Sources of viscous soluble fibre, such as barley and oats, have often been included in the weaning diet of the pig to accelerate development of the large intestine. Inclusion of a non-fermentable, viscous compound, sodium carboxymethylcellulose (CMC), in a low-fibre weaning diet was tested to assess the influence of digesta viscosity on the gut in the absence of increased fermentation. Two CMC sources, of low and high viscosity, were added to cooked rice-based diet at 40 g/kg total diet. A third control rice diet did not contain any CMC. Diets were fed for 13 d following weaning at 3 weeks of age. Addition of CMC to the diet significantly increased the intestinal viscosity of digesta within the small (P<0·001) and large (P<0·05) intestine. No simple association was found between increases in intestinal viscosity and effects on intestinal morphology and whole-body growth. The average empty-body-weight gain and the small intestinal villus height increased with low-viscosity CMC, but decreased with the high-viscosity CMC group. The full large intestinal weight increased in all pigs fed CMC. Dietary CMC (both low- and high-viscosity) increased the percentage moisture of digesta and faeces, and was associated with increased faecal shedding of enterotoxigenic haemolytic Escherichia coli. Feed ingredients in weaning diets that excessively increase the viscosity of the intestinal digesta may be detrimental to pig health and production.

To assess the acclimation of pigs to heat stress, the effects of high (33°C) or thermoneutral (23°C) constant temperatures on feeding behaviour and components of energy balance were studied in group-housed young pigs. Three groups of five pigs were used at each temperature. After 1 week of adaptation, voluntary feed intake (VFI) and heat production (HP) were recorded for thirteen consecutive days. Animals were fed ad libitum. Fasting HP was measured on the last day. Average initial body weights (BW) were 21·4 and 20·9 kg at 23 and 33°C respectively. Feeding behaviour was measured individually and rate of feed intake and characteristics of feeding behaviour were calculated. The O2 consumption, CO2 production and physical activity of the group were used to calculate total HP (HPtot) and its components, i.e. fasting HP (HPfas), HP due to physical activity (HPact) and thermic effect of feed (TEF). The BW gain and VFI were reduced by 37 and 30 % respectively at 33°C. The decrease in VFI corresponded to reduced consumption time (-34 %) and size of the meals (-32 %). Feeding behaviour was mostly diurnal (66 % of the VFI), and the rate of feed intake (28 g/min) was not affected by temperature. Daily HPtot, HPfas and TEF, expressed per kg metabolic weight (BW0·60), were significantly decreased at 33°C by 22, 18 and 35 % respectively, whereas HPact was not affected; TEF expressed per g feed was not affected (2 kJ/g). The decrease in HPtot at 33°C was caused by a reduction in TEF and HPfas (kJ/d per/kg BW0·60), which are both related to reduction in VFI.

Reductions in red meat and increases in cereals in the diet may compromise the intake and bioavailability of Zn. In this cross-sectional study of 330 premenopausal New Zealand women aged 18–40 years, we have assessed the inter-relationships among dietary intakes (via computer-administered food-frequency questionnaire), biochemical Zn status, and anthropometric indices, and compared our results with earlier data. Fasting serum (12·00 (SD 1·36) ΜMOL/L) AND HAIR ZN (2·71 (sd 0·36) μmol/g) were lower than those for young Dunedin, New Zealand, women in 1973 (non-fasting serum Zn 18·6 (sd 4·6) μmol/l, hair Zn 2·99 (sd 0·35) μmol/g). Further, our mean serum Zn was at the 25th percentile of the US National Health and Nutrition Examination Survey (NHANES) (1976–1980) reference sample for women aged 20–44 years. Meat–poultry–fish contributed only 28 % total Zn in the present study, a level comparable with that from cereals–nuts–legumes (27 %), compared to about 40 % in 1989. Significant negative correlations existed between serum Zn and dietary [phytate]:[Zn] molar ratios (r -0·163, P<0·01); 35 % had diets with [phytate]:[Zn] >15, a level said to compromise Zn status. Mean serum Zn of a subgroup of non-oral contraceptive users free of infection was higher in the red-meat eaters (n 149) compared with non-red-meat eaters (n 48) (12·2 v. 11·8 μmol/g, P<0·05). In contrast, serum Zn was lower in those with dietary [phytate]:[Zn] ratios >15 v. <15 (i.e. 11·9 v. 12·3 μmol/l, P=0·04). We postulate that the lower biochemical Zn status of these New Zealand women may be associated in part with changes in food selection patterns, which have led to a reduction in the bioavailability of dietary Zn.

Due to the potential oestrogenic effects of secoisolariciresinol diglycoside (SDG), the mammalian lignan precursor in flaxseed (Linum usitatissimum), we hypothesized that exposure to purified SDG during early life would have a positive effect on developing bone. This present study determined whether exposure to SDG purified from flaxseed during suckling via mother's milk or continuously to adolescence (postnatal day (PND) 50) or adulthood (PND 132) increased bone mineral content (BMC) or bone strength in female rat offspring. Offspring were exposed to basal diet (BD) or one of two doses of SDG (50S, 100S) equivalent to that in a 50 or 100 g flaxseed/kg diet during lactation only or through to PND 50 or 132. At PND 50 and 132, femurs were analysed for BMC by dual energy X-ray absorptiometry and biomechanical strength by a 3-point bending test. Compared with BD group, rats exposed to continuous 50S or 100S diet had stronger femurs at PND 50 without changes in BMC. At PND 132 there were no differences in femur strength despite the fact that continuous exposure to BD resulted in a higher (P<0·05) BMC than rats exposed to 100S during lactation only or to 50S or 100S during lactation through to adulthood. In conclusion, female rat bone is more sensitive to the oestrogen-like action of lignans during early life when endogenous levels of sex hormones are low, but by adulthood the improved bone strength does not persist. Importantly, exposure to purified lignan does not have negative effects on bone strength.

The fatty acid composition and lipid content of white muscle, liver and mesenteric fat, in reared v. wild sharpsnout sea bream (Diplodus puntazzo) were compared. The mesenteric fat index (100×mesenteric fat weight/body weight) and the lipid contents of both white muscle and liver proved consistently higher in farmed v. wild sharpsnout sea bream (79·0 (SE 13·1) V. 38·7 (se 5·1) g/kg, 188·4 (se 30·0) v. 58·2 (se 3·9) g/kg and 27·2 (se 3·7) v. 17·3 (se 1·9) g/kg, respectively). The higher values of linoleic, eicosapentaenoic, docosahexaenoic and n-3 series acids in reared fish muscle make reared sharpsnout more favourable for human consumption. In reared fish mesenteric fat, polyunsaturated fatty acids reached higher levels (32·54 (se 0·71) g/100 g total fatty acids than those found in wild fish (26·08 (se 1·38) g/100 g total fatty acids or even present in the diet (28·34 g/100 g total fatty acids). Compared with cultured fish, wild sharpsnout displayed a higher content of n-3 fatty acids in liver fat (31·67 (se 1·13) g/100 g total fatty acids), but lower in mesenteric fat (20·35 (se 1·41) g/100 g total fatty acids). Atherogenic index values were similar for wild and reared fish in all tissues, while the index of thrombogenicity of muscle and mesenteric fat (0·353 (se 0·012) and 0·402 (se 0·021) respectively) was significantly increased in wild fish probably due to the omnivorous habits of the species and/or to seasonal food variations. Depending on the time of the year or the season, reared fish could be more suitable for human consumption than wild fish.

Body weight, weight changes and BMI are easily obtainable indicators of nutritional status, but they do not provide information on the amount of fat-free and fat masses. The purpose of the present study was to determine if fat-free mass (FFM) and fat mass were depleted in patients with normal BMI or serum albumin at hospital admission. A group of 995 consecutive patients were evaluated for malnutrition by BMI, serum albumin, and 50 kHz bioelectrical impedance analysis and compared with 995 healthy adults, matched for age and height, and then compared with FFM and fat mass percentiles previously determined in 5225 healthy adults. A BMI of ≤20 kg/m2 was noted in 17·3 % of patients and serum albumin of ≤35 g/l was found in 14·9 % of patients. In contrast, 31 % of all patients were below the tenth percentile for FFM, compared with 10·1 % of controls (χ2, P=0·0001), while 73 % of patients with BMI ≤20 kg/m2 and 31 % of patients with BMI 20–24·9 kg/m2 fell below the tenth percentile for FFM. Furthermore, the FFM was lower in patients than controls and the differences with age in FFM (lower) and fat mass (higher) were greater in patients than in controls. BMI and albumin significantly underestimated the prevalence of malnutrition in patients at hospital admission compared with body composition measurements. Optimal nutritional assessment should therefore include objective measurement of FFM and fat mass.

High intake of fish has been associated with reduced risk of CHD. The high content of n-3 polyunsaturated fatty acids (PUFA) in fish has been suggested to be a protective factor. In addition, fish is the entirely dominating source of methylmercury for the general population, and the concentration of Hg in erythrocytes (Ery-Hg) is often used as an index of fish consumption. Our aim was to study the relationships between a first-ever myocardial infarction, Ery-Hg, activity of gluthathione peroxidase in erythrocytes (Ery-GSH-Px) and plasma concentration of the n-3 PUFA eicosapentaenoic and docosahexaenoic acids (P-PUFA). In a population-based prospective nested case–control study within Northern Sweden seventy-eight cases of a first-ever myocardial infarction were compared with 156 controls with respect to Ery-Hg, P-PUFA and Ery-GSH-Px. Both Ery-Hg and P-PUFA, but not Ery-GSH-Px, were significantly (P<0·0001) higher in subjects reporting high fish intake (at least one meal per week) than in those with lower intake. This finding suggests that Ery-Hg and P-PUFA reflect previous long-term fish intake. Low risk of myocardial infarction was associated with high Ery-Hg or high P-PUFA. In a multivariate model the risk of myocardial infarction was further reduced in subjects with both high Ery-Hg and high P-PUFA (odds ratio 0·16, 95 % CI 0·04, 0·65). In conclusion, there is a strong inverse association between the risk of a first myocardial infarction and the biomarkers of fish intake, Ery-Hg and P-PUFA, and this association is independent of traditional risk factors.

Low-fat diets, in which carbohydrates replace some of the fat, decrease serum cholesterol. This decrease is due to decreases in LDL-cholesterol but in part to possibly harmful decreases in HDL-cholesterol. High-oil diets, in which oils rich in monounsaturated fat replace some of the saturated fat, decrease serum cholesterol mainly through LDL-cholesterol. We used these two diets to investigate whether a change in HDL-cholesterol would change flow-mediated vasodilation, a marker of endothelial function. We fed thirty-two healthy volunteers two controlled diets in a 2×3·5 weeks' randomised cross-over design to eliminate variation in changes due to differences between subjects. The low-fat diet contained 59·7 % energy (en%) as carbohydrates and 25·7 en% as fat (7·8 en% as monounsaturates); the oil-rich diet contained 37·8 en% as carbohydrates and 44·4 en% as fat (19·3 en% as monounsaturates). Average (SD) SERUM HDL-CHOLESTEROL AFTER THE LOW-FAT DIET WAS 0·21 (sd 0·12) mmol/l (8·1 mg/dl) lower than after the oil-rich diet. Serum triacylglycerols were 0·22 (sd 0·28) mmol/l (19·5 mg/dl) higher after the low-fat diet than after the oil-rich diet. Serum LDL and homocysteine concentrations remained stable. Flow-mediated vasodilation was 4·8 (SD 2·9) after the low-fat diet and 4·1 (SD 2·7) after the oil-rich diet (difference 0·7 %; 95 % CI -0·6, 1·9). Thus, although the low-fat diet produced a lower HDL-cholesterol than the high-oil diet, flow-mediated vasodilation, an early marker of cardiovascular disease, was not impaired.

A study was undertaken to examine the effect of different dietary carnitine (200 and 1000 mg/kg diet) and fat (90 and 190 g/kg diet) supplementation on growth and fatty acid concentrations of fish fed either with a low- (13 g/kg) or a high-lysine (21 g/kg) diet. African catfish (22·7 g/fish), Clarias gariepinus Burchell, juveniles were stocked (sixteen aquaria, twenty-five fish per aquarium) and fed for a maximum of 74 d. Dietary lysine had a clear effect on growth performance and feed conversion ratios, but dietary carnitine supplements had no effect. High-carnitine supplements increased total carnitine content (P<0·0004) and reduced tissue free carnitine: acyl-carnitine ratio (P<0·05) compared with low-carnitine supplements. High-fat supplements decreased liver carnitine concentrations. Clear effects on liver fatty acid concentrations were observed in high-carnitine-fed fish compared with low-carnitine-fed fish. The primary liver fatty acids affected were 18:2n-6 (linoleic acid), 20:5n-3 (eicosapentanoic acid) and 22:6n-3 (docosahexanoic acid). The whole-body fatty acid balance suggested that 20:5n-3 disappeared (apparently by β-oxidation) more readily than 18:2n-6 and/or 22:6n-3. From 774 mg 20:5n-3 eaten by high-lysine–high-fat–low-carnitine fish, 58 % was not assimilated into body tissues. High-carnitine-fed fish showed an increase in 20:5n-3 oxidation by 7 % compared with low-carnitine fish. Although dietary carnitine did not improve body growth, these results support the hypothesis that carnitine can enhance the mobilisation of long-chain fatty acids towards oxidation.

The objective of the present study was to assess the relationship between anthropometric measurements and risk factors for non-communicable diseases (NCD) in South African black women. A cross-sectional sample of 1040 apparently healthy black female volunteers, 15–70 years old, was recruited from thirty-seven randomly selected sites in the North West Province, stratified according to level of urbanisation. We analysed the association between BMI, waist:hip (WHR), waist circumference (WC) and skinfold measurements and the following risk factors for NCD: blood pressure, serum lipids, fasting serum glucose and insulin and plasma fibrinogen, by using age-adjusted correlation analyses and stepwise regression analysis. Of the subjects, 28·6 % were obese (BMI>30). After adjustment for age and smoking status, BMI correlated significantly with diastolic blood pressure (r 0·21, P=0·037), serum triacylglycerols (TG) (r 0·30, P=0·003), fasting glucose (r 0·29, P=0·005) and log fasting insulin (r 0·24, P=0·02). There was a significant negative correlation between BMI and HDL-cholesterol (r -0·38, P<0·001). Similar but stronger correlations were found between both WC and WHR and these risk factors. Together with age, WC was a significant predictor of TG, HDL-cholesterol and fasting glucose in regression analysis, while subscapular skinfold was a significant predictor of diastolic blood pressure and fasting glucose concentration. Triceps skinfold was a significant predictor of total serum cholesterol, LDL-cholesterol, plasma fibrinogen and the insulin sensitivity index. Measures of obesity, particularly WC, are associated with the risk for NCD in black South African women, in which a high rate of obesity has been found.

The effects of red wine and ethanol on plasma lipoproteins and the fatty acid composition of kidney lipids and erythrocytes phospholipids were studied. Lipid peroxidation is one of the main deleterious effects of oxidant attack on biomolecules, due to the disruption of the structural integrity of membranes. The vulnerability of the kidney to oxidative damage has been partly attributed to its high content of long-chain polyunsaturated fatty acids. Antioxidants, such as flavonoids, would be a means of reducing the risk of oxidative damage to membranes. Nutritional sources rich in antioxidants, including those provided by wine, are expected to attenuate the effects of oxidative challenges. Adult rats were fed red wine rich in flavonols, ethanol (125 ml/l), or alcohol-free red wine. The control group drank water. After 10 weeks, blood samples served to measure plasma lipoproteins and antioxidant capacity. Kidney lipids and erythrocyte phospholipids were extracted. The samples were assayed by GLC. Energy intake did not differ between all the groups, but the weight gain of the ethanol group was less than the other three groups. Blood HDL and triacylglycerols were increased by both ethanol and red wine. Ethanol decreased arachidonic and docosahexaenoic acids in both kidney lipids and erythrocyte phospholipids, as compared with either water, red wine or alcohol-free red wine groups. These results indicate that non-alcoholic components of red wine could contribute to avoiding the unfavourable effects of ethanol on plasma lipoproteins, kidney lipids and membrane erythrocyte phospholipids.

The aim of the present cross-sectional study was to examine the agreement and disagreement between a 7 d diet diary (7DD) and a self-administered machine-readable food-frequency questionnaire (FFQ) asking about diet in the previous year, and to validate both methods with biomarkers of nutrient intake. The subjects were an age- and employment-grade-stratified random subsample of London-based civil servants (457 men and 403 women), aged 39–61 years, who completed both a 7DD and a FFQ at phase 3 follow-up (1991–1993) of the Whitehall II study. Mean daily intakes of dietary energy, total fat, saturated, monounsaturated and polyunsaturated fatty acids, linoleic acid, total carbohydrate excluding fibre, sugars, starch, dietary fibre, protein, vitamin C, vitamin E (as α-tocopherol equivalents), folate, carotenes (as total β-carotene activity), Fe, Ca, Mg, K and alcohol were measured. Serum cholesteryl ester fatty acids (CEFA), plasma α-tocopherol and β-carotene were also measured as biomarkers. Estimates of mean energy intake from the two methods were similar in men, and some 10 % higher according to the FFQ in women. Compared with the 7DD, the FFQ tended to overestimate plant-derived micronutrient intakes (carotenes from FFQ v. 7DD men 2713 (SD 1455) V. 2180 (sd 1188) μg/d, women 3100 (sd 1656) v. 2221 (sd 1180) μg/d, both differences P<0·0001) and to underestimate fat intake. Against plasma β-carotene/cholesterol, carotene intake was as well estimated by the FFQ as the 7DD (Spearman rank correlations, men 0·32 v. 0·30, women 0·27 v. 0·22, all P≤0·0001, energy-adjusted data). Ranking of participants by other nutrient intakes tended to be of the same order according to the two dietary methods, e.g. rank correlations for CEFA linoleic acid against FFQ and 7DD estimates respectively, men 0·38 v. 0·41, women 0·53 v. 0·62, all P≤0·0001, energy-adjusted % fat). For α-tocopherol there were no correlations between plasma level and estimated intakes by either dietary method. Quartile agreement for energy-adjusted nutrient intakes between the two self-report methods was in the range 37–50 % for men and 32–44 % for women, and for alcohol, 57 % in both sexes. Disagreement (misclassification into extreme quartiles of intake) was in the range 0–6 % for both sexes. The dietary methods yielded similar prevalences (about 34 %) of low energy reporters. The two methods show satisfactory agreement, together with an expected level of systematic differences, in their estimates of nutrient intake. Against the available biomarkers, the machine-readable FFQ performed well in comparison with the manually coded 7DD in this study population. For both methods, regression-based adjustment of nutrient intake to mean dietary energy intake by gender appears on balance to be the optimal approach to data presentation and analysis, in view of the complex problem of low energy reporting.

The purpose of the present study was to evaluate whether micronutrient supplementation improved the nutritional status of women with poor diets during the inter-pregnancy interval. Fifty-five women who had given birth to a low birth weight baby (<2·5 kg), and who planned to have a further pregnancy, were recruited to a prospective randomised study in East London, UK. Of the fifty-five mothers recruited, forty-four (78 %) met fewer than four of sixteen dietary reference values according to the information provided in a 7 d diet diary, and were categorised as having an ‘inadequate’ diet. Half of the mothers in the ‘inadequate’-diet group were randomly assigned to receive a micronutrient and a single cell oil supplement containing docosahexaenoic acid. All participants received dietary advice based on analysis of their diet diaries, and general lifestyle advice on preparing for pregnancy. Mothers had a blood sample taken at 3 and 9 months post-partum to measure their folate, Fe stores and fatty acid status. Mean serum and erythrocyte folate levels increased significantly between 3 and 9 months post-partum in both the adequate-diet group and the supplemented group. At 9 months post-partum, over half of the unsupplemented, inadequate-diet group remained severely deficient in folate (serum folate <230 nmol/l) and had low serum ferritin levels (<15 μg/l). The high prevalence of inadequate diets in this inner-city population and the low motivation of women to participate in a nutrition programme suggests that consideration should be given to the provision of free folate and Fe supplements to all women in this and similar populations, or at least to women who have delivered a low birth weight baby, who plan further pregnancies.

Sources of viscous soluble fibre, such as barley and oats, have often been included in the weaning diet of the pig to accelerate development of the large intestine. Inclusion of a non-fermentable, viscous compound, sodium carboxymethylcellulose (CMC), in a low-fibre weaning diet was tested to assess the influence of digesta viscosity on the gut in the absence of increased fermentation. Two CMC sources, of low and high viscosity, were added to cooked rice-based diet at 40 g/kg total diet. A third control rice diet did not contain any CMC. Diets were fed for 13 d following weaning at 3 weeks of age. Addition of CMC to the diet significantly increased the intestinal viscosity of digesta within the small (P<0·001) and large (P<0·05) intestine. No simple association was found between increases in intestinal viscosity and effects on intestinal morphology and whole-body growth. The average empty-body-weight gain and the small intestinal villus height increased with low-viscosity CMC, but decreased with the high-viscosity CMC group. The full large intestinal weight increased in all pigs fed CMC. Dietary CMC (both low- and high-viscosity) increased the percentage moisture of digesta and faeces, and was associated with increased faecal shedding of enterotoxigenic haemolytic Escherichia coli. Feed ingredients in weaning diets that excessively increase the viscosity of the intestinal digesta may be detrimental to pig health and production.

To test the hypothesis that inclusion of whole cereals in the diet of broiler chickens reduces the severity of a coccidial infection, the effects were investigated, in birds infected with Eimeria acervulina, of feeding a complete pelleted control food, the control food diluted with whole wheat (400 g/kg), or a diet in which ground wheat in the pellets was substituted with whole wheat (400 g/kg) so as to achieve the same composition as the control diet. In the weeks prior to and after infection, (days 14–20 and days 21–27), no significant differences in performance were observed between birds fed the complete pellet and substituted feeds. Birds fed the diluted feeds had significantly lower gains per unit of feed and numerically lower weight gains than the birds fed the control and substituted feeds during this period. No significant differences in performance were observed between treatments from days 28–34. Diluting feeds with whole wheat (400 g/kg) significantly increased ileal digesta viscosity levels, compared with feeding the control and substituted feeds in birds dissected on day 21, although no significant differences in digesta viscosity levels were observed in birds dissected on day 35. Whole wheat feeding, either by dilution or substitution, significantly increased gizzard sizes in birds dissected on day 21 and day 35. Neither dilution nor substitution of feeds with whole wheat (400 g/kg) significantly affected the level of an Eimeria acervulina infection, as measured by daily and total faecal oocyst yields.

There is concern that replacement of dietary fat with carbohydrate may not reduce the overall risk of CHD because this replacement strategy elevates postprandial plasma triacylglycerol (TAG) concentrations. The present study was designed to test the hypothesis that daily exercise can offset the augmented postprandial lipaemia associated with a short-term high-carbohydrate diet. Nine healthy, normolipidaemic men aged 33 (SD 4) YEARS CONSUMED A TEST MEAL (G/KG BODY MASS; 1·2 FAT, 1·1 CARBOHYDRATE, 0·2 PROTEIN) ON THREE OCCASIONS: AFTER 3 D ON A TYPICAL WESTERN DIET (46, 38 AND 16 % ENERGY FROM CARBOHYDRATE, FAT AND PROTEIN RESPECTIVELY); AFTER 3 D ON AN ISOENERGETIC HIGH-CARBOHYDRATE DIET (CORRESPONDING VALUES: 70, 15 AND 15 % ENERGY); AFTER 3 D ON THE SAME HIGH-CARBOHYDRATE DIET WITH 30 MIN MODERATE EXERCISE DAILY. FASTING PLASMA TAG CONCENTRATION WAS HIGHER AFTER THE HIGH-CARBOHYDRATE DIET (1·15 (se 0·16) mmol/l) than after the Western diet (0·83 (se 0·10) mmol/l; P=0·03). Similarly, postprandial lipaemia (6 h total area under plasma TAG concentration v. time curve) was higher after the high-carbohydrate diet (12·54 (se 2·07) mmol/l·h) than after the Western diet (9·30 (se 1·30) mmol/l·h; P=0·004). The addition of exercise to the high-carbohydrate diet significantly reduced postprandial lipaemia (9·95 (se 1·94) mmol/l·h; P=0·01 when compared with the high-carbohydrate diet) but not fasting TAG concentration (1·02 (se 0·24) mmol/l). In conclusion, daily exercise prevented the augmentation of postprandial lipaemia attributable to the short-term high-carbohydrate diet and, thus, exercise may be a powerful adjunct to dietary change.

Soya products contain high levels of the isoflavones genistein and daidzein, and their glucosides, and may lower the risk of cardiovascular disease, osteoporosis and cancer. The present cross-sectional study investigated plasma concentrations of daidzein and genistein and their correlations with dietary soya consumption in four groups of twenty premenopausal British women. The women were selected from the Oxford arm of the European Prospective Investigation into Cancer and Nutrition using data from food-frequency questionnaires (FFQ) to guarantee a wide variation in soya consumption, and to investigate the utility of the question related to soya milk consumption compared with the utility of the question related to other soya foods. Dietary intakes of isoflavones were additionally assessed by 7 d food diaries. Plasma concentrations of daidzein and genistein were measured by time-resolved fluoroimmunoassay. Geometric mean plasma concentrations (nmol/l) were for the four groups, which were based on increasing soya intake, 4·9, 8·4, 39·2 and 132 for daidzein and 14·3, 16·5, 119 and 378 for genistein. The Spearman correlation coefficients for plasma isoflavone concentrations with estimated dietary intakes were between 0·66 and 0·80 for the diary-based estimates and between 0·24 and 0·74 for the FFQ-based estimates. The correlations for soya milk intakes were clearly higher than the correlations for intakes of other soya foods. We conclude that both the food diary and the FFQ estimate dietary soya isoflavone intakes sufficiently well to use them in epidemiological studies, and that plasma concentrations of daidzein and genistein in Western women who consumed soya products as a part of their regular diet were close to those in Asian populations.

The efficacy of insulin in stimulating whole-body glucose disposal (insulin sensitivity) was quantified using direct methodology in thirty lacto-ovo vegetarians and in thirty meat-eaters. All subjects were adult, lean (BMI <23 kg/m2), healthy and glucose tolerant. Lacto-ovo vegetarians were more insulin sensitive than meat-eaters, with a steady-state plasma glucose (mmol/l) of 4·1 (95 % CI 3·5, 5·0) v. 6·9 (95 % CI 5·2, 7·5; P=0·0028) respectively. In addition, lacto-ovo vegetarians had lower body Fe stores, as indicated by a serum ferritin concentration (μg/l) of 35 (95 % CI 21, 49) compared with 72 (95 % CI 45, 100) for meat-eaters (P=0·0012). To test whether or not Fe status might modulate insulin sensitivity, body Fe was lowered by phlebotomy in six male meat-eaters to levels similar to that seen in vegetarians, with a resultant approximately 40 % enhancement of insulin-mediated glucose disposal (P=0·0008). Our results demonstrate that lacto-ovo vegetarians are more insulin sensitive and have lower Fe stores than meat-eaters. In addition, it seems that reduced insulin sensitivity in meat-eaters is amenable to improvement by reducing body Fe. The latter finding is in agreement with results from animal studies where, no matter how induced, Fe depletion consistently enhanced glucose disposal.