Knowledge about the association between alcohol and Barrett’s oesophagus and reflux oesophagitis is conflicting. In this case–control study we evaluated the role of specific alcoholic beverages (red and white wine, beer and liquors) in 339 Barrett’s oesophagus and 462 oesophagitis patients compared with 619 endoscopic controls with other disorders, recruited in twelve Italian endoscopic units. Data on alcohol and other individual characteristics were obtained from structured questionnaires. No clear, monotonic significant dose–response relationship was pointed out for red wine. However, a generalised U-shaped trend of Barrett’s oesophagus/oesophagitis risk due to red wine consumption particularly among current drinkers was found. Similar results were also found for white wine. Liquor/spirit consumption seemed to bring about a 1·14–2·30 risk excess, although statistically non-significant, for current Barrett’s oesophagus/oesophagitis drinkers. Statistically significant decreasing dose–response relationships were found in Barrett’s oesophagus for frequency and duration of beer consumption. Similar, but less clear downward tendencies were also found for oesophagitis patients. In conclusion, although often not statistically significant, our data suggested a reduced risk of Barrett’s oesophagus and oesophagitis with a low/moderate intake of wine and beer consumption. A non-significant increased risk of Barrett’s oesophagus/oesophagitis was observed with a higher intake of any type of heavy alcohol consumption, but no conclusion can be drawn owing to the high number of non-spirit drinkers and to the small number of drinkers at higher alcohol intake levels.

Carotenoids are found in abundance in fruit and vegetables, and may be involved in the positive association of these foods with bone health. This study aimed to explore the associations of dietary carotenoid intakes and plasma concentrations with bone density status and osteoporotic fracture risk in a European population. Cross-sectional analyses (n 14 803) of bone density status, using calcaneal broadband ultrasound attenuation (BUA) and longitudinal analyses (n 25 439) of fracture cases were conducted on data from the prospective European Prospective Investigation into Cancer and Nutrition-Norfolk cohort of middle-aged and older men and women. Health and lifestyle questionnaires were completed, and dietary nutrient intakes were derived from 7-d food diaries. Multiple regression demonstrated significant positive trends in BUA for women across quintiles of dietary α-carotene intake (P=0·029), β-carotene intake (P=0·003), β-cryptoxanthin intake (P=0·031), combined lutein and zeaxanthin intake (P=0·010) and lycopene intake (P=0·005). No significant trends across plasma carotenoid concentration quintiles were apparent (n 4570). The Prentice-weighted Cox regression showed no trends in fracture risk across dietary carotenoid intake quintiles (mean follow-up time 12·5 years), except for a lower risk for wrist fracture in women with higher lutein and zeaxanthin intake (P=0·022); nevertheless, inter-quintile differences in fracture risk were found for both sexes. Analysis of plasma carotenoid data (mean follow-up time 11·9 years) showed lower hip fracture risk in men across higher plasma α-carotene (P=0·026) and β-carotene (P=0·027) quintiles. This study provides novel evidence that dietary carotenoid intake is relevant to bone health in men and women, demonstrating that associations with bone density status and fracture risk exist for dietary intake of specific carotenoids and their plasma concentrations.

Accurate data on dietary intake are important for public health, nutrition and agricultural policy. The National Sample Survey is widely used by policymakers in India to estimate nutritional outcomes in the country, but has not been compared with other dietary data sources. To assess relative differences across available Indian dietary data sources, we compare intake of food groups across six national and sub-national surveys between 2004 and 2012, representing various dietary intake estimation methodologies, including Household Consumption Expenditure Surveys (HCES), FFQ, food balance sheets (FBS), and 24-h recall (24HR) surveys. We matched data for relevant years, regions and economic groups, for ages 16–59. One set of national HCES and the 24HR showed a decline in food intake in India between 2004–2005 and 2011–2012, whereas another HCES and FBS showed an increase. Differences in intake were smallest between the two HCES (1 % relative difference). Relative to these, FFQ and FBS had higher intake (13 and 35 %), and the 24HR lower intake (−9 %). Cereal consumption had high agreement across comparisons (average 5 % difference), whereas fruit and nuts, eggs, meat and fish and sugar had the least (120, 119, 56 and 50 % average differences, respectively). Spearman’s coefficients showed high correlation of ranked food group intake across surveys. The underlying methods of the compared data highlight possible sources of under- or over-estimation, and influence their relevance for addressing various research questions and programmatic needs.

DHA from diet or endogenous synthesis has been proposed to affect infant development, however, results are inconclusive. In this study, we aim to verify previously observed fatty acid desaturase gene cluster (FADS) SNP-specific associations with erythrocyte DHA status in 9-month-old children and sex-specific association with developmental outcomes. The study was performed in 166 children (55 % boys) of obese mothers. Erythrocyte fatty acid composition was analysed in blood-samples obtained at 9 months of age, and developmental outcomes assessed by the Ages and Stages Questionnaire at 3 years. Erythrocyte DHA level ranged from 4·4 to 9·9 % of fatty acids, but did not show any association with FADS SNP or other potential determinants. Regression analysis showed associations between erythrocyte DHA and scores for personal–social skills (β 1·8 (95 % CI 0·3, 3·3), P=0·019) and problem solving (β 3·4 (95 % CI 1·2, 5·6), P=0·003). A tendency was observed for an association in opposite direction between minor alleles (G-variant) of rs1535 and rs174575 and personal–social skills (P=0·062 and 0·068, respectively), which became significant when the SNP were combined based on their previously observed effect on erythrocyte DHA at 9 months of age (β 2·6 (95 % CI 0·01, 5·1), P=0·011). Sex–SNP interaction was indicated for rs174575 genotype on fine motor scores (P=0·016), due to higher scores among minor allele carrying girls (P=0·043), whereas no effect was seen among boys. In conclusion, DHA-increasing FADS SNP and erythrocyte DHA status were consistently associated with improved personal–social skills in this small cohort of children of obese mothers irrespective of sex, but the sample was too small to verify potential sex-specific effects.

Previous studies have yielded conflicting results on the associations of maternal Fe intake with birth outcomes. This study aimed to investigate the associations between maternal Fe intake (total Fe from diet and supplements, dietary total Fe, haeme Fe, non-haeme Fe and Fe supplements use) and adverse birth outcomes in Shaanxi Province of Northwest China. In all, 7375 women were recruited using a stratified multistage random sampling method at 0–12 months (median 3; 10th–90th percentile 0–7) after delivery. Diets were collected by a validated FFQ and maternal characteristics were obtained via a standard questionnaire. The highest tertile of haeme Fe intake compared with the lowest tertile was negatively associated with low birth weight (LBW) (OR 0·68; 95 % CI 0·49, 0·94), small for gestational age (SGA) (OR 0·76; 95 % CI 0·62, 0·94) and birth defects (OR 0·55; 95 % CI 0·32, 0·89). Maternal haeme Fe intake was associated with a lower risk of intra-uterine growth retardation (IUGR) (medium tertile v. lowest tertile: OR 0·78; 95 % CI 0·61, 0·95; highest tertile v. lowest tertile: OR 0·76; 95 % CI 0·59, 0·93; Ptrend=0·045). The OR of LBW associated with Fe supplements use were as follows: during pregnancy: 0·72 (95 % CI 0·50, 0·95); in the second trimester: 0·67 (95 % CI 0·42, 0·98); in the third trimester: 0·47 (95 % CI 0·24, 0·93). We observed no associations of total Fe, dietary total Fe or non-haeme Fe intake with birth outcomes. The results suggest that maternal haeme Fe intake is associated with a reduced risk of LBW, SGA, IUGR and birth defects, and Fe supplements use during pregnancy reduces LBW risk.

This study aimed to examine the association between vitamin B6, folate and vitamin B12 biomarkers and plasma fatty acids in European adolescents. A subsample from the Healthy Lifestyle in Europe by Nutrition in Adolescence study with valid data on B-vitamins and fatty acid blood parameters, and all the other covariates used in the analyses such as BMI, Diet Quality Index, education of the mother and physical activity assessed by a questionnaire, was selected resulting in 674 cases (43 % males). B-vitamin biomarkers were measured by chromatography and immunoassay and fatty acids by enzymatic analyses. Linear mixed models elucidated the association between B-vitamins and fatty acid blood parameters (changes in fatty acid profiles according to change in 10 units of vitamin B biomarkers). DHA, EPA) and n-3 fatty acids showed positive associations with B-vitamin biomarkers, mainly with those corresponding to folate and vitamin B12. Contrarily, negative associations were found with n-6:n-3 ratio, trans-fatty acids and oleic:stearic ratio. With total homocysteine (tHcy), all the associations found with these parameters were opposite (for instance, an increase of 10 nmol/l in red blood cell folate or holotranscobalamin in females produces an increase of 15·85 µmol/l of EPA (P value <0·01), whereas an increase of 10 nmol/l of tHcy in males produces a decrease of 2·06 µmol/l of DHA (P value <0·05). Positive associations between B-vitamins and specific fatty acids might suggest underlying mechanisms between B-vitamins and CVD and it is worth the attention of public health policies.

Anaemia is prevalent in developing countries and is commonly Fe deficiency related. We aimed to assess the association between Fe status, Fe intake and mortality among Chinese adults. We prospectively studied 8291 adults aged 20–98 years with a mean follow-up of 9·9 years. All participants were measured for Hb at baseline in 2002. Food intake, measured by 3-d weighed food record (n 2832), and fasting serum ferritin were measured. We documented 491 deaths (including 192 CVD and 165 cancer deaths) during 81 527 person-years of follow-up. There was a U-shaped association between Hb levels and all-cause mortality. Compared with the second quartile of Hb (121 g/l), the first (105) and fourth quartile (144) had hazard ratios (HR) of 2·29 (95 % CI 1·51, 3·48) and 2·31 (95 % CI 1·46, 3·64) for all-cause mortality in women. In men, compared with third quartile of Hb (143 g/l), first (122) and fourth quartiles (154) had 61 and 65 % increased risk of all-cause mortality. Anaemia was associated with an increased risk of all-cause and CVD mortality in men but not in women after adjusting for potential confounders. Low and high Fe intake as percentage of Chinese recommended nutrient intake (RNI) were positively associated with all-cause mortality in women but not in men. In women, across quartiles of relative Fe intake, HR for all-cause mortality were 2·55 (95 % CI 0·99, 6·57), 1·00, 3·12 (95 % CI 1·35, 7·18) and 2·78 (95 % CI 1·02, 7·58). Both low and high Hb levels are related to increased risk of all-cause mortality. Both low and high intake of Fe as percentage of RNI was positively associated with mortality in women.

The deficiency of Se, an essential micronutrient, has been implicated in adverse pregnancy outcomes. Our study was designed to determine total serum Se, selenoproteins (extracellular glutathione peroxidase (GPx-3), selenoprotein P (SeP)), selenoalbumin (SeAlb) and selenometabolites in healthy women and their newborns at delivery. This cross-sectional study included eighty-three healthy mother–baby couples. Total Se and Se species concentrations were measured in maternal and umbilical cord sera by an in-series coupling of two-dimensional size-exclusion and affinity HPLC. Additional measurements of serum SeP concentration and of serum GPx-3 enzyme activity were carried out using ELISA. Total Se concentration was significantly higher in maternal serum than in cord serum (68·9 (sd 15·2) and 56·1 (sd 14·6) µg/l, respectively; P<0·01). There were significant correlations between selenoprotein and SeAlb concentrations in mothers and newborns, although they also showed significant differences in GPx-3 (11·2 (sd 3·7) v. 10·5 (sd 3·5) µg/l; P<0·01), SeP (42·5 (sd 9·5) v. 28·1 (sd 7·7) µg/l; P<0·01) and SeAlb (11·6 (sd 3·6) v. 14·1 (sd 4·3) µg/l; P<0·01) concentrations in maternal and cord sera, respectively. Serum GPx-3 activity and concentration were positively correlated in mothers (r 0·33; P=0·038) but not in newborns. GPx-3 activity in cord serum was significantly correlated with gestational age (r 0·44; P=0·009). SeAlb concentration was significantly higher in babies, whereas SeP and GPx-3 concentrations were significantly higher in mothers. The differences cannot be explained by simple diffusion; specific transfer mechanisms are probably involved. GPx-3 concentrations in mothers, at delivery, are related to maternal Se status, whereas the GPx-3 activity in cord serum depends on gestational age.

Arterial wall thickening, stimulated by low-grade systemic inflammation, underlies many cardiovascular events. As diet is a significant moderator of systemic inflammation, the dietary inflammatory index (DIITM) has recently been devised to assess the overall inflammatory potential of an individual’s diet. The primary objective of this study was to assess the association of the DII with common carotid artery–intima-media thickness (CCA–IMT) and carotid plaques. To substantiate the clinical importance of these findings we assessed the relationship of DII score with atherosclerotic vascular disease (ASVD)-related mortality, ischaemic cerebrovascular disease (CVA)-related mortality and ischaemic heart disease (IHD)-related mortality more. The study was conducted in Western Australian women aged over 70 years (n 1304). Dietary data derived from a validated FFQ (completed at baseline) were used to calculate a DII score for each individual. In multivariable-adjusted models, DII scores were associated with sub-clinical atherosclerosis: a 1 sd (2·13 units) higher DII score was associated with a 0·013-mm higher mean CCA–IMT (P=0·016) and a 0·016-mm higher maximum CCA–IMT (P=0·008), measured at 36 months. No relationship was seen between DII score and carotid plaque severity. There were 269 deaths during follow-up. High DII scores were positively associated with ASVD-related death (per sd, hazard ratio (HR): 1·36; 95 % CI 1·15, 1·60), CVA-related death (per sd, HR: 1·30; 95 % CI 1·00, 1·69) and IHD-related death (per sd, HR: 1·40; 95 % CI 1·13, 1·75). These results support the hypothesis that a pro-inflammatory diet increases systemic inflammation leading to development and progression of atherosclerosis and eventual ASVD-related death.

The effect of fruit and vegetable intake on breast cancer prognosis is controversial. Thus, a meta-analysis was carried out to explore their associations. A comprehensive search was conducted in PubMed, Web of Science, OVID, ProQuest and Chinese databases from inception to April 2016. The summary hazard ratios (HR) and 95 % CI were estimated using a random effects model if substantial heterogeneity existed and using a fixed effects model if not. Subgroup analyses and sensitivity analyses were also performed. In total, twelve studies comprising 41 185 participants were included in the meta-analysis. Comparing the highest with the lowest, the summary HR for all-cause mortality were 1·01 (95 % CI 0·72, 1·42) for fruits and vegetables combined, 0·96 (95 % CI 0·83, 1·12) for total vegetable intake, 0·99 (95 % CI 0·89, 1·11) for cruciferous vegetable intake and 0·88 (95 % CI 0·74, 1·05) for fruit intake; those for breast cancer-specific mortality were 1·05 (95 % CI 0·77, 1·43) for total vegetable intake and 0·94 (95 % CI 0·69, 1·26) for fruit intake; and those for breast cancer recurrence were 0·89 (95 % CI 0·53, 1·50) for total vegetable intake and 0·98 (95 % CI 0·76, 1·26) for cruciferous vegetable intake. This meta-analysis found no significant associations between fruit and vegetable intake and breast cancer prognosis.

Sufficient I intake is important for the synthesis of thyroid hormones, which play an important role in normal growth and development. Our aim was to estimate habitual I intake for the Dutch population and the risk of inadequate or excessive intakes. Further, we aimed to provide an insight into the dietary sources of I and the association with socio-demographic factors. Data from the Dutch National Food Consumption Survey 2007–2010 (n 3819; 7–69 years), and from the Dutch food and supplement composition tables were used to estimate habitual I intake with a calculation model. Contribution of food groups to I intake were computed and multiple linear regression was used to examine associations of intakes with socio-demographic factors. A total of ≤2 % of the population had an intake below the estimated average requirement or above the upper level. The main sources of I were bread containing iodised salt (39 %), dairy products (14 %) and non-alcoholic drinks (6 %). I intake (natural sources only, excluding iodised salt and supplements) was positively associated with (parental) education, which could at least partly be attributed to a higher consumption of dairy products. Among children, the consumption of bread, often containing iodised bakery salt, was positively associated with parental education. The I intake of the Dutch population (7–69 years) seems adequate, although it has decreased since the period before 2008. With the current effort to reduce salt intake and changing dietary patterns (i.e. less bread, more organic foods) it is important to keep a close track on the I status, important sources and potential risk groups.

The number of days of data required to accurately estimate usual nutrient intake of children is not well established. This study aims to calculate the variability and the number of days required to estimate usual nutrient intake in children aged 13–32 months. This cross-sectional study, which is part of the BRISA Project in São Luís, Maranhão, Brazil, involved 231 children from April 2011 to January 2013. Socio-economic and demographic data were collected using a questionnaire, and 3 non-consecutive days of food consumption were collected using a 24-h dietary recall (24HDR) survey. Intrapersonal and interpersonal variability and variance ratio (VR) were obtained for each nutrient using the Multiple Source Method® program (version 1.0.1). The number of days (d) needed was calculated using the formula proposed by Black et al. for different correlation coefficients (r) (i.e. 0·7, 0·8 or 0·9). For the vast majority of nutrients, intrapersonal and interpersonal variability values of <1 were observed, with even smaller intrapersonal variabilities, resulting in low VR (<1). More days were needed to estimate intakes of soluble fibre (12), insoluble fibre (11), total fibre (10), vitamin C (9) and PUFA (7), while fewer days were needed for energy, carbohydrate, SFA, Ca, Fe, P and Zn (all had 2 d for r 0·9). However, most nutrients required one, two or three 24HDR for r 0·7, 0·8 or 0·9.

Education interventions that stimulate complementary feeding practices can improve the nutritional status of children and may protect against future chronic diseases. We assessed the long-term effectiveness of dietary intervention during the 1st year of life on insulin resistance levels, and investigated the relationship between insulin resistance and weight changes over time. A randomised field trial was conducted among 500 mothers who gave birth to full-term infants between October 2001 and June 2002 in a low-income area in São Leopoldo, Brazil. Mother–child pairs were randomly assigned to intervention (n 200) and control groups (n 300), and the mothers in the intervention group received dietary counselling on breast-feeding and complementary feeding of their children during the 1st year of life. Fieldworkers blinded to assignment assessed socio-demographic, dietary and anthropometric data during follow-up at ages 1, 4 and 8 years. Blood tests were performed in 305 children aged 8 years to measure fasting serum glucose and insulin concentrations and the homoeostasis model assessment index of insulin resistance (HOMA-IR). At the age of 8 years, the intervention group showed no changes in glucose and insulin concentrations or HOMA-IR values (change 0·07; 95 % CI −0·06, 0·21 for girls; and change −0·07; 95 % CI −0·19, 0·04 for boys) compared with study controls. Insulin resistance was highly correlated, however, with increases in BMI between birth and 8 years of age. Although this dietary intervention had no impact on glucose profile at age 8 years, our findings suggest that BMI changes in early childhood can serve as an effective marker of insulin resistance.

As biomarkers of dietary intake or disease risk factor, n-3 fatty acid (FA) can be measured in plasma phospholipids (PL), total lipids (TL) or erythrocytes. However, the numeric relationships between n-3 FA in these lipid pools are not clear. Our goal was to derive conversion ratios for plasma and erythrocyte n-3 FA. Potential studies were identified through systematic literature search in PubMed, Embase and the Cochrane Library of Systematic reviews (1950 to October 2014). In all, fifty-six studies reporting n-3 in healthy individuals were included, of which thirty-four articles reported plasma PL and erythrocytes, and twenty-two reported plasma TL and erythrocytes. Meta-regressions were performed to quantify the ratio between plasma and erythrocyte n-3 FA weight percentages, controlling for covariates including age, sex and study design. The conversion ratios from plasma PL to erythrocytes for EPA, DHA, DPA and total n-3 PUFA are 0·75, 1·16, 2·32 and 1·22; the corresponding conversion ratios from plasma TL to erythrocytes are 1·00, 2·10, 3·85 and 2·08, respectively. The conversion ratios were validated using reported values from the literature and measured data from fifty individuals. The relative error of the predicted results were within 10 % of the mean reported values except for EPA, and the individual measured data except for DPA, in plasma TL. The conversion ratios between plasma PL and erythrocytes were more stable compared with plasma TL. Such conversion ratios will be useful for nutritionists or public health professionals to assess FA profiles of different populations using data collected with different methodologies.

Stroke is a leading cause of morbidity and mortality. The role of PUFA in reducing the risk of stroke is uncertain. The concentrations of PUFA in the human body are determined both by dietary intake and by activities of desaturase enzymes. Desaturase enzymes have been associated with chronic diseases, but little is known about their association with stroke risk. We investigated the associations of Δ-6-desaturase (D6D) and Δ-5-desaturase (D5D) activities with stroke risk factors and risk of stroke among 1842 men from the prospective, population-based Kuopio Ischaemic Heart Disease Risk Factor Study, aged 42–60 years and free of CVD at baseline in 1984–1989. ANCOVA and Cox regression models were used for the analyses. Whole serum desaturase activities were estimated as product:precursor ratios – γ-linolenic acid:linoleic acid for D6D and arachidonic acid:dihomo-γ-linolenic acid for D5D. Higher D6D activity was associated with higher systolic and diastolic blood pressure, BMI, serum insulin and TAG concentrations and worse homoeostatic model assessment (HOMA) indices. In contrast, higher D5D activity was associated with lower systolic and diastolic blood pressure, BMI, serum insulin, LDL-cholesterol, TAG and C-reactive protein concentrations, higher HDL-cholesterol concentration, and better HOMA indices. During the mean follow-up of 21·2 years, 202 stroke cases occurred. Neither D6D activity (multivariable-adjusted extreme-quartile hazard ratios (HR) 1·18; 95 % CI 0·80, 1·74) nor D5D activity (HR 1·06; 95 % CI 0·70, 1·60) were associated with stroke risk. In conclusion, higher D5D activity was favourably associated and higher D6D activity unfavourably associated with several stroke risk factors, but not with the risk of incident stroke.

Probiotic yogurt and milk supplemented with probiotics have been investigated for their role in ‘low-grade’ inflammation but evidence for their efficacy is inconclusive. This study explores the impact of probiotic yogurt on metabolic and inflammatory biomarkers, with a parallel study of gut microbiota dynamics. The randomised cross-over study was conducted in fourteen healthy, young men to test probiotic yogurt compared with milk acidified with 2 % d-(+)-glucono-δ-lactone during a 2-week intervention (400 g/d). Fasting assessments, a high-fat meal test (HFM) and microbiota analyses were used to assess the intervention effects. Baseline assessments for the HFM were carried out after a run-in during which normal milk was provided. No significant differences in the inflammatory response to the HFM were observed after probiotic yogurt compared with acidified milk intake; however, both products were associated with significant reductions in the inflammatory response to the HFM compared with the baseline tests (assessed by IL6, TNFα and chemokine ligand 5) (P<0·001). These observations were accompanied by significant changes in microbiota taxa, including decreased abundance of Bilophila wadsworthia after acidified milk (log 2-fold-change (FC)=–1·5, Padj=0·05) and probiotic yogurt intake (FC=–1·3, Padj=0·03), increased abundance of Bifidobacterium species after acidified milk intake (FC=1·4, Padj=0·04) and detection of Lactobacillus delbrueckii spp. bulgaricus (FC=7·0, Padj<0·01) and Streptococcus salivarius spp. thermophilus (FC=6·0, Padj<0·01) after probiotic yogurt intake. Probiotic yogurt and acidified milk similarly reduce postprandial inflammation that is associated with a HFM while inducing distinct changes in the gut microbiota of healthy men. These observations could be relevant for dietary treatments that target ‘low-grade’ inflammation.

Adolescence represents an important period for the development of executive functions, which are a set of important cognitive processes including attentional control. However, very little is known regarding the associations of nutrition with components of executive functions in adolescence. Thus, the aim of this study was to investigate associations of dietary patterns and macronutrient composition with attention capacity in European adolescents. This cross-sectional study included 384 (165 boys and 219 girls) adolescents, aged 12·5–17·5 years, from five European countries in the Healthy Lifestyle in Europe by Nutrition in Adolescence study. Attention capacity was examined using the d2 Test of Attention. Dietary intake was assessed through two non-consecutive 24 h recalls using a computer-based self-administered tool. Three dietary patterns (diet quality index, ideal diet score and Mediterranean diet score) and macronutrient/fibre intakes were calculated. Linear regression analysis was conducted adjusting for age, sex, BMI, maternal education, family affluence scale, study centre and energy intake (only for Mediterranean diet score). In these adjusted regression analyses, higher diet quality index for adolescents and ideal diet score were associated with a higher attention capacity (standardised β=0·16, P=0·002 and β=0·15, P=0·005, respectively). Conversely, Mediterranean diet score or macronutrient/fibre intake were not associated with attention capacity (P>0·05). Our results suggest that healthier dietary patterns, as indicated by higher diet quality index and ideal diet score, were associated with attention capacity in adolescence. Intervention studies investigating a causal relationship between diet quality and attention are warranted.

No study has analysed how protein intake from early childhood to young adulthood relate to adult BMI in a single cohort. To estimate the association of protein intake at 2, 11, 15, 19 and 22 years with age- and sex-standardised BMI at 22 years (early adulthood), we used linear regression models with dietary and anthropometric data from a Filipino birth cohort (1985–2005, n 2586). We used latent growth curve analysis to identify trajectories of protein intake relative to age-specific recommended daily allowance (intake in g/kg body weight) from 2 to 22 years, then related trajectory membership to early adulthood BMI using linear regression models. Lean mass and fat mass were secondary outcomes. Regression models included socioeconomic, dietary and anthropometric confounders from early life and adulthood. Protein intake relative to needs at age 2 years was positively associated with BMI and lean mass at age 22 years, but intakes at ages 11, 15 and 22 years were inversely associated with early adulthood BMI. Individuals were classified into four mutually exclusive trajectories: (i) normal consumers (referent trajectory, 58 % of cohort), (ii) high protein consumers in infancy (20 %), (iii) usually high consumers (18 %) and (iv) always high consumers (5 %). Compared with the normal consumers, ‘usually high’ consumption was inversely associated with BMI, lean mass and fat mass at age 22 years whereas ‘always high’ consumption was inversely associated with male lean mass in males. Proximal protein intakes were more important contributors to early adult BMI relative to early-childhood protein intake; protein intake history was differentially associated with adulthood body size.