HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Brunner, E. J Articles by Marmot, M. G Search for Related Content PubMed PubMed Citation Articles by Brunner, E. J Articles by Marmot, M. G Agricola Articles by Brunner, E. J Articles by Marmot, M. G

Dietary patterns and 15-y risks of major coronary events, diabetes, and mortality^1,2,3

¹ From the Department of Epidemiology and Public Health, University College London, London, United Kingdom (EJB, AM, DRW, MS, MJS, and MGM), and the Helsinki Collegium for Advanced Studies, University of Helsinki, Finland (PM)

² The Whitehall II study was supported by grants from the UK Medical Research Council (MRC), the British Heart Foundation, the Health and Safety Executive, the Department of Health, the National Heart Lung and Blood Institute (HL36310), the National Institute on Aging (AG13196), the Agency for Health Care Policy Research (HS06516), and the MacArthur Foundation Research Network on Socio-economic Status and Health. DRW is supported by an MRC New Investigator Award. MS is supported by a Department of Health career development award. MJS is supported by the British Heart Foundation. MGM is supported by an MRC Research Professorship. No funding source had direct influence over the design, conduct, or reporting of this study.

³ Reprints not available. Address correspondence to E Brunner, Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, London WC1E 6BT, United Kingdom. E-mail: e.brunner{at}ucl.ac.uk.

	ABSTRACT

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Background: Few studies have examined the long-term effect of habitual diet on risks of incident diabetes, coronary heart disease, and mortality.

Objective: We analyzed the prospective relation of dietary patterns with incident chronic disease and mortality during 15 y of follow-up in the Whitehall II study.

Design: We conducted a prospective analysis (106 633 person-years at risk) among men and women (n = 7731) with a mean age of 50 y at the time of dietary assessment (127-item food-frequency questionnaire). Coronary death or nonfatal myocardial infarction and incident diabetes were verified by record tracing and oral-glucose-tolerance tests.

Results: Cluster analysis identified 4 dietary patterns at baseline. The patterns were termed unhealthy (white bread, processed meat, fries, and full-cream milk; n = 2665), sweet (white bread, biscuits, cakes, processed meat, and high-fat dairy products; n = 1042), Mediterranean-like (fruit, vegetables, rice, pasta, and wine; n = 1361), and healthy (fruit, vegetables, whole-meal bread, low-fat dairy, and little alcohol; n = 2663). Compared with the unhealthy pattern, the healthy pattern reduced the risk of coronary death or nonfatal myocardial infarction and diabetes; hazard ratios (95% CI) were 0.71 (0.51, 0.98) and 0.74 (0.58, 0.94), respectively, after adjustment for age, sex, ethnicity, dietary energy misreporting, social position, smoking status, and leisure-time physical activity. Dietary pattern was not associated with all-cause mortality. Residual confounding by socioeconomic factors was unlikely to account for the observed dietary effects.

Conclusions: The healthy eating pattern reduced risks of diabetes and major coronary events. Such dietary patterns offer considerable health benefits to individuals and contribute to public health.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Long-term prospective studies show increased survival (ie, low mortality rates) among elderly persons who follow Mediterranean or prudent diets (1-3). These observational studies of the diet as a whole are particularly significant because large trials of antioxidant and multivitamin supplementation do not show a preventive effect for cardiovascular disease or cancer (4-6). It is unclear whether we should reject the trial evidence because it is based on dietary supplements or accept that the salutary effect of a prudent diet is in doubt, despite the effectiveness of shorter term interventions involving real diets (7, 8). It is possible that the positive observational findings involving self-selected eating patterns are the result of confounding (9). If this were the case, socioeconomic factors and health behaviors correlated with diet, rather than diet itself, would account for the mortality reduction. To increase confidence in the dietary explanation, disease-specific protective effects need to be shown in studies that pay attention to the confounding issue.

We therefore identified dietary patterns in a younger cohort of healthy British adults to study their relations to incident major coronary events and diabetes as well as mortality. Participants are well-characterized for socioeconomic position and health behaviors, enabling effective adjustment for potential confounding factors within this single population (9). Follow-up for coronary events and incident diabetes has been carried out by means of screening contacts every 5 y and verification of medical records for cases of myocardial infarction and coronary death.

	SUBJECTS AND METHODS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Participants were recruited to the Whitehall II study in 1985–1988 (phase 1) from 20 civil service departments in London. The response rate was 73% (n = 10 308). Informed consent was obtained at phase 1 and was renewed at each contact. At phase 3 (1991–1993), 8637 participants completed a further questionnaire or attended the research clinic (8104 attended the clinic). Details of the study are reported elsewhere (10, 11) The University College London Research Ethics Committee approved the study. Phase 3 respondents were sent a food-frequency questionnaire, and 7935 (92%) returned it fully completed. Two groups, defined below, were excluded from analysis: members of outlying dietary clusters (n = 124) and energy misreporters (n = 80), yielding 7731 participants in the dataset. Follow-up for incident type 2 diabetes required an oral-glucose-tolerance test at the phase 3 baseline (prevalent diabetes: known, n = 80; newly detected, n = 149) and at least one subsequent research clinic (n = 6961). Of these, 6610 were assigned to a dietary cluster, as above.

Outcomes
Participants were flagged by the National Health Service Central Registry, who notified us of the date and cause of all deaths up to 31 July 2006. Deaths were classified as coronary if ICD9 (International Classification of Diseases, 9th edition) codes 410–414 or ICD10 (International Classification of Diseases, 10th edition) codes I20–I25 were present on the death certificate. Gastrointestinal tract-related cancer mortality was defined as fatal neoplasms of the lip, oral cavity, pharynx, or digestive organs. Cases of nonfatal myocardial infarction were identified from 12-lead electrocardiograms obtained at study phases 3, 5, and 7 and questionnaire items on chest pain and doctor's diagnosis. Details of physician investigations, diagnoses, and interventions were sought from medical records for all potential cases for final ascertainment. Classification, following the World Health Organization Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) methods, was carried out independently by 2 trained coders, with adjudication in the event of disagreement. Self-report in the absence of verification was not classified as myocardial infarction. Participants’ date of censoring depended on the last attended phase: 30 September 2004 for those who attended phase 7.

Incident cases of diabetes were identified by self-report of doctor's diagnosis and diabetic medication and 2-h 75-g oral-glucose-tolerance test at phases 3, 5, and 7 according to the 1999 World Health Organization classification. Individuals with prevalent self-reported or oral-glucose-tolerance test-diagnosed diabetes at phase 3 were excluded from the corresponding analysis. Incident diabetes was dated at the day of study visit for those first identified through the oral-glucose-tolerance test. For those identified by self-report, the midpoint between the first instance of self-reported diabetes and the previous phase was used. Person-time of exposure was censored at the midpoint between the last known visit and the first missing visit for those lost to follow-up. Participants with an intermediate missing phase were assumed to have continuous follow-up time. For those who had not developed diabetes up to phase 7, follow-up (mean duration 11.6 y) was censored on 30 September 2004 (phase 7 closing date).

Follow-up for mortality was essentially complete (99.9% flagged at the National Health Service Central Registry). Of those participating in the study who had no history of myocardial infarction at the baseline phase, 87% were included in the multivariate analysis of incident coronary heart disease (CHD). For incident diabetes, the corresponding proportion was 82%. Compared with those included in the diet-diabetes analysis, those without follow-up were older (mean difference: 0.88 y) and had a higher body mass index (mean difference: 0.63, in kg/m²).

Dietary clusters
Usual dietary intake was assessed by using a 127-item food-frequency questionnaire that has been validated against a 7-d diary (12). For each food item, participants were asked to report their frequency of eating a common unit or portion size in 1 of 9 categories during the previous year. Dietary patterns were identified as in a previous article (13) by using cluster analysis (PROC FASTCLUS; SAS Institute Inc, Cary, NC) with responses to all food-frequency items, except tea and coffee consumption, as input variables. The FASTCLUS procedure minimizes the sum of squared Euclidean distances between the observations and the cluster means. A total of 124 (1.5%) individuals were excluded when we did not allow FASTCLUS to assign outlying (distant) observations to a cluster. After we excluded outliers, the cluster solution was robust: we obtained similar cluster solutions (average 91.4% concordance) with 5 random halves of the observations. Twenty-two food-frequency questionnaire items with the strongest correlation with clusters were examined to assign cluster names. In sex-specific cluster analysis, lack of concordance with the combined data was due largely to movement between the 2 healthy and the 2 unhealthy clusters, which were merged to yield a total of 4 clusters, termed healthy, Mediterranean-like, unhealthy, and sweet (Table 1).

Evaluation of energy misreporting
Recorded total energy intake (EI) was evaluated by examining the ratio of EI to estimated energy expenditure (EE). The ratio EI/EE will be 1 if there is no energy misreporting and <1 if there is underreporting (15). Energy expenditure was based on estimated basal metabolic rate (BMR) (16). A physical activity level of at least 1.55 x BMR (sedentary) was assumed for all participants. The energy cost of reported leisure-time physical activity was added by using metabolic equivalent (MET) values for energy expenditure/kg body wt (3 MET/h of moderate activity, 5 MET/h of vigorous activity). Weight was missing for 261 men and 149 women, who were assigned the mean weight at phase 3 for calculating energy expenditure (men: 78.2 kg; women: 67.3 kg). Mean calculated physical activity level based on EE/BMR was 1.63 (range: 1.55–2.99) and mean EI/EE was 0.79 (range: 0.14–3.40). A total of 80 participants (1.0%) with a log EI/EE value outside 3 SD of the log mean were excluded from the analysis.

Employment grade and other covariates
Employment grade within the Civil Service, in 6 levels, was used as the measure of adult socioeconomic position. Annual salary in August 1992 was in the range of £6483 to £87 620. Father's social class was coded from the question, "what is/was your father's main job " and additional questions at phase 1 on training, employment status, and supervisory responsibility. Ethnicity was based on self-report (white, South Asian, Afro-Caribbean, or other). Weight, height, and waist circumference were measured in standardized fashion with participants dressed in a cloth gown and underclothes. Smoking habit (never, exsmoker, or current), leisure-time physical activity (hours of moderate and hours of vigorous activity per week), and blood pressure medication were self-reported. Serum total and HDL cholesterol and triacylglycerols were measured in fresh serum (10).

Statistical methods
Heterogeneity in food and micronutrient intakes across clusters was tested with the Kruskal-Wallis test or analysis of variance for energy and energy-yielding nutrients. Tests of proportions used the chi-square method. STATA 9.2 was used to fit Cox proportional hazards regression models by using follow-up time as the time variable and with initial adjustment for age (4 groups) at baseline, ethnicity, and energy misreporting. Tests for the proportionality assumption were carried out by using scaled Schoenfeld residuals (STATA stphtest command), (17), which was met for each outcome (P > 0.1). The reference group was the unhealthy cluster. Further adjustments for covariates were made by fitting models that included appropriate factors such as employment grade. For models involving serial adjustments, observations with incomplete data for any of the relevant covariates were excluded.

	RESULTS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
The distribution of energy, food, and nutrient intakes across the 4 dietary clusters is shown in Table 2. Mean energy intake was highest in the sweet and lowest in the unhealthy pattern in men and women. Energy misreporting based on EI/EE differed significantly between dietary patterns, but was not associated with healthiness. Underreporting was most prominent with the unhealthy and healthy patterns and was least prominent with the sweet pattern. There were clear differences in food intakes across clusters. Fruit and vegetable intake was almost two-fold higher among those with a healthy or Mediterranean-like pattern than among those with a unhealthy pattern. Cereal consumption followed a similar distribution, and differences were particularly noticeable for the high- and low-fiber bread/breakfast cereal subgroups. Median intake of oily fish was 50% greater in the Mediterranean-like and healthy patterns than in the sweet and unhealthy patterns. The unhealthy pattern included a high intake of red meat and a low intake of low-fat dairy products. The sweet pattern had the highest intakes of confectionary, biscuits, cakes, puddings, and high-fat dairy products. The Mediterranean-like pattern was associated with relatively high intakes of rice, pasta, and alcohol (median alcohol intake: men, 1.2 standard drinks/d; women, 0.8 standard drinks/d). Median alcohol intake was lower in the healthy pattern (men, 0.4 drinks/d; women, 0.2 drinks/d). Nutrient intakes reflected food intakes, with a low ratio of unsaturated to saturated fat in the sweet pattern and a high ratio in the healthy pattern. The Mediterranean-like and healthy patterns had relatively high antioxidant and fiber densities.

Table 3

Figure 1

View larger version (20K): [in this window] [in a new window]   FIGURE 1.. Survival and disease-free follow-up according to dietary pattern. Adjusted for age, sex, ethnicity, and energy misreporting. 95% CIs (unadjusted) are shown for each curve. CHD, coronary heart disease; MI, myocardial infarction. *High rates of diabetes incidence were obtained when the participants attended the screening clinic.

Table 4

Models for coronary events and diabetes with additional adjustment for baseline degree of obesity, waist circumference, and biomedical risk factors, after employment grade, smoking, and physical activity were controlled for, are shown in Table 5. Adjustment for obesity had a small attenuating effect on the hazard ratio for the healthy pattern with both outcomes. Further adjustment for blood pressure and lipid levels weakened the association between the healthy pattern and incident diabetes, but this was not the case for coronary events.

	DISCUSSION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

Our study provides support for a holistic concept of nutrition. Large trials of several antioxidant vitamin supplements have failed to show protection from cardiovascular disease and cancer, despite plausible mechanisms (4-6). For example, the heart protection study of 20 536 high-risk adults showed no benefit in terms of mortality from, or incidence of, vascular disease or cancer of vitamin E, vitamin C, or β-carotene supplementation, despite good compliance with the regimen and raised plasma vitamin concentrations (18). The US Physicians Study (>80 000 low-risk men) found similarly that supplements are ineffective alternatives to healthy eating in this respect. Among self-selected users of supplementation (vitamins C and E or multivitamins), there was no evidence of cardiovascular disease risk reduction (19). In contrast, trials of real diets combining salt reduction, increased fruit and vegetable consumption and other dietary modifications of the type associated with disease protection in the present study were effective in reducing blood pressure (7) and preventing diabetes (8).

Previous findings
Cluster analysis has been used before among British (20), Irish (21), Swedish (22), and American (23, 24) adults and in the multinational European Prospective Investigation into Cancer and Nutrition study (25). Typically, the method identifies 2 to 6 groups that share food or nutrient intake profiles, depending on the source population and analysis method. Our cross-sectional analysis (13) showed that the unhealthy pattern was linked to higher waist-hip ratios, lower HDL cholesterol, and higher triacylglycerol concentrations than was the healthy pattern, after control for employment grade.

We are aware of one previous prospective analysis that sought to link dietary cluster membership to survival and cardiovascular disease events (26). This inconclusive study used energy-yielding nutrients rather than food sources to identify clusters. Studies of the Mediterranean diet (1, 2), studies of other predefined healthy eating indexes, and factor analysis have been used to show that a prudent diet is associated with reduced risks of all-cause mortality (3, 27), self-reported diabetes (28), and CHD (29). In the present study, self-reports were complemented with repeat screening clinic visits, and incident disease detection was likely to have been more complete.

Methodologic considerations
The health status of the Whitehall II cohort is well characterized through the use of repeat screening and questionnaire contacts and record tracing (11). Measurement of social position is a particular strength of the study, and we show that after we accounted for potential confounding by several childhood and adult socioeconomic indicators and marital status, the inverse associations of the healthy dietary pattern with incident coronary events and diabetes remained strong (9). In addition, a sensitivity analysis in which socioeconomic differences within the cohort were constrained by omitting office support workers gave no support to the proposition that residual confounding might be responsible for the observed dietary effects. Selection bias is unlikely to have had a major impact on our findings. Although most clearly of concern for the diabetes outcome, differences between those included and excluded in the analysis were small on the basis of age and body mass index comparisons.

Confounding of the dietary effects by nondietary health behaviors and physiologic factors is a further important consideration. Adjustment for smoking status and moderate and vigorous leisure-time physical activity reduced the observed effect of the healthy pattern by about one-sixth, which is consistent with the health effects of these behaviors and their association with dietary behavior (2). A degree of residual confounding of the dietary effects may remain in the fully adjusted model as a result of measurement imprecision. Conversely, imprecision in assignment to dietary clusters would lead to underestimation of dietary effects. Further adjustments for degree of obesity, blood pressure, and serum lipid concentrations (after control for employment grade, smoking and physical activity) produced minor changes in hazard ratios for the healthy dietary pattern in relation to coronary events and diabetes. Controlling for risk factors for CHD and diabetes may constitute overadjustment, because diet-related reduction of disease risk is likely to be mediated at least in part by these biological factors.

In prospective studies, it is possible that the presence of disease at baseline may influence dietary and other behaviors over the period of follow-up, thereby confounding the observed dietary effects. By excluding participants with diabetes at baseline from the analysis of incident CHD, and vice versa, we found no evidence that such confounding operated.

Food-frequency questionnaires are prone to measurement error, and isolating the effect of a single nutrient on a health outcome in multivariate analysis is problematic (9, 30). The cluster analysis approach uses information on many food items to identify common dietary patterns, and those we observe have face validity. Energy misreporting was not systematically associated with the healthiness of the pattern, and in addition, adjustment for misreporting resulted in reassuringly small changes in hazard ratios. Adjustment for misreporting by using energy intake instead of the ratio of energy intake to energy expenditure produced similar results. Further, our validation study using biomarkers and a 7-d estimated food diary suggests the food-frequency questionnaire performed well in our cohort of conscientious civil servants (12).

This study lacked power to investigate dietary prevention of cancer. Relatively few deaths in the cohort were attributed to neoplastic disease, and of those, approximately one-third were classified as having important dietary causes on the basis of present evidence (31). Collaborative studies with pooling of cases by tumor type across cohorts will be needed to permit prospective investigations of the dietary causes of specific cancers.

Implications
Our study supports the view that dietary pattern has protective or detrimental effects on chronic disease risk. Dietary change as a result of health promotion advice is effective among motivated, high-risk individuals (32) but tends to be modest among others (33). Our findings add to the body of work showing the importance of dietary quality for population health.

Conclusions
A healthy dietary pattern that was a variant of actual food consumption habits was associated with lower rates of major coronary events and incident diabetes. Models involving adjustments for major socioeconomic and behavioral confounders suggest the effects observed are directly attributable to diet. Analyses with further adjustments for obesity, blood pressure, and lipids, each of which is likely to mediate the effects of a prudent diet, suggested that the observed diet-CHD and diet-diabetes associations are partially independent of these baseline characteristics. Measurement errors make assessment of the effects of individual nutrients difficult, and here we did not attempt to do this. Our study provides evidence for the benefits of a prudent dietary pattern, rather than the advantages or risks associated with individual foods or nutrients.

	ACKNOWLEDGMENTS

 
We thank all participating civil service departments and their welfare, personnel, and establishment officers; the Occupational Health and Safety Agency; the Council of Civil Service Unions; all participating civil servants in the Whitehall II study; and all members of the Whitehall II study team.

The contributions of the authors were as follows—EJB: conceived the study, analyzed the data, and wrote the article; AM: prepared the dietary data; DRW: prepared the diabetes data and survival curves; PM: conducted the cluster analysis; MS: prepared the area deprivation data; and MJS: prepared the mortality and coronary event data and advised on statistical methods. All authors commented on drafts of the paper. No author had any conflicts of interest.

	REFERENCES

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Trichopoulou A, Orfanos P, Norat T, et al. Modified Mediterranean diet and survival: EPIC-elderly prospective cohort study. BMJ 2005;330:991–5.[Abstract/Free Full Text]
2. Knoops KTB, de Groot LCPM, Kromhout D, et al. Mediterranean diet, lifestyle factors, and 10 year mortality in elderly European men and women. JAMA 2004;292:1433–9.[Abstract/Free Full Text]
3. Waijers PMCM, Ocke MC, van Rossum CTM, et al. Dietary patterns and survival in older Dutch women. Am J Clin Nutr 2006;83:1170–6.[Abstract/Free Full Text]
4. Vivekananthan DP, Penn MS, Sapp SK, Hsu A, Topol EJ. Use of antioxidant vitamins for the prevention of cardiovascular disease: meta-analysis of randomised trials. Lancet 2003;361:2017–23.[Medline]
5. Hercberg S, Galan P, Preziosi P, et al. The SU.VI.MAX Study: a randomized, placebo-controlled trial of the health effects of antioxidant vitamins and minerals. Arch Intern Med 2004;164:2335–42.[Abstract/Free Full Text]
6. Lee IM, Cook NR, Gaziano JM, et al. Vitamin E in the primary prevention of cardiovascular disease and cancer: the Women's Health Study: a randomized controlled trial. JAMA 2005;294:56–65.[Abstract/Free Full Text]
7. Sacks FM, Svetkey LP, Vollmer WM, et al. Effects on blood pressure of reduced dietary sodium and the dietary approaches to stop hypertension (DASH) diet. N Engl J Med 2001;344:3–10.[Abstract/Free Full Text]
8. Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–50.[Abstract/Free Full Text]
9. Lawlor DA, Davey Smith G, Bruckdorfer KR, Kundu D, Ebrahim S. Those confounded vitamins: what can we learn from the differences between observational versus randomised trial evidence? Lancet 2004;363:1724–7.[Medline]
10. Brunner EJ, Marmot MG, Nanchahal K, et al. Social inequality in coronary risk: central obesity and the metabolic syndrome. Evidence from the Whitehall II study. Diabetologia 1997;40:1341–9.[Medline]
11. Marmot MG, Brunner EJ. Cohort profile: the Whitehall II study. Int J Epidemiol 2005;34:251–6.[Free Full Text]
12. Brunner EJ, Stallone D, Juneja M, Bingham S, Marmot M. Dietary assessment in Whitehall II: comparison of 7d diet diary and food frequency questionnaire and validity against biomarkers. Br J Nutr 2001;86:405–14.[Medline]
13. Martikainen P, Brunner E, Marmot M. Socioeconomic differences in dietary patterns among middle-aged men and women. Soc Sci Med 2003;56:1397–410.[Medline]
14. Slimani N, Ferrari P, Ocke M, et al. Standardization of the 24-hour diet recall calibration method used in the European Prospective Investigation into Cancer and Nutrition (EPIC): general concepts and preliminary results. Eur J Clin Nutr 2000;54:900–17.[Medline]
15. Black AE. Critical evaluation of energy intake using the Goldberg cut-off for energy intake: basal metabolic rate. A practical guide to its calculation, use and limitations. Int J Obes 2000;24:1119–30.[Medline]
16. Schofield WN, Schofield C, James WPT. Basal metabolic rate. Hum Nutr Clin Nutr 1985;39C(suppl 1):1–96.
17. Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 1994;81:515–26.[Abstract/Free Full Text]
18. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of antioxidant vitamin supplementation in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:23–33.[Medline]
19. Muntwyler J, Hennekens CH, Manson JE, Buring JE, Gaziano JM. Vitamin supplement use in a low-risk population of US male physicians and subsequent cardiovascular mortality. Arch Intern Med 2002;162:1472–6.[Abstract/Free Full Text]
20. Pryer JA, Nichols R, Elliott P, Thakrar B, Brunner E, Marmot M. Dietary patterns among a national random sample of British adults. J Epidemiol Community Health 2001;55:29–37.[Abstract/Free Full Text]
21. Villegas R, Salim A, Flynn A, Perry IJ. Prudent diet and the risk of insulin resistance. Nutr Metab Cardiovasc Dis 2004;14:334–43.[Medline]
22. Wirfalt E, Mattisson I, Gullberg B, Berglund G. Food patterns defined by cluster analysis and their utility as dietary exposure variables: a report from the Malmo diet and cancer study. Public Health Nutr 2000;3:159–73.[Medline]
23. Millen BE, Quatromoni PA, Nam BH, O'Horo CE, Polak JF, D'Agostino RB. Dietary patterns and the odds of carotid atherosclerosis in women: the Framingham Nutrition Studies. Prev Med 2002;35:540–7.[Medline]
24. Popkin BM, Barclay DV, Nielsen SJ. Water and food consumption patterns of U.S. adults from 1999 to 2001. Obes Res 2005;13:2146–52.[Medline]
25. Bamia C, Orfanos P, Ferrari P, et al. Dietary patterns among older Europeans: the EPIC-Elderly study. Br J Nutr 2005;94:100–13.[Medline]
26. Diehr P, Beresford SA. The relation of dietary patterns to future survival, health, and cardiovascular events in older adults. J Clin Epidemiol 2003;56:1224–35.[Medline]
27. Huijbregts PPCW, Feskens EJM, Kromhout D. Dietary patterns and cardiovascular risk factors in elderly men: The Zutphen elderly study. Int J Epidemiol 1995;24:313–20.[Abstract/Free Full Text]
28. van Dam RM, Rimm EB, Willett WC, Stampfer MJ, Hu FB. Dietary patterns and risk for type 2 diabetes mellitus in U.S. men. Ann Intern Med 2002;136:201–9.[Abstract/Free Full Text]
29. Hu FB, Rimm EB, Stamper MJ, Ascherio A, Spiegelman D, Willett WC. Prospective study of major dietary patterns and risk of coronary heart disease in men. Am J Clin Nutr 2000;72:912–21.[Abstract/Free Full Text]
30. Day NE, Wong MY, Bingham S, et al. Correlated measurement error–implications for nutritional epidemiology. Int J Epidemiol 2004;33:1373–81.[Abstract/Free Full Text]
31. Key TJ, Schatzkin A, Willett WC, Allen NE, Spencer EA, Travis RC. Diet, nutrition and the prevention of cancer. Public Health Nutr 2004;7:187–200.[Medline]
32. Brunner EJ, Rees K, Burke M, Thorogood M. Dietary advice for reducing cardiovascular risk. Cochrane Database Syst Rev 2007;CD002128.
33. Howard BV, Van Horn L, Hsia J, et al. Low-fat dietary pattern and risk of cardiovascular disease: the Women's Health Initiative Randomized Controlled Dietary Modification Trial. JAMA 2006;295:655–66.[Abstract/Free Full Text]

This article has been cited by other articles:

				S. Eilat-Adar, M. Mete, E. D. Nobmann, J. Xu, R. R. Fabsitz, S. O. E. Ebbesson, and B. V. Howard Dietary Patterns are Linked to Cardiovascular Risk Factors but Not to Inflammatory Markers in Alaska Eskimos J. Nutr., December 1, 2009; 139(12): 2322 - 2328. [Abstract] [Full Text] [PDF]

				W. You, G. Zhang, B. M. Davy, A. Carlson, and B.-H. Lin Food Consumed Away from Home Can Be a Part of a Healthy and Affordable Diet J. Nutr., October 1, 2009; 139(10): 1994 - 1999. [Abstract] [Full Text] [PDF]

				S. M. Nilsen, S. Krokstad, T. L. Holmen, and S. Westin Adolescents' health-related dietary patterns by parental socio-economic position, The Nord-Trondelag Health Study (HUNT) Eur J Public Health, September 18, 2009; (2009) ckp137v1. [Abstract] [Full Text] [PDF]

				J. L. Kuk, C. I. Ardern, T. S. Church, J. R. Hebert, X. Sui, and S. N. Blair Ideal Weight and Weight Satisfaction: Association With Health Practices Am. J. Epidemiol., August 15, 2009; 170(4): 456 - 463. [Abstract] [Full Text] [PDF]

				D. S. Lee, M. Chiu, D. G. Manuel, K. Tu, X. Wang, P. C. Austin, M. Y. Mattern, T. F. Mitiku, L. W. Svenson, W. Putnam, et al. Trends in risk factors for cardiovascular disease in Canada: temporal, socio-demographic and geographic factors Can. Med. Assoc. J., August 4, 2009; 181(3-4): E55 - E66. [Abstract] [Full Text] [PDF]

				M. L. Kwan, E. Weltzien, L. H. Kushi, A. Castillo, M. L. Slattery, and B. J. Caan Dietary Patterns and Breast Cancer Recurrence and Survival Among Women With Early-Stage Breast Cancer J. Clin. Oncol., February 20, 2009; 27(6): 919 - 926. [Abstract] [Full Text] [PDF]

				A. Tremblay and J.-A. Gilbert Milk Products, Insulin Resistance Syndrome and Type 2 Diabetes J. Am. Coll. Nutr., February 1, 2009; 28(Supplement_1): 91S - 102S. [Abstract] [Full Text] [PDF]

				WRITING GROUP MEMBERS, D. Lloyd-Jones, R. Adams, M. Carnethon, G. De Simone, T. B. Ferguson, K. Flegal, E. Ford, K. Furie, A. Go, et al. Heart Disease and Stroke Statistics--2009 Update: A Report From the American Heart Association Statistics Committee and Stroke Statistics Subcommittee Circulation, January 27, 2009; 119(3): e21 - e181. [Full Text] [PDF]

				W. K Al-Delaimy Commentary: Lactose and ischaemic heart disease: a weak 28-year-old hypothesis Int. J. Epidemiol., December 1, 2008; 37(6): 1214 - 1216. [Full Text] [PDF]

				K. Esposito and D. Giugliano Mediterranean dietary patterns and chronic diseases Am. J. Clinical Nutrition, October 1, 2008; 88(4): 1179 - 1180. [Full Text] [PDF]

				E. Brunner, D. Witte, M. Shipley, M. Marmot, A. Mosdol, and P. Martikainen Reply to K Esposito and D Giugliano Am. J. Clinical Nutrition, October 1, 2008; 88(4): 1180 - 1181. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Brunner, E. J Articles by Marmot, M. G Search for Related Content PubMed PubMed Citation Articles by Brunner, E. J Articles by Marmot, M. G Agricola Articles by Brunner, E. J Articles by Marmot, M. G