Studies have suggested that meal timing plays a role in nutritional health, but this subject has not been sufficiently studied in pregnant women. We analysed the effect that timing of food intake has on eating patterns, diet quality and weight gain in a prospective cohort study with 100 pregnant women. Data were collected once per trimester: 4th–12th, 20th–26th and 30th–37th weeks. Food intake was evaluated using three 24-h dietary recalls, which were used to assess eating patterns and diet quality. Distribution of energy and macronutrient intake throughout the day was considered eating patterns. Diet quality was assessed using the Brazilian Healthy Eating Index-Revised. Weight gain was evaluated during each trimester. Women were classified as early or late timing of the first and last eating episodes if these values were below or above the median of the population, respectively (first eating episode = 08.38 hours; last eating episode = 20.20 hours). Generalised estimating equation models adjusted for confounders were used to determine the effects of timing of the first and last eating episodes (groups) and gestational trimesters (time) (independent variable) on eating patterns, diet quality and weight gain (dependent variables). Early eaters of the first eating episode have a higher percentage of energy and carbohydrate intake in morning and a lower at evening meals. They also have a better diet quality for fruit components when compared with late eaters of the first eating episode. Our results emphasise the importance of considering meal timing in the nutritional antenatal guidelines to promote maternal–fetal health.

Diet during pregnancy is related to several maternal and infant health outcomes; however, the relationship between maternal dietary glycaemic index (GI) and glycaemic load (GL) and gestational weight gain (GWG) or newborn birth weight is controversial. The purpose of the present study was to investigate the relationship between maternal dietary GI and GL and GWG and birth weight. A cohort of adult pregnant women with usual obstetric risk was followed in Botucatu, SP, Brazil. Two 24-h dietary recalls were collected in each gestational trimester (<14, 24–27, 31–34 weeks), one in person and the other by telephone. GI and GL were determined using the software Nutrition Data System for Research. GWG was obtained from medical records and evaluated as the weekly GWG between the second and third gestational trimesters. Newborn birth weight z-score in relation to gestational age was evaluated according to Intergrowth-21st Project recommendations. A multiple linear regression model, adjusted for potential confounders, showed a one-point increase in the GI resulted in a mean decrease of 12·9 (95 % CI –21·48, –4·24) g in weekly GWG; GL was not associated with this outcome. The birth weight z-score was not associated with GI (P = 0·763) or GL (P = 0·317). In conclusion, in a cohort of pregnant women considered at usual risk for obstetric complications, maternal dietary GI was negatively associated with weekly GWG in the second and third gestational trimesters. No association was observed between GL and GWG, and neither GI nor GL was associated with birth weight z-score.

Maintenance haemodialysis (MHD) is the use of a machine to filter wastes, salts and fluid from blood for at least 3 months to prolong the life of patients with advanced kidney failure. Although low dietary energy intake (DEI) has been observed in MHD patients, few studies have related DEI to the risk of mortality. To explore this relationship, a study included 1039 MHD patients from eight centres was conducted. DEI was assessed by three 24-h diet recalls and was normalised to ideal body weight (IBW). All-cause mortality and CVD mortality were the primary and secondary outcomes, respectively. During a median follow-up of 28 months, a U-shaped relationship was observed between DEI and all-cause or CVD mortality. The risk of all-cause mortality decreased significantly with the increase of DEI in participants with DEI <167·4 kJ/kg IBW per d (hazard ratio (HR) 0·98; 95 % CI 0·96, 1·00) and increased significantly with the increase of DEI in those with DEI ≥167·4 kJ/kg IBW per d (HR 1·12; 95 % CI 1·04, 1·20). Similarly, the risk of CVD mortality decreased with the increase of DEI in participants with DEI <152·7 kJ/kg IBW per d (HR 0·96; 95 % CI 0·93, 0·99) and increased with the increase of DEI in participants with DEI ≥152·7 kJ/kg IBW per d (HR 1·11; 95 % CI 1·04, 1·18). In summary, there was a U-shaped association between DEI and all-cause or CVD mortality, with a turning point at about 167·4 and 152·7 kJ/kg IBW per d, respectively, in MHD patients.

Epidemiological studies suggest that consumption of potatoes is associated with increased risk of cardiometabolic diseases. However, few clinical trials have empirically tested this. The aim of this single-blind, randomised, crossover study was to evaluate the effect of daily potato consumption, compared with refined grains, on risk factors for cardiometabolic diseases. It was hypothesised that no difference in cardiometabolic endpoints would be detected between conditions, but diet quality would improve with potato consumption. Healthy participants on self-selected diets received one potato-based side dish or one refined grain-based side dish daily, for 4 weeks, separated by a minimum 2-week break. Dishes were isoenergetic, carbohydrate-matched and prepared without excess saturated fat or Na. Participants were instructed to consume the side dish with a meal in place of carbohydrates habitually consumed. Lipids/lipoproteins, markers of glycaemic control, blood pressure, weight and pulse wave velocity were measured at baseline and condition endpoints. Diet quality was calculated, based on 24-h recalls, using the Healthy Eating Index (HEI)-2015. Fifty adults (female n 34; age 40 (sd 13) years; BMI 24·5 (sd 3·6) kg/m2) completed the present study. No between-condition differences were detected for fasting plasma glucose (–0·05 mmol/l, 95 % CI –0·14, 0·04; P = 0·15), the primary outcome or any other outcomes. Compared with refined grains, the HEI-2015 score (3·5, 95 % CI 0·6, 6·4; P = 0·01), K (547 mg, 95 % CI 331, 764, P < 0·001) and fibre (2·4 g, 95 % CI 0·6, 4·2, P = 0·01) were higher following the potato condition. Consuming non-fried potatoes resulted in higher diet quality, K and fibre intake, without adversely affecting cardiometabolic risk.

The National Institute for Health and Care Excellence (NICE) has acknowledged the value of waist-to-height ratio (WHtR) as an indicator for ‘early health risk’. We used recent UK data to explore whether classification based on WHtR identifies more adults at cardiometabolic risk than the ‘matrix’ based on BMI and waist circumference, currently used for screening. Data from the Health Survey for England (4112 adults aged 18+ years) were used to identify cardiometabolic risk, indicated by raised glycated Hb, dyslipidaemia and hypertension. HbA1c, total/HDL-cholesterol and systolic blood pressure (BP) were more strongly associated with WHtR than the ‘matrix’. In logistic regression models for HbA1c ≥ 48 mmol/mol, total/HDL-cholesterol > 4 and hypertension (BP > 140/90 mmHg or on medication), WHtR had a higher predictive value than the ‘matrix’. AUC was significantly greater for WHtR than the ‘matrix’ for raised HbA1c and hypertension. Of adults with raised HbA1c, 15 % would be judged as ‘no increased risk’ using the ‘matrix’ in contrast to 3 % using WHtR < 0·5. For hypertension, comparative values were 23 and 9 %, and for total/HDL-cholesterol > 4, 26 and 13 %. Nearly one-third of the ‘no increased risk’ group in the ‘matrix’ had WHtR ≥ 0·5 and hence could be underdiagnosed for cardiometabolic risk. WHtR has the potential to be a better indicator of cardiometabolic risks associated with central obesity than the current NICE ‘matrix’. The cut-off WHtR 0·5 in early screening translates to a simple message, ‘your waist should be less than half your height’, that allows individuals to be aware of their health risks.

The first positive genome-wide association study on gestational length and preterm delivery showed the involvement of an Se metabolism gene. In the present study, we examine the association between maternal intake of Se and Se status with gestational length and preterm delivery in 72 025 women with singleton live births from the population-based, prospective Norwegian Mother, Father and Child Cohort Study (MoBa). A self-reported, semi-quantitative FFQ answered in pregnancy week 22 was used to estimate Se intake during the first half of pregnancy. Associations were analysed with adjusted linear and Cox regressions. Se status was assessed in whole blood collected in gestational week 17 (n 2637). Median dietary Se intake was 53 (interquartile range (IQR) 44–62) µg/d, supplements provided additionally 50 (IQR 30–75) µg/d for supplement users (n 23 409). Maternal dietary Se intake was significantly associated with prolonged gestational length (β per sd = 0·25, 95 % CI, 0·07, 0·43) and decreased risk of preterm delivery (n 3618, hazard ratio per sd = 0·92, 95 % CI, 0·87, 0·98). Neither Se intake from supplements nor maternal blood Se status was associated with gestational length or preterm delivery. Hence, the present study showed that maternal dietary Se intake but not intake of Se-containing supplements, during the first half of pregnancy was significantly associated with decreased risk of preterm delivery. Further investigations, preferably in the form of a large randomised controlled trial, are needed to elucidate the impact of Se on pregnancy duration.

We assessed the effects of increased Ca consumption from fat-free milk in an energy-restricted diet and educational activities in the metabolic control of overweight type 2 diabetes mellitus (T2DM) patients. Fourteen subjects with T2DM (BMI 29·4 (sd 4·5) kg/m2, low habitual Ca consumption (<600 mg/d)) were included in this randomised, crossover clinical trial. Subjects were randomly allocated to one of the two interventions: drink containing 700 mg of Ca (DAIR) or drink containing 0 mg of Ca (CONT) for ninety consecutive days each. Energy-restricted diets (–500 kcal/d; –2092 kJ/d), containing 800 mg of Ca from dietary sources/d, were prescribed for both groups. Questionnaires were applied at baseline and at the end of the study to assess the subjects’ knowledge on the disease and on self-care, biochemical variables and physical activity. Blood pressure, food intake, body composition and anthropometry were assessed at baseline, days 45 and 90. There was a higher reduction of body fat %, waist circumference, hip circumference, neck circumference, waist:hip ratio, sagittal abdominal diameter, diastolic/systolic blood pressure and an increase in fat-free mass % in DAIR than in CONT. Uric acid, fasting glucose, Hb1Ac, parathyroid hormone and alanine aminotransferase concentrations reduced and vitamin D concentration increased after 90 d in DAIR compared with CONT. The consumption of energy-restricted diet containing 1200 mg Ca/d seems to favour metabolic control in subjects with T2DM. The educational activities increased the knowledge on the disease care.

Depression is an important public health problem. The aim of the study is to explore the associations of total protein intake and protein sources with the risk of depressive symptoms. This cross-sectional study used data from the National Health and Nutrition Examination Survey for the years 2007–2014. Dietary protein intake was obtained from two 24-h dietary recall interviews. Depressive symptoms were assessed by a nine-item Patient Health Questionnaire. Logistic regression models and restricted cubic spline models were used to estimate the associations of total protein intake (g/kg per d) and protein sources with the risk of depressive symptoms. A total of 17 845 individuals aged 18 years and older were included in this study. Total protein intake was inversely associated with the risk of depressive symptoms. The full-adjusted OR of depressive symptoms was 0·34 (95 % CI 0·17, 0·68) for quartile (Q) 4 v. Q1 of total protein intake. For protein intake from milk and milk products, the association with depressive symptoms was significant both for Q2 v. Q1 (OR 0·61; 95 % CI 0·41, 0·93) and Q3 v. Q1 (OR 0·37; 95 % CI 0·24, 0·59) in the full-adjusted model. In the dose–response analysis, the shape of the associations of total protein intake and protein intake from milk and milk products with the risk of depressive symptoms was approximately L-shaped and U-shaped, respectively. The present study demonstrated that total protein intake and protein intake from milk and milk products might reduce the risk of depressive symptoms in US adults.

BMI, waist circumference (WC) and waist-to-height ratio (WHtR) can be used for discriminating children and adolescents at risk of CVD. However, consensus on how to use these anthropometric indicators is lacking for children and adolescents in Asia. Discrete criteria are promoted internationally, but continuous variables could be used. Data from a survey of 10 949 Vietnamese school-aged children (6–18 years) were used to evaluate the performance of anthropometric indicators to identify elevated blood pressure (BP), dyslipidaemia or at least three cardiovascular risk factors (CVRF). Weight, height, WC and BP were measured using standardised protocols; 1009 participants who had blood lipids were analysed. AUC was used to assess the performance, and the Youden index to identify optimal cut-offs. The prevalence of elevated BP, dyslipidaemia and CVRF was 26·5, 49·3 and 12·2 %, respectively. BMI, WC and WHtR had low capacity to identify elevated BP and dyslipidaemia (AUC range 0·61–0·66) but moderate capacity to identify CVRF (0·72–0·74). Optimal BMIZ cut-offs to identify elevated BP, dyslipidaemia and CVRF were 0·40, 1·01 and 1·1 sd; for WC z-score, they were 0·06, 0·49 and 0·62 sd; for WHtR, optimal cut-offs were close to 0·5. A BMIZ cut-off of 1·0 sd and a WHtR cut-off of 0·5 would, therefore, be useful criteria to identify Vietnamese children who are likely to have CVRF. However, further validation of these criteria in other studies of Asian children and adolescents is needed.

The aim of the study was to investigate whether the interaction of physical activity (PA) and protein intake is associated with physical function (PF). The women from the Osteoporosis Risk Factor and Fracture Prevention Study (n 610) completed a questionnaire on lifestyle factors and PA and underwent PF and body composition measurements at baseline (BL) and over 3 years of follow-up (3y-FU). PA was categorised according to WHO cut-off PA = 0, 0 < PA < 2·5 and PA ≥ 2·5 h/week. Protein intake was calculated from the 3-d food record at baseline and categorised according to the Nordic Nutrition Recommendations <1·1 and ≥1·1 g/kg body weight (BW). The results showed in univariate ANOVA at the baseline and at the 3-year follow-up, women with high PA ≥ 2·5 h/week and protein intake ≥ 1·1 g/kg BW had higher grip strength adjusted for BMI, higher mean number of chair rises, faster mean walking speed, higher modified mean short physical performance battery score and lower mean fat mass compared with other interaction groups. High PA and protein intake were associated with lower BMI despite significantly higher energy intake. In conclusion, higher PA and protein intake interaction was associated with greater PF and lower fat mass, but the association with relative skeletal muscle index and muscle mass was inconclusive. The present study gives noteworthy information for preventing sarcopenia.

In the present study, the aim was to investigate the correlation between the acute and habitual dietary intake of flavanones, their main food sources and the concentrations of aglycones naringenin and hesperetin in 24 h urine in a European population. A 24-h dietary recall (24-HDR) and a 24-h urine sample were collected the same day from a subsample of 475 people from four different countries of the European Prospective Investigation into Cancer and Nutrition study. Acute and habitual dietary data were captured through a standardised 24-HDR and a country/centre-specific validated dietary questionnaire (DQ). The intake of dietary flavanones was estimated using the Phenol-Explorer database. Urinary flavanones (naringenin and hesperetin) were analysed using tandem MS with a previous enzymatic hydrolysis. Weak partial correlation coefficients were found between urinary flavanone concentrations and both acute and habitual dietary flavanone intakes (Rpartial = 0·14–0·17). Partial correlations were stronger between urinary excretions and acute intakes of citrus fruit and juices (Rpartial ∼ 0·6) than with habitual intakes of citrus fruit and juices (Rpartial ∼ 0·24). In conclusion, according to our results, urinary excretion of flavanones can be considered a good biomarker of acute citrus intake. However, low associations between habitual flavanone intake and urinary excretion suggest a possible inaccurate estimation of their intake or a too sporadic intake. For assessing habitual exposures, multiple urinary collections may be needed. These results show that none of the approaches tested is ideal, and the use of both DQ and biomarkers can be recommended.

Little information is known about the influence of altitude on child growth in Ethiopia, where most people live in highlands. We investigated the relation of residential altitude with growth faltering (stunting) of infants and young children in Ethiopia. We also examined whether the altitude–growth relationship was independent of the influence of the dietary and non-dietary determinants of growth. We used the data of 26 976 under-5-year-old children included in the Ethiopian Demographic and Health Surveys, conducted from 2005 to 2016. The samples were recruited following a two-stage cluster sampling strategy. Stunting was defined by height-for-age <−2 z-scores. The relationship between residential altitude and stunting was examined by running multiple logistic regression analysis, controlling the effect of covariate dietary and non-dietary variables. The residential altitude of the study participants ranged from −116 to 4500 m above sea level (masl). There was a significant and progressive increase in the prevalence and odds of stunting with increasing altitude (P < 0·001), irrespective of the dietary and non-dietary predictors of stunting. The prevalence of stunting was lowest in lowlands (39 %) and highest in highlands (47 %). Compared with altitude <1000 masl, the odds of stunting was 1·41 times higher at altitude ≥2500 masl (OR 1·41, 95 % CI 1·16, 1·71) and 1·29 times higher at altitude 2000–2499 masl (OR 1·29, 95 % CI 1·11, 1·49). Children living in highlands might be at a higher risk of poor growth. Further studies are warranted to understand the mechanism behind the observed altitude–stunting link and identify strategies to compensate for the growth-faltering effect of living in highlands.

Starting university is an important time with respect to dietary changes. This study reports a novel approach to assessing student diet by utilising student-level food transaction data to explore dietary patterns. First-year students living in catered accommodation at the University of Leeds (UK) received pre-credited food cards for use in university catering facilities. Food card transaction data were obtained for semester 1, 2016 and linked with student age and sex. k-Means cluster analysis was applied to the transaction data to identify clusters of food purchasing behaviours. Differences in demographic and behavioural characteristics across clusters were examined using χ2 tests. The semester was divided into three time periods to explore longitudinal changes in purchasing patterns. Seven dietary clusters were identified: ‘Vegetarian’, ‘Omnivores’, ‘Dieters’, ‘Dish of the Day’, ‘Grab-and-Go’, ‘Carb Lovers’ and ‘Snackers’. There were statistically significant differences in sex (P < 0·001), with women dominating the Vegetarian and Dieters, age (P = 0·003), with over 20s representing a high proportion of the Omnivores and time of day of transactions (P < 0·001), with Dieters and Snackers purchasing least at breakfast. Many students (n 474, 60·4 %) changed dietary cluster across the semester. This study demonstrates that transactional data present a feasible method for dietary assessment, collecting detailed dietary information over time and at scale, while eliminating participant burden and possible bias from self-selection, observation and attrition. It revealed that student diets are complex and that simplistic measures of diet, focusing on narrow food groups in isolation, are unlikely to adequately capture dietary behaviours.

This study aimed to investigate the association between 25-hydroxyvitamin D (25(OH)D) level and low back pain (LBP) among adolescents while adjusting for potential confounders pertinent to this age group including the weight of school bags, BMI and physical activity. A cross-sectional study was conducted on 760 randomly selected adolescents in middle schools. Data on LBP and the risk factors for LBP were collected from parents by a self-administered questionnaire and from adolescents by face-to-face interview. Blood samples were tested in an accredited laboratory; and 25(OH)D was measured using liquid chromatography-tandem MS. The lifetime prevalence and the 6-month prevalence of LBP were 32·28 (95 % CI 28·97, 35·73) % and 21·26 (95 % CI 18·40, 24·33) %, respectively. There was no difference in the geometric mean of 25(OH)D between those with and without LBP in the past 6 months (28·50 nmol/l and 30·82 nmol/l, respectively; P = 0·122). There was no association between 25(OH)D and LBP in the univariable or multivariable analysis whether 25(OH)D fitted as a continuous or as a categorical variable. We found no association between vitamin D level and LBP in adolescents in an area with high prevalence of vitamin D deficiency. Although it is important to have sufficient vitamin D levels during adolescence for several other health benefits, we concluded that vitamin D is not a major determinant for LBP among adolescents in our setting.

Prior data on long-term association between legume consumption and hypertension risk are sparse. We aimed to evaluate whether total legume and subtype intakes prospectively associate with hypertension incidence among 8758 participants (≥30 years) from the China Health and Nutrition Survey 2004–2011. Diet was assessed by interviews combining 3-d 24-h food recalls and household food inventory weighing at each survey round. Incident hypertension was identified by self-reports or blood pressure measurements. We applied multivariable Cox regressions to estimate hazard ratios (HR) with corresponding 95 % CI for hypertension across increasing categories of cumulatively averaged legume intakes. For 35 990 person-years (median 6·0 years apiece), we documented 944 hypertension cases. After covariate adjustment, higher total legume intakes were significantly associated with lower hypertension risks, with HR comparing extreme categories being 0·56 (95 % CI 0·43, 0·71; Ptrend < 0·001). Then we found that intakes of dried legumes (HR 0·53 (95 % CI 0·43, 0·65); Ptrend < 0·001) and fresh legumes (HR 0·67 (95 % CI 0·55, 0·81); Ptrend < 0·001) were both related to decreased hypertension hazards. However, further dried legume classification revealed that negative association with hypertension substantially held for soyabean (HR 0·51 (95 % CI 0·41, 0·62); Ptrend < 0·001) but not non-soyabean intakes. In stratified analyses, the association of interest remained similar within strata by sex, BMI, physical activity, smoking and drinking status; rather, significant heterogeneity showed across age strata (Pinteraction = 0·02). Total legume consumption among the over-65s was related to a more markedly reduced hypertension risk (HR 0·47 (95 % CI 0·30, 0·73); Ptrend < 0·001). Our findings suggest an inverse association of all kinds of legume (except non-soyabean) intakes with hypertension risks.

During pregnancy, mothers-to-be should adapt their diet to meet increases in nutrient requirements. Pregnant women appear to be keener to adopt healthier diets, but are not always successful. The objective of the present study was to determine whether a guided, stepwise and tailored dietary counselling programme, designed using an optimisation algorithm, could improve the nutrient adequacy of the diet of pregnant women, beyond generic guidelines. Pregnant women (n 80) who attended Notre-Dame-de-Bon-Secours Maternity Clinic were randomly allocated to the control or intervention arm. Dietary data were obtained twice from an online 3-d dietary record. The nutrient adequacy of the diet was calculated using the PANDiet score, a 100-point diet quality index adapted to the specific nutrient requirements for pregnancy. Women were supplied with generic dietary guidelines in a reference booklet. In the intervention arm, they also received nine sets of tailored dietary advice identified by an optimisation algorithm as best improving their PANDiet score. Pregnant women (n 78) completed the 12-week dietary follow-up. Initial PANDiet scores were similar in the control and intervention arms (60·4 (sd 7·3) v. 60·3 (sd 7·3), P = 0·92). The PANDiet score increased in the intervention arm (+3·6 (sd 9·3), P = 0·02) but not in the control arm (−0·3 (sd 7·3), P = 0·77), and these changes differed between arms (P = 0·04). In the intervention arm, there were improvements in the probabilities of adequacy for α-linolenic acid, thiamin, folate and cholesterol intakes (P < 0·05). Tailored dietary counselling using a computer-based algorithm is more effective than generic dietary counselling alone in improving the nutrient adequacy of the diet of French women in mid-pregnancy.

The EHU12/24 (code of a survey from the University of the Basque Country/Euskal Herriko Unibertsitatea, UPV/EHU) study was designed to investigate the risk factors of overweight/obesity-related lifestyles, particularly those associated with diet, their psychosocial influences and the interactions among these factors. This observational cohort study was carried out according to a standardised protocol and involved a representative sample of the UPV/EHU student population. Anthropometric measurements, direct behavioural determinants, such as physical activity and diet, and indirect determinants, such as social/psychological factors, are considered. In this paper, we present the survey design, instruments, measurements and related quality management. We describe the study sample in terms of its socio-economic and demographic factors and knowledge area and summarise the methodology used to collect the data and obtain the anthropometric measurements. The participants were 603 students (59·5 % female) aged 18–28 years. The crude participation proportion was 53·5 %. Regarding the knowledge area, the lowest response proportions were obtained from the Health Sciences (38·6 %) compared with the Non-Health Sciences (48·3 %) (P = 0·003). The mean age was 20·9 years, and 83·1 % of the sample were from Basque Country. Regarding the socio-economic characteristics, there were significant differences by sex and knowledge area in most studied variables. Moreover, the Health Sciences students were more likely younger, from outside Basque Country, to have parents with university degrees and to have a higher social status. In conclusion, the EHU12/24 cohort provides valuable data for analysing the complexity and multidimensionality of obesity in university students.

Current evidence indicates that maternal diets before and during pregnancy could influence rates of preterm birth, low birth weight (LBW) and small for gestational age (SGA) births. However, findings have been inconsistent. This review summarised evidence concerning the effects of maternal diets before and during pregnancy on preterm birth, LBW and SGA. Systematic electronic database searches were carried out using PubMed, Embase, Scopus and Cochrane library using the preferred reporting items for systematic reviews and meta-analyses guidelines. The review included forty eligible articles, comprising mostly of prospective cohort studies, with five randomised controlled trials. The dietary patterns during pregnancy associated with a lower risk of preterm birth were commonly characterised by high consumption of vegetables, fruits, whole grains, fish and dairy products. Those associated with a lower risk of SGA also had similar characteristics, including high consumption of vegetables, fruits, legumes, seafood/fish and milk products. Results from a limited number of studies suggested there was a beneficial effect on the risk of preterm birth of pre-pregnancy diet quality characterised by a high intake of fruits and proteins and less intake of added sugars, saturated fats and fast foods. The evidence was mixed for the relationship between maternal dietary patterns during pregnancy and LBW. These findings indicate that better maternal diet quality during pregnancy, characterised by a high intake of vegetables, fruits, whole grains, dairy products and protein diets, may have a synergistic effect on reducing the risk of preterm birth and SGA.

Few studies have investigated the association between maternal dietary patterns (DP) during pregnancy, derived from reduced-rank regression (RRR), and fetal growth. This study aims to identify DP during pregnancy associated with macro- and micronutrient intakes, using the RRR method, and to examine their relationship with birth weight (BW). We used data of 7194 women from a large-scale cross-sectional survey in Northwest China. Dietary protein, carbohydrate, haem Fe density and the ratio of PUFA and MUFA:SFA were used as the intermediate variables in the RRR model to extract DP. Generalised estimating equation models were applied to evaluate the associations between DP and BW and related outcomes (including BW z-score, low birth weight (LBW) and small for gestational age (SGA)). Four DP during pregnancy were identified. Socio-demographically disadvantaged pregnant women were more likely to have lower BW and lower adherence to DP1 (high legumes, soyabean products, vegetables and animal-source foods, with relative low wheat and oils). Women with medium and high adherence to DP1 had significantly increased BW (medium 28·6 (95 % CI 7·1, 50·1); high 25·2 (95 % CI 2·7, 47·6)) and BW z-score and had significantly reduced risks of LBW and SGA. The associations were stronger among women with babies <3100 g. There is no association between other DP and outcomes. Higher adherence to the DP that was high in legumes, soyabean products, vegetables and animal-source foods was associated with improved BW in the Chinese pregnant women, particularly among those with disadvantageous socio-demographic conditions.