Studies that investigated complex actual eating behaviours of the general population and their relation to cardiometabolic risk markers are sparse. We aimed to identify dietary patterns within a nationally representative sample of 4025 German adults by factor analysis based on validated dietary history interviews. Furthermore, we evaluated associations of the derived dietary patterns with abnormalities clustered within the metabolic syndrome and related metabolic markers by logistic regression models and ANCOVA. A high adherence to the ‘processed foods’ pattern reflected a high intake of refined grains, processed meat, red meat, high-sugar beverages, eggs, potatoes, beer, sweets and cakes, snacks and butter, whereas a high adherence to the ‘health-conscious’ pattern represented a high intake of vegetables, vegetable oils, legumes, fruits, fish and whole grains. For subjects in the highest compared with those in the lowest quintile of the processed foods pattern, the occurrence of abdominal obesity was 88 (95 % CI 31, 169) % higher, hypertension was 34 (95 % CI − 4, 86) % higher, hypertriacylglycerolaemia was 59 (95 % CI 11, 128 ) % higher and the metabolic syndrome was 64 (95 % CI 10, 143) % higher when adjusted for age, sex, energy intake, socio-economic status, sport activity and smoking. Furthermore, subjects in the highest quintile had statistically significantly higher uric acid concentrations and lower folate concentrations (P for trend < 0·05). In contrast, subjects in the highest quintile of the health-conscious pattern had a 30 (95 % CI 10, 46) % lower occurrence of hypertension, higher folate concentrations and lower homocysteine and fibrinogen concentrations (P for trend < 0·05). These data strengthen the findings from non-representative studies and emphasise the importance of healthy overall food patterns for preventing metabolic disturbances.

Ruminants represent an important source of methane (CH4) emissions; therefore, CH4 mitigation by diet supplementation is a major goal in the current ruminant research. The objective of the present study was to use a rumen simulation technique to evaluate the CH4-mitigating potential of pure compounds in comparison with that achieved with garlic oil, a known anti-methanogenic supplement. A basal diet (15 g DM/d) consisting of ryegrass hay, barley and soyabean meal (1:0·7:0·3) was incubated with the following additives: none (negative control); garlic oil (300 mg/l incubation liquid; positive control); allyl isothiocyanate (75 mg/l); lovastatin (150 mg/l); chenodeoxycholic acid (150 mg/l); 3-azido-propionic acid ethyl ester (APEE, 150 mg/l); levulinic acid (300 mg/l); 4-[(pyridin-2-ylmethyl)-amino]-benzoic acid (PABA, 300 mg/l). Fermentation profiles (SCFA, microbial counts and N turnover) and H2 and CH4 formation were determined. Garlic oil, allyl isothiocyanate, lovastatin and the synthetic compound APEE decreased the absolute daily CH4 formation by 91, 59, 42 and 98 %, respectively. The corresponding declines in CH4 emitted per mmol of SCFA were 87, 32, 40 and 99 %, respectively, compared with the negative control; the total SCFA concentration was unaffected. Garlic oil decreased protozoal numbers and increased bacterial counts, while chenodeoxycholic acid completely defaunated the incubation liquid. In vitro, neutral-detergent fibre disappearance was lower following chenodeoxycholic acid and PABA treatments ( − 26 and − 18 %, respectively). In conclusion, garlic oil and APEE were extremely efficient at mitigating CH4 without noticeably impairing microbial nutrient fermentation. Other promising substances were allyl isothiocyanate and lovastatin.

The National Diet and Nutrition Survey (NDNS) is a cross-sectional survey designed to gather data representative of the UK population on food consumption, nutrient intakes and nutritional status. The objectives of the present paper were to identify and describe food consumption and nutrient intakes in the UK from the first year of the NDNS rolling programme (2008–09) and compare these with the 2000–01 NDNS of adults aged 19–64 years and the 1997 NDNS of young people aged 4–18 years. Differences in median daily food consumption and nutrient intakes between the surveys were compared by sex and age group (4–10 years, 11–18 years and 19–64 years). There were no changes in energy, total fat or carbohydrate intakes between the surveys. Children aged 4–10 years had significantly lower consumption of soft drinks (not low calorie), crisps and savoury snacks and chocolate confectionery in 2008–09 than in 1997 (all P < 0·0001). The percentage contribution of non-milk extrinsic sugars to food energy was also significantly lower than in 1997 in children aged 4–10 years (P < 0·0001), contributing 13·7–14·6 % in 2008–09 compared with 16·8 % in 1997. These changes were not as marked in older children and there were no changes in these foods and nutrients in adults. There was still a substantial proportion (46 %) of girls aged 11–18 years and women aged 19–64 years (21 %) with mean daily Fe intakes below the lower reference nutrient intake. Since previous surveys there have been some positive changes in intakes especially in younger children. However, further attention is required in other groups, in particular adolescent girls.

Although there are indications for modulatory effects of PUFA on associations between SNP and obesity risk, scientific evidence in human subjects is still scarce. The present analyses investigated interaction effects between SNP in candidate genes for obesity and PUFA in erythrocyte membranes on obesity risk. Within the second Bavarian Food Consumption Survey (cross-sectional, population-based), 568 adults provided blood samples. Fatty acid composition of erythrocyte membranes was analysed by means of GC. Genotyping was performed for twenty-one genes, including cytokines, adipokines, neurotransmitters and transcription factors. In addition, plasma IL-6 concentrations were analysed. For the statistical analysis, a logistic regression model assuming additive genetic effects was chosen. About 20 % of the study participants were classified as obese (BMI ≥ 30 kg/m2). Several significant gene–PUFA interactions were found, indicating regulatory effects of PUFA by gene variants of IL-2, IL-6, IL-18, TNF receptor family member 1B and 21, leptin receptor and adiponectin on obesity risk. After stratification by genotype, the strongest effects were found for rs2069779 (IL-2) and all tested PUFA as well as for rs1800795 (IL-6) and linoleic or arachidonic acid. The obesity risk of minor allele carriers significantly decreased with increasing fatty acid content. The genetic PUFA–IL-6 interaction was also reflected in plasma IL-6 concentrations. If replicated in a prospective study with sufficient statistical power, the results would indicate a beneficial effect of high PUFA supply for a substantial proportion of the population with respect to obesity risk.

Anthocyanidins are bioactive flavonoids with potential health-promoting effects. These may vary among single anthocyanidins considering differences in their bioavailability and some of the mechanisms involved. The aim of the present study was to estimate the dietary intake of anthocyanidins, their food sources and the lifestyle factors (sex, age, BMI, smoking status, educational level and physisical activity) involved among twenty-seven centres in ten European countries participating in the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Anthocyanidin intake and their food sources for 36 037 subjects, aged between 35 and 74 years, in twenty-seven redefined centres were obtained using standardised 24 h dietary recall software (EPIC-SOFT). An ad hoc food composition database on anthocyanidins (cyanidin, delphinidin, malvidin, pelargonidin, peonidin, petunidin) was compiled using data from the US Department of Agriculture and Phenol-Explorer databases and was expanded by adding recipes, estimated values and cooking factors. For men, the total anthocyanidin mean intake ranged from 19·83 (se 1·53) mg/d (Bilthoven, The Netherlands) to 64·88 (se 1·86) mg/d (Turin, Italy), whereas for women the range was 18·73 (se 2·80) mg/d (Granada, Spain) to 44·08 (se 2·45) mg/d (Turin, Italy). A clear south to north gradient intake was observed. Cyanidins and malvidins were the main anthocynidin contributors depending on the region and sex. Anthocyanidin intake was higher in non-obese older females, non-smokers, and increased with educational level and physical activity. The major food sources were fruits, wine, non-alcoholic beverages and some vegetables. The present study shows differences in both total and individual anthocyanidin intakes and various lifestyle factors throughout Europe, with some geographical variability in their food sources.

Flavonols, flavanones and flavones (FLAV) are sub-classes of flavonoids that exert cardioprotective and anti-carcinogenic properties in vitro and in vivo. We aimed to estimate the FLAV dietary intake, their food sources and associated lifestyle factors in ten European countries participating in the European Prospective Investigation into Cancer and Nutrition (EPIC) study. FLAV intake and their food sources for 36 037 subjects, aged between 35 and 74 years, in twenty-seven study centres were obtained using standardised 24 h dietary recall software (EPIC-SOFT). An ad hoc food composition database on FLAV was compiled using data from US Department of Agriculture and Phenol-Explorer databases and was expanded using recipes, estimations and flavonoid retention factors in order to increase its correspondence with the 24 h dietary recall. Our results showed that the highest FLAV-consuming centre was the UK health-conscious group, with 130·9 and 97·0 mg/d for men and women, respectively. The lowest FLAV intakes were 36·8 mg/d in men from Umeå and 37·2 mg/d in women from Malmö (Sweden). The flavanone sub-class was the main contributor to the total FLAV intake ranging from 46·6 to 52·9 % depending on the region. Flavonols ranged from 38·5 to 47·3 % and flavones from 5·8 to 8·6 %. FLAV intake was higher in women, non-smokers, increased with level of education and physical activity. The major food sources were citrus fruits and citrus-based juices (especially for flavanones), tea, wine, other fruits and some vegetables. We concluded that the present study shows heterogeneity in intake of these three sub-classes of flavonoids across European regions and highlights differences by sex and other sociodemographic and lifestyle factors.

There are concerns that weight-loss (WL) diets based on very low carbohydrate (LC) intake have a negative impact on antioxidant status and biomarkers of cardiovascular and metabolic health. Obese men (n 16) participated in a randomised, cross-over design diet trial, with food provided daily, at approximately 8·3 MJ/d (approximately 70 % of energy maintenance requirements). They were provided with two high-protein diets (30 % of energy), each for a 4-week period, involving a LC (4 % carbohydrate) and a moderate carbohydrate (MC, 35 % carbohydrate) content. Body weight was measured daily, and weekly blood samples were collected. On average, subjects lost 6·75 and 4·32 kg of weight on the LC and MC diets, respectively (P < 0·001, sed 0·350). Although the LC and MC diets were associated with a small reduction in plasma concentrations of retinol, vitamin E (α-tocopherol) and β-cryptoxanthin (P < 0·005), these were still above the values indicative of deficiency. Interestingly, plasma vitamin C concentrations increased on consumption of the LC diet (P < 0·05). Plasma markers of insulin resistance (P < 0·001), lipaemia and inflammation (P < 0·05, TNF-α and IL-10) improved similarly on both diets. There was no change in other cardiovascular markers with WL. The present data suggest that a LC WL diet does not impair plasma indices of cardiometabolic health, at least within 4 weeks, in otherwise healthy obese subjects. In general, improvements in metabolic health associated with WL were similar between the LC and MC diets. Antioxidant supplements may be warranted if LC WL diets are consumed for a prolonged period.

Previous lines of experimental evidence have suggested that Phaseolus vulgaris extracts reduce food intake, body weight, lipid accumulation, hedonic properties of food, carbohydrate absorption and metabolism, and glycaemia in rats. The present study was designed to assess the effect of multiple cycles of repeated treatments with a standardised P. vulgaris dry extract on daily food intake and body weight in genetically obese Zucker fa/fa rats (Expt 1). Additionally, the study tested the effect of acute treatment with P. vulgaris dry extract on postprandial glycaemia in Zucker fa/fa rats (Expt 2). In Expt 1, P. vulgaris dry extract was administered daily, at doses of 50 and 500 mg/kg, in three 5 d treatment periods followed by three 20 d off-treatment periods. Administration of P. vulgaris dry extract resulted in dose-dependent decreases in daily food intake and body weight in each treatment phase. Reductions in food intake were of comparable magnitude in each treatment phase. In Expt 2, food-deprived rats were acutely treated with 50 and 500 mg P. vulgaris dry extract per kg immediately before access to a fixed amount of a starch-enriched chow. Treatment with P. vulgaris dry extract resulted in a dose-dependent suppression of glycaemia. These results extend previous data on the anorectic and hypoglycaemic effects of the P. vulgaris dry extract to a validated animal model of obesity. Together with data published previously in the literature, these results strengthen the hypothesis that potentially effective, novel pharmacotherapies for obesity and related disorders may originate from extracts and derivatives of P. vulgaris.

Diet is important for the prevention of CVD, and diets high in MUFA might be more cardioprotective than low-fat diets. We hypothesise that inflammation and endothelial cell function will be improved most favourably by a high-MUFA diet compared with a low-fat diet. This was tested in a parallel randomised intervention trial on overweight individuals (aged 28·2 (sd 4·6) years) assigned to a diet moderate in the amount of fat (35–45% of energy; >20% of fat as MUFA; MUFA diet, n 39), a low-fat (20–30% of energy) diet (LF diet, n 43) or a control diet (35 % of energy as fat, n 24) for 6 months after weight loss. Protein constituted 10–20 % of energy in all diets. Food was provided free of charge. Fasting blood samples were collected before and after the intervention and analysed for C-reactive protein (CRP), IL-6, intercellular adhesion molecule, von Willebrand factor (vWF) and tissue factor pathway inhibitor. vWF concentrations tended to fall on the LF diet (4·78 (sd 16·44) %; P = 0·07). Concentrations of IL-6 were reduced by the MUFA (0·37 (sd 0·74) pg/ml; P < 0·01) and LF (0·47 (sd 0·69) pg/ml; P < 0·001) diets, and CRP was reduced on all diets (MUFA: 0·48 (sd 1·93) mg/l (P < 0·01); LF: 1·46 (sd 2·89) mg/l (P < 0·001); control: 1·20 (sd 1·97) mg/l (P < 0·01)). No significant differences were observed between changes induced by the different diets. Our findings suggest that in overweight subjects after weight loss, the MUFA and LF diets have similar long-term effects on inflammation and endothelial cell function.

Epidemiological studies show that adherence to a Mediterranean diet (MD) increases longevity; however, few studies are restricted to Mediterranean populations or explore the effect of a MD pattern that directly incorporates olive oil. Therefore the relationship between adherence to the MD and mortality was studied within the the Spanish cohort of the European Prospective Investigation into Cancer and Nutrition (EPIC-Spain). The EPIC-Spain analysis included 40 622 participants (37·7 % males) aged 29–69 years who were recruited from five Spanish regions in 1992–1996. During a mean follow-up of 13·4 years, 1855 deaths were documented: 913 from cancer, 399 from CVD, 425 from other causes and 118 from unknown causes of death. Risk of all-cause and cause-specific mortality was assessed according to the level of adherence to a relative MD (rMED) score, measured using an 18-unit scale incorporating nine selected dietary components. A high compared with a low rMED score was associated with a significant reduction in mortality from all causes (hazard ratio (HR) 0·79; 95 % CI 0·69, 0·91), from CVD (HR 0·66; 95 % CI 0·49, 0·89), but not from overall cancer (HR 0·92; 95 % CI 0·75, 1·12). A 2-unit increase in rMED score was associated with a 6 % (P < 0·001) decreased risk of all-cause mortality. A high olive oil intake and moderate alcohol consumption contributed most to this association. In this Spanish cohort, following an olive oil-rich MD was related to a significant reduction in all-cause mortality, and reduced the risk of mortality from CVD. These results support the important role that the MD pattern has on reducing mortality in Mediterranean countries.

Weight rebound after successful weight loss is a well-known phenomenon in humans and dogs, possibly due to the fact that energy restriction improves metabolic efficiency, reducing post-weight-loss maintenance energy requirements (MER). The aim of the present study was to estimate post-weight-loss MER in obese pet dogs that had successfully lost weight and did not subsequently rebound. A total of twenty-four obese dogs, successfully completing a weight management programme at the Royal Canin Weight Management Clinic, University of Liverpool (Wirral, UK), were included. In all dogs, a period of >14 d of stable weight ( < 1 % change) was identified post-weight loss, when food intake was constant and activity levels were stable (assessed via owners' diary records). Post-weight-loss MER was indirectly estimated by determining dietary energy consumption during this stable weight period. Multivariable linear regression was used to identify factors that were associated with post-weight-loss MER. The mean length of stable weight after weight loss was 54 (sd 34·1) d. During this time, MER was 285 (sd 54·8) kJ/kg0·75 per d. The rate of prior weight loss and food intake during the weight-loss phase was positively associated with post-weight-loss MER, while the amount of lean tissue lost was negatively associated with post-weight-loss MER. MER are low after weight loss in obese pet dogs (typically only 10 % more than required during weight-loss MER), which has implications for what should constitute the optimal diet during this period. Preserving lean tissue during weight loss may maximise post-weight-loss MER and help prevent rebound.

Ca status in the uterus during pregnancy has been suggested to affect fetal growth and size at birth. In Bangladesh, low Ca levels in pregnant women and low birth weight in infants are common. The present study explored the association between Ca levels in cord blood and newborn size at birth (birth weight and birth length) in Bangladesh. Samples and data included 223 women with live-born singleton deliveries in rural Bangladesh. Newborn weight and length were measured at birth. From cord blood obtained at delivery, Ca, 25-hydroxy vitamin D, bone-specific alkaline phosphatase and intact parathyroid hormone levels were determined. An association between size at birth and Ca levels in cord blood was found (birth weight, P = 0·022; birth length, P = 0·001). Associations between Ca and newborn size were further analysed using multivariate regression analyses. After adjusting for several covariates of characteristics in mothers and newborns (gestational weeks at birth, sex of newborn, socio-economic status, maternal height, BMI, age and season at birth), birth length still exhibited a significant relationship with Ca levels in cord blood (birth length, P = 0·030). The present study indicates that Ca status in cord blood might be associated with the birth length of newborns. Ca levels during gestation may affect fetal growth.

The present study investigates the effect of strawberry antioxidants in beverage form on meal-induced postprandial inflammatory and insulin responses in human subjects. Overweight adults (n 24) consumed a high-carbohydrate, moderate-fat meal (HCFM) accompanied by either a strawberry or a placebo beverage in a cross-over design. Postprandial changes in plasma anthocyanins, their metabolites, insulin, glucose and inflammatory markers were assessed for 6 h. The postprandial concentrations of pelargonidin sulfate and pelargonidin-3-O-glucoside were significantly increased when the strawberry beverage was consumed concurrently with the HCFM compared with the placebo beverage (P < 0·001). The strawberry beverage significantly attenuated the postprandial inflammatory response as measured by high-sensitivity C-reactive protein and IL-6 (P < 0·05) induced by the HCFM. It was also associated with a reduction in postprandial insulin response (P < 0·05). Collectively, these data provide evidence for favourable effects of strawberry antioxidants on postprandial inflammation and insulin sensitivity.

In healthy older subjects, the glycaemic response to carbohydrate-containing meals is dependent on gastric emptying and intestinal absorption; when the latter is slowed, the magnitude of the rise in glucose is attenuated. The oligosaccharide α-cyclodextrin has been reported to diminish the glycaemic response to starch in young adults; this effect has been attributed to the inhibition of pancreatic amylase. We examined the effects of α-cyclodextrin on gastric emptying of, and the glycaemic and insulinaemic responses to, oral sucrose in healthy older subjects; as sucrose is hydrolysed by intestinal disaccharides, any effect(s) of α-cyclodextrin would not be attributable to amylase inhibition. A total of ten subjects (seven males and three females, age 68–76 years) were studied on 2 d. Gastric emptying, blood glucose and serum insulin were measured after ingestion of a 300 ml drink containing 100 g sucrose, labelled with 99mTc-sulphur colloid, with or without 10 g α-cyclodextrin. Gastric emptying was slowed slightly by α-cyclodextrin; this effect was evident between 135 and 195 min and was associated with a slight increase (P < 0·05) in distal stomach retention. After α-cyclodextrin, blood glucose was slightly less (P < 0·05) at 60 min, and serum insulin was less (P < 0·0005) at 90 and 120 min. There was no difference in the incremental areas under the curve (iAUC) for blood glucose, but there was a trend for the iAUC for serum insulin to be lower (P = 0·09) after α-cyclodextrin. We conclude that in a dose of 10 g, α-cyclodextrin has modest effects to slow gastric emptying of, and modify the glycaemic and insulinaemic responses to, oral sucrose, probably due to delayed intestinal carbohydrate absorption.

The objective of the present study was to assess the effect of adding specific prebiotics to standard formula feeding on the number of fever episodes in the first year of life. In the present randomised, double-blind, placebo-controlled trial in seven centres in five West European countries, 830 healthy term infants, without a first-degree family history of allergic disease, of mothers who indicated to give only formula feeding were randomised either to receive a standard non-hydrolysed cows' milk-based formula to which a mixture of specific oligosaccharides was added (prebiotics group (PG)), or to receive a similar formula without oligosaccharides (control group (CG)). A separate reference group consisted of 300 breast-fed infants. The primary outcome was the number of fever episodes prospectively documented by the parents. There was no difference in the number of fever episodes between the PG (median value 1·19; 25th–75th percentile 0·09–2·34) and CG (1·16; 25th–75th percentile 0·06–2·38). The median number of fever episodes in the separate breast-feeding reference group was 1·24 (25th–75th percentile 0·51–3·45). There was no effect of adding specific prebiotics to standard formula feeding in reducing the number of fever episodes in the present study.

Glucosinolates, phytochemicals found in cruciferous vegetables, are metabolised to bioactive isothiocyanates (ITC) by certain bacteria in the human gut. Substantial individual variation in urinary ITC excretion has been observed in previous cruciferous vegetable-feeding studies. We hypothesised that individual differences in gut microbial community contribute to the observed variation in glucosinolate metabolism, i.e. gut microbiota composition between high- and low-ITC excreters differs. We recruited twenty-three healthy individuals and fed them a standardised meal containing 200 g of cooked broccoli. After the meal, 24 h urinary ITC excretion was measured. Study participants with the highest (n 5) and lowest (n 5) ITC excretion provided faecal samples for ex vivo bacterial cultivation with 50 μm-glucoraphanin, the major glucosinolate found in broccoli. When grown ex vivo, faecal bacteria from the selected high-ITC excreters were able to degrade more glucoraphanin than those from the low-ITC excreters (P = 0·05). However, bacterial fingerprints of faecal and ex vivo culture microbiota revealed no statistically significant differences between the high- and low-ITC excreters in terminal restriction fragment length polymorphism analysis of the bacterial 16S ribosomal RNA gene. In conclusion, glucosinolate degradation by faecal bacteria ex vivo may be associated with in vivo bacterial glucosinolate metabolism capacity, but no direct link to specific bacterial species could be established, possibly due to the complexity and functional redundancy of the gut microbiota.

It is well accepted that n-3 long-chain PUFA intake is positively associated with a range of health benefits. However, while benefits have been clearly shown, especially for CVD, the mechanisms for prevention/benefit are less understood. Analysis of plasma and erythrocyte phospholipids (PL) have been used to measure the status of the highly unsaturated fatty acids (HUFA), especially EPA (20 : 5n-3) and DHA (22 : 6n-3), although the time and complexity of the process places limitations on the sample numbers analysed. An assay has been developed using whole blood, collected by finger prick, and stored on absorbant paper, subjected to direct methylation and fatty acids quantified by automated GC. Tests on fatty acid stability show that blood samples are stable when stored at − 20°C for 1 month although some loss of HUFA was seen at 4°C. A total of fifty-one patients, including twenty-seven who consumed no fatty acid supplements, provided a blood sample for analysis. Concentrations of all major fatty acids were measured in erythrocyte PL and whole blood. The major HUFA, including EPA, DHA and arachidonic acid (ARA; 20 : 4n-6), as well as the ARA:EPA ratio and the percentage n-3 HUFA/total HUFA all showed good correlations, between erythrocyte PL and whole blood. Values of r2 ranged from 0·48 for ARA to 0·95 for the percentage of n-3 HUFA/total HUFA. This assay provides a non-invasive, rapid and reliable method of HUFA quantification with the percentage of n-3 HUFA value providing a potential blood biomarker for large-scale nutritional trials.

A community-based cross-sectional study was carried out among Aboriginal schoolchildren aged 7–12 years living in remote areas in Pos Betau, Pahang, Malaysia to investigate the potential determinants influencing the cognitive function and educational achievement of these children. Cognitive function was measured by intelligence quotient (IQ), while examination scores of selected school subjects were used in assessing educational achievement. Blood samples were collected to assess serum Fe status. All children were screened for soil-transmitted helminthes. Demographic and socio-economic data were collected using pre-tested questionnaires. Almost two-thirds (67·6 %) of the subjects had poor IQ and most of them (72·6 %) had insufficient educational achievement. Output of the stepwise multiple regression model showed that poor IQ was significantly associated with low household income which contributed the most to the regression variance (r2 0·059; P = 0·020). Low maternal education was also identified as a significant predictor of low IQ scores (r2 0·042; P = 0·043). With educational achievement, Fe-deficiency anaemia (IDA) was the only variable to show significant association (r2 0·025; P = 0·015). In conclusion, the cognitive function and educational achievement of Aboriginal schoolchildren are poor and influenced by household income, maternal education and IDA. Thus, effective and integrated measures to improve the nutritional and socio-economic status of rural children would have a pronounced positive effect on their education.

There are no published data regarding the overall dietary glycaemic index (GI) and glycaemic load (GL) of Australian children and adolescents. We therefore aim to describe the dietary GI and GL of participants of the 2007 Australian National Children's Nutrition and Physical Activity Survey (2007ANCNPAS), and to identify the main foods contributing to their GL. Children, aged 2–16 years, who provided two 24 h recalls in the 2007ANCNPAS were included. A final dataset of 4184 participants was analysed. GI of each food item was assigned using a previously published method. GL was calculated, and food groups contributing to the GL were described by age group and sex. The weighted mean dietary GI and GL of the participants were 54 (sd 5) and 136 (sd 44), respectively. Among the nutrients examined, Ca had the highest inverse relationship with GI (P < 0·001), while percentage energy from starch was most positively associated with GI. The association between fibre density and GI was modest, and percentage energy from sugar had an inverse relationship with GI. Daily dietary GL contributed by energy-dense and/or nutrient-poor (EDNP) items in subjects aged 14–16 years was more than doubled that of subjects aged 2–3 years. To conclude, Australian children and adolescents were having a high-GI dietary pattern characterised by high-starchy food intake and low Ca intake. A significant proportion of their dietary GL was from EDNP foods. Efforts to reduce dietary GI and GL in children and adolescents should focus on energy-dense starchy foods.

The duration of the period of time during which diet should be recorded for sufficiently accurate results on the usual intake of an individual is an especially challenging issue in prospective studies among children. We set out to describe nutrient intake variability in preschoolers and to determine the number of record days required (D) to estimate intake of energy and thirty-two nutrients. The diet and the use of dietary supplements were assessed with three consecutive daily food records including one weekend day in 1639 children participating in the population-based birth cohort of the Type 1 Diabetes Prediction and Prevention Project (DIPP) in Finland. Variance ratios and D stratified by sex and age groups were calculated for 455 (1-year-old), 471 (3-year-old) and 713 (6-year-old) children (born between 1998 and 2003). Within:between variance ratios and D increase with increasing age, and are slightly higher for girls. Vitamin A, cholesterol, n-3 and n-6 fatty acids, β-carotene and folate intakes require the most replicates. Including supplemental intake has an impact on the variance estimates according to the proportion of supplement users. In the DIPP Nutrition Study with 3 d food records, the correlation coefficients between observed and true intakes of energy and thirty-two nutrients averaged 0·91 in 1-year-old children, 0·79 in 3-year-old children and 0·74 in 6-year-old children. For providing accurate nutrient intake estimates, three replicates of food records are reasonable in 1-year-old children but must be questioned for several nutrients in 3- and 6-year-old children. The accuracy of ranking boys is greater than that for girls.