Epidemiological studies have suggested that flavonoids exhibit preventive effects on degenerative diseases. However, lack of sufficient data on flavonoid intake has limited evaluating the proposed effects in populations. Therefore, we aimed to estimate the total and individual flavonoid intakes among Korean adults and determine the major dietary sources of these flavonoids. We constructed a flavonoid database of common Korean foods, based on the food list reported in the 24-h recall of the Korea National Health and Nutrition Examination Survey (KNHANES) 2007–2012, using data from the Korea Functional Food Composition Table, US Department of Agriculture flavonoid database, Phenol-Explorer database and other analytical studies. This database, which covers 49 % of food items and 76 % of food intake, was linked with the 24-h recall data of 33 581 subjects aged ≥19 years in the KNHANES 2007–2012. The mean daily intake of total flavonoids in Korean adults was 318·0 mg/d, from proanthocyanidins (22·3 %), flavonols (20·3 %), isoflavones (18·1 %), flavan-3-ols (16·2 %), anthocyanidins (11·6 %), flavanones (11·3 %) and flavones (0·3 %). The major contributing food groups to the flavonoid intake were fruits (54·4 %), vegetables (20·5 %), legumes and legume products (16·2 %) and beverages and alcohols (3·1 %), and the major contributing food items were apples (21·9 %), mandarins (12·5 %), tofu (11·5 %), onions (9·6 %) and grapes (9·0 %). In the regression analysis, the consumption of legumes and legume products, vegetables and fruits predicted total flavonoid intake the most. The findings of this study could facilitate further investigation on the health benefits of flavonoids and provide the basic information for establishing recommended flavonoid intakes for Koreans.

The aim of this study was to explore the effect of single-dose albendazole and vitamin A intervention on the anaemic status and Fe metabolism of pre-school children. This study was a randomised, placebo-controlled and double-blinded intervention trial. All eligible anaemic pre-school children were randomly divided into three groups: group 1 received no intervention, which served as the control group, group 2 received 400 mg single-dose albendazole administration and group 3 received a 60000 μg vitamin A capsule combined with 400 mg single-dose albendazole at the beginning of the study. The follow-up period was for 6 months. Anthropometry and biochemical index about Fe metabolism were measured before and after intervention. A total of 209 pre-school anaemic children were randomly divided into three intervention groups (sixty-four, sixty-two and sixty for groups 1, 2 and 3, respectively). The mean age of the children in the study was 4·4 (sd 0·7) years and 50·5 % of the children were female (94/186). After a follow-up period of 6 months, the levels of serum retinol, ferritin, transferrin receptor-ferritin index and body total Fe content of children in group 3 were significantly higher compared with children in groups 1 and 2 (P<0·05). Moreover, the proportion of vitamin A deficiency, marginal vitamin A deficiency and Fe deficiency among children in group 3 were markedly lower compared with children in groups 1 and 2 (P<0·05). Albendazole plus vitamin A administration showed more efficacy on the improvement of serum retinol and Fe metabolic status.

Demand for organic milk is partially driven by consumer perceptions that it is more nutritious. However, there is still considerable uncertainty over whether the use of organic production standards affects milk quality. Here we report results of meta-analyses based on 170 published studies comparing the nutrient content of organic and conventional bovine milk. There were no significant differences in total SFA and MUFA concentrations between organic and conventional milk. However, concentrations of total PUFA and n-3 PUFA were significantly higher in organic milk, by an estimated 7 (95 % CI −1, 15) % and 56 (95 % CI 38, 74) %, respectively. Concentrations of α-linolenic acid (ALA), very long-chain n-3 fatty acids (EPA+DPA+DHA) and conjugated linoleic acid were also significantly higher in organic milk, by an 69 (95 % CI 53, 84) %, 57 (95 % CI 27, 87) % and 41 (95 % CI 14, 68) %, respectively. As there were no significant differences in total n-6 PUFA and linoleic acid (LA) concentrations, the n-6:n-3 and LA:ALA ratios were lower in organic milk, by an estimated 71 (95 % CI −122, −20) % and 93 (95 % CI −116, −70) %. It is concluded that organic bovine milk has a more desirable fatty acid composition than conventional milk. Meta-analyses also showed that organic milk has significantly higher α-tocopherol and Fe, but lower I and Se concentrations. Redundancy analysis of data from a large cross-European milk quality survey indicates that the higher grazing/conserved forage intakes in organic systems were the main reason for milk composition differences.

Imbalances in dietary fat intakes are linked to several chronic diseases. This study describes dietary intakes and food sources of fat and fatty acids in 1051 Irish adults (aged 18–90 years), using data from the 2011 national food consumption survey, the National Adult Nutrition Survey. It also compares current intakes for 18–64-year-olds with those reported in the last such survey in 2001, the North/South Ireland Food Consumption Survey. Dietary fat intakes were estimated using data from 4-d semi-weighed (2011) and 7-d estimated (2001) food diaries. In 2011, intakes for 18–64-year-olds were as follows: total fat, 34·1 (sd 6·1) % total energy (%TE); SFA, 13·3 (sd 3·3) %TE; MUFA, 12·5 (sd 2·6) %TE; PUFA, 6·1 (sd 2·2) %TE; and trans-fat, 0·511 (sd 0·282) %TE. Apart from MUFA, intakes decreased (P<0·001) compared with 2001. There was no statistically significant difference in intakes of EPA and DHA by 18–64-year-olds in 2011 (269·0 (sd 515·0) mg/d) and 2001 (279·1 (sd 497·5) mg/d). In 2011, adults aged >65 years had the highest intakes of SFA; however, intakes were typically higher than UK-recommended values for all groups. In contrast, intakes of long-chain n-3 fatty acids were lowest in younger age groups. Intakes of trans-fat were well within UK-recommended levels. Although there have been some improvements in the profile of intakes since 2001, imbalances persist in the quantity and quality of dietary fat consumed by Irish adults, most notably for total and SFA and for younger age groups for long-chain n-3 fatty acids.

Brown seaweeds such as Ascophyllum nodosum are a rich source of phlorotannins (oligomers and polymers of phloroglucinol units), a class of polyphenols that are unique to Phaeophyceae. At present, there is no information on the bioavailability of seaweed polyphenols and limited evidence on their bioactivity in vivo. Consequently, we investigated the gastrointestinal modifications in vitro of seaweed phlorotannins from A. nodosum and their bioavailability and effect on inflammatory markers in healthy participants. In vitro, some phlorotannin oligomers were identified after digestion and colonic fermentation. In addition, seven metabolites corresponding to in vitro-absorbed metabolites were identified. Urine and plasma samples contained a variety of metabolites attributed to both unconjugated and conjugated metabolites (glucuronides and/or sulphates). In both urine and plasma, the majority of the metabolites were found in samples collected at late time points (6–24 h), suggesting colonic metabolism of high-molecular-weight phlorotannins, with three phlorotannin oligomers (hydroxytrifuhalol A, 7-hydroxyeckol, C-O-C dimer of phloroglucinol) identified in urine samples. A significant increase of the cytokine IL-8 was also observed. Our study shows for the first time that seaweed phlorotannins are metabolised and absorbed, predominantly in the large intestine, and there is a large inter-individual variation in their metabolic profile. Three phlorotannin oligomers present in the capsule are excreted in urine. Our study is the first investigation of the metabolism and bioavailability of seaweed phlorotannins and the role of colonic biotransformation. In addition, IL-8 is a possible target for phlorotannin bioactivity.

Recent literature suggests that Ca supplements have adverse effects on cardiovascular health. The effects of a Ca-rich supplement administered alone or in combination with short-chain fructo-oligosaccharides (scFOS) on serum lipids in postmenopausal women were examined using secondary data from a 24-month double-blind randomised controlled study. A total of 300 postmenopausal women were randomly assigned to daily supplements of 800 mg of Ca (2·4 g Aquamin) (Ca), 800 mg of Ca with 3 g of scFOS (CaFOS) or control (maltodextrin) (MD). A full lipid profile, body composition, blood pressure and a range of cytokines were measured at baseline and after 24 months. Intention-to-treat ANCOVA assessed treatment effects between the groups. A significant time-by-treatment effect was observed for LDL and total cholesterol for the Ca and CaFOS groups, with both groups having lower LDL and total cholesterol concentrations compared with MD after 24 months. The control group had mean (5·2 mmol/l) total cholesterol concentrations above the normal range (≤5 mmol/l) at 24 months, whereas values remained within the normal range in the treatment groups. There was no significant treatment effect on HDL-cholesterol, TAG, body composition, blood pressure or cytokine concentrations at 24 months, with the exception of IL-4, where there was a significant increase in the CaFOS group compared with the placebo. This study demonstrates a lipid-lowering effect of both the Ca-rich supplement alone and the supplement with scFOS. At the 4-year follow-up, there was no significant difference between the groups for reported diagnosed cardiovascular conditions.

We investigated whether dietary and urinary Na is associated with adiposity in Korean children and adolescents (10–18 years), a population with a high salt intake. Study subjects were Korean children and adolescents who participated in the cross-sectional nationally representative Korea National Health and Nutrition Examination Survey (2010–2011). This study used measures of dietary (24-h dietary recall) and urinary Na (Na:creatinine ratio) and three methods to determine obesity (BMI, waist circumference (WC) and total body per cent fat (TBPF)). Higher Na intake was significantly associated with obesity, adjusting for the covariates. Subjects in the highest tertile of urinary Na excretion had a significantly higher OR for higher adiposity compared with those in the lowest tertile (multivariate-adjusted OR 3·13 (95 % CI 1·81, 5·50) for BMI, 2·15 (95 % CI 1·27, 3·66) for WC and 1·92 (95 % CI 1·29, 2·86) for TBPF, respectively). Na intake estimated by the 24-h recall method also showed significant association with adiposity (multivariate-adjusted OR 2·79 (95 % CI 1·66, 4·68) for BMI and 2·14 (95 % CI 1·25, 3·67) for WC, respectively). The significant associations between Na and adiposity remained significant after additionally adjusting for sugar-sweetened beverage (SSB) consumption. Our results revealed a significant positive association between urinary and dietary Na and adiposity in Korean children and adolescents, independent of SSB consumption.

Severe mental illness is characterised by a 20-year mortality gap due to cardiometabolic disease. Poor diet in those with severe mental illness is an important and modifiable risk factor. The present study aimed to (i) examine baseline nutritional intake in youth with first-episode psychosis (FEP), (ii) evaluate the feasibility and acceptability of nutritional intervention early in FEP and (iii) to evaluate the effectiveness of early dietary intervention on key nutritional end points. Participants were recruited over a 12-month period from a community-based programme specifically targeting young people aged 15–25 years with newly diagnosed FEP. Individual dietetic consultations and practical group sessions were offered as part of a broader lifestyle programme. Dietary assessments were conducted before and at the end of the 12-week intervention. Participants exceeded recommended energy and Na intakes at baseline. Retention within the nutrition intervention was 67 %, consistent with other interventions offered to FEP clients. There was a 47 % reduction in discretionary food intake (−94 g/d, P<0·001) and reductions in daily energy (−24 %, P<0·001) and Na (−26 %, P<0·001) intakes. Diet quality significantly improved, and the mean change was 3·6 (95 % CI 0·2, 6·9, P<0·05), although this finding was not significant after Bonferroni’s correction. Increased vegetable intake was the main factor contributing to improved diet quality. Nutrition intervention delivered shortly after initiation of antipsychotic medication is feasible, acceptable and effective in youth with FEP. Strategies to prevent weight gain and metabolic decline will contribute to prevent premature cardiometabolic disease in this vulnerable population.

Obesity in young adults is an increasing health problem in Australia and many other countries. Evidence-based information is needed to guide interventions that reduce the obesity-promoting elements in tertiary-education environments. In a food environmental audit survey, 252 outlets were audited across seven institutions: three universities and four technical and further education institutions campuses. A scoring instrument called the food environment-quality index was developed and used to assess all food outlets on these campuses. Information was collated on the availability, accessibility and promotion of foods and beverages and a composite score (maximum score=148; higher score indicates healthier outlets) was calculated. Each outlet and the overall campus were ranked into tertiles based on their ‘healthiness’. Differences in median scores for each outcome measure were compared between institutions and outlet types using one-way ANOVA with post hoc Scheffe’s testing, χ2 tests, Kruskal–Wallis H test and the Mann–Whitney U test. Binomial logistic regressions were used to compare the proportion of healthy v. unhealthy food categories across different types of outlets. Overall, the most frequently available items were sugar-sweetened beverages (20 % of all food/drink items) followed by chocolates (12 %), high-energy (>600 kJ/serve) foods (10 %), chips (10 %) and confectionery (10 %). Healthy food and beverages were observed to be less available, accessible and promoted than unhealthy options. The median score across all outlets was 72 (interquartile range=7). Tertiary-education food environments are dominated by high-energy, nutrient-poor foods and beverages. Interventions to decrease availability, accessibility and promotion of unhealthy foods are needed.

Mechanistic hypotheses suggest that vitamin D and the closely related parathyroid hormone (PTH) may be involved in prostate carcinogenesis. However, epidemiological evidence is lacking for PTH and inconsistent for vitamin D. Our objectives were to prospectively investigate the association between vitamin D status, vitamin D-related gene polymorphisms, PTH and prostate cancer risk. A total of 129 cases diagnosed within the Supplémentation en Vitamines et Minéraux Antioxydants cohort were included in a nested case–control study and matched to 167 controls (13 years of follow-up). 25-Hydroxyvitamin D (25(OH)D) and PTH concentrations were assessed from baseline plasma samples. Conditional logistic regression models were computed. Higher 25(OH)D concentration was associated with decreased risk of prostate cancer (ORQ4 v. Q1 0·30; 95 % CI 0·12, 0·77; Ptrend=0·007). PTH concentration was not associated with prostate cancer risk (Ptrend=0·4) neither did the studied vitamin D-related gene polymorphisms. In this prospective study, prostate cancer risk was inversely associated with 25(OH)D concentration but not with PTH concentration. These results bring a new contribution to the understanding of the relationship between vitamin D and prostate cancer, which deserves further investigation.

The present study explored the relationships between fried food consumption and metabolic risk factors and hypertension in Korean adults. The study was based on the fifth Korean National Health and Nutrition Examination Survey between 2010 and 2011. A total of 9221 Korean adults aged ≥19 years were studied. Fried food consumption was assessed using a validated FFQ. Metabolic risk factors such as waist circumference, fasting plasma glucose (FPG), TAG, HDL-cholesterol and systolic and diastolic blood pressure (SBP and DBP) were measured. Hypertension was defined as SBP≥140 mmHg, DBP≥90 mmHg or current use of antihypertensive medication. Adjusted OR for elevated blood pressure significantly increased in men (OR 1·62; 95 % CI 1·11, 2·37; Ptrend=0·0447) and women (OR 2·20; 95 % CI 1·21, 4·00; Ptrend=0·0403) with a greater than twice a week consumption of fried food compared with those who rarely consumed fried food. However, fried food consumption was not associated with other metabolic risk factors (abdominal obesity, high FPG, hypertriacylglycerolaemia, low HDL-cholesterol and the metabolic syndrome). The adjusted OR for hypertension increased by 2·4-fold in women (OR 2·37; 95 % CI 1·19, 4·72; Ptrend=0·0272) with a greater than twice a week fried food consumption compared with those who rarely consumed it. No significant association was found between fried food consumption and hypertension in men. This study suggests that frequent fried food consumption is associated with hypertension in Korean women. Further studies are needed to investigate the effect of different types of fried foods on hypertension.

Food and nutrient intake data are scarce in very old adults (85 years and older) – one of the fastest growing age segments of Western societies, including the UK. Our primary objective was to assess energy and macronutrient intakes and respective food sources in 793 85-year-olds (302 men and 491 women) living in North-East England and participating in the Newcastle 85+ cohort Study. Dietary information was collected using a repeated multiple-pass recall (2×24 h recalls). Energy, macronutrient and NSP intakes were estimated, and the contribution (%) of food groups to nutrient intake was calculated. The median energy intake was 6·65 (interquartile ranges (IQR) 5·49–8·16) MJ/d – 46·8 % was from carbohydrates, 36·8 % from fats and 15·7 % from proteins. NSP intake was 10·2 g/d (IQR 7·3–13·7). NSP intake was higher in non-institutionalised, more educated, from higher social class and more physically active 85-year-olds. Cereals and cereal products were the top contributors to intakes of energy and most macronutrients (carbohydrates, non-milk extrinsic sugars, NSP and fat), followed by meat and meat products. The median intakes of energy and NSP were much lower than the estimated average requirement for energy (9·6 MJ/d for men and 7·7 MJ/d for women) and the dietary reference value (DRV) for NSP (≥18 g/d). The median SFA intake was higher than the DRV (≤11 % of dietary energy). This study highlights the paucity of data on dietary intake and the uncertainties about DRV for this age group.

Irregularity in eating patterns could be a potential cardiometabolic risk factor. We aimed to study the associations of irregular intake of energy at meals in relation to cardiometabolic risk factors 10 and 17 years later. Variability of energy intake data – derived from 5-d estimated diet diaries of cohort members of the National Survey for Health and Development collected at ages 36 (n 1416), 43 (n 1505) and 53 years (n 1381) – was used as a measure for irregularity. Associations between meal irregularity scores with cardiometabolic risk factors measured 10 and 17 years later were investigated using linear mixed models and logistic regression models. The results showed that irregularity scores changed significantly over the years (P<0·05). At age 36 years, subjects with a more irregular intake of energy at lunch (OR 1·42; 95 % CI 1·05, 1·91) and between meals (OR 1·35; 95 % CI 1·01, 1·82) had an increased risk for the metabolic syndrome 17 years later; at lunch was also associated with an increased waist circumference (OR 1·58; 95 % 1·27, 1·96) and TAG levels (OR 1·33; 95 % CI 1·02, 1·72). At age 43 years, subjects with a more irregular intake at breakfast had an increased risk of the metabolic syndrome 10 years later (OR 1·53; 95 % CI 1·15, 2·04), as well as an increased BMI (OR 1·66; 95 % CI 1·31, 2·10), waist circumference (OR 1·53; 95 % CI 1·23, 1·90) and diastolic blood pressure (OR 1·42; 95 % CI 1·13, 1·78). In conclusion, subjects with a more irregular intake of energy, mostly at breakfast and lunch, appeared to have an increased cardiometabolic risk 10 and 17 years later.

Dietary patterns derived by statistical procedures is a way to identify overall dietary habits in specific populations. The aim of this study was to identify and characterise dietary patterns in Swedish adults using data from the national dietary survey Riksmaten adults 2010–11 (952 women, 788 men). Principal component analyses were used and two patterns were identified in both sexes: a healthy pattern loading positively on vegetables, fruits, fish and seafood, and vegetable oils, and negatively on refined bread and fast food, and a Swedish traditional pattern loading positively on potatoes, meat and processed meat, full-fat milk products, sweet bakery products, sweet condiments and margarine. In addition, a light-meal pattern was identified in women with positive loadings on fibre-rich bread, cheese, rice, pasta and food grain dishes, substitute products for meat and dairy products, candies and tea. The healthy pattern was positively correlated to dietary fibre (r 0·51–0·58) and n-3 (r 0·25–0·31) (all P<0·0001), and had a higher nutrient density of folate, vitamin D and Se. The Swedish traditional and the light-meal pattern were positively correlated to added sugar (r 0·20–0·25) and the Swedish traditional also to SFA (r 0·13–0·21) (all P<0·0001); both patterns were in general negatively correlated to micronutrients. Dietary pattern scores were associated with, for example, age, physical activity, education and income. In conclusion, we identified three major dietary patterns among Swedish adults. The patterns can be further used for examining the association between whole diet and health outcomes.

FFQ comprising food items, intake frequency categories and portion sizes have been used in large-scale observational studies to assess long-term dietary exposure. Although gender is an important influence on food choice and portion size, gender differences are not often analysed during FFQ development. This study investigated whether gender differences were considered sufficiently when developing FFQ, which affects the results of validation studies. A PubMed search using combinations of ‘FFQ’, ‘Food Frequency Questionnaire’, ‘Validation’ and ‘Validity’ identified 246 validation studies available in English, published between January 1983 and May 2014, which included healthy male and female adults. The development process of the 196 FFQ used in the 246 validation studies was examined. Of these, twenty-one FFQ (10·7 %) considered gender during item selection or portion size determination, and were therefore classified as gender specific (GS), but 175 (89·3 %) did not consider gender, and were classified as ‘not gender specific (NGS)’. When the ratios between intake levels obtained using the FFQ and a reference method for energy and seven nutrients were compared between the GS group and the NGS group, more significant differences were observed in women than in men (four v. one nutrient). Intake of three nutrients was significantly underestimated in both sexes in the GS group. In the NGS group, nutrient intakes were significantly overestimated more often in women than in men (four v. one). These results indicate that not considering gender in FFQ development causes greater inaccuracy in dietary intake assessment in women than in men. Results of nutritional epidemiological studies should be re-evaluated for their validity, especially if the studies used NGS-FFQ.

A comprehensive estimation of polyphenol intake is needed to gain a better understanding of the association between polyphenol-rich food intake and the potential effects of this intake on chronic diseases. The aim of this study was to estimate the intake of polyphenols and the major dietary contributors in the population of Sao Paulo. Data were obtained from the Health Survey-São Paulo (ISA-Capital 2008) and were reported for 1103 adults and elderly adults. Food intake was estimated by one 24-h dietary recall (24HR). Polyphenol intake was calculated by matching food consumption data from the 24HR with the polyphenol content in foods listed in the Phenol-Explorer database. The mean total intake of polyphenols was 377·5 (se 15·3) mg/d. The main polyphenol classes were phenolic acids (284·8 (se 15·9) mg/d) and flavonoids (54·6 (se 3·5) mg/d). Intakes were higher in the elderly adults than in other adults (P<0·001) and higher in individuals with lower educational level (P=0·01) and current smokers (P=0·02). The main dietary contributors for total polyphenols were coffee (70·5 %), citrus fruits (4·6 %) and tropical fruits (3·4 %). Coffee was the major source of polyphenols, providing 266·2 (se 16·5) mg/d, and contributed 92·3 % of the phenolic acids and 93·1 % of the alkylmethoxyphenols. These findings will be useful for assessing the potential role on health of polyphenols and specific polyphenol-rich foods, such as coffee, and enable a comparison with people from other countries.

Despite the potential of declared serving size to encourage appropriate portion size consumption, most countries including Australia have not developed clear reference guidelines for serving size. The present study evaluated variability in manufacturer-declared serving size of discretionary food and beverage products in Australia, and how declared serving size compared with the 2013 Australian Dietary Guideline (ADG) standard serve (600 kJ). Serving sizes were obtained from the Nutrition Information Panel for 4466 packaged, discretionary products in 2013 at four large supermarkets in Sydney, Australia, and categorised into fifteen categories in line with the 2013 ADG. For unique products that were sold in multiple package sizes, the percentage difference between the minimum and the maximum serving size across different package sizes was calculated. A high variation in serving size was found within the majority of food and beverage categories – for example, among 347 non-alcoholic beverages (e.g. soft drinks), the median for serving size was 250 (interquartile range (IQR) 250, 355) ml (range 100–750 ml). Declared serving size for unique products that are available in multiple package sizes also showed high variation, particularly for chocolate-based confectionery, with median percentage difference between minimum and maximum serving size of 183 (IQR 150) %. Categories with a high proportion of products that exceeded the 600 kJ ADG standard serve included cakes and muffins, pastries and desserts (≥74 % for each). High variability in declared serving size may confound interpretation and understanding of consumers interested in standardising and controlling their portion selection. Future research is needed to assess if and how standardising declared serving size might affect consumer behaviour.

We evaluated the effect of maternal vitamin E supplementation on the α-tocopherol concentrations of colostrum, transitional milk and mature milk of women who had given birth prematurely. This longitudinal randomised-controlled trial divided eighty-nine women into two groups: a control group and a supplemented group. Blood and breast milk were collected from all the participants after delivery. Next, each woman in the supplemented group received 400 IU of RRR-α-tocopheryl acetate. Further breast milk samples were collected 24 h after the first collection, as well as 7 and 30 d after delivery. α-Tocopherol concentrations were determined by HPLC. The baseline α-tocopherol concentrations in the maternal serum of the two groups were similar: 1159·8 (sd 292·4) μg/dl (27·0 (SD 6·8) μmol/l) for the control group and 1128·3 (sd 407·2) μg/dl (26·2 (SD 9·5) μmol/l) for the supplemented group. None of the women was vitamin E deficient. Breast milk α-tocopherol concentrations increased by 60 % 24 h after supplementation in the intervention group and did not increase at all in the control group. α-Tocopherol concentration of the transitional milk in the supplemented group was 35 % higher compared with the control group. α-Tocopherol concentrations of the mature milk in both groups were similar. Maternal supplementation with 400 IU of RRR-α-tocopherol increased the vitamin E concentrations of the colostrum and transitional milk, but not of the mature milk. This study presents relevant information for the design of strategies to prevent and combat vitamin E deficiency in the risk group of preterm infants.

There is increasing evidence for the role of nutrition in the prevention of depression. This study aims to describe changes in diet quality over 12 years among participants in the Australian Longitudinal Study on Women’s Health in relation to changes in depressive symptoms. Women born between 1946 and 1951 were followed-up for 12 years (2001–2013). Dietary intake was assessed using the Dietary Questionnaire for Epidemiological Studies (version 2) in 2001, 2007 and every 2–3 years after that until 2013. Diet quality was summarised using the Australian Recommended Food Score (ARFS). Depressive symptoms were measured using the ten-item Centre for Epidemiologic Depression Scale at every 2–3-year intervals during 2001–2013. Linear mixed models were used to examine trends in diet quality and its sub-components. The same model including time-varying covariates was used to examine associations between diet quality and depressive symptoms adjusting for confounders. Sensitivity analyses were carried out using the Mediterranean dietary pattern (MDP) index to assess diet quality. Minimal changes in overall diet quality and its sub-components over 12 years were observed. There was a significant association between baseline diet quality and depression (β=−0·24, P=0·001), but this was lost when time-varying covariates were added (β=−0·04, P=0·10). Sensitivity analyses showed similar performance for both ARFS and MDP in predicting depressive symptoms. In conclusion, initial associations seen when using baseline measures of diet quality and depressive symptoms disappear when using methods that handle time-varying covariates, suggesting that previous studies indicating a relationship between diet and depression may have been affected by residual confounding.

We evaluated the relationship between urine concentrations of phyto-oestrogens (isoflavones and lignans) and risk of incident type 2 diabetes in middle-aged and elderly Chinese residing in Singapore. Urine metabolites of isoflavones and lignans were assayed by HPLC among 564 diabetes cases and 564 matched controls in a case–control study nested within the Singapore Chinese Health Study cohort. Participants were free of diagnosed diabetes, CVD and cancer at morning urine collections during 1999–2004. Cases were participants who reported to have physician-diagnosed diabetes at follow-up visits during 2006–2010, whereas controls were randomly selected among those who remained free of diabetes and were matched to the index cases by age, sex, dialect group and date of urine collection. Conditional logistic regression models were used to calculate OR and 95 % CI with adjustment for potential confounders. The mean age of the participants at the time of urine collection was 59·8 years, and the average interval between urine collection and diabetes diagnosis was 4·0 years. The multivariate-adjusted OR for diabetes were 1·00 (reference), 0·76 (95 % CI 0·52, 1·11), 0·78 (95 % CI 0·53, 1·14) and 0·79 (95 % CI 0·54, 1·15) across quartiles of urine isoflavones (Pfor trend=0·54), and were 1·00 (reference), 0·87 (95 % CI 0·60, 1·27), 1·10 (95 % CI 0·77, 1·56) and 0·93 (95 % CI 0·63, 1·37) for lignans (Pfor trend=0·93). The results were similar in men and women, as well as for individual metabolites of isoflavones (genistein, daidzein, glycitin and equol) or lignans (enterodiol and enterolactone). The present study did not find a significant association between urine phyto-oestrogen metabolites and risk of type 2 diabetes in Chinese adults.