In epidemiologic analytical studies, the primary goal is to obtain a valid and precise estimate of the effect of the exposure of interest on a given outcome in the population under study. A crucial source of violation of the internal validity of a study involves bias arising from confounding, which is always a challenge in observational research, including life course epidemiology. The increasingly popular approach of meta-analyzing individual participant data from several observational studies also brings up to discussion the problem of confounding when combining data from different populations. In this study, we review and discuss the most common sources of confounding in life course epidemiology: (i) confounding by indication, (ii) impact of baseline selection on confounding, (iii) time-varying confounding and (iv) mediator–outcome confounding. We also discuss the issue of addressing confounding in the context of an individual participant data meta-analysis.

Exposures to endocrine disrupting chemicals and metals are near ubiquitous worldwide, and their potential impact on children is a major public health concern. This pilot study was designed to characterize exposures to phthalates, phenols and metals among pregnant women in the first trimester, and to examine associations with fetal biometrics and birth weight. A total of 41 chemicals and elements were analyzed in urine from 56 mothers with full-term newborns from the Michigan Mother–Infant Pairs study. Bivariate analyses identified predictors of exposure biomarkers. Associations between birth weight, Fenton z-scores and second trimester fetal biometrics with toxicants were examined via multivariable linear regression. An average of 30 toxicants were detected in maternal urine. Fast food consumption was associated with several phthalate metabolites, phenols and metals, and canned food consumption with bisphenol F (P <0.05). Mono (3-carboxypropyl) phthalate was significantly associated with higher birth weight and Fenton z-score while the opposite was observed for bisphenol S. Estimated femur length from ultrasonography was significantly inversely associated with arsenic, barium and lead. While limited by sample size, this study is one of the first to evaluate birth outcomes with respect to emerging endocrine disrupting chemicals and to examine associations between toxicants and fetal biometrics. Exposure assessment was provided by the National Institute of Environmental Health Sciences’ Children’s Health Exposure Analysis Resource (NIEHS CHEAR), a resource available to children’s studies with the goal of combining data across cohorts in an effort to characterize the impact of toxicants on child health from birth and beyond.

Numerous studies have investigated the risk of developing asthma due to early-life experiences and environmental exposures. However, the influence of intrauterine growth restriction and postnatal undernutrition on childhood wheezing/asthma remains unclear. Thus, we examined the effects of both small for gestational age (SGA) and postnatal stunted growth on ever asthma among children in the rural areas in Bangladesh.

Multiple follow-up studies were conducted in a cohort of randomized clinical trial of nutrition interventions during pregnancy (the MINIMat trial). Overall, 1208 and 1697 children were followed-up for asthma at 4.5 and 10 years, respectively. Anthropometric measurements were obtained at various intervals from birth to 10 years of age. Ever asthma was identified using the International Study of Asthma and Allergies in Childhood (ISAAC) questionnaire.

Results showed that SGA was significantly associated with increased risk of ever asthma at 4.5 and 10 years after adjusting for sex, body mass index, socioeconomic status, family history of asthma, gestational age at birth, mother’s parity, mother’s age at birth and intervention trial arm [odds ratio (OR)=1.97 (95% confidence interval (CI): 1.34–2.90) and 1.86 (95% CI: 1.18–2.72)]. For the postnatal effect of undernutrition, stunting at 1 and 2 years was significantly associated with ever asthma at 4.5 and 10 years [1 year: OR=1.77 (95% CI: 1.22–2.57) and OR=1.72 (95% CI: 1.16–2.56), 2 years: OR=1.49 (95% CI: 1.06–2.10) and OR=1.41 (95% CI: 1.02–1.96)].

In conclusion, SGA and undernutrition during infancy has an influence on childhood asthma among children in Bangladesh, indicating the need for nutritional interventions early in life.

Caesarean delivery has been linked to a number of inflammatory conditions in childhood and adolescence. Yet the mechanisms underlying these associations and their generalizability across contexts with different postnatal feeding and pathogenic exposures remain unclear. This study tests the association between delivery type and three measures of immune function, inflammation, morbidity and leukocyte proportions, in Ecuadorian infants and children aged 6 months to 2 years. Data were collected from mother–child pairs participating in a nationally representative health and nutrition survey Encuesta Nacional de Salud y Nutricion (ENSANUT-ECU) conducted in 2012. The analytic sample includes 828 mothers and infants with delivery information and measured biomarkers. Poisson regression models were used to examine the association between delivery type and markers of immune function, controlling for maternal and infant characteristics, including age, sex, sociodemographic characteristics and medical indications. 40.8% (n=338) of sample infants and children were delivered by caesarean. Compared to those born vaginally, infants born by caesarean were less likely to have elevated C-reactive protein (CRP) [CRP>2 mg/l; risk ratio (RR): 0.76, 95% confidence interval (CI): 0.58–1.00] and more likely to have illness symptoms (RR: 1.22, 95% CI: 1.01–1.46) and elevated basophils (RR: 1.83, 95% CI: 1.03–3.25). No other immune cell proportions differed by delivery type. The results suggest that differences in the perinatal exposures accompanying caesarean delivery may alter immune development and function, particularly in the inflammatory response to infection and in cells involved in the allergic response, across infancy and early childhood. Understanding the pathways linking perinatal exposures to immune development is important for preventing the development of inflammatory conditions.

Barker et al. proposed that low birth weight predisposes to higher death rates in adult life. We previously confirmed this fact in a cohort of young adults who were born in a remote Australian Aboriginal community between 1956 and 1985. We now present data in these same people with four more years of follow-up and a greater number of deaths. The fates of participants were documented from age 15 years until death, start of dialysis, or until the end of 2010 and causes of death were derived from clinic narratives and dialysis records. Rates of natural deaths were compared by birth cohorts and birth weight, and hazard ratios were calculated using Cox proportional hazards methods, by birth weight and adjusted for birth cohort and sex. Over follow-up of 19,661 person-years, 61 people died of natural causes between age 15 and the censor date. Low birth weights (<2.5 kg) were associated with higher rates of natural death, with HR (95% CI) 1.76 (1.1–2.9, P=0.03), after adjustment for year of birth and sex. The effect was particularly prominent for deaths at <41 years of age, and with deaths from respiratory conditions/sepsis and unusual causes. A predisposing effect of low birth weight on adult deaths was confirmed. This phenomenon, occurring in the context of dramatically improved survivals of lower birth weight infants and children since the early 1960s, helps explain the current epidemic of chronic disease in Aboriginal people. Birth weights continue to improve, so excess deaths from this source should progressively be minimized.

Imprinted genes uniquely drive and support fetoplacental growth by controlling the allocation of maternal resources to the fetus and affecting the newborn’s growth. We previously showed that alterations of the placental imprinted gene expression are associated with suboptimal perinatal growth and respond to environmental stimuli including socio-economic determinants. At the same time, maternal psychosocial stress during pregnancy (MPSP) has been shown to affect fetal growth. Here, we set out to test the hypothesis that placental imprinted gene expression mediates the effects of MPSP on fetal growth in a well-characterized birth cohort, the Stress in Pregnancy (SIP) Study. We observed that mothers experiencing high MPSP deliver infants with lower birthweight (P=0.047). Among the 109 imprinted genes tested, we detected panels of placental imprinted gene expression of 23 imprinted genes associated with MPSP and 26 with birthweight. Among these genes, five imprinted genes (CPXM2, glucosidase alpha acid (GAA), GPR1, SH3 and multiple ankyrin repeat domains 2 (SHANK2) and THSD7A) were common to the two panels. In multivariate analyses, controlling for maternal age and education and gestational age at birth and infant gender, two genes, GAA and SHANK2, each showed a 22% mediation of MPSP on fetal growth. These data provide new insights into the role that imprinted genes play in translating the maternal stress message into a fetoplacental growth pattern.

An adverse early life environment is associated with increased cardiovascular disease in offspring. Work in animal models has shown that maternal undernutrition (UN) during pregnancy leads to hypertension in adult offspring, with effects thought to be mediated in part via altered renal function. We have previously shown that growth hormone (GH) treatment of UN offspring during the pre-weaning period can prevent the later development of cardiometabolic disorders. However, the mechanistic basis for these observations is not well defined. The present study examined the impact of GH treatment on renal inflammatory markers in adult male offspring as a potential mediator of these reversal effects. Female Sprague-Dawley rats were fed either a chow diet fed ad libitum (CON) or at 50% of CON intake (UN) during pregnancy. All dams were fed the chow diet ad libitum during lactation. CON and UN pups received saline (CON-S/UN-S) or GH (2.5 µg/g/day; CON-GH/UN-GH) from postnatal day 3 until weaning (p21). Post-weaning males were fed a standard chow diet for the remainder of the study (150 days). Histological analysis was performed to examine renal morphological characteristics, and gene expression of inflammatory and vascular markers were assessed. There was evidence of renal hypotrophy and reduced nephron number in the UN-S group. Tumour necrosis factor-α, monocyte chemoattractant protein-1 (MCP-1), intercellular adhesion molecular-1 and vascular cell adhesion molecule-1 gene expression was increased in UN-S offspring and normalized in the UN-GH group. These findings indicate that pre-weaning GH treatment has the potential to normalize some of the adverse renal and cardiovascular sequelae that arise as a consequence of poor maternal nutrition.

Epigenetic changes represent a potential mechanism underlying associations of early-life exposures and later life health outcomes. Population-based cohort studies starting in early life are an attractive framework to study the role of such changes. DNA methylation is the most studied epigenetic mechanism in population research. We discuss the application of DNA methylation in early-life population studies, some recent findings, key challenges and recommendations for future research. Studies into DNA methylation within the Developmental Origins of Health and Disease framework generally either explore associations between prenatal exposures and offspring DNA methylation or associations between offspring DNA methylation in early life and later health outcomes. Only a few studies to date have integrated prospective exposure, epigenetic and phenotypic data in order to explicitly test the role of DNA methylation as a potential biological mediator of environmental effects on health outcomes. Population epigenetics is an emerging field which has challenges in terms of methodology and interpretation of the data. Key challenges include tissue specificity, cell type adjustment, issues of power and comparability of findings, genetic influences, and exploring causality and functional consequences. Ongoing studies are working on addressing these issues. Large collaborative efforts of prospective cohorts are emerging, with clear benefits in terms of optimizing power and use of resources, and in advancing methodology. In the future, multidisciplinary approaches, within and beyond longitudinal birth and preconception cohorts will advance this complex, but highly promising, the field of research.

Recent work has implicated one type of horizontal strabismus (exotropia) as a risk factor for schizophrenia. This new insight raises questions about a potential common developmental origin of the two diseases. Seasonality of births is well established for schizophrenia. Seasonal factors such as light exposure affect eye growth and can cause vision abnormalities, but little is known about seasonality of births in strabismus. We examined birth seasonality in people with horizontal strabismus in a retrospective study in Washoe County, Nevada, and re-examined similar previously obtained data from Osaka, Japan. We then compared seasonal patterns of births between strabismus, refractive error, schizophrenia and congenital toxoplasmosis. Patients with esotropia had a significant seasonality of births, with a deficit in March, then increasing to an excess in September, while patients with exotropia had a distinctly different pattern, with an excess of births in July, gradually decreasing to a deficit in November. These seasonalities were statistically significant with either χ2 or Kolmogorov–Smirnov-type statistics. The birth seasonality of esotropia resembled that for hyperopia, with an increase in amplitude, while the seasonality for myopia involved a phase-shift. There was no correlation between seasonality of births between strabismus and congenital toxoplasmosis. The pattern of an excess of summer births for people with exotropia was remarkably similar to the well-established birth seasonality of one schizophrenia subtype, the deficit syndrome, but not schizophrenia as a whole. This suggests a testable hypothesis: that exotropia may be a risk factor primarily for the deficit type of schizophrenia.

Blood pressure (BP) tracks from childhood to adulthood, and early BP trajectories predict cardiovascular disease risk later in life. Excess postnatal weight gain is associated with vascular changes early in life. However, to what extent it is associated with children’s BP is largely unknown. In 853 healthy 5-year-old children of the Wheezing-Illnesses-Study-Leidsche-Rijn (WHISTLER) birth cohort, systolic (SBP) and diastolic BP (DBP) were measured, and z scores of individual weight gain rates adjusted for length gain rates were calculated using at least two weight and length measurements from birth until 3 months of age. Linear regression analyses were conducted to investigate the association between weight gain rates adjusted for length gain rates and BP adjusted for sex and ethnicity. Each standard deviation increase in weight gain rates adjusted for length gain rates was associated with 0.9 mmHg (95% CI 0.3, 1.5) higher sitting SBP after adjustment for confounders. Particularly in children in the lowest birth size decile, high excess weight gain was associated with higher sitting SBP values compared to children with low weight gain rates adjusted for length gain rates. BMI and visceral adipose tissue partly explained the association between excess weight gain and sitting SBP (β 0.5 mmHg, 95% CI −0.3, 1.3). Weight gain rates adjusted for length gain rates were not associated with supine SBP or DBP. Children with excess weight gain, properly adjusted for length gain, in the first three months of life, particularly those with a small birth size, showed higher sitting systolic BP at the age of 5 years.

Adverse exposures during fetal life and the postnatal period influence physical, cognitive and emotional development, and predispose to an increased risk of various chronic diseases throughout the life course. Findings from large observational studies in various populations and experimental animal studies have identified different modifiable risk factors in early life. Adverse maternal lifestyle factors, including overweight, unhealthy diet, sedentary behavior, smoking, alcohol consumption and stress in the preconception period and during pregnancy, are the most common modifiable risk factors leading to a suboptimal in-utero environment for fetal development. In the postnatal period, breastfeeding, infant growth and infant dietary intake are important modifiable factors influencing long-term offspring health outcomes. Despite the large amount of findings from observational studies, translation to lifestyle interventions seems to be challenging. Currently, randomized controlled trials focused on the influence of lifestyle interventions in these critical periods on short-term and long-term maternal and offspring health outcomes are scarce, have major limitations and do not show strong effects on maternal and offspring outcomes. New and innovative approaches are needed to move from describing these causes of ill-health to start tackling them using intervention approaches. Future randomized controlled lifestyle intervention studies and innovative observational studies, using quasi-experimental designs, are needed focused on the effects of an integrated lifestyle advice from preconception onwards on pregnancy outcomes and long-term health outcomes in offspring on a population level.

The objective of this study was to investigate the impact of the most commonly cited factors that may have influenced infants’ gut microbiota profiles at one year of age: mode of delivery, breastfeeding duration and antibiotic exposure. Barcoded V3/V4 amplicons of bacterial 16S-rRNA gene were prepared from the stool samples of 52 healthy 1-year-old Australian children and sequenced using the Illumina MiSeq platform. Following the quality checks, the data were processed using the Quantitative Insights Into Microbial Ecology pipeline and analysed using the Calypso package for microbiome data analysis. The stool microbiota profiles of children still breastfed were significantly different from that of children weaned earlier (P<0.05), independent of the age of solid food introduction. Among children still breastfed, Veillonella spp. abundance was higher. Children no longer breastfed possessed a more ‘mature’ microbiota, with notable increases of Firmicutes. The microbiota profiles of the children could not be differentiated by delivery mode or antibiotic exposure. Further analysis based on children’s feeding patterns found children who were breastfed alongside solid food had significantly different microbiota profiles compared to that of children who were receiving both breastmilk and formula milk alongside solid food. This study provided evidence that breastfeeding continues to influence gut microbial community even at late infancy when these children are also consuming table foods. At this age, any impacts from mode of delivery or antibiotic exposure did not appear to be discernible imprints on the microbial community profiles of these healthy children.

Fetal growth restriction (FGR) is a pregnancy condition in which fetal growth is suboptimal for gestation, and this population is at increased risk for type 2 diabetes as adults. In humans, maternal malnutrition and placental insufficiency are the most common causes of FGR, and both result in fetal undernutrition. We hypothesized that maternal nutrient restriction (MNR) in mice will cause FGR and alter glucose metabolism in adult offspring. Pregnant CD-1 mice were subjected to MNR (70% of average ad libitum) or control (ad libitum) from E6.5 to birth. Following birth, mice were fostered by mothers on ad libitum feeds. Weight, blood glucose, glucose tolerance and tissue-specific insulin sensitivity were assessed in male offspring. MNR resulted in reduced fetal sizes but caught up to controls by 3 days postnatal age. As adults, glucose intolerance was detected in 19% of male MNR offspring. At 6 months, liver size was reduced (P = 0.01), but pAkt-to-Akt ratios in response to insulin were increased 2.5-fold relative to controls (P = 0.004). These data suggest that MNR causes FGR and long-term glucose intolerance in a population of male offspring similar to human populations. This mouse model can be used to investigate the impacts of FGR on tissues of importance in glucose metabolism.

Fetal programming by androgen excess is hypothesized as one of the main factors contributing to the development of polycystic ovary syndrome (PCOS). PCOS is more than a reproductive disorder, as women with PCOS also show metabolic and other endocrine alterations. Since both ovarian and reproductive functions depend on energy balance, the alterations in metabolism may be related to reproductive alterations. The present study aimed to evaluate the effect of androgen excess during prenatal life on ovarian fuel sensors and its consequences on steroidogenesis. To this end, pregnant rats were hyperandrogenized with testosterone and the following parameters were evaluated in their female offspring: follicular development, PPARG levels, adipokines (including leptin, adiponectin, and chemerin as ovarian fuel sensors), serum gonadotropins (LH and FSH), the mRNA of their ovarian receptors, and the expression of steroidogenic mediators. At 60 days of age, the prenatally hyperandrogenized (PH) female offspring displayed both an irregular ovulatory phenotype and an anovulatory phenotype with altered follicular development and the presence of cysts. Both PH groups showed altered levels of both proteins and mRNA of PPARG and a different expression pattern of the adipokines studied. Although serum gonadotropins were not impaired, there were alterations in the mRNA levels of their ovarian receptors. The steroidogenic mediators Star, Cyp11a1, Cyp17a1, and Cyp19a1 were altered differently in each of the PH groups. We concluded that androgen excess during prenatal life leads to developmental programming effects that affect ovarian fuel sensors and steroidogenesis in a phenotype-specific way.

Complications of pregnancy remain key drivers of morbidity and mortality, affecting the health of both the mother and her offspring in the short and long term. There is lack of detailed understanding of the pathways involved in the pathology and pathogenesis of compromised pregnancy, as well as a shortfall of effective prognostic, diagnostic and treatment options. In many complications of pregnancy, such as in preeclampsia, there is an increase in uteroplacental vascular resistance. However, the cause and effect relationship between placental dysfunction and adverse outcomes in the mother and her offspring remains uncertain. In this review, we aim to highlight the value of gestational hypoxia-induced complications of pregnancy in elucidating underlying molecular pathways and in assessing candidate therapeutic options for these complex disorders. Chronic maternal hypoxia not only mimics the placental pathology associated with obstetric syndromes like gestational hypertension at morphological, molecular and functional levels, but also recapitulates key symptoms that occur as maternal and fetal clinical manifestations of these pregnancy disorders. We propose that gestational hypoxia provides a useful model to study the inter-relationship between placental dysfunction and adverse outcomes in the mother and her offspring in a wide array of examples of complicated pregnancy, such as in preeclampsia.

Environmental factors during perinatal life can lead to changes in the mammary gland, making it susceptible to cancer in adulthood. Breastfeeding has a special importance since it takes place at a critical period of growth and development of the newborn. We aimed to analyze if an appropriate lactation protects the offspring against mammary carcinogenesis during adult life and explore the mechanisms involved in the protective effect. One-day-old Sprague-Dawley female rats were randomly distributed in litters of three (L3), eight (L8) or 12 (L12) pups per dam, to induce a differential consumption of breast milk. At 55 days of age, the animals were treated with a single dose of dimethylbenzanthracene to study tumor latency, incidence and progression. Histological, immunohistochemical and Western blot studies were performed. We observed lower incidence and higher latency in L3 compared to the other groups. The mitotic index and expression of proliferating cell nuclear antigen (PCNA) was significantly augmented in tumors of L12 rats compared to L3 and L8, while the apoptotic index was augmented in tumors of L3 v. L12. Cleaved caspase 8 was significantly higher in tumors from L3 compared to L12. Tumors developed in L3 have a greater number of apoptotic bodies and a greater expression of caspase 8. These results demonstrate that the animals that maintained a higher intake of maternal milk (L3) presented lower incidence and greater tumor latency. Lower consumption of breast milk (L12) would increase tumor mitosis and the expression of PCNA, explaining the higher tumor incidence observed in this group.

Maternal gut microbiota is thought to be one of the important factors in the developmental origins of health and disease (DOHaD) concept, but the effects of maternal gut microbiota on foetal growth are not well known. In this study, the association between maternal gut microbiota and foetal growth was investigated. Maternal and newborn information, as well as stool samples at the third trimester of pregnancy, were obtained from 51 mother–newborn pairs from the Chiba study of Mother and Child Health (C-MACH). Gut microbiota was analysed by 16S rRNA sequencing of stool samples and short-chain fatty acids (SCFAs) in stool were analysed by gas chromatography-tandem mass spectrometry. After adjustment for covariates, it was found that maternal gut microbial diversity had a positive association with head circumference in newborn males (Chao 1: adjusted r = 0.515, p = 0.029). Genus Parabacteroides and genus Eggerthella showed negative associations with newborn head circumference and weight, respectively in males (genus Parabacteroides: adjusted r = −0.598, p = 0.009, genus Eggerthella: adjusted r = −0.481, p = 0.043). On the other hand, genus Streptococcus showed a negative association with newborn height in females (adjusted r = −0.413, p = 0.040). In addition, hexanoate was involved in the association between maternal gut microbiota and newborn anthropometrics in the univariate analysis, but not in the multivariate analysis. These data suggest that maternal gut microbiota has sex-specific effects on foetal growth. Maternal gut microbiota is an important factor for optimal intrauterine growth.

Maternal adversity and fetal glucocorticoid exposure has long-term effects on cardiovascular, metabolic and behavioral systems in offspring that can persist throughout the lifespan. These data, along with other environmental exposure data, implicate epigenetic modifications as potential mechanisms for long-term effects of maternal exposures on adverse health outcomes in offspring. Advances in microarray, sequencing and bioinformatic approaches have enabled recent studies to examine the genome-wide epigenetic response to maternal adversity. Studies of maternal exposures to xenobiotics such as arsenic and smoking have been performed at birth to examine fetal epigenomic signatures in cord blood relating to adult health outcomes. However, there have been no epigenomic studies examining these effects in animal models. On the other hand, to date, only a few studies of the effects of maternal psychosocial stress have been performed in human infants, and the majority of animal studies have examined epigenomic outcomes in adulthood. In terms of maternal exposure to excess glucocorticoids by synthetic glucocorticoid treatment, there has been no epigenetic study performed in humans and only a few studies undertaken in animal models. This review emphasizes the importance of examining biomarkers of exposure to adversity throughout development to identify individuals at risk and to target interventions. Thus, research performed at birth will be reviewed. In addition, potential subject characteristics associated with epigenetic modifications, technical considerations, the selection of target tissues and combining human studies with animal models will be discussed in relation to the design of experiments in this field of study.

While birthweight of offspring is associated with pre-pregnancy body mass index (BMI) and later risk of obesity, its mediating effect between the association of maternal pre-pregnancy BMI and offspring’s childhood anthropometrics has rarely been investigated. This study aimed to examine whether offspring birthweight is a mediator in the association between pre-pregnancy BMI and offspring’s childhood anthropometrics. The study included 1,618 mother–child pairs from the Australian Longitudinal Study on Women’s Health and Mothers and their Children’s Health Study. Children’s anthropometrics [mean age 8.6 (s.d. =3.0) years] were calculated from the mothers’ self-reported child weight and height measures. G-computation was used to estimate the natural direct and indirect (via birthweight) effects of pre-pregnancy BMI. In the fully adjusted model for maternal sociodemographic and lifestyle factors, the natural direct effects of pre-pregnancy obesity on child BMI-for-age, height-for-age, weight-for-age and weight-for-height outcomes were, β (95% confidence interval, CI), 0.75 (0.55, 0.95), 0.13 (−0.07, 0.32), 0.62 (0.44, 0.80) and 0.57 (0.24, 0.90), respectively. The corresponding natural indirect effects were 0.04 (−0.04, 0.12), −0.01 (−0.09, 0.07), −0.01 (−0.08, 0.07) and 0.09 (−0.05, 0.23). Similar results were observed for pre-pregnancy overweight and pre-pregnancy BMI as a continuous scale. Most of the effect of pre-pregnancy obesity on childhood weight-related anthropometric outcomes appears to be via a direct effect, not mediated through offspring’s birthweight.