Iron is essential for the functioning of all cells and organs, most critically for the developing brain in the fundamental neuronal processes of myelination, energy and neurotransmitter metabolism. Iron deficiency, especially in the first 1000 days of life, can result in long-lasting, irreversible deficits in cognition, motor function and behaviour. Pregnant women, infants and young children are most vulnerable to iron deficiency, due to their high requirements to support growth and development, coupled with a frequently inadequate dietary supply. An unrecognised problem is that even if iron intake is adequate, common pregnancy-related and lifestyle factors can affect maternal–fetal iron supply in utero, resulting in an increased risk of deficiency for the mother and her fetus. Although preterm birth, gestational diabetes mellitus and intrauterine growth restriction are known risk factors, more recent evidence suggests that maternal obesity and delivery by caesarean section further increase the risk of iron deficiency in the newborn infant, which can persist into early childhood. Despite the considerable threat that early-life iron deficiency poses to long-term neurological development, life chances and a country's overall social and economic progress, strategies to tackle the issue are non-existent, too limited or totally inappropriate. Prevention strategies, focused on improving the health and nutritional status of women of reproductive age are required. Delayed cord clamping should be considered a priority. Better screening strategies to enable the early detection of iron deficiency during pregnancy and early-life should be prioritised, with intervention strategies to protect maternal health and the developing brain.

A multi-disciplinary expert group met to discuss vitamin D deficiency in the UK and strategies for improving population intakes and status. Changes to UK Government advice since the 1st Rank Forum on Vitamin D (2009) were discussed, including rationale for setting a reference nutrient intake (10 µg/d; 400 IU/d) for adults and children (4+ years). Current UK data show inadequate intakes among all age groups and high prevalence of low vitamin D status among specific groups (e.g. pregnant women and adolescent males/females). Evidence of widespread deficiency within some minority ethnic groups, resulting in nutritional rickets (particularly among Black and South Asian infants), raised particular concern. Latest data indicate that UK population vitamin D intakes and status reamain relatively unchanged since Government recommendations changed in 2016. Vitamin D food fortification was discussed as a potential strategy to increase population intakes. Data from dose–response and dietary modelling studies indicate dairy products, bread, hens’ eggs and some meats as potential fortification vehicles. Vitamin D3 appears more effective than vitamin D2 for raising serum 25-hydroxyvitamin D concentration, which has implications for choice of fortificant. Other considerations for successful fortification strategies include: (i) need for ‘real-world’ cost information for use in modelling work; (ii) supportive food legislation; (iii) improved consumer and health professional understanding of vitamin D’s importance; (iv) clinical consequences of inadequate vitamin D status and (v) consistent communication of Government advice across health/social care professions, and via the food industry. These areas urgently require further research to enable universal improvement in vitamin D intakes and status in the UK population.

Micronutrient deficiency persists throughout the world, and although the burden is higher in low-resource settings, it is also prevalent in wealthy countries, a phenomenon termed ‘hidden hunger’. Due to their high requirements for vitamins and minerals relative to their energy intake, young women and children are particularly vulnerable to hidden hunger. As they share several risk factors and impact on overlapping outcomes, we consider how deficiency of iron, iodine and vitamin D can have profound impacts on perinatal health and infant development. We review the epidemiology of these micronutrient deficiencies during pregnancy, including social, environmental and dietary risk factors. We identify the main challenges in defining nutritional status of these nutrients using validated diagnostic criteria linked with meaningful clinical outcomes. Public health strategies are urgently required to improve the overall health and nutritional status of women of reproductive age. Obesity prevention and early detection of malnutrition with standardised screening methods would detect pregnant women at increased risk of iron deficiency. Development of sensitive, individual biomarkers of iodine status is required to protect maternal health and fetal/infant brain development. Risk assessments of vitamin D requirements during pregnancy need to be revisited from the perspective of fetal and neonatal requirements. International consensus on standardised approaches to micronutrient assessment, analysis and reporting as well as sensitive, clinically validated infant and child neuro-behavioural outcomes will enable progression of useful observational and intervention studies.

In parallel with increased public awareness of the health and environmental benefits of consuming a plant-based diet, the numbers of people who identify as vegan has increased sharply. The question of whether vegetarian and vegan diets are appropriate for children is a longstanding and unresolved controversy. The more restrictive the diet and the younger the child, the greater the risk of nutritional deficiency. Nutrients of potential concern are protein quantity and quality, iron, zinc, selenium, calcium, riboflavin, vitamins A, D, B12 and essential fatty acids. Although intakes and status of some nutrients (e.g. vitamin D and iron) are low in many children, vegan children are particularly susceptible due to inadequate supply and/or excess dietary fibre as well as other components that limit bioavailability. Although position papers from North America state that well-planned vegetarian and vegan diets, supplemented appropriately, are suitable for all life stages, European statements include strong recommendations to parents that vegan diets should not be adopted by children without medical and dietetic supervision. Case histories of malnutrition and serious harm persist, including irreversible neurological damage due to vitamin B12 deficiency among un-supplemented children. The evidence available to evaluate the nutritional appropriateness of vegetarian diets for children is inadequate and dated. Although nutritionally adequate vegetarian diets are more easily achieved, successful provision of a complete vegan diet for a young child requires substantial commitment, expert guidance, planning, resources and supplementation.

Recognised as the optimum infant feeding method, breastfeeding affords substantial health benefits to mother and infant. Exclusive breastfeeding is recommended to 6 months, with continuation thereafter. In Ireland, data indicates that 60% breastfeed to any extent at hospital discharge and 40% do so at 3 months. However, nationally collected data is limited in detail, particularly with regard to breastfeeding continuation. The ongoing Cork-based COMBINE birth cohort study recruited 456 participants between 2015 and 2017. Through interviewer-led questionnaires, research midwives collected feeding data at hospital discharge, 1, 2, 4, 6 and 9 months of age. Infant feeding was classed as breast (breastmilk as main milk source), combination (both breastmilk and infant formula daily) or infant formula only feeding. Adherence to the World Health Organisation (WHO) recommendation to exclusively breastfeed (breastmilk only, no infant formula, supplementary fluids or solid foods) was examined longitudinally. To explore secular changes, breastfeeding rates in COMBINE were compared to the BASELINE birth cohort study, which recruited participants (n = 2116) between 2008 and 2011 in the same setting. In COMBINE, 75% of mothers provided any breastmilk (breast or combination) at hospital discharge, of whom 44% breastfed only. Two-fifths (40%), one-third (33%) and one-quarter (24%) breastfed at 1, 4 and 6 months, respectively. Combination feeding of breastmilk and infant formula was prevalent at discharge (31%) and 1 month (20%). Advice from healthcare professionals (31%) and concern baby was hungry (30%) were commonly reported reasons for this practice and throughout the study, mothers who combination fed were more likely to stop breastfeeding altogether than those who breastfed (all P < 0.001). Infant formula only feeding increased from 25% at discharge to 40% at 1 month, 49% at 2 months and 74% at 9 months. Half (45%) of mothers who breastfed did so for less than 4 months and insufficient milk/growth faltering was the most commonly cited reason for cessation (27%). The rate of WHO-defined exclusive breastfeeding was 40% from birth to discharge, 22% to 1 month, 15% to 4 months and 2% to 6 months. The breastfeeding (breastmilk as main milk source) rate at discharge did not differ between our two birth cohorts (44 vs. 40%, P = 0.23), but was significantly higher in COMBINE compared to BASELINE at 2 (36 vs. 27%) and 6 (24 vs. 12%) months (both P < 0.05). While these data provide evidence of some progress towards longer breastfeeding durations, there remains much scope to improve infant feeding practices in Ireland.

The high prevalence of maternal deficiency and the low vitamin D content of breastmilk places newborns and infants at particular risk of vitamin D deficiency. In response to an increase in the incidence of nutritional rickets, the Food Safety Authority of Ireland published an interim infant vitamin D supplementation policy in 2007, which was implemented by the Health Service Executive in Ireland in May 2010. This recommends that all infants be given a 5μg exclusive vitamin D3 supplement daily from birth to 12 months. As adherence is not monitored nationally and the policy has not been evaluated, the aim of this study was to conduct a detailed evaluation of supplementation practices across two maternal-infant cohort studies. Data from the prospective BASELINE (recruited 2008–2011) and COMBINE (recruited 2015–2017) birth cohorts, based in Cork, Ireland, were used to examine supplementation practices. After supplementation policy implementation, BASELINE collected vitamin D supplement use data (n = 1528) at 2, 6 and 12 months. In COMBINE, 7 study visits from birth to 12 months allowed continuous collection of detailed longitudinal supplementation data in 364 participants. Use of supplemental vitamin D was higher in COMBINE than BASELINE at 2 (93 vs. 49%), 6 (89 vs. 64%) and 12 (72 vs. 44%) months (all P < 0.001). In COMBINE, 92% initiated vitamin D supplementation at birth and the median supplementation duration was 51 (40, 52) weeks, although there was a wide range (3–52 weeks). 94% of COMBINE parents used a vitamin D3 only supplement and 88% used the recommended 5μg dose. Half (51%) always supplemented daily; a further 33% supplemented at least 3–6 times/week. Full policy adherence was defined as the provision of a 5μg vitamin D3 supplement daily from birth; 64% adhered fully to 2 months and 52% did so to 6 months. By 12 months, 30% had adhered fully to the policy and a further 16% gave 5μg frequently for the full 12 months. This data indicates a high level of awareness of the supplementation policy amongst new mothers, with substantially higher rates of supplementation in our current cohort compared with BASELINE. While most parents gave an exclusive 5μg vitamin D3 supplement, frequency and duration were the key barriers to full policy adherence. Given the lack of a maternal vitamin D supplementation policy in Ireland and high prevalence of low vitamin D status at birth, supplementation of infants with vitamin D remains a vital public health policy.

Dietary protein is critical for normal growth and development through its contribution of essential amino acids. In high-resource settings, protein intakes are generally adequate. Previous studies have suggested that higher protein intakes in infancy and early childhood are associated with higher body mass index in later childhood.

This analysis was performed in a subgroup of 2-year-old children (n = 468) with detailed dietary intake data (2-d weighed food record) participating in an extensively-characterised, prospective birth cohort, the Cork BASELINE Birth Cohort Study (n = 2183). Body weight and height were measured at 2- and 5-years, and indices of body composition (fat mass, fat free mass) were obtained at 5 years in a subgroup of 295 children. Total protein intake was estimated at 2-years and the contribution of animal and plant sources to total protein intake was quantified. Children were split into thirds of protein intake (as % total energy, %TE) and anthropometry and body composition at 5 years was compared across groups. The relationship between protein intake (total and animal sources) and BMI-SDS, fat mass index-SDS and fat free mass index-SDS was explored in multivariate linear regression models, adjusted for common confounders (energy intake, gestational age, duration of breastfeeding, birthweight-SDS, maternal education).

Mean (SD) daily protein intake was 41.3 (11.2) g/d, representing 16.1 %TE (girls: 16.2 %TE; boys: 15.9 %TE, P = 0.202) and all children met EFSA average requirement and population reference intake thresholds. Meat (25%), cows’ milk (22%), breads (7%), yoghurt (6%), breakfast cereals (6%) and cheese (5%) were key sources of protein intake in toddlers. The majority of dietary protein intake was from animal sources (69%).

Girls with high %TE from protein [mean (SD): 19.7 (2.6) %TE] were significantly heavier and taller at 5 years than their counterparts [low: 13.1 (1.4) %TE; medium: 16.0 (0.7) %TE], and had significantly higher fat free mass, fat free mass index and total body bone area (indicative of body size), but not fat mass. When adjusted for confounders in multivariate regression models, no association was observed between total protein intake and BMI-SDS, fat mass index-SDS or fat free mass index-SDS at 5 years.

Children with high animal protein intakes (as %TE) had significantly higher fat free mass than their low and medium group counterparts; however, when adjusted for height and other confounders, no association was observed.

Protein intakes in early childhood were associated with larger body size, but not BMI or body composition at 5-years.

Little consideration has been given to the long-term consequences of iron deficiency in new-born infants. Fetal iron accretion is compromised by multiple pregnancy complications including preterm birth, gestational diabetes mellitus and fetal growth restriction, while our work has identified increased risks from maternal lifestyle factors such as smoking and obesity for low iron stores at birth. Early-life events, including C-section delivery, also add to this cumulative risk of neonatal iron deficiency, predisposing infants to iron deficiency later in infancy and early childhood. This study aimed to investigate the effect of neonatal iron deficiency on neurological development up to 5 years of age in term-born participants of a maternal-infant birth cohort in Ireland. In the Cork BASELINE Birth Cohort, 697 maternal-infant dyads with prospectively collected lifestyle and clinical data from 15 weeks’ gestation had umbilical cord serum ferritin concentrations measured. Neurological assessments were performed at 2 (Bayley Scales of Infant Development and Child Behaviour Checklist [CBCL]) and 5 (Kaufman Brief Intelligence Test and CBCL) years of age. In the cohort, median [IQR] cord ferritin concentrations were 200.9 [139.0,265.8] μg/L; 7.5% had neonatal iron deficiency (< 76μg/L). Using the risk factors for neonatal iron deficiency that we previously identified (smoking, obesity, C-section delivery, SGA) in this cohort, as selection criteria, we conducted an a priori sensitivity analysis in 306 children. Of the 306 children identified as high-risk, 12.4% had neonatal iron deficiency. Those with neonatal iron deficiency had higher median [IQR] CBCL internal (9.0 [5.3,12.0] vs. 5.0 [3.0,10.0]), external (7.5 [4.0,14.8] vs. 5.0 [2.0,10.0]) and total problem (24.5 [15.3,40.8] vs. 16.0 [10.0,30.0], all P < 0.05) scores at 5 years compared to those without neonatal deficiency. This adverse effect was especially apparent in children of obese mothers (n = 85) who were iron deficient at birth, with a total problem score at 5 years of 42.0 [24.5,54.5] compared to 16.0 [8.8,29.3] in those not deficient (P = 0.008). Associations were robust to adjustment for confounding factors. No effect on cognition or intelligence at 2 or 5 years was observed in this cohort. This study has identified behavioural consequences of neonatal iron deficiency. Interventions targeting the fetal/neonatal period could, therefore, represent a key opportunity for prevention of iron deficiency and its associated long-term neurological consequences. A dual approach is required, comprising public health strategies targeting prevention, through improving nutrition and health in women, and the development of screening strategies for early detection of iron deficiency in new-borns.

Breast-feeding initiation and continuation rates in the UK and Ireland are low relative to many European countries. As a core outcome of the prospective Cork Nutrition and Development Maternal-Infant Cohort (COMBINE) study (Cork, Ireland), we aimed to describe infant milk feeding practices in detail and examine the prevalence and impact of combination feeding of breast milk and infant formula on breast-feeding duration. COMBINE recruited 456 nulliparous mothers (2015–2017) for maternal–infant follow-up via interview at hospital discharge (median 3 (interquartile range (IQR) 2, 4) d (n 453)), 1 (n 418), 2 (n 392), 4 (n 366), 6 (n 362) and 9 (n 345) months of age. Median maternal age was 32 (IQR 29, 34) years, 97 % of mothers were of white ethnicity, 79 % were Irish-born and 75 % were college-educated. Overall, 75 % breastfed to any extent at discharge and 44 % breastfed solely. At 1, 2, 4, 6 and 9 months, respectively, 40, 36, 33, 24 and 19 % breastfed solely. Combination feeding of breast milk and infant formula was common at discharge (31 %) and 1 month (20 %). Reasons for combination feeding at 1 month included perceived/actual hunger (30 %), healthcare professional advice (31 %) and breast-feeding difficulties (13 %). Of mothers who breastfed to any extent at discharge, 45 % stopped within 4 months. Mothers who combination fed were more likely to cease breast-feeding than those who breastfed solely (relative risk 2·3 by 1 month and 12·0 by 2 months). These granular data provide valuable insight to early milk feeding practices and indicate that supporting early breast-feeding without formula use may be key to the successful continuation of breast-feeding.

Breastfeeding may reduce obesity risk, but this association could be confounded by breastfeeding families’ characteristics. We investigated if body composition differs at birth among infants who were either exclusively breast- or formula-fed. We hypothesized the two groups would differ in body composition, even at birth, prior to their post-natal feeding experience. Healthy primiparous carrying singleton pregnancy were recruited at 15 weeks’ gestation. PEA POD® measured body composition within 72 hours of delivery and infant feeding was prospectively captured. Out of the 1,152 infants recruited, 117 (10.2%) and 239 (20.7%) went on to be either exclusively breast- or formula-fed, respectively. Breastfed infants were heavier at birth, but their percentage fat mass (FM) was lower than that of exclusively formula-fed infants (covariate adjusted β = −1.91 percentage points of FM; 95% CI −2.82 to −1.01). Differences in intra-uterine exposures, irrespective of early diet, may partly explain an infant’s obesity risk.

Despite the inverse association between skin colour and efficiency of cutaneous vitamin D synthesis, in addition to the widely accepted racial disparity in vitamin D status, populations of ethnic minority are understudied in terms of setting target serum 25-hydroxyvitamin D concentrations and corresponding dietary requirements for vitamin D. In minority groups, prevention of vitamin D deficiency on a population basis is challenging due to the lack of clarity surrounding the metabolism and transport of vitamin D. Authoritative agencies have been unable to define pregnancy-specific dietary recommendations for vitamin D, owing to an absence of sufficient evidence to confirm whether nutritional requirements for vitamin D are altered during pregnancy. While the question of setting race- and pregnancy-specific dietary reference values for vitamin D has not been addressed to date, endemic vitamin D deficiency has been reported among gravidae worldwide, specifically among ethnic minorities and white women resident at high latitude. In light of the increased risk of nutritional rickets among infants of ethnic minority, coupled with growing evidence for potential non-skeletal roles of vitamin D in perinatal health, determination of the dietary vitamin D requirement that will prevent deficiency during pregnancy is a research priority. However, systematic approaches to establishing dietary requirements are limited by the quality of the available evidence and the under-representation of minority groups in clinical research. This review considers the evidence for racial differences in vitamin D status and response to vitamin D supplementation, with particular application to pregnancy-specific requirements among ethnic minorities resident at high latitudes.

Fe deficiency in early childhood is associated with long-term consequences for cognitive, motor and behavioural development; however explorations in healthy children from low risk, high-resource settings have been limited. We aimed to explore associations between Fe status and neurodevelopmental outcomes in low risk, healthy 2-year-olds. This study was a secondary analysis of a nested case–control subgroup from the prospective, maternal-infant Cork Babies after Screening for Pregnancy Endpoints: Evaluating the Longitudinal Impact using Neurological and Nutritional Endpoints (BASELINE) Birth Cohort Study. At 2 years, serum ferritin, Hb and mean corpuscular volume (MCV) were measured and neurodevelopment was assessed using the Bayley Scales of Infant and Toddler Development (n 87). Five children had Fe deficiency (ferritin <12 µg/l) and no child had Fe deficiency anaemia (Hb<110 g/l+ferritin<12 µg/l). Children with microcytosis (MCV<74 fl, n 13) had significantly lower mean cognitive composite scores (88·5 (sd 13·3) v. 97·0 (sd 7·8), P=0·04, Cohen’s d effect size=0·8) than those without microcytosis. The ferritin concentration which best predicted microcytosis was calculated as 18·4 µg/l (AUC=0·87 (95% CI 0·75, 0·98), P<0·0001, sensitivity 92 %, specificity 75 %). Using 18·5 µg/l as a threshold, children with concentrations <18·5 µg/l had significantly lower mean cognitive composite scores (92·3 (sd 10·5) v. 97·8 (sd 8·1), P=0·012, Cohen’s d effect size=0·6) compared with those with ferritin ≥18·5 µg/l. All associations were robust after adjustment for potential confounding factors. Despite a low prevalence of Fe deficiency using current diagnostic criteria in this healthy cohort, microcytosis was associated with lower cognitive outcomes at 2 years. This exploratory study emphasises the need for re-evaluation of the diagnostic criteria for Fe deficiency in young children, with further research in adequately powered studies warranted.

The current study was aiming to report the prevalence of suboptimal vitamin D status among schoolchildren in Greece and investigate the role of sex, urbanisation and seasonality on vitamin D status. A sample of 2386 schoolchildren (9–13 years old) from four distinct prefectures was examined. The prevalence of 25-hydroxyvitamin D (25(OH)D) concentration <30 and <50 nmol/l (vitamin D deficiency and insufficiency respectively) was 5·2 and 52·5 %, respectively. Girls had a higher prevalence of 25(OH)D<30 (7·2 v. 3·2 %) and 50 nmol/l (57·0 v. 48·0 %) than boys (P<0·001). The highest prevalence rates of 25(OH)D<30 and 50 nmol/l (9·1 and 73·1 %, respectively) were observed during spring (April to June), whereas the lowest (1·5 and 31·9 %, respectively) during autumn (October to December). The prevalence of 25(OH)D<50 nmol/l was higher in urban/semi-urban than rural regions, particularly during spring months (74·6 v. 47·2 %; P<0·001). Female sex, urban/semi-urban region of residence and spring months were found to increase the likelihood of vitamin D deficiency and insufficiency, with the highest OR observed for spring months (7·47; 95 % CI 3·23, 17·3 and 5·14; 95 % CI 3·84, 6·89 for 25(OH)D<30 and 50 nmol/l respectively). In conclusion, despite the southerly latitude, the prevalence of low vitamin D status among primary schoolchildren in Greece is comparable to or exceeds the prevalence reported among children and adolescents on a European level. Sub-populations at highest risk are girls in urban/semi-urban areas during spring months, thus indicating the need for effective initiatives to support adequate vitamin D status in these population groups.

The goal of preterm nutrition in achieving growth and body composition approximating that of the fetus of the same postmenstrual age is difficult to achieve. Current nutrition recommendations depend largely on expert opinion, due to lack of evidence, and are primarily birth weight based, with no consideration given to gestational age and/or need for catch-up growth. Assessment of growth is based predominately on anthropometry, which gives insufficient attention to the quality of growth. The present paper provides a review of the current literature on the nutritional management and assessment of growth in preterm infants. It explores several approaches that may be required to optimise nutrient intakes in preterm infants, such as personalising nutritional support, collection of nutrient intake data in real-time, and measurement of body composition. In clinical practice, the response to inappropriate nutrient intakes is delayed as the effects of under- or overnutrition are not immediate, and there is limited nutritional feedback at the cot-side. The accurate and non-invasive measurement of infant body composition, assessed by means of air displacement plethysmography, has been shown to be useful in assessing quality of growth. The development and implementation of personalised, responsive nutritional management of preterm infants, utilising real-time nutrient intake data collection, with ongoing nutritional assessments that include measurement of body composition is required to help meet the individual needs of preterm infants.

Nutritional requirements for vitamin D during pregnancy have been inadequately described, and there are conflicting data on the impact of gestation on vitamin D status. In the present study, we conducted a longitudinal analysis of total and free (unbound) serum 25-hydroxyvitamin D (25(OH)D), vitamin D-binding protein (DBP) and albumin concentrations in a random sample of thirty women from the Screening for Pregnancy Endpoints Ireland pregnancy cohort study at 15, 20, 24, 28, 32, 36 and 40 weeks of gestation and at 2 months postpartum. Concentrations of serum 25(OH)D, DBP and albumin were determined, and free 25(OH)D was calculated from the concentrations of total 25(OH)D, DBP and albumin. Serum albumin concentration decreased during pregnancy (P< 0·001), with a nadir at 36 weeks (P< 0·005), during which the concentration was approximately 80 % of the postnatal concentration. Serum DBP concentration increased during pregnancy and at 28 weeks of gestation, which was almost double the postnatal level (P< 0·001). Total and free 25(OH)D concentrations decreased (both P< 0·005) as pregnancy progressed, and both were lowest at 36 weeks of gestation. At 15 weeks, 10 and 63 % of the women had serum 25(OH)D concentration < 30 and 50 nmol/l, respectively, which increased to 53 and 80 % at 36 weeks of gestation. The time course of decreasing concentrations of 25(OH)D during gestation among women recruited during May–July, who delivered between October and November, and among those recruited in August–September, who delivered between February and March, was similar. The lower percentage of free 25(OH)D during pregnancy is mainly due to increased DBP.

It has been suggested that vitamin D2 is not very prevalent in the human food chain. However, data from a number of recent intervention studies suggest that the majority of subjects had measurable serum 25-hydroxyvitamin D2 (25(OH)D2) concentrations. Serum 25(OH)D2, unlike 25(OH)D3, is not directly influenced by exposure of skin to sun and thus has dietary origins; however, quantifying dietary vitamin D2 is difficult due to the limitations of food composition data. Therefore, the present study aimed to characterise serum 25(OH)D2 concentrations in the participants of the National Adult Nutrition Survey (NANS) in Ireland, and to use these serum concentrations to estimate the intake of vitamin D2 using a mathematical modelling approach. Serum 25(OH)D2 concentration was measured by a liquid chromatography–tandem MS method, and information on diet as well as subject characteristics was obtained from the NANS. Of these participants, 78·7 % (n 884) had serum 25(OH)D2 concentrations above the limit of quantification, and the mean, maximum, 10th, 50th (median) and 90th percentile values of serum 25(OH)D2 concentrations were 3·69, 27·6, 1·71, 2·96 and 6·36 nmol/l, respectively. To approximate the intake of vitamin D2 from these serum 25(OH)D2 concentrations, we used recently published data on the relationship between vitamin D intake and the responses of serum 25(OH)D concentrations. The projected 5th to 95th percentile intakes of vitamin D2 for adults were in the range of 0·9–1·2 and 5–6 μg/d, respectively, and the median intake ranged from 1·7 to 2·3 μg/d. In conclusion, the present data demonstrate that 25(OH)D2 concentrations are present in the sera of adults from this nationally representative sample. Vitamin D2 may have an impact on nutritional adequacy at a population level and thus warrants further investigation.