Long-term observational cohorts provide the opportunity to investigate the potential impact of dietary patterns on death. We aimed to investigate all-cause death according to the consumption of selected food groups, and then to identify those independently associated with reduced mortality.
Population survey of middle-aged men randomly selected in the period 1995–1997 from the general population of three French areas and followed over a median of 14.8 years. Dietary data were collected through a 3-day food record. Cox modeling was used to assess the risk of death according to selected foods groups after extensive adjustment for confounders, including a diet quality index.
The study population comprised 960 men (mean age 55.5 ±6.2 years). After a median follow-up of 14.8 (interquartile range 14.3–15.2) years, 150 (15.6%) subjects had died. Food groups that remained independently predictive of a lower risk of death after extensive adjustment were an above-median consumption of milk (adjusted relative risk: 0.61, 95% confidence interval (CI): 0.43–0.86, P-value=0.005), fruits and vegetables (0.68, 0.46–0.98, P-value=0.041) and a moderate consumption of yogurts and cottage cheese (0.50, 95% CI: 0.31–0.81, P-value=0.005), other cheeses (0.62, 0.39–0.97, P-value=0.036) and bread (0.57, 0.37–0.89, P-value=0.014). Besides, there was a nonsignificant trend for a higher risk of death associated with highest sodium intakes.
Consumption of food groups that largely match recommendations is associated with a reduced risk of all-cause death in men. A diet providing moderate amounts of diverse food groups appears associated with the highest life expectancy.
Recommendations for a healthy diet have been developed in numerous countries,1, 2, 3, 4, 5 in part, based on randomized clinical trials showing the impact of foods and nutrients on body weight, blood pressure and biological parameters.6 However, long-term ‘hard’ clinical end points will always be difficult to investigate through randomized clinical trials, although some have been conducted,7, 8 because ensuring perfect compliance to a specific diet over a long period is almost impossible. Consequently, observational cohorts with long-term follow-up on clinical end points are needed. As dietary patterns are closely linked to cultural habits that vary across countries,9 analyses conducted in cohorts from different origins provide valuable information to help adapt national recommendations to local food culture.
The aim of this analysis, using quantitative dietary data from a population survey of middle-aged men in three areas of France followed over a median period of 15 years, was to investigate all-cause death according to the consumption of different food groups and then to identify selected food groups that are independently associated with a reduced risk of death.
Materials and methods
Baseline data collection, including food consumption record, was performed between 1995 and 1997 at the three French centers involved in the MONICA (MONItoring of trends and determinants in CArdiovascular disease) Project.10, 11 People living in areas of Lille (northern France), Strasbourg (northeastern France) and Toulouse (southwestern France) were invited to participate. Participants were selected randomly from the population, using data from polling lists of French inhabitants aged ⩾18 years in the three areas.12, 13 Sixty percent of the individuals contacted agreed to participate. A dietary record was performed in the subsample of 991 men aged 45–64 years. The data from 31 (3.1%) of these individuals could not be analyzed because of incomplete baseline evaluations, leaving a study sample size of 960 men. Participation was voluntary and financial compensation was not given to the participants.
Vital status (and date of death when applicable) was obtained for all the participants on 31 December 2010 through the French national database, which records all the deaths in French nationals.14 Permission to use these data was obtained as required by the French law.
The study protocol was approved by the local ethics committee and was conducted in accordance with the French law on human biomedical research. All the subjects provided informed consent to participate in the survey.
Each subject answered an extensive questionnaire during a face-to-face guided interview with trained medical staff. The data were collected on age, socioeconomic status (number of completed years of schooling and eligibility to pay income tax), history of medication use, cardiovascular risk factors, chronic diseases (coded according to the International Classification of Diseases) and lifestyle factors including tobacco use, alcohol consumption and physical activity. The levels of physical activity were assessed and categorized as light (less than once per week), moderate (⩾20 min once or twice each week) and high (⩾20 min on ⩾3 occasions each week). The Framingham risk equation was used to assess the level of cardiovascular risk.15 Serious chronic conditions included a history of cardiovascular disease, cancer (excluding in situ malignant tumors), cirrhosis, hemorrhagic stroke, chronic cardiac failure and chronic renal failure.
Clinical and laboratory measurements
Trained research nurses took the clinical measurements using standard procedures, and included height, weight, waist and hip circumferences and waist:hip ratio. Blood pressure was measured twice, with the subject in the sitting position, on the right arm and after 5 min of rest. Measurements were rounded up to the nearest 2 mm Hg and the mean value was used. Plasma glucose, triglycerides, total cholesterol and high-density lipoprotein cholesterol concentrations were measured in blood samples drawn after a ⩾10-h overnight fast. Low-density lipoprotein cholesterol was calculated according to the Friedewald formula.16
Assessment of dietary and alcohol intakes
The intakes of food and alcohol were assessed using a 3-day food record method, the details of which have been published previously.11, 12, 17, 18 In brief, subjects kept a record of the types and amounts of all foods and beverages (alcoholic and nonalcoholic) consumed throughout 3 consecutive days in the week following the clinical examination. A dietitian interviewed subjects subsequently to verify the reliability of the data. Photographs and household measures were used to estimate the portion sizes. The nutrient value of foods was determined using food composition tables, as previously described.11, 12, 18 Intakes were expressed in grams, milligrams or milliliters per day, by calculating the average consumption over the 3-day record. Intakes were also expressed in grams, milligrams or milliliters per 1000 kJ of daily energy intake and per day. Intakes were described in terms of macronutrients, micronutrients and food groups. Alcohol consumption was divided into no consumption (subjects who had never regularly consumed alcohol), moderate intake (⩽30 g/day or three glasses), high intake (>30 g/day or more than three glasses) and former consumption (subject who used to regularly consume alcohol but were no longer consumers).
Patients’ characteristics were described at baseline. Then, unadjusted death rates were compared between quartiles of food consumption using the chi-squared test. Assumptions were checked and satisfied. All the tests were two sided.
Baseline dietary predictors of all-cause death were determined by Cox proportional hazard modeling. Subjects were censored either on their date of death or on 31 December 2010, if they were still alive at that point. Multivariate models were systematically adjusted for center, age, payment of income tax, obesity (body mass index ⩾30 kg/m2), alcohol consumption (no consumption, moderate intake, high intake or former consumption), smoking habits (never, past and current smoking), physical activity (light, moderate and high), presence of a serious chronic condition and a diet quality index evaluating adherence to the French guidelines for healthy food choices (Programme National Nutrition Santé).19, 20 These adjusting covariates were chosen as they were significantly associated with all-cause death in univariate Cox models or as they have been described as potential determinants of all-cause death in the literature. To avoid overadjustment, when two potential covariates were strongly linked (such as payment of income tax and educational level, or Framingham risk score and serious chronic condition), only one covariate was considered for model adjustment. The proportional hazard assumption was tested for each covariate and could not be rejected.
Multivariate Cox models were first built to investigate the association of each food group with all-cause death. The lowest quartile of food consumption was systematically chosen as the reference to compute the hazard ratio (HR) for all-cause death (for alcohol and legume intakes, no consumption was chosen as the reference class). Each model was adjusted as described above. This analysis identified food groups with at least one quartile of consumption associated with a significantly reduced risk of all-cause death, or with a trend (P-value <0.10) for a reduced risk with higher consumptions. These were milk, yogurts and cottage cheese, all cheeses, except cottage cheese, fruits and vegetables, cereals, bread, sugar and sweets and sodium intake. These food groups were finally considered jointly as explanatory variables, in one single adjusted Cox model assessing risk of all-cause death. Consumptions of cereals, sugar and sweets and sodium were removed from this final model as they were no longer significantly associated with death after adjustment for the other dietary covariates. Correspondences between the quartiles of food consumption associated with a reduced risk of death and the number of servings per day were estimated, based on the following assumptions: 150±75 ml for a serving of milk, 120±60 g for a serving of yogurt or cottage cheese, 30±15 g for a serving of cheese (except cottage cheese), 100±50 g for a serving of fruits and vegetables and 30±15 g for a serving of bread.
Of the 976 men who responded to the survey, 960 had complete data and comprise the study population. The baseline characteristics of these subjects are shown in Table 1, and the distribution of dietary consumptions at baseline is shown in Table 2. The median-declared calorie intake was 9856 kJ/day, 38.2% provided by carbohydrates, 36.7% by lipids, 15.3% by proteins and 7.5% by alcohol. Median intakes of meat (168 g/day), yogurts and cottage cheese (41 g/day), other cheeses (40 g/day), fruits (138 g/day), vegetables (180 g/day) and bread (120 g/day) approximately corresponded to 1.5 servings of meat per day, less than one yogurt or portion of cottage cheese, one portion of cheese, one to two fruits, one to two servings of vegetables and four portions of bread per day. Over one-third of the sample (35.4%) did not consume fish or seafood during the 3-day food record. Median sodium intake was 2061 mg/day, corresponding to 5.9 g/day of salt.
Mortality and predictors of death
Vital status was recorded for all subjects in 2010, after a median follow-up of 14.8 years (interquartile range: 14.3–15.2; 13 394 person-years). Of the 960 men enrolled, 150 (15.6%) had died (50% from cancer, 31.3% from a cardiovascular cause and 18.7% from other causes).
Table 3 shows subjects’ baseline characteristics associated with mortality. Death rates were higher in northern France, in older people, among subjects who did not complete secondary school and in those who did not pay income tax. Obesity (body mass index ⩾30 kg/m2), smoking (current or past), hypertension (assessed by the use of antihypertensive drugs), a history of cardiovascular disease, presence of a serious chronic condition and intermediate-to-high Framingham risk score were all associated with an increased risk of death, whereas a reduced risk was observed among subjects who were moderately active. The diet quality score was inversely associated with death (lower risk with better adherence to recommendations).
The raw rates and adjusted HRs of death according to the quartile of food group intake are given in Table 4. Men with a consumption of milk in the upper two quartiles had a reduced risk of all-cause death. For yogurts and cottage cheese, all other cheeses, cereals and bread, quartile 3 was characterized by a significantly reduced risk of death. Consumptions of fruits and vegetables and sugar and sweets also tended to be associated with a lower mortality risk (P-value for trend: 0.07 and 0.039, respectively). Finally, high sodium intake was associated with higher risk of death.
Independent predictors of all-cause death are shown in Table 5. After adjustment for confounders, food groups that remained predictive of a lower risk of death were highest consumptions of milk, and fruits and vegetables (upper third and fourth quartiles compared with first and second) and moderate consumptions of yogurts and cottage cheese, other cheeses and bread (third quartile compared with other quartiles). Consumptions of cereals, sugar and sweets and sodium were no longer significantly associated with risk of death, although the relationship was border significant for subjects in the highest quartile of sodium intake compared with other people (HR: 1.46; 95% CI: 0.99–2.14; P-value 0.053). Sensitivity analyses were conducted after exclusion of people with a serious chronic condition at baseline and people who died during the first year (only five subjects). These exclusions did not significantly affect the results (⩽15% change in the HRs associated with the dietary variables).
Corresponding numbers of usual portion sizes are given in Table 6, which shows the distribution of the approximate number of servings per day of milk, yogurts and cottage cheese, other cheeses, fruits and vegetables and bread, in these categories associated with a reduced risk of death. The median consumptions in these categories were one glass of milk, one yogurt or cottage cheese, two servings of other cheeses, five servings of fruits or vegetables and five servings of bread per day.
Our data indicate an association between selective food groups and improved longevity. Men in the upper two quartiles for the consumption of milk (corresponding to a median intake of one glass per day), or fruits and vegetables (five servings), and those in the third quartile for the consumption of yogurts and cottage cheese (one serving), other cheeses (two servings) and bread (five servings), had significantly reduced relative risks of all-cause death over 15 years of follow-up. Our findings show close concordance with, and offer support for, current recommendations1, 2, 3, 4, 5 for a healthy diet, which includes the daily intake of three dairy products, four to six servings of fruits or vegetables and starchy food at every meal. As in previous studies, we report a north–south gradient in all-cause death,21, 22 which is believed to reflect patterns of behavior in relation to food intakes but also to social inequalities, with a lower average socioeconomic level and reduced access to health care in northern France compared with southern France.18, 21, 22, 23 As expected, other non-dietary independent predictors of death were age, low income and smoking. Risk of death was nonsignificantly increased among subjects with a serious chronic condition at baseline, likely because of a lack of power, as only 82 subjects suffered from a serious chronic condition.
Milk, dairy and cheese
The results from published epidemiological data on the impact of dairy products on incident diseases and mortality vary.24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34 Some studies have indicated that death from cardiovascular disease and cancer may be modified by the intake of dairy products. In line with our results, recent studies and meta-analyses have indicated a lack of harm from the consumption of milk and dairy foods,24, 25, 26, 27 and one has suggested a benefit in terms of reduced cardiovascular risk.28 Several mechanisms have been proposed by which dairy products could offer cardiovascular health benefits: the lowering effect of calcium on blood pressure, through decreasing vascular resistance and sodium retention,35 the angiotensin-converting enzyme inhibitor effect of bioactive peptides on hypertension36 and the reduced fat absorption associated with calcium intake because of soap formation in the gut in the presence of fatty acids and calcium.24
Several international guidelines advocate the consumption of low-fat as opposed to high-fat dairy products.2, 5 In accordance, we have recently shown that low-fat dairy products are associated with lower levels of low-density lipoprotein cholesterol.37 The rationale for promoting low-fat dairies is the fear that saturated fat intake may be a factor in the development of high cholesterol, obesity and heart disease. The present data do not support a deleterious impact of cheese (which is considered rich in saturated fats) on mortality, as long as it is consumed moderately (two servings per day). Besides, unlike what has been described in a recent Swedish cohort,38 our data do not support either a deleterious impact of milk consumption on mortality. This discrepancy is possibly explained by the fact that the median milk consumption reached only one to two glasses per day, for the greatest consumers, in our French cohort, whereas this level of consumption was considered as a small intake in the Swedish cohort.38
Fruits, vegetables and bread
The results of our study show a 32% relative risk reduction in all-cause death for men in the upper two quartiles of consumption of fruits and vegetables and a 43% relative risk reduction for those in the third quartile for bread intake (corresponding to a median consumption of five slices of bread per day). In France, white bread is preferentially consumed as a side dish, sandwiches being more casually consumed.
Foods high in dietary fiber such as fruits and vegetables, through their bulk and relative low energy density, may help in weight control and are associated with a reduced risk of all-cause death,39, 40 coronary heart disease and cardiovascular disease,41 metabolic syndrome, diabetes mellitus42 and some cancers.43 Vegetables and fruits offer a source of vitamins, minerals and other bioactive compounds, including phytochemicals,44 which have potential health benefits through, for example, their blood pressure reducing or antioxidant properties.45 Meta-analyses are fairly consistent in their findings of a reduced risk of coronary heart disease, but particularly of stroke, in individuals who report a diet high in fruits and vegetables.46, 47
Other foods and nutrients
Although our study did not show a significant association for all-cause death and the level of fish consumption, we did see a nonsignificant inverse trend for all-cause death and fish intake and a positive but nonsignificant association between high sodium intake and high risk of death, likely related to increased blood pressure levels. Of interest, the DASH (Dietary Approaches to Stop Hypertension) diet demonstrated a reduction in blood pressure for patients randomized to a diet rich in fruits, vegetables and low-fat dairy products and reduced saturated and total fat compared with a control diet (rich in fruits and vegetables but not in low-fat dairy products), with no change in either sodium intake or body weight.48 The combination of a diet with reduced sodium intake further reduced systolic blood pressure compared with the control diet.49
Finally, the lower risk of death observed among high consumers of sugar and sweets is quite unexpected. However, the relationship is markedly attenuated by adjustment for other food intakes, in particular for dairy products, which are one of the main sources of sugars after sweet products and substitutes, fruits and cereals, the latter ones being frequently consumed with milk products.50
Strengths and limitations
This observational study is subject to several limitations. The study was restricted to men aged 45–64 years; therefore, the results cannot be extrapolated to the female population or to older or younger men. Men who agreed to participate in the study may be more concerned about following a healthy lifestyle and may consequently exhibit different patterns of dietary intake compared with those who declined. The study may be subject to bias from a healthy cohort effect. The individuals’ dietary habits may have been associated with other confounding factors not included in the study. In particular, subjects who make healthy food choices may also follow healthier non-dietary lifestyle habits. Even if we adjusted extensively for confounders, the lack of randomization makes it impossible to fully control for potential biases. Another limitation is the lack of reassessment of food intakes during follow-up. Besides, using a food diary for collecting dietary intakes may induce changes in dietary habits, leading to record intakes that are not representative of the usual diet. Finally, quite a large number of associations were tested between food groups and mortality. This may have affected the validity of the results by providing false associations (increase in type I error).
The main strengths of the study include the multicenter prospective design and the long-term follow-up. Also, the end point of all-cause death is regarded as a ‘hard’ end point; no subjects were lost to follow-up; and we adjusted for the main determinants of death.
In conclusion, our long-term study shows that consumption of selected food groups that match recommendations in terms of daily intakes of dairy products, fruits and vegetables and, to a certain extent, bread and starchy food, is associated with a reduced risk of all-cause death in men. More generally, a diet providing moderate amounts of diverse food groups seems to be associated with the highest life expectancy. However, as data are non-randomized, the analyses should be considered as exploratory and require further investigation in randomized trials.
Perk J, De Backer G, Gohlke H, Graham I, Reiner Z, Verschuren M et al. European Guidelines on cardiovascular disease prevention in clinical practice (version 2012). The Fifth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of nine societies and by invited experts). Eur Heart J 2012; 33: 1635–1701.
American Heart Association Nutrition Committee, Lichtenstein AH, Appel LJ, Brands M, Carnethon M, Daniels S, Franch HA et al. Diet and lifestyle recommendations revision 2006: a scientific statement from the American Heart Association Nutrition Committee. Circulation 2006; 114: 82–96.
Hercberg S, Chat-Yung S, Chaulia M . The French National Nutrition and Health Program: 2001-2006-2010. Int J Public Health 2008; 53: 68–77.
French Ministry of Health. French National Nutrition and Health Program—2011-2015. Available from http://social-sante.gouv.fr/IMG/pdf/PNNS_UK_INDD_V2.pdf (accessed 14 April 2015.
Eckel RH, Jakicic JM, Ard JD, Hubbard VS, de Jesus JM, Lee IM et al. 2013 AHA/ACC Guideline on Lifestyle Management to Reduce Cardiovascular Risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014; 129: S76–S99.
Svetkey LP, Simons-Morton D, Vollmer WM, Appel LJ, Conlin PR, Ryan DH et al. Effects of dietary patterns on blood pressure: subgroup analysis of the Dietary Approaches to Stop Hypertension (DASH) randomized clinical trial. Arch Intern Med 1999; 159: 285–293.
Schaefer EJ . Lipoproteins, nutrition, and heart disease. Am J Clin Nutr 2002; 75: 191–212.
Estruch R, Ros E, Salas-Salvado J, Covas MI, Corella D, Aros F et al. Primary prevention of cardiovascular disease with a Mediterranean diet. New Engl J Med 2013; 368: 1279–1290.
Naska A, Fouskakis D, Oikonomou E, Almeida MD, Berq MA, Gedrich K et al. Dietary patterns and their socio-demographic determinants in 10 European countries: data from the DAFNE databank. Eur J Clin Nutr 2006; 60: 181–190.
Kuulasmaa K, Tunstall-Pedoe H, Dobson A, Fortmann S, Sans S, Tolonen H et al. Estimation of contribution of changes in classic risk factors to trends in coronary-event rates across the WHO MONICA Project populations. Lancet 2000; 355: 675–687.
Ruidavets JB, Bongard V, Dallongeville J, Arveiler D, Ducimetiere P, Perret B et al. High consumptions of grain, fish, dairy products and combinations of these are associated with a low prevalence of metabolic syndrome. J Epidemiol Community Health 2007; 61: 810–817.
Perrin AE, Simon C, Hedelin G, Arveiler D, Schaffer P, Schlienger JL . Ten-year trends of dietary intake in a middle-aged French population: relationship with educational level. Eur J Clin Nutr 2002; 56: 393–401.
Ruidavets JB, Bataille V, Dallongeville J, Simon C, Bingham A, Amouyel P et al. Alcohol intake and diet in France, the prominent role of lifestyle. Eur Heart J 2004; 25: 1153–1162.
Institut National de la Statistique et des Etudes Economiques (INSEE). Internet. Available from http://www.insee.fr/fr/methodes/default.asp?page=definitions/rnipp.htm (accessed 14 April 2015).
Wilson PW, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB . Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97: 1837–1847.
Friedewald WT, Levy RI, Fredrickson DS . Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972; 18: 499–502.
Evans AE, Ruidavets JB, McCrum EE, Cambou JP, McClean R, Douste-Blazy P et al. Autres pays, autres coeurs? Dietary patterns, risk factors and ischaemic heart disease in Belfast and Toulouse. QJM 1995; 88: 469–477.
Perrin AE, Dallongeville J, Ducimetiere P, Ruidavets JB, Schlienger JL, Arveiler D et al. Interactions between traditional regional determinants and socio-economic status on dietary patterns in a sample of French men. Br J Nutr 2005; 93: 109–114.
Kesse-Guyot E, Castetbon K, Estaquio C, Czernichow S, Galan P, Hercberg S . Association between the French nutritional guideline-based score and 6-year anthropometric changes in a French middle-aged adult cohort. Am J Epidemiol 2009; 170: 757–765.
Kesse-Guyot E, Touvier M, Henegar A, Czernichow S, Galan P, Hercberg S et al. Higher adherence to French dietary guidelines and chronic diseases in the prospective SU.VI.MAX cohort. Eur J Clin Nutr 2011; 65: 887–894.
Berard E, Bongard V, Arveiler D, Amouyel P, Wagner A, Dallongeville J et al. Ten-year risk of all-cause mortality: assessment of a risk prediction algorithm in a French general population. Eur J Epidemiol 2011; 26: 359–368.
Ferrieres J . The French paradox: lessons for other countries. Heart 2004; 90: 107–111.
Cottel D, Dallongeville J, Wagner A, Ruidavets JB, Arveiler D, Ferrieres J et al. The North-East-South gradient of coronary heart disease mortality and case fatality rates in France is consistent with a similar gradient in risk factor clusters. Eur J Epidemiol 2000; 16: 317–322.
Elwood PC, Pickering JE, Givens DI, Gallacher JE . The consumption of milk and dairy foods and the incidence of vascular disease and diabetes: an overview of the evidence. Lipids 2010; 45: 925–939.
Louie JC, Flood VM, Burlutsky G, Rangan AM, Gill TP, Mitchell P . Dairy consumption and the risk of 15-year cardiovascular disease mortality in a cohort of older australians. Nutrients 2013; 5: 441–454.
Patterson E, Larsson SC, Wolk A, Akesson A . Association between dairy food consumption and risk of myocardial infarction in women differs by type of dairy food. J Nutr 2013; 143: 74–79.
Bonthuis M, Hughes MC, Ibiebele TI, Green AC, van der Pols JC . Dairy consumption and patterns of mortality of Australian adults. Eur J Clin Nutr 2010; 64: 569–577.
Elwood PC, Givens DI, Beswick AD, Fehily AM, Pickering JE, Gallacher J . The survival advantage of milk and dairy consumption: an overview of evidence from cohort studies of vascular diseases, diabetes and cancer. J Am Coll Nutr 2008; 27: 723S–734S.
Elwood PC, Pickering JE, Fehily AM . Milk and dairy consumption, diabetes and the metabolic syndrome: the Caerphilly prospective study. J Epidemiol Community Health 2007; 61: 695–698.
Ness AR, Smith GD, Hart C . Milk coronary heart disease and mortality. J Epidemiol Community Health 2001; 55: 379–382.
Tholstrup T . Dairy products and cardiovascular disease. Curr Opin Lipidol 2006; 17: 1–10.
Elwood PC, Pickering JE, Hughes J, Fehily AM, Ness AR . Milk drinking, ischaemic heart disease and ischaemic stroke II. Evidence from cohort studies. Eur J Clin Nutr 2004; 58: 718–724.
Soedamah-Muthu SS, Ding EL, Al-Delaimy WK, Hu FB, Engberink MF, Willett WC et al. Milk and dairy consumption and incidence of cardiovascular diseases and all-cause mortality: dose-response meta-analysis of prospective cohort studies. Am J Clin Nutr 2011; 93: 158–171.
Cho E, Smith-Warner SA, Spiegelman D, Beeson WL, van den Brandt PA, Colditz GA et al. Dairy foods, calcium, and colorectal cancer: a pooled analysis of 10 cohort studies. J Natl Cancer Inst 2004; 96: 1015–1022.
Zemel EB . Calcium modulation of hypertension and obesity: mechanisms and implications. J Am Coll Nutr 2001; 20 (Suppl 5), 428S–435S.
Fitzgerald RJ, Meisel H . Milk protein-derived peptide inhibitors of angiotensine-I-converting enzyme. Br J Nutr 2000; 84 (Suppl 1), 33S–37S.
Kai SH, Bongard V, Simon C, Ruidavets JB, Arveiler D, Dallongeville J, Wagner A, Amouyel P, Ferrieres J . Low-fat and high-fat dairy products are differently related to blood lipids and cardiovascular risk score. Eur J Prev Cardiol 2014; 21: 1557–1567.
Michaelsson K, Wolk A, Langenskiold S, Basu S, Warensjo Lemming E et al. Milk intake and risk of mortality and fractures in women and men: cohort studies. Br Med J 2014; 349: g6015.
Mozaffarian D, Hao T, Rimm EB, Willett WC, Hu FB . Changes in diet and lifestyle and long-term weight gain in women and men. New Engl J Med 2011; 364: 2392–2404.
National Research Council Dietary Reference Intakes: the Essential Guide to Nutrient Requirements. National Academies Press: Washington, DC, USA, 2006.
Threapleton DE, Greenwood DC, Evans CE, Cleghorn CL, Nykjaer C, Woodhead C et al. Dietary fibre intake and risk of cardiovascular disease: systematic review and meta-analysis. Br Med J 2013; 347: f6879.
Weickert MO, Pfeiffer AF . Metabolic effects of dietary fiber consumption and prevention of diabetes. J Nutr 2008; 138: 439–442.
Aune D, Chan DS, Lau R, Vieira R, Greenwood DC, Kampman E et al. Dietary fibre, whole grains, and risk of colorectal cancer: systematic review and dose-response meta-analysis of prospective studies. Br Med J 2011; 343: d6617.
Rodriguez-Casado A . The health potential of fruits and vegetables phytochemicals: notable examples. Crit Rev Food Sci Nutr; e-pub ahead of print 16 September 2014; doi:10.1080/10408398.2012.755149.
He FJ, MacGregor GA . Fortnightly review: beneficial effects of potassium. Br Med J 2001; 323: 497–501.
Dauchet L, Amouyel P, Hercberg S, Dallongeville J . Fruit and vegetable consumption and risk of coronary heart disease: a meta-analysis of cohort studies. J Nutr 2006; 136: 2588–2593.
He FJ, Nowson CA, MacGregor GA . Fruit and vegetable consumption and stroke: meta-analysis of cohort studies. Lancet 2006; 367: 320–326.
Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. New Engl J Med 1997; 336: 1117–1124.
Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D et al. Effects on blood pressure of reduced dietary sodium and the Dietary Approaches to Stop Hypertension (DASH) diet. DASH-Sodium Collaborative Research Group. New Engl J Med 2001; 344: 3–10.
Sette S, Le Donne C, Piccinelli R, Mitura L, Ferrari M, Lecercq C . The third National Food Consumption Survey, INRAN-SCAI 2005-06: major dietary sources of nutrients in Italy. Int J Food Sci Nutr 2013; 64: 1014–1021.
We thank all the investigators of the MONICA Project for their contribution to the compilation, validation and analysis of the data. We are grateful to the Institut National de la Statistique et des Etudes Economiques (INSEE) and the three regional health centers for their collaboration. Sophie Rushton-Smith provided writing support and was funded by the authors. The study received an unrestricted grant from the Centre National Interprofessionnel de l’Economie Laitière (CNIEL).
The authors declare no conflict of interest.
About this article
Cite this article
Bongard, V., Arveiler, D., Dallongeville, J. et al. Food groups associated with a reduced risk of 15-year all-cause death. Eur J Clin Nutr 70, 715–722 (2016). https://doi.org/10.1038/ejcn.2016.19
High vs. low-fat dairy and milk differently affects the risk of all-cause, CVD, and cancer death: A systematic review and dose-response meta-analysis of prospective cohort studies
Critical Reviews in Food Science and Nutrition (2021)
Fruit and Vegetable Intake and Mortality: Results From 2 Prospective Cohort Studies of US Men and Women and a Meta-Analysis of 26 Cohort Studies
Potato consumption and risk of all cause, cancer and cardiovascular mortality: a systematic review and dose-response meta-analysis of prospective cohort studies
Critical Reviews in Food Science and Nutrition (2020)
Yogurt Intake Reduces All-Cause and Cardiovascular Disease Mortality: A Meta-Analysis of Eight Prospective Cohort Studies
Chinese Journal of Integrative Medicine (2020)
Canadian Journal of Cardiology (2019)