Statistical methods to study timing of vulnerability with sparsely sampled data on environmental toxicants

Sex-specific associations of early postnatal blood copper levels with neurodevelopment at 2 years of age

Copper (Cu) is an essential trace element; however, it can be harmful in excess. Previous studies have shown that prenatal Cu levels may affect childhood neurodevelopment; however, studies focused on early postnatal Cu levels are limited. We studied 843 children born in Wuhan City and investigated the associations between early life Cu levels and neurodevelopment in 2-year-old children. Blood samples collected from children at 12 and 24 months of age were used to analyze Cu levels. Neurodevelopment was scored using the Bayley Scale of Children at 24 months of age. We found that a higher Cu level at 12 months of age was positively associated with mental development index (MDI) in boys (β = 6.75, 95 %CI: 1.12, 12.38). Further non-linear analysis showed an inverted U-shape association between Cu level at 20 months and PDI in boys, indicating that Cu levels may have an optimal concentration for neurodevelopment (p for overall association = 0.01, p for non-linear association < 0.01). In addition, all meaningful results mentioned above were observed only in boys, and a statistically significant sex-related modifying effect was observed (p < 0.05). In conclusion, this study repeated measures early life Cu levels and suggested sex-specific associations between early life Cu levels and neurodevelopment in 2-year-old children.

A spatially varying distributed lag model with application to an air pollution and term low birth weight study

Distributed lag models have been used to identify critical pregnancy periods of exposure (i.e. critical exposure windows) to air pollution in studies of pregnancy outcomes. However, much of the previous work in this area has ignored the possibility of spatial variability in the lagged health effect parameters that may result from exposure characteristics and/or residual confounding. We develop a spatially varying Gaussian process model for critical windows called 'SpGPCW' and use it to investigate geographic variability in the association between term low birth weight and average weekly concentrations of ozone and PM_2:5 during pregnancy by using birth records from North Carolina. SpGPCW is designed to accommodate areal level spatial correlation between lagged health effect parameters and temporal smoothness in risk estimation across pregnancy. Through simulation and a real data application, we show that the consequences of ignoring spatial variability in the lagged health effect parameters include less reliable inference for the parameters and diminished ability to identify true critical window sets, and we investigate the use of existing Bayesian model comparison techniques as tools for determining the presence of spatial variability. We find that exposure to PM_2:5 is associated with elevated term low birth weight risk in selected weeks and counties and that ignoring spatial variability results in null associations during these periods. An R package (SpGPCW) has been developed to implement the new method.

Early-life triclosan exposure and parent-reported behavior problems in 8-year-old children

BACKGROUND

Triclosan exposure may decrease circulating thyroxine levels or cause neuron apoptosis, which in turn may adversely affect neurodevelopment. However, few studies have examined the association of early life triclosan exposure with child behavior.

OBJECTIVE

To quantify the association between early-life triclosan exposure and child behavior at age 8-years in 202 mother-child pairs from the HOME study (Cincinnati, OH; enrolled: 2003-2006).

METHODS

We quantified urinary triclosan concentrations up to 3 times in mothers (16-weeks, 26-weeks, and delivery) and up to 6 times in children (1, 2, 3, 4, 5, and 8 years). Parents rated children's problem behaviors at age 8-years using the Behavioral Assessment System for Children-2 (BASC-2). Adjusting for covariates and accounting for exposure measurement error, we estimated changes in behavior problem scores per 10-fold increase in mean gestational and childhood triclosan concentrations. In addition, we estimated sex-specific associations.

RESULTS

Child sex modified the association of gestational and childhood triclosan with several BASC-2 scales (sex × triclosan p-values < 0.2). In boys, increasing gestational triclosan was associated with higher behavioral symptom index (β: 4.5; 95% CI: 1.0, 8.1), externalizing problems (β: 5.0; 95% CI: 1.2, 9.0), attention problem (β: 6.6; 95% CI: 2.4, 11), hyperactivity (β: 6.4; 95% CI: 2.1, 11), and somatization (β: 3.8; 95% CI: 0.3, 7.3) scores. In contrast, triclosan-BASC-2 associations in girls were generally null and not statistically significant. We observed similar patterns of associations between childhood triclosan and these same behavioral scores; however, their magnitude decreased substantially after adjusting for gestational triclosan and associations were not statistically significant.

CONCLUSION

In this cohort, increasing gestational and childhood urinary triclosan concentrations were associated with higher behavior problem scores in 8-year old boys, but not girls.

Statistical Methodology in Studies of Prenatal Exposure to Mixtures of Endocrine-Disrupting Chemicals: A Review of Existing Approaches and New Alternatives

BACKGROUND

Prenatal exposures to endocrine-disrupting chemicals (EDCs) during critical developmental windows have been implicated in the etiologies of a wide array of adverse perinatal and pediatric outcomes. Epidemiological studies have concentrated on the health effects of individual chemicals, despite the understanding that EDCs act together via common mechanisms, that pregnant women are exposed to multiple EDCs simultaneously, and that substantial toxicological evidence of adverse developmental effects has been documented. There is a move toward multipollutant models in environmental epidemiology; however, there is no current consensus on appropriate statistical methods.

OBJECTIVES

We aimed to review the statistical methods used in these studies, to identify additional applicable methods, and to determine the strengths and weaknesses of each method for addressing the salient statistical and epidemiological challenges.

METHODS

We searched Embase, MEDLINE, and Web of Science for epidemiological studies of endocrine-sensitive outcomes in the children of mothers exposed to EDC mixtures during pregnancy and identified alternative statistical methods from the wider literature.

DISCUSSION

We identified 74 studies and analyzed the methods used to estimate mixture health effects, identify important mixture components, account for nonmonotonicity in exposure–response relationships, assess interactions, and identify windows of exposure susceptibility. We identified both frequentist and Bayesian methods that are robust to multicollinearity, performing shrinkage, variable selection, dimension reduction, statistical learning, or smoothing, including methods that were not used by the studies included in our review.

CONCLUSIONS

Compelling motivation exists for analyzing EDCs as mixtures, yet many studies make simplifying assumptions about EDC additivity, relative potency, and linearity, or overlook the potential for bias due to asymmetries in chemical persistence. We discuss the potential impacts of these choices and suggest alternative methods to improve analyses of prenatal exposure to EDC mixtures. https://doi.org/10.1289/EHP2207.

Identifying Vulnerable Periods of Neurotoxicity to Triclosan Exposure in Children

BACKGROUND

Exposure to triclosan, an endocrine disrupting chemical, may affect thyroid hormone homeostasis and adversely affect neurodevelopment.

OBJECTIVE

Using a longitudinal pregnancy and birth cohort, we investigated associations between triclosan exposures during different time windows, and cognitive test scores at 8 y of age in 198 children from the HOME Study.

METHODS

We quantified triclosan in urine samples from mother-child pairs up to nine times between the second trimester of gestation and 8 y of age. The Wechsler Intelligence Scale for Children-IV [i.e., Full-Scale Intelligence Quotient (IQ)] assessment was administered to HOME Study children at 8 y of age. We estimated covariate-adjusted triclosan-IQ associations at each visit. We also tested whether associations between triclosan concentrations and cognitive test scores varied among exposure at different time periods.

RESULTS

Full-Scale IQ was not significantly associated with urinary triclosan concentrations during gestation or childhood but was significantly associated with a 10-fold increase in maternal urinary triclosan concentration at delivery [-4.5 points (95% CI: -7.0, -2.0)]. Perceptual Reasoning Index (PRI) scores were significantly decreased in association with urinary triclosan concentrations at delivery and at 2 y of age. Associations between repeated triclosan concentrations and cognitive test scores significantly varied among exposure at different time periods for Full-Scale IQ, PRI, Verbal Comprehension Index, and Working Memory (triclosan-visit interaction p≤0.04).

CONCLUSION

Urinary triclosan concentrations at delivery, but not during mid to late pregnancy and childhood, were associated with significantly lower children's cognitive test scores at 8 y of age in this cohort of U.S. children. https://doi.org/10.1289/EHP2777.

Faster ticking rate of the epigenetic clock is associated with faster pubertal development in girls

Epigenetic age is an indicator of biological aging, capturing the impact of environmental and behavioral influences across time on cellular function. Deviance between epigenetic age and chronological age (AgeAccel) is a predictor of health. Pubertal timing has similarly been associated with cancer risk and mortality rate among females. We examined the association between AgeAccel and pubertal timing and adolescent breast composition in the longitudinal Growth and Obesity Cohort Study. AgeAccel was estimated in whole blood using the Horvath method at breast Tanner 2 (B2) and 4 (B4). Total breast volume, absolute fibro-glandular volume (FGV), and %FGV were evaluated at B4 using dual X-ray absorptiometry. The impact of AgeAccel (mean: 0; SD: 3.78) across puberty on the time to breast development (thelarche), menarche, and pubertal tempo (thelarche to menarche) was estimated using accelerated failure time models; generalized estimating equations were used to evaluate associations with breast density. A five-year increase in average adolescent AgeAccel was associated with a significant decrease in time to menarche [hazard ratio (HR): 1.37; 95% confidence interval (CI): 1.04, 1.80] adjusting for birth weight, maternal pre-pregnancy body mass index, maternal height, maternal education, B2 height, fat percentage, and cell composition. AgeAccel displayed a stronger inverse association with pubertal tempo (HR: 1.48; 95% CI: 1.10, 1.99). A five-year increase in AgeAccel was associated with 5% greater %FGV, adjusting for B4 percent body fat, and maternal traits (95% CI: 1.01, 1.10). Our study provides unique insight into the influence of AgeAccel on pubertal development in girls, which may have implications for adult health.

Challenges to studying the health effects of early life environmental chemical exposures on children's health

Epidemiological studies play an important role in quantifying how early life environmental chemical exposures influence the risk of childhood diseases. These studies face at least four major challenges that can produce noise when trying to identify signals of associations between chemical exposure and childhood health. Challenges include accurately estimating chemical exposure, confounding from causes of both exposure and disease, identifying periods of heightened vulnerability to chemical exposures, and determining the effects of chemical mixtures. We provide recommendations that will aid in identifying these signals with more precision.

Potential for Bias When Estimating Critical Windows for Air Pollution in Children's Health

Evidence supports an association between maternal exposure to air pollution during pregnancy and children's health outcomes. Recent interest has focused on identifying critical windows of vulnerability. An analysis based on a distributed lag model (DLM) can yield estimates of a critical window that are different from those from an analysis that regresses the outcome on each of the 3 trimester-average exposures (TAEs). Using a simulation study, we assessed bias in estimates of critical windows obtained using 3 regression approaches: 1) 3 separate models to estimate the association with each of the 3 TAEs; 2) a single model to jointly estimate the association between the outcome and all 3 TAEs; and 3) a DLM. We used weekly fine-particulate-matter exposure data for 238 births in a birth cohort in and around Boston, Massachusetts, and a simulated outcome and time-varying exposure effect. Estimates using separate models for each TAE were biased and identified incorrect windows. This bias arose from seasonal trends in particulate matter that induced correlation between TAEs. Including all TAEs in a single model reduced bias. DLM produced unbiased estimates and added flexibility to identify windows. Analysis of body mass index z score and fat mass in the same cohort highlighted inconsistent estimates from the 3 methods.

Prenatal Nitrate Exposure and Childhood Asthma. Influence of Maternal Prenatal Stress and Fetal Sex

RATIONALE

Impact of ambient pollution upon children's asthma may differ by sex, and exposure dose and timing. Psychosocial stress can also modify pollutant effects. These associations have not been examined for in utero ambient nitrate exposure.

OBJECTIVES

We implemented Bayesian-distributed lag interaction models to identify sensitive prenatal windows for the influence of nitrate (NO3-) on child asthma, accounting for effect modification by sex and stress.

METHODS

Analyses included 752 mother-child dyads. Daily ambient NO3- exposure during pregnancy was derived using a hybrid chemical transport (Geos-Chem)/land-use regression model and natural log transformed. Prenatal maternal stress was indexed by a negative life events score (high [>2] vs. low [≤2]). The outcome was clinician-diagnosed asthma by age 6 years.

MEASUREMENTS AND MAIN RESULTS

Most mothers were Hispanic (54%) or black (29%), had a high school education or less (66%), never smoked (80%), and reported low prenatal stress (58%); 15% of children developed asthma. BDILMs adjusted for maternal age, race, education, prepregnancy obesity, atopy, and smoking status identified two sensitive windows (7-19 and 33-40 wk gestation), during which increased NO3- was associated with greater odds of asthma, specifically among boys born to mothers reporting high prenatal stress. Cumulative effects of NO3- across pregnancy were also significant in this subgroup (odds ratio = 2.64, 95% confidence interval = 1.27-5.39; per interquartile range increase in ln NO3-).

CONCLUSIONS

Prenatal NO3- exposure during distinct sensitive windows was associated with incident asthma in boys concurrently exposed to high prenatal stress.

Early life bisphenol A exposure and neurobehavior at 8years of age: Identifying windows of heightened vulnerability

BACKGROUND

Early life BPA exposure could affect neurobehavior, but few studies have investigated whether there are developmental periods when the fetus or child is more vulnerable to these potential effects.

OBJECTIVES

We explored windows of vulnerability to BPA exposure in a multiethnic cohort of 228 mothers and their children from Cincinnati, Ohio.

METHODS

We measured urinary BPA concentrations at up to two prenatal and six postnatal time points from the 2nd trimester of pregnancy until the child was age 8years. At age 8years, we administered the Behavioral Assessment System for Children-2 (BASC-2), Behavior Rating Inventory of Executive Function, and Wechsler Intelligence Scale for Children-IV. We estimated covariate-adjusted differences in composite scores from each instrument using a multiple informant model designed to identify heightened windows of vulnerability.

RESULTS

Among all children, there was not strong evidence that the associations between BPA and neurobehavior varied by the timing of exposure (Visit x BPA p-values≥0.16). However, child sex modified the associations of repeated BPA measures with BASC-2 scores (Visit x Sex x BPA p-values=0.02-0.23). For example, each 10-fold increase in prenatal BPA was associated with more externalizing behaviors in girls (β=6.2, 95% CI: 0.8, 11.6), but not boys (β=-0.8, 95% CI: -5.0, 3.4). In contrast, a 10-fold increase in 8-year BPA was associated with more externalizing behaviors in boys (β=3.9, 95% CI: 0.6, 7.2), but not girls (β=0.3, 95% CI: -3.5, 4.1).

CONCLUSIONS

We found that sex-dependent associations between BPA and child neurobehavior may depend on the timing of BPA exposure.

Prenatal and postnatal polybrominated diphenyl ether (PBDE) exposure and measures of inattention and impulsivity in children

Exposure to polybrominated diphenyl ethers (PBDEs) during fetal development may be associated with deficits in attention and impulse control. However, studies examining postnatal PBDE exposures and inattention and impulsivity have been inconsistent. Using data from 214 children in the Health Outcomes and Measures of the Environment (HOME) Study, a prospective pregnancy and birth cohort with enrollment from 2003 to 2006 in the Greater Cincinnati Area, we investigated the relationship of both prenatal and postnatal PBDE exposures with attention and impulse control. Serum PBDEs were measured at 16±3weeks of gestation and during childhood at 1, 2, 3, 5, and 8years. We assessed children's attention and impulse control using the Conners' Continuous Performance Test-Second Edition (CPT-II) at 8years. We used multiple informant models to estimate associations of repeated PBDE measures with inattention and impulsivity. There was a pattern of associations between PBDEs and poorer performance on CPT-II measures of attention. For BDE-153, adverse associations extended to exposures at preschool and kindergarten ages; ten-fold increases in exposure were associated with higher omission errors (BDE-153 at 3years: β=4.0 [95% CI: -2.4, 10.5]; at 5years: β=4.6 [95% CI: -2.8, 12.0]; at 8years: β=4.1 [95% CI: -3.4, 11.5]). Longer hit reaction times, indicated by the exponential part of the hit reaction curve, were also observed with 10-fold increases in BDE-153 during the prenatal period and throughout childhood (Prenatal: β=15.0 milliseconds (ms) [95% CI: -15.8, 45.8]; 5years: β=20.6ms [95% CI: -20.8, 61.9]; 8years: β=28.6ms [95% CI: -12.1, 69.4]). Significant impairment in discriminability, as indicated by detectability (d'), between targets and non-targets was also noted with 5 and 8-year PBDE concentrations. Associations between PBDEs and inattention significantly differed by child sex, with males performing more poorly than females with regard to omission errors and measures of reaction times. Collectively, these results do not strongly support that PBDEs are associated with poorer impulse and attention control among 8year old children. However, there may be a possible relationship between prenatal and concurrent PBDEs and inattention, which requires additional research.

Early-Life Phthalate Exposure and Adiposity at 8 Years of Age

BACKGROUND

Early-life phthalate exposure may influence child adiposity, but prior studies have not determined if there are periods of enhanced vulnerability to phthalates.

OBJECTIVE

To examine the relationship between child adiposity at 8 y of age and repeated urinary biomarkers of phthalate exposure from gestation through childhood to determine if there are distinct periods of vulnerability.

METHODS

In 219 mother-child pairs from Cincinnati, Ohio, we quantified nine urinary phthalate metabolites up to two times prenatally and six times from 1-8 y of age. We measured child body mass index (BMI), waist circumference, and percent body fat at 8 y of age. To identify periods of vulnerability, we used two statistical methods to estimate phthalate-adiposity associations at each visit, test differences in phthalate-adiposity associations across visits, and model trajectories of phthalate concentrations for children at different levels of adiposity.

RESULTS

Prenatal phthalate concentrations were not associated with excess child adiposity. Monobenzyl phthalate (MBzP) concentrations during pregnancy and childhood were inversely associated with adiposity. The associations of di(2-ethylhexyl) phthalate (∑DEHP) metabolites and monoethyl phthalate (MEP) with child adiposity depended on the timing of exposure. A 10-fold increase in ∑DEHP at 1 and 5 y was associated with a 2.7% decrease [95% confidence interval (CI): -4.8, -0.5] and 2.9% increase (95% CI: 0.3, 5.5) in body fat, respectively. MEP concentrations at 5 and 8 y of age were associated with higher child adiposity, but earlier childhood concentrations were not.

CONCLUSION

In this cohort, we did not find evidence of an obesogenic effect of prenatal phthalate exposure. Positive associations between postnatal MEP and ∑DEHP concentrations depended on the timing of exposure. https://doi.org/10.1289/EHP1022.

Statistical methods for modeling repeated measures of maternal environmental exposure biomarkers during pregnancy in association with preterm birth

BACKGROUND

It is of critical importance to evaluate the role of environmental chemical exposures in premature birth. While a number of studies investigate this relationship, most utilize single exposure measurements during pregnancy in association with the outcome. The studies with repeated measures of exposure during pregnancy employ primarily cross-sectional analyses that may not be fully leveraging the power and additional information that the data provide.

METHODS

We examine 9 statistical methods that may be utilized to estimate the relationship between a longitudinal exposure and a binary, non-time-varying outcome. To exemplify these methods we utilized data from a nested case-control study examining repeated measures of urinary phthalate metabolites during pregnancy in association with preterm birth.

RESULTS

The methods summarized may be useful for: 1) Examining sensitive windows of exposure in association with an outcome; 2) Summarizing repeated measures to estimate the relationship between average exposure and an outcome; 3) Identifying acute exposures that may be relevant to the outcome; and 4) Understanding the contribution of temporal patterns in exposure levels to the outcome of interest. In the study of phthalates, changes in urinary metabolites over pregnancy did not appear to contribute significantly to preterm birth, making summary of average exposure across gestation optimal given the current design.

CONCLUSIONS

The methods exemplified may be of great use in future epidemiologic research projects intended to: 1) Elucidate the complex relationships between environmental chemical exposures and preterm birth; 2) Investigate biological mechanisms in prematurity using repeated measures of maternal factors throughout pregnancy; and 3) More generally, address the relationship between a longitudinal predictor and a binary, non-time-varying outcome.

Windows of lead exposure sensitivity, attained height, and body mass index at 48 months

OBJECTIVE

To examine longitudinal associations of prenatal, infancy, and early childhood lead exposure during sensitive periods with height and body mass index (BMI).

STUDY DESIGN

A total of 773 participants were recruited between 1994 and 2005 in Mexico City. Lead exposure history categories were constructed for the prenatal period (maternal patellar lead concentration) and for infancy and childhood (mean child blood lead concentration at birth to 24 months and 30-48 months, respectively). Linear regression models were used to study lead exposure history with height and BMI at 48 months.

RESULTS

Mean height at age 48 months was significantly lower in children with a blood lead level exceeding the median during infancy (-0.84 cm; 95% CI, -1.42 to -0.25) than in children with a level below the median. Prenatal lead exposure was not associated with height at 48 months. Results for attained BMI generally trended in the same direction as for height.

CONCLUSION

Our findings suggest an effect of lead exposure early in life on height attainment at 48 months, with the exposure window of greatest sensitivity in infancy.