Recent studies have focused on habitual intake without addressing meal-specific intakes. We aimed to identify meal-specific dietary patterns and their contribution to habitual dietary patterns. This cross-sectional study was conducted on 838 adults, both sexes who attended the health centres in Tehran. Dietary data were recorded by three 24-h dietary recalls (24hDR). Dietary patterns were identified by using principal component analysis on meal-specific and overall food intakes. Intraclass correlation (ICC) was used as a measurement of consistency across meals and days. Correlation analysis and linear regression (partial R 2) were used for meals contribution. Four habitual dietary patterns were derived from average dietary intake of 3-d 24hDR labelled as ‘Western’, ‘Healthy’, ‘Traditional’ and ‘Legume and broth’. Also, we identified two major dietary patterns on each meal level (factor 1 and 2 for breakfast, lunch, afternoon snack and dinner). The highest contribution of energy intake was observed in lunch (25·7 %), followed by dinner (20·81 %). Consistency of food groups was the highest across days (ICC tea = 0·58) and breakfasts (ICC tea = 0·60). Dinner had a strong correlation coefficient with the ‘Western’ habitual dietary pattern then followed by lunch. Similarly, dinner and lunch contributed the most (r and partial R 2) to the ‘Western’ habitual dietary pattern. Our results suggest that habitual dietary patterns to several extents are formed at meal levels, and dinner has a greater contribution to the habitual dietary patterns in Iranian people. This may help planning for local dietary guidelines according to the time of eating to promote public health.

Evidence suggests that processed red meat consumption is a risk factor for CVD and type 2 diabetes (T2D). This analysis investigates the association between dietary patterns, their processed red meat contributions, and association with blood biomarkers of CVD and T2D, in 786 Irish adults (18–90 years) using cross-sectional data from a 2011 national food consumption survey. All meat-containing foods consumed were assigned to four food groups (n 502) on the basis of whether they contained red or white meat and whether they were processed or unprocessed. The remaining foods (n 2050) were assigned to twenty-nine food groups. Two-step and k-means cluster analyses were applied to derive dietary patterns. Nutrient intakes, plasma fatty acids and biomarkers of CVD and T2D were assessed. A total of four dietary patterns were derived. In comparison with the pattern with lower contributions from processed red meat, the dietary pattern with greater processed red meat intakes presented a poorer Alternate Healthy Eating Index (21·2 (sd 7·7)), a greater proportion of smokers (29 %) and lower plasma EPA (1·34 (sd 0·72) %) and DHA (2·21 (sd 0·84) %) levels (P<0·001). There were no differences in classical biomarkers of CVD and T2D, including serum cholesterol and insulin, across dietary patterns. This suggests that the consideration of processed red meat consumption as a risk factor for CVD and T2D may need to be re-assessed.

Diet is related to many chronic disease conditions such as the metabolic syndrome (MetS). We set out to compare behaviour-related with disease-related patterns and their association with the MetS in a German cross-sectional study. A total of 905 participants of a Northern German cohort (aged 25–82 years) completed a FFQ, underwent anthropometric assessments and provided a blood sample. Dietary patterns were derived by principal component analysis (PCA) and reduced-rank regression (RRR) from forty-two food groups. Components of the MetS were used as response variables for the RRR analysis. Simplified patterns comprising ten food groups were generated. Logistic regression analysis was performed to evaluate the likelihood of having the MetS across the quartiles of simplified pattern scores. We identified two similar dietary patterns derived by PCA and RRR characterised by high intakes of potatoes, various vegetables, red and processed meat, fats, sauce and bouillon. Comparing simplified patterns, an increased RRR pattern score was associated with a higher OR (2·18, 95 % CI 1·25, 3·81) of having the MetS than an increased PCA pattern score (OR 1·92, 95 % CI 1·21, 3·03). Comparing concordant food groups by both dietary pattern methods, a diet high in legumes, beef, processed meat and bouillon was also positively associated with the prevalence of the MetS after adjustment for potential confounders (OR 1·71, 95 % CI 1·04, 2·79). We identified a behaviour-related pattern that was positively associated with the MetS. The application of both dietary pattern methods may be advantageous to obtain information for designing and realising dietary guidelines. Prospective studies are needed to confirm the results.

This paper aims to describe different approaches for studying the overall diet with advantages and limitations. Studies of the overall diet have emerged because the relationship between dietary intake and health is very complex with all kinds of interactions. These cannot be captured well by studying single dietary components. Three main approaches to study the overall diet can be distinguished. The first method is researcher-defined scores or indices of diet quality. These are usually based on guidelines for a healthy diet or on diets known to be healthy. The second approach, using principal component or cluster analysis, is driven by the underlying dietary data. In principal component analysis, scales are derived based on the underlying relationships between food groups, whereas in cluster analysis, subgroups of the population are created with people that cluster together based on their dietary intake. A third approach includes methods that are driven by a combination of biological pathways and the underlying dietary data. Reduced rank regression defines linear combinations of food intakes that maximally explain nutrient intakes or intermediate markers of disease. Decision tree analysis identifies subgroups of a population whose members share dietary characteristics that influence (intermediate markers of) disease. It is concluded that all approaches have advantages and limitations and essentially answer different questions. The third approach is still more in an exploration phase, but seems to have great potential with complementary value. More insight into the utility of conducting studies on the overall diet can be gained if more attention is given to methodological issues.

Few studies have investigated the absorption of phylloquinone (vitamin K1). We recruited twelve healthy, non-obese adults. On each study day, fasted subjects took a capsule containing 20 μg of 13C-labelled phylloquinone with one of three meals, defined as convenience, cosmopolitan and animal-oriented, in a three-way crossover design. The meals were formulated from the characteristics of clusters identified in dietary pattern analysis of data from the National Diet and Nutrition Survey conducted in 2000–1. Plasma phylloquinone concentration and isotopic enrichment were measured over 8 h. Significantly more phylloquinone tracer was absorbed when consumed with the cosmopolitan and animal-oriented meals than with the convenience meal (P = 0·001 and 0·035, respectively). Estimates of the relative availability of phylloquinone from the meals were: convenience meal = 1·00; cosmopolitan meal = 0·31; animal-oriented meal = 0·23. Combining the tracer data with availability estimates for phylloquinone from the meals provides overall relative bioavailability values of convenience = 1·00, cosmopolitan = 0·46 and animal-oriented = 0·29. Stable isotopes provide a useful tool to investigate further the bioavailability of low doses of phylloquinone. Different meals can affect the absorption of free phylloquinone. The meal-based study design used in the present work provides an approach that reflects more closely the way foods are eaten in a free-living population.

The aim of the present study was to determine whether under-reporting rates vary between dietary pattern clusters. Subjects were sixty-five Brazilian women. During 3 weeks, anthropometric data were collected, total energy expenditure (TEE) was determined by the doubly labelled water method and diet was measured. Energy intake (EI) and the daily frequency of consumption per 1000 kJ of twenty-two food groups were obtained from a FFQ. These frequencies were entered into a cluster analysis procedure in order to obtain dietary patterns. Under-reporters were defined as those who did not lose more than 1 kg of body weight during the study and presented EI:TEE less than 0·82. Three dietary pattern clusters were identified and named according to their most recurrent food groups: sweet foods (SW), starchy foods (ST) and healthy (H). Subjects from the healthy cluster had the lowest mean EI:TEE (SW = 0·86, ST = 0·71 and H = 0·58; P = 0·003) and EI − TEE (SW = − 0·49 MJ, ST = − 3·20 MJ and H = − 5·08 MJ; P = 0·008). The proportion of under-reporters was 45·2 (95 % CI 35·5, 55·0) % in the SW cluster; 58·3 (95 % CI 48·6, 68·0) % in the ST cluster and 70·0 (95 % CI 61·0, 79) % in the H cluster (P = 0·34). Thus, in Brazilian women, under-reporting of EI is not uniformly distributed among dietary pattern clusters and tends to be more severe among subjects from the healthy cluster. This cluster is more consistent with both dietary guidelines and with what lay individuals usually consider ‘healthy eating’.