The beneficial association of the Mediterranean diet (MedDiet) with longevity has been consistently demonstrated, but the associations of MedDiet components have not been accordingly evaluated. We performed an updated meta-analysis of prospective cohort studies published up to 31 December 2017, to quantify the association of adherence to MedDiet, expressed as an index/score (MDS) and of its components with all-cause mortality. We estimated summary relative risks (SRR) and 95 % CI using random effects models. On the basis of thirty studies (225 600 deaths), SRR for the study-specific highest/lowest and per 1sd MDS increment were 0·79 (95 % CI 0·77, 0·81, Ι2=42 %, P-heterogeneity 0·02) and 0·92 (95 % CI 0·90, 0·94, Ι2 56 %, P-heterogeneity <0·01), respectively. Inversely, statistically significant associations were evident in stratified analyses by country, MDS range and publication year, with some evidence for heterogeneity across countries overall (P-heterogeneity 0·011), as well as across European countries (P=0·018). Regarding MDS components, relatively stronger and statistically significant inverse associations were highlighted for moderate/none-excessive alcohol consumption (0·86, 95 % CI 0·77, 0·97) and for above/below-the-median consumptions of fruit (0·88, 95 % CI 0·83, 0·94) and vegetables (0·94, 95 % CI 0·89, 0·98), whereas a positive association was apparent for above/below-the-median intake of meat (1·07, 95 % CI 1·01, 1·13). Our meta-analyses confirm the inverse association of MedDiet with mortality and highlight the dietary components that influence mostly this association. Our results are important for better understanding the role of MedDiet in health and proposing dietary changes to effectively increase adherence to this healthy dietary pattern.

Improvements in colorectal cancer (CRC) detection and treatment have led to greater numbers of CRC survivors, for whom there is limited evidence on which to provide dietary guidelines to improve survival outcomes. Higher intake of red and processed meat and lower intake of fibre are associated with greater risk of developing CRC, but there is limited evidence regarding associations with survival after CRC diagnosis. Among 3789 CRC cases in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, pre-diagnostic consumption of red meat, processed meat, poultry and dietary fibre was examined in relation to CRC-specific mortality (n 1008) and all-cause mortality (n 1262) using multivariable Cox regression models, adjusted for CRC risk factors. Pre-diagnostic red meat, processed meat or fibre intakes (defined as quartiles and continuous grams per day) were not associated with CRC-specific or all-cause mortality among CRC survivors; however, a marginal trend across quartiles of processed meat in relation to CRC mortality was detected (P 0·053). Pre-diagnostic poultry intake was inversely associated with all-cause mortality among women (hazard ratio (HR)/20 g/d 0·92; 95 % CI 0·84, 1·00), but not among men (HR 1·00; 95 % CI 0·91, 1·09) (Pfor heterogeneity=0·10). Pre-diagnostic intake of red meat or fibre is not associated with CRC survival in the EPIC cohort. There is suggestive evidence of an association between poultry intake and all-cause mortality among female CRC survivors and between processed meat intake and CRC-specific mortality; however, further research using post-diagnostic dietary data is required to confirm this relationship.

Dietary habits play an important role in healthy ageing. We have investigated the role of dietary patterns on overall mortality in a large series of Italian elderly, recruited in five EPIC cohorts in Northern (Varese and Turin), Central (Florence) and Southern Italy (Naples and Ragusa).A total of 5611 subjects (72·6 % women) aged 60 years or older, enrolled in 1993–1998, were prospectively followed (median 6·2 years), with 152 deaths (98 women). Four major dietary patterns were identified by using an exploratory factor analysis based on dietary information collected at enrolment. The associations between these dietary patterns and overall mortality were evaluated by Cox models adjusted for potential confounders. The ‘Olive Oil & Salad’ pattern, characterised by a high consumption of olive oil, raw vegetables, soups and poultry, emerged as being inversely associated with overall mortality in both crude and adjusted models. After adjustment for gender, age and caloric intake, overall mortality was reduced by approximately 50 % in the highest quartile and a significant trend emerged (P = 0·008). This association persisted after adjusting for several additional confounders (hazard ratio (HR) 0·50; 95 % CI 0·29, 0·86; P for trend =  0·02). An association of the ‘Pasta & Meat’ pattern (characterised by pasta, tomato sauce, red meat, processed meat, added animal fat, white bread and wine) with increased overall mortality was also suggested, but only for the highest quartile in a multivariate model. Dietary recommendations aimed at the Italian elderly population should support a dietary pattern characterised by a high consumption of olive oil, raw vegetables and poultry.

Overall dietary patterns have been associated with health and longevity. We used principal component (PC) and cluster analyses to identify the prevailing dietary patterns of 99 744 participants, aged 60 years or older, living in nine European countries and participating in the European Prospective Investigation into Cancer and Nutrition (EPIC-Elderly cohort) and to examine their socio-demographic and lifestyle correlates. Two PC were identified: PC1 reflects a ‘vegetable-based’ diet with an emphasis on foods of plant origin, rice, pasta and other grain rather than on margarine, potatoes and non-alcoholic beverages. PC2 indicates a ‘sweet- and fat-dominated’ diet with a preference for sweets, added fat and dairy products but not meat, alcohol, bread and eggs. PC1 was associated with a younger age, a higher level of education, physical activity, a higher BMI, a lower waist:hip ratio and never and past smoking. PC2 was associated with older age, less education, never having smoked, a lower BMI and waist:hip ratio and lower levels of physical activity. Elderly individuals in southern Europe scored positively on PC1 and about zero on PC2, whereas the elderly in northern Europe scored negatively on PC1 and variably on PC2. The results of cluster analysis were compatible with the indicated dietary patterns. ‘Vegetable-based’ and a ‘sweet- and fat-dominated’ diets are prevalent among the elderly across Europe, and there is a north–south gradient regarding their dietary choices. Our study contributes to the identification of groups of elderly who are likely to have different prospects for long-term disease occurrence and survival.

Dietary patterns are comprehensive variables of dietary intake appropriate to model the complex exposure in nutritional research. The objectives of this study were to identify dietary patterns by applying two statistical methods, principal component analysis (PCA) and reduced rank regression (RRR), and to assess their ability to predict all-cause mortality. Motivated by previous studies we chose percentages of energy from different macronutrients as response variables in the RRR analysis. We used data from 9356 German elderly subject enrolled in the European Prospective Investigation into Cancer and Nutrition study. The first RRR pattern, subjects which explained 30·8 % of variation in energy sources and especially much variation in intake of saturated fat, monounsaturated fat and carbohydrates was a significant predictor of all-cause mortality. The pattern score had high positive loadings in all types of meat, butter, sauces and eggs, and was inversely associated with bread and fruits. After adjustment for other known risk factors, the relative risks from the lowest to highest quintiles of the first RRR pattern score were 1·0, 1·01, 0·96, 1·32, 1·61 (P for trend: 0·0004). In contrast, the first two PCA patterns explaining 19·7 % of food intake variation but only 7·0 % of variation in energy sources were not related to mortality. These results suggest that variation in macronutrients is meaningful for mortality and that the RRR method is more appropriate than the classic PCA method to identify dietary patterns relevant to mortality.