Adverse birth outcomes, such as early gestational age and low birth weight, can have lasting effects on morbidity and mortality, with impacts that persist into adulthood. Identifying the maternal factors that contribute to adverse birth outcomes in the next generation is thus a priority. Epigenetic clocks, which have emerged as powerful tools for quantifying biological aging and various dimensions of physiological dysregulation, hold promise for clarifying relationships between maternal biology and infant health, including the maternal factors or states that predict birth outcomes. Nevertheless, studies exploring the relationship between maternal epigenetic age and birth outcomes remain few. Here, we attempt to replicate a series of analyses previously reported in a US-based sample, using a larger similarly aged sample (
Infants born preterm or small for gestational age have elevated rates of morbidity and mortality. Using birth certificate records in Texas from 2002 to 2004 and Environmental Protection Agency air pollution estimates, we relate the quantile functions of birth weight and gestational age to ozone exposure and multiple predictors, including parental age, race, and education level. We introduce a semi-parametric Bayesian quantile approach that models the full quantile function rather than just a few quantile levels. Our multilevel quantile function model establishes relationships between birth weight and the predictors separately for each week of gestational age and between gestational age and the predictors separately across Texas Public Health Regions. We permit these relationships to vary nonlinearly across gestational age, spatial domain and quantile level and we unite them in a hierarchical model via a basis expansion on the regression coefficients that preserves interpretability. Very low birth weight is a primary concern, so we leverage extreme value theory to supplement our model in the tail of the distribution. Gestational ages are recorded in completed weeks of gestation (integer-valued), so we present methodology for modeling quantile functions of discrete response data. In a simulation study we show that pooling information across gestational age and quantile level substantially reduces MSE of predictor effects. We find that ozone is negatively associated with the lower tail of gestational age in south Texas and across the distribution of birth weight for high gestational ages. Our methods are available in the R package BSquare.
more » « less- NSF-PAR ID:
- 10485088
- Publisher / Repository:
- Oxford University Press
- Date Published:
- Journal Name:
- Biometrics
- Volume:
- 71
- Issue:
- 2
- ISSN:
- 0006-341X
- Format(s):
- Medium: X Size: p. 508-519
- Size(s):
- ["p. 508-519"]
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract n = 296) of participants of a long-running study in the Philippines. New pregnancies were identified prospectively, dried blood spot samples were collected during the third trimester, and information was obtained on gestational age at delivery and offspring weight after birth. Genome-wide DNA methylation was assessed with the Infinium EPIC array. Using a suite of 15 epigenetic clocks, we only found one significant relationship: advanced age on the epigenetic clock trained on leptin predicted a significantly earlier gestational age at delivery (β = − 0.15,p = 0.009). Of the other 29 relationships tested predicting gestational age and offspring birth weight, none were statistically significant. In this sample of Filipino women, epigenetic clocks capturing multiple dimensions of biology and health do not predict birth outcomes in offspring. -
more » « less
Abstract Objectives Maternal experiences before pregnancy predict birth outcomes, a key indicator of health trajectories, but the timing and pathways for these effects are poorly understood. Here we test the hypothesis that maternal pre‐adult growth patterns predict pregnancy glucose and offspring fetal growth in Cebu, Philippines.
Methods Using multiple regression and path analysis, gestational age‐adjusted birthweight and variables reflecting infancy, childhood, and post‐childhood/adolescent weight gain (conditional weights) were used to predict pregnancy HbA1c and offspring birth outcomes among participants in the Cebu Longitudinal Health and Nutrition Survey.
Results Maternal early/mid‐childhood weight gain predicted birth weight, length, and head circumference in female offspring. Late‐childhood/adolescent weight gain predicted birth length, birth weight, skinfold thickness, and head circumference in female offspring, and head circumference in male offspring. Pregnancy HbA1c did not mediate relationships between maternal growth and birth size parameters.
Discussion In Cebu, maternal growth patterns throughout infancy, childhood, and adolescence predict fetal growth via a pathway independent of circulating glucose, with stronger impacts on female than male offspring, consistent with a role of developmental nutrition on offspring fetal growth. Notably, the strength of relationships followed a pattern opposite to what occurs in response to acute pregnancy stress, with strongest effects on head circumference and birth length and weakest on skinfolds. We speculate that developmental sensitivities are reversed for stable, long‐term nutritional cues that reflect average local environments. These findings are relevant to public health and life‐history theory as further evidence of developmental influences on health and resource allocation across the life course.
-
more » « less
Abstract Objectives Establishment and development of the infant gastrointestinal microbiome (GIM) varies cross‐culturally and is thought to be influenced by factors such as gestational age, birth mode, diet, and antibiotic exposure. However, there is little data as to how the composition of infants' households may play a role, particularly from a cross‐cultural perspective. Here, we examined relationships between infant fecal microbiome (IFM) diversity/composition and infants' household size, number of siblings, and number of other household members.
Materials and methods We analyzed 377 fecal samples from healthy, breastfeeding infants across 11 sites in eight different countries (Ethiopia, The Gambia, Ghana, Kenya, Peru, Spain, Sweden, and the United States). Fecal microbial community structure was determined by amplifying, sequencing, and classifying (to the genus level) the V1–V3 region of the bacterial 16S rRNA gene. Surveys administered to infants' mothers identified household members and composition.
Results Our results indicated that household composition (represented by the number of cohabitating siblings and other household members) did not have a measurable impact on the bacterial diversity, evenness, or richness of the IFM. However, we observed that variation in household composition categories did correspond to differential relative abundances of specific taxa, namely:
Lactobacillus ,Clostridium ,Enterobacter , andKlebsiella .Discussion This study, to our knowledge, is the largest cross‐cultural study to date examining the association between household composition and the IFM. Our results indicate that the social environment of infants (represented here by the proxy of household composition) may influence the bacterial composition of the infant GIM, although the mechanism is unknown. A higher number and diversity of cohabitants and potential caregivers may facilitate social transmission of beneficial bacteria to the infant gastrointestinal tract, by way of shared environment or through direct physical and social contact between the maternal–infant dyad and other household members. These findings contribute to the discussion concerning ways by which infants are influenced by their social environments and add further dimensionality to the ongoing exploration of social transmission of gut microbiota and the “old friends” hypothesis.
-
more » « less
Abstract Objective We investigated the relationship between early life growth patterns and blood telomere length (TL) in adulthood using conditional measures of lean and fat mass growth to evaluate potentially sensitive periods of early life growth.
Methods This study included data from 1562 individuals (53% male; age 20‐22 years) participating in the Cebu Longitudinal Health and Nutrition Survey, located in metropolitan Cebu, Philippines. Primary exposures included length‐for‐age
z ‐score (HAZ) and weight‐for‐agez ‐score (WAZ) at birth and conditional measures of linear growth and weight gain during four postnatal periods: 0‐6, 6‐12, and 12‐24 months, and 24 months to 8.5 years. TL was measured at ~21 years of age. We estimated associations using linear regression.Results The study sample had an average gestational age (38.5 ± 2 weeks) and birth size (HAZ = –0.2 ± 1.1, WAZ = –0.7 ± 1.0), but by age 8.5 years had stunted linear growth (HAZ = –2.1 ± 0.9) and borderline low weight (WAZ = –1.9 ± 1.0) relative to World Health Organization references. Heavier birth weight was associated with longer TL in early adulthood (
P = .03), but this association was attenuated when maternal age at birth was included in the model (P = .07). Accelerated linear growth between 6 and 12 months was associated with longer TL in adulthood (P = .006), whereas weight gain between 12 and 24 months was associated with shorter TL in adulthood (P = .047).Conclusions In Cebu, individuals who were born heavier have longer TL in early adulthood, but that birthweight itself may not explain the association. Findings suggest that childhood growth is associated with the cellular senescence process in adulthood, implying early life well‐being may be linked to adult health.
-
We derive the properties and demonstrate the desirability of a model-based method for estimating the spatially-varying effects of covariates on the quantile function. By modeling the quantile function as a combination of I-spline basis functions and Pareto tail distributions, we allow for flexible parametric modeling of the extremes while preserving non-parametric flexibility in the center of the distribution. We further establish that the model guarantees the desired degree of differentiability in the density function and enables the estimation of non-stationary covariance functions dependent on the predictors. We demonstrate through a simulation study that the proposed method produces more efficient estimates of the effects of predictors than other methods, particularly in distributions with heavy tails. To illustrate the utility of the model we apply it to measurements of benzene collected around an oil refinery to determine the effect of an emission source within the refinery on the distribution of the fence line measurements.more » « less
