the other hand, the second RCT, that examined men with metabolic syndrome, found no effect of a diet treatment that matched adherence to the authors’ a priori diet score and changes in LDL-C, HDL-C, or triglycerides, although there was a significant decrease in total cholesterol in the diet treatment group (Jacobs, 2009 [a priori diet score]). The third study in this group was a prospective cohort study that showed a favorable association between adherence to a Mediterranean diet and total and HDL-C (van der Laar, 2012 [aMed]).
 
Three studies examined blood lipids as components of metabolic syndrome. One study found an association between adherence to a dietary pattern and higher HDL-C levels (Zamora, 2011 [DQI]), one study reported adherence to a Mediterranean diet was associated with favorable changes in HDL-C and triglycerides (Rumawas, 2009 [MSDPS]), and one study found no association between adherence to a Mediterranean diet and blood lipids (Tortosa, 2007 [MDS]). 
 

Themes and Key Findings for Food Components of Scores

Food Components of Diet Pattern Scores

Twenty studies assessed the association between individual food components of a dietary pattern score and CVD endpoint outcomes (Akbaraly, 2011 [AHEI]; Belin, 2011 [AHEI, DMI]; Buckland, 2009 [rMed]; Camoes, 2010 [DASH]; Chiuve, 2012; Dauchet, 2007 [DASH]; Estruch, 2006 and 2013 [Med]; Fitzgerald, 2012 [DASH]; Folsom, 2007 [DASH[; Gardener, 2011 [MDS]; Hansen-Krone, 2012 [Smart Diet Score]; Hoevenaar-Blom, 2012 [MDS]; Martínez-González, 2011 [MDS]; McCullough, 2000a and 2000b [HEI-f]; Misirli, 2012 [MDS]; Núñez-Córdoba 2009 [MDS]; Seymour, 2003 [DQI]; Steffen, 2005 [Food Index]). Eight studies used components from Mediterranean diet scores (5 MDS, 1 rMed, 2 Med); four studies used components from DASH diets; six studies used components from dietary guidelines-related scores (2 AHEI, 1 AHEI-2010 and HEI-2005, 2 HEI-f, 1 DQI), and three studies used components from the authors’ customized diet pattern scores (1 Smart Diet Score, 1 Food Index, 1 Diet Modification Index).
 
Fruits, vegetables, and nuts (either individually or in combination as fruits & vegetables or fruits & nuts) were cited by nine studies as having a protective association with clinical outcomes, including CVD (Belin, 2011 [AHEI, DMI]; Estruch, 2013 [Med]; Fitzgerald, 2012 [DASH]; Martinez-Gonzalez, 2011 [MDS]; McCullough, 2000a [HEIf]; and Seymour, 2003 [DQI] only in women); CHD (Buckland, 2009 [rMed]); stroke (Estruch, 2013 [Med]; Misirli, 2012 [MDS]); and MI (Hansen-Krone, 2012 [Smart Diet]). One study specified dark-green and orange vegetables as having a protective association against CHD (Chiuve, 2012 [HEI, AHEI]). One study reported on nuts and soy as having a favorable association with risk of CVD (Akbaraly, 2011 [AHEI]). Four studies reported a favorable association between fruits, vegetables, and/or nuts and intermediate outcomes or CVD risk factors, including HTN (Camoes, 2010 [DASH]), BP (Dauchet, 2007 [DASH]), Estruch, 2006 [Med], Steffen, 2005 [Food Index]) or blood lipids (Estruch, 2006 [Med]). Legumes were reported in two studies as having a favorable association with either decreased risk of stroke (Misirli, 2012 [MDS]) or HTN (Núñez-Córdoba, 2009 [MDS]). Whole grains were reported in three studies as having a favorable association with CVD (McCullough, 2000a [(HEIf]), CHD (Chiuve, 2012 [HEI, AHEI], or BP (Steffen, 2005 [Food Index]).
 
Several studies reported a favorable association between olive oil (MUFA) or PUFA consumption and risk of CVD (Estruch, 2013 [Med]), CHD (Buckland, 2009 [rMed]), stroke (Estruch, 2013 [Med]; Misirli, 2012 [MDS]) or MI (Hansen-Krone, 2012 [Smart Diet]). One study reported a favorable association between healthy oils (plant and fish oils) and CHD risk in women only (Chiuve, 2012 [HEI, AHEI]). As well, these components were favorably associated with improved BP and blood lipids (Estruch, 2006 [Med]). Fish consumption was favorably associated with risk of CVD (Gardener, 2011 [MDS], CHD (Buckland, 2009 [rMed]), or MI (Hansen-Krone, 2012 [Smart Diet]). The studies that assessed plant oils and fish were predominantly using a Mediterranean diet score, although the HEI-2005 included plant and fish oils in their healthy oils component.
 
Alcohol consumption was reported by seven studies as having a favorable association with risk of CVD (Akbaraly, 2011 [AHEI]; Belin, 2011 [AHEI]; Gardener, 2011 [MDS]; Hoevenaar-Blom, 2012 [MDS]); CHD (Buckland, 2009 [rMed]; Chiuve, 2012 [HEI, AHEI]); or HTN (Núñez-Córdoba, 2009 [MDS]). Studies that assessed alcohol used components of either a Mediterranean diet score (3 MDS, 1 rMed) or the AHEI. One study that used the HEI-2005 reported a favorable association between a composite solid fats/alcoholic beverages/added sugars (SoFAAS) component, the benefits of which were driven by alcohol intake (Chiuve, 2012 [HEI, AHEI]).
 
Certain food components were also considered for their unfavorable association with risk of CVD. Meat, including red and processed meat, was associated with increased risk of CVD (Belin, 2011 [AHEI]), CHD (Buckland, 2009 [rMed], CHD only in women (Chiuve, 2012 [HEI, AHEI]), and elevated BP (Steffen, 2005 [Food Index]). Dairy consumption, on the other hand, was reported both negatively and positively in different studies. One study reported an unfavorable association between total dairy and CHD risk (Buckland, 2009 [rMed]); however, two studies reported a favorable association between total dairy and HTN (Folsom, 2007 [DASH] or elevated BP (Steffen, 2005 [Food Index]). 
 

Components across All Scores and Indices

Although the dietary pattern scores that were used in this large body of evidence were numerous, those scores adherence to which was associated with decreased risk of CVD clinical endpoints were many fewer, and can be compared to assess commonalities across them. Scores that were associated with decreased risk of CVD, CHD, stroke, MI, and HF were selected to examine commonalities in components across scores related to clinically significant outcomes, rather than intermediate markers (table 4-B-I-1 Comparison of Dietary Components across Diet Pattern Scores). The scores include the MDS (18 associated outcomes), aMed (5 associated outcomes), AHEI (7 associated outcomes, 3 AHEI 2010), RFS (6 associated outcomes), HEI-2005 (3 associated outcomes), and DASH (6 associated outcomes). Regarding a DASH score, the score developed by Fung (2008) was associated with CVD endpoints in all six studies; by comparison, the DASH score of Folsom was not significantly associated with CVD, CHD, stroke death, or HTN (Folsom, 2007) or heart failure (Levitan, 2009a). Overall, the above scores were the most commonly applied scores associated with a protective outcome for CVD risk in healthy adults. This relatively small number of dietary pattern scores, that are also uncomplicated versions, had similarities in that food groups were aligned and described in similar ways, although they were scored differently. Many of these scores have been tested in the same cohort with similar results.
 
Two of the scores assessed adherence to a Mediterranean style diet (MDS, aMed). Although these were a priori, hypothesis-driven scores, they were also population-based in that dichotomous points (0, 1) were assigned to individuals who were above or below the population median. The aMed score was based on the original MDS of Trichopolou and modified by excluding potatoes from the vegetable group, separating fruit and nuts into two groups, eliminating the dairy group, including whole-grain products only, including only red and processed meats for the meat group, and assigning alcohol intake of 5-15 g/d for 1 point (Fung et al., 2005). The Healthy Eating Index (HEI) was developed to measure compliance with U.S. dietary guidance, initially using the 1995 Dietary Guidelines for Americans (DGAs) and updated for 2005 (and currently 2010, but not utilized in studies reported in this review). In this body of evidence, the updated HEI-2005 was used to assess CVD outcomes, as well as to compare predictive utility with an alternative version, the Alternative HEI (AHEI) (Chiuve, 2012). Additionally, a variation on the original HEI was used by McCullough to assess CVD risk in men and women separately (McCullough, 2000a and 2000b). In the studies included here, the HEI and AHEI scores (including updated HEI-2005 and AHEI-2010) were based on dietary guidance with individuals ranked across a gradient of high to low scores, based on recommended intakes, independent of the population. The AHEI-2010 introduced detailed changes from the original, including the recommendation of ~5 or 6 servings per day of 100% whole-grain products for women and men, respectively, and including long chain n-3 fats (EPA and DHA) as well as PUFA intake as percent energy as positive components. The DASH score developed by Fung reflected adherence to a DASH-style diet and was operationalized based on quintiles of intake of food group and nutrient components, rather than absolute intake. The Recommended Food Score (RFS) of Kant (2000) was based on consumption of dietary guidelines-recommended foods assessed by summing individual food items derived from the FFQ tool used in the respective study, with a summary score that added the food items consumed at least once per week. The total RFS scores varied from study to study, as the number of FFQ items varied, although the most commonly used RFS version was the 23-item version of Kant (2000). Overall, although the Mediterranean, HEI and AHEI, DASH, and RFS score methods differed, they were all employed to address, and successfully predicted, the association between an a priori defined dietary pattern and CVD risk, at the level of endpoint outcomes. It is interesting to note that the AHEI and updated AHE-2010, the aMed score, and the DASH-Fung score have all been tested in women in the Nurses’ Health study and were favorably associated with CVD, CHD, and/or stroke risk. Related to this, when the HEI-2005 and AHEI-2010 were compared in both the Nurses’ Health Study and the Health Professionals Follow-Up Study, and risk was adjusted for the other respective model, the HEI-2005 lost significance in association with CVD, CHD, and stroke risk, whereas the AHEI-2010 did not (Chiuve, 2012).
 
Taken together, the positive components of the scores that were associated with decreased CVD risk were fruits and nuts, vegetables, whole grains, legumes, unsaturated fats, and fish. Alcohol was commonly included as a positive component when consumed in moderation in all of the scores except for HEI, RFS, and DASH. Meat (red and processed meat) was a negative component in the Mediterranean scores, AHEI scores, and DASH, although the original AHEI emphasized a white/red meat ratio where the white meat included poultry and fish. The RFS of Kant included chicken or turkey as positive components in their food item-based score. There was variation in the method by which dairy was assessed. Total dairy (high fat) was a negative component in the MDS, was omitted from the aMed, but was a positive component when meeting recommended intakes for the HEI-2005 or as low-fat dairy in the RFS and DASH-Fung scores. Certain scores also included sugars (HEI-2005 included sugar in a combined category of solid fats, alcoholic beverages and added sugars [SoFAAS]) or sugar-sweetened beverages (AHEI-2010 included sugar sweetened beverages and fruit juice), as well as trans fats, and sodium as negative components.

Table 4-B-I-1  Comparison of Dietary Components across Diet Pattern Scores
 

Qualitative Assessment of the Collected Evidence

Quality and Quantity

Quality assessment for the studies included in this systematic review involved determining the validity of each study. Validity was assessed by examining the scientific soundness of study design and execution to avoid potential bias in the findings related to outcomes; this included selection, performance, attrition, detection, and reporting bias. For this body of evidence, the preponderance of the evidence consisted of positive quality studies (45 out of 55 studies), indicating a low risk of bias overall. This was a relatively large body of evidence with studies that directly addressed the question. The majority of these studies were prospective cohort studies with large numbers of participants in nationally recognized cohorts.

Consistency
  • CVD Incidence and Mortality: The evidence of a protective association between a dietary pattern score and risk of CVD was consistent in the majority of studies that used either a Mediterranean diet score or dietary guidelines index in healthy adult populations. The body of evidence for CVD endpoint clinical outcomes was relatively large (29 studies), with many fewer studies on intermediate outcomes. Typically, studies that measured CVD as a primary outcome did not include all diseases of the circulatory system (i.e., studies did not include the entire spectrum of ICD disorders for this category). Total CVD most commonly included CHD and stroke, both fatal and nonfatal events. These studies were consistent in finding a decreased risk of CVD associated with adherence to either a Mediterranean diet or dietary guidelines-related pattern. Out of 17 studies, 15 studies reported a favorable association. This included studies in the United States (9 studies), Spain (2 studies), France, the Netherlands, Greece, Japan, and the 11 European countries of the HALE study. All of these studies measured a composite fatal and non-fatal CVD, with the exception of two studies that assessed CVD mortality alone. The one RCT reported that a Mediterranean diet decreased risk of CVD in high risk subjects, and the trial (PREDIMED) was stopped early for meeting benefit requirements. This category also included studies with the primary objective of assessing total mortality, but that also assessed cause-specific mortality including CVD. These studies commonly assessed CVD as a larger composite category than the above studies, most often described as “diseases of the circulatory system” and including the complete spectrum, or a large spectrum, of ICD disorders for the category. These studies were less consistent, and this may relate to the larger number of cardiovascular diseases included in the CVD outcome. Out of twelve studies, five reported a favorable association, four reported different risk outcomes based on gender, and three found no association.
  • CHD Incidence and Mortality: The evidence of a protective association between a dietary pattern score and risk of CHD was consistent in the majority of studies in healthy adult populations. Most of the studies assessed adherence to a Mediterranean diet, with fewer studies that assessed dietary guidelines-related or DASH scores. The body of evidence for CHD endpoint clinical outcomes was also relatively large. Out of 16 studies that assessed CHD, including nonfatal and fatal outcomes, nine studies reported a favorable association between adherence to a dietary pattern and CHD risk, one study found a favorable association between adherence to the AHEI, but not HEI, and three studies reported no association. In studies that examined CHD mortality secondary to total mortality, all three studies found a favorable association between dietary pattern adherence and CHD mortality. Similar to the studies on CVD, the prospective cohort studies were conducted in the United States and several European countries with large, well-characterized cohorts.
  • Stroke: The evidence of a protective association between a dietary pattern score and risk of stroke was consistent in the majority of studies in healthy adult populations. Out of 13 studies that assessed stroke, including nonfatal and fatal outcomes, 10 studies reported a favorable association. This included one RCT, the above-mentioned PREDIMED trial. Two studies that compared different dietary pattern scores found a favorable association between one or more scores and stroke risk and one study reported no association.
  • Myocardial Infarction: The evidence of a protective association between a dietary pattern score and risk of MI was inconsistent and included fewer studies than the above health outcomes. Out of six studies, two studies reported a favorable association between dietary pattern and MI risk, one study reported different risk outcomes based on gender, two studies found no association, and one study found a favorable association with one score, but an unfavorable association with a second score.
  • Heart Failure: The three studies that examined heart failure were consistent in showing a favorable association between adherence to a dietary pattern and risk of heart failure; two of these studies included the ICD categories for heart failure and hypertensive heart failure.

Impact

This body of evidence directly addressed the exposures and health outcomes of interest for this systematic review, with a large number of studies measuring endpoint clinical outcomes. Overall, several large prospective cohort studies found a decreased risk of CVD, CHD, stroke, or heart failure associated with adherence to a Mediterranean diet or dietary guidelines score. Although not clinical trials, these cohort studies reported results that are applicable in free-living populations. For example, a recent study from the Nurses’ Health and Health Professionals Follow-up cohorts reported that a change from lowest to highest AHEI adherence reduced risk of CVD, CHD, and stroke by 20, 31, and 14 percent, respectively, over 24 years (Chuive, 2012 (HEI, AHEI)). It is notable that in this study dietary intake was assessed at regular intervals over the long follow-up period. Additionally, although there were only a few clinical trials in this body of evidence, the PREDIMED study reported that a Mediterranean diet (plus olive oil or nuts) decreased risk of clinical endpoint outcomes, CVD and stroke, in high risk subjects and, as already indicated, this trial was stopped early for meeting benefit requirements.

Generalizability/External Validity              

Studies were conducted in the United States, Europe, Japan, and Australia and included many large, well-characterized cohorts. Nineteen studies were conducted in the United States and the remaining studies were conducted mainly in Europe, with Spain, Sweden, and Greece heavily represented. In addition, two studies were conducted in Japan and one in Australia. Taken together, the prospective cohort studies were primary prevention studies of general populations and most studies were conducted with healthy adults. Subgroup analysis was conducted in many studies on men and women; however, there were no consistent differences between men and women in the association between dietary pattern and CVD outcomes. There were only a few studies that specifically identified and analyzed race/ethnic subgroups of their cohorts and none reported differences based on race or ethnicity. Overall, therefore, there is a relatively large body of evidence on adherence to dietary patterns in the United States and European populations of healthy adults. This is primarily among Caucasian populations, however, with little additional analysis of racial or ethnic subpopulations. Given the robust evidence involving U.S. cohorts and endpoint clinical outcomes, the generalizability to the U.S. population, and relevance of this body of evidence to U.S. policy on dietary patterns and risk of CVD, is compelling.

Limitations of the Evidence

Common limitations of studies on dietary patterns using a priori scores involve the use of different scores, differences between scores that are based on median population intakes versus indices that are based on recommended intakes, scores that use similar weights for each component assuming equivalent effects on health, the use of different confounding factors (or lack of sufficient adjustment), and problems associated with use of different FFQs and validation related to other methods of diet assessment. It should be said, however, that in this relatively large body of evidence, a limited number of scores were used, oftentimes less complicated versions of these scores, and in a number of cases the different scores were tested in the same cohorts. Overall, this makes the comparison of food components across these scores feasible. Additionally, a very common limitation in many prospective cohort studies is that dietary intake is based on a single dietary assessment at baseline, with no follow-up assessment of dietary intake over the period of the study. However, this body of evidence had notable exceptions including Chiuve (2010 [aMed/NHS] and 2011 [HEI, AHEI/NHS, HPFS]), and Fung (2008 [DASH/NHS] and 2009 [aMed/NHS]) that measured dietary intakes at regular intervals across the period of follow-up of the respective studies. Therefore, these studies did take into account the fact that diets change over time due to trends in the food supply, as well as the fact that population-level and individual-level food choices change over time.

Research Recommendations

The studies covered in this systematic review provide results that improve some of the problems involved in dietary patterns research. For example, the need for consensus on a single score or index that is applicable across populations is less problematic in this body of evidence than for some other outcomes, as a relatively small number of uncomplicated scores have been used to successfully predict CVD risk in large U.S. and European populations. Further quantitative analysis/comparisons of these scores and their respective components by meta-analysis would be particularly useful. Although a large number of the studies assessed food group components and their association with CVD outcomes, many did not, and more precise determination of the benefits and risks of individual components (e.g., alcohol) would be helpful for policy recommendations. In addition, component analysis could be improved by determining interaction terms across components that would be needed to maintain a dietary patterns approach. Methodologically, research in this area could be improved by measuring dietary intake at regular intervals over the course of a prospective study, rather than just at baseline (although a few of the large cohort studies in this body of evidence did this). Determining the best approach to weighing and scoring individual food components would also improve the rigor in application of scores to assess dietary pattern adherence. Additionally, studies in this body of evidence that assessed gender differences in the relationship between adherence to a dietary pattern and CVD risk found inconsistent results. Further research is needed to clarify this. There were also very few studies that identified racial/ethnic subgroups within their cohorts and analyzed these groups separately related to CVD risk and this warrants additional research. Assessment of dietary patterns at earlier and later stages of the life cycle is also recommended. Lastly, behavioral issues related to timing, frequency, and size of meals need further consideration.


Abbreviations

Scores & Indices: Dietary Approaches to Stop Hypertension (DASH); Diet Quality Index (DQI); DQI-Swedish Nutrition Recommendations (DQI-SNR); Healthy Eating Index (HEI); Alternate Healthy Eating Index (AHEI); Mediterranean Diet Score (MDS); Alternate Mediterranean Diet Score (aMed); Mediterranean Adequacy Index (MAI); Mediterranean Style Diet Pattern Score (MSDPS); Relative Mediterranean Diet Score (rMED); modified MDS (MMDS); Programme National Nutrition Santé Guideline Score (PNNS-GS); Recommended Food Score (RFS)
Cohorts: Coronary Artery Risk Development in Young Adults (CARDIA); European Prospective Investigation into Cancer and Nutrition (EPIC); Framingham Offspring and Spouse (FOS); Health Professionals Follow-Up Study (HPFS); Healthy Ageing: a Longitudinal study in Europe (HALE); Multi-Ethnic Study of Atherosclerosis (MESA); Nurses’ Health Study (NHS); Prevencion con Dieta Mediterranea (PREDIMED); Seguimiento Universidad de Navarra (SUN); SUpplementation en VItamines et Minereaux AntioXydants (SU.VI.MAX); Women’s Health Initiative (WHI)

Table 4-B-I-2. Overview Table: Cardiovascular Disease
Table 4-B-I-3. Overview Table: Coronary Heart Disease
Table 4-B-I-4.  Overview Table: Stroke, Myocardial Infarction and Heart Failure
Table 4-B-I-5  Overview Table: Hypertension & Blood Pressure and Blood Lipids

REFERENCES (Included Studies)

See Search Plan for CVD
 

Additonal References 

  1. 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 Apr 17;336(16):1117-24. PMID: 9099655.
  2. Sacks FM, Obarzanek E, Windhauser MM, Svetkey LP, Vollmer WM, McCullough M, Karanja N, Lin PH, Steele P, Proschan MA, et al. Rationale and design of the Dietary Approaches to Stop Hypertension trial (DASH). A multicenter controlled-feeding study of dietary patterns to lower blood pressure. Ann Epidemiol. 1995 Mar;5(2):108-18. PMID: 7795829.
  3. Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D, Obarzanek E, Conlin PR, Miller ER 3rd, Simons-Morton DG, Karanja N, Lin PH; DASH-Sodium Collaborative Research Group. Effects on blood pressure of reduced dietary sodium and the Dietary Approaches to Stop Hypertension (DASH) diet. DASH-Sodium Collaborative Research Group. N Engl J Med. 2001 Jan 4;344(1):3-10. PMID: 11136953.



Research Design and Implementation
For a summary of the Research Design and Implementation results, click here.
Worksheets
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 Aug; 141(8): 1,552-1,558.

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 years of follow-up: Results from the Whitehall II cohort. Am J Clin Nutr. 2011 Jul; 94(1): 247-253.

Belin RJ, Greenland P, Allison M, Martin L, Shikany JM, Larson J, Tinker L, Howard BV, Lloyd-Jones D, Van Horn, L. Diet quality and the risk of cardiovascular disease: The Women's Health Initiative (WHI). Am J Clin Nutr. 2011; 94(1): 49-57.

Buckland G, Agudo A, Travier N, Huerta JM, Cirera L, Tormo MJ, Navarro C, Chirlaque MD, Moreno-Iribas C, Ardanaz E, Barricarte A, Etxeberria J, Marin P, Quirós JR, Redondo ML, Larrañaga N, Amiano P, Dorronsoro M, Arriola L, Basterretxea M, Sanchez MJ, et al. Adherence to the Mediterranean diet reduces mortality in the Spanish cohort of the European Prospective Investigation into Cancer and Nutrition (EPIC-Spain). Br J Nutr. 2011 Nov; 106(10): 1,581-1,591.

Buckland G, Gonzalez CA, Agudo A, Vilardell M, Berenguer A, et al. Adherence to the Mediterranean Diet and risk of coronary heart disease in the Spanish EPIC Cohort Study. Am J Epidemiol. 2009; 170(12): 1,518-1,529.

Camoes M, Oliveira A, Pereira M, Severo M, Lopes C. Role of physical activity and diet in incidence of hypertension: A population-based study in Portugese adults. Eur J Clin Nutr. 2010 Dec; 64(12): 1,441-1,449.

Chiuve SE, Fung TT, Rexrode KM, Spiegelman D, Manson JE, Stampfer MJ, Albert CM. Adherence to a low-risk, healthy lifestyle and risk of sudden cardiac death among women. JAMA. 2011 Jul 6; 306(1): 62-69.

Chiuve SE, Fung TT, Rimm EB, Hu FB, McCullough ML, Wang M, Stampher MJ, Willett WC. Alternative dietary indices both strongly predict risk of chronic disease. J Nutr. 2012 Jun; 142(6): 1,009-1,018.

Dauchet L, Kesse-Guyot E, Czernichow S, Bertrais S, Estaquio C, Péneau S, Vergnaud AC, Chat-Yung S, Castetbon K, Deschamps V, Brindel P, Hercberg S. Dietary patterns and blood pressure change over 5-y follow-up in the SU.VI.MAX cohort. Am J Clin Nutr. 2007 Jun;85(6):1650-6. 

Drake I, Gulber B, Sonestedt E, Wallstrom P, Margaretha P, Hlebowicz J, Nilsson J, Hedblad B and Wirfalt E. Scoring models of a diet qualty index and the predictive capability of mortality in a population-based cohort of Swedish men and women. Pub Health Nutr. 2013 Mar; 16(3): 468-478.

Estruch R, Martínez-González MA, Corella D, Salas-Salvadó J, Ruiz-Gutiérrez V, Covas MI, Fiol M, Gómez-Gracia E, López-Sabater MC, Vinyoles E, Arós F, Conde M, Lahoz C, Lapetra J, Sáez G, Ros E; PREDIMED Study Investigators. Effects of a Mediterranean-style diet on cardiovascular risk factors: a randomized trial. Ann Intern Med. 2006; 145(1): 1-11.

Fidanza F, Alberti A, Lanti M, Menotti A. Mediterranean Adequacy Index: correlation with 25-year mortality from coronary heart disease in the Seven Countries Study. Nutr Metab Cardiovasc Dis. 2004 Oct;14(5):254-8.

Fitzgerald KC, Chiuve SE, Buring JE, Ridker PM, Glynn RJ. Comparison of associations of adherence to a Dietary Approaches to Stop Hypertension (DASH)-style diet with risks of cardiovascular disease and venous thromboembolism. J Thromb Haemost. 2012 Feb; 10(2): 189-198.

Folsom AR, Parker ED, Harnack LJ. Degree of concordance with DASH diet guidelines and incidence of hypertension and fatal cardiovascular disease. Am J Hypertens. 2007 Mar; 20 (3): 225-232.

Forman JP, Stampfer MJ, Curhan GC. Diet and lifestyle risk factors associated with incident hypertension in women. JAMA. 2009 Jul 22;302(4):401-11.

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(7):713-20.

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 Mar 3; 119 (8): 1,093-1,100. Epub 2009 Feb 16. Erratum in: Circulation. 2009 Mar 31; 119 (12): e379.

Gardener H, Wright CB, Gu Y, Demmer RT, Boden-Albala B, Elkind MSV, Sacco RL, Scarmeas N. Mediterranean-style diet and risk of ischemic stroke, myocardial infarction, and vascular death: The Northern Manhattan Study. Am J Clin Nutr. 2011 Dec; 94(6): 1,458-1,464.

Hansen-Krone IJ, Enga KF, Njølstad I, Hansen JB, Braekkan SK. Heart healthy diet and risk of myocardial infarction and venous thromboembolism. The Tromsø Study. Thromb Haemost. 2012 Sep; 108(3): 554-560.

Jacobs DR Jr, Sluik D, Rokling-Andersen MH, Anderssen SA, Drevon CA. Association of one-year changes in diet pattern with cardiovascular disease risk factors and adipokines: Results from the one-year randomized Oslo Diet and Exercise Study. Am J Clin Nutr. 2009 Feb; 89(2): 509-517.

Kaluza J, Håkansson N, Brzozowska A, Wolk A. Diet quality and mortality: A population-based prospective study of men. Eur J Clin Nutr. 2009 Apr; 63(4): 451-457.

Kant AK, Schatzkin A, Braubard BI, Schairer C. A prospective study of diet quality and mortality in women. JAMA. 2000; 283: 2,109-2,115.

Kesse-Guyot E, Touvier M, Henegar A, Czernichow S, Galan P, Hercberg S, Castetbon K. Higher adherence to French dietary guidelines and chronic diseases in the prospective SU.VI.MAX cohort. Eur J Clin Nutr. 2011 Aug; 65(8): 887-894. 

Knoops KT, de Groot LC, Kromhout D, Perrin AE, Moreiras-Varelas, Menotti A, van Staveren WA. Mediterrean diet, lifestyle factors, and 10-year mortality in elderly European men and women: the HALE project. JAMA. 2004 Sep 22; 292(12):1433-9.

Levitan EB, Wolk A, Mittleman MA.Relation of consistency with the dietary approaches to stop hypertension diet and incidence of heart failure in men aged 45 to 79 years. Am J Cardiol. 2009 Nov 15; 104(10): 1,416-1,420.

Levitan EB, Wolk A, Mittleman MA. Consistency with the DASH diet and incidence of heart failure. Arch Intern Med. 2009 May 11;169(9):851-7.

Martínez-González MA, Guillén-Grima F, De Irala J, Ruíz-Canela M, Bes-Rastrollo M, Beunza JJ, López Del Burgo C, Toledo E, Carlos S, Sánchez-Villegas A. Mediterranean diet is associated with a reduction in premature mortality among middle-aged adults. J Nutr. 2012 Sep; 142(9): 1,672-1,678. 

Martínez-González MA, García-López M, Bes-Rastrollo M, Toledo E, Martínez-Lapiscina EH, Delgado-Rodriguez M, Vazquez Z, Benito S, Beunza JJ. Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort. Nutr Metab Cardiovasc Dis. 2011 Apr; 21(4): 237-244.

McCullough ML, Feskanich D, Rimm EB, Giovannucci EL, Ascherio A, Variyam JN, Spiegelman D, Stampfer MJ, Willett WC. Adherence to the Dietary Guidelines for Americans and risk of major chronic disease in men. Am J Clin Nutr. 2000 Nov; 72(5): 1,223-1,231.

McCullough ML, Feskanich D, Stampfer MJ, Rosner BA, Hu FB, Hunter DJ, Variyam JN, Colditz GA, Willett WC. Adherence to the Dietary Guidelines for Americans and risk of major chronic disease in women. Am J Clin Nutr. 2000 Nov; 72(5): 1,214-1,222.

McCullough ML, Feskanich D, Stampfer MJ, Giovannucci EL, Rimm EB, Hu FB, Spiegleman 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: 1,261-1,271.

McCullough ML, Willett W. Evaluating adherence to recommended diets in adults: The Alternate Healthy Eating Index. Public Health Nutr. 2006 Feb; 9(1a): 152-157.

Menotti A, Alberti-Fidanza A, Fidanza F. The association of the Mediterranean Adequacy Index with fatal coronary events in an Italian middle-aged male population followed for 40 years. Nutr Metab Cardiovasc Dis. 2012 Apr; 22(4): 369-375.

Michels KB, Wolk A. A prospective study of variety of health foods and mortality in women. Int J Epidemiol. 2002 Aug; 31(4): 847-854.

Mitrou PN, Kipnis V, Thiébaut AC, Reedy J, Subar AF, Wirfält E, Flood A, Mouw T, Hollenbeck AR, Leitzmann MF, Schatzkin A. Mediterranean dietary pattern and prediction of all-cause mortality in a US population: results from the NIH-AARP Diet and Health Study. Arch Intern Med. 2007 Dec 10;167(22):2461-8.

Moore LL, Bradlee ML, Singer MR, Qureshi MM, Buendia JR, Daniels SR. Dietary Approaches to Stop Hypertension (DASH) eating pattern and risk of elevated blood pressure in adolescent girls. Br J Nutr. 2012 Jan; 16: 1-8.

Nakamura Y, Ueshima H, Okamura T, Kadowaki T, Hayakawa T, Kita Y, Abbott RD, Okayama A. National Integrated Project for Prospective Observation of Non-Communicable Diseases and its Trends in the Aged, 1980 Research Group. A Japanese diet and 19-year mortality: National integrated project for prospective observation of non-communicable diseases and its trends in the aged, 1980. Br J Nutr. 2009; 101(11):1,696-1,705. 

Nilsson LM, Winkvist A, Brustad M, Jansson JH, Johansson I, Lenner P, Lindahl B, Van Guelpen B. A traditional Sami diet score as a determinant of mortality in a general northern Swedish population. Int J Circumpolar Health. 2012 May 4; 71(0): 1-12.  

 Núñez-Córdoba JM, Valencia-Serrano F, Toledo E, Alonso A, Martínez-González MA. The Mediterranean diet and incidence of hypertension: the Seguimiento Universidad de Navarra (SUN) Study. Am J Epidemiol. 2009 Feb 1;169(3):339-46. Epub 2008 Nov 26.

Oba S, Nagata C, Nakamura K, Fujii K, Kawachi T, Takatsuka N, Shimizu H. Diet based on the Japanese Food Guide Spinning Top and subsequent mortality among men and women in a general Japanese population. J Am Diet Assoc. 2009 Sep; 109(9): 1,540-1,547.

Osler M, Helms Andreasen A, Heitmann B, Høidrup S, Gerdes U, Mørch Jørgensen L, Schroll M. Food intake patterns and risk of coronary heart disease: a prospective cohort study examining the use of traditional scoring techniques. Eur J Clin Nutr. 2002 Jul;56(7):568-74

Osler M, Schroll M. Diet and mortality in a cohort of elderly people in a north European community. Int J Epidemiol. 1997 Feb; 26(1): 155-159.

Panagiotakos DB, Pitsavos C, Chrysohoou C, Skoumas I, Stefanadis C; ATTICA Study. Five-year incidence of cardiovascular disease and its predictors in Greece: the ATTICA study. Vasc Med. 2008; 13(2): 113-121.

Rumawas ME, Meigs JB, Dwyer JT, McKeown NM, Jacques PF. Mediterranean-style dietary pattern, reduced risk of metabolic syndrome traits, and incidence in the Framingham Offspring Cohort. Am J Clin Nutr. 2009; 90(6): 1,608-1,614.

Russell J, Flood V, Rochtchina E, Gopinath B, Allman-Farinelli M, Bauman A, Mitchell P. Adherence to dietary guidelines and 15-year risk of all-cause mortality. Br J Nutr. 2012 May; 9:1-9.

Salas-Salvado J , Fernandez-Ballart J, Ros E, Martinez-Gonzalez MA, Fito M, Estruch R ,Corella D, Fiol D, Gomez-Gracia E, Aros F, Flores G, Lapetra J, Lamuela-Raventos R, Ruiz-Gutierrez V, Bullo M, Basora J, Covas M. Effect of a Mediterranean diet supplemented with nuts on metabolic syndrome status; one-year results of the PREDIMED randomized trial. Arch Intern Med. 2008; 168 (22): 2,449-2,458. 

Seymour JD, Calle EE, Flagg EW, Coates RJ, Ford ES, Thun MJ; American Cancer Society. Diet Quality Index as a predictor of short-term mortality in the American Cancer Society Cancer Prevention Study II Nutrition Cohort. Am J Epidemiol. 2003 Jun 1;157(11):980-8

Sjögren P, Becker W, Warensjö E, Olsson E, Byberg L, Gustafsson IB, Karlström B, Cederholm T. Mediterranean and carbohydrate-restricted diets and mortality among elderly men: A cohort study in Sweden. Am J Clin Nutr. 2010 Oct; 92(4): 967-974.    

Steffen LM, Kroenke CH, Yu X, Pereira MA, Slattery ML, Van Horn L, Gross MD, Jacobs DR Jr. Associations of plant food, dairy product and meat intakes with 15-year incidence of elevated blood pressure in young black and white adults: The Coronary Artery Risk Development in Young Adults (CARDIA) Study. Am J Clin Nutr. 2005 Dec; 82 (6): 1,169-1,177.

Tognon G, Nilsson LM, Lissner L, Johansson I, Hallmans G, Lindahl B, Winkvist A. The Mediterranean diet score and mortality are inversely associated in adults living in the subarctic region. J Nutr. 2012 Aug; 142(8): 1,547-1,553.

Tortosa A, Bes-Rastrollo M, Sanchez-Villegas A, Basterra-Gortari FJ, Nuñez-Cordoba JM, Martinez-Gonzalez MA. Diabetes Care. 2007 Nov; 30(11): 2,957-2,959.

Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterrean diet and survival in a Greek population. N Engl J Med. 2003 Jun 26;348(26):2599-608.

van de Laar RJ, Stehouwer CD, van Bussel BC, Prins MH, Twisk JW, Ferreira I. Adherence to a Mediterranean dietary pattern in early life is associated with lower arterial stiffness in adulthood: The Amsterdam Growth and Health Longitudinal Study. J Internal Medicine. 2012; Jul 19. doi: 10.1111/j.1365-2796.2012.02577.x. [Epub ahead of print].

von Ruesten A, Illner AK, Buijsse B, Heidemann C, Boeing H. Adherence to recommendations of the German food pyramid and risk of chronic diseases: results from the EPIC-Potsdam study. Eur J Clin Nutr. 2010 Nov; 64(11): 1,251-1,259.

Zamora D, Gordon-Larsen P, He K, Jacobs DR Jr, Shikany JM, Popkin BM. Are the 2005 Dietary Guidelines for Americans associated with reduced risk of type 2 diabetes and cardiometabolic risk factors? Twenty-year findings from the CARDIA study. Diabetes Care. 2011 May; 34(5): 1,183-1,185.