Fish-oil supplements are marketed as enhancing intelligence and cognitive performance. However, empirical data concerning the utility of these products in healthy term infants are mixed, particularly with respect to lasting effects into childhood. We evaluated whether fish-oil supplementation during infancy leads to better neurocognitive/behavioural development at 6 years. We conducted a double-blind randomised controlled trial of supplementation with n-3 long-chain PUFA in 420 healthy term infants. Infants received either fish oil (containing at least 250 mg DHA and at least 60 mg EPA) or placebo (olive oil) daily from birth to 6 months of age. Neurodevelopmental follow-up was conducted at a mean age of 6 years (sd 7 months), whereby 335 children were assessed for language, executive functioning, global intelligence quotient and behaviour. No significant differences were observed between the groups for the main neurocognitive outcomes. However in parent-report questionnaire, fish-oil supplementation was associated with negative externalising (P = 0·035, d = 0·24) and oppositional/defiant behaviour (P = 0·006, d = 0·31), particularly in boys (P = 0·01, d = 0·45; P = 0·004, d = 0·40). Our results provide evidence that fish-oil supplementation to predominantly breast-fed infants confers no significant cognitive or behavioural benefit to children at 6 years.

Diet has direct and indirect effects on health through inflammation and the gut microbiome. We investigated total dietary inflammatory potential via the literature-derived index (Dietary Inflammatory Index (DII®)) with gut microbiota diversity, composition and function. In cancer-free patient volunteers initially approached at colonoscopy and healthy volunteers recruited from the medical centre community, we assessed 16S ribosomal DNA in all subjects who provided dietary assessments and stool samples (n 101) and the gut metagenome in a subset of patients with residual fasting blood samples (n 34). Associations of energy-adjusted DII scores with microbial diversity and composition were examined using linear regression, permutational multivariate ANOVA and linear discriminant analysis. Spearman correlation was used to evaluate associations of species and pathways with DII and circulating inflammatory markers. Across DII levels, α- and β-diversity did not significantly differ; however, Ruminococcus torques, Eubacterium nodatum, Acidaminococcus intestini and Clostridium leptum were more abundant in the most pro-inflammatory diet group, while Akkermansia muciniphila was enriched in the most anti-inflammatory diet group. With adjustment for age and BMI, R. torques, E. nodatum and A. intestini remained significantly associated with a more pro-inflammatory diet. In the metagenomic and fasting blood subset, A. intestini was correlated with circulating plasminogen activator inhibitor-1, a pro-inflammatory marker (rho = 0·40), but no associations remained significant upon correction for multiple testing. An index reflecting overall inflammatory potential of the diet was associated with specific microbes, but not overall diversity of the gut microbiome in our study. Findings from this preliminary study warrant further research in larger samples and prospective cohorts.

Cranberries are high in polyphenols, and epidemiological studies have shown that a high-polyphenol diet may reduce risk factors for diabetes and CVD. The present study aimed to determine if short-term cranberry beverage consumption would improve insulin sensitivity and other cardiovascular risk factors. Thirty-five individuals with obesity and with elevated fasting glucose or impaired glucose tolerance participated in a randomised, double-blind, placebo-controlled, parallel-designed pilot trial. Participants consumed 450 ml of low-energy cranberry beverage or placebo daily for 8 weeks. Changes in insulin sensitivity and cardiovascular risk factors including vascular reactivity, blood pressure, RMR, glucose tolerance, lipid profiles and oxidative stress biomarkers were evaluated. Change in insulin sensitivity via hyperinsulinaemic–euglycaemic clamp was not different between the two groups. Levels of 8-isoprostane (biomarker of lipid peroxidation) decreased in the cranberry group but increased in the placebo group (–2·18 v. +20·81 pg/ml; P = 0·02). When stratified by baseline C-reactive protein (CRP) levels, participants with high CRP levels (>4 mg/l) benefited more from cranberry consumption. In this group, significant differences in the mean change from baseline between the cranberry (n 10) and the placebo groups (n 7) in levels of TAG (–13·75 v. +10·32 %; P = 0·04), nitrate (+3·26 v. −6·28 µmol/l; P = 0·02) and 8-isoprostane (+0·32 v. +30·8 pg/ml; P = 0·05) were observed. These findings indicate that 8 weeks of daily cranberry beverage consumption may not impact insulin sensitivity but may be helpful in lowering TAG and changing certain oxidative stress biomarkers in individuals with obesity and a proinflammatory state.

The aim of the present work was to compare the capacity to modulate the intestinal microbiota and the production of metabolites after 14 d administration of a commercial dietary supplement and a manufactured ice cream, both containing the same quantity of inulin and the same viable counts of Lactobacillus acidophilus LA-5 and Bifidobacterium animalis BB-12, using the Simulator of the Human Intestinal Microbial Ecosystem (SHIME®) model. Samples of the colonic contents were evaluated microbiologically by real-time quantitative PCR (qRT-PCR) and next-generation sequencing and chemically by the production of SCFA (acetate, propionate and butyrate) and ammonium ions ($\text{NH}_4^ + $). Statistical analyses were carried out for all the variables using the two-way ANOVA followed by the Tukey multiple comparisons test (P < 0·05) for metabolite production, qRT-PCR and the bioinformatics analysis for microbiota diversity. Dietary supplement and ice cream were able to deliver the probiotic L. acidophilus and B. animalis to the simulated colon and modulate the microbiota, increasing beneficial micro-organisms such as Bifidobacterium spp., Bacteroides spp. and Faecalibacterium spp. for dietary supplement administration, and Lactobacillus spp. for ice cream supplementation. However, the ice cream matrix was probably more favourable for the maintenance of the metabolic activity of the probiotics in the SHIME® model, due to the larger amounts of acetate, propionate, butyrate and ammonium ions obtained after 14 d of supplementation. In conclusion, both ways of probiotic supplementation could be efficient, each with its own particularities.

Human milk oligosaccharides, such as 2′-fucosyllactose (2′-FL), and galacto-oligosaccharides (GOS), a prebiotic carbohydrate mixture, are being increasingly added to infant formulas, necessitating the understanding of their impact on the oral microbiota. Here, for the first time, the effects of 2′-FL and GOS on the planktonic growth and adhesion characteristics of the caries-associated oral pathogen Streptococcus mutans were assessed, and the results were compared against the effects of xylitol, lactose and glucose. There were differences in S. mutans growth between 2′-FL and GOS. None of the three S. mutans strains grew with 2′-FL, while they all grew with GOS as well as lactose and glucose. Xylitol inhibited S. mutans growth. The adhesion of S. mutans CI 2366 to saliva-coated hydroxyapatite was reduced by 2′-FL and GOS. Exopolysaccharide-mediated adhesion of S. mutans DSM 20523 to a glass surface was decreased with 2′-FL, GOS and lactose, and the adhesion of strain CI 2366 strain was reduced only by GOS. Unlike GOS, 2′-FL did not support the growth of any S. mutans strain. Neither 2′-FL nor GOS enhanced the adhesive properties of the S. mutans strains, but they inhibited some of the tested strains. Thus, the cariogenic tendency may vary between infant formulas containing different types of oligosaccharides.

Here, we explored the influences of dietary inulin (INU) supplementation on growth performance and intestinal health in a porcine model. Thirty-two male weaned pigs (with an average body weight of 7·10 (sd 0·20) kg) were randomly assigned to four treatments and fed with a basal diet (BD) or BD containing 2·5, 5·0 and 10·0 g/kg INU. After a 21-d trial, pigs were killed for collection of serum and intestinal tissues. We show that INU supplementation had no significant influence on the growth performance in weaned pigs. INU significantly elevated serum insulin-like growth factor-1 concentration but decreased diamine oxidase concentration (P < 0·05). Interestingly, 2·5 and 5·0 g/kg INU supplementation significantly elevated the villus height in jejunum and ileum (P < 0·05). Moreover, 2·5 and 5·0 g/kg INU supplementation also elevated the villus height to crypt depth (V:C) in the duodenum and ileum and improved the distribution and abundance of tight-junction protein zonula occludens-1 in duodenum and ileum epithelium. INU supplementation at 10·0 g/kg significantly elevated the sucrase activity in the ileum mucosa (P < 0·05). INU supplementation decreased the expression level of TNF-α but elevated the expression level of GLUT 2 and divalent metal transporter 1 in the intestinal mucosa (P < 0·05). Moreover, INU increased acetic and butyric acid concentrations in caecum (P < 0·05). Importantly, INU elevated the Lactobacillus population but decreased the Escherichia coli population in the caecum (P < 0·05). These results not only indicate a beneficial effect of INU on growth performance and intestinal barrier functions but also offer potential mechanisms behind the dietary fibre-regulated intestinal health.

Several prospective studies have shown an association between cows’ milk consumption and the risk of islet autoimmunity and/or type 1 diabetes. We wanted to study whether processing of milk plays a role. A population-based birth cohort of 6081 children with HLA-DQB1-conferred risk to type 1 diabetes was followed until the age of 15 years. We included 5545 children in the analyses. Food records were completed at the ages of 3 and 6 months and 1, 2, 3, 4 and 6 years, and diabetes-associated autoantibodies were measured at 3–12-month intervals. For milk products in the food composition database, we used conventional and processing-based classifications. We analysed the data using a joint model for longitudinal and time-to-event data. By the age of 6 years, islet autoimmunity developed in 246 children. Consumption of all cows’ milk products together (energy-adjusted hazard ratio 1·06; 95 % CI 1·02, 1·11; P = 0·003), non-fermented milk products (1·06; 95 % CI 1·01, 1·10; P = 0·011) and fermented milk products (1·35; 95 % CI 1·10, 1·67; P = 0·005) was associated with an increased risk of islet autoimmunity. The early milk consumption was not associated with the risk beyond 6 years. We observed no clear differences based on milk homogenisation and heat treatment. Our results are consistent with the previous studies, which indicate that high milk consumption may cause islet autoimmunity in children at increased genetic risk. The study did not identify any specific type of milk processing that would clearly stand out as a sole risk factor apart from other milk products.

This study investigated metabolic, endocrine, appetite and mood responses to a maximal eating occasion in fourteen men (mean: age 28 (sd 5) years, body mass 77·2 (sd 6·6) kg and BMI 24·2 (sd 2·2) kg/m2) who completed two trials in a randomised crossover design. On each occasion, participants ate a homogenous mixed-macronutrient meal (pizza). On one occasion, they ate until ‘comfortably full’ (ad libitum) and on the other, until they ‘could not eat another bite’ (maximal). Mean energy intake was double in the maximal (13 024 (95 % CI 10 964, 15 084) kJ; 3113 (95 % CI 2620, 3605) kcal) compared with the ad libitum trial (6627 (95 % CI 5708, 7547) kJ; 1584 (95 % CI 1364, 1804) kcal). Serum insulin incremental AUC (iAUC) increased approximately 1·5-fold in the maximal compared with ad libitum trial (mean: ad libitum 43·8 (95 % CI 28·3, 59·3) nmol/l × 240 min and maximal 67·7 (95 % CI 47·0, 88·5) nmol/l × 240 min, P < 0·01), but glucose iAUC did not differ between trials (ad libitum 94·3 (95 % CI 30·3, 158·2) mmol/l × 240 min and maximal 126·5 (95 % CI 76·9, 176·0) mmol/l × 240 min, P = 0·19). TAG iAUC was approximately 1·5-fold greater in the maximal v. ad libitum trial (ad libitum 98·6 (95 % CI 69·9, 127·2) mmol/l × 240 min and maximal 146·4 (95 % CI 88·6, 204·1) mmol/l × 240 min, P < 0·01). Total glucagon-like peptide-1, glucose-dependent insulinotropic peptide and peptide tyrosine–tyrosine iAUC were greater in the maximal compared with ad libitum trial (P < 0·05). Total ghrelin concentrations decreased to a similar extent, but AUC was slightly lower in the maximal v. ad libitum trial (P = 0·02). There were marked differences on appetite and mood between trials, most notably maximal eating caused a prolonged increase in lethargy. Healthy men have the capacity to eat twice the energy content required to achieve comfortable fullness at a single meal. Postprandial glycaemia is well regulated following initial overeating, with elevated postprandial insulinaemia probably contributing.

Whole-grain wheat, in particular coloured varieties, may have health benefits in adults with chronic metabolic disease risk factors. Twenty-nine overweight and obese adults with chronic inflammation (high-sensitivity C-reactive protein) > 1·0 mg/l) replaced four daily servings of refined grain food products with bran-enriched purple or regular whole-wheat convenience bars (approximately 41–45 g fibre, daily) for 8 weeks in a randomised, single-blind parallel-arm study where body weight was maintained. Anthropometrics, blood markers of inflammation, oxidative stress, and lipaemia and metabolites of anthocyanins and phenolic acids were compared at days 1, 29 and 57 using repeated-measures ANOVA within groups and ANCOVA between groups at day 57, with day 1 as a covariate. A significant reduction in IL-6 and increase in adiponectin were observed within the purple wheat (PW) group. TNF-α was lowered in both groups and ferulic acid concentration increased in the regular wheat (RW) group. Comparing between wheats, only plasma TNF-α and glucose differed significantly (P < 0·05), that is, TNF-α and glucose decreased with RW and PW, respectively. Consumption of PW or RW products showed potential to improve plasma markers of inflammation and oxidative stress in participants with evidence of chronic inflammation, with modest differences observed based on type of wheat.

The Ethiopian great famine was one of the severe forms of global famines ever documented in Africa as well as in the recent history of the world. Earlier famine studies, as natural experiments, had tested the association between prenatal famine exposure and the metabolic syndrome and reported heterogeneous findings. Hence, this study aimed at evaluating the effects of prenatal exposure to the 1983–1985 Ethiopian great famine on the metabolic syndrome in adults. Self-reported birth date and age of the study subjects were used to classify the status of famine exposure. The International Diabetes Federation criterion was used to assess the metabolic syndrome. Multivariable logistic regression models were fitted to examine relationship between prenatal famine exposure and the metabolic syndrome. The findings showed that, adjusted for covariates, adults who had prenatal exposure to famine were 2·94 times more likely to develop the metabolic syndrome compared with non-exposed groups (adjusted OR (AOR) 2·94, 95 % CI 1·66, 5·27). More specifically, famine exposure during prenatal life was associated with increased waist circumference (AOR 2·27 cm, 95 % CI 0·28, 4·26), diastolic blood pressure (AOR 2·47 mmHg, 95 % CI 0·84, 4·11), TAG (AOR 0·20 mmol/l, 95 % CI 0·10, 0·28) and fasting blood glucose (AOR 0·24 mmol/l, 95 % CI 0·04, 0·43) compared with the control groups. Higher proportion of the metabolic syndrome, risky anthropometric and dyslipidaemic parameters were observed among exposed groups. This finding adds further evidence on fetal origin of adult diseases hypothesis. The finding may imply that one potential means of preventing adulthood metabolic syndrome is to optimise maternal nutrition during pregnancy.

The European Society for Clinical Nutrition and Metabolism (ESPEN) guidelines recommend the Royal Free Hospital-Nutritional Prioritizing Tool (RFH-NPT) to identify malnutrition risk in patients with liver disease. However, little is known about the application of the RFH-NPT to screen for the risk of malnutrition in China, where patients primarily suffer from hepatitis virus-related cirrhosis. A total of 155 cirrhosis patients without liver cancer or uncontrolled co-morbid illness were enrolled in this prospective study. We administered the Nutritional Risk Screening 2002 (NRS-2002), RFH-NPT, Malnutrition Universal Screening Tool (MUST) and Liver Disease Undernutrition Screening Tool (LDUST) to the patients within 24 h after admission and performed follow-up observations for 1·5 years. The RFH-NPT and NRS-2002 had higher sensitivities (64·8 and 52·4 %) and specificities (60 and 70 %) than the other tools with regard to screening for malnutrition risk in cirrhotic patients. The prevalence of nutritional risk was higher under the use of the RFH-NPT against the NRS-2002 (63 v. 51 %). The RFH-NPT tended more easily to detect malnutrition risk in patients with advanced Child–Pugh classes (B and C) and lower Model for End-stage Liver Disease scores (<15) compared with NRS-2002. RFH-NPT score was an independent predictive factor for mortality. Patients identified as being at high malnutrition risk with the RFH-NPT had a higher mortality rate than those at low risk; the same result was not obtained with the NRS-2002. Therefore, we suggest that using the RFH-NPT improves the ability of clinicians to predict malnutrition risk in patients with cirrhosis primarily caused by hepatitis virus infection at an earlier stage.

Preoperative nutritional status plays an important role in predicting postoperative outcomes. Prognostic Nutritional Index (PNI) and Controlling Nutritional Status (CONUT) are good tools to assess patients’ nutritional status. They have been used in predicting outcomes in various malignancies, but few studies have focused on pancreatic adenocarcinoma (PDAC) patients. Totally, 306 PDAC patients were enrolled. The propensity score matching (PSM) method was introduced to eliminate the baseline inequivalence. Patients with different PNI (or CONUT) scores showed inequivalence baseline characteristics, and patients with compromised nutritional status were related with a more advanced tumour stage. After PSM, the baseline characteristics were well balanced. Both low PNI (≤45) and high CONUT (≥3) were independent risk factors for poor overall survival (P < 0·05), and the result remained the same after PSM. Survival analysis demonstrated both patients with low PNI and high CONUT score were associated with poorer survival, and the result remained the same after PSM. The results of AUC indicated that CONUT might have a higher sensitivity and specificity in predicting complications and survival. Preoperative low PNI (≤45) and high CONUT (≥3) scores might be reliable predictors of prognosis and surgical complications in PDAC patients. Compared with PNI, CONUT might be more effective.

Evidence linking fasting plasma total homocysteine (tHcy) and methylenetetrahydrofolate reductase (MTHFR) 677C>T genotype with hypertension is inconsistent. Differences in B vitamin status, other lifestyle factors or their consideration in analyses might explain this. We investigated these associations in the absence of mandatory fortification with folic acid and B vitamin supplement use. A cross-sectional study was conducted in 788 adults, aged 18–75 years, randomly selected from three Catalonian town population registers. Fasting plasma folate, cobalamin, tHcy, erythrocyte folate, erythrocyte glutathione reductase activation coefficient (EGRAC, functional riboflavin status indicator; increasing EGRAC indicates worsening riboflavin status), MTHFR 677C>T and solute carrier family 1 (SLC19A1) 80 G>A genotypes were determined. Medical history and lifestyle habits were recorded. Principal tHcy determinants differed between women (age, plasma folate, plasma cobalamin, cigarettes/d) and men (MTHFR 677TT genotype, plasma folate, plasma cobalamin and CT genotype). The MTHFR 677C>T polymorphism–tHcy association (β standardised regression coefficients) was stronger in male smokers (0·52, P < 0·001) compared with non-smokers (0·21, P = 0·001) and weaker in participants aged >50 years (0·19, P = 0·007) compared with ≤50 years (0·31, P < 0·001). Hypertension was more probable in the third tHcy tertile compared with other tertiles (OR 1·9; 95 % CI 1·2, 3·0), and in participants aged ≤50 years, for the MTHFR 677TT genotype compared with the CC genotype (OR 4·1; 95 % CI 1·0, 16·9). EGRAC was associated with increased probability of hypertension in participants aged >50 years (OR 6·2; 95 % CI 1·0, 38·7). In conclusion, moderately elevated tHcy and the MTHFR 677CT genotype were associated with hypertension. The MTHFR 677C>T genotype–hypertension association was confined to adults aged ≤50 years.

Most childhood disabilities are caused by congenital factors such as birth defects. The present study aims to evaluate the effect of periconceptional nutrition intervention on the prevention of congenital disability among Chinese children using the National Birth Defects Intervention Project as a natural experiment. We obtained individual-level data from the Second National Sample Survey on Disability, a nationally representative survey, and 110 365 children born between September 1999 and August 2003 were included for analysis. Difference-in-differences estimates of the project effects on congenital disability were captured by exploiting temporal variation in the timing of project exposure across four birth cohorts along with geographical variation in project category at the province level. The findings contribute to an emerging body of evidence showing that prenatal micronutrient intervention before and during early pregnancy could substantially reduce the risk of congenital disability in childhood (OR 0·73; 95 % CI 0·57, 0·94). The National Birth Defects Intervention Project improved the awareness of reproductive health and disability prevention in the population. It highlights the need for a potential policy change focusing on early-life health investment in China.

Although legumes are rich in protein and fibre, and low in saturated fat and Na, traditional legume-based recipes include substantial amounts of processed meat, salt and potatoes, which could counteract the potential benefits of legumes. This prospective study aimed to assess the longitudinal association of consumption of different types of legumes, and traditional legume-based recipes, with unhealthy ageing in older adults. Data were taken from 2505 individuals aged ≥60 years from the Seniors-ENRICA cohort. Habitual legume consumption was assessed in 2008–2010 with a validated diet history. Unhealthy ageing was measured in the 2013, 2015 and 2017 follow-up waves, with a fifty-two-item multidimensional health deficit accumulation index (DAI) which ranges from 0 (best) to 100 (worst health). The mean age was 68·7 years, with 53·1 % of women. Among study participants, 78·4 % reported consumption of legumes, with a mean intake of 57·9 g/d. Multivariable-adjusted linear regression models did not show an association between total legume consumption and the DAI over a 7-year follow-up (non-standardised coefficient for the second and highest v. the lowest tertile of consumption: 0·94 (95 % CI −0·30, 2·17) and 0·18 (95 % CI −1·07, 1·43), respectively; Ptrend = 0·35). Similar results were observed for the 3-year and 5-year follow-ups and, separately, for lentils, beans, chickpeas and traditional legume-based recipes. According to the results obtained, consumption of legumes and traditional legume-based recipes is not associated with unhealthy ageing and can be part of a healthy diet in old age.

No data exist on the associations of dietary tomato and lycopene consumption with total and cause-specific mortality. Using the National Health and Nutrition Examination Surveys 1999–2010, we evaluted the long-term impact of tomato and lycopene intake on total and cause-specific (CHD and cerebrovascular disease) mortality. We also assessed the changes in cardio-metabolic risk factors according to tomato and lycopene intake. Vital status to 31 December 2011 was ascertained. Cox proportional hazard regression models (followed by propensity score matching) were used to investigate the link between tomato and lycopene consumption total, CHD and cerebrovascular mortality. Among the 23 935 participants included (mean age = 47·6 years, 48·8 % men), 3403 deaths occurred during 76·4 months of follow-up. Tomato intake was inversely associated with total (risk ratio (RR) 0·86, 95 % CI 0·81, 0·92), CHD (RR 0·76, 95 % CI 0·70, 0·85) and cerebrovascular (RR 0·70, 95 % CI 0·62, 0·81) mortality. Similar inverse associations were found between lycopene consumption, total (RR 0·76, 95 % CI 0·72, 0·81), CHD (RR 0·73, 95 % CI 0·65, 0·83) and cerebrovascular (RR 0·71, 95 % CI 0·65, 0·78) mortality; these associations were independent of anthropometric, clinical and nutritional parameters. Age and obesity did not affect the association of tomato and lycopene consumption with total, CHD and cerebrovascular mortality. C-reactive protein significantly moderated the link between lycopene and tomato intake with total, CHD and cerebrovascular mortality. ANCOVA showed that participants with a higher tomato and lycopene consumption had a more cardio-protective profile compared with those with a lower intake. Our results highlighted the favourable effect of tomato and lycopene intake on total and cause-specific mortality as well as on cardio-metabolic risk factors. These findings should be taken into consideration for public health strategies.

The aim of the study was to assess the accuracy and readability of Internet prenatal nutrition advice. Between August and December 2018, 130 Internet pages returned from Google searches on foods to avoid, foods to eat and supplements use were compared with UK government advice for pregnant women. Readability was assessed using the Flesch Readability Ease (FRE) tool. Descriptive and non-parametric tests were used. Spearman’s correlation explored associations between accuracy and readability. Kruskal–Wallis tests with Bonferroni correction were used for multiple pairwise tests and Mann–Whitney U tests for two-sample differences in medians. A total of 130 Internet pages were examined: 48 % from publishers, 27 % from other commercial organisations, 22 % from charities and 3 % from governments. Eighty-three (64 %) pages contained inaccurate and accurate advice, twenty-three (18 %) were accurate and complete, twenty-one (16 %) were inaccurate, and three (2 %) lacked any relevant advice. The median percentage accuracy of all advice was 83 (lower quartile, upper quartile: 48, 100). Median FRE was 55 (46, 61) ‘fairly difficult’. Eighty-seven pages (67 %) scored below the recommended FRE for public Internet pages. There was a weak positive correlation between accuracy and readability of Internet pages (rho = 0·241, P = 0·006). Accuracy of Internet pages did not differ by dietary theme. Pages on supplements were the most difficult to read. Internet pages from publishers and other commercial organisations were significantly less accurate than those from not-for-profit organisations (median percentage difference –8 (–29, 0·00), P = 0·019). Much pregnancy-related dietary advice online is inaccurate and difficult to read. Advice should be developed in consultation with qualified nutritionists and dietitians.

The present study aims to investigate the effect of wholegrain and legume consumption on the incidence of age-related cataract in an older Australian population-based cohort. The Blue Mountains Eye Study (BMES) is a population-based cohort study of eye diseases among older adults aged 49 years or older (1992–1994, n 3654). Of 2334 participants of the second examination of the BMES (BMES 2, 1997–2000), 1541 (78·3 % of survivors) were examined 5 years later (BMES 3) who had wholegrain and legume consumption estimated from the FFQ at BMES 2. Cataract was assessed using photographs taken during examinations following the Wisconsin cataract grading system. Multivariable-adjusted logistic regression models were used to assess associations with the 5-year incidence of cataract from BMES 2 (baseline) to BMES 3. The 5-year incidence of cortical, nuclear and posterior subcapsular (PSC) cataract was 18·2, 16·5 and 5·9 %, respectively. After adjustment for age, sex and other factors, total wholegrain consumption at baseline was not associated with incidence of any type of cataract. High consumption of legumes showed a protective association for incident PSC cataract (5th quintile: adjusted OR 0·37; 95 % CI 0·15, 0·92). There was no significant trend of this association across quintiles (P = 0·08). In this older Australian population, we found no associations between wholegrain intake at baseline and the 5-year incidence of three cataract types. However, intake of legumes in the highest quintile, compared with the lowest quintile, may protect against PSC formation, a finding needing replication in other studies.

From 2008, the UK’s National Diet and Nutrition Survey (NDNS) changed the method of dietary data collection from a 7-d weighed diary to a 4-d unweighed diary, partly to reduce participant burden. This study aimed to test whether self-reported energy intake changed significantly over the 4-d recording period of the NDNS rolling programme. Analyses used data from the NDNS years 1 (2008/2009) to 8 (2015/2016) inclusive, from participants aged 13 years and older. Dietary records from participants who reported unusual amounts of food and drink consumed on one or more days were excluded, leaving 6932 participants. Mean daily energy intake was 7107 kJ (1698 kcal), and there was a significant decrease of 164 kJ (39 kcal) between days 1 and 4 (P < 0·001). There was no significant interaction of sex or low-energy reporter status (estimated from the ratio of reported energy intake:BMR) with the change in reported energy intake. The decrease in reported energy intake on day 4 compared with day 1 was greater (P < 0·019) for adults with higher BMI (>30 kg/m2) than it was for leaner adults. Reported energy intake decreased over the 4-d recording period of the NDNS rolling programme suggesting that participants change their diet more, or report less completely, with successive days of recording their diet. The size of the effect was relatively minor, however.